ChangeLog revision 1.1.1.8
1 commit 80514b7b1b88f46f3965ff023035dea83b57e652 2 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3 Date: Fri Sep 14 17:48:49 2018 +0200 4 5 Bump version for 18.1.0 release 6 7 commit 2d58830c3feafc54dccc0b7bf761a466437d4a09 8 Author: Michel Dnzer <michel.daenzer (a] amd.com> 9 Date: Wed Sep 5 18:23:29 2018 +0200 10 11 Fix uninitialized use of local variable pitch in radeon_setup_kernel_mem 12 13 Fixes server reset. 14 15 Pointed out by clang: 16 17 ../../src/radeon_kms.c:2721:9: warning: variable 'pitch' is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized] 18 if (!info->front_buffer) { 19 ^~~~~~~~~~~~~~~~~~~ 20 ../../src/radeon_kms.c:2765:27: note: uninitialized use occurs here 21 pScrn->displayWidth = pitch / cpp; 22 ^~~~~ 23 24 commit 5d5d883496842da84d9418e91cb13454751da625 25 Author: Michel Dnzer <michel.daenzer (a] amd.com> 26 Date: Fri Sep 7 18:16:22 2018 +0200 27 28 Bail early from drm_wait_pending_flip if there's no pending flip 29 30 No need to process any events in that case. 31 32 (Ported from amdgpu commit ca5eb9894fff153c0a1df7bdc4a4745713309e27) 33 34 commit 4c7d5e50a5e469a541bc463cecb505fe850c0824 35 Author: Michel Dnzer <michel.daenzer (a] amd.com> 36 Date: Wed Sep 5 11:29:43 2018 +0200 37 38 Don't use xorg_list_for_each_entry_safe for signalled flips 39 40 drm_wait_pending_flip can get called from drm_handle_event, in which 41 case xorg_list_for_each_entry_safe can end up processing the same entry 42 in both. To avoid this, just process the first list entry until the list 43 is empty. 44 45 (Ported from amdgpu commit 26770be44b89b83bf39c28f2fe284c8cb92ed0c0) 46 47 commit 87b9a3e516d19dd1b89a64f6cac990fae53fc1b3 48 Author: Michel Dnzer <michel.daenzer (a] amd.com> 49 Date: Wed Sep 5 11:27:25 2018 +0200 50 51 Always delete entry from list in drm_queue_handler 52 53 We left entries without a handler hook in the list, so the list could 54 keep taking longer to process and use up more memory. 55 56 (Ported from amdgpu commit 7eea3e2cd74eed22e982319144e18ae5b1087b78) 57 58 commit de88ea2755611bdcb18d91d8234d2ab5be8ff2e9 59 Author: Michel Dnzer <michel.daenzer (a] amd.com> 60 Date: Fri Aug 24 11:00:16 2018 +0200 61 62 glamor: Handle ihandle == -1 in radeon_glamor_set_shared_pixmap_backing 63 64 commit 84bde2df5b453f8aab35aa18b0cf1f20b8d20488 65 Author: Michel Dnzer <michel.daenzer (a] amd.com> 66 Date: Fri Aug 24 10:58:51 2018 +0200 67 68 EXA: Handle ihandle == -1 in RADEONEXASharePixmapBacking 69 70 commit 1799680f7bd84e0618f34f4c7486799521ddaf83 71 Author: Michel Dnzer <michel.daenzer (a] amd.com> 72 Date: Fri Aug 24 10:56:58 2018 +0200 73 74 Handle ihandle == -1 in radeon_set_shared_pixmap_backing 75 76 It means to stop using the shared pixmap backing. 77 78 Fixes crash when changing PRIME slave output configuration. 79 80 commit 3c42bd04800211f64a17640c7ce7c4111ffd4978 81 Author: Michel Dnzer <michel.daenzer (a] amd.com> 82 Date: Fri Aug 24 10:55:49 2018 +0200 83 84 EXA: Handle NULL BO pointer in radeon_set_pixmap_bo 85 86 commit cba8fe4d64819aaa8ba516aa68dbe6d2aa153046 87 Author: Michel Dnzer <michel.daenzer (a] amd.com> 88 Date: Thu Aug 23 18:59:18 2018 +0200 89 90 Use AC_CONFIG_MACRO_DIR instead of AC_CONFIG_MACRO_DIRS 91 92 Older versions of autoconf only supported the former. 93 94 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 95 96 commit 7b01c10137aba24c8f61dd9b2a19ea257ad24371 97 Author: Michel Dnzer <michel.daenzer (a] amd.com> 98 Date: Thu Aug 23 18:42:28 2018 +0200 99 100 Add m4 directory 101 102 Although normally it only warns about it, under some circumstances, 103 aclocal can error out if this directory doesn't exist. 104 105 Reported-by: John Lumby <johnlumby (a] hotmail.com> 106 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 107 108 commit 60cd28bbbe4f9db391b4f27c5ff9289d5c2bef40 109 Author: Michel Dnzer <michel.daenzer (a] amd.com> 110 Date: Thu Aug 16 18:06:31 2018 +0200 111 112 Remove drmmode_crtc_private_rec::present_vblank_* related code 113 114 Not needed anymore with the more robust mechanisms for preventing nested 115 drmHandleEvent calls introduced in the previous changes. 116 117 (Ported from amdgpu commit 85cd8eef0cbed7b409b07f58d76dacd34aa3ddea) 118 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 119 120 commit c42f6e2e61d166c8d3ef3fcad175d7050a00288b 121 Author: Michel Dnzer <michel.daenzer (a] amd.com> 122 Date: Fri Jul 20 16:56:22 2018 +0200 123 124 Defer vblank event handling while waiting for a pending flip 125 126 This is to avoid submitting more flips while we are waiting for pending 127 ones to complete. 128 129 (Ported from amdgpu commit e52872da69ecc84dafb3355839e35b0383f0d228) 130 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 131 132 commit ba83a866af5a3784fc4822614375cc081e93197c 133 Author: Michel Dnzer <michel.daenzer (a] amd.com> 134 Date: Thu Aug 16 17:44:45 2018 +0200 135 136 Add radeon_drm_handle_event wrapper for drmHandleEvent 137 138 Instead of processing DRM events directly from drmHandleEvent's 139 callbacks, there are three phases: 140 141 1. drmHandleEvent is called, and signalled events are re-queued to 142 _signalled lists from its callbacks. 143 2. Signalled page flip completion events are processed. 144 3. Signalled vblank events are processed. 145 146 This should make sure that we never call drmHandleEvent from one of its 147 callbacks, which would usually result in blocking forever. 148 149 (Ported from amdgpu commit 739181c8d3334ff14b5a607895dfdeb29b0d9020) 150 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 151 152 commit 93621e408c17dd9e082236c17f051c06558d7f4d 153 Author: Michel Dnzer <michel.daenzer (a] amd.com> 154 Date: Thu Aug 16 17:27:06 2018 +0200 155 156 Add radeon_drm_wait_pending_flip function 157 158 Replacing the drmmode_crtc_wait_pending_event macro. 159 160 (Ported from amdgpu commit 6029794e8a35417faf825491a89b85f713c77fc1) 161 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 162 163 commit ecdf0b7ec9378bc386ce8276f16fb16d09f72812 164 Author: Michel Dnzer <michel.daenzer (a] amd.com> 165 Date: Thu Aug 16 17:14:42 2018 +0200 166 167 Move DRM event queue related initialization to radeon_drm_queue_init 168 169 And make radeon_drm_queue_handler not directly accessible outside of 170 radeon_drm_queue.c. 171 172 (Ported from amdgpu commit 0148283984c77f7a6e97026edc3093497547e0a4) 173 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 174 175 commit 824189b3da9edc33e1a4f5c6130a043da73c1a4c 176 Author: Michel Dnzer <michel.daenzer (a] amd.com> 177 Date: Thu Aug 16 18:22:27 2018 +0200 178 179 Use correct FB handle in radeon_do_pageflip 180 181 We were always using the handle of the client provided FB, which 182 prevented RandR transforms from working, and could result in a black 183 screen. 184 185 Bugzilla: https://bugs.freedesktop.org/107519 186 Fixes: 740f0850f1e4 "Store FB for each CRTC in drmmode_flipdata_rec" 187 (Ported from amdgpu commit f6cd72e64e85896b6d155bee0930e59771dcb701) 188 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 189 190 commit 740f0850f1e40403c8dd727e074eae36caeb1f63 191 Author: Michel Dnzer <michel.daenzer (a] amd.com> 192 Date: Thu Aug 2 18:49:48 2018 +0200 193 194 Store FB for each CRTC in drmmode_flipdata_rec 195 196 We were only storing the FB provided by the client, but on CRTCs with 197 TearFree enabled, we use a separate FB. This could cause 198 drmmode_flip_handler to fail to clear drmmode_crtc->flip_pending, which 199 could result in a hang when waiting for the pending flip to complete. We 200 were trying to avoid that by always clearing drmmode_crtc->flip_pending 201 when TearFree is enabled, but that wasn't reliable, because 202 drmmode_crtc->tear_free can already be FALSE at this point when 203 disabling TearFree. 204 205 Now that we're keeping track of each CRTC's flip FB separately, 206 drmmode_flip_handler can reliably clear flip_pending, and we no longer 207 need the TearFree hack. 208 209 (Ported from amdgpu commit 9b6782c821e0bdc53336d98f87ddde752faf7902) 210 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 211 212 commit ef2a6b818fa47ad571bb0bc105aa8193d51a2190 213 Author: Michel Dnzer <michel.daenzer (a] amd.com> 214 Date: Thu Aug 2 18:48:04 2018 +0200 215 216 glamor: Set RADEON_CREATE_PIXMAP_DRI2 for DRI3 pixmaps 217 218 Not doing this resulted in falling back to software for DRI3 client 219 presentation operations with ShadowPrimary. 220 221 (Ported from amdgpu commit 2989d40ef74d9966e8e8df2ef7727b2cc48d4960) 222 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 223 224 commit db28d35ce9fd07a2a4703f3df0633d4c8291ff9b 225 Author: Michel Dnzer <michel.daenzer (a] amd.com> 226 Date: Fri Jul 27 11:30:57 2018 +0200 227 228 glamor: Use glamor_egl_create_textured_pixmap_from_gbm_bo when possible 229 230 Inspired by the modesetting driver. 231 232 Fixes screen pixmap corruption with Xorg < 1.20, and as a bonus, 233 simplifies the code slightly. 234 235 Bugzilla: https://bugs.freedesktop.org/107385 236 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 237 238 commit 7de4ea15880565c6536ed6859fd32ec89ce0b4c0 239 Author: Michel Dnzer <michel.daenzer (a] amd.com> 240 Date: Fri Jul 27 09:13:12 2018 +0200 241 242 Bump version to 18.0.99 243 244 Forgot to do this after the 18.0.1 release. 245 246 commit fed411290b696ecf9f0e799ec4bad054e4a95017 247 Author: Michel Dnzer <michel.daenzer (a] amd.com> 248 Date: Tue Jul 24 17:28:46 2018 +0200 249 250 Use strcpy for RandR output property names 251 252 Instead of strncpy with the string length. Avoids new warnings with GCC 253 8: 254 255 ../../src/drmmode_display.c: In function drmmode_output_create_resources: 256 ../../src/drmmode_display.c:2240:2: warning: strncpy output truncated before terminating nul copying 8 bytes from a string of the same length [-Wstringop-truncation] 257 strncpy(tearfree_prop->name, "TearFree", 8); 258 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 259 ../../src/drmmode_display.c:2244:2: warning: strncpy output truncated before terminating nul copying 3 bytes from a string of the same length [-Wstringop-truncation] 260 strncpy(tearfree_prop->enums[0].name, "off", 3); 261 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 262 ../../src/drmmode_display.c:2245:2: warning: strncpy output truncated before terminating nul copying 2 bytes from a string of the same length [-Wstringop-truncation] 263 strncpy(tearfree_prop->enums[1].name, "on", 2); 264 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 265 ../../src/drmmode_display.c:2247:2: warning: strncpy output truncated before terminating nul copying 4 bytes from a string of the same length [-Wstringop-truncation] 266 strncpy(tearfree_prop->enums[2].name, "auto", 4); 267 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 268 269 (Ported from amdgpu commit f3b2ed37d683f8616a0a31ff63133ddb8fe1a4a3) 270 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 271 272 commit d258413c956cd46a237d4b5b94a35817ca7ce8f4 273 Author: Michel Dnzer <michel.daenzer (a] amd.com> 274 Date: Mon Jul 23 18:54:40 2018 +0200 275 276 Remove drmmode_terminate_leases 277 278 The RandR screen private is already freed when our CloseScreen runs, so 279 this can't do anything useful. This cleanup has to be done by the X 280 server itself. 281 282 (Ported from amdgpu commit 5f06d6b8ba570b500956ad26fee711d5ac427818) 283 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 284 285 commit ffdbd46b0abcf8cbe73eb8930ddcd7fd2485d9b5 286 Author: Michel Dnzer <michel.daenzer (a] amd.com> 287 Date: Mon Jul 23 11:02:32 2018 +0200 288 289 Hardcode "non-desktop" RandR property name 290 291 It's a bit silly to require current randrproto just for this definition, 292 which can't really change anyway. 293 294 Suggested-by: Qiang Yu <qiang.yu (a] amd.com> 295 (Ported from amdgpu commit ae8e02c6fc4ef5d5340b8cd4739e66b19b9e3386) 296 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 297 298 commit 64bd009d17986c71726724859f100ad40a6e3096 299 Author: Michel Dnzer <michel.daenzer (a] amd.com> 300 Date: Thu Jul 19 11:33:42 2018 +0200 301 302 Don't allocate drmmode_output->props twice 303 304 This was accidentally duplicated when porting RandR lease support from 305 amdgpu. 306 307 Trivial. 308 309 commit 499d2f9d5d301ef1efd4ffc2952677609ef05122 310 Author: Michel Dnzer <michel.daenzer (a] amd.com> 311 Date: Tue Jul 17 11:43:12 2018 +0200 312 313 glamor: Invalidate cached GEM handle in radeon_set_pixmap_bo 314 315 We continued using the stale cached handle, causing issues e.g. when 316 resizing the screen via RandR. 317 318 Reported-by: iive on IRC 319 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 320 321 commit d9a139bc6d453d33593685de67d1425611f3ab6f 322 Author: Michel Dnzer <michel.daenzer (a] amd.com> 323 Date: Mon Jul 16 12:51:26 2018 +0200 324 325 glamor: Fix glamor_block_handler argument in radeon_glamor_finish 326 327 I missed this before because this code isn't compiled with current 328 xserver. 329 330 Trivial. 331 332 commit f533b1f654952cee794de49d28d01947a6571daf 333 Author: Keith Packard <keithp (a] keithp.com> 334 Date: Wed Jul 11 19:13:28 2018 +0200 335 336 Add RandR leases support 337 338 Signed-off-by: Keith Packard <keithp (a] keithp.com> 339 (Ported from xserver commit e4e3447603b5fd3a38a92c3f972396d1f81168ad) 340 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 341 (Ported from amdgpu commit 61040bdfa360975614fb47aa7ea1b3a1abac3427) 342 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 343 344 commit b4f0f44a9f67f2eafd4a2b5ab919e6ea7fa2acf7 345 Author: Keith Packard <keithp (a] keithp.com> 346 Date: Wed Jul 11 19:10:20 2018 +0200 347 348 modesetting: Create CONNECTOR_ID properties for outputs [v2] 349 350 This lets a DRM client map between X outputs and kernel connectors. 351 352 v2: 353 Change CONNECTOR_ID to enum -- Adam Jackson <ajax (a] nwnk.net> 354 355 Signed-off-by: Keith Packard <keithp (a] keithp.com> 356 (Ported from xserver commit 023d4aba8d45e9e3630b944ecfb650c081799b96) 357 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 358 (Ported from amdgpu commit ab7e39c5a03e24c3ce3ee2f22ada7572bc2d9aa7) 359 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 360 361 commit 612bda0a5e769f23478b364cf89620222cfac349 362 Author: Keith Packard <keithp (a] keithp.com> 363 Date: Wed Jul 11 19:07:59 2018 +0200 364 365 modesetting: Record non-desktop kernel property at PreInit time 366 367 Save any value of the kernel non-desktop property in the xf86Output 368 structure to avoid non-desktop outputs in the default configuration. 369 370 [Also bump randrproto requirement to a version that defines 371 RR_PROPERTY_NON_DESKTOP - ajax] 372 373 Signed-off-by: Keith Packard <keithp (a] keithp.com> 374 (Ported from xserver commit b91c787c4cd2d20685db69426c539938c556128a) 375 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 376 (Ported from amdgpu commit 14db71a606128c4a207f43298809af279b77e2a8) 377 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 378 379 commit 4b3e5f81c6032003237bb05c3ab96423c39524fc 380 Author: Michel Dnzer <michel.daenzer (a] amd.com> 381 Date: Wed Jul 11 19:04:26 2018 +0200 382 383 Call drmmode_crtc_gamma_do_set from drmmode_setup_colormap 384 385 Instead of from drmmode_set_mode_major. There's no need to re-set the 386 gamma LUT on every modeset, the kernel should preserve it. 387 388 (Ported from amdgpu commit baea4fa492f635cdfe746a84be2e337d9aeae8a9) 389 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 390 391 commit 16ddc109660dbe813c457765eda66422fb745a3e 392 Author: Michel Dnzer <michel.daenzer (a] amd.com> 393 Date: Wed Jul 11 19:03:12 2018 +0200 394 395 Remove #if 0'd code 396 397 This has always been disabled, no need to keep it. 398 399 (Ported from amdgpu commit 19a40758be04e1d451a030f452efb49e8aaad541) 400 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 401 402 commit 447ef0458fac047919d021f2ba1753a647f5c503 403 Author: Michel Dnzer <michel.daenzer (a] amd.com> 404 Date: Wed Jul 11 19:01:31 2018 +0200 405 406 Check dimensions passed to drmmode_xf86crtc_resize 407 408 When enabling a secondary GPU output, Xorg can try resizing the screen 409 beyond the limit advertised by the driver, leading to drmModeAddFB 410 failing and primary GPU outputs turning off. Check for this and bail 411 instead. 412 413 (Ported from amdgpu commit 940c8b39f79789d4d5ddb8ab8d25a8ae05932756) 414 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 415 416 commit 6a0c01bbd4ed48c696c38952ee33ce21afec9f91 417 Author: Michel Dnzer <michel.daenzer (a] amd.com> 418 Date: Wed Jul 11 18:59:58 2018 +0200 419 420 Use drmmode_crtc_dpms in drmmode_set_desired_modes 421 422 Simplifies the latter slightly. 423 424 (Ported from amdgpu commit 74124f2c17dbb4b752707bb7eee398ae099e8a2c) 425 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 426 427 commit 59441ee3dae15e0e81ed20688b0ba6dba12d7917 428 Author: Michel Dnzer <michel.daenzer (a] amd.com> 429 Date: Wed Jul 11 18:58:20 2018 +0200 430 431 Call drmmode_do_crtc_dpms from drmmode_crtc_dpms as well 432 433 Leo pointed out that drmmode_do_crtc_dpms wasn't getting called when 434 turning off an output with 435 436 xrandr --output <output> --off 437 438 This meant that the vblank sequence number and timestamp wouldn't be 439 saved before turning off the CRTC in this case. 440 441 Reported-by: Leo (Sunpeng) Li <sunpeng.li (a] amd.com> 442 (Ported from amdgpu commit ceeacb455cd058492a493aac954deab8455804b5) 443 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 444 445 commit 4050b0ad51b1c65945c6474981d1228888738cd4 446 Author: Michel Dnzer <michel.daenzer (a] amd.com> 447 Date: Wed Jul 11 18:42:58 2018 +0200 448 449 Replace 'foo == NULL' with '!foo' 450 451 Shorter and sweeter. :) 452 453 (Ported from amdgpu commit e8e688f3852fb06b0c34ed5bce47c9493bcd1613) 454 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 455 456 commit cf8bc72e3473cef2b511e2c938eb00aca82de909 457 Author: Michel Dnzer <michel.daenzer (a] amd.com> 458 Date: Tue Jul 10 18:57:49 2018 +0200 459 460 Wait for pending flips in drmmode_output_set_tear_free 461 462 This prevents a nested call to drmHandleEvent, which would hang. 463 464 Fixes hangs when disabling TearFree on a CRTC while a DRI3 client is 465 page flipping. 466 467 (Ported from amdgpu commit 04947b83cce3a7782e59dece2c7797cc396c1e05) 468 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 469 470 commit f01d8cf2bd9681b8f5f0e2eddec0a79614389771 471 Author: Michel Dnzer <michel.daenzer (a] amd.com> 472 Date: Tue Jul 10 18:53:54 2018 +0200 473 474 Refactor drmmode_output_set_tear_free helper 475 476 Preparation for the following fix, no functional change intended. 477 478 (Ported from amdgpu commit fa30f4601de7a44edfb4a95873bd648946fd4292) 479 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 480 481 commit ee7e15746148122abf6728fb0f59d6c8ae329e4e 482 Author: Michel Dnzer <michel.daenzer (a] amd.com> 483 Date: Tue Jul 10 18:53:06 2018 +0200 484 485 Set drmmode_crtc->scanout_id = 0 when TearFree is disabled 486 487 When disabling TearFree, drmmode_crtc->scanout_id could remain as 1, 488 but drmmode_set_mode_major would destroy drmmode_crtc->scanout[1], so 489 scanout_do_update() would keep bailing, and the scanout buffer would 490 stop being updated. 491 492 Fixes freeze after disabling TearFree on a CRTC with active RandR 493 rotation or other transform. 494 495 (Ported from amdgpu commit 7db0c8e9d7586cff4312d4b93684d35de3e6376f) 496 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 497 498 commit e9d2d149481e2a9c7cba50d43c6a5146124be3f2 499 Author: Michel Dnzer <michel.daenzer (a] amd.com> 500 Date: Tue Jul 10 18:52:08 2018 +0200 501 502 Simplify drmmode_handle_transform 503 504 Set crtc->driverIsPerformingTransform for any case we can handle before 505 calling xf86CrtcRotate. We already clear it afterwards when the latter 506 clears crtc->transform_in_use. 507 508 This should allow our separate scanout buffer mechanism to be used in 509 more cases. 510 511 (Cherry picked from amdgpu commit 8e544b4a0de6717feb4abf00052d57c5b726b5ce) 512 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 513 514 commit 89d38e976fd6ba6e026cda7c5d05971f1f177f13 515 Author: Michel Dnzer <michel.daenzer (a] amd.com> 516 Date: Tue Jul 10 18:51:00 2018 +0200 517 518 Don't call scanout_flip/update with a legacy RandR scanout buffer 519 520 It means we are not using our own scanout buffers. 521 522 Fixes crash when TearFree is supposed to be enabled, but 523 drmmode_handle_transform doesn't set crtc->driverIsPerformingTransform. 524 525 Bugzilla: https://bugs.freedesktop.org/105736 526 (Ported from amdgpu commit 463477661c88cab3a87746499e5838c5b9f9a13b) 527 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 528 529 commit 9f2409c4d3a9ecad8e769783e307e57a75b1255e 530 Author: Michel Dnzer <michel.daenzer (a] amd.com> 531 Date: Tue Jul 10 18:49:05 2018 +0200 532 533 Simplify drmmode_crtc_scanout_update 534 535 Use our own BoxRec for the extents, and RegionEmpty for clearing the 536 scanout damage region. 537 538 (Ported from amdgpu commit 72c3e9c7308fbcdf85708b72f9be14a5f2f8e7b5) 539 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 540 541 commit e07c38649280b3f4361005bc4c256f2145a72537 542 Author: Michel Dnzer <michel.daenzer (a] amd.com> 543 Date: Tue Jul 10 18:48:16 2018 +0200 544 545 Update RandR CRTC state if set_mode_major fails in set_desired_modes 546 547 Without this, RandR would report the CRTC and its outputs as enabled, 548 even though they were actually off due to the failure. 549 550 (Cherry picked from amdgpu commit 4dcda0b48d62944c841cd9540f4ad4c7ac8dee47) 551 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 552 553 commit 05390ae36a0abb19f1d533ff95a3fceaafdf79c8 554 Author: Michel Dnzer <michel.daenzer (a] amd.com> 555 Date: Tue Jul 10 18:45:51 2018 +0200 556 557 Abort scanout_update_pending event when possible 558 559 We don't need to wait for a non-TearFree scanout update before scanning 560 out from the screen pixmap or before flipping, as the scanout update 561 won't be visible anyway. Instead, just abort it. 562 563 (Ported from amdgpu commit 36d01989cd842588f12fdae5b2cba5fdcf9c91dd) 564 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 565 566 commit 290291a11598ba9aa594417998502b0e0ac96970 567 Author: Michel Dnzer <michel.daenzer (a] amd.com> 568 Date: Tue Jul 10 18:43:28 2018 +0200 569 570 Track DRM event queue sequence number in scanout_update_pending 571 572 Preparation for next change, no behaviour change intended. 573 (Ported from amdgpu commit 04a5c5f7cfacad8d9ccffe81e388cc3da2036cb5) 574 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 575 576 commit d14ff6b77824d0e32728869907e88d6866243101 577 Author: Michel Dnzer <michel.daenzer (a] amd.com> 578 Date: Tue Jul 10 18:41:43 2018 +0200 579 580 Ignore RADEON_DRM_QUEUE_ERROR (0) in radeon_drm_abort_entry 581 582 This allows a following change to be slightly simpler. 583 (Ported from amdgpu commit 8fcc3a9b43d3907052a83a96e5a2423afab5ad3f) 584 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 585 586 commit 731d4b386a55cdb468e37b69d41284150952cf8c 587 Author: Emil Velikov <emil.velikov (a] collabora.com> 588 Date: Tue Jul 10 18:39:31 2018 +0200 589 590 Do not export the DriverRec RADEON 591 592 Unused externally and should not be exported. 593 594 Signed-off-by: Emil Velikov <emil.velikov (a] collabora.com> 595 (Ported from amdgpu commit 7fb8b49895e225b3908c8bd186539de23afe91d1) 596 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 597 598 commit cc1d0824a8a7ef29f8911b95695f7cb1b4abe9b8 599 Author: Jim Qu <Jim.Qu (a] amd.com> 600 Date: Tue Jul 10 18:36:42 2018 +0200 601 602 Wait for pending scanout update before calling drmmode_crtc_scanout_free 603 604 There is a case that when set screen from reverse to normal, the old 605 scanout damage is freed in modesetting before scanout update handler, 606 so it causes segment fault issue. 607 608 Signed-off-by: Jim Qu <Jim.Qu (a] amd.com> 609 610 [ Michel Dnzer: Only call drmmode_crtc_wait_pending_event before 611 drmmode_crtc_scanout_free is actually called, slightly tweak commit 612 message ] 613 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 614 (Ported from amdgpu commit 9f6a8905611b5b1d8fcd31bebbc9af7ca1355cc3) 615 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 616 617 commit 413622bcccc9d36a9c87679f379dc84b21a3c655 618 Author: Michel Dnzer <michel.daenzer (a] amd.com> 619 Date: Tue Jul 10 17:23:24 2018 +0200 620 621 Swap pixmap privates in radeon_dri2_exchange_buffers 622 623 Instead of only the BOs. 624 625 This matches what amdgpu does, and fixes issues with DRI2 page flipping. 626 627 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 628 629 commit 3c4c0213c11d623cba7adbc28dde652694f2f758 630 Author: Michel Dnzer <michel.daenzer (a] amd.com> 631 Date: Fri Jun 29 17:57:03 2018 +0200 632 633 glamor: Use GBM for BO allocation when possible 634 635 Inspired by amdgpu. This avoids various issues due to a GEM handle 636 lifetime conflict between us and Mesa with current glamor. 637 638 Bugzilla: https://bugs.freedesktop.org/105381 639 Tested-by: Konstantin Kharlamov <hi-angel (a] yandex.ru> 640 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 641 642 commit b85b7b11f5b5e792f21951b881bd8433d3d70858 643 Author: Michel Dnzer <michel.daenzer (a] amd.com> 644 Date: Fri May 18 18:49:35 2018 +0200 645 646 Add struct radeon_buffer 647 648 Inspired by amdgpu, preparation for the following change. For now, this 649 is mostly a wrapper around struct radeon_bo, no functional change 650 intended. 651 652 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 653 654 commit 37ba075b34130c41bb7a2261bd666af5b29ffaf3 655 Author: Michel Dnzer <michel.daenzer (a] amd.com> 656 Date: Fri Jun 15 18:23:53 2018 +0200 657 658 Refactor radeon_finish helper 659 660 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 661 662 commit 6c986e997159ad0086f940294b244fc4c30b61fc 663 Author: Jammy Zhou <Jammy.Zhou (a] amd.com> 664 Date: Fri Jun 15 18:41:34 2018 +0200 665 666 Remove throttling from radeon_dri2_copy_region2 667 668 Throttling should be handled by the client-side drivers. 669 670 Signed-off-by: Jammy Zhou <Jammy.Zhou (a] amd.com> 671 (Ported from amdgpu commit 8a34a8149860ac15e83ccdbd8d9a527d8d3e5997) 672 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 673 674 commit 90b94d40449f665f2d12874598062a5e5e5b64cd 675 Author: Michel Dnzer <michel.daenzer (a] amd.com> 676 Date: Fri Jun 15 18:27:46 2018 +0200 677 678 Move flush from radeon_scanout_do_update to its callers 679 680 No functional change intended. 681 682 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 683 684 commit 5da2bf43e64f2d702fe05016263f617c10f8f134 685 Author: Michel Dnzer <michel.daenzer (a] amd.com> 686 Date: Mon May 28 18:35:32 2018 +0200 687 688 Factor out radeon_surface_initialize helper 689 690 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 691 692 commit ba5d5402b3e2e807d7e47205ac83f930b6c8caf5 693 Author: Michel Dnzer <michel.daenzer (a] amd.com> 694 Date: Thu May 17 12:36:29 2018 +0200 695 696 glamor: Don't store radeon_surfaces in pixmaps 697 698 Only EXA needs them. 699 700 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 701 702 commit eec4a41925127ae490f0a5156a881a08d521e28e 703 Author: Michel Dnzer <michel.daenzer (a] amd.com> 704 Date: Thu May 17 18:25:37 2018 +0200 705 706 Only initialize libdrm_radeon surface manager for >= R600 707 708 Not used with older GPUs. 709 710 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 711 712 commit ac18a993a3a68629805b8f272a339e25444c0897 713 Author: Michel Dnzer <michel.daenzer (a] amd.com> 714 Date: Wed May 23 12:18:36 2018 +0200 715 716 EXA: Remove old RADEONEXACreatePixmap hook 717 718 Not used by any supported version of xserver. 719 720 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 721 722 commit 25836375e0aab17a2e6bd91acb1c19b837ae2109 723 Author: Michel Dnzer <michel.daenzer (a] amd.com> 724 Date: Thu May 17 18:18:42 2018 +0200 725 726 Drop unused drmmode_create_bo_pixmap surface parameter 727 728 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 729 730 commit 65c9dfea4e841b7d6f795c7489fede58c5e9631f 731 Author: Michel Dnzer <michel.daenzer (a] amd.com> 732 Date: Tue May 15 18:59:30 2018 +0200 733 734 glamor: Bail CreatePixmap on unsupported pixmap depth 735 736 Fixes crash in that case. 737 738 Bugzilla: https://bugs.freedesktop.org/106293 739 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 740 741 commit 3dcfce8d0f495d09d7836caf98ef30d625b78a13 742 Author: Michel Dnzer <michel.daenzer (a] amd.com> 743 Date: Thu May 17 09:50:50 2018 +0200 744 745 Bail from dri2_create_buffer2 if we can't get a pixmap 746 747 We would store the NULL pointer and continue, which would lead to a 748 crash down the road. 749 750 Bugzilla: https://bugs.freedesktop.org/106293 751 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 752 753 commit fdba5373522c008f43e9ca5a9e76e35c6e7ad73d 754 Author: Michel Dnzer <michel.daenzer (a] amd.com> 755 Date: Thu Mar 15 17:19:59 2018 +0100 756 757 Bump version for 18.0.1 release 758 759 commit e0f5020117d0a6340f4208ec52bd3e3b3aa99a1b 760 Author: Michel Dnzer <michel.daenzer (a] amd.com> 761 Date: Wed Mar 14 16:54:52 2018 +0100 762 763 Pass extents to radeon_scanout_do_update by value 764 765 radeon_scanout_extents_intersect could leave the scanout damage region 766 in an invalid state, triggering debugging checks in pixman: 767 768 *** BUG *** 769 In pixman_region_append_non_o: The expression r->x1 < r->x2 was false 770 Set a breakpoint on '_pixman_log_error' to debug 771 772 (Ported from amdgpu commit 8af989546907ad9fb491d940e1936d3bfc89276b) 773 774 commit b915e8e6fb956e983c2ce2a5565c20838c85707c 775 Author: Michel Dnzer <michel.daenzer (a] amd.com> 776 Date: Wed Mar 7 17:28:55 2018 +0100 777 778 Wrap the whole miPointerScreenFuncRec, instead of only Set/MoveCursor 779 780 We were clobbering entries in mi's global miSpritePointerFuncs struct, 781 which cannot work correctly with multiple primary screens. Instead, 782 assign a pointer to our own wrapper struct to PointPriv->spriteFuncs. 783 784 Fixes crashes with multiple primary screens. 785 786 Fixes: 1fe8ca75974c ("Keep track of how many SW cursors are visible on 787 each screen") 788 Reported-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 789 790 commit 7e18aea984e186dd2ab1144470b0c0e832562ad1 791 Author: Michel Dnzer <michel.daenzer (a] amd.com> 792 Date: Tue Mar 6 18:15:45 2018 +0100 793 794 Bump version for 18.0.0 release 795 796 commit 504b8721b17a672caf1ed3eab087027c02458cab 797 Author: Michel Dnzer <michel.daenzer (a] amd.com> 798 Date: Mon Mar 5 18:41:44 2018 +0100 799 800 Only change Set/MoveCursor hooks from what we expect 801 802 Since xf86CursorCloseScreen runs after RADEONCloseScreen_KMS, 803 PointPriv->spriteFuncs doesn't point to the same struct in the latter as 804 in RADEONCursorInit_KMS. So we were restoring info->Set/MoveCursor to 805 the wrong struct. Then in the next server generation, 806 info->Set/MoveCursor would end up pointing to 807 drmmode_sprite_set/move_cursor, resulting in an infinite loop if one of 808 them was called. 809 810 To avoid this, only change the Set/MoveCursor hooks if their values 811 match our expectations, otherwise leave them as is. This is kind of a 812 hack, but the alternative would be invasive and thus risky changes to 813 the way we're wrapping CloseScreen, and it's not even clear that can 814 work without changing xserver code. 815 816 Fixes: 1fe8ca75974c ("Keep track of how many SW cursors are visible on 817 each screen") 818 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 819 820 commit e0d23092337efe4c1e406c9124ca8d8856863e65 821 Author: Keith Packard <keithp (a] keithp.com> 822 Date: Fri Mar 2 12:13:43 2018 +0100 823 824 modesetting: Update property values at detect and uevent time 825 826 We were updating the link-status property when a uevent came in, but 827 we also want to update the non-desktop property, and potentially 828 others as well. We also want to check at detect time in case we don't 829 get a hotplug event. 830 831 This patch updates every property provided by the kernel, sending 832 changes to DIX so it can track things as well. 833 834 Signed-off-by: Keith Packard <keithp (a] keithp.com> 835 836 (Ported from amdgpu commit 374cb8fef4fdbb648af089ee80803ec78321f1b2) 837 838 commit aef8fe6e9e94ebecbbb6adf940160d3a7ab4d5ac 839 Author: Daniel Martin <consume.noise (a] gmail.com> 840 Date: Fri Mar 2 12:12:09 2018 +0100 841 842 modesetting: Reset output_id if drmModeGetConnector failed 843 844 If drmModeGetConnector() fails in drmmode_output_detect(), we have to 845 reset the output_id to -1 too. 846 847 Yet another spot leading to a potential NULL dereference when handling 848 the mode_output member as output_id was != -1. Though, this case should 849 be very hard to hit. 850 851 Signed-off-by: Daniel Martin <consume.noise (a] gmail.com> 852 853 (Ported from amdgpu commit 10054b6c3d9a755b30abb43020121b9631fa296d) 854 855 commit 1ef12a92ca53cd6fbfed3ef8628616445b007c9a 856 Author: Daniel Martin <consume.noise (a] gmail.com> 857 Date: Fri Mar 2 12:08:09 2018 +0100 858 859 modesetting: Use helper to fetch drmModeProperty(Blob)s 860 861 Replace the various loops to lookup drmModeProperty(Blob)s by 862 introducing helper functions. 863 864 Signed-off-by: Daniel Martin <consume.noise (a] gmail.com> 865 866 (Ported from amdgpu commit fb58e06acd6c6bd59de2dbdadbca27eb1dd0025b) 867 868 commit ace1e3871c39a86becdf473a47fdc096e8ef39f5 869 Author: Michel Dnzer <michel.daenzer (a] amd.com> 870 Date: Wed Feb 21 17:53:59 2018 +0100 871 872 Disable all unused CRTCs before setting desired modes 873 874 This might avoid modeset failures in some cases where a CRTC which isn't 875 used by Xorg was enabled before. 876 877 (Ported from amdgpu commit e3aae7a24296f640c0153d1459f3e0820485468a) 878 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 879 880 commit f2915eb7a4beb6140ebec183aeac02111fc6ba77 881 Author: Christoph Haag <haagch (a] frickel.club> 882 Date: Thu Mar 1 15:07:00 2018 +0100 883 884 fix include order for present.h configure test 885 886 xorg-server.h defines _XSERVER64 which is used in X.h to choose the 887 correct definition of XID 888 889 this prevents a failure in the present.h configure test that disables 890 DRI3 on X.Org 1.20 891 892 Reviewed-and-Tested-by: Michel Dnzer <michel.daenzer (a] amd.com> 893 894 commit f777e909fd23e063729ecb038997db445f82ef5e 895 Author: Michel Dnzer <michel.daenzer (a] amd.com> 896 Date: Fri Feb 16 16:55:03 2018 +0100 897 898 Don't bail from drmmode_set_desired_modes immediately 899 900 If we fail to find or set the mode for a CRTC, keep trying for the 901 remaining CRTCs, and only return FALSE if we failed for all CRTCs that 902 should be on. 903 904 (Ported from amdgpu commit f5ac5f385f41d1547cfd7ccc8bb35a537a8fffeb) 905 906 commit 2d63748667d0bcfcd9999e4e41f775eba87897a0 907 Author: Michel Dnzer <michel.daenzer (a] amd.com> 908 Date: Thu Feb 15 17:47:14 2018 +0100 909 910 Simplify depth 30 "kernel too old" error message 911 912 We know it's depth 30, no need to have xf86DrvMsg fill it in. 913 914 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 915 916 commit 680b4ab43f00b1f635f9359281e13a031d09ad24 917 Author: Michel Dnzer <michel.daenzer (a] amd.com> 918 Date: Wed Jan 24 12:29:17 2018 +0100 919 920 If glamor is too old for depth 30, fall back to EXA or ShadowFB 921 922 Instead of not starting up at all. 923 924 Corresponding to amdgpu commit 37c7260bdef3a53b0f0295a531f33938e9aad8cf. 925 926 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 927 928 commit 16954fab59f95bdca4540c4496b2ec0759c447e9 929 Author: Michel Dnzer <michel.daenzer (a] amd.com> 930 Date: Wed Feb 14 18:20:04 2018 +0100 931 932 Call RADEONFreeRec from RADEONFreeScreen_KMS even if info == NULL 933 934 It's safe now. 935 936 (Ported from amdgpu commit c9bd1399a13cea2e1331af2c826ca054b88db071) 937 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 938 939 commit 6b44b8c584f3b63712858be11277842c97d03426 940 Author: Michel Dnzer <michel.daenzer (a] amd.com> 941 Date: Wed Feb 14 18:06:18 2018 +0100 942 943 Don't call RADEONFreeRec from RADEONPreInit_KMS 944 945 If the latter fails, Xorg will call RADEONFreeScreen_KMS, which calls 946 the former. 947 948 (Ported from amdgpu commit 103b7285845b786929fb509083c57e074c48f9be) 949 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 950 951 commit 8a55e66bfd77dcb2553a43e86215b9d3c87b03f0 952 Author: Michel Dnzer <michel.daenzer (a] amd.com> 953 Date: Wed Feb 14 18:13:16 2018 +0100 954 955 Don't use RADEONEntPriv in RADEONFreeRec 956 957 It crashes if info == NULL. 958 959 (Ported from amdgpu commits fb8444e731765588c0ff1e9053c1c7b73f5f0907 & 960 cfccf4c4e7e5c73fe4040fabeb1b43283cf29b33) 961 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 962 963 commit 703cbb7177c6f69264836e975c1921e552fd60a4 964 Author: Michel Dnzer <michel.daenzer (a] amd.com> 965 Date: Wed Feb 14 17:28:10 2018 +0100 966 967 glamor: Allow depth 30 with Xorg >= 1.19.99.1 968 969 Corresponding to amdgpu commit 6aee5770fb913713bb1b9a1af8f0d0892a66f21a. 970 971 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 972 973 commit 733f606dd6ca8350e6e7f0858bfff5454ddc98ed 974 Author: Michel Dnzer <michel.daenzer (a] amd.com> 975 Date: Thu Jan 25 16:59:30 2018 +0100 976 977 Always use screen depth/bpp for KMS framebuffers 978 979 DRI clients can use depth 32 pixmaps while the screen is depth 24, in 980 which case page flipping would fail. 981 982 Reported-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 983 Reviewed-and-Tested-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 984 985 commit 83124426040bc1f2dba74c8eeb5a5a9c7c64e664 986 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 987 Date: Wed Jan 24 18:19:38 2018 +0100 988 989 Make XvMC extension initialize at depth 30. 990 991 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 992 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 993 994 commit 2d171c892d373b45ef9f1ee068648437e2529a78 995 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 996 Date: Wed Jan 24 18:19:37 2018 +0100 997 998 Add missing depth 30 -> cpp=4 handling for DRI2. 999 1000 Missing this apparently didn't cause any ill side effects, only a bit of 1001 confusion when looking at xtrace's of DRI2 clients. 1002 1003 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 1004 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 1005 1006 commit 574bfab4bf1fcd95163a8f33cea2889189429d30 1007 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 1008 Date: Mon Jan 22 03:14:37 2018 +0100 1009 1010 Support exa screen color depth 30 on Linux 3.16 and later. (v2) 1011 1012 Linux 3.16's radeon-kms introduced 10 bpc fb support. 1013 Currently, as of X-Server 1.19, this works with exa accel, 1014 but not yet when using glamor acceleration. 1015 1016 v2: Style fixes, check for glamor first, as suggested 1017 by Michel. 1018 1019 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 1020 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 1021 1022 commit ee7e6c25989e4faf1adfa4754b7edd5a9a4722ce 1023 Author: Fredrik Hglund <fredrik (a] kde.org> 1024 Date: Mon Jan 22 03:14:36 2018 +0100 1025 1026 exa: Accelerate ARGB2101010 pictures 1027 1028 Tested by Mario on a Radeon HD 4000 series for the r600 1029 exa path, and on Radeon HD 5770 for the evergreen exa 1030 path. 1031 1032 Signed-off-by: Fredrik Hglund <fredrik (a] kde.org> 1033 Reviewed-and-Tested-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 1034 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 1035 1036 commit 1f1d4b1fa7d4b22dd8553f7e71251bf17ca7a7b1 1037 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 1038 Date: Mon Jan 22 03:14:35 2018 +0100 1039 1040 Skip xf86HandleColormaps() at color depth 30. (v2) 1041 1042 The hardware gamma luts get bypassed at color 1043 depth 30 anyway, so skip their setup. 1044 1045 v2: Also don't hook up drmmode_crtc_gamma_set() 1046 to .gamma_set for a screen with depth 30, 1047 so the server can spare itself the effort of 1048 updating and setting unused cluts. 1049 Suggested by Michel. 1050 1051 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 1052 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 1053 1054 commit 21f6753462464acfd3c452393328c977a375ce26 1055 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 1056 Date: Mon Jan 22 03:14:34 2018 +0100 1057 1058 Define per x-screen individual drmmode_crtc_funcs 1059 1060 This allows to en-/disable some functions depending on individual screen 1061 settings. 1062 1063 Prep work for more efficient depth 30 support. 1064 1065 Suggested-by: Michel Dnzer <michel.daenzer (a] amd.com> 1066 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 1067 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 1068 1069 commit 1fe8ca75974c5241c3b556b757f9c966c9ce5002 1070 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1071 Date: Wed Dec 27 17:13:30 2017 +0100 1072 1073 Keep track of how many SW cursors are visible on each screen 1074 1075 And use this to determine when we cannot use page flipping for DRI 1076 clients. We previously did this based on whether the HW cursor cannot 1077 be used on at least one CRTC, which had at least two issues: 1078 1079 * Even while the HW cursor cannot be used, no SW cursor may actually be 1080 visible (e.g. because all cursors are disabled), in which case we can 1081 use page flipping for DRI clients anyway 1082 * Even while the HW cursor can be used, there may be SW cursors visible 1083 from non-core pointer devices, in which case we cannot use page 1084 flipping for DRI clients anyway 1085 1086 (Ported from amdgpu commit 69e20839bfeb3ee0b0a732d72de0a32d6c5435fc) 1087 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1088 1089 commit b4ce8913ac392df339081f6f3e28bf79a239ee75 1090 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1091 Date: Wed Dec 27 17:05:26 2017 +0100 1092 1093 Move cursor related ScreenInit calls into RADEONCursorInit_KMS 1094 1095 And bail if xf86_cursors_init fails. 1096 1097 (Ported from amdgpu commit dfccaa7043ccb157a1f8be7313123792bb7e7001) 1098 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1099 1100 commit ee907e9b9563609cb4a4817e17215b50bb1c7177 1101 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1102 Date: Fri Dec 1 15:49:11 2017 +0100 1103 1104 Add radeon_dirty_src_drawable helper 1105 1106 Allows tidying up radeon_dirty_src_equals and redisplay_dirty slightly. 1107 1108 (Cherry picked from amdgpu commit 1d65ac395971571094df21ca0408d5972c6b56ec) 1109 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 1110 1111 commit fccfd75ecf9ccfa628a8ff9c1fbb7f60c0f10ea5 1112 Author: Joakim Tjernlund <joakim.tjernlund (a] infinera.com> 1113 Date: Fri Nov 24 17:27:09 2017 +0100 1114 1115 Fix non GLAMOR build 1116 1117 Non GLAMOR builds needs pRADEONEnt so move the USE_GLAMOR define 1118 down to make pRADEONEnt defined in this case. 1119 1120 Signed-off-by: Joakim Tjernlund <joakim.tjernlund (a] infinera.com> 1121 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 1122 1123 commit 15350bd5538c21cd4c6b406d565ca765b7e0218a 1124 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1125 Date: Wed Nov 15 18:01:29 2017 +0100 1126 1127 Use correct ScrnInfoPtr in redisplay_dirty 1128 1129 We used the destination pixmap's screen for flushing drawing commands. 1130 But when we are the master screen, the destination pixmap is from the 1131 slave screen. 1132 1133 Fixes crash when the slave screen isn't using the same acceleration 1134 architecture as us. 1135 1136 Bugzilla: https://bugs.freedesktop.org/103613 1137 Fixes: 01b040b4a807 ("Adapt to PixmapDirtyUpdateRec::src being a 1138 DrawablePtr") 1139 (Ported from amdgpu commit 3a4f7422913093ed9e26b73ecd7f9e773478cb1e) 1140 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1141 1142 commit 6441210bc53978a2c46c572e9174215b14f9e813 1143 Author: Daniel Martin <consume.noise (a] gmail.com> 1144 Date: Fri Oct 20 10:05:35 2017 +0200 1145 1146 modesetting: Check crtc before searching link-status property 1147 1148 No need to lookup the link-status property if we don't have a crtc. 1149 1150 Signed-off-by: Daniel Martin <consume.noise (a] gmail.com> 1151 (Ported from xserver commit 8d7f7e24261e68459e6f0a865e243473f65fe7ad) 1152 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1153 1154 commit cafcb55768b58dc63a8c5682000eba21ad904689 1155 Author: Keith Packard <keithp (a] keithp.com> 1156 Date: Mon Sep 25 16:18:22 2017 -0700 1157 1158 modesetting: Skip no-longer-present connectors when resetting BAD links 1159 1160 Outputs may have NULL mode_output (connector) pointers if the 1161 connector disappears while the server is running. Skip these when 1162 resetting outputs with BAD link status. 1163 1164 (Ported from xserver commit 37f4e7651a2fd51efa613a08a1e705553be33e76) 1165 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1166 1167 commit 0cd14777b7e39619f231b841a2d73f3122458030 1168 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1169 Date: Fri Oct 20 17:20:30 2017 +0200 1170 1171 Always call drmModeFreeProperty after drmModeGetProperty 1172 1173 We were not doing so in all cases, leaking memory allocated by the 1174 latter. 1175 1176 (Cherry picked from amdgpu commit f6b39bcd45cb06976ba8a3600df77fc471c63995) 1177 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1178 1179 commit 84c7d9c1c71d9cee9f50f96671032529b606cf04 1180 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1181 Date: Fri Oct 20 17:19:09 2017 +0200 1182 1183 Call TimerFree for timer created in LeaveVT 1184 1185 We were leaking the memory allocated by TimerSet. 1186 1187 (Ported from amdgpu commit 84aad09f18fed6b52b0c073f0bbd675a6de07807) 1188 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1189 1190 commit d588015f4f60ce270bcfd8433215fc981a84f143 1191 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1192 Date: Fri Oct 20 17:13:36 2017 +0200 1193 1194 Free memory returned by xf86GetEntityInfo 1195 1196 We were leaking it. 1197 1198 (Ported from amdgpu commit cfccf4c4e7e5c73fe4040fabeb1b43283cf29b33) 1199 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1200 1201 commit ec90c94881239791d0a66a02fc6d003c73a6c1f3 1202 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1203 Date: Fri Oct 20 16:51:16 2017 +0200 1204 1205 Free pRADEONEnt memory in RADEONFreeRec 1206 1207 We were leaking it. 1208 1209 (Inspired by amdgpu commit 9d84934309e4ccd9a43c73d958b8ff10ef2fc990) 1210 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1211 1212 commit b93ae9b159790e8a979e9d5d567a0e6fccd7d08a 1213 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1214 Date: Thu Oct 19 18:57:45 2017 +0200 1215 1216 Unreference pixmap's FB with EXA as well in radeon_set_pixmap_bo 1217 1218 Not doing so resulted in DRI2 page flips not actually changing the FB 1219 being scanned out, showing intermittent flicker of the "back" buffer 1220 rendering. 1221 1222 Bugzilla: https://bugs.freedesktop.org/102643 1223 Fixes: 55e513b978b2 "Use reference counting for tracking KMS framebuffer 1224 lifetimes" 1225 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1226 1227 commit 61113768bd79ec5c2164835e2e048e8f2a42ce9a 1228 Author: Darren Salt <devspam (a] moreofthesa.me.uk> 1229 Date: Thu Oct 19 18:17:07 2017 +0200 1230 1231 Clarify when TearFree is automatically enabled. 1232 1233 (Ported from amdgpu commit 2f72be038d22c54620e436af30121dd89f79a003) 1234 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1235 1236 commit 7d435354099119234d443b07e2df1c7b9f97cf3c 1237 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1238 Date: Wed Oct 4 11:11:33 2017 +0200 1239 1240 Bail if there's a problem with ShadowFB 1241 1242 If we hit a problem while setting up ShadowFB, just carrying on trying 1243 to set up HW acceleration instead is unlikely to work. 1244 1245 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1246 1247 commit b71836c9962f8f2eadf0c2188e03faed737c18da 1248 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1249 Date: Wed Oct 4 11:05:49 2017 +0200 1250 1251 Post-release version bump 1252 1253 commit 34da04daec82077571558ac3fe1ec0c1203a01ad 1254 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1255 Date: Tue Oct 3 12:47:14 2017 +0200 1256 1257 Fix VT switching with ShadowFB 1258 1259 We were trying to call acceleration specific functions from LeaveVT. 1260 Instead, memset the scanout buffer to all 0 in LeaveVT and allocate a 1261 new one in EnterVT. 1262 1263 Bugzilla: https://bugs.freedesktop.org/102948 1264 Fixes: 06a465484101 ("Make all active CRTCs scan out an all-black 1265 framebuffer in LeaveVT") 1266 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 1267 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1268 1269 commit f9aca8c1a7a423b9e738c88400ae821bdfbcdb1c 1270 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1271 Date: Fri Sep 8 17:02:59 2017 +0900 1272 1273 Bump version for 7.10.0 release 1274 1275 commit 5cdd334b3402c2431deb3a87a8d04ef590da53ee 1276 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1277 Date: Wed Aug 30 18:48:19 2017 +0900 1278 1279 Require xserver >= 1.13 1280 1281 xserver 1.13.0 was released on September 6th, 2012, almost 5 years ago. 1282 1283 This allows cleaning up a bunch of backwards compatibility code. 1284 1285 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1286 Reviewed-by: Emil Velikov <emil.l.velikov (a] gmail.com> 1287 1288 commit 9d9c565c84601f4c6c73ad769f86491088683f7a 1289 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1290 Date: Tue Aug 29 18:07:51 2017 +0900 1291 1292 Use a timer for unreferencing the all-black FB 1293 1294 The timer fires 1 second after LeaveVT. This gives the next DRM master 1295 enough time to set up scanout of its own buffers. 1296 1297 Fixes prolonged intermittent black screen when switching from Xorg to 1298 e.g. the GDM Wayland mode login VT. 1299 1300 Fixes: 06a465484101 ("Make all active CRTCs scan out an all-black 1301 framebuffer in LeaveVT") 1302 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1303 1304 commit e4a3df19d588a4310fcb889ef34e205d0e92e4d7 1305 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1306 Date: Mon Aug 28 17:54:23 2017 +0900 1307 1308 Remove drmmode_scanout_free 1309 1310 Not used anymore. 1311 1312 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1313 1314 commit 06a465484101f21e99d3a0a62fb03440bcaff93e 1315 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1316 Date: Wed Aug 23 18:00:15 2017 +0900 1317 1318 Make all active CRTCs scan out an all-black framebuffer in LeaveVT 1319 1320 And destroy all other FBs. This is so that other DRM masters can only 1321 get access to this all-black FB, not to any other FB we created, while 1322 we're switched away and not DRM master. 1323 1324 Fixes: 55e513b978b2 ("Use reference counting for tracking KMS 1325 framebuffer lifetimes") 1326 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1327 1328 commit 7f0cd68d1b0c132e32ae736371bce3e12ed33c7a 1329 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1330 Date: Mon Aug 28 17:53:19 2017 +0900 1331 1332 Create radeon_master_screen helper 1333 1334 Preparatory, no functional change intended yet. 1335 1336 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1337 1338 commit 20f6b56fdb74d88086e8e094013fedbb14e50a24 1339 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1340 Date: Thu Aug 24 17:10:29 2017 +0900 1341 1342 Create radeon_pixmap_get_fb_ptr helper 1343 1344 Preparatory, no functional change intended yet. 1345 1346 Also inline radeon_pixmap_create_fb into radeon_pixmap_get_fb, since 1347 there's only one call-site anymore. 1348 1349 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1350 1351 commit 4bc992c31059eb50e22df4ebf5b92d08411f41ef 1352 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1353 Date: Wed Aug 23 17:24:53 2017 +0900 1354 1355 Create drmmode_set_mode helper 1356 1357 Preparatory, no functional change intended yet. 1358 1359 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1360 1361 commit 3f6210ca2c8ef60d59efc8139151d3b9838bb875 1362 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1363 Date: Thu Jun 22 18:37:26 2017 +0900 1364 1365 Create radeon_pixmap_clear helper 1366 1367 Preparatory, no functional change intended yet. 1368 1369 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1370 1371 commit 0472a605e0ec8fec1892bbc3a84698b7ef9c5296 1372 Author: Martin Peres <martin.peres (a] linux.intel.com> 1373 Date: Mon Apr 10 16:48:21 2017 +0300 1374 1375 modesetting: re-set the crtc's mode when link-status goes BAD 1376 1377 Despite all the careful planning of the kernel, a link may become 1378 insufficient to handle the currently-set mode. At this point, the 1379 kernel should mark this particular configuration as being broken 1380 and potentially prune the mode before setting the offending connector's 1381 link-status to BAD and send the userspace a hotplug event. This may 1382 happen right after a modeset or later on. 1383 1384 Upon receiving a hot-plug event, we iterate through the connectors to 1385 re-apply the currently-set mode on all the connectors that have a 1386 link-status property set to BAD. The kernel may be able to get the 1387 link to work by dropping to using a lower link bpp (with the same 1388 display bpp). However, the modeset may fail if the kernel has pruned 1389 the mode, so to make users aware of this problem a warning is outputed 1390 in the logs to warn about having a potentially-black display. 1391 1392 This patch does not modify the current behaviour of always propagating 1393 the events to the randr clients. This allows desktop environments to 1394 re-probe the connectors and select a new resolution based on the new 1395 (currated) mode list if a mode disapeared. This behaviour is expected in 1396 order to pass the Display Port compliance tests. 1397 1398 (Ported from xserver commit bcee1b76aa0db8525b491485e90b8740763d7de6) 1399 1400 [ Michel: Bump libdrm dependency to >= 2.4.78 for 1401 DRM_MODE_LINK_STATUS_BAD ] 1402 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1403 1404 commit d822a0f47070374ad0c1a97b559bae27724dc52a 1405 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1406 Date: Thu Aug 17 16:20:52 2017 +0900 1407 1408 Make radeon_scanout_do_update take a PixmapPtr instead of a DrawablePtr 1409 1410 All callers were already passing in a pixmap. 1411 1412 This allows simplifying the rotated scanout case slightly. 1413 1414 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1415 1416 commit 3e24770b1b472fc15df56d06f5f04778c9db63dd 1417 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1418 Date: Wed Aug 9 13:02:34 2017 +0900 1419 1420 Use xorg_list_append for the DRM event list 1421 1422 We were adding entries at the start of the list, i.e. the list was 1423 ordered from most recently added to least recently added. However, the 1424 corresponding DRM events are generally expected to arrive in the same 1425 order as they are queued, which means that radeon_drm_queue_alloc would 1426 generally have to traverse the whole list to find the entry 1427 corresponding to an arrived event. Fix this by adding entries at the end 1428 of the list. 1429 1430 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1431 1432 commit 36ce7920136c0d723c9397a84e7dd5926a9c7943 1433 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1434 Date: Wed Aug 9 19:01:15 2017 +0900 1435 1436 Consolidate radeon_scanout_flip_abort/handler helpers 1437 1438 While at it, make them use crtc->driver_private. 1439 1440 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1441 1442 commit d314cbfb228bb4b8762714f98d0c114a8ee3f061 1443 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1444 Date: Mon Jul 31 17:12:59 2017 +0900 1445 1446 Always allow DRI2 page flipping with TearFree 1447 1448 Even if TearFree is enabled for the CRTC we're synchronizing to. 1449 1450 v2: 1451 * Move manpage hunk here from previous change. 1452 1453 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> # v1 1454 1455 commit 4445765af5b97d0cfd10889fe6d6f58f2ce85659 1456 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1457 Date: Thu Feb 9 13:01:03 2017 +0900 1458 1459 Always allow Present page flipping with TearFree 1460 1461 Even if TearFree is active for the the CRTC we're synchronizing to. In 1462 that case, for Present flips synchronized to vertical blank, the other 1463 scanout buffer is immediately synchronized and flipped to during the 1464 target vertical blank period. For Present flips not synchronized to 1465 vertical blank, we simply use the MSC and timestamp values of the last 1466 vertical blank period for timing purposes, and let the normal TearFree 1467 mechanism handle display updates. 1468 1469 v2: 1470 * Move manpage hunk to next change, since TearFree can still prevent 1471 DRI2 page flipping with this change. 1472 1473 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> # v1 1474 1475 commit 65e0c5ea1b4adff21d673dbf54af99704c429627 1476 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1477 Date: Wed Feb 8 17:52:28 2017 +0900 1478 1479 Pass extents to radeon_scanout_do_update 1480 1481 Preparation for following change, no functional change intended yet. 1482 1483 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1484 1485 commit 1443270e52e8562bd8dc3603f301963bd4027cef 1486 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1487 Date: Tue Feb 7 18:58:23 2017 +0900 1488 1489 Add source drawable parameter to radeon_scanout_do_update 1490 1491 Preparation for following changes, no functional change intended yet. 1492 1493 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1494 1495 commit 7c10ee9c88378d773c0bcf651fdc5d9f2c6dc5e5 1496 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1497 Date: Thu Aug 3 18:16:36 2017 +0900 1498 1499 Handle multiple "pending" Present flips 1500 1501 The xserver Present code can submit a flip in response to notifying it 1502 that a vblank event arrived. This can happen before the completion event 1503 of the previous flip is processed. In that case, we were clearing the 1504 drmmode_crtc->flip_pending field prematurely. 1505 1506 Prevent this by only clearing drmmode_crtc->flip_pending when it matches 1507 the framebuffer being scanned out since the flip whose completion event 1508 we're processing. 1509 1510 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1511 1512 commit e6d7dc2070f4d21a6900916bb70a31839112882c 1513 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1514 Date: Thu Aug 3 17:51:20 2017 +0900 1515 1516 Wait for pending flips synchronously before turning off a CRTC 1517 1518 Allows removing drmmode_clear_pending_flip and the pending_dpms_mode 1519 field and cleaning up the code considerably. 1520 1521 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1522 1523 commit f87acdbfb1b0b6d2769764772a52ea8b81675e20 1524 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1525 Date: Fri Jul 28 12:28:53 2017 +0900 1526 1527 Create drmmode_crtc_wait_pending_event helper macro 1528 1529 Preparation for following change, no functional change intended yet. 1530 1531 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1532 1533 commit 99f1d7a474af3683fe1a66f50c0bb8935478ff0a 1534 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1535 Date: Mon Aug 14 12:23:04 2017 +0900 1536 1537 Create drmmode_wait_vblank helper 1538 1539 Allows cleaning up the code considerably. 1540 1541 v2: 1542 * Fix "drmWaiVBlank" typo, add blank line for readability (Slava Abramov) 1543 * Rename in/out sequence parameters to "target_seq" and "result_seq", 1544 hopefully that will be clearer. 1545 1546 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> # v1 1547 1548 commit 49cc61ab970ee28d4509b4e2dd0a57165136889f 1549 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1550 Date: Tue Aug 1 18:45:53 2017 +0900 1551 1552 Pass reference CRTC to radeon_do_pageflip directly 1553 1554 Simplifies the code slightly. 1555 1556 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1557 1558 commit 3f120fa1d5d921656a367751bc079e020e9ab105 1559 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1560 Date: Thu Aug 3 17:42:51 2017 +0900 1561 1562 Remove drmmode_crtc->scanout_destroy[] array 1563 1564 No longer necessary since we're reference counting framebuffers. 1565 1566 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com 1567 1568 commit 01b040b4a80754b8c8a305cbe968416182b54246 1569 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1570 Date: Tue Aug 15 17:15:27 2017 +0900 1571 1572 Adapt to PixmapDirtyUpdateRec::src being a DrawablePtr 1573 1574 commit 5309bde0c4e28adf2b167191c6d7011a19e31eed 1575 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1576 Date: Thu Jul 27 16:11:19 2017 +0900 1577 1578 Allow DRI page flipping when some CRTCs use separate scanout buffers 1579 1580 As long as the CRTC we're synchronizing to doesn't. 1581 1582 v2: 1583 * Remove redundant checks from can_exchange which still prevented DRI2 1584 page flipping 1585 1586 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1587 1588 commit 9bc3eef74452d924f9101c024f66ad9b14c404c8 1589 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1590 Date: Thu Jul 27 15:46:41 2017 +0900 1591 1592 Add drmmode_crtc_can_flip helper 1593 1594 To reduce code duplication between DRI2 and Present. No functional 1595 change intended yet. 1596 1597 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1598 1599 commit c2d26890691ec105858f086b63170ad94c6f7f05 1600 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1601 Date: Wed Jul 12 12:13:03 2017 +0900 1602 1603 Use root window (pixmap) instead of screen pixmap for scanout updates 1604 1605 Preparation for following changes, no functional change intended yet. 1606 1607 v2: 1608 * Add drmmode_screen_damage_destroy callback to prevent use-after-free 1609 on server shutdown 1610 1611 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1612 1613 commit 377c9dc7e51cc8ebd06816b08682a8098a49508a 1614 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1615 Date: Thu Jul 27 15:27:54 2017 +0900 1616 1617 autogen.sh: Pass -f to autoreconf 1618 1619 To ensure that any existing copies of autotools files will be replaced 1620 with the current versions. 1621 1622 (Cherry picked from amdgpu commit a47c0093338d80d84e7033ad15d051925d542ca0) 1623 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 1624 1625 commit bf0a759e3bced3a93bead033a26871cc340bd309 1626 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1627 Date: Thu Jul 27 15:26:23 2017 +0900 1628 1629 Makefile.am: Set ACLOCAL_AMFLAGS = -I m4 1630 1631 Suggested by one of the tools called by autoreconf. 1632 1633 (Cherry picked from amdgpu commit 842bad4b951296ca25f47b50cb358e502bf30ebb) 1634 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 1635 1636 commit e628f604d032d32971a1ca8670469e88939e3de2 1637 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1638 Date: Thu Jul 27 15:24:37 2017 +0900 1639 1640 Add AC_CONFIG_MACRO_DIRS([m4]) to configure.ac 1641 1642 Suggested by one of the tools called by autoreconf. 1643 1644 (Cherry picked from amdgpu commit 227b399badaad9bbef0be5a776ce008d0d243449) 1645 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 1646 1647 commit 7d7abf99b5441ddb04dbee99bc8fa7abc30d4c46 1648 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1649 Date: Wed Jul 26 18:13:35 2017 +0900 1650 1651 Only handle reflection in the driver with Xorg < 1.16 1652 1653 Xorg doesn't handle the hardware cursor correctly in that case for 1654 rotation and general transforms, and we can't force the SW cursor. 1655 1656 Fixes: f2bc882f1c10 ("Handle rotation in the driver also with Xorg 1657 1.12-1.18") 1658 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1659 1660 commit 94dc2b80f3ef0b2c17c20501d824fb0447d52e7a 1661 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1662 Date: Tue Mar 14 16:57:17 2017 +0900 1663 1664 If a TearFree flip fails, fall back to non-TearFree operation 1665 1666 In order to avoid possible freeze / log file spam in that case. 1667 1668 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99769 1669 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1670 1671 commit aff267ee36cc6a703a532f91f82adc1ba1425ff3 1672 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1673 Date: Tue Mar 14 16:23:45 2017 +0900 1674 1675 Use drmmode_crtc->scanout_id instead of 0 to check for scanout buffer 1676 1677 Preparation for following change, no functional change intended. 1678 1679 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1680 1681 commit df88d51c5005476e6fbfca3f31b54be079ec5aaa 1682 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1683 Date: Tue Jun 27 19:15:18 2017 +0900 1684 1685 Use pRADEONEnt->fd exclusively for the DRM file descriptor 1686 1687 This brings us closer to amdgpu. 1688 1689 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1690 1691 commit 19626bce4e5e31c863eedb503ea3884ac3f60bea 1692 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1693 Date: Tue Jun 27 18:32:28 2017 +0900 1694 1695 Improve drmmode_fb_reference debugging code 1696 1697 If a reference count is <= 0, call FatalError with the call location 1698 (in case it doesn't get resolved in the backtrace printed by 1699 FatalError). 1700 1701 (Ported from amdgpu commit 1b6ff5fd9933c00ec1ec90dfc62e0b531927749b) 1702 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1703 1704 commit bc46ffdf71ab3dfa0f95572529e818f2b619d380 1705 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1706 Date: Tue Jun 27 18:19:31 2017 +0900 1707 1708 Increase reference count of FB assigned to drmmode_crtc->flip_pending 1709 1710 Otherwise, it could happen that we destroy the FB before the flip 1711 completes, resulting in use-after-free and most likely a crash. 1712 1713 (Ported from amdgpu commit af7221e1c4d2dbdfd488eb0976a835584ea8441c) 1714 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1715 1716 commit c9dd28cb0c9c3de676eadac61e727732510f6b9b 1717 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1718 Date: Thu Jun 22 12:38:55 2017 +0900 1719 1720 Only call drmmode_scanout_free for non-GPU screens in LeaveVT 1721 1722 Destroying the scanout buffers of GPU screens resulted in a crash when 1723 switching back to the Xorg VT. 1724 1725 Fixes: 4cfa4615f79f ("Use drmmode_crtc_scanout_* helpers for RandR 1.4 1726 scanout pixmaps") 1727 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1728 1729 commit 78fad9ca1635cca04b89807e3db029477cf46681 1730 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1731 Date: Mon Jun 26 18:58:30 2017 +0900 1732 1733 Include xf86Pci.h for DRICreatePCIBusID with xserver Git master 1734 1735 The declaration has been moved there from dri.h. 1736 1737 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1738 1739 commit ef95e3a3cc9189c8906fe5b4d70bfa9f1055d7ce 1740 Author: Adam Jackson <ajax (a] redhat.com> 1741 Date: Tue Jun 13 09:32:44 2017 -0400 1742 1743 modesetting: Validate the atom for enum properties 1744 1745 The client could have said anything here, and if what they said doesn't 1746 actually name an atom NameForAtom() will return NULL, and strcmp() will 1747 be unhappy about that. 1748 1749 [copied from xserver d4995a3936ae283b9080fdaa0905daa669ebacfc] 1750 1751 Signed-off-by: Adam Jackson <ajax (a] redhat.com> 1752 Reviewed-and-Tested-by: Michel Dnzer <michel.daenzer (a] amd.com> 1753 1754 commit d80d01a73c2eaba2e3649b7bc0a3541b3ff782f6 1755 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1756 Date: Thu May 18 11:12:34 2017 +0900 1757 1758 Update URLs 1759 1760 * Point to the amd-gfx mailing list 1761 * Specify the component in all bugzilla URLs 1762 * Use https:// for all HTML URLs 1763 1764 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1765 1766 commit 2b7d77b90108911777a11ecaa63435552000c958 1767 Author: Eric Anholt <eric (a] anholt.net> 1768 Date: Tue May 16 11:04:29 2017 -0700 1769 1770 Use plain glamor_egl_create_textured_screen(). 1771 1772 Since 5064ffab631 (2014), glamor's implementation of _ext just drops the 1773 back_pixmap arg, which we were passing NULL (the default) to anyway. 1774 1775 Signed-off-by: Eric Anholt <eric (a] anholt.net> 1776 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 1777 1778 commit 3a9f7e87c30b9f6334d38dfdd1cb15b28fe37fcb 1779 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1780 Date: Thu May 11 18:57:25 2017 +0900 1781 1782 Remove unused struct members from drmmode_display.h 1783 1784 (Ported from amdgpu commit 462ac3341e5bfbded9086d3d9043821d19352b3e) 1785 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1786 1787 commit 7dc68e26755466f9056f8c72195ab8690660693d 1788 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1789 Date: Tue May 2 11:47:58 2017 +0900 1790 1791 Simplify tracking of PRIME scanout pixmap 1792 1793 Remember the shared pixmap passed to drmmode_set_scanout_pixmap for each 1794 CRTC, and just compare against that. 1795 1796 Fixes leaving stale entries in ScreenRec::pixmap_dirty_list under some 1797 circumstances, which would usually result in use-after-free and a crash 1798 down the line. 1799 1800 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1801 1802 commit 55e513b978b2afc52b7cafc5bfcb0d1dc78d75f6 1803 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1804 Date: Sat Apr 29 17:05:58 2017 +0900 1805 1806 Use reference counting for tracking KMS framebuffer lifetimes 1807 1808 References are held by the pixmaps corresponding to the FBs (so 1809 the same KMS FB can be reused as long as the pixmap exists) and by the 1810 CRTCs scanning out from them (so a KMS FB is only destroyed once it's 1811 not being scanned out anymore, preventing intermittent black screens and 1812 worse issues due to a CRTC turning off when it should be on). 1813 1814 v2: 1815 * Only increase reference count in drmmode_fb_reference if it was sane 1816 before 1817 * Make drmmode_fb_reference's indentation match the rest of 1818 drmmode_display.h 1819 1820 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1821 1822 commit 944391b0052466b71bf9919b56139dc197a7e072 1823 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1824 Date: Wed Feb 8 18:14:30 2017 +0900 1825 1826 Pass pixmap instead of handle to radeon_do_pageflip 1827 1828 This brings us in line with amdgpu and prepares for the following 1829 change, no functional change intended. 1830 1831 (Ported from amdgpu commit e463b849f3e9d7b69e64a65619a22e00e78d297b) 1832 1833 v2: 1834 * Be more consistent with the amdgpu code, which should make porting 1835 the following change to amdgpu easier 1836 1837 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1838 1839 commit f32c45194ac6f82cbe42d255ed72f857018778e0 1840 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1841 Date: Tue May 9 11:18:00 2017 +0900 1842 1843 Apply gamma correction to HW cursor 1844 1845 The display hardware CLUT we're currently using for gamma correction 1846 doesn't affect the HW cursor, so we have to apply it manually when 1847 uploading the HW cursor data. 1848 1849 This currently only works in depth 24/32. 1850 1851 (Ported from amdgpu commit 82fa615f38137add75f9cd4bb49c48dd88de916f) 1852 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1853 1854 commit 4c91f36d3058180b5a2d6a23e9b82f5c933d8716 1855 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1856 Date: Mon Apr 17 18:32:31 2017 +0900 1857 1858 Don't set modes before RADEONWindowExposures_oneshot is called 1859 1860 The root window contents may be undefined before that, so we don't want 1861 to show anything yet. 1862 1863 Fixes a crash on startup with rotation and virtual resolution set in 1864 xorg.conf. 1865 1866 Bugzilla: https://bugs.freedesktop.org/100276 1867 Fixes: cc9d6b7db9c2 ("Move DPMS check from radeon_scanout_do_update to 1868 radeon_scanout_flip") 1869 (Ported from amdgpu commit 981bac185cfd74ae50dffc28f57cf34623a9595f) 1870 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1871 1872 commit d6c29250dbc293c932ca14997eb23399610b106d 1873 Author: Daniel Stone <daniels (a] collabora.com> 1874 Date: Fri Apr 7 14:27:58 2017 +0100 1875 1876 Set correct DRM event context version 1877 1878 DRM_EVENT_CONTEXT_VERSION is the latest context version supported by 1879 whatever version of libdrm is present. We were blindly asserting we 1880 supported whatever version that may be, even if we actually didn't. 1881 1882 Set the version as 2, which should be bumped only with the appropriate 1883 version checks. 1884 1885 Signed-off-by: Daniel Stone <daniels (a] collabora.com> 1886 (Ported from xserver commit 0c8e6ed85810e96d84173a52d628863802a78d82) 1887 v2: Remove second paragraph of commit log, we always initialize 1888 page_flip_handler2 = NULL (Emil Velikov) 1889 Reviewed-by: Emil Velikov <emil.l.velikov (a] gmail.com> 1890 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> # v1 1891 1892 commit 23985d3ef3150ef8cad30b92ef0cd8ba6cfa1f4d 1893 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1894 Date: Thu Mar 23 18:06:45 2017 +0900 1895 1896 manpage: Don't put "'" at the beginning of a line 1897 1898 It caused the whole line to be dropped. 1899 1900 Fixes: 58cd1600057e ("Allow toggling TearFree at runtime via output 1901 property") 1902 Reported-by: Andy Furniss <adf.lists (a] gmail.com> 1903 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> (amdgpu commit) 1904 (Ported from amdgpu commit 165b51447643ce37f391f25ca6aecb8d76fabaa3) 1905 1906 commit dc30f5e27ef8a8e980adb21c1fea7c4f60c7e896 1907 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1908 Date: Thu Mar 23 18:06:23 2017 +0900 1909 1910 Post-release version bump 1911 1912 commit 66227060872219c1a0dc84fa4cea6264f7548446 1913 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1914 Date: Thu Mar 16 16:38:15 2017 +0900 1915 1916 Bump version for 7.9.0 release 1917 1918 commit 9a71445094b728f3d78db8f6808b4782ee19a453 1919 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1920 Date: Wed Mar 8 17:42:56 2017 +0900 1921 1922 Pass TRUE to drmmode_set_desired_modes the first time for GPU screens 1923 1924 This is the only place we call drmmode_set_desired_modes for GPU screens 1925 during server startup. Without this change, the display outputs of 1926 secondary GPUs may stay on even while Xorg isn't using them. 1927 1928 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1929 1930 commit 67ae5e00a748ad52cf92738d401afff2947b1891 1931 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1932 Date: Wed Mar 8 11:20:30 2017 +0900 1933 1934 Skip some initialization steps for GPU screens 1935 1936 Xorg doesn't use the following functionality of GPU screens, so don't 1937 bother initializing it: 1938 1939 * DRI page flipping 1940 * DRI3 / Present / SYNC fences 1941 * XVideo / XvMC 1942 * Root window with background None 1943 1944 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1945 1946 commit f0e7948e1c0e984fc27f235f365639e9cf628291 1947 Author: Jochen Rollwagen <joro-2013 (a] t-online.de> 1948 Date: Sun Mar 5 11:32:38 2017 +0100 1949 1950 Fix build for XServer 1.13 1951 1952 Latest git build stops with the error message 1953 1954 radeon_kms.c: In function 'RADEONWindowExposures_oneshot': 1955 radeon_kms.c:1644:45: error: expected expression before 'RegionPtr' 1956 pScreen->WindowExposures(pWin, pRegion, RegionPtr pBSRegion); 1957 1958 This patch fixes the build. 1959 1960 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 1961 1962 commit d63881623f0686a66a2e3e3c1f84e496aa52ec6b 1963 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1964 Date: Thu Mar 2 16:22:51 2017 +0900 1965 1966 Don't call radeon_cs_flush_indirect & radeon_bo_wait in drmmode_copy_fb 1967 1968 RADEONWindowExposures_oneshot takes care of it. 1969 1970 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 1971 1972 commit cc9d6b7db9c2078be1e530a64af6d517c6a42024 1973 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1974 Date: Wed Mar 1 17:35:59 2017 +0900 1975 1976 Move DPMS check from radeon_scanout_do_update to radeon_scanout_flip 1977 1978 When radeon_scanout_do_update is called from 1979 drmmode_crtc_scanout_update, drmmode_crtc->pending_dpms_mode may still 1980 be != DPMSModeOn, e.g. during server startup. 1981 1982 Fixes intermittently showing garbage with TearFree enabled. 1983 1984 commit 0a12bf1085505017068dfdfd31d23133e51b45b9 1985 Author: Michel Dnzer <michel.daenzer (a] amd.com> 1986 Date: Wed Mar 1 18:00:40 2017 +0900 1987 1988 Call drmmode_set_desired_modes from a WindowExposures hook 1989 1990 This is the earliest opportunity where the root window contents are 1991 guaranteed to be initialized, and prevents drmmode_set_mode_major from 1992 getting called before drmmode_set_desired_modes via RADEONUnblank -> 1993 drmmode_crtc_dpms. Also, in contrast to the BlockHandler hook, this is 1994 called when running Xorg with -pogo. 1995 1996 Fixes intermittently showing garbage on server startup or after server 1997 reset. 1998 1999 As a bonus, this avoids trouble due to higher layers (e.g. the tigervnc 2000 Xorg module) calling RADEONBlockHandler_oneshot repeatedly even after 2001 we set pScreen->BlockHandler = RADEONBlockHandler_KMS. 2002 2003 v2: 2004 * Drop spaces between XORG_VERSION_NUMERIC arguments 2005 * Call radeon_bo_wait after radeon_cs_flush_indirect 2006 2007 Bugzilla: https://bugs.freedesktop.org/99457 2008 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> (v1) 2009 2010 commit df2d749a4db33298c8ce9f2cfb77c20c5538c9cc 2011 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2012 Date: Thu Mar 2 16:05:42 2017 +0900 2013 2014 Fix bogus indentation 2015 2016 Trivial. 2017 2018 Fixes: 58cd1600057e ("Allow toggling TearFree at runtime via output property") 2019 2020 commit 9035b6abea557828e672ee455f0c84e43da0906f 2021 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2022 Date: Tue Feb 28 18:13:42 2017 +0900 2023 2024 present: Flush before flipping 2025 2026 This isn't necessary for DRI clients, but the Present extension can also 2027 be used for presenting normal pixmaps rendered to via the X11 protocol. 2028 2029 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2030 2031 commit 0a4eb0e12f0c9c653cf4cea6fd62e1a507eb261c 2032 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2033 Date: Fri Feb 10 12:52:02 2017 +0900 2034 2035 present: Use async flip for unflip if possible 2036 2037 In that case, unflip operations should finish faster in general. 2038 2039 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2040 2041 commit 9a951a3e551db58ba50e7a594521ceac54d90615 2042 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2043 Date: Tue Feb 28 18:06:55 2017 +0900 2044 2045 present: Also flush before using a flip to unflip 2046 2047 Not doing so might result in intermittently scanning out stale contents 2048 of the screen pixmap. 2049 2050 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2051 2052 commit f2bc882f1c1082bed9f496cfab6c8f07a76bc122 2053 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2054 Date: Tue Feb 28 17:31:14 2017 +0900 2055 2056 Handle rotation in the driver also with Xorg 1.12-1.18 2057 2058 We cannot use the HW cursor in that case, but in turn we get more 2059 efficient and less teary updates of rotated outputs. 2060 2061 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2062 2063 commit ae921a3150f69c38b5b3c88a9e37d54fdf0d5093 2064 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2065 Date: Tue Feb 28 17:47:09 2017 +0900 2066 2067 Fold drmmode_crtc_scanout_allocate into drmmode_crtc_scanout_create 2068 2069 Not used anywhere else anymore. 2070 2071 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2072 2073 commit 987a34adb319923ad36e2b47a26837248f187c3e 2074 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2075 Date: Tue Feb 28 17:17:20 2017 +0900 2076 2077 Call drmmode_crtc_scanout_create in drmmode_crtc_shadow_allocate as well 2078 2079 Calling drmmode_crtc_scanout_allocate in drmmode_crtc_shadow_allocate 2080 resulted in drmmode_crtc_scanout_create called from 2081 drmmode_crtc_shadow_create passing an uninitialized pitch value to 2082 drmmode_create_bo_pixmap. 2083 2084 Fixes issues such as failure to allocate the scanout pixmap or visual 2085 corruption and GPUVM faults when attempting to use rotation with Xorg 2086 <1.19. 2087 2088 Bugzilla: https://bugs.freedesktop.org/99916 2089 Fixes: ea30d856ba5e ("Pass pitch from drmmode_crtc_scanout_allocate to drmmode_create_bo_pixmap") 2090 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2091 2092 commit 244d4bc7f8c8f6bc90f49556c0b9344c8aa40295 2093 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2094 Date: Thu Feb 9 18:55:04 2017 +0900 2095 2096 present: Wait for screen pixmap BO idle before setting modes for unflip 2097 2098 To make sure the screen pixmap contents are up to date when it starts 2099 being scanned out. 2100 2101 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2102 2103 commit 3ff29e5a14451916bc66b4e0028e9a317f0723f8 2104 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2105 Date: Thu Feb 9 18:50:40 2017 +0900 2106 2107 present: Only call drmModeRmFB after setting modes for unflip 2108 2109 Fixes display intermittently blanking when a modeset is used for unflip. 2110 2111 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2112 2113 commit 0c29deb5a97d9a57e994cc0053c49ddf7aca6ecb 2114 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2115 Date: Tue Feb 21 16:56:27 2017 +0900 2116 2117 Use drmmode_crtc_scanout_free in drmmode_fini 2118 2119 We were leaking drmmode_crtc->scanout_damage, which caused trouble on 2120 server reset. Fixes server reset with active separate scanout pixmaps. 2121 2122 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2123 2124 commit 58cd1600057e41aade0106d4acf78e23eac6e44f 2125 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2126 Date: Thu Dec 1 18:37:57 2016 +0900 2127 2128 Allow toggling TearFree at runtime via output property 2129 2130 Option "TearFree" now sets the default value of the output property. 2131 See the manpage update for details. 2132 2133 TearFree is now enabled by default for outputs using rotation or other 2134 RandR transforms, and for RandR 1.4 slave outputs. 2135 2136 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2137 2138 commit 305e2cbf335837a2ab6a24e9ff65815afe038296 2139 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2140 Date: Tue Feb 21 15:55:13 2017 +0900 2141 2142 Factor out drmmode_crtc_scanout_update helper 2143 2144 Cleanup in preparation for following change, no functional change 2145 intended. 2146 2147 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2148 2149 commit 649644a88347a6d03de68f8c41db03a82deeb23b 2150 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2151 Date: Tue Feb 14 17:36:31 2017 +0900 2152 2153 Factor out radeon_prime_dirty_to_crtc helper 2154 2155 Cleanup in preparation for the following change, no functional change 2156 intended. 2157 2158 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2159 2160 commit 1351e48efe7a2c28eab447e16f36a00fbd02ae48 2161 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2162 Date: Tue Feb 7 18:43:56 2017 +0900 2163 2164 Don't destroy current FB if drmModeAddFB fails 2165 2166 It would probably result in a black screen. 2167 2168 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2169 2170 commit 481394e3c9f9f7d88bb66fe9ae8834c87952a8ab 2171 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2172 Date: Tue Feb 7 18:39:05 2017 +0900 2173 2174 Fix flip event data leak if calloc or drmModeAddFB fails 2175 2176 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2177 2178 commit 80fc78633962c1ec996b1721bfb757b98dc85bfc 2179 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2180 Date: Tue Jan 31 16:25:37 2017 +0900 2181 2182 Don't handle Option "SwapbuffersWait" at all with glamor 2183 2184 It never had any effect with glamor. 2185 2186 v2: Better formatting, no functional change from v1. 2187 2188 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> (v1) 2189 2190 commit d611d21d9bf9aaa100ff7a5762e5980634846207 2191 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2192 Date: Tue Jan 24 18:05:45 2017 +0900 2193 2194 Enable tiling by default with glamor on PALM 2195 2196 The DFS check is only relevant for EXA. 2197 2198 The lack of tiling prevented DRI3 clients from using page flipping, 2199 resulting in tearing. 2200 2201 Bugzilla: https://bugs.freedesktop.org/99491 2202 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2203 2204 commit 55a33205cb9bd8091b4bcd478c2d1d191248681b 2205 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2206 Date: Tue Jan 24 18:22:49 2017 +0900 2207 2208 Add 10-radeon.conf xorg.conf.d snippet 2209 2210 This instructs Xorg >= 1.16 to try loading the radeon driver for devices 2211 managed by the radeon kernel driver, even if the ati wrapper driver 2212 isn't available. 2213 2214 Copied from the amdgpu driver. 2215 2216 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2217 2218 commit 8e6a4e96b7b27559e186f71b5547abb0a80b96dd 2219 Author: Mihail Konev <k.mvc (a] ya.ru> 2220 Date: Thu Jan 26 14:00:22 2017 +1000 2221 2222 autogen: add default patch prefix 2223 2224 Signed-off-by: Mihail Konev <k.mvc (a] ya.ru> 2225 2226 commit e062933b5489705bdcba7a266d14ba6fd3e767ba 2227 Author: Emil Velikov <emil.l.velikov (a] gmail.com> 2228 Date: Mon Mar 9 12:00:52 2015 +0000 2229 2230 autogen.sh: use quoted string variables 2231 2232 Place quotes around the $srcdir, $ORIGDIR and $0 variables to prevent 2233 fall-outs, when they contain space. 2234 2235 Signed-off-by: Emil Velikov <emil.l.velikov (a] gmail.com> 2236 Reviewed-by: Peter Hutterer <peter.hutterer (a] who-t.net> 2237 Signed-off-by: Peter Hutterer <peter.hutterer (a] who-t.net> 2238 2239 commit 4fe183111f82d412760b175b3c9c8583acba804a 2240 Author: Peter Hutterer <peter.hutterer (a] who-t.net> 2241 Date: Tue Jan 24 10:32:07 2017 +1000 2242 2243 autogen.sh: use exec instead of waiting for configure to finish 2244 2245 Syncs the invocation of configure with the one from the server. 2246 2247 Signed-off-by: Peter Hutterer <peter.hutterer (a] who-t.net> 2248 Reviewed-by: Emil Velikov <emil.velikov (a] collabora.com> 2249 2250 commit a1787e4615a76ab0bb12498be66591c86a6b07ae 2251 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2252 Date: Wed Jan 18 19:07:11 2017 +0900 2253 2254 ati: Support loading the amdgpu driver from the ati wrapper 2255 2256 If .../share/X11/xorg.conf.d/10-amdgpu.conf doesn't exist, but the ati 2257 wrapper is loaded, it will otherwise try to use the radeon driver even 2258 for GPUs driven by the amdgpu kernel driver. This can only fail, 2259 potentially in bad ways. 2260 2261 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2262 2263 commit 41b82c776b571e1556f300d77c715aebdd87d8e6 2264 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2265 Date: Tue Jan 24 18:14:23 2017 +0900 2266 2267 .editorconfig: src/ati.c only uses spaces for indentation 2268 2269 Trivial. 2270 2271 commit ea30d856ba5e7274c8ea499293b8b0e721b8e082 2272 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2273 Date: Fri Jan 6 17:36:05 2017 +0900 2274 2275 Pass pitch from drmmode_crtc_scanout_allocate to drmmode_create_bo_pixmap 2276 2277 radeon_bo_get_tiling can fail, e.g. for linear BOs. 2278 2279 Fixes TearFree not working in those cases. 2280 2281 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97873 2282 Fixes: 040a7b80e1fc ("Explicitly set the fbcon pixmap pitch again") 2283 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2284 2285 commit 720dd07e79b5f5cbbf3ef8bd92300578c485d583 2286 Author: Jammy Zhou <Jammy.Zhou (a] amd.com> 2287 Date: Mon Dec 19 17:27:19 2016 +0900 2288 2289 Use render node for DRI3 if available 2290 2291 Signed-off-by: Jammy Zhou <Jammy.Zhou (a] amd.com> 2292 (ported from amdgpu commit ea558e645786b08d75307716036045170e97b43e) 2293 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2294 [ Second attempt, let's see if there's any fallout this time... ] 2295 2296 commit 92fc387a6256ce3c95d62142735783ea07162c54 2297 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2298 Date: Thu Dec 15 12:44:47 2016 +0900 2299 2300 Simplify drmmode_handle_uevents 2301 2302 No functional change intended. 2303 2304 Reviewed-by: Jim Qu <Jim.Qu (a] amd.com> 2305 (Cherry picked from amdgpu commit edd276185d42962a13faf9ec9eeebc754ef284e7.) 2306 2307 commit 404c64752774d23a2db9aa3134f6ada3155708c9 2308 Author: jimqu <Jim.Qu (a] amd.com> 2309 Date: Wed Dec 14 11:46:16 2016 +0900 2310 2311 udev_monitor_receive_device() will block when hotplug monitor 2312 2313 udev_monitor_receive_device() will block and wait for the event of udev 2314 use select() to ensure that this will not block. 2315 2316 Signed-off-by: JimQu <Jim.Qu (a] amd.com> 2317 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 2318 (Cherry picked from amdgpu commit 732cf4d3a248b288532ad0f3443da49e08dc7507) 2319 2320 commit a89c0bced423ce3e09f66edafc35642a7cc03b2f 2321 Author: Jochen Rollwagen <joro-2013 (a] t-online.de> 2322 Date: Fri Dec 2 10:56:39 2016 +0100 2323 2324 Calculate log base 2 in radeon.h based on clz for all platforms 2325 2326 This commit replaces the inline assembler code (for x86 platforms) and 2327 loop (for non-x86 platforms) in RADEONLog2 with a one-liner version 2328 based on clz (count leading zeroes). 2329 2330 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 2331 2332 commit e2942449171fe628a7726e59bcaab65e27d88563 2333 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2334 Date: Mon Nov 21 18:47:53 2016 +0900 2335 2336 Call radeon_drm_abort_entry on failure to flip to a scanout pixmap 2337 2338 Fixes leaking the corresponding struct radeon_drm_queue list entry in 2339 that case. 2340 2341 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2342 2343 commit 14c3f59f5157885ad8f941f0bad6c0c5e3db12f8 2344 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2345 Date: Mon Nov 21 18:45:30 2016 +0900 2346 2347 Call ValidateGC after ChangeClip in radeon_sync_scanout_pixmaps 2348 2349 The wrong order meant that the clipping region wasn't actually applied, 2350 so it always copied the full contents from the other scanout pixmap. 2351 2352 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2353 2354 commit a995f5830916a0fee5126263d1bfe48632be3a15 2355 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2356 Date: Mon Nov 21 18:39:00 2016 +0900 2357 2358 Fix radeon_scanout_extents_intersect for GPU screens 2359 2360 Fixes incorrect screen updates with TearFree enabled on PRIME slave 2361 outputs which are not located at (0, 0). 2362 2363 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2364 2365 commit e543ef3a2fb304cbe3a965fb780632af2e4186f4 2366 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2367 Date: Mon Nov 21 18:33:25 2016 +0900 2368 2369 Take current scanout_id into account everywhere involved with TearFree 2370 2371 Fixes various potential issues with TearFree enabled, e.g. outputs 2372 freezing after display configuration changes. 2373 2374 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2375 2376 commit f130b10e63f7526360b41aa0918b4940f63f662a 2377 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2378 Date: Fri Nov 25 18:12:40 2016 +0900 2379 2380 Add radeon_is_gpu_screen helper 2381 2382 This will hopefully decrease the chance of accidentally breaking the 2383 build against xserver < 1.13 in the future. 2384 2385 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2386 2387 commit 487aa62a2a23b86e4ea4714fdfd465c9e513141f 2388 Author: Jochen Rollwagen <joro-2013 (a] t-online.de> 2389 Date: Thu Nov 24 18:44:01 2016 +0100 2390 2391 fix build for xserver < 1.13 2392 2393 same procedure every few patches..... 2394 2395 Fixes: 13c6bc5e3827 ("Don't install Flush/EventCallback for GPU screens") 2396 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 2397 2398 commit 5fea5ef2f07eee4a0f94baab427010b936f1d4b4 2399 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2400 Date: Thu Nov 24 18:45:22 2016 +0900 2401 2402 Pass fb_id into drmmode_page_flip_target_absolute/relative 2403 2404 drmmode->fb_id isn't what we need in the TearFree case. Fixes TearFree 2405 freezing with 2406 2407 (WW) RADEON(0): flip queue failed in radeon_scanout_flip: No such file or directory 2408 2409 in the log file. 2410 2411 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98834 2412 Fixes: 1106b2f773ad ("Use DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flags when available") 2413 2414 commit 13c6bc5e382765fe567091e1c616c0a26eec04ca 2415 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2416 Date: Mon Aug 1 18:11:57 2016 +0900 2417 2418 Don't install Flush/EventCallback for GPU screens 2419 2420 Their purpose is to flush GPU rendering commands corresponding to damage 2421 events, but there can be no damage events corresponding to GPU screen 2422 rendering operations. 2423 2424 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2425 2426 commit f11531c99fcd6473f58b4d10efaf3efd84304d8e 2427 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2428 Date: Mon Nov 21 18:15:05 2016 +0900 2429 2430 Enable glamor by default with >= R600 and Xorg >= 1.18.3 2431 2432 glamor should now perform at least as well as EXA in general, and this 2433 allows DRI3 to be enabled by default for >= R600. 2434 2435 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2436 2437 commit 1106b2f773ad0611c729b27f4c192a26b43ef1e7 2438 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2439 Date: Wed Jul 6 15:25:33 2016 +0900 2440 2441 Use DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flags when available 2442 2443 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2444 2445 commit aab80d2203efb8569b26dd02a1c651c792e1ef0c 2446 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2447 Date: Fri Nov 18 12:11:38 2016 +0900 2448 2449 Post-release version bump 2450 2451 commit 5cbe1ee8e499e1b6b2646e341946292721d07e69 2452 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2453 Date: Thu Nov 17 11:27:16 2016 +0900 2454 2455 Bump version for 7.8.0 release 2456 2457 commit 68023e01842ed7e5c798be77c25993bbe6a5a966 2458 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2459 Date: Thu Nov 17 11:32:37 2016 +0900 2460 2461 Update manpage entry for Option "TearFree" 2462 2463 It's now effective for arbitrary transforms as well. 2464 2465 commit 9760ef33cba5795eddeda4d5c2fcbe2dcce21689 2466 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2467 Date: Tue Nov 8 13:02:43 2016 +0900 2468 2469 Use pRADEONEnt to find both screens of a GPU in radeon_mode_hotplug 2470 2471 Fixes misbehaviour when hotplugging DisplayPort connectors on secondary 2472 GPUs. 2473 2474 Fixes: c801f9f10a5d ("Handle Zaphod mode correctly in radeon_mode_hotplug") 2475 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98626 2476 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2477 2478 commit 35bec4937d89b48a79acfcb4f814b7370cb631b2 2479 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2480 Date: Tue Nov 8 13:01:58 2016 +0900 2481 2482 Refactor radeon_mode_hotplug 2483 2484 Preparation for the next change, no functional change intended. 2485 2486 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2487 2488 commit be9f67339a19a6649eab52aa2e0c8971cd9b4727 2489 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2490 Date: Mon Nov 7 16:34:37 2016 +0900 2491 2492 Reindent code in radeon_dri2_create_buffer2 2493 2494 Fixes warning about misleading indentation from recent versions of gcc: 2495 2496 ../../src/radeon_dri2.c: In function radeon_dri2_create_buffer2: 2497 ../../src/radeon_dri2.c:224:2: warning: this if clause does not guard... [-Wmisleading-indentation] 2498 if (flags & RADEON_CREATE_PIXMAP_TILING_MACRO) 2499 ^~ 2500 ../../src/radeon_dri2.c:227:6: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the if 2501 if (aligned_width == front_width) 2502 ^~ 2503 2504 No functional change intended. 2505 2506 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2507 2508 commit 350a2645a1b127227ff294c0b62d20000d0fd48a 2509 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2510 Date: Tue Nov 1 16:30:42 2016 +0900 2511 2512 Check Xorg version at runtime instead of build time in two places 2513 2514 This means that all possible paths can be handled as intended, no matter 2515 which Xorg version the driver happened to be compiled against. 2516 2517 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2518 2519 commit 5df36de39952c3a26cb2fbc125f298139a9dd5bc 2520 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2521 Date: Fri Aug 26 18:24:08 2016 +0900 2522 2523 Require xserver 1.10 or newer 2524 2525 1.10.0 was released in February 2011. 2526 2527 We've been accidentally requiring 1.10 or newer since 121a6de72da5 ("Keep 2528 track of damage event related flushes per-client v2"). 2529 2530 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2531 2532 commit 22b5ce9548393ba2ff73ee234ecd82eeaf0ef6c4 2533 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2534 Date: Tue Oct 25 17:28:03 2016 +0900 2535 2536 Consume all available udev events at once 2537 2538 We get multiple udev events for actions like docking a laptop into its 2539 station or plugging a monitor to the station. By consuming as many 2540 events as we can, we reduce the number of output re-evalutions. 2541 2542 It depends on the timing how many events can be consumed at once. 2543 2544 (Inspired by xserver commit 363f4273dd4aec3e26cc57ecb6c20f27e6c813d8) 2545 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2546 2547 commit 82d3c8f5500d2a6fb1495e217a0b79c396f1534c 2548 Author: Hans de Goede <hdegoede (a] redhat.com> 2549 Date: Tue Oct 25 16:56:40 2016 +0900 2550 2551 PRIME: Fix swapping of provider sink / source capabilities 2552 2553 When a card has import capability it can be an offload _sink_, not a 2554 source and vice versa for export capability. 2555 2556 This went unnoticed sofar because most gpus have both import and export 2557 capability. 2558 2559 Signed-off-by: Hans de Goede <hdegoede (a] redhat.com> 2560 (Ported from xserver commit 94a1c77259ce39ba59ad87615df39b570ffab435) 2561 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2562 2563 commit 6c940446ddadf418ee4959e46fa552b6c1cf6704 2564 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2565 Date: Tue Oct 25 16:42:03 2016 +0900 2566 2567 Always call PixmapStopDirtyTracking in drmmode_set_scanout_pixmap 2568 2569 Otherwise, we may leak screen->pixmap_dirty_list entries if 2570 drmmode_set_scanout_pixmap is called repatedly with ppix != NULL, which 2571 can happen from RRReplaceScanoutPixmap. 2572 2573 (Inspired by xserver commit b773a9c8126222e5fed2904d012fbf917a9f22fd) 2574 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2575 2576 commit 61df12e2377cbb19a19ca9d5624df8959822da9f 2577 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2578 Date: Wed Oct 19 18:55:33 2016 +0900 2579 2580 Don't rely on randr_crtc->scanout_pixmap in drmmode_set_scanout_pixmap 2581 2582 RRReplaceScanoutPixmap may set randr_crtc->scanout_pixmap = NULL before 2583 we get here. 2584 2585 (Inspired by xserver commit f4c37eeee7953df1fe0e3196eda452acf0078e61) 2586 v2: Always return TRUE in the if (!ppix) block. 2587 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> (v1) 2588 2589 commit 49cf3b5032a7ce40afe514b7092440e3e19e05aa 2590 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2591 Date: Thu Oct 20 15:33:09 2016 +0900 2592 2593 Saynara, AM_MAINTAINER_MODE! 2594 2595 If --enable-maintainer-mode got lost from config.status for any reason, 2596 builds would fail in mysterious ways after changing between different 2597 Git commits. 2598 2599 There are more reasons for dropping it in the automake manual: 2600 2601 https://www.gnu.org/software/automake/manual/html_node/maintainer_002dmode.html 2602 2603 I'm not aware of any reason why --disable-maintainer-mode would ever be 2604 useful with this project. 2605 2606 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2607 2608 commit 4b17de1cad12e7be12f2c71e5149bfc04564702d 2609 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2610 Date: Wed Oct 19 18:09:14 2016 +0900 2611 2612 Remove PCI IDs and bus type from ati_pciids.csv 2613 2614 This cleans up the list of unique chipsets a little further. 2615 2616 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2617 2618 commit 1ce1b1656acc6211deb2091ff7f28d51b6daf86b 2619 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2620 Date: Wed Oct 19 18:08:36 2016 +0900 2621 2622 Order unique chipsets according to first appearance in ati_pciids.csv 2623 2624 Instead of lexically. This makes it more likely for similar generations 2625 to be close to each other in the list of unique chipsets. 2626 2627 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2628 2629 commit 7c66be961de3477fc685eb71b46265adbae923f0 2630 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2631 Date: Mon Oct 17 18:38:55 2016 +0900 2632 2633 Enable HW cursor support with PRIME slave output & Xorg > 1.18.99.901 2634 2635 Supported since Xorg 1.18.99.2, but buggy until 1.18.99.901. 2636 2637 (Ported from amdgpu commit 7cc04035c55788261cda89a915c433c2add6cad9) 2638 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2639 2640 commit d016f20b14ffe4f90f7457375b2d9f5cf0f5650f 2641 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2642 Date: Thu Oct 13 16:33:45 2016 +0900 2643 2644 Restore location of radeon_set_pixmap_bo call in drmmode_create_bo_pixmap 2645 2646 radeon_set_pixmap_bo sets the tiling flags, so it needs to be called 2647 before radeon_get_pixmap_tiling_flags. Fixes a regression with EXA and 2648 TearFree. 2649 2650 Fixes: 8523a733b6a5 ("Propagate failure from radeon_set_pixmap_bo") 2651 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98097 2652 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2653 2654 commit 937398f6cd50db105d5e53ab553eb284d44121f4 2655 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2656 Date: Thu Oct 6 18:52:37 2016 +0900 2657 2658 Rotate and reflect cursor hotspot position for drmModeSetCursor2 2659 2660 We were always passing the hotspot position in the X screen coordinate 2661 space, but drmModeSetCursor2 needs it in the CRTC coordinate space. The 2662 wrong hotspot position would cause the kernel driver to adjust the 2663 HW cursor position incorrectly when the hotspot position changed. 2664 2665 (Ported from amdgpu commit d42773eb45baff5933730e26878a0b45fcf07b65) 2666 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2667 2668 commit 3fc839ff49f01c24eb94d5e1f0ab4065de47bc17 2669 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2670 Date: Wed Sep 21 17:50:36 2016 +0900 2671 2672 Only list each unique chipset family once in the log file 2673 2674 (Ported from amdgpu commit 6a1ba044c2b71081e6060d0c096917d6238f2145) 2675 2676 commit 9a5eb1915a240e3021ac38a6dacafef81e7bb743 2677 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2678 Date: Wed Sep 21 17:48:38 2016 +0900 2679 2680 Clear damage in radeon_scanout_update if it doesn't intersect the CRTC 2681 2682 There's no need to test that same damage again. 2683 2684 (Ported from amdgpu commit a576430526cbc404de64b30e1377a356644e8024) 2685 2686 commit 792a30dbe5f7d7815ea14032dbab55b5de8eabca 2687 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2688 Date: Wed Sep 21 17:43:57 2016 +0900 2689 2690 Remove w/h parameters from radeon_scanout_extents_intersect 2691 2692 We can use the dimensions of the CRTC's mode instead. 2693 2694 (Ported from amdgpu commit ede7f2bcae63be65e05e3029bfe7c742e5978932) 2695 2696 commit 80d794f11a5f047663897e4349da75d89ceff7c7 2697 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2698 Date: Wed Sep 21 17:38:57 2016 +0900 2699 2700 Make the dedicated scanout mechanism work with arbitrary transforms 2701 2702 This makes TearFree work with arbitrary transforms, and makes transforms 2703 work better even without TearFree, with xserver >= 1.12. 2704 2705 (Ported from amdgpu commit bf000ea7ef91f5ecb59fc3c1ab8ed9eddcc0841d) 2706 2707 commit 09ba7b3b7b1b724aa386dbf2d33f567d7f26326b 2708 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2709 Date: Wed Sep 21 17:00:22 2016 +0900 2710 2711 Remove drmmode_load_palette 2712 2713 Not used by any supported version of xserver. 2714 2715 (Ported from amdgpu commits 1091f28e1fa239ee1a973d84a8376fa4a95d7247 2716 and 5a4d3267ac3823fe58b51b0b9075b82375d7180c) 2717 2718 commit 730f443955cf2bcedd760c9fef5b11c81eacdd84 2719 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2720 Date: Wed Sep 21 16:48:49 2016 +0900 2721 2722 present: Separate checks for flips vs unflips 2723 2724 All unflip checks apply to flips as well, but not vice versa. 2725 2726 (Ported from amdgpu commit 4d506c23c9a628204fa23607931557b07ada3e31) 2727 2728 commit 343d28672411a510de21fa57986324ef8bcdd1b6 2729 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2730 Date: Wed Sep 21 16:47:17 2016 +0900 2731 2732 dri2: Don't allow flipping when using a dedicated scanout buffer 2733 2734 Fixes issues when mixing rotation and page flipping with current xserver 2735 Git master. 2736 2737 (Ported from amdgpu commit decabd574f90d3df397c80ec931b3fde8a4afb49) 2738 2739 commit 83e4781d15d66d6fa0e153eb4cd0e3d9e2ebed9f 2740 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2741 Date: Wed Sep 21 16:44:50 2016 +0900 2742 2743 present: Don't allow flipping when using a dedicated scanout buffer 2744 2745 Fixes issues when mixing rotation and page flipping with current xserver 2746 Git master. 2747 2748 (Ported from amdgpu commit 3ed28ce7cd26f89969617ba901ff253091d0d469) 2749 2750 commit 8419db3de6157875f9a840773350ecd29cd42d83 2751 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2752 Date: Wed Sep 21 16:40:42 2016 +0900 2753 2754 Make sure drmmode_crtc->scanout[] are destroyed when not needed 2755 2756 We failed to do this when going back to scanning out directly from the 2757 screen pixmap. 2758 2759 As a bonus, since we now destroy drmmode_crtc->scanout[] after setting 2760 the new scanout buffer, we may avoid the CRTC turning off intermittently 2761 in this case. 2762 2763 (Ported from amdgpu commit 9c3324715fd395fd486ea341654d78f4f298b97f) 2764 2765 commit 956e03d6a6b9478dd8e77c81f426c0d371c5d7a1 2766 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2767 Date: Wed Sep 21 16:37:53 2016 +0900 2768 2769 Simplify drmmode_set_mode_major error handling 2770 2771 Initialize ret = FALSE and only set it to TRUE when we've succeeded. 2772 2773 (Ported from amdgpu commit 3bce0519a4008cf87c0e31a7a579e10f5dcdd2f3) 2774 2775 commit ace455d1db4db3f9b8b434a3747acfa7e1273db4 2776 Author: Hans de Goede <hdegoede (a] redhat.com> 2777 Date: Wed Sep 21 16:31:51 2016 +0900 2778 2779 Only add main fb if necessary 2780 2781 If we're doing reverse-prime; or doing rotation the main fb is not used, 2782 and there is no reason to add it in this case. 2783 2784 Signed-off-by: Hans de Goede <hdegoede (a] redhat.com> 2785 (Ported from xserver commit 4313122dea0df9affc280ee698e929489061ccc6) 2786 (Ported from amdgpu commit a3ca1500703837cbb8d49c554199a25dea7d5e1e) 2787 2788 commit eea370dc4f96e680cc29b416fe62d6e2ea09bc91 2789 Author: Hans de Goede <hdegoede (a] redhat.com> 2790 Date: Wed Sep 21 16:27:11 2016 +0900 2791 2792 Remove unnecessary fb addition from drmmode_xf86crtc_resize 2793 2794 drmmode_set_mode_major() is the only user of drmmode->fb_id and will 2795 create it if necessary. 2796 2797 Signed-off-by: Hans de Goede <hdegoede (a] redhat.com> 2798 (Ported from xserver commit 877453212166fdc912e0d687cdecee11aba563b5) 2799 (Ported from amdgpu commit 9ca1c24235ff5ab2e028333fc326e2eff008c574) 2800 2801 commit cf52c76484a4adfa3e0b2271ea0c00f46d32ca48 2802 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2803 Date: Tue Sep 13 18:21:37 2016 +0900 2804 2805 Free priv in amdgpu_set_pixmap_bo also if priv->bo == NULL 2806 2807 Fixes memory leak when destroying pixmaps with priv->bo == NULL. 2808 2809 Reported-by: Qiang Yu <qiang.yu (a] amd.com> 2810 (Ported from amdgpu commit 7f7f9825caf3983902491da27c16d14cd8bf9b7d) 2811 2812 commit 50abcc19e29bb50451d031b2c9e99d691d4faf70 2813 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2814 Date: Tue Sep 13 18:18:37 2016 +0900 2815 2816 glamor: Fix leak of pixmap private when replacing BO 2817 2818 Reported-by: Qiang Yu <qiang.yu (a] amd.com> 2819 (Ported from amdgpu commit 397aedafee437c125b8ac1feafb1c3b466842aeb) 2820 2821 commit 8523a733b6a5de6116a6332fefc871d4c32652d8 2822 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2823 Date: Mon Sep 12 18:54:33 2016 +0900 2824 2825 Propagate failure from radeon_set_pixmap_bo 2826 2827 (Ported from amdgpu commits c315c00e44afc91a7c8e2eab5af836d9643ebb88 2828 and 0d42082108c264568e2aadd15ace70e72388bc65) 2829 2830 commit 53be26b00e83f871f0afd39caa5a7a1d6ec4aea1 2831 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2832 Date: Wed Aug 31 16:46:56 2016 +0900 2833 2834 Add support for ScreenPtr::SyncSharedPixmap 2835 2836 This allows deferring shared pixmap updates between different drivers. 2837 2838 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2839 2840 commit 156b5bc00da2eecb71cad517136f8cd74cf5d2c9 2841 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2842 Date: Tue Sep 20 17:50:20 2016 +0900 2843 2844 Only define transform_region function for XF86_CRTC_VERSION >= 4 2845 2846 We're not using it with older xserver. 2847 2848 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2849 2850 commit 80cc892ee1ce54fad3cb7dd11bd9df18c359136f 2851 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2852 Date: Tue Sep 20 17:45:16 2016 +0900 2853 2854 Use local implementation of RegionDuplicate for older xserver 2855 2856 It was only added in xserver 1.15. Fixes build against older xserver. 2857 2858 Reported-by: Pali Rohr <pali.rohar (a] gmail.com> 2859 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2860 2861 commit 12d30eeb9711bd2b1609d6bbb74c4a1760596f72 2862 Author: Qiang Yu <Qiang.Yu (a] amd.com> 2863 Date: Wed Sep 14 17:03:03 2016 +0900 2864 2865 DRI2: Fix radeon_dri2_exchange_buffers width/height copy'n'paste error 2866 2867 Signed-off-by: Qiang Yu <Qiang.Yu (a] amd.com> 2868 (Ported from amdgpu commit 73c8dc000ad6b2b53ba3aa7155f5e8f6b55623b7) 2869 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2870 2871 commit 11cc6843aa3b745aa1361f1a65e465e16696b914 2872 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2873 Date: Wed Sep 14 17:01:13 2016 +0900 2874 2875 DRI2: Add interpolated_vblanks in radeon_dri2_get_crtc_msc 2876 2877 We need that in radeon_dri2_drawable_crtc as well for priv->vblank_delta 2878 to work as intended. 2879 2880 radeon_dri2_get_msc was already doing this. 2881 2882 Fixes hangs in some cases when using VDPAU via DRI2 and moving the 2883 window between CRTCs. 2884 2885 (Ported from amdgpu commit abd1a7901c95e4bc78415cf1b7923623b9177152) 2886 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2887 2888 commit 380daff43cfefe2a8e4d496eaf65673f2815c878 2889 Author: Qiang Yu <Qiang.Yu (a] amd.com> 2890 Date: Wed Sep 14 16:59:38 2016 +0900 2891 2892 Fix radeon_mode_hotplug crash on multi GPU platform. 2893 2894 On multi GPU platform, some screen is created by other GPU DDX. 2895 2896 Signed-off-by: Qiang Yu <Qiang.Yu (a] amd.com> 2897 (Ported from amdgpu commit 978242977e5dc905e1d5a46b1b0d34b356c7af26) 2898 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2899 2900 commit 4a5fa37f74c233c6b9c6a08306688628a8e216e8 2901 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2902 Date: Wed Sep 14 16:56:22 2016 +0900 2903 2904 Destroy all dedicated scanout buffers during CloseScreen 2905 2906 Fixes leaking active scanout buffers across a server reset, which also 2907 fixes server reset with glamor and active scanout buffers. 2908 2909 (Ported from amdgpu commit d96dabc71b1b32dc4b422a9633cdd4e0e95da052) 2910 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2911 2912 commit cc2555835cd4c5fd2ae4f999a4bf7c18cdb1dda4 2913 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2914 Date: Mon Sep 12 18:26:57 2016 +0900 2915 2916 glamor: Reallocate linear pixmap BO if necessary for DRI2 PRIME 2917 2918 Fixes corruption when using DRI2 PRIME render offloading with the master 2919 screen using this driver. 2920 2921 (Ported from amdgpu commit 0007c2f018ba663303d91d847e7c085269a23062) 2922 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2923 2924 commit b3e5259e60157fdbdf46ee59b1b78995c2b15f72 2925 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2926 Date: Mon Sep 12 18:12:06 2016 +0900 2927 2928 Move DRI2's local fixup_glamor helper to radeon_glamor_set_pixmap_bo 2929 2930 So it can be used outside of the DRI2 code. 2931 2932 (Ported from amdgpu commit 5518bf5d793439b5bab369e5fc18de9a4a3b9dd6) 2933 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2934 2935 commit 38632bbd5ff80a9cf8ce584b2bc499d17d15befe 2936 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2937 Date: Mon Sep 12 18:05:51 2016 +0900 2938 2939 Consolidate get_drawable_pixmap helper 2940 2941 There were two static helpers for the same purpose. Consolidate them 2942 into a single inline helper which can be used anywhere. 2943 2944 (Ported from amdgpu commit 641f4647b7f51dfd2da330376cd10fa9702b6423) 2945 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2946 2947 commit e91858e435672c32f9c4a854b3dec048199d6f7f 2948 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2949 Date: Mon Sep 12 18:02:07 2016 +0900 2950 2951 glamor: Fix radeon_glamor_share_pixmap_backing for priv->bo == NULL 2952 2953 Fixes crash when running a compositor and DRI_PRIME client via DRI2. 2954 2955 Reported-by: Qiang Yu <qiang.yu (a] amd.com> 2956 (Ported from amdgpu commit b36c77695ba77b59a0ccd868454e3af4fc04d5ff) 2957 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2958 2959 commit 38797a33117222dadbc89e5f21ed8cd5deef9bea 2960 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2961 Date: Fri Sep 2 11:08:28 2016 +0900 2962 2963 Make TearFree effective with PRIME slave scanout 2964 2965 TearFree can now prevent tearing with any possible display 2966 configuration. 2967 2968 Note that there may still be inter-GPU tearing if the primary GPU uses 2969 a different driver. 2970 2971 v2: 2972 * Also test dirty->slave_dst in radeon_prime_scanout_do_update 2973 2974 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> [v1] 2975 2976 commit eda1f3df6aaed683036369fe8820da4dac3c2ae2 2977 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2978 Date: Thu Sep 1 12:54:13 2016 +0900 2979 2980 Synchronize scanout pixmaps for TearFree 2981 2982 Copy the damaged areas which are still valid in the other scanout pixmap 2983 from there, then only copy the remaining damaged area from the screen 2984 pixmap. 2985 2986 This is slightly more efficient (only needs one Damage record instead of 2987 two, and only needs to copy each screen update across PCIe once with 2988 ShadowPrimary and a discrete GPU), and will be significantly more 2989 efficient for PRIME with the following change. 2990 2991 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 2992 2993 commit 2f6e5fb15f1a9ce523c85550e493f8bda9d0c00f 2994 Author: Michel Dnzer <michel.daenzer (a] amd.com> 2995 Date: Fri Sep 2 17:23:16 2016 +0900 2996 2997 Move up radeon_scanout_extents_intersect 2998 2999 Will be needed higher up by the following changes. No functional change. 3000 3001 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3002 3003 commit 5a57005178fc13b6f7e513458ca6dae72a3e5783 3004 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3005 Date: Thu Sep 1 17:19:27 2016 +0900 3006 3007 Factor out transform_region helper 3008 3009 While we're at it, fix leaking the memory allocated for xRectangles. 3010 3011 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3012 3013 commit 99232f64db52812a843cd616d263d3a6b90eef3d 3014 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3015 Date: Tue Dec 1 17:58:13 2015 +0900 3016 3017 Only copy from screen pixmap to shared pixmap on demand for slave scanout 3018 3019 Only copy once for each time we update the corresponding scanout pixmap. 3020 This can significantly reduce the bandwidth usage when there are 3021 frequent updates to the screen pixmap. 3022 3023 This initial implementation only works when both the master and slave 3024 screens use this driver. 3025 3026 v2: 3027 * Reduce churn in radeon_prime_scanout_update_handler 3028 * Clear the correct damage in radeon_dirty_update 3029 3030 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3031 3032 commit a92c27484703abc7c410b6ae0e4b8d1efbbb8e6f 3033 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3034 Date: Fri Aug 26 18:09:03 2016 +0900 3035 3036 Fix build against xserver < 1.13 3037 3038 pScreen->isGPU was only introduced in 1.13. 3039 3040 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97490 3041 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3042 3043 commit 4bd2d01552f18153afa03a8947b22eebf3d67c6b 3044 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3045 Date: Fri Aug 26 12:02:38 2016 +0900 3046 3047 Also handle disabled CRTCs in drmmode_clear_pending_flip 3048 3049 If disabling a CRTC had to be deferred due to a pending flip in 3050 drmmode_crtc_dpms, there may no longer be any outputs associated with 3051 the CRTC when we get here. So we have to check for !crtc->enabled and 3052 call drmmode_crtc_dpms in that case as well. 3053 3054 Fixes: 9090309e057d ("Wait for pending flips to complete before turning 3055 off an output or CRTC") 3056 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3057 3058 commit a36fdaff40d5b4795a1400c348a80eee94892212 3059 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3060 Date: Wed Aug 24 22:52:11 2016 +0900 3061 3062 Don't override crtc parameter value in drmmode_flip_handler/abort 3063 3064 When overriding the crtc parameter value of the last pending CRTC, 3065 drmmode_clear_pending_flip would work on the wrong CRTC, and the last 3066 pending CRTC's flip_pending flag might never get cleared. This would 3067 prevent that CRTC from properly turning off and back on again. 3068 3069 Fixes: 9090309e057d ("Wait for pending flips to complete before turning 3070 off an output or CRTC") 3071 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97392 3072 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3073 3074 commit e520ce0ec0adf91ddce5c932d4b3f9477fd49304 3075 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3076 Date: Thu Aug 25 18:22:34 2016 +0900 3077 3078 Also call drmmode_clear_pending_flip from radeon_scanout_flip_abort 3079 3080 Not doing so could break DPMS with TearFree. 3081 3082 Reported-and-Tested-by: furkan on IRC 3083 Fixes: 9090309e057d ("Wait for pending flips to complete before turning 3084 off an output or CRTC") 3085 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3086 3087 commit b0867063abb197b9134166706d99fcbe5f204bb5 3088 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3089 Date: Mon Nov 30 18:54:12 2015 +0900 3090 3091 Track damage accurately for RandR 1.4 slave scanout 3092 3093 This further reduces the PCIe bandwidth usage. 3094 3095 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3096 3097 commit ad0a0656dd0e74683e6d7789decba827aa29c221 3098 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3099 Date: Fri Nov 27 16:31:21 2015 +0900 3100 3101 Handle RandR 1.4 slave dirty updates via radeon_drm_queue 3102 3103 This reduces PCIe bandwidth usage and tearing. 3104 3105 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3106 3107 commit 4cfa4615f79f64062e5e771cd45dd7048f48b4f6 3108 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3109 Date: Fri Nov 27 16:53:30 2015 +0900 3110 3111 Use drmmode_crtc_scanout_* helpers for RandR 1.4 scanout pixmaps 3112 3113 This should allow using multiple CRTCs via RandR 1.4 even with xserver 3114 < 1.17. It also simplifies the code a little, and paves the way for 3115 following changes. 3116 3117 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3118 3119 commit 121a6de72da5fcf9a32408eff36b2235f3dfbcfe 3120 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3121 Date: Mon Jul 11 12:51:46 2016 +0900 3122 3123 Keep track of damage event related flushes per-client v2 3124 3125 This further reduces the compositing slowdown due to flushing overhead, 3126 by only flushing when the X server actually sends XDamageNotify events 3127 to a client, and there hasn't been a flush yet in the meantime. 3128 3129 v2: Use ScreenPrivateKey, fixes invalid memory access with GPU screens 3130 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3131 3132 commit 9090309e057dc703d1a5bffd88e6cae14108cfc3 3133 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3134 Date: Tue May 10 18:45:30 2016 +0900 3135 3136 Wait for pending flips to complete before turning off an output or CRTC 3137 3138 At least with older kernels, the flip may never complete otherwise, 3139 which can result in us hanging in drmmode_set_mode_major. 3140 3141 Fixes: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/1577170 3142 3143 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3144 3145 commit 9a1afbf61fbb2827c86bd86d295fa0848980d60b 3146 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3147 Date: Mon Jul 11 12:22:09 2016 +0900 3148 3149 Use EventCallback to avoid flushing every time in the FlushCallback 3150 3151 We only need to flush for XDamageNotify events. 3152 3153 Significantly reduces compositing slowdown due to flushing overhead, in 3154 particular with glamor. 3155 3156 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3157 3158 commit 94fe42f29e0b00a26e810581d6c438ac6d8ecd8a 3159 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3160 Date: Tue Jul 12 17:36:27 2016 +0900 3161 3162 Don't enable DRI3 by default with EXA 3163 3164 It doesn't work correctly in all cases, see e.g. 3165 https://bugs.freedesktop.org/show_bug.cgi?id=95475 . I'm not sure this 3166 is fixable, given EXA's architecture. 3167 3168 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3169 3170 commit 6d91fb4fc701895473ff675f440a8eef655e80ca 3171 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3172 Date: Tue Jul 12 15:57:28 2016 +0900 3173 3174 Don't enable micro-tiling for scanout buffers on pre-R600 3175 3176 The display engine didn't support it. 3177 3178 Fixes display corruption with options "TearFree" and "ShadowPrimary" 3179 (and rotation or transforms with current xserver) on pre-R600. 3180 3181 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3182 3183 commit a37af701768b12d86868a831a79f1e02ee4968cf 3184 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3185 Date: Wed Jul 6 12:46:01 2016 +0900 3186 3187 Add explicit RADEON_DRM_QUEUE_ERROR define 3188 3189 Should make the radeon_drm_queue_alloc error handling clearer, and gets 3190 rid of a compile warning about it returning NULL. 3191 3192 Reviewed-by: Alexandre Demers <alexandre.f.demers (a] gmail.com> 3193 3194 commit 024afff27eb9f4c60043214099ddb42bbfb3d289 3195 Author: Keith Packard <keithp (a] keithp.com> 3196 Date: Mon Jul 18 21:24:23 2016 -0700 3197 3198 Adapt to video API 22 3199 3200 Deal with drm fd wakeup 3201 3202 Signed-off-by: Keith Packard <keithp (a] keithp.com> 3203 Reviewed-and-Tested-by: Michel Dnzer <michel.daenzer (a] amd.com> 3204 3205 commit eac4db4f0cb22a23d6e69161dcaacc7d53978aae 3206 Author: Adam Jackson <ajax (a] redhat.com> 3207 Date: Tue Jul 19 10:03:56 2016 -0400 3208 3209 Adapt Block/WakeupHandler signature for ABI 23 3210 3211 Signed-off-by: Adam Jackson <ajax (a] redhat.com> 3212 3213 commit 3be841d0ae7d505cef325993205b12d15e98dba9 3214 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3215 Date: Fri Jun 24 11:12:49 2016 +0900 3216 3217 Only use RandR APIs if RandR is enabled 3218 3219 Fixes crash with Xinerama enabled, which disables RandR. 3220 3221 Fixes: https://bugs.debian.org/827984 3222 3223 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3224 3225 commit 7835558acdce318130ba4a09ef936fd675e3197d 3226 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3227 Date: Mon Mar 28 18:49:15 2016 +0900 3228 3229 Adapt to XF86_CRTC_VERSION 7 3230 3231 Now the HW cursor can be used with TearFree rotation. 3232 3233 This also allows always using the separate scanout pixmap mechanism for 3234 rotation, so that should be much smoother even without TearFree enabled. 3235 3236 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3237 3238 commit 0945db4d902056bda3d9ad4a4de2dfa685d10a70 3239 Author: Tan Hu <tan.hu (a] zte.com.cn> 3240 Date: Fri May 27 17:05:15 2016 +0800 3241 3242 EXA/6xx/7xx: accelerate PictOpOver with component alpha 3243 3244 Subpixel text rendering is typically done with a solid src and 3245 a pixmap mask. Traditionally, this cannot be accelerated in a single 3246 pass and requires two passes [1]. However, we can cheat a little 3247 with a constant blend color. 3248 3249 We can use: 3250 const.A = src.A / src.A 3251 const.R = src.R / src.A 3252 const.G = src.G / src.A 3253 const.B = src.B / src.A 3254 3255 dst.A = const.A * (src.A * mask.A) + (1 - (src.A * mask.A)) * dst.A 3256 dst.R = const.R * (src.A * mask.R) + (1 - (src.A * mask.R)) * dst.R 3257 dst.G = const.G * (src.A * mask.G) + (1 - (src.A * mask.G)) * dst.G 3258 dst.B = const.B * (src.A * mask.B) + (1 - (src.A * mask.B)) * dst.B 3259 3260 This only needs a single source value. src.A is cancelled down in 3261 the right places. 3262 3263 [1] http://anholt.livejournal.com/32058.html 3264 3265 r6xx still be used on some machine, 3266 Ported from commit 4375a6e75e5d41139be7031a0dee58c057ecbd07. 3267 3268 Signed-off-by: Tan Hu <tan.hu (a] zte.com.cn> 3269 Reviewed-by: Grigori Goronzy <greg (a] chown.ath.cx> 3270 3271 commit 9b9ad669c748f53247e53fa3f3b03a77da5e5cb3 3272 Author: Tan Hu <tan.hu (a] zte.com.cn> 3273 Date: Fri May 27 17:05:14 2016 +0800 3274 3275 EXA/6xx/7xx: fast solid pixmap support 3276 3277 Solid pixmaps are currently implemented with scratch pixmaps, which 3278 is slow. This replaces the hack with a proper implementation. The 3279 Composite shader can now either sample a src/mask or use a constant 3280 value. 3281 3282 r6xx still be used on some machine, 3283 Ported from commit 94d0d14914a025525a0766669b556eaa6681def7. 3284 3285 Signed-off-by: Tan Hu <tan.hu (a] zte.com.cn> 3286 Reviewed-by: Grigori Goronzy <greg (a] chown.ath.cx> 3287 3288 commit aa07b365d7b0610411e118f105e49daff5f5a5cf 3289 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3290 Date: Tue May 31 18:16:26 2016 +0900 3291 3292 Add .editorconfig file 3293 3294 Basically a conversion from .dir-locals.el, but also correctly handles 3295 files which predominantly use tabs for indentation. 3296 3297 Also, EditorConfig supports many more editors and IDEs. 3298 3299 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 3300 3301 commit 040a7b80e1fcbaa93ac17f7113d696d9b853cf8a 3302 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3303 Date: Tue Apr 12 18:18:43 2016 +0900 3304 3305 Explicitly set the fbcon pixmap pitch again 3306 3307 The kernel driver returns 0 as the pitch of the fbcon BO via the 3308 DRM_RADEON_GEM_GET_TILING ioctl, so we ended up using an incorrect 3309 pitch in some cases. 3310 3311 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94901 3312 3313 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3314 3315 commit 1181b9c582f10b6c523e4b2988e2ce87ecf3d367 3316 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3317 Date: Tue May 10 19:02:20 2016 +0900 3318 3319 Enable DRI3 by default when building for Xorg >= 1.18.3 3320 3321 Seems to work well enough in general now. 3322 3323 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3324 3325 commit c801f9f10a5d72d935faf21e72f7e7808fb4f05f 3326 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3327 Date: Wed Apr 13 11:12:38 2016 +0900 3328 3329 Handle Zaphod mode correctly in radeon_mode_hotplug 3330 3331 We need to scan both screens of the entity for existing connectors, and 3332 enumerate DVI & HDMI connectors consistently regardless of which screen 3333 they're assigned to. 3334 3335 Fixes crash when hot-(un)plugging connectors in Zaphod mode. 3336 3337 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93415 3338 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3339 3340 commit 1ca677309720e2f6c953c9e76f5b34c22a4416c6 3341 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3342 Date: Tue Mar 15 16:47:35 2016 +0900 3343 3344 present: Support async flips 3345 3346 The xserver Present code only calls radeon_present_flip with 3347 sync_flip=FALSE if radeon_present_screen_init sets 3348 PresentCapabilityAsync, and the latter only sets it if the kernel driver 3349 advertises support for async flips. 3350 3351 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3352 3353 commit 90a915c62d012e99193833aecc93974e68880c60 3354 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3355 Date: Tue Mar 15 16:42:16 2016 +0900 3356 3357 Add support for async flips to radeon_do_pageflip 3358 3359 Will be used by the next change. No functional change here. 3360 3361 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3362 3363 commit ba8b6288c8e6fc4be5d7144ecbe9a1f241881674 3364 Author: Qiang Yu <Qiang.Yu (a] amd.com> 3365 Date: Mon Apr 11 16:35:37 2016 +0900 3366 3367 Remove RR_Capability_SinkOutput for GPU without CRTC 3368 3369 Signed-off-by: Qiang Yu <Qiang.Yu (a] amd.com> 3370 (Ported from amdgpu commit a0bbb373f902e0ffc14570c85faec7e44134f62e) 3371 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3372 3373 commit bd4c72c8625996d842824ce4963f2d759fe2954a 3374 Author: Tom St Denis <tom.stdenis (a] amd.com> 3375 Date: Fri Apr 8 10:22:11 2016 -0400 3376 3377 dri3: Return NULL from radeon_dri3_pixmap_from_fd if calloc fails. 3378 3379 Signed-off-by: Tom St Denis <tom.stdenis (a] amd.com> 3380 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 3381 3382 commit 3300ea01fbdf770dc084eebbf2854fba35144220 3383 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3384 Date: Mon Apr 11 16:11:57 2016 +0900 3385 3386 Post 7.7.0 release version bump 3387 3388 commit df6662864b56d1aeeea4c67d9b668e64197afb20 3389 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3390 Date: Thu Apr 7 11:44:05 2016 +0900 3391 3392 Bump version for 7.7.0 release 3393 3394 commit c12614cfdc5729041c204e3ead1dbfde38011f0c 3395 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3396 Date: Mon Apr 4 14:59:41 2016 +0900 3397 3398 Update manpage entry for Option "TearFree" 3399 3400 It's now effective for rotation as well. 3401 3402 (Ported from amdgpu commit faf9d720b7d650f5f1ea657a874d08eac3972e60) 3403 3404 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3405 3406 commit 4693b1bd5b5c381e8b7b68a6f7f0c6696d6a68df 3407 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3408 Date: Thu Mar 31 17:02:55 2016 +0900 3409 3410 Identify DRM event queue entries by sequence number instead of by pointer 3411 3412 If the memory for an entry was allocated at the same address as that for 3413 a previously cancelled entry, the handler could theoretically be called 3414 prematurely, triggered by the DRM event which was submitted for the 3415 cancelled entry. 3416 3417 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3418 3419 commit 83734317e6bdaeebb4462a63f541e73a1d7c2f77 3420 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3421 Date: Wed Mar 30 11:44:09 2016 +0900 3422 3423 Update pixmap pitch in radeon_set_pixmap_bo 3424 3425 Stop second guessing it in drmmode_crtc_scanout_create. 3426 3427 Fixes display corruption in some cases with TearFree enabled. 3428 3429 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94751 3430 3431 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3432 3433 commit 0b3aac1de9db42bfca545fa331e4985836682ec7 3434 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3435 Date: Mon Mar 28 18:43:36 2016 +0900 3436 3437 DRI3: Refuse to open DRM file descriptor for ssh clients (v2) 3438 3439 Fixes hangs when attempting to use DRI3 on display connections forwarded 3440 via SSH. 3441 3442 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93261 3443 3444 v2: Don't do this for Xorg > 1.18.99.1 since the corresponding xserver 3445 change has landed in Git master. 3446 3447 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> (v1) 3448 3449 commit 8a6cd4bda05b9569b3dd0a5a75b2cc385b9ecba9 3450 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3451 Date: Fri Mar 25 11:50:15 2016 +0900 3452 3453 Revert "Use render node for DRI3 if available" 3454 3455 This reverts commit cd94248ffa7d8fe0b57476f79e7e860dee66d1b0. 3456 3457 It broke VDPAU<->GL interop with DRI3 enabled, because the Gallium VDPAU 3458 code doesn't support DRI3 yet. We can consider re-enabling this once 3459 there is a Mesa release where the Gallium VDPAU code supports DRI3. 3460 3461 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94675 3462 3463 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3464 3465 commit ed31cffba0d1bd4b14e5348a1456e4377277059c 3466 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3467 Date: Thu Mar 24 18:08:22 2016 +0900 3468 3469 Call RADEONBlockHandler_KMS before setting initial modes 3470 3471 Doing it the other way around meant that there was still a possibility 3472 for the front buffer contents to be uninitialized when they start being 3473 scanned out. 3474 3475 (Ported from amdgpu commit 4a60b4b1851a3cbc2d8ad9048d68eeb6947cf132) 3476 3477 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3478 3479 commit e592f32f8b5f5873fcc18b10a69dd5e4ccf11073 3480 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3481 Date: Wed Mar 23 18:25:48 2016 +0900 3482 3483 Require xserver 1.9 or newer 3484 3485 1.9.0 was released in August 2010. 3486 3487 We were already unintentionally relying on things not available in 1.8 3488 for at least a year, and nobody has complained. 3489 3490 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3491 3492 commit bde466e5d44cad64b4e4eceaa5de80fdbf86356e 3493 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3494 Date: Wed Mar 23 18:16:42 2016 +0900 3495 3496 Fix build against older versions of xserver 3497 3498 Also slightly clean up the error handling in radeon_scanout_do_update. 3499 3500 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94614 3501 3502 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3503 3504 commit b2b079be0e7ef2704c3df9ee25fe46166e450e83 3505 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3506 Date: Wed Mar 23 17:29:47 2016 +0900 3507 3508 Use radeon_alloc_pixmap_bo for allocating scanout BOs 3509 3510 This enables tiling for scanout BOs used for TearFree / ShadowPrimary, 3511 and simplifies the code a little. 3512 3513 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3514 3515 commit d21ac4669a8b2cdd4eec5e5a94d1950b7423b8b5 3516 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3517 Date: Wed Mar 23 17:27:27 2016 +0900 3518 3519 DRI3 only works with acceleration 3520 3521 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94214 3522 3523 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3524 3525 commit eb1a2c57e6c45d7ca57b2e2be643194243b7aae1 3526 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3527 Date: Wed Mar 23 17:23:31 2016 +0900 3528 3529 TearFree can only work with acceleration 3530 3531 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3532 3533 commit d670c5c9851b4eff21c845d26c7d7e4eb5ee0fa9 3534 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3535 Date: Fri Mar 18 16:46:08 2016 +0900 3536 3537 Check for xf86CursorResetCursor 3538 3539 If it's available, Xorg calls it on each mode configuration change. It 3540 does what xf86_reload_cursors does (and more), so we don't need to call 3541 the latter anymore. 3542 3543 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3544 3545 commit 7f3d0780ca65a90117c2a61362dbc0899bd9c0b0 3546 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3547 Date: Wed Mar 9 18:28:47 2016 +0900 3548 3549 Don't try DRI2/Present flipping while the HW cursor can't be used 3550 3551 Flipping doesn't interact correctly with SW cursor: A flip makes the SW 3552 cursor disappear. It will only appear again when the cursor is moved, 3553 but it will be surrounded by corruption, because the SW cursor code 3554 will restore stale screen contents at the old cursor location before 3555 drawing the cursor at the new location. 3556 3557 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3558 3559 commit 3de480e83c0a1824838d662d6d67c9fe85277298 3560 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3561 Date: Wed Mar 9 16:44:32 2016 +0900 3562 3563 Factor out HW cursor checking code into drmmode_can_use_hw_cursor 3564 3565 And add a check for RandR 1.4 multihead. 3566 3567 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3568 3569 commit 2e0b458e747c7df81dd5f9579a762262af3350e1 3570 Author: Ryan Kennedy <Ryan.Kennedy (a] amd.com> 3571 Date: Sun Mar 20 17:57:56 2016 -0400 3572 3573 Use correct tiling in drmmode_create_bo_pixmap 3574 3575 The callers were using 0 for the tiling flags, causing surfaces to be 3576 incorrectly treated as linear. This issue was causing corruption on Cayman 3577 just after login with GNOME3. The kernel was rejecting the cs in 3578 drmmode_copy_fb due to invalid texture pitch. 3579 3580 Signed-off-by: Ryan Kennedy <Ryan.Kennedy (a] amd.com> 3581 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 3582 3583 commit a03271de5ecdaa7790d1316e993c4450b91fe936 3584 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3585 Date: Wed Mar 9 18:23:10 2016 +0900 3586 3587 present: Return rotated CRTCs from radeon_present_get_crtc 3588 3589 Sync-to-vblank works fine with rotation. We're still checking for 3590 rotation in radeon_present_check_flip. 3591 3592 Returning NULL from here resulted in the xserver present code falling 3593 back to the fake CRTC running at 1 fps. 3594 3595 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3596 3597 commit 40191d82370eb7e58bd34c44966cbf44c3703229 3598 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3599 Date: Wed Mar 9 18:15:57 2016 +0900 3600 3601 present: Clear drmmode->fb_id before calling set_mode_major for unflip 3602 3603 Without this, drmmode_set_mode_major may just re-set the FB for the 3604 last flipped BO, in which case the display will probably freeze. 3605 3606 Reproduction recipe: Enable rotation while a fullscreen client is 3607 flipping. 3608 3609 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3610 3611 commit 798c4fd16d339b1ad5fd729cc884be084c60e38b 3612 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3613 Date: Fri Dec 25 18:57:42 2015 +0900 3614 3615 Make Option "TearFree" effective for rotated/reflected outputs as well (v2) 3616 3617 Support varies by xserver version: 3618 3619 < 1.12: No support for the driver handling rotation/reflection 3620 1.12-1.15: Support for driver handling rotation/reflection, but there's 3621 a bug preventing the HW cursor from being visible everywhere 3622 it should be on rotated outputs, so we can only support 3623 TearFree for reflection. 3624 >= 1.16: While the bug above is still there (fixes pending review), 3625 the driver can force SW cursor for rotated outputs, so we 3626 can support TearFree for rotation as well. 3627 3628 v2: Don't set crtc->driverIsPerformingTransform after xf86CrtcRotate if 3629 it wasn't set before. Fixes breaking rotation with TearFree disabled. 3630 3631 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> (v1) 3632 3633 commit eb611a2e4ecce7a1ab85fd72b9b78e3269311dd5 3634 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3635 Date: Tue Feb 2 18:04:11 2016 +0900 3636 3637 Consolidate pScreen usage in drmmode_set_mode_major 3638 3639 We were already relying on pScrn->pScreen being non-NULL in some cases, 3640 which is supposedly always true ever since this function is no longer 3641 getting called from ScreenInit. 3642 3643 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3644 3645 commit 06602171386e538081c298645fb7ca1a70fe80cc 3646 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3647 Date: Thu Feb 4 17:13:57 2016 +0900 3648 3649 Remove check for XF86_CRTC_VERSION 3 3650 3651 We require xserver >= 1.8, which was already at version 3. 3652 3653 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3654 3655 commit a88985f5d1e39caca49ceb65678aaa9cb622a0d2 3656 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3657 Date: Mon Feb 29 18:12:43 2016 +0900 3658 3659 Deal with modesets and page flips crossing on a CRTC 3660 3661 If we set a mode while a flip is pending, the kernel driver may program 3662 the flip to the hardware after the modeset. If that happens, the hardware 3663 will display the BO from the flip, whereas we will assume it displays the 3664 BO from the modeset. In other words, the display will most likely freeze, 3665 at least until another modeset. 3666 3667 Prevent this condition by waiting for a pending flip to finish before 3668 setting a mode. 3669 3670 Fixes display freezing when setting rotation or a transform with 3671 TearFree enabled. 3672 3673 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3674 3675 commit f5d968cbba3c9b7ec202161f2157d8d64778c817 3676 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3677 Date: Thu May 21 12:54:31 2015 +0900 3678 3679 Make DRM event queue xf86CrtcPtr based instead of ScrnInfoPtr based 3680 3681 This allows for a minor simplification of the code. 3682 3683 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3684 3685 commit e87365117acbd80b7d80fbb5eb30890ef7153291 3686 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3687 Date: Thu Feb 25 16:49:16 2016 +0900 3688 3689 DRI2: Also clear dri2_flipping when client disconnects before event 3690 3691 Fixes the following problem: 3692 3693 With DRI3 enabled, run glxgears with LIBGL_DRI3_DISABLE=1, make it 3694 fullscreen and press Escape while it's still fullscreen. This could 3695 result in dri2_flipping not getting cleared, spuriously preventing apps 3696 using DRI3 from flipping. 3697 3698 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3699 3700 commit d5dbb07db22d5420c81dfebc060f0dd86e7b8a20 3701 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3702 Date: Thu Feb 25 18:02:12 2016 +0900 3703 3704 Remove radeon_scanout_flip_handler 3705 3706 No longer necessary now that radeon_drm_queue_handler can handle 3707 e->handler == NULL. 3708 3709 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3710 3711 commit 3989766edde85d1abe7024577b98fc9b007bc02a 3712 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3713 Date: Thu Feb 25 16:43:48 2016 +0900 3714 3715 drm_queue: Don't abort events immediately from radeon_drm_abort_client 3716 3717 Keep them around until the DRM event arrives, but then call the abort 3718 functions instead of the handler functions. 3719 3720 This is a prerequisite for the following fix. 3721 3722 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3723 3724 commit 7bc089ee9740a3ef91f16e36786d8ed64e165acd 3725 Author: jimqu <Jim.Qu (a] amd.com> 3726 Date: Tue Feb 16 17:05:48 2016 +0900 3727 3728 glamor: Return NullPixmap on failure to create shareable pixmap 3729 3730 If we were asked to create a shareable pixmap, it doesn't make sense 3731 to return a pixmap which isn't shareable. Doing so caused trouble down 3732 the line such as a crash with older versions of glamor when trying to 3733 use GLX pixmaps of bpp < 32 via DRI2. 3734 3735 Signed-off-by: JimQu <jim.qu (a] amd.com> 3736 (ported from amdgpu commit 2fcb7dadd3c71cd405cbbaafc777697538ca9c29) 3737 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3738 3739 commit 4cc32031467157ab8788f5c684fb1ac67ae96ff9 3740 Author: jimqu <Jim.Qu (a] amd.com> 3741 Date: Tue Feb 16 17:03:24 2016 +0900 3742 3743 Move radeon_glamor_destroy_pixmap before radeon_glamor_create_pixmap 3744 3745 The next commit will call the former from the latter. No functional 3746 change. 3747 3748 Signed-off-by: JimQu <jim.qu (a] amd.com> 3749 (ported from amdgpu commit 5269a2228bff6023c1a7f3e8534027e1d7addc25) 3750 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3751 3752 commit a9b181083d56b5de9b6822d0ed89fcc86de777a6 3753 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3754 Date: Tue Feb 16 16:59:40 2016 +0900 3755 3756 Remove pci_dev test from radeon_get_scrninfo 3757 3758 The pci_dev parameter can never be NULL since we only support KMS. 3759 3760 Reported-by: Tom St Denis <tom.stdenis (a] amd.com> 3761 (ported from amdgpu commit 6e42c58375a4c3229da93c27bbd104af145c6163) 3762 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3763 3764 commit 2c7f068b0a0493daae86e5eacdf8d9812fcd28dc 3765 Author: Mykola Lysenko <Mykola.Lysenko (a] amd.com> 3766 Date: Mon Feb 15 18:21:39 2016 +0900 3767 3768 Initialize drmmode_crtc dpms_mode to DPMSModeOff 3769 3770 This disables query of disabled pipes for drmWaitVBlank on X start 3771 3772 Signed-off-by: Mykola Lysenko <Mykola.Lysenko (a] amd.com> 3773 (ported from amdgpu commit 4eb9cedca080b30c57ded349a397620ee7d0cd46) 3774 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3775 3776 commit 9358e44f03c95396b7f7a4f7398f5b24d2559bb2 3777 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3778 Date: Mon Feb 15 18:20:31 2016 +0900 3779 3780 sync: Check if miSyncShmScreenInit symbol is resolved at runtime 3781 3782 It may be disabled in the Xorg build, either explicitly or because the 3783 xshmfence library isn't available. 3784 3785 (ported from amdgpu commit 1d0b0c1794e65e581a48aa9fb19679d928d82a5d) 3786 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3787 3788 commit cd94248ffa7d8fe0b57476f79e7e860dee66d1b0 3789 Author: Jammy Zhou <Jammy.Zhou (a] amd.com> 3790 Date: Tue Feb 16 16:36:33 2016 +0900 3791 3792 Use render node for DRI3 if available 3793 3794 Signed-off-by: Jammy Zhou <Jammy.Zhou (a] amd.com> 3795 (ported from amdgpu commit ea558e645786b08d75307716036045170e97b43e) 3796 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3797 3798 commit d41fcccf2f11d0393d252e76bbbe6d233c4ac443 3799 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3800 Date: Wed Jan 20 11:20:30 2016 +0900 3801 3802 Set RADEON_GEM_NO_CPU_ACCESS flag for BOs which don't need CPU access 3803 3804 Failing to do this was resulting in the kernel driver unnecessarily 3805 leaving open the possibility of CPU access to those BOs. 3806 3807 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3808 3809 commit 9483a3d777919b224f70c3b4d01e4b320a57db31 3810 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3811 Date: Fri Jan 15 16:48:46 2016 +0900 3812 3813 Fix RandR CRTC transforms 3814 3815 Currently, Xorg will only transform the cursor as of the first time the 3816 cursor image changes after a transform is set. 3817 3818 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=80678 3819 3820 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3821 3822 commit b16856b25086ffb27365ac2249b8da921066ce62 3823 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3824 Date: Thu Jan 14 17:32:54 2016 +0900 3825 3826 Set the RandR primary output on startup if Xorg hasn't 3827 3828 Fixes xrandr (XRRGetOutputPrimary) not reporting any output as primary 3829 after startup. 3830 3831 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3832 3833 commit c7cf00487cd6d4a5d0f39d5b92ff04f6420d6a32 3834 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3835 Date: Fri Nov 27 12:52:28 2015 +0900 3836 3837 Build RandR 1.4 provider name from chipset name and bus ID 3838 3839 Instead of just "radeon", it's now e.g. "KAVERI @ pci:0000:00:01.0". 3840 3841 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3842 3843 commit ff9a6b6f079a8419f4e6fadfee778060618bf735 3844 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3845 Date: Fri Dec 11 17:21:27 2015 +0900 3846 3847 HAS_DIRTYTRACKING_ROTATION also supports multiple CRTCs 3848 3849 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3850 3851 commit 78fbca095ae9887a2d3de48bb07975e2d1126e68 3852 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3853 Date: Tue Dec 1 15:38:13 2015 +0900 3854 3855 Load fb module before glamoregl/shadow modules 3856 3857 Fixes unresolved symbols on some systems. 3858 3859 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93105 3860 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3861 3862 commit b19417e2fddf4df725951aea5ad5e9558338f59e 3863 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3864 Date: Fri Nov 20 18:48:56 2015 +0900 3865 3866 Don't advertise any PRIME offloading capabilities without acceleration 3867 3868 Acceleration is required even for display offloading. Trying to enable 3869 display offloading without acceleration resulted in a crash. 3870 3871 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3872 3873 commit 10b7c3def58bb34acc38f076bc230e25b454ab79 3874 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3875 Date: Tue Nov 10 16:31:09 2015 +0900 3876 3877 glamor: Deal with glamor_egl_destroy_textured_pixmap being removed 3878 3879 When it's not available, it's safe to call down to the glamor 3880 DestroyPixmap hook instead. 3881 3882 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3883 3884 commit 535e5438b2c32f774b9c8c27ee0289b4749548ef 3885 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3886 Date: Tue Nov 10 17:31:52 2015 +0900 3887 3888 glamor: Restore all ScreenRec hooks during CloseScreen 3889 3890 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3891 3892 commit dfb52777e43f3855476cff812d64de687a199674 3893 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3894 Date: Thu Nov 12 13:49:04 2015 +0900 3895 3896 Post 7.6.1 release version bump 3897 3898 commit 99cb8c3faf1a4ce368b7500f17a2a7868c15e8e8 3899 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3900 Date: Thu Nov 12 13:38:14 2015 +0900 3901 3902 Set version for 7.6.1 release 3903 3904 commit 4a2114a717ec8d78cacd8fe158e132d62013faac 3905 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3906 Date: Thu Nov 12 13:32:25 2015 +0900 3907 3908 Remove duplicate free(output_ids) call 3909 3910 Fixes double-free regression introduced by commit 875ad48e ("Simplify 3911 drmmode_set_mode_major() and avoid leaking memory."). 3912 3913 commit 960a9f37096b53ec9603a99c7b1f3679f915e03f 3914 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3915 Date: Thu Nov 12 13:20:18 2015 +0900 3916 3917 Post 7.6.0 release version bump 3918 3919 commit 9aae4b71fb402841439e3267fcb7f410d8429c3a 3920 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3921 Date: Thu Nov 12 11:56:48 2015 +0900 3922 3923 Bump version for 7.6.0 release 3924 3925 commit 98291869ac4a542a0b478920586407ff9d2c8ef0 3926 Author: Tom St Denis <tom.stdenis (a] amd.com> 3927 Date: Wed Nov 11 16:01:41 2015 +0900 3928 3929 Clean up radeon_dri2_create_buffer2() 3930 3931 Remove the depth_pixmap variable from the function and clear 3932 out any dead/odd behaviour that results. 3933 3934 Signed-off-by: Tom St Denis <tom.stdenis (a] amd.com> 3935 (ported from amdgpu commit 6000aef4e2f0a121b94023484406fb6f04688f74) 3936 3937 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3938 3939 commit c6fc7e309a8a922f94a1f5f3e8bfb9058cff7ad1 3940 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3941 Date: Wed Nov 11 15:59:06 2015 +0900 3942 3943 Properly handle drmModeAddFB failure in drmmode_crtc_scanout_allocate 3944 3945 We were printing an error message, but not propagating the failure. That 3946 would probably lead to trouble down the road. 3947 3948 (ported from amdgpu commit 21e72fb2418b5cc7fc849a9cf951186e209036b0) 3949 3950 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3951 3952 commit 4e4f4d53da0539ef9feb8766230a6e9927ae005b 3953 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3954 Date: Wed Nov 11 15:57:21 2015 +0900 3955 3956 Eliminate redundant data parameter from drmmode_crtc_scanout_create 3957 3958 drmmode_crtc_scanout_create just needs to call 3959 drmmode_crtc_scanout_allocate when scanout->bo is NULL. 3960 3961 This makes it clearer to the reader / compiler that 3962 drmmode_crtc_scanout_create doesn't dereference scanout->bo when it's 3963 NULL. 3964 3965 (ported from amdgpu commit 8da1d0c870e1081d77925807d6e3bbc61a23f54f) 3966 3967 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3968 3969 commit 421a7e797bdd58d83e81af7a6512cc715a3df514 3970 Author: Michel Dnzer <michel.daenzer (a] amd.com> 3971 Date: Wed Nov 11 15:51:58 2015 +0900 3972 3973 Don't advertise rotation support without hardware acceleration v2 3974 3975 Rotation currently doesn't work without acceleration (doesn't actually 3976 rotate with Option "NoAccel", crashes with Option "AccelMethod" "none" 3977 or when glamor fails to initialize) and would probably be too slow 3978 anyway. 3979 3980 v2: Also remove now dead code checking for ShadowFB from 3981 drmmode_crtc_scanout_allocate(). 3982 3983 (ported from amdgpu commit dc40582d5ff94d812cbc08f95cf14b80cd0f410d) 3984 3985 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 3986 3987 commit 875ad48e7b5cdb7beefbf18dddcbee3ed22b5446 3988 Author: Tom St Denis <tom.stdenis (a] amd.com> 3989 Date: Wed Nov 11 15:51:19 2015 +0900 3990 3991 Simplify drmmode_set_mode_major() and avoid leaking memory. 3992 3993 The function would leak the memory allocated for output_ids. This 3994 patch addresses that as well as simplifies the logic somewhat. 3995 3996 Signed-off-by: Tom St Denis <tom.stdenis (a] amd.com> 3997 (ported from amdgpu commit 460560502a1bdf26d06f3c30df46fa9f28ffb9e5) 3998 3999 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4000 4001 commit 789d7d6a04cca6b36fb088a074027807ccb8dd61 4002 Author: Tom St Denis <tom.stdenis (a] amd.com> 4003 Date: Wed Nov 11 15:48:51 2015 +0900 4004 4005 Clean up allocation in RADEONInitVideo() 4006 4007 The allocation of the adapters should use the correct sizeof (even if 4008 allocating an array of pointers). 4009 4010 Signed-off-by: Tom St Denis <tom.stdenis (a] amd.com> 4011 (ported from amdgpu commit db3bb2061b9ac16b0922d9afae99874820356a04) 4012 4013 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4014 4015 commit d88fa0dd5d37604de8efb05853738cfaca6a3166 4016 Author: Tom St Denis <tom.stdenis (a] amd.com> 4017 Date: Wed Nov 11 15:46:50 2015 +0900 4018 4019 Simplify pick best crtc to fold two loops into one 4020 4021 This patch folds the two for loops from radeon_pick_best_crtc() into 4022 one to reduce the LOC and make the routine easier to read. 4023 4024 Signed-off-by: Tom St Denis <tom.stdenis (a] amd.com> 4025 (ported from amdgpu commit 3055724aef76a624718f26d5f0f9e9d567ffbcfb) 4026 4027 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4028 4029 commit dbbcd75b3c80aba77673904d46bca97779fd8a8d 4030 Author: Tom St Denis <tom.stdenis (a] amd.com> 4031 Date: Wed Nov 11 12:54:54 2015 +0900 4032 4033 dri2: Avoid calculation with undefined msc value 4034 4035 If the get_msc() call fails for any reason we should avoid updating the 4036 vblank counter delta with undefined data. 4037 4038 Signed-off-by: Tom St Denis <tom.stdenis (a] amd.com> 4039 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> (minor fixups) 4040 (ported from amdgpu commit 8823c3d4c6db70cff7699b31088f2d92db8faaf4) 4041 4042 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4043 4044 commit 7186a8713ba004de4991f21c1a9fc4abc62aeff4 4045 Author: Stephen Chandler Paul <cpaul (a] redhat.com> 4046 Date: Fri Oct 23 09:59:36 2015 -0400 4047 4048 Handle failures in setting a CRTC to a DRM mode properly 4049 4050 This fixes a bug where running the card out of PPLL's when hotplugging 4051 another monitor would result in all of the displays going blank and 4052 failing to work properly until X was restarted or the user switched to 4053 another VT. 4054 4055 [Michel Dnzer: Pass errno instead of -ret to strerror()] 4056 4057 Signed-off-by: Stephen Chandler Paul <cpaul (a] redhat.com> 4058 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 4059 4060 commit 548e97b3b7d1e94075a54ca2bb4eb683025098a7 4061 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4062 Date: Fri Oct 16 16:26:58 2015 +0900 4063 4064 Call xf86CrtcRotate from initial drmmode_set_desired_modes call 4065 4066 Fixes various problems when rotation is specified in xorg.conf. 4067 4068 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92475 4069 4070 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4071 4072 commit ce9914af8d8d5243977023ec7b09c605f9cca8b4 4073 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4074 Date: Fri Oct 2 18:07:18 2015 +0900 4075 4076 Only align screen / scanout pixmap height where necessary 4077 4078 When using glamor acceleration, the pixmap's header has to have a height 4079 that matches exactly what the actual height is minus the GPU memory 4080 alignment. Otherwise CRTCs scanning out from the main scanout buffer 4081 (e.g. every CRTC that isn't rotated or transformed in some way) won't 4082 always work. This results in a bug where rotating one monitor in a 4083 multi-monitor setup won't always work properly. Easiest way to reproduce 4084 this: 4085 4086 - Have two monitors (I've gotten this working with a 1920x1080 and 4087 1280x1024, along with two 1920x1080s) 4088 - Rotate one of them from 0 to 90, then rotate the same monitor from 4089 90 to 180. The monitor that hasn't been rotated won't properly 4090 update, and will stay on a blank screen 4091 4092 This doesn't seem to make any difference when using EXA for 4093 acceleration. 4094 4095 Compared to Stephen Chandler's patch, this drops the height alignment 4096 in most places and only keeps it where it's really necessary. 4097 4098 Reported-and-Tested-by: Stephen Chandler Paul <cpaul (a] redhat.com> 4099 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4100 4101 commit 0288a4b87b65ba54f37fbeeea3cb32238deee92e 4102 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4103 Date: Wed Aug 19 16:54:14 2015 +0900 4104 4105 DRI2: Keep MSC monotonic when moving window between CRTCs 4106 4107 This mirrors the DRI3 implementation in xserver. Fixes VDPAU video 4108 playback hanging when moving the window between CRTCs. 4109 4110 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=66384 4111 4112 (Ported from amdgpu commit 63948ea091a9b324327ade7ec4fc5d67ca7e6f6f) 4113 4114 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4115 4116 commit 1d886b526dc49f32dc6744b7a882894bdac4e846 4117 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4118 Date: Wed Aug 19 18:11:45 2015 +0900 4119 4120 DRI2: Don't ignore rotated CRTCs in amdgpu_dri2_drawable_crtc 4121 4122 Waiting for vblank interrupts works fine with rotated CRTCs. The only 4123 case we can't handle with rotation is page flipping, which is handled 4124 in can_exchange(). 4125 4126 This fixes gnome-shell hanging on rotation, probably because 4127 amdgpu_dri2_get_msc returned MSC/UST 0 for rotated CRTCs. 4128 4129 (Ported from amdgpu commit 7b3212e33cd36fb6f122774df27b56ec4e1a22b8) 4130 4131 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4132 4133 commit 270da55340766074cabff8af4258e29fe2f0fc81 4134 Author: Alex Deucher <alexander.deucher (a] amd.com> 4135 Date: Mon Aug 10 15:32:34 2015 -0400 4136 4137 add new OLAND pci id 4138 4139 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 4140 4141 commit f123096877f0a0044f3d0315285441f2b5790c29 4142 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4143 Date: Mon Aug 10 23:34:39 2015 +0200 4144 4145 Make selection between DRI2 and DRI3 consistent with other drivers. (v2) 4146 4147 Add Option "DRI" to allow selection of maximum DRI level. 4148 4149 This allows the user to select the maximum level of DRI 4150 implementation to use, DRI2 or DRI3. It replaces the old 4151 option "DRI3" which had exactly the same purpose, but 4152 differs from the method used in both intel ddx and nouveau ddx. 4153 Make this consistent before a new stable driver is released. 4154 4155 v2: Retain handling of old Option "DRI3" for backwards 4156 compatibility, but Option "DRI" will take precedence 4157 over "DRI3" if both are provided. 4158 4159 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4160 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 4161 4162 commit 3791fceabf2cb037467dc41c15364e9f9ec1e47e 4163 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4164 Date: Thu Aug 6 16:27:01 2015 +0900 4165 4166 Wait for scanout BO initialization to finish before setting mode 4167 4168 This should avoid intermittent artifacts which could sometimes be visible 4169 when setting a new scanout pixmap, e.g. on server startup or when 4170 changing resolutions. 4171 4172 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4173 4174 commit 4e3dfa69e4630df2e0ec0f5b81d61159757c4664 4175 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4176 Date: Thu Aug 6 16:16:38 2015 +0900 4177 4178 Only call drmmode_copy_fb (at most) once on server startup 4179 4180 It doesn't make sense to copy the screen contents from console when VT 4181 switching back to Xorg or when Xorg resets. 4182 4183 Fixes intermittent artifacts when VT switching back from console to the 4184 gdm login screen. 4185 4186 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4187 4188 commit 09c7cdb923965f9a1ea11d2f449bc02114408938 4189 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4190 Date: Thu Jul 30 15:58:54 2015 +0900 4191 4192 glamor: Move declaration of struct radeon_pixmap out of #if/#else blocks 4193 4194 Reviewed-by: Dave Airlie <airlied (a] redhat.com> 4195 4196 commit 936582fde0db461c5c3d78ce4f5f4c93a88a489d 4197 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4198 Date: Thu Jul 30 15:56:36 2015 +0900 4199 4200 Remove unused local variable pRADEONEnt 4201 4202 ../../src/radeon_probe.c: In function 'radeon_get_scrninfo': 4203 ../../src/radeon_probe.c:157:22: warning: variable 'pRADEONEnt' set but not used [-Wunused-but-set-variable] 4204 RADEONEntPtr pRADEONEnt; 4205 ^ 4206 4207 Reviewed-by: Dave Airlie <airlied (a] redhat.com> 4208 4209 commit 5510cd6027d2387efdf33575e3bfc424cb11bfd8 4210 Author: Dave Airlie <airlied (a] redhat.com> 4211 Date: Mon Jul 27 09:22:57 2015 +1000 4212 4213 radeon: move radeon_pixmap forward declaration into other block 4214 4215 There is already a radeon_pixmap forward decl here, the #else 4216 block is missing one. 4217 4218 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4219 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 4220 4221 commit b32a0a3de84a44b9af4f1ca8be19f10d7fa31b12 4222 Author: Dave Airlie <airlied (a] redhat.com> 4223 Date: Wed Jul 22 13:35:07 2015 +1000 4224 4225 radeon: cleanup the entity rec 4226 4227 Some of these were set, some of them were 4228 always opposites, so clean things up. 4229 4230 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4231 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 4232 4233 commit fcb32231a38f9461d12720cbf72f63502197a711 4234 Author: Emil Velikov <emil.l.velikov (a] gmail.com> 4235 Date: Tue Jul 21 23:55:20 2015 +0100 4236 4237 Do not link radeon_drv.so against libpciaccess 4238 4239 It's only used/needed by ati_drv.so 4240 4241 Signed-off-by: Emil Velikov <emil.l.velikov (a] gmail.com> 4242 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 4243 4244 commit b6d871bf299c7d0f106c07ee4d8bd3b2337f53cc 4245 Author: Dave Airlie <airlied (a] gmail.com> 4246 Date: Thu Jul 9 19:25:36 2015 +1000 4247 4248 radeon: adopt for new X server dirty tracking APIs. 4249 4250 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 4251 4252 commit 95f5d09e3667ded027ae648c97eb4737d8bf67c5 4253 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4254 Date: Thu May 28 11:24:42 2015 +0900 4255 4256 present: Handle DPMS off in radeon_present_get_ust_msc 4257 4258 The DRM_IOCTL_WAIT_VBLANK ioctl may return an error during DPMS off, 4259 which would trigger an error message in drmmode_crtc_get_ust_msc. 4260 4261 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4262 4263 commit 211862b777d0be251a4662f5dd24f2d400544c09 4264 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4265 Date: Fri May 1 18:20:01 2015 +0900 4266 4267 present: Look at all CRTCs to determine if we can flip 4268 4269 Inspired by modesetting driver change by Kenneth Graunke. 4270 4271 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4272 4273 commit 802d33e474a82262d9cdf11b03568b0c4929cd0d 4274 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4275 Date: Fri May 1 18:54:16 2015 +0900 4276 4277 present: Fall back to modeset for unflip operation 4278 4279 It's not always possible to use the page flip ioctl for this, e.g. 4280 during DPMS off. We were previously just skipping the unflip in that 4281 case, which could result in hangs when setting DPMS off while a 4282 fullscreen Present app is running, e.g. at the GNOME3 lock screen. 4283 4284 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4285 4286 commit a8ed62010d5012dfb27773595c446b217f3c00c5 4287 Author: Piotr Redlewski <predlewski (a] gmail.com> 4288 Date: Mon Jul 6 19:57:07 2015 +0200 4289 4290 Do not try to enable already enabled CRTCs in DPMS hook 4291 4292 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91227 4293 4294 Signed-off-by: Piotr Redlewski <predlewski (a] gmail.com> 4295 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 4296 4297 commit fc07c370dac4649511eb7322d05a61605a539f39 4298 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4299 Date: Sun Jun 28 02:45:58 2015 +0200 4300 4301 Fix inconsistent default eg_tile_split in evergreen_accel.c 4302 4303 eg_tile_split() maps the default: switch-case to tile split 4304 flag 6, like for a tile split size of 4096 Bytes. All other 4305 instances of tile split mappings or reverse mappings in both 4306 the ddx and in Mesa's radeon gallium drivers assign the default: 4307 case to tile split flag 4, consistent with a default of 1024 4308 Bytes. 4309 4310 Make this one outlier consistent with other instances of 4311 mappings in the code to avoid surprises. 4312 4313 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4314 4315 commit ddaba449e8d6fe9fc0d97085e4045843fd8d7af9 4316 Author: Piotr Redlewski <predlewski (a] gmail.com> 4317 Date: Wed Jun 24 18:44:39 2015 +0200 4318 4319 Disable CRTCs when disabling the outputs 4320 4321 When turning the outputs off (DPMSModeOff), CRTCs stayed enabled. This led 4322 to higher gpu temperatures than with fglrx driver. 4323 4324 v2: when entering DPMS also disable active CRTCs 4325 v3: use drmmode_set_mode_major() for enabling CRTCs when leaving DPMS 4326 4327 Signed-off-by: Piotr Redlewski <predlewski (a] gmail.com> 4328 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 4329 4330 commit acc11877423ecd81a6e0a7f38466f80e43efee20 4331 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4332 Date: Tue Jun 16 17:34:57 2015 +0900 4333 4334 Skip disabled CRTCs in radeon_scanout_(do_)update 4335 4336 The vblank / page flip ioctls don't work as expected for a disabled CRTC. 4337 4338 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4339 4340 commit fc9fadaebbc8aead6e030d93a9ccd84561f8f59e 4341 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4342 Date: Mon Jun 22 00:09:17 2015 +0200 4343 4344 Don't set TILE_SPLIT flags if surface.tile_split == 0. 4345 4346 On pre-Evergreen hw, libdrm's r6_surface_best() helper 4347 for the surface managers radeon_surface_best() routine 4348 is a no-op and therefore doesn't assign any tile_split 4349 settings to created surfaces, so it leaves 4350 surface.tile_split on its "undefined" value of 0. 4351 4352 Mesa's DRI3/Present backend creates DRI3 Pixmaps via 4353 the DRIImage extension and the radeon gallium driver 4354 implementation of that extension uses the libdrm 4355 surface manager for backing bo creation and treats 4356 an undefined surface.tile_split==0, as returned by 4357 the surface manager for pre-evergreen, as a signal 4358 to not assign any tile_split flags to the DRI3 Pixmaps 4359 bo. 4360 4361 The ddx also uses libdrm surface manager to create the 4362 x-screen pixmap, but so far treated the returned undefined 4363 surface.tile_split==0 by mapping it to eg_tile_split()'s 4364 default tile_split flags, which are different from Mesa's 4365 tiling flags for DRI3 pixmaps. Under DRI3/Present this 4366 causes a mismatch of src pixmap and destination root 4367 pixmaps tiling flags and thereby prevents page flipping 4368 for pixmap presents. 4369 4370 Change the ddx code to treat surface.tile_split==0 the 4371 same way as the radeon gallium driver to avoid mismatched 4372 tiling flags and thereby allow DRI3/Present page-flip to 4373 work on pre-Evergreen hw. 4374 4375 Tested on RV730 and Evergreen "Juniper". 4376 4377 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4378 4379 commit 49f5b0bc301414df049e00d226034e3d6e56421b 4380 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4381 Date: Fri May 1 19:02:36 2015 +0900 4382 4383 Don't attempt a DRI2/Present page flip while the other one is flipping 4384 4385 Fixes corrupted display and hangs when switching between DRI2 and DRI3 4386 fullscreen apps, e.g. a compositor using DRI3 and a fullscreen app using 4387 DRI2 or vice versa. 4388 4389 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4390 4391 commit afab7839fc15722dbaa7203d00fe7f6ce5336b9d 4392 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4393 Date: Wed Jun 10 20:55:08 2015 +0200 4394 4395 Allow/Fix use of multiple ZaphodHead outputs per x-screen. (v2) 4396 4397 Defining multiple ZaphodHead outputs per x-screen in a 4398 multiple x-screen's per gpu configuration caused all 4399 outputs except one per x-screen to go dark, because 4400 there was a fixed mapping x-screen number -> crtc number, 4401 limiting the number of crtc's per x-screen to one. 4402 4403 On a ZaphodHead's setup, be more clever and assign 4404 as many crtc's to a given x-screen as there are 4405 ZaphodHeads defined for that screen, assuming 4406 there are enough unused crtc's available. 4407 4408 Tested on a triple display setup with different combos 4409 of one, two or three ZaphodHeads per one, two or three 4410 x-screens. 4411 4412 This is a port of similar code from xf86-video-nouveau. 4413 4414 v2: Implement suggestions by Michel Dnzer: Less verbose 4415 debug output, more clear warning message on crtc allocation 4416 failure. Move clearing of per gpu assigned_crtc mask to 4417 CloseScreen, indeed testing shows no need for the more 4418 complex new server generation check from v1. 4419 4420 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4421 (v1) Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4422 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 4423 4424 commit d7c82731a8bf3d381bc571b94d80d9bb2dd6e40d 4425 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4426 Date: Thu May 21 16:01:16 2015 +0900 4427 4428 DRI2: Fix handling of drmmode_crtc_get_ust_msc return code 4429 4430 Fixes regression introduced in commit 4431 76c2923ac5c7230a8b2f9f8329c308d28b44d9c0 ("DRI2: Split out helper for 4432 getting UST and MSC of a specific CRTC"). 4433 4434 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4435 4436 commit e58fc380ccf2a581d28f041fd74b963626ca5404 4437 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4438 Date: Thu May 28 10:59:22 2015 +0900 4439 4440 glamor: Only wait for GPU writes in radeon_glamor_prepare_access_cpu_ro 4441 4442 We don't need to wait for GPU reads to finish before reading with the 4443 CPU. 4444 4445 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4446 4447 commit 5f82a720374c9c1caebb42bfbeea1f0cf8847d28 4448 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4449 Date: Thu May 21 15:46:32 2015 +0900 4450 4451 present: Remove get_drmmode_crtc helper 4452 4453 It was getting the drmmode_crtc_private_ptr in a roundabout way. 4454 4455 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4456 4457 commit d64a13ebe0ecd241ee3260dbffd8f4a01e254183 4458 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4459 Date: Wed May 20 09:58:38 2015 +0900 4460 4461 Replace SyncFence typedef with declaration of struct _SyncFence 4462 4463 Fixes build failure in some environments: 4464 4465 CC radeon_sync.lo 4466 In file included from radeon_sync.c:28: 4467 /home/tinderbox/xorg-build/include/xorg/misync.h:31: error: redefinition of typedef 'SyncFence' 4468 radeon.h:93: note: previous declaration of 'SyncFence' was here 4469 make[2]: *** [radeon_sync.lo] Error 1 4470 4471 Reported-Tested-and-Acked-by: David Airlie <airlied (a] redhat.com> 4472 4473 commit 818c180c8932233b214a35ba0647af82f7bcec3d 4474 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4475 Date: Tue May 12 17:43:57 2015 +0900 4476 4477 glamor: Deal with glamor_glyphs_init being removed from xserver 4478 4479 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4480 4481 commit 7c4b78ab10b82c6dba9f72034ff7583859cca63d 4482 Author: Alex Deucher <alexander.deucher (a] amd.com> 4483 Date: Tue May 12 13:21:24 2015 -0400 4484 4485 add new bonaire pci id 4486 4487 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 4488 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 4489 4490 commit 80eede245d1eda27eaba108b0761a24bfd69aff6 4491 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4492 Date: Sat May 9 07:24:36 2015 +0200 4493 4494 present: Fix present notify timestamps and counts. 4495 4496 Invalid (msc,ust)=(0,0) were returned to the server 4497 because a wrong crtc_id was used to specify which 4498 crtc should deliver pageflip completion data. Fix 4499 it in accordance with the dri2 implementation. 4500 4501 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4502 4503 commit e3be8b0a8cf484ff16597413a6172788178e80c8 4504 Author: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4505 Date: Sat May 9 07:24:35 2015 +0200 4506 4507 present: Move check for async flips 4508 4509 Check for and reject currently unsupported async_flip 4510 inside radeon_present_check_flip() instead of inside 4511 radeon_present_flip(). 4512 4513 This way the server can detect early that async flips 4514 aren't supported by the ddx and can deal with this 4515 correctly by using its non-vsync'ed CopyRegion fallback. 4516 4517 Signed-off-by: Mario Kleiner <mario.kleiner.de (a] gmail.com> 4518 4519 commit 1584dc545c78e0bce8d4b4b9f26b568e2c211453 4520 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4521 Date: Tue Apr 28 17:14:11 2015 +0900 4522 4523 Skip disabled CRTCs in drmmode_set_desired_modes() even if set_hw == FALSE 4524 4525 Not skipping a disabled CRTC results in a crash. 4526 4527 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90187 4528 4529 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4530 4531 commit 340baf3804b7d9de8082badfc715a9bee7856e8f 4532 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4533 Date: Mon Apr 27 15:07:06 2015 +0900 4534 4535 Don't call radeon_get_pixmap_private() when using EXA 4536 4537 It only works when using glamor. 4538 4539 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90169 4540 4541 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4542 4543 commit 80f3d727f93cb6efedd2b39338d2301035965fe2 4544 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4545 Date: Wed Apr 22 13:33:15 2015 +0900 4546 4547 On screen resize, clear the new buffer before displaying it 4548 4549 Fixes garbage being intermittently visible during a screen resize. 4550 4551 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=27757#c7 4552 4553 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4554 4555 commit 3999bf88cdb192fe2f30b03bd2ed6f6a3f9f9057 4556 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4557 Date: Thu Apr 2 18:29:38 2015 +0900 4558 4559 Make drmmode_copy_fb() work with glamor as well 4560 4561 Needed for Xorg -background none. 4562 4563 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4564 4565 commit a4a8cdbcc10c1c5f07485a2af9e9e81e490c3e1d 4566 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4567 Date: Thu Apr 2 17:46:34 2015 +0900 4568 4569 Update scanout pixmap contents before setting a mode with it 4570 4571 This ensures the scanout pixmaps used for Option "TearFree" and Option 4572 "ShadowPrimary" have been initialized when their initial mode is set. 4573 4574 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4575 4576 commit 673e1c7637687c74fc9bdeeeffb7ace0d04b734f 4577 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4578 Date: Thu Apr 2 17:54:33 2015 +0900 4579 4580 Defer initial modeset until the first BlockHandler invocation 4581 4582 This ensures that the screen pixmap contents have been initialized when 4583 the initial modes are set. 4584 4585 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=27757 4586 4587 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4588 4589 commit 37874a4eeace5df04b02c8fc28f67b824e3f0f5f 4590 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4591 Date: Tue Apr 21 17:19:15 2015 +0900 4592 4593 Defer initial drmmode_copy_fb call until root window creation 4594 4595 That's late enough for acceleration to be fully initialized, but still 4596 early enough to set pScreen->canDoBGNoneRoot. 4597 4598 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4599 4600 commit 39c497f3efca5ca08343b884f44c93215dcdef31 4601 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4602 Date: Thu Apr 2 18:10:42 2015 +0900 4603 4604 Only copy fbcon BO contents if bgNoneRoot is TRUE 4605 4606 Otherwise, the X server will initialize the screen pixmap contents 4607 anyway. 4608 4609 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4610 4611 commit 1af044d7eee211fd4b248c236280274a68334da5 4612 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4613 Date: Wed Apr 22 18:43:02 2015 +0900 4614 4615 DRI2: Clear old->devPrivate.ptr in fixup_glamor 4616 4617 It doesn't point to the memory of the newly allocated BO. Fixes crash 4618 running piglit with Option "ShadowPrimary" enabled. 4619 4620 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4621 4622 commit 43159ef400c3b18b9f4d3e6fa1c4aef2d60d38fe 4623 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4624 Date: Mon Apr 20 18:44:36 2015 +0900 4625 4626 Add Option "TearFree" v4 4627 4628 Avoids tearing by flipping between two scanout BOs per (non-rotated) CRTC 4629 4630 v2: 4631 * Fix condition for TearFree log message (Richard Wilbur) 4632 * Log warning message about DRI page flipping being enabled because of 4633 TearFree (or ShadowPrimary) also when building without glamor support 4634 4635 v3: 4636 * Only override fb_id/x/y if all scanout pixmaps have been successfully 4637 allocated 4638 4639 v4: 4640 * Make log warning clearer if drmModePageFlip returns an error 4641 4642 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> (v1) 4643 4644 commit ed401f5b4f07375db17ff05e294907ec95fc946d 4645 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4646 Date: Fri Apr 3 12:11:00 2015 +0900 4647 4648 glamor: Remove the stride member of struct radeon_pixmap 4649 4650 Its value was always the same as that of the PixmapRec devKind member. 4651 4652 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4653 4654 commit ae92d1765fa370a8d94c2856ad6c45d273ec3c69 4655 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4656 Date: Wed Mar 18 16:23:24 2015 +0900 4657 4658 glamor: Add Option "ShadowPrimary" v2 4659 4660 When this option is enabled, most pixmaps (including the screen pixmap) 4661 are allocated in system RAM and mostly accessed by the CPU. Changed areas 4662 of the screen pixmap are copied to dedicated per-CRTC scanout pixmaps 4663 regularly, triggered by the vblank interrupt. 4664 4665 v2: 4666 * Set region data pointer to NULL for keeping only the extents 4667 * Move pRegion and pBox local variable declarations closer to their uses 4668 in drmmode_set_mode_major() 4669 4670 commit eea79472a84672ee4dc7adc4487cec6a4037048a 4671 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4672 Date: Wed Apr 1 15:51:52 2015 +0900 4673 4674 glamor: Add wrappers for the X server rendering hooks 4675 4676 They can choose between using the GPU or CPU for the operation. 4677 4678 commit 051d46382656ffc3e6cac1aab3aee7efdf5b623a 4679 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4680 Date: Thu Mar 19 17:34:27 2015 +0900 4681 4682 glamor: Add radeon_pixmap parameter to radeon_glamor_create_textured_pixmap 4683 4684 commit 2fa021f77372ca93375a3d13a0c43a9089674899 4685 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4686 Date: Thu Mar 19 17:38:47 2015 +0900 4687 4688 glamor: Remove unused function radeon_glamor_pixmap_is_offscreen 4689 4690 commit e96349ba6281fd18b8bf9c76629128276b065e6c 4691 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4692 Date: Wed Apr 1 16:04:13 2015 +0900 4693 4694 Add RADEON_CREATE_PIXMAP_SCANOUT flag 4695 4696 It means that the pixmap is used for scanout exclusively. 4697 4698 commit 9be7dd382e86d2b804de81d4e2af7431b2e16843 4699 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4700 Date: Wed Apr 1 12:40:16 2015 +0900 4701 4702 Split out struct drmmode_scanout for rotation shadow buffer information 4703 4704 Will be used for other kinds of dedicated scanout buffers as well. 4705 4706 commit c32b0530302739f6512755bccf281c2300617376 4707 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4708 Date: Thu Mar 19 17:46:48 2015 +0900 4709 4710 Rename scanout_pixmap_x field to prime_pixmap_x 4711 4712 To avoid confusion with upcoming changes. 4713 4714 commit 5921ba4ca705a0d919515626088f3948cc4848c1 4715 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4716 Date: Tue Mar 31 15:14:52 2015 +0900 4717 4718 present: Don't flip between BOs with different tiling parameters 4719 4720 The kernel driver doesn't handle that correctly yet. 4721 4722 Fixes or at least avoids issues with OpenGL fullscreen apps with DRI3 4723 enabled and using PRIME or with (2D) tiling disabled. 4724 4725 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89720 4726 4727 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4728 4729 commit 428e416e7cb04a1e0527da39cfebf70218879a77 4730 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4731 Date: Thu Apr 2 10:34:03 2015 +0900 4732 4733 Add radeon_get_pixmap_tiling_flags helper 4734 4735 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4736 4737 commit 98fb4199e63fedd4607cddee64bf602d6398df81 4738 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4739 Date: Tue Mar 31 12:25:18 2015 +0900 4740 4741 Only enable SYNC extension fences and the Present extension along with DRI3 4742 4743 This avoids some trouble with the Gallium nine state tracker, which uses 4744 the Present extension even when DRI3 is disabled. 4745 4746 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89772 4747 4748 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4749 4750 commit f8b0f23e9f4af9f9097ee5e72d53b45173163c41 4751 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4752 Date: Fri Mar 27 12:34:55 2015 +0900 4753 4754 DRI2: Use radeon_get_pixmap_handle 4755 4756 Now we can share pixmaps with no struct radeon_bo via DRI2. 4757 4758 Fixes VDPAU video playback freezing when using an OpenGL compositor with 4759 DRI3 enabled. 4760 4761 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89755 4762 4763 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4764 Tested-by: Nick Sarnie <commendsarnex (a] gmail.com> 4765 4766 commit ccbda955ebae1d457d35293833f12791e0f9fb0b 4767 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4768 Date: Fri Mar 27 12:16:44 2015 +0900 4769 4770 Move get_pixmap_handle helper to radeon_bo_helper.c 4771 4772 No functional change. 4773 4774 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4775 Tested-by: Nick Sarnie <commendsarnex (a] gmail.com> 4776 4777 commit de5ddd09db82141b263338dcf0c28e01f58268ee 4778 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4779 Date: Thu Mar 26 16:33:02 2015 +0900 4780 4781 Move radeon_drm_handler/abort_proc fields to drmmode_flipdata_rec 4782 4783 Their values are the same for all DRM flip ioctl calls within a single 4784 radeon_do_pageflip() call. 4785 4786 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4787 4788 commit e8c0f6319fbf4c3ea11e22ab1a68837031bdec8c 4789 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4790 Date: Thu Mar 26 16:27:35 2015 +0900 4791 4792 Simplify radeon_do_pageflip() error handling slightly more 4793 4794 We don't need the local variable old_fb_id. 4795 4796 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4797 4798 commit 8fc22360d5520469c82092ccb0fcf2af330c573f 4799 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4800 Date: Thu Mar 26 15:58:01 2015 +0900 4801 4802 Increase robustness against DRM page flip ioctl failures v3 4803 4804 Centralize cleanup, only clean up things that have been allocated for 4805 the failed ioctl call. 4806 4807 Fixes double-free after a flip ioctl failure. 4808 4809 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89681 4810 4811 v2: Only call drmModeRmFB for flipdata->old_fb_id on receipt of last DRM 4812 page flip event. Fixes Black screen on making glxgears fullscreen with 4813 DRI3 enabled. 4814 v3: Avoid double-free of flipdata in the unlikely case that calloc fails 4815 for flipcarrier, but only for the second or later CRTC. 4816 4817 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> (v2) 4818 4819 commit b16609b453bb1a181198cf27778f205dc23fb642 4820 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4821 Date: Thu Mar 26 17:15:21 2015 +0900 4822 4823 glamor: Handle GLAMOR_* flags removed from xserver 4824 4825 The behaviour is the same as when the removed flags were passed in. 4826 4827 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4828 4829 commit 391900a670addec39515f924265bfa9f8bfa9ec0 4830 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4831 Date: Wed Mar 18 12:56:07 2015 +0900 4832 4833 glamor: Avoid generating GEM flink names for BOs shared via DRI3 v2 4834 4835 We can't create our own struct radeon_bo representation in this case 4836 because destroying that would make the GEM handle inaccessible to glamor 4837 as well. So just get the handle directly via dma-buf. 4838 4839 v2: Close dma-buf file descriptor, pointed out by Axel Davy. 4840 4841 Reviewed-by: Axel Davy <axel.davy (a] ens.fr> 4842 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4843 4844 commit dfad91fffb5bd013785223b42d78886df839eacf 4845 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4846 Date: Thu Mar 19 12:37:01 2015 +0900 4847 4848 Present: Add radeon_present_get_pixmap_handle helper 4849 4850 Reviewed-by: Axel Davy <axel.davy (a] ens.fr> 4851 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4852 4853 commit 7b4fc4a677d252d01c2bf80d162bc35814059eaa 4854 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4855 Date: Thu Mar 19 12:15:52 2015 +0900 4856 4857 Make radeon_do_pageflip take a BO handle directly 4858 4859 Reviewed-by: Axel Davy <axel.davy (a] ens.fr> 4860 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4861 4862 commit af6076241c0d322b295a4e898407ae2472bd8eb4 4863 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4864 Date: Tue Mar 24 13:37:01 2015 +0900 4865 4866 Adapt radeon_sync.c for misyncshm.h vs misync.h as well 4867 4868 This should have been part of the previous commit. Without this, SYNC 4869 extension fences (and by extension DRI3) were disabled because it checked 4870 for HAVE_MISYNC_H, but configure now only defines HAVE_MISYNCSHM_H. 4871 4872 commit aa7825eb29cdf6ac9d7b28ad18186807ff384687 4873 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4874 Date: Mon Mar 23 10:12:17 2015 +0900 4875 4876 configure.ac: Check for misyncshm.h again 4877 4878 We need this header, which was added after the other misync*.h headers, 4879 along with DRI3 support. 4880 4881 commit 6291baaed261e36a63dc001307427fe00ba82259 4882 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4883 Date: Thu Mar 19 00:12:06 2015 +0900 4884 4885 EXA: Return NULL from radeon_get_pixmap_bo if there is no driver private 4886 4887 This mirrors what the glamor part of the function does. Fixes a crash 4888 running glxgears_pixmap with DRI3 enabled, reported by "marvin24" on IRC. 4889 4890 commit f68d9b5ba0c91a725b5eec9386c61bea8824c299 4891 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4892 Date: Wed Mar 18 18:29:27 2015 +0900 4893 4894 dri3: Use screen->CreatePixmap instead of fbCreatePixmap directly 4895 4896 Fixes crash with EXA reported by "marvin24" on IRC. 4897 4898 commit 4b0997e56dec0053cb2cb793e0f4ae35055ff7e6 4899 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4900 Date: Wed Mar 18 12:30:09 2015 +0900 4901 4902 glamor: Add glamor_fd_from_pixmap define for standalone glamor tree 4903 4904 commit af1862a37570fa512a525ab47d72b30400d2e2d6 4905 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4906 Date: Wed Mar 18 11:05:40 2015 +0900 4907 4908 Always include misync.h before other misync headers 4909 4910 Older versions of xserver didn't include misync.h from other misync 4911 headers as needed. 4912 4913 commit fcd37f65f485291084c174666bd605e215bf1398 4914 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4915 Date: Wed Mar 18 10:39:03 2015 +0900 4916 4917 DRI3: Use open hook instead of open_client 4918 4919 We don't need the client pointer, and this allows it to work with older 4920 versions of xserver as well. 4921 4922 commit f940fd741b15f03393037c5bb904cd74f012de9d 4923 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4924 Date: Wed Mar 18 10:06:47 2015 +0900 4925 4926 Initialize boolean variable before calling xf86GetOptValBool 4927 4928 We were just lucky that it contained 0 in most cases. 4929 4930 commit 4a98f60117c387a228d5cbaadb6e298fb4e865df 4931 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4932 Date: Tue Mar 17 16:47:11 2015 +0900 4933 4934 Add xorg_list_for_each_entry_safe fallback in radeon_list.h 4935 4936 Fixes another build failure against older versions of xserver reported 4937 by "Pali" on IRC. 4938 4939 commit 694e04720b886060fe3eefdce59741f218c8269f 4940 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4941 Date: Tue Mar 17 16:32:09 2015 +0900 4942 4943 Simplify includes in radeon_{dri3,present}.c 4944 4945 Just include radeon.h first. Fixes build failures against older versions 4946 of xserver because compat-api.h wasn't getting picked up. Reported by 4947 "Pali" on IRC. 4948 4949 commit 64e1e4dbdd3caee6f5d8f6b6c094b4533fa94953 4950 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4951 Date: Tue Mar 10 16:53:11 2015 +0900 4952 4953 Add DRI3 support v2 4954 4955 Must be enabled with 4956 4957 Option "DRI3" 4958 4959 in xorg.conf. 4960 4961 v2: Adapt to v2 of patches 11/12. 4962 4963 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4964 4965 commit 69ae0194778fe4276895839db92383f63f0b5de4 4966 Author: David Heidelberger <david.heidelberger (a] ixit.cz> 4967 Date: Fri Mar 6 17:57:22 2015 +0900 4968 4969 Handle tiling in radeon_set_shared_pixmap_backing 4970 4971 [ Michel Dnzer: Fixups for glamor ] 4972 Signed-off-by: David Heidelberger <david.heidelberger (a] ixit.cz> 4973 Signed-off-by: Axel Davy <axel.davy (a] ens.fr> 4974 4975 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4976 4977 commit 3c65fb849e1ba9fb6454bcaa55b696548902f3fc 4978 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4979 Date: Tue Mar 10 16:52:54 2015 +0900 4980 4981 Add support for the Present extension v2 4982 4983 v2: Fix up for struct radeon_drm_queue -> radeon_drm_queue_entry. 4984 Swapped order of patches 11 & 12 because the Present extension uses 4985 SYNC fences. 4986 4987 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4988 4989 commit 8fc9a241ab59ffbcdc178d6415332c88a54e85fe 4990 Author: Michel Dnzer <michel.daenzer (a] amd.com> 4991 Date: Thu Mar 5 18:18:56 2015 +0900 4992 4993 Add support for SYNC extension fences v2 4994 4995 v2: Swapped order of patches 11 & 12 because the Present extension uses 4996 SYNC fences. 4997 4998 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 4999 5000 commit 4a35e2f33d9cdfb608423046391311109f96fb6b 5001 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5002 Date: Thu Mar 5 18:34:07 2015 +0900 5003 5004 Fold radeon_glamor_flush into radeon_cs_flush_indirect 5005 5006 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5007 5008 commit 4b8adebb80158bcf81ada83bb88517febe931b12 5009 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5010 Date: Thu Mar 5 18:40:23 2015 +0900 5011 5012 Move #include "radeon_glamor.h" from radeon.h to where it's needed 5013 5014 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5015 5016 commit 76c2923ac5c7230a8b2f9f8329c308d28b44d9c0 5017 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5018 Date: Wed Mar 4 10:30:19 2015 +0900 5019 5020 DRI2: Split out helper for getting UST and MSC of a specific CRTC 5021 5022 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5023 5024 commit 6c3a721cde9317233072b573f9502348dcd21b16 5025 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5026 Date: Tue Mar 10 16:52:18 2015 +0900 5027 5028 DRI2: Use helper functions for DRM event queue management v3 5029 5030 This is mostly in preparation for Present support, but it also simplifies 5031 the DRI2 specific code a little. 5032 5033 v2: Fix up for struct radeon_drm_queue -> radeon_drm_queue_entry. 5034 v3: Removed excess 0s from conversion from microseconds to seconds, 5035 thanks to Richard Wilbur <richard.wilbur (a] gmail.com> for the catch! 5036 5037 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5038 5039 commit c3fa22a479e61d1899fa9d327d9c4e2a7f64b0c1 5040 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5041 Date: Wed Mar 11 17:47:59 2015 +0900 5042 5043 DRI2: Move radeon_dri2_flip_event_handler 5044 5045 In preparation for the next change, which will modify it to a static 5046 function which needs to be in the new place. No functional change. 5047 5048 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5049 5050 commit 65045112fdc8a9fa36e0e00f46739a6152b775ff 5051 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5052 Date: Wed Mar 11 17:33:54 2015 +0900 5053 5054 DRI2: Remove superfluous assignments to *_info->frame 5055 5056 That field is only used for page flipping. 5057 5058 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5059 5060 commit ad27f16f308079d06a2b1c788b3cb0947531253a 5061 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5062 Date: Wed Mar 11 17:30:11 2015 +0900 5063 5064 DRI2: Simplify blit fallback handling for scheduled swaps 5065 5066 Also use radeon_dri2_schedule_event when possible. 5067 5068 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5069 5070 commit b4af8a327ed8420f0ff4ea0f113f4a59406ed4d3 5071 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5072 Date: Mon Mar 2 18:59:54 2015 +0900 5073 5074 Add DRM event queue helpers v2 5075 5076 v2: Rename struct radeon_drm_queue to struct radeon_drm_queue_event, 5077 thanks to Richard Wilbur <richard.wilbur (a] gmail.com> for the suggestion. 5078 Also changed the corresponding parameter and local variable names from 5079 'q' to 'e'. 5080 5081 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5082 5083 commit 7c3470f4b659206ed23f761948936ede3a2dba3d 5084 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5085 Date: Fri Mar 6 18:51:29 2015 +0900 5086 5087 Move xorg_list backwards compatibility to new radeon_list.h header 5088 5089 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5090 5091 commit 7388d0b6c54b9d536fdb161e3aa61b326627b939 5092 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5093 Date: Thu Mar 5 17:35:06 2015 +0900 5094 5095 Require at least xserver 1.8 5096 5097 So we can rely on the list.h header. 5098 5099 xserver 1.8 was released in April 2010. 5100 5101 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5102 5103 commit b8ec9ed4fe86952763b963c86f0af0dcae69aa6c 5104 Author: Jerome Glisse <jglisse (a] redhat.com> 5105 Date: Thu Feb 19 14:47:41 2015 -0500 5106 5107 Proper leak fix, previous leak fix was bogus. 5108 5109 Signed-off-by: Jrme Glisse <jglisse (a] redhat.com> 5110 5111 commit 63dc36dc49f93cb00111b497ab6805194bc9d240 5112 Author: Jerome Glisse <jglisse (a] redhat.com> 5113 Date: Thu Feb 19 11:43:08 2015 -0500 5114 5115 Avoid leaking memory on output. 5116 5117 Signed-off-by: Jrme Glisse <jglisse (a] redhat.com> 5118 5119 commit c80ea1e3e8c5f155cedadbe4588870fb1f410ab4 5120 Author: Rolf Eike Beer <eb (a] emlix.com> 5121 Date: Tue Jan 13 12:11:26 2015 +0100 5122 5123 radeon: remove unneeded inclusion of sarea.h 5124 5125 None of the structs or defines from that header is used. 5126 5127 Signed-off-by: Rolf Eike Beer <eb (a] emlix.com> 5128 5129 commit 04da199231bb3f11cf17f94574a8df05855a7b82 5130 Author: Adam Jackson <ajax (a] redhat.com> 5131 Date: Wed Dec 17 14:03:58 2014 -0500 5132 5133 Remove dead accelDFS flag 5134 5135 Signed-off-by: Adam Jackson <ajax (a] redhat.com> 5136 5137 commit c9f8f642fd495937400618a4fc25ecae3f8888fc 5138 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5139 Date: Mon Nov 17 12:32:29 2014 +0900 5140 5141 Prefer drmModeSetCursor2 over drmModeSetCursor 5142 5143 The former includes information about the position of the hotspot within 5144 the cursor image. 5145 5146 Copied from xf86-video-modesetting. 5147 5148 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5149 5150 commit 2f11dcd43966cf2ee26e61960fd72e6644f5e037 5151 Author: Dave Airlie <airlied (a] redhat.com> 5152 Date: Mon Nov 10 13:49:29 2014 +1000 5153 5154 radeon: add support for DP 1.2 display hotplug (v2) 5155 5156 This allows for dynamic creation of conneectors when the 5157 kernel tells us. 5158 5159 v2: fix dpms off crash 5160 5161 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5162 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 5163 5164 commit c88424d1f4aaa78b569e5d44f0b4a47de2f422f4 5165 Author: Dave Airlie <airlied (a] redhat.com> 5166 Date: Mon Nov 10 14:17:54 2014 +1000 5167 5168 radeon: move output name creation to its own function 5169 5170 The secondary indent is deliberate to make the next patch more 5171 parseable for mst support. 5172 5173 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5174 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 5175 5176 commit 32b003cb7657e07d5af6338ad44d768eda87fd33 5177 Author: Dave Airlie <airlied (a] redhat.com> 5178 Date: Mon Nov 10 14:12:34 2014 +1000 5179 5180 radeon: stop caching mode resources 5181 5182 This is step one towards MST connector hotplug support, 5183 it stop caching the mode resources structure, and 5184 just passes a pointer to it around. 5185 5186 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5187 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 5188 5189 commit 7c7b38e0b375b6e8853ad2d1092302ea83f6f570 5190 Author: David Heidelberger <david.heidelberger (a] ixit.cz> 5191 Date: Sun Oct 12 16:34:21 2014 +0200 5192 5193 radeon/vdpau: don't report VDPAU for < r300 5194 5195 Signed-off-by: David Heidelberger <david.heidelberger (a] ixit.cz> 5196 5197 commit c74de9fec13fac2c836bb2a07ae6f90e1d61e667 5198 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5199 Date: Wed Aug 6 11:08:00 2014 +0900 5200 5201 PRIME: Don't advertise offload capabilities when acceleration is disabled 5202 5203 Xorg tends to crash if the user tries to actually use the offload 5204 capabilities with acceleration disabled. 5205 5206 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57200 5207 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5208 5209 commit 636a6382f80412ae4e28abe0cd9f98f84cd52291 5210 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5211 Date: Wed Oct 15 12:02:17 2014 +0900 5212 5213 Remove duplicate OPTION_PAGE_FLIP entry 5214 5215 Reported by 'amanual' on IRC 5216 5217 commit c854b4479ec5122390ede81b655b259efd9b1800 5218 Author: Alex Deucher <alexdeucher (a] gmail.com> 5219 Date: Thu Oct 2 10:22:22 2014 -0400 5220 5221 radeon: bump version post release 5222 5223 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 5224 5225 commit 068a59e010ce6bfcd54f5a18cc08c55c54b8618d 5226 Author: Alex Deucher <alexander.deucher (a] amd.com> 5227 Date: Wed Oct 1 23:38:52 2014 -0400 5228 5229 radeon: bump version for release 5230 5231 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5232 5233 commit 5fee565e420d3efbf72fdf4f63c3d5d93d8ceddb 5234 Author: Alex Deucher <alexander.deucher (a] amd.com> 5235 Date: Wed Oct 1 23:36:31 2014 -0400 5236 5237 radeon: require libdrm_radeon 2.4.58 for latest SI/CI pci ids 5238 5239 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5240 5241 commit 56c7fb8ada4cd9cf096c6b06a8d368d286d74e68 5242 Author: Thomas DeRensis <tderensis (a] gmail.com> 5243 Date: Wed Sep 10 21:20:32 2014 -0400 5244 5245 radeon: fix build warnings regarding const qualifier 5246 5247 Signed-off-by: Thomas DeRensis <tderensis (a] gmail.com> 5248 5249 commit b9a9b244336ed50d7df4f36135f875a3bb6ca948 5250 Author: Thomas DeRensis <tderensis (a] gmail.com> 5251 Date: Thu Sep 4 21:45:19 2014 -0400 5252 5253 radeon: fix a leak in radeon_vbo_get_bo() 5254 5255 The dma_bo struct was never freed in the error path. Reported by cppcheck. 5256 5257 Signed-off-by: Thomas DeRensis <tderensis (a] gmail.com> 5258 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 5259 5260 commit ba5fcc4d6b6eae3f2c55c6229a53d24d160789c1 5261 Author: Alex Deucher <alexander.deucher (a] amd.com> 5262 Date: Thu Aug 21 11:32:45 2014 -0400 5263 5264 radeon/kms: fix mullins pci id 5265 5266 strange errant change from: 5267 39fef269f521c92a2a31c80447e9401bacb3797a 5268 5269 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5270 5271 commit efef34930650d6d80f7b527f4cee76d9e5954ace 5272 Author: Alex Deucher <alexander.deucher (a] amd.com> 5273 Date: Thu Aug 21 11:30:17 2014 -0400 5274 5275 radeon/kms: add new SI pci ids 5276 5277 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5278 5279 commit 39fef269f521c92a2a31c80447e9401bacb3797a 5280 Author: Alex Deucher <alexander.deucher (a] amd.com> 5281 Date: Thu Aug 21 11:27:49 2014 -0400 5282 5283 radeon/kms: add new CIK pci ids 5284 5285 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5286 5287 commit 7dab6b2fe9cf690b0dae1c127d0374322bd8278e 5288 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5289 Date: Wed Aug 20 17:51:35 2014 +0900 5290 5291 Revert "glamor: Set environment variable RADEON_THREAD=0" 5292 5293 This reverts commit 4b5060f357a3cb248c9359c92c1e9c42ef6434c8. 5294 5295 Further testing shows that disabling the thread is only a moderate win in 5296 some cases, but a much bigger loss in some other cases. 5297 5298 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5299 5300 commit 1ff383360886c5cee1bb75abcc20675aca0336a6 5301 Author: Alex Deucher <alexander.deucher (a] amd.com> 5302 Date: Tue Aug 12 12:27:12 2014 -0400 5303 5304 radeon: fix warnings when building against older xservers 5305 5306 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5307 5308 commit 94202cbfbca05a503acdc1cca2f8409d141173af 5309 Author: Alex Deucher <alexdeucher (a] gmail.com> 5310 Date: Fri Aug 1 21:55:40 2014 +0200 5311 5312 radeon: enable hawaii accel conditionally (v3) 5313 5314 Only if the kernel has the new CP firmware. 5315 5316 v2: check value of ACCEL_WORKING2 5317 v3 (Andreas Boll): 5318 - check for value 2 or 3 of ACCEL_WORKING2 5319 - update man page 5320 5321 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> (v2) 5322 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5323 Signed-off-by: Andreas Boll <andreas.boll.dev (a] gmail.com> 5324 5325 commit 03930edd49f6b8a8d79910c7be5408b47db9649b 5326 Author: Andreas Boll <andreas.boll.dev (a] gmail.com> 5327 Date: Mon Aug 4 16:23:13 2014 +0200 5328 5329 radeon: remove definitions already present in radeon_drm.h 5330 5331 Signed-off-by: Andreas Boll <andreas.boll.dev (a] gmail.com> 5332 5333 commit 91849fba0742ef61ba327e71fc3ce8f754af0a6f 5334 Author: Andreas Boll <andreas.boll.dev (a] gmail.com> 5335 Date: Mon Aug 4 16:23:12 2014 +0200 5336 5337 radeon: drop radeon_drm.h 5338 5339 Now we use libdrm's radeon_drm.h. 5340 5341 Signed-off-by: Andreas Boll <andreas.boll.dev (a] gmail.com> 5342 5343 commit b4658901bf1d619f8ff9b5f94344894f935ee6c8 5344 Author: Andreas Boll <andreas.boll.dev (a] gmail.com> 5345 Date: Mon Aug 4 16:23:11 2014 +0200 5346 5347 radeon: move RADEON_TILING_{MASK, LINEAR} from radeon_drm.h to radeon.h 5348 5349 This allows us to drop radeon_drm.h from xf86-video-ati and use instead 5350 radeon_drm.h from libdrm. 5351 5352 Signed-off-by: Andreas Boll <andreas.boll.dev (a] gmail.com> 5353 5354 commit 656bae7361c1e018553ef6d6d8c9efad616a4513 5355 Author: Andreas Boll <andreas.boll.dev (a] gmail.com> 5356 Date: Mon Aug 4 16:23:10 2014 +0200 5357 5358 radeon: drop redundant radeon_drm.h includes 5359 5360 Already included via radeon.h. 5361 5362 Signed-off-by: Andreas Boll <andreas.boll.dev (a] gmail.com> 5363 5364 commit fbf575cb010e558a87fad0de45660738c8180896 5365 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5366 Date: Wed Jul 30 17:03:24 2014 +0900 5367 5368 Add Emacs .dir-local.el file 5369 5370 Based on the one from the Gallium radeon winsys, but enabling tabs for 5371 indentation. 5372 5373 commit 4b5060f357a3cb248c9359c92c1e9c42ef6434c8 5374 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5375 Date: Thu Jun 19 18:27:59 2014 +0900 5376 5377 glamor: Set environment variable RADEON_THREAD=0 5378 5379 Reviewed-by: Marek Olk <marek.olsak (a] amd.com> 5380 5381 commit 9b54caf6509a9c02dd17c9c43d6be8f7ddc98054 5382 Author: Hans de Goede <hdegoede (a] redhat.com> 5383 Date: Mon Jul 28 15:55:12 2014 +0200 5384 5385 configure: Include xorg-server.h before glamor.h 5387 5388 glamor.h cannot be included without first including xorg-server.h, this also 5389 applies to including it from configure snippets. 5390 5391 Without this the configure glamor checks fail on systems with the latest 5392 glibc, throwing this error: 5393 5394 In file included from /usr/include/xorg/misc.h:115:0, 5395 from /usr/include/xorg/screenint.h:50, 5396 from /usr/include/xorg/scrnintstr.h:50, 5397 from /usr/include/xorg/glamor.h:32, 5398 from conftest.c:61: 5399 /usr/include/xorg/os.h:579:2: error: expected identifier or '(' before 5400 '__exten 5401 strndup(const char *str, size_t n); 5402 ^ 5403 5404 This is caused by HAVE_STRNDUP not being set (it is set from xorg-server.h), 5405 causing os.h to redefine it. 5406 5407 Signed-off-by: Hans de Goede <hdegoede (a] redhat.com> 5408 5409 commit c4ae0e2cbcc0e2ebf9f13ee92d59b5120254a1dc 5410 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5411 Date: Mon Jun 30 10:20:12 2014 +0900 5412 5413 Handle CRTC DPMS from output DPMS hooks 5414 5415 This fixes at least two issues: 5416 5417 The CRTC DPMS hook isn't called after a modeset, so the vertical blank 5418 interrupt emulation code considered the CRTC disabled after a modeset. As 5419 a side effect, page flipping was no longer used after a modeset. 5420 5421 This change also makes sure the vertical blank interrupt emulation code 5422 runs before the hardware CRTC is disabled and after it's enabled from the 5423 output DPMS hook. The wrong order could cause gnome-shell to hang after 5424 a suspend/resume and/or DPMS off/on cycle. 5425 5426 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 5427 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5428 5429 commit c0c3cac8613e31e310242695d0955b452f116e25 5430 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 5431 Date: Wed Jun 25 16:21:57 2014 +0200 5432 5433 bump version post release 5434 5435 commit 906a0ec9224146098bb4581486129d2934d36495 5436 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 5437 Date: Wed Jun 25 16:10:21 2014 +0200 5438 5439 bump version for release 5440 5441 commit cc615d06db0332fc6e673b55632bcc7bf957b44b 5442 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5443 Date: Wed Jun 4 16:58:53 2014 +0900 5444 5445 Rename Option "NoAccel" to "Accel" 5446 5447 Removes the need for a double negation when forcing acceleration on. 5448 5449 Note that this change is backwards compatible, as the option parser 5450 automagically handles the 'No' prefix. 5451 5452 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5453 5454 commit 851b2cf8714618843725f6d067915375485ade9d 5455 Author: Adam Jackson <ajax (a] redhat.com> 5456 Date: Wed May 21 09:34:32 2014 -0400 5457 5458 kms: Use own thunk function instead of shadowUpdatePackedWeak 5459 5460 I plan to delete the Weak functions from a future server. 5461 5462 Signed-off-by: Adam Jackson <ajax (a] redhat.com> 5463 5464 commit b2dba2906f0b2284f17f53fd5251ba0f03d52a8b 5465 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5466 Date: Thu May 15 16:07:53 2014 +0900 5467 5468 Don't disable acceleration on >= SI on attempts to force EXA 5469 5470 Also make this case clear in the log file: 5471 5472 (WW) RADEON(0): EXA not supported, using glamor 5473 5474 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5475 5476 commit be1469cc23aba46daf3293b3d09c5f2e792e7f42 5477 Author: Alex Deucher <alexander.deucher (a] amd.com> 5478 Date: Fri May 2 17:48:07 2014 -0400 5479 5480 radeon: enable tiling for mullins 5481 5482 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5483 5484 commit 2ae8e4b8d12f5c9bab6655eb8cd3c5c1d5cfb10e 5485 Author: Samuel Li <samuel.li (a] amd.com> 5486 Date: Tue Nov 12 15:30:42 2013 -0500 5487 5488 radeon: add Mullins pci ids. 5489 5490 Signed-off-by: Samuel Li <samuel.li (a] amd.com> 5491 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5492 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 5493 5494 commit 5c86a3461597647224c33d5190b4531aeeb2655f 5495 Author: Samuel Li <samuel.li (a] amd.com> 5496 Date: Thu Apr 17 15:17:28 2014 -0400 5497 5498 radeon: add support for Mullins. 5499 5500 Signed-off-by: Samuel Li <samuel.li (a] amd.com> 5501 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5502 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 5503 5504 commit fdb90ffc50acbb7d5ba0598470f9feeac6ce55fc 5505 Author: Alex Deucher <alexander.deucher (a] amd.com> 5506 Date: Fri May 2 17:45:45 2014 -0400 5507 5508 radeon: require libdrm_radeon 2.4.54 for mullins support 5509 5510 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5511 5512 commit 06e3c8c53ef4bd159f5864eabf726438d008b49a 5513 Author: Dave Airlie <airlied (a] redhat.com> 5514 Date: Wed Apr 23 13:39:42 2014 +1000 5515 5516 radeon: fix use-after-free in modesetting cleanup 5517 5518 noticed while looking at something else. 5519 5520 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 5521 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 5522 5523 commit dbac18c361f9e514ecb40d0617f9d68b65a542e0 5524 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5525 Date: Mon Apr 28 17:51:56 2014 +0900 5526 5527 Revert "Adapt to load_cursor_argb signature change in xserver 1.15.99.902" 5528 5529 This reverts commit 48d3dbc8a0d3bfde88f46e402e530438f9317715. 5530 5531 xserver Git has been updated to be backwards compatible with the 5532 previous API. 5533 5534 commit c84230d686c078aac1dc98d82153f8b02521b2e1 5535 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5536 Date: Fri Apr 25 09:17:51 2014 +0900 5537 5538 dri2: Handle PRIME for source buffer as well in radeon_dri2_copy_region2 5539 5540 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=77810 5541 5542 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5543 5544 commit 48d3dbc8a0d3bfde88f46e402e530438f9317715 5545 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5546 Date: Tue Apr 15 17:45:35 2014 +0900 5547 5548 Adapt to load_cursor_argb signature change in xserver 1.15.99.902 5549 5550 Apart from the compiler warning below, not doing this may result in 5551 accidentally using software cursors. 5552 5553 ../../src/drmmode_display.c:808:5: warning: initialization from incompatible pointer type [enabled by default] 5554 .load_cursor_argb = drmmode_load_cursor_argb, 5555 ^ 5556 ../../src/drmmode_display.c:808:5: warning: (near initialization for 'drmmode_crtc_funcs.load_cursor_argb') [enabled by default] 5557 5558 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5559 5560 commit 409786a8f780d78a83bf0bddea5d37117ff6fa39 5561 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5562 Date: Thu Apr 10 11:43:04 2014 +0900 5563 5564 glamor: Fix test for creating shared pixmaps 5565 5566 The pixmap usage hint is not a bitmask in general. The test for 5567 CREATE_PIXMAP_USAGE_SHARED was incorrectly triggering for a glamor internal 5568 usage hint being added in the xserver tree. 5569 5570 Tested-by: Ed Tomlinson <edtoml (a] gmail.com> 5571 5572 commit aecf1c4e5f4718adcfb85836830d065d3f4f97a5 5573 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5574 Date: Thu Apr 10 15:10:56 2014 +0900 5575 5576 dri2: Fix conflicting CreatePixmap usage flag definitions 5577 5578 RADEON_CREATE_PIXMAP_TILING_MICRO_SQUARE was the same as 5579 RADEON_CREATE_PIXMAP_DRI2. 5580 5581 Disambiguate the definitions and rearrange them to try and prevent this 5582 from happening again. 5583 5584 Tested-by: Ed Tomlinson <edtoml (a] gmail.com> 5585 5586 commit b50da3b96c212086cb58501dbe988d64f1f35b6d 5587 Author: Hans de Goede <hdegoede (a] redhat.com> 5588 Date: Fri Apr 11 09:44:37 2014 +0200 5589 5590 Fix building on older servers without xf86platformBus.h 5591 5592 Signed-off-by: Hans de Goede <hdegoede (a] redhat.com> 5593 5594 commit ed0cfbb4fe77146b0b38f777bc28f3a4ea6da07f 5595 Author: Hans de Goede <hdegoede (a] redhat.com> 5596 Date: Fri Mar 7 13:27:30 2014 +0100 5597 5598 Add support for server managed fds 5599 5600 Signed-off-by: Hans de Goede <hdegoede (a] redhat.com> 5601 5602 commit 3d7861fe112f25874319d4cdc12b745fbcd359cf 5603 Author: Hans de Goede <hdegoede (a] redhat.com> 5604 Date: Mon Mar 17 10:38:13 2014 +0100 5605 5606 Add radeon_get_drm_master_fd helper function 5607 5608 This is a preparation patch for adding server-managed-fd support without it 5609 turning into a goto fest. 5610 5611 Signed-off-by: Hans de Goede <hdegoede (a] redhat.com> 5612 5613 commit a63342ad15408071437c80b411d14196f3288aed 5614 Author: Hans de Goede <hdegoede (a] redhat.com> 5615 Date: Mon Mar 17 10:36:55 2014 +0100 5616 5617 radeon_open_drm_master get rid of unnecessary goto 5618 5619 Signed-off-by: Hans de Goede <hdegoede (a] redhat.com> 5620 5621 commit bdc412044f6ced056cd57320d1b2ee0d967c2191 5622 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5623 Date: Thu Mar 13 16:40:19 2014 +0900 5624 5625 Build against glamor in the xserver tree if available 5626 5627 commit 921a153f9964ca452e1241f76c7f7d653f42ceaf 5628 Author: Alex Deucher <alexander.deucher (a] amd.com> 5629 Date: Tue Mar 11 12:24:13 2014 -0400 5630 5631 update man page to reflect tiling changes for CI parts 5632 5633 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5634 5635 commit 5eee8a4d5c86bb1cc34d8caf2f2b64b53c241fa5 5636 Author: Alex Deucher <alexander.deucher (a] amd.com> 5637 Date: Mon Mar 10 16:20:09 2014 -0400 5638 5639 radeon: enable tiling by default on CIK 5640 5641 Now that mesa 10.1 is released, we can enable this by 5642 default for CIK parts. Tiling improves memory bandwidth 5643 utilization. 5644 5645 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5646 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 5647 5648 commit 691ec3d99c30111a4789830dfccb6eb5d3c40187 5649 Author: Alex Deucher <alexander.deucher (a] amd.com> 5650 Date: Mon Mar 10 16:17:34 2014 -0400 5651 5652 radeon: require libdrm 2.4.51 5653 5654 Required for proper tiling support on CIK parts. 5655 5656 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5657 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 5658 5659 commit 515bcf14d514f9dcaaf30fd0bf1ef6dd6ba9a0cd 5660 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5661 Date: Tue Mar 4 12:34:26 2014 +0900 5662 5663 Allow enabling glamor on R500 (and R300) class 3D engines as well. 5664 5665 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=75709 5666 5667 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5668 5669 commit ea6d0affe52d82becadab6fb1c87f9261b0605a2 5670 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5671 Date: Mon Feb 24 13:12:21 2014 +0900 5672 5673 Only log debugging output about initializing colormaps when we're doing so 5674 5675 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5676 5677 commit 8da17f30c70f4494ce22ad781a1cee17041812f3 5678 Author: Jrme Glisse <jglisse (a] redhat.com> 5679 Date: Mon Feb 24 14:50:25 2014 -0500 5680 5681 evergreen: fix shader constant upload on ppc 5682 5683 The number of dword we have to swap is (16*4) ie 16 vectors of 5684 4 floats each not 16 floats. Never hit this issue before because 5685 we never had more than 4 constant vector. 5686 5687 Signed-off-by: Jrme Glisse <jglisse (a] redhat.com> 5688 5689 commit cadb6b493942a84bfeb298751dce0dee39257a06 5690 Author: Alex Deucher <alexander.deucher (a] amd.com> 5691 Date: Fri Feb 21 08:33:21 2014 -0500 5692 5693 radeon: don't install colormap handling if there are no crtcs 5694 5695 Fixes a crash on cards with 0 crtcs. 5696 5697 Discussion: 5698 http://lists.freedesktop.org/archives/dri-devel/2014-February/054186.html 5699 5700 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5701 5702 commit 7cd972a85705341dd8306eefc558ed9e5def05d7 5703 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 5704 Date: Thu Feb 20 11:14:03 2014 +0100 5705 5706 return immediately in preinit when called with PROBE_DETECT 5707 5708 This fixes a crash with Xorg -configure. 5709 5710 Bug: 5711 https://bugs.launchpad.net/ubuntu/+source/xorg/+bug/1278046 5712 5713 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 5714 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5715 5716 commit 8de6f7b2f476d3baa9c7e2bb3544e4bafaad46b7 5717 Author: Alex Deucher <alexander.deucher (a] amd.com> 5718 Date: Fri Jan 24 11:04:30 2014 -0500 5719 5720 bump version post release 5721 5722 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5723 5724 commit 0333f5bda27dc0ec2edc180c7a4dc9a432f13f97 5725 Author: Alex Deucher <alexander.deucher (a] amd.com> 5726 Date: Fri Jan 24 10:19:49 2014 -0500 5727 5728 radeon: bump version for release 5729 5730 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5731 5732 commit fc4167f2a85d9cba65078d8fc6f08c7a619ad66e 5733 Author: Alex Deucher <alexander.deucher (a] amd.com> 5734 Date: Fri Jan 24 10:17:08 2014 -0500 5735 5736 Require glamor 0.6.0 5737 5738 This is required for Xv support and a number of important 5739 performance improvements. 5740 5741 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5742 5743 commit f2a0a5cf6c5a21e2a02280e110a4eb8e6609dace 5744 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5745 Date: Wed Jan 22 11:04:42 2014 +0900 5746 5747 Don't require the glamoregl module to be pre-loaded with xserver >= 1.15 5748 5749 The issues with loading it on demand have been fixed in xserver 1.15. 5750 5751 Inspired by Jrme Glisse on IRC. 5752 5753 commit 3213df16d61302148be0088c8f93c6a5a88558f1 5754 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5755 Date: Wed Jan 8 11:30:59 2014 +0900 5756 5757 dri2: Make last_vblank_seq local unsigned to match dpms_last_seq 5758 5759 Without this, I was occasionally running into gnome-shell hangs due to 5760 wildly off vblank sequence values. Doesn't seem to happen anymore with 5761 this change. 5762 5763 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5764 5765 commit bcc454ea2fb239e13942270faec7801270615b9c 5766 Author: Alex Deucher <alexander.deucher (a] amd.com> 5767 Date: Mon Jan 6 09:52:50 2014 -0500 5768 5769 radeon/exa: Always use a scratch surface for UTS to vram 5770 5771 If we don't, we may hit a buffer that crosses the 5772 visible vram boundary resulting in a sigbus when the 5773 CPU accesses the buffer beyond the PCI aperture. 5774 This will introduce an extra copy in certain cases. 5775 5776 This is based on Michel's patch from bug 44099 updated 5777 for all asic families. 5778 5779 Bug: 5780 https://bugs.freedesktop.org/show_bug.cgi?id=44099 5781 5782 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5783 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 5784 5785 commit 04ef035c9315b4a6fbf1b14720be87cee4099a9f 5786 Author: Alex Deucher <alexander.deucher (a] amd.com> 5787 Date: Tue Dec 24 15:14:35 2013 -0500 5788 5789 drm/radeon: fix SUMO2 pci id 5790 5791 0x9649 is sumo2, not sumo. 5792 5793 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5794 5795 commit bfbff3b246db509c820df17b8fcf5899882ffcfa 5796 Author: Robert Millan <rmh (a] freebsd.org> 5797 Date: Fri Dec 20 11:03:14 2013 +0000 5798 5799 radeon: Restore kernel module load on FreeBSD. 5800 5801 Since the introduction of a call to drmCheckModesettingSupported() 5802 in radeon_kernel_mode_enabled(), with abort condition if such call 5803 fails, the drmOpen() call in radeon_open_drm_master() no longer 5804 takes the responsibility of loading the radeon kernel module. 5805 5806 However at least on FreeBSD (and GNU/kFreeBSD), X is still relied 5807 on to load the modules it needs. This commit restores the old 5808 behaviour of loading kernel modules on these systems. 5809 5810 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=72852 5811 Signed-off-by: Robert Millan <rmh (a] freebsd.org> 5812 5813 commit 796c9a0cb587f528326bede11fa3f3eb7d3edaf1 5814 Author: Robert Millan <rmh (a] freebsd.org> 5815 Date: Fri Dec 20 10:55:09 2013 +0000 5816 5817 radeon: Set first parameter of drmOpen() to NULL 5818 5819 Since the introduction of a call to drmCheckModesettingSupported() 5820 in radeon_kernel_mode_enabled(), with abort condition if such call 5821 fails, the first argument to drmOpen() call in radeon_open_drm_master() 5822 has become a no-op red herring. 5823 5824 Such argument (a kernel module name) is supposed to result in load 5825 of specified kernel module. However, this will never happen. The 5826 problem is that if the code containing drmOpen() call is reached, it 5827 means that drmCheckModesettingSupported() check has previously 5828 succeeded, which implies the module is already loaded. 5829 5830 So, drmOpen() will never load a kernel module. But it gives the 5831 impression that it will. 5832 5833 In order to avoid this confusion, this commit replaces it with NULL, 5834 like xf86-video-intel driver does. 5835 5836 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=72849 5837 Signed-off-by: Robert Millan <rmh (a] freebsd.org> 5838 5839 commit d571d6af70ef27efd1ed6420eb892bdde963ed7a 5840 Author: Alex Deucher <alexander.deucher (a] amd.com> 5841 Date: Tue Sep 24 11:39:10 2013 -0400 5842 5843 radeon/kms: add Hawaii pci ids 5844 5845 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5846 5847 commit e38a92e00d015a6b80a1f3a16d58c61f084b066f 5848 Author: Alex Deucher <alexander.deucher (a] amd.com> 5849 Date: Tue Sep 24 11:35:00 2013 -0400 5850 5851 radeon: add support for Hawaii 5852 5853 Disabled by default until the acceleration code stablizes. 5854 5855 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5856 5857 commit e4cd0f4392ea11c93088ad429f36eaaf9bcbf505 5858 Author: Alex Deucher <alexander.deucher (a] amd.com> 5859 Date: Wed Oct 30 15:21:26 2013 -0400 5860 5861 radeon: enable tiling on SI by default (v2) 5862 5863 Now that mesa 9.2 is out with support for tiling 5864 on SI asics, we can enable it here. Tiling improves 5865 memory bandwidth utilization. 5866 5867 V2: update man page 5868 5869 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5870 5871 commit 3b38701a72fa1cad1e4610a2f4330b3da4cc6391 5872 Author: Vadim Girlin <vadimgirlin (a] gmail.com> 5873 Date: Fri Nov 1 10:36:39 2013 -0400 5874 5875 radeon: disable 2D tiling on buffers < 128 pixels 5876 5877 Seems to run into alignment problems with certain 5878 card configurations. 5879 5880 bug: 5881 https://bugs.freedesktop.org/show_bug.cgi?id=70675 5882 5883 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5884 5885 commit 0c921edf0162fed616cea9d02e168b719243bcd2 5886 Author: Jerome Glisse <jglisse (a] redhat.com> 5887 Date: Wed Oct 30 13:01:02 2013 -0400 5888 5889 radeon/glamor: with new pixmap for dri2 drawable no need to create new texture 5890 5891 When creating a new pixmap/bo for dri2 drawable there is no need to create a 5892 new texture, instead the texture associated with the new pixmap should be use 5893 otherwise there is a missmatch between the bo backing the texture for the 5894 drawable and the bo used by dri2 client. 5895 5896 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 5897 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 5898 5899 commit f1dc677e79cd7a88d7379a934ebc7d87a3b18805 5900 Author: Christopher James Halse Rogers <raof (a] ubuntu.com> 5901 Date: Mon Sep 23 12:25:29 2013 -0700 5902 5903 EXA/evergreen: Paranoia around linear tiling. (v2) 5904 5905 The last two bytes of tiling_mode contain the actual tiling mode; the rest are 5906 extra tiling configuration bits. These configuration bits are not necessarily 5907 zero for a linear buffer, so mask them out before checking for linearity 5908 5909 v2: Also fix up evergreen_textured_videofuncs.c 5910 5911 agd5f: remove trailing whitespace 5912 5913 Signed-off-by: Christopher James Halse Rogers <raof (a] ubuntu.com> 5914 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5915 5916 commit 67fb82a3f0759b171fea21b475a70fa825693570 5917 Author: Alex Deucher <alexander.deucher (a] amd.com> 5918 Date: Tue Oct 1 09:35:30 2013 -0400 5919 5920 radeon: fix the non-glamor build harder... 5921 5922 I need to stop pushing patches first thing in the morning. 5923 5924 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5925 5926 commit afc0374fdec3e24ece25805724459463e9a19f5e 5927 Author: Alex Deucher <alexander.deucher (a] amd.com> 5928 Date: Tue Oct 1 09:32:02 2013 -0400 5929 5930 drm/radeon: fix non-glamor build 5931 5932 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5933 5934 commit 2d791370dfc5570eb74d7a1fb3baf4d4c8ecf243 5935 Author: Dave Airlie <airlied (a] redhat.com> 5936 Date: Mon Sep 23 07:57:15 2013 +0100 5937 5938 radeon: use glamor Xv support if present. 5939 5940 This creates adaptors using glamor if possible. 5941 5942 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 5943 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5944 5945 commit cbb99f659ee7b18ded0008a606e41ded38c1a194 5946 Author: Alex Deucher <alexander.deucher (a] amd.com> 5947 Date: Mon Sep 30 08:55:16 2013 -0400 5948 5949 Revert "radeon: add glamor Xv support (v2)" 5950 5951 This causes problems if glamor doesn't have Xv support 5952 enabled. I just noticed that Dave has a better version, 5953 so use that instead. 5954 5955 This reverts commit 4fc1fa920584ace2c84d75af82d06962d0c84ec8. 5956 5957 commit 4fc1fa920584ace2c84d75af82d06962d0c84ec8 5958 Author: Alex Deucher <alexander.deucher (a] amd.com> 5959 Date: Thu Sep 26 11:02:55 2013 -0400 5960 5961 radeon: add glamor Xv support (v2) 5962 5963 v2: guard new glamor Xv bits with USE_GLAMOR 5964 5965 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5966 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 5967 5968 commit 41dfe327ac8740ac2cd84def96b5947224e422e7 5969 Author: Alex Deucher <alexander.deucher (a] amd.com> 5970 Date: Thu Sep 26 11:00:49 2013 -0400 5971 5972 radeon: fix limit handling for cards with >4G of ram 5973 5974 We can overflow the 32-bit limit. 5975 5976 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5977 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 5978 5979 commit c45e728107269c6f51599dad4f6a02ccfef703f1 5980 Author: Michel Dnzer <michel.daenzer (a] amd.com> 5981 Date: Wed Sep 18 10:57:52 2013 +0200 5982 5983 DRI2: Install client callback only once 5984 5985 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=60182 5986 Acked-by: Alex Deucher <alexander.deucher (a] amd.com> 5987 5988 commit b955ddd0d41801e4ca0c30a70a5d0b27c3f366c8 5989 Author: Alex Deucher <alexander.deucher (a] amd.com> 5990 Date: Mon Sep 16 10:58:22 2013 -0400 5991 5992 glamor: require 0.5.1 or newer 5993 5994 0.5.1 fixes a number of issues. 5995 5996 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 5997 5998 commit 282587cd0709850e7bacb1d8307065d95dc2c97d 5999 Author: Alex Deucher <alexander.deucher (a] amd.com> 6000 Date: Mon Sep 16 09:41:13 2013 -0400 6001 6002 radeon: enable glamor by default (v4) 6003 6004 It's required for newer asics, so enable it by 6005 default. 6006 6007 v2: update the autoconf help. Require --disable-glamor 6008 to build without glamor support. 6009 v3: default to yes rather than auto 6010 v4: fix help text 6011 6012 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6013 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 6014 6015 commit fdb7563a5cbc736b09c2864b67a93b475c98b2bd 6016 Author: Alex Deucher <alexander.deucher (a] amd.com> 6017 Date: Thu Jan 24 21:17:11 2013 -0500 6018 6019 radeon/kms: add berlin pci ids 6020 6021 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6022 6023 commit 8927d33f76ee12bc618fecfc59fc7ff1fcedcd5e 6024 Author: Mark Kettenis <kettenis (a] openbsd.org> 6025 Date: Thu Aug 22 10:32:46 2013 -0400 6026 6027 Fix shadowfb on big-endian machines 6028 6029 For shadowfb, the framebuffer needs to have the RADEON_TILING_SURFACE 6030 flag set, otherwise the appropriate byte swapping won't happen. 6031 6032 See https://bugs.freedesktop.org/show_bug.cgi?id=66663 6033 6034 Signed-off-by: Mark Kettenis <kettenis (a] openbsd.org> 6035 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6036 6037 commit fa83d3d1636c315bc43dd622d407eb058e3ef976 6038 Author: Alex Deucher <alexander.deucher (a] amd.com> 6039 Date: Tue Aug 20 09:29:23 2013 -0400 6040 6041 radeon: disallow glamor on pre-R600 asics 6042 6043 I'm not sure they can handle the shaders properly, especially 6044 only older parts like r300. 6045 6046 This will avoid display corruption problems reported by people 6047 using glamor on older asics by falling back to EXA if they try 6048 and enable glamor. 6049 6050 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6051 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 6052 6053 commit d0323622ee9b97a5f246baffbb2c65930a78ed14 6054 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6055 Date: Wed Aug 7 11:27:07 2013 +0200 6056 6057 bump version post release 6058 6059 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6060 6061 commit 9c97cca5c24409ca8447c99f051a12fd2d494e79 6062 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6063 Date: Wed Aug 7 10:48:17 2013 +0200 6064 6065 radeon: bump version for release 6066 6067 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6068 6069 commit 16270cfb202ab67dd152644ef019b2f1ee4d0341 6070 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6071 Date: Wed Aug 7 10:29:33 2013 +0200 6072 6073 add bicubic_table.py to EXTRA_DIST 6074 6075 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6076 6077 commit 2cb9197ca7a337c911f38b5de562a2364b922b86 6078 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6079 Date: Wed Aug 7 10:28:52 2013 +0200 6080 6081 kill unused radeon_driver.c 6082 6083 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6084 6085 commit c5cbfcf575b0b4aea6f797558ae974c1453c8e07 6086 Author: Alex Deucher <alexander.deucher (a] amd.com> 6087 Date: Tue Jul 30 10:08:25 2013 -0400 6088 6089 drmmode: add support for multi-screen reverse optimus 6090 6091 Initial reverse optimus didn't consider multiple screens, so 6092 this overhauls the code to use the new X server interface, 6093 and allows for multiple outputs on the dGPU to be used with 6094 the iGPU doing the rendering. Ported from Dave's nouveau 6095 patch. 6096 6097 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6098 6099 commit 429d5b797769895eb4f5fef816ce4e2f3a342031 6100 Author: Dave Airlie <airlied (a] redhat.com> 6101 Date: Tue Jan 8 15:56:37 2013 +1000 6102 6103 radeon: add support for reverse prime (v2) 6104 6105 This adds support for reverse prime configurations 6106 6107 v2: fix compilation with older xservers 6108 6109 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 6110 6111 commit 4de9356a2900ae0fb380a2350791ef045629cd05 6112 Author: Alex Deucher <alexander.deucher (a] amd.com> 6113 Date: Mon Aug 5 17:57:16 2013 -0400 6114 6115 radeon: fix naming clashes with multiple GPUs (v3) 6116 6117 The compat naming code for UMS causes problems 6118 with multiple GPU as you may end up with the same 6119 output name on multiple GPUs. Adjust the naming on 6120 secondary GPUs to avoid conflicts. 6121 6122 v2: integrate Dave's fixes for nouveau 6123 v3: keep compat with existing naming on primary GPU 6124 6125 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6126 6127 commit 2ae6bb18fefddb309920fa69c9b56c3a7f3db7b4 6128 Author: Grigori Goronzy <greg (a] chown.ath.cx> 6129 Date: Wed Jul 31 12:01:20 2013 +0200 6130 6131 EXA/evergreen/ni: replace magic number 6132 6133 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 6134 6135 commit 6a278369c05a298a4367306d986467a9ceacae8c 6136 Author: Raul Fernandes <rgfernandes (a] gmail.com> 6137 Date: Tue Jul 30 09:26:05 2013 -0400 6138 6139 EXA/6xx/7xx: optimize non-overlapping Copy 6140 6141 In case dst and src rectangles of a Copy operation in the same surface 6142 don't overlap, it is safe to skip the scratch surface. This is a 6143 common case. 6144 6145 Based on evergreen/ni patch from Grigori Goronzy. 6146 6147 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6148 6149 commit 4375a6e75e5d41139be7031a0dee58c057ecbd07 6150 Author: Grigori Goronzy <greg (a] chown.ath.cx> 6151 Date: Mon Jul 22 02:30:28 2013 +0200 6152 6153 EXA/evergreen/ni: accelerate PictOpOver with component alpha 6154 6155 Subpixel text rendering is typically done with a solid src and a 6156 pixmap mask. Traditionally, this cannot be accelerated in a single 6157 pass and requires two passes [1]. However, we can cheat a little 6158 with a constant blend color. 6159 6160 We can use: 6161 6162 const.A = src.A / src.A 6163 const.R = src.R / src.A 6164 const.G = src.G / src.A 6165 const.B = src.B / src.A 6166 6167 dst.A = const.A * (src.A * mask.A) + (1 - (src.A * mask.A)) * dst.A 6168 dst.R = const.R * (src.A * mask.R) + (1 - (src.A * mask.R)) * dst.R 6169 dst.G = const.G * (src.A * mask.G) + (1 - (src.A * mask.G)) * dst.G 6170 dst.B = const.B * (src.A * mask.B) + (1 - (src.A * mask.B)) * dst.B 6171 6172 This only needs a single source value. src.A is cancelled down in 6173 the right places. 6174 6175 [1] http://anholt.livejournal.com/32058.html 6176 6177 commit 94d0d14914a025525a0766669b556eaa6681def7 6178 Author: Grigori Goronzy <greg (a] chown.ath.cx> 6179 Date: Thu Jul 18 16:06:23 2013 +0200 6180 6181 EXA/evergreen/ni: fast solid pixmap support 6182 6183 Solid pixmaps are currently implemented with scratch pixmaps, which 6184 is slow. This replaces the hack with a proper implementation. The 6185 Composite shader can now either sample a src/mask or use a constant 6186 value. 6187 6188 commit 5bb04351c43a91a1d60348b7293544da05d75e72 6189 Author: Grigori Goronzy <greg (a] chown.ath.cx> 6190 Date: Fri Jul 27 17:31:53 2012 +0200 6191 6192 EXA/evergreen/ni: optimize non-overlapping Copy 6193 6194 In case dst and src rectangles of a Copy operation in the same surface 6195 don't overlap, it is safe to skip the scratch surface. This is a 6196 common case. 6197 6198 commit c08e09b7bec441c4bf93b4cae4de1260754bf940 6199 Author: Grigori Goronzy <greg (a] chown.ath.cx> 6200 Date: Sat May 18 13:46:03 2013 +0200 6201 6202 Fix RADEON_FALLBACK logging 6203 6204 commit c16c59f8f9b6aa7a4a6a6465582ad98f02a3606a 6205 Author: Mark Kettenis <kettenis (a] openbsd.org> 6206 Date: Sun Jul 7 13:44:13 2013 +0200 6207 6208 Always go through DFS/UTS when byte swapping is needed. 6209 6210 Before commit ef9bfb262db7004bef3704e5d914687e50d3fca4 and 6211 e5bd99faa3b6629a55168386d5dfa936ee4e97ae, byte swapping for the front buffer 6212 used to be done in hardware. Now that this no longer happens we need to let 6213 DFS/UTS ihandle the byte swapping. 6214 6215 See https://bugs.freedesktop.org/show_bug.cgi?id=66663 . 6216 6217 Signed-off-by: Mark Kettenis <kettenis (a] openbsd.org> 6218 6219 commit 1239dbbd8c8d9b55756c1de52cad353171a06522 6220 Author: Alex Deucher <alexander.deucher (a] amd.com> 6221 Date: Sun Jul 7 20:22:10 2013 -0400 6222 6223 radeon: bump libdrm requirement to 2.4.46 for CIK support 6224 6225 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6226 6227 commit 869e0b187b2d07e637c4482fe50d9f1b0c09f4e4 6228 Author: Alex Deucher <alexander.deucher (a] amd.com> 6229 Date: Thu Jan 24 21:12:28 2013 -0500 6230 6231 radeon/kms: add kabini pci ids 6232 6233 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6234 6235 commit 0edcefc4c97e4572431c38fe170032470e22e0b7 6236 Author: Alex Deucher <alexander.deucher (a] amd.com> 6237 Date: Fri Jun 7 14:56:26 2013 -0400 6238 6239 radeon/kms: add bonaire pci ids 6240 6241 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6242 6243 commit ebc32b27af25b23604e725eb50d844a8d26116bb 6244 Author: Alex Deucher <alexander.deucher (a] amd.com> 6245 Date: Fri Aug 31 18:17:24 2012 -0400 6246 6247 radeon: update cursor handling for CIK 6248 6249 CIK asics have 128x128 hw cursors 6250 6251 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6252 6253 commit 4adaea996454b1ad5185f9c0f37667dbfc266495 6254 Author: Alex Deucher <alexander.deucher (a] amd.com> 6255 Date: Fri Jun 7 14:47:36 2013 -0400 6256 6257 radeon: add family ids for CIK 6258 6259 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6260 6261 commit 3626ab147b6749a355e7d21710447f13b9796c77 6262 Author: Leo Liu <leo.liu (a] amd.com> 6263 Date: Fri Jun 14 10:27:20 2013 -0400 6264 6265 radeon: fix a memory leak in get_modes() callback. 6266 6267 leak happens when looping xrandr prop. 6268 6269 Signed-off-by: Leo Liu <leo.liu (a] amd.com> 6270 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 6271 6272 commit c0f7d03cb29166c6c00dd6fb698ba3f70c59dbdf 6273 Author: Alex Deucher <alexdeucher (a] gmail.com> 6274 Date: Mon Jun 17 11:38:27 2013 -0400 6275 6276 man: fix up previous commit 6277 6278 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 6279 6280 commit 03b5edc9dd3859eed2504155970613aaf88df852 6281 Author: Eric S. Raymond <esr (a] thyrsus.com> 6282 Date: Fri Jun 7 13:23:14 2013 -0400 6283 6284 Fix some incomplete list entries. 6285 6286 This necessarily involved adding some content, which I was able to 6287 Google for. Please review, correct, and make more specific. 6288 6289 Bodiless .TP entries screw up doclifter's page parsing. Please 6290 don't do that anymore. 6291 6292 commit 7ce43639a8e53eefb94e5255a39c1232b6bdbf66 6293 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6294 Date: Thu Jun 6 09:58:27 2013 +0200 6295 6296 autogen.sh: Restore passing --enable-maintainer-mode to configure 6297 6298 Looks like this was accidentally dropped when the GNOME Build API was 6299 implemented. 6300 6301 commit 365e571d9de0b5979425c34210789afca4ea5f68 6302 Author: Samuel Li <samuel.li (a] amd.com> 6303 Date: Wed Mar 20 11:59:58 2013 -0400 6304 6305 radeon: use direct mapping for fast fb access. 6306 6307 Signed-off-by: Samuel Li <samuel.li (a] amd.com> 6308 6309 commit bd2557ea5ef84b975060e929d5ece53ec464336f 6310 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6311 Date: Wed May 8 22:39:48 2013 -0400 6312 6313 DRI2: add interpolated blanks to frame number in event handlers 6314 6315 The 'frame' argument passed to event handlers is the plain 6316 CRTC vblank counter that stops progressing when the 6317 associated display is in DPMS-off mode. If we have a 6318 DPMS-off period the frame counter and MSC will 6319 diverge, which can cause some higher-level functions 6320 to return incorrect values. 6321 6322 This patch fixes the problem by adding interpolated 6323 vblanks to the frame counter before using it in handler 6324 functions. 6325 6326 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6327 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6328 6329 commit 385a92b4fbe8f53b359ef6c463704414d00476fa 6330 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6331 Date: Wed May 8 22:39:47 2013 -0400 6332 6333 DRI2: support scheduling emulated events with zero delay 6334 6335 Now that we fully emulating a running CRTC through DPMS-off state 6336 it is possible to come up with a zero delay when scheduling 6337 a swap or MSC-wait (e.g., if a call into respective wait function 6338 was entered very late). This patch wraps the TimerSet function 6339 into our own radeon_dri2_schedule_event such that the latter 6340 calls the event right away if zero delay is specified. 6341 6342 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6343 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6344 6345 commit e87b52e6ad41ad7a87e43b818d80e7d522d9c68d 6346 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6347 Date: Wed May 8 22:39:46 2013 -0400 6348 6349 DRI2: make wait_msc and get_msc work on disabled CRTCs 6350 6351 Now that the running CRTC is emulated through DPMS-off 6352 states, wait_msc and get_msc must also work on disabled 6353 CRTCs. When CRTC is disabled, we must extrapolate the 6354 MSC count from present time and last MSC when CRTC was 6355 running and also use timers facility in wait_msc. 6356 6357 v2: CRTC-private now stores frame rate instead of nominal 6358 vblank period. 6359 6360 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6361 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6362 6363 commit f1584152bbeda98cc1907359f1c7954a84c8837d 6364 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6365 Date: Wed May 8 22:39:45 2013 -0400 6366 6367 DRI2: move rename and rework radeon_dri2_deferred_swap 6368 6369 radeon_dri2_deferred_swap will be used to generate 6370 real events (not just fallbacks) so now it needs to 6371 generate real timestamp and frame counter. Also 6372 this function will be used both by schedule_swap 6373 and wait_msc, so give it a more generic name: 6374 radeon_dri2_deferred_event 6375 6376 v2: - Extrapolate the frame number from the time 6377 of actual execution of the function instead 6378 of using the MSC calculated (extrapolated) 6379 at event scheduling time. 6380 - CRTC-private now stores frame rate instead of 6381 nominal vblank period. 6382 6383 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6384 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6385 6386 commit 84bce7943b735aee19e26be4d47fdf140564f3f5 6387 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6388 Date: Wed May 8 22:39:44 2013 -0400 6389 6390 DRI2: track CRTC in event record 6391 6392 Tracking the CRTC associated with an event will save us 6393 some lookups later in event handlers. 6394 6395 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6396 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6397 6398 commit e40d5390b3efdea3e02267413350410d8e6a2970 6399 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6400 Date: Wed May 8 22:39:43 2013 -0400 6401 6402 DRI2: hook up vblank extrapolation to schedule_swap 6403 6404 This patch hooks up swap-scheduling function with 6405 vblank-extrapolation function. Rather than waiting for 6406 fixed time, we calculate exactly how much we should wait 6407 and what we should update target_msc to using 6408 radeon_dri2_extrapolate_msc_delay helper function 6409 and schedule the swap completion using DIX's timer facility. 6410 6411 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6412 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6413 6414 commit ffaa5abf207415159cdb28e90da49b95f497ef61 6415 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6416 Date: Wed May 8 22:39:42 2013 -0400 6417 6418 DRI2: add vblank extrapolation function 6419 6420 Implement a helper function that will be called when emulating 6421 the running CRTC. The function should be called only when CRTC 6422 is in DPMS-off state. It will look at the vblank count and the 6423 time that was recorded last time the CRTC was running and 6424 calculate how long one must wait (from present time) until 6425 the target_msc is reached if the CRTC were running. 6426 6427 v2: - CRTC-private now stores frame rate instead of nominal 6428 vblank period. 6429 - DIX's timer facility can sometimes wake up the scheduled 6430 functions more than a millisecond earlier. To avoid 6431 generating an old MSC, we have to add more margin when 6432 converting the delay in microseconds to milliseconds. 6433 6434 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6435 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6436 6437 commit 80ae2291d082b57c70d27a80182f00d760fb3d3a 6438 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6439 Date: Wed May 8 22:39:41 2013 -0400 6440 6441 DRI2: adjust MSC by the number of interpolated vblanks 6442 6443 Normally, MSC value equals the vblank count of the CRTC 6444 on which the drawable resides. However, we are now interpolating 6445 vblanks through DPMS-off state, so if the CRTC is turned 6446 off at some point, the vlbank count does not change, but 6447 MSC does change by the amount that equals the cumulative 6448 number of interpolated vblanks. 6449 6450 So each time we interact with the CRTC's vblank facility, 6451 we have to adjust the vblank counter: in the request, we 6452 have to subtract the number of interpolated vblanks from 6453 MSC and in the reply we have to add the number of interpolated 6454 vblanks to the MSC. 6455 6456 This patch only makes the MSC correct when CRTC is on. 6457 The subsequent patch will construct the MSC when CRTC is not 6458 running. 6459 6460 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6461 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6462 6463 commit 5fd2eb5d12cea32927a9e6c6ce4afd18aa7d046a 6464 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6465 Date: Wed May 8 22:39:40 2013 -0400 6466 6467 drmmode: calculate interpolated vblanks while in dpms-off state 6468 6469 This adds provisions for interpolating vblanks while the CRTC 6470 is in DPMS-off state. When entering DPMS-off state, we 6471 record the last vblank time, sequence number and frame rate 6472 in CRTC-private structure. 6473 6474 When going back to DPMS-on state we read the current 6475 time and calculate how long we have been off. Then we derive 6476 how many vblanks that would have been had the CRTC remained 6477 running. These are the interpolated vblanks. 6478 6479 Finally, we accumulate the number of interpolated vblanks 6480 in CRTC-private structure to get the number of interpolated 6481 vblanks over the system lifetime. 6482 6483 v2: Track frame rate instead of vblank period. The former 6484 eliminates some roundoff errors. 6485 6486 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6487 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6488 6489 commit e41ad30d09be4962cfb1942b0b9f63875dbb2d2a 6490 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6491 Date: Wed May 8 22:39:39 2013 -0400 6492 6493 drmmode: remove dead code from drmmode_crtc_dpms 6494 6495 We are about to modify this function, so let's clean it 6496 up first. 6497 6498 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6499 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6500 6501 commit 373671d2eed30e30b3bdee7e40426cf58c127234 6502 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6503 Date: Wed May 8 22:39:38 2013 -0400 6504 6505 drmmode: add drmmode_get_current_ust function 6506 6507 The new helper function retrieves current time in the format 6508 that is compatible with vblank timestamps. 6509 6510 v2: - fix an incorrect statement in a comment 6511 - add a #define so that don't depend on libdrm patches that 6512 have not yet been accepted upstream 6513 6514 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6515 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6516 6517 commit 006fbbd1d38a089b50ab3197d32815689ed249fa 6518 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6519 Date: Wed May 8 22:39:37 2013 -0400 6520 6521 DRI2: make populate_vbl_request_type external 6522 6523 We'll need to use this function in other .c files so 6524 remove static annotation from it. Also, add radeon_ 6525 prefix because this function is now part of global namespace, 6526 so it would be nice to know where it is coming from. 6527 6528 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6529 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6530 6531 commit 34660d8757ff16acd1686ff2872f4600d92a68bf 6532 Author: Ilija Hadzic <ilijahadzic (a] gmail.com> 6533 Date: Wed May 8 22:39:36 2013 -0400 6534 6535 DRI2: change signature of populate_vbl_request_type 6536 6537 This function no longer uses info argument, so nuke it. 6538 6539 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6540 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6541 6542 commit ab6e85d1f5dd4657591b314b85fd2453f99c3712 6543 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6544 Date: Fri Mar 22 10:51:05 2013 +0100 6545 6546 glamor: Keep old private in radeon_pixmap_set_bo 6547 6548 commit 0a84f2760eccf07781ae6c3f55d2d246c155c9ce 6549 Author: Alex Deucher <alexander.deucher (a] amd.com> 6550 Date: Mon May 13 17:06:00 2013 -0400 6551 6552 radeon: add Hainan pci ids 6553 6554 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6555 6556 commit 6cee1eaebfcef2b2d2c51ec30c31260fd98b564e 6557 Author: Alex Deucher <alexander.deucher (a] amd.com> 6558 Date: Mon May 13 17:01:30 2013 -0400 6559 6560 radeon: add hainan chip family 6561 6562 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6563 6564 commit 8a9e9b21cbb44c23d37bf18c983e2c3f0fe33e87 6565 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6566 Date: Thu Apr 25 17:10:12 2013 +0200 6567 6568 glamor: Fix build against xserver < 1.13 6569 6570 Those didn't define CREATE_PIXMAP_USAGE_SHARED. 6571 6572 commit 6216b5aa3793abe71b864170bc9d4d745f027328 6573 Author: Alex Deucher <alexander.deucher (a] amd.com> 6574 Date: Thu Apr 25 14:19:37 2013 -0400 6575 6576 radeon: add new richland pci ids 6577 6578 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6579 6580 commit fd5903f15b36555bf1ccc1c3865c9fffd499128a 6581 Author: Alex Deucher <alexander.deucher (a] amd.com> 6582 Date: Thu Apr 25 14:17:51 2013 -0400 6583 6584 radeon: add new SI pci ids 6585 6586 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6587 6588 commit b2826580e4d246d45871b413e4deecb070162577 6589 Author: Jerome Glisse <jglisse (a] redhat.com> 6590 Date: Mon Apr 8 13:45:18 2013 -0400 6591 6592 radeonsi: add 2d tiling support 6593 6594 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 6595 6596 commit 6e74aacc5e5da3b51744153dad1645caa6ea4ce3 6597 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6598 Date: Wed Mar 27 11:57:29 2013 +0100 6599 6600 glamor: Only enable for depth >= 24 6601 6602 It can't handle lower depths yet. 6603 6604 Reported-by: Igor Vagulin <igor.vagulin (a] gmail.com> 6605 6606 commit 33d8408eec806355c2e55726679ec50ef3b769f1 6607 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6608 Date: Tue Mar 12 19:03:06 2013 +0100 6609 6610 glamor: Enable by default on SI 6611 6612 We should now handle its major initialization failure cases gracefully. 6613 6614 Reviewed-by: Jerome Glisse <jglisse (a] redhat.com> 6615 6616 commit ad1d8874fab30257ba8ec8bee8ffa42a8fa9b5c0 6617 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6618 Date: Fri Mar 15 18:28:27 2013 +0100 6619 6620 glamor: Initial PRIME pixmap sharing hooks. 6621 6622 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57200 6623 6624 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6625 Tested-by: Christoph Haag <haagch.christoph (a] googlemail.com> 6626 Tested-by: linedot <linedot (a] xcpp.org> 6627 6628 commit 35b384a00a0c76ea556f55787fccc95ecd51bc0c 6629 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6630 Date: Thu Mar 21 15:34:21 2013 +0100 6631 6632 drmmode: handle failure of drmModeGetConnector 6633 6634 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6635 6636 commit 751c7259cf23a29fdd2653881e2db63ce3cf1ca5 6637 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6638 Date: Thu Mar 21 10:40:42 2013 +0100 6639 6640 Move fd cleanup to correct place. 6641 6642 Oops, just after previous commit I realized that it 6643 needs to be done from RADEONFreeRec, because when preinit fails 6644 info pointer can be null in FreeScreen. 6645 6646 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6647 6648 commit 1643b8cd2dc53ed36916e11003590c7037b4ddd6 6649 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6650 Date: Thu Mar 21 10:33:19 2013 +0100 6651 6652 radeon: add refcounts to fix up zaphod open/close. 6653 6654 Oops, turns out my previous commits were buggy. 6655 Adding proper refcounts will handle this correctly. 6656 6657 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6658 6659 commit a4d1bc43d4d10e2ca1e512c1010e962c587d6c16 6660 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6661 Date: Wed Mar 20 17:13:44 2013 +0100 6662 6663 Cleanup drmmode in CloseScreen. 6664 6665 Prevents a use-after-free. 6666 6667 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6668 6669 commit 3cb3ab8704813d7c14b9e547a30e37598d70c85a 6670 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6671 Date: Wed Mar 20 17:12:29 2013 +0100 6672 6673 Close drm_fd in CloseScreen. 6674 6675 This prevents a small fd leak. 6676 6677 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6678 6679 commit 462d8d3b86038438ab854072e546cc7a9efacdd7 6680 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6681 Date: Tue Mar 12 18:59:19 2013 +0100 6682 6683 glamor: Bail if the glamoregl module wasn't loaded early 6684 6685 Trying to initialize glamor in that case crashes. 6686 6687 Reviewed-by: Jerome Glisse <jglisse (a] redhat.com> 6688 6689 commit 3e5350bedf50a1a140e0eba08c39ed2451fb9115 6690 Author: Alex Deucher <alexander.deucher (a] amd.com> 6691 Date: Fri Mar 8 14:07:42 2013 -0500 6692 6693 radeon: add Richland pci ids 6694 6695 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6696 6697 commit bdbf766d8170f6ea6c16b9780ef5a4007730588b 6698 Author: Alex Deucher <alexander.deucher (a] amd.com> 6699 Date: Fri Mar 8 14:00:09 2013 -0500 6700 6701 radeon: mark trinity APUs with IGP flag 6702 6703 Was missing on trinity. 6704 6705 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6706 6707 commit 5ac335eaae135c0e66cc4643908f41174e2f1123 6708 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6709 Date: Fri Mar 8 17:37:41 2013 +0100 6710 6711 Only parse Option "AccelMethod" if glamor was enabled at build time 6712 6713 Will hopefully make it more obvious when it wasn't. 6714 6715 commit f1059e3f0934ccf1ae183099dbd4de700a7c81ac 6716 Author: Colin Walters <walters (a] verbum.org> 6717 Date: Wed Jan 4 17:37:06 2012 -0500 6718 6719 autogen.sh: Implement GNOME Build API 6720 6721 http://people.gnome.org/~walters/docs/build-api.txt 6722 6723 Signed-off-by: Adam Jackson <ajax (a] redhat.com> 6724 6725 commit 343b01c9bf35b125cd0c3df8db7c01a5fb227bda 6726 Author: Alex Deucher <alexander.deucher (a] amd.com> 6727 Date: Thu Jan 24 19:57:23 2013 -0500 6728 6729 radeon/kms: add oland pci ids 6730 6731 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6732 6733 commit 9980f1e0e0a5df5fb0c791687e5c98c13089d7d1 6734 Author: Alex Deucher <alexander.deucher (a] amd.com> 6735 Date: Thu Jan 24 19:52:04 2013 -0500 6736 6737 radeon: add oland chip family 6738 6739 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 6740 6741 commit d89e545826969910226cb96e73b64f18e92d2064 6742 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6743 Date: Wed Jan 30 17:21:05 2013 +0100 6744 6745 bump version post release 6746 6747 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6748 6749 commit 21a4f3eaca98c1c54bffcdf3926c631cec9c409d 6750 Author: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6751 Date: Wed Jan 30 16:55:27 2013 +0100 6752 6753 radeon: bump version for release 6754 6755 And fixup distcheck.. 6756 6757 Signed-off-by: Maarten Lankhorst <maarten.lankhorst (a] canonical.com> 6758 6759 commit 102ee4a24e4c9ba0c0b65d3ca8599dc7e0216c1e 6760 Author: Alex Deucher <alexdeucher (a] gmail.com> 6761 Date: Fri Jan 11 09:52:32 2013 -0500 6762 6763 radeon/r200: program RE_WIDTH_HEIGHT properly 6764 6765 Values are inclusive. Mesa already sets these 6766 correctly. Also bump EXA/Xv limits from 2047 to 6767 2048. 6768 6769 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 6770 6771 commit 47689ef8b878a56ea094e21a6090b228e67d605e 6772 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6773 Date: Fri Jan 11 15:41:48 2013 +0100 6774 6775 Make radeon_glamor_create_textured_pixmap() return TRUE without USE_GLAMOR. 6776 6777 Otherwise rotation won't work when glamor support isn't built in. 6778 6779 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6780 6781 commit 1a8c05fd5559be842d8a038df73de34d4b261c15 6782 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6783 Date: Fri Jan 11 13:22:24 2013 +0100 6784 6785 Make CRTC shadow pixmaps usable by glamor. 6786 6787 Fixes rotation with glamor. 6788 6789 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6790 6791 commit e5e22d788b103c04d3460a221a3b9dc3be69e0a4 6792 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6793 Date: Thu Jan 10 17:00:16 2013 +0100 6794 6795 DRI2/r[345]xx: 16 bit depth buffers use square micro tiling. 6796 6797 Fixes piglit sanity.tests (with 3D drivers which allocate depth buffers from 6798 the X server). 6799 6800 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6801 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6802 6803 commit f5d2682584c74587890a9795ece7a8917214a83f 6804 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6805 Date: Thu Jan 10 17:00:15 2013 +0100 6806 6807 DRI2: Properly use buffer depth passed in by client. 6808 6809 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6810 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6811 6812 commit 4e35b2f530e2ca8c7b7220cacd05c661de43d20d 6813 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6814 Date: Thu Jan 10 12:10:52 2013 +0100 6815 6816 Drop support for X servers older than 1.7. 6817 6818 The main purpose is to drop DRI2 compatibility code paths which are getting 6819 awkward to deal with and at this point are probably only tested lightly if at 6820 all. xserver 1.7 was released more than 3 years ago, and 1.6.2 was already 6821 required before. 6822 6823 Mostly mechanical, there's probably potential for more cleanups. 6824 6825 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 6826 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6827 6828 commit 857d729eb3672443f89c0516f8226eacb37dad2b 6829 Author: Dave Airlie <airlied (a] redhat.com> 6830 Date: Mon Jan 7 14:13:33 2013 +1000 6831 6832 radeon: fix damage reporting for slave pixmaps 6833 6834 This is definitely more correct, not sure if it'll fix any bugs. 6835 6836 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 6837 6838 commit 6981a5c087165b126c15ba0025cffdba218ab652 6839 Author: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6840 Date: Wed Dec 19 10:35:43 2012 -0500 6841 6842 DRI2: limit the swap rate when CRTC is in DPMS-off state 6843 6844 If drawable is displayed on a CRTC and relevant CRTC is in 6845 DPMS off state, defer the swap by a fixed (hard-coded) time. 6846 6847 This patch fixes a bug that caused an application to render 6848 at uncontrolled rate when CRTC goes into DPMS "off" state, 6849 thus thrashing the GPU and CPU and likely offsetting the 6850 power savings achieved by shutting off the display. 6851 6852 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6853 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 6854 6855 commit 3657672207322be651cdb94a811337b7c5668c84 6856 Author: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6857 Date: Wed Dec 19 10:35:42 2012 -0500 6858 6859 DRI2: change signature of radeon_dri2_drawable_crtc 6860 6861 Return pointer to the CRTC instead of CRTC ID and expose 6862 consider_disabled option in arguments of this function. 6863 6864 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6865 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 6866 6867 commit 61d0aec40e2521488c2fe43e7a6823e5c87d94d7 6868 Author: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6869 Date: Wed Dec 19 10:35:41 2012 -0500 6870 6871 video: add option to include disabled CRTCs in best CRTC search 6872 6873 This patch adds an option called consider_disabled to 6874 radeon_pick_best_crtc function. If this option is set 6875 and searching for best-fit CRTC yields nothing, the search 6876 will be widened to include CRTCs in DPMS "off" state. 6877 The new option is not used yet; it will be in the patches 6878 to follow. 6879 6880 v2: Do not look at the CRTC mode, we only care about its box. 6881 6882 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 6883 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 6884 6885 commit a60d2152e928a7011fc7c44a885a34c3cdd4f0fe 6886 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6887 Date: Thu Dec 27 12:13:57 2012 +0100 6888 6889 glamor: Enable full 2D acceleration for SI as well. 6890 6891 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6892 6893 commit 04ebe14fe775d18b534e89b50b90ee293ae1910e 6894 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6895 Date: Thu Dec 27 11:54:27 2012 +0100 6896 6897 glamor: Release the drawable after passing to glamor_push_pixels 6898 6899 Ported from xf86-video-intel commit ca5c028c2b4d9bf02002acd484054fe427ea8d09 . 6900 6901 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6902 6903 commit cee7d61b144162974238b282d76a70aa2f14cc0d 6904 Author: Alex Deucher <alexdeucher (a] gmail.com> 6905 Date: Tue Dec 18 15:49:47 2012 -0500 6906 6907 exa/Xv: fix coordinate limits on AVIVO IGPs 6908 6909 RS6xx asics are r4xx derived, but seem to have r3xx 6910 limitations as far as clipping is concerned. Spotted 6911 by Michel on IRC. 6912 6913 Fixes: 6914 https://bugs.freedesktop.org/show_bug.cgi?id=58469 6915 6916 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 6917 6918 commit 3c7d024c2671dac541b8f2daed55040f4fd5d62d 6919 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6920 Date: Wed Dec 12 10:43:55 2012 +0100 6921 6922 Remove dead option definitions. 6923 6924 These were only used with UMS. 6925 6926 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6927 6928 commit bdbd84a6c99943f4e012c92ad9cb920cbd676621 6929 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6930 Date: Mon Dec 10 23:35:41 2012 +0100 6931 6932 Remove local variables total_size_bytes. 6933 6934 Not actually used for anything. Refactoring gone wrong? 6935 6936 Also remove a comment that doesn't seem to make sense anymore. 6937 6938 Stumbled over this due to a warning by clang: 6939 6940 ../../src/radeon_kms.c:1562:6: warning: variable 'total_size_bytes' is uninitialized when used here [-Wuninitialized] 6941 total_size_bytes += (64 * 4 * 64); 6942 ^~~~~~~~~~~~~~~~ 6943 ../../src/radeon_kms.c:1557:25: note: initialize the variable 'total_size_bytes' to silence this warning 6944 int total_size_bytes; 6945 ^ 6946 = 0 6947 6948 Reviewed-by: Alex Deucher <aleander.deucher (a] amd.com> 6949 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6950 6951 commit cf0a81547bcde32bdd2b080c0a546a49dd2eb628 6952 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6953 Date: Tue Dec 11 11:48:33 2012 +0100 6954 6955 Remove unused local variable 'path'. 6956 6957 ../../src/radeon_probe.c: In function 'radeon_platform_probe': 6958 ../../src/radeon_probe.c:270:11: error: unused variable 'path' [-Werror=unused-variable] 6959 6960 Reviewed-by: Alex Deucher <aleander.deucher (a] amd.com> 6961 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6962 6963 commit d268038a5bdd1655f7533c8dd2edf74fbc82bbee 6964 Author: Michel Dnzer <michel.daenzer (a] amd.com> 6965 Date: Mon Jul 2 09:07:14 2012 +0200 6966 6967 Remove dead code flagged by gcc -Wunused-but-set-variable. 6968 6969 Reviewed-by: Alex Deucher <aleander.deucher (a] amd.com> 6970 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 6971 6972 commit 793e1b0ea579cce7489aee57a95f20384d6ecce8 6973 Author: Alex Deucher <alexdeucher (a] gmail.com> 6974 Date: Thu Dec 6 09:21:19 2012 -0500 6975 6976 radeon: fix ring count in R300PrepareTexturedVideo 6977 6978 Fixes: 6979 https://bugs.freedesktop.org/show_bug.cgi?id=31364 6980 6981 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 6982 6983 commit f402805b22e4f4f735924d59c69474b24e623a25 6984 Author: Jerome Glisse <jglisse (a] redhat.com> 6985 Date: Fri Nov 30 10:07:51 2012 -0500 6986 6987 radeon: avoid copying over itself ddx fb 6988 6989 In some rare case it seems that the old fb could already be the 6990 ddx fb. As copying from the same bo will trigger a ttm deadlck 6991 detection which will result in infinite kernel/userspace loop. 6992 Avoid doing any copy as anyway it's useless. 6993 6994 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 6995 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 6996 6997 commit 60cd6ceaf44b506433aebf6b3a639a17604dfddd 6998 Author: Alex Deucher <alexdeucher (a] gmail.com> 6999 Date: Wed Nov 21 18:42:56 2012 -0500 7000 7001 radeon: add new SI pci id 7002 7003 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7004 7005 commit 53af6aa99dfb67b5223c28d5129a79694165048d 7006 Author: Jerome Glisse <jglisse (a] redhat.com> 7007 Date: Wed Nov 14 12:15:47 2012 -0500 7008 7009 radeon: avoid segfault when pixmap exceed GPU capabilities 7010 7011 We might get a request for a pixmap/drawable that is too big 7012 for GPU capabilities in the dri2 get buffer path. In face of 7013 such things just return NULL to dri2 get buffer request. The 7014 GL driver should then use something like transparent black to 7015 avoid something too ugly on the screen. 7016 7017 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 7018 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 7019 7020 commit 71eeb0833489ece745f1e5dc41379ec63161ee81 7021 Author: Alex Deucher <alexdeucher (a] gmail.com> 7022 Date: Tue Nov 6 09:43:55 2012 -0500 7023 7024 bump verion post release 7025 7026 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7027 7028 commit cf377c991f4c1cd4eee1a8530ba9fdf6f9f63301 7029 Author: Alex Deucher <alexdeucher (a] gmail.com> 7030 Date: Tue Nov 6 09:12:39 2012 -0500 7031 7032 radeon: bump version for release 7033 7034 7.0.0, RIP UMS 7035 7036 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7037 7038 commit 20bfc652ce40008ea561db2984bccf137409c7fd 7039 Author: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 7040 Date: Mon Sep 24 17:19:09 2012 -0400 7041 7042 radeon/radeon_platform_probe: fix Zaphod mode breakage 7043 7044 Using radeon_platform_probe function breaks the Zaphod mode because 7045 it attempts to call xf86AddEntityToScreen multiple times, but nobody 7046 calls xf86SetEntityShared prior to that. Consequently, calls for all 7047 but first device instance fail. 7048 7049 Prior to introduction of platform bus, the logic was that the Probe 7050 function would make the entity sharable, which would cause Xserver 7051 to later make it shared prior to adding it to screen. With the 7052 platform bus loading, add to screen happens in the probe 7053 function so we have to make it shared there. 7054 7055 v2: do not make the entity shared if it was not previously marked 7056 sharable. Should fix: 7057 7058 https://bugs.freedesktop.org/show_bug.cgi?id=56680 7059 https://bugs.freedesktop.org/show_bug.cgi?id=56663 7060 7061 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 7062 7063 commit b2807b53083af5db0b85645666b0127be16623c7 7064 Author: Alex Deucher <alexdeucher (a] gmail.com> 7065 Date: Fri Nov 2 17:57:18 2012 -0400 7066 7067 man: note that 2D tiling requires Mesa 9.0 or newer 7068 7069 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7070 7071 commit 1038e29162d869e2ce37f9b8f5f7c8ea37c3e53d 7072 Author: Alex Deucher <alexdeucher (a] gmail.com> 7073 Date: Fri Nov 2 10:04:46 2012 -0400 7074 7075 Revert "radeon/radeon_platform_probe: fix Zaphod mode breakage" 7076 7077 This reverts commit 39eac3104c2f08b4d78aab3f88fd104301eb4711. 7078 7079 This breaks non-zaphod configurations with xserver 1.13. 7080 See: 7081 https://bugs.freedesktop.org/show_bug.cgi?id=56680 7082 https://bugs.freedesktop.org/show_bug.cgi?id=56663 7083 7084 commit d748894c84e891662cde6d7a5748f74a5aa5175e 7085 Author: Alex Deucher <alexdeucher (a] gmail.com> 7086 Date: Fri Nov 2 08:50:16 2012 -0400 7087 7088 man: update man pages for trinity (ARUBA) 7089 7090 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7091 7092 commit 6740e8e90b934611c3bb0f9e8732d66e25b8c727 7093 Author: Andreas Boll <andreas.boll.dev (a] gmail.com> 7094 Date: Fri Nov 2 13:30:44 2012 +0100 7095 7096 radeon/man: ColorTiling2D is by default on for r6xx-cayman 7097 7098 Signed-off-by: Andreas Boll <andreas.boll.dev (a] gmail.com> 7099 7100 commit 39eac3104c2f08b4d78aab3f88fd104301eb4711 7101 Author: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 7102 Date: Mon Sep 24 17:41:32 2012 -0400 7103 7104 radeon/radeon_platform_probe: fix Zaphod mode breakage 7105 7106 Using radeon_platform_probe function breaks the Zaphod mode because 7107 it attempts to call xf86AddEntityToScreen multiple times, but nobody 7108 calls xf86SetEntityShared prior to that. Consequently, calls for all 7109 but first device instance fail. 7110 7111 Prior to introduction of platform bus, the logic was that the Probe 7112 function would make the entity sharable, which would cause Xserver 7113 to later make it shared prior to adding it to screen. With the 7114 platform bus loading, add to screen happens in the probe 7115 function so we have to make it shared there. 7116 7117 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 7118 7119 commit 2e96aa641bc8a5c263a5b643961383cdf530d044 7120 Author: Marek Olk <maraeo (a] gmail.com> 7121 Date: Sun Oct 28 12:50:49 2012 -0400 7122 7123 radeon: turn on 2D tiling by default on r6xx-cayman 7124 7125 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7126 7127 commit e830c039479539735d0836082abef24f6841c803 7128 Author: Marek Olk <maraeo (a] gmail.com> 7129 Date: Sun Oct 28 12:49:15 2012 -0400 7130 7131 radeon: log if 2D tiling is enabled 7132 7133 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7134 7135 commit ce1b745dcb60dc516ad999756240b78e72a7aa54 7136 Author: Alex Deucher <alexander.deucher (a] amd.com> 7137 Date: Tue Oct 16 13:05:46 2012 -0400 7138 7139 radeon: add new pci ids for SI 7140 7141 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 7142 7143 commit e3519be285bdab50ef2193f98826a351a4414fae 7144 Author: Dave Airlie <airlied (a] redhat.com> 7145 Date: Tue Oct 16 16:09:23 2012 +1000 7146 7147 radeon/dri2: fix the window->pixmap math properly. 7148 7149 Okay kwin did things different, and I really hope this time this is the 7150 correct transformation operation. It works here with gears under kwin 7151 and under mutter and under xcompmgr. 7152 7153 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7154 7155 commit bd9e2c064d93afbe4ab84febb2a43ae6a95ab17b 7156 Author: Dave Airlie <airlied (a] redhat.com> 7157 Date: Thu Oct 11 14:36:11 2012 +1000 7158 7159 radeon/dri2: fix damage reporting on pageflip 7160 7161 when you use MPX or slave usb devices, the damage posting was incorrect, 7162 and resulted in misrendered boxes on the screen. 7163 7164 Fix it by reporting damage correctly. 7165 7166 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7167 7168 commit e8cb0b721e6ea251f85c799ca0563bfa59a2d37c 7169 Author: Adam Jackson <ajax (a] redhat.com> 7170 Date: Wed Sep 26 09:27:54 2012 -0400 7171 7172 Remove mibstore.h 7173 7174 Signed-off-by: Adam Jackson <ajax (a] redhat.com> 7175 7176 commit 8637f772347c958fa3beea6c0dc9c22255db70e5 7177 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7178 Date: Thu Sep 13 19:50:35 2012 +0200 7179 7180 glamor: Force acceleration for DRI2 copies. 7181 7182 Improves performance of non-pageflipped 3D apps on SI. 7183 7184 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7185 7186 commit 40b87f0e9db2642ca8a70e994fb4d3742e199640 7187 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7188 Date: Wed Sep 12 18:44:10 2012 +0200 7189 7190 glamor: Fix DRI2 crash introduced by PRIME changes. 7191 7192 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7193 7194 commit 646ffd49d2cf0d3130301e8ad2f49f620bd9ef1b 7195 Author: Dave Airlie <airlied (a] redhat.com> 7196 Date: Mon Sep 10 15:31:06 2012 +1000 7197 7198 radeon: fix build against old servers 7199 7200 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=54711 7201 7202 Reported-by: wolput (a] onsneteindhoven.nl 7203 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7204 7205 commit 7c7f2775604da61537c24c249884aaa220e06e13 7206 Author: Dave Airlie <airlied (a] redhat.com> 7207 Date: Mon Sep 10 11:27:45 2012 +1000 7208 7209 radeon: don't fail to load if we have no modes and gpu driver. 7210 7211 This ensures radeon loads on output less GPUs as a GPU driver. 7212 7213 There are server bugs that also need to be fixed. 7214 7215 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7216 7217 commit 7fe16dd037ac688726869816888bd77fe4356f8e 7218 Author: Dave Airlie <airlied (a] gmail.com> 7219 Date: Wed Sep 5 10:41:51 2012 +1000 7220 7221 radeon: fix typo in glamor paths 7222 7223 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=54519 7224 7225 Reported-By: scaty (a] dcinformatique.com 7226 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7227 7228 commit cdc985e9190ce54dc6274ebc0e87ab954b15bb0d 7229 Author: Dave Airlie <airlied (a] gmail.com> 7230 Date: Tue Sep 4 18:18:50 2012 +1000 7231 7232 radeon: fix build against older servers 7233 7234 Another silly one. 7235 7236 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7237 7238 commit bd0e9c51407cb99f1fe4f0acfedc37ee0c354040 7239 Author: Dave Airlie <airlied (a] gmail.com> 7240 Date: Tue Sep 4 18:17:06 2012 +1000 7241 7242 radeon/dri2: fix build against older servers. 7243 7244 Older servers won't have DRI2UpdatePrime, so fix build. 7245 7246 bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=54433 7247 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7248 7249 commit 75c1f62f44d7b3709c1fa103d59733866967b841 7250 Author: Dave Airlie <airlied (a] redhat.com> 7251 Date: Mon Sep 3 13:09:02 2012 +1000 7252 7253 radeon: add capability and provider support 7254 7255 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7256 7257 commit ca8b61e916998458203dfa4b9d0c997fa7d90ba9 7258 Author: Dave Airlie <airlied (a] redhat.com> 7259 Date: Mon Sep 3 13:08:37 2012 +1000 7260 7261 radeon: allocated shared pixmaps as untiled gtt. 7262 7263 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7264 7265 commit 8c1bf9d8fe3948b72795984e625ef46b2f0bf654 7266 Author: Dave Airlie <airlied (a] redhat.com> 7267 Date: Mon Sep 3 13:08:05 2012 +1000 7268 7269 radeon: add pixmap sharing hooks. 7270 7271 This hooks into EXA and the dirty tracking to add sharing and output 7272 offload support. 7273 7274 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7275 7276 commit 3add8df8122697acfe126d4857f3946ce44b8305 7277 Author: Dave Airlie <airlied (a] redhat.com> 7278 Date: Mon Sep 3 13:07:07 2012 +1000 7279 7280 radeon: add dri2 offload support. 7281 7282 This adds the new dri2 interfaces required for radeon to be offloaded to. 7283 7284 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7285 7286 commit f7502a11c8ef9c453ceb40d26109977116df88c2 7287 Author: Dave Airlie <airlied (a] redhat.com> 7288 Date: Mon Sep 3 12:56:56 2012 +1000 7289 7290 radeon: add shared support to pixmaps. 7291 7292 this just adds the interface and shared support to the pixmap. 7293 7294 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7295 7296 commit f71139a2afe8fffb628331402bf829a6d67c9fff 7297 Author: Dave Airlie <airlied (a] redhat.com> 7298 Date: Mon Sep 3 12:37:22 2012 +1000 7299 7300 radeon: add platform bus loading support. 7301 7302 This allows the radeon driver to be loaded via the platform bus mechanism. 7303 7304 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7305 7306 commit 9911e72ea79e000ee0492d19a434aa98b83b5066 7307 Author: Dave Airlie <airlied (a] redhat.com> 7308 Date: Fri Aug 17 17:02:51 2012 +1000 7309 7310 radeon: damage full screen after pageflip 7311 7312 in order for offload to work we need to cause a full 7313 screen damage to be registered after pageflip. 7314 7315 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7316 7317 commit 0873b8cef21abf1e7286731758f0db1e7f2e7c97 7318 Author: Dave Airlie <airlied (a] redhat.com> 7319 Date: Fri Aug 3 16:45:50 2012 +1000 7320 7321 ati: fix stub driver loader to load all drivers properly 7322 7323 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7324 7325 commit cdf522f5992adf7b76f9153c777e6dddcf7758a6 7326 Author: Adam Jackson <ajax (a] redhat.com> 7327 Date: Wed Aug 15 13:38:28 2012 -0400 7328 7329 Implement ->driverFunc 7330 7331 Copied from fbdev, makes it so we can run without iopl. 7332 7333 Signed-off-by: Adam Jackson <ajax (a] redhat.com> 7334 7335 commit 763f387ccac2399f1d58ec112c8bed45196bea5e 7336 Author: Adam Jackson <ajax (a] redhat.com> 7337 Date: Wed Aug 15 12:49:10 2012 -0400 7338 7339 Don't bother to include vbe.h 7340 7341 The only VBE code here went away when we dropped UMS. 7342 7343 Signed-off-by: Adam Jackson <ajax (a] redhat.com> 7344 7345 commit aabcf5ea30195274c79ae94c0fe6edc965f5c123 7346 Author: Tomas Chvatal <tchvatal (a] suse.cz> 7347 Date: Mon Aug 13 22:58:18 2012 +0200 7348 7349 Add switch for udev to disable automagic detection. 7350 7351 Signed-off-by: Tomas Chvatal <tchvatal (a] suse.cz> 7352 7353 commit aef90993621331f01d17b95a751d4924d77eed0c 7354 Author: Alex Deucher <alexdeucher (a] gmail.com> 7355 Date: Mon Aug 6 11:00:45 2012 -0400 7356 7357 radeon: add some new SI pci ids 7358 7359 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7360 7361 commit 6ef1ad6a46348d3aecd8d1f5e94431ca2298853c 7362 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7363 Date: Fri Jul 13 11:15:25 2012 +0200 7364 7365 Deal more gracefully with DRI2 being unavailable at build or run time. 7366 7367 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7368 7369 commit ef8a404391036d8aa814dbda2407c789b8a64b92 7370 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7371 Date: Thu Jul 5 20:14:48 2012 +0200 7372 7373 Initial SI support. 7374 7375 Defaults to shadowfb. 3D acceleration is available with glamor. 2D 7376 acceleration is disabled until the radeonsi driver can handle glamor's 7377 shaders. 7378 7379 v2: add chip flags (Alex Deucher) 7380 7381 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7382 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 7383 7384 commit e9edd2f5002c642b59f028b3ec076d604ae8ce9d 7385 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7386 Date: Wed Jun 20 08:40:07 2012 +0200 7387 7388 Initial glamor support. 7389 7390 Enable at build time with --enable-glamor and runtime with 7391 7392 Option "AccelMethod" "glamor" 7393 7394 The most notable lack of functionality is XVideo. Use something like VDPAU for 7395 now. 7396 7397 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7398 7399 commit 9eac8021f3d33a63156f9f5d43a220e88bb3f8db 7400 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7401 Date: Wed Jun 27 14:48:47 2012 +0200 7402 7403 EXA: Factor out pixmap BO allocation into a helper function. 7404 7405 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7406 7407 commit 060c7836e7f7777bacca4e23f57c5985beab33bc 7408 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7409 Date: Wed Jul 4 11:49:18 2012 +0200 7410 7411 Some cosmetic fixups for the radeon manpage. 7412 7413 ATI -> ATI/AMD 7414 PCIE -> PCIe 7415 7416 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7417 7418 commit c3a9f64fe40c78331b9340c3832d344f38b11564 7419 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7420 Date: Wed Jul 4 11:46:30 2012 +0200 7421 7422 Drop UMS specific parts from the radeon manpage. 7423 7424 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7425 7426 commit 5ebebfdb4fadada157d1369f35a8610c0fa72e02 7427 Author: Dave Airlie <airlied (a] redhat.com> 7428 Date: Fri Jun 29 15:03:43 2012 +0100 7429 7430 radeon: fix conflict with fix in master. 7431 7432 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7433 7434 commit ff12ddf613ad42e4566d6e10fc6cf33a4e445736 7435 Author: Dave Airlie <airlied (a] redhat.com> 7436 Date: Fri Jun 29 15:00:44 2012 +0100 7437 7438 configure: bump to 6.99.99 7439 7440 this is in prep for 7.0 release. 7441 7442 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7443 7444 commit 50689ec8dbd4a68527b2ac16cecac298b8d441d0 7445 Merge: 6a60fcdf0 ae682a6a1 7446 Author: Dave Airlie <airlied (a] redhat.com> 7447 Date: Fri Jun 29 14:59:47 2012 +0100 7448 7449 Merge remote-tracking branch 'origin/kms-only' 7450 7451 This merges the removal of the UMS code and subsequent simplifications. 7452 7453 Hell yes. 7454 7455 * origin/kms-only: (24 commits) 7456 radeon: fix radeonchipsets properly 7457 radeon: fix RADEONChipsets 7458 radeon: migrate remainder of radeon_driver.c to rest of driver 7459 radeon: avoid including large static struct twice. 7460 radeon: drop some more unused macros/inlines 7461 radeon: drop another unused struct member. 7462 radeon: more unused stuff 7463 radeon: drop cardType shouldn't matter to userspace. 7464 radeon: drop radeonGetPixmapOffset 7465 radeon: move more functions into texture video 7466 radeon: drop legacy memory, merge into radeon video 7467 radeon: drop unused txoffset vars. 7468 radeon: drop vb_mc_addr, not needed anymore 7469 radeon: drop offset member of r600 accel object 7470 radeon: move macros into radeon.h 7471 radeon: drop more wrapper macros. 7472 radeon: further macro cleanups. 7473 radeon: make exa copy funcs static. 7474 radeon: drop FUNC_NAME macro and ONCE_ONLY stuff 7475 radeon: drop radeon_commonfuncs. 7476 ... 7477 7478 Conflicts: 7479 configure.ac 7480 7481 commit 6a60fcdf060a0f553d8d4f7939c2a05aa04bedfa 7482 Author: Dave Airlie <airlied (a] redhat.com> 7483 Date: Fri Jun 29 14:57:58 2012 +0100 7484 7485 configure: bump version after release. 7486 7487 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7488 7489 commit 008eb209d506ed7195fd544dfa91be9783946fd5 7490 Author: Dave Airlie <airlied (a] redhat.com> 7491 Date: Fri Jun 29 14:46:55 2012 +0100 7492 7493 configure: bump version number to 6.14.6 pre-release 7494 7495 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7496 7497 commit 72fea2635a56c4f572c07fc50fc58f703a21eb4f 7498 Author: Dave Airlie <airlied (a] redhat.com> 7499 Date: Fri Jun 29 14:46:17 2012 +0100 7500 7501 radeon: require libdrm 2.4.36 for KMS support. 7502 7503 This is due to some commits to the surface manager that fix bugs. 7504 7505 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7506 7507 commit 179b035835bfa99283d44ba2d5ee17e686196535 7508 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7509 Date: Wed Jun 27 19:26:51 2012 +0200 7510 7511 Fix up displayWidth vs. virtualX confusion in radeon_setup_kernel_mem(). 7512 7513 It was using the pitch (displayWidth) for the virtual width (virtualX). This 7514 prevented using page flipping in some cases, as displayWidth was already 7515 overaligned for virtualX, so the DRI2 front and back buffers ended up having 7516 a different pitch. 7517 7518 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7519 Reviewed-by: Jerome Glisse <jglisse (a] redhat.com> 7520 7521 commit 5dc5012320536f7b1e86a81ed3b823e3626828e1 7522 Author: Dave Airlie <airlied (a] redhat.com> 7523 Date: Mon Jun 25 09:23:04 2012 +0100 7524 7525 radeon: dixPrivateKeyRegistered is only in server 1.9 7526 7527 just fallback to the old behaviour on older servers. 7528 7529 should fix: https://bugs.freedesktop.org/show_bug.cgi?id=51388 7530 Review-over-irc: Michel Dnzer <michel.daenzer (a] amd.com> 7531 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7532 7533 commit ae682a6a1781ab023ab36e3154d6b3e88da63d64 7534 Author: Dave Airlie <airlied (a] redhat.com> 7535 Date: Mon Jun 18 13:11:31 2012 +0100 7536 7537 radeon: fix radeonchipsets properly 7538 7539 commit 6514bcdc0fc199b4f3a8b4273bd29e0fe255d232 7540 Author: Dave Airlie <airlied (a] redhat.com> 7541 Date: Mon Jun 18 13:02:03 2012 +0100 7542 7543 radeon: fix RADEONChipsets 7544 7545 reported on irc by okias. 7546 7547 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7548 7549 commit dd4ab5aba9047d522dbcfe8a341c368312e1a471 7550 Author: Dave Airlie <airlied (a] redhat.com> 7551 Date: Sat Jun 16 08:47:32 2012 +0100 7552 7553 radeon: migrate remainder of radeon_driver.c to rest of driver 7554 7555 This helps make a few more things static and the driver generally 7556 smaller. 7557 7558 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7559 7560 commit a19237fad2492d70777167c631f4e6e1fae0e908 7561 Author: Dave Airlie <airlied (a] redhat.com> 7562 Date: Sat Jun 16 08:19:35 2012 +0100 7563 7564 radeon: avoid including large static struct twice. 7565 7566 Just extern this in the one file that needs access. 7567 7568 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7569 7570 commit c8a5c5a0dd9adc46b03b898c9c1719dbbde05ba1 7571 Author: Dave Airlie <airlied (a] redhat.com> 7572 Date: Fri Jun 15 19:52:14 2012 +0100 7573 7574 radeon: drop some more unused macros/inlines 7575 7576 These are no longer used anywhere. 7577 7578 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7579 7580 commit 3097f273443b5b8fce097e2b1e09a17b02d06a40 7581 Author: Dave Airlie <airlied (a] redhat.com> 7582 Date: Fri Jun 15 19:45:15 2012 +0100 7583 7584 radeon: drop another unused struct member. 7585 7586 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7587 7588 commit cb97e75acab84b67b7b7358860788638efc9b344 7589 Author: Dave Airlie <airlied (a] redhat.com> 7590 Date: Fri Jun 15 19:38:45 2012 +0100 7591 7592 radeon: more unused stuff 7593 7594 Remove all CurrentLayout stuff. 7595 7596 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7597 7598 commit 81593e7deb688fa3108a0589c1418459ec0df4de 7599 Author: Dave Airlie <airlied (a] redhat.com> 7600 Date: Fri Jun 15 18:52:52 2012 +0100 7601 7602 radeon: drop cardType shouldn't matter to userspace. 7603 7604 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7605 7606 commit 997b0dddfeb180043ece5eff0ac0d6a1a751f9b5 7607 Author: Dave Airlie <airlied (a] redhat.com> 7608 Date: Fri Jun 15 18:43:01 2012 +0100 7609 7610 radeon: drop radeonGetPixmapOffset 7611 7612 no longer needed. 7613 7614 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7615 7616 commit 639061edb3ac35a24e4d6633c9b6384434dcac64 7617 Author: Dave Airlie <airlied (a] redhat.com> 7618 Date: Fri Jun 15 18:38:48 2012 +0100 7619 7620 radeon: move more functions into texture video 7621 7622 make more thing static. 7623 7624 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7625 7626 commit 85711a0a6a553232b603dbbb2b2d234c35391578 7627 Author: Dave Airlie <airlied (a] redhat.com> 7628 Date: Fri Jun 15 17:26:50 2012 +0100 7629 7630 radeon: drop legacy memory, merge into radeon video 7631 7632 This merges these two functions into textured video code. 7633 7634 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7635 7636 commit 445ca1d66ce7b37762731fe67407cc1b978bb129 7637 Author: Dave Airlie <airlied (a] redhat.com> 7638 Date: Fri Jun 15 17:13:48 2012 +0100 7639 7640 radeon: drop unused txoffset vars. 7641 7642 These vars aren't being used anymore. 7643 7644 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7645 7646 commit 832244de9e6d7aa1b8fbeb29c9a6a86923744d60 7647 Author: Dave Airlie <airlied (a] redhat.com> 7648 Date: Fri Jun 15 17:09:00 2012 +0100 7649 7650 radeon: drop vb_mc_addr, not needed anymore 7651 7652 This field is totally unused now, so drop it. 7653 7654 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7655 7656 commit 6559b6e28bf111b64c3aeb5ec1260acfeb3e0b7f 7657 Author: Dave Airlie <airlied (a] redhat.com> 7658 Date: Fri Jun 15 17:05:54 2012 +0100 7659 7660 radeon: drop offset member of r600 accel object 7661 7662 This is always 0 now. 7663 7664 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7665 7666 commit 857179e538c9752fab239a21312d0fcc80022151 7667 Author: Dave Airlie <airlied (a] redhat.com> 7668 Date: Fri Jun 15 16:55:23 2012 +0100 7669 7670 radeon: move macros into radeon.h 7671 7672 Drop separate header file, there isn't many separate macros now. 7673 7674 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7675 7676 commit 148e889149f5ecc3dc2473f2a9e01572eeb037d3 7677 Author: Dave Airlie <airlied (a] redhat.com> 7678 Date: Fri Jun 15 16:52:24 2012 +0100 7679 7680 radeon: drop more wrapper macros. 7681 7682 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7683 7684 commit ad8629e3323efe484dd1421d94dd934e1bfafe56 7685 Author: Dave Airlie <airlied (a] redhat.com> 7686 Date: Fri Jun 15 16:49:13 2012 +0100 7687 7688 radeon: further macro cleanups. 7689 7690 drop some more of the wrapper macros. 7691 7692 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7693 7694 commit 96644f96f1a2df1d7d1ca8cbdc9a105affdb4f32 7695 Author: Dave Airlie <airlied (a] redhat.com> 7696 Date: Fri Jun 15 16:34:53 2012 +0100 7697 7698 radeon: make exa copy funcs static. 7699 7700 These are no longer used outside this area. 7701 7702 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7703 7704 commit 83904b820b81d226940ec6d663246f243f92e7ba 7705 Author: Dave Airlie <airlied (a] redhat.com> 7706 Date: Fri Jun 15 16:30:41 2012 +0100 7707 7708 radeon: drop FUNC_NAME macro and ONCE_ONLY stuff 7709 7710 Drop all the CP vs MMIO macros that name functions. 7711 7712 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7713 7714 commit b422d8085aab4e01a1ac4ffb247c73fe51aafa0d 7715 Author: Dave Airlie <airlied (a] redhat.com> 7716 Date: Fri Jun 15 16:14:54 2012 +0100 7717 7718 radeon: drop radeon_commonfuncs. 7719 7720 modify all the macros and drop it into radeon_accel.c 7721 7722 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7723 7724 commit 29dd3f84eecd2892a438f004403521de8986a91d 7725 Author: Dave Airlie <airlied (a] redhat.com> 7726 Date: Fri Jun 15 16:11:03 2012 +0100 7727 7728 radeon: misc cleanups. 7729 7730 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7731 7732 commit 678c1f3425c0fa7e98795c8e351da77d04652e02 7733 Author: Dave Airlie <airlied (a] redhat.com> 7734 Date: Fri Jun 15 16:06:27 2012 +0100 7735 7736 make pci chipsets static 7737 7738 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7739 7740 commit 95cbbe999418c72dc407069d31e759d70f49ab16 7741 Author: Dave Airlie <airlied (a] redhat.com> 7742 Date: Fri Jun 15 16:01:53 2012 +0100 7743 7744 drop RING_LOCALS/ACCEL_PREAMBLE 7745 7746 these were unused now. 7747 7748 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7749 7750 commit 18d5ae3bd9075ac1a2ee21b071ac133e2e634b62 7751 Author: Dave Airlie <airlied (a] redhat.com> 7752 Date: Fri Jun 15 10:05:03 2012 +0100 7753 7754 radeon: drop all UMS/DRI1/XAA/overlay support. 7755 7756 This overhauls the radeon driver and removes all the old UMS-only code, 7757 it drops all the UMS, DRI1, XAA, overlay Xv, video capture, tv tuners 7758 7759 There are probably a lot more cleanups that will fall out of this afterwards. 7760 7761 So far this is compile/build tested. 7762 7763 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7764 7765 commit 248e912c487636d7352cfad43c03fc9f19fc2215 7766 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7767 Date: Fri Jun 8 12:18:48 2012 +0200 7768 7769 UMS: Fix CRTC DPMS state check. 7770 7771 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7772 7773 commit 58883711136fa6671d4f3250d4ee973e2953ebb4 7774 Author: Alex Deucher <alexdeucher (a] gmail.com> 7775 Date: Thu Jun 7 11:47:21 2012 -0400 7776 7777 configure: bump version post release 7778 7779 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7780 7781 commit 3a9103f855c67bbcd7b97ee2e96836353e0d789d 7782 Author: Alex Deucher <alexdeucher (a] gmail.com> 7783 Date: Thu Jun 7 11:27:08 2012 -0400 7784 7785 configure: bump for release 7786 7787 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7788 7789 commit f1693ee09e9b251eb75e3e5fe0ee7699fec6fb98 7790 Author: Alex Deucher <alexdeucher (a] gmail.com> 7791 Date: Thu Jun 7 11:23:31 2012 -0400 7792 7793 configure: bump libdrm_radeon requirement 7794 7795 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7796 7797 commit 77a056c1725c9c44a328ed324687bdf35144e9e2 7798 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7799 Date: Wed Jun 6 12:29:37 2012 +0200 7800 7801 Don't page-flip or wait on a CRTC while we're VT-switched away. 7802 7803 We don't know what the CRTC state is. 7804 7805 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7806 7807 commit 1d9f1eb7e65c57c52759ccbfa51e8bdd5a10acac 7808 Author: Michel Dnzer <michel.daenzer (a] amd.com> 7809 Date: Wed Jun 6 12:16:08 2012 +0200 7810 7811 Don't wait on a CRTC which has been disabled via DPMS (bug #49761). 7812 7813 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=49761 . 7814 7815 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 7816 7817 commit 4603285aa8efaf2614f15a38254465cec2075f11 7818 Author: Alex Deucher <alexdeucher (a] gmail.com> 7819 Date: Tue Jun 5 09:48:59 2012 -0400 7820 7821 radeon: add new PCI ids 7822 7823 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 7824 7825 commit 9307609420b4b209767d2057b4803bcb16cc1455 7826 Author: Dave Airlie <airlied (a] redhat.com> 7827 Date: Sat Jun 2 17:16:34 2012 +0100 7828 7829 radeon: improve smooth startup fallbacks. 7830 7831 If we can't handover the framebuffer, memset it to black. 7832 7833 mostly ported from nouveau. 7834 7835 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7836 7837 commit 89a4c79a43a2b6963e41d7812e8fe5beedb556af 7838 Author: Dave Airlie <airlied (a] redhat.com> 7839 Date: Sat Jun 2 17:14:14 2012 +0100 7840 7841 radeon: fix smooth startup with tiling enabled. 7842 7843 We need to use the surface we worked out when we allocated the front bo, 7844 not work out a new surface from scratch. 7845 7846 This fixes smooth handover from plymouth to gdm on F17. 7847 7848 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7849 7850 commit c1b9b2c9d71a87a0b98ccdb542da8a30c8c4d99a 7851 Author: Dave Airlie <airlied (a] redhat.com> 7852 Date: Fri Jun 1 12:22:06 2012 +0100 7853 7854 radeon: fix compat api for newest planned API. 7855 7856 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7857 7858 commit 68c51d148b22ee8cbe98b5d5441e3494804ee808 7859 Author: Dave Airlie <airlied (a] redhat.com> 7860 Date: Fri Jun 1 12:21:34 2012 +0100 7861 7862 drmmode: drop flags arg to adjust frame. 7863 7864 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7865 7866 commit b1e9c3081d2b975682a62389aa29063b79f9655c 7867 Author: Dave Airlie <airlied (a] redhat.com> 7868 Date: Thu May 24 17:54:41 2012 +0100 7869 7870 radeon: move include down further 7871 7872 fixes tinderbox build. 7873 7874 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7875 7876 commit 5ed959585f35a4eff2cd803ae222b46d164f210d 7877 Author: Dave Airlie <airlied (a] redhat.com> 7878 Date: Thu May 24 08:17:11 2012 +0100 7879 7880 radeon: update compat-api.h for block handler + enable/disable fb. 7881 7882 This updates the compat stuff for the latest block handler code, 7883 and the enable/disable interface. 7884 7885 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7886 7887 commit 2de17efdcb502f91fb145130c12be3cb164f03c5 7888 Author: Dave Airlie <airlied (a] redhat.com> 7889 Date: Thu May 24 08:16:39 2012 +0100 7890 7891 radeon/xvmc: fix build with new API. 7892 7893 This was missing the compat include. 7894 7895 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7896 7897 commit a1f47768309aa8b14b13f05ba01bb490024c4cbc 7898 Author: Dave Airlie <airlied (a] redhat.com> 7899 Date: Thu May 24 07:55:14 2012 +0100 7900 7901 radeon: fix a few more RHDAtomBiosFunc uses 7902 7903 I missed these in my initial search/replace for some reason. 7904 7905 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7906 7907 commit c26e40cf60eb6a2bffb5ec7cba1f1bbffcbf1d53 7908 Author: Dave Airlie <airlied (a] redhat.com> 7909 Date: Wed May 23 12:12:54 2012 +0100 7910 7911 radeon/atombios: rewrite atombios parser code to not use xf86Screens. 7912 7913 Just pass the ScrnInfoPtr around instead. 7914 7915 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7916 7917 commit 14122214edabd2a574b66577289a7c8732f4e953 7918 Author: Dave Airlie <airlied (a] redhat.com> 7919 Date: Wed May 23 11:57:35 2012 +0100 7920 7921 radeon/generic_bus: stop passing scrnIndex 7922 7923 Just pass a pointer to the screen, removes usage of xf86Screens lookup 7924 7925 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7926 7927 commit 2101e485480e3ffd6a17ed12f5287b8b1e4e0223 7928 Author: Dave Airlie <airlied (a] redhat.com> 7929 Date: Tue May 22 16:17:55 2012 +0100 7930 7931 compat for new server API 7932 7933 commit 9d2549057ffced996bfaeb8df4901ef57da636c7 7934 Author: Dave Airlie <airlied (a] redhat.com> 7935 Date: Wed May 16 15:45:53 2012 +0100 7936 7937 ati: convert to new screen conversion APIs 7938 7939 The compat header takes care of the old server vs new server. 7940 7941 this commit was autogenerated from util/modular/x-driver-screen-scrn-conv.sh 7942 7943 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7944 7945 commit 80a45bfc26e20d1d481f13cb15c0720d165a8306 7946 Author: Dave Airlie <airlied (a] redhat.com> 7947 Date: Wed May 16 15:42:59 2012 +0100 7948 7949 radeon: add compat-api.h 7950 7951 commit 84c3900b5179a8d2cc5f925e94ed5b2bbdee1ad9 7952 Author: Dave Airlie <airlied (a] redhat.com> 7953 Date: Wed May 23 08:49:45 2012 +0100 7954 7955 radeon/xaa: drop scrnIndex parameter to some functions 7956 7957 This isn't needed, and makes api changes later easier. 7958 7959 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 7960 7961 commit 3fb694b308ebadd1b849836059b6b56bb19385f7 7962 Author: Thierry Vignaud <thierry.vignaud (a] gmail.com> 7963 Date: Wed May 16 14:43:53 2012 +0200 7964 7965 UMS/EXA: Add reminder for potential solid picture performance issue. 7966 7967 commit 4b9bad959438725e4434e2aa4f142542d8dcc260 7968 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 7969 Date: Fri May 11 05:25:32 2012 +0200 7970 7971 radeon: use GB_GR and BG_RG formats for packed yuv video for r600+ 7972 7973 Those formats were invented for exactly that purpose so use them. 7974 This saves some code and also some hw resources (only need one 7975 sampler instead of two for packed yuv). 7976 Only tested on EG. 7977 7978 commit 37786e9027b8c8d1f9ec9928915784dd28853766 7979 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 7980 Date: Fri May 11 05:25:33 2012 +0200 7981 7982 radeon: avoid rounding errors in texture coords for textured xv on EG+ 7983 7984 make sure the division is done with floats, otherwise the coordinate 7985 can be wrong up to 1 texel. 7986 Particularly visible with clipping and small source scaled up (since one 7987 texel can be a shift of several pixels) but could be seen even unscaled. 7988 Should provide more accurate coords without clipping too depending on the 7989 scale factor probably. 7990 This is a straight port of 688c8a54a00b01e73a11970ad2abe858f8c7c5c4 7991 when I apparently forgot the eg code... 7992 7993 commit faea3aafa8c9a7c1bc866ffcd847972c5569c8bc 7994 Author: Anisse Astier <anisse (a] astier.eu> 7995 Date: Thu May 10 17:22:59 2012 +0200 7996 7997 Fail more gracefully when drm surface manager can't be initialized 7998 7999 Should make bugs like https://bugs.freedesktop.org/show_bug.cgi?id=48138 8000 easier to diagnose. 8001 8002 [ Michel Dnzer: Appended newline to error message. ] 8003 8004 Signed-off-by: Anisse Astier <anisse (a] astier.eu> 8005 Singed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8006 8007 commit b0b7d8d26fd107df342b5c87b0a38e5bb08101a9 8008 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8009 Date: Wed May 9 11:08:49 2012 +0200 8010 8011 EXA/UMS: Synchronize to the GPU before writing solid colour to scratch pixmap. 8012 8013 UMS doesn't do this automagically. It's a big hammer that will probably suck 8014 for performance, but I don't have any better ideas right now. 8015 8016 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8017 8018 commit 6bda7ceda645e838723883d133d614def1511d16 8019 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8020 Date: Thu May 3 15:07:30 2012 +0200 8021 8022 EXA: Attempt to fix solid picture acceleration with UMS. 8023 8024 Only compile tested, but should fix 8025 https://bugs.freedesktop.org/show_bug.cgi?id=49182 . 8026 8027 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8028 8029 commit b33cf9cf9dce5add8aaccab10ba5aaa64247c8ce 8030 Author: Alex Deucher <alexdeucher (a] gmail.com> 8031 Date: Wed May 2 09:55:16 2012 -0400 8032 8033 radeon: add MacModel entry for SAM440ep embedded board 8034 8035 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8036 8037 commit 0bda305f7ab2a4720b3fea3f318ab2a73be151e5 8038 Author: Alex Deucher <alexander.deucher (a] amd.com> 8039 Date: Fri Apr 13 13:28:10 2012 -0400 8040 8041 EXA: Support acceleration of solid pictures on R2xx. 8042 8043 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 8044 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 8045 8046 commit a7754b076e355fef3aea082b6e2d3aefbb8e7a9b 8047 Author: Alex Deucher <alexander.deucher (a] amd.com> 8048 Date: Fri Apr 13 13:24:46 2012 -0400 8049 8050 EXA: Support acceleration of solid pictures on R1xx. 8051 8052 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 8053 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 8054 8055 commit cac2d7ed54918579418fc762558497d3ec547fad 8056 Author: Alex Deucher <alexander.deucher (a] amd.com> 8057 Date: Sat Apr 14 08:53:39 2012 -0400 8058 8059 EXA: Support acceleration of solid pictures on Evergreen/NI. 8060 8061 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 8062 Reviewed-by: Michel Dnzer <michel.daenzer (a] amd.com> 8063 8064 commit d88b9700137ee407c483f263bb55c77cd6f92fef 8065 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8066 Date: Fri Apr 13 09:35:38 2012 +0200 8067 8068 EXA: Support acceleration of solid pictures on R3xx-R7xx. 8069 8070 Allocate 1x1 scratch pixmaps to hold the solid picture colours. 8071 8072 This works around https://bugs.freedesktop.org/show_bug.cgi?id=47266 and might 8073 improve performance in other cases as well. 8074 8075 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8076 8077 commit 66b586b9b9cdaf70f0fcd547b5a04f044d848d44 8078 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8079 Date: Mon Apr 16 15:33:36 2012 +0200 8080 8081 RADEONCopySwap: Fix RADEON_HOST_DATA_SWAP_16BIT case. 8082 8083 It was the same code as for RADEON_HOST_DATA_SWAP_32BIT. This caused bus errors 8084 on FreeBSD/PPC, but I'm not sure how it could not cause problems anywhere... 8085 8086 Reported-by: Andreas Tobler <andreast (a] fgznet.ch> 8087 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8088 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 8089 8090 commit 1e656cd5973bbcf85b63406638ede0bccd65e28f 8091 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8092 Date: Mon Apr 16 15:44:01 2012 +0200 8093 8094 Make radeon_setup_kernel_mem failures more graceful and verbose. 8095 8096 So that bugs like https://bugs.freedesktop.org/show_bug.cgi?id=48138 can be 8097 diagnosed more easily. 8098 8099 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8100 8101 commit d282719a9c2fb0ee32830aa75b8dfbb9392954ed 8102 Author: Jerome Glisse <jglisse (a] redhat.com> 8103 Date: Wed Apr 4 17:08:30 2012 -0400 8104 8105 r6xx-r9xx: force 1D tiling for buffer with height < 64 8106 8107 Due to some old kernel issue, height is 8 aligned insided the ddx 8108 For buffer with height btw 57 & 63 this lead ddx to believe it can 8109 allocate a 2D tiled surface while mesa will not align height and 8110 will assume 1D tiled leading to disagreement and rendering issue. 8111 This patch force buffer with height < 64 to be 1D tiled. 8112 8113 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 8114 8115 commit de2419e5509398328903ca61f4ea80852cba6bc4 8116 Author: Alex Deucher <alexdeucher (a] gmail.com> 8117 Date: Thu Mar 29 00:28:39 2012 -0400 8118 8119 configure: bump version post release 8120 8121 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8122 8123 commit 9425c50e93903fb64d9e569cfdc1e2c35d16ce25 8124 Author: Alex Deucher <alexdeucher (a] gmail.com> 8125 Date: Thu Mar 29 00:19:12 2012 -0400 8126 8127 configure: bump version for release 8128 8129 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8130 8131 commit dc18d771713ecd893c7d5833da6e0661093161dc 8132 Author: Alex Deucher <alexdeucher (a] gmail.com> 8133 Date: Wed Mar 28 17:32:53 2012 -0400 8134 8135 configure: bump libdrm requirement for TN support 8136 8137 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8138 8139 commit 03535904a3e1542b3924d0a062c4b022ca196888 8140 Author: Alex Deucher <alexdeucher (a] gmail.com> 8141 Date: Tue Mar 27 09:48:28 2012 -0400 8142 8143 radeon: man page updates 8144 8145 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8146 8147 commit 6ed191c33caa33c12c2c6dafcba3a5ab1bf4a02f 8148 Author: Alex Deucher <alexdeucher (a] gmail.com> 8149 Date: Tue Mar 20 19:57:53 2012 -0400 8150 8151 radeon/kms: add TN pci ids 8152 8153 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8154 8155 commit 55d65fcf33eb383e3fbc7a1d469ab68a70a7ab37 8156 Author: Alex Deucher <alexdeucher (a] gmail.com> 8157 Date: Tue Mar 20 19:54:57 2012 -0400 8158 8159 radeon/kms: add support for TN (trinity) APUs 8160 8161 - KMS only 8162 - Includes full EXA/Xv support 8163 8164 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8165 8166 commit b5cf9bd693cf4090956add4c33c4fae9c3069a03 8167 Author: Marek Olk <maraeo (a] gmail.com> 8168 Date: Wed Mar 7 11:01:40 2012 -0500 8169 8170 r6xx: initialize SX_MISC 8171 8172 If Mesa set it to 1, the DDX would not render anything = the monitor would 8173 basically freeze. 8174 8175 agd5f: update emit count as well. 8176 8177 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8178 8179 commit 355dc4295912c153f5333421594fa90aa119a056 8180 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8181 Date: Tue Mar 6 15:52:40 2012 +0100 8182 8183 DRI2: Unreference buffers immediately when event wait info is invalidated. 8184 8185 Deferring this could result in trying to unreference buffers from a previous 8186 server generation, i.e. accessing freed memory. 8187 8188 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8189 Tested-by: Christian Knig <Christian.koenig (a] amd.com> 8190 8191 commit fe51469b2e02e4d565050bab077985270fb58a9b 8192 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8193 Date: Tue Mar 6 15:52:40 2012 +0100 8194 8195 Re-register DRM FD wakeup handler for each server generation. 8196 8197 Fixes hang when trying to use DRI2 swap scheduling after a server reset. 8198 8199 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8200 Tested-by: Christian Knig <Christian.koenig (a] amd.com> 8201 8202 commit 878454ae8d8e96dd27a19d0b30940d014c4cd7e2 8203 Author: Hans Verkuil <hverkuil (a] xs4all.nl> 8204 Date: Fri Feb 24 09:35:39 2012 -0500 8205 8206 Fix ConnectorTable crash in radeon_output.c 8207 8208 The sam440ep PPC board requires a ConnectorTable xorg.conf option, but putting 8209 in that option causes the radeon driver to crash. I finally traced it to a 8210 copy-and-paste bug in radeon_output.c as a result of a major rework in commit 8211 82f12e5a40c1fbcb91910a0f8b725c34fff02aae. 8212 8213 The actual crash occurred in RADEONPrintPortMap(). 8214 8215 Signed-off-by: Hans Verkuil <hverkuil (a] xs4all.nl> 8216 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8217 8218 commit 688c8a54a00b01e73a11970ad2abe858f8c7c5c4 8219 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 8220 Date: Sat Feb 18 21:12:34 2012 +0100 8221 8222 radeon: avoid rounding errors in texture coords for textured xv 8223 8224 make sure the division is done with floats, otherwise the coordinate 8225 can be wrong up to 1 texel. 8226 Particularly visible with clipping and small source scaled up (since one 8227 texel can be a shift of several pixels) but could be seen even unscaled. 8228 Should provide more accurate coords without clipping too depending on the 8229 scale factor probably. 8230 Changed for r100-r600, though only tested on r300. 8231 8232 commit 2778b56252124ef6f636a493d2e1457b43911c37 8233 Author: Jerome Glisse <jglisse (a] redhat.com> 8234 Date: Mon Feb 13 20:42:57 2012 -0500 8235 8236 radeon: r6xx-eg use linear general when using scratch bo 8237 8238 In path where we need to use scratch bo as temporary area, 8239 consider it as linear buffer. Not linear aligned. Fix some 8240 case such as in bugs: 8241 8242 https://bugs.freedesktop.org/show_bug.cgi?id=45827 8243 8244 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 8245 8246 commit c66ae235700f5efe64eb168327551b8f1d153c9c 8247 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8248 Date: Mon Feb 13 10:43:58 2012 +0100 8249 8250 Handle new xorg_list API. 8251 8252 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=45937 8253 8254 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8255 8256 commit 7ff277e22c629308915307bbee96eb25ff77f8b9 8257 Author: Alex Deucher <alexdeucher (a] gmail.com> 8258 Date: Fri Feb 10 13:04:59 2012 -0500 8259 8260 radeon: fix crash in drmmode_create_bo_pixmap() 8261 8262 Only init surface on r6xx+. Return NULL rather than 8263 FALSE. 8264 8265 Fixes: 8266 https://bugs.freedesktop.org/show_bug.cgi?id=45829 8267 8268 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8269 8270 commit 60b949f34df5db05e0e102cc3daa33469aa50cfc 8271 Author: Alex Deucher <alexdeucher (a] gmail.com> 8272 Date: Fri Feb 10 13:11:26 2012 -0500 8273 8274 radeon/kms: reusing fd message is not an error 8275 8276 It's standard behavior. 8277 8278 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8279 8280 commit e20284409937d784847339b5d466a95012d85940 8281 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8282 Date: Fri Feb 3 12:21:59 2012 +0100 8283 8284 EXA/r6xx+: Only set write domain or read domains, not both. (Bug #43893) 8285 8286 Avoids an accounting bug in libdrm_radeon 2.4.31 or older. 8287 8288 See https://bugs.freedesktop.org/show_bug.cgi?id=43893 8289 8290 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8291 8292 commit 5a7f64c5170ca424c9dca739662018e30df13413 8293 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8294 Date: Wed Feb 8 11:35:35 2012 +0100 8295 8296 Fix UMS build failure. 8297 8298 And some UMS specific warnings. 8299 8300 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8301 8302 commit 87f776b966f4200c97a989536d4b71822ae4c0b3 8303 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8304 Date: Wed Feb 8 11:10:27 2012 +0100 8305 8306 Remove unused local variable 'height'. 8307 8308 Pointed out by gcc -Wunused-variable. 8309 8310 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8311 8312 commit f63262e2f112a348c45f0dcecd891c8b6d9c5ee8 8313 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8314 Date: Wed Feb 8 10:28:45 2012 +0100 8315 8316 evergreen: Initialize source surface member for textured video. 8317 8318 Fixes crash reported by Ole Salscheider on IRC. 8319 8320 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8321 8322 commit c8f104d38870f14049402bbc14f662c151caeeef 8323 Author: Jerome Glisse <jglisse (a] redhat.com> 8324 Date: Tue Feb 7 15:04:37 2012 -0500 8325 8326 radeon: fix tiling for weird resolution 8327 8328 Should also fix xv for some case. 8329 8330 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 8331 8332 commit 615033f2b5e3817e335e9d022fc9fdcf8ac8b11a 8333 Author: Jerome Glisse <jglisse (a] redhat.com> 8334 Date: Tue Dec 13 11:08:19 2011 -0500 8335 8336 r600-evergreen: use common surface allocator for tiling v11 8337 8338 Use libdrm common surface code so mesa,ddx have same idea 8339 about tiling surface and what their pitch should be and 8340 the alignment constraint. 8341 8342 v2 fix remaining issue add new option to conditionaly enable 8343 v3 fix fbcon copy and r600 exa copy path 8344 v4 fix non tiled path 2D tiling on GPU >= R600, set it to false 8345 as default 8346 v5 adapt to pixel/element size split of libdrm/radeon 8347 v6 update to properly handle falling back to 1d tiled 8348 v6 final fix to tile split value on evergreen and newer 8349 v7 fix default array mode on r6xx, fix height alignment issue 8350 on evergreen 8351 v8 fix tile split value 8352 v9 add stencil tile split support, simplify dri2 for stencil 8353 with evergreen 8354 v10 Try to fix xv path regarding tiling. Adapt to libdrm API 8355 change. Try to fix case where there is no surface which 8356 means non tiled bo. 8357 v11 check for proper libdrm 8358 8359 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 8360 8361 commit 36c190671081967bac6fff48aaf66d67b639a48c 8362 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8363 Date: Wed Feb 1 13:21:02 2012 +0100 8364 8365 Fix vline range calculations. 8366 8367 The range passed in is in pixmap coordinates, so the CRTC offset needs to be 8368 added to the clamping limits and subtracted from the clamped range for 8369 pre-AVIVO display engines. 8370 8371 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8372 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 8373 8374 commit bb0e1531ac6949d38025d7dcb19234fee33b2acf 8375 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8376 Date: Wed Feb 1 13:07:11 2012 +0100 8377 8378 Check for empty vline ranges after clamping. 8379 8380 The clamping could turn a previously non-empty range into an empty one. 8381 8382 Also, start == stop means the range is empty. 8383 8384 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8385 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 8386 8387 commit 13b3aed4ef9afbcbaea1dcf0ed1acb162b240a3f 8388 Author: Alex Deucher <alexdeucher (a] gmail.com> 8389 Date: Tue Jan 10 09:35:09 2012 -0500 8390 8391 EXA/r6xx+: fix rop setting for overlapping copies 8392 8393 Need to use GXCopy for the src to temp copy, then 8394 the original rop for the temp to dest copy. 8395 8396 Noticed by: Frank Huang 8397 8398 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8399 8400 commit eb6d769a087b2ed5952f477fc3f0b0625810a287 8401 Author: Egbert Eich <eich (a] suse.de> 8402 Date: Tue Nov 15 18:50:56 2011 +0100 8403 8404 DPMS: Split non-modeset CRTC DPMS function. 8405 8406 RADEONRestore() calls crtc->funcs->dpms() after most of the mode setting 8407 subsystems have been restored. This function enables the CRTCs but does 8408 more: it calls DRM pre- and post-modeset ioctls and sets up the palettes 8409 (LUTs). 8410 None of these two things are needed. Accessing the palette registers after 8411 restoring the PLLs can even lead to lockups. 8412 Thus the CRTC DPMS function is split into two parts: one that just enables 8413 /disables the CRTC and one which wraps this function and does the rest. 8414 Now the inner function can be called directly from RADEONRestore() as 8415 there is no need to go thru the RandR hooks in this function while the 8416 RandR hook uses the wrappering function so the full functionality is 8417 preserved from an RandR point of view. 8418 8419 Signed-off-by: Egbert Eich <eich (a] freedesktop.org> 8420 Reviewed-by: Alex Deucher <alexdeucher (a] gmail.com> 8421 8422 commit ac51e331895b216d288bc7bd108a38b362214668 8423 Author: Egbert Eich <eich (a] suse.de> 8424 Date: Mon Nov 14 19:10:01 2011 +0100 8425 8426 UMS: Fix lockups in palette save/restore on pre-AVIVO chips. 8427 8428 The reintroduction of palette save/restore in 5efdf514 causes some 8429 pre-AVIVO chips to lock up. An investigation revealed that accessing 8430 palette registers when the associated PLL is not running is causing 8431 this. With UMS the PLL setup that is saved has been done by the BIOS 8432 typically. 8433 A similar issue was observed when VGA palette save/restore had 8434 been reinitroduced with 80eee856: 8435 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=480312 8436 and has been worked around for Linux without further investigation 8437 by 87e66ce7. 8438 To fix the issue we now 8439 a. introduce 'on-demand' palette saving (ie the palette is 8440 saved before it is first altered). This guarantees that 8441 the palette register are only associated when the associated 8442 CRTC is active and thus the PLLs are powered up and running. 8443 b. move palette restore before PLL restore. 8444 c. eliminate generic VGA palette save/restore which seems to be 8445 unneeded when the palette is restored natively. 8446 It is believed that this caused the behavior described in 8447 https://bugs.freedesktop.org/show_bug.cgi?id=18407#c27 8448 8449 Signed-off-by: Egbert Eich <eich (a] freedesktop.org> 8450 Reviewed-by: Alex Deucher <alexdeucher (a] gmail.com> 8451 8452 commit 0a8d04eeac95f4db9d03ee31070bd825a7feb0b2 8453 Author: Matthieu Herrb <matthieu.herrb (a] laas.fr> 8454 Date: Sun Jan 1 18:27:54 2012 +0100 8455 8456 Update for new vgaHW API. 8457 8458 Signed-off-by: Matthieu Herrb <matthieu.herrb (a] laas.fr> 8459 Reviewed-by: Mark Kettenis <kettenis (a] openbsd.org> 8460 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8461 8462 commit ae45d7e6d8e6844cd4586c9ee97c21b257fa788f 8463 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8464 Date: Wed Dec 28 11:48:36 2011 +0100 8465 8466 DRI2: Can't use page flipping for pixmaps. (bug #42913) 8467 8468 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=42913 . 8469 8470 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8471 8472 commit aacbd629b02cbee3f9e6a0ee452b4e3f21376bd3 8473 Author: Alex Deucher <alexdeucher (a] gmail.com> 8474 Date: Mon Dec 12 09:32:30 2011 -0500 8475 8476 radeon: add some new pci ids 8477 8478 fixes: 8479 https://bugs.freedesktop.org/show_bug.cgi?id=43739 8480 8481 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8482 8483 commit bc54e415e2fd344939c5c788ea0686133a7e2c69 8484 Author: Dave Airlie <airlied (a] redhat.com> 8485 Date: Tue Dec 6 15:47:45 2011 +0000 8486 8487 radeon: add original radeon to always tiled. 8488 8489 and actually enable it for M7, previous commit only did one function. 8490 8491 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 8492 8493 commit ba46c7b0cf72d157748981eb3224d5eefb6200aa 8494 Author: Dave Airlie <airlied (a] redhat.com> 8495 Date: Tue Dec 6 13:42:49 2011 +0000 8496 8497 radeon: refine always tiled depth check 8498 8499 So it appears the M7 family always tiles its depth buffer also. 8500 8501 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 8502 8503 commit 98b2d5fe1722a43c4bbe7711ed7180a3fb65305f 8504 Author: Dave Airlie <airlied (a] redhat.com> 8505 Date: Mon Dec 5 18:44:28 2011 +0000 8506 8507 radeon: r200 depth buffers are always tiled 8508 8509 When we do the allocations we need to make sure the always tiled 8510 nature is taken into account. 8511 8512 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 8513 8514 commit 7dcefc69d9fbceae27cd03083c815e01a19b527e 8515 Author: Alex Deucher <alexdeucher (a] gmail.com> 8516 Date: Mon Dec 5 09:21:48 2011 -0500 8517 8518 Xv: Evergreen+ asics support 16k surfaces 8519 8520 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8521 8522 commit 72e386d42516e7cd3c2cbf2fffc9174cd3ec8451 8523 Author: Alex Deucher <alexander.deucher (a] amd.com> 8524 Date: Wed Nov 30 19:38:35 2011 -0500 8525 8526 radeon: add some new pci ids 8527 8528 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 8529 8530 commit 3853c3020d05175ae180b9a188dec7c425bdd0b8 8531 Author: Dave Airlie <airlied (a] redhat.com> 8532 Date: Mon Nov 28 18:38:30 2011 +0000 8533 8534 fixup xinerama since 9151f3b1c2ebcc34e63195888ba696f2183ba5e2 8535 8536 since the driver would call RRFirstOutput without checking if randr has 8537 been enabled, and it would crash in privates code. 8538 8539 reported by vereteran on #radeon 8540 8541 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 8542 Acked-on-irc-by: Michel Dnzer <michel.daenzer (a] amd.com> 8543 8544 commit d669c34f140c000f88c4b4e464e44e6c8694f581 8545 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 8546 Date: Mon Nov 21 11:35:40 2011 +1100 8547 8548 ddx/evergreen: Fix endian of ALU constants 8549 8550 The constants are written directly into a buffer object shared with the 8551 card and we "forget" to swap them. This patch fixes it by doing the swap 8552 in evergreen_set_alu_consts() in-place (ie, it modifies the buffer), 8553 which should be fine with the way we use it in the ddx. 8554 8555 This makes everything work fine on my caicos card on a G5 including some 8556 quik tests with Xv, gnome3 shell, etc... 8557 8558 Thanks a lot to Jerome Glisse for holding my hand through debugging that 8559 (and finding the actual bug). 8560 8561 Signed-off-by: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 8562 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8563 8564 commit 422bdd4fe6cb728e1dd08a56f6ee2d0f009cbfcb 8565 Author: Alex Deucher <alexdeucher (a] gmail.com> 8566 Date: Mon Nov 14 09:39:16 2011 -0500 8567 8568 radeon: add missing FireMV pci id 8569 8570 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8571 8572 commit 534fb6e413a909a9d1afd57d1c711844b7c5ddf4 8573 Author: Dave Airlie <airlied (a] redhat.com> 8574 Date: Fri Nov 11 10:26:51 2011 +0000 8575 8576 ati: enable bg none when fbcon succeeds and we are built against ABI after 10. 8577 8578 One less patch to keep carrying in Fedora. 8579 8580 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 8581 8582 commit 89452c08048c98fb5cc3dc551b3824be40d52cf2 8583 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8584 Date: Tue Nov 8 11:23:11 2011 +0100 8585 8586 UMS: Guard references to PCITAG / pciTag with XSERVER_LIBPCIACCESS (bug #42690) 8587 8588 Should fix https://bugs.freedesktop.org/show_bug.cgi?id=42690 . 8589 8590 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8591 8592 commit 5ec34ed95948f7164184551615c1fc4c3eef3b98 8593 Author: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 8594 Date: Thu Nov 3 20:16:47 2011 -0400 8595 8596 DRI/DRI2: remove hard-coded limitation to 6 crtcs 8597 8598 DRM's hard limit to the number of CRTCs is 32. ATI DDX unnecessarily 8599 clips this limit to 6 by hard coding initial assumption for 8600 output->possible_crtcs mask to 0x7f (before it gets trimmed down to 8601 what's really possible for a given output) and by allocating only 6 8602 entries for for cursor_bo[] array in RADEONInfoRec. 8603 8604 Fix this and thus allow the ATI DDX to deal with as many CRTCs 8605 as the DRM allows (32), so it is ready if anything with >6 CRTCs 8606 comes out. 8607 8608 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 8609 8610 commit 4853ab2cdc3b97948c7cd69eaf4fff54f59774fc 8611 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8612 Date: Fri Nov 4 12:15:53 2011 +0100 8613 8614 Turn compile time check into runtime check. 8615 8616 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8617 8618 commit bcdb54fe16ebf2e239b84eebf20e8adfe5094bff 8619 Author: Alex Deucher <alexdeucher (a] gmail.com> 8620 Date: Thu Oct 20 11:11:35 2011 -0400 8621 8622 check for xserver 1.9.4.901 to enable tiling by default 8623 8624 Previous xservers had a bug in the EXA code which caused 8625 display corruption in some cases. 8626 8627 See: 8628 https://bugs.freedesktop.org/show_bug.cgi?id=33929 8629 8630 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8631 8632 commit edde94cba5321e6e51e7fd4d79dde5abc4944495 8633 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8634 Date: Wed Nov 2 13:40:05 2011 +0100 8635 8636 Bump version post release. 8637 8638 commit 93459f842c2d8dc178a1954b8e05150fcb96ac9a 8639 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8640 Date: Wed Nov 2 12:51:15 2011 +0100 8641 8642 Bump version for 6.14.3 release. 8643 8644 commit fe3ce559b7f07d39ded39abb38576846d6eb185b 8645 Author: Jeremy Huddleston <jeremyhu (a] apple.com> 8646 Date: Tue Nov 1 20:27:03 2011 -0700 8647 8648 Build fix for -Werror=int-to-pointer-cast 8649 8650 Signed-off-by: Jeremy Huddleston <jeremyhu (a] apple.com> 8651 8652 commit d3d6a5da07aaec5961e51c9a8f90c1490ee101b6 8653 Author: Tormod Volden <debian.tormod (a] gmail.com> 8654 Date: Sat Oct 29 23:08:46 2011 +0200 8655 8656 radeon: do not include xf86PciInfo.h 8657 8658 We already use atipciids.h instead most places. 8659 8660 Signed-off-by: Tormod Volden <debian.tormod (a] gmail.com> 8661 8662 commit 70da7001e81363ed6ef2c4727c512daf53ae29fe 8663 Author: Jeremy Huddleston <jeremyhu (a] apple.com> 8664 Date: Sat Oct 29 20:15:09 2011 -0700 8665 8666 Use malloc/calloc/realloc/free directly 8667 8668 Signed-off-by: Jeremy Huddleston <jeremyhu (a] apple.com> 8669 8670 commit a3bb07efb1757c33d70e2e1928219d12a4dd6498 8671 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8672 Date: Tue Oct 25 17:43:58 2011 +0200 8673 8674 EXA >= R6xx / KMS: Avoid running out of CS space at inconvenient times. 8675 8676 Otherwise we may end up with things not properly set up at the beginning of the 8677 next CS. 8678 8679 Fixes http://bugs.debian.org/645007 . 8680 8681 In contrast to the Composite code for < R6xx, this isn't necessary with UMS, 8682 as the draw packet only uses constant space in the indirect buffer, and nothing 8683 else can mess with the 3D state between indirect buffers. 8684 8685 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8686 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 8687 8688 commit 23788c4a5e3b6affb9b183e1393edd0e5ca4550e 8689 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8690 Date: Thu Oct 27 19:12:22 2011 +0200 8691 8692 EXA < R6xx: Make sure 2D state is re-emitted after running out of CS space. 8693 8694 Otherwise it's basically luck what the 2D state ends up being at the beginning 8695 of the next CS. 8696 8697 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8698 8699 commit 6e0e1a821accc6ca95f4134e49b66a6b168c1934 8700 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8701 Date: Tue Oct 25 17:39:57 2011 +0200 8702 8703 Make radeon_dri2_create_buffer(s) more robust. (Bug #30047) 8704 8705 In particular, handle and propagate failure to allocate GPU accessible memory, 8706 instead of crashing. Fixes https://bugs.freedesktop.org/show_bug.cgi?id=30047 . 8707 8708 Also take care not to leak resources in error paths. 8709 8710 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8711 8712 commit 856583dbca9319c77fed40daa9956e81a0068f9e 8713 Author: Jerome Glisse <jglisse (a] redhat.com> 8714 Date: Thu Oct 20 14:17:14 2011 -0400 8715 8716 radeon/kms: fallback to vesa if GPU is not supported by UMS 8717 8718 For GPU not supported by UMS, test in probe so that we properly 8719 fallback to vesa. 8720 8721 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 8722 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 8723 8724 commit f6f1b4f7789cfef763c063e671b76b60be8bdd8e 8725 Author: Maarten Lankhorst <m.b.lankhorst (a] gmail.com> 8726 Date: Tue Oct 18 14:30:39 2011 +0200 8727 8728 dri2: Add vdpau driver name entry 8729 8730 libvdpau has a driver loading mechanism that looks for a dri2 driver 8731 first before falling back to nvidia, so lets use that. 8732 8733 Allows use of libvdpau_rx00 without having to set things up separately, 8734 similar to the patch to xf86-video-nouveau. 8735 8736 Signed-off-by: Maarten Lankhorst <m.b.lankhorst (a] gmail.com> 8737 Reviewed-by: Christian Knig <deathsimple (a] vodafone.de> 8738 Reviewed-by: Michel Dnzer <michel (a] daenzer.net> 8739 Tested-by: Michel Dnzer <michel (a] daenzer.net> 8740 8741 commit a6b2bd2d184f10d4c56c4ee17186aedb238a36ec 8742 Author: Alex Deucher <alexdeucher (a] gmail.com> 8743 Date: Wed Sep 21 17:00:16 2011 -0400 8744 8745 UMS: fix DDIA enable on some rs690 systems 8746 8747 DVOOutputControl checks the value of of bios scratch reg 3 8748 on some tables and assumes the encoder is already enabled 8749 if the DFP2_ACTIVE bit is set. Clear that bit so the table 8750 sets the DDIA enable bit properly. 8751 8752 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8753 8754 commit d78860ba53d9bfcf6c28e1cfd2d970709b5e20fa 8755 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8756 Date: Fri Aug 12 11:21:33 2011 +0200 8757 8758 Only call radeon_dri2_close_screen() if DRI2 was enabled. 8759 8760 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8761 8762 commit 11330ca5dc61a70fe4507e63230f9133ca22d891 8763 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8764 Date: Fri Aug 12 11:21:32 2011 +0200 8765 8766 Remove dead variable remain_size_bytes. 8767 8768 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8769 8770 commit fcf0cca9c0ab0f692b222f619aee8f1cdad3b519 8771 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8772 Date: Tue Sep 20 12:34:05 2011 +0200 8773 8774 KMS Color Tiling requires xserver which supports EXA_MIXED_PIXMAPS. 8775 8776 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8777 8778 commit c96e6fb8a5f5be2319fdb4c431c1ba5279997fe8 8779 Author: Alex Deucher <alexdeucher (a] gmail.com> 8780 Date: Sat Sep 17 08:26:12 2011 -0400 8781 8782 man: note that the list of marketing names is non-exhaustive 8783 8784 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8785 8786 commit 1a51fce6f6ab169c882a86b936909c0820f27a68 8787 Author: Alex Deucher <alexdeucher (a] gmail.com> 8788 Date: Thu Sep 15 17:52:25 2011 -0400 8789 8790 update man page with new marking names 8791 8792 Note that the driver support all currently 8793 shipping asics and the names in the man page 8794 are just a sampling. 8795 8796 Fixes: 8797 https://bugs.freedesktop.org/show_bug.cgi?id=40808 8798 8799 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8800 8801 commit 64f237a4cf3ef5bcb3163e88e1447ff275a1eefa 8802 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8803 Date: Thu Aug 18 19:11:08 2011 +0200 8804 8805 Convert register ranges for >= r6xx from enums to defines. 8806 8807 Avoids lots of "comparison between 'enum <anonymous>' and 'enum <anonymous>'" 8808 warnings with newer versions of gcc. See 8809 https://bugs.freedesktop.org/show_bug.cgi?id=38238 . 8810 8811 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8812 8813 commit 95991fcce45f0dd904f76b9f98f9c7ed5708e4fa 8814 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8815 Date: Wed Aug 17 11:10:34 2011 +0200 8816 8817 Bail if we're trying to start up in UMS mode on KMS. 8818 8819 Ideally, the display manager will start the X server again, and everything 8820 will be fine and dandy. But in the worst case, at least we won't hit the 8821 hardware behind the KMS driver's back. 8822 8823 (This change intentionally makes (ab)use of the fact that Bool is defined as 8824 int). 8825 8826 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8827 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 8828 8829 commit f95a41b7851565c282d22f8d679db1377428f165 8830 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8831 Date: Thu Aug 11 11:22:57 2011 +0200 8832 8833 video: Don't round up bottom/right edge for clipping source width/height. 8834 8835 It's not necessary: If the top/left edge was rounded down, this will be 8836 compensated by the subtraction. 8837 8838 Worse, if the original source width/height is odd, rounding up may result in 8839 reading past the end of the source data. 8840 8841 Fixes http://bugs.debian.org/637258 . 8842 8843 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8844 Reviewed-by: Alex Deucher <alexander.deucher (a] amd.com> 8845 8846 commit 93fc0843a1e31dc9237433bc2bf17df79e956d26 8847 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8848 Date: Wed Aug 10 17:44:37 2011 +0200 8849 8850 Change my e-mail address to something that still works, and always will, I hope. 8851 8852 commit 9151f3b1c2ebcc34e63195888ba696f2183ba5e2 8853 Author: Michel Dnzer <michel.daenzer (a] amd.com> 8854 Date: Tue Aug 9 19:13:26 2011 +0200 8855 8856 Prefer the CRTC of the primary output for synchronization. 8857 8858 See https://bugs.freedesktop.org/show_bug.cgi?id=39696 . 8859 8860 Signed-off-by: Michel Dnzer <michel.daenzer (a] amd.com> 8861 8862 commit 3b9fdc807dd7e52af0576299cefba596040f6f2f 8863 Author: Alex Deucher <alexdeucher (a] gmail.com> 8864 Date: Wed Aug 3 16:20:13 2011 -0400 8865 8866 r5xx+: Fix vline setup with crtc offsets 8867 8868 On r5xx+, vline is relative to to the viewport, not 8869 the scanlines. Based on initial patch and investigation 8870 from Herbert Ptzl (Bertl) on IRC. 8871 8872 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8873 8874 commit 104b2d7c071f29266b1bc4184a74e9714d14febc 8875 Author: Alex Deucher <alexander.deucher (a] amd.com> 8876 Date: Mon Aug 1 10:05:30 2011 -0400 8877 8878 kms: fix possible leak in pageflip code 8879 8880 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 8881 8882 commit d29bab632e9ecccba518d4107d52620bf75eb1cf 8883 Author: Ville Syrjala <syrjala (a] sci.fi> 8884 Date: Wed May 4 23:51:27 2011 +0300 8885 8886 kms: Move flip_count and co. to a per swap structure 8887 8888 If multiple drawables are doing page flipping, the global drmmode 8889 structure can't be used to keep per swap information. For example 8890 flip_count can increase prematurely due to another swap request, 8891 and then the previous swap request never gets completed, leading to a 8892 stuck client. Move the relevant pieces of data to a strucuture that 8893 gets allocated once per swap request and shared by all involved CRTCs. 8894 8895 Signed-off-by: Ville Syrjala <syrjala (a] sci.fi> 8896 8897 commit 9493563c1ef4b51af0ee8a44cb4e7c5bb280347e 8898 Author: Ville Syrjala <syrjala (a] sci.fi> 8899 Date: Wed May 4 23:51:26 2011 +0300 8900 8901 dri2: Update front buffer pixmap and name before exchanging buffers 8902 8903 Buffer exchange assumes that the front buffer pixmap and name 8904 information is accurate. That may not be the case eg. if the window 8905 has been (un)redirected since the buffer was created. 8906 8907 Signed-off-by: Ville Syrjala <syrjala (a] sci.fi> 8908 8909 commit 8c9266ed2da22a510243f9a952c14d4423f48a2b 8910 Author: Alex Deucher <alexander.deucher (a] amd.com> 8911 Date: Fri Jul 15 10:44:57 2011 -0400 8912 8913 radeon: add some new NI pci ids 8914 8915 Signed-off-by: Alex Deucher <alexander.deucher (a] amd.com> 8916 8917 commit f59c3b294b0f715fc96e2bbe25893f2b31aa488b 8918 Author: Christian Knig <deathsimple (a] vodafone.de> 8919 Date: Thu Jul 14 11:49:06 2011 +0200 8920 8921 Register XvMC video decoding acceleration 8922 8923 commit e8d0d437957b15252dfad775796a3949ed50dbcf 8924 Author: Dave Airlie <airlied (a] redhat.com> 8925 Date: Tue Jul 12 11:43:25 2011 -0400 8926 8927 evergreen: Emit SQ_LDS_RESOURCE_MGMT 8928 8929 Avoids rendering problems when compute changes this reg. 8930 8931 Fixes: 8932 https://bugs.freedesktop.org/show_bug.cgi?id=39119 8933 8934 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8935 8936 commit 9bb31158466e6168116d841d12c8b4303f11c4a6 8937 Author: Alex Deucher <alexdeucher (a] gmail.com> 8938 Date: Wed Jun 22 12:24:28 2011 -0400 8939 8940 evergreen: fix num_banks for 2D tiling config 8941 8942 The field is encoded. 8943 8944 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8945 8946 commit 122b471f734aa07427b01d4bec35ff1ac28290b5 8947 Author: Alex Deucher <alexdeucher (a] gmail.com> 8948 Date: Fri Jun 17 04:02:16 2011 -0400 8949 8950 dri2: fix copy pasto in a6154c00c64932332e8f6e334661ffd579cfd894 8951 8952 Reported-by: Nils Wallmnius <nils.wallmenius (a] gmail.com> 8953 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8954 8955 commit 122bedcbcf45cb583cf51b2fd04ed2805e0ca60b 8956 Author: Alex Deucher <alexdeucher (a] gmail.com> 8957 Date: Thu Jun 16 12:57:11 2011 -0400 8958 8959 dri2: missing bit from a6154c00c64932332e8f6e334661ffd579cfd894 8960 8961 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8962 8963 commit a6154c00c64932332e8f6e334661ffd579cfd894 8964 Author: Alex Deucher <alexdeucher (a] gmail.com> 8965 Date: Thu Jun 16 12:48:43 2011 -0400 8966 8967 dri2/eg+: fix size and alignment of depth/stencil buffers 8968 8969 Base alignment may be 256B or 512B depending on the group 8970 size. Also need to check against front size for virtualX. 8971 8972 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8973 8974 commit 41eb1fbb3d9da64feb4a96df7c575e44136b1538 8975 Author: Alex Deucher <alexdeucher (a] gmail.com> 8976 Date: Mon Jun 13 12:50:39 2011 -0400 8977 8978 kms/man: update ColorTiling info 8979 8980 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8981 8982 commit 481baa5c74271cd7ce38bae3965d2bc4b8809058 8983 Author: Alex Deucher <alexdeucher (a] gmail.com> 8984 Date: Mon Jun 13 12:44:07 2011 -0400 8985 8986 kms: enable ColorTiling by default on r6xx-cayman asics 8987 8988 Currently only 1D tiling as 2D tiling still has some corner 8989 cases to fix up. 8990 8991 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 8992 8993 commit cbcc57b0fa6f581be777bef648f2bf3efe7443ee 8994 Author: Alex Deucher <alexdeucher (a] gmail.com> 8995 Date: Mon Apr 4 12:52:00 2011 -0400 8996 8997 radeon: add llano pci ids 8998 8999 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9000 9001 commit 36afd1e1055eeadb2396dadcc68b214655bd90a9 9002 Author: Alex Deucher <alexdeucher (a] gmail.com> 9003 Date: Tue May 31 16:03:36 2011 -0400 9004 9005 radeon: add support for llano APUs 9006 9007 - KMS only 9008 - Includes full EXA/Xv support 9009 9010 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9011 9012 commit 557f46dc2f18734ecf1f18dee7e951e0bf062e63 9013 Author: Dave Airlie <airlied (a] redhat.com> 9014 Date: Fri May 27 07:22:08 2011 +1000 9015 9016 bump version after release 9017 9018 commit 2fca40ea65d9f2a6f8451c324bb4b82786f34f76 9019 Author: Dave Airlie <airlied (a] redhat.com> 9020 Date: Thu May 26 12:52:21 2011 +1000 9021 9022 radeon: bump version for release of 6.14.2 9023 9024 commit fab868c5f7d8cafdb0176d2751f216819a5ba66a 9025 Author: Alex Deucher <alexdeucher (a] gmail.com> 9026 Date: Wed May 25 01:09:12 2011 -0400 9027 9028 EXA: make evergreen_fix_scissor_coordinates static 9029 9030 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9031 9032 commit fe5c42f5155361006b687da824181418f688809f 9033 Author: Alex Deucher <alexdeucher (a] gmail.com> 9034 Date: Tue May 24 22:32:01 2011 -0400 9035 9036 cayman: endian fixes for shaders 9037 9038 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9039 9040 commit 470ecd02347c32e79316046d01a7d5dad0e2fe99 9041 Author: Alex Deucher <alexdeucher (a] gmail.com> 9042 Date: Tue May 24 18:31:47 2011 -0400 9043 9044 EXA/Xv: add workarounds for eg/cayman scissors bugs 9045 9046 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9047 9048 commit b913e7ba2b60d47a6660699210cc3cf6f5dc52c2 9049 Author: Alex Deucher <alexdeucher (a] gmail.com> 9050 Date: Tue May 24 18:02:52 2011 -0400 9051 9052 Revert "cayman: add scissors workaround." 9053 9054 Needs a proper workaround for a hw bug. 9055 9056 This reverts commit b77d374b0d11f48c33cfffdb4157c4ec4b05ea15. 9057 9058 commit 82cb33c3f0e1ba802d7a94f3159b3c5c86cd4043 9059 Author: Dave Airlie <airlied (a] redhat.com> 9060 Date: Wed May 18 14:49:17 2011 +1000 9061 9062 cayman: enable all accel 9063 9064 commit b77d374b0d11f48c33cfffdb4157c4ec4b05ea15 9065 Author: Dave Airlie <airlied (a] redhat.com> 9066 Date: Tue May 24 15:53:58 2011 +1000 9067 9068 cayman: add scissors workaround. 9069 9070 wasted a lot of time getting to this. 9071 9072 commit ffeab7a7058298e15294a3b2c740c731e36dda1d 9073 Author: Alex Deucher <alexdeucher (a] gmail.com> 9074 Date: Mon Apr 18 18:16:51 2011 -0400 9075 9076 cayman: fix dword counts default state 9077 9078 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9079 9080 commit 3cbfae361bf5e779d3364f0f31cfd25bd0f59e65 9081 Author: Alex Deucher <alexdeucher (a] gmail.com> 9082 Date: Wed Mar 2 20:48:19 2011 -0500 9083 9084 cayman: add spi state to default state 9085 9086 changed in e3145801b80fd4be4cf770128876e86e89bda66f 9087 9088 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9089 9090 commit b8ade97c9d0fa5aacb0e3166868bb72e9bc679a6 9091 Author: Alex Deucher <alexdeucher (a] gmail.com> 9092 Date: Wed Mar 2 20:44:19 2011 -0500 9093 9094 cayman: first pass at exa/Xv shaders 9095 9096 Main differences with evergreen: 9097 - 4-way rather than 5-way 9098 - END_OF_PROGRAM bit removed from CF istructions, use 9099 CF_INST_END instead. 9100 - MEGA_FETCH* fields removed from VTX commands 9101 - no more VC, all fetches go through the TC 9102 9103 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9104 9105 commit 01b646ed800732985c1638b147716641a99082f9 9106 Author: Alex Deucher <alexdeucher (a] gmail.com> 9107 Date: Wed Mar 2 20:39:38 2011 -0500 9108 9109 cayman: add a default state function 9110 9111 The rest of the state functions can be shared 9112 with evergreen. I've noted where there are 9113 differences. 9114 9115 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9116 9117 commit 42eecc6f4fb1570769490bdaeac06817c6c36a7e 9118 Author: Alex Deucher <alexdeucher (a] gmail.com> 9119 Date: Wed Mar 2 20:20:56 2011 -0500 9120 9121 cayman: add 3D register headers 9122 9123 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9124 9125 commit e1d28e011f4a5139cbc778973c63158ed2746716 9126 Author: Alex Deucher <alexdeucher (a] gmail.com> 9127 Date: Wed Mar 2 20:13:50 2011 -0500 9128 9129 kms/cayman: stub out exa support 9130 9131 Just fallbacks for now. 9132 9133 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9134 9135 commit 21e44a20b8b1b64079ee77f45aaa5010206ed7b6 9136 Author: Michel Dnzer <daenzer (a] vmware.com> 9137 Date: Thu May 12 09:23:38 2011 +0200 9138 9139 UMS: Fix comparison of unsigned variable against < 0. 9140 9141 Pointed out by clang: 9142 9143 ../../src/radeon_crtc.c:242:18: error: comparison of unsigned expression < 0 is always false [-Werror,-Wtautological-compare] 9144 error = error < 0 ? 0xffffffff : error; 9145 ~~~~~ ^ ~ 9146 9147 If a UMS regression is bisected to this commit, the assignment should probably 9148 just be removed, as it's a no-op in the current form. 9149 9150 commit 3b893d81982c9381393c92625e308541e0071b05 9151 Author: Michel Dnzer <daenzer (a] vmware.com> 9152 Date: Fri May 13 09:10:02 2011 +0200 9153 9154 KMS: Fix output properties logic error. 9155 9156 Pointed out by clang: 9157 9158 ../../src/drmmode_display.c:1023:30: error: use of logical && with constant operand; switch to bitwise & or remove constant [-Werror,-Wconstant-logical-operand] 9159 if (props && (props->flags && DRM_MODE_PROP_ENUM)) { 9160 ^ ~~~~~~~~~~~~~~~~~~ 9161 9162 Reviewed-by: Alex Deucher <alexdeucher (a] gmail.com> 9163 9164 commit f83d58cf5b33686139067f8f898b8e566ba5c253 9165 Author: Nicolas Kaiser <nikai (a] nikai.net> 9166 Date: Fri May 13 00:56:31 2011 +0200 9167 9168 man: fix typos 9169 9170 Signed-off-by: Nicolas Kaiser <nikai (a] nikai.net> 9171 9172 commit 90abffbd30f44b9cf76a6e28103ddcb5419b4522 9173 Author: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 9174 Date: Fri May 6 09:45:23 2011 -0400 9175 9176 DRI2: fix high-crtc/vblank oversight/bug 9177 9178 improvements to high-crtc handling done in 9179 f0b7d7b449cc77bb2b281d81108507f8bc2e6018 introduced a bug that caused 9180 the populate_vbl_request_type to never use the high-crtc field even 9181 when it should. The reason is that the offending patch put the code 9182 under #ifdef DRM_VBLANK_HIGH_CRTC_MASK which is not visible outside the 9183 enum type, so #else was always taken in compilation type. This patch 9184 fixes it by basing #ifdef on (pre-processor visible) 9185 DRM_VBLANK_HIGH_CRTC_SHIFT constant 9186 9187 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 9188 9189 commit 62a4cd180fe884dca24586d453395472516e6496 9190 Author: Alex Deucher <alexdeucher (a] gmail.com> 9191 Date: Wed May 4 01:13:55 2011 -0400 9192 9193 fusion: fix tiling enable logic 9194 9195 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9196 9197 commit 76638ca687b02d3b1494b9868f817fd4fd892c64 9198 Author: Alex Deucher <alexdeucher (a] gmail.com> 9199 Date: Wed May 4 01:06:22 2011 -0400 9200 9201 fusion: enable tiling if DFS works 9202 9203 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9204 9205 commit a6d2dba6573a3512d550d7e442bf42ea03012bbc 9206 Author: Dave Airlie <airlied (a] redhat.com> 9207 Date: Wed May 4 10:44:43 2011 +1000 9208 9209 radeon: add add hw DFS support for fusion 9210 9211 Fusion had a bug setting up the VM on earlier kernels so we need to work 9212 around that and only enable accel on a new enough kernel. 9213 9214 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 9215 9216 commit 859e052af49e68a826b77a9135c7f067dc331a06 9217 Author: Alex Deucher <alexdeucher (a] gmail.com> 9218 Date: Tue May 3 15:15:04 2011 -0400 9219 9220 radeon: add some new pci ids 9221 9222 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9223 9224 commit 8f8bbf628c6eed037f57bc8c155f0ecdacbebad1 9225 Author: Alex Deucher <alexdeucher (a] gmail.com> 9226 Date: Sun May 1 13:19:15 2011 -0400 9227 9228 man: add cayman to man page 9229 9230 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9231 9232 commit 903e90c31cf0319be9297529aa7b8daa1756cf63 9233 Author: Alex Deucher <alexdeucher (a] gmail.com> 9234 Date: Wed Apr 20 03:10:08 2011 -0400 9235 9236 EXA/Xv: used cached bo tiling flags for accel setup on 6xx+ 9237 9238 This avoids calling into the kernel for each bo in the accel 9239 code. This is a follow on to: 9240 cc7d1fa39da40a532fcdbe6c7924ca47a879e66a 9241 9242 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9243 9244 commit 982c22f16c8eeee9be81779fbfe17d8d3f9b6897 9245 Author: Adam Jackson <ajax (a] redhat.com> 9246 Date: Thu Apr 14 16:04:50 2011 -0400 9247 9248 R520: Fix textures larger than 2k 9249 9250 Ported from the equivalent fix in Mesa. 9251 9252 commit cc7d1fa39da40a532fcdbe6c7924ca47a879e66a 9253 Author: Michel Dnzer <daenzer (a] vmware.com> 9254 Date: Mon Apr 4 17:37:12 2011 +0200 9255 9256 EXA: Cache BO tiling flags. 9257 9258 Calling into the kernel every time is quite expensive, and nobody else should 9259 ever change the tiling flags. 9260 9261 There's still more to do along the same lines for >= R6xx. 9262 9263 commit f0b7d7b449cc77bb2b281d81108507f8bc2e6018 9264 Author: Michel Dnzer <daenzer (a] vmware.com> 9265 Date: Tue Apr 5 13:36:01 2011 +0200 9266 9267 DRI2: Some cleanups for the scheduling mess. 9268 9269 * Fix build against libdrm that doesn't define *_VBLANK_HIGH_CRTC*. 9270 * If we have more than two CRTCs but can't use DRM_VBLANK_HIGH_CRTC_MASK, don't 9271 enable scheduling in the first place rather than relying on 9272 DRM_VBLANK_SECONDARY magically doing something sensible for higher CRTCs. 9273 * Only set up client state tracking when scheduling is enabled. 9274 * Only declare pRADEONEnt when it's needed, and break long lines. 9275 9276 commit 7acf9bc833de539fa2259a051c66a99445a54bc4 9277 Author: Alex Deucher <alexdeucher (a] gmail.com> 9278 Date: Mon Apr 4 11:08:37 2011 -0400 9279 9280 radeon: add some additional ontario pci ids 9281 9282 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9283 9284 commit 0af6386f85a0f1ba14864a3334164733a10a6cb8 9285 Author: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 9286 Date: Thu Mar 24 13:33:27 2011 -0400 9287 9288 xf86-video-ati: (revised #2) add support for vblank on crtc > 1 9289 9290 Hi Alex, 9291 9292 Enclosed is a revised version of two patches sent on Mar 18 and Mar 22, 9293 respectively. Details summarized in these two threads: 9294 http://lists.freedesktop.org/archives/dri-devel/2011-March/009463.html 9295 http://lists.freedesktop.org/archives/dri-devel/2011-March/009582.html 9296 9297 This patch reconciles the DDX with the change in libdrm sent to this list 9298 earlier today. Specifically, it refers to a symbol that has been renamed 9299 from DRM_CAP_HIGH_CRTC to DRM_CAP_VBLANK_HIGH_CRTC. It *supersedes* the 9300 previous patch (i.e. apply it to the master branch as it exists at the 9301 time of this writing, not as an incremental patch to the one sent previously). 9302 9303 Regards, 9304 9305 Ilija 9306 9307 Signed-off-by: Ilija Hadzic <ihadzic (a] research.bell-labs.com> 9308 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9309 9310 commit fe2e0ad3ffa58f40311319c950b842e2928a5740 9311 Author: matthew green <mrg (a] eterna.com.au> 9312 Date: Mon Mar 21 12:17:58 2011 -0400 9313 9314 bug fix for r6xx/r7xx UMS 9315 9316 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9317 9318 commit 83978ad5fe37581e2b2f3fbd9c073d91b2ae1d50 9319 Author: Alex Deucher <alexdeucher (a] gmail.com> 9320 Date: Thu Mar 17 19:07:43 2011 -0400 9321 9322 bump version post release 9323 9324 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9325 9326 commit 38d9368e59b2990bf32a028ece2132451b402350 9327 Author: Alex Deucher <alexdeucher (a] gmail.com> 9328 Date: Thu Mar 17 18:59:55 2011 -0400 9329 9330 bump for release 9331 9332 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9333 9334 commit cdfc007eccc9b130fc08012deef304488eb6e469 9335 Author: Dave Airlie <airlied (a] redhat.com> 9336 Date: Wed Mar 16 10:40:42 2011 +1000 9337 9338 radeon/exa: correct function name 9339 9340 this corrects the function name so it matches the contents. 9341 9342 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 9343 9344 commit b93d7658f669f6dc1cfacebcfe955a1e113a537c 9345 Author: Dave Airlie <airlied (a] redhat.com> 9346 Date: Wed Mar 16 10:37:22 2011 +1000 9347 9348 radeon/exa: fix scaling check 9349 9350 This caused a regression in firefox, as these numbers are xFixed 9351 values hence 1 is actually 65536. 9352 9353 Should fix: https://bugzilla.redhat.com/show_bug.cgi?id=685084 9354 9355 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 9356 9357 commit 8567b8cd9e136ef69906e02286b3752db2404741 9358 Author: Alex Deucher <alexdeucher (a] gmail.com> 9359 Date: Tue Mar 15 13:13:06 2011 -0400 9360 9361 APU: no tiling yet 9362 9363 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9364 9365 commit 4d350497012fa31a417ada662006e2d64db2a4b5 9366 Author: Dave Airlie <airlied (a] redhat.com> 9367 Date: Tue Mar 15 10:32:19 2011 +1000 9368 9369 radeon: exa shaders don't handle scaling either. 9370 9371 rendercheck tsrccoords test fails. 9372 9373 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 9374 9375 commit 6319a33cb7bd8dba1eaf47bee1e2608cee5d7600 9376 Author: Alex Deucher <alexdeucher (a] gmail.com> 9377 Date: Wed Mar 2 19:38:04 2011 -0500 9378 9379 kms: add cayman pci ids 9380 9381 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9382 9383 commit fd65def45a57e3242596274d270791df8abc8e29 9384 Author: Alex Deucher <alexdeucher (a] gmail.com> 9385 Date: Thu Jan 13 17:02:31 2011 -0500 9386 9387 kms: no accel yet for cayman 9388 9389 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9390 9391 commit 585e2a3b803b6d3e070f2427e9cb49bab0949239 9392 Author: Alex Deucher <alexdeucher (a] gmail.com> 9393 Date: Thu Jan 13 16:58:01 2011 -0500 9394 9395 kms: add cayman chip family 9396 9397 KMS only. No UMS support for cayman. 9398 9399 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9400 9401 commit 6835e22256ba7a74c02504383f63baaba7e87156 9402 Author: Alex Deucher <alexdeucher (a] gmail.com> 9403 Date: Wed Mar 2 12:20:45 2011 -0500 9404 9405 r6xx+: truncate point sampled coordinates 9406 9407 By default the hardware rounds texcoords. However, 9408 for point sampled textures, the expected behavior is 9409 to truncate. When we have point sampled textures, 9410 set the truncate bit in the sampler. 9411 9412 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9413 9414 commit 8a8862e4165dc5be779bce5e327a31da4826db3b 9415 Author: Sedat Dilek <sedat.dilek (a] googlemail.com> 9416 Date: Sat Feb 26 21:04:22 2011 +0100 9417 9418 kms: Fix warning XNFprintf is deprecated (v2) 9419 9420 The first take of my patch missed the breakage against pre-1.10 9421 xservers and now I know XNFasprintf() is not a 1:1 replacement 9422 for XNFprintf(). 9423 The idea for the fix seemed to be so obvious, but as usual "really" 9424 test the patch you are sending to any ML. 9425 I am awfully sorry for the overhasty bad hack. 9426 Thanks to Alan and Alex for the help to fix it correctly, now. 9427 9428 Tested-By: Sedat Dilek <sedat.dilek (a] gmail.com> 9429 Signed-off-by: Sedat Dilek <sedat.dilek (a] gmail.com> 9430 9431 commit 88304fe9c0ce69012d44a77a5368c35236511dac 9432 Author: Alex Deucher <alexdeucher (a] gmail.com> 9433 Date: Mon Feb 28 11:29:19 2011 -0500 9434 9435 kms: don't rotate if acceleration is not enabled 9436 9437 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9438 9439 commit 99141761e947e2153ed1a3bc485dc5716a22acb1 9440 Author: Alex Deucher <alexdeucher (a] gmail.com> 9441 Date: Mon Feb 28 11:20:44 2011 -0500 9442 9443 kms: don't enable tiling if accel is off 9444 9445 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9446 9447 commit 7928c120fb629a782846bcd5734d9fa14cbb099d 9448 Author: Dave Airlie <airlied (a] redhat.com> 9449 Date: Mon Feb 28 14:11:57 2011 +1000 9450 9451 xv: fix height alignments for U/V planes 9452 9453 The kernel CS checks showed we were incorrectly aligning the 9454 U/V allocations for when we used them as texture sources. 9455 9456 This should fix 9457 https://bugs.freedesktop.org/show_bug.cgi?id=34567 9458 9459 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 9460 9461 commit e55520c0e582f2a500b3b01c2cb78750e15256ca 9462 Author: Michel Dnzer <daenzer (a] vmware.com> 9463 Date: Sat Feb 26 15:27:59 2011 +0100 9464 9465 Revert "kms: Fix warning XNFprintf is deprecated" 9466 9467 This reverts commit cb4dc7ab66016d0c980800f8dbf71bbe993889d3. 9468 9469 Looks like this caused https://bugs.freedesktop.org/attachment.cgi?id=43852 , 9470 but even if it didn't, the following needs to be addressed first: 9471 9472 cc1: warnings being treated as errors 9473 ../../src/radeon_kms.c: In function radeon_open_drm_master: 9474 ../../src/radeon_kms.c:413: error: passing argument 1 of XNFasprintf from incompatible pointer type 9475 /usr/local/include/xorg/Xprintf.h:57: note: expected char ** but argument is of type char * 9476 ../../src/radeon_kms.c:413: error: passing argument 2 of XNFasprintf makes pointer from integer without a cast 9477 /usr/local/include/xorg/Xprintf.h:57: note: expected const char * restrict but argument is of type uint16_t 9478 ../../src/radeon_kms.c:413: error: assignment makes pointer from integer without a cast 9479 make[2]: *** [radeon_kms.lo] Error 1 9480 9481 commit cb4dc7ab66016d0c980800f8dbf71bbe993889d3 9482 Author: Sedat Dilek <sedat.dilek (a] gmail.com> 9483 Date: Fri Feb 25 17:21:52 2011 -0500 9484 9485 kms: Fix warning XNFprintf is deprecated 9486 9487 Noticed when building ati/radeon ddx against xserver 1.10-rc3. 9488 9489 [ build.log ] 9490 ../../src/radeon_kms.c:412:5: warning: 'XNFprintf' is deprecated (declared at /usr/include/xorg/os.h:273) 9491 9492 agd5f: don't break the build against pre-1.10 xservers. 9493 9494 Signed-off-by: Sedat Dilek <sedat.dilek (a] gmail.com> 9495 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9496 9497 commit ecfdb209afe2aafc378baab8c511f5df7b000270 9498 Author: Sedat Dilek <sedat.dilek (a] googlemail.com> 9499 Date: Fri Feb 25 21:48:14 2011 +0100 9500 9501 UMS: Fix build against xserver 1.10-rc3 9502 9503 This issue was introduced due to last minute backout of RandR-1.4 9504 in xserver 1.10-rc3. 9505 9506 Switch to "#ifdef RANDR_14_INTERFACE" as suggested by Keith Packard. 9507 See also <http://lists.x.org/archives/xorg-devel/2011-February/019643.html>. 9508 9509 Note: 9510 The ddx needs a rebuild as the X video driver ABI changed to version 10.0. 9511 9512 Reported-by: Alex Deucher <alexdeucher (a] gmail.com> 9513 CC: Keith Packard <keithp (a] keithp.com> 9514 Signed-off-by: Sedat Dilek <sedat.dilek (a] gmail.com> 9515 9516 commit 04293a3a84acd96ccdf8e5e865cf973325ca43a1 9517 Author: Alex Deucher <alexdeucher (a] gmail.com> 9518 Date: Thu Feb 24 02:33:41 2011 -0500 9519 9520 dri2: disable pageflipping for transformed displays 9521 9522 Based on Ben's nouveau patch. 9523 9524 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9525 9526 commit 91070cfd75d5607c4a72ace780f830f0ddb40e84 9527 Author: Alex Deucher <alexdeucher (a] gmail.com> 9528 Date: Wed Feb 23 00:55:28 2011 -0500 9529 9530 Xv: use aligned height to calculate Xv buffer size 9531 9532 Make sure it's large enough for the tiling mode so the 9533 CS checker doesn't complain. Should fix: 9534 https://bugs.freedesktop.org/show_bug.cgi?id=34567 9535 9536 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9537 9538 commit acd54a48a6680956ed903fc890204c0c910b6712 9539 Author: Mark Kettenis <kettenis (a] openbsd.org> 9540 Date: Sun Feb 20 22:51:04 2011 +0100 9541 9542 Properly restore CLOCK_CNTL_INDEX register 9543 9544 RADEONRestore() restores the CLOCK_CNTL_INDEX register before calling 9545 RADEONRestorePLLRegisters(), which modifies the CLOCK_CNTL_INDEX register 9546 to access the PLL registers. As a result we may end up with the wrong clock 9547 being selected when exiting X. This happens on platforms where the driver 9548 doesn't save and restore the VGA state. 9549 9550 Signed-off-by: Mark Kettenis <kettenis (a] openbsd.org> 9551 9552 commit d7c2ba4286ea32401c0026cb92ec167f7b5753d8 9553 Author: Alex Deucher <alexdeucher (a] gmail.com> 9554 Date: Wed Feb 16 18:30:34 2011 -0500 9555 9556 man: add btc chips to the list 9557 9558 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9559 9560 commit d0f8bffbbc26818eccdd305ef142387dc4975bba 9561 Author: Alex Deucher <alexdeucher (a] gmail.com> 9562 Date: Wed Feb 16 15:09:20 2011 -0500 9563 9564 kms: add tiling support for evergreen/NI 9565 9566 - requires 2.6.37 or newer 9567 - 1D only at the moment 9568 - disabled by default 9569 9570 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9571 9572 commit 0471d8412acd82e281a35fc4c6bb2d53b1ff5802 9573 Author: Alex Deucher <alexdeucher (a] gmail.com> 9574 Date: Wed Feb 16 15:07:35 2011 -0500 9575 9576 kms: EXA/Xv tiling fixes 9577 9578 - properly set tiling flags for temp surfaces 9579 - fix CB non_disp_tiling bits on evergreen 9580 9581 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9582 9583 commit 4d7e1498f7d9eb50e2eddabca193fc27bde24f0e 9584 Author: Alex Deucher <alexdeucher (a] gmail.com> 9585 Date: Tue Feb 15 11:24:23 2011 -0500 9586 9587 fix Xv on pre-r6xx asics 9588 9589 broken by: 9590 a9c1358099:"Xv: fix textured video alignment" 9591 9592 Reported by: 9593 Artur Skawina <art.08.09 (a] gmail.com> 9594 9595 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9596 9597 commit a9a59717d11af37a2dda5555f6a83c5b65449527 9598 Author: Alex Deucher <alexdeucher (a] gmail.com> 9599 Date: Sat Feb 12 05:34:50 2011 -0500 9600 9601 kms: use worst case base/pitch align if we don't have drm tiling info 9602 9603 To avoid CS rejection. 9604 9605 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9606 9607 commit 5f9b7fdb4ce27f8c6a64134bb4d5ebeb1c958cca 9608 Author: Alex Deucher <alexdeucher (a] gmail.com> 9609 Date: Sat Feb 12 05:19:12 2011 -0500 9610 9611 kms: always get the tiling info even if we aren't tiling 9612 9613 We need the proper group size to calculate alignment so the 9614 ddx and drm agree. 9615 9616 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9617 9618 commit 355f0f7a67ce239b7fd2f2e81bcaaa88b2a09f03 9619 Author: Alex Deucher <alexdeucher (a] gmail.com> 9620 Date: Sat Feb 12 04:52:52 2011 -0500 9621 9622 kms: fix rotate pitch align 9623 9624 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9625 9626 commit f1dc419c989addc4737aed06ec8b8acdb4d40063 9627 Author: Alex Deucher <alexdeucher (a] gmail.com> 9628 Date: Fri Feb 11 17:21:10 2011 -0500 9629 9630 kms: evergreen/ni big endian accel support 9631 9632 Based on 6xx/7xx patches from Cdric Cano. 9633 9634 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9635 9636 commit e8dc728a549323f1babe337b9d42ad504af1ca39 9637 Author: Cdric Cano <ccano (a] interfaceconcept.com> 9638 Date: Fri Feb 11 17:06:47 2011 -0500 9639 9640 ums: atombios endian fixes 9641 9642 agd5f: fix a few more cases. 9643 9644 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9645 9646 commit 48ffad957f1dbca909515ffa00629f4caa68706b 9647 Author: Cdric Cano <ccano (a] interfaceconcept.com> 9648 Date: Fri Feb 11 17:00:31 2011 -0500 9649 9650 kms: 6xx/7xx big endian accel support 9651 9652 agd5f: minor cleanups 9653 9654 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9655 9656 commit 151b22bd7c3b1002a7261538611fb2b468815c86 9657 Author: Cdric Cano <ccano (a] interfaceconcept.com> 9658 Date: Fri Feb 11 16:33:16 2011 -0500 9659 9660 kms/6xx+: endian swap cursor uploads 9661 9662 agd5f: remove kms surface for cursor since 9663 we now do the swap in the cursor load to avoid 9664 breaking pre-6xx cards with BE. 9665 9666 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9667 9668 commit 309fac041f998e20338d320aabaeda5cc1bb6348 9669 Author: Alex Deucher <alexdeucher (a] gmail.com> 9670 Date: Fri Feb 11 14:32:45 2011 -0500 9671 9672 kms/exa/xv: fix alignment checking in accel state setup 9673 9674 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9675 9676 commit 25fe3ec4c6dfcdf6ba72a032bc9cb2c84d84029a 9677 Author: Alex Deucher <alexdeucher (a] gmail.com> 9678 Date: Fri Feb 11 14:03:56 2011 -0500 9679 9680 kms/exa: UTS/DFS base alignment fixes 9681 9682 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9683 9684 commit 3d10278ce511f5dabb68ed86ee43eaaf43585983 9685 Author: Alex Deucher <alexdeucher (a] gmail.com> 9686 Date: Fri Feb 11 13:51:11 2011 -0500 9687 9688 Xv: fix textured video alignment harder 9689 9690 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9691 9692 commit a9c135809945fb372ce390f06918b9fd8e455b3f 9693 Author: Alex Deucher <alexdeucher (a] gmail.com> 9694 Date: Fri Feb 11 13:21:43 2011 -0500 9695 9696 Xv: fix textured video alignment 9697 9698 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9699 9700 commit e5bd99faa3b6629a55168386d5dfa936ee4e97ae 9701 Author: Alex Deucher <alexdeucher (a] gmail.com> 9702 Date: Fri Feb 11 12:14:55 2011 -0500 9703 9704 kms: remove RADEON_TILING_SURFACE flag on front bo 9705 9706 Should have been removed in: 9707 ef9bfb262db7004bef3704e5d914687e50d3fca4 9708 9709 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9710 9711 commit 92ffce1892d24a98d789aad57a4230cadb0c812a 9712 Author: Alex Deucher <alexdeucher (a] gmail.com> 9713 Date: Fri Feb 11 03:12:43 2011 -0500 9714 9715 EXA/r6xx+: properly account for height alignment in copy temp buffer 9716 9717 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9718 9719 commit 39104c6e8461cf49c1bb03a18858ad75a9d98b46 9720 Author: Alex Deucher <alexdeucher (a] gmail.com> 9721 Date: Thu Feb 10 14:31:43 2011 -0500 9722 9723 remove EVERGREENSetAccelState() 9724 9725 It was a duplicate of the R600 variant. 9726 9727 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9728 9729 commit 7ac3a2e0bcdadff7c7172a9f833f526b526da16b 9730 Author: Alex Deucher <alexdeucher (a] gmail.com> 9731 Date: Thu Feb 10 14:24:50 2011 -0500 9732 9733 6xx+: switch to linear aligned rather than linear general 9734 9735 linear aligned is supposedly more performant, but more 9736 importantly, linear general only works on the CB without 9737 slices. The texture blocks technically don't support 9738 linear general although, I think linear general gets 9739 upgraded to linear aligned in the hw which is why it 9740 currently works. 9741 9742 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9743 9744 commit e3145801b80fd4be4cf770128876e86e89bda66f 9745 Author: Alex Deucher <alexdeucher (a] gmail.com> 9746 Date: Thu Feb 10 14:14:55 2011 -0500 9747 9748 evergreen/NI: consolidate spi setup 9749 9750 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9751 9752 commit be67ded05621aff9c85525372fd119071d3278ec 9753 Author: Alex Deucher <alexdeucher (a] gmail.com> 9754 Date: Thu Feb 10 14:06:38 2011 -0500 9755 9756 6xx/7xx: consolidate spi setup 9757 9758 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9759 9760 commit 2c5ae1724307e0dba5d0306fe27c1e15a7390a2f 9761 Author: Alex Deucher <alexdeucher (a] gmail.com> 9762 Date: Thu Feb 10 13:47:53 2011 -0500 9763 9764 evergreen/ni: consolidate CB state handling 9765 9766 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9767 9768 commit c52d817b51e13447802fe338be2d247ffba2a669 9769 Author: Alex Deucher <alexdeucher (a] gmail.com> 9770 Date: Thu Feb 10 13:35:02 2011 -0500 9771 9772 6xx/7xx: consolidate remaining CB state 9773 9774 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9775 9776 commit 5f9bc127d302404432b631e3e774192950f57b42 9777 Author: Alex Deucher <alexdeucher (a] gmail.com> 9778 Date: Thu Feb 10 03:28:01 2011 -0500 9779 9780 kms/r6xx+: clean up pitch/height alignment in EXA UTS/DFS 9781 9782 commit a4899db96029acde6cd400fc0541693a487898e7 9783 Author: Alex Deucher <alexdeucher (a] gmail.com> 9784 Date: Sat Feb 5 14:11:42 2011 -0500 9785 9786 UMS: fix spelling in error message 9787 9788 Noticed by: Warren Block <wblock (a] wonkity.com> 9789 9790 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9791 9792 commit 84450ea9221f32296c1f786a6aa126a08bb9756f 9793 Author: Alex Deucher <alexdeucher (a] gmail.com> 9794 Date: Sat Feb 5 14:05:48 2011 -0500 9795 9796 UMS/DCE3.2: fix segfault 9797 9798 atombios_pick_dig_encoder() needs to be called 9799 with a valid crtc assigned. 9800 9801 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9802 9803 commit a27b5dbd9fba689f1814c0888d1b7459b6c6b1a1 9804 Author: Alex Deucher <alexdeucher (a] gmail.com> 9805 Date: Thu Feb 3 19:06:22 2011 -0500 9806 9807 Bump version post release 9808 9809 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9810 9811 commit 0a1a0513a61f392580bde39cca4880f2c19abc8d 9812 Author: Alex Deucher <alexdeucher (a] gmail.com> 9813 Date: Thu Feb 3 18:09:17 2011 -0500 9814 9815 bump version for release 9816 9817 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9818 9819 commit ef9bfb262db7004bef3704e5d914687e50d3fca4 9820 Author: Alex Deucher <alexdeucher (a] gmail.com> 9821 Date: Tue Feb 1 19:24:44 2011 -0500 9822 9823 kms/pre-6xx: fix pageflipping with tiling 9824 9825 Use UTS/DFS to tile/untile as appropriate for sw access on 9826 pre-6xx. 9827 9828 Fixes: 9829 https://bugs.freedesktop.org/show_bug.cgi?id=33738 9830 9831 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9832 9833 commit bb16dd77321e5a64a3cb2d2ca9982117799ac1a8 9834 Author: Alex Deucher <alexdeucher (a] gmail.com> 9835 Date: Tue Feb 1 19:20:00 2011 -0500 9836 9837 kms/r6xx+: fix tiling and pageflipping harder 9838 9839 Thanks for Michel for final fix. Fixes: 9840 https://bugs.freedesktop.org/show_bug.cgi?id=33738 9841 for r6xx+ asics. A similar approach for pre-r6xx 9842 asics is pending. 9843 9844 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9845 9846 commit 9c124f781049ef11a8b19894a29b7f62975b3011 9847 Author: Alex Deucher <alexdeucher (a] gmail.com> 9848 Date: Tue Feb 1 18:59:40 2011 -0500 9849 9850 radeon: remove device 0x4243 9851 9852 0x4243 is a PCI bridge, not a gpu. 9853 9854 See: 9855 https://bugs.freedesktop.org/show_bug.cgi?id=33815 9856 9857 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 9858 9859 commit 66eb81b62e5ae8e1d7bd44ed8a179e5ec1ca69af 9860 Author: Michel Dnzer <daenzer (a] vmware.com> 9861 Date: Mon Jan 31 12:26:20 2011 +0100 9862 9863 UMS: Slightly improve xserver version check. 9864 9865 This will ensure the xserver definitions necessary to compile the new code are 9866 there. xf86CrtcSetMode will be unresolved at runtime with a few xserver 9867 commits, but I'm afraid this is the best we can do. 9868 9869 See https://bugs.freedesktop.org/show_bug.cgi?id=33719 . 9870 9871 commit 3dc28c86eb57cac819e4ed650acfe1f7df6ef565 9872 Author: Alex Deucher <alexdeucher (a] gmail.com> 9873 Date: Thu Jan 27 13:09:26 2011 -0500 9874 9875 ums: fix build against xserver >= 1.10 9876 9877 xf86CrtcSetMode was removed in: 9878 http://cgit.freedesktop.org/xorg/xserver/commit/?id=8b35118c03590a7ad3786d3284bafb3f40fcb8cc 9879 9880 commit f58e1354b78bf6b70120bddfe1566da3b0723f72 9881 Author: Jerome Glisse <jglisse (a] redhat.com> 9882 Date: Wed Jan 26 16:13:30 2011 -0500 9883 9884 dp: fix displayport support by syncing with KMS code 9885 9886 Warning the dp clock value are divided by 10 in ddx (10 times 9887 bigger than kernel value) this is somethings very picky. 9888 9889 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 9890 9891 commit 0f90133ac61c5e1d80b6fa9256e079349940029f 9892 Author: Alex Deucher <alexdeucher (a] gmail.com> 9893 Date: Mon Jan 24 12:40:22 2011 -0500 9894 9895 vbo: warn if we add a bo with a pointer to the wait list 9896 9897 commit 42529603ecf86fcfd0a8f3495d9db9ce1ee9b4c3 9898 Author: Alex Deucher <alexdeucher (a] gmail.com> 9899 Date: Mon Jan 24 12:30:56 2011 -0500 9900 9901 vbo: always flush the cbuf bo in flush_indirect() 9902 9903 Always flush the cbuf in case we end up with a cbuf mapped in 9904 Prepare*(), but never end up issuing a draw call since the cbuf 9905 may be in use by multiple ops. 9906 9907 The CS check for the cbuf bo is no longer necessay because 9908 the cbuf bo is NULL at this point due to the radeon_vbo_put() 9909 just prior. 9910 9911 Fixes: 9912 https://bugs.freedesktop.org/show_bug.cgi?id=32188 9913 9914 commit 9b0e03e98ab739efb4031f81cc4a1a50b3d87a42 9915 Author: Alex Deucher <alexdeucher (a] gmail.com> 9916 Date: Mon Jan 24 00:13:25 2011 -0500 9917 9918 vbo: avoid extra cs revalidation in vbo alloc 9919 9920 Make sure we have a valid bo before revalidating. 9921 9922 commit fadee0409a8e13b78bbccb83dd70f590fee23d57 9923 Author: Alex Deucher <alexdeucher (a] gmail.com> 9924 Date: Fri Jan 21 17:30:28 2011 -0500 9925 9926 vbo: balance radeon_vbo_get() and radeon_vbo_put() 9927 9928 bo_ref() and bo_map() in radeon_vbo_get() 9929 bo_unmap() and bo_unref() in radeon_vbo_put() 9930 9931 rather than doing the bo_map() separately in 9932 radeon_vbo_space(). 9933 9934 commit af7d81625a8cf873e6efc881489b3eda9861bd03 9935 Author: Alex Deucher <alexdeucher (a] gmail.com> 9936 Date: Fri Jan 21 17:01:25 2011 -0500 9937 9938 evergreen: use vb_offset rather than vb_start_op for cbuf offset 9939 9940 commit 4817fac5f728b777939e2e2bee16b842c9e1367b 9941 Author: Alex Deucher <alexdeucher (a] gmail.com> 9942 Date: Fri Jan 21 15:59:32 2011 -0500 9943 9944 vbo: remove useless radeon_vb_discard() 9945 9946 We already reset vb_start_op to -1 in the 9947 UMS/KMS ib discard functions. 9948 9949 commit 0a03f03a65aad925ba2d9c76b1d3356184607bf9 9950 Author: Alex Deucher <alexdeucher (a] gmail.com> 9951 Date: Thu Jan 20 17:07:32 2011 -0500 9952 9953 evergreen/ni: fix Xv artifacts 9954 9955 Port of the 6xx/7xx fix to evergreen. Bad texture size 9956 for texture cache flush. 9957 9958 commit f41cf83750ba9a2e0797fdb21ea9104b85ce53aa 9959 Author: Marton Balint <cus (a] fazekas.hu> 9960 Date: Thu Jan 20 17:04:56 2011 -0500 9961 9962 r6xx/7xx: fix Xv artifacts 9963 9964 bad textures size for cache flushes. 9965 9966 Fixes: 9967 https://bugs.freedesktop.org/show_bug.cgi?id=22007 9968 9969 commit 57fbddfc21d8c6794f378489b764cc2a0ad4a48c 9970 Author: Michel Dnzer <daenzer (a] vmware.com> 9971 Date: Tue Jan 18 16:07:52 2011 +0100 9972 9973 Fix crash freeing KMS video memory. 9974 9975 Where's that brown paper bag? :} 9976 9977 commit 6548bb9836253c586023ffe5ad1497ddabaa50fc 9978 Author: Michel Dnzer <daenzer (a] vmware.com> 9979 Date: Tue Jan 18 10:23:41 2011 +0100 9980 9981 Fix KMS textured video leaks (bug #33193). 9982 9983 v2: Fix radeon_legacy_free_memory() argument type error pointed out by 9984 Marton Balint, refactor video memory freeing logic into helper function. 9985 9986 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=33193 . 9987 9988 commit edc3496b55577ee8509ddd9188e6f2bcdf7169a1 9989 Author: Dave Airlie <airlied (a] redhat.com> 9990 Date: Mon Jan 17 07:55:29 2011 +1000 9991 9992 radeon: add back flush in block handler. 9993 9994 this is also needed for certain things like stipple rendering. 9995 9996 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 9997 9998 commit 63d7dece3f4be8b14012b3a2bedd850831c0437b 9999 Author: Alex Deucher <alexdeucher (a] gmail.com> 10000 Date: Thu Jan 13 11:24:26 2011 -0500 10001 10002 rs880 fix typo in HD 4250 string 10003 10004 Noticed by Nigel Taylor 10005 10006 Fixes: 10007 https://bugs.freedesktop.org/show_bug.cgi?id=33057 10008 10009 commit c5b3db18d888552328e9718ea022794fc5bde352 10010 Author: Alex Deucher <alexdeucher (a] gmail.com> 10011 Date: Tue Jan 11 16:21:18 2011 -0500 10012 10013 kms: fix pitch aligment for scanout 10014 10015 Display has slightly stricter pitch alignment requirements 10016 than other blocks. Factor that in when aligning pitch. 10017 10018 Fixes: 10019 https://bugs.freedesktop.org/show_bug.cgi?id=32997 10020 10021 commit bbd7adce889359b5eb3239b73e904b3ede283e12 10022 Author: Alex Deucher <alexdeucher (a] gmail.com> 10023 Date: Tue Jan 11 15:41:03 2011 -0500 10024 10025 radeon: fix yet another pitch align 10026 10027 commit af2e6d7d2f1b3d8f8f6b0acfb2b7b0cfaff7bcdb 10028 Author: Alex Deucher <alexdeucher (a] gmail.com> 10029 Date: Tue Jan 11 14:42:17 2011 -0500 10030 10031 radeon: fix pitch align in pageflip code 10032 10033 commit e5d0a400d08da2358fac9c2ad12042f125525736 10034 Merge: 0e432dff9 e27e9b4e5 10035 Author: Alex Deucher <alexdeucher (a] gmail.com> 10036 Date: Mon Jan 10 18:16:49 2011 -0500 10037 10038 Merge branch 'kms-pflip' of git+ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 10039 10040 commit 0e432dff9e06a183acaeb20db29cbd03ff0f4b82 10041 Author: Alex Deucher <alexdeucher (a] gmail.com> 10042 Date: Thu Jan 6 20:56:45 2011 -0500 10043 10044 NI: add pci ids 10045 10046 commit 34d61785b9bc13287bd7ab8bdd8a1b99a3df4440 10047 Author: Alex Deucher <alexdeucher (a] gmail.com> 10048 Date: Tue Nov 23 20:54:57 2010 -0500 10049 10050 Add EXA/Xv acceleration support for NI chips 10051 10052 commit 97322c00faf08461b941edf97efe86d8b082a0ce 10053 Author: Alex Deucher <alexdeucher (a] gmail.com> 10054 Date: Wed Nov 24 12:18:04 2010 -0500 10055 10056 Bail on NI cards with UMS 10057 10058 NI cards require KMS. 10059 10060 commit c678b79f99238473df6ff1bedbae840950cdf88c 10061 Author: Alex Deucher <alexdeucher (a] gmail.com> 10062 Date: Tue Nov 23 20:45:25 2010 -0500 10063 10064 add NI family ids 10065 10066 commit e27e9b4e50ad42885ad2e25be897cdf29aa59712 10067 Author: Mario Kleiner <mario.kleiner (a] tuebingen.mpg.de> 10068 Date: Sun Dec 19 03:56:17 2010 +0100 10069 10070 ddx/ati: Bugfix for pageflip consistency check. 10071 10072 target_msc wasn't passed from vblank event handler 10073 to pageflip routine, due to a missing initalization. 10074 Now fixed. 10075 10076 Signed-off-by: Mario Kleiner <mario.kleiner (a] tuebingen.mpg.de> 10077 10078 commit 30c5f217f91ed1ee15b5fff0002521d98682efb7 10079 Author: Alex Deucher <alexdeucher (a] gmail.com> 10080 Date: Fri Dec 17 13:40:40 2010 -0500 10081 10082 exa/dfs/ontario: always use the CPU for DFS for now 10083 10084 There are issues with rendering to gart memory that haven't 10085 been sorted out completely yet. 10086 10087 commit e97079c1b621433b4aa3ecfdd2aa0884aac5df76 10088 Author: Alex Deucher <alexdeucher (a] gmail.com> 10089 Date: Thu Dec 16 18:49:55 2010 -0500 10090 10091 DCE4.1: fix DIGx routing 10092 10093 Works more like DCE4.0. With only two DIGx blocks 10094 the routing is hardcoded based on link. 10095 This fixes blank screen issues when changing crtc 10096 routing due to incorrect crtc to dig mapping. 10097 10098 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 10099 10100 commit 3b1fb12dec7df4e9434b2c51568ac8d997638922 10101 Author: Alex Deucher <alexdeucher (a] gmail.com> 10102 Date: Sun Dec 12 19:16:47 2010 -0500 10103 10104 man: minor formatting fixes 10105 10106 commit b1af52e522158788a9613d6f1a2309a3987ad59f 10107 Author: Alex Deucher <alexdeucher (a] gmail.com> 10108 Date: Tue Dec 7 19:17:09 2010 -0500 10109 10110 man: add rs880 info 10111 10112 commit f9bbb26dd97254b66de11bb2abd821aa293ecba5 10113 Author: Matt Turner <mattst88 (a] gmail.com> 10114 Date: Thu Dec 2 17:00:11 2010 -0500 10115 10116 Add missing call to vbeFree (bug 4417) 10117 10118 Also kill unneeded vbe.h include from radeon_bios.c. 10119 10120 Signed-off-by: Matt Turner <mattst88 (a] gmail.com> 10121 10122 commit 9fc16ff3f8806971cb495e8838af1e5f9149d34f 10123 Author: Alex Deucher <alexdeucher (a] gmail.com> 10124 Date: Fri Dec 3 02:04:28 2010 -0500 10125 10126 radeon: remove duplicated define 10127 10128 xfixedtofloat macro was defined several times 10129 10130 commit f074762fff4253a457cb48519bb33a2e2c90f8b9 10131 Author: Alex Deucher <alexdeucher (a] gmail.com> 10132 Date: Fri Dec 3 01:25:41 2010 -0500 10133 10134 radeon/exa: fallback for non-affine transforms 10135 10136 We need to implement a texture lookup with perspective 10137 divide for non-affine transforms. For now just fallback. 10138 Should fix: 10139 https://bugs.freedesktop.org/show_bug.cgi?id=31799 10140 although it appears either EXA or the xserver gets this 10141 wrong too. 10142 10143 commit 774b114baa932078f8bf91ea302850a8c54c2c07 10144 Author: Nicolas Joly <njoly (a] pasteur.fr> 10145 Date: Thu Dec 2 16:02:44 2010 -0500 10146 10147 ati: remove leftover word in ati.man 10148 10149 fixes: 10150 https://bugs.freedesktop.org/show_bug.cgi?id=28297 10151 10152 commit 69639ef377a9d6701cdef902f8a1c5e0b58cf833 10153 Author: Alex Deucher <alexdeucher (a] gmail.com> 10154 Date: Thu Dec 2 02:12:24 2010 -0500 10155 10156 radeon/kms: pageflipping man page updates 10157 10158 commit f48af8a6cfa1ac665f07b8f9712e94b77bc4f5e9 10159 Author: Mario Kleiner <mario.kleiner (a] tuebingen.mpg.de> 10160 Date: Tue Nov 23 04:41:53 2010 +0100 10161 10162 ddx/ati: Fix reporting of pageflip completion events on multi-head. 10163 10164 When a drawable is page-flipped on multiple crtc's (fullscreen 10165 drawable on mirror-mode or multi-head x-screen), only one pageflip 10166 event is finally delivered, after the last participating crtc signals 10167 flip completion, this to avoid visual corruption. 10168 10169 Old code returned vblank count and timestamps of flip completion 10170 of this last crtc, instead of the values of the "master crtc", the 10171 one that was used for initially scheduling/triggering the pagflip 10172 via vblank events. (master = radeon_dri2_drawable_crtc(drawable)) 10173 10174 This patch makes sure that the pageflip completion values of the 10175 "master" crtc are returned, otherwise client applications will 10176 get confused by the random (msc, ust) values returned by whichever 10177 crtc was the last to complete its flip. Without this, the returned 10178 values change randomly and jump forward and backward in time and 10179 count. 10180 10181 The patch also implements a consistency check on returned vblank 10182 count values of pageflip completion. Impossible values are detected, 10183 a x-warning is logged and returned (msc,ust) values are marked invalid, 10184 so clients could perform error handling. Such a warning would indicate 10185 bugs in the pageflip completion routine of future kms drivers or the 10186 ddx to aid driver debugging. 10187 10188 Signed-off-by: Mario Kleiner <mario.kleiner (a] tuebingen.mpg.de> 10189 10190 commit 122536ee0aeb1eef1a9d80d5e464dcb423dc2837 10191 Author: Mario Kleiner <mario.kleiner (a] tuebingen.mpg.de> 10192 Date: Mon Nov 22 04:11:07 2010 +0100 10193 10194 ddx/ati: Add option "SwapbuffersWait" to control vsync of DRI2 swaps. 10195 10196 A new optional kms driver option "SwapbuffersWait" is defined 10197 for xorg.conf, which defaults to "on". If "on", DRI2 bufferswaps 10198 will be synchronized to vsync, otherwise not. 10199 10200 This currently only affects copy-swaps, not pageflipped swaps. 10201 It also requires a swap_interval setting of zero by the OpenGL 10202 client. 10203 10204 Ideally, we'd provide a way for dri2 to pass the current swap 10205 interval to the ddx so we could change this dynamically. 10206 10207 Signed-off-by: Mario Kleiner <mario.kleiner (a] tuebingen.mpg.de> 10208 10209 commit 0de680730294bd623f6b3e189faa7b88a09d3a2a 10210 Author: Alex Deucher <alexdeucher (a] gmail.com> 10211 Date: Wed Nov 17 17:39:39 2010 -0500 10212 10213 radeon/kms: add pageflip support 10214 10215 requires radeon drm 2.8.0 or higher 10216 10217 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 10218 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 10219 10220 commit fccdca8db34010f566bd068c74cdef0f4a8cb7f5 10221 Author: Alex Deucher <alexdeucher (a] gmail.com> 10222 Date: Wed Nov 17 17:37:25 2010 -0500 10223 10224 radeon/kms: allow tiled front buffer on 6xx/7xx 10225 10226 Use UTS/DFS to tile/untile as appropriate for sw access. 10227 Also enables pageflipping with tiling enabled. 10228 10229 commit 035f7f3ab529ca19b853066792af8a23d08a0f53 10230 Author: Dave Airlie <airlied (a] redhat.com> 10231 Date: Wed Dec 1 10:23:14 2010 +1000 10232 10233 radeon: add some missing math.h and assert.h includes 10234 10235 commit 0f924e83d7be6fca7e024acdf973089efdf3b627 10236 Author: Dave Airlie <airlied (a] redhat.com> 10237 Date: Wed Dec 1 10:23:33 2010 +1000 10238 10239 radeon: make sure EXA is off when EXA is disabled with --disable-exa 10240 10241 commit e142e55c56d8440d7185b594c696ec9eeb699cda 10242 Author: Alex Deucher <alexdeucher (a] gmail.com> 10243 Date: Tue Nov 30 15:51:21 2010 -0500 10244 10245 evergreen: set default group_bytes to 256 10246 10247 512 seems to cause issues on certain cards. 10248 10249 Fixes: 10250 https://bugs.freedesktop.org/show_bug.cgi?id=31724 10251 10252 commit 90f831361844f1b80b3f6bb718ff5ac584d73d48 10253 Author: Alex Deucher <alexdeucher (a] gmail.com> 10254 Date: Mon Nov 29 18:09:05 2010 -0500 10255 10256 evergreen: use dot4 for transforms 10257 10258 commit 3cae361d0448b6e231c80f53d64bdbbdd74dc4cf 10259 Author: Alex Deucher <alexdeucher (a] gmail.com> 10260 Date: Mon Nov 29 17:44:47 2010 -0500 10261 10262 6xx/7xx: clean up gpr/const handling in shaders 10263 10264 commit d9bcac516f2a810acb300b29169e56a2df0b47ac 10265 Author: Alex Deucher <alexdeucher (a] gmail.com> 10266 Date: Mon Nov 29 17:23:30 2010 -0500 10267 10268 r6xx/r7xx use dot4 for transforms 10269 10270 commit 5d3f33729be0639cef17372345b2dab6127e39d9 10271 Author: Alex Deucher <alexdeucher (a] gmail.com> 10272 Date: Wed Nov 24 10:59:09 2010 -0500 10273 10274 radeon/kms: fix output enumeration 10275 10276 If a system has a mix of DVI-I, and DVI-D connectors, or a mix 10277 of HDMI-A and HDMI-B connectors, the enumeration was per 10278 connector type, but the naming was the same for multiple 10279 connectors (DVI-[0-n] for DVI-I, -D, -A, and HDMI-[0-n] 10280 for HDMI-A, and -B), you'd end up with multiple connectors 10281 with the same name. Fix this by tracking the number of 10282 connectors across similar connector types. 10283 10284 commit dc470b5382ffa27bd4a673b21fb6b978d3716ca0 10285 Author: Alex Deucher <alexdeucher (a] gmail.com> 10286 Date: Tue Nov 23 01:46:15 2010 -0500 10287 10288 radeon/man: divide driver options into UMS and KMS 10289 10290 commit 4cbb7100df4b983d77c20a77431cf2f5101bde36 10291 Author: Alex Deucher <alexdeucher (a] gmail.com> 10292 Date: Tue Nov 23 01:22:04 2010 -0500 10293 10294 update man page to include ontario 10295 10296 commit 718dfd443897d70d97fc19afddfbcfe1e39a9c59 10297 Author: Alex Deucher <alexdeucher (a] gmail.com> 10298 Date: Mon Nov 22 13:11:50 2010 -0500 10299 10300 Ontario: add workaround for small pixmap issues with DFS 10301 10302 On past asics these kind a problems were usually due to a bad 10303 backend map config. I suspect in this case, and the previous 10304 cases it may be related to using linear buffers rather than 10305 tiled. 10306 10307 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 10308 10309 commit 61c97b0812c820564c7bc9d64a998c109bcd46b3 10310 Author: Alex Deucher <alexdeucher (a] gmail.com> 10311 Date: Mon Nov 22 13:10:16 2010 -0500 10312 10313 Add EXA/Xv acceleration support for Ontario Fusion APUs 10314 10315 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 10316 10317 commit 0659f35dfbeda7dd3ff58da1deb6484fabadc332 10318 Author: Alex Deucher <alexdeucher (a] gmail.com> 10319 Date: Mon Nov 22 13:07:00 2010 -0500 10320 10321 Add Ontario fusion APU pci ids 10322 10323 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 10324 10325 commit fa0cec331e6c94cebf82270dcd6c179b15851c5e 10326 Author: Alex Deucher <alexdeucher (a] gmail.com> 10327 Date: Mon Nov 22 12:56:45 2010 -0500 10328 10329 ontario: add UMS modesetting support 10330 10331 Accel not enabled yet. 10332 10333 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 10334 10335 commit 7ad1f01573e930aa6238eaeb0e11708ae548dfce 10336 Author: Jerome Glisse <jglisse (a] redhat.com> 10337 Date: Fri Nov 19 10:27:40 2010 -0500 10338 10339 evergreen: fix typo for fetch resource offset 10340 10341 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 10342 10343 commit 3455a3b58532ea3ad901a317126968ab6cbb21b7 10344 Author: Alex Deucher <alexdeucher (a] gmail.com> 10345 Date: Wed Nov 17 17:32:41 2010 -0500 10346 10347 radeon/kms: fix buffer base alignment for tiling 10348 10349 On r6xx+, 2D tiling can require larger than 4k base alignment. 10350 10351 commit beb7fecd0191e38fb238134ba612985062cf9770 10352 Author: Dave Airlie <airlied (a] redhat.com> 10353 Date: Mon Nov 15 11:59:58 2010 +1000 10354 10355 dce4: improve the VT switching code. 10356 10357 add in lots more blocks of regs to save/restore 10358 10359 commit 51e51f864a691e51b5193d102398ced667e0beb1 10360 Author: Dave Airlie <airlied (a] redhat.com> 10361 Date: Thu Nov 11 18:48:40 2010 +1000 10362 10363 evergreen: remove warnings since last commit 10364 10365 commit abbf73ee990512ac16ca77e8bb23288495e1f9f4 10366 Author: Dave Airlie <airlied (a] redhat.com> 10367 Date: Thu Nov 11 09:38:15 2010 +1000 10368 10369 evergreen: add UMS VT switch support. 10370 10371 This isn't perfect, but it brings back text VTs here on the 10372 DAC and DVI outputs. 10373 10374 commit e1dfaf93d06bc5eafdbc2e1823d19204ce8f242c 10375 Author: Dave Airlie <airlied (a] redhat.com> 10376 Date: Thu Nov 11 09:15:27 2010 +1000 10377 10378 avivo: use arrays to store the crtc/pll info for save/restore 10379 10380 this is just prep work for evergreen VT save/restore 10381 10382 commit 0c2834e67df6e143a7d15f373faaddca1dda6b18 10383 Author: Michel Dnzer <daenzer (a] vmware.com> 10384 Date: Fri Nov 5 09:30:40 2010 +0100 10385 10386 Fix incorrect CS size in one textured video path. (Bug #31364) 10387 10388 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=31364 . 10389 10390 commit 2a7b080727f61559cf8444c93a5088f4155a8fdf 10391 Author: Alex Deucher <alexdeucher (a] gmail.com> 10392 Date: Fri Oct 29 16:56:45 2010 -0400 10393 10394 radeon/kms: more alignment fixes 10395 10396 commit bf60af579382a0d48f7a65a4cec88759cc8b683c 10397 Author: Dave Airlie <airlied (a] redhat.com> 10398 Date: Thu Oct 28 13:06:54 2010 +1000 10399 10400 dri2: reference count the client privates key/callback 10401 10402 This lets multi-screen work better, but still having issues after server 10403 recycle, but it doesn't crash at least. 10404 10405 commit fb22d0c06a7dc42216230e198ff443d8035e9d21 10406 Author: Dave Airlie <airlied (a] redhat.com> 10407 Date: Thu Oct 28 11:10:03 2010 +1000 10408 10409 radeon: only register block handler once per fd. 10410 10411 should fix https://bugs.freedesktop.org/show_bug.cgi?id=29726 10412 10413 the problem is of course the second head instance tries to access the 10414 fd and fails, however I think this might break syncing on the second 10415 head but not sure, but its better than just hanging up the X server 10416 10417 commit 040140b5d7cbeaf5d83687bc87e4a236ff9f0cbd 10418 Author: Alex Deucher <alexdeucher (a] gmail.com> 10419 Date: Wed Oct 27 20:16:17 2010 -0400 10420 10421 drm/radeon: unify fb bo alignment harder 10422 10423 More duplicated paths discoved... 10424 10425 commit f07f9b7b61c05f2de1d61bb0e2f71bd017c8d36a 10426 Author: Alex Deucher <alexdeucher (a] gmail.com> 10427 Date: Wed Oct 27 12:37:42 2010 -0400 10428 10429 kms/radeon: unify fb bo alignment handling 10430 10431 Previously there were 3 different paths with what should 10432 have had duplicated code: 10433 - EXACreatePixmap2 10434 - Initial front buffer creation 10435 - Randr resize 10436 10437 This patch attempts to unify the alignment across all 3. 10438 10439 This may fix tiling issues in some cases and should make 10440 buffer pitches match for pageflipping. 10441 10442 commit d31046ba6c8eee9b7decc3875697d37c38bc38f3 10443 Author: Michel Dnzer <daenzer (a] vmware.com> 10444 Date: Wed Oct 27 12:12:04 2010 +0200 10445 10446 Re-use result of GetScreenPixmap call. 10447 10448 Fixes compile warning due to local variable ppix being unused when building 10449 against current xserver Git. 10450 10451 commit 4dd7f835c17b00707f1d8d4e36a24380ba52761e 10452 Author: Michel Dnzer <daenzer (a] vmware.com> 10453 Date: Wed Oct 27 12:09:03 2010 +0200 10454 10455 Fix another stray xfree() call. 10456 10457 commit f815e9c3c61b1ed73e7dd9383587efcdd8dfc07f 10458 Author: Michel Dnzer <daenzer (a] vmware.com> 10459 Date: Wed Oct 27 11:29:12 2010 +0200 10460 10461 Don't try to delete DRI2 event list entries that were never added. (Bug #31086) 10462 10463 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=31086 . 10464 10465 commit df92245e1fd0ead2581b675cd96ac8bcc8cb65d7 10466 Author: Alex Deucher <alexdeucher (a] gmail.com> 10467 Date: Sun Oct 24 14:27:27 2010 -0400 10468 10469 r100: add some missing bits in SE_COORD_FMT 10470 10471 Texture coordinates work fine with or without these, 10472 but this should be more correct I think although 10473 I don't think it matters since we aren't sending w 10474 anyway. 10475 10476 commit fc7aa6e3bbb86eaa2f34b827629f7e82cde65ac7 10477 Author: Alex Deucher <alexdeucher (a] gmail.com> 10478 Date: Sun Oct 24 11:56:48 2010 -0400 10479 10480 UMS/radeon: fix typo in restore palette 10481 10482 commit 5efdf514ab0e5439114e8dd7a49105812155eb69 10483 Author: Alex Deucher <alexdeucher (a] gmail.com> 10484 Date: Sun Oct 24 11:45:19 2010 -0400 10485 10486 UMS/radeon: fix screen dimming on VT switch 10487 10488 Save and restore the palettes on VT switch. The restore 10489 has to be done after the vga restore to work properly as 10490 determined by Jonathan Kollasch. 10491 10492 Fixes: 10493 https://bugs.freedesktop.org/show_bug.cgi?id=18407 10494 10495 commit 1e3ede62c39307d0210cb2044badd619a4f44fec 10496 Author: Alex Deucher <alexdeucher (a] gmail.com> 10497 Date: Sun Oct 24 11:24:07 2010 -0400 10498 10499 UMS/atom: default to DVI mode if we don't have monitor hdmi info 10500 10501 Fixes: 10502 https://bugs.freedesktop.org/show_bug.cgi?id=30330 10503 10504 commit 51f26a7f01b68e373867c03881868ca8830664d6 10505 Author: Mark Schreiber <mark7 (a] alumni.cmu.edu> 10506 Date: Sat Apr 10 10:55:08 2010 -0700 10507 10508 Correct spelling and grammar 10509 10510 commit 0f575984594e6977b4a7cb794dad8c6e56c850f1 10511 Author: Alex Deucher <alexdeucher (a] gmail.com> 10512 Date: Tue Oct 19 19:21:39 2010 -0400 10513 10514 Xv: limit overlay to 2047x2047 10515 10516 Fixes: 10517 https://bugs.freedesktop.org/show_bug.cgi?id=15391 10518 10519 commit 0f184630cd4760199430320215e4280438fc4ce5 10520 Author: Alex Deucher <alexdeucher (a] gmail.com> 10521 Date: Tue Oct 19 17:13:06 2010 -0400 10522 10523 radeon: fix compilation on xserver >= 1.10 10524 10525 Fixes: 10526 https://bugs.freedesktop.org/show_bug.cgi?id=30451 10527 10528 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 10529 10530 commit 179b0e15924ae2757efaece6488ae835ba6436e3 10531 Author: Dave Airlie <airlied (a] redhat.com> 10532 Date: Tue Oct 19 13:37:22 2010 +1000 10533 10534 r600: for 2d tiles we need to align to channels no banks. 10535 10536 commit 809cacbc23ab37aa7952665f9b2e37c03a34f173 10537 Author: Adam Jackson <ajax (a] redhat.com> 10538 Date: Mon Oct 18 12:32:11 2010 -0400 10539 10540 kms: Let the server know if we got more than 1 EDID block 10541 10542 Otherwise things like xf86MonitorIsHDMI() won't work right. 10543 10544 Signed-off-by: Adam Jackson <ajax (a] redhat.com> 10545 10546 commit 042037e485396981cf4a420c247104ba5d016ca0 10547 Author: Alex Deucher <alexdeucher (a] gmail.com> 10548 Date: Thu Oct 14 17:00:51 2010 -0400 10549 10550 evergreen: requires same cursor fix as avivo chips 10551 10552 commit 0f37458c253aec46d98db6661c4def711df8fc17 10553 Author: Adam Jackson <ajax (a] redhat.com> 10554 Date: Thu Oct 14 14:37:34 2010 -0400 10555 10556 Don't allow the config file to override BIOS location 10557 10558 Signed-off-by: Adam Jackson <ajax (a] redhat.com> 10559 10560 commit 2918df3a312fc600e83f84c298b9653d2eb3a8d5 10561 Author: Alex Deucher <alexdeucher (a] gmail.com> 10562 Date: Wed Oct 13 17:16:04 2010 -0400 10563 10564 evergreen: increase XV_CRTC Xv attribute range 10565 10566 Cover all 6 possible crtcs. 10567 10568 commit 0d1f9fd03d0196dda5c1ce34b2e68b007c1874f2 10569 Author: Alex Deucher <alexdeucher (a] gmail.com> 10570 Date: Sun Oct 10 12:06:22 2010 -0400 10571 10572 DCE3+: UMS modesetting fixes 10573 10574 - add mising LVTMA case statement for DCE3.0 dig encoder 10575 - some DCE4 systems have EN/DISABLE_OUTPUT actions 10576 10577 commit acf4024aa0ef192355b2bd8281507b1c2e6b7de4 10578 Author: Alex Deucher <alexdeucher (a] gmail.com> 10579 Date: Fri Oct 8 16:30:36 2010 -0400 10580 10581 ATOM DDC fixes for UMS 10582 10583 - Header size was already subtraced from table size 10584 - Only hw capable ddc pads are shared with aux 10585 10586 commit 74fd2b91477106a26a2d9fb4b11c885910996041 10587 Author: Alex Deucher <alexdeucher (a] gmail.com> 10588 Date: Thu Oct 7 17:33:00 2010 -0400 10589 10590 DCE3.2 UMS: fix duallink 10591 10592 Should fix: 10593 https://bugs.freedesktop.org/show_bug.cgi?id=30686 10594 10595 commit 8c544272f218fd8bd93a898b96afedc2498b27bb 10596 Author: Alex Deucher <alexdeucher (a] gmail.com> 10597 Date: Thu Oct 7 17:25:47 2010 -0400 10598 10599 fix the non-kms build 10600 10601 Should fix: 10602 https://bugs.freedesktop.org/show_bug.cgi?id=30685 10603 10604 commit 81360adffb2a66b9a95a38671f9227a9718c9841 10605 Author: Oldich Jedlika <oldium.pro (a] seznam.cz> 10606 Date: Mon Sep 6 21:14:52 2010 +0200 10607 10608 radeon: proper DRI2 pending events handling when client gone. (v6) 10609 10610 Properly handle asynchronous DRI2 events for disconnected clients. 10611 Track client's pending requests and mark them as invalid when the 10612 client disconnects. 10613 10614 This is based on the version from Alban Browaeys in bug #29065. 10615 10616 v1 (Alban Browaeys): Based upon a detailed explanation from Oldich 10617 Jedlika and comments from Christopher James Halse Rogers. 10618 on http://lists.x.org/archives/xorg-driver-ati/2010-August/016780.html . 10619 10620 v2: Updated version to apply on master. Removed unnecessary 10621 client_index field from _DRI2FrameEvent. Added freeing/removing from 10622 list to failed paths of radeon_dri2_schedule_wait_msc and 10623 radeon_dri2_schedule_swap. 10624 10625 v3: Adopt to older xorg-server that doesn't have dixRegisterPrivateKey. 10626 10627 v4: Conditional include of list.h, unreachable return removed. 10628 10629 v5: Distribute list.h as xorg_list.h, remove xorg-server version check. 10630 Use the version from xorg-server when available (checked in 10631 configure.ac). 10632 10633 v6: Removed xorg_list.h, made DRI2 scheduling features dependent on 10634 list.h presence. 10635 10636 commit adee138f007e26307f1aab7f8fe066150c12e55d 10637 Author: Jesse Adkins <jesserayadkins (a] gmail.com> 10638 Date: Tue Sep 28 13:29:50 2010 -0700 10639 10640 Purge cvs tags. 10641 10642 Signed-off-by: Jesse Adkins <jesserayadkins (a] gmail.com> 10643 Signed-off-by: Alan Coopersmith <alan.coopersmith (a] oracle.com> 10644 10645 commit f1bf9cbb471d60ee4b0422bc4e896e3b86f4d75f 10646 Author: Alex Deucher <alexdeucher (a] gmail.com> 10647 Date: Wed Oct 6 11:44:42 2010 -0400 10648 10649 radeon: man page updates 10650 10651 - tv-out is not supported on carding using 10652 the rage theatre chip for tv-out 10653 - add evergreen cards to the list 10654 10655 commit 0921ecc1c751df0dd56e0b1d0b78ab53d7164904 10656 Author: Alex Deucher <alexdeucher (a] gmail.com> 10657 Date: Mon Oct 4 12:34:47 2010 -0400 10658 10659 evergreen: use EXACreatePixmap2 if available 10660 10661 commit 5bdb6434975584eef90eb9e5955b9c2a14b7f327 10662 Author: Alex Deucher <alexdeucher (a] gmail.com> 10663 Date: Mon Oct 4 12:31:51 2010 -0400 10664 10665 evergreen: port Karl's UTS/DFS changes to evergreen 10666 10667 commit be8f45cbd313b68ad663f303c64edb4525b8f981 10668 Merge: e843faf35 ff5f466e6 10669 Author: Alex Deucher <alexdeucher (a] gmail.com> 10670 Date: Mon Oct 4 12:36:56 2010 -0400 10671 10672 Merge branch 'evergreen_accel' of git+ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 10673 10674 commit ff5f466e6ca0ee541a255facff6c9198976dfd52 10675 Author: Alex Deucher <alexdeucher (a] gmail.com> 10676 Date: Fri Oct 1 15:46:43 2010 -0400 10677 10678 evergreen/dri2: fix depth allocation for depth+stencil 10679 10680 evergreen uses separate allocations for depth and stencil, 10681 so to handle that, create a depth buffer large enough to 10682 handle both. This is required for using the stencil 10683 buffer in mesa. 10684 10685 commit e843faf355c864beab81e74f0e39f8ad53d4c2bf 10686 Author: Alex Deucher <alexdeucher (a] gmail.com> 10687 Date: Thu Sep 30 19:30:35 2010 -0400 10688 10689 radeon: fix fbLocation for >32 bit MC addresses 10690 10691 If the fbLocation was at an address >32 bits, we'd fail. 10692 Change fbLocation to uint64_t and properly cast when needed. 10693 10694 commit 886febc882053e09294225e85b102f965041b62b 10695 Author: Alex Deucher <alexdeucher (a] gmail.com> 10696 Date: Thu Sep 30 19:20:17 2010 -0400 10697 10698 r6xx: fix bad mask when setting up HDP_NONSURFACE_BASE 10699 10700 This fails for MC addresses >32 bits 10701 10702 commit 16589b862feb796e15dc6e471d25604a4f5e7c1c 10703 Author: Alex Deucher <alexdeucher (a] gmail.com> 10704 Date: Mon Sep 27 18:28:43 2010 -0400 10705 10706 bump version post release 10707 10708 commit cc5005af61f45a3552f7358dc5aa711e42f5af54 10709 Author: Alex Deucher <alexdeucher (a] gmail.com> 10710 Date: Mon Sep 27 18:20:53 2010 -0400 10711 10712 bump version for release 10713 10714 commit 7f8820fcec8c90bf2f823170bd08a23e8b4ff7af 10715 Author: Michael Cree <mcree (a] orcon.net.nz> 10716 Date: Mon Sep 27 13:39:12 2010 -0400 10717 10718 Fix some unaligned 32bit accesses in the AtomBios code. 10719 10720 On the Alpha architecture unaligned 32bit accesses incur a software 10721 trap to the kernel and pollute the kernel logs. Fixed by use of the 10722 ldl_u() interface. 10723 10724 Signed-off-by: Michael Cree <mcree (a] orcon.net.nz> 10725 10726 commit c4f834cdfbe96aa47ac5fb039f9dd7aa9730c8a3 10727 Author: Nicolas Reinecke <nr (a] das-labor.org> 10728 Date: Mon Sep 27 13:33:55 2010 -0400 10729 10730 radeon: Convert remaining x(c)alloc/xfree to m/calloc/free. 10731 10732 Fixes deprecation warnings missed out by 10733 f7a91ece264af9f3fd2fc18e99aefcda93ce9f5c 10734 10735 commit f8fb9312d791af1f77020e8c2d35bb30841ed9aa 10736 Author: Karl Tomlinson <karlt+@karlt.net> 10737 Date: Sun Aug 22 22:46:33 2010 +1200 10738 10739 RADEONPrepareAccess_CS: fallback to DFS when pixmap is in VRAM 10740 10741 This avoids costly CPU VRAM reads and lets EXA manage a system memory cache 10742 of the portions of pixmaps needed for unaccelerated operations. 10743 10744 https://bugs.freedesktop.org/show_bug.cgi?id=27139 10745 10746 commit 35c4ff936601ee083f51510a5192fb97d622a483 10747 Author: Karl Tomlinson <karlt+@karlt.net> 10748 Date: Sun Aug 22 22:28:06 2010 +1200 10749 10750 radeon: complete UTS and DFS even when a scratch BO is not necessary 10751 10752 Turns on the big-endian paths even for little-endian systems, and adds 10753 similar paths to the r6xx/r7xx functions. 10754 10755 This makes UTS and DFS reliable, which will let PrepareAccess (with 10756 mixed pixmaps) choose to fail based on whether the pixmap is in VRAM 10757 (to avoid CPU reads). 10758 10759 commit d46381a3a6bf10903803f5acaa7aa0ce06373b96 10760 Author: Karl Tomlinson <karlt+@karlt.net> 10761 Date: Sun Aug 22 21:02:45 2010 +1200 10762 10763 radeon: complete big endian UTS and DFS even when scratch allocation fails. 10764 10765 On big endian systems, PrepareAccess will fail when byte-swapping is 10766 required so UploadToScreen and DownloadFromScreen cannot rely on 10767 fallback to PrepareAccess. 10768 10769 When scratch BO space allocation fails, this patch merely adds simple 10770 fallback to direct CPU access without any GPU blit. This sometimes 10771 requires a CS flush even in UploadToScreen. 10772 (No allocation retry after a flush is added here.) 10773 10774 commit 4ced4e1eff67946e306c0c67c9ed59dd5f3c4ba9 10775 Author: Karl Tomlinson <karlt+@karlt.net> 10776 Date: Sun Aug 22 20:04:42 2010 +1200 10777 10778 RADEONDownloadFromScreenCS: flush CS writes before mapping BO for read 10779 10780 If unflushed CS operations write to the pixmap BO, then these need to be 10781 flushed before mapping the BO for read. This currently only affects big 10782 endian systems and only when the operation writes to the GTT domain. 10783 10784 commit a4eef8faffbb1ea2f742273ee855f4e6f992e5c8 10785 Author: Karl Tomlinson <karlt+@karlt.net> 10786 Date: Sat Aug 21 22:29:34 2010 +1200 10787 10788 FinishAccess_CS: set bo_mapped to FALSE on unmap 10789 10790 This is actually only necessary when PrepareAccess may behave differently on 10791 different calls with the same pixmap, which currently doesn't happen. 10792 10793 However resetting bo_mapped is necessary to let PrepareAccess (with mixed 10794 pixmaps) choose to fail based on whether the pixmap is in VRAM (to avoid CPU 10795 reads). 10796 10797 commit bfebe039af0c0282d04eb6234b6e6d1e02097146 10798 Author: Karl Tomlinson <karlt+@karlt.net> 10799 Date: Sat Aug 21 21:44:39 2010 +1200 10800 10801 DownloadFromScreenCS: download via a scratch BO if pixmap domain is unknown 10802 10803 radeon_bo_is_busy() may return without setting the domain out-parameter. 10804 If this happens, then download via a scratch GTT BO to avoid CPU VRAM read. 10805 10806 commit e26a59e9db8067882327f872e3d2d760ce4c66f3 10807 Author: Mathieu Brard <mathieu (a] mberard.eu> 10808 Date: Thu Sep 9 19:32:38 2010 -0400 10809 10810 Fix NULL possible deref in evergreen_cp_wait_vline_sync 10811 10812 commit b90cb61ccb0f4f80e0627141f223354a9371d47d 10813 Author: Alex Deucher <alexdeucher (a] gmail.com> 10814 Date: Tue Sep 7 11:51:16 2010 -0400 10815 10816 radeon: set interlaced and doublescan enabled for randr outputs 10817 10818 interlaced used to work without setting these parameters. Changes 10819 in the xserver seem to require them now. 10820 10821 Should fix: 10822 https://bugs.freedesktop.org/show_bug.cgi?id=29591 10823 10824 commit 7cc0d3311f23ad569af004e0e4a0e2efbac107e5 10825 Author: Alex Deucher <alexdeucher (a] gmail.com> 10826 Date: Fri Sep 3 00:32:27 2010 -0400 10827 10828 evergreen: fix Xv 10829 10830 VS const buffer offset was wrong. 10831 10832 fixes: 10833 https://bugs.freedesktop.org/show_bug.cgi?id=29788 10834 10835 commit 7c6ce4e62693f446d7d3c8a86502ccc03c0e55b0 10836 Author: Alex Deucher <alexdeucher (a] gmail.com> 10837 Date: Thu Sep 2 20:34:56 2010 -0400 10838 10839 evergreen: add additional default state 10840 10841 Note, you also need a drm patch to fix the GPU hangs: 10842 drm/radeon/kms/evergreen: fix gpu hangs in userspace accel code 10843 10844 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 10845 10846 commit 2b98ec1f7e931019a4ab699a56d5dfaa395946fb 10847 Author: Alex Deucher <alexdeucher (a] gmail.com> 10848 Date: Wed Sep 1 13:24:19 2010 -0400 10849 10850 r6xx/r7xx: fix typos 10851 10852 some stray - signs 10853 10854 commit 966ac1be81da76c8aa4ea46b63f3ca5358a2c021 10855 Author: Alex Deucher <alexdeucher (a] gmail.com> 10856 Date: Fri Aug 27 18:22:21 2010 -0400 10857 10858 evergreen: work around bad data in some i2c tables 10859 10860 The 7th entry in a lot of evergreen i2c gpio tables is partially 10861 zeroed. Fix the entry. 10862 10863 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 10864 10865 commit 91f707d308d4bbf16c3d62d046cf280fef5a8f4c 10866 Author: Heikki Lindholm <holin (a] iki.fi> 10867 Date: Fri Aug 27 02:26:24 2010 -0400 10868 10869 xv: fix non-kms/non-dri Xv column ordering on big endian systems 10870 10871 Column order is wrong on big endian systems, primarly because of a 10872 bits / bytes mix up with the bpp variable. Fix tested with r100 and 10873 r300, screen depth 16 and 32 with YV12 and YUY2 (overlay, textured video), 10874 RGBA and RGBT (overlay). 10875 10876 Should fix: https://bugs.freedesktop.org/show_bug.cgi?id=29041 10877 10878 Signed-off-by: Heikki Lindholm <holin (a] iki.fi> 10879 10880 commit 6a2c8587a4e05a8be2a2e975a6660942cfe115d6 10881 Author: Christopher James Halse Rogers <christopher.halse.rogers (a] canonical.com> 10882 Date: Fri Aug 27 13:14:33 2010 +1000 10883 10884 dri2: Reference count DRI2 buffers 10885 10886 When a client calls ScheduleSwap we set up a kernel callback when the 10887 relevent vblank event occurs. However, it's possible for the client 10888 to go away between calling ScheduleSwap and the vblank event, 10889 resulting in the buffers being destroyed before they're passed to 10890 radeon_dri2_frame_event_handler. 10891 10892 Add reference-counting to the buffers and take a reference in 10893 radeon_dri2_schedule_swap to ensure the buffers won't be destroyed 10894 before the vblank event is dealt with. 10895 10896 This parallels the approach taken by the Intel DDX in commit 10897 0d2392d44aae95d6b571d98f7ec323cf672a687f. 10898 10899 Fixes: http://bugs.freedesktop.org/show_bug.cgi?id=29065 10900 10901 v2: Don't write completion events to the client if it has quit. 10902 v3: Don't try to unref the NULL buffers from a DRI2_WAITMSC event. 10903 Take a ref in schedule_swap earlier, so the offscreen fallback 10904 doesn't incorrectly destroy the buffers. 10905 10906 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers (a] canonical.com> 10907 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 10908 10909 commit e9928fe036e9382fd7bc353f3f05531445f08977 10910 Author: Dave Airlie <airlied (a] redhat.com> 10911 Date: Wed Aug 25 10:42:39 2010 +1000 10912 10913 remove explicit buffer submit from copy region 10914 10915 port of 0be3e95c844247746742805830860ace9f546d99 from intel driver. 10916 10917 Remove explicit batchbuffer submit in DRI2 copyregion 10918 10919 Now that we submit from the flush callback chain, we know we'll always 10920 submit before the client receives the reply or event that blocks it from 10921 rendering the next frame. 10922 10923 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 10924 10925 commit 9f13049ddf06f6f2138851a548cfb82f12a52f42 10926 Author: Dave Airlie <airlied (a] redhat.com> 10927 Date: Wed Aug 25 08:56:37 2010 +1000 10928 10929 radeon: add correct flushing for direct rendered 10930 10931 this is a port of 69d65f9184006eac790efcff78a0e425160e95aa from the Intel 10932 driver. 10933 10934 Submit batch buffers from flush callback chain 10935 10936 There are a few cases where the server will flush client output buffers 10937 but our block handler only catches the most common (before going into select 10938 If the server flushes client buffers before we submit our batch buffer, 10939 the client may receive a damage event for rendering that hasn't happened yet 10940 10941 Instead, we can hook into the flush callback chain, which the server will 10942 invoke just before flushing output. This lets us submit batch buffers 10943 before sending out events, preserving ordering. 10944 10945 Fixes 28438: [bisected] incorrect character in gnome-terminal under compiz 10946 https://bugs.freedesktop.org/show_bug.cgi?id=28438 10947 10948 Signed-off-by: Kristian Hgsberg <krh (a] bitplanet.net> 10949 10950 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 10951 10952 commit 141cbc283fddeb67c2a6c47a0f0f5c2aa2bfb430 10953 Author: Alex Deucher <alexdeucher (a] gmail.com> 10954 Date: Tue Aug 24 18:01:05 2010 -0400 10955 10956 evergreen Xv: remove stray line that caused a segfault 10957 10958 fixes: 10959 https://bugs.freedesktop.org/show_bug.cgi?id=29788 10960 10961 commit f170dddd3538a587f2363ef5fa10a4b484e762da 10962 Author: Alex Deucher <alexdeucher (a] gmail.com> 10963 Date: Mon Aug 23 18:58:12 2010 -0400 10964 10965 evergreen: use vbo pool for constant buffers 10966 10967 commit eede93b057bbdddcde5da3220a3f8c6f73784a7e 10968 Author: Alex Deucher <alexdeucher (a] gmail.com> 10969 Date: Mon Aug 23 13:16:27 2010 -0400 10970 10971 radeon: move vbo data to a separate struct 10972 10973 this way we can share the vbo code with const buffers 10974 10975 commit d8abf27dbd14f5eb746c5e8b8b1436ad292d8ec6 10976 Author: Alex Deucher <alexdeucher (a] gmail.com> 10977 Date: Mon Aug 23 01:31:27 2010 -0400 10978 10979 evergreen: fix dword counts in default state 10980 10981 commit 5a7c9d94733a0db1d3565447acc9f0e751db5950 10982 Author: Alex Deucher <alexdeucher (a] gmail.com> 10983 Date: Mon Aug 23 00:43:48 2010 -0400 10984 10985 radeon: fix legacy lvds dpms sequence 10986 10987 Take from my kms commit. 10988 10989 Should fix: 10990 https://bugs.freedesktop.org/show_bug.cgi?id=19459 10991 10992 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 10993 10994 commit 6930d2c981221757b1e11ef194809f085753a611 10995 Author: Alex Deucher <alexdeucher (a] gmail.com> 10996 Date: Fri Aug 20 16:55:21 2010 -0400 10997 10998 Add initial EXA and Xv support for evergreen 10999 11000 Based on the r6xx/r7xx code updated for evergreen. 11001 Still causes GPU hangs in some cases. We haven't 11002 tracked down why yet. Might be related to constant 11003 buffer persistence. 11004 11005 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11006 11007 commit bdd41fecdb19c83c6c7b793016b61d38065dcd13 11008 Author: Alex Deucher <alexdeucher (a] gmail.com> 11009 Date: Fri Aug 20 01:17:13 2010 -0400 11010 11011 evergreen: set encoder type to DVI for HDMI 11012 11013 Fixes: 11014 http://bugs.freedesktop.org/show_bug.cgi?id=27452 11015 11016 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11017 11018 commit ad8ea1f6e5fcb4f163622cf5eab953ea812b5829 11019 Author: Alex Deucher <alexdeucher (a] gmail.com> 11020 Date: Thu Aug 19 17:04:35 2010 -0400 11021 11022 DCE3+: switch pads to ddc mode when doing i2c 11023 11024 The pins for ddc and aux are shared so you need to switch the 11025 mode when doing ddc. The ProcessAuxChannel table already sets 11026 the pin mode to DP. This should fix unreliable ddc issues 11027 on DP ports using non-DP monitors. 11028 11029 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11030 11031 commit fd686668289258ffaf6b81057545e50612aac6a8 11032 Author: Dave Airlie <airlied (a] redhat.com> 11033 Date: Thu Aug 12 12:59:18 2010 +1000 11034 11035 radeon: fixup non-kms build 11036 11037 commit 5a9865d90c23c4ce0f46d380ea9119eac87a99eb 11038 Author: Dave Airlie <airlied (a] redhat.com> 11039 Date: Thu Aug 12 11:18:06 2010 +1000 11040 11041 evergreen: fix multi-head cursor support 11042 11043 commit 8351bb9f085fde5dc47b115375efcc61adc23859 11044 Author: Dave Airlie <airlied (a] redhat.com> 11045 Date: Tue Aug 10 15:05:45 2010 +1000 11046 11047 radeon: take 8/10 encoding into a/c on displayport link 11048 11049 commit b91e0efcb24eea32e6978c146c5409eeeeac0a62 11050 Author: Dave Airlie <airlied (a] redhat.com> 11051 Date: Tue Aug 10 15:05:25 2010 +1000 11052 11053 dce32: remove rmx workaround 11054 11055 commit 94bc1b7156cd0866566dc44a823c7e051bb45175 11056 Author: Dave Airlie <airlied (a] redhat.com> 11057 Date: Tue Aug 10 13:24:52 2010 +1000 11058 11059 atombios: fixup set crtc source like KMS 11060 11061 This removes a bunch of strict aliasing warnings and fixes the 11062 codepaths up like the latest KMS code, including a workaround for a bug 11063 on evergreen. 11064 11065 commit 9bc716eb62d4e0eed2902b92437a42634eef6ba1 11066 Author: Dave Airlie <airlied (a] redhat.com> 11067 Date: Tue Aug 10 13:23:21 2010 +1000 11068 11069 atombios: move adjust pixel clock around to follow KMS code flow 11070 11071 This reworks the pixel clock adjusting code to follow the KMS style, 11072 also fixes warnings in this code. 11073 11074 commit 31de43bf9d9eb93cc2b2150474ea7404beabe49d 11075 Author: Dave Airlie <airlied (a] redhat.com> 11076 Date: Tue Aug 10 13:21:39 2010 +1000 11077 11078 displayport: retry on timeout 11079 11080 this is ported from KMS 11081 11082 commit 9ef67335583d36080d227e8bce1966afe08e0486 11083 Author: Dave Airlie <airlied (a] redhat.com> 11084 Date: Tue Aug 10 13:21:01 2010 +1000 11085 11086 evergreen: don't call YUV table on evergreen 11087 11088 commit 1cce55e8ba43e7958cb67147aeaeb068826ab99f 11089 Author: Dave Airlie <airlied (a] redhat.com> 11090 Date: Tue Aug 10 13:20:13 2010 +1000 11091 11092 evergreen: add support to parse firmware info for ext dp clk 11093 11094 commit bbffd67d3296344e8735b007cdee83146d38369c 11095 Author: Dave Airlie <airlied (a] redhat.com> 11096 Date: Tue Aug 10 13:14:54 2010 +1000 11097 11098 atombios: realign digital transmitter/encoder setup with kms 11099 11100 commit 6244153467665f5007e2fc7786b4bcc4b0b96030 11101 Author: Dave Airlie <airlied (a] redhat.com> 11102 Date: Tue Aug 10 12:57:22 2010 +1000 11103 11104 update atombios.h to latest from kernel 11105 11106 commit bb7c77ca75e857f90791b0dd1c04c8e2f19d0e3c 11107 Author: Alex Deucher <alexdeucher (a] gmail.com> 11108 Date: Mon Aug 9 22:18:53 2010 -0400 11109 11110 atom: upstream parser update 11111 11112 fixes digital output problems on evergreen asics 11113 11114 commit cc0a167ff2db9b097891883ded147af1d67e4407 11115 Author: Alex Deucher <alexdeucher (a] gmail.com> 11116 Date: Tue May 25 18:17:15 2010 -0400 11117 11118 r6xx/r7xx: default to micro (1D) tiling for now 11119 11120 SW de-tiling used in the r600c 3D driver has issues with 11121 certain configurations. 11122 11123 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11124 11125 commit b5bfdbd70d9671250957ccd41dfc8818850d257e 11126 Author: Alex Deucher <alexdeucher (a] gmail.com> 11127 Date: Thu Aug 5 17:26:28 2010 -0400 11128 11129 r6xx/r7xx: add support for tiling with kms (v3) 11130 11131 Requires radeon drm 2.6.0 and updated mesa. 11132 11133 v2: - fix lockup due to not emiting DB_DEPTH_INFO 11134 https://bugs.freedesktop.org/show_bug.cgi?id=28342 11135 - fix drm minor version to deal with evergreen accel 11136 v3: rebase on current ddx git 11137 11138 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11139 11140 commit a3c59c6f6be7067421e348142da0ca13428dcd57 11141 Author: Alex Deucher <alexdeucher (a] gmail.com> 11142 Date: Wed Aug 4 12:07:36 2010 -0400 11143 11144 radeon: add new pci ids 11145 11146 commit a456587b77ae357750179a50f8db2a17c0f2738e 11147 Author: Alex Deucher <alexdeucher (a] gmail.com> 11148 Date: Mon Aug 2 14:24:41 2010 -0400 11149 11150 r6xx/r7xx: move syrface sync emit to the functions that emit surface info 11151 11152 reduces code duplication. 11153 11154 commit 8eba977cab1878ba247da8160771d41194d8014f 11155 Author: Alex Deucher <alexdeucher (a] gmail.com> 11156 Date: Thu Jun 10 13:34:59 2010 -0400 11157 11158 EXA: move more common helpers to exa_shared 11159 11160 commit 71c1a2704af23b61439cee5ce784f7fe267a8a26 11161 Author: Alex Deucher <alexdeucher (a] gmail.com> 11162 Date: Thu Jun 10 11:52:43 2010 -0400 11163 11164 remove unused reg headers from radeon_exa_shared.c 11165 11166 commit d73aef78919005369af1b60df138439b4b6105c3 11167 Author: Alex Deucher <alexdeucher (a] gmail.com> 11168 Date: Mon Aug 2 13:27:12 2010 -0400 11169 11170 r6xx/r7xx: set VGT_MAX_VTX_INDX to a larger value 11171 11172 commit dacaf5d827b58c39f9e5a7ac0530f9ea6e257347 11173 Author: Dave Airlie <airlied (a] redhat.com> 11174 Date: Mon Aug 2 08:33:51 2010 +1000 11175 11176 fix make distcheck 11177 11178 commit c79ce215a01b45fc63b483da167ae37ec7aefad6 11179 Author: Dave Airlie <airlied (a] redhat.com> 11180 Date: Sun Aug 1 16:51:48 2010 +1000 11181 11182 radeon/r600: restructure exa + vbo to provide more sharing 11183 11184 This is a precursor for r300/500 vbo support. 11185 11186 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11187 11188 commit 82254b59268140c4102ae3cd713743ae2be15c00 11189 Author: Alex Deucher <alexdeucher (a] gmail.com> 11190 Date: Fri Jul 30 17:15:05 2010 -0400 11191 11192 r6xx/r7xx: unify composite mask and non-mask pixel shader 11193 11194 commit 1c17f3a192f644e8e38b5cfb1470f49434bfba27 11195 Author: Alex Deucher <alexdeucher (a] gmail.com> 11196 Date: Fri Jul 30 16:34:54 2010 -0400 11197 11198 r6xx/r7xx: clean up composite vertex shader 11199 11200 keep CF, ALU, Fetch instructions in separate groups 11201 11202 commit f9d6c0de231357f96e2e0de71e6c9221bcb36bd4 11203 Author: Gaetan Nadon <memsize (a] videotron.ca> 11204 Date: Fri Jul 23 13:28:42 2010 -0400 11205 11206 The local copy of the modes code is no longer required. 11207 11208 The server 1.2 as shipped in the tarball on the web does not contain the 11209 modes code. It was added just after and found in git branch server-1.2-branch. 11210 11211 The modes code was initially included in version ati 6.8.0 and fails to compile 11212 with server 1.2 as it requires randr 1.2. The modes code is included in server 11213 versions 1.3 and later, so there is no need to provide an unknown version of 11214 the modes code in the ati driver tarball. It will never be used. 11215 11216 This patch makes the ati driver requiring server 1.3 or later. 11217 Version 6.8.0 configures and builds ok on server 1.3 11218 Master branch post 6.13.1 configures and builds ok on server 1.3 11219 11220 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11221 11222 commit 0028419acb0762eeb950de5fe702c93e70301612 11223 Author: Gaetan Nadon <memsize (a] videotron.ca> 11224 Date: Sat Jul 24 10:51:18 2010 -0400 11225 11226 config: add AM_PROG_CC_C_O for per-target compilation flags 11227 11228 Per-target compilation flags (theatre200_drv_la_CFLAGS) are required 11229 when multiple targets which require different compiler flags, 11230 are build in the same makefile. 11231 11232 Automake issues a command with -c and -o flags which not all compilers 11233 support. The object fles are prefixed with theatre200_drv_la. 11234 The macro AM_PROG_CC_C_O must then be used to provide this feature 11235 on compilers that do not have it. If not, a warning is issued at make time. 11236 11237 This macros checks for compiler support and if missing, uses a "compile" 11238 script it generates in the package root directory. 11239 11240 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11241 11242 commit aca0a8669b538d58f018f95c9b22e6b3ec1ffe50 11243 Author: Gaetan Nadon <memsize (a] videotron.ca> 11244 Date: Wed Jul 21 16:49:04 2010 -0400 11245 11246 config: add comments for main statements 11247 11248 commit 66e614f7115efeec237b3b916d9637e8b3e8985c 11249 Author: Gaetan Nadon <memsize (a] videotron.ca> 11250 Date: Wed Jul 21 16:07:00 2010 -0400 11251 11252 config: replace deprecated use of AC_OUTPUT with AC_CONFIG_FILES 11253 11254 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11255 11256 commit 16e5510c90ef1ba2bbaab78d18943f080b86d809 11257 Author: Gaetan Nadon <memsize (a] videotron.ca> 11258 Date: Wed Jul 21 14:37:41 2010 -0400 11259 11260 config: replace deprecated AC_HELP_STRING with AS_HELP_STRING 11261 11262 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11263 11264 commit b36d8c09e91382f4cfa71635374ec88f5b676d1c 11265 Author: Gaetan Nadon <memsize (a] videotron.ca> 11266 Date: Wed Jul 21 14:05:22 2010 -0400 11267 11268 config: replace deprecated AM_CONFIG_HEADER with AC_CONFIG_HEADERS 11269 11270 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11271 11272 commit cd9351b04c2d6982b28c647a63d550eb3e1937eb 11273 Author: Alex Deucher <alexdeucher (a] gmail.com> 11274 Date: Wed Jul 21 13:48:24 2010 -0400 11275 11276 r6xx/r7xx: group op variable state 11277 11278 Group the op variable state into one emit block, re-order 11279 to reduce dwords emitted. 11280 11281 commit 5f838c664e8010f4e51afecd4100d73a96fe1209 11282 Author: Gaetan Nadon <memsize (a] videotron.ca> 11283 Date: Wed Jul 21 09:27:42 2010 -0400 11284 11285 config: complete AC_INIT m4 quoting 11286 11287 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11288 11289 commit 48ec2e65c268b426ab9a3e214d174447cf5b5936 11290 Author: Gaetan Nadon <memsize (a] videotron.ca> 11291 Date: Wed Jul 21 08:33:38 2010 -0400 11292 11293 config: remove unrequired AC_SUBST for LIBDRM and LIBUDEV 11294 11295 These macros are called by the PKG_CHECK_MODULES macro. 11296 11297 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11298 11299 commit 1aabb7eb7d8f06c7481151145db3b9a722ce4ef0 11300 Author: Gaetan Nadon <memsize (a] videotron.ca> 11301 Date: Tue Jul 20 21:54:11 2010 -0400 11302 11303 config: remove unrequired AC_SUBST([DRI_CFLAGS]) 11304 11305 This macro is called by PKG_CHECK_MODULES 11306 11307 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11308 11309 commit 8f92b349821a3ee5ed8df55273d905b9605385aa 11310 Author: Gaetan Nadon <memsize (a] videotron.ca> 11311 Date: Tue Jul 20 21:44:57 2010 -0400 11312 11313 config: remove unrequired AC_SUBST([XORG_CFLAGS]) 11314 11315 This macro is called by PKG_CHECK_MODULES 11316 11317 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11318 11319 commit 5b483b832f9c4a5b92ffb7f72a470669201f1fba 11320 Author: Gaetan Nadon <memsize (a] videotron.ca> 11321 Date: Tue Jul 20 20:24:42 2010 -0400 11322 11323 config: remove unrequired AC_HEADER_STDC 11324 11325 Autoconf says: 11326 "This macro is obsolescent, as current systems have conforming 11327 header files. New programs need not use this macro". 11328 11329 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11330 11331 commit 6574e3a16eb3631ee7e00ee60a8c9ba95c8b84ef 11332 Author: Gaetan Nadon <memsize (a] videotron.ca> 11333 Date: Tue Jul 20 19:41:30 2010 -0400 11334 11335 config: remove AC_PROG_CC as it overrides AC_PROG_C_C99 11336 11337 XORG_STRICT_OPTION from XORG_DEFAULT_OPTIONS calls 11338 AC_PROG_C_C99. This sets gcc with -std=gnu99. 11339 If AC_PROG_CC macro is called afterwards, it resets CC to gcc. 11340 11341 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11342 11343 commit 593eff2924c2ad161d8b490fbf6d7e433fbe2a80 11344 Author: Gaetan Nadon <memsize (a] videotron.ca> 11345 Date: Tue Jul 20 18:45:18 2010 -0400 11346 11347 config: update AC_PREREQ statement to 2.60 11348 11349 Unrelated to the previous patches, the new value simply reflects 11350 the reality that the minimum level for autoconf to configure 11351 all x.org modules is 2.60 dated June 2006. 11352 11353 ftp://ftp.gnu.org/gnu/autoconf/autoconf-2.60.tar.gz 11354 11355 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11356 11357 commit c2ab6ffc25aa6759cbbb4c1fbbd4a136b38983bf 11358 Author: Gaetan Nadon <memsize (a] videotron.ca> 11359 Date: Tue Jul 20 16:15:29 2010 -0400 11360 11361 config: upgrade to util-macros 1.8 for additional man page support 11362 11363 Use MAN_SUBST now supplied in XORG_MANPAGE_SECTIONS 11364 The value of MAN_SUBST is the same for all X.Org packages. 11365 11366 commit cdeb1949c820242f05a8897d3ddd0718f204dacf 11367 Author: Jerome Glisse <jglisse (a] redhat.com> 11368 Date: Thu Jul 15 16:21:41 2010 -0400 11369 11370 kms: don't call cursor helper if using software cursor 11371 11372 Fix : 11373 https://bugzilla.redhat.com/show_bug.cgi?id=601713 11374 https://bugzilla.redhat.com/show_bug.cgi?id=598358 11375 11376 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 11377 11378 commit 06691376b1ee963c711420edaf5a03eab6f5658f 11379 Author: Dave Airlie <airlied (a] redhat.com> 11380 Date: Wed Jul 7 13:15:03 2010 +1000 11381 11382 fix build on non-kms 11383 11384 commit 052cf0169ae70d5448af6dc4db840b2fc195569b 11385 Author: Dave Airlie <airlied (a] redhat.com> 11386 Date: Wed Jul 7 11:10:46 2010 +1000 11387 11388 configure.ac: bump version post release 11389 11390 commit ad999e633ff41d27eed9d2c6535e163a7181b0bd 11391 Author: Dave Airlie <airlied (a] redhat.com> 11392 Date: Wed Jul 7 10:49:22 2010 +1000 11393 11394 set version for release 11395 11396 commit 37b348059b1c15d7b381cd3df3db52bd9ee6613e 11397 Author: Alex Deucher <alexdeucher (a] gmail.com> 11398 Date: Wed Jun 30 12:56:48 2010 -0400 11399 11400 remove rv100 quirk 11401 11402 Some RV100 cards with 2 VGA ports show up with DVI+VGA, however 11403 some boards with DVI+VGA have the same subsystem ids. Better 11404 to have a VGA port show up as DVI than having a non-useable 11405 DVI port. 11406 11407 reported by DHR in irc. 11408 11409 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11410 11411 commit c8ea942bd0e9d1c055d50b94440aa4dae425f20b 11412 Author: Cooper Yuan <cooperyuan (a] gmail.com> 11413 Date: Tue Jun 29 20:34:57 2010 +0800 11414 11415 Remove HDP_SOFT_RESET function, there is no need to reset HDP block. 11416 11417 This commit can fix an issue reported on DELL server, system gets hang 11418 during soft resetting while another application tries to access PCI 11419 configuration space. 11420 11421 commit 139b38bf67ec10d876cc56df833541d497ae4fa4 11422 Author: Alex Deucher <alexdeucher (a] gmail.com> 11423 Date: Thu Jun 24 14:55:09 2010 -0400 11424 11425 r6xx shader: use ADDR() for CF_DWORD0 11426 11427 no change in functionality 11428 11429 commit 801e83227a59a29eea425ea612083bbf2b536c30 11430 Author: Wolfram <bugzilla1 (a] malloc.de> 11431 Date: Mon Jun 21 18:59:19 2010 -0400 11432 11433 r6xx/r7xx: fix ums cmd buffer leak 11434 11435 Fixes: 11436 https://bugs.freedesktop.org/show_bug.cgi?id=27957 11437 11438 commit b13d719080b75fc6db4d15d2d323b8fce8b7ad06 11439 Author: Alex Deucher <alexdeucher (a] gmail.com> 11440 Date: Mon Jun 21 18:49:44 2010 -0400 11441 11442 r6xx/r7xx: fix miscount in state emit 11443 11444 commit c3c5c8e2cc91b51a24effdffb85281216eed731d 11445 Author: Alex Deucher <alexdeucher (a] gmail.com> 11446 Date: Mon Jun 21 14:30:17 2010 -0400 11447 11448 r6xx/r7xx accel: add back some additional default state 11449 11450 This adds back everything removed in c29157bbf5b0dd26857675282ab094082fbaed0d 11451 except CB_FOG_*, CB_CLEAR_* and the VPORT transforms. Those shouldn't 11452 be needed as we aren't using fog or viewport transforms. We probably don't 11453 need all the state that was added back either but I can't reproduce any 11454 problems here, so it's hard to say which parts are problematic. 11455 11456 Should fix: 11457 https://bugs.freedesktop.org/show_bug.cgi?id=28629 11458 and several corruption reports on #radeon. 11459 11460 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11461 11462 commit 800cb2088fec698d0626063a9ab198ff534938c0 11463 Author: Michel Dnzer <daenzer (a] vmware.com> 11464 Date: Mon Jun 21 08:15:14 2010 +0200 11465 11466 DRI2: Fix up confusion between windows and pixmaps. 11467 11468 Fixes crashes with xserver master, where looking up a pixmap private on a 11469 window no longer happens to work. 11470 11471 commit f7a91ece264af9f3fd2fc18e99aefcda93ce9f5c 11472 Author: Michel Dnzer <daenzer (a] vmware.com> 11473 Date: Mon Jun 14 08:52:16 2010 +0200 11474 11475 Convert x(c)alloc/xfree to m/calloc/free. 11476 11477 Fixes deprecation warnings with xserver master and should also work with older 11478 xservers. 11479 11480 commit ea37d24b1b6d4cbcf73e680846de25b72af216e3 11481 Author: Dave Airlie <airlied (a] redhat.com> 11482 Date: Mon Jun 21 13:55:15 2010 +1000 11483 11484 radeon: fix support for 1.9 server master. 11485 11486 This moves pixmap and mode set into CSR where its allowed. Should work fine on 11487 old servers also. 11488 11489 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11490 11491 commit fdd8ecafd054f65842351aee6ee6fba7af6613b2 11492 Author: Alex Deucher <alexdeucher (a] gmail.com> 11493 Date: Wed Jun 16 19:02:10 2010 -0400 11494 11495 r6xx/r7xx: macro safety fixes 11496 11497 commit 4651d77211b508cb6b76931807780e317f232220 11498 Author: Alex Deucher <alexdeucher (a] gmail.com> 11499 Date: Wed Jun 16 12:28:36 2010 -0400 11500 11501 radeon: fix depth 16 with ums 11502 11503 Fixes: 11504 https://bugs.freedesktop.org/show_bug.cgi?id=28494 11505 https://bugzilla.redhat.com/show_bug.cgi?id=554967 11506 11507 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11508 11509 commit 1e1d6a515428b6884fea586d180346fc74ef75a1 11510 Author: Alex Deucher <alexdeucher (a] gmail.com> 11511 Date: Wed Jun 16 12:20:03 2010 -0400 11512 11513 r3xx-r5xx Xv: disable bicubic filtering by default 11514 11515 - makes Xv more consistent with r1xx/r2xx/r6xx/r7xx 11516 - Xv attributes like brightness, contrast, hue, etc. only work 11517 when bicubic is disabled. 11518 - avoids performance issues on some systems when sampling from textures 11519 in GART with kms. 11520 - It can be re-enabled with using xvattr 11521 11522 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11523 11524 commit c29157bbf5b0dd26857675282ab094082fbaed0d 11525 Author: Alex Deucher <alexdeucher (a] gmail.com> 11526 Date: Tue Jun 15 19:39:42 2010 -0400 11527 11528 r6xx/r7xx: remove unnecessary state emit 11529 11530 No need to emit state that doesn't impact that we use for EXA/Xv. 11531 11532 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11533 11534 commit 35280545fcfb911f189d7657bd1040a28450fe7b 11535 Author: Alex Deucher <alexdeucher (a] gmail.com> 11536 Date: Tue Jun 15 19:21:32 2010 -0400 11537 11538 r6xx/r7xx: reorg default and clipping state emit 11539 11540 Group ordered registers to save command buffer space. 11541 Reduces the default and clipping state from 256 to 160 dwords. 11542 11543 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11544 11545 commit a2528a734c1d4e8639c49e5d222e3630a93ffbfd 11546 Author: Alex Deucher <alexdeucher (a] gmail.com> 11547 Date: Tue Jun 15 17:00:34 2010 -0400 11548 11549 r6xx/r7xx accel: remove some duplicate emits and minor clean up 11550 11551 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11552 11553 commit 5f093357f18eb9bea641394ab86a92a1766d8f2e 11554 Author: Gaetan Nadon <memsize (a] videotron.ca> 11555 Date: Sat Jun 12 15:11:46 2010 -0400 11556 11557 COPYING: replace stub file with actual copyright notices 11558 11559 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 11560 11561 commit 0c2118586d3edeecc2473b5d685472df4b5e70fa 11562 Author: Dave Airlie <airlied (a] redhat.com> 11563 Date: Fri Jun 11 15:53:18 2010 +1000 11564 11565 radeon: fixup last fix, use CURSOR_WIDTH not RADEON_CURSOR_WIDTH. 11566 11567 commit 9a117f768cb8261327fd3c324da9c98875785cc1 11568 Author: Dave Airlie <airlied (a] redhat.com> 11569 Date: Fri Jun 11 14:28:24 2010 +1000 11570 11571 radeon: set proper stride for cursor in tiling flags. 11572 11573 definitely incorrect, but hope it doesn't break anything. 11574 11575 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11576 11577 commit b6346ede94f0d0b11ee04770cf52508cb0a5e6c6 11578 Author: Dave Airlie <airlied (a] redhat.com> 11579 Date: Thu Jun 10 20:13:58 2010 -0400 11580 11581 drmmode: fix big endian issue with properties. 11582 11583 On a power machine with an rn50, this was causing load detection to be turned off after the first X run, subsequent X runs would find nothing connected. 11584 11585 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11586 11587 commit 1971dc6d758dea13b9fb6d1c8f516c165628b5e5 11588 Author: Alex Deucher <alexdeucher (a] gmail.com> 11589 Date: Tue Jun 8 11:18:28 2010 -0400 11590 11591 radeon: fix rn50 cloning with kms 11592 11593 Since they only have one crtc sometimes the xserver doesn't assign 11594 a crtc to one of the outputs even though both outputs have common modes 11595 which results in only one monitor being enabled. Assign a crtc in 11596 preinit so that both outputs light up. 11597 11598 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11599 11600 commit 426114b4a99d37b394efe3336968bb0ab9b6e9ae 11601 Author: Dave Airlie <airlied (a] redhat.com> 11602 Date: Tue Jun 8 11:34:35 2010 +1000 11603 11604 xv: fix Xv on M6/RV100 under KMS. 11605 11606 pRADEONEnt->HasCRTC2 wasn't setup under KMS. 11607 11608 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11609 11610 commit f64bf0de8e2de7c1bf9cc0c614603dd23c9060ad 11611 Author: Alex Deucher <alexdeucher (a] gmail.com> 11612 Date: Thu Jun 3 14:48:21 2010 -0400 11613 11614 evergreen: explicitly disable accel on evergreen 11615 11616 Previously we relied on the drm not having accel enabled 11617 to make sure evergreen used shadowfb, now we when we enable 11618 accel in the drm, we need to make sure the ddx doesn't try 11619 and use it until it's implemented. 11620 11621 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11622 11623 commit 428125c095b18c2a2864c1aef24ac0f384b6be54 11624 Author: Marek Olk <maraeo (a] gmail.com> 11625 Date: Sun May 16 20:11:17 2010 +0000 11626 11627 r3xx-r5xx: enable color tiling by default on KMS 11628 11629 Signed-off-by: Marek Olk <maraeo (a] gmail.com> 11630 11631 commit 6712ce4f4715c8ce7c2fcddb52ca9b103bdd634c 11632 Author: Marek Olk <maraeo (a] gmail.com> 11633 Date: Sun May 16 20:09:31 2010 +0000 11634 11635 rs740: follow macro-switch too 11636 11637 Signed-off-by: Marek Olk <maraeo (a] gmail.com> 11638 11639 commit 842fa162e9d3bbad2bc44d3732bbc8e5a54402ea 11640 Author: Alex Deucher <alexdeucher (a] gmail.com> 11641 Date: Tue May 18 11:21:01 2010 -0400 11642 11643 dri2: use radeon_pick_best_crtc() 11644 11645 It's already exposed and used by exa and Xv and has 11646 the same functionality. radeon_covering_crtc() can 11647 be dropped eventually. 11648 11649 commit 64e6858aaf5d0e39ecc9f9804689012f3a38660a 11650 Author: Alex Deucher <alexdeucher (a] gmail.com> 11651 Date: Mon May 17 12:50:44 2010 -0400 11652 11653 r2xx exa: fix typo in 27c3326863deae36bc35e2c3b73fffd400208ff1 11654 11655 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11656 11657 commit 27c3326863deae36bc35e2c3b73fffd400208ff1 11658 Author: Alex Deucher <alexdeucher (a] gmail.com> 11659 Date: Mon May 17 09:53:32 2010 -0400 11660 11661 r1xx/r2xx EXA: handle RepeatNone properly with transforms 11662 11663 Fallback with unsupported ops like r3xx, r6xx. 11664 11665 Should fix fdo bug 28142. 11666 11667 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 11668 11669 commit 30591320ec46e491ba20904cc64f3405b51c6505 11670 Author: Jerome Glisse <jglisse (a] redhat.com> 11671 Date: Fri May 7 17:35:09 2010 +0200 11672 11673 kms: add support for the MSC swap & sync API 11674 11675 This patch is mostly a port over of Intel ddx code for 11676 MSC support. It needs a radeon KMS module with version 11677 2.4 which has the query for hw crtc id. 11678 11679 Signed-off-by: Jesse Barnes <jbarnes (a] virtuousgeek.org> 11680 Signed-off-by: Jerome Glisse <jglisse (a] redhat.com> 11681 11682 commit 766024dcc61c83490540910ce752f9bfe6dddba4 11683 Author: Marek Olk <maraeo (a] gmail.com> 11684 Date: Mon May 3 22:56:27 2010 +0200 11685 11686 r3xx-r5xx: fix texturing with small macrotiled pixmaps 11687 11688 Pixmaps smaller than a macrotile cannot be used as textures because hardware 11689 automatically switches to macro-linear and therefore sampled pixels are 11690 messed up. This behavior is known as MACRO_SWITCH. 11691 11692 The only sane workaround seems to be not using macrotiling for small pixmaps. 11693 11694 The function RADEONMacroSwitch has been ported from r300g and implements 11695 MACRO_SWITCH the same way it's implemented in hardware. It's been well tested 11696 in r300g. 11697 11698 This commit also fixes blit-based framebuffer reads, which are used for tiled 11699 surfaces in r300g, when ColorTiling is enabled. 11700 11701 Signed-off-by: Marek Olk <maraeo (a] gmail.com> 11702 11703 commit 6aedd57f6c2f08d0151a8bd1c5893d40d3db709a 11704 Author: Alex Deucher <alexdeucher (a] gmail.com> 11705 Date: Tue May 4 14:30:52 2010 -0400 11706 11707 radeon: fix pll_out_min default for pre-avivo 11708 11709 Fixes fdo bug 27960 11710 11711 commit b2ecd1bb507ed4e0d9ba6ebd498c51da2e8136fa 11712 Author: Michel Dnzer <daenzer (a] vmware.com> 11713 Date: Fri Apr 23 18:28:22 2010 +0200 11714 11715 FB/MMIO_cnt should be ints, not pointers. 11716 11717 Pointed out by compiler warnings. 11718 11719 https://bugs.freedesktop.org/show_bug.cgi?id=27817 11720 11721 commit 1ad41054989f7c3edec373ccf09dceda6d7c94b4 11722 Author: Alex Deucher <alexdeucher (a] gmail.com> 11723 Date: Thu Apr 22 16:48:21 2010 -0400 11724 11725 r3xx: fix gb pipe setup for SE cards 11726 11727 Noticed by Tormod Volden 11728 11729 commit 59e287d7c484b4addd4a06d013670577639c7ae2 11730 Author: Alex Deucher <alexdeucher (a] gmail.com> 11731 Date: Thu Apr 22 02:05:40 2010 -0400 11732 11733 r3xx-r5xx: fix vertex units 11734 11735 Noticed by Tormod Volden. 11736 11737 RV3xx is 2, RV560,RV570 is 8 11738 11739 commit 2a6b409496f26da0436972b5feae6ea035dde08d 11740 Author: Owain Ainsworth <zerooa (a] googlemail.com> 11741 Date: Mon Apr 19 14:37:33 2010 +0100 11742 11743 Reference count shared driver mappings. 11744 11745 With MMIO it wasn't *such* a bit deal if we leaked the smallish mapping. 11746 with FB it could be a larger deal. So instead of worrying about this, 11747 reference count the mappings in the entity structure and unmap them when 11748 no one cares anymore. 11749 11750 Prompted by a discussion with airlied 11751 11752 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11753 11754 commit 761f0de5556e46f166280476185977f720efe586 11755 Author: Owain G. Ainsworth <zerooa (a] googlemail.com> 11756 Date: Mon Apr 19 04:05:36 2010 +0100 11757 11758 Make consistent use of fbOffset and share fb mappings. 11759 11760 What we were doing previously was mapping the framebuffer for zaphod for 11761 only this driver instances chunk, however, fbOffset was (rightly) set to 11762 the offset into the whole framebuffer we were using. 11763 11764 Since in some cases we did operations on the FB virtual address + 11765 fbOffset (for example zeroing the framebuffer on entervt) we were 11766 actually pissing all over ourselves in those cases. 11767 11768 Fix this by implementing shared fb mappings like we do for MMIO already, 11769 and whenever we wish to refer to our area of FB space we always use 11770 fbOffset. Fixes zaphod for some users on r600 chipsets, my 4870 is still 11771 behaving strangely on screen 0, but I suspect that is another bug. 11772 11773 Once calculation (in PreInitAccel) is now wrong because of this, however 11774 dri on zaphod does now happen so this is irrelavent, add a comment to 11775 that effect. 11776 11777 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11778 11779 commit 2059d628c5fb03222a62502cc1b5724bf296a7b3 11780 Author: Owain G. Ainsworth <zerooa (a] googlemail.com> 11781 Date: Mon Apr 19 04:05:35 2010 +0100 11782 11783 Kill per-context SAREA support. 11784 11785 It has never been used, and since the world is changing it almost 11786 certainly never will be. Good riddance. 11787 11788 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11789 11790 commit c4b46f56279cc18c8114235ca41cfb41ffede67f 11791 Author: Owain G. Ainsworth <zerooa (a] googlemail.com> 11792 Date: Mon Apr 19 04:05:34 2010 +0100 11793 11794 Remove remnants of the ShowCache option. 11795 11796 The rest of it died a long time ago. 11797 11798 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11799 11800 commit 03b34c5fb12c4df180d4df113d2b654f6afd7230 11801 Author: Dave Airlie <airlied (a] redhat.com> 11802 Date: Mon Apr 19 17:58:07 2010 +1000 11803 11804 radeon: add FireMV 2400 pci id. 11805 11806 From RH#581927 - thanks to Carsten Clasohm 11807 11808 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11809 11810 commit 6c07816025f70e522986614c29c300ea13e5d932 11811 Author: Alex Deucher <alexdeucher (a] gmail.com> 11812 Date: Fri Apr 16 18:50:17 2010 -0400 11813 11814 r5xx texvid: deal with large numbers of verts 11815 11816 should fix fdo bug 25884 11817 11818 commit 5ee320a6b1b4c65fe592c8ac4d1981799242d59e 11819 Author: Alex Deucher <alexdeucher (a] gmail.com> 11820 Date: Fri Apr 16 18:42:56 2010 -0400 11821 11822 r3xx texvid: deal with large numbers of verts 11823 11824 should fix fdo bug 25884 11825 11826 commit 47af3f4f266232517486238917d82fc5ca9c82e6 11827 Author: Alex Deucher <alexdeucher (a] gmail.com> 11828 Date: Fri Apr 16 20:50:18 2010 -0400 11829 11830 r2xx texvid: deal with large numbers of verts 11831 11832 should fix fdo bug 25884 11833 11834 commit 57577d5cd0641b7cad02242478699bcfece59227 11835 Author: Alex Deucher <alexdeucher (a] gmail.com> 11836 Date: Fri Apr 16 20:40:56 2010 -0400 11837 11838 r1xx texvid: deal with large numbers of verts 11839 11840 should fix fdo bug 25884 11841 11842 commit 31a888e78fa403f2bddacee098a29f36eaa969cb 11843 Author: Alex Deucher <alexdeucher (a] gmail.com> 11844 Date: Fri Apr 16 15:52:59 2010 -0400 11845 11846 r5xx textured video: split into prepare and draw functions 11847 11848 commit f7ebe4a127965bb9e46d20c60331bcc4dab76fc5 11849 Author: Alex Deucher <alexdeucher (a] gmail.com> 11850 Date: Fri Apr 16 15:46:24 2010 -0400 11851 11852 r3xx textured video: split into prepare and draw functions 11853 11854 commit 10799b4ce0c18c5a7d92d688549e8e777344f111 11855 Author: Alex Deucher <alexdeucher (a] gmail.com> 11856 Date: Fri Apr 16 15:38:49 2010 -0400 11857 11858 r2xx textured video: split into prepare and draw functions 11859 11860 commit bd1cf42201a1f918080bf34786a6de327cb31b5f 11861 Author: Alex Deucher <alexdeucher (a] gmail.com> 11862 Date: Fri Apr 16 15:31:40 2010 -0400 11863 11864 r1xx textured video: split into prepare and draw functions 11865 11866 commit c1136f94b80346065893f8a43c0fbf60c8c9b057 11867 Author: Alex Deucher <alexdeucher (a] gmail.com> 11868 Date: Fri Apr 16 15:23:14 2010 -0400 11869 11870 Xv: track vtx_count, is_planar in port priv 11871 11872 commit 2f680b631357661511456f07b7b4ba1aa2cbf30c 11873 Author: Alex Deucher <alexdeucher (a] gmail.com> 11874 Date: Fri Apr 16 13:43:41 2010 -0400 11875 11876 radeon/kms: fix possible crtc mask for evergreen 11877 11878 commit 4656f5dff1ed72fa2c7a1484305f2aef7b65ff2b 11879 Author: Alex Deucher <alexdeucher (a] gmail.com> 11880 Date: Fri Apr 16 13:36:25 2010 -0400 11881 11882 radeon/kms: fix crash when using more than two heads 11883 11884 fixes fdo bug 27692 11885 11886 commit a69e749d0562887af6bd236c38802472e54640c4 11887 Author: Dave Airlie <airlied (a] redhat.com> 11888 Date: Fri Apr 16 13:58:54 2010 +1000 11889 11890 kms: add uevent support. 11891 11892 When we get a hotplug event from the kernel we should notify the client side to reconfigure displays. 11893 11894 based on work by ajax in intel driver. 11895 11896 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11897 11898 commit 22a46dddd375b2b9399e12fdf168fa5292ff17a4 11899 Author: Alex Deucher <alexdeucher (a] gmail.com> 11900 Date: Tue Apr 13 11:08:59 2010 -0400 11901 11902 radeon: add support for pll algo selection 11903 11904 tv-out on atom systems is very particular about it's 11905 dividers. force it to use the old algo. 11906 11907 Should fix fdo bug 27593. 11908 11909 commit c1b817c45477c21234abaaebe78feb7ea4fd92b1 11910 Author: Alex Deucher <alexdeucher (a] gmail.com> 11911 Date: Mon Apr 12 16:32:04 2010 -0400 11912 11913 atom: disable TV encoder when VGA is in use 11914 11915 Switching between TV and VGA caused problems on some 11916 systems since the TV encoder was left enabled. 11917 11918 Should fix fdo bug 25520. 11919 11920 commit 878814aeab7031ff6169fcaf2be869b7595c8699 11921 Author: Alex Deucher <alexdeucher (a] gmail.com> 11922 Date: Mon Apr 12 11:16:14 2010 -0400 11923 11924 Xv overlay: set scalerwidth to 1920 on r3xx/r4xx chips 11925 11926 reported on phoronix forums. 11927 11928 commit 276fe739f1caa95778213054375a65288b80f320 11929 Author: Alex Deucher <alexdeucher (a] gmail.com> 11930 Date: Fri Apr 9 20:03:09 2010 -0400 11931 11932 radeon: add some new pci ids 11933 11934 commit eb5665688ef9b52f03f61546351d0848cab54740 11935 Author: Alex Deucher <alexdeucher (a] gmail.com> 11936 Date: Thu Apr 8 00:31:52 2010 -0400 11937 11938 radeon: don't setup Xv on rn50 11939 11940 It has no overlay or 3D engine. 11941 11942 fixes fdo bug 27528 11943 11944 commit 3c6c6afcbfc6f2c582e13a7c37fe2014b0875b22 11945 Author: Alex Deucher <alexdeucher (a] gmail.com> 11946 Date: Mon Apr 5 12:12:56 2010 -0400 11947 11948 change version post release 11949 11950 commit fb7911912e60b2cdbc2152b96847775b767ca3ef 11951 Author: Alex Deucher <alexdeucher (a] gmail.com> 11952 Date: Mon Apr 5 11:59:24 2010 -0400 11953 11954 bump version for release 11955 11956 commit 476a1c6e8b42807b897b8c6f8550ec42b2d5c10e 11957 Author: Alex Deucher <alexdeucher (a] gmail.com> 11958 Date: Thu Apr 1 12:42:56 2010 -0400 11959 11960 radeon: fix possible gpio i2c table overrun 11961 11962 GPIO_I2C_INFO does not always have ATOM_MAX_SUPPORTED_DEVICE 11963 entries. Should fix Novell bug 589022. 11964 11965 commit 6baa96c44ca93b88acf5233335cee233e59d5af4 11966 Author: Dave Airlie <airlied (a] redhat.com> 11967 Date: Thu Apr 1 11:24:32 2010 +1000 11968 11969 r600/exa: further cleanup, use the object struct in the accel state. 11970 11971 This cleans up the accel state objects as well. 11972 11973 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11974 11975 commit 87f49a24019967552c5478d6af89ba438b4261c8 11976 Author: Dave Airlie <airlied (a] redhat.com> 11977 Date: Thu Apr 1 10:52:58 2010 +1000 11978 11979 r600: cleanup wasteful variables. 11980 11981 Just assign directly to the structs. 11982 11983 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11984 11985 commit 478319e55d52fce241d040912fea952071e3b742 11986 Author: Dave Airlie <airlied (a] redhat.com> 11987 Date: Thu Apr 1 10:45:52 2010 +1000 11988 11989 r600: reduce function call overhead. 11990 11991 Create a small accel object that can be used to reduce the amount 11992 of parameters passed to SetAccelState. This can be cleaned up a lot more. 11993 11994 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 11995 11996 commit bc93395b3eb5e3511c1b62af90693269f4fa6e13 11997 Author: Alex Deucher <alexdeucher (a] gmail.com> 11998 Date: Wed Mar 31 22:30:24 2010 -0400 11999 12000 r6xx+ EXA: fix swapped domains in kms UTS 12001 12002 Noticed by Dave. 12003 12004 Should fix fdo bug 27284 12005 12006 commit 3a33b1a92c328b690bdc1aafa0011e4326109815 12007 Author: Owain G. Ainsworth <zerooa (a] googlemail.com> 12008 Date: Tue Mar 30 17:14:02 2010 +0100 12009 12010 Add support for MacModel autodetection on OpenBSD. 12011 12012 This allows radeon-equipped macs to work without an xorg.conf on openbsd. 12013 Patch originally written by Mark Kettenis (kettenis (a] openbsd.org) 12014 12015 commit c7eeda8c3f3514ba95ebf2893fbe124bf526b3df 12016 Author: Alex Deucher <alexdeucher (a] gmail.com> 12017 Date: Fri Mar 26 01:38:36 2010 -0400 12018 12019 radeon: add support eDP connectors with and LCD device tag 12020 12021 should fix fdo bug 27322 12022 12023 commit ff323e11485fa5030ac278c836bf80a535c22249 12024 Author: Alex Deucher <alexdeucher (a] gmail.com> 12025 Date: Fri Mar 26 01:24:16 2010 -0400 12026 12027 radeon: fix name string for eDP 12028 12029 commit 53ac06161eb2b8cffb1b88e24a9a21cfd12e5883 12030 Author: Alex Deucher <alexdeucher (a] gmail.com> 12031 Date: Tue Mar 23 13:34:38 2010 -0400 12032 12033 r6xx+ EXA/Xv: add a R600SetAccelState function 12034 12035 This moves CS bo checking and alignment checks into 12036 a central location. It also cleans up the code. 12037 12038 commit 7a044472dfea7cf05ba4c2b87be30e94e2ae0b62 12039 Author: Alex Deucher <alexdeucher (a] gmail.com> 12040 Date: Tue Mar 23 01:27:22 2010 -0400 12041 12042 r6xx+ EXA: always use a temp surface for overlapping copies 12043 12044 The scanline based fallback code was complex and added a lot of 12045 overhead. It also didn't work with kms. 12046 12047 commit d33bddaedae81126ab7f0023af1c6443bb0b4c4f 12048 Author: Alex Deucher <alexdeucher (a] gmail.com> 12049 Date: Mon Mar 22 18:15:23 2010 -0400 12050 12051 r6xx+ EXA: always use the accel_state state in DoPrepareCopy 12052 12053 commit dda3f5a99e7a2dc5d57860f4d07df3498e1e21df 12054 Author: Alex Deucher <alexdeucher (a] gmail.com> 12055 Date: Mon Mar 22 18:05:36 2010 -0400 12056 12057 r6xx EXA/Xv: track src/dst domains 12058 12059 Much of the code is shared, so track the src/dst 12060 domains so we make sure the uses consistent domains 12061 for each bo. 12062 12063 commit 5c256808cb5fea955eea96ffe9196473715156aa 12064 Author: Alex Deucher <alexdeucher (a] gmail.com> 12065 Date: Wed Mar 17 23:47:50 2010 -0400 12066 12067 XAA: disable render accel 12068 12069 It's been reported broken for a while. Should fix 12070 fdo bug 27151, others. 12071 12072 commit a0683be5cc082bdbdd3bc4e9b52f39f423650946 12073 Author: Dave Airlie <airlied (a] redhat.com> 12074 Date: Thu Mar 18 12:36:25 2010 +1000 12075 12076 radeon: avoid using DRI1 init path on DRI2 driver. 12077 12078 I was playing with multi-seat and found this code, fixed 12079 it up to be sane and more DRI2 like. 12080 12081 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 12082 12083 commit 080a5414593e9b59ed923f26aa6057747b0c868f 12084 Author: Alex Deucher <alexdeucher (a] gmail.com> 12085 Date: Tue Mar 16 12:33:39 2010 -0400 12086 12087 kms: fix ums naming compat for DisplayPort 12088 12089 commit 819b4015349b5d8c5ffa5f979097599774fce5bb 12090 Author: Alex Deucher <alexdeucher (a] gmail.com> 12091 Date: Mon Mar 15 13:47:29 2010 -0400 12092 12093 dump version for rc release 12094 12095 commit d3482a947e3731be4ed0b00b4e3079470700dc4c 12096 Author: Michael Cree <mcree (a] orcon.net.nz> 12097 Date: Fri Mar 12 22:23:31 2010 +1300 12098 12099 Fix some word accesses in AtomBios to work on all architectures. 12100 12101 The UINT16LE_TO_CPU(), etc., macros are used in the AtomBios code to 12102 fix up endian issues but they do not address bad alignment or assist 12103 architectures that cannot perform hardware byte or word accesses. 12104 This patch inserts use of the ldw_u(), etc., interface of the Xserver 12105 into certain AtomBios accesses to address alignment issues. 12106 12107 This resolves Debian bug 572311, namely that the driver when compiled 12108 for generic Alpha architecture (i.e. doesn't use the byte-word extension) 12109 resulted in no display output on certain Radeon cards. 12110 12111 Signed-off-by: Michael Cree <mcree (a] orcon.net.nz> 12112 12113 commit 488c9fd8300505cc6c0c2f8f0f00849f27cc5d63 12114 Author: Alex Deucher <alexdeucher (a] gmail.com> 12115 Date: Mon Mar 15 12:25:57 2010 -0400 12116 12117 r6xx/r7xx: fix domain handling in accel code 12118 12119 Noticed by Pauli and Michel on IRC. 12120 12121 Improves GetImage performace by a factor of ~10. 12122 12123 commit 2ace2591d92fb6d3ce7a6453edb04b36a6c49a32 12124 Author: Alex Deucher <alexdeucher (a] gmail.com> 12125 Date: Mon Mar 15 12:03:58 2010 -0400 12126 12127 radeon: remove some leftover debugging output 12128 12129 commit 67e81c8f17ddde6eba633d2a5aef528e1d598d89 12130 Author: Andrzej Hajda <andrzej.hajda (a] wp.pl> 12131 Date: Wed Mar 10 18:19:35 2010 -0500 12132 12133 radeon: add support for pal on legacy IGP chips 12134 12135 Based on my initial non-working patch. 12136 12137 Fixes some element of fdo bug 12007 12138 12139 commit 3a44f1cb0d2bb748692b1024003de8ee88ca77a5 12140 Author: Alex Deucher <alexdeucher (a] gmail.com> 12141 Date: Tue Mar 9 09:44:01 2010 -0500 12142 12143 atom: i2c gpio fixes 12144 12145 Basically a port of my kms patch. This allows us 12146 to remove some quirks. 12147 12148 commit d85cb40f516c67305e818302bec7ee817df4144c 12149 Author: Matt Turner <mattst88 (a] gmail.com> 12150 Date: Sun Mar 7 14:24:35 2010 -0500 12151 12152 Don't check for Xinerama. 12153 12154 It doesn't seem to be used anywhere, so don't require it. 12155 12156 CC: Jerome Glisse <jglisse (a] redhat.com> 12157 CC: Alex Deucher <alexdeucher (a] gmail.com> 12158 CC: Dave Airlie <airlied (a] redhat.com> 12159 Signed-off-by: Matt Turner <mattst88 (a] gmail.com> 12160 12161 commit e7b41f8cb082ed462d29bf3fc440072424cbd852 12162 Author: Alex Deucher <alexdeucher (a] gmail.com> 12163 Date: Fri Mar 5 19:16:11 2010 -0500 12164 12165 radeon: disable frac fb div with new pll code 12166 12167 fixes fdo bug 26897 12168 12169 commit 14aff767490c253cbcdd411f812e50b91673119e 12170 Author: Alex Deucher <alexdeucher (a] gmail.com> 12171 Date: Wed Mar 3 13:31:19 2010 -0500 12172 12173 radeon: add new RS880 pci id 12174 12175 commit e6dc886634b38e4a36af7b5f0b23299d5acd7244 12176 Author: Dave Airlie <airlied (a] redhat.com> 12177 Date: Tue Mar 2 10:25:15 2010 +1000 12178 12179 radeon: bump configure.ac 12180 12181 commit 4975658f05c387b39b3e96a292a7683f17645c2c 12182 Author: Dave Airlie <airlied (a] redhat.com> 12183 Date: Sat Feb 27 16:47:19 2010 +1000 12184 12185 pciids: hopefully fix HP 12186 12187 commit e76b90b399c3cc0f0998c0209300c46f97505498 12188 Author: Alex Deucher <alexdeucher (a] gmail.com> 12189 Date: Fri Feb 26 15:01:28 2010 -0500 12190 12191 rv740: disable dfs workaround for drm 1.32+ 12192 12193 rv740 pipe setup was fixed in 2.6.33 12194 12195 commit 2de0af5f5b806f9dbfdb1e9b6a5cf96d9433961d 12196 Author: Matt Turner <mattst88 (a] gmail.com> 12197 Date: Wed Feb 24 22:46:28 2010 -0500 12198 12199 Use RADEON_ALIGN instead of open coding it. 12200 12201 Also fix some RADEON_ALIGN(x, 63), which would return incorrect results 12202 for odd x. Though this shouldn't happen, it's still not right. You 12203 wouldn't ever write (x + 62) & ~62 which is clearly wrong (and what it 12204 expands to). 12205 12206 CC: Jerome Glisse <jglisse (a] redhat.com> 12207 CC: Alex Deucher <alexdeucher (a] gmail.com> 12208 CC: Dave Airlie <airlied (a] redhat.com> 12209 Signed-off-by: Matt Turner <mattst88 (a] gmail.com> 12210 12211 commit c7e81d2f3a372e0d5f751dd0c5091aec2b56d936 12212 Author: Matt Turner <mattst88 (a] gmail.com> 12213 Date: Wed Feb 24 22:46:27 2010 -0500 12214 12215 Use/define RADEON_GPU_PAGE_SIZE instead of sprinkling 4096 everywhere. 12216 12217 Also, define RADEON_BUFFER_ALIGN in terms of it, and replace some 12218 RADEON_ALIGN(x, RADEON_BUFFER_ALIGN) with RADEON_ALIGN(x, 12219 RADEON_GPU_PAGE_SIZE) since this is really what was intended. 12220 12221 CC: Jerome Glisse <jglisse (a] redhat.com> 12222 CC: Alex Deucher <alexdeucher (a] gmail.com> 12223 CC: Dave Airlie <airlied (a] redhat.com> 12224 Signed-off-by: Matt Turner <mattst88 (a] gmail.com> 12225 12226 commit c0a5c9403dff254e1669df606a4193794270aeff 12227 Author: Dave Airlie <airlied (a] redhat.com> 12228 Date: Fri Feb 26 19:17:28 2010 +1000 12229 12230 pci: add HP vendor id 12231 12232 commit 3970cc5563900f08ba5236b23f765c926616c345 12233 Author: Cooper Yuan <cooperyuan (a] gmail.com> 12234 Date: Fri Feb 26 16:14:12 2010 +0800 12235 12236 add more triple-head server ID 12237 12238 commit b499eee7506f7478649e0000e9f8fcd00ac9d88d 12239 Author: James Le Cuirot <chewi (a] aura-online.co.uk> 12240 Date: Thu Feb 25 21:13:18 2010 -0500 12241 12242 clean up kms zaphod handling 12243 12244 Taken from James patch on bug 24523 12245 12246 agd5f: adapt to the current code 12247 12248 commit 433c8617341f5768255826435a2b09afba684f02 12249 Author: Alex Deucher <alexdeucher (a] gmail.com> 12250 Date: Thu Feb 25 21:03:13 2010 -0500 12251 12252 update ZaphodHeads option 12253 12254 Now that the screen section mapping is fixed, 12255 Make the option per-instance and allow multiple 12256 outputs to be specified; e.g., DVI-0 and HDMI-0 12257 associated with instance 0 and LVDS and VGA-0 12258 associated with instance 1. 12259 12260 commit f3e20b01c77f02bdf3483ceee27a1f00d240c33d 12261 Author: James Le Cuirot <chewi (a] aura-online.co.uk> 12262 Date: Thu Feb 25 19:10:19 2010 -0500 12263 12264 zaphod: Fix entity instance assignment. 12265 12266 Each screen needs to have a pointer back to its 12267 device's entity instance, Currently Screen1 is 12268 pointing to Screen0's instance so when it comes 12269 to match up the screen sections with the detected 12270 screens, Screen0 gets chosen in both cases. 12271 12272 See bug 24523 12273 12274 commit 1b7e9a2e50f77819b3aff4e37ba39eaec69ff932 12275 Author: Alex Deucher <alexdeucher (a] gmail.com> 12276 Date: Tue Feb 23 11:39:55 2010 -0500 12277 12278 radeon: fixes for zaphodheads option 12279 12280 Needed for systems with more than two outputs. 12281 Both KMS and non-KMS zaphod work on systems with 12282 more than two outputs now. 12283 12284 commit bd83e5f1c75b8c00b43846c7225568917fc9a8ce 12285 Author: Dave Airlie <airlied (a] redhat.com> 12286 Date: Tue Feb 23 19:27:24 2010 +1000 12287 12288 zaphod: hopefully fix kms + zaphod 12289 12290 commit 0c3468d812e3790ce03d9e76779ae81e7b7b82d5 12291 Author: Alex Deucher <alexdeucher (a] gmail.com> 12292 Date: Mon Feb 22 17:34:51 2010 -0500 12293 12294 AVIVO: set frac fb divider pll flag 12295 12296 matches previous behavior 12297 12298 commit 212e152536c4325e6799018891d9aee132681f48 12299 Author: Alex Deucher <alexdeucher (a] gmail.com> 12300 Date: Mon Feb 22 17:27:24 2010 -0500 12301 12302 radeon: update new pll algo 12303 12304 - add support for pre-avivo chips 12305 - add support for fixed post/ref dividers 12306 - add support for non-fractional fb dividers 12307 12308 By default avivo chips use the new algo and 12309 pre-avivo chips use the old algo. Use the 12310 "NewPLL" option to toggle between them (set to 12311 TRUE for the new algo, FALSE for the old). 12312 12313 commit e68d3a3890fc81c51f2006b5548da1e8756ad2fd 12314 Author: Alex Deucher <alexdeucher (a] gmail.com> 12315 Date: Thu Feb 18 14:02:00 2010 -0500 12316 12317 RS600: add connector quirk 12318 12319 System lists DVI port as HDMI. 12320 12321 fixes fdo bug 26605 12322 12323 commit a3b730eceb522c7ac1ef3dd6f6c7d773118d03f7 12324 Author: Jerome Glisse <jglisse (a] redhat.com> 12325 Date: Thu Feb 18 14:13:48 2010 +0100 12326 12327 r6xx/kms: when reseting BO force default state emission 12328 12329 In KMS world each cs need to fully initialize the 3D engine 12330 when we were erasing cs in IBDiscard for r6xx we weren't 12331 forcing reemission of default state. This patch force this 12332 emission. Fix : 12333 12334 http://bugs.freedesktop.org/show_bug.cgi?id=26603 12335 http://bugzilla.kernel.org/show_bug.cgi?id=15284 12336 12337 commit 579cdcf9b4e38c791a497b747a055fc0a07d8dd6 12338 Author: Alex Deucher <alexdeucher (a] gmail.com> 12339 Date: Wed Feb 17 12:22:48 2010 -0500 12340 12341 radeon: add ZaphodHeads option 12342 12343 Allows users that want to use zaphod mode to select 12344 which xrandr outputs are assigned to which head. E.g., 12345 12346 Option "ZaphodHeads" "LVDS,VGA-0" 12347 will assign LVDS to the first zaphod driver instance 12348 and VGA-0 to the second instance. 12349 12350 commit 47136fa347d1756523239746b4c74cd5278a1118 12351 Author: Michel Dnzer <daenzer (a] vmware.com> 12352 Date: Mon Feb 15 16:44:39 2010 +0100 12353 12354 EXA: Fix order of coordinates passed to radeon_pick_best_crtc(). 12355 12356 The wrong order prevented tearing avoidance from working with EXA/DRI2. 12357 12358 commit 6e95905f759b844373be62ec513c78153ed00222 12359 Author: Michel Dnzer <daenzer (a] vmware.com> 12360 Date: Mon Feb 15 17:02:16 2010 +0100 12361 12362 Allocate BOs for 'video offscreen surfaces' and 'TV-in' in VRAM. 12363 12364 I think these can only really work in VRAM, though right now there's probably 12365 nothing to prevent them from getting evicted... 12366 12367 commit 78e7047c5235b09858b66dd3688d39aaa27d7589 12368 Author: Pauli Nieminen <suokkos (a] gmail.com> 12369 Date: Mon Feb 15 13:40:37 2010 +0200 12370 12371 Allocate Xv buffers to GTT. 12372 12373 KMS doesn't have acceleration for upload to vram. memcpy/memmove to VRAM 12374 directly is very slow (40M/s in benchmark) which causes visible problems 12375 to video. 12376 12377 Allocating video buffer in GTT will give good performance (350-450M/s) 12378 for memmove operation. This is nice performance boost for Xv under KMS. 12379 12380 Signed-off-by: Pauli Nieminen <suokkos (a] gmail.com> 12381 12382 commit 3ec25e59854b6b03ad763bc374d3475a50f562d8 12383 Author: Pauli Nieminen <suokkos (a] gmail.com> 12384 Date: Mon Feb 15 13:45:28 2010 +0200 12385 12386 Use memcpy instead of memove for RADEONSwapCopy. 12387 12388 memcpy has about 25% better performance than memmove when 12389 destination is GTT (wc caching). Changing memmove to memcpy 12390 in SwapCopy will reduce CPU time spent moving data between 12391 host and GPU. 12392 12393 memcpy will specially help Xv when playing high resolution 12394 videos. 12395 12396 The swap path doesn't support overlapping memory copies either 12397 so extra safety is not worth the performance hit in commonly 12398 used code path. 12399 12400 Signed-off-by: Pauli Nieminen <suokkos (a] gmail.com> 12401 12402 [ Michel Dnzer: Fixed up whitespace ] 12403 12404 commit 221ef11b31756deb7134801730e76c040e841f5c 12405 Author: Alex Deucher <alexdeucher (a] gmail.com> 12406 Date: Fri Feb 12 14:15:10 2010 -0500 12407 12408 r6xx/r7xx: implement EXA vline support 12409 12410 commit 8ad4025affe1fb2f417e3a3031d74f83be5df253 12411 Author: Alex Deucher <alexdeucher (a] gmail.com> 12412 Date: Fri Feb 12 11:34:47 2010 -0500 12413 12414 radeon: consolidate crtc selector for vline wait 12415 12416 Use the Xv version as it takes into account the area 12417 covered by the op rather than just picking the largest 12418 crtc area. 12419 12420 commit 4f9d1714a7382594b834d446bbe502663f6a2778 12421 Author: Alex Deucher <alexdeucher (a] gmail.com> 12422 Date: Fri Feb 5 12:27:51 2010 -0500 12423 12424 r500: fix relocs for Xv planar video 12425 12426 fixes fdo bug 26445 12427 12428 commit b7ca1ab11ac0e4e4be5a1e7789ae8633b460a2d4 12429 Author: Dave Airlie <airlied (a] redhat.com> 12430 Date: Fri Feb 5 13:37:20 2010 +1000 12431 12432 rs400/rs480: mc idle bit is bit 2 like original radeon 12433 12434 backport from a KMS fix, the rs400/480 mc idle is bit 2 not bit 4. 12435 12436 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 12437 12438 commit 97c387e269672b146a95b45fbef5c8c18e473e36 12439 Author: Gaetan Nadon <memsize (a] videotron.ca> 12440 Date: Fri Jan 29 16:08:17 2010 -0500 12441 12442 config: remove dead LINUXDOC macro usage 12443 12444 This module does not generate LINUXDOC documentation. 12445 12446 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 12447 12448 commit 05fa2dee0c63d56fd523ee1ebab479bcb9f5e89d 12449 Author: Alex Deucher <alexdeucher (a] gmail.com> 12450 Date: Thu Feb 4 13:13:58 2010 -0500 12451 12452 evergreen: minor clean up in transmitter setup 12453 12454 both link and encoder must be set for linkb 12455 12456 commit 76eea5e02776f7beeb8f4491a374bd36792eda92 12457 Author: Alex Deucher <alexdeucher (a] gmail.com> 12458 Date: Thu Feb 4 10:52:14 2010 -0500 12459 12460 r600: reduce number of cache flushes 12461 12462 We don't need to flush so often. Next step 12463 would be to move the flushing to the drm and only 12464 flush after each command buffer rather than each 12465 draw. 12466 12467 commit 8d63d70f7ebaf9d250f0449d3720ef47516c05df 12468 Author: Alex Deucher <alexdeucher (a] gmail.com> 12469 Date: Wed Feb 3 15:22:22 2010 -0500 12470 12471 evergreen: use external clock source for DP PHY 12472 12473 DP CRTC clock always comes from DCPLL. This frees 12474 up PPll1/2 for non-DP-mode PHYs and CRTCs 12475 12476 commit ecbc26431914216a8b207e81451282ea07c8b92f 12477 Author: Alex Deucher <alexdeucher (a] gmail.com> 12478 Date: Wed Feb 3 03:52:07 2010 -0500 12479 12480 evergreen: blank/unblank DP in dpms calls 12481 12482 commit 710a2fd07235349f084ec40626cc28fbae523d27 12483 Author: Alex Deucher <alexdeucher (a] gmail.com> 12484 Date: Wed Feb 3 03:00:24 2010 -0500 12485 12486 evergreen: pll fixes 12487 12488 SetPixelClock needs to be called for DP outputs with a 12489 valid PLL. 12490 12491 commit 9ab2377ee23c5dc360ba135a77aa6b181a1862e6 12492 Author: Alex Deucher <alexdeucher (a] gmail.com> 12493 Date: Tue Feb 2 15:15:36 2010 -0500 12494 12495 evergreen: DP requires coherent mode 12496 12497 commit 634da80056525a48ec17ffb81e2fb59b168bea9e 12498 Author: Alex Deucher <alexdeucher (a] gmail.com> 12499 Date: Tue Feb 2 15:09:57 2010 -0500 12500 12501 evergreen: fix units on frac_fb_div 12502 12503 commit a887818f491f6c7315c56c4e0d0b702c4c6aa4ac 12504 Author: Alex Deucher <alexdeucher (a] gmail.com> 12505 Date: Mon Feb 1 11:01:47 2010 -0500 12506 12507 evergreen: add pci ids 12508 12509 commit cb2772b69480268c059766c4f6b209ce590ede0e 12510 Author: Alex Deucher <alexdeucher (a] gmail.com> 12511 Date: Fri Jan 29 12:59:46 2010 -0500 12512 12513 evergreen: add atombios crtc/pll functions 12514 12515 commit bd8e04cb7b39f38b6958273582a9b324a9f0759a 12516 Author: Alex Deucher <alexdeucher (a] gmail.com> 12517 Date: Mon Feb 1 10:07:43 2010 -0500 12518 12519 evergreen: add atom support for digital outputs 12520 12521 analog is already supported by the existing code. 12522 12523 commit 8ad40d3c32ad5b91725bd37fcade6bed504df421 12524 Author: Alex Deucher <alexdeucher (a] gmail.com> 12525 Date: Tue Jan 26 15:39:44 2010 -0500 12526 12527 evergreen: add crtc set base/format support 12528 12529 commit 2c9cf0a07ff9e5a4989861bc2fdfd71d841013a8 12530 Author: Alex Deucher <alexdeucher (a] gmail.com> 12531 Date: Thu Dec 10 15:29:14 2009 -0500 12532 12533 evergreen: add lut support 12534 12535 commit 65246545c3dd9bfef73e8a0f200bd5909b89a167 12536 Author: Alex Deucher <alexdeucher (a] gmail.com> 12537 Date: Tue Jan 26 12:10:02 2010 -0500 12538 12539 evergreen: add hw cursor support 12540 12541 commit c05cad56b69d239fa2e69905d15f4f08b9db4c55 12542 Author: Alex Deucher <alexdeucher (a] gmail.com> 12543 Date: Thu Jan 28 01:28:52 2010 -0500 12544 12545 evergreen: add base asic support 12546 12547 commit ed63e1b1abe8810b5da6b4140892337eef08a9ea 12548 Author: Alex Deucher <alexdeucher (a] gmail.com> 12549 Date: Fri Jan 29 12:42:33 2010 -0500 12550 12551 evergreen: add register and utility defines 12552 12553 commit 019260ec4a9f9fbf2ac63a2ca3314aa308471f7e 12554 Author: Alex Deucher <alexdeucher (a] gmail.com> 12555 Date: Thu Dec 10 14:57:39 2009 -0500 12556 12557 evergreen: add chip enums 12558 12559 commit e86ae564d8b94d2cc8d3c7cceb452a197ab6387a 12560 Author: Alex Deucher <alexdeucher (a] gmail.com> 12561 Date: Mon Feb 1 12:03:23 2010 -0500 12562 12563 atom: update to the latest upstream atombios.h changes 12564 12565 commit 77b13a02c70842a58e0590d0243f0ae016c5a640 12566 Author: Alex Deucher <alexdeucher (a] gmail.com> 12567 Date: Mon Feb 1 12:16:20 2010 -0500 12568 12569 r500: fix CS section size missmatch in Xv code 12570 12571 commit 5c11264517d1b9808415d47e9dedc345610f0f86 12572 Author: Dave Airlie <airlied (a] linux.ie> 12573 Date: Fri Jan 29 15:10:33 2010 +1000 12574 12575 radeon: further digital encoder cleanups. 12576 12577 this just cleans up a few more bits of the digital encoder setup. 12578 12579 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 12580 12581 commit 872a11fa9efb0b76cef83089ea1bb80792b94930 12582 Author: Alex Deucher <alexdeucher (a] gmail.com> 12583 Date: Thu Jan 28 01:19:23 2010 -0500 12584 12585 radeon: clean up dig encoder handling 12586 12587 - switch the var name to dig_encoder 12588 - quiet coherent messages 12589 - clean up dig encoder selection 12590 12591 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 12592 12593 commit c1a73ff51b371ecc6f57a81a33f644344b4e3f70 12594 Author: Dave Airlie <airlied (a] linux.ie> 12595 Date: Thu Jan 28 12:58:29 2010 +1000 12596 12597 radeon: add support for picking the digitial encoder 12598 12599 Testing on a W500 laptop found the LVDS + DisplayPort would conflict 12600 over the choice of digital encoder, and only one would work. 12601 12602 This patch fixes the DCE3 case where LVTMA requires the second digital 12603 encoder and lets DP/DVI use the first in most cases. 12604 12605 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 12606 12607 commit 95b56275d80dcee48a7927df124602c34cd72235 12608 Author: Alex Deucher <alexdeucher (a] gmail.com> 12609 Date: Wed Jan 27 15:48:25 2010 -0500 12610 12611 avivo: add some hotplug detect regs 12612 12613 commit 5a4327f7784361933484895c9af751ccfa242d48 12614 Author: Alex Deucher <alexdeucher (a] gmail.com> 12615 Date: Tue Jan 26 16:06:00 2010 -0500 12616 12617 radeon: minor pll updates 12618 12619 add new fixed post divider option 12620 12621 commit cbd642af7c76469d701471daea0c8d167567ccf8 12622 Author: Alex Deucher <alexdeucher (a] gmail.com> 12623 Date: Tue Jan 26 12:03:45 2010 -0500 12624 12625 radeon: use mmio bar size rather than hardcoded number for register ops 12626 12627 newer asics have larger mmio bars 12628 12629 commit e5933fd74923a07aa401bdb4ecd8d1d641148440 12630 Author: Alex Deucher <alexdeucher (a] gmail.com> 12631 Date: Mon Jan 25 13:18:47 2010 -0500 12632 12633 KMS: fix no accel option 12634 12635 NO_ACCEL wasn't handled correctly with KMS. 12636 12637 fixes fdo bug 26198 12638 12639 commit ba99a82ced1a6101365876723d22bcfec4607537 12640 Author: Dave Airlie <airlied (a] redhat.com> 12641 Date: Mon Jan 25 13:50:13 2010 +1000 12642 12643 radeon/kms: only enable EXA pixmaps under kms if we support render accel 12644 12645 I've seen RN50s with 64MB of RAM that are slow as molasses, this 12646 should fix them. 12647 12648 RH bug 556400 12649 12650 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 12651 12652 commit de40d36264364434541377c7f65f3836dc514fc0 12653 Author: Kusanagi Kouichi <slash (a] ac.auone-net.jp> 12654 Date: Fri Jan 22 17:18:51 2010 +0900 12655 12656 Remove unnecessary checks. 12657 12658 Signed-off-by: Kusanagi Kouichi <slash (a] ac.auone-net.jp> 12659 12660 commit 30a19b75cc82b8e04c45e6684b84f9a4ccc0505b 12661 Author: Alan Coopersmith <alan.coopersmith (a] sun.com> 12662 Date: Tue Jan 19 14:54:04 2010 -0800 12663 12664 Fix configure with --disable-dri 12665 12666 AM_CONDITIONAL calls can't be made inside if statements, since 12667 automake won't know what to do when the if isn't true. 12668 12669 Signed-off-by: Alan Coopersmith <alan.coopersmith (a] sun.com> 12670 12671 commit bd011784c0609f1fd03dcb784744c8f5cdef33da 12672 Author: Alex Deucher <alexdeucher (a] gmail.com> 12673 Date: Tue Jan 19 11:59:38 2010 -0500 12674 12675 ATOM: Upstream parser updates 12676 12677 commit 95d63e408cc88b6934bec84a0b1ef94dfe8bee7b 12678 Author: Jerome Glisse <jglisse (a] redhat.com> 12679 Date: Mon Jan 18 10:46:50 2010 +0100 12680 12681 r6xx/r7xx: emit relocation for FRAG & TILE buffer 12682 12683 FRAG & TILE buffer are unused but still they need 12684 to be associated with a valid relocation so that 12685 userspace can't try to abuse them to overwritte 12686 GART and then try to write anywhere in system 12687 memory. 12688 12689 commit bbaf71fb46e3e0acdadcb91e9b1c73396bd9aa8e 12690 Author: Alan Coopersmith <alan.coopersmith (a] sun.com> 12691 Date: Fri Jan 15 15:09:27 2010 -0800 12692 12693 Update Sun license notices to current X.Org standard form 12694 12695 Signed-off-by: Alan Coopersmith <alan.coopersmith (a] sun.com> 12696 12697 commit 3d158716a60d5a113a541ea2f680b81a1be41ad2 12698 Author: Dave Airlie <airlied (a] redhat.com> 12699 Date: Thu Jan 14 11:21:33 2010 +1000 12700 12701 displayport: fix DDC on DVI->DP convertors. 12702 12703 Seems we have to not do auxch DDC if we aren't talking to a DP 12704 sink. 12705 12706 commit 6a363f68415d37c302151581f2a86855dba39b67 12707 Author: Alex Deucher <alexdeucher (a] gmail.com> 12708 Date: Mon Jan 11 15:13:45 2010 -0500 12709 12710 rv100: fix typo in fix for bug 25992 12711 12712 Noticed by Maarten Maathuis. 12713 12714 commit 7b01e1ee29f681bf1735ecded6445d12beeb52d8 12715 Author: Alex Deucher <alexdeucher (a] gmail.com> 12716 Date: Mon Jan 11 12:40:06 2010 -0500 12717 12718 rv100: reject modes >135 Mhz with DVI 12719 12720 Due to heat issues. fixes bug 25992 12721 12722 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 12723 12724 commit 74da9a66a420a40bb5d562f07e97ce6ac2eb010f 12725 Author: Alex Deucher <alexdeucher (a] gmail.com> 12726 Date: Fri Jan 8 16:51:43 2010 -0500 12727 12728 radeon/kms: add new strings for eDP and TV 12729 12730 commit 61977f61f516b54d1fd9df34d0f0892fbf5227d7 12731 Author: Alex Deucher <alexdeucher (a] gmail.com> 12732 Date: Thu Jan 7 02:15:31 2010 -0500 12733 12734 radeon: fallback to i2c for edid if aux fails 12735 12736 In case of a DP->DVI adapter for example. 12737 12738 commit 958d073869404f60e56dc0cc70b3e7de85904694 12739 Author: Alex Deucher <alexdeucher (a] gmail.com> 12740 Date: Thu Jan 7 02:12:00 2010 -0500 12741 12742 radeon: add initial support for eDP (embedded DisplayPort) 12743 12744 Should fix fdo bug 25931 12745 12746 commit 944ae73f7cd5f9a1622b38e06a58565812cb102f 12747 Author: Alex Deucher <alexdeucher (a] gmail.com> 12748 Date: Thu Jan 7 01:52:35 2010 -0500 12749 12750 radeon: updated ObjectID.h 12751 12752 commit 48aa5064aff4b9adf768e480df2312d4375e9c40 12753 Author: Alex Deucher <alexdeucher (a] gmail.com> 12754 Date: Mon Jan 4 11:05:44 2010 -0500 12755 12756 radeon: fix LVDS power sequence on Mac cards 12757 12758 Noticed by John R. Dunning. Fix taken from radeonfb. 12759 12760 I'm not sure if this sequence would be useful on any PC 12761 laptops or not so make it mac specific for now. 12762 12763 commit 36bd69affc996c92c40b7360a7fbaa1a3a46abfd 12764 Author: Gaetan Nadon <memsize (a] videotron.ca> 12765 Date: Wed Dec 16 15:52:18 2009 -0500 12766 12767 configure.ac: remove unused sdkdir=$(pkg-config...) statement 12768 12769 The sdkdir variable isn't use, so remove the statement. 12770 12771 Acked-by: Dan Nicholson <dbn.lists (a] gmail.com> 12772 12773 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 12774 12775 commit 9d0f3af7278dc939fd4e6f3ea69d9f488a9fbed7 12776 Author: Alex Deucher <alexdeucher (a] gmail.com> 12777 Date: Tue Dec 22 08:45:27 2009 -0500 12778 12779 radeon: add cvt timing if we only have panel w/h 12780 12781 fixes mac laptops without an edid 12782 12783 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 12784 12785 commit 6e1f5553c6d7e3b5d089af2e3d587efe95936855 12786 Author: Alex Deucher <alexdeucher (a] gmail.com> 12787 Date: Mon Dec 21 17:32:40 2009 -0500 12788 12789 ATOM: add new power table defs 12790 12791 commit 4b05c47ac657f9a93d76221269761ed64c81f716 12792 Author: Dave Airlie <airlied (a] redhat.com> 12793 Date: Thu Dec 17 14:30:23 2009 +1000 12794 12795 radeon: use new libdrm api 12796 12797 commit 1d620800bd79290967fa487decf798e318c45f25 12798 Author: Ingmar Vanhassel <ingmar (a] exherbo.org> 12799 Date: Sun Dec 20 21:04:56 2009 +0100 12800 12801 Replace shave with automake silent-rules 12802 12803 Which are enabled by default in Xorg's util-macros 1.3 if automake-1.11 12804 is used. 12805 12806 Signed-off-by: Ingmar Vanhassel <ingmar (a] exherbo.org> 12807 12808 commit 52f6a2ff682ad1f8e26ef1cd7d380493dcd57752 12809 Author: Gaetan Nadon <memsize (a] videotron.ca> 12810 Date: Wed Dec 16 15:29:53 2009 -0500 12811 12812 src/Makefile.am: missing 3 header files in tarball 12813 12814 radeon_dummy_bufmgr.h 12815 radeon_vbo.h 12816 simple_list.h 12817 Replace leading spaces with tab for radeon_atomwrapper.h 12818 12819 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 12820 12821 commit 3a30210d50b27f8772fc5045133940246764fce9 12822 Author: Matthias Hopf <mhopf (a] suse.de> 12823 Date: Tue Dec 15 10:53:48 2009 -0500 12824 12825 fix 200M freezes on VT switch if CRTC is disabled 12826 12827 It appears that RS4xx chips need to have the crtc 12828 enabled when the timing is programmed. 12829 12830 agd5f: minor fixes/cleanup of the original patch 12831 12832 commit 299d395bd3f294239dee58ab7d607d7d2c657f61 12833 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 12834 Date: Sun Dec 13 14:51:17 2009 -0800 12835 12836 dri: be more useful if disabling DRI due to KMS/UMS mismatch. 12837 12838 It's only fair. 12839 12840 Compile-tested only. 12841 12842 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 12843 12844 commit 0e5c9d87b5d7e0751df71cc8958ca5ccaed25104 12845 Author: Alex Deucher <alexdeucher (a] gmail.com> 12846 Date: Thu Dec 10 14:27:43 2009 -0500 12847 12848 ATOM: split set base/format into separate functions 12849 12850 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 12851 12852 commit 46630da5fd6f45bb8ea150b870162997480d69c7 12853 Author: Alex Deucher <alexdeucher (a] gmail.com> 12854 Date: Wed Dec 9 12:55:25 2009 -0500 12855 12856 AVIVO: add new PLL code 12857 12858 This should hopefully help the problems with flickering 12859 and blinking monitors reported on some systems. If there 12860 are problems, the old PLL algorithm can be selected with: 12861 Option "NewPLL" "FALSE" 12862 in the device section of your X config. 12863 12864 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 12865 12866 commit f082b1693d6f7f763ccf5a8436a89890ca2c6129 12867 Author: Matthijs Kooijman <matthijs (a] stdin.nl> 12868 Date: Thu Dec 3 12:10:14 2009 -0500 12869 12870 radeon: fix crtc2 dpms 12871 12872 noticed by Matthijs Kooijman on fdo bug 22140 12873 12874 commit 3a96fbf3b84522cf1ba1b176e82b662222331c14 12875 Author: Alex Deucher <alexdeucher (a] gmail.com> 12876 Date: Wed Dec 2 18:02:33 2009 -0500 12877 12878 radeon: only read RADEON_PPLL_REF_DIV for ref div on pre-avivo 12879 12880 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 12881 12882 commit f03450796d2e9247a1228c4e2abb1dfad7aecddf 12883 Author: Kusanagi Kouichi <slash (a] ac.auone-net.jp> 12884 Date: Wed Dec 2 01:36:37 2009 -0500 12885 12886 radeon: Lift hardcoded limit from RADEONQueryImageAttributes 12887 12888 The dimension of an XvImage is limited to 2048 x 2048 even if an adaptor 12889 supports larger image. 12890 XvCreateImage and XvShmCreateImage lower the width or height of an image. 12891 XvPutImage and XvShmPutImage return BadValue. 12892 The cause is that 2048 is hardcoded in RADEONQueryImageAttributes. 12893 12894 commit efbc2c80ab02879edf3b7b3d65b16c45ddce5017 12895 Merge: 88a50a30d a612813c5 12896 Author: Dave Airlie <airlied (a] redhat.com> 12897 Date: Wed Dec 2 14:30:36 2009 +1000 12898 12899 Merge remote branch 'origin/displayport' 12900 12901 commit 88a50a30df11a06263209340a42251851f8e2334 12902 Author: Dave Airlie <airlied (a] redhat.com> 12903 Date: Mon Nov 30 15:32:12 2009 +1000 12904 12905 r600: fix multi-operation in single batch support. 12906 12907 This ports the mesa DMA buffer handling with the 3 lists, 12908 12909 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 12910 12911 commit b2597deea3a3953ff50d54ff37e3c043eac409f4 12912 Author: Kusanagi Kouichi <slash (a] ac.auone-net.jp> 12913 Date: Mon Nov 30 11:10:50 2009 -0500 12914 12915 R600: fix some warnings 12916 12917 commit 4d90dc3cb248e9d61c0c490bba80f6727fafd318 12918 Author: Alex Deucher <alexdeucher (a] gmail.com> 12919 Date: Mon Nov 30 11:05:38 2009 -0500 12920 12921 atom: pull misc mode info for lvds panel mode from bios tables 12922 12923 sync polarity, etc. This will likely fix LVDS problems 12924 on some laptops. 12925 12926 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 12927 12928 commit a612813c5f329f00a9271e7bb69abd2630a49403 12929 Author: Alex Deucher <alexdeucher (a] gmail.com> 12930 Date: Fri Nov 27 17:20:30 2009 -0500 12931 12932 radeon: clean up DP code 12933 12934 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 12935 12936 commit a8dbf7c23481501987971a9c0b6cb0760f86127f 12937 Author: Dave Airlie <airlied (a] redhat.com> 12938 Date: Fri Nov 27 10:16:00 2009 +1000 12939 12940 r600/accel: cleanup and merge a lot of the accel functions 12941 12942 All of the drawing ops were the exact same modulo the vtx size, 12943 this along with the vertex buffer wrapping code could all be consolidated 12944 into a smaller set of functions. 12945 12946 This also adds 2 VBO which we switch between, and merges a #define to 12947 enable the multiple operations in one CS under KMS mode. 12948 12949 Multi-operation still isn't working though. 12950 12951 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 12952 12953 commit 971e463b2cfdc1ba1c78a24fb439d33dd9155dfc 12954 Author: Alex Deucher <alexdeucher (a] gmail.com> 12955 Date: Thu Nov 26 12:45:15 2009 -0500 12956 12957 dce3: CV1/TV1OutputControl tables didn't go away until dce3.2 12958 12959 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 12960 12961 commit 95385f0906f371a1ed6e60eb3e597e699c7a3222 12962 Author: Jerome Glisse <jglisse (a] redhat.com> 12963 Date: Thu Nov 26 12:21:21 2009 +0100 12964 12965 kms: Fix resizing when acceleration is disabled 12966 12967 When acceleration is disabled we need to reallocate 12968 a new shadow framebuffer and we should also avoid 12969 calling any EXA function as EXA is disabled in such 12970 case. 12971 12972 commit 8195385d3cedf70d89ca576371af6bca5045bbd2 12973 Author: Dave Airlie <airlied (a] redhat.com> 12974 Date: Thu Nov 26 15:43:54 2009 +1000 12975 12976 dce3: no CV1 or TV1 Output Control tables anymore. 12977 12978 According to the DCE3 docs we should only use DAC1/2 not 12979 CV1/TV OutputControl, also my rv730 bios doesn't have any 12980 CV1/TV tables. 12981 12982 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 12983 12984 commit 381b14d5422131ea8ff942ab06a328f6beca0d3f 12985 Author: Dave Airlie <airlied (a] redhat.com> 12986 Date: Thu Nov 26 15:28:15 2009 +1000 12987 12988 atombios: CV mode retrival was broken. 12989 12990 a) crev and frev are reversed 12991 b) my rv730 bios only has one mode in it, so 12992 bounds check the table size. 12993 12994 commit a67a13c315a9235be052662e2a3270686cccc0f0 12995 Author: Dave Airlie <airlied (a] redhat.com> 12996 Date: Thu Nov 26 13:41:36 2009 +1000 12997 12998 radeon: fix typo in unused as of yet CV code 12999 13000 commit cd9bfb5892eebef6dff14ac414dff074db5ddee4 13001 Author: Dave Airlie <airlied (a] redhat.com> 13002 Date: Thu Nov 26 06:43:13 2009 +1000 13003 13004 fixes displayport for me here 13005 13006 commit 256de862a3c017dcb4ed0d8689294a552eda0d18 13007 Author: Dave Airlie <airlied (a] redhat.com> 13008 Date: Thu Nov 26 06:35:39 2009 +1000 13009 13010 displayport: cleanup call dp from dpms on instead. 13011 13012 commit 05551295c5e0946745163f17e5c1d3d41b94bcbf 13013 Author: Kristian Hgsberg <krh (a] bitplanet.net> 13014 Date: Tue Nov 24 21:09:03 2009 -0500 13015 13016 dri2: Use drmGetDeviceNameFromFd() instead of open coded loop. 13017 13018 commit 0061c4db1d3aecdca13efb5133ab8784dc37df95 13019 Author: Dave Airlie <airlied (a] redhat.com> 13020 Date: Wed Nov 25 15:39:10 2009 +1000 13021 13022 Revert "r600: enable multiple operations in one CS" 13023 13024 turn this off for now, on my rv635 desktop, I started getting blanks 13025 in places in firefox and the odd bit of font corruption, need to 13026 track that down. 13027 13028 This reverts commit 985a065518b1d33599de33f7fe082d3302db58a6. 13029 13030 commit 8b28534bcf877557a5681fa3b4f107c74615d3de 13031 Author: Dave Airlie <airlied (a] redhat.com> 13032 Date: Wed Nov 25 13:24:44 2009 +1000 13033 13034 radeon/exa: change option to determine exa pixmap usage. 13035 13036 This moves to a boolean instead of using VRAM sizing. 13037 13038 as per Michel's suggestions on list. 13039 13040 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 13041 13042 commit f0acb16fa19844453adc6db3399977fba7c8a0db 13043 Author: Dave Airlie <airlied (a] redhat.com> 13044 Date: Wed Nov 25 12:04:47 2009 +1000 13045 13046 r600/xv: drop inited 3d false in xv code 13047 13048 commit 985a065518b1d33599de33f7fe082d3302db58a6 13049 Author: Dave Airlie <airlied (a] redhat.com> 13050 Date: Wed Nov 25 11:54:08 2009 +1000 13051 13052 r600: enable multiple operations in one CS 13053 13054 This switches on multiple ops in a single CS under KMS/DRI2. 13055 13056 It gets for on a Pentium D 3 + rv740 from 330,000 to 500,000 13057 with x11perf -aa10text. 13058 13059 It also knocks a couple of seconds of gtkperf -a 13060 13061 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 13062 13063 commit af816ac752820255f245793b53a7cca5a4a49cd4 13064 Author: Dave Airlie <airlied (a] redhat.com> 13065 Date: Wed Nov 25 11:53:07 2009 +1000 13066 13067 r600: fixup problems with EXA operation reset for multiple ops 13068 13069 To put multiple ops into one CS, you can't just discard the whole 13070 IB. This add supports for reset the CS cdw to the correct place 13071 after an op discards. 13072 13073 Still doesn't enable the final accel bits. 13074 13075 commit 3d8dcbc29323a3c644100bec13aa93f024653bd3 13076 Author: Dave Airlie <airlied (a] redhat.com> 13077 Date: Wed Nov 25 11:43:57 2009 +1000 13078 13079 kms: flush cs on close screen. 13080 13081 this is needed for server recycle. 13082 13083 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 13084 13085 commit 19f1a357944f9d8f4567a2691a68067ec033ccb7 13086 Author: Dave Airlie <airlied (a] redhat.com> 13087 Date: Wed Nov 25 11:41:52 2009 +1000 13088 13089 radeon: fix check for no work in operation 13090 13091 commit 3a460a14b9603159f10d89da27b559c36a184e27 13092 Author: Dave Airlie <airlied (a] redhat.com> 13093 Date: Wed Nov 25 10:33:17 2009 +1000 13094 13095 r600: refactor code to help future acceleration speedups. 13096 13097 This changes the vertex buffer index to be an offset, and 13098 records the start of the vb for each operation and uses 13099 that to set the operations up. 13100 13101 This still flushes after each operation to make sure we have 13102 no regressions in non-kms/kms cases. 13103 13104 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 13105 13106 commit 797a3f0c71c94477eec565ea2c95553c6f66d9fd 13107 Author: Dave Airlie <airlied (a] redhat.com> 13108 Date: Wed Nov 25 10:07:59 2009 +1000 13109 13110 kms: allow prepare copy to fail without dying 13111 13112 commit ba76acc76e0d0a51a1c488e4d5494a71e155a05b 13113 Author: Dave Airlie <airlied (a] redhat.com> 13114 Date: Tue Nov 24 15:38:47 2009 +1000 13115 13116 radeon/kms: drop special r600 indirect flush 13117 13118 This just merges it with the main kms cs flush 13119 13120 commit 3c527d752d3cced219c6113acb932511b16e35fb 13121 Author: Dave Airlie <airlied (a] redhat.com> 13122 Date: Tue Nov 24 15:11:12 2009 +1000 13123 13124 r600: enable mixed pixmaps 13125 13126 commit b9eb2380df5145c85b2ef694a04628c055c47429 13127 Author: Alex Deucher <alexdeucher (a] gmail.com> 13128 Date: Mon Nov 23 12:12:02 2009 -0500 13129 13130 DCE3.2: fix uniphy2 dvi issues 13131 13132 In some cases the atom transmitter table sets the 13133 golden value of this reg differently which some monitors 13134 don't like. I haven't had time to dig further, so this 13135 works around it for now. 13136 13137 Fixes fdo bug 24313 13138 13139 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13140 13141 commit d0dd5122d8d75b7c4a72df07dc7f562b2e75dd5e 13142 Author: Gaetan Nadon <memsize (a] videotron.ca> 13143 Date: Mon Nov 23 09:25:05 2009 -0500 13144 13145 Makefile.am: add ChangeLog and INSTALL on MAINTAINERCLEANFILES 13146 13147 Now that the INSTALL file is generated. 13148 Allows running make maintainer-clean. 13149 13150 commit 3d7888aa0493f46d826f4d80dea36f1f642f3c6b 13151 Author: Dave Airlie <airlied (a] redhat.com> 13152 Date: Mon Nov 23 15:26:15 2009 +1000 13153 13154 radeon: fix warning in printf 13155 13156 commit 3a1a8b7b73424a769baf917bba2223ea6515b596 13157 Author: Dave Airlie <airlied (a] redhat.com> 13158 Date: Mon Nov 23 15:07:10 2009 +1000 13159 13160 radeon: surround mixed pixmaps usage with ifdef check for old servers 13161 13162 commit e20af9c94982ec6487dae42c141d33cf6a7a2eb4 13163 Author: Dave Airlie <airlied (a] redhat.com> 13164 Date: Fri Nov 20 12:15:02 2009 +1000 13165 13166 radeon: avoid using hw pixmaps when we have little VRAM. 13167 13168 This patch returns NULL for pixmap creation when we are using 13169 mixed pixmaps and the pixmap has a size. 13170 13171 The size check is necessary for the front buffer. 13172 13173 We add a flag to force pixmap creation for certain pixmaps 13174 that need to be hw, like the DRI2 and Xv ones. 13175 13176 Idea from Michel and workarounds from Ben Skeggs. 13177 13178 v2: add Option "EXALowVRAM" to allow configuring this, value in MBs. 13179 13180 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 13181 13182 commit 543338842ba7fe7da4902947d386ea1cabd1cdf0 13183 Author: Alex Deucher <alexdeucher (a] gmail.com> 13184 Date: Fri Nov 20 16:38:43 2009 -0500 13185 13186 dp fix timing 13187 13188 aux channel atom command table delay is 10 usec units 13189 this gets the link training working, but alas, no image. 13190 13191 commit dd3eab848cf352bb96c3d01fe6028d8a4a8e451e 13192 Author: Alex Deucher <alexdeucher (a] gmail.com> 13193 Date: Fri Nov 20 01:06:49 2009 -0500 13194 13195 r300: render target limit is 2560 13196 13197 limit was increased to 4021 in r4xx. 13198 13199 fixes fdo bug 25191. 13200 13201 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13202 13203 commit df35d40d4ec260cdab57a9b36e4e96b2a6c2a827 13204 Author: Alex Deucher <alexdeucher (a] gmail.com> 13205 Date: Thu Nov 19 20:01:42 2009 -0500 13206 13207 DP: hack that makes DP work. 13208 13209 Something's wrong with the status check. 13210 13211 commit f42412ec4dbd21f562a7fe4239c0a1f17ef1b8f2 13212 Author: Alex Deucher <alexdeucher (a] gmail.com> 13213 Date: Thu Nov 19 19:40:54 2009 -0500 13214 13215 wip 13216 13217 commit eb9bc133fc426e67b397e661bfd22bf62009d9d3 13218 Author: Dave Airlie <airlied (a] redhat.com> 13219 Date: Fri Nov 20 09:23:31 2009 +1000 13220 13221 kms: recalculate the flush limits after screen resize. 13222 13223 When we resize the front buffer we need to reduce the flush limits 13224 appropriately. 13225 13226 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 13227 13228 commit f7f58ef4c042e492618665a6c5555e8e67387ab3 13229 Author: Dave Airlie <airlied (a] redhat.com> 13230 Date: Fri Nov 20 09:22:39 2009 +1000 13231 13232 radeon: r100/r200 have a 2047 scissor limit 13233 13234 We were overflowing this in my case with a 2704 width desktop, 13235 so videos were stopping around 700 bytes across the screen. 13236 13237 can I haz shatter already? 13238 13239 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 13240 13241 commit 54c09778d79de1516ee511d6d7d5b54dd4bf9130 13242 Author: Alex Deucher <alexdeucher (a] gmail.com> 13243 Date: Thu Nov 19 13:32:30 2009 -0500 13244 13245 DP wip 13246 13247 - use coherent 13248 - fix up lane/clock numbers 13249 13250 commit 02d017a0552db2c8a48506c5b26c48a1a2ab75b6 13251 Author: Gaetan Nadon <memsize (a] videotron.ca> 13252 Date: Wed Oct 28 14:41:41 2009 -0400 13253 13254 INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206 13255 13256 Automake 'foreign' option is specified in configure.ac. 13257 Remove from Makefile.am 13258 13259 commit 396ee9b7e7e3b1dd64532270c4e32cdb33857596 13260 Author: Gaetan Nadon <memsize (a] videotron.ca> 13261 Date: Wed Oct 28 14:09:09 2009 -0400 13262 13263 INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206 13264 13265 Add missing INSTALL file. Use standard GNU file on building tarball 13266 README may have been updated 13267 Remove AUTHORS file as it is empty and no content available yet. 13268 Remove NEWS file as it is empty and no content available yet. 13269 13270 commit 1dcda0c4d5d506e18a7989ff05f2ad8569a73f23 13271 Author: Gaetan Nadon <memsize (a] videotron.ca> 13272 Date: Mon Oct 26 12:54:21 2009 -0400 13273 13274 Several driver modules do not have a ChangeLog target in Makefile.am #23814 13275 13276 The git generated ChangeLog replaces the hand written one. 13277 Update configure.ac to xorg-macros level 1.3. 13278 Use XORG_DEFAULT_OPTIONS which replaces four XORG_* macros 13279 Update Makefile.am to add ChangeLog target if missing 13280 Remove ChangeLog from EXTRA_DIST or *CLEAN variables 13281 This is a pre-req for the INSTALL_CMD 13282 13283 commit 5b517bcc353e7111d2365a213505f27cfae30924 13284 Author: Gaetan Nadon <memsize (a] videotron.ca> 13285 Date: Thu Oct 22 12:34:16 2009 -0400 13286 13287 .gitignore: use common defaults with custom section # 24239 13288 13289 Using common defaults will reduce errors and maintenance. 13290 Only the very small or inexistent custom section need periodic maintenance 13291 when the structure of the component changes. Do not edit defaults. 13292 13293 commit 605c869076985b69350c893cf2bbcc90de8dc6bf 13294 Author: Alex Deucher <alexdeucher (a] gmail.com> 13295 Date: Wed Nov 18 01:51:12 2009 -0500 13296 13297 more dp cleanup 13298 13299 - clean up handling of linkb, etc. 13300 - add support for ATOM_TRANSMITTER_ACTION_SETUP_VSEMPH 13301 to existing transmitter control function 13302 - fix lots of warnings 13303 13304 commit bce1911f66f446b793125ac8cd5463bf6292629e 13305 Author: Alex Deucher <alexdeucher (a] gmail.com> 13306 Date: Wed Nov 18 00:04:33 2009 -0500 13307 13308 radeon: pull previous displayport work into master 13309 13310 commit 437113124bbd6fb166825169eabec4dfde900dd9 13311 Author: Alex Deucher <alexdeucher (a] gmail.com> 13312 Date: Tue Nov 17 14:16:22 2009 -0500 13313 13314 radeon: deal with connectors sourced to the same encoder 13315 13316 Some systems have multiple connectors connected to the 13317 same encoder; e.g., DVI and HDMI connected to the same 13318 encoder with the same ddc line. Since we expose 13319 connectors as xrandr outputs, randr treats them separately 13320 which results in it trying to source the same encoder to 13321 different crtcs. If we have an HDMI and DVI-D port on the 13322 same encoder, pick the one to be considered connected 13323 based on the edid (HDMI if edid indicates HDMI, DVI 13324 otherwise). Also, don't turn off (dpms) encoders that are 13325 shared. 13326 13327 Fixes fdo bug 21767. 13328 13329 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13330 13331 commit 9b8605d4e2bf7703c7848f3e2022a8c7328a3acd 13332 Author: Alex Deucher <alexdeucher (a] gmail.com> 13333 Date: Tue Nov 17 11:45:09 2009 -0500 13334 13335 radeon: fix potential memory leak in ddc code 13336 13337 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13338 13339 commit ce8299962003de572122561a6f7f61eaccf633b2 13340 Author: Dave Airlie <airlied (a] redhat.com> 13341 Date: Tue Nov 17 15:23:30 2009 +1000 13342 13343 kms: check for 0 mask for cloned outputs. 13344 13345 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 13346 13347 commit 06d68f7fa3870f9601e837e1834d33e8395008b1 13348 Author: Dave Airlie <airlied (a] redhat.com> 13349 Date: Tue Nov 17 15:10:04 2009 +1000 13350 13351 kms: add cloning support. 13352 13353 We have to translate from the kernel encoder cloning to the randr 13354 connector cloning, we do this by constructing an encoder mask per output 13355 and an encoder cloning mask per output then comparing them. 13356 13357 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 13358 13359 commit 7587ce16ab0380337fe801f457c1d1d9b4141cc5 13360 Author: Alex Deucher <alexdeucher (a] gmail.com> 13361 Date: Mon Nov 16 19:16:03 2009 -0500 13362 13363 r600: remove un-needed format conversions 13364 13365 we do the normalization and xforms in the vertex 13366 shader. 13367 13368 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13369 13370 commit ba605b5c9ff48765f2b66960aa4cc81b6235d012 13371 Author: Alex Deucher <alexdeucher (a] gmail.com> 13372 Date: Mon Nov 16 19:00:52 2009 -0500 13373 13374 r600: fix num format in vtx fetch 13375 13376 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13377 13378 commit 0c4710c67a2fee2061fc3da43c9f908585693cfa 13379 Author: Alex Deucher <alexdeucher (a] gmail.com> 13380 Date: Thu Nov 12 11:11:17 2009 -0500 13381 13382 radeon: man page updates 13383 13384 commit 2af2744c80f3307e6d84edba9479be5bd36e1d03 13385 Author: Alex Deucher <alexdeucher (a] gmail.com> 13386 Date: Wed Nov 11 13:12:24 2009 -0500 13387 13388 radeon: fix support for external tmds on some legacy chips 13389 13390 Should fix bug 11801. 13391 13392 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13393 13394 commit f17657fdc83d8f4c0386d2c7dade98de5b94acbe 13395 Author: Alex Deucher <alexdeucher (a] gmail.com> 13396 Date: Mon Nov 9 16:59:11 2009 -0500 13397 13398 r600/r700: typo, fix mask of DB_ALPHA_TO_MASK 13399 13400 noticed by glisse. 13401 13402 commit 0ee7763fb33bf7fbc4ae1e8372cbc74578bfb720 13403 Author: Alex Deucher <alexdeucher (a] gmail.com> 13404 Date: Mon Nov 9 15:09:34 2009 -0500 13405 13406 IGP: some IGP chips report as AGP 13407 13408 Set bus type appropriately. fixes bug 25002 13409 13410 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13411 13412 commit da1fcddaade58e371c7bee555ce32944c689f810 13413 Author: Alex Deucher <alexdeucher (a] gmail.com> 13414 Date: Thu Nov 5 14:13:49 2009 -0500 13415 13416 radeon: properly fix unused label warnings 13417 13418 commit 6e496f46fdc90d0a6cef41e9d45a6458aaed1eca 13419 Author: Alex Deucher <alexdeucher (a] gmail.com> 13420 Date: Thu Nov 5 14:11:55 2009 -0500 13421 13422 Revert "Fix warning about unused defined labels in radeon_exa_funcs.c." 13423 13424 This reverts commit b6ac42cce2156880ac18b32e15acbff40b2f53e9. 13425 13426 This patch is very broken. 13427 13428 commit b6ac42cce2156880ac18b32e15acbff40b2f53e9 13429 Author: Tom Chvtal <scarabeus (a] gentoo.org> 13430 Date: Sat Oct 31 23:07:07 2009 +0100 13431 13432 Fix warning about unused defined labels in radeon_exa_funcs.c. 13433 13434 Signed-off-by: Tom Chvtal <scarabeus (a] gentoo.org> 13435 13436 commit 87d7235790866f9c19ef08972d5237f09d940cd9 13437 Author: Alex Deucher <alexdeucher (a] gmail.com> 13438 Date: Thu Nov 5 10:23:03 2009 -0500 13439 13440 EXA: fallback if no pMaskPicture->pDrawable 13441 13442 A solid or gradient mask could be used for blending 13443 the source picture onto the destination picture. 13444 13445 Fixes fdo bug 24838 13446 13447 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13448 13449 commit 873897c54429e230a23011a9dd2f1069f4eb7752 13450 Author: Alex Deucher <alexdeucher (a] gmail.com> 13451 Date: Thu Nov 5 00:51:11 2009 -0500 13452 13453 DCE3+: call transmitter init on mode set 13454 13455 Generally this is done at post, but might not always 13456 be done with softboot or for connectors on docking 13457 stations. 13458 13459 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13460 13461 commit 3028374488cc0f34942ac372c8d05cf15898a613 13462 Author: Alex Deucher <alexdeucher (a] gmail.com> 13463 Date: Wed Nov 4 18:48:03 2009 -0500 13464 13465 ATOM/DCE3+: fix up usPixelClock calculation for Transmitter tables 13466 13467 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13468 13469 commit 52973e8f2c4104de60d00c0e5c29a31f77bf3de7 13470 Author: Alex Deucher <alexdeucher (a] gmail.com> 13471 Date: Sun Nov 1 13:58:03 2009 -0500 13472 13473 radeon/atom: fix tv-out on r5xx 13474 13475 fixes bug 24796 13476 13477 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13478 13479 commit 19b4f5f9b647cd470616b8ec1d34f1e4afa357ad 13480 Author: Alex Deucher <alexdeucher (a] gmail.com> 13481 Date: Tue Oct 27 11:55:20 2009 -0400 13482 13483 ATOM: fix up DVO for DCE 3.x 13484 13485 AdjustDisplayPll takes care of this for us. 13486 13487 commit 5a0019126a57138ee506d9a66738c9e8b75cbb96 13488 Author: Alex Deucher <alexdeucher (a] gmail.com> 13489 Date: Tue Oct 27 11:39:06 2009 -0400 13490 13491 radeon: fix DVO on AVIVO chips 13492 13493 DVO generally requires 2x ppll on AVIVO cards due to the 13494 way the tmds chip is wired up. Because of this, disable 13495 cloning of DVO outputs. 13496 13497 fixes fdo bug 21857. 13498 13499 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13500 13501 commit 2a6eeec72d6bb04a32225883f431c1d2e8cff123 13502 Author: Alex Deucher <alexdeucher (a] gmail.com> 13503 Date: Tue Oct 27 11:18:44 2009 -0400 13504 13505 atom: loosen pll min output limits 13506 13507 Limiting the pll output range is a good thing generally as 13508 it limits the number of possible pll combinations for a given 13509 frequency presumably to the ones that work best on each card. 13510 That's why the limits are in the bios tables. However, certain 13511 duallink DVI monitors seem to like pll combinations that would 13512 be limited by this at least on pre-DCE 3.0 r6xx hardware. This 13513 might need to be adjusted per family or per clock range in the 13514 future. 13515 13516 Fixes fdo bug 24727. 13517 13518 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13519 13520 commit f0d9d80fee4176eaba9435a9539f29d0eefe2a87 13521 Author: Luca Tettamanti <kronos.it (a] gmail.com> 13522 Date: Sat Oct 24 16:45:23 2009 -0400 13523 13524 Disable color tiling on r600/r700 13525 13526 Not supported yet. 13527 13528 commit ac499c3c0ea92f83ebd8127bb8cb5de625de6c94 13529 Author: Alex Deucher <alexdeucher (a] gmail.com> 13530 Date: Fri Oct 23 10:48:55 2009 -0400 13531 13532 radeon: clamp the internal FB map to the aperture size v2 13533 13534 We don't use the invisible memory yet and on cards with 13535 large amounts of vram this can cause the top of GART 13536 calculation to overflow. 13537 13538 Fixes bug fdo bug 24301: 13539 http://bugs.freedesktop.org/show_bug.cgi?id=24301 13540 13541 v2: only clamp cards with more than 512 MB. This seems 13542 to cause problems on some older cards due to the way the 13543 drm and ddx set up the internal memory map. 13544 13545 commit e57b54daf318ff59315a1c9ed21934f288c5c782 13546 Author: Alex Deucher <alexdeucher (a] gmail.com> 13547 Date: Thu Oct 22 16:07:58 2009 -0400 13548 13549 R7xx: program additional CUR/GRPH regs for 40 bit addresses 13550 13551 The *_HIGH regs are reversed. The secondary ones are in the 13552 primary block and vice versa. 13553 13554 commit 4cf06dfba617529291ce4b4c306c4fc1bba110ee 13555 Author: Alex Deucher <alexdeucher (a] gmail.com> 13556 Date: Thu Oct 22 02:08:45 2009 -0400 13557 13558 ATOM: use indirect addressing for regs >= 0x10000 13559 13560 commit 66b194a78c470cb3978f310828dd96c3f3e96944 13561 Author: Alex Deucher <alexdeucher (a] gmail.com> 13562 Date: Wed Oct 21 15:58:18 2009 -0400 13563 13564 ATOM: fix up get clock info 13565 13566 Newer revisions of the atom firmware table have 13567 changed. This helps select better pll dividers 13568 in some cases. Noticed by Mathias Froehlich. 13569 13570 commit bd89b7501f294ac645390ef144df569953c81dc4 13571 Author: Alex Deucher <alexdeucher (a] gmail.com> 13572 Date: Sun Oct 18 16:21:36 2009 -0400 13573 13574 radeon: fixup CustomEDID option 13575 13576 - add support for analog option to force edid to analog 13577 mode 13578 - fix compilation on older xservers 13579 - updated man page 13580 13581 commit 255ac8d54e3dc12e579e05edb41b40946a00eab8 13582 Author: Alex Deucher <alexdeucher (a] gmail.com> 13583 Date: Tue Oct 13 12:27:52 2009 -0400 13584 13585 radeon: add XV_CRTC attribute for textured video 13586 13587 This attribute allows the user to override which 13588 crtc is synced with when XV_VSYNC is enabled. This 13589 is useful for clone modes where the user can might want 13590 to override the default. 13591 13592 commit 801d900add4731f3c424ebb780ad2dbd7a1c6a25 13593 Author: Alex Deucher <alexdeucher (a] gmail.com> 13594 Date: Mon Oct 12 10:00:57 2009 -0400 13595 13596 ATOM: minor cleanups to pll setup 13597 13598 commit f4407962cd7b272e0860319f11f6a6583ef226c2 13599 Author: Csillag Kristof <csillag.kristof (a] united-consult.hu> 13600 Date: Fri Oct 9 18:00:09 2009 -0400 13601 13602 Fix some issues with CustomEDID support 13603 13604 - Fixes a memory allocation problem introduced with the previous 13605 version of the CustomEDID patch (now in GIT), and 13606 - Enhances the syntax of the CustomEDID option so that I can force 13607 digital output, even if the EDID was acquired over an analog link. 13608 13609 commit de55995e82c3875f70b6394fff440d695d062113 13610 Author: Alex Deucher <alexdeucher (a] gmail.com> 13611 Date: Fri Oct 9 11:07:30 2009 -0400 13612 13613 r600 EXA: fix up mask reg mixup 13614 13615 commit a168caf5349b757873ad89d977772feaa7fb6e58 13616 Author: Alex Deucher <alexdeucher (a] gmail.com> 13617 Date: Fri Oct 9 10:32:26 2009 -0400 13618 13619 r600 EXA: fix some compile warnings 13620 13621 commit fb8dda2928bd0def2b34c8c8e414129e650d6eb5 13622 Author: Alex Deucher <alexdeucher (a] gmail.com> 13623 Date: Fri Oct 9 10:30:21 2009 -0400 13624 13625 r600 EXA: properly check planemask 13626 13627 We only support per-component masks. 13628 13629 commit 1b25bad2d8ebe481ffb4b0fd3a85ab636bd4fec2 13630 Author: Dave Airlie <airlied (a] linux.ie> 13631 Date: Fri Oct 9 20:09:52 2009 +1000 13632 13633 Revert "radeon: clamp the internal FB map to the aperture size" 13634 13635 This reverts commit 5f846360c46f5a989f5d0fde6d251cdbd61d4968. 13636 13637 Numerous reports of system hangs since this, I'm guessing 13638 some sort of conflict with the drm memory setup. This code 13639 has always been fragile between kernel/userspace drivers. 13640 13641 commit a956d478043b44c49aed39c7aa3d576ef32cadc5 13642 Author: Dave Airlie <airlied (a] linux.ie> 13643 Date: Fri Oct 9 20:08:31 2009 +1000 13644 13645 Revert "kms: setup colormap and gamma correctly." 13646 13647 This reverts commit 60d9685abddccec17c1a9a5ec48cbe9c92543e0f. 13648 13649 which breaks colormap on 1.6 X server 13650 13651 Conflicts: 13652 13653 src/drmmode_display.c 13654 13655 commit a493feb94c0d84aaf5bfd84c6940d7f03974dd86 13656 Author: Dave Airlie <airlied (a] redhat.com> 13657 Date: Fri Oct 9 15:05:25 2009 +1000 13658 13659 kms: don't use scratch pixmaps at all 13660 13661 use our own wrapper to allocate a pixmap for wrapping a bo in. 13662 13663 commit 0caffbd6fd6b356b9b7d715c015285eaa91e3e36 13664 Author: Dave Airlie <airlied (a] redhat.com> 13665 Date: Fri Oct 9 15:04:56 2009 +1000 13666 13667 drmmode: reload cursors on modeset 13668 13669 commit 02e12ae6be7bc1976f63848fa2854d320d5ab36e 13670 Author: Alex Deucher <alexdeucher (a] gmail.com> 13671 Date: Wed Oct 7 19:56:58 2009 -0400 13672 13673 ATOM: rework crtc modeset 13674 13675 - clean up tv timing handling 13676 - unify SetCRTCTiming and SetCRTCDTDTiming interfaces 13677 13678 commit d499eeaf22f77a1294e99aa38a50aa6810bb684a 13679 Author: Alex Deucher <alexdeucher (a] gmail.com> 13680 Date: Wed Oct 7 17:43:11 2009 -0400 13681 13682 ATOM: reorder crtc dpms based on bios recommendations 13683 13684 commit 4ccd2a21f9c50515b4246f35454f76ef49006c76 13685 Author: Alex Deucher <alexdeucher (a] gmail.com> 13686 Date: Wed Oct 7 16:30:03 2009 -0400 13687 13688 ATOM: add support for AdjustDisplayPll table 13689 13690 Depending on the output and clock, this table will 13691 adjust the pixelclock accordingly. 13692 13693 commit 5f846360c46f5a989f5d0fde6d251cdbd61d4968 13694 Author: Alex Deucher <alexdeucher (a] gmail.com> 13695 Date: Wed Oct 7 17:01:55 2009 -0400 13696 13697 radeon: clamp the internal FB map to the aperture size 13698 13699 We don't use the invisible memory yet and on cards with 13700 large amounts of vram this can cause the top of GART 13701 calculation to overflow. 13702 13703 Fixes bug fdo bug 24301: 13704 http://bugs.freedesktop.org/show_bug.cgi?id=24301 13705 13706 commit 186bf719c394116341abff471fb4b4803650da9a 13707 Author: Dave Airlie <airlied (a] redhat.com> 13708 Date: Wed Oct 7 17:23:36 2009 +1000 13709 13710 radeon: fix rotation since mixed pixmaps. 13711 13712 Passing null in here stops the mixed pixmap code doing it wrong, 13713 but I'm not 100% sure how it is meant to work, we have the 13714 same issue with the real front bo by the looks of it. 13715 13716 commit e08411af1aa8c7d4233ba593b84360397cdbb307 13717 Author: Roel Kluin <roel.kluin (a] gmail.com> 13718 Date: Tue Oct 6 18:31:17 2009 -0400 13719 13720 radeon: Fix duplicated bit settings 13721 13722 [agd5f: adapted from kms patch] 13723 13724 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13725 13726 commit e59ae08270711512e64b70f79b6476cc2c52d230 13727 Author: Michel Dnzer <daenzer (a] vmware.com> 13728 Date: Tue Oct 6 00:13:09 2009 +0200 13729 13730 Don't call radeon_cs_flush_indirect() without KMS. 13731 13732 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=24316 . 13733 13734 commit 457646d734558672402b6bcd15cadb66741edbed 13735 Author: Jerome Glisse <jglisse (a] redhat.com> 13736 Date: Mon Oct 5 15:52:42 2009 +0200 13737 13738 kms: assume accel is working is info ioctl return -EINVAL 13739 13740 -EINVAL should only happen if the info ioctl doesn't support 13741 the get accel query. This patch assume that on such case accel 13742 is working. 13743 13744 commit 9d596562496863d65850306d2126d8df98464de4 13745 Author: Dave Airlie <airlied (a] linux.ie> 13746 Date: Mon Oct 5 18:35:20 2009 +1000 13747 13748 kms: only touch gamma stuff on 1.7 13749 13750 for correct colormap operation you probably want 1.7 13751 13752 commit 876f43d4682df4d47d3e158f75e0ca6040f41c70 13753 Author: Alex Deucher <alexdeucher (a] gmail.com> 13754 Date: Mon Oct 5 02:23:03 2009 -0400 13755 13756 R1/2xx EXA: Minimise number of draw primitives used for Composite operations. 13757 13758 commit 86cafb8affea448bdf58300044e755201b822d2a 13759 Author: Dave Airlie <airlied (a] redhat.com> 13760 Date: Mon Oct 5 14:17:25 2009 +1000 13761 13762 kms: don't use scratch pixmaps when copying fbcon. 13763 13764 scratch pixmaps seem to interact badly with mixed pixmaps, it appears 13765 some state may be getting left around in the privates somewhere, since 13766 scratch pixmap headers don't get destroyed. 13767 13768 commit f8471512ea9f1d38140dfe98a0f832e9f935f51b 13769 Author: Csillag Kristof <csillag.kristof (a] united-consult.hu> 13770 Date: Sun Oct 4 18:18:56 2009 -0400 13771 13772 radeon: add support for Custom EDID 13773 13774 Allows you to specify an edid per output from a file 13775 to override what is detected by DDC. Useful for 13776 problematic monitors or KVM switches that block 13777 DDC. Specifying an EDID that is not compatible with 13778 your monitor could damage your monitor so use with 13779 caution. 13780 13781 agd5f: cache the custom edid at startup so we don't 13782 have to read it from file every time the output is 13783 queried. 13784 13785 commit 60d9685abddccec17c1a9a5ec48cbe9c92543e0f 13786 Author: Dave Airlie <airlied (a] redhat.com> 13787 Date: Mon Oct 5 12:41:30 2009 +1000 13788 13789 kms: setup colormap and gamma correctly. 13790 13791 We need to program the gamma at modeset. no need to use 13792 old cmap interface at all. 13793 13794 commit b0b10e548fd69b3ff67a0f9b71456b89c61cb4d6 13795 Author: Alex Deucher <alexdeucher (a] gmail.com> 13796 Date: Sun Oct 4 17:07:32 2009 -0400 13797 13798 R6/7xx EXA: Add support for BGRA picture formats. 13799 13800 Also add mysteriously missing case statements for ABGR destination formats. 13801 13802 commit 8404c37b2ff95aba31db97676f106ec2f30fda7f 13803 Author: Dave Airlie <airlied (a] linux.ie> 13804 Date: Sun Oct 4 18:18:51 2009 +1000 13805 13806 radeon: fix build without kms 13807 13808 commit 9460ea864b12ec1fbd11c5d9a20bb5a4279d9d3d 13809 Author: Michel Dnzer <daenzer (a] vmware.com> 13810 Date: Sat Oct 3 16:33:33 2009 +0200 13811 13812 Fix KMS on big endian machines. 13813 13814 Requires at least xserver 1.7 to work properly. 13815 13816 Also make sure the front buffer is and stays tiled if colour tiling is enabled. 13817 13818 commit aee6b6f2c120baf477b4845ddc1a94637c31db2f 13819 Author: Michel Dnzer <daenzer (a] vmware.com> 13820 Date: Sat Oct 3 16:33:33 2009 +0200 13821 13822 EXA: Bail earlier from PrepareCopy hook on trivially unsupportable conditions. 13823 13824 commit 30031b169def5caadb394225919a6079a0957142 13825 Author: Michel Dnzer <daenzer (a] vmware.com> 13826 Date: Sat Oct 3 16:33:33 2009 +0200 13827 13828 R3/5xx EXA: Add support for BGRA picture formats. 13829 13830 Also add mysteriously missing case statements for ABGR destination formats. 13831 13832 commit 4b4ce36081ca151c24e028c54b59986f41731a73 13833 Author: Michel Dnzer <daenzer (a] vmware.com> 13834 Date: Sat Oct 3 16:33:32 2009 +0200 13835 13836 R3/5xx EXA: Minimise number of draw primitives used for Composite operations. 13837 13838 This should reduce the kernel CS checker overhead, if nothing else. 13839 13840 I'll leave porting this to other chipset families to others who can test it. 13841 13842 commit eade1e5be159c9f2965d611925596d33cab11d6d 13843 Author: Michel Dnzer <daenzer (a] vmware.com> 13844 Date: Sat Oct 3 16:33:32 2009 +0200 13845 13846 EXA: Don't always flush when switching between 2D/3D engines with KMS. 13847 13848 This seems to work fine here now and help 2D performance quite a bit, let's see 13849 if it breaks anybody else's setup... 13850 13851 commit 926e414fc835ace141c066830d11d8ce32dbb06c 13852 Author: Michel Dnzer <daenzer (a] vmware.com> 13853 Date: Sat Oct 3 16:33:32 2009 +0200 13854 13855 KMS: Double-buffer textured video source image upload. 13856 13857 In order to avoid stalling on previous frame. 13858 13859 OTOH without KMS we can't do this but have to wait for the previous frame to 13860 finish rendering. 13861 13862 commit 6a3b75fae4147244212298ff55c7ab36c3d80d30 13863 Author: Michel Dnzer <daenzer (a] vmware.com> 13864 Date: Sat Oct 3 16:33:31 2009 +0200 13865 13866 KMS: Don't hide HW cursor when it's updated. 13867 13868 Inspired by the intel driver; might avoid/reduce HW cursor flicker in some 13869 cases. 13870 13871 commit c4ba15b4871c70cced7f1761e24be692cdb28d2b 13872 Author: Michel Dnzer <daenzer (a] vmware.com> 13873 Date: Sat Oct 3 16:33:31 2009 +0200 13874 13875 Fix some compiler warnings. 13876 13877 commit cc45856a18dd3e6f7e44d9eb507b31419da70977 13878 Author: Kenneth Graunke <kenny (a] whitecape.org> 13879 Date: Fri Oct 2 14:31:36 2009 -0400 13880 13881 radeon: Remove gatos message from driver startup 13882 13883 Remove a message on driver startup which directs people to the old GATOS 13884 website; said site contains highly out of date software which is no longer 13885 necessary. 13886 13887 commit d302481493ff61eb7f8648b7afe7407b6b653cf7 13888 Author: Jerome Glisse <jglisse (a] redhat.com> 13889 Date: Wed Sep 16 15:18:40 2009 +0200 13890 13891 radeon/kms: fallback to shadowfb if kernel report acceleration is off 13892 13893 This will fallback to shadowfb is GPU accel failed somewhere during 13894 KMS kernel init. 13895 13896 commit 7968e1fb89f6b59d1654df48249bf4b81990c008 13897 Author: Alex Deucher <alexdeucher (a] gmail.com> 13898 Date: Fri Sep 25 19:05:38 2009 -0400 13899 13900 radeon: fix segfault in MMIO path in RADEONInit3DEngine 13901 13902 fixes fdo bug 24158. 13903 13904 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13905 13906 commit 0380a9cf98d34e88dd81370ade5525680ec89c02 13907 Author: Alex Deucher <alexdeucher (a] gmail.com> 13908 Date: Fri Sep 25 16:46:43 2009 -0400 13909 13910 kms/r600: add support for vline relocs 13911 13912 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13913 13914 commit 3efecebb10de7f7bacf9f8c57ae20fd508097294 13915 Author: Alex Deucher <alexdeucher (a] gmail.com> 13916 Date: Fri Sep 25 14:44:41 2009 -0400 13917 13918 radeon: fix vline handling for kms 13919 13920 drm crtc ids do not correspond to actual hw crtcs, 13921 as such the vline stuff was never enabled for Xv. 13922 13923 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13924 13925 commit 9733dcde0a21b7503aa20254724f2910b541b990 13926 Author: Dave Airlie <airlied (a] linux.ie> 13927 Date: Sat Sep 26 06:15:58 2009 +1000 13928 13929 r600: fix build 13930 13931 commit ff18595eace42ddfc931f4f948cb5adf59ebcf52 13932 Author: Dave Airlie <airlied (a] linux.ie> 13933 Date: Sat Sep 26 06:04:51 2009 +1000 13934 13935 r600/xv: fix dst bo write domain 13936 13937 commit 718a611f3cc7f5eea95cd73a9dcc7913c922fcbc 13938 Author: Dave Airlie <airlied (a] linux.ie> 13939 Date: Sat Sep 26 06:03:51 2009 +1000 13940 13941 radeon: fix zaphod 13942 13943 commit 86c367be517e992aabc1bcfea2a2dde1a1ae520c 13944 Author: Alex Deucher <alexdeucher (a] gmail.com> 13945 Date: Fri Sep 25 13:03:44 2009 -0400 13946 13947 r600: fix bo accounting for Xv 13948 13949 fixes Xv after changes in: 13950 8f80e37eed3ec028718b4e71bbb9b598847fd94e 13951 13952 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13953 13954 commit 8f80e37eed3ec028718b4e71bbb9b598847fd94e 13955 Author: Dave Airlie <airlied (a] linux.ie> 13956 Date: Fri Sep 25 11:37:38 2009 +1000 13957 13958 r600: space check wasn't correct. 13959 13960 the r6xx space check code was incorrect for the DFS case, since 13961 the dst bo was in GTT but the hardcoded cp_start function 13962 assumed the bos were where it thought. 13963 13964 Ripped out assumptions and replaced with code more like other radeons. 13965 13966 commit 90669f6cb5ada9067d6fccc7f54c69b367862f42 13967 Author: Alex Deucher <alexdeucher (a] gmail.com> 13968 Date: Thu Sep 24 10:40:40 2009 -0400 13969 13970 radeon: remove old pre-randr rotation cruft 13971 13972 leftover from the randr 1.2 conversion. 13973 13974 Signed-off-by: Alex Deucher <alexdeucher (a] gmail.com> 13975 13976 commit 97a4e747bfac14f34646c55ddf639e8fe22f2f55 13977 Author: Michael Olbrich <m.olbrich (a] pengutronix.de> 13978 Date: Sun Sep 20 14:19:41 2009 +0200 13979 13980 use AC_CHECK_HEADER instead of AC_CHECK_FILE 13981 13982 AC_CHECK_FILE is not possible when cross-compiling. Use AC_CHECK_HEADER / 13983 AC_PREPROC_IFELSE instead. 13984 13985 [ Michel Dnzer: Shuffled things around slightly to make it work on my setup ] 13986 13987 Signed-off-by: Michael Olbrich <m.olbrich (a] pengutronix.de> 13988 Signed-off-by: Michel Dnzer <daenzer (a] vmware.com> 13989 13990 commit 579070ed9f0218a975e840331b6fe9fa8fd05789 13991 Author: Alex Deucher <alexdeucher (a] gmail.com> 13992 Date: Sat Sep 19 19:15:14 2009 -0400 13993 13994 kms: disable dga 13995 13996 Can be dangerous with kms: 13997 http://lists.freedesktop.org/archives/xorg-devel/2009-September/002113.html 13998 13999 commit da7487f6ac7b32ed7866af53da2925aa52ed7aae 14000 Author: Alex Deucher <alexdeucher (a] gmail.com> 14001 Date: Sat Sep 19 13:08:02 2009 -0400 14002 14003 kms: init dga using xf86DiDGAInit 14004 14005 fixes cursor problems in games like UT2004 and Penumbra Overture 14006 14007 see fdo bug 24034 14008 14009 commit 52279251fae9df99c569c16e2522bbd346d8ec38 14010 Author: Dave Airlie <airlied (a] redhat.com> 14011 Date: Fri Sep 18 14:29:32 2009 +1000 14012 14013 avivo: disable VGA rendering core when starting X. 14014 14015 this thing can be rendering to VRAM when we don't expect it. 14016 turn it off. 14017 14018 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 14019 14020 commit 2c46bafcb77c4125a27c18ad6ca2f6de5f143a2d 14021 Author: Pauli Nieminen <suokkos (a] gmail.com> 14022 Date: Thu Sep 17 15:13:51 2009 +0300 14023 14024 Fix configuration to disable KMS if not finding libdrm_radeon. 14025 14026 Most of KMS building was disabled if LIBDRM_RADEON was not detected. But at 14027 least configure was reporting KMS enabled even tough it was realy disabled. 14028 14029 commit ee9f6d802988a4d7f7985687036db854f4a14fe5 14030 Author: Pauli Nieminen <suokkos (a] gmail.com> 14031 Date: Mon Sep 14 16:36:48 2009 +0300 14032 14033 radeon: Add missing libdrm_radeon cflags to compiler paramaters. 14034 14035 This fixes compilation if libdrm_radeon is installed to non-standard location like /opt. 14036 14037 commit ac853ca0a950d6147cd6a0b4a1b089f5d2f080ea 14038 Author: Alex Deucher <alexdeucher (a] gmail.com> 14039 Date: Fri Sep 11 14:25:31 2009 -0400 14040 14041 atom: fix typo in asus quirks 14042 14043 Should be DVI-I, not DVI-D 14044 14045 commit a88f60e7b789cd23a79e6e18a7e3fa008d15aa77 14046 Author: Alex Deucher <alexdeucher (a] gmail.com> 14047 Date: Thu Sep 10 13:42:31 2009 -0400 14048 14049 r600: add CS UTS and DFS hooks 14050 14051 commit 0bb0ff0e55aaf8e64017d204e376a14f0a7384a3 14052 Author: Dave Airlie <airlied (a] linux.ie> 14053 Date: Thu Sep 10 11:48:00 2009 +1000 14054 14055 r6xx: fix EXA crash with gtkperf -a 14056 14057 unmapping was getting unbalanced. 14058 14059 commit 10a58d54857484b6e89763ecf463f54dc290c702 14060 Author: Dave Airlie <airlied (a] redhat.com> 14061 Date: Wed Sep 9 14:24:49 2009 +1000 14062 14063 kms: add property support. 14064 14065 this adds support to the userspace DDX to get properties from the kernel 14066 and expose them over randr. Its most liberated from the Intel driver. 14067 14068 commit fad9ee6cdae5ae45e6406e3071a84276f1b59a85 14069 Author: Alex Deucher <alexdeucher (a] gmail.com> 14070 Date: Tue Sep 8 17:50:50 2009 -0400 14071 14072 radeon: replace dga code with DiDGA 14073 14074 commit 651fe5a47435c6a5cc1cdfb33c5c6601d692adcb 14075 Author: Adam Jackson <ajax (a] redhat.com> 14076 Date: Tue Sep 8 13:21:07 2009 +1000 14077 14078 radeon: fix dri2 struct zeroing 14079 14080 commit 6990f2ac6478bf92929a4400ef84fb2142699204 14081 Merge: 917f2d7cd 853f4c3d1 14082 Author: Dave Airlie <airlied (a] redhat.com> 14083 Date: Tue Sep 8 11:26:32 2009 +1000 14084 14085 Merge branch 'r6xx-cs' 14086 14087 commit 853f4c3d1ea8f975ab2855f18d3ae336a4095091 14088 Author: Dave Airlie <airlied (a] redhat.com> 14089 Date: Tue Sep 8 11:25:39 2009 +1000 14090 14091 r600: more alignment fixups + vb map/unmap 14092 14093 I'm not so sure the vb map/unmap is a good idea, I think 14094 it pretty much locksteps the cpu/gpu, so we should really 14095 work out if we really need to flush this often, since 14096 mesa doesn't have to and we are just doing 3D ops. 14097 14098 commit 917f2d7cd1815279710939edc27133ca6609460f 14099 Author: Alex Deucher <alexdeucher (a] gmail.com> 14100 Date: Sun Sep 6 10:53:29 2009 -0400 14101 14102 RV280: add agp quirk 14103 14104 deb bug 545040 14105 14106 commit 78fcbf577ad6eba6399cc39f74b7ce5f9c8e265e 14107 Author: Michel Dnzer <daenzer (a] vmware.com> 14108 Date: Thu Sep 3 14:55:05 2009 +0200 14109 14110 EXA: Check for solid/gradient pictures the same way for all generations. 14111 14112 In particular, also catch them for >= R300. 14113 14114 commit 794ae743c305331741e7cf45dd71c755fe325ed5 14115 Author: Michel Dnzer <daenzer (a] vmware.com> 14116 Date: Wed Sep 2 09:19:17 2009 +0200 14117 14118 EXA: Fix Composite restart dst/src switcheroo. 14119 14120 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=22636 . 14121 14122 Now, where's that brown paper bag? :} 14123 14124 commit c4ab50c5eafca3d04129a72453496eb8afb26b34 14125 Author: Alex Deucher <alexdeucher (a] gmail.com> 14126 Date: Tue Sep 1 13:43:00 2009 -0400 14127 14128 radeon: add some notes about DVO setup 14129 14130 commit 22fbd6476f4b3513378e848affc4dd269e9d4498 14131 Author: Dave Airlie <airlied@itt42.(none)> 14132 Date: Tue Sep 1 15:22:57 2009 +1000 14133 14134 r100/r200: seriously WRAP is illegal for rectangular textures 14135 14136 unless we scale all the bos up to POT 14137 14138 commit 8f4196e88855f10762254fca9e0a0988e7b5562f 14139 Author: Alex Deucher <alexdeucher (a] gmail.com> 14140 Date: Mon Aug 31 19:41:59 2009 -0400 14141 14142 r6xx/r7xx: various CS fixes from Dave 14143 14144 commit a32384f9b9ed73c14484a380dfa14652ed478ad6 14145 Author: Alex Deucher <alexdeucher (a] gmail.com> 14146 Date: Mon Aug 31 13:26:05 2009 -0400 14147 14148 r1xx: fix textured video 14149 14150 Copy/paste bug from when kms support was added. 14151 14152 commit 3d6bf0f8cf128ee67d448491be44d345547b75c5 14153 Author: Pauli Nieminen <suokkos (a] gmail.com> 14154 Date: Fri Aug 28 17:05:44 2009 +0300 14155 14156 radeon: Fix DRI2BufferPtr to be DRI2Buffer2Ptr for xserver 1.6. 14157 14158 Bump requirement for xserver in KMS mode to 1.6.2 for DRI2Buffer2Ptr. 14159 14160 commit 6521dbf3884185ca5692d35d879d251890cc4f08 14161 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 14162 Date: Sat Aug 29 12:11:03 2009 +0200 14163 14164 r100/r200: fix section size mismatch for textured video 14165 14166 this fixes the driver complaining though reportedly it still does 14167 not work (on r100) 14168 14169 commit 39dfac41eebbcdacb1ae0c17cea325ea2966c4a5 14170 Author: Michel Dnzer <daenzer (a] vmware.com> 14171 Date: Fri Aug 28 12:20:04 2009 +0200 14172 14173 EXA: RENDER repeat fix and more cleanup. 14174 14175 Apparently changing pPict->repeatType causes badness, see 14176 http://bugs.freedesktop.org/show_bug.cgi?id=23560 . 14177 14178 Now we just use a normalized repeatType value derived from pPict->repeat and 14179 pPict->repeatType everywhere. 14180 14181 commit 7d36f13afa5da742aa6c989a2652e92c96e5f80e 14182 Author: Michel Dnzer <daenzer (a] vmware.com> 14183 Date: Thu Aug 27 09:52:11 2009 +0200 14184 14185 KMS: Remove explicit radeon_bo_wait calls before radeon_bo_map. 14186 14187 libdrm_radeon always takes care of this implicitly now. 14188 14189 commit 066c90e8f5b93c80195dc43486ad3efa083a1f3a 14190 Author: Michel Dnzer <daenzer (a] vmware.com> 14191 Date: Thu Aug 27 09:52:11 2009 +0200 14192 14193 KMS: Don't use a blit for UploadToScreen if the pixmap BO will be idle anyway. 14194 14195 Direct CPU writes should be at least as fast in that case. 14196 14197 commit 174b61bb786a841cebd354e23b4b0caa85b541aa 14198 Author: Michel Dnzer <daenzer (a] vmware.com> 14199 Date: Thu Aug 27 09:52:11 2009 +0200 14200 14201 KMS: DownloadFromScreen improvements. 14202 14203 * Drop superfluous RADEONDownloadFromScreenGTT function, EXA does the same 14204 thing when we return FALSE. 14205 * Take unflushed operations into account for determining which GEM domain the 14206 pixmap BO will end up in. 14207 * Only use a blit if it ends up in VRAM. 14208 14209 commit 7623e169e1f7d5afbd4108de03f28098bca519db 14210 Author: Michel Dnzer <daenzer (a] vmware.com> 14211 Date: Thu Aug 27 09:52:11 2009 +0200 14212 14213 EXA: Allocate pixmap BOs in VRAM. 14214 14215 This is the intention with EXA in xserver Git, and IME it improves performance 14216 significantly even with older versions. 14217 14218 commit b90b823790657707b5bf7249095833bef959b0e8 14219 Author: Michel Dnzer <daenzer (a] vmware.com> 14220 Date: Thu Aug 27 09:52:11 2009 +0200 14221 14222 DRI2: Set tiling for depth/stencil buffers with all DRI2 interface versions. 14223 14224 commit 27d9664f8862d1948dd32021f8e9bcb94f23e4ef 14225 Author: Michel Dnzer <daenzer (a] vmware.com> 14226 Date: Thu Aug 27 09:52:10 2009 +0200 14227 14228 EXA: RENDER repeat cleanups. 14229 14230 We can't rely on the server setting pPict->repeatType to something sensible 14231 when pPict->repeat is FALSE, but we can do it ourselves. 14232 14233 commit 9b5517fcf7f2397a1a814d396fc9dc24769a6d9f 14234 Author: Michel Dnzer <daenzer (a] vmware.com> 14235 Date: Thu Aug 27 09:52:10 2009 +0200 14236 14237 Throttle DRI2 buffer swaps / frontbuffer flushes. 14238 14239 commit e87f0f50f31a59ca1f60d4582d4a57ed00854fb7 14240 Author: Alex Deucher <alexdeucher (a] gmail.com> 14241 Date: Wed Aug 26 02:13:38 2009 -0400 14242 14243 r6xx/r7xx: set EXA_HANDLES_PIXMAPS 14244 14245 commit 3212c26b90c0f6f1a7248b4da3ed985a9c2e9381 14246 Author: Alex Deucher <alexdeucher (a] gmail.com> 14247 Date: Wed Aug 26 01:42:10 2009 -0400 14248 14249 r6xx/r7xx: more WIP 14250 14251 commit 5a08e68cc254fb255e631b456e331c32456ef0e7 14252 Author: Alex Deucher <alexdeucher (a] gmail.com> 14253 Date: Tue Aug 25 19:24:41 2009 -0400 14254 14255 r6xx/r7xx: fix some define problems in Xv code 14256 14257 commit 9aa214e125b7927d62b9fe124a851d0373c24d7e 14258 Author: Alex Deucher <alexdeucher (a] gmail.com> 14259 Date: Tue Aug 25 18:45:49 2009 -0400 14260 14261 r6xx/r7xx: fix reloc for vtx buffer 14262 14263 commit bba51187055932ecd466f5f817428d6c773747b9 14264 Author: Alex Deucher <alexdeucher (a] gmail.com> 14265 Date: Tue Aug 25 18:37:15 2009 -0400 14266 14267 R6xx/r7xx: unmap vb bo when done 14268 14269 commit 599adfc1f5e6d708be7ad30f4871de3046775727 14270 Author: Alex Deucher <alexdeucher (a] gmail.com> 14271 Date: Tue Aug 25 18:13:14 2009 -0400 14272 14273 r6xx/r7xx: fix flipped domains 14274 14275 commit 2e83cca8d7efaf1a6836cfb9ea5893fd9d70175f 14276 Author: Alex Deucher <alexdeucher (a] gmail.com> 14277 Date: Tue Aug 25 18:05:43 2009 -0400 14278 14279 r6xx/r7xx: more cs exa wip 14280 14281 commit 65852de027989c105246fa4e4eed432f29525a22 14282 Author: Alex Deucher <alexdeucher (a] gmail.com> 14283 Date: Tue Aug 25 17:29:42 2009 -0400 14284 14285 r6xx/r7xx EXA: WIP 14286 14287 commit 69ec7a35e2a0a3d802ec093a6aab2d7ed2cc88be 14288 Author: Alex Deucher <alexdeucher (a] gmail.com> 14289 Date: Tue Aug 25 16:14:02 2009 -0400 14290 14291 r6xx/r7xx: first pass at kms accel support 14292 14293 Adapted from various patches from Dave and Jerome. 14294 14295 commit 04692f80678ffea15ac33820381ce45865e697b8 14296 Author: Dave Airlie <airlied (a] linux.ie> 14297 Date: Wed Aug 26 05:31:56 2009 +1000 14298 14299 radeon: afaics these aren't radeon regs but r128 regs. 14300 14301 so remove them from radeon use. 14302 14303 commit 9cf965bbc977f0523437c0ecf1d7363b17de2468 14304 Author: Alex Deucher <alexdeucher (a] gmail.com> 14305 Date: Tue Aug 25 12:47:34 2009 -0400 14306 14307 R6xx/r7xx: add begin/end batch macros 14308 14309 commit b6368cc572c79bce9a9366242c727c13cab3f006 14310 Author: Alex Deucher <alexdeucher (a] gmail.com> 14311 Date: Tue Aug 25 12:14:33 2009 -0400 14312 14313 r6xx/r7xx: move more common state to default state setup 14314 14315 commit fc74e1194c980d978667e02c60a29a761a694bde 14316 Author: Alex Deucher <alexdeucher (a] gmail.com> 14317 Date: Tue Aug 25 11:39:56 2009 -0400 14318 14319 RV740: disable small DFS transfers 14320 14321 Seems problematic on all rv740 chips, so until we 14322 can find a solution, disable them. 14323 14324 commit 1c909ecd88e75c6e16af7aa068e32a7a60b0af37 14325 Author: Jerome Glisse <jglisse (a] redhat.com> 14326 Date: Tue Aug 25 11:21:41 2009 +0200 14327 14328 radeon: Fix X recycling with KMS 14329 14330 On X recycling CloseScreen drop master so grab it again 14331 in InitScreen and avoid to reallocate buffer & various 14332 others cs/bo utilities on X recycling. 14333 14334 commit 39aaf9addc923852598ba6fb914459671b78e492 14335 Author: Dave Airlie <airlied (a] linux.ie> 14336 Date: Mon Aug 24 18:59:22 2009 +1000 14337 14338 r600: use exa fallback code for solid/copy codepaths. 14339 14340 Solid seem to try and handle 1bpp which really no. 14341 14342 commit e4b2d57ddca8739ec0b452565713398d036be0d5 14343 Author: Dave Airlie <airlied (a] redhat.com> 14344 Date: Mon Aug 24 14:11:24 2009 +1000 14345 14346 r600: fix bad unit in texture setup 14347 14348 commit bac224912c750dc1c85ff2d9b8526dad6c23b572 14349 Author: Dave Airlie <airlied (a] redhat.com> 14350 Date: Sat Aug 22 21:17:59 2009 +1000 14351 14352 radeon: don't spec any initial placement for pixmaps. 14353 14354 allow the first use to decide placement. 14355 14356 commit 77f98717d825162da106c6898cdbcbdf5c984ae6 14357 Author: Dave Airlie <airlied (a] redhat.com> 14358 Date: Sat Aug 22 21:16:25 2009 +1000 14359 14360 exa/cs: add DFS from GTT optimisation 14361 14362 This uses the new libdrm busy interface, once I had this in place 14363 I added a error if this happened and it does on my desktop here, 14364 so may as well add the optimisation that used to be in my old KMS tree. 14365 14366 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 14367 14368 commit b1b77a4d6fb7404af9568644e1a8e050fdfa956e 14369 Author: Alex Deucher <alexdeucher (a] gmail.com> 14370 Date: Wed Aug 19 13:29:04 2009 -0400 14371 14372 rs600: add support for DisplayPriority HIGH 14373 14374 Still haven't implemented the actual watermark 14375 calculation, but this forces the display requests 14376 to urgent if the user specifies DisplayPriority HIGH. 14377 14378 commit 72e0d1b2cb11a67b8e4be4c74913ee44dc051c5b 14379 Author: Dave Airlie <airlied (a] redhat.com> 14380 Date: Wed Aug 19 19:01:30 2009 +1000 14381 14382 radeon/kms: add initial colortiling support (disabled by default). 14383 14384 This requires an X server from git with createpixmap2 support fixed up in it. 14385 14386 On 1.6 and previous it won't do any tiling, if you are running git server 14387 please upgrade to latest git. 14388 14389 Option "AllowColorTiling" "true" to enable and do some testing 14390 14391 commit 479a6daefe46f985c415b0d000b1b1b820f3924e 14392 Author: Alex Deucher <alexdeucher (a] gmail.com> 14393 Date: Tue Aug 18 02:53:29 2009 -0400 14394 14395 AVIVO: add support for DisplayPriority HIGH 14396 14397 If the DisplayPriority option is set to HIGH, 14398 force the diplay mem requests to urgent. 14399 14400 commit 8fd6f69c84103b277de70181988e843da23c85b0 14401 Author: Dave Airlie <airlied (a] redhat.com> 14402 Date: Mon Aug 17 15:50:21 2009 +1000 14403 14404 radeon: move detected monitor type debugging. 14405 14406 We should print this after DAC detection to be useful for TV 14407 14408 commit d25185f33d06a8162469e18688e95ee19a5d94d8 14409 Author: Dave Airlie <airlied (a] redhat.com> 14410 Date: Mon Aug 17 15:31:16 2009 +1000 14411 14412 atom: recall crtc source script after tv dpms on 14413 14414 For some reason we lost the DACB source when we dpms tv back on. 14415 14416 commit a971f1cfbbda1e3d16fb41f4a07f3b43fd59bc0d 14417 Author: Dave Airlie <airlied (a] redhat.com> 14418 Date: Mon Aug 17 15:25:17 2009 +1000 14419 14420 atombios: store dac detect result into save register. 14421 14422 We end up writing back an empty save register over the dac 14423 detection results so the atombios table can no longer 14424 figure out the S-video vs composite so sets the wrong one up. 14425 14426 this gets tv-out on my r580 with a composite connector working. 14427 14428 commit efef7c01ad38e078de2fa3f9e528e4ef7d05d00a 14429 Author: Dave Airlie <airlied (a] redhat.com> 14430 Date: Mon Aug 17 13:44:20 2009 +1000 14431 14432 atom/tvout: set the correct scaler up for tv out 14433 14434 If we need to use the second scaler actually set the second scaler 14435 up. 14436 14437 commit 032a8a06315d12e66240e7ef74b5b165f6d0d1a9 14438 Author: Dave Airlie <airlied (a] redhat.com> 14439 Date: Mon Aug 17 12:18:23 2009 +1000 14440 14441 kms: make tv out match the connector name 14442 14443 commit 36e51c4d63836863dd7a17cbf6e5a9b7c1bbd31a 14444 Author: Dave Airlie <airlied (a] redhat.com> 14445 Date: Mon Aug 17 12:17:31 2009 +1000 14446 14447 r100/r200: dont emit wrong clamp modes. 14448 14449 for rect textures you don't want to use wrap clamping which 14450 is the default. 14451 14452 commit 595f7c3a60ca982ca2836a0174e1f36f9d1a4fec 14453 Author: Dave Airlie <airlied@itt42.(none)> 14454 Date: Mon Aug 17 09:38:09 2009 +1000 14455 14456 radeon/tv: fix typo in crtc register modifications 14457 14458 commit 92baeafde8f0c1eed2b6769e96f8093e3d6d7478 14459 Author: Alex Deucher <alexdeucher (a] gmail.com> 14460 Date: Tue Aug 11 15:29:17 2009 -0400 14461 14462 RV280: add agp quirk 14463 14464 from Hartmut Niemann 14465 14466 commit 04288fdb8b8c62fc2eb32fd467fdf05151ad121a 14467 Author: Jerome Glisse <jglisse (a] redhat.com> 14468 Date: Tue Aug 11 19:32:52 2009 +0200 14469 14470 radeon/r600: abstract vertex buffer for r600 & r700 hardware 14471 14472 Abstract vertex buffer, preliminary work before others change 14473 to bring cs support along IB support. 14474 14475 commit 72daee2a18bfb809260d1bcbe80cc14afea3b0fa 14476 Author: Adam Jackson <ajax (a] redhat.com> 14477 Date: Mon Aug 10 13:59:49 2009 -0400 14478 14479 Fix misleading filenames in DRI setup failure messages 14480 14481 commit cd99d9f0d715f1f74de4fe22f2fd30046f2c7568 14482 Author: Michel Dnzer <daenzer (a] vmware.com> 14483 Date: Fri Aug 7 11:56:50 2009 +0200 14484 14485 Properly let radeon_cs_flush_indirect() re-emit 2D state if necessary. 14486 14487 Also remove a superfluous assignment. 14488 14489 commit 577ff3ce922e457cc32f80d4365cb1da81552e72 14490 Author: Michel Dnzer <daenzer (a] vmware.com> 14491 Date: Fri Aug 7 11:56:49 2009 +0200 14492 14493 Add support for EXA_MIXED_PIXMAPS in xserver master. 14494 14495 commit e755fa56d03aa338d0c6345ed41e32aa3115ad4b 14496 Author: Michel Dnzer <daenzer (a] vmware.com> 14497 Date: Fri Aug 7 11:56:49 2009 +0200 14498 14499 Minor fixes for KMS EXA DownloadFromScreen hook. 14500 14501 Check we can handle the bpp, and remove superfluous flush. 14502 14503 commit 4cebafae81c1738a1c330d52a2c5248869f06411 14504 Author: Alex Deucher <alexdeucher (a] gmail.com> 14505 Date: Fri Aug 7 11:56:40 2009 +0200 14506 14507 Add KMS EXA UploadToScreen hook. 14508 14509 Fixups by Michel Dnzer: Doesn't seem to be a win in terms of raw numbers, but 14510 should allow more pipelining and avoiding BO memory waste for small glyph 14511 pixmaps. 14512 14513 commit 9243791322e36b9231e6a3f04024ad66325385e3 14514 Author: Dave Airlie <airlied (a] redhat.com> 14515 Date: Fri Aug 7 14:07:32 2009 +1000 14516 14517 r200: fixup scissors for DDX. 14518 14519 a) turn of R200_RE_CNTL - SCISSOR_ENABLE - this save us emitting R200_RE_TOP_LEFT, note scissor is still enabled. 14520 b) disable aux scissors. 14521 14522 commit bd03977e320591ca55b1a2fbb32414c53cb3f72e 14523 Author: Alex Deucher <alexdeucher (a] gmail.com> 14524 Date: Tue Aug 4 15:21:44 2009 -0400 14525 14526 radeon: fix compile with !kms 14527 14528 commit 4fc1e67e606daf9227e74518c54315b60cd38301 14529 Author: Alex Deucher <alexdeucher (a] gmail.com> 14530 Date: Tue Aug 4 15:17:39 2009 -0400 14531 14532 radeon: re-add asus connector quirks 14533 14534 fixes bug 19943 14535 14536 commit 22074cf0e58fddba743924532625e6fca49b6bdc 14537 Author: Jerome Glisse <jglisse (a] redhat.com> 14538 Date: Tue Aug 4 21:09:25 2009 +0200 14539 14540 radeon/kms: add simple DownloadFromScreen implementation 14541 14542 What we want to do is add userspace object support to radeon 14543 kernel modesetting. Also this DFS is dumb and might endup doing 14544 blit from GTT to GTT. 14545 14546 commit fce31b61a88522733863a9b4e9f1c935c439cb4e 14547 Author: Alex Deucher <alexdeucher (a] gmail.com> 14548 Date: Tue Aug 4 11:16:05 2009 -0400 14549 14550 RS880: enable accel 14551 14552 commit 6b1b4b9214d58175727f7992fc42bc7244c42bf2 14553 Author: Alex Deucher <alexdeucher (a] gmail.com> 14554 Date: Mon Aug 3 15:55:04 2009 -0400 14555 14556 radeon: add some new r7xx pci ids 14557 14558 commit f564460e94c9d0f1cf3ff4b8535481b2b8b4e9c1 14559 Author: Alex Deucher <alexdeucher (a] gmail.com> 14560 Date: Sat Aug 1 16:53:47 2009 -0400 14561 14562 radeon: use XAA in some cases 14563 14564 Use XAA in low memory situations or when the DRI is disabled. 14565 Using shadowfb might also be a viable option, maybe even a better option... 14566 fixes bug 21683 14567 14568 commit c71b2f050e8996787eae463eddbfdb5864ffa65a 14569 Author: Bryce Harrington <bryce (a] canonical.com> 14570 Date: Sat Aug 1 16:50:43 2009 -0400 14571 14572 radeon: AGPMode quirk needed for SiS 14573 14574 fixes bug 23065 14575 14576 commit e3659ed06fc5bb8817f1dbd7c2d6bc94c67b30f7 14577 Author: Bryce Harrington <bryce (a] canonical.com> 14578 Date: Sat Aug 1 16:48:48 2009 -0400 14579 14580 radeon: AGPMode quirk needed for IBM Thinkpad T40 with Mobility M7 LW 14581 14582 fixes bug 23064 14583 14584 commit 2391531ed6b7c11ddd5ab91b2369821cc5f8b8a7 14585 Author: Bryce Harrington <bryce (a] canonical.com> 14586 Date: Sat Aug 1 16:41:41 2009 -0400 14587 14588 radeon: AGPMode quirk needed for HP Omnibook 6200 14589 14590 fixed bug 23063 14591 14592 commit 474eda02257152ced52364f38cbad24c20aebbc0 14593 Author: Alex Deucher <alexdeucher (a] gmail.com> 14594 Date: Sat Aug 1 16:06:52 2009 -0400 14595 14596 radeon: reload bicubic Xv texture on VT switch 14597 14598 fixed bicubic filtering after VT switch or suspend/resume 14599 14600 commit 447a2ce1b88aa2d6d5713e93174c4002617720f7 14601 Author: Dave Airlie <airlied (a] redhat.com> 14602 Date: Thu Jul 30 09:54:01 2009 +1000 14603 14604 radeon: fix r600 shadow fb mode with cursor allocation in wrong place 14605 14606 From RHEL QE testing, we could end up with the cursor at 0 since 14607 we think EXA is in use when really it isn't. the info->useEXA = FALSe 14608 might be unnecessary but better to be explicit 14609 14610 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 14611 14612 commit 2b9a5be3bb22fad1d52ec9eea7733c5e956250b7 14613 Author: Alex Deucher <alexdeucher (a] gmail.com> 14614 Date: Wed Jul 29 03:34:11 2009 -0400 14615 14616 radeon: fix typo in object header to connector type conversion 14617 14618 Should fix bug 19943 14619 14620 commit 63c873cbd4d1d21d9f688028c0900c79fadc42c1 14621 Author: Dave Airlie <airlied (a] redhat.com> 14622 Date: Tue Jul 28 15:22:40 2009 +1000 14623 14624 ati: change to using ABI version check 14625 14626 commit 21a621c297ac71c65c239ea960c38706e718b91c 14627 Author: Dave Airlie <airlied (a] redhat.com> 14628 Date: Tue Jul 28 13:32:28 2009 +1000 14629 14630 ati: update for resources/RAC API removal 14631 14632 commit 5ad2519c2a7c0df389b2cd7cf7151c7e4b7252a8 14633 Author: Michel Dnzer <daenzer (a] vmware.com> 14634 Date: Mon Jul 27 14:22:02 2009 +0200 14635 14636 KMS: Explicitly specify VRAM and GTT domains when allocating BOs for pixmaps. 14637 14638 Without this, apparently they are initially allocated in system RAM (or 14639 possibly GTT), so any GPU rendering to them first triggers a copy of the whole 14640 uninitialized data to VRAM... 14641 14642 We may want to explore more sophisticated schemes in the future, but for now 14643 this seems to improve KMS 2D performance quite a bit. 14644 14645 commit 57f2c83a22f27567506c555af431f89e6031204c 14646 Author: Alex Deucher <alexdeucher (a] gmail.com> 14647 Date: Sat Jul 25 14:19:38 2009 -0400 14648 14649 radeon: Set PCI/PCIE bus type properly per asic 14650 14651 - r1xx-rv350 chips have the old pci gart 14652 - rv380+ chips have newer pcie gart 14653 14654 Select the right kind regardless of whether the user selects 14655 PCI or PCIE. 14656 14657 commit 2afc46fa74ce7730f766a3370d323c6b59694186 14658 Author: Jerome Glisse <jglisse (a] redhat.com> 14659 Date: Thu Jul 23 20:13:15 2009 +0200 14660 14661 radeon: fix KMS shadowfb for r6xx and newer hw. 14662 14663 We need to map front buffer and also to avoid EXA initialization 14664 when falling back to shadowfb on newer r6xx/r7xx when KMS is enabled. 14665 14666 commit 328e4c816348b6f867df30b39856dbe78c8e0dcc 14667 Author: Alex Deucher <alexdeucher (a] gmail.com> 14668 Date: Thu Jul 23 13:28:36 2009 -0400 14669 14670 RS780: fill in MC access functions 14671 14672 Also, r6xx/r7xx don't have indirect MC space, so 14673 don't try to access it. 14674 14675 commit f96d5b255425fbd02be2cad26edb590d474a5640 14676 Author: Alex Deucher <alexdeucher (a] gmail.com> 14677 Date: Thu Jul 23 13:25:04 2009 -0400 14678 14679 R6xx/R7xx: clip rendering to destination surface dimensions 14680 14681 commit e372f845b0defaf2d2c9ef3cbbf7498e09d9372e 14682 Author: Michel Dnzer <daenzer (a] vmware.com> 14683 Date: Sun Jul 19 16:05:29 2009 +0200 14684 14685 Guard reference to CRTC active field. 14686 14687 Fixes build against older xserver. 14688 14689 commit e38305aebdc95f80f5b4b3e5ba541ea67dc05f01 14690 Author: Michel Dnzer <daenzer (a] vmware.com> 14691 Date: Sat Jul 18 22:23:45 2009 +0200 14692 14693 Also drop DRM master in KMS CloseScreen. 14694 14695 The LeaveVT hook isn't always called when the server dies, e.g. when quitting 14696 from the GDM greeter. This may cause existing servers to fall over if the dying 14697 server process still exists when they try to re-acquire master and set a mode. 14698 14699 Also use drmSet/DropMaster() rather than ioctl() directly. 14700 14701 commit a43c660a00147bfae5ca601f4720b2680b75211f 14702 Author: Michel Dnzer <daenzer (a] vmware.com> 14703 Date: Sat Jul 18 22:21:59 2009 +0200 14704 14705 Set CRTC active flag to TRUE after a successful KMS mode set. 14706 14707 Otherwise some things like changing the colour map won't work properly. 14708 14709 commit f32069f42b3e4643f15148d0b96164def00dcc74 14710 Author: Michel Dnzer <daenzer (a] vmware.com> 14711 Date: Sat Jul 18 22:20:08 2009 +0200 14712 14713 Set the STALL bit in the CRTC_GUI_TRIG_VLINE registers. 14714 14715 As recommended by the register reference when using the WAIT_CRTC_VLINE bit in 14716 the WAIT_UNTIL register, as we are. 14717 14718 commit 27bd9fc9b2c0eeff488c1f26f1355a7fa3655520 14719 Author: Michel Dnzer <daenzer (a] vmware.com> 14720 Date: Sat Jul 18 22:14:27 2009 +0200 14721 14722 Enable vsync for DRI2 region copies. 14723 14724 commit 4b1b28865ffb17f58a70d955212fd51cb1fb55ab 14725 Author: Alex Deucher <alexdeucher (a] gmail.com> 14726 Date: Thu Jul 16 17:09:33 2009 -0400 14727 14728 R6xx/r7xx: reduce the amount of default state setup 14729 14730 Most of the default state setup was extraneous or duplicated 14731 in the accel ops. There's no need to reset all the consts 14732 everytime you reset the default state for example. This leads 14733 to a 10x reduction in default state. 14734 14735 commit 7e6557b81026d8a4d6c837839e68c28b151f8a97 14736 Author: Peter Hutterer <peter.hutterer (a] who-t.net> 14737 Date: Thu Jul 16 11:33:57 2009 +1000 14738 14739 Update to xextproto 7.1 support. 14740 14741 DPMS header was split into dpms.h (client) and dpmsconst.h (server). Drivers 14742 need to include dpmsconst.h if xextproto 7.1 is available. 14743 14744 Signed-off-by: Peter Hutterer <peter.hutterer (a] who-t.net> 14745 14746 commit 95d431e8b260dd43cf6d93b90eecab1dc0ec7c26 14747 Author: Dave Airlie <airlied (a] redhat.com> 14748 Date: Wed Jul 15 12:07:01 2009 +1000 14749 14750 radeon: emit colorpitch relocs. 14751 14752 This causes relocs to be emitted for the colorpitch. 14753 14754 Shouldn't have no effect on current kernels, but will keep DDX 14755 going on kms kernels when tiling lands 14756 14757 Also contains a missing reloc space for r200 14758 14759 commit 8312763977b22e108b7d003cbf3256e0165e4439 14760 Author: Alex Deucher <alexdeucher (a] gmail.com> 14761 Date: Tue Jul 14 18:13:27 2009 -0400 14762 14763 Add an r420 AGP quirk 14764 14765 fixes bug 22726 14766 14767 commit 5473eeeb45468fd200fec49a8506281628a5e4b5 14768 Author: Alex Deucher <alexdeucher (a] gmail.com> 14769 Date: Tue Jul 14 14:05:13 2009 -0400 14770 14771 R3/4/5xx: only upload the bicubic texture once 14772 14773 Upload the bicubic texture once during textured video init 14774 rather than once per frame. Suggested by Michel Daenzer on 14775 IRC. 14776 14777 commit 0485f27bc3d75cb6ab320e8164dbe6ea2713c78e 14778 Author: Alex Deucher <alexdeucher (a] gmail.com> 14779 Date: Tue Jul 14 13:11:12 2009 -0400 14780 14781 R3/4/5xx: fix bicubic Xv filtering with KMS changes 14782 14783 fixes bug 22730 14784 14785 commit ca4a9efef987108266ae59afbb5142d4d2000c8e 14786 Author: Alex Deucher <alexdeucher (a] gmail.com> 14787 Date: Mon Jul 13 11:28:41 2009 -0400 14788 14789 Add configure option to disable kms support 14790 14791 Useful to building with mixed kms/non-kms setups 14792 14793 commit 035e8d1d5593c12828bb079de4e663cf1b1f1674 14794 Author: Edward O'Callaghan <eocallaghan (a] auroraux.org> 14795 Date: Sun Jul 12 13:45:29 2009 +0200 14796 14797 Fix a warning 14798 14799 Signed-off-by: Nicolai Hhnle <nhaehnle (a] gmail.com> 14800 14801 commit 76af48c43f829e7aebacc9f2a623823fa26ee22b 14802 Author: Gaetan Nadon <memsize (a] videotron.ca> 14803 Date: Wed Jul 8 09:58:56 2009 -0400 14804 14805 xf86-video-ati: use XORG_CHANGELOG macro to create ChangeLog. #22611 14806 14807 Build break: Makefile.am: command not found: git-log 14808 Adding the macro in configure.ac and use it in Makefile.am 14809 Refer to: https://bugs.freedesktop.org/show_bug.cgi?id=22611 14810 Tested: running autogen.sh, make and 'make dist' 14811 14812 Signed-off-by: Gaetan Nadon <memsize (a] videotron.ca> 14813 Signed-off-by: Peter Hutterer <peter.hutterer (a] who-t.net> 14814 14815 commit 43db263d301082e84e9bc304816bcbb206fe280e 14816 Author: Alex Deucher <alexdeucher (a] gmail.com> 14817 Date: Thu Jul 9 12:48:06 2009 -0400 14818 14819 RV280: another AGP quirk 14820 14821 fdo bug 12544 14822 14823 commit 69b5e5496f10a9f566d2e563862c96cb41952eb6 14824 Author: Alex Deucher <alexdeucher (a] gmail.com> 14825 Date: Thu Jul 9 12:42:45 2009 -0400 14826 14827 RV280: Add an AGP quirk 14828 14829 lp bug 370205 via fdo bug 12544 14830 14831 commit a6d6c900e60a3685ee9e93c34eb7f6d237c45fd8 14832 Author: Alex Deucher <alexdeucher (a] gmail.com> 14833 Date: Thu Jul 9 11:30:57 2009 -0400 14834 14835 RS740: fixup display latency setting as per rs690 14836 14837 Both chips program the same in this regard 14838 14839 commit ea407570d39ace3162d372eda56bf791bfd80c24 14840 Author: Dave Airlie <airlied (a] redhat.com> 14841 Date: Thu Jul 9 10:46:23 2009 +1000 14842 14843 radeon: add shave support. 14844 14845 clean up output a lot 14846 14847 commit 8c03c1fdb5ea35570064946557050c87ca30582a 14848 Author: Alex Deucher <alexdeucher (a] gmail.com> 14849 Date: Wed Jul 8 18:34:57 2009 -0400 14850 14851 R6xx/R7xx: fix hangs on x2 cards with PM options 14852 14853 Changing the PCIE lanes on x2 cards results in a hang, 14854 so for now, disable it. 14855 14856 Fixes fdo bug 22669 14857 14858 commit 9645838c57f6b40837fdce23ce7f9faefb3d9966 14859 Author: Michel Dnzer <daenzer (a] vmware.com> 14860 Date: Wed Jul 8 19:18:02 2009 +0200 14861 14862 Further non-KMS fixes / cleanups. 14863 14864 Only compile tested. 14865 14866 commit 2d86ec4ddb1fb83be95ed9cdcbabf4d568a6d7b6 14867 Author: Alex Deucher <alexdeucher (a] gmail.com> 14868 Date: Wed Jul 8 12:33:34 2009 -0400 14869 14870 Fix the build for the !XF86DRM_MODE case 14871 14872 compile tested only. 14873 14874 commit a68afc2a80b8581cc3d67e9b5a95a10b0d4ead0e 14875 Author: Michel Dnzer <daenzer (a] vmware.com> 14876 Date: Wed Jul 8 09:15:28 2009 +0200 14877 14878 Fix RADEON_TRACE_FALL build. 14879 14880 commit 9ca0aeaad01b7c682396b30af48668f287a59d6e 14881 Author: Michel Dnzer <daenzer (a] vmware.com> 14882 Date: Wed Jul 8 09:13:36 2009 +0200 14883 14884 Pass on CS relocation offset values. 14885 14886 Not actually used ATM, but just in case... 14887 14888 commit f718dd794477fbdd5eeb4183f43ff25517c63034 14889 Author: Michel Dnzer <daenzer (a] vmware.com> 14890 Date: Wed Jul 8 09:11:18 2009 +0200 14891 14892 Don't set EXA UploadTo/DownloadFromScreen hooks when they'd always return FALSE. 14893 14894 commit 42e76d0b24b76fbcc6b08a4ea5003dbec368645c 14895 Author: Michel Dnzer <daenzer (a] vmware.com> 14896 Date: Wed Jul 8 08:51:10 2009 +0200 14897 14898 Don't always flush CS when switching from 2D to 3D. 14899 14900 This works fine here and doubles x11perf -aa10text scores, but may cause 14901 problems on other setups according to Dave. Let's see... 14902 14903 commit 94fe049c35187ddc3ec0472e9309c3a22879f415 14904 Author: Michel Dnzer <daenzer (a] vmware.com> 14905 Date: Wed Jul 8 08:36:22 2009 +0200 14906 14907 KMS CS fixes, take two. 14908 14909 Avoid flushing CS in the middle of (setting up state for) an operation, 14910 properly finish/restart the operation. 14911 14912 commit 0519f15af2fe5e7e5a6c94e7203a96fe363c643e 14913 Author: Michel Dnzer <daenzer (a] vmware.com> 14914 Date: Tue Jul 7 10:04:09 2009 +0200 14915 14916 Fix DRI2 with current xserver Git. 14917 14918 commit 7ad11fec2579807371360cb680e593f723ebaa1a 14919 Author: Dave Airlie <airlied (a] redhat.com> 14920 Date: Mon Jul 6 18:43:08 2009 +1000 14921 14922 radeon: add make dist headers 14923 14924 commit e68fbf743ec88e0fa3a9f1ffe83daf251e250b21 14925 Author: Dave Airlie <airlied (a] redhat.com> 14926 Date: Mon Jul 6 18:32:32 2009 +1000 14927 14928 git: fix git-log to git log 14929 14930 commit 5eeb1fd19c6797cc76ce308570dffe34cb82b24d 14931 Author: Dave Airlie <airlied (a] linux.ie> 14932 Date: Mon Jul 6 17:45:50 2009 +1000 14933 14934 radeon: oops fix compile without libdrm 14935 14936 commit 98824f37fc74dc1fec2677515d0dc3cfc263f00a 14937 Author: Dave Airlie <airlied (a] redhat.com> 14938 Date: Mon Jul 6 15:29:53 2009 +1000 14939 14940 radeon: move state changes into flush function 14941 14942 commit 503b51343267248b4c6994b88238dd5384528f03 14943 Author: Dave Airlie <airlied (a] redhat.com> 14944 Date: Mon Jul 6 15:20:13 2009 +1000 14945 14946 radeon: taken from kms-support + CS fixes commit. 14947 14948 re-emit 2D state and reset 3D state emit on flush 14949 14950 commit d78d50591851d14543f0935d051a59ef29751bbc 14951 Author: Dave Airlie <airlied (a] redhat.com> 14952 Date: Mon Jul 6 14:45:07 2009 +1000 14953 14954 radeon: port to new space checking in libdrm interface 14955 14956 This uses a new libdrm interface which shares code with mesa. 14957 14958 It also fixes the bo to flush when full instead of never flushing. 14959 14960 It survives gtkperf -a here which the driver didn't before now 14961 14962 commit e1200cb89218930d01330ba0114e013438655cce 14963 Author: Dave Airlie <airlied (a] redhat.com> 14964 Date: Mon Jul 6 15:10:29 2009 +1000 14965 14966 Revert "KMS CS fixes." 14967 14968 This reverts commit 11cf8a28eb46f48d101888552266921bee8b63fc. 14969 14970 I'll bring this back in a few minutes, I've got a big change introducing 14971 a new cs space accounting scheme which this messes up and I think 14972 probably conflicts with what I've done. 14973 14974 I promise i'll go over this and pick the bits that are still useful :) 14975 14976 commit 11cf8a28eb46f48d101888552266921bee8b63fc 14977 Author: Michel Dnzer <daenzer (a] vmware.com> 14978 Date: Sun Jul 5 15:14:59 2009 +0200 14979 14980 KMS CS fixes. 14981 14982 * Flush CS before running out of space. 14983 * Don't flush CS every time when switching between 2D and 3D engine, but make 14984 sure to re-initialize 2D/3D state after flushing. 14985 14986 Doesn't seem quite perfect yet (some rendering seems to get dropped on the floor 14987 sometimes), but this should fix a lot of CS failures people have been seeing. 14988 14989 commit 8d2f712eaf1e569fd92bbe2db5aceb43b7b367d1 14990 Author: Dave Airlie <airlied (a] linux.ie> 14991 Date: Fri Jul 3 10:40:59 2009 +1000 14992 14993 radeon: don't set up overlay under KMS until we figure out how. 14994 14995 commit 2a622168606d98c404708b879e6417c8f7d7305c 14996 Author: Michel Dnzer <daenzer (a] vmware.com> 14997 Date: Thu Jul 2 15:03:53 2009 +0200 14998 14999 Test BO pointer before passing it to radeon_bo_unref(). 15000 15001 commit 36a4ea5a9c7062f4c8aa7480bb7488ce2062390f 15002 Author: Michel Dnzer <daenzer (a] vmware.com> 15003 Date: Thu Jul 2 14:52:44 2009 +0200 15004 15005 Warning fixes. 15006 15007 commit 65465d56c1bf943113518a5b2f57b3170213bab9 15008 Author: Dave Airlie <airlied (a] linux.ie> 15009 Date: Thu Jul 2 18:38:38 2009 +1000 15010 15011 radeon: fix up configure check better 15012 15013 commit ad17108b72c9adf773a122b4c054c7b92e6a8b9d 15014 Author: Dave Airlie <airlied (a] linux.ie> 15015 Date: Thu Jul 2 17:25:57 2009 +1000 15016 15017 radeon: fix it properly. 15018 15019 I blame the flu 15020 15021 commit e1b9cb9ea665eaa84e64a386299efa0819d55df4 15022 Author: Dave Airlie <airlied (a] redhat.com> 15023 Date: Thu Jul 2 17:18:47 2009 +1000 15024 15025 radeon: fix compile without kms 15026 15027 commit de427fac6912a625f7e4d8b982a1f7a1ba80dfbe 15028 Author: Dave Airlie <airlied (a] redhat.com> 15029 Date: Thu Jul 2 15:41:43 2009 +1000 15030 15031 radeon/dri2: fix dri2 - needs newest libdrm_radeon 15032 15033 commit 2b671adba54ebff1756dd990123f4bade6e35d61 15034 Author: Dave Airlie <airlied (a] redhat.com> 15035 Date: Thu Jul 2 15:25:19 2009 +1000 15036 15037 radeon/dri2: call proper gem bo function 15038 15039 commit 851b2b8cab883e4422745cca1827aa8ac9f193a3 15040 Author: Dave Airlie <airlied (a] redhat.com> 15041 Date: Thu Jul 2 15:17:13 2009 +1000 15042 15043 radeon/kms: add textured video BO size checks 15044 15045 commit 9d092060236e8587697fdd57a7a01f3c0dddd680 15046 Author: Dave Airlie <airlied (a] redhat.com> 15047 Date: Thu Jul 2 15:17:03 2009 +1000 15048 15049 kms: drop DGA 15050 15051 commit 207174ea964c561076bdd5297a2fb6f136b4b645 15052 Author: Dave Airlie <airlied (a] redhat.com> 15053 Date: Thu Jul 2 15:16:47 2009 +1000 15054 15055 radeon: fix CS imbalance 15056 15057 commit 4d24a9d2c0be0c025266cc8cdf7e8d7bb065cdb1 15058 Author: Dave Airlie <airlied (a] redhat.com> 15059 Date: Thu Jul 2 14:55:45 2009 +1000 15060 15061 radeon: enable KMS build 15062 15063 commit 9bcd3ae7966eb7a17b922bcd33e0d6f016245614 15064 Author: Dave Airlie <airlied (a] redhat.com> 15065 Date: Thu Jul 2 14:55:18 2009 +1000 15066 15067 radeon/kms: fixup 3d state emission for KMS 15068 15069 This fixes font rendering under KMS (should have looked at dmesg) 15070 15071 commit ac1607bb23f0c7584648407a3ac3fb27142e186e 15072 Author: Dave Airlie <airlied (a] redhat.com> 15073 Date: Thu Jul 2 14:55:05 2009 +1000 15074 15075 kms: no need to reference fifo slots 15076 15077 commit 1782ce28953184776c90eb1255208a3e0ad245f0 15078 Author: Dave Airlie <airlied (a] redhat.com> 15079 Date: Tue Jun 30 16:55:26 2009 +1000 15080 15081 radeon: add KMS support (still disabled) 15082 15083 This adds DRI2 + KMS + driver pixmaps support to the driver. 15084 15085 I've decided to just do a completely separate KMS driver file 15086 instead of hacking the crap out of radeon_driver.c. So now 15087 I do the KMS check in radeon_probe.c time and set the DDX 15088 pointed up to a completely different set at this stage. 15089 15090 This avoids a lot of if (kms) type crap in the code at 15091 the expense of making sure we make changes to both files 15092 if necessary. 15093 15094 This code is still disabled in configure.ac as I broke EXA composite 15095 rendering somehow in KMS mode 15096 15097 commit ac4bd24a66c1bdda0293f770a3f891e2b88cc8ee 15098 Author: Alex Deucher <alexdeucher (a] gmail.com> 15099 Date: Tue Jun 30 16:00:06 2009 -0400 15100 15101 EXA: fix the build on servers without EXA_HANDLES_PIXMAPS 15102 15103 commit 49a0ca19ee4109cdd39a43fbf4bedf3f0557d6ec 15104 Author: Michel Dnzer <daenzer (a] vmware.com> 15105 Date: Tue Jun 30 17:22:50 2009 +0200 15106 15107 Warning fixes. 15108 15109 commit e932836691aeaec37794fdaed2dabb22710fd171 15110 Author: Dave Airlie <airlied (a] redhat.com> 15111 Date: Tue Jun 30 16:24:37 2009 +1000 15112 15113 radeon: initial preparation for kms patch. 15114 15115 This patch contains most of the changes to the EXA and texture video 15116 accel code. 15117 15118 It adds a few bits of pixmap support but doesn't actually do anything 15119 useful KMS yet. 15120 15121 Testing this should not have any regressions over what we have already, 15122 biggest worries are r6xx, I've fixed a textured video one, but no idea 15123 what other might lurk 15124 15125 It won't build against libdrm radeon yet either 15126 15127 commit bb04b450ed00ca4b1aa44c33085567d47b33b547 15128 Author: Fredrik Hglund <fredrik (a] kde.org> 15129 Date: Mon Jun 29 22:38:26 2009 +0200 15130 15131 Check if the composite op is supported in R200CheckComposite. 15132 15133 commit f39cafc5c4a93a16ac93756473ebb1f33c7881db 15134 Author: Alex Deucher <alexdeucher (a] gmail.com> 15135 Date: Sun Jun 28 20:42:49 2009 -0400 15136 15137 radeon: call hdmi-b connectors dvi in randr 15138 15139 Technically HDMI-B is a glorfied DL DVI, but this is confusing 15140 to users, so call it DVI when reporting randr outputs. This 15141 allows us to remove two quirks where the bios was actually 15142 correct. 15143 15144 commit 407e676b04117c7ca3dcc0cc072f3552b03c64e5 15145 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 15146 Date: Sun Jun 28 16:54:32 2009 -0400 15147 15148 R6xx/R7xx: fix pixel centers 15149 15150 Make sure we are using GL mode (centers at 0.5) rather 15151 than D3D mode (centers at 0.0). This also fixes fdo 15152 bug 21963. 15153 15154 commit 68001981f22173ff949720055dba89291f284474 15155 Author: Alex Deucher <alexdeucher (a] gmail.com> 15156 Date: Sun Jun 28 16:20:10 2009 -0400 15157 15158 Add some missing M96 pci ids 15159 15160 fixes fdo bug 22404 15161 15162 commit 06d5d8d7fd92b58b2467c782e57a80a3520a2c90 15163 Author: Adam Jackson <ajax (a] redhat.com> 15164 Date: Thu Jun 25 13:51:36 2009 -0400 15165 15166 Add a (harmlessly) missing break statement. 15167 15168 commit 7d1d968fdc8de1e0ab0be916a1f3a4a398af47aa 15169 Author: Adam Jackson <ajax (a] redhat.com> 15170 Date: Thu Jun 25 13:49:18 2009 -0400 15171 15172 Cosmetic cleanup to digital output detection 15173 15174 No functional change, but this logic would need extending anyway for 15175 DisplayID, so unify it now. 15176 15177 commit 96581746eea2f4e29a04f9266ec9649f25480424 15178 Author: Adam Jackson <ajax (a] redhat.com> 15179 Date: Wed Jun 24 17:05:47 2009 -0400 15180 15181 atom: Fix phantom VGA connector when HDMI-B present. 15182 15183 HDMI-B is really a glorified DL-DVI connector, the analog pins work fine. 15184 15185 commit 496adc4dc7cf53fc6acfecb1158f1491d5f36168 15186 Author: Michel Dnzer <daenzer (a] vmware.com> 15187 Date: Fri Jun 19 14:37:01 2009 +0200 15188 15189 Revert "Build fixes for -Werror that aren't ready for prime time yet." 15190 15191 This reverts commit ab20ea1530ec8a3a9fd4089775c4e157e50fe1bf. 15192 15193 Whoops, pushed wrong branch... 15194 15195 commit ab20ea1530ec8a3a9fd4089775c4e157e50fe1bf 15196 Author: Michel Dnzer <daenzer (a] vmware.com> 15197 Date: Sun Jun 14 16:50:07 2009 +0200 15198 15199 Build fixes for -Werror that aren't ready for prime time yet. 15200 15201 commit 77e3537d312175a25f0e21cc07c3a96f78c3b35a 15202 Author: Markus Gapp <markus.gapp (a] gmx.net> 15203 Date: Sat Jun 13 11:51:29 2009 -0400 15204 15205 Add quirk for asus hd3450 board 15206 15207 Fixes bug 22266 15208 15209 commit e8e0bb1ba29761c94d473e42ca6ada161cfff557 15210 Author: Alex Deucher <alexdeucher (a] gmail.com> 15211 Date: Fri Jun 12 11:07:43 2009 -0400 15212 15213 Fix count in 3d state setup after 43374c7420e378918bec062f4cbd581f16adb6f0 15214 15215 commit 808c90a24c48da7fa97e15e2f12be5bb8fd8cc96 15216 Author: Alex Deucher <alexdeucher (a] gmail.com> 15217 Date: Thu Jun 11 16:09:40 2009 -0400 15218 15219 Only enable frac fb divs on rs600/rs690/rs740 for now 15220 15221 fixes bug 21413 15222 15223 commit d095bb3157faaee10b0130ea9b1032e0b533aaee 15224 Author: Alex Deucher <alexdeucher (a] gmail.com> 15225 Date: Thu Jun 11 12:34:15 2009 -0400 15226 15227 Re-enable frac feedback dividers on AVIVO hardware 15228 15229 Preferring a freq <= target frequency seems to fix the 15230 previous issues with frac fb divs. Appears to work fine 15231 on all hardware I've tested on. See fdo bug 22229 for 15232 details. Should fix bugs: 22229, 21553, 21413, and 15233 possibly other ones related to unstable images at certain 15234 resolutions. 15235 15236 commit 55fbdbae83d1563b472f49d0436c9298e390be66 15237 Author: Alex Deucher <alexdeucher (a] gmail.com> 15238 Date: Thu Jun 11 12:31:53 2009 -0400 15239 15240 Add PLL flag to prefer frequencies <= the target freq 15241 15242 This appears to be needed when using fractional feedback 15243 dividers. Based on a patch from Tom Hirst. See fdo 15244 bug 22229 for more details. 15245 15246 commit 43374c7420e378918bec062f4cbd581f16adb6f0 15247 Author: Alex Deucher <alexdeucher (a] gmail.com> 15248 Date: Tue Jun 9 14:39:54 2009 -0400 15249 15250 r1xx/r2xx: clip 3D rendering to destination buffer 15251 15252 commit 74b0a38c75c1d6a6344015d32d84b43d9e80df1d 15253 Author: Alex Deucher <alexdeucher (a] gmail.com> 15254 Date: Tue Jun 9 14:26:21 2009 -0400 15255 15256 Consolidate indentical RADEONDoneSolid/RADEONDoneCopy functions 15257 15258 commit f6dff8836729cc9d29264080c9cfa0e1df6182cf 15259 Author: Alex Deucher <alexdeucher (a] gmail.com> 15260 Date: Tue Jun 9 13:31:15 2009 -0400 15261 15262 r3xx/r4xx/r5xx: clip 3D rendering to destination buffer 15263 15264 commit 48dbf09e1309c815ac8b733145a0310083a29933 15265 Author: Alex Deucher <alexdeucher (a] gmail.com> 15266 Date: Tue Jun 9 13:12:40 2009 -0400 15267 15268 r3xx/r4xx/r5xx: refactor exa prepare composite 15269 15270 - consolidate code duplicated between r3xx/r4xx and r5xx 15271 - make pixel shader setup easier to read and less 15272 redundant 15273 15274 commit ec1a3ccdf61c2226617b571cc4a209a542e9d7e3 15275 Author: Alex Deucher <alexdeucher (a] gmail.com> 15276 Date: Mon Jun 8 15:31:57 2009 -0400 15277 15278 R6xx/r7xx: don't clamp tex coords in composite VS 15279 15280 Fixes broken repeat modes. see fdo bugs 21818, 22139 15281 15282 commit 32c218c5633fb7ffeb19e717bf645c1b25dd4ef5 15283 Author: Alan Coopersmith <alan.coopersmith (a] sun.com> 15284 Date: Thu Jun 4 07:23:03 2009 -0700 15285 15286 Add link to patch submission instructions to README 15287 15288 Signed-off-by: Alan Coopersmith <alan.coopersmith (a] sun.com> 15289 15290 commit 39f736d6cb65fc7c3fd769478023fc87c1af5979 15291 Author: Alex Deucher <alexdeucher (a] gmail.com> 15292 Date: Thu Jun 4 09:57:22 2009 -0400 15293 15294 atom: connector quirk 15295 15296 Some r6xx cards have a VGA and an HDMI port with 15297 a shared ddc line listed as an HDMI port with both 15298 analog and digital encoders. 15299 15300 Fixes bug 19943 15301 15302 commit 7599dc40855c33a5fbd8e9bbc2b4cd62752fb7df 15303 Author: Alex Deucher <alexdeucher (a] gmail.com> 15304 Date: Fri May 29 16:29:50 2009 -0400 15305 15306 Add RV740 (HD4770) support 15307 15308 Verified by ernstp on IRC 15309 15310 commit ee0424078a3ea26f58c2998c9e58e065fb63f1a0 15311 Author: Alex Deucher <alexdeucher (a] gmail.com> 15312 Date: Wed May 27 18:37:58 2009 -0400 15313 15314 radeon: remove dead code 15315 15316 commit b34df233115c0d82d7bcf82e041afbc55981ce82 15317 Author: Alex Deucher <alexdeucher (a] gmail.com> 15318 Date: Tue May 26 17:47:01 2009 -0400 15319 15320 R3xx/R4xx: increase EXA render target limits 15321 15322 now that we are using 1/12 subpixel mode 15323 15324 commit 1457bc089d1395d18cf582a4d52444005f013753 15325 Author: Matthieu Herrb <matthieu.herrb (a] laas.fr> 15326 Date: Tue May 26 17:21:07 2009 -0400 15327 15328 radeon: fix segfault if Option "NoAccel" is set 15329 15330 commit 4079f97958cf1d5fd76a83aa61b93d6289bb9b1e 15331 Author: Michel Dnzer <daenzer (a] vmware.com> 15332 Date: Sat May 23 13:51:00 2009 +0200 15333 15334 Bump version to mark Git snapshot. 15335 15336 The fourth version component isn't visible in the log file. 15337 15338 commit 88a9e98341d96e5e7f48b69aed597d1bada6313a 15339 Author: Alex Deucher <alexdeucher (a] gmail.com> 15340 Date: Tue May 19 17:11:34 2009 -0400 15341 15342 Pre-atom: Improve engine clock setup code 15343 15344 Also remove SetMemoryClk() for pre-atom cards for now 15345 as it requires quite a bit more asic specific work. 15346 To set the mclk we'll need to use the mem reset/dll tables 15347 in the bios. 15348 15349 commit 1429133ffdbfe046249fcaaa4cbee432a1012e55 15350 Author: Michel Dnzer <daenzer (a] vmware.com> 15351 Date: Mon May 18 17:59:00 2009 +0200 15352 15353 EXA: Set new EXA_SUPPORTS_OFFSCREEN_OVERLAPS flag when we have a 2D engine. 15354 15355 commit e9cca1e7dc521d3756bb2d97f1d703b9ad65cb47 15356 Author: Alex Deucher <alexdeucher (a] gmail.com> 15357 Date: Mon May 18 10:48:25 2009 -0400 15358 15359 r4xx: Add some missing pci ids 15360 15361 Noticed by Romain in bug 21798 15362 15363 commit 76fffd51027820b3a6ea10de2be67d5fd4f3e843 15364 Author: Alex Deucher <alexdeucher (a] gmail.com> 15365 Date: Fri May 15 16:35:13 2009 -0400 15366 15367 R100: switch to normalized texture coordinates 15368 15369 for consistency with the other asics 15370 15371 commit 1a1eef31e36f98b7a30e27e567c1b66e0e4cf7dc 15372 Author: Alex Deucher <alexdeucher (a] gmail.com> 15373 Date: Thu May 14 15:48:54 2009 -0400 15374 15375 radeon: try harder when checking if a card is posted 15376 15377 fixes debian bug 524280, possibly others. 15378 15379 commit 22e39392297fa11003df90c175db3c449d8f9853 15380 Author: Michel Dnzer <daenzer (a] vmware.com> 15381 Date: Thu May 14 10:24:07 2009 +0200 15382 15383 Pre-R600 EXA: Fix coordinate limits off-by-one error. 15384 15385 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=21598 . 15386 15387 commit fa09b058c7a17689989e600ffd465856a058579d 15388 Author: Alex Deucher <alexdeucher (a] gmail.com> 15389 Date: Wed May 13 16:10:40 2009 -0400 15390 15391 R6xx/R7xx Xv: normalize texture coordinates in the vertex shader 15392 15393 commit 026b6f820d6caea17d2a082193e850713d5770a8 15394 Author: Alex Deucher <alexdeucher (a] gmail.com> 15395 Date: Wed May 13 15:48:32 2009 -0400 15396 15397 R6xx/R7xx: do EXA transforms in the vertex shader 15398 15399 commit cd89241396d1931b04cfbdd8d553be16dbf9c360 15400 Author: Alex Deucher <alexdeucher (a] gmail.com> 15401 Date: Tue May 12 17:30:02 2009 -0400 15402 15403 R3xx-R5xx: do EXA transforms in the vertex shader 15404 15405 commit 15ee78d37f8b64c3e6d234e7ab37a31e82327b6b 15406 Author: Alex Deucher <alexdeucher (a] gmail.com> 15407 Date: Mon May 11 02:22:13 2009 -0400 15408 15409 RV770: add missing pci id 15410 15411 Fixes bug 21209 15412 15413 commit 5ff51ea83f9fdd836e830c77e3561eaef4b83183 15414 Author: Alex Deucher <alexdeucher (a] gmail.com> 15415 Date: Fri May 8 13:19:26 2009 -0400 15416 15417 R7xx: add support for reading disabled bios images 15418 15419 commit 76fdae9018330aa5af0d57339baff3ca4834e1b5 15420 Author: Alex Deucher <alexdeucher (a] gmail.com> 15421 Date: Thu May 7 02:59:32 2009 -0400 15422 15423 R3xx/R4xx tex vid: no need to check if IS_R300_3D 15424 15425 function is r3xx/r4xx specific already 15426 15427 commit 23a8ada06a5b1ccc975e89f129c00382ecbf29df 15428 Author: Alex Deucher <alexdeucher (a] gmail.com> 15429 Date: Thu May 7 02:34:22 2009 -0400 15430 15431 R3xx/R4xx tex vid: increase the guardband limit for rendering with a tri 15432 15433 limit goes from 2880 to 4021 now that we've switched to 1/12 subpixel mode. 15434 15435 This allows us to render with a clipped tri at higher resolutions preventing 15436 diagonal tearing. 15437 15438 commit 0c3541f7013c53fb8096ada63cc7b3577524aa4c 15439 Author: Alex Deucher <alexdeucher (a] gmail.com> 15440 Date: Wed May 6 12:24:35 2009 -0400 15441 15442 AVIVO: set default cursor mode in cursor init 15443 15444 Should fix bug 21557 15445 15446 commit da58e351b6398fa18b1d9c1a57a6e49b285f610f 15447 Author: Alex Deucher <alexdeucher (a] gmail.com> 15448 Date: Mon May 4 16:03:53 2009 -0400 15449 15450 AVIVO: fix cursoe corruption bands for real 15451 15452 Don't have to leave both cursors enabled, just have to use 15453 the same mode for both cursors whether or not they are enabled. 15454 15455 commit 2c8e130f73c680d4a7381b2ef37982b82c6ee478 15456 Author: Alex Deucher <alexdeucher (a] gmail.com> 15457 Date: Mon May 4 13:40:17 2009 -0400 15458 15459 radeon: switch to EXA by default 15460 15461 commit 8155a62df6817358a12cc2d613c1470ca55996ef 15462 Author: Alex Deucher <alexdeucher (a] gmail.com> 15463 Date: Mon May 4 12:57:36 2009 -0400 15464 15465 AVIVO: move cusor offscreen when disabling 15466 15467 Adapted from Yang's patch. Setting size to zero is 15468 actually a 1x1 cursor. 15469 15470 commit a1c64ea5224009779ccad66b0f84d861eae966ac 15471 Author: Alex Deucher <alexdeucher (a] gmail.com> 15472 Date: Sun May 3 19:42:44 2009 -0400 15473 15474 r3xx-r5xx: switch to 1/12 subpixel precision 15475 15476 - based on similar patch in Jerome's cs ddx tree 15477 - also fix clipping offsets 15478 - should eventually allow for 4k render targets 15479 - mesa driver uses 1/12 mode, this avoids changing the 15480 subpixel mode when switching between ddx and mesa 15481 15482 commit 3880bc3c95764acccd62cfa438932458df95daf6 15483 Author: Alex Deucher <alexdeucher (a] gmail.com> 15484 Date: Sun May 3 19:01:28 2009 -0400 15485 15486 Remove some old debugging info 15487 15488 commit 3362381aa3a97c1da564f8a479ed4e7b0d6368b8 15489 Author: Alex Deucher <alexdeucher (a] gmail.com> 15490 Date: Sun May 3 18:52:00 2009 -0400 15491 15492 r2xx-r4xx: fix typo in last i2c fix and clarify hw i2c pin sel 15493 15494 commit e01b8dc827ce1913927e1a4fad05a99a2dd9e805 15495 Author: Alex Deucher <alexdeucher (a] gmail.com> 15496 Date: Sun May 3 13:09:39 2009 -0400 15497 15498 RS690: gpio/connector fixes 15499 15500 Fixes bug 21521 15501 15502 commit 1f70c9f05df9017d87b37f887e1eccd6d0568a02 15503 Author: Alex Deucher <alexdeucher (a] gmail.com> 15504 Date: Thu Apr 30 19:18:13 2009 -0400 15505 15506 R3xx-R5xx: don't set TX_OFFSET_RS in RS_INST_COUNT 15507 15508 Isn't necessary and seems to cause problems for RS690 users 15509 15510 commit de17d36c1f710305870e3c636cae3f742f1cb6d6 15511 Author: Alex Deucher <alexdeucher (a] gmail.com> 15512 Date: Thu Apr 30 19:13:09 2009 -0400 15513 15514 AVIVO: first pass at display watermark setup 15515 15516 This helps with flickering and blanking when 15517 there is contention for MC bandwidth. 15518 15519 commit 7d9f643ae3d07e51e644a5979ca90bc2c102bc89 15520 Author: Alex Deucher <alexdeucher (a] gmail.com> 15521 Date: Wed Apr 29 12:38:15 2009 -0400 15522 15523 radeon: fix several dpms issues 15524 15525 - atom dpms was unblanking for standby/suspend 15526 - return if r600+ in radeon_crtc_modeset_ioctl() 15527 - remove seprate standby/suspend handling in legacy 15528 crtc dpms. we turn the outputs off, so turn the 15529 crtcs off too. 15530 - disable the crtcs in legacy crtc dpms 15531 - move radeon_crtc_modeset_ioctl() calls to radeon_crtc_dpms() 15532 so they will get called for both atom and legacy paths 15533 15534 Should fix bug 21321 15535 15536 commit 35183ef528411eb0122cf48550d93f921d291d12 15537 Author: Alex Deucher <alexdeucher (a] gmail.com> 15538 Date: Wed Apr 29 01:55:10 2009 -0400 15539 15540 AVIVO: better fix for cursor flickering/corruption issues 15541 15542 Should prevent ghost cursor from being left on screen. 15543 15544 commit 53e75cbd0ba8f39799d7b776cdc59d60aa2a6916 15545 Author: Alex Deucher <alexdeucher (a] gmail.com> 15546 Date: Wed Apr 29 01:05:31 2009 -0400 15547 15548 R6xx/r7xx: fix CURSOR_SWAPPING_* macros 15549 15550 r6xx/r7xx have different swapper regs 15551 15552 commit 4e948e2aaa59267a9aa797396add2c6b7b4c1e7c 15553 Author: Alex Deucher <alexdeucher (a] gmail.com> 15554 Date: Tue Apr 28 19:28:25 2009 -0400 15555 15556 AVIVO: fix cursor corruption/flickering issues 15557 15558 When both crtcs are enabled, both cursors need to be enabled 15559 or you get corruption bands. 15560 15561 Thanks to Yang Zhao for figuring this out. 15562 15563 commit 4d0beaf02b28d5a5dac4ff6053593a2f979f92ec 15564 Author: Alex Deucher <alexdeucher (a] gmail.com> 15565 Date: Tue Apr 28 12:33:24 2009 -0400 15566 15567 RV410: SE variants only have 1 quad pipe 15568 15569 Should fix EXA corruption with the 3D engine. 15570 Mesa and drm patches forthcoming. 15571 15572 Reported by Kano on IRC. 15573 15574 commit abd565ae71d9844d5afcd658628ca7778ce26ab1 15575 Author: Alex Deucher <alexdeucher (a] gmail.com> 15576 Date: Mon Apr 27 14:21:38 2009 -0400 15577 15578 AVIVO: disable fractional feedback divs 15579 15580 Doesn't seem to work well in all cases. Needs more investigation. 15581 fixes bug 21413 15582 15583 commit 93626b1277f57bd559cf71b263221209b02e0046 15584 Author: Alex Deucher <alexdeucher (a] gmail.com> 15585 Date: Mon Apr 27 03:35:24 2009 -0400 15586 15587 AVIVO: make sure cursor width isn't negative 15588 15589 commit 32af0f097bccc4499d67e969d46ba35e7487c5e2 15590 Author: Dave Airlie <airlied (a] linux.ie> 15591 Date: Sat Apr 25 19:39:34 2009 +1000 15592 15593 Revert "r300/r500: set R clamp mode." 15594 15595 This reverts commit dab3d61619d91ae6a3da3e9b29b8258313653813. 15596 15597 commit b2c8f57bccbd46164907049a26ad2dd209d7b108 15598 Author: Dave Airlie <airlied (a] linux.ie> 15599 Date: Sat Apr 25 19:39:19 2009 +1000 15600 15601 Revert "r3xx/r5xx: fix typo in last commit" 15602 15603 This reverts commit 1a7b982457ead5f21c6e62a409ca0dbf86005607. 15604 15605 commit 253731dce6ab25122f2eda559233c098536b7143 15606 Author: Dave Airlie <airlied (a] linux.ie> 15607 Date: Sat Apr 25 19:39:05 2009 +1000 15608 15609 Revert "radeon: attempt to fix R clamping for repeat again" 15610 15611 This reverts commit 5998f262a52cb85b334fcc1fe24c7b2ae474ce93. 15612 15613 commit 5998f262a52cb85b334fcc1fe24c7b2ae474ce93 15614 Author: Dave Airlie <airlied (a] linux.ie> 15615 Date: Fri Apr 24 17:30:57 2009 +1000 15616 15617 radeon: attempt to fix R clamping for repeat again 15618 15619 commit 7ed27b926f28a1f088eace5034a7d9985f13752d 15620 Author: Alex Deucher <alexdeucher (a] gmail.com> 15621 Date: Thu Apr 23 14:06:06 2009 -0400 15622 15623 r2xx/r3xx/r4xx: further i2c fixups 15624 15625 - hw i2c engine has pin selection on r2xx/r2xx/r3xx chips 15626 - also switch hw i2c pin sel for external tmds 15627 15628 commit 0da80f34b5aa5999d2030ffbc5187328fa4e4ae5 15629 Author: Alex Deucher <alexdeucher (a] gmail.com> 15630 Date: Wed Apr 22 18:42:36 2009 -0400 15631 15632 radeon: minor cleanup and warning fix 15633 15634 commit 1a7b982457ead5f21c6e62a409ca0dbf86005607 15635 Author: Alex Deucher <alexdeucher (a] gmail.com> 15636 Date: Thu Apr 23 10:09:22 2009 -0400 15637 15638 r3xx/r5xx: fix typo in last commit 15639 15640 Noticed by osiris on IRC 15641 15642 commit dab3d61619d91ae6a3da3e9b29b8258313653813 15643 Author: Dave Airlie <airlied (a] redhat.com> 15644 Date: Thu Apr 23 15:46:44 2009 +1000 15645 15646 r300/r500: set R clamp mode. 15647 15648 On rs690 without this set, we appear to end up fetching from well 15649 outside the texture sometimes into memory that isn't mapped. 15650 15651 This "shouldn't" have any effect. 15652 15653 commit bbfd4625478e37d10fd58b23f46153f349370a52 15654 Author: Alex Deucher <alexdeucher (a] gmail.com> 15655 Date: Wed Apr 22 14:47:16 2009 -0400 15656 15657 radeon PM: update quirks 15658 15659 commit e3c564114eea4d032f920fcebd4040f1934c21b0 15660 Author: Alex Deucher <alexdeucher (a] gmail.com> 15661 Date: Wed Apr 22 14:31:44 2009 -0400 15662 15663 AVIVO: fix for cursor corruption issue 15664 15665 commit 24e4b73b4fbbb2c790e6120ede3caaa4e7e58359 15666 Author: Alex Deucher <alexdeucher (a] gmail.com> 15667 Date: Wed Apr 22 11:42:18 2009 -0400 15668 15669 radeon pll: add support for fractional feedback divs 15670 15671 Allows us to hit dot clocks much closer, especially on 15672 chips with non-27 Mhz reference clocks like most IGP chips. 15673 15674 This should fix most flickering and blanking problems with 15675 non-exact dot clocks. 15676 15677 commit efa0825a86a8dc0f03ebb42c576ed26189e9d4bb 15678 Author: Alex Deucher <alexdeucher (a] gmail.com> 15679 Date: Tue Apr 21 13:51:11 2009 -0400 15680 15681 Tex vid: fix top clipping 15682 15683 commit aa59b5b55e6ac64940fb3f0fa61f7b8fe36ba745 15684 Author: Alex Deucher <alexdeucher (a] gmail.com> 15685 Date: Tue Apr 21 12:06:24 2009 -0400 15686 15687 RS690: add quirk for acer board 15688 15689 thanks to Gino Badouri for testing 15690 15691 commit 0e5164d968316bbb59b4484a7df087854738068f 15692 Author: Alex Deucher <alexdeucher (a] gmail.com> 15693 Date: Tue Apr 21 04:47:43 2009 -0400 15694 15695 Tex vid: rework part 2 15696 15697 - use pPriv->w/h directly for tex coords 15698 - take src x/y offset into account when calculating tex coords 15699 - when copying data, only optimize for y clipping. In order 15700 to deal with the x clipping optimization, the copy routines 15701 or tex coords would need to be fixed. This should fix clipping 15702 problems with the current code. 15703 15704 commit 06a1b4169b506f69dd807380d9ce7210b3a3bc23 15705 Author: Alex Deucher <alexdeucher (a] gmail.com> 15706 Date: Tue Apr 21 01:21:51 2009 -0400 15707 15708 R4xx: avoid a PM related hang 15709 15710 commit 228b9ac87668241b09980fca17e9d866891285e6 15711 Author: Alex Deucher <alexdeucher (a] gmail.com> 15712 Date: Tue Apr 21 01:02:34 2009 -0400 15713 15714 radeon PM: add some quirks 15715 15716 commit b6febf7368d3df3f1d3fee6a87aa71c54e77cdcc 15717 Author: Alex Deucher <alexdeucher (a] gmail.com> 15718 Date: Mon Apr 20 18:04:18 2009 -0400 15719 15720 Tex vid: cleanup and unify common code 15721 15722 commit d096d1189c8dc5e19f68cc605ccffe1a8fef048b 15723 Author: Alex Deucher <alexdeucher (a] gmail.com> 15724 Date: Mon Apr 20 13:05:18 2009 -0400 15725 15726 Endian fixes for object table 15727 15728 commit 101ce28ec4af696549b7d6defebb5b4ad412176c 15729 Author: Alex Deucher <alexdeucher (a] gmail.com> 15730 Date: Sun Apr 19 12:13:18 2009 -0400 15731 15732 Remove old messages 15733 15734 R5xx/r6xx support isn't really experimental anymore 15735 15736 commit db177c70ce88af19c8e05eb916a15f3e124876eb 15737 Author: Alex Deucher <alexdeucher (a] gmail.com> 15738 Date: Fri Apr 17 01:05:15 2009 -0400 15739 15740 Update Xv info in man page 15741 15742 commit 00266177bed2dc8693df497ca3ec19f2dc4adc05 15743 Author: Alex Deucher <alexdeucher (a] gmail.com> 15744 Date: Thu Apr 16 20:33:28 2009 -0400 15745 15746 R3xx/R5xx: only apply Xv attributes if bicubic is disabled 15747 15748 Provides consistent output 15749 15750 commit 85323a7f84381fef7fad20c7f7ec601637af9aa7 15751 Author: Alex Deucher <alexdeucher (a] gmail.com> 15752 Date: Thu Apr 16 11:10:51 2009 -0400 15753 15754 R3xx/R4xx: set tex caching for Y texture when doing planar rendering 15755 15756 Doesn't affect performance, but docs indicate its the right 15757 thing to do. 15758 15759 commit 5ea5df22c038fc8f00984acc760e9d8c962bf902 15760 Author: Alex Deucher <alexdeucher (a] gmail.com> 15761 Date: Wed Apr 15 20:26:34 2009 -0400 15762 15763 Tex vid: remove remnants of XV_HWPLANAR 15764 15765 no longer needed as bicubic is the only thing that uses 15766 the old csc code. 15767 15768 commit 9091b3f5f13dbea83ffd89679dac600e9f280bb2 15769 Author: Alex Deucher <alexdeucher (a] gmail.com> 15770 Date: Wed Apr 15 20:07:42 2009 -0400 15771 15772 R3xx/R4xx: fix up planar shader 15773 15774 We were overwriting the coord fetch address with the first 15775 tex fetch. Seemed to work however, luck I guess. Reorder 15776 the fetches to write to temp0 last. 15777 15778 commit 14c13faeb9f9b7717a25fcc1ca97d46cc6ee0031 15779 Author: Alex Deucher <alexdeucher (a] gmail.com> 15780 Date: Wed Apr 15 19:53:12 2009 -0400 15781 15782 R5xx: add shader-based csc 15783 15784 - native planar support 15785 - Xv attributes 15786 15787 commit 832efc7b90f5eb2da99512fcb902ab4838d2dcd1 15788 Author: Alex Deucher <alexdeucher (a] gmail.com> 15789 Date: Tue Apr 14 17:56:49 2009 -0400 15790 15791 R3xx/R4xx: Implement shader-based csc for packed formats 15792 15793 commit 32625118c27041265d25811c00d25ab7e82fb340 15794 Author: Alex Deucher <alexdeucher (a] gmail.com> 15795 Date: Tue Apr 14 16:38:40 2009 -0400 15796 15797 tex vid: fix attribute setup typo for XV_COLORSPACE 15798 15799 commit adf0912006b4f1597784dbfcc563d5c6d1c5667d 15800 Author: Alex Deucher <alexdeucher (a] gmail.com> 15801 Date: Tue Apr 14 16:16:33 2009 -0400 15802 15803 R6xx/R7xx: implement Xv attributes 15804 15805 - brightness, contrast, hue, etc. 15806 - TODO: implement gamma 15807 15808 commit 8810fe92b5aed08888584c6914482586b59f71ab 15809 Author: Alex Deucher <alexdeucher (a] gmail.com> 15810 Date: Tue Apr 14 11:26:35 2009 -0400 15811 15812 R200: clean code, always use shader based csc 15813 15814 - consolidate common r2xx csc shader code 15815 - always use shader based csc for both packed and planar 15816 formats 15817 - always use native planar csc on r1xx 15818 15819 commit 17685fefba68d188c7c0fe7a079180ec0722c046 15820 Author: Alex Deucher <alexdeucher (a] gmail.com> 15821 Date: Mon Apr 13 20:06:11 2009 -0400 15822 15823 Tex vid: general cleanup 15824 15825 - convert macros to more meaningful VTX_OUT_4 and VTX_OUT_6 15826 names to reflect that they actually do 15827 - fix indenting 15828 15829 commit 093ab4c9a33b0b396b78c061c3321dc044bdccdc 15830 Author: Alex Deucher <alexdeucher (a] gmail.com> 15831 Date: Mon Apr 13 19:48:35 2009 -0400 15832 15833 R1xx: add support for native planar textured Xv 15834 15835 commit ec0cb51df81c6c9a1de640d227fa9c9c33161083 15836 Author: Alex Deucher <alexdeucher (a] gmail.com> 15837 Date: Mon Apr 13 17:21:20 2009 -0400 15838 15839 R2xx tex vid: append verts for clip boxes 15840 15841 rather than sending a new draw packet for each rect 15842 15843 commit fde075a30a8ee2c333aa1bbe8fbd177258b085ba 15844 Author: Alex Deucher <alexdeucher (a] gmail.com> 15845 Date: Mon Apr 13 17:13:51 2009 -0400 15846 15847 R1xx tex vid: append verts for clip boxes 15848 15849 rather than sending a new draw packet for each rect 15850 15851 commit 12839fc17a2cca4ac14b9757bdaa63ba4679f96f 15852 Author: Alex Deucher <alexdeucher (a] gmail.com> 15853 Date: Mon Apr 13 17:04:31 2009 -0400 15854 15855 Tex vid: split by family 15856 15857 commit a30737b337edb31528174b483c9094941a5d41bb 15858 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 15859 Date: Mon Apr 13 15:36:07 2009 -0400 15860 15861 r200/r300: implement brightness/contrast/hue/saturation/gamma controls for textured video 15862 15863 This implements 15864 contrast/brightness/hue/saturation controls for r200/r300 plus gamma (same 15865 gamma value for all channels used though separate values would be trivial) 15866 control for r300. 15867 Some issues left: 15868 - only r200/r300 15869 - still can't be combined with bicubic 15870 - controls will silently cease to work if the format used is packed and not 15871 planar (except for rv250) 15872 - gamma range is from 100 to 10000 corresponding to 0.1 and 10.0 like used in 15873 overlay. However, usable range is far smaller. Over 2.0 picture gets dark 15874 pretty quickly, and below 0.6 or so black seems to turn into purple (I've 15875 verified that even with gamma 1.0 black actually often seems to be RGB 1/0/1 so 15876 this explains this since that gets amplified by low gamma values - not sure if 15877 this is a rounding problem somewhere, bogus reference values or is somehow 15878 expected). 15879 - gamma adds a bit too many instructions for my taste (7) though the 15880 alternative (3 texture lookups + some swizzling instructions) doesn't seem any 15881 better. 15882 15883 commit 9dd33cc8346073300312cd5f68de72d9bea1dd6b 15884 Author: Dave Airlie <airlied (a] redhat.com> 15885 Date: Fri Apr 17 10:30:21 2009 +1000 15886 15887 radeon: tweak ddc fix for all r300s. 15888 15889 Alex seems to think this might affect some other chips 15890 15891 commit a846ec14dd65bb12f314f83ffb45c4c4be564a5e 15892 Author: Dave Airlie <airlied (a] redhat.com> 15893 Date: Fri Apr 17 10:24:15 2009 +1000 15894 15895 radeon: fix DDC on rv410 VGA ports for non-atom use. 15896 15897 The rv410 hw i2c block appears to hold down whatever i2c lines 15898 it is connected to when in reset, so when doing DDC change hw i2c to 15899 point at different i2c lines. 15900 15901 commit efa1476fc0765b53298624198137dc5cc71d37a8 15902 Author: Alex Deucher <alexdeucher (a] gmail.com> 15903 Date: Thu Apr 16 16:43:02 2009 -0400 15904 15905 PM: fix reversed logic in pm block handler 15906 15907 commit 2b95de17781959457a809c8fecc6bbb08336c83f 15908 Author: Alex Deucher <alexdeucher (a] gmail.com> 15909 Date: Thu Apr 16 12:52:41 2009 -0400 15910 15911 RS400/RS480: attempt to fix ClockGating on RS4xx chips 15912 15913 commit c0215d2f2a27755fb04d20aaa4320a5e7da7ee04 15914 Author: Alex Deucher <alexdeucher (a] gmail.com> 15915 Date: Thu Apr 16 11:58:19 2009 -0400 15916 15917 RS690/RS740: fix connector enumeration in some cases 15918 15919 Should fix Masta-G's issue reported on IRC. 15920 15921 commit 573339114660f06e5f257648ba413bab0d6083ce 15922 Author: Alex Deucher <alexdeucher (a] gmail.com> 15923 Date: Thu Apr 16 11:19:01 2009 -0400 15924 15925 R3xx/R4xx EXA: set explicit tex cache regions 15926 15927 When using a mask, set explicit cache partitions for 15928 each texture. Gives 1% performance improvement in 15929 x11perf with anti-aliased text. 15930 15931 commit eea9800890b56bac9c07b7bd9c9e33fae2938af3 15932 Author: Alex Deucher <alexdeucher (a] gmail.com> 15933 Date: Thu Apr 16 10:41:28 2009 -0400 15934 15935 Revert "DCE3/3.2: cleanup dpms after previous changes" 15936 15937 This reverts commit f8c7d6a6162196a743f6885ecaf63ba50de1722a. 15938 15939 This is apparently still needed for some setups, however, I can't 15940 reproduce this locally anymore. 15941 15942 commit 099cb27d760af61b967239d960bc77f71e828655 15943 Author: Alex Deucher <alexdeucher (a] gmail.com> 15944 Date: Wed Apr 15 16:23:01 2009 -0400 15945 15946 Change ForceLowPowerMode to 4 pcie lanes 15947 15948 This fixes corruption for some users 15949 15950 commit 5abcea88d1f3307a7d08ae7757c4b535df1503bf 15951 Author: Alex Deucher <alexdeucher (a] gmail.com> 15952 Date: Wed Apr 15 15:52:04 2009 -0400 15953 15954 PM: don't touch clock gating regs unless ClockGating option is set 15955 15956 Fixes an M26 hang reported by tormod on IRC 15957 15958 commit bae88f81e2a01c0d547132b5b91e717ff9d0497d 15959 Author: Alex Deucher <alexdeucher (a] gmail.com> 15960 Date: Wed Apr 15 14:38:35 2009 -0400 15961 15962 man page: fix missing break 15963 15964 commit 4b3a3785f7b06205eeec6ca6b8913ecc2b94bd1f 15965 Author: Alex Deucher <alexdeucher (a] gmail.com> 15966 Date: Sat Apr 4 11:39:35 2009 -0400 15967 15968 Unify ForceLowPowerMode and DynamicPM Options 15969 15970 Force lower power mode and switch to a lower mode 15971 when idle. 15972 15973 commit 63cb57c5eb0d6bc083f54eb100cb972b0e7bca69 15974 Author: Alex Deucher <alexdeucher (a] gmail.com> 15975 Date: Wed Apr 1 16:22:02 2009 -0400 15976 15977 Add DynamicPM Option 15978 15979 Dyanmically switch between power states. Switch to a low 15980 power state when the system is idle (DPMS off). 15981 15982 commit a34a8b37afbea6ed4bf8ca42364195e174250c48 15983 Author: Alex Deucher <alexdeucher (a] gmail.com> 15984 Date: Wed Apr 1 15:23:17 2009 -0400 15985 15986 Set default low power PCIE lanes to 2 15987 15988 1 lane seems to cause occasional corruption when 15989 blitting to/from gart memory. 15990 15991 commit 1f0dc778dc25f4f85fedd73c55c847cab2c79fc5 15992 Author: Alex Deucher <alexdeucher (a] gmail.com> 15993 Date: Wed Apr 1 15:12:27 2009 -0400 15994 15995 Add support for setting the number PCIE lanes 15996 15997 commit 7e10b6222e8f44a3ecc6aaea55a7a7680d133bb3 15998 Author: Alex Deucher <alexdeucher (a] gmail.com> 15999 Date: Wed Apr 1 15:05:38 2009 -0400 16000 16001 Add ForceLowPowerMode Option 16002 16003 Force the chip to a low power mode at the expense 16004 of performance. 16005 16006 commit 533d01a5933cd491bbc09cd463ea62475abf4bf2 16007 Author: Alex Deucher <alexdeucher (a] gmail.com> 16008 Date: Tue Mar 31 17:24:29 2009 -0400 16009 16010 radeon: Add functions to set sclk/mclk using atombios 16011 16012 commit adb099409768e695b9928fa6aa5760f93dadd9af 16013 Author: Alex Deucher <alexdeucher (a] gmail.com> 16014 Date: Tue Mar 31 17:13:11 2009 -0400 16015 16016 radeon: Add functions to set sclk/mclk on r1xx-r4xx 16017 16018 commit 68e2a959ccc3d1a5d0731f1b55fdf1b2412635b2 16019 Author: Alex Deucher <alexdeucher (a] gmail.com> 16020 Date: Tue Mar 31 17:02:45 2009 -0400 16021 16022 Add unified ClockGating Option 16023 16024 - replaces DynamicClocks Option as the name was misleading 16025 - unified interface for atom and com based bioses 16026 - fix up clock gating code for newer r3xx asics 16027 16028 commit 336cee0e6d19068582b189b2c747d92cb2139d2f 16029 Author: Dave Airlie <airlied (a] redhat.com> 16030 Date: Mon Apr 13 18:11:48 2009 +1000 16031 16032 radeon: legacy bandwidth calcs for later r3xx/r4xx cards. 16033 16034 This is an attempt to rationalise the code using the register info 16035 files available to me here. 16036 16037 Further info is required: 16038 r350 check for stop_req > 15 then subtract 10: what other chips need it? 16039 get rs480 working instead of using magic. 16040 16041 commit 367ba7af76e7f63f3186a6e8d9bbe21aa31fc6d1 16042 Author: Alex Deucher <alexdeucher (a] gmail.com> 16043 Date: Fri Apr 10 09:37:15 2009 -0400 16044 16045 radeon: fix tmds_pll randr property 16046 16047 commit 15f1a2f31a7bc33f276a65283dbad0c106bee73e 16048 Author: Alex Deucher <alexdeucher (a] gmail.com> 16049 Date: Thu Apr 9 20:17:03 2009 -0400 16050 16051 radeon: remove old crtc size heuristics 16052 16053 IIRC, the old randr code used to use this to use for front buffer sizing, 16054 but it has since been changed. 16055 16056 commit ce9077b2f5a5e749c0078dfb79d26ef8711a823a 16057 Author: Alex Deucher <alexdeucher (a] gmail.com> 16058 Date: Thu Apr 9 19:59:19 2009 -0400 16059 16060 R6xx/R7xx: cleanup macro 16061 16062 commit 6532aeddacdeda3aa534264741c8648e1c449e76 16063 Author: Michel Dnzer <daenzer (a] vmware.com> 16064 Date: Thu Apr 9 09:41:56 2009 +0200 16065 16066 Clear the framebuffer before initializing modes. 16067 16068 Fixes garbage being visible shortly on server startup or when VT switching back 16069 to X. 16070 16071 commit 0c19053975925fed637aaa41d440e0cab8f6b596 16072 Author: Alex Deucher <alexdeucher (a] gmail.com> 16073 Date: Wed Apr 8 13:34:23 2009 -0400 16074 16075 radeon: chip family typo 16076 16077 commit 2888dd9ae9689b1cd72115dc0ceea1f5957299b0 16078 Author: Alex Deucher <alexdeucher (a] gmail.com> 16079 Date: Wed Apr 8 13:10:01 2009 -0400 16080 16081 R6xx/R7xx: set proper 3D client driver name for r600 16082 16083 commit 61a083116a2c89e145aa19da951243536e07542e 16084 Author: Alex Deucher <alexdeucher (a] gmail.com> 16085 Date: Wed Apr 8 10:44:30 2009 -0400 16086 16087 Bump version post release 16088 16089 commit 1b02b93895c31a0c9d641e47b46dce43b40edd97 16090 Author: Alex Deucher <alexdeucher (a] gmail.com> 16091 Date: Wed Apr 8 10:34:44 2009 -0400 16092 16093 Bump for release 16094 16095 commit da74b94a5da8f0bcf4ae99b20e594d2f0f4fc6f6 16096 Author: Alex Deucher <alexdeucher (a] gmail.com> 16097 Date: Tue Apr 7 18:59:06 2009 -0400 16098 16099 ATOM: code cleanup 16100 16101 - remove cases that aren't possible (e.g., no TMDSA on DCE3 cards) 16102 - condense duplicate cases 16103 16104 commit a24aa7250d09ae68ceaaba987314388b5e52096c 16105 Author: Alex Deucher <alexdeucher (a] gmail.com> 16106 Date: Tue Apr 7 18:52:08 2009 -0400 16107 16108 DCE3/3.2: further fixes 16109 16110 disable the dig transmitter and encoder before setup and enable 16111 16112 commit 284e1943f6726ed08e796f99e49fbd90ac7fa8a4 16113 Author: Alex Deucher <alexdeucher (a] gmail.com> 16114 Date: Wed Apr 1 15:07:01 2009 -0400 16115 16116 Cast info->sclk to int 16117 16118 commit b580e388e0e31843308c783c1ca859672dc6ca0e 16119 Author: Alex Deucher <alexdeucher (a] gmail.com> 16120 Date: Wed Apr 1 14:59:43 2009 -0400 16121 16122 Fix missing ')' in PCIE macro 16123 16124 commit 8ec27be3c8043a6ddc816b0b1480dd19f8d73832 16125 Author: Alex Deucher <alexdeucher (a] gmail.com> 16126 Date: Tue Apr 7 11:33:44 2009 -0400 16127 16128 radeon: fix Xv vsync for multi-head 16129 16130 commit f8c7d6a6162196a743f6885ecaf63ba50de1722a 16131 Author: Alex Deucher <alexdeucher (a] gmail.com> 16132 Date: Sat Apr 4 16:05:20 2009 -0400 16133 16134 DCE3/3.2: cleanup dpms after previous changes 16135 16136 commit a707d355c3c6ff92252c5a060a1fc32d97547552 16137 Author: Alex Deucher <alexdeucher (a] gmail.com> 16138 Date: Sat Apr 4 15:20:01 2009 -0400 16139 16140 DCE3/3.2: fix up transmitter/encoder setup 16141 16142 Should fix bug 21050 16143 16144 commit 215e12f9c0e8ac62c23af1add776ef88f9a0dc54 16145 Author: Eduard Fuchs <edfuchs (a] uni-kassel.de> 16146 Date: Thu Apr 2 12:36:59 2009 -0400 16147 16148 Enable byte swapping for r6xx/r7xx Hardware 16149 16150 Patch from Eduard Fuchs with some cleanup from me. 16151 16152 Tested at 32 bpp on MPC8641HPCN board (PowerPC) with 16153 HD2400 PCIe card 16154 16155 commit 0ded9fab6bc3fecc1976ae8369b788659e4d7793 16156 Author: Alex Deucher <alexdeucher (a] gmail.com> 16157 Date: Thu Apr 2 12:09:00 2009 -0400 16158 16159 Add RV790 (HD 4890) support 16160 16161 commit 0c6f20a9404c6a4210c5180bdca61c13fc186754 16162 Author: Alex Deucher <alexdeucher (a] gmail.com> 16163 Date: Tue Mar 31 15:51:11 2009 -0400 16164 16165 Add PCIE register accessors 16166 16167 commit 229d4e20e5adcdf9207944c19ca721477936ff02 16168 Author: Alex Deucher <alexdeucher (a] gmail.com> 16169 Date: Tue Mar 31 15:39:42 2009 -0400 16170 16171 Unify mc_idle() 16172 16173 commit f44b486848094b75479abef61c1aa075aede835c 16174 Author: Alex Deucher <alexdeucher (a] gmail.com> 16175 Date: Tue Mar 31 15:33:11 2009 -0400 16176 16177 COMBIS: Clarify case 0 for LCD DDC table 16178 16179 commit e6f3c48899a92174a55f6fe7327412bc88039d87 16180 Author: Alex Deucher <alexdeucher (a] gmail.com> 16181 Date: Tue Mar 31 15:30:40 2009 -0400 16182 16183 COMBIOS: fix default sclk/mclk from bios 16184 16185 commit 75bf5272936093bc024559e5744d97d8e79943b1 16186 Author: Alex Deucher <alexdeucher (a] gmail.com> 16187 Date: Tue Mar 31 15:17:37 2009 -0400 16188 16189 mac mini: 3rd time's the charm 16190 16191 fixes 10fbbac4bed59e12ad794ac873dd415d9ee4e340 16192 fixes bugs 20796 and 20979 16193 16194 commit 10fbbac4bed59e12ad794ac873dd415d9ee4e340 16195 Author: Alex Deucher <alexdeucher (a] gmail.com> 16196 Date: Tue Mar 31 15:14:27 2009 -0400 16197 16198 Mac mini: fix connector setup for real 16199 16200 fixes f061308e7abcf93d1612681b935387fee23e0a24 16201 fixes bugs 20796 and 20979 16202 16203 commit 8f2a03bfc86422669a46c2853050979f099a17a8 16204 Author: Alex Deucher <alexdeucher (a] gmail.com> 16205 Date: Mon Mar 30 11:30:34 2009 -0400 16206 16207 r1xx-r3xx TV: fix white noise 16208 16209 should fix bug 13872 16210 16211 commit f061308e7abcf93d1612681b935387fee23e0a24 16212 Author: Alex Deucher <alexdeucher (a] gmail.com> 16213 Date: Fri Mar 27 01:14:12 2009 -0400 16214 16215 mac mini: fix connector setup regression 16216 16217 fixes bug 20796 16218 16219 commit 47134ca77988f8c3b42d55ac4fcbb5dc182878d4 16220 Author: Alex Deucher <alexdeucher (a] gmail.com> 16221 Date: Thu Mar 26 17:18:47 2009 -0400 16222 16223 radeon: re-enable r/g/b intensity attributes for overlay 16224 16225 Noticed by sroland on IRC. 16226 16227 commit e2b5e48e2652daf03fcae852d8368f29ea4abcd9 16228 Author: Alex Deucher <alexdeucher (a] gmail.com> 16229 Date: Thu Mar 26 13:42:25 2009 -0400 16230 16231 R6xx/R7xx EXA: rework composite pixel shader 16232 16233 - move to vram storage 16234 - move swizzle logic to tex setup 16235 16236 commit 24dbd3f93c640f819a9f5029f5a49e41ab875bd4 16237 Author: Alex Deucher <alexdeucher (a] gmail.com> 16238 Date: Thu Mar 26 12:01:07 2009 -0400 16239 16240 R6xx/R7xx: clean up logic in EXA Comp PS setup 16241 16242 commit 076e4e3ac1c81086aecb4e70dd30fd5d26ec7bc8 16243 Author: Alex Deucher <alexdeucher (a] gmail.com> 16244 Date: Thu Mar 26 11:36:26 2009 -0400 16245 16246 R6xx/R7xx: clean up bool const code 16247 16248 3 regs: 1 bit per bool, 32 bools per ps/vs/gs 16249 16250 commit d2c3964fe04be42fe538f36439ed5ffca96e436a 16251 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 16252 Date: Wed Mar 18 01:55:12 2009 +0100 16253 16254 fix textured video allocation bug 16255 16256 size needs to be calculated after dstPitch adjustments, got already fixed 16257 for earlier than R600 chips by planar textured yuv patches, clean this up. 16258 16259 commit 18e56eb179fde28477487c63e6f9ebf7579e2cd5 16260 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 16261 Date: Thu Mar 5 02:07:46 2009 +0100 16262 16263 don't convert planar yuv to packed for r200 16264 16265 uses 3 textures for planar yuv and does yuv->rgb conversion in the shader. 16266 Similar to r300 code, but might have precision issues - hardware alu should 16267 have enough precision but hardware consts are only 8bit and we'd want 16268 at least 11. 16269 This also enables textured video on rv250 (and also supports packed yuv 16270 on that chip by using basically the same shader with packed data). 16271 16272 commit 58530bf4912800f9e09ebaea42a13cff8a80c19e 16273 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 16274 Date: Sat Feb 21 04:46:31 2009 +0100 16275 16276 don't convert planar yuv to packed for r300 16277 16278 uses 3 textures for planar yuv and does yuv->rgb conversion in the shader. 16279 small performance advantage, but manual texture cache setting is necessary 16280 otherwise it may be measurably slower (but probably not relevant) in some 16281 cases. 16282 Unlike some other drivers, using MADs instead of DP3s, since this requires 16283 less instructions due to no MOVs are required, the end result is the same 16284 though the constants need to be different. 16285 Use of this is user settable for now (XV_HWPLANAR attrib). 16286 16287 commit 97e19d96ba65a3df2fa3bbf73cfcc01b6dc3e796 16288 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 16289 Date: Tue Dec 30 22:23:39 2008 +0100 16290 16291 clip fixes 16292 16293 This fixes some oddities observed when the video is only partly visible. 16294 Instead of recalculating the geometry of the video, always use the same. 16295 Also fixes a assignment present twice, and another issue (bring in line with 16296 what the overlay code does). 16297 16298 commit 7aaf5f08e4147b75c4975e6d680d3f5a93424164 16299 Author: Alex Deucher <alexdeucher (a] gmail.com> 16300 Date: Mon Mar 23 12:55:02 2009 -0400 16301 16302 R4xx ATOM: fix crtc routing for CRT1 16303 16304 should fix bug 20814 16305 16306 commit 2fe6b393d9dbb193d4f0a3351ad76618cda07e2c 16307 Author: Alex Deucher <alexdeucher (a] gmail.com> 16308 Date: Sun Mar 22 01:31:30 2009 -0400 16309 16310 radeon: add new chip ids 16311 16312 commit 1ddee7cd6fd267b2fc86f21af27c5425eb0835a4 16313 Author: Alex Deucher <alexdeucher (a] gmail.com> 16314 Date: Fri Mar 20 13:44:00 2009 -0400 16315 16316 DCE3.2: use RMX for for non-native modes on DVI 16317 16318 DVI seems to have issues with low dotclocks, so 16319 use the scaler instead. 16320 16321 Fixes bug 20754 16322 16323 commit 44081e8fd2bf974272d5447b45dd7b206cf939c1 16324 Author: Alex Deucher <alexdeucher (a] gmail.com> 16325 Date: Thu Mar 19 20:48:16 2009 -0400 16326 16327 R6xx/R7xx EXA: fix maxPitchBytes 16328 16329 should now allow accel up the hw max of 8192x8192 16330 16331 commit e8e2367b66d9a4f5977d95f52b37cdbad75e409a 16332 Author: Alex Deucher <alexdeucher (a] gmail.com> 16333 Date: Thu Mar 19 20:09:32 2009 -0400 16334 16335 Add new pci ids 16336 16337 commit 68e83f10adb23fc359c1da881f5d666cd08b4067 16338 Author: Alex Deucher <alexdeucher (a] gmail.com> 16339 Date: Wed Mar 18 20:58:19 2009 -0400 16340 16341 bump version post release 16342 16343 commit a196aae19603c8651dc502f5cb0df4d55befe163 16344 Author: Alex Deucher <alexdeucher (a] gmail.com> 16345 Date: Wed Mar 18 20:52:12 2009 -0400 16346 16347 bump for release 16348 16349 commit 537da7018b0fd2ad263fc5cdc55b3b85cf8e9205 16350 Author: Alex Deucher <alexdeucher (a] gmail.com> 16351 Date: Wed Mar 18 19:22:55 2009 -0400 16352 16353 ATOM: fix up tv-out support 16354 16355 - fix purple hue when using ATOMTvOut option 16356 - fix TV load detection on newer r6xx/r7xx chips 16357 - disable CV for now 16358 16359 commit c0e2513ab128ddd5be0ed626d9e31777a98983ef 16360 Author: Alex Deucher <alexdeucher (a] gmail.com> 16361 Date: Wed Mar 18 14:08:30 2009 -0400 16362 16363 R6xx/R7xx AGP: disable gart data transfers 16364 16365 UTS/DFS/Xv 16366 16367 Doesn't seem to be reliable on AGP. 16368 16369 fixes bugs: 20436, 20236, several reports on ML and IRC 16370 16371 commit e3e1f0c7b892b094b292081021a0e97932b8e079 16372 Author: Fabio <fabio.ped (a] libero.it> 16373 Date: Wed Mar 18 12:13:53 2009 -0400 16374 16375 man page update 16376 16377 commit 6ae54d3ea023968887bf0cb0ec1be2115621d554 16378 Author: Alex Deucher <alexdeucher (a] gmail.com> 16379 Date: Wed Mar 18 11:50:22 2009 -0400 16380 16381 AVIVO: Fix dac load detection 16382 16383 commit 04e36b794c216e4fa77c780fd989df1ef1c7f0d9 16384 Author: Fabio <fabio.ped (a] libero.it> 16385 Date: Tue Mar 17 11:29:26 2009 -0400 16386 16387 man page updates 16388 16389 from bug 20648 16390 16391 commit a6855c370194b6df307ea33724fe17a85d67607e 16392 Author: Alex Deucher <alexdeucher (a] gmail.com> 16393 Date: Mon Mar 16 13:52:42 2009 -0400 16394 16395 radeon: add another AGP quirk 16396 16397 bug 20647 16398 16399 commit a55ced5ee20c07e743c7c0978803fd10589c1531 16400 Author: Alex Deucher <alexdeucher (a] gmail.com> 16401 Date: Sun Mar 15 12:08:04 2009 -0400 16402 16403 Fix fallout from r6xx/r7xx EXA merge in IB handling 16404 16405 Noticed Christiaan van Dijk. 16406 Should fix bug 20510 16407 16408 commit a065eeb9b301ab452cc8fbd3cac1fd9a3841c4e1 16409 Author: Alex Deucher <alexdeucher (a] gmail.com> 16410 Date: Sat Mar 14 18:40:28 2009 -0400 16411 16412 atom: fix up bios scartch register usage 16413 16414 the bios scratch registers keep 3 types of state for each output: 16415 1. dpms 16416 2. active 16417 3. connected 16418 16419 some of the command tables use the active flags. We used to set the active 16420 flags based on dpms state, but instead set them based on whether they are 16421 connected or not. 16422 16423 commit 5c4ef924d082fd3b279f62fda7a97f563b009d38 16424 Author: Alex Deucher <alexdeucher (a] gmail.com> 16425 Date: Sat Mar 14 18:16:49 2009 -0400 16426 16427 radeon: fix typo in bios scratch reg setup 16428 16429 commit 96fce5a56269509a935e8ee621a5e17cc9005ddc 16430 Author: Alex Deucher <alexdeucher (a] gmail.com> 16431 Date: Fri Mar 13 19:29:58 2009 -0400 16432 16433 Bump post-release 16434 16435 commit 8da1b82a70de7b3ef38905bd77c58a20fd44ddd6 16436 Author: Alex Deucher <alexdeucher (a] gmail.com> 16437 Date: Fri Mar 13 19:23:19 2009 -0400 16438 16439 bump for release 16440 16441 commit 28d9d699dcdff394aa561505f7f54f08db800828 16442 Author: Alex Deucher <alexdeucher (a] gmail.com> 16443 Date: Fri Mar 13 18:18:31 2009 -0400 16444 16445 radeon: man page updates 16446 16447 commit 06963796d00b5cb6704412a7d4417fcefae4aa4d 16448 Author: Adam Jackson <ajax (a] redhat.com> 16449 Date: Fri Mar 13 18:14:43 2009 -0400 16450 16451 uniphy: start adding DisplayPort setup 16452 16453 commit 1db81afe4f8db55e02a65cd22ad84c6bbb27e8ef 16454 Author: Alex Deucher <alexdeucher (a] gmail.com> 16455 Date: Fri Mar 13 17:51:56 2009 -0400 16456 16457 radeon: add support for 30 bit LUTs 16458 16459 All radeons have them. Thanks to Yang Zhao for figuring 16460 this out. 16461 16462 commit 5bb326c64ad3bf5c93943280dcbcb25c8910d1aa 16463 Author: Alex Deucher <alexdeucher (a] gmail.com> 16464 Date: Fri Mar 13 14:10:58 2009 -0400 16465 16466 R6xx/R7xx: EXA VSync Option not supported yet 16467 16468 commit 8c2d40894448329e721360811c124998a17ba2ba 16469 Author: Alex Deucher <alexdeucher (a] gmail.com> 16470 Date: Fri Mar 13 14:03:43 2009 -0400 16471 16472 R6xx/R7xx: fix up vline stuff along the lines of previous chips 16473 16474 Avoids an additional function call. 16475 16476 commit 176c1b43fc3ee48e87c134446754360b3683c24b 16477 Author: Alex Deucher <alexdeucher (a] gmail.com> 16478 Date: Fri Mar 13 13:28:42 2009 -0400 16479 16480 R6xx/R7xx: return in RADEONWaitForIdleMMIO() if accel is off 16481 16482 Don't wait for engine idle if we haven't initialized the engine. 16483 Behavior should be consistent with versions of the driver prior 16484 to accel support being added. 16485 16486 Should fix bug 20645 16487 16488 commit d538a4508248adf2debaf0cc870a2d1c48abd986 16489 Author: Alex Deucher <alexdeucher (a] gmail.com> 16490 Date: Fri Mar 13 13:22:03 2009 -0400 16491 16492 R6xx/r7xx: clarify accel messages 16493 16494 We only support EXA and and only with DRI. 16495 16496 commit 1b7090a4c0d6a38701fbfb928fac8e15d40fb81f 16497 Author: Adam Jackson <ajax (a] redhat.com> 16498 Date: Fri Mar 13 17:41:59 2009 -0400 16499 16500 Add some more cases to the DVI single-link checks. 16501 16502 commit 104e13da74de37477e3a31a5de96c5ebb8fe888b 16503 Author: Cooper Yuan <cooperyuan (a] gmail.com> 16504 Date: Fri Mar 13 11:47:43 2009 -0400 16505 16506 radeon: save bios scratch registers in Preinit() 16507 16508 The display detect routines can change these which causes 16509 havok with some bioses. 16510 16511 commit f70d7a0fd9ac3659d93511ab362eb80a7ff49777 16512 Author: Dave Airlie <airlied (a] redhat.com> 16513 Date: Fri Mar 13 12:56:29 2009 +1000 16514 16515 r600: reload shaders into VRAM on resume 16516 16517 As VRAM gets zeroed out over s/r, we need to reload the 16518 shaders. 16519 16520 commit 1fe0dcfe98dfe30499ba53c1f208c4ef10bef001 16521 Author: Yang Zhao <yang (a] yangman.ca> 16522 Date: Thu Mar 12 18:50:32 2009 -0400 16523 16524 R6xx/R7xx shader: Fix OFFSET_[XYZ] macro for TEX_DWORD2 to accept floats 16525 16526 Values for OFFSET_[XYZ] are 5-bits two's-complement fixed-point with 16527 one-bit after decimal point. Values in [-8.0, 7.5] are valid. Inputs 16528 that do not exactly land on 0.5 increments are rounded towards 0 to the 16529 nearest increment. 16530 16531 commit f356ca7795531e0c080775efb598141eb7db44ba 16532 Author: Alex Deucher <alexdeucher (a] gmail.com> 16533 Date: Thu Mar 12 03:42:40 2009 -0400 16534 16535 RS600: fix page table size for rs600 as well 16536 16537 commit ef69336584c45e35aabb797a06c4030e75d66507 16538 Author: Dave Airlie <airlied (a] redhat.com> 16539 Date: Thu Mar 12 14:54:18 2009 +1000 16540 16541 r600: fix sizing of PCI GART table for r600 16542 16543 commit 945ccbbd4fa2b65ccdfb23716c178c95b036734d 16544 Author: Alex Deucher <alexdeucher (a] gmail.com> 16545 Date: Tue Mar 10 09:53:14 2009 -0400 16546 16547 radeon: clean more thoroughly in RADEONFreeRec() 16548 16549 commit 03ec2099fdc859b23e5fa56ac8696ec045f1b110 16550 Author: (Yu-yeon Oh) <yuyeon_oh (a] tmax.co.kr> 16551 Date: Tue Mar 10 09:51:08 2009 -0400 16552 16553 radeon_driver.c small memory bug 16554 16555 commit 60dac878cc8626b0821c4c7a6a6dfa12a5597cfa 16556 Author: Alex Deucher <alexdeucher (a] gmail.com> 16557 Date: Mon Mar 9 20:48:39 2009 -0400 16558 16559 R4xx: add R4xxATOM option 16560 16561 If set, enables the use of atombios for modesetting 16562 on r4xx cards. 16563 16564 commit 758fcc33c232c32e7b66261fca51de161d1a4800 16565 Author: Dave Airlie <airlied (a] redhat.com> 16566 Date: Tue Mar 10 10:34:13 2009 +1000 16567 16568 r600: enable DRI by default 16569 16570 fallback to shadowfb if DRI fails pre init 16571 16572 commit 7735e411954fd2cb6105c30ebd5a1b209974ce22 16573 Author: Alex Deucher <alexdeucher (a] gmail.com> 16574 Date: Mon Mar 9 17:49:25 2009 -0400 16575 16576 R6xx/r7xx: clarify EXA message 16577 16578 commit 1d355798f3d421cabf2a3a0ba32a580483476ab0 16579 Author: Alex Deucher <alexdeucher (a] gmail.com> 16580 Date: Mon Mar 9 15:13:01 2009 -0400 16581 16582 AVIVO: fix panning 16583 16584 low 2 bits of X need to be 0. 16585 16586 commit d1add18f3d238c755f9875b5e5c211e4af742482 16587 Author: root <root (a] benzylpiperazine.boston.devel.redhat.com> 16588 Date: Fri Mar 6 17:53:59 2009 -0500 16589 16590 output: Filter out dual-link modes from DP->DVI connections 16591 16592 There's not enough pins on a DP->DVI connector for this to possibly work. 16593 16594 commit 04481bf450650d1ea5aadabc9213b899b0fe8119 16595 Author: root <root (a] benzylpiperazine.boston.devel.redhat.com> 16596 Date: Fri Mar 6 17:32:15 2009 -0500 16597 16598 atom: Enable DisplayPort source to DVI/HDMI sink 16599 16600 DP source to DP sink still doesn't work yet, of course. 16601 16602 commit 0ba73e040b94590867f8b1071a26da2526a3c375 16603 Author: Adam Jackson <ajax (a] redhat.com> 16604 Date: Fri Mar 6 15:38:00 2009 -0500 16605 16606 output: Allow for multiple DisplayPort outputs. 16607 16608 commit 8f7c17a6b77ca368f067c2acb803cc8f9ef19351 16609 Author: Tormod Volden <debian.tormod (a] gmail.com> 16610 Date: Fri Mar 6 09:42:39 2009 -0500 16611 16612 RV350: Add AGPMode quirk for Thinkpad 16613 16614 fixes bug 15096 (and Ubuntu bug #195051) 16615 16616 commit 5dc4b69fbdd0e8ca48cc00321f193260a8f10bd0 16617 Author: Alex Deucher <alexdeucher (a] gmail.com> 16618 Date: Thu Mar 5 12:26:04 2009 -0500 16619 16620 radeon: add a few more default common modes for lvds 16621 16622 commit 3b0fc22ad9e5c0f120a74c4f3d8e48c270f0ff29 16623 Author: Alex Deucher <alexdeucher (a] gmail.com> 16624 Date: Wed Mar 4 19:04:34 2009 -0500 16625 16626 R6xx/R7xx: wire up DMAForXv option like older asics 16627 16628 Select between SW and HW-assisted uploads 16629 16630 commit 52f06ace04ad8141effc45fb6a0107a05bb46a73 16631 Author: Alex Deucher <alexdeucher (a] gmail.com> 16632 Date: Wed Mar 4 18:51:44 2009 -0500 16633 16634 R6xx/R7xx: disable XV_BICUBIC attribute 16635 16636 It's not implemented yet. 16637 16638 commit b1fd883b59b85fed8782e035890098908902f4ce 16639 Author: Alex Deucher <alexdeucher (a] gmail.com> 16640 Date: Wed Mar 4 11:17:16 2009 -0500 16641 16642 RN50: fix up cloning on servers 16643 16644 Since the rn50 only has one crtc, but often has multiple outputs, 16645 occasionally one of the outputs doesn't get a crtc assigned eventhough 16646 the outputs may have common modes. If we see an output with modes, but 16647 no crtc, assign it so all the attached outputs light up. 16648 16649 fixes bugs 19457 and similar ones 16650 16651 commit 592bcac52f113a95923a8f1cb8427e7552d5670b 16652 Author: Alex Deucher <alexdeucher (a] gmail.com> 16653 Date: Wed Mar 4 10:53:20 2009 -0500 16654 16655 ATOM: don't use fixed ref div for LVDS 16656 16657 The bios doesn't provide one. If there are problems 16658 with certain panels, we may need to adjust the PLL flags 16659 in certain cases. 16660 16661 Should fix bug 20458 16662 16663 commit e88b0eac8cab8b54b769fd7b4ad3b3b65a90de09 16664 Author: Alex Deucher <alexdeucher (a] gmail.com> 16665 Date: Wed Mar 4 02:52:36 2009 -0500 16666 16667 radeon: just add some common modes for LVDS 16668 16669 avoids needing to hack around with the edid 16670 16671 commit 2bb319fd80d20a3d4e5cc9416891bc7e4c3e431e 16672 Author: Alex Deucher <alexdeucher (a] gmail.com> 16673 Date: Wed Mar 4 02:50:45 2009 -0500 16674 16675 Revert "radeon: adjust LVDS so that default modes get added" 16676 16677 This reverts commit 1a2b16561d19ec9c027c562902f5fc086c856994. 16678 16679 commit 1a2b16561d19ec9c027c562902f5fc086c856994 16680 Author: Alex Deucher <alexdeucher (a] gmail.com> 16681 Date: Tue Mar 3 20:21:17 2009 -0500 16682 16683 radeon: adjust LVDS so that default modes get added 16684 16685 we can scale, so add the default modes. for panels 16686 with an EDID, set the continous freq bit. for panels without 16687 and EDID, add a FAKE edid with the continous freq bit set. 16688 16689 Based on similar code in the Intel driver. 16690 16691 commit 71117970df36cbe689ef15e9a6cca24439b4cd62 16692 Author: Alex Deucher <alexdeucher (a] gmail.com> 16693 Date: Tue Mar 3 19:40:30 2009 -0500 16694 16695 AVIVO: add aspect scaling mode 16696 16697 No luck yet for aspect on pre-avivo chips 16698 16699 commit d586a2c6f821c821a4a7708a3382acb63187534f 16700 Author: Alex Deucher <alexdeucher (a] gmail.com> 16701 Date: Tue Mar 3 12:17:30 2009 -0500 16702 16703 Rotation: don't rotate if acceleration is not active 16704 16705 commit fdc64948119e2218921b4dc0a492522442668875 16706 Author: Alex Deucher <alexdeucher (a] gmail.com> 16707 Date: Mon Mar 2 16:33:46 2009 -0500 16708 16709 R6xx/r7xx: remove some unneeded code I missed in the last commit 16710 16711 commit 4ad1c4decfee653dbbc1ea2ca4270487be622382 16712 Author: Alex Deucher <alexdeucher (a] gmail.com> 16713 Date: Mon Mar 2 15:38:35 2009 -0500 16714 16715 R6xx/R7xx EXA: cleanup composite texture setup 16716 16717 commit b7164ac4ad55e5d0fc474df8ae762b469b91ba30 16718 Author: Alex Deucher <alexdeucher (a] gmail.com> 16719 Date: Mon Mar 2 04:08:09 2009 -0500 16720 16721 R6xx/R7xx EXA: combine composite mask/non-mask VS 16722 16723 Also fix set_bool_const() 16724 the CF bool consts are not contiguous by shader type 16725 There are 96 boolean constants (32 each for PS, VS, GS) and 16726 they are ordered as follows: 16727 ps, vs, gs ... ps, vs, gs 16728 16729 commit fa98f424de739be2c6005b740a74bbf1ee968a8b 16730 Author: Alex Deucher <alexdeucher (a] gmail.com> 16731 Date: Mon Mar 2 02:28:57 2009 -0500 16732 16733 R6xx/R7xx Xv: combine packed and planar shaders 16734 16735 use a bool const to select the tex fetch routine 16736 16737 commit ccde35c3eda3fff0de29eb8c6fdc392629724a34 16738 Author: Christian Koenig <deathsimple (a] vodafone.de> 16739 Date: Sun Mar 1 23:38:37 2009 -0500 16740 16741 R6xx/R7xx: move shaders to r600_shader.c and fixup Xv PS 16742 16743 patches from Christian Koenig with some adjustments from me 16744 16745 commit a8e631c1b1c9b46602aeca66f8e7e68154d0bfc8 16746 Author: Alex Deucher <alexdeucher (a] gmail.com> 16747 Date: Sun Mar 1 22:30:31 2009 -0500 16748 16749 R6xx/R7xx: code cleanups 16750 16751 commit 528061d51bb4248d6fabec9579dead32a730467a 16752 Author: Alex Deucher <alexdeucher (a] gmail.com> 16753 Date: Sun Mar 1 18:28:13 2009 -0500 16754 16755 R6xx/R7xx: write vertexes directly to the IB 16756 16757 Reduces the vertex buffer setup overhead 16758 16759 commit 96a0bafa16dbb67a3d10fa6fe45be13930495ddc 16760 Author: Alex Deucher <alexdeucher (a] gmail.com> 16761 Date: Sun Mar 1 17:58:36 2009 -0500 16762 16763 R6xx/R7xx: switch emit functions to macros 16764 16765 This improves performance due to decreased function call 16766 overhead. 16767 16768 commit cbd579c41a16bddcbb8793eac2a41f2e958bef4d 16769 Author: Alex Deucher <alexdeucher (a] gmail.com> 16770 Date: Sun Mar 1 13:16:20 2009 -0500 16771 16772 Print a message when we have a shared DDC line 16773 16774 commit a7f465f73363fce409870f62173d518b1bc02ae6 16775 Author: Alex Deucher <alexdeucher (a] gmail.com> 16776 Date: Sun Mar 1 12:39:31 2009 -0500 16777 16778 R300: Add AGP quirk 16779 16780 bug 7770 16781 16782 commit 017bc7f34de2e40919f0e711205b78ee4e3643f6 16783 Author: Tormod Volden <debian.tormod (a] gmail.com> 16784 Date: Sat Feb 28 10:33:35 2009 -0500 16785 16786 M9+: Add AGP quirk for Sony Vaio 16787 16788 commit 49b57767d0d2c041517b0764c2ed2d2ba5a7092c 16789 Author: Bryce Harrington <bryce (a] bryceharrington.org> 16790 Date: Fri Feb 27 20:31:03 2009 -0500 16791 16792 Quirk for RV280 on 82865G/PE/P DRAM Controller/Host-Hub 16793 16794 Solves issue where display suffers from line corruption. 16795 (See https://edge.launchpad.net/ubuntu/+bug/300304) 16796 16797 commit 150abd4123664eab65ea43585dbdb9b3151c46af 16798 Author: Tormod Volden <lists.tormod (a] gmail.com> 16799 Date: Fri Feb 27 11:23:32 2009 -0500 16800 16801 Janitor: cosmetic clean-up of AGPMode quirk table 16802 16803 commit c7d344f0ad255761d5ae30da59cf0a8b532dfc68 16804 Author: Michel Dnzer <daenzer (a] vmware.com> 16805 Date: Fri Feb 27 17:05:01 2009 +0100 16806 16807 EXA: Make sure Prepare/FinishAccess hooks can handle EXA_PREPARE_AUX* indices. 16808 16809 And signal this to the EXA core using the new EXA_SUPPORTS_PREPARE_AUX flag. 16810 16811 commit f4d57105fc869e9318f447a1514f1b284a0a4ec0 16812 Author: Michel Dnzer <daenzer (a] vmware.com> 16813 Date: Fri Feb 27 09:23:24 2009 +0100 16814 16815 Revert "EXA: Adapt to EXA changes in xserver Git." 16816 16817 This reverts commit de358736dc696559ba99c71cf5b2a97508201630. 16818 16819 commit 3285a4af17b19f2d1c24eebfb6aa51cddcf3ca50 16820 Author: Alex Deucher <alexdeucher (a] gmail.com> 16821 Date: Thu Feb 26 17:31:13 2009 -0500 16822 16823 DCE3.2: fix up Save()/Restore() 16824 16825 commit 4fc978f3d83ef6370c9715e7ca5350640ccbfe10 16826 Author: Alex Deucher <alexdeucher (a] gmail.com> 16827 Date: Thu Feb 26 12:10:08 2009 -0500 16828 16829 R6xx/R7xx: use shadowfb if DRI is disabled 16830 16831 commit 000756e052a291230e5c95e48b69a5aa9c4fab0e 16832 Merge: 22d7746bf 8373f4399 16833 Author: Alex Deucher <alexdeucher (a] gmail.com> 16834 Date: Thu Feb 26 11:44:13 2009 -0500 16835 16836 Merge branch 'r6xx-r7xx-support' of git+ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati 16837 16838 commit 22d7746bfc1bcbb3269afac88971254088ea9ece 16839 Author: Alex Deucher <alexdeucher (a] gmail.com> 16840 Date: Thu Feb 26 11:00:46 2009 -0500 16841 16842 RS600: enable the DRI by default 16843 16844 commit 8373f4399b03961f2c928a9275d47e9f41bd92bb 16845 Author: Alex Deucher <alexdeucher (a] gmail.com> 16846 Date: Wed Feb 25 09:16:38 2009 -0500 16847 16848 R6xx/R7xx EXA: same surface and same coords equals nop 16849 16850 should fix bug 20305 16851 16852 commit c74727015453ff3c3d6d06b812ebca9eb19a9767 16853 Author: Alex Deucher <alexdeucher (a] gmail.com> 16854 Date: Tue Feb 24 11:08:46 2009 -0500 16855 16856 R6xx/R7xx EXA: init copy_area to NULL 16857 16858 commit 95ce13572dc2d9f5dd6cf55c23411e275c0aadf1 16859 Author: Alex Deucher <alexdeucher (a] gmail.com> 16860 Date: Tue Feb 24 10:51:13 2009 -0500 16861 16862 R6xx/R7xx EXA: Optimize temp surface for overlapping copies 16863 16864 - allocate temp surface in PrepareCopy() 16865 - fall back to old OverlapCopy() path if we are not able 16866 to allocate a temp surface 16867 16868 commit 1a7db3fc2a0277d724d60d028064d8ef75019c28 16869 Author: Mark van Doesburg <mark.vandoesburg (a] hetnet.nl> 16870 Date: Tue Feb 24 10:44:19 2009 -0500 16871 16872 R6xx/R7xx EXA: use a temp surface for overlapping copy 16873 16874 commit 957c2ba3459af152ad16a3f6dd4631044f9df6f9 16875 Author: Alex Deucher <alexdeucher (a] gmail.com> 16876 Date: Tue Feb 24 09:43:58 2009 -0500 16877 16878 RS600: fix MC addr mask 16879 16880 Noticed by osiris on IRC 16881 16882 commit de358736dc696559ba99c71cf5b2a97508201630 16883 Author: Michel Dnzer <daenzer (a] vmware.com> 16884 Date: Tue Feb 24 08:30:23 2009 +0100 16885 16886 EXA: Adapt to EXA changes in xserver Git. 16887 16888 The PrepareAccess hook can now get six different indices. 16889 16890 Also remove superfluous instances of #include "exa.h". 16891 16892 Note that you may need to run autoheader manually to get config.h.in updated. 16893 16894 commit 752b0218af36a45b333a12f66dfdce8c1e662aa8 16895 Author: Michel Dnzer <daenzer (a] vmware.com> 16896 Date: Tue Feb 24 08:13:06 2009 +0100 16897 16898 EXA: Pass pScrn and info into RadeonCompositeTile. 16899 16900 Saves some cycles looking them up. 16901 16902 commit e2e508bc08d993fc108297e9cd2fa0ce183282bc 16903 Author: Michel Dnzer <daenzer (a] vmware.com> 16904 Date: Tue Feb 24 08:12:42 2009 +0100 16905 16906 Only call RADEONWaitForVLine if it might actually do anything useful. 16907 16908 Otherwise it and RADEONBiggerCrtcArea show up pretty high in x11perf -aa*text 16909 profiles. 16910 16911 commit 5ab8699c97c050424cf813b9a60550aba32446a2 16912 Author: Michel Dnzer <daenzer (a] vmware.com> 16913 Date: Tue Feb 24 08:12:23 2009 +0100 16914 16915 Fix compile warning when building without EXA. 16916 16917 commit 132e61741185f4a1114e66a34bdbf21e08e4d5d8 16918 Author: Alex Deucher <alexdeucher (a] gmail.com> 16919 Date: Mon Feb 23 12:13:41 2009 -0500 16920 16921 radeon: re-enable load detection output attribute for TV/CV 16922 16923 commit 918d511f6f26d2fdeb63d6e32cd348707b4a2140 16924 Author: Alex Deucher <alexdeucher (a] gmail.com> 16925 Date: Mon Feb 23 10:22:19 2009 -0500 16926 16927 RS600: fix up MC setup 16928 16929 Please test if you have an RS600 16930 16931 commit 27f8ca2cce65be2bcb3375231886d5444d251808 16932 Author: Alex Deucher <alexdeucher (a] gmail.com> 16933 Date: Thu Feb 19 21:18:07 2009 -0500 16934 16935 R6xx/R7xx: add wait for idle MMIO path 16936 16937 commit 1f783117f79f95f1744681ccc7d80dc0e7b53057 16938 Author: Alex Deucher <alexdeucher (a] gmail.com> 16939 Date: Thu Feb 19 21:07:06 2009 -0500 16940 16941 RBBM_GUICNTL is pre-r6xx only 16942 16943 commit f3933ed200dbf61b05a114f7697f2bb68d92abe7 16944 Author: Alex Deucher <alexdeucher (a] gmail.com> 16945 Date: Thu Feb 19 20:12:33 2009 -0500 16946 16947 Don't write new HDP location until we've written the new FB location 16948 16949 commit 7f6801abb4bbd79830999d99ccdbd672a9984df6 16950 Author: Alex Deucher <alexdeucher (a] gmail.com> 16951 Date: Thu Feb 19 19:02:52 2009 -0500 16952 16953 radeon: one more 32 -> 64 just to be safe 16954 16955 missed this in cadae2a8d9979b2d61eae7d1f7a9ab8f5eb28e83 16956 16957 commit ca22390c8a5256bbc8eeff775d5eebf0e892d890 16958 Author: Alex Deucher <alexdeucher (a] gmail.com> 16959 Date: Thu Feb 19 18:58:40 2009 -0500 16960 16961 R6xx/R7xx: fix up a few more paths 16962 16963 commit cadae2a8d9979b2d61eae7d1f7a9ab8f5eb28e83 16964 Author: Alex Deucher <alexdeucher (a] gmail.com> 16965 Date: Thu Feb 19 18:50:49 2009 -0500 16966 16967 Fix MC setup on systems with more than 512 MB of VRAM 16968 16969 many thanks to app4des on IRC for helping me sort this out. 16970 16971 commit 42492c70c481c88911337eefb97ba9d030adf7b6 16972 Author: Alex Deucher <alexdeucher (a] gmail.com> 16973 Date: Thu Feb 19 10:20:28 2009 -0500 16974 16975 R6xx/R7xx: wait for MC idle when changing the MC 16976 16977 should fix bug 19984 16978 16979 commit e6475282486f4895bc68f6b093ecbb1aa6d25f72 16980 Author: Alex Deucher <alexdeucher (a] gmail.com> 16981 Date: Wed Feb 18 20:21:13 2009 -0500 16982 16983 R6xx/R7xx Xv: fix some missing bits from last commit 16984 16985 commit adff8906c9899dde7711382577a63f4a726437ca 16986 Author: Alex Deucher <alexdeucher (a] gmail.com> 16987 Date: Wed Feb 18 20:01:57 2009 -0500 16988 16989 R6xx/R7xx EXA/Xv: properly deal with running out of vertex buffer space 16990 16991 As noted by mhopf, if VGT_MAX/MIN_INDX, etc. regs change, you need to re-emit 16992 CB blocks to avoid a hang. So, just set the VGT_MAX_INDX to a reasonably large value 16993 in the default state and don't touch them when drawing. When we run out of VB space, 16994 just draw the current buffer, grab a new one, and continue. 16995 16996 commit cf85d4a1d43a3209c7ca9307aede2c2c243f7130 16997 Author: Alex Deucher <alexdeucher (a] gmail.com> 16998 Date: Wed Feb 18 19:02:56 2009 -0500 16999 17000 R6xx/R7xx: reset 3D state after VT switch 17001 17002 commit 6545f0a2967414fa518a78440060b8b69c0146ee 17003 Author: Alex Deucher <alexdeucher (a] gmail.com> 17004 Date: Wed Feb 18 18:15:18 2009 -0500 17005 17006 R6xx/R7xx: fixup accel paths 17007 17008 commit 61311b1de019c5a2344928a47347a966c229b08a 17009 Author: Alex Deucher <alexdeucher (a] gmail.com> 17010 Date: Wed Feb 18 17:27:05 2009 -0500 17011 17012 Bump version post release 17013 17014 commit 0d3d1f1f7b450dabd3ad7d2df26ad1c72709b29b 17015 Author: Alex Deucher <alexdeucher (a] gmail.com> 17016 Date: Wed Feb 18 17:24:57 2009 -0500 17017 17018 bump for release 17019 17020 commit 76a32f8391317513538dafee49cbb61c0d756356 17021 Author: Alex Deucher <alexdeucher (a] gmail.com> 17022 Date: Wed Feb 18 16:31:44 2009 -0500 17023 17024 R6xx: Connector quirk for asus board 17025 17026 bug 19943 17027 17028 commit 97b8482dba4e99088b59d3ab1396be98a70e823e 17029 Author: Alex Deucher <alexdeucher (a] gmail.com> 17030 Date: Wed Feb 18 13:27:15 2009 -0500 17031 17032 Set default RMX type to FULL on LVDS 17033 17034 commit 7d22b1799b34010d34c3600d6cb02ffe839a0780 17035 Author: Alex Deucher <alexdeucher (a] gmail.com> 17036 Date: Wed Feb 18 12:55:40 2009 -0500 17037 17038 DCE3.2+: allow output cloning 17039 17040 commit 1a237a40958c006c56b80850bd77b2ac6c17e030 17041 Author: Alex Deucher <alexdeucher (a] gmail.com> 17042 Date: Wed Feb 18 11:46:20 2009 -0500 17043 17044 ATOM: reset crtc initialized flag on CloseScreen() 17045 17046 additional fix needed for bug 16781 17047 17048 commit 9a108f0a0b7203458673ce6221e747a166d39617 17049 Author: Alex Deucher <alexdeucher (a] gmail.com> 17050 Date: Tue Feb 17 19:52:27 2009 -0500 17051 17052 ATOM: don't unblank uninitialized crtcs 17053 17054 If the crtc timing isn't setup, you might get stuck in a loop 17055 in the BlankCRTC command table 17056 17057 fixes bug 16781 17058 17059 commit 3dff20e276615e8b77177689a4a5f8d91b3e8eac 17060 Author: Alex Deucher <alexdeucher (a] gmail.com> 17061 Date: Tue Feb 17 13:14:38 2009 -0500 17062 17063 R6xx/R7xx EXA: switch to surface sync packet 17064 17065 commit 31bd6d28dd35021e0e706d4f094989deb856c26a 17066 Author: Alex Deucher <alexdeucher (a] gmail.com> 17067 Date: Tue Feb 17 11:24:02 2009 -0500 17068 17069 Fix crtc routing on pre-DCE3.2 systems 17070 17071 This should fix peterz and bug 20074 17072 17073 commit 3c9e00c7f2ed494976713e8f77b729e0d76409db 17074 Author: Alex Deucher <alexdeucher (a] gmail.com> 17075 Date: Mon Feb 16 16:18:42 2009 -0500 17076 17077 RV350: add another AGP quirk 17078 17079 commit 036a17bec5dabbb91640b907811da060dac154d6 17080 Author: Alex Deucher <alexdeucher (a] gmail.com> 17081 Date: Mon Feb 16 13:10:45 2009 -0500 17082 17083 RV280: add another agp quirk 17084 17085 commit 00ac9d2f8d58725d0d3e6b1bf7f728d9ae7e6109 17086 Author: David Miller <davem (a] davemloft.net> 17087 Date: Sun Feb 15 18:26:14 2009 +1000 17088 17089 GART: Save/restore GART table consistently. 17090 17091 Always save the GART table with the RADEON_SURFACE_CNTL register 17092 zero'd out to make sure we always use the same endinanness. 17093 17094 fixed up as per Michel's suggestion for endianness. 17095 17096 Signed-off-by: David S. Miller <davem (a] davemloft.net> 17097 Signed-off-by: Dave Airlie <airlied (a] linux.ie> 17098 17099 commit 2de90c8349409eb892edc57fd24b61b679eb7719 17100 Author: David Miller <davem (a] davemloft.net> 17101 Date: Sun Feb 15 18:24:01 2009 +1000 17102 17103 DRI: Fix page size used in RADEONDRIGetPciAperTableSize(). 17104 17105 The ATI GART has a fixed size of 4096 bytes. So using 17106 the system page size here is wrong. 17107 17108 Signed-off-by: David S. Miller <davem (a] davemloft.net> 17109 Signed-off-by: Dave Airlie <airlied (a] linux.ie> 17110 17111 commit 7cde00f49649e25fd5816927c7a5e28b608fabcd 17112 Author: Alex Deucher <alexdeucher (a] gmail.com> 17113 Date: Thu Feb 12 17:39:44 2009 -0500 17114 17115 R6xx/R7xx: switch to drm for wait for idle 17116 17117 THIS REQUIRES AN UPDATED DRM 17118 17119 commit e22cd4011b9be437ba89bff568e7fb82b4907d99 17120 Author: Yang Zhao <yang (a] yangman.ca> 17121 Date: Thu Feb 12 14:46:53 2009 -0500 17122 17123 R6xx/R7xx EXA: Further optimizations to overlapping copy 17124 17125 Diagonal overlapping copies can be reduced to either horizontal- or 17126 vertical-only offset, and the one with fewer copies is picked. 17127 17128 commit da08b760bcf3d04d775c4440fafec10657bb1863 17129 Author: Alex Deucher <alexdeucher (a] gmail.com> 17130 Date: Thu Feb 12 13:53:11 2009 -0500 17131 17132 R6xx/R7xx EXA: handle running out of vertex space in the copy path 17133 17134 commit e3be312b0b73982c24f1f5d9cf76d7caafae0853 17135 Author: Alex Deucher <alexdeucher (a] gmail.com> 17136 Date: Thu Feb 12 13:48:36 2009 -0500 17137 17138 R6xx/R7xx EXA: properly handle non repeat cases in the texture setup 17139 17140 commit 3ff1eb7e50fef22adb6280dd7de77c3ecafdb451 17141 Author: Michel Dnzer <daenzer (a] vmware.com> 17142 Date: Thu Feb 12 19:24:41 2009 +0100 17143 17144 EXA: If making a pixmap offscreen fails, return ~0ULL as texture offset. 17145 17146 This allows AIGLX to fall back to the non-zero-copy code rather than messing up 17147 the 3D driver. 17148 17149 commit c19aa4fc8da7ac4745624098b146bcc42c0436dd 17150 Author: Michel Dnzer <daenzer (a] vmware.com> 17151 Date: Thu Feb 12 19:22:12 2009 +0100 17152 17153 EXA: The source tiling code can't handle RepeatReflect yet. 17154 17155 commit adc35636aa1e936ec74cba8931ccffe2b5c11656 17156 Author: Alex Deucher <alexdeucher (a] gmail.com> 17157 Date: Thu Feb 12 13:10:53 2009 -0500 17158 17159 R1xx/R2xx EXA: fix non repeat texture setup 17160 17161 commit c6e1bf75c399807716b99486ab3301aeba60fa79 17162 Author: Alex Deucher <alexdeucher (a] gmail.com> 17163 Date: Thu Feb 12 12:49:24 2009 -0500 17164 17165 R3xx-R5xx EXA: fix texture setup for non-repeat case 17166 17167 fixes bug 19923 17168 17169 commit 0804dde9012cf33f614a1c2ee6d5d0def040a9af 17170 Author: Alex Deucher <alexdeucher (a] gmail.com> 17171 Date: Thu Feb 12 12:02:57 2009 -0500 17172 17173 AVIVO: fix dualhead/rotation for real 17174 17175 commit 3e5ac32f747bef70903e9cdf41652a827eece707 17176 Author: Alex Deucher <alexdeucher (a] gmail.com> 17177 Date: Wed Feb 11 17:40:20 2009 -0500 17178 17179 Revert "R6xx/R7xx: handle running out of vertex buffer space" 17180 17181 This reverts commit 4fd7228de7ad88edf825dbc3039df877795a9479. 17182 17183 Causes lockups. 17184 17185 commit bd141aa73a77f68301715fb3b5664e2082327a80 17186 Author: Alex Deucher <alexdeucher (a] gmail.com> 17187 Date: Wed Feb 11 15:31:47 2009 -0500 17188 17189 R6xx/R7xx Xv: switch packed over to Yang's new shader code 17190 17191 commit cb4a1ceaa32d6847b146a31291772e1183972ee7 17192 Author: Yang Zhao <yang (a] yangman.ca> 17193 Date: Wed Feb 11 15:18:00 2009 -0500 17194 17195 R6xx/R7xx Xv: Planar - Properly scale Y'CbCr values before converting to RGB 17196 17197 According to MPEG-2 spec, Y' and Cb/Cr values are scaled to [16, 235] 17198 and [16, 240], respectively, when packed into bytes. Properly take care 17199 of the reverse scaling before translating to RGB. 17200 17201 Conversion matrix has been simplified to remove 3rd column, as the fitting 17202 to [-0.5, 0.5] can be done with scaling. 17203 17204 Redundant MOV instructions were also removed, and now only 3 GPRs are required. 17205 17206 commit 4fd7228de7ad88edf825dbc3039df877795a9479 17207 Author: Alex Deucher <alexdeucher (a] gmail.com> 17208 Date: Wed Feb 11 14:46:02 2009 -0500 17209 17210 R6xx/R7xx: handle running out of vertex buffer space 17211 17212 draw current VB and start a new one. 17213 17214 commit ec60ef094762901ede4df5cde55c3f162e8c667b 17215 Author: Alex Deucher <alexdeucher (a] gmail.com> 17216 Date: Wed Feb 11 13:05:27 2009 -0500 17217 17218 R6xx/R7xx: be more verbose about what function ran out of VB space 17219 17220 commit e7dedbc355970407eddb38370b58a8c96d204c1d 17221 Author: Alex Deucher <alexdeucher (a] gmail.com> 17222 Date: Wed Feb 11 12:45:14 2009 -0500 17223 17224 R6xx/R7xx: Move engine idle to sync functions 17225 17226 commit 4a759a907ecd571460c7ff64cd6288380359b04b 17227 Author: Alex Deucher <alexdeucher (a] gmail.com> 17228 Date: Wed Feb 11 12:29:50 2009 -0500 17229 17230 adjust alignment 17231 17232 commit d98359b20e1eb8eae893581ab199c299c8352dc4 17233 Author: Alex Deucher <alexdeucher (a] gmail.com> 17234 Date: Wed Feb 11 12:03:56 2009 -0500 17235 17236 Revert "Radeon EXA: wait for the engine to be idle before sw access" 17237 17238 This reverts commit 0c4694c58f2ec152b19023626bc7ced891d9da65. 17239 17240 This patch just adds latency and thus isn't a real solution 17241 17242 commit d6c50b221cbab6e726948e2310c1def9fc38da64 17243 Author: Alex Deucher <alexdeucher (a] gmail.com> 17244 Date: Wed Feb 11 11:05:02 2009 -0500 17245 17246 R6xx/R7xx Xv: add support for packed uploads 17247 17248 commit 8e437e996cc3f2c424c342701f4aa6bcf72ad08e 17249 Author: Alex Deucher <alexdeucher (a] gmail.com> 17250 Date: Wed Feb 11 10:53:50 2009 -0500 17251 17252 R6xx/R7xx Xv: Add native support for packed formats 17253 17254 commit 0c4694c58f2ec152b19023626bc7ced891d9da65 17255 Author: Alex Deucher <alexdeucher (a] gmail.com> 17256 Date: Tue Feb 10 17:08:37 2009 -0500 17257 17258 Radeon EXA: wait for the engine to be idle before sw access 17259 17260 Really we need proper fencing with timestamps but this should 17261 help in the short term. 17262 17263 This should help with minor corruption issues and possibly 17264 lockups. 17265 17266 commit 13ab61d90bb58112bef4ce7267ef127935315f80 17267 Author: Fabio <fabio.ped (a] libero.it> 17268 Date: Tue Feb 10 10:50:30 2009 -0500 17269 17270 man page updates 17271 17272 bug 20030 17273 17274 commit 615f2f202cee5f0dd2c6b29329f843bb1c9c70d5 17275 Author: Alex Deucher <alexdeucher (a] gmail.com> 17276 Date: Tue Feb 10 10:17:47 2009 -0500 17277 17278 Bump version post RC release 17279 17280 commit 1b4afc1c9f8458f9ab3434418f4ccf959f532ac1 17281 Author: Alex Deucher <alexdeucher (a] gmail.com> 17282 Date: Mon Feb 9 19:50:37 2009 -0500 17283 17284 R6xx/R7xx: Add checks to make sure we don't overrun VB space 17285 17286 commit ff6efe513e0c3148812d6a9578d2265522417053 17287 Author: Alex Deucher <alexdeucher (a] gmail.com> 17288 Date: Mon Feb 9 17:43:24 2009 -0500 17289 17290 ATOM: warning fixes 17291 17292 commit 87ad9006f3dafb45e131c9a3d6df713321f39998 17293 Author: Alex Deucher <alexdeucher (a] gmail.com> 17294 Date: Mon Feb 9 17:40:41 2009 -0500 17295 17296 RV350: add AGP quirk 17297 17298 bug 19981 17299 17300 commit 610b8b65fd7aa97f9c03b2ff6ae607aa959eab0b 17301 Author: Alex Deucher <alexdeucher (a] gmail.com> 17302 Date: Mon Feb 9 16:33:20 2009 -0500 17303 17304 Bump for rc release 17305 17306 commit e85b7a1c5948b5123eddf9dfbb9e002d74fbe9fb 17307 Author: Alex Deucher <alexdeucher (a] gmail.com> 17308 Date: Mon Feb 9 14:13:16 2009 -0500 17309 17310 R6xx/R7xx Xv: fix cache flush buffer size for planar 17311 17312 commit 2a893bac1faffd28dce6d9a9693108196c71e6d3 17313 Author: Alex Deucher <alexdeucher (a] gmail.com> 17314 Date: Mon Feb 9 13:18:17 2009 -0500 17315 17316 R6xx/R7xx Xv: add accelerated uploads for planar formats 17317 17318 commit 231aee18a73805be2f6c962e94a8345dd89fd0df 17319 Author: Alex Deucher <alexdeucher (a] gmail.com> 17320 Date: Mon Feb 9 13:02:27 2009 -0500 17321 17322 R6xx/R7xx Xv: implement native shader for planar formats 17323 17324 commit 6c76bfe8105e3cf4e7e6ea1bfe1235be2079110f 17325 Author: Alex Deucher <alexdeucher (a] gmail.com> 17326 Date: Mon Feb 9 09:47:22 2009 -0500 17327 17328 R6xx/R7xx UTS: move actual upload to separate function 17329 17330 So it can be shared with Xv 17331 17332 commit 04859362ce64ed4ae48305a83c92059899c1db4b 17333 Author: Alex Deucher <alexdeucher (a] gmail.com> 17334 Date: Sun Feb 8 20:13:06 2009 -0500 17335 17336 Add some missing r6xx/r7xx pci ids 17337 17338 commit 132e4c575dc4675f4995e45f08c53c26bffd999a 17339 Author: Yang Zhao <yang (a] yangman.ca> 17340 Date: Sat Feb 7 13:54:51 2009 -0500 17341 17342 R6xx/R7xx EXA: Optimize overlapping copy 17343 17344 Overlapping copy is now done in chunks proportional to the 17345 non-overlapping area. 17346 17347 Diagonal overlaps are also handled properly. 17348 17349 commit 8e9ef8ff581892cbe1b7ea56d48b9a1abd70179d 17350 Author: Pierre Ossman <pierre (a] ossman.eu> 17351 Date: Sat Feb 7 18:57:47 2009 +0100 17352 17353 Xv vsync support on r6xx/r7xx cards. 17354 17355 commit 2222f0fd700f100b2e91fac2babe7d1b53f56c3e 17356 Author: Pierre Ossman <pierre (a] ossman.eu> 17357 Date: Sat Feb 7 18:56:42 2009 +0100 17358 17359 Fix bad range adjustment in VLINE code. 17360 17361 commit 8ce6c024e31f6a3f5ae6c882738b1e64ae2944a1 17362 Author: Alex Deucher <alexdeucher (a] gmail.com> 17363 Date: Sat Feb 7 10:32:01 2009 -0500 17364 17365 R6xx/R7xx Xv: fix typos in cache flushing commands 17366 17367 commit bba5d806cc46328ad9e4c521bfeeb2a59f96862b 17368 Author: Alex Deucher <alexdeucher (a] gmail.com> 17369 Date: Fri Feb 6 19:44:17 2009 -0500 17370 17371 AVIVO: better fix for rotation 17372 17373 This should handle initial rotation as well. 17374 17375 commit 28cf3492f431f325cb08d7c6aade43d8886df3f6 17376 Author: Alex Deucher <alexdeucher (a] gmail.com> 17377 Date: Fri Feb 6 19:20:18 2009 -0500 17378 17379 AVIVO: fix rotation 17380 17381 When rotation is active, virtualX and virtualY change. 17382 17383 commit 1b45936ae614244aa49b1a5d3c7fc39773c4f9b6 17384 Author: Alex Deucher <alexdeucher (a] gmail.com> 17385 Date: Fri Feb 6 18:29:55 2009 -0500 17386 17387 R6xx/R7xx EXA: Fix typo in DFS 17388 17389 noticed by pzad in IRC 17390 17391 commit c06d89e16d5b2553142e8641e66080e1770c1563 17392 Author: Alex Deucher <alexdeucher (a] gmail.com> 17393 Date: Fri Feb 6 18:07:22 2009 -0500 17394 17395 R6xx/R7xx EXA: fallback on overlapping blits for now 17396 17397 Leave this disabled until we get a proper solution. 17398 17399 commit 1d5fc3febf3470b94c423a1eda5e0683856909df 17400 Author: Alex Deucher <alexdeucher (a] gmail.com> 17401 Date: Fri Feb 6 14:43:30 2009 -0500 17402 17403 Revert "r6xx/r7xx EXA: Optimize overlapping copy" 17404 17405 This reverts commit 0dfadc1843e0d14b9cc1ee19a72f4fd60a2c495b. 17406 17407 This doesn't always work properly. 17408 17409 commit def317e22d072405cd95ddb19d17eacd784ffd9e 17410 Author: Alex Deucher <alexdeucher (a] gmail.com> 17411 Date: Fri Feb 6 11:04:59 2009 -0500 17412 17413 R6xx/R7xx EXA: add accelerated UTS/DFS hooks 17414 17415 I'm not sure how much of a win these are. I need to 17416 do some benchmarking. 17417 17418 commit 0dfadc1843e0d14b9cc1ee19a72f4fd60a2c495b 17419 Author: Yang Zhao <yang (a] yangman.ca> 17420 Date: Fri Feb 6 10:29:39 2009 -0500 17421 17422 r6xx/r7xx EXA: Optimize overlapping copy 17423 17424 When source and destination blocks are only offset horizontally, it 17425 appears to be unnecessary to perform careful, segment-by-segment copy. 17426 The code path that does this is taken out completely. 17427 17428 For the case where offset is only vertical, copying is now done by 17429 height of the non-overlapping area each time, instead of always 17430 line-by-line. 17431 17432 commit 3d17bd199423e92d201c20f047d5e699942af976 17433 Author: Alex Deucher <alexdeucher (a] gmail.com> 17434 Date: Thu Feb 5 19:25:03 2009 -0500 17435 17436 Revert "R6xx/R7xx EXA: improve overlapping copy performance" 17437 17438 This reverts commit b24827c9d211e8a35da53b665385a7733d19910d. 17439 17440 This seems to cause corruption in some cases. 17441 17442 commit bf74055f543e7f0664741620fb1fe827ebc12711 17443 Author: Alex Deucher <alexdeucher (a] gmail.com> 17444 Date: Thu Feb 5 17:31:43 2009 -0500 17445 17446 r6xx/r7xx EXA: fix corruption when doing sw access 17447 17448 need to wait until the engine is idle. Ideally we wait 17449 on a timestamp shadowed in memory, but polling the 17450 GRBM_STATUS reg will do for now. 17451 17452 commit b24827c9d211e8a35da53b665385a7733d19910d 17453 Author: Alex Deucher <alexdeucher (a] gmail.com> 17454 Date: Thu Feb 5 16:21:37 2009 -0500 17455 17456 R6xx/R7xx EXA: improve overlapping copy performance 17457 17458 send vertices for each line of the copy, but only draw once 17459 17460 commit 729fe756f809a41494dc161252d82313956e4a4b 17461 Author: Alex Deucher <alexdeucher (a] gmail.com> 17462 Date: Thu Feb 5 16:08:35 2009 -0500 17463 17464 r6xx/r7xx EXA: cleanup overlapping copy 17465 17466 commit 16e01a5796a8e8b86ad25d3aa45b9e7044dc72d1 17467 Author: Alex Deucher <alexdeucher (a] gmail.com> 17468 Date: Tue Feb 3 19:46:10 2009 -0500 17469 17470 Fix encoder accounting 17471 17472 Should fix bug 19924 17473 17474 commit d7bf7b9c17add31dc9a115b56a563c7f9bf2cdb6 17475 Author: Alex Deucher <alexdeucher (a] gmail.com> 17476 Date: Tue Feb 3 18:28:38 2009 -0500 17477 17478 EXA: fix and re-enable Solid() on R7xx 17479 17480 missing last bit in alu clause in solid PS 17481 17482 commit 9abb09d5f43320bd27b336612d39d264fe93549f 17483 Author: Alex Deucher <alexdeucher (a] gmail.com> 17484 Date: Tue Feb 3 11:17:13 2009 -0500 17485 17486 Fix bad rv710 pci id 17487 17488 commit e5b916770946a9eebcb4bd1e6f698220db8c718a 17489 Author: Alex Deucher <alexdeucher (a] gmail.com> 17490 Date: Tue Feb 3 10:48:41 2009 -0500 17491 17492 Allow rotation on r6xx/r7xx 17493 17494 commit d1f071c7f1dad6babfbcfcc2cb2b722a4987f372 17495 Author: Alex Deucher <alexdeucher (a] gmail.com> 17496 Date: Tue Feb 3 10:44:10 2009 -0500 17497 17498 Initial R6xx/R7xx EXA and textured video support 17499 17500 commit c88c3ef6f3db266c1aacba5297b8dfc8b66bf00e 17501 Author: Wolke Liu <wolke.liu (a] amd.com> 17502 Date: Mon Feb 2 17:01:34 2009 -0500 17503 17504 AVIVO: Save/restore vga pll registers 17505 17506 This fixes some VT switch issues on some chips 17507 17508 commit 6fac3cefd1f46161c1e276ba40e72da2823aa9f6 17509 Author: Alex Deucher <alexdeucher (a] gmail.com> 17510 Date: Sat Jan 31 14:21:02 2009 -0500 17511 17512 Return NULL for encoder if no active device is assigned 17513 17514 fixes bug 19855 17515 17516 commit db178c852d00e9b85513d750b5f8f7495f9e3360 17517 Author: Alan Coopersmith <alan.coopersmith (a] sun.com> 17518 Date: Fri Jan 30 20:37:41 2009 -0800 17519 17520 Add README with pointers to mailing list, bugzilla & git repos 17521 17522 commit fa8e5a4fc236f8f15f462cb0d6164b194a65a118 17523 Author: Michel Dnzer <daenzer (a] vmware.com> 17524 Date: Fri Jan 30 16:16:26 2009 +0100 17525 17526 EXA: Accelerate Composite of RepeatPad/Reflect pictures when possible. 17527 17528 commit 3a6e958c59b4b44fe620c8abe760ed1abf0007b1 17529 Author: airlied <airlied@roo.(none)> 17530 Date: Fri Jan 30 07:56:14 2009 +1000 17531 17532 rs780: include RS780 in the InitMemory to leave alone 17533 17534 This stops the MC_FB_LOCATION getting reassigned to 0, which 17535 is really bad thing to happen. 17536 17537 I've had reported memory corruption on these cards so hopefully this 17538 fixes it. 17539 17540 commit bd8021d46e9066e4cd116c03a7b7adcfe2557aff 17541 Merge: 4f88dd83f 20d5dd387 17542 Author: Alex Deucher <alexdeucher (a] gmail.com> 17543 Date: Wed Jan 28 14:33:14 2009 -0500 17544 17545 Merge branch 'atom-rework' and fix conflicts 17546 17547 commit 20d5dd387da555e895e2b73fb53e2b026dd91003 17548 Author: Alex Deucher <alexdeucher (a] gmail.com> 17549 Date: Wed Jan 28 14:12:26 2009 -0500 17550 17551 Move encoder specific data to encoder dev_priv 17552 17553 commit 4f88dd83f290fb38e41256a7f1804008c0f28139 17554 Merge: 1c5678fc4 2ca970c7a 17555 Author: Alex Deucher <alexdeucher (a] gmail.com> 17556 Date: Wed Jan 28 13:13:13 2009 -0500 17557 17558 Merge branch 'atom-tvout' of ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati 17559 17560 Conflicts: 17561 17562 src/radeon_atombios.c 17563 17564 commit 1a62360e8260eaf5f98fa6be3f89f64c8cae7d32 17565 Author: Alex Deucher <alexdeucher (a] gmail.com> 17566 Date: Mon Jan 26 10:11:28 2009 -0500 17567 17568 Encoders not assigned yet, use supported devices 17569 17570 need to move this to encoder setup 17571 17572 commit 44d9ceacb1fa6e1859d5d8bd1c55405aea24ad85 17573 Author: Alex Deucher <alexdeucher (a] gmail.com> 17574 Date: Mon Jan 26 03:47:32 2009 -0500 17575 17576 Fix legacy output setup 17577 17578 don't have assigned encoders at this point. Need to 17579 eventually move this stuff to encoder dev_priv. 17580 17581 commit 0c24cbcea56c8f36374aca0f8ba849adfbf0ecfe 17582 Author: Alex Deucher <alexdeucher (a] gmail.com> 17583 Date: Mon Jan 26 03:24:19 2009 -0500 17584 17585 ATOM: print useful output info for DPMS events 17586 17587 commit 2f92b885457c5cee2098ebbf8c6fe5a5eb98ec48 17588 Author: Alex Deucher <alexdeucher (a] gmail.com> 17589 Date: Mon Jan 26 03:12:44 2009 -0500 17590 17591 Warning fixes 17592 17593 commit 4cbf6189f3a7945a1141b0451e4f7881ceaaadcd 17594 Author: Alex Deucher <alexdeucher (a] gmail.com> 17595 Date: Mon Jan 26 02:59:51 2009 -0500 17596 17597 Move legacy output setup functions to legacy_output.c 17598 17599 commit 93ed767255b60fbdf5b416b9bd06c366036a5141 17600 Author: Alex Deucher <alexdeucher (a] gmail.com> 17601 Date: Mon Jan 26 02:40:41 2009 -0500 17602 17603 Fix off by one when printing encoder name 17604 17605 - also cleanup some debugging output 17606 17607 commit b515eee4b1cafae78ed39cd91c18b4f5e671b20b 17608 Author: Alex Deucher <alexdeucher (a] gmail.com> 17609 Date: Mon Jan 26 01:50:18 2009 -0500 17610 17611 Additional output cleanup 17612 17613 commit 1c5678fc452dbc93135e9ece50a4cff6583f6078 17614 Author: Michel Dnzer <daenzer (a] vmware.com> 17615 Date: Sat Jan 24 11:10:00 2009 +0100 17616 17617 Drop memcpy fallbacks from EXA UploadToScreen and DownloadFromScreen hooks. 17618 17619 While in theory it's possible for the PrepareAccess hook to fail on big endian 17620 platforms, in practice it isn't at this point because there's no other users of 17621 the surface registers. 17622 17623 commit 5cbd6082b9f8d7009ee9ef5e036673f99e1bfb05 17624 Author: Michel Dnzer <daenzer (a] vmware.com> 17625 Date: Sat Jan 24 11:07:12 2009 +0100 17626 17627 Don't transform EXA Composite mask coordinates when there's no mask. 17628 17629 commit 7e54c4b8acafead8de5d0dd6fdc1b619f832dd6f 17630 Author: Thomas Jaeger <ThJaeger (a] gmail.com> 17631 Date: Fri Jan 23 20:03:27 2009 -0500 17632 17633 Fall back to software for unsupported repeat modes 17634 17635 commit 289b4a377cfff0c2a8ec938951cfbae1e270570e 17636 Author: Alex Deucher <alexdeucher (a] gmail.com> 17637 Date: Fri Jan 23 20:00:43 2009 -0500 17638 17639 Remove OutputType and other cruft 17640 17641 commit 5f98508effbdc1fd62d0df15798722c147d88da1 17642 Author: Maciej Cencora <m.cencora (a] gmail.com> 17643 Date: Fri Jan 23 19:33:25 2009 -0500 17644 17645 Make sure gb_num_pipes is initialized when DRI is disabled 17646 17647 Found with valgrind 17648 17649 commit 0faab02cc732414af3cfb2014d3ddc1139aff063 17650 Author: Alex Deucher <alexdeucher (a] gmail.com> 17651 Date: Fri Jan 23 17:35:01 2009 -0500 17652 17653 Remove some unused cruft 17654 17655 commit 025082160c53c01a44f4d681cc63bf26ac4561d5 17656 Author: Alex Deucher <alexdeucher (a] gmail.com> 17657 Date: Fri Jan 23 17:31:18 2009 -0500 17658 17659 track encoder state 17660 17661 commit 82f12e5a40c1fbcb91910a0f8b725c34fff02aae 17662 Author: Alex Deucher <alexdeucher (a] gmail.com> 17663 Date: Fri Jan 23 16:21:33 2009 -0500 17664 17665 Remove TMDSType, DACType, LVDSType from output rec 17666 17667 commit 7b42b57d748f77ff6ef6f8ade7cbc983a98d7204 17668 Author: Alex Deucher <alexdeucher (a] gmail.com> 17669 Date: Wed Jan 21 16:28:00 2009 -0500 17670 17671 few more logic pasto's bits I missed 17672 17673 commit fef1c12fbe4ceb4676258850d1fa5cac994f7c21 17674 Author: Alex Deucher <alexdeucher (a] gmail.com> 17675 Date: Wed Jan 21 15:34:01 2009 -0500 17676 17677 Fix compilation with RADEON_TRACE_FALL set 17678 17679 noticed by curaga on phoronix forum 17680 17681 commit 1305376b009d17d65b4935896dc35cb7530f7c52 17682 Author: Alex Deucher <alexdeucher (a] gmail.com> 17683 Date: Wed Jan 21 15:20:47 2009 -0500 17684 17685 Move active_device setup to detect() 17686 17687 Also add encoder debugging printouts 17688 17689 commit 9e0cc0ca5d7bc68756ae92ee3b5b2392366be71a 17690 Author: Alex Deucher <alexdeucher (a] gmail.com> 17691 Date: Wed Jan 21 14:47:08 2009 -0500 17692 17693 Fix logic cut and paste error 17694 17695 commit 12010ae9296635ffa5f42ce0cd14482aaa290cf7 17696 Author: Alex Deucher <alexdeucher (a] gmail.com> 17697 Date: Wed Jan 21 13:52:11 2009 -0500 17698 17699 More legacy rework 17700 17701 commit 3f8bbede8bf639050f773780888b89cbdcb425b8 17702 Author: Alex Deucher <alexdeucher (a] gmail.com> 17703 Date: Wed Jan 21 13:45:11 2009 -0500 17704 17705 fix legacy crtc routing and add some debugging info 17706 17707 commit 80e0162ea2c1f793964731268251aba2d3243127 17708 Author: Alex Deucher <alexdeucher (a] gmail.com> 17709 Date: Wed Jan 21 13:28:14 2009 -0500 17710 17711 ATOM: fix encoder init 17712 17713 commit aea27d9caabb0e54e868a21e6e547733a4ca2709 17714 Author: Alex Deucher <alexdeucher (a] gmail.com> 17715 Date: Wed Jan 21 12:33:49 2009 -0500 17716 17717 Switch legacy output code to use new encoder objects 17718 17719 commit 9b44a4b57b00b150d041a5d82cb2df42159c062b 17720 Author: Alex Deucher <alexdeucher (a] gmail.com> 17721 Date: Wed Jan 21 11:32:24 2009 -0500 17722 17723 ATOM: more output cleanup 17724 17725 commit 9e283fa9de34407fa6f0d5a8006a611b104f4bf2 17726 Author: Alex Deucher <alexdeucher (a] gmail.com> 17727 Date: Wed Jan 21 03:16:20 2009 -0500 17728 17729 Fixup encoder setup on pre-ATOM chips 17730 17731 commit 5850abbdcd22b70bdbccd3642f8ba1a27e38e7cd 17732 Author: Alex Deucher <alexdeucher (a] gmail.com> 17733 Date: Wed Jan 21 02:28:47 2009 -0500 17734 17735 clean up encoder setup 17736 17737 - make sure not to duplicate encoders 17738 - use single encoder array 17739 17740 commit 152d1d4c179c262be4d0a5618400aa13b4820342 17741 Author: Alex Deucher <alexdeucher (a] gmail.com> 17742 Date: Tue Jan 20 00:58:18 2009 -0500 17743 17744 First pass at converting legacy code to encoder objects 17745 17746 commit 7cc7ba8bf9ce2f0eebfd9b48aa648c416c333ff4 17747 Author: Alex Deucher <alexdeucher (a] gmail.com> 17748 Date: Mon Jan 19 18:42:12 2009 -0500 17749 17750 ATOM: round 1 of output rework 17751 17752 commit 25f3878858e6efb486888e5271b60d4ba08c73c6 17753 Author: Alex Deucher <alexdeucher (a] gmail.com> 17754 Date: Mon Jan 19 15:54:42 2009 -0500 17755 17756 start to re-org outputs 17757 17758 commit cbeb310d35e5ea25ba7a45a2c555468391cdf29e 17759 Author: Alex Deucher <alexdeucher (a] gmail.com> 17760 Date: Fri Jan 16 19:53:49 2009 -0500 17761 17762 ATOM: switch to define for external tmds 17763 17764 commit d0d157abced43f822885fa26f14a4924766594cd 17765 Author: Alex Deucher <alexdeucher (a] gmail.com> 17766 Date: Fri Jan 16 19:51:07 2009 -0500 17767 17768 ATOM: combine DAC setup functions 17769 17770 plus a few other clean-ups 17771 17772 commit f6f969338f8187529c7ae9b2adc980b71a3f5837 17773 Author: Alex Deucher <alexdeucher (a] gmail.com> 17774 Date: Fri Jan 16 18:10:34 2009 -0500 17775 17776 DCE30: LVTMA requires DIG2 encoder 17777 17778 Fixes bug 19544 17779 17780 commit 549ae63aecc2624a69b41ce4db75883d03d59c23 17781 Author: Alex Deucher <alexdeucher (a] gmail.com> 17782 Date: Thu Jan 15 19:26:40 2009 -0500 17783 17784 RV280 Add another AGP quirk 17785 17786 commit 0e69068a8ed82e3fb7349d8224933451955e5f16 17787 Author: Tormod Volden <debian.tormod (a] gmail.com> 17788 Date: Tue Jan 13 20:31:17 2009 -0500 17789 17790 [PATCH] Add yet another AGP quirk for RV280 17791 17792 commit b437da34cb4f7c637674fb432ad81b9bf32244a3 17793 Author: Christiaan van Dijk <dvbmail (a] xs4all.nl> 17794 Date: Mon Jan 12 11:54:15 2009 -0500 17795 17796 R3xx/R4xx: Maximize the use of clipped triangles for Xv rendering 17797 17798 While trying to get tear free video on my RS690 I found some problems in the code for rendering the 17799 video as a clipped triangle. The size limitations are quite strict and scaling the video output window 17800 creates all kinds of weird effects. This patch maximizes the use of this function and resolves the 17801 scaling issues. There could be a problem with zero target height/width, not sure if this is handled 17802 elsewhere. 17803 17804 commit 3d106b990fa8daba96e590c57b164640a2faae7f 17805 Author: Alan Coopersmith <alan.coopersmith (a] sun.com> 17806 Date: Fri Jan 9 16:26:04 2009 -0800 17807 17808 Remove xorgconfig & xorgcfg from See Also list in man page 17809 17810 commit 14d29a223028c183ba11e717a3737285d0ac8de1 17811 Author: Alex Deucher <alexdeucher (a] gmail.com> 17812 Date: Wed Jan 7 17:04:42 2009 -0500 17813 17814 RV280: add another AGP quirk 17815 17816 commit 1795706a6bb13272a58e8fa1bd5ad2603d553dde 17817 Author: Nicos Gollan <gtdev (a] spearhead.de> 17818 Date: Wed Jan 7 14:25:24 2009 -0500 17819 17820 [PATCH] Fixed enumerations in radeon-output.c 17821 17822 commit f69ad5aa17bb907105c1657803628d52b5ff3f11 17823 Author: Alex Deucher <alexdeucher (a] gmail.com> 17824 Date: Wed Jan 7 11:33:53 2009 -0500 17825 17826 Bump version post release 17827 17828 commit a0dd5d7ee3f038a9bfe051db8dbfac4934a81581 17829 Author: Alex Deucher <alexdeucher (a] gmail.com> 17830 Date: Tue Jan 6 19:08:28 2009 -0500 17831 17832 Bump version for release 17833 17834 commit c3fb8bb279959512a4ced644a64cc660c5cd97bd 17835 Author: Alex Deucher <alexdeucher (a] gmail.com> 17836 Date: Tue Jan 6 17:41:22 2009 -0500 17837 17838 ATOM: rework encoder/transmitter setup 17839 17840 commit 9bb6b8ceaf772e1ce63610812a12260ee3860c27 17841 Author: Alex Deucher <alexdeucher (a] gmail.com> 17842 Date: Tue Jan 6 16:25:26 2009 -0500 17843 17844 ATOM: refactor output dpms 17845 17846 commit 467995d960bdefb1911eabfc2a047cf816013101 17847 Author: Alex Deucher <alexdeucher (a] gmail.com> 17848 Date: Tue Jan 6 15:18:41 2009 -0500 17849 17850 ATOM: Adjust PLL setup for recent atom changes 17851 17852 commit ee6310174c44809c761dbdb423e56e61392a0e10 17853 Author: Alex Deucher <alexdeucher (a] gmail.com> 17854 Date: Tue Jan 6 14:59:22 2009 -0500 17855 17856 ATOM: handle cases where TMDS uses linkb 17857 17858 fixes bug 18564 17859 17860 commit 690999b682c0f954e27d9f1b3a3d8db4ef865c21 17861 Author: Alex Deucher <alexdeucher (a] gmail.com> 17862 Date: Tue Jan 6 14:55:49 2009 -0500 17863 17864 ATOM: rework object table parsing 17865 17866 commit e23baa2fa8b25b68cd473aec7a3d8cf3ee7b0ea8 17867 Author: Alex Deucher <alexdeucher (a] gmail.com> 17868 Date: Tue Jan 6 14:38:06 2009 -0500 17869 17870 R100: Fix typo in e55598cc1b35d398c9eb2c3946df71456775f621 17871 17872 Noticed by Andreas Stenglein 17873 17874 commit d5cb24075928690377fd2dfaa32abfebff9990df 17875 Author: Alex Deucher <alexdeucher (a] gmail.com> 17876 Date: Sun Jan 4 21:04:52 2009 -0500 17877 17878 ATOM: add format rev 2 for digitial command tables 17879 17880 should fix bug 18911 17881 17882 commit 1813456af3def38c6f53103054976f3092491746 17883 Author: Alex Deucher <alexdeucher (a] gmail.com> 17884 Date: Sun Jan 4 21:01:05 2009 -0500 17885 17886 ATOM: fail on more tables with unhandled revisions 17887 17888 commit 90ce00a3dbf37833a71db234372695460c8f0342 17889 Author: Alex Deucher <alexdeucher (a] gmail.com> 17890 Date: Sun Jan 4 19:13:53 2009 -0500 17891 17892 AGP: Print both device and subsytem pci ids for agp 17893 17894 helpful for adding quirks with libpciaccess 17895 17896 commit 006d736d890a6b2f1d7eb8ee20ccf467a939516f 17897 Author: Alex Deucher <alexdeucher (a] gmail.com> 17898 Date: Sun Jan 4 19:08:37 2009 -0500 17899 17900 Another AGP quirk 17901 17902 debian bug 510208 17903 17904 commit c39ca66a7e216b7492fc5e3680ac535831cfb9a1 17905 Author: Alex Deucher <alexdeucher (a] gmail.com> 17906 Date: Tue Dec 30 19:05:58 2008 -0500 17907 17908 R2xx: add another AGP quirk 17909 17910 from http://www.phoronix.com/forums/showthread.php?t=14626 17911 17912 commit 877be1b268f275b8ad83d5e245a0f0d3b1a05091 17913 Author: Fabio <fabio.ped (a] libero.it> 17914 Date: Mon Dec 29 18:11:10 2008 -0500 17915 17916 man page updates 17917 17918 bug 19329 17919 17920 commit 837f7e8a48ea35d21de24b620830a2c6db854861 17921 Author: Dave Airlie <airlied (a] linux.ie> 17922 Date: Tue Dec 30 06:50:54 2008 +1000 17923 17924 atombios: endian fix in r600 object table handling 17925 17926 This fixes the r600 object table on r600. 17927 17928 commit 45f8b911659dba7f43b9b70bde4b9c3be51b0b5b 17929 Author: Alex Deucher <alexdeucher (a] gmail.com> 17930 Date: Mon Dec 29 10:25:33 2008 -0500 17931 17932 RS600: remove gpio quirk for ddc 17933 17934 Doesn't seem necessary on RS600. I'll try and find out more. 17935 17936 commit e913990030c536427bfecdb7b3d1e18f1beb90d2 17937 Author: Alex Deucher <alexdeucher (a] gmail.com> 17938 Date: Mon Dec 29 10:22:20 2008 -0500 17939 17940 Revert "radeon: maybe special handling only needed for DVI port not HDMI one." 17941 17942 This reverts commit c877450a75d3fd4a2f200af197c691669966f6f4. 17943 17944 This breaks RS690 setups. I think for now we just apply this quirk to 17945 RS690/RS740. I'll follow up with the bios folks about RS600. 17946 17947 commit c877450a75d3fd4a2f200af197c691669966f6f4 17948 Author: Dave Airlie <airlied (a] linux.ie> 17949 Date: Mon Dec 29 09:05:45 2008 +1000 17950 17951 radeon: maybe special handling only needed for DVI port not HDMI one. 17952 17953 fixes ciaranm's rs600 board on radeon DDC 17954 17955 commit 2a9bc6be3246ec230b181391221e3a46beb94a23 17956 Author: Dave Airlie <airlied (a] linux.ie> 17957 Date: Mon Dec 29 09:05:16 2008 +1000 17958 17959 radeon: a-bit motherboard has no DVI output 17960 17961 thanks to ciaranm for finding this. 17962 17963 commit eccf3fed2be54f5c02fefef32e39c4084acbbe22 17964 Author: Alex Deucher <alexdeucher (a] gmail.com> 17965 Date: Fri Dec 26 19:51:58 2008 -0500 17966 17967 Make sure to include radeon_drm.h in the tarball 17968 17969 commit e55598cc1b35d398c9eb2c3946df71456775f621 17970 Author: Alex Deucher <alexdeucher (a] gmail.com> 17971 Date: Fri Dec 26 19:31:55 2008 -0500 17972 17973 R2xx EXA: limit composite to 2047 like r1xx 17974 17975 See bug 19269 17976 17977 commit 631123d144d088d4f77a0599c34deaa31d551d71 17978 Author: Dave Airlie <airlied (a] redhat.com> 17979 Date: Tue Dec 23 10:08:46 2008 +1000 17980 17981 radeon: update to 6.9.0.91 17982 17983 do a prerelease for end of year. 17984 17985 commit 047f7603174f6047090f4fafbe6488d697eb78bc 17986 Author: Dave Airlie <airlied (a] linux.ie> 17987 Date: Tue Dec 23 08:16:25 2008 +1000 17988 17989 radeon: pass distcheck 17990 17991 commit c0c33dab44e6966b1702d4e8cfba3537fc6e2d5c 17992 Author: Patrick Haller <patrick.haller (a] haller-consult.com> 17993 Date: Mon Dec 22 03:06:23 2008 -0500 17994 17995 Fix off by one in EXA composite limit checking 17996 17997 Patch from Patrick, with some updates from me: 17998 - fix r200 limits 17999 - note about r100 limits 18000 18001 commit d01a609a2b07da4ca0f182e79459432584ec61f6 18002 Author: Alex Deucher <alexdeucher (a] gmail.com> 18003 Date: Mon Dec 22 02:18:43 2008 -0500 18004 18005 R3xx-R5xx: better fix for xv primitive tearing issues 18006 18007 R5xx: always use single clipped triangle 18008 R3xx/R4xx: use single clipped triangle up to guardband limit, 18009 then use quad. 18010 18011 commit 4d009d14c9c72f6e1bc8100ec8d9b8037ee4e514 18012 Merge: d52882d81 4e96278b5 18013 Author: Dave Airlie <airlied (a] redhat.com> 18014 Date: Mon Dec 22 16:00:01 2008 +1000 18015 18016 Merge branch 'atom-tvout' 18017 18018 commit 4e96278b581e296c1203d97a6d7aa3bff3977222 18019 Author: Dave Airlie <airlied (a] redhat.com> 18020 Date: Mon Dec 22 15:58:35 2008 +1000 18021 18022 atombios/tv: add an option to enable atom tv-out for users. 18023 18024 This code is still experimental but we will allow users to enable it for experimental reasons 18025 18026 commit d52882d8188830dd52fa112dadcf5ea7f3e5fd5e 18027 Author: Dave Airlie <airlied (a] redhat.com> 18028 Date: Mon Dec 22 09:27:47 2008 +1000 18029 18030 radeon: setup 3D engine even when no DRI. 18031 18032 This should fix some missing font issues in EXA without DRI cases. 18033 18034 commit 2346fd7cbd90dcdce2b361d374a53da064fa6a12 18035 Author: Dave Airlie <airlied (a] redhat.com> 18036 Date: Thu Dec 18 16:35:13 2008 +1000 18037 18038 radeon: add all new pci ids for rv730/rv710 families 18039 18040 commit 2ca970c7aa1f3f212f03bf3287d6b60eeafd68ae 18041 Author: Alex Deucher <alexdeucher (a] gmail.com> 18042 Date: Wed Dec 17 13:07:19 2008 -0500 18043 18044 Allow arbitrary tv-out modes 18045 18046 commit 5da147eec70abb7c9125a8940089fd3ff8bdea95 18047 Author: Alex Deucher <alexdeucher (a] gmail.com> 18048 Date: Wed Dec 17 05:37:58 2008 -0500 18049 18050 missed one in last commit 18051 18052 commit 6bfbb930c549e67f68969f7f0483380417f755b5 18053 Author: Alex Deucher <alexdeucher (a] gmail.com> 18054 Date: Wed Dec 17 05:36:09 2008 -0500 18055 18056 Make sure we hit the right bios reg 18057 18058 commit 6ed0e68060fd2775e2d4027ff1533729e87848ba 18059 Author: Alex Deucher <alexdeucher (a] gmail.com> 18060 Date: Wed Dec 17 04:54:50 2008 -0500 18061 18062 properly handle EnableYUV 18063 18064 A bit of a hack. we really need to restructure the 18065 bios scratch reg handling. 18066 18067 commit f21e673aced1bb5138bbb80e6d9724530eb95ebf 18068 Author: Alex Deucher <alexdeucher (a] gmail.com> 18069 Date: Wed Dec 17 03:17:45 2008 -0500 18070 18071 Fix colors on tv-out 18072 18073 Call EnableYUV to set the transform and enable/disable it. 18074 should also fix green tint with tv connected at boot issues. 18075 18076 commit b2b43905a5385a8bb0b59b8e50952863d8dacb59 18077 Author: Alex Deucher <alexdeucher (a] gmail.com> 18078 Date: Wed Dec 17 00:59:52 2008 -0500 18079 18080 Pre-avivo: fix FP setup 18081 18082 - make sure to clear various shadow timing bits 18083 - crtc1 select bit was set wrong. 18084 - should fix bugs like 19100 18085 18086 commit 3754cc23abac99880eda87d06aa42e16081b6c8a 18087 Author: Alex Deucher <alexdeucher (a] gmail.com> 18088 Date: Wed Dec 17 00:56:19 2008 -0500 18089 18090 R2xx: switch EXA composite to rects rather quads 18091 18092 For conistency with Xv path and to reduce potential diagonal tearing 18093 18094 commit be12cf8fcc7a5a3c3ee2c630474e2f9d9368f4d4 18095 Author: Dave Airlie <airlied (a] redhat.com> 18096 Date: Wed Dec 17 15:28:25 2008 +1000 18097 18098 radeon: r500 tv-out force scaler values to nice set that looks correct 18099 18100 commit 01f7a50d76f706a456de7a4eb8323939d905c82b 18101 Author: Dave Airlie <airlied (a] redhat.com> 18102 Date: Wed Dec 17 13:43:21 2008 +1000 18103 18104 r500: re-enable TV out 18105 18106 commit 45b9ff5e250d69de0df41f68b33524f7e2ce13b7 18107 Author: Dave Airlie <airlied (a] redhat.com> 18108 Date: Wed Dec 17 13:42:10 2008 +1000 18109 18110 radeon: r500 PAL timings are slightly incorrect 18111 18112 this stablises the PAL image at least 18113 18114 commit 435d996949185201a97493e131f9920a32863acd 18115 Author: Dave Airlie <airlied (a] redhat.com> 18116 Date: Wed Dec 17 11:50:48 2008 +1000 18117 18118 radeon: tv timings have crev/frev wrong way around for some reason. 18119 18120 commit b9549e89cbdf8fff4aa7bd9f3adfb7d5732508cc 18121 Author: Dave Airlie <airlied (a] redhat.com> 18122 Date: Wed Dec 17 11:40:53 2008 +1000 18123 18124 atombios: re-enable TV dpms 18125 18126 commit db85ff2e0c52a030546c0cc8ed0b851418f3e045 18127 Author: Dave Airlie <airlied (a] redhat.com> 18128 Date: Wed Dec 17 11:37:17 2008 +1000 18129 18130 atom/r600: re-enable TV outputs 18131 18132 commit ff7b52c9a48f2e417acf6bec3b72546ad9b33471 18133 Author: Dave Airlie <airlied (a] redhat.com> 18134 Date: Wed Dec 17 11:36:38 2008 +1000 18135 18136 radeon: use latest atombios constants for TV encoders in set_crtc_source 18137 18138 commit 1e0c7d3f0bf6be67b976fdea8fa2df0b4b16435f 18139 Author: Dave Airlie <airlied (a] redhat.com> 18140 Date: Wed Dec 17 11:35:06 2008 +1000 18141 18142 atom: spc3 only requires ucEncoderMode for DIG outputs 18143 18144 commit 105cfd0b8696347ef71ad9daa941abb308bd6c2c 18145 Author: Dave Airlie <airlied (a] redhat.com> 18146 Date: Wed Dec 17 11:31:15 2008 +1000 18147 18148 atom/tv: enable scaler can set the TV output scaler up. 18149 18150 Add the tv standard to the scaler enable for TV use. 18151 18152 commit 2a768abbd2f18abe203571213260bccc225f70fd 18153 Author: Dave Airlie <airlied (a] redhat.com> 18154 Date: Wed Dec 17 11:29:07 2008 +1000 18155 18156 atom/tv: fixup the tv out default if bios doesn't specify 18157 18158 commit b7fdd5da08bfec705691363b8890699c794dce53 18159 Author: Dave Airlie <airlied (a] redhat.com> 18160 Date: Wed Dec 17 11:24:56 2008 +1000 18161 18162 atombios: fixup parsing of TV tables on r500 and r600 cards. 18163 18164 The TV tables changed between for later cards. 18165 18166 commit e3b08cb59716b525b3063b184eab70e32b9d66f0 18167 Author: Dave Airlie <airlied (a] redhat.com> 18168 Date: Tue Dec 16 15:43:16 2008 +1000 18169 18170 atombios: correct fix for previous issue. 18171 18172 I committed an old patch. 18173 18174 commit 1431bc6900b0e0d7f486b5b3ba281873472b3fdc 18175 Author: Dave Airlie <airlied (a] redhat.com> 18176 Date: Tue Dec 16 15:31:34 2008 +1000 18177 18178 atombios: fix issue with Clearing of MM_INDEX 0. 18179 18180 While playing with tv-out I discovered this issue with clearing 18181 the MM_INDEX register. 18182 18183 commit f72367aff2d09161d30066a00673ba291b292a36 18184 Author: Dave Airlie <airlied (a] redhat.com> 18185 Date: Tue Dec 16 12:32:24 2008 +1000 18186 18187 radeon: add initial rv730 pciids. 18188 18189 Lots more to come, this just adds a few to start. 18190 18191 commit 4d2429f3ec4d7d3e1512b268c7ff5cbff8d186b4 18192 Author: Dave Airlie <airlied (a] redhat.com> 18193 Date: Tue Dec 16 12:19:00 2008 +1000 18194 18195 DCE32: add support for DCE3.2 digital outputs. 18196 18197 This adds support for the digital outputs on the RV730. 18198 18199 commit 4d3792fd1b9f733107582436540f483d8993199f 18200 Author: Dave Airlie <airlied (a] redhat.com> 18201 Date: Fri Dec 12 10:34:57 2008 +1000 18202 18203 radeon: fixes from Alex for some output engines 18204 (cherry picked from commit 1e8ac6ea7b35ccbc3649c96ead60f69158f66ebc) 18205 18206 commit f892154dbd0d8b6dd8abdf2196ab0c7c18327412 18207 Author: Alex Deucher <alexdeucher (a] gmail.com> 18208 Date: Mon Dec 15 19:41:32 2008 -0500 18209 18210 R3xx-R5xx: switch back to quad rendering for Xv 18211 18212 The large clipped triangle trick hits the guardband limits on 18213 large displays. This gets textured video working again 18214 until we sort out how to deal with the situation better. 18215 18216 See bug 19046 18217 18218 commit 62344e460f5d0cac81f94c535cda77d3d76528a9 18219 Author: Alex Deucher <alexdeucher (a] gmail.com> 18220 Date: Mon Dec 15 11:30:42 2008 -0500 18221 18222 Remove remnants of unused DDCMode option 18223 18224 commit f1efee6a11dc9436276fb12cfb90aa79ada1192b 18225 Author: Alex Deucher <alexdeucher (a] gmail.com> 18226 Date: Mon Dec 15 11:27:17 2008 -0500 18227 18228 Remove mention of MergedFB from server log 18229 18230 noticed by Fabio, bug 19098 18231 18232 commit 94bffac6968057cdedc792f8540fe44cb6ecfd7c 18233 Author: Fabio <fabio.ped (a] libero.it> 18234 Date: Mon Dec 15 11:22:56 2008 -0500 18235 18236 man page updates 18237 18238 from bug 19098 with some minor fixes from me 18239 18240 commit c7899905ec824071f9cabbd1e271c4d76010b733 18241 Author: Fabio <fabio.ped (a] libero.it> 18242 Date: Mon Dec 15 10:58:23 2008 -0500 18243 18244 Remove duplicate ATI in some device names 18245 18246 from bug 19098 18247 18248 commit 89a8fd5337c77abf1f3166e69ed061d85b5c20f0 18249 Author: Dave Airlie <airlied (a] redhat.com> 18250 Date: Mon Dec 15 14:41:10 2008 +1000 18251 18252 r600: don't idle engine 18253 (cherry picked from commit b77729448e487d37707ce028d8edc59352d1cc2d) 18254 18255 commit 4ce255a9db8027f2471ba50e57724d3e4d5b16ba 18256 Author: Fabio <fabio.ped (a] libero.it> 18257 Date: Fri Dec 12 09:53:33 2008 -0500 18258 18259 Fix formatting in man page 18260 18261 fixes bug 19039 18262 18263 commit 0e99017cb9a6d46b891705df73678c2705439532 18264 Author: Joseph Adams <joeyadams3.14159 (a] gmail.com> 18265 Date: Thu Dec 11 00:57:03 2008 -0500 18266 18267 Add more eMac modes, fix crash in last patch 18268 18269 commit fb46c30d316ab3ec54c54f3aec91d5164070a423 18270 Author: Alex Deucher <alexdeucher (a] gmail.com> 18271 Date: Wed Dec 10 17:16:56 2008 -0500 18272 18273 Add support for Apple eMacs 18274 18275 Thanks to Joseph Adams for helping me sort this out. 18276 Verified on eMac G4/1.0 with radeon 7500 18277 Later eMacs with radeon 9200 or 9600 chips may have 18278 different ddc setups. Need to verify. 18279 18280 commit c0bcea9150ef215fa614733cef9a5b71a55a33bd 18281 Author: Alex Deucher <alexdeucher (a] gmail.com> 18282 Date: Wed Dec 10 11:40:47 2008 -0500 18283 18284 Add another AGP quirk 18285 18286 fixes bug 12544 18287 18288 commit 607b02d6abf78147902a5944bd006897ad95e982 18289 Author: Alex Deucher <alexdeucher (a] gmail.com> 18290 Date: Wed Dec 10 03:54:35 2008 -0500 18291 18292 Add randr 1.3 panning support 18293 18294 commit 24007acbf36bcfd09cb5e74ce4a8d918d393cf34 18295 Author: Benjamin Close <Benjamin.Close (a] clearchain.com> 18296 Date: Wed Dec 10 02:17:56 2008 -0500 18297 18298 [PATCH] xf86misc extension has been removed, update driver to reflect it 18299 18300 commit 69e445f7f2ebd299070e44457b434bbc28b7e05c 18301 Author: Dave Airlie <airlied (a] linux.ie> 18302 Date: Tue Dec 9 18:59:11 2008 +1000 18303 18304 radeon: memset the pll setting function 18305 18306 Thanks to Peter Zijlstra for making me debug this and providing 18307 feedback. 18308 18309 commit c72b732a2f5597cd72bd4ebe207799ae07f2681e 18310 Author: Adrian Friedli <adi (a] koalatux.ch> 18311 Date: Mon Dec 8 10:42:46 2008 -0500 18312 18313 Add another AGP quirk 18314 18315 commit 979ad04d901676cb41aea5cf47a8a2e9c19064dd 18316 Author: Alex Deucher <alexdeucher (a] gmail.com> 18317 Date: Fri Dec 5 17:53:57 2008 -0500 18318 18319 Add missing .TP in man page 18320 18321 commit b89d434011787fb71593395f462d142375783e35 18322 Author: Pierre Ossman <pierre (a] ossman.eu> 18323 Date: Fri Dec 5 17:52:02 2008 -0500 18324 18325 Fix scissor setup for Xv 18326 18327 We need to adjust the scissoring for each clip box, so move the setup 18328 to inside the while loop. 18329 18330 commit 83c1115333cbe763f86422fa5aabf4ab21951e68 18331 Author: Alex Deucher <alexdeucher (a] gmail.com> 18332 Date: Fri Dec 5 11:52:18 2008 -0500 18333 18334 R200: Remove scissor setup from init3d() 18335 18336 We aren't using them, so no need to set them up. 18337 18338 commit 7593d5399cdb2f42ecbcb5193580385b286a9db9 18339 Author: Alex Deucher <alexdeucher (a] gmail.com> 18340 Date: Fri Dec 5 11:18:15 2008 -0500 18341 18342 set coherent mode consistently (enabled for now) 18343 18344 It's set to enabled in the output setup, but then was set to 18345 disabled when teh output properties were setup. Set it up 18346 consistently. 18347 18348 commit ad2579f8898251105a6b36b745afd1ce1dab103e 18349 Author: Dave Airlie <airlied (a] linux.ie> 18350 Date: Fri Dec 5 20:03:53 2008 +1000 18351 18352 radeon: damn you bios, HDMI cannot have DACs. 18353 18354 commit bae30856f2cb3884387ef347327af6ff00b75854 18355 Author: Alex Deucher <alexdeucher (a] gmail.com> 18356 Date: Fri Dec 5 08:15:59 2008 +0100 18357 18358 Make VSync for EXA and Xv configurable 18359 18360 commit 33638d9e388b330e2f4eb4debd05ba09924cf176 18361 Author: Pierre Ossman <pierre (a] ossman.eu> 18362 Date: Thu Dec 4 23:29:31 2008 +0100 18363 18364 Optimise RADEONWaitForVLine 18365 18366 Only avoid the vlines we are rendering to, instead of the entire 18367 screen. This way we don't stall the card for longer than we 18368 absolutely have to. 18369 18370 EXA calls fixed by Alex Deucher. 18371 18372 commit d1690f5cc096e2f735c8b407c370a1c1cd7a8afc 18373 Author: Pierre Ossman <pierre (a] ossman.eu> 18374 Date: Wed Dec 3 21:49:37 2008 +0100 18375 18376 Improve tearing avoidance for Xvideo in two steps 18377 18378 - Fix up VLINE handling to trigger whenever scanout is outside the 18379 visible area. 18380 18381 - Render the video as a scissored triangle as R300+ cannot render a 18382 quad in a single pass. 18383 18384 commit 4d98acbca2e630056bf56cdcd0e23007fded2ced 18385 Author: Alex Deucher <alexdeucher (a] gmail.com> 18386 Date: Wed Jul 16 03:14:06 2008 -0400 18387 18388 First pass at tear-free accel 18389 18390 if the dest pixmap is the front buffer, stall the pipe 18391 until the vline is outside the active area. 18392 For EXA, pick crtc based on the larger mode area; 18393 ideally we'd have one pixmap per crtc. 18394 For Xv, use dst window area to determine crtc. 18395 18396 commit 82d67b70d0fb747cf7b1eb0b64244ae56cb84572 18397 Author: Pierre Ossman <pierre (a] ossman.eu> 18398 Date: Thu Dec 4 15:38:13 2008 -0500 18399 18400 Switch r200 Xv to use rect lists rather than quads to avoid tearing 18401 18402 Patch from Pierre with fixes from me. 18403 18404 commit 9eccd0ad0827a1b8ec9abb2cebf5768d204a5ce9 18405 Author: Alex Deucher <alexdeucher (a] gmail.com> 18406 Date: Thu Dec 4 12:28:12 2008 -0500 18407 18408 Fix up US setup after r3xx bicubic merge 18409 18410 commit e13fba853ba19e6b0f081b9b3d9fa76c38a0f82b 18411 Merge: d29633708 79bbdd984 18412 Author: Alex Deucher <alexdeucher (a] gmail.com> 18413 Date: Thu Dec 4 12:25:29 2008 -0500 18414 18415 Merge branch 'bicubic' of git://git.infradead.org/users/drzeus/xf86-video-ati 18416 18417 commit d29633708b5451f5541e88371d831ae03019a7f5 18418 Author: Dave Airlie <airlied (a] redhat.com> 18419 Date: Thu Dec 4 09:08:48 2008 +1000 18420 18421 radeon: apply atom quirks to object table 18422 18423 commit 941ac0d3ab839dc5aaaef2f3b7f5c7085b2894c5 18424 Author: Alex Deucher <alexdeucher (a] gmail.com> 18425 Date: Wed Dec 3 17:31:25 2008 -0500 18426 18427 Make sure 3D state is emited when switching to 3D IB 18428 18429 - fixes bug 18864 18430 18431 commit 79bbdd984c925e37f5b3db2605339f1a21377fcf 18432 Author: Pierre Ossman <pierre (a] ossman.eu> 18433 Date: Wed Dec 3 20:33:36 2008 +0100 18434 18435 Change the XV_BICUBIC attribute to a tristate, where the third state 18436 automatically turns the filter on and off as appropriate. 18437 18438 commit bbdf821c179242266fd86edf9e688a6be9cf985f 18439 Author: Pierre Ossman <pierre (a] ossman.eu> 18440 Date: Wed Dec 3 19:42:50 2008 +0100 18441 18442 Make the R300 bicubic shader program a bit easier to understand by 18443 having comments that describe the operation, not how it has to be 18444 implemented in the hardware. 18445 18446 commit 8f4b22932c91f1d51fcbf1d1a99d852fcdaafdcc 18447 Author: Pierre Ossman <pierre (a] ossman.eu> 18448 Date: Wed Dec 3 19:31:31 2008 +0100 18449 18450 Now that we have a bicubic code path that messes up the pixel shader, 18451 we need to restore it in the non-bicubic code. 18452 18453 commit 0851daa6e7db77e440188d774ef8b3236146bcb9 18454 Author: Pierre Ossman <pierre (a] ossman.eu> 18455 Date: Wed Dec 3 19:17:36 2008 +0100 18456 18457 Fix macros so that they are safe to use with expressions as parameters. 18458 18459 commit e04bcb0066781f61e97787856261e3380036d304 18460 Author: Pierre Ossman <pierre (a] ossman.eu> 18461 Date: Wed Dec 3 19:16:03 2008 +0100 18462 18463 Fix node setup on R300 bicubic pixel shader. It was backwards and used 18464 the wrong base for sizes. 18465 18466 commit 927aee181f3637bda1b761caf93fbc0652e1207b 18467 Author: Alex Deucher <alexdeucher (a] gmail.com> 18468 Date: Wed Dec 3 01:39:38 2008 -0500 18469 18470 Only flush IB when direct rendering is enabled 18471 18472 commit ca74ebf8813b7fab8d08465ef9600f001c948e94 18473 Author: Alex Deucher <alexdeucher (a] gmail.com> 18474 Date: Tue Dec 2 22:26:21 2008 -0500 18475 18476 move 3d state init after RADEON_SWITCH_TO_3D() 18477 18478 commit fc079c5267baf431bbecee7744e484783d393152 18479 Author: Dave Airlie <airlied (a] redhat.com> 18480 Date: Tue Dec 2 22:11:57 2008 -0500 18481 18482 Don't mix 2D and 3D in the same IB 18483 18484 fix from radeon-gem-cs 18485 18486 commit 678693aabd0b4c1812ab44bf5abb4c8394b254bb 18487 Author: Dave Airlie <airlied (a] redhat.com> 18488 Date: Tue Dec 2 22:02:21 2008 -0500 18489 18490 Stability fixes from radeon-gem-cs 18491 18492 commit f231e4d9b2c683dc16f42cf09484fae926a892c1 18493 Author: Dave Airlie <airlied (a] redhat.com> 18494 Date: Wed Dec 3 11:39:04 2008 +1000 18495 18496 radeon: fix HDMI having DACs in some BIOSes 18497 18498 commit fe73d9a7dfe8ec5c8f1a8dc08e14b4e138aa9276 18499 Author: Bryce Harrington <bryce (a] canonical.com> 18500 Date: Tue Dec 2 19:47:45 2008 -0500 18501 18502 Add another AGP quirk 18503 18504 commit 6e0c48958c6a50a639368e369ce72d813256aee4 18505 Author: Alex Deucher <alexdeucher (a] gmail.com> 18506 Date: Mon Dec 1 18:50:40 2008 -0500 18507 18508 Add quirk for r200 card with the primary dac wired to both ports 18509 18510 refactor load detection setup as a byproduct 18511 should fix bug 18719 18512 18513 commit fa496d7b0397d9be57db90d0860928e9ced73cca 18514 Author: Pierre Ossman <pierre (a] ossman.eu> 18515 Date: Mon Dec 1 08:29:29 2008 +0100 18516 18517 Fix comments for R500 fragment shader to reflect the code. 18518 18519 commit 0d95fca114194664d9ab612c90d3222912cf41ff 18520 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 18521 Date: Sat Aug 30 12:38:43 2008 -0700 18522 18523 Fix indentation on IS_R300_3D Xv code. 18524 18525 commit 36a62b8c47def3f23a0b735875f59d23ed3c095a 18526 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 18527 Date: Sat Aug 30 11:15:20 2008 -0700 18528 18529 r3xx: Various Xv fixes. 18530 18531 commit 01ed5462fad56f5b7bf867041d00443bdf81c653 18532 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 18533 Date: Thu Aug 7 18:08:54 2008 -0700 18534 18535 Switch to Mesa-style 24-bit float packing. 18536 Seems like Mesa's got it right, so we should follow suit. 18537 18538 commit 85bbe889045aa434f71a3eb0ba8153b7d2cc5023 18539 Author: Maciej Cencora <m.cencora (a] gmail.com> 18540 Date: Thu Aug 7 07:06:46 2008 -0700 18541 18542 Enable bicubic Xv on r3xx+. Still some buggies in the actual display, but looks pretty good. 18543 18544 commit bf6263cff50bb85c19ff6dfebb739333fa8aca18 18545 Author: Maciej Cencora <m.cencora (a] gmail.com> 18546 Date: Thu Aug 7 07:00:36 2008 -0700 18547 18548 Add r3xx-specific bicubic filtering code. FP, RS, mostly, everything else is shared with r5xx. 18549 18550 commit da021c36bbdf3bca31ee50ebe01cdb9495c09b36 18551 Author: Dave Airlie <airlied (a] linux.ie> 18552 Date: Sun Nov 30 18:44:24 2008 +1000 18553 18554 radeon_drm.h: remove kernel defines 18555 18556 remove these we should update from an installed kernel copy not from the 18557 original kernel copy. 18558 18559 commit 783cdb7374941bb1d2b63eea375fbf2f1b808cc3 18560 Author: Dave Airlie <airlied (a] redhat.com> 18561 Date: Thu Nov 27 15:29:06 2008 +1000 18562 18563 r100/r200: EXA misrenders dst == a8 + dst alpha use. 18564 18565 Fedora BZ#469556 18566 18567 I've played with various dst and texture formats and rewritten large 18568 parts of this to no avail, so I'm punting for now. 18569 18570 commit 2d7cfc421713286edd5121b56999246f9ae70286 18571 Author: Eygene Ryabinkin <rea-fbsd (a] codelabs.ru> 18572 Date: Tue Nov 25 13:25:33 2008 -0500 18573 18574 Catch unsubstituted macro for PCI region base access 18575 18576 also fix up some warnings. 18577 18578 commit af7690b9c300d37c3a0c51e27e0ad2ca009224a2 18579 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 18580 Date: Tue Nov 25 02:57:30 2008 -0500 18581 18582 [PATCH] radeon: Fix PCI usage of 32-bit driver on 64-bit platform 18583 18584 The radeon driver is storing PCI addresses in unsigned long's which 18585 won't work well on 32-bit platforms with 64-bit physical address space 18586 such as PowerPC 4xx. This fixes it by using unsigned long long instead. 18587 18588 commit 065938617c0feab17f4274a5350de02a692ba065 18589 Author: Ramon van der Stelt <rvanderstelt (a] xs4all.nl> 18590 Date: Tue Nov 25 02:23:46 2008 -0500 18591 18592 Interlaced mode fixes 18593 18594 see bug 12626 18595 18596 commit 36a7dc6ea1e1929e986ab1159497c71521cb2f10 18597 Author: Bryce Harrington <bryce (a] canonical.com> 18598 Date: Tue Nov 25 01:04:37 2008 -0500 18599 18600 Additional AGP quirks 18601 18602 bug 18693 18603 18604 commit 2acb8e4ed220fccb21b22a53c5142ab3a35d32ef 18605 Author: Alex Deucher <alexdeucher (a] gmail.com> 18606 Date: Tue Nov 25 15:07:09 2008 +1000 18607 18608 radeon: fixup shared DDC lines for some rv610 cards. 18609 18610 Some cards share DDC between a HDMI and VGA output, and expect the 18611 driver to use load detect or EDID to figure it out. 18612 18613 airlied- shipped in RHEL5 18614 18615 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 18616 18617 commit b01e35168de40d192fd7a9ce6884b9c7419afbd4 18618 Author: Alex Deucher <alexdeucher (a] gmail.com> 18619 Date: Mon Nov 24 21:06:42 2008 -0500 18620 18621 Fix up posted logic 18622 18623 noticed by benh on IRC 18624 18625 commit d7a03e93372b17d642b42f5490464616ae7c99c1 18626 Author: Alex Deucher <alexdeucher (a] gmail.com> 18627 Date: Mon Nov 24 15:52:05 2008 -0500 18628 18629 Get hardcoded edid from the bios for servers that support it 18630 18631 should fix bug 11300 18632 18633 commit 3858e31fa9d77d52a5cabb02f53f06385eab40f7 18634 Author: Alex Deucher <alexdeucher (a] gmail.com> 18635 Date: Mon Nov 24 14:39:41 2008 -0500 18636 18637 Don't support rotation is accel is not available 18638 18639 Fixes bug 18068 18640 18641 commit 3d27876d91ef69d71f17b6cf8c07c26bf12857b4 18642 Author: Dave Airlie <airlied (a] redhat.com> 18643 Date: Mon Nov 10 15:47:13 2008 +1000 18644 18645 radeon: always align dstPitch to 64 bytes for textured video 18646 18647 bz# 18454 18648 18649 commit 6c8b352264540ccc4622a7c0b704ce452dcd707d 18650 Author: Dave Airlie <airlied (a] redhat.com> 18651 Date: Wed Oct 29 15:53:57 2008 +1000 18652 18653 pciid: add HD3300 - 790GX chipset. 18654 18655 RH BZ (#466706) 18656 18657 commit 902eaf768142c6c7dcc487e10775027b84cd1f9a 18658 Author: Alex Deucher <alexdeucher (a] gmail.com> 18659 Date: Thu Nov 6 15:46:43 2008 -0500 18660 18661 Check for LVDS on all IGP chips 18662 18663 - fixes bug 18395 18664 18665 commit 44abaa0ea8e5e3bbe833fa73e6d011618bb6d7e7 18666 Author: Adam Jackson <ajax (a] redhat.com> 18667 Date: Mon Nov 3 15:09:28 2008 -0500 18668 18669 Fix HDMI output setup on DIG1/DIG2. 18670 18671 The only reliable way to detect an HDMI sink is to check for the 18672 appropriate EDID extension block, so this will only work right on 18673 servers with xf86DoEEDID(). RANDR 1.2 will call this internally for 18674 us if it exists, but pre-1.6 servers are out of luck. 18675 18676 commit 937b7ac2a259cf504a19dcf62a58b1db1afb8eb9 18677 Author: Bryce Harrington <bryce (a] canonical.com> 18678 Date: Wed Oct 29 01:31:50 2008 -0400 18679 18680 Add AGP quirk table 18681 18682 - set known good default agp modes for problematic chip/bridge combos 18683 - should fix bug 17360, others 18684 18685 commit 67bc2ae7f227a2afedc8e699cf0458fa7960f301 18686 Author: Julien Cristau <jcristau (a] debian.org> 18687 Date: Sun Oct 26 16:05:04 2008 +0100 18688 18689 Bug#7148: Fix manpage formatting 18690 18691 commit 98fdd78a1229584b922f816f331a45bfc178c961 18692 Author: Dave Airlie <airlied (a] linux.ie> 18693 Date: Tue Oct 21 18:36:22 2008 +1000 18694 18695 atombios: oops typo 18696 18697 commit 763dff6c117a10b6a57ba1854c5b798359789e44 18698 Author: Dave Airlie <airlied (a] linux.ie> 18699 Date: Tue Oct 21 18:35:44 2008 +1000 18700 18701 atombios: set missing pixel clock 18702 18703 commit 435cf7da68186f2601c4b888296117d4f652c625 18704 Author: Alex Deucher <alexdeucher (a] gmail.com> 18705 Date: Wed Oct 15 10:52:14 2008 -0400 18706 18707 Add LCD dithering quirk for macbook pro 18708 18709 fixes bug 17897. 18710 18711 commit c0e6cb6d0eeef8f2ea60d840e1cd668fa92cd7f9 18712 Author: Alex Deucher <alexdeucher (a] gmail.com> 18713 Date: Fri Oct 10 11:09:27 2008 -0400 18714 18715 Fix grey level dithering setup 18716 18717 commit 9eb2b6cd28224f0cbbabf3df451696ae1a845b9c 18718 Author: Dave Airlie <airlied (a] redhat.com> 18719 Date: Fri Oct 10 16:18:05 2008 +1000 18720 18721 radeon: hopefully fix plls for 30" monitors. 18722 18723 Only do the HIGH trick on the DCE3 cards. Tested on rv610/630/770. 18724 18725 commit 0975e007ec7933acf9cd8ad2f281313ef8d32ac9 18726 Author: Dave Airlie <airlied (a] redhat.com> 18727 Date: Thu Oct 9 17:16:28 2008 +1100 18728 18729 radeon: rename radeon_memory to radeon_legacy_memory. 18730 18731 This is a setup patch for the kms/memory management code. 18732 18733 commit d744dc013cf5ca3141c0fdf55e36d713fe9f53ba 18734 Author: Adam Jackson <ajax (a] redhat.com> 18735 Date: Wed Oct 8 15:50:25 2008 -0400 18736 18737 Fix connector table mapping for lvtma/kaleidoscope/uniphy. 18738 18739 Fixes DPMS on LVDS. 18740 18741 commit dc795ba66a8c0eb2c3fdb86bd42dfc17e0aa3b6e 18742 Author: Alex Deucher <alexdeucher (a] gmail.com> 18743 Date: Tue Oct 7 12:32:47 2008 -0400 18744 18745 Fixups from last DFP/LCD refactor 18746 18747 - fix DFP1 -> LCD1 18748 - use _INDEX rather than _SUPPORT defs 18749 18750 commit 4a9e8d9fe24278b56aef51677fd35289d01584a7 18751 Author: Alex Deucher <alexdeucher (a] gmail.com> 18752 Date: Mon Oct 6 18:00:08 2008 -0400 18753 18754 set grey levels correctly for temporal dithering 18755 18756 commit 5708624d1151877dd66a462d4a75fed6774604a7 18757 Author: Alex Deucher <alexdeucher (a] gmail.com> 18758 Date: Mon Oct 6 15:11:56 2008 -0400 18759 18760 Turn coherent mode off by default on DVI 18761 18762 commit a5c5ce96279d01eb519bfb92b94c06a58acb7f07 18763 Author: Alex Deucher <alexdeucher (a] gmail.com> 18764 Date: Mon Oct 6 15:07:31 2008 -0400 18765 18766 Refactor atom LCD/DFP output setup 18767 18768 Simplify the code, properly handle dithering, coherent mode, encoding, etc. 18769 Should fix bug 17897. 18770 18771 commit f9826a5694b7adb6920eb5bdf45d840d8fb14d53 18772 Author: Alex Deucher <alexdeucher (a] gmail.com> 18773 Date: Fri Oct 3 15:24:12 2008 -0400 18774 18775 Add support for DDC via atom commands for RV410 18776 18777 The atom calls use the hw i2c engine for DDC. For some 18778 reason, sw i2c doesn't seem to work on the VGA GPIO on 18779 RV410 chips, so we use atom in that case. 18780 18781 This fixes the longstanding VGA DDC problems on RV410/M26 18782 chips. 18783 18784 commit 23c17c0121d43c2fd11bb5cfbaa4b02abac2b16d 18785 Author: Alex Deucher <alexdeucher (a] gmail.com> 18786 Date: Fri Oct 3 14:30:33 2008 -0400 18787 18788 man page updates 18789 18790 Update render accel and page flipping information. 18791 fixes bug 17887. 18792 18793 commit 4ac6e499089aec3c79af5dbede5fe631b105ce3d 18794 Author: Alex Deucher <alexdeucher (a] gmail.com> 18795 Date: Wed Oct 1 18:53:15 2008 -0400 18796 18797 proper fix for initial rotation 18798 18799 The initial modeset was setting the MC regs, so just set them 18800 rather than the full modeset. 18801 18802 commit 346228b20c69a965c9e7d67157f42c1d36a5b6c3 18803 Author: Alex Deucher <alexdeucher (a] gmail.com> 18804 Date: Wed Oct 1 14:38:27 2008 -0400 18805 18806 Workaround to make initial rotation work 18807 18808 Acceleration needs to be initialized before setting the mode 18809 for initial rotation to work. Changing the order in RADEONScreenInit() 18810 doesn't work because RADEONDRIKernelInit() hangs in the ioctl to 18811 initialize the CP if the mode is not set prior (even if the heads 18812 are turned off after setting the mode). The workaround is to set the 18813 modes twice, once before accel and once after. 18814 18815 This needs to be looked into further. 18816 18817 commit c359c2a31caf9f75b9fc6b6bcbc3e9dc1fe404ba 18818 Author: Michel Dnzer <michel (a] tungstengraphics.com> 18819 Date: Tue Sep 30 11:04:54 2008 +0200 18820 18821 Pass base offset into RADEONDisplayVideo() explicitly. 18822 18823 This prevents the base offset from flipping back and forth if double buffering 18824 is enabled and the two buffers start on different sides of a 4 MB boundary. 18825 18826 Should fix http://bugs.freedesktop.org/show_bug.cgi?id=17254 ; please don't 18827 reopen that bug report without attaching a log file captured after reproducing 18828 the problem with this fix. 18829 18830 commit ba21d5ae7fbbf5828c44c41655aff8b182453e2e 18831 Author: Michel Dnzer <michel (a] tungstengraphics.com> 18832 Date: Tue Sep 30 11:03:04 2008 +0200 18833 18834 Update GEN_INT_CNTL register value after calling DRM modeset ioctl. 18835 18836 Fixes sync-to-vblank hangs after mode switch with DRM vblank-rework. 18837 18838 commit 3110646daf05fa09faaa70066cc8c3db1bb07cec 18839 Author: Michel Dnzer <michel (a] tungstengraphics.com> 18840 Date: Sun Sep 14 11:46:36 2008 +0200 18841 18842 Fix -Werror build. 18843 18844 commit d82f2938f69402c641a1c8362fdc513419b27659 18845 Author: Alex Deucher <alexdeucher (a] gmail.com> 18846 Date: Fri Sep 26 13:51:24 2008 -0400 18847 18848 atombios updates from upstream 18849 18850 commit d0d58b39e49c5ce00bc8bd12f721ad8c7f554b91 18851 Author: Alex Deucher <alexdeucher (a] gmail.com> 18852 Date: Wed Sep 24 11:54:34 2008 -0400 18853 18854 radeon: Further pll adjustments 18855 18856 commit d100a6af8f828eb94f8ba6e8a96c24389b5cf46f 18857 Author: Alex Deucher <alexdeucher (a] gmail.com> 18858 Date: Fri Sep 19 14:04:59 2008 -0400 18859 18860 cleanup macbook quirk 18861 18862 commit 1218e38c75351a20f7414ba6fcb3545689339442 18863 Author: Alex Deucher <alexdeucher (a] gmail.com> 18864 Date: Fri Sep 19 13:52:07 2008 -0400 18865 18866 more macbook fixes 18867 18868 commit 83e1f132965deadedda7a622184eb400cd4c120c 18869 Author: Alex Deucher <alexdeucher (a] gmail.com> 18870 Date: Fri Sep 19 09:36:59 2008 -0400 18871 18872 Fix copy/paste error in the last commit 18873 18874 DVI-D -> DVI-I 18875 18876 commit 4af825f7b13cba8c3b204e72a07cab5009331240 18877 Author: Alex Deucher <alexdeucher (a] gmail.com> 18878 Date: Fri Sep 19 00:29:20 2008 -0400 18879 18880 radeon: Add quirk for macbook 18881 18882 - should fix bug 17625 18883 18884 commit 38aa450fce6a91b35c191fd07112490a62cc29c6 18885 Author: Alex Deucher <alexdeucher (a] gmail.com> 18886 Date: Tue Aug 26 21:50:23 2008 -0400 18887 18888 fix some fallout from the common allocator 18889 18890 should fix bug 17317 18891 18892 commit a3cc1d7a421456186024c5c069e403d374a0f0b9 18893 Author: Dave Airlie <airlied (a] linux.ie> 18894 Date: Tue Aug 26 08:03:20 2008 +1000 18895 18896 radeon: fix powerpc build 18897 18898 commit 6cebfe257f7ddad855ee743e4eb899bd6fac7f46 18899 Author: Alex Deucher <alexdeucher (a] gmail.com> 18900 Date: Fri Jul 11 19:32:06 2008 -0400 18901 18902 Switch EXA path back to static cursor allocation 18903 18904 pre-AVIVO cards have address limits for the cursor offset 18905 18906 commit 4dff54a3c8d7c9f2d6ec50354ff0b92f1b7fcbdf 18907 Author: Alex Deucher <alexdeucher (a] gmail.com> 18908 Date: Fri Jul 11 17:28:03 2008 -0400 18909 18910 Switch cursors over to generic allocator 18911 18912 commit 5b1978a4796bcc31ac2f01d303dc8f8f44323025 18913 Author: Alex Deucher <alexdeucher (a] gmail.com> 18914 Date: Mon Aug 25 10:05:28 2008 -0400 18915 18916 Bicubic fixes from the last cherry-pick 18917 18918 commit 7daba77ded1c718e93ae8c372a39a6e85228d513 18919 Author: Alex Deucher <alexdeucher (a] gmail.com> 18920 Date: Fri Jul 11 02:02:38 2008 -0400 18921 18922 Convert randr, Xv to a common allocator 18923 18924 commit 610fe1a937da78f4ac813ac919c158dde8f42442 18925 Author: Alex Deucher <alexdeucher (a] gmail.com> 18926 Date: Thu Jul 10 22:47:35 2008 -0400 18927 18928 Switch to using radeon_drm.h from the drm 18929 18930 modelled after Matthias' similar rhd change 18931 18932 commit a6561f2ec673b38907f7181235386f32e60c32ba 18933 Author: Alex Deucher <alexdeucher (a] gmail.com> 18934 Date: Thu Jul 10 22:31:13 2008 -0400 18935 18936 Move DRI to a separate struct 18937 18938 based on Matthias' similar work in rhd 18939 18940 commit 75ef8dc214715d3c5c50996b293933842903ba65 18941 Author: Egbert Eich <eich (a] freedesktop.org> 18942 Date: Thu Jul 10 21:49:55 2008 -0400 18943 18944 Cleanups from rhd port 18945 18946 - remove unused vars 18947 - remove static exa render vars 18948 18949 commit 71ad140fa11f3a504c38d6bddf40e3a3c0a20e60 18950 Author: Alex Deucher <alexdeucher (a] gmail.com> 18951 Date: Thu Jul 10 21:24:16 2008 -0400 18952 18953 Move accel state to a separate struct 18954 18955 commit 5b2e095c31b88d8495a4f86e6cb46b49fa4acd65 18956 Author: Alex Deucher <alexdeucher (a] gmail.com> 18957 Date: Thu Jul 10 20:07:44 2008 -0400 18958 18959 Move CP into a separate struct 18960 18961 commit 2145309230e36aee7758bd244deb1e72ada0c065 18962 Author: Alex Deucher <alexdeucher (a] gmail.com> 18963 Date: Mon Aug 25 08:47:59 2008 -0400 18964 18965 match textured video macro names with other accel code 18966 18967 OUT_VIDEO_REG() -> OUT_ACCEL_REG() 18968 etc. 18969 18970 commit 01daef0f095fbbaee701d5fe97f3dd7838b5f915 18971 Author: Alex Deucher <alexdeucher (a] gmail.com> 18972 Date: Mon Aug 25 08:26:16 2008 -0400 18973 18974 Additional cleanups and re-arragement following bicubic merge 18975 18976 commit 0a51d08c24af040fe48690662b1a912acad51700 18977 Author: Alex Deucher <alexdeucher (a] gmail.com> 18978 Date: Mon Aug 25 06:43:14 2008 -0400 18979 18980 Whitespace cleanup from bucubic merge 18981 18982 commit c0170fef510b53a6ca2c6ea7a99119235229c929 18983 Author: Maciej Cencora <m.cencora (a] gmail.com> 18984 Date: Thu Aug 7 06:53:39 2008 -0700 18985 18986 Add needed FP registers, etc. for r3xx bicubic Xv. 18987 18988 commit d9c38326cf70f57ab777ffdf9520b8cdea9d9cb6 18989 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 18990 Date: Tue Aug 5 18:45:38 2008 +0200 18991 18992 Fix typos. 18993 18994 commit d38ceba62aa5cee76baa342ce7719a983a79f09e 18995 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 18996 Date: Tue Aug 5 16:21:57 2008 +0200 18997 18998 Fix bicubic fp calculation. 18999 19000 commit 6f9c7fde75edc0399559f975db9a6c4dc22714ae 19001 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 19002 Date: Tue Aug 5 15:38:42 2008 +0200 19003 19004 Fixed bicubic fragment program comments. 19005 19006 commit 48b09ca40ccb28b5584069316fd38786a78c1dd3 19007 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19008 Date: Sat Aug 2 02:36:21 2008 -0700 19009 19010 Switch from 32-bit floats to 16-bit half-floats. 19011 Massive bandwidth savings, or so I'm told. Yay? 19012 19013 commit ebbb7fb634fcadf28ff99c1df2c3db89fd56932d 19014 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19015 Date: Sat Aug 2 01:45:43 2008 -0700 19016 19017 Change floats to uint32_t hex. 19018 Useful for moving to 16-bit half-floats. 19019 19020 commit 3b46162d8a90be3524cb513d42b9ad3d0bea45f5 19021 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 19022 Date: Fri Aug 1 15:24:01 2008 +0200 19023 19024 Move some ALU instructions after the TEX instructions, so we can do something usefull while we are waiting for the texture values. 19025 19026 commit 2ecdec4bafc97212dde4d6908ee4ccf618adc0e1 19027 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 19028 Date: Fri Aug 1 04:03:12 2008 +0200 19029 19030 Another uneeded SEM_AQUIRE. 19031 19032 commit e93b5d1b80d6203f63543b7b678e2f1d9221b5b3 19033 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 19034 Date: Fri Aug 1 03:00:26 2008 +0200 19035 19036 Smarter usage of the texture semaphore. 19037 19038 commit 286f7aa18f19bd21b263701adab38b736dbeda0f 19039 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19040 Date: Thu Jul 31 17:01:44 2008 -0700 19041 19042 Enable bicubic filtering for all r5xx HW. 19043 19044 commit 242aa4f630b4c60aefa3c12dc459a4d4d0b334a0 19045 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 19046 Date: Fri Aug 1 01:11:28 2008 +0200 19047 19048 Remove one constant. 19049 19050 commit 413eacb0538977b0b3c92df074d40510f4539abc 19051 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 19052 Date: Fri Aug 1 00:52:09 2008 +0200 19053 19054 Heavy optimizations. 19055 19056 commit a0c4a949cb49e5ac1e857aef08a8742b9f7b49da 19057 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 19058 Date: Thu Jul 31 21:19:18 2008 +0200 19059 19060 Remove uneeded negations. 19061 19062 commit abb2b2e757d92591ab9277824b9d9746ba98f875 19063 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 19064 Date: Thu Jul 31 20:31:03 2008 +0200 19065 19066 Implement LPR in one instruction. 19067 19068 commit c370b74bec13194573348583c38adf710b880e79 19069 Author: Dennis Kasprzyk <onestone (a] opencompositing.org> 19070 Date: Thu Jul 31 19:50:49 2008 +0200 19071 19072 Set helper texture filter correctly. 19073 19074 commit 5e85d5a5d0c50b29086ec0c219c8b52d25dbc2e9 19075 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19076 Date: Wed Jul 30 19:45:15 2008 -0700 19077 19078 Update bicubic tables. 19079 Now including the 2048x1 texture as well. 19080 19081 commit 17e5e9573e59c3d82d51c261b9c5005f6aec7d43 19082 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19083 Date: Wed Jul 30 11:57:25 2008 -0700 19084 19085 Force R580-only for bicubic. 19086 Initial reading of docs suggest RV560 and RV570 can't handle it, 19087 but they're welcome to test. 19088 19089 commit 83b52473d0e7102265918b07be053fcda17a14b4 19090 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19091 Date: Wed Jul 30 01:03:57 2008 -0700 19092 19093 Bump bicubic cutoff to R580. 19094 RV530 chipsets can't handle fullscreen bicubic... 19095 19096 commit 69a4998d1286bcdd7bfe874cd5628bc1cc232bae 19097 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19098 Date: Wed Jul 30 01:03:46 2008 -0700 19099 19100 Oops, made a mistake with vertices. 19101 19102 commit a4ec30a677906ec2ff9824c7ddca586655f6d1a8 19103 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19104 Date: Wed Jul 30 00:54:32 2008 -0700 19105 19106 Merge upstream changes to vertices, and also add Xv attributes for textured video, including bicubic filtering. 19107 19108 commit f3b81c7582aed307fa44e134ee161cd8a3158657 19109 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19110 Date: Mon Jul 28 19:50:10 2008 -0700 19111 19112 Fix constants. 19113 19114 commit 7dbb7023ba023ec1a38be63af9c9f49e40222b7b 19115 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19116 Date: Mon Jul 28 17:30:41 2008 -0700 19117 19118 Finally got the fragment program fully working for bicubic filtering on r5xx. 19119 19120 commit 0e4dd73b9ebc6f608eeff945b4d463a00c02e07c 19121 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19122 Date: Mon Jul 28 12:16:50 2008 -0700 19123 19124 More bicubic FP buggies. 19125 19126 commit a87647e0c27e0950f4d0d8203a1242a994ad3419 19127 Author: Dennis Kasprzyk <onestone (a] compiz-fusion.org> 19128 Date: Sun Jul 27 10:43:01 2008 -0700 19129 19130 Fix texture size, texture filter, vertex offsets, etc. 19131 19132 commit 232aa3e943fef4c4037b255c3b64a0aaff90ab5c 19133 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19134 Date: Sat Jul 26 10:56:56 2008 -0700 19135 19136 Make vertices emit properly. 19137 *bangs head against wall* 19138 19139 commit 8c84f67b93d926095633830aa8d95930a48b1c7b 19140 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19141 Date: Sun Jul 20 14:25:44 2008 -0700 19142 19143 Try to get tex coords from the VAP to RS properly when bicubic filtering is enabled. 19144 I'm soo bad at this... :c 19145 19146 commit a4a7d5f5967c51c394229de5eccaec44cfec8f50 19147 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19148 Date: Sat Jul 19 13:02:02 2008 -0700 19149 19150 Upload pixel shader to card for r5xx. This was ridiculous. Also it doesn't work yet... 19151 19152 commit 20c1db2d7c110ab5c1117a57b169baa1ab070518 19153 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19154 Date: Fri Jul 18 13:48:08 2008 -0700 19155 19156 Upload bicubic filter to card. This was a LOT easier than I had feared, to be honest. 19157 19158 commit a760e628134c6d7d42ec3c98118b6e5f6fcd3e7f 19159 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19160 Date: Fri Jul 18 12:21:20 2008 -0700 19161 19162 Allocate memory for the bicubic filter texture. 19163 19164 commit b6c9e2bb5365de82315c6814f915e57b0c4fa444 19165 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19166 Date: Fri Jul 18 11:24:14 2008 -0700 19167 19168 Fixed typos in the bicubic texture tables. Whoops, looks like I'm still asleep. 19169 19170 commit ca51f4f37e1dbf53bf7ffc0e8f612e9609e11209 19171 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 19172 Date: Fri Jul 18 11:06:34 2008 -0700 19173 19174 Add bicubic texture table, as well as the script used to (re)generate it. To regenerate, just run "python bicubic_table.py > bicubic_table.h". 19175 19176 commit 1cf7a5494fa94e8d9f30f9b2905dfbe6d4faa445 19177 Author: Bryce Harrington <bryce (a] bryceharrington.org> 19178 Date: Wed Aug 20 09:46:59 2008 -0400 19179 19180 radeon: Fix pasto in connector table setup for vga powerbooks 19181 19182 fixes bug 17214 19183 19184 commit a55e85f742d1334bf88e4681e553f025d2de38df 19185 Author: Michel Dnzer <michel (a] tungstengraphics.com> 19186 Date: Thu Aug 14 15:21:51 2008 -0400 19187 19188 Make sure video offerlay offsets don't exceed the hardware limit of 128 MB. 19189 19190 Always set the overlay base address such that the buffer offsets are as small 19191 as possible. This could still break in theory if the buffers were more than 19192 128 MB apart, but in reality this can't happen ATM because we always allocate 19193 a single memory area for all buffers. 19194 19195 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=16845 . 19196 19197 commit 92ee21df344a989778e37369c7beb3904a00ead6 19198 Author: Alex Deucher <alexdeucher (a] gmail.com> 19199 Date: Thu Aug 14 14:49:45 2008 -0400 19200 19201 PLL adjustments 19202 19203 Seems higher dotclocks prefer a higher FB div. 19204 Someone with a lot of should try and find out where 19205 the div sweet spots are for various dotclock ranges. 19206 fixes bug 17125 19207 19208 commit 0d5e0347af4322713075193154b8a348de4a0b52 19209 Author: Alex Deucher <alexdeucher (a] gmail.com> 19210 Date: Wed Aug 13 14:17:34 2008 -0400 19211 19212 Remove reset of 3D scissor registers when using the CP in the ddx 19213 19214 They should only affect 3D and init3d() should take care of that case 19215 noticed by libv on IRC. 19216 19217 commit 7e456e9c427d48151b166bcac08c3e8a8b180f01 19218 Author: Dave Airlie <airlied (a] linux.ie> 19219 Date: Tue Aug 12 08:18:30 2008 +1000 19220 19221 radeon: add 0x9441 for hd 4870 x2 19222 19223 thanks to Michael Larabel for testing 19224 19225 commit 4dbdeea7c9316575fba26b41fd347452e42cdcf2 19226 Author: Alex Deucher <alexdeucher (a] gmail.com> 19227 Date: Fri Aug 8 17:39:48 2008 -0400 19228 19229 Further cleanup and unification of i2c code 19230 19231 - unify the ddc and i2c code 19232 - add gpio mask support for legacy chips 19233 - remove the magic gpio dance for ancient monitors 19234 (if you have an ancient monitor that ddc stops 19235 working on let me know). This should speed up DDC 19236 on legacy chips. 19237 19238 -- radeon sw gpio i2c -- 19239 4 sets of gpio registers for clk and data and corresponding bit masks 19240 mask - locks the gpio for use by sw 19241 get - reads the value off the gpio pad 19242 put - sets the gpio direction to output 19243 a - "other stuff" On legacy chips you clear them if you want 19244 to use a gpio for i2c. In some cases they are used for the 19245 output value when the gpio in the output state. 19246 19247 commit 268c848130ec1770bb645a74197b6aca7fc95abc 19248 Author: Alex Deucher <alexdeucher (a] gmail.com> 19249 Date: Fri Aug 8 15:50:07 2008 -0400 19250 19251 Fix VT switching on M6 chips 19252 19253 Some M6 chips have a faulty MEM_SIZE register that in 19254 some cases reports 0 on 8 MB cards. On EnterVT we check 19255 the MEM_SIZE reg as a check to see if the card is posted or 19256 not. Since this reg returns 0, the driver attempts to post 19257 the card which can lead to a hang. Switch this to check if 19258 either crtc is active as is done in the bios init code. 19259 fixes bug 13994 19260 19261 commit 33f88f7fc90d9d93fdcbba9ad59dd70a6596bc3f 19262 Author: Alex Deucher <alexdeucher (a] gmail.com> 19263 Date: Thu Aug 7 17:57:07 2008 -0400 19264 19265 RS4xx DDC fixes take 3 19266 19267 commit df0d1ef53100f0a19c5b5fdc349f5186c8d9bd87 19268 Author: Alex Deucher <alexdeucher (a] gmail.com> 19269 Date: Wed Aug 6 19:13:56 2008 -0400 19270 19271 RS4xx: Fix up ddc gpio 19272 19273 - I mixed up the regs previously 19274 19275 commit 18429390440a829fb24ed3afd99ccf8278138496 19276 Author: Alex Deucher <alexdeucher (a] gmail.com> 19277 Date: Wed Aug 6 00:35:15 2008 -0400 19278 19279 Remove un-needed dac check for single crtc cards 19280 19281 Connector tables should be setup properly already and this 19282 check breaks systems with 2 DACs and 1 crtc like the es1000. 19283 19284 commit 001c535687d8588873037ee5363d0a709f44b418 19285 Author: Alex Deucher <alexdeucher (a] gmail.com> 19286 Date: Tue Aug 5 17:14:48 2008 -0400 19287 19288 IGP: fix typo in IGP quirk handling 19289 19290 commit eb65ddf70d182b6457e1ef5ebb820456039e8f6d 19291 Author: Alex Deucher <alexdeucher (a] gmail.com> 19292 Date: Tue Aug 5 15:01:33 2008 -0400 19293 19294 IGP: attempt to get external TMDS working 19295 19296 RS4xx chips have a tmds init table in the mobile info table 19297 19298 commit 8b8990917809b9a35c6e9c1b9e3b12ff81c6dbb3 19299 Author: Alex Deucher <alexdeucher (a] gmail.com> 19300 Date: Tue Aug 5 12:16:06 2008 -0400 19301 19302 IGP: Updated quirk handling 19303 19304 - Hopefully finally fix DDC on RS4xx chips 19305 - RS3xx/RS4xx do not have internal TMDS 19306 - general quirk handling cleanup 19307 - we don't currently support mm gpio for external TMDS 19308 so return false. 19309 19310 commit cb0deba5412a575d36f2f99377120b123506c946 19311 Author: Calvin Fong <hoiwai930 (a] gmail.com> 19312 Date: Mon Aug 4 02:04:15 2008 -0400 19313 19314 IGP: add support for NTSC tv-out on legacy IGP chips 19315 19316 Patch from Calvin with some cleanups from me. 19317 No support for PAL yet. 19318 19319 commit 942b18aca91819fa65d853cd15ffd1cd720cbd68 19320 Author: Dave Airlie <airlied (a] redhat.com> 19321 Date: Mon Aug 4 14:38:21 2008 +1000 19322 19323 radeon: make r600 use i2c table lookup for ddc. 19324 19325 This may fix DDC on rv770 cards. 19326 19327 commit 1f3eee3682f3598a303c9c3accfbe01b245cacf9 19328 Author: Alex Deucher <alexdeucher (a] gmail.com> 19329 Date: Tue Jul 29 20:29:32 2008 -0400 19330 19331 Change prim types for exa and textured video to help avoid tearing 19332 19333 - r1xx - switch from tri fan to rect list 19334 - r2xx/r3xx/r4xx/r5xx - switch from tri fan to quad list 19335 19336 commit 5b5441f8cc119db0d1e03dd35bd06015a26270dd 19337 Author: Brice Goglin <bgoglin (a] debian.org> 19338 Date: Tue Jul 29 23:31:14 2008 +0200 19339 19340 Increase default GART size to 32MB on >=R300 19341 19342 commit 0a505297f09aefb1b4432176a263bfdf6f256f77 19343 Author: Alex Deucher <alexdeucher (a] gmail.com> 19344 Date: Mon Jul 28 17:12:41 2008 -0400 19345 19346 Fix error in driver connector table for powerbook w/ vga 19347 19348 commit d5799ac53c3e1b2ca1da97780b171a44060c3aad 19349 Author: Alex Deucher <alexdeucher (a] gmail.com> 19350 Date: Mon Jul 28 11:09:10 2008 -0400 19351 19352 Add quirk for oem x300 card 19353 19354 - debian bug 492457 19355 19356 commit 64dbadddcf6d069c0472f37afeab89d3e31e937d 19357 Author: Owain Gordon Ainsworth <oga (a] openbsd.org> 19358 Date: Mon Jul 28 10:40:58 2008 +1000 19359 19360 radeon: don't call ioctl unless DRI is initialised 19361 19362 commit 1c5858484da4fb1c9bc3ac3b4d7a97863ab99730 19363 Author: Alex Deucher <alexdeucher (a] gmail.com> 19364 Date: Mon Jul 21 23:47:45 2008 -0400 19365 19366 First pass at InitDispBandwidth() for AVIVO chips 19367 19368 - support for LB allocation 19369 - MC priority bumps for display1/2 on RV515 variants and RS690 19370 If you are having display underflow problems (flickering on sides of 19371 screen in high res modes, etc.) on RV515 or RS690 boards, try setting: 19372 Option "DisplayPriority" "HIGH" in your config. 19373 - still no support for full display watermark programming yet 19374 19375 Something similar might be useful in rhd as well. 19376 19377 commit b0378bb145c8a915c943bef7d17f2cdecfccc891 19378 Author: Alex Deucher <alexdeucher (a] gmail.com> 19379 Date: Mon Jul 21 13:47:09 2008 -0400 19380 19381 Interlaced mode fixups for AVIVO chips 19382 19383 commit c18fad622a3c4f9572051120d83af68b625b5686 19384 Author: Alex Deucher <alexdeucher (a] gmail.com> 19385 Date: Mon Jul 21 10:36:48 2008 -0400 19386 19387 Clear display priority bits before resetting them 19388 19389 commit dc231ff8e063313d2bcf5acccad67a9f8a7e3314 19390 Author: Alex Deucher <alexdeucher (a] gmail.com> 19391 Date: Mon Jul 21 10:30:41 2008 -0400 19392 19393 Clean up legacy display watermark setup 19394 19395 - makes crtc1 and crtc2 watermark setup independant. 19396 - fixes the case where only crtc2 is active 19397 19398 commit f9034214f070fe3054cd9cefd7034252de234f38 19399 Author: Michel Dnzer <michel (a] tungstengraphics.com> 19400 Date: Mon Jul 21 09:09:02 2008 +0200 19401 19402 Call DRM modeset ioctl after the IRQ has been initialized. 19403 19404 This lets the DRM know it can safely disable the vblank interrupts. 19405 19406 commit ecb6347a3d7071890600c98c2addef3a8ca260ee 19407 Author: Alex Deucher <alexdeucher (a] gmail.com> 19408 Date: Sat Jul 19 11:34:16 2008 -0400 19409 19410 Add oem quirk for external tmds setup on Dell Inspiron 8600 19411 19412 Noticed by fnord42 on IRC. 19413 19414 commit df53d12a06fad41f00cff849458cb358ab5e2098 19415 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 19416 Date: Thu Jul 17 14:37:05 2008 +1000 19417 19418 Fix console switch on R500 19419 19420 This patch fixes the console switch for me on R5xx. 19421 19422 There are two aspects to it: 19423 19424 - Fix the ordering of avivo_restore() to better match what's 19425 happening in the driver & ATOM, properly locking/unlocking and 19426 only enabling the CRTCs after everything has been properly 19427 programmed. 19428 19429 - Don't ASIC_INIT if the card has any CRTC enabled. This is the 19430 best I came up with for avoiding spurrious ASIC_INIT on cards that 19431 -are- POSTed but don't have the BIOS coming from c0000 on x86. The 19432 problem with spurrious ASIC_INIT is that we do it before we do 19433 RADEONSave(), so that screws up the console switch. 19434 19435 Note that I think we also should save/restore the palette, I don't think 19436 we do. right now, it's a minor issue for me because I fixed offb to be 19437 able to set it on AVIVO's but it might still have to be done in the long 19438 run. 19439 19440 Tested with a VGA analog setup on DACA and a DVI setup on TMDS 0. I 19441 haven't tested any other combo but that should get us going. 19442 19443 Cheers, 19444 Ben. 19445 19446 Signed-off-by: Dave Airlie <airlied (a] redhat.com> 19447 19448 commit e8c07270e1fc2ee455320370692aaf90abac2773 19449 Author: Alex Deucher <alexdeucher (a] gmail.com> 19450 Date: Tue Jul 15 17:50:32 2008 -0400 19451 19452 ATOM: fix for r4xx cards with no VRAM_Usage tables 19453 19454 fixes bug 15954 19455 19456 commit da41e71efd41907d9347a902720bce60b32550e5 19457 Author: Alex Deucher <alexdeucher (a] gmail.com> 19458 Date: Mon Jul 14 13:21:38 2008 -0400 19459 19460 ATOM: Add support for UpdateCRTC_DoubleBufferRegisters cmd table 19461 19462 locks/unlocks the crtc/grph/mode regs before updating the crtc 19463 19464 commit a5e0cf13dc7ace6cf0e44e18b73b9a9266e740ab 19465 Author: Julien Cristau <jcristau (a] debian.org> 19466 Date: Sun Jul 6 12:17:28 2008 +0200 19467 19468 Link with -lpciaccess and -ldrm if needed 19469 19470 This makes sure the driver ends up with a DT_NEEDED reference to 19471 the libraries it's using. 19472 19473 commit f8da849fd89322a54cc4d722767da755358eab70 19474 Author: Michel Dnzer <michel (a] tungstengraphics.com> 19475 Date: Sun Jul 13 11:31:03 2008 +0200 19476 19477 Only declare local variable when it's used. 19478 19479 commit a6db4dc65aec560f3b1f37ff73ec6b0bd8207295 19480 Author: Dave Airlie <airlied (a] linux.ie> 19481 Date: Sat Jul 12 18:47:07 2008 +1000 19482 19483 atombios: fix typo in mode conversion for panel 19484 19485 commit 19a0d5879851eff23a3e7d5cdea55bd784051115 19486 Author: Dave Airlie <airlied (a] linux.ie> 19487 Date: Sat Jul 12 17:04:25 2008 +1000 19488 19489 atombios: use macro to get offset into table 19490 19491 commit a9746114369d516072d841ec356ec3ba3d0be71a 19492 Author: Dave Airlie <airlied (a] linux.ie> 19493 Date: Sat Jul 12 10:46:36 2008 +1000 19494 19495 pciid: add radeon HD3850. 19496 19497 pci id from legume on #radeon 19498 19499 commit ab14f725676e4d4e45278c64b03fe2d328a3e439 19500 Author: Alex Deucher <alexdeucher (a] gmail.com> 19501 Date: Fri Jul 11 19:05:00 2008 -0400 19502 19503 R300: NUM_FPU adjustments for VAP_CNTL 19504 19505 commit 810c28cc2660b73e03e4d27be97988fc5321a06f 19506 Author: Alex Deucher <alexdeucher (a] gmail.com> 19507 Date: Fri Jul 11 15:31:57 2008 -0400 19508 19509 Make sure cursor offsets are properly aligned when using EXA 19510 19511 commit 7e67d0163579a44f104e8f354a929ac9b33e4c21 19512 Author: Alex Deucher <alexdeucher (a] gmail.com> 19513 Date: Fri Jul 11 14:30:17 2008 -0400 19514 19515 Fix cursor with multi-head and rotation 19516 19517 commit 9086d008fb8c3cde0633b90ce19ffbf4eded388d 19518 Author: Alex Deucher <alexdeucher (a] gmail.com> 19519 Date: Tue Jul 8 09:57:04 2008 -0400 19520 19521 Add quirk for Falcon NW laptop 19522 19523 commit 61f82ace0210251beb0bcc492218a75a193e1deb 19524 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 19525 Date: Tue Jul 8 21:58:43 2008 +1000 19526 19527 atombios: add support for other endians. 19528 19529 This is a cleaned up (in as much as atombios can be..) of benh's patch. 19530 19531 airlied - removed benh's debugging for now, it might need to be put back later.. 19532 19533 commit b4d1a47b5c6eafda5e274398eebe2701b030f22e 19534 Author: Dave Airlie <airlied (a] linux.ie> 19535 Date: Tue Jul 8 11:09:01 2008 +1000 19536 19537 rv770 initial support 19538 19539 adds pci ids and one register from AMD code 19540 19541 commit 7ae4cec8cc8c90aee5dc4fa7abcce22321d4f4eb 19542 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 19543 Date: Mon Jul 7 14:39:47 2008 -0400 19544 19545 clamp tex coords (r100/r200) for textured video 19546 19547 fixes bug 14904 19548 19549 commit 879f9a37856642b337e3c0233e38fd5443338eef 19550 Author: Alex Deucher <alexdeucher (a] gmail.com> 19551 Date: Fri Jul 4 13:54:49 2008 -0400 19552 19553 Revert "rotate: fix rotation in conf file." 19554 19555 This reverts commit 22c6ca0d5d226c725ba448e5db63fdba9e32332c. 19556 19557 There seems to be an ordering issue in that causes a nasty hang 19558 if the DRI is enabled. This needs more investigation. 19559 19560 commit 22c6ca0d5d226c725ba448e5db63fdba9e32332c 19561 Author: Dave Airlie <airlied (a] redhat.com> 19562 Date: Thu Jul 3 20:14:59 2008 +1000 19563 19564 rotate: fix rotation in conf file. 19565 19566 need to set modes after init accel architecture 19567 19568 reported-by: Christoph Bartoschek <bartoschek (a] or.uni-bonn.de> 19569 19570 commit ce1a3edd52b647744cfa4433301befb437d211bb 19571 Author: Dave Airlie <airlied (a] redhat.com> 19572 Date: Thu Jul 3 20:05:54 2008 +1000 19573 19574 radeon: drop all use of CPMode. 19575 19576 We never test the other codepath and I don't think I've ever recommended it 19577 for anyone. 19578 19579 commit c037b4ce8769ad840a257e22b1e4ad58b8ed96fa 19580 Author: Wolke Liu <wolke.lui (a] amd.com> 19581 Date: Tue Jul 1 10:45:39 2008 -0400 19582 19583 Add pci id for FireMV 2400 19584 19585 commit 72992668ca96717bf69bcea95a5deeaf7c57e826 19586 Author: Alex Deucher <alex (a] botchco.com> 19587 Date: Mon Jun 30 14:09:09 2008 -0400 19588 19589 RS4xx: enable the DRI by default on all variants 19590 19591 Some RS4xx chips had the DRI disabled by default. 19592 With the recent drm and ddx changes these chips have been 19593 reported to work now with the DRI. 19594 19595 commit 0378c8ed88c829f09b5c6d51314325eaf2685fdb 19596 Author: Michel Dnzer <michel (a] tungstengraphics.com> 19597 Date: Sat Jun 28 14:06:45 2008 +0200 19598 19599 Restore versioning of interface for Mesa DRI driver. 19600 19601 This doesn't have anything to do with the package version. 19602 19603 commit cfaa23d925e3c062cf87ea844566ac11ea02d69e 19604 Author: Alex Deucher <alex (a] botchco.com> 19605 Date: Fri Jun 27 20:36:54 2008 -0400 19606 19607 RADEON: IGP VGA fixes take 2 19608 19609 It seems only RS300/350/400/480 are missing the primary dac 19610 See bug 15708 19611 19612 commit 31c27ffcb3c4c5334cf726ecd4e293a678b2a1ea 19613 Author: Alex Deucher <alex (a] botchco.com> 19614 Date: Fri Jun 27 20:29:04 2008 -0400 19615 19616 Revert "IGP: attempt to fix VGA on IGP chips" 19617 19618 This reverts commit e78e8a21b4040cd7f1983c241c860d9209398396. 19619 19620 commit af79a77e6d8206b18bba97cf4e0770921dc439a7 19621 Author: Brice Goglin <Brice.Goglin (a] ens-lyon.org> 19622 Date: Fri Jun 27 17:09:13 2008 +0200 19623 19624 use PACKAGE_VERSION_* for radeon version 19625 19626 now radeon shows verion 6.9.0 instead of 4.3.0. 19627 19628 RADEON_VERSION_MAJOR_TILED is dropped from ddxDriverMajorVersion 19629 since RADEON_VERSION_MAJOR is bigger now. 19630 19631 commit c83fbdfa076c107012b7dfbbfbbb2feede00542b 19632 Author: Alex Deucher <alex (a] botchco.com> 19633 Date: Thu Jun 26 19:48:45 2008 -0400 19634 19635 Bump for 6.9.0 release 19636 19637 commit 32e1d1daf4f68ea37624afaa2bc1ea7742e1163b 19638 Author: Brad Smith <brad (a] comstyle.com> 19639 Date: Thu Jun 26 10:22:03 2008 -0400 19640 19641 RADEON: sys/endian.h needs sys/types.h on BSDs 19642 19643 See bug 16512 19644 19645 commit dca522355a9039eca6efaba3b36397b246800f94 19646 Author: Brad Smith <brad (a] comstyle.com> 19647 Date: Thu Jun 26 10:13:59 2008 -0400 19648 19649 RADEON: fix copy/paste error in accel code 19650 19651 commit f4292e110105910d0c19bf2db28e2682b27af2c7 19652 Author: Dave Airlie <airlied (a] redhat.com> 19653 Date: Fri Jun 20 15:31:43 2008 +1000 19654 19655 Revert "Revert "ATOM: disable TV-out for now"" 19656 19657 Oops I really didn't mean to do this, I was testing something and it slipped 19658 past. 19659 19660 This reverts commit dd18caa4b2efc430eaae0c4362b65641f9bef440. 19661 19662 commit bb1cfcbfbc1ace1f257fc3bf6401501f1a6da827 19663 Author: Alex Deucher <alex (a] botch2.com> 19664 Date: Wed Jun 25 15:50:16 2008 -0400 19665 19666 RADEON: adjustments to Jerome's last commit 19667 19668 - Flush caches and wait for idle after drawing 19669 - Make sure 3D is idle too (after composite or textured video) 19670 19671 commit 07be302f6dba61dd15918963fae66502230c74b7 19672 Merge: 8c9b8de03 52459745e 19673 Author: Jerome Glisse <glisse (a] freedesktop.org> 19674 Date: Wed Jun 25 10:30:45 2008 +0200 19675 19676 Merge branch 'master' of git://anongit.freedesktop.org/xorg/driver/xf86-video-ati 19677 19678 commit 8c9b8de0373797cba6f0a27e6b6f461e1070fef9 19679 Author: Jerome Glisse <glisse (a] freedesktop.org> 19680 Date: Wed Jun 25 10:28:26 2008 +0200 19681 19682 radeon: flush & wait for 2d & dma idle after 2d blit 19683 19684 This should help to avoid 2d & 3d engine to step on each 19685 other dma transaction. 19686 19687 commit 52459745ec05de88adbc087e9566fe6d97ef424b 19688 Author: Dave Airlie <airlied (a] linux.ie> 19689 Date: Wed Jun 25 17:09:11 2008 +1000 19690 19691 ati: hopefully fix byteswap mess for those other OSes. 19692 19693 commit 67a6ac0001bc9d062aa426384a11a41fa7a1c09a 19694 Author: Alex Deucher <alex (a] botch2.com> 19695 Date: Tue Jun 24 21:06:37 2008 -0400 19696 19697 bump for rc release 19698 19699 commit bd68507d2d66e03d8bcde5f6e7ea9b2dbfe8b8a0 19700 Author: Alex Deucher <alex (a] botch2.com> 19701 Date: Tue Jun 24 20:59:58 2008 -0400 19702 19703 RADEON: warning fix 19704 19705 commit 8e534d69c9b19fc085f7c5ca9e18f5ea04f6fc12 19706 Author: Alex Deucher <alex (a] botch2.com> 19707 Date: Tue Jun 24 20:08:35 2008 -0400 19708 19709 RADEON: cleanups 19710 19711 - fix some warnings 19712 - RS400 and RS480 are separate families now (update default tmds and dac2 19713 tables) 19714 19715 commit e78e8a21b4040cd7f1983c241c860d9209398396 19716 Author: Alex Deucher <alex (a] botch2.com> 19717 Date: Tue Jun 24 19:53:28 2008 -0400 19718 19719 IGP: attempt to fix VGA on IGP chips 19720 19721 VGA has never worked on some IGP chips. While the chip only has 19722 one DAC, it appears to use a mix of Primary DAC and TVDAC controls. 19723 19724 See bug 15708 19725 19726 commit faa4b4b8dbe9b8a452cfa60d53874350bb04e0cb 19727 Author: Ji Paleek <jpalecek (a] web.de> 19728 Date: Tue Jun 24 19:20:50 2008 -0400 19729 19730 RADEON: remove extraneous line from new pll code 19731 19732 commit 72feaa37ea07620f5f2ead438dbc72a1c8883cd3 19733 Author: Ji Paleek <jpalecek (a] web.de> 19734 Date: Mon Jun 23 15:53:58 2008 -0400 19735 19736 RADEON: PLL tweaks 19737 19738 Patch from Ji Paleek (see debian bug 465864) with some tweaks 19739 by me. 19740 19741 - abort rather than programming bad dividers if no pll dividers can be found 19742 - improve the pll selection algorithm 19743 - in general, prefer lower ref dividers 19744 19745 I've tested this patch on a wide variety of chips (r1xx-r6xx) and clocks. 19746 19747 commit 9c2f909ea437a63a408d2398ecabe0b378dbb982 19748 Author: Alex Deucher <alex (a] botch2.com> 19749 Date: Mon Jun 23 10:38:15 2008 -0400 19750 19751 RADEON: adjust randr crtc/output prepare/commit ordering 19752 19753 This fixes some occasional mode change problems with multiple heads active. 19754 It seems radeons generally like to turn on the whole output/crtc setup 19755 in one shot. 19756 19757 commit aea9bf75cf0774afd3e65fcf3fd3851f5fb21ca3 19758 Author: Alex Deucher <alex (a] botch2.com> 19759 Date: Sat Jun 21 10:57:05 2008 -0400 19760 19761 0x1002 0x5657 is actually RV410 19762 19763 See bug 14289 19764 19765 com_bios_scratch.diff 19766 19767 commit 38ce8a984f96056b7345bcc06505ba27e0e6d5b4 19768 Author: Dave Airlie <airlied (a] redhat.com> 19769 Date: Fri Jun 20 14:16:22 2008 +1000 19770 19771 legacy: use entity MMIO for dpms as this can crossover between zaphod infos 19772 19773 commit 1a7d9bc53512b0a5240176c949e6abf1ae2fb1fd 19774 Author: Dave Airlie <airlied (a] redhat.com> 19775 Date: Fri Jun 20 14:14:21 2008 +1000 19776 19777 atombios: use MMIO from the entity not the info 19778 19779 commit ef624b88903b1a87ef5b6388e18291f75776b93d 19780 Author: Alex Deucher <alex (a] botch2.com> 19781 Date: Thu Jun 19 18:20:52 2008 -0400 19782 19783 RADEON: fix read past the end of an array 19784 19785 commit cfe814a481d8cf2005d738a0ca9782f1ed4177f5 19786 Author: Dave Airlie <airlied (a] linux.ie> 19787 Date: Fri Jun 20 07:51:27 2008 +1000 19788 19789 r600: don't add fb offset here to make shadowfb work. 19790 19791 discovered on irc with wpwrak. 19792 19793 commit 5b323a2cbbc412b6dd4e782438b64ee996558d05 19794 Merge: dd18caa4b eed241553 19795 Author: Dave Airlie <airlied (a] linux.ie> 19796 Date: Fri Jun 20 07:49:16 2008 +1000 19797 19798 Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 19799 19800 commit eed241553748125e902c44dcc8cf8e0605be071b 19801 Author: Alex Deucher <alex (a] botch2.com> 19802 Date: Tue Jun 17 17:42:31 2008 -0400 19803 19804 radeon: X_ERROR -> X_WARNING for num gb pipes 19805 19806 This just means your drm is old, not fatal or anything 19807 19808 commit dd18caa4b2efc430eaae0c4362b65641f9bef440 19809 Author: Dave Airlie <airlied (a] linux.ie> 19810 Date: Sat Jun 14 18:37:56 2008 +1000 19811 19812 Revert "ATOM: disable TV-out for now" 19813 19814 This reverts commit effa245914823371e052cd9aa1143a02350891e7. 19815 19816 commit b416e97a1f16ef700ba1aaca317dee82b7a3cf64 19817 Author: Alex Deucher <alex (a] botch2.com> 19818 Date: Fri Jun 13 02:21:31 2008 -0400 19819 19820 RADEON: man page updates 19821 19822 commit 88ce87d79069662de28ecb3c10c140919927d584 19823 Author: Alex Deucher <alex (a] botchco.com> 19824 Date: Thu Jun 12 15:10:47 2008 -0400 19825 19826 Bump for 6.9.0rc1 release 19827 19828 commit ada41723fdc8cbeeda12bb4ae2d16e8d1ba215de 19829 Author: Alex Deucher <alex (a] botch2.com> 19830 Date: Wed Jun 11 14:29:36 2008 -0400 19831 19832 R3xx: use DSTCACHE_CTLSTAT rather than RB2D_DSTCACHE_CTLSTAT 19833 19834 According to the hw guys, you should use DSTCACHE_CTLSTAT to 19835 flush the 2D dst cache rather than RB2D_DSTCACHE_CTLSTAT. 19836 19837 commit 3d70c398afa0831387af81559453b77c218a54f9 19838 Author: Dave Airlie <airlied (a] redhat.com> 19839 Date: Thu Jun 12 06:18:26 2008 +1000 19840 19841 atom: fix another endian define 19842 19843 commit f1794e4c34d801371461afed37e7bb7d1fc79305 19844 Author: Dave Airlie <airlied (a] redhat.com> 19845 Date: Thu Jun 12 06:10:30 2008 +1000 19846 19847 atom: fixup includes in cd operations to get correct endian bits in atombios.h 19848 19849 commit f29976d12cc2020cd80c90fa3fd4b25c031559ba 19850 Author: Dave Airlie <airlied (a] ppcg5.localdomain> 19851 Date: Wed Jun 11 19:33:02 2008 +1000 19852 19853 radeon: add lots of endian bits into the atombios parser 19854 19855 commit 09eb220971b5d2bfd7d1ff6f552060967a133152 19856 Author: Dave Airlie <airlied (a] ppcg5.localdomain> 19857 Date: Wed Jun 11 19:21:22 2008 +1000 19858 19859 atombios: fixup atombios bitfields for endianness 19860 19861 reverse the bits for big endian 19862 19863 commit b80a3e85bf9dbfc8d96a745876d75d38bea97c1a 19864 Author: Alex Deucher <alex (a] botchco.com> 19865 Date: Mon Jun 9 20:29:47 2008 -0400 19866 19867 R5xx: fix RADEONSetAgpBase() for R5xx chips 19868 19869 Is there any reason to still do this in the ddx? 19870 Maybe real old drms? 19871 19872 commit ab7936708dfbee4d3676262f8010e001b73a4a38 19873 Author: Alex Deucher <alex (a] botchco.com> 19874 Date: Mon Jun 9 17:05:04 2008 -0400 19875 19876 RS4xx: RC410 chips are actually RS400 based 19877 19878 commit a54ca3c699c103e2e3df87101dca929a3eee4377 19879 Author: Dave Airlie <airlied (a] linux.ie> 19880 Date: Thu Jun 5 18:31:53 2008 +1000 19881 19882 radeon: fix typo in memory calcs 19883 19884 noticed by ghepeu on irc. 19885 19886 commit 6e4e6d2a8f29f92efc219dca24ea31d1f37d5a0f 19887 Author: Alex Deucher <alex (a] botchco.com> 19888 Date: Tue Jun 3 20:32:20 2008 -0400 19889 19890 RADEON: minor cleanups 19891 19892 commit 9d3afbf5fa4110928a9f965df2733c79db92ea99 19893 Author: Michel Dnzer <michel (a] tungstengraphics.com> 19894 Date: Tue Jun 3 11:40:49 2008 +0200 19895 19896 Call DRM_IOCTL_MODESET_CTL ioctl to avoid problems with DRM post vblank-rework. 19897 19898 commit effa245914823371e052cd9aa1143a02350891e7 19899 Author: Alex Deucher <alex (a] botchco.com> 19900 Date: Sun Jun 1 20:47:50 2008 -0400 19901 19902 ATOM: disable TV-out for now 19903 19904 It only sort of works in NTSC mode and isn't handled correctly 19905 in all cases yet. Until we sort out the remaining details leave it 19906 disabled. 19907 See bugs: 16186, 16178, 16185 19908 19909 commit 8504c6b0e40477ee544ad7f5366d569bdc53d6f0 19910 Author: Alex Deucher <alex (a] botch2.com> 19911 Date: Fri May 30 11:19:03 2008 -0400 19912 19913 RADEON: update RADEONGetVRAMType() for newer chips 19914 19915 commit 7cb695a329a9b543ad61af08c2d7d6eaf56de35e 19916 Author: Alex Deucher <alex (a] botch2.com> 19917 Date: Fri May 30 10:10:56 2008 -0400 19918 19919 RADEON: add Int10 option 19920 19921 On some radeons you can't read the bios without initializing int10. 19922 On ATOM-based secondary cards, intitializing int10 tends to hang the card. 19923 19924 commit 714b2c63ec1c4a1410a6d521c03e9d1f90937c01 19925 Author: Alex Deucher <alex (a] botch2.com> 19926 Date: Thu May 29 02:31:00 2008 -0400 19927 19928 RADEON: don't enable legacy bios table based init yet 19929 19930 Needs more work and we need to figure out how best to decide 19931 when to use it. 19932 19933 commit e20b08525a64888287ec4a369d8f7dbde95c655d 19934 Author: Alex Deucher <alex (a] botch2.com> 19935 Date: Wed May 28 14:43:47 2008 -0400 19936 19937 ATOM: fail of we are not able to get a bios image 19938 19939 commit d4fa17b426b0cafac3184b2ea9b675e2ff154ee9 19940 Author: Alex Deucher <alex (a] botch2.com> 19941 Date: Wed May 28 09:50:02 2008 -0400 19942 19943 R1xx: Gamma fixes for overlay 19944 19945 commit c7e6a50f60a04bbbf1c1ddbf4738a5f75c94f950 19946 Author: Alex Deucher <alex (a] botchco.com> 19947 Date: Wed May 28 09:05:19 2008 -0400 19948 19949 RADEON: only attempt to read vbios from legacy ISA space if it's primary 19950 19951 commit faea008806802ec0e045754ec1eca492ebae320e 19952 Author: Alex Deucher <alex (a] botch2.com> 19953 Date: Tue May 27 18:36:01 2008 -0400 19954 19955 R3/4/5xx: use get_param to get the num_gb_pipes from the drm 19956 19957 commit 965a5dbcd9dc4bf1cdd7f2bbdec15e9733b2e090 19958 Author: Alex Deucher <alex (a] botch2.com> 19959 Date: Tue May 27 16:48:41 2008 -0400 19960 19961 RADEON: improve support for secondary cards 19962 19963 this should fix bugs 16115, 16035 19964 19965 commit 5f951a5573f0c7572230c9aa4d3f75d67f91ed71 19966 Author: Alex Deucher <alex (a] botch2.com> 19967 Date: Wed May 21 00:35:14 2008 -0400 19968 19969 ATOM: remove duplicate code 19970 19971 commit edce33e87fb79a540d8c285f205d4c3f2c2bc9f4 19972 Author: Dave Airlie <airlied (a] redhat.com> 19973 Date: Wed May 28 06:43:40 2008 +1000 19974 19975 radeon: fix surface access on avivo chips. 19976 19977 This should fix VT switch on vesafb 19978 19979 commit 307bd65a25ee425d8359dd30572b002ce2338f91 19980 Author: Michael Babej <mbabej (a] redhat.com> 19981 Date: Tue May 27 19:13:49 2008 +1000 19982 19983 r600: fix vt switch issue 19984 19985 RH BZ 441492 19986 19987 commit bf48c9eb8ec592515be9d1732d60283af715674d 19988 Author: Dave Airlie <airlied (a] linux.ie> 19989 Date: Sat May 24 11:51:20 2008 +1000 19990 19991 atombios: for LVDS set use ref div as per legacy 19992 19993 commit 04500c8419b1aeaeac7968492b82e1d7cad1d05b 19994 Author: Alan Coopersmith <alan.coopersmith (a] sun.com> 19995 Date: Tue May 20 20:34:31 2008 -0700 19996 19997 Strip ^M's from src/AtomBios/includes/ObjectID.h 19998 19999 commit 521a0488d1c414209c3534dc242000faa332a441 20000 Author: Alan Coopersmith <alan.coopersmith (a] sun.com> 20001 Date: Tue May 20 20:33:37 2008 -0700 20002 20003 On non-gcc compilers, use C99's __func__ instead of gcc's __FUNCTION__ 20004 20005 commit b7c80d0c86646105d2bce5d4d59ba6c45aa7cafc 20006 Author: Tobias Diedrich <ranma+freedesktop (a] tdiedrich.de> 20007 Date: Tue May 20 19:16:54 2008 -0400 20008 20009 R200/R300: fix gamma setup for overlay 20010 20011 More pending. See bug 16001 20012 20013 commit 69423482e2e94637142a9ba675589a1449a346a8 20014 Author: Alex Deucher <alex (a] botch2.com> 20015 Date: Tue May 20 18:57:13 2008 -0400 20016 20017 RADEON: cleanup connection detection and remove duplicate code 20018 20019 commit 55e4469f59c82bb5762673de5f3f27d18b0bd9a3 20020 Author: Alex Deucher <alex (a] botch2.com> 20021 Date: Tue May 20 17:46:58 2008 -0400 20022 20023 RADEON: enable cloning on multi-crtc cards 20024 20025 commit 12f185634071980041aaac6265d89708b181b215 20026 Author: Alex Deucher <alex (a] botch2.com> 20027 Date: Tue May 20 16:05:41 2008 -0400 20028 20029 ATOM: Ignore invalid connector entries 20030 20031 commit 19e97f74e39fc2b35727708ac429de33d0b70162 20032 Author: Egbert Eich <eich (a] freedesktop.org> 20033 Date: Tue May 20 11:16:15 2008 +0200 20034 20035 Change RMX code to follow the programming algorithm suggested by ATI. 20036 20037 ATI provides the following algorithm to calculate the RMX scaling ratios 20038 in its programming specs: 20039 when RMX_AUTO_RATIO_HORZ_INC set to 1,Horizontal auto ratio result trucated, 20040 and then incremented by 1. 20041 Horz_Ratio = ( ((Active display width in characters (including overscan) + 1) 20042 / (Panel width in characters)) x 4096 + 1 ) 20043 else 20044 Horz_Ratio = ( ((Active display width in characters (including overscan)) 20045 / (Panel width in characters)) x 4096 + 1 ) 20046 20047 when RMX_AUTO_RATIO_VERT_INC set to 1, Vertical auto ratio result trucated, 20048 and then incremented by 1. 20049 Vert_Ration = ( ((Active display width in characters (including overscan) + 1) 20050 / (Panel width in characters)) x 4096 + 1) 20051 else 20052 Vert_Ration = ( ((Active display width in characters (including overscan)) 20053 / (Panel width in characters)) x 4096 + 1) 20054 20055 This patch implements this behavor. Additionally it avoids the use of floats. 20056 20057 commit a4f3d0088ba763ed8eab1d331959b5ecde8262e8 20058 Author: Alex Deucher <alex (a] botch2.com> 20059 Date: Tue May 20 11:42:53 2008 -0400 20060 20061 R3/4/5xx: fixup logic from last commit 20062 20063 Spotted by otaylor and MrCooper 20064 20065 commit 130e55738047f2a073bcc47e3e1400f7b694a81b 20066 Author: Alex Deucher <alex (a] botch2.com> 20067 Date: Tue May 20 11:32:42 2008 -0400 20068 20069 R3/4/5xx: fix EXA rotation 20070 20071 xrandr uses PictOpSrc for rotation which we were falling back on since 20072 render semanties require alpha=0 for REPEAT_NONE when there is no alpha 20073 channel and there is a transform. If the dst has no alpha channel we 20074 should be ok I think. 20075 20076 Otayler and I discussed this on IRC. the more general fix would be 20077 to clip the dst to the src and mask for bounded ops like in/add/over 20078 in the pixel exact transform case. 20079 20080 commit cc9f510770700228e5d597c872e926a4e99bd950 20081 Author: Corbin Simpson <MostAwesomeDude (a] gmail.com> 20082 Date: Wed May 14 16:49:01 2008 -0400 20083 20084 RADEON: missing stdint.h includes 20085 20086 commit 71fa57f871dba03260dba2180ce1dab44048ac1a 20087 Author: Alex Deucher <alex@cube.(none)> 20088 Date: Wed May 14 13:36:38 2008 -0400 20089 20090 Add RS600 support 20091 20092 commit 582c1a1b2c7b1032e9f9f54ca36100c57f580c5c 20093 Author: Alex Deucher <alex@cube.(none)> 20094 Date: Tue May 13 20:50:25 2008 -0400 20095 20096 RS4xx: Split out RS400 and RS480 as separate families 20097 20098 RS400 (intel based IGP) and RS480 (AMD based IGP) have different 20099 MC setups and need to be handled differently 20100 20101 commit 708e7c98f636738fbcc47a597bc94b309a4dc1c4 20102 Author: Michel Dnzer <michel (a] tungstengraphics.com> 20103 Date: Tue May 13 11:00:38 2008 +0200 20104 20105 Reinstate CARD* types that are part of external interfaces. 20106 20107 The compiler pointed this out... 20108 20109 commit 908b7b940e3ef296836bc94680ebb8ab67650e31 20110 Author: Matt Turner <mattst88 (a] gmail.com> 20111 Date: Mon May 12 12:06:33 2008 -0400 20112 20113 Replace CARD{8,16,32} with uint{8,16,32}_t 20114 20115 As has been done with xf86-video-intel, replace all CARD* datatypes with 20116 uint*_t datatypes available from stdint.h. 20117 20118 commit 87e66ce76430890ab4939ffcd42f72b9288eb598 20119 Author: Avi Rozen <avi.rozen (a] gmail.com> 20120 Date: Mon May 12 08:33:38 2008 -0400 20121 20122 RADEON: fix lockup on start 20123 20124 see debian bug 480312 20125 20126 commit 94bf8f01bd43cb103fffecfe04d04a214f892baf 20127 Author: Dave Airlie <airlied (a] redhat.com> 20128 Date: Mon May 12 20:02:51 2008 +1000 20129 20130 radeon: fix set_cursor_colours remove fb offset 20131 20132 commit 1d0f1d31e2ed1d91ee87cb3e02ce48c8c07aa418 20133 Author: Dave Airlie <airlied (a] redhat.com> 20134 Date: Mon May 12 15:44:35 2008 +1000 20135 20136 radeon: rs485 vs rs485m.. mobile? non-mobile? bi? 20137 20138 So it appears we have in the wild two chipsets with the same PCI ID (0x1002:0x5974) 20139 that are mobile and non-mobile. 20140 20141 the RH bug references is a desktop system. 20142 The Dell Vostro 1100 also has this chipset with LVDS bits. 20143 20144 commit 25e0c3945a51ae8c14b3a847ec75a256e1397f24 20145 Author: Alex Deucher <alexdeucher (a] gmail.com> 20146 Date: Mon May 12 10:28:33 2008 +1000 20147 20148 radeon: add initial support for cloning outputs. (single-CRTC cards only) 20149 20150 airlied - This code was originally written by Alex, and I've modified it to 20151 only run on single-CRTC cards for now until we can test it some more. 20152 20153 commit 94405eb1c9e4e0ababc6aef03b753d6ed9eb7838 20154 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 20155 Date: Fri May 9 05:28:44 2008 -0400 20156 20157 RADEON: fixup ifdef from last commit 20158 20159 commit 18e20bc22a55ecfee9798c01079d7b24d19f0051 20160 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 20161 Date: Thu May 8 08:45:51 2008 -0400 20162 20163 RADEON: disable MMIO paths for EXA composite/texvid on IGP/R5xx 20164 20165 The MMIO paths eventually lead to a hang on r5xx/IGP. I haven't 20166 been able to find out why yet. 20167 20168 commit c3532268875fd24e6519bea2fb1b814d612bbdb4 20169 Author: Dave Airlie <airlied (a] linux.ie> 20170 Date: Wed May 7 02:37:18 2008 +1000 20171 20172 radeon: fix zaphod EXA with texture video 20173 20174 commit ffc437f3606ab8ceba1ff152e4bb08988a58b54c 20175 Author: Dave Airlie <airlied (a] linux.ie> 20176 Date: Wed May 7 02:30:28 2008 +1000 20177 20178 avivo: fix zaphod cursor in theory 20179 20180 commit e36ef14e3a1087e1fe41baa26ade2937f396001f 20181 Author: Dave Airlie <airlied (a] linux.ie> 20182 Date: Wed May 7 01:39:28 2008 +1000 20183 20184 radeon: fix textured-xv on zaphod 20185 20186 commit 8fc19bee27c0f151d2ab3354f6ac0992b358436d 20187 Author: Dave Airlie <airlied (a] linux.ie> 20188 Date: Wed May 7 01:38:24 2008 +1000 20189 20190 radeon: zaphod: fix render accel for EXA 20191 20192 commit fc41b9042a5220a8419cc7b69ca3850cae6b903c 20193 Author: Dave Airlie <airlied (a] linux.ie> 20194 Date: Wed May 7 01:32:23 2008 +1000 20195 20196 radeon: fix EXA pixmap offset on zaphod 20197 20198 commit 4568cb820d567c6909a4be956d7e79b91232535e 20199 Author: Dave Airlie <airlied (a] linux.ie> 20200 Date: Wed May 7 01:19:39 2008 +1000 20201 20202 radeon: zaphod fixes for pciaccess not allowing multiple MMIO maps 20203 20204 commit ca81fa086b21633a7fd926833fb6d1d4fa080646 20205 Author: Dave Airlie <airlied (a] linux.ie> 20206 Date: Wed May 7 01:12:01 2008 +1000 20207 20208 radeon: zaphod fix for cursor on second head 20209 20210 We don't need to add fbOffset here as the mmap we have of the framebuffer 20211 starts half way. 20212 20213 commit 24b60c8965f6a0b3f0c2bb1e7236b4d6642c5918 20214 Author: Julien Cristau <jcristau (a] debian.org> 20215 Date: Fri May 2 15:30:45 2008 -0400 20216 20217 Add a test for __GLIBC__ to CD_Common_Types.h. 20218 20219 Atombios redefines the standard types but the definitions conflict 20220 with the ones from glibc (causes build failures on GNU/Hurd 20221 and GNU/kFreeBSD). 20222 20223 commit f051359ac09c6b9416e39b9ca7d9dc0880aa1557 20224 Author: thegraze <thegraze (a] googlemail.com> 20225 Date: Fri May 2 14:02:16 2008 -0400 20226 20227 ATOM: add support for DragonFlyBSD 20228 20229 commit 3d469cbc3225d890a895dac7cbc1ab7e08054b48 20230 Author: Alex Deucher <alex@cube.(none)> 20231 Date: Wed Apr 30 18:33:04 2008 -0400 20232 20233 RADEON: lock the cursors when updating 20234 20235 this should fix occasional corruption seen when updating 20236 the cursor. 20237 20238 commit 445b71021843665ba32f37b2ce5c9d2857c07cc7 20239 Author: Alex Deucher <alex@cube.(none)> 20240 Date: Tue Apr 29 21:01:41 2008 -0400 20241 20242 RADEON: assorted fixes 20243 20244 - free rotate pixmaps on VT switch 20245 - save crtc/output status so we only turn on 20246 crtcs/outputs if they are off 20247 - show/hide cursors when changing modes 20248 20249 commit 070cce5255a5c311f9d8b85ec54bd56655014933 20250 Author: Stephan Wolf <stephan (a] letzte-bankreihe.de> 20251 Date: Mon Apr 28 11:26:37 2008 -0400 20252 20253 R3xx+: further fix for IGP chips 20254 20255 see bug 15538 20256 20257 commit 211e0041c7fc2df494b77428553943a2b526ee4e 20258 Author: Alex Deucher <alex@cube.(none)> 20259 Date: Sun Apr 27 21:08:00 2008 -0400 20260 20261 IGP: fix EXA composite corruption 20262 20263 commit 656b06bdde129ca4fc370f5a2cf7311c9179b0ff 20264 Author: Alex Deucher <alex (a] botch2.com> 20265 Date: Sun Apr 27 20:20:49 2008 -0400 20266 20267 RADEON: remove duplicate register define 20268 20269 Also add more bit defs to wait_until register 20270 20271 commit 8a9820a3aa49bc667f90ac291a27e4d7b4ae38b3 20272 Author: Alex Deucher <alex (a] botch2.com> 20273 Date: Sun Apr 27 19:02:22 2008 -0400 20274 20275 RADEON: decrease crtc/output verbosity 20276 20277 commit c5d62fa0e8f52c3264ff9db3ff10cdf5a806bfc0 20278 Author: Owen Taylor <otaylor (a] huygens.home.fishsoup.net> 20279 Date: Thu Apr 17 13:14:53 2008 +0200 20280 20281 Emulate repeats by drawing in tiles 20282 20283 When we can't turn on hardware repeats, because the texture 20284 is non-power-of-two, or has padding at the ends of lines, 20285 try to draw the image in multiple tiles rather than falling 20286 back to software. (We can only do this when there is no 20287 transform.) 20288 20289 commit eeb7b74bb6c813b0e3afa4b704f6ffb0d0aab92b 20290 Author: Owen Taylor <otaylor (a] huygens.home.fishsoup.net> 20291 Date: Thu Apr 17 13:14:25 2008 +0200 20292 20293 Turn on wrapping when repeating on R100 + R200 20294 20295 Actually enable repeats for R100 and R200. This corresponds 20296 to a R300 change made in the patch in: 20297 http://bugs.freedesktop.org/show_bug.cgi?id=15333 20298 20299 commit e511f39dfef503006cf249b9f6934091eaade9b5 20300 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 20301 Date: Thu Apr 17 05:04:34 2008 -0400 20302 20303 R300+: move more common code into init3d() 20304 20305 - pre-load r3xx tex instructions 20306 - setup RS instructions in init3d() 20307 20308 commit 99435b7c18d931ea620044d0fdb4cc93dfcc6331 20309 Author: Owen Taylor <otaylor (a] redhat.com> 20310 Date: Thu Apr 17 02:46:11 2008 -0400 20311 20312 Radeon: Omit mask coordinates 20313 20314 Adapted from Owen's patch on bug 15546 20315 This fixes the slowness with aatext on r300 20316 and may speed up other chips marginally. 20317 20318 commit 37614e1db9a595fbe8a21d7a045895e11d272db9 20319 Author: Alex Deucher <alex (a] botch2.com> 20320 Date: Tue Apr 15 09:48:16 2008 -0400 20321 20322 fix up some things from the last commit 20323 20324 commit 1286fe5ce1c77453d57817b9b26b1bdb32ca7bc8 20325 Author: Alex Deucher <alex@botch2.(none)> 20326 Date: Mon Apr 14 20:02:14 2008 -0400 20327 20328 R300+: properly setup vap_cntl 20329 20330 this fixes tcl/pvs on RV515 among other things 20331 20332 commit f72a4b805db26f10f69330b88459cbeae661189b 20333 Author: Alex Deucher <alex (a] botch2.com> 20334 Date: Mon Apr 14 14:10:40 2008 -0400 20335 20336 EXA: Don't wait for 3D idle after each Composite() 20337 20338 wait in CompositeDone() instead 20339 20340 commit 4cd4acf1092aeb696b086a382a033aee471d2de9 20341 Author: Alex Deucher <alex (a] botch2.com> 20342 Date: Mon Apr 14 11:50:59 2008 -0400 20343 20344 R300: move more common code to init3d() 20345 20346 commit 3c523c9a07402e17dff588fad842224c57e98223 20347 Author: Alex Deucher <alex (a] botch2.com> 20348 Date: Mon Apr 14 11:21:42 2008 -0400 20349 20350 R3xx+: 3D engine documentation and minor cleanups 20351 20352 - document the R300 exa/textured video code 20353 - minor cleanups of textured video code to clarify meaning 20354 20355 commit ce025bbb2496d4de94b8d4ac450c64441b64ee04 20356 Author: Alex Deucher <alex (a] botch2.com> 20357 Date: Sat Apr 12 21:22:03 2008 -0400 20358 20359 R300+: consolidate some tcl/non-tcl paths 20360 20361 - Move more code to init3d() 20362 - MMIO textured video seems more reliable now on newer chips 20363 20364 commit 11b54a319c7c9dd52e3fb13372697059dafe1cd3 20365 Author: Alex Deucher <alex (a] botch2.com> 20366 Date: Sat Apr 12 16:50:22 2008 -0400 20367 20368 R3xx+: fix XAA + textured video on non-TCL path 20369 20370 commit dd15a2f5906725116b8cd9954243099055e88e37 20371 Author: Alex Deucher <alex (a] botch2.com> 20372 Date: Sat Apr 12 16:49:03 2008 -0400 20373 20374 R3xx+: more fixes to 2D/3D engine init 20375 20376 commit f3e68d4b7afd2e23675bf6361c496814c9cb4b94 20377 Author: Alex Deucher <alex (a] botch2.com> 20378 Date: Fri Apr 11 10:59:07 2008 -0400 20379 20380 Fix exa glyph corruption on newer chips 20381 20382 commit b59686d6427cbf8b35e36b020cbbc6a0c5149b22 20383 Author: Alex Deucher <alex (a] botch2.com> 20384 Date: Fri Apr 11 10:15:25 2008 -0400 20385 20386 R300+: pre-load vertex programs in init3D() 20387 20388 commit acc5833a35ad6c29a57f659607afb27eebdc2ea5 20389 Author: Alex Deucher <alex (a] botch2.com> 20390 Date: Thu Apr 10 17:52:52 2008 -0400 20391 20392 R3xx+: consolidate more tcl code 20393 20394 commit 6f8f75bd19ef1919c0291141675be2d0e29b3251 20395 Author: Alex Deucher <alex (a] botch2.com> 20396 Date: Thu Apr 10 17:08:50 2008 -0400 20397 20398 R3xx+: consolidate some common 3D code 20399 20400 commit 4b9234e1c4f7c7f419cb4245d64f3f9756c98bb6 20401 Author: Alex Deucher <alex (a] botch2.com> 20402 Date: Thu Apr 10 16:58:22 2008 -0400 20403 20404 R3xx+: tcl wip 20405 20406 commit 865c463e3afb4759758f569132be8bf1386da5cc 20407 Author: Alex Deucher <alex (a] botch2.com> 20408 Date: Thu Apr 10 16:51:04 2008 -0400 20409 20410 R300+: textured video tcl cleanup 20411 20412 commit 79c8d4ca36a1c3e5fe759d4ccc379c36af8f1676 20413 Author: Alex Deucher <alex (a] botch2.com> 20414 Date: Thu Apr 10 16:28:18 2008 -0400 20415 20416 RADEON: cleanup 20417 20418 commit c4821a287d29a65f3bcb7d60dc72ec13c0384008 20419 Author: Alex Deucher <alex (a] botch2.com> 20420 Date: Thu Apr 10 16:20:17 2008 -0400 20421 20422 Revert "R3xx/R5xx: move more VAP, etc. state setup into common init3d() function" 20423 20424 This reverts commit 305a3310963a5dd07b3495015b06aa8c7c4e6b02. 20425 20426 Conflicts: 20427 20428 src/radeon_commonfuncs.c 20429 src/radeon_exa_render.c 20430 src/radeon_textured_videofuncs.c 20431 20432 commit 0032c80bf30bab189204e3e6929e18a19d753138 20433 Author: Alex Deucher <alex (a] botch2.com> 20434 Date: Thu Apr 10 14:35:00 2008 -0400 20435 20436 RADEON: store tcl status in driver rec 20437 20438 commit 9e2ffe66d106abe34a670d2edc9905ed62c485e8 20439 Author: Alex Deucher <alex (a] botch2.com> 20440 Date: Thu Apr 10 14:24:04 2008 -0400 20441 20442 R3xx+: use the right register for engine flush 20443 20444 commit e1a9f26c2d2cbca9ad159e723ec95b95be1ef349 20445 Author: Alex Deucher <alex (a] botch2.com> 20446 Date: Thu Apr 10 14:12:15 2008 -0400 20447 20448 R3xx+: minor textured video fixes 20449 20450 - set shader output swizzling correctly 20451 - flush the right cache register on r3xx+ 20452 20453 commit d79040906cd25bd494feb5901f465bbd050aa923 20454 Author: Alex Deucher <alex (a] botch2.com> 20455 Date: Thu Apr 10 13:59:58 2008 -0400 20456 20457 R3xx+: EXA/textured video fixes 20458 20459 - get pipe config based on GB_PIPE_SELECT where applicable 20460 (adapted from a similar patch from Dave) 20461 - only flush the dst cache after submitting vertices, freeing 20462 the cache lines stalls the pipe 20463 - no need to wait for 3D idle after submitting vertices 20464 - fix PURGE_CACHE() and PURGE_ZCACHE() for r3xx+ 20465 - fix depth 16 with EXA composite 20466 20467 commit 0a96173cc38e506728d4c3f2dd383ba56e856578 20468 Author: Michel Dnzer <michel (a] tungstengraphics.com> 20469 Date: Mon Apr 7 18:15:34 2008 +0200 20470 20471 Increase default CP timeout. 20472 20473 Helps avoid spurious timeouts causing problems, see 20474 http://bugs.freedesktop.org/show_bug.cgi?id=15203 . 20475 20476 commit 255fbf465f5e7db2609a5a151bfa810249db52a0 20477 Author: Owen W. Taylor <otaylor (a] fishsoup.net> 20478 Date: Thu Apr 3 02:25:41 2008 -0400 20479 20480 Fix rendering of transformed sources for REPEAT_NONE with EXA on >= R300. 20481 20482 Use the border color when possible, otherwise fall back to software. 20483 20484 commit bc0407e53237d7968808110bc0243076377acf6e 20485 Author: Alex Deucher <alex@cube.(none)> 20486 Date: Fri Apr 4 18:40:16 2008 -0400 20487 20488 ATOMBIOS: Add support for DynamicClocks option 20489 20490 This patch adds support for dynamic clock gating and static 20491 power management using the atom command tables. In some cases 20492 the bios may already set this up during post, so YMMV. 20493 20494 I was only able to test on desktop cards, so I haven't tested 20495 to see how much (if any) power this saves or how it affects the 20496 thermal footprint. 20497 20498 commit 5f5e21bb50555c56bd371576074c28c929307ff1 20499 Author: Alex Deucher <alex@cube.(none)> 20500 Date: Fri Apr 4 14:29:45 2008 -0400 20501 20502 RADEON: warning fixes 20503 20504 commit c8e9a973aaded24aad567a0e36d0c78a05d6b2fd 20505 Author: Alex Deucher <alex@cube.(none)> 20506 Date: Fri Apr 4 14:26:19 2008 -0400 20507 20508 RADEON: add some quirks 20509 20510 commit 091963a635b79884afe77c026eabb48972fbe175 20511 Author: Alex Deucher <alex (a] botch2.com> 20512 Date: Thu Apr 3 22:35:16 2008 -0400 20513 20514 Minor cleanup 20515 20516 commit 950e9860643c20acde0eca4e4ff26baacc1f2b69 20517 Author: Alex Deucher <alex (a] botch2.com> 20518 Date: Thu Apr 3 22:11:48 2008 -0400 20519 20520 Revert "RADEON: memmap rework 1" 20521 20522 This reverts commit dd8ee1b444f4b973a1e0fadca5f943f2162b5e94. 20523 20524 Conflicts: 20525 20526 src/radeon.h 20527 src/radeon_driver.c 20528 20529 This rework seems to have caused more trouble than it was worth. 20530 20531 commit 88a1fe4a94c5d11aff22734b21c89890e4428cd5 20532 Author: Alex Deucher <alex (a] botch2.com> 20533 Date: Thu Apr 3 22:04:43 2008 -0400 20534 20535 Revert "RADEON: remove driver rec copies of mc info, use save rec directly" 20536 20537 This reverts commit be0858a84fbdf74c0b844f462933a221d48c707d. 20538 20539 Conflicts: 20540 20541 src/radeon_driver.c 20542 20543 commit c40a7aa3989576a8144213e2f31b892d21df8686 20544 Author: Owen W. Taylor <otaylor (a] fishsoup.net> 20545 Date: Thu Apr 3 14:43:55 2008 -0400 20546 20547 R3xx/R5xx: Fix pitch and clamp mode for repeating textures 20548 20549 - We can always use TXPITCH on a R300 even when repeating, 20550 (previous check for pitch matching width was also wrong) 20551 - Fix clamp mode for repeating textures to be WRAP 20552 20553 commit a8593482c1f2e0f2dbac06c2e5325ba8c83ed9ff 20554 Author: Dave Airlie <airlied (a] redhat.com> 20555 Date: Wed Apr 2 09:58:05 2008 +1000 20556 20557 atombios: fix the dual-head hopefully. 20558 20559 tested on r600 with DVI and VGA 20560 20561 commit 61d883d116fab3e9b513432d65e705afc5bb39f1 20562 Author: Dave Airlie <airlied (a] redhat.com> 20563 Date: Wed Apr 2 09:57:38 2008 +1000 20564 20565 Revert "Revert "atombios: fixup the width/height to use the mode values not the scrn ones"" 20566 20567 This reverts commit fc9af578997b6f22ee8b17e83f37d98689291b0e. 20568 20569 I see your revert and raise you one... 20570 20571 commit fc9af578997b6f22ee8b17e83f37d98689291b0e 20572 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 20573 Date: Tue Apr 1 09:25:45 2008 -0400 20574 20575 Revert "atombios: fixup the width/height to use the mode values not the scrn ones" 20576 20577 This reverts commit c2b1c8b706a6c7c1fd0af80091958473133d54e7. 20578 20579 These registers hold surface size. Using the mode values 20580 breaks dualhead. 20581 20582 commit 959509dd54de053f526b534e379a46934127231f 20583 Author: Dave Airlie <airlied (a] redhat.com> 20584 Date: Mon Mar 31 14:29:44 2008 +1000 20585 20586 radeon: use correct DDC interfaces so quirks get applied 20587 20588 Radeon seemed to mess up applying certain quirks, hopefully this will fix it. 20589 20590 commit 18f5f1cd2f52afed89fc11ade0920f3dfea87306 20591 Author: Dave Airlie <airlied (a] redhat.com> 20592 Date: Mon Mar 31 14:11:49 2008 +1000 20593 20594 radeon: split quirks into separate function and new quirk for IBM RN50 20595 20596 Add a connector table quirk for the IBM RN50. 20597 20598 commit c2b1c8b706a6c7c1fd0af80091958473133d54e7 20599 Author: Dave Airlie <airlied (a] redhat.com> 20600 Date: Sun Mar 30 11:44:14 2008 +1000 20601 20602 atombios: fixup the width/height to use the mode values not the scrn ones 20603 20604 this fixes it properly, legacy appears to be okay. 20605 20606 commit c5edea3d8c9254d3a21e390b8309e39e4c9635db 20607 Author: Dave Airlie <airlied (a] redhat.com> 20608 Date: Sun Mar 30 11:11:22 2008 +1000 20609 20610 r500/r600: fix rotation to fill screen 20611 20612 I'm not 100% sure this is the correct fix (maybe we shouldn't be using scrn 20613 virtualX/Y)... this will fix it for now until I get more time. 20614 20615 commit 9c62c820ba45ebc14d5f36f5d7885863800b6adb 20616 Author: Michel Dnzer <michel (a] tungstengraphics.com> 20617 Date: Fri Mar 28 12:37:29 2008 +0100 20618 20619 Include config.h, so FGL_LINUX can actually be defined when it's tested... 20620 20621 commit a00d9260a85b94a522c442aee24bc5ea4dc31c5c 20622 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 20623 Date: Thu Mar 27 20:03:13 2008 -0400 20624 20625 RADEON: fix lid issues on AVIVO chips for real this time :) 20626 20627 commit f0e89c09074b2c7e641f73692bb39b0bf68eb49c 20628 Author: Alex Deucher <alex (a] botch2.com> 20629 Date: Thu Mar 27 19:15:18 2008 -0400 20630 20631 Revert "RADEON: attempt to fix lid issues" 20632 20633 This reverts commit 9b4473c1d830b88866dd22e8174a07195bd6fcf4. 20634 This doesn't help. 20635 20636 commit 1442d396b938049b83f009a78ddabe2bf85641b6 20637 Author: Dave Airlie <airlied (a] redhat.com> 20638 Date: Thu Mar 27 14:02:51 2008 +1000 20639 20640 radeon: size bios to max of bar vs 64k. 20641 20642 reported by dwmw2: rhbz 438299 20643 20644 commit de2f609ff0004ef8b74727bfebc2c74fb91205ea 20645 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 20646 Date: Wed Mar 26 18:35:21 2008 -0400 20647 20648 AVIVO: no need to call PreinitXv() on AVIVO chips as they have no overlay 20649 20650 commit 75884c257bc2bcfa5b498a77d4c403f09face036 20651 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 20652 Date: Wed Mar 26 18:16:47 2008 -0400 20653 20654 XAA: update message about render so as to not confuse users 20655 20656 commit 9b4473c1d830b88866dd22e8174a07195bd6fcf4 20657 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 20658 Date: Wed Mar 26 18:01:29 2008 -0400 20659 20660 RADEON: attempt to fix lid issues 20661 20662 On some laptops the bios attempts to re-program the chip 20663 when a lid event comes in. This should hopefully prevent 20664 the bios from doing that. 20665 20666 commit 8b144830fe9b4a0cee4745023de5e7d387070f60 20667 Author: Alex Deucher <alex@samba.(none)> 20668 Date: Tue Mar 25 01:15:05 2008 -0400 20669 20670 RV250: disable textured video due to HW bug 20671 20672 The YUV->RGB conversion in the texture engine is broken 20673 on RV250 so the colors come out wrong. 20674 20675 commit 1789f11ab91633d3928f8b71988d51ff44bda9d1 20676 Author: Alex Deucher <alex@samba.(none)> 20677 Date: Mon Mar 24 19:03:30 2008 -0400 20678 20679 R3xx/R5xx: flush PVS state before enabling pvs-bypass 20680 20681 commit 305a3310963a5dd07b3495015b06aa8c7c4e6b02 20682 Author: Alex Deucher <alex@samba.(none)> 20683 Date: Mon Mar 24 14:25:03 2008 -0400 20684 20685 R3xx/R5xx: move more VAP, etc. state setup into common init3d() function 20686 20687 Also some minor code cleanups 20688 20689 commit 399b1d405e602c62d6deebea6d7e1f38886cd8e2 20690 Author: Alex Deucher <alex@samba.(none)> 20691 Date: Mon Mar 24 13:04:57 2008 -0400 20692 20693 R3xx/R5xx: use non VAP/TCP for textured video 20694 20695 Just extra state to emit. 20696 20697 commit cd77ec18f32a7b36acb655c927bbfd7044019f97 20698 Author: Dave Airlie <airlied (a] redhat.com> 20699 Date: Mon Mar 24 18:42:21 2008 +1000 20700 20701 r300: don't bother with VAP/TCL for render. 20702 20703 We just send more data to the card to process per transaction, without getting 20704 any actual gains, as we already pre-compute the vertices without needing 20705 any clipping or transforms from the card. 20706 20707 Perhaps some stuff could be done on-card, but so far the code is a lot 20708 faster if we avoid sending this extra info. 20709 20710 pre: 150000 glyphs/sec 20711 post: 185000 glyphs/sec 20712 20713 commit 301c6739b88676a0c78fc72194e993f894b8dc28 20714 Author: Alex Deucher <alex (a] botch2.com> 20715 Date: Sun Mar 23 11:14:02 2008 -0400 20716 20717 RS4xx: Revert back to previous fifo settings for now 20718 20719 Setup of these registers needs more investigation. 20720 20721 commit 9bea60b3eb378de5e1d44cc02a2763f4feae7882 20722 Author: Alex Deucher <alex (a] botch2.com> 20723 Date: Sat Mar 22 11:46:15 2008 -0400 20724 20725 RS4xx: more work on disp/disp2 fifo setup 20726 20727 commit 90f11c3986c28daa7b600b9662da145af325d264 20728 Author: Alex Deucher <alex (a] botch2.com> 20729 Date: Sat Mar 22 11:29:51 2008 -0400 20730 20731 RS4xx: missed this on the last commit. 20732 20733 commit 6d5066a451017a2683addc9e2496987626795dda 20734 Author: Alex Deucher <alex@samba.(none)> 20735 Date: Fri Mar 21 16:21:54 2008 -0400 20736 20737 RS4xx: attempt to set up disp/disp2 fifos correctly 20738 20739 If you have an XPRESS chip, please test!!! 20740 20741 commit fb1cffac05ae20c8365b25a2042b0ae961880faf 20742 Author: Alex Deucher <alex@samba.(none)> 20743 Date: Fri Mar 21 15:24:36 2008 -0400 20744 20745 RS4xx: attempt to fix TMDS/DVO support 20746 20747 XPRESS chips added a second set of FP control registers. 20748 I don't have the hw to test however. 20749 20750 commit 5e3b21284482df9974c9a58f248f0100def2bb0c 20751 Author: Alex Deucher <alex@samba.(none)> 20752 Date: Wed Mar 19 19:15:05 2008 -0400 20753 20754 Disable the setting of HARDWARE_CURSOR_BIT_ORDER_MSBFIRST 20755 20756 See bug 11796 20757 20758 commit 17cd42ed31814ba329a6a68edd0d75390a7da40e 20759 Author: Matt Turner <mattst88 (a] gmail.com> 20760 Date: Wed Mar 19 18:17:10 2008 -0400 20761 20762 Enable BSR in Log2 functions 20763 20764 This patch edits RADEONLog2 and ATILog2 to use the x86 BSR instruction instead 20765 of looping through bits. It should provide a somewhat of a speed increase in 20766 this function on x86 and AMD64 architectures. 20767 20768 Note: the BSR instruction was added with the 80386 CPU and is therefore not 20769 compatible with earlier CPUs, though I highly doubt it's even possible to use a 20770 286 in conjunction with a Radeon. 20771 20772 The inline assembly also works with Intel's compiler (icc). 20773 20774 commit c83827b4d2b6f03c54429e757a756eb99ff8be6b 20775 Author: Paulo Cesar Pereira de Andrade <pcpa (a] mandriva.com.br> 20776 Date: Wed Mar 19 17:58:34 2008 -0400 20777 20778 [PATCH] Compile warning fixes. 20779 20780 Minor changes to avoid declarations mixed with code. 20781 Ansified functions with empty prototype to specify they don't 20782 receive arguments. 20783 Added some prototypes to radeon.h, and major reorder on radeon.h 20784 adding prototypes in alphabetical order and specifying to file that 20785 defines it. 20786 20787 commit bed9754ad21d6c0a7f61067b04ba31c430a7cecb 20788 Merge: 55e446b5b f71ac0e40 20789 Author: Alex Deucher <alex@samba.(none)> 20790 Date: Wed Mar 19 16:06:41 2008 -0400 20791 20792 Merge branch 'master' of ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati into r3xx-render 20793 20794 commit 55e446b5bc091e6c7b3c2e9ae20b45130555c246 20795 Author: Alex Deucher <alex@samba.(none)> 20796 Date: Wed Mar 19 13:15:32 2008 -0400 20797 20798 R3xx/R5xx: Make sure to clamp the output of the FS 20799 20800 commit b6aa4279cbe68cc8e4523795e9714fb798b62d98 20801 Author: Alex Deucher <alex@samba.(none)> 20802 Date: Wed Mar 19 12:45:01 2008 -0400 20803 20804 R5xx: bump textured video limits to 4096 20805 20806 commit 4a445a3e8c4c5ecd9d4ef8daa26906c3ceaa94a1 20807 Author: Alex Deucher <alex@samba.(none)> 20808 Date: Wed Mar 19 12:31:51 2008 -0400 20809 20810 RADEON: add new macros to distinguish between R3xx and R5xx 3D 20811 20812 commit 85d0c9e8d22ccc72bec87b3fd44da5d7609293e0 20813 Author: Alex Deucher <alex@samba.(none)> 20814 Date: Wed Mar 19 12:07:33 2008 -0400 20815 20816 RADEON: fixed textured video with XAA and tiling 20817 20818 commit f5951db7b3522e0fe6af7f46a170c9c9a60a9bff 20819 Author: Alex Deucher <alex@samba.(none)> 20820 Date: Wed Mar 19 12:01:50 2008 -0400 20821 20822 RV515: fix textured video and EXA Composite 20823 20824 There seems to be an issue with the PVS setup on RV515, but 20825 bypassing it seems to work fine. 20826 20827 commit 13573879fe56368ad06234712b677c23fabc56c6 20828 Author: Dave Airlie <airlied (a] redhat.com> 20829 Date: Wed Mar 19 15:06:47 2008 +1000 20830 20831 r500: make it work from startup. 20832 20833 I'm not sure why this worked or what is going wrong here, really the 20834 VAP internal architecture escapes me :) 20835 20836 commit d331dd64d644a18ec99a2136cd0943b5edca1f03 20837 Author: Alex Deucher <alex@samba.(none)> 20838 Date: Tue Mar 18 19:44:26 2008 -0400 20839 20840 R3xx/R5xx: remove extra return after last commit 20841 20842 commit bc34df7a9c35cdd38c49d5c22471f3f487a33d6e 20843 Author: Alex Deucher <alex@samba.(none)> 20844 Date: Tue Mar 18 19:39:47 2008 -0400 20845 20846 R3xx/R5xx: switch an ErrorF() to RADEONFALLBACK() 20847 20848 commit 6f03f8fe0ecf4181dcf125049cf63bece0451fb2 20849 Author: Alex Deucher <alex@samba.(none)> 20850 Date: Tue Mar 18 19:36:05 2008 -0400 20851 20852 R3xx: we only use 2 temps, not 3 20853 20854 commit 8bb71ab4a3eb4fb6ef7f709e87c8df387cb70ee3 20855 Author: Tilman Sauerbeck <tilman (a] code-monkey.de> 20856 Date: Tue Mar 18 14:36:08 2008 -0400 20857 20858 R3xx/R5xx: fix up a8-src-something_with_colors 20859 20860 commit c362591d9b496df30668543158e4de44de742dc3 20861 Author: Alex Deucher <alex@samba.(none)> 20862 Date: Tue Mar 18 11:15:17 2008 -0400 20863 20864 R3xx/R5xx: remove some cruft 20865 20866 commit 89fe6d2c7d7471e6088558130f6e49f46c31dd47 20867 Author: Dave Airlie <airlied (a] linux.ie> 20868 Date: Tue Mar 18 09:43:43 2008 -0400 20869 20870 R5xx: fix typ in r5xx render accel 20871 20872 This gets render working on r5xx 20873 20874 commit 79b40ebcd8dedfc83e484c1024beeeaccc6124f3 20875 Author: Alex Deucher <alex@samba.(none)> 20876 Date: Tue Mar 18 02:46:49 2008 -0400 20877 20878 R5xx: first pass at render support (untested) 20879 20880 commit 71292c8f193230255d1d980c2e996bb01d04fab6 20881 Author: Alex Deucher <alex@samba.(none)> 20882 Date: Tue Mar 18 00:45:37 2008 -0400 20883 20884 R5xx: bump tex/dst limits to 4096 20885 20886 commit 30b52f8aa6a471455284f59b5b27252743892b13 20887 Author: Alex Deucher <alex@samba.(none)> 20888 Date: Mon Mar 17 23:20:10 2008 -0400 20889 20890 R3xx/R5xx: whitespace cleanup and cruft removal 20891 20892 commit 9c9f1b538ed710c3066775fba0a8e936b63087b1 20893 Author: Alex Deucher <alex@samba.(none)> 20894 Date: Mon Mar 17 23:01:37 2008 -0400 20895 20896 R3xx: get masks working and cleanup 20897 20898 RS offset was wrong for mask texture 20899 20900 commit ef94febd74f8ee63081b61e42f093a5a2b8fbf1e 20901 Author: Alex Deucher <alex@samba.(none)> 20902 Date: Mon Mar 17 22:27:19 2008 -0400 20903 20904 R3xx: minor adjustments 20905 20906 commit f71ac0e40b9d950bcb3bba42a75d41f45b6ed1bf 20907 Author: Alban Browaeys <prahal (a] yahoo.com> 20908 Date: Mon Mar 17 20:48:48 2008 -0400 20909 20910 RADEON: Revert to old behavior when resetting the memmap on VT switch 20911 20912 Not sure why this needs to be done twice. Should fix bug 14980 20913 Probably needs more investigation. 20914 20915 commit bedbbf196dc97ee5142e7dfae16fb6f317fca5a7 20916 Author: Alex Deucher <alex@samba.(none)> 20917 Date: Mon Mar 17 20:16:25 2008 -0400 20918 20919 R3xx: some progress 20920 20921 commit af0e626c132de2dd9958fec657fcc85d4c0fe5e1 20922 Author: Alex Deucher <alex@samba.(none)> 20923 Date: Mon Mar 17 18:07:12 2008 -0400 20924 20925 R3xx: fix errant w 20926 20927 commit 29ea5bfc0eb3194e2454fc3ee863df54f0300880 20928 Author: Alex Deucher <alex@cube.(none)> 20929 Date: Mon Mar 17 16:41:57 2008 -0400 20930 20931 RADEON: fix typo in RADEONAdjustMemMapRegisters() 20932 20933 commit ab317e85c5ab1a249a510c34aeb3a908be1a66fc 20934 Author: Alex Deucher <alex@cube.(none)> 20935 Date: Mon Mar 17 15:28:09 2008 -0400 20936 20937 RADEON: make sure var is initialized properly in RADEONAdjustMemMapRegisters() 20938 20939 commit 208d307227e15f37a6af5194398ed23266ff743a 20940 Author: Dave Airlie <airlied (a] linux.ie> 20941 Date: Sun Mar 16 19:39:23 2008 +1000 20942 20943 radeon: the 0x5974 appears to be a mobility chip... 20944 20945 After debugging with partymola on #radeon, adding this allowed his 20946 Dell Vostro 1000 to work properly 20947 20948 commit 9bc7c2ec4048e1677547c1d60c51ccb954f7589a 20949 Author: Alex Deucher <alex@samba.(none)> 20950 Date: Fri Mar 14 20:12:22 2008 -0400 20951 20952 R3xx: odds and ends... 20953 20954 still not working. 20955 - swizzle US output for BGR formats 20956 - no need to write to temps in ALU ops, 20957 write to output only 20958 - flush the PVS before updating 20959 20960 commit 96bea7906c4706fcd57a9cd8f1ce3feab6ac676d 20961 Author: Alex Deucher <alex@samba.(none)> 20962 Date: Fri Mar 14 15:59:36 2008 -0400 20963 20964 R3xx: theoretical support for component alpha 20965 20966 masks are still broken so... 20967 20968 commit cffe3dcc8991cd7c457a9c1a9f41055aa9ea3436 20969 Author: Alex Deucher <alex@samba.(none)> 20970 Date: Fri Mar 14 14:37:43 2008 -0400 20971 20972 R3xx: VS WIP 20973 20974 commit b73f52a50dfd6ff8d92f04d6b510c39582c6ac67 20975 Author: Alex Deucher <alex@samba.(none)> 20976 Date: Fri Mar 14 14:20:49 2008 -0400 20977 20978 R3xx/R5xx: enable VS for mask texture 20979 20980 commit 569a14ca9be1e18fe9921edc816ac3dc32d6cca7 20981 Author: Alex Deucher <alex@samba.(none)> 20982 Date: Fri Mar 14 13:32:12 2008 -0400 20983 20984 R3xx/R5xx: Fix magic numbers in vertex shaders 20985 20986 commit 4878997529601d62e257aa1c9112bd460561de73 20987 Author: Alex Deucher <alex@samba.(none)> 20988 Date: Thu Mar 13 21:23:40 2008 -0400 20989 20990 R3xx: make sure to set the FS code size correctly 20991 20992 commit 22f46b88ef05afb6a6b6d70007ac4980a446430e 20993 Author: Alex Deucher <alex@samba.(none)> 20994 Date: Thu Mar 13 20:25:33 2008 -0400 20995 20996 R3xx: attempt to setup the rasterizer properly for mask texture 20997 20998 Not working yet 20999 21000 commit 081fc9e892fa3d2e07b7db65b2e2719646255463 21001 Author: Alex Deucher <alex@samba.(none)> 21002 Date: Thu Mar 13 18:38:26 2008 -0400 21003 21004 R3xx: more mask work 21005 21006 commit 2bf0236c03538ace3ce6d0e68f0829fc47d1385b 21007 Author: Alex Deucher <alex@samba.(none)> 21008 Date: Thu Mar 13 18:32:00 2008 -0400 21009 21010 R3xx: enable composite for non-mask cases 21011 21012 commit 74286ba41302107d2fc626fee2181f7c4bc18164 21013 Author: Alex Deucher <alex@samba.(none)> 21014 Date: Thu Mar 13 18:25:32 2008 -0400 21015 21016 R3xx: add basic mask support 21017 21018 commit a2bbe10d866567911b68f222b4758624bfe9bf84 21019 Author: Alex Deucher <alex@samba.(none)> 21020 Date: Thu Mar 13 18:16:53 2008 -0400 21021 21022 R300: setup source selects and output swizzling 21023 21024 commit b9974ecce7d1932595226004858b08a7a6b188dc 21025 Author: Alex Deucher <alex@samba.(none)> 21026 Date: Thu Mar 13 17:35:38 2008 -0400 21027 21028 R3xx: set the texture id and add some register info 21029 21030 commit 0ef700b7da5e554a0d0d166f3fde85ff45c9eb1f 21031 Author: Alex Deucher <alex@samba.(none)> 21032 Date: Thu Mar 13 17:02:25 2008 -0400 21033 21034 R3xx/R5xx: enable blending 21035 21036 commit b35c09a597c93a1d9f06ef0091c96822b0653f98 21037 Author: Dave Airlie <airlied (a] redhat.com> 21038 Date: Thu Mar 13 18:42:29 2008 +1000 21039 21040 xv: fixup XAA on r500 textured video 21041 21042 the XAA area should never end up tiled. This may break with nooffscreen pixmaps 21043 21044 commit d4446461c3630caff166456c351ace34f57cc119 21045 Author: Matt Turner <mattst88 (a] gmail.com> 21046 Date: Tue Mar 11 21:20:53 2008 -0400 21047 21048 Properly fix uninitialized variables warnings 21049 21050 According to commit 9fd13e6773371c82b9799a5bda7c96ffa5cafe8c to 21051 xf86-video-intel by Kristian Hgsberg, there is a better way to fix the 21052 possibly initialized variables warnings. This patch will use Kristian's fix. 21053 21054 commit 20adfd7390d9b1f100e0c4a14f175377b8335c82 21055 Author: Alex Deucher <alex@cube.(none)> 21056 Date: Tue Mar 11 20:09:35 2008 -0400 21057 21058 RADEON: enable output attributes that require a modeset immediately 21059 21060 This should fix bug 14915 21061 21062 commit 53ba7f5771b0b53fb0d3bc29d64bdd3813756d10 21063 Author: Alex Deucher <alex@cube.(none)> 21064 Date: Tue Mar 11 19:12:40 2008 -0400 21065 21066 RADEON: fix vblank interrupts after VT switch or suspend/resume 21067 21068 commit e946c097f0438afbea6f3dd37ee39d67d415708c 21069 Author: Matt Turner <mattst88 (a] gmail.com> 21070 Date: Tue Mar 11 19:07:58 2008 -0400 21071 21072 [PATCH] Fix a few warnings 21073 21074 commit 8e160508520c0a24ca90aad182f296461ca0d9b6 21075 Author: Alex Deucher <alex@cube.(none)> 21076 Date: Tue Mar 11 18:11:13 2008 -0400 21077 21078 DCE3: add support for PCIEPHY (untested) 21079 21080 commit fbded88a2925f9f049936dad0736721e7b84a6ee 21081 Author: Alex Deucher <alex@cube.(none)> 21082 Date: Tue Mar 11 14:10:31 2008 -0400 21083 21084 ATOM: remove some cruft 21085 21086 commit 3263f6e4a410281d620c288a92bb4521f7b6fc06 21087 Author: Alex Deucher <alex@cube.(none)> 21088 Date: Tue Mar 11 14:05:48 2008 -0400 21089 21090 DCE3: enable DPMS on DIG ports 21091 21092 commit eb90e235b58c94f3d4d75394725ab2fe246a42ff 21093 Author: Alex Deucher <alex@cube.(none)> 21094 Date: Tue Mar 11 13:53:54 2008 -0400 21095 21096 DCE3: adjust PLL for DCE3 chips 21097 21098 this fixes stability issues on digital outputs and certain modes. 21099 21100 commit 552615ccc5360baafb8bb41698c1ca27816fd4b2 21101 Author: Alex Deucher <alex@cube.(none)> 21102 Date: Tue Mar 11 13:38:29 2008 -0400 21103 21104 ATOMBIOS: enable load detection by default on both DACs 21105 21106 Load detection is reliable with atom, so enable it by default 21107 on both DACA and DACB, rather than just DACA. 21108 21109 commit 78b10487cf222c96f8944ba25e2ea970506b3535 21110 Author: Alex Deucher <alex@cube.(none)> 21111 Date: Tue Mar 11 13:16:00 2008 -0400 21112 21113 DCE3: add output attribute to enable/disable coherent mode 21114 21115 Enabled by default. The TMDS transmitter can be programmed 21116 slightly differently depending on the chips in the panel. If you 21117 have problems with tmds on a particular panel, try disabling it. 21118 21119 commit d20be31c46fbec623af4c3628a7c603ceacf500f 21120 Author: Alex Deucher <alex@botch2.(none)> 21121 Date: Mon Mar 10 21:05:43 2008 -0400 21122 21123 RV550: MC setup is like RV515 not RV530 21124 21125 commit 38606b08b68842fbcc81c233009c1117269f3be9 21126 Author: Matthieu Herrb <matthieu (a] bluenote.herrb.net> 21127 Date: Sat Mar 8 23:22:59 2008 +0100 21128 21129 Makefile.am: nuke RCS Id 21130 21131 commit 9d710ee1a44cf2f3a948fbdbe17ef09521cbe744 21132 Author: Alex Deucher <alex@cube.(none)> 21133 Date: Fri Mar 7 15:09:14 2008 -0500 21134 21135 AVIVO: clean up some unused variables 21136 21137 commit c28c30c9f3d7bfebfd56a5c982c96f0090982054 21138 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 21139 Date: Fri Mar 7 14:10:49 2008 -0500 21140 21141 RADEON: Fix crash in last commit 21142 21143 commit c3a3635865e380c784a226c8ead069d4716d6b75 21144 Author: Dave Airlie <airlied (a] redhat.com> 21145 Date: Thu Mar 6 20:17:45 2008 -0500 21146 21147 RADEON: fix tiling/interlaced interaction with randr 1.2 21148 21149 commit df1b94dc4eb1f35b636dbf2ec0ab1c2da9937c0d 21150 Author: Alex Deucher <alex@botch2.(none)> 21151 Date: Thu Mar 6 19:22:08 2008 -0500 21152 21153 DCE3: Ignore outputs with DIN connectors for now 21154 21155 commit cb2dc19387c7b6494c47c76d683cf38a48700768 21156 Author: Alex Deucher <alex@cube.(none)> 21157 Date: Thu Mar 6 18:33:12 2008 -0500 21158 21159 AVIVO: fix typo from a previous commit 21160 21161 Leave tv dpms hook disabled or you may get bad interactions 21162 with the shared DAC 21163 21164 commit 77355de48057e5e7e0d5b3f3cf5a7a92220a53b1 21165 Author: Alex Deucher <alex@cube.(none)> 21166 Date: Thu Mar 6 17:46:00 2008 -0500 21167 21168 AVIVO: don't add outputs for invalid connectors 21169 21170 commit 600dbe080997a01ceaf6be86723189d518bc1281 21171 Merge: 594743a99 5b7875d0c 21172 Author: Alex Deucher <alex@cube.(none)> 21173 Date: Thu Mar 6 17:31:37 2008 -0500 21174 21175 Merge branch 'master' of ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati 21176 21177 commit 594743a99811a8b0f391412892414fcd158eeb56 21178 Author: Alex Deucher <alex@cube.(none)> 21179 Date: Thu Mar 6 17:30:21 2008 -0500 21180 21181 AVIVO: fix up memsize detection for IGP chips 21182 21183 commit 5b7875d0cbfbdbcd1515c4e942d30de298b49dff 21184 Author: Doug Chapman <doug.chapman (a] hp.com> 21185 Date: Thu Mar 6 14:31:06 2008 -0500 21186 21187 Bug #14826: Fix a bogus check around xf86SetOperatingState. 21188 21189 commit 651fe23f4c650ed91843dec48db24e18e8b91219 21190 Merge: 3de2dc88c 766f464df 21191 Author: Adam Jackson <ajax (a] redhat.com> 21192 Date: Thu Mar 6 14:28:27 2008 -0500 21193 21194 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 21195 21196 commit 41171c25cd235bafad26bcbabced16ead4b8c54b 21197 Author: Alex Deucher <alex@cube.(none)> 21198 Date: Thu Mar 6 14:05:18 2008 -0500 21199 21200 DCE3.0: add support for crtc memreq table 21201 21202 commit 766f464dfdfccadef23e4232f2bce5db22195513 21203 Author: Alex Deucher <alex@cube.(none)> 21204 Date: Thu Mar 6 13:35:43 2008 -0500 21205 21206 RADEON: take 2 on proper pragma pack support for bsds 21207 21208 See bug 14594. Based on suggestion by Henry Zhao 21209 21210 commit a842ce9ca6494e724a7828cead9b61c9ef02b6aa 21211 Author: Alex Deucher <alex@cube.(none)> 21212 Date: Thu Mar 6 12:32:18 2008 -0500 21213 21214 DCE3.0: Minor fixups 21215 21216 commit 8a1ba374033591c725a78923aa30829e4de2a5ae 21217 Author: Alex Deucher <alex@cube.(none)> 21218 Date: Thu Mar 6 09:53:51 2008 -0500 21219 21220 RADEON: option to override TVDAC adj values from bios with driver defaults 21221 21222 If you have a washed out image on the tv dac, try this option. 21223 Option "DefaultTVDACAdj" "TRUE" 21224 21225 commit 0ed48f8f651a28e189f9fee8c6b593da0178d21c 21226 Author: Alex Deucher <alex@cube.(none)> 21227 Date: Wed Mar 5 18:41:01 2008 -0500 21228 21229 AVIVO: Initial support for DCE 3.0 using atombios 21230 21231 DACs are working well, DIG support (DVI, HDMI, LVDS, etc.) 21232 still has some issues. 21233 21234 commit 2901e99f1942842856cd39c1dcc8b22f3cf7d9e3 21235 Author: Alex Deucher <alex@botch2.(none)> 21236 Date: Wed Mar 5 10:40:06 2008 -0500 21237 21238 RADEON: fix fetching of dac2 adj values from newer bios tables 21239 21240 commit 74eb981287d76836327830bd51272f605a07e0cc 21241 Author: Alex Deucher <alex@botch2.(none)> 21242 Date: Mon Mar 3 12:02:44 2008 -0500 21243 21244 ATOMBIOS: fix atombios parser support on *bsd 21245 21246 bsd requires a different pragma pack than Linux. 21247 See bug 14594. 21248 21249 commit f7769ea86e265f347eb58c517ccb5ef8b35eec27 21250 Author: Paulo Cesar Pereira de Andrade <pcpa (a] mandriva.com.br> 21251 Date: Sun Mar 2 14:49:21 2008 -0500 21252 21253 [PATCH] Ensure symbols used by other modules are visible. 21254 21255 The xf86-video-ati drivers are one of the cases where LoaderSymbol is 21256 widely used in some obscure ways. This patch fixes the problem, and 21257 allows compiling with -fvisibility=hidden. 21258 21259 commit a4398ac3ad77216f2c8aa628425bef5f2912a0a9 21260 Author: Alex Deucher <alex@cube.(none)> 21261 Date: Sat Mar 1 18:52:26 2008 -0500 21262 21263 RS6xx: change isIGP checks to CHIP_FAMILY_RS690 21264 21265 these paths are only relevant on RS6xx chips 21266 21267 commit 67d4d04836c05293b844bc505f303cfb04c0f8a4 21268 Author: Alex Deucher <alex@cube.(none)> 21269 Date: Sat Mar 1 18:33:18 2008 -0500 21270 21271 RADEON: use xf86SetDesiredModes() in screeninit and enterVT 21272 21273 this should restore the proper output state on VT switches 21274 21275 commit be0858a84fbdf74c0b844f462933a221d48c707d 21276 Author: Maciej Cencora <m.cencora (a] gmail.com> 21277 Date: Sat Mar 1 18:11:58 2008 -0500 21278 21279 RADEON: remove driver rec copies of mc info, use save rec directly 21280 21281 info->mc_* were used and the immediately copied into info->ModeReg 21282 ones. Just use the ModeReg copies directly. 21283 21284 commit dd8ee1b444f4b973a1e0fadca5f943f2162b5e94 21285 Author: Alex Deucher <alex@cube.(none)> 21286 Date: Sat Mar 1 16:23:51 2008 -0500 21287 21288 RADEON: memmap rework 1 21289 21290 Don't restore memmap regs on every mode switch. 21291 Just do memmap save/restore/setup on server start and VT switch. 21292 21293 commit 1f6a23000001f3d1c21b5c04f94714a8caa7aa8b 21294 Author: Alex Deucher <alex@cube.(none)> 21295 Date: Sat Mar 1 15:53:42 2008 -0500 21296 21297 RADEON: only restore legacy dac regs on legacy radeons 21298 21299 commit dee6cef8e62d0651c00319e03eea92940fd24aa4 21300 Author: Alex Deucher <alex@botch2.(none)> 21301 Date: Sat Mar 1 14:39:32 2008 -0500 21302 21303 RS4xx: enable exa render accel and textured video 21304 21305 RS6xx paths seem to work fine on RS4xx 21306 21307 commit 129f737efe4e8d1a368e7db4b063bdcd9339cb09 21308 Author: Alex Deucher <alex@cube.(none)> 21309 Date: Sat Mar 1 14:32:30 2008 -0500 21310 21311 AVIVO: save/restore regs by block 21312 21313 Save/Restore the entire block for each output. 21314 This should fix VT switch problems. 21315 21316 commit b069aadaa63a95d7a71b5cfbab83577b49501094 21317 Author: Alex Deucher <alex@botch2.(none)> 21318 Date: Fri Feb 29 22:36:02 2008 -0500 21319 21320 AVIVO: LVDS panels need dithering enabled 21321 21322 Fixes bug 14760 21323 21324 commit fe87bdee815372b4b4d7d4c705e34681625b90f2 21325 Author: Alex Deucher <alex@botch2.(none)> 21326 Date: Fri Feb 29 13:10:13 2008 -0500 21327 21328 AVIVO: disable pageflipping on avivo chips until we have proper drm support 21329 21330 commit fb3678c7f511d539a51cd090cb8b5041d7d2ba26 21331 Author: Alex Deucher <alex@botch2.(none)> 21332 Date: Fri Feb 29 13:01:21 2008 -0500 21333 21334 R5xx: fix register count when sending fragment program for textured video 21335 21336 commit a66d37d1a896ec934989592457c2beff8e6f1639 21337 Author: Alex Deucher <alex@botch2.(none)> 21338 Date: Fri Feb 29 04:07:05 2008 -0500 21339 21340 fix off-by-one in last commit 21341 21342 commit e56062960be0c8d3947861dd5e0691fce6516b99 21343 Author: Alex Deucher <alex@cube.(none)> 21344 Date: Thu Feb 28 19:16:39 2008 -0500 21345 21346 AVIVO: save/restore scaler regs 21347 21348 commit ae1c39a9b3e666404d0931679c9078c2e125a8bc 21349 Author: Alex Deucher <alex@cube.(none)> 21350 Date: Thu Feb 28 18:53:55 2008 -0500 21351 21352 RS6xx: rework output parsing 21353 21354 Turns out it's not as complex as I originially thought. 21355 IGP chips just have non-standard GPIO entires for DDC. 21356 21357 commit d8d6c9fe4ae7e1ab67dd041a251e901d97c29ed6 21358 Author: Alex Deucher <alex@botch2.(none)> 21359 Date: Thu Feb 28 17:01:14 2008 -0500 21360 21361 RS6xx: fix typos in previous commit 21362 21363 Noted by Maciej Cencora on IRC 21364 21365 commit 46547ae8bdbc5c10f1fd028b95ec4c5c31a5b318 21366 Author: Alex Deucher <alex@cube.(none)> 21367 Date: Thu Feb 28 14:29:30 2008 -0500 21368 21369 AVIVO: disable dithering on DFPs 21370 21371 This should fix the color banding some people have noticed. 21372 Also save/restore DDIA regs on RS6xx 21373 21374 commit 72a53d6f20ac29b3baddb7d8af04f19b76d2e04f 21375 Author: Michel Dnzer <michel (a] tungstengraphics.com> 21376 Date: Thu Feb 28 17:38:04 2008 +0100 21377 21378 Handle EXA coordinate limits more cleverly. 21379 21380 Generally set the 2D engine limits, and only enforce the 3D engine limits in the 21381 CheckComposite hook. This should still prevent useless migration of pixmaps the 21382 3D engine can't handle but allows for basic acceleration of bigger ones. 21383 21384 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=14708 . 21385 21386 commit 5249f450a2487475a95531603cc8668db2c21c33 21387 Author: Michel Dnzer <michel (a] tungstengraphics.com> 21388 Date: Thu Feb 28 12:23:58 2008 +0100 21389 21390 Fix 16 bit packed YUV XVideo playback on big endian systems with DRI disabled. 21391 21392 http://bugs.freedesktop.org/show_bug.cgi?id=14668 21393 21394 commit e40d75fd8b2aece9dae8076fac822a4a83025fb2 21395 Author: Alex Deucher <alex@samba.(none)> 21396 Date: Wed Feb 27 22:53:10 2008 -0500 21397 21398 R500: fragment program clean up and magic number conversion 21399 21400 commit 140dadba36b2191f0e18e41dd987785abd5f55d2 21401 Author: Alex Deucher <alex@samba.(none)> 21402 Date: Wed Feb 27 22:21:12 2008 -0500 21403 21404 R300: fix up magic numbers in fragment program 21405 21406 commit e521476bb5e2dfabc93747e43eb911a8a101357e 21407 Author: Alex Deucher <alex@botch2.(none)> 21408 Date: Wed Feb 27 21:26:55 2008 -0500 21409 21410 R300/R400: bump up the clip limits for textured video 21411 21412 This allows up to 2560x2560 (hw limit) 21413 21414 commit 10db46f11d7e1c055c9ad6034c65ad163dad17dc 21415 Author: Alex Deucher <alex@samba.(none)> 21416 Date: Wed Feb 27 15:28:50 2008 -0500 21417 21418 AVIVO: make sure we select the right LUT for each crtc 21419 21420 commit ea944f38dcfd871b27345698afea1cb986ecb049 21421 Author: Alex Deucher <alex@samba.(none)> 21422 Date: Wed Feb 27 14:37:52 2008 -0500 21423 21424 R300+: update RADEONCP_REFRESH() to reflect new location of scissor regs 21425 21426 commit b865faf95666e2172c3eec143f77fe9c524e4983 21427 Author: Alex Deucher <alex@samba.(none)> 21428 Date: Wed Feb 27 14:05:44 2008 -0500 21429 21430 R100/R200: move r100/r200 specific 3D setup into appropriate blocks 21431 21432 R3xx+ doesn't have these regs. 21433 21434 commit ee485ee0fc550414cd39cdb8fd2769fe0bc996a8 21435 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21436 Date: Wed Feb 27 18:41:31 2008 +0200 21437 21438 radeon: pass distcheck 21439 21440 commit ff110c781299fb729e93572ba3d630ca59e8127d 21441 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21442 Date: Wed Feb 27 18:41:17 2008 +0200 21443 21444 add warning for r128 21445 21446 commit b4a344f36332a12a51c9e896d93e9cb0672d31ef 21447 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21448 Date: Wed Feb 27 18:40:50 2008 +0200 21449 21450 add warning for mach64 21451 21452 commit 81ce4097c12ab64a52d0992f981f588f5a297b7b 21453 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21454 Date: Wed Feb 27 18:40:23 2008 +0200 21455 21456 drop r128 21457 21458 commit 873ff2033f1d0b84c91c48e829aa945f41b490b7 21459 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21460 Date: Wed Feb 27 18:40:09 2008 +0200 21461 21462 drop r128 (build system) 21463 21464 commit 2b8b2676e7b2fd2952f5307a934ef0a11006f838 21465 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21466 Date: Wed Feb 27 18:39:49 2008 +0200 21467 21468 drop mach64 21469 21470 commit 05bbbec11a04e7f6321b92c9dbec4ad4c0d130ff 21471 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21472 Date: Wed Feb 27 18:39:32 2008 +0200 21473 21474 drop mach64 (build system) 21475 21476 commit 68888189cf8d460ef6f8f2f1431a6ffe9fcd8134 21477 Author: Alex Deucher <alex@samba.(none)> 21478 Date: Wed Feb 27 03:09:28 2008 -0500 21479 21480 RADEON: fix textured video uploads 21481 21482 commit 7cfad216a3f1486be194c972b9b6bcf070358586 21483 Author: Alex Deucher <alex@samba.(none)> 21484 Date: Wed Feb 27 02:13:10 2008 -0500 21485 21486 RADEON: fix clipping for textured video 21487 21488 commit a2dca1d68d751def34ef3c6f836574173737bf76 21489 Author: Alex Deucher <alex@samba.(none)> 21490 Date: Tue Feb 26 16:38:29 2008 -0500 21491 21492 RADEON: Convert textured video to use pipelined uploads 21493 21494 commit b4fa1ce9d2da04d94521a82d3c2e95f0fe985ccc 21495 Author: Alex Deucher <alex@samba.(none)> 21496 Date: Tue Feb 26 14:45:25 2008 -0500 21497 21498 R300/R500: fix up clipping 21499 21500 Based on patch from Peter Zubaj. 21501 21502 commit 44e527a117ab0a363135ff066c7f7e0c12e3dc89 21503 Author: Alex Deucher <alex@samba.(none)> 21504 Date: Tue Feb 26 14:33:24 2008 -0500 21505 21506 R300: fix cordinate clamping in render code 21507 21508 Based on Peter's fix for textured video 21509 21510 commit 00ec17ad53d7ad43f19c9b723794ac1b8ef86826 21511 Author: Peter Zubaj <pzubaj at marticonet.sk> 21512 Date: Tue Feb 26 14:26:14 2008 -0500 21513 21514 R300: Fix clamping and pitch for textured video 21515 21516 commit 4207793f9b6c891cb17ba715223d2002668885e3 21517 Author: Alex Deucher <alex@samba.(none)> 21518 Date: Tue Feb 26 13:16:22 2008 -0500 21519 21520 R300/R500: set the number graphics pipes properly 21521 21522 This should fix the checkerboard issues on r300/r350 cards. 21523 21524 commit 823d8bf49e4c0a8c839354179942c5a1944c178d 21525 Author: Alex Deucher <alex@samba.(none)> 21526 Date: Tue Feb 26 12:25:14 2008 -0500 21527 21528 R300/R500: clean up magic numbers in textured video 21529 21530 commit 8ea75b268f11794f4a7e7bac52cb256490ed3fd2 21531 Author: Dave Airlie <airlied (a] linux.ie> 21532 Date: Tue Feb 26 16:29:19 2008 +1000 21533 21534 regs: fix spelling properly 21535 21536 commit f2816064a6c2c4c35ccba74b9aa80547e25c012e 21537 Author: Dave Airlie <airlied (a] linux.ie> 21538 Date: Tue Feb 26 16:28:24 2008 +1000 21539 21540 regs: fix spelling mistake 21541 21542 pointed out by plaes on irc 21543 21544 commit 9d2ca30b90607085578dde1f314db663bd5f82ec 21545 Author: Alex Deucher <alex@samba.(none)> 21546 Date: Mon Feb 25 17:34:00 2008 -0500 21547 21548 R300/R500: clean up magic numbers in render code 21549 21550 commit 153ad6fcf704cbf9f811d9986cd4baf04e82c9d2 21551 Author: Dave Airlie <airlied (a] linux.ie> 21552 Date: Mon Feb 25 07:10:48 2008 +1000 21553 21554 fixup check for EXA composite pointed out by Alan Swanson 21555 21556 commit 85e470e64f629de72e361c77770e2e29998d1bf4 21557 Merge: 27ddb39b1 1b84c76f2 21558 Author: Alex Deucher <alex@samba.(none)> 21559 Date: Sun Feb 24 05:37:22 2008 -0500 21560 21561 Merge master and fix conflicts 21562 21563 Merge branch 'master' of ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati 21564 21565 Conflicts: 21566 21567 src/radeon_commonfuncs.c 21568 21569 commit 27ddb39b12a0b54e099fd5274c4c91f08e2d2822 21570 Author: Alex Deucher <alex@samba.(none)> 21571 Date: Sun Feb 24 05:30:11 2008 -0500 21572 21573 R300: clean up magic numbers in RADEONInit3DEngine 21574 21575 commit d4c20f33ad6a1f88615cd7e09ad3638896873f9e 21576 Author: Alex Deucher <alex@samba.(none)> 21577 Date: Sun Feb 24 04:46:10 2008 -0500 21578 21579 R300: replace magic numbers in cache flush 21580 21581 commit e52f1c8d2647b81d891ec0728dd582941a76c83f 21582 Author: Alex Deucher <alex@samba.(none)> 21583 Date: Sun Feb 24 04:43:18 2008 -0500 21584 21585 R300: fill in some more 3D bitfields 21586 21587 commit 1b84c76f27c8d24cb42beae26abf000721901c1c 21588 Author: Dave Airlie <airlied (a] redhat.com> 21589 Date: Sun Feb 24 19:20:36 2008 +1100 21590 21591 rs690: initial textured video support 21592 21593 commit 68158124366db883a901e960fe5f05f8df5efa42 21594 Author: Dave Airlie <airlied (a] redhat.com> 21595 Date: Tue Feb 19 19:51:18 2008 +1100 21596 21597 rs690: initial rs690 rotate 21598 21599 suffers same problem as r500 with clipping 21600 21601 commit e614bb6965588bf09dcb87f5e08e67120ec9847f 21602 Author: Dave Airlie <airlied (a] redhat.com> 21603 Date: Sun Feb 24 01:46:05 2008 -0500 21604 21605 r500: convert fragprog to use register values 21606 21607 commit 6ce9ee47c75620b2e5d211c5d59d17271a6a7b19 21608 Author: Dave Airlie <airlied (a] redhat.com> 21609 Date: Sat Feb 23 22:49:34 2008 -0500 21610 21611 r500: add textured video Xv adapter support 21612 21613 commit 9aaf8b33b22b6ba112869558ae54e021b9487ad2 21614 Author: Dave Airlie <airlied (a] redhat.com> 21615 Date: Sat Feb 23 22:16:25 2008 -0500 21616 21617 r500: initial rotate support - not fully working yet. 21618 21619 Just an example of how to setup and run the r500 3D engine for rotation. 21620 this rotates for me but I get some strange clipping on the bottom of my screen 21621 21622 commit 05dc3e4fc19d056ce99a7b110665adab2ca1ea21 21623 Author: Adam Jackson <ajax (a] redhat.com> 21624 Date: Sat Feb 23 20:29:51 2008 -0500 21625 21626 Clarify R500 US setup. 21627 21628 commit 080606ad528972623e1ed1124d8efe7705a73446 21629 Author: Adam Jackson <ajax (a] redhat.com> 21630 Date: Sat Feb 23 20:21:17 2008 -0500 21631 21632 Add the R500 US index/data pair. 21633 21634 commit edfb3b6bbf0ee17ace8e6ba704a6f54e249fec63 21635 Author: Alex Deucher <alex@samba.(none)> 21636 Date: Sat Feb 23 19:59:33 2008 -0500 21637 21638 RADEON: no textured video yet on XPRESS chips 21639 21640 Still need to sort out the VAP and PVS stuff 21641 21642 commit 4146bfe5d00e40a86d17826fac50d04b2469621d 21643 Author: Alex Deucher <alex@samba.(none)> 21644 Date: Sat Feb 23 19:21:52 2008 -0500 21645 21646 R500: fix typo in new r5xx regs 21647 21648 commit d9be9f34b0d3313e7b22b2a8bb0a8924ad3116bf 21649 Author: Alex Deucher <alex@samba.(none)> 21650 Date: Sat Feb 23 19:06:30 2008 -0500 21651 21652 RADEON: add textured video support for r1xx-r4xx radeons 21653 21654 Based on the kdrive ati video code by Eric Anholt. 21655 R3xx/R4xx still have some clipping issues in certain situations 21656 21657 commit 9dc4acad79196e9d5d94dd710773bfa83456d47f 21658 Author: Alex Deucher <alex@cube.(none)> 21659 Date: Sat Feb 23 18:29:00 2008 -0500 21660 21661 RS6xx: gpio entry for DDIA varies depending on the number of DFP ports 21662 21663 commit ed0a93edf28155308e7ab9d8705581bb38455ea0 21664 Author: Adam Jackson <ajax (a] redhat.com> 21665 Date: Sat Feb 23 15:02:17 2008 -0500 21666 21667 Fix R500_US_CONFIG. 21668 21669 commit bc2bd6f841b51aeed3b6b4a47dbe758c200bc5a6 21670 Author: Adam Jackson <ajax (a] redhat.com> 21671 Date: Sat Feb 23 14:34:18 2008 -0500 21672 21673 Add R500 unified shader register block. 21674 21675 commit a7b5c3bb74fc4de5e38a75ac31656445ce823464 21676 Author: Alex Deucher <alex@cube.(none)> 21677 Date: Fri Feb 22 19:35:11 2008 -0500 21678 21679 RS6xx: fix DDC on DDIA output (usually HDMI port) 21680 21681 commit 3327a681e21101cc6f6e162f4e29f9937b69ccc3 21682 Author: Alex Deucher <alex@cube.(none)> 21683 Date: Fri Feb 22 17:05:56 2008 -0500 21684 21685 ATOM: properly set up DDIA output on RS6xx boards 21686 21687 commit 1d0e9ab8b9451101b1b91943546f6c5833c21b3f 21688 Author: Michel Dnzer <michel (a] tungstengraphics.com> 21689 Date: Wed Feb 20 10:21:49 2008 +0100 21690 21691 radeon: Fix typo flagged by gcc -Wall. 21692 21693 commit b5bd442b60dbc72fe4c1e928ab864aeb0fd7a3cb 21694 Author: Alex Deucher <alex@botch2.(none)> 21695 Date: Tue Feb 19 20:47:40 2008 -0500 21696 21697 R100: fix render accel for transforms 21698 21699 Not sure why we had a separate broken path for r100 vertex 21700 submission. 21701 21702 commit a0a73208a21546ac120fb9a463261836c9ea7b55 21703 Author: Alex Deucher <alex@botch2.(none)> 21704 Date: Tue Feb 19 20:11:19 2008 -0500 21705 21706 RADEON: restore clock gating and CP clock errata on VT switch 21707 21708 This may help people with hangs on resume 21709 21710 commit b77e2aff7453a9f370beba37ca3c25b92b3f97ff 21711 Author: Alex Deucher <alex@botch2.(none)> 21712 Date: Tue Feb 19 19:55:41 2008 -0500 21713 21714 RADEON: fix DDC types 5 and 6 21715 21716 commit af82172a82f2bdf96e571def659a1c70f92dfdbf 21717 Author: Alex Deucher <alex@botch2.(none)> 21718 Date: Tue Feb 19 19:39:35 2008 -0500 21719 21720 RADEON: update man page with supported chips 21721 21722 commit 1302d1b019c8091986aaf9c86b25c1f36a037788 21723 Author: Alex Deucher <alex@botch2.(none)> 21724 Date: Tue Feb 19 00:34:31 2008 -0500 21725 21726 Bump for 6.8 release 21727 21728 commit 830d29e8b280c11e798fb5bf5278bf24ecbac7bf 21729 Author: Alex Deucher <alex@cube.(none)> 21730 Date: Mon Feb 18 21:17:12 2008 -0500 21731 21732 ATOM: fix RS6xx connector table parsing and fix some warnings 21733 21734 commit 03aa4cc6d6e8c715a1c1d677cc1845223505b358 21735 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 21736 Date: Mon Feb 18 20:19:58 2008 -0500 21737 21738 bring back to life planar-to-packed conversion for rs4xx 21739 21740 Here's a patch to bring back the code for converting planar yuv to packed yuv, 21741 if a RS400 family chip is used (though I've no idea if they really all fail 21742 with planar yuv). 21743 21744 fixes bug 12744 21745 21746 commit 66b0b3f43bb714bc007169d5ba6dabe34cfc008b 21747 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21748 Date: Mon Feb 18 10:33:59 2008 +0200 21749 21750 r128/radeon: probably correct fix for non pci access code 21751 21752 commit d055b9e800ae50d08cca9db75fc666ce1da9ab52 21753 Author: Dave Airlie <airlied (a] linux.ie> 21754 Date: Sat Feb 16 09:13:43 2008 +1000 21755 21756 r128/radeon: hopeful fix for non pci access code 21757 21758 commit c773bc6a314327da29e21e4ebac6fa7f3e98a0a7 21759 Author: Dave Airlie <airlied (a] linux.ie> 21760 Date: Sat Feb 16 09:05:07 2008 +1000 21761 21762 r128/radeon: fix build without pciaccess 21763 21764 commit 690a52da5248f47a587a878d05fce9784957970b 21765 Author: Dave Airlie <airlied (a] linux.ie> 21766 Date: Sat Feb 16 08:33:36 2008 +1000 21767 21768 mach64: fix non pciaccess build 21769 21770 commit fc85188fd95bf78b7f965cdde3e22b644c74ff81 21771 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21772 Date: Fri Feb 15 18:36:10 2008 +0200 21773 21774 ati: convert to pci probe 21775 21776 add pciids for each subdriver, make no use of the match_data functionality. 21777 21778 thanks to Alex Deucher for reviewing and testing. 21779 21780 commit 665bd7e2f61cac3e029bbad5024034e5136deec1 21781 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21782 Date: Fri Feb 15 18:34:56 2008 +0200 21783 21784 ati wrapper: translate Device lines 21785 21786 commit a596c1618f72179a45289a50a1f9e89462ce9667 21787 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21788 Date: Fri Feb 15 18:34:31 2008 +0200 21789 21790 r128: do not compile in PciChipsets twice 21791 21792 commit 99cd8ff9a7e15fc2b4e55f8bc020f584173a8c2d 21793 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21794 Date: Fri Feb 15 18:33:51 2008 +0200 21795 21796 ati: drop duplicate pci-id defines 21797 21798 commit 146b01b51069dc227d0b125babb3f6957c9b9de2 21799 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21800 Date: Fri Feb 15 18:33:29 2008 +0200 21801 21802 mach64: clean probe a little 21803 21804 do not report I/O ports now, they are reported later. 21805 21806 commit f47d461331a032f9bdcf6f63336e848778cec6cc 21807 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21808 Date: Fri Feb 15 18:32:56 2008 +0200 21809 21810 mach64: minor cosmetic, I2C type 21811 21812 commit 2c66f2e812195167df9ca113044d46deece776ac 21813 Author: Dave Airlie <airlied (a] clockmaker.usersys.redhat.com> 21814 Date: Fri Feb 15 10:59:30 2008 +1000 21815 21816 make distcheck pass 21817 21818 commit ed87f367ddab7366f84917b43b31d76df4ce1110 21819 Author: Alex Deucher <alex@botch2.(none)> 21820 Date: Wed Feb 13 12:53:46 2008 -0500 21821 21822 RADEON: disable LVDS if panel size is 0x0 21823 21824 if we can't get the panel size from the bios or edid 21825 or a user specified option, assume it's not connected. 21826 21827 commit 422d7f441fdbb318d22d44db661ac9bd61387bd2 21828 Author: Alex Deucher <alex@cube.(none)> 21829 Date: Tue Feb 12 15:35:46 2008 -0500 21830 21831 R6xx: when both crtcs are in use make sure they are both enabled 21832 21833 sometimes setting a mode on one crtc can cuase a blank screen on the other. 21834 make sure they are both enabled if they should be. 21835 21836 commit 860f5af75274cb236f536e1da09da6bd9a579530 21837 Author: Alex Deucher <alex@cube.(none)> 21838 Date: Tue Feb 12 14:46:49 2008 -0500 21839 21840 R6xx: fix up use of bios scratch regs to reflect the new offsets 21841 21842 commit 32f2119b43a0faf6069d8cc0816f0d9f7914c07f 21843 Author: Alex Deucher <alex@cube.(none)> 21844 Date: Tue Feb 12 14:20:02 2008 -0500 21845 21846 R6xx: bios scratch regs moved. 21847 21848 commit 8d64be6ebd7f50d4bcb587afeee8252c1367dc77 21849 Author: Alex Deucher <alex@samba.(none)> 21850 Date: Tue Feb 12 12:53:09 2008 -0500 21851 21852 RADEON: make sure we always set up the FP crtc shadow registers for crtc0 21853 21854 The behavior changed when I added rmx center mode support. In cases where 21855 crtc0 drives a DAC this can lead to a blank screen. 21856 21857 commit 810d192ee046077a894e0fb5f2dfd6a7c0130766 21858 Author: Alex Deucher <alex@cube.(none)> 21859 Date: Mon Feb 11 19:23:01 2008 -0500 21860 21861 R6xx: add missing objects 21862 21863 commit 85043439426e534e561259ce98bebdd8508b36a9 21864 Author: Alex Deucher <alex@botch2.(none)> 21865 Date: Mon Feb 11 16:36:58 2008 -0500 21866 21867 R6xx: make sure we set up the HDP base properly 21868 21869 commit 9ab5d2ec7c583c74f364d7cfbb54bcd2cd8ae2f5 21870 Author: Alex Deucher <alex@botch2.(none)> 21871 Date: Mon Feb 11 15:26:51 2008 -0500 21872 21873 RADEON: always restore crtc1 before crtc0 when using both crtcs 21874 21875 In some rare cases restoring crtc0 first will result in a blank screen 21876 on crtc1. If you are having issues with a blank screen on crtc1 21877 that used to work on 6.6.3 or before, this should help. 21878 21879 commit e33edca75bd9df0aa19a33e74c38a6d02610befd 21880 Author: Alex Deucher <alex@botch2.(none)> 21881 Date: Mon Feb 11 00:33:12 2008 -0500 21882 21883 RADEON: Implement proper (hopefully) BIOS/DRIVER interaction for ATOM bios 21884 21885 Tested on atom-based Desktop cards. It'd nice to get some testing 21886 on atom-based laptops. 21887 21888 commit 6524e33435a786f7de0064cdd1b04c1120d21593 21889 Author: Alex Deucher <alex@botch2.(none)> 21890 Date: Sun Feb 10 18:52:52 2008 -0500 21891 21892 RADEON: Implement proper (hopefully) BIOS/DRIVER interaction for COM bios 21893 21894 Tested on my M10-based laptop. 21895 21896 commit 8606c1bd175893c77e8c758246a1aed166c8be51 21897 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21898 Date: Fri Feb 8 19:07:03 2008 +0200 21899 21900 mach64: factor out BIOS panel info parsing 21901 21902 commit 9f33218c80f5a6d6d9464aa3db8ae25a4759f351 21903 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21904 Date: Fri Feb 8 19:06:41 2008 +0200 21905 21906 mach64: minor cosmetic, LCD panel id 21907 21908 commit 8cd5a465a03834b4b2f3a19f8d36fc9d2e18c6d4 21909 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21910 Date: Fri Feb 8 19:06:27 2008 +0200 21911 21912 mach64: minor cosmetic, DAC type 21913 21914 commit 7f4db96123fdcba9874226352802d702c2a498bd 21915 Author: Dave Airlie <airlied (a] redhat.com> 21916 Date: Fri Feb 8 10:47:10 2008 +1000 21917 21918 add rv670 pciids 21919 21920 commit e8899b9978291c62a65f468c92f340f65ad5479d 21921 Author: Alex Deucher <alex@botch2.(none)> 21922 Date: Thu Feb 7 19:27:38 2008 -0500 21923 21924 R6xx: fix ddc after my i2c rework 21925 21926 Seems r6xx does something different for its i2c table, 21927 revert to the old behavior for now. 21928 21929 commit 435de6c4e46ff2bebd4cee58888a66b936cd3fdf 21930 Author: Alex Deucher <alex@samba.(none)> 21931 Date: Thu Feb 7 19:14:13 2008 -0500 21932 21933 RADEON: sync up with latest upstream versions 21934 21935 atombios.h 21936 ObjectID.h 21937 21938 commit 692789a293970f70b88ccb6adcf0676d8b495ae2 21939 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21940 Date: Thu Feb 7 18:03:37 2008 +0200 21941 21942 mach64: factor out BIOS multimedia parsing 21943 21944 commit 933328ffd6d1d872a18d3de8624c4df845a64588 21945 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21946 Date: Thu Feb 7 18:03:03 2008 +0200 21947 21948 mach64: complement hint for sparc 21949 21950 commit 956c8c81f3ff434930a0cb17b027b2f8e4eeabb2 21951 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21952 Date: Thu Feb 7 18:02:31 2008 +0200 21953 21954 mach64: consolidate refclk #2 21955 21956 commit dce4cc26a8e2bf53805ec63763243f3ff6b4a6d3 21957 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21958 Date: Thu Feb 7 18:02:17 2008 +0200 21959 21960 mach64: consolidate refclk #1 21961 21962 commit f7ed807f0d82a7446ebc4acdd4e94df44a675f19 21963 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21964 Date: Thu Feb 7 18:01:59 2008 +0200 21965 21966 mach64: cosmetic 21967 21968 commit cda1cd198f33c26ef1b51532a2126468369743b8 21969 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21970 Date: Thu Feb 7 18:01:33 2008 +0200 21971 21972 mach64: factor out BIOS clock parsing 21973 21974 commit 73ff279469be9c7cbf9f533b85fcb553694ff413 21975 Author: George Sapountzis <gsap7 (a] yahoo.gr> 21976 Date: Thu Feb 7 18:00:55 2008 +0200 21977 21978 mach64: BIOSBase is no longer used 21979 21980 commit caea326cc6f1932bb299f451be013651a5749ea7 21981 Author: Dave Airlie <airlied (a] linux.ie> 21982 Date: Wed Feb 6 06:36:13 2008 +1000 21983 21984 r300: move fragprog setup code to prepare composite for now 21985 21986 commit 470cd6a401c6a3e8fea981a8fe97c28be3cfb81d 21987 Author: Dave Airlie <airlied (a] linux.ie> 21988 Date: Wed Feb 6 06:04:13 2008 +1000 21989 21990 r300: remove r300 specific vertex emission 21991 21992 Set the vertex and fragment engine to expect the mask coords. 21993 21994 commit f65e8dfac23adfa199026765fe3a1ea08cf4da67 21995 Author: Alex Deucher <alex@cube.(none)> 21996 Date: Sun Feb 3 00:09:59 2008 -0500 21997 21998 RADEON: rework i2c handling 21999 22000 Split out clk, data, and lock regs and masks. some cards use different 22001 regs and masks for each. For cards with ATOMBIOS, use the i2c bios 22002 table to grab the i2c data. 22003 22004 commit a38a903debc0a50dbc73f59dc2741bbea76d2bd9 22005 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22006 Date: Fri Feb 1 02:59:07 2008 -0500 22007 22008 RADEON: don't restore LVDS_PLL_CNTL for now 22009 22010 seems to cause problems with resume for some users. 22011 this needs further investigation. 22012 see bug 12596 22013 22014 commit 5d7bea2b62c86868baf1960665a40249a43bafc5 22015 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22016 Date: Fri Feb 1 02:53:22 2008 -0500 22017 22018 RADEON: remove redundant RADEONDisableDisplays() 22019 22020 use RADEONBlank() instead 22021 22022 commit 73b437ce232c94c0067a0d2f70538b6e1e8c07a7 22023 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22024 Date: Fri Feb 1 02:47:06 2008 -0500 22025 22026 RADEON: remove unused "aspect" scaler option 22027 22028 commit f1fb9e4daa29bc379f653f847254db1496b625fd 22029 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22030 Date: Fri Feb 1 02:39:18 2008 -0500 22031 22032 RADEON: Implement "center" mode for RMX on legacy radeons 22033 22034 commit bcd590103e04bfdb4f12413beacebf344f07e88e 22035 Author: Alex Deucher <alex@botch2.(none)> 22036 Date: Tue Jan 29 12:12:54 2008 -0500 22037 22038 RADEON: update man page to reflect AGP 1x default revert 22039 22040 commit 09348a83d06ba9d3129499c4daedd44a68771530 22041 Author: Alex Deucher <alex@botch2.(none)> 22042 Date: Tue Jan 29 12:09:24 2008 -0500 22043 22044 Seems the default is more reliable... we can't win. 22045 22046 Revert "radeon: Default to 1x again with non-v3 AGP cards." 22047 22048 This reverts commit b653e5a628bfa4dfb168e96f93f41eb910f409fb. 22049 22050 commit 0c26806245381b925b73ff9f3b747e3dcf0ebd6f 22051 Author: Alex Deucher <alex@botch2.(none)> 22052 Date: Tue Jan 29 10:26:48 2008 -0500 22053 22054 RADEON: Add new RV380 pci id 22055 22056 bug 14289 22057 22058 commit ce77ed78a877023da72dbe51609aef9a07e250b1 22059 Author: Maciej Cencora <m.cencora (a] gmail.com> 22060 Date: Mon Jan 28 19:02:56 2008 -0500 22061 22062 RS690: Implement MC idle check 22063 22064 commit b7de4ff52cfbdd85ee65000613632e21b92af24e 22065 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22066 Date: Sat Jan 26 19:28:05 2008 +0200 22067 22068 mach64: add hint for sparc and minor cosmetic. 22069 22070 commit 09d713aa0ed6367b4457420b3c2832fe8eca9b00 22071 Author: Alex Deucher <alex@botch2.(none)> 22072 Date: Wed Jan 23 18:05:41 2008 -0500 22073 22074 Revert "RADEON: adjust PAL mode hstart" 22075 22076 This reverts commit 719a9a376e34d99032af75e3f7b002670ccb816b. 22077 22078 This breaks TV out on some cards. 22079 22080 commit 719a9a376e34d99032af75e3f7b002670ccb816b 22081 Author: Alex Deucher <alex@botch2.(none)> 22082 Date: Sun Jan 20 18:40:53 2008 -0500 22083 22084 RADEON: adjust PAL mode hstart 22085 22086 Most people seem to get a more aligned picture with this setting 22087 22088 commit b2db8657fb888cff6d64c6dcb182caac389776ce 22089 Author: Alex Deucher <alex@botch2.(none)> 22090 Date: Sun Jan 20 18:33:22 2008 -0500 22091 22092 RADEON: re-work i2c for DDC 22093 22094 Unify the radeon/avivo paths and grab the data/clk masks from bios 22095 if available 22096 22097 commit a43003b24022a833e604f41b2873c0350b34181c 22098 Author: Alex Deucher <alex@botch2.(none)> 22099 Date: Sat Jan 19 18:49:53 2008 -0500 22100 22101 RADEON: get dac2 adj values from the bios tables 22102 22103 commit d4596c52ac9994be26e9ec2d7d57b3892c34abdb 22104 Author: Alex Deucher <alex@botch2.(none)> 22105 Date: Sat Jan 19 17:17:26 2008 -0500 22106 22107 RADEON: grab pll_in_min/pll_in_max from bios tables if available 22108 22109 commit 9a0947c812d0d38d1bca6a91140ac210831a6cb4 22110 Author: Alex Deucher <alex@botch2.(none)> 22111 Date: Sat Jan 19 13:57:50 2008 -0500 22112 22113 ATOM: Use LVDS edid from bios if available 22114 22115 commit 7238258c12def8ef273e5362f716d165f720c5a5 22116 Author: Kusanagi Kouichi <slash (a] ma.neweb.ne.jp> 22117 Date: Sat Jan 19 15:04:21 2008 +0100 22118 22119 radeon: Partial fix for XVideo RGB image distortions. 22120 22121 commit 32be3cf9d6c34e60ff8c3d6cfe9f73f1869c50e4 22122 Author: Brice Goglin <Brice.Goglin (a] ens-lyon.org> 22123 Date: Fri Jan 18 14:42:14 2008 -0500 22124 22125 RADEON: print the name of the output when printing the EDID 22126 22127 commit 4ba9430ee97dbce8f77db8de6ce9b753a75e453d 22128 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22129 Date: Fri Jan 18 18:18:18 2008 +0200 22130 22131 mach64: workaround for corruption at upper-left 22132 22133 commit possible workaround, it's reported multiple times ... 22134 22135 commit 12c00111b68c9cf4872a424258c6f8b7247aac47 22136 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22137 Date: Fri Jan 18 18:17:40 2008 +0200 22138 22139 use stand-alone drivers for each chip family. 22140 22141 Do not load the ati wrapper when the user specifies the sub-driver name in 22142 the Driver line of xorg.conf. Also, for -configure cause the wrapper to fail 22143 and let each sub-driver speak for themselves. 22144 22145 commit 24c7d134cd450f9e2cca85e4a2fc3253d250be04 22146 Author: Alex Deucher <alex@botch2.(none)> 22147 Date: Fri Jan 18 09:50:38 2008 -0500 22148 22149 RADEON: make sure EXA Composite is actually disabled on XPRESS chips. 22150 22151 commit eaf425436008092abe81208321a2b3b6698a5d79 22152 Author: Dave Airlie <airlied (a] redhat.com> 22153 Date: Fri Jan 18 20:11:57 2008 +1000 22154 22155 fixup register 6594 save/restore 22156 22157 commit 3de2dc88cf26ff5932f11cecdf975777b8aa2a4a 22158 Author: Adam Jackson <ajax (a] redhat.com> 22159 Date: Wed Jan 16 14:55:05 2008 -0500 22160 22161 Bump CRTC size limits on AVIVO chips so 30" displays work without tweaking. 22162 22163 Note that the CRTC size limits we're using right now are _not_ the 22164 hardware limits, they're just heuristics until we can resize the front 22165 buffer properly. 22166 22167 commit 2a89a31481d71a56a9930073cf99d3ae7b4290e1 22168 Author: Alex Deucher <alex@samba.(none)> 22169 Date: Thu Jan 17 15:08:17 2008 -0500 22170 22171 RADEON: use radeon_output->Flags for tracking RMX rather than mode->Flags 22172 22173 commit 495e3119250ffb48489debbaabe560d23753cc43 22174 Author: Alex Deucher <alex@samba.(none)> 22175 Date: Thu Jan 17 14:56:19 2008 -0500 22176 22177 AVIVO: Add support for RMX 22178 22179 Both centered and expansion modes are supported. Select 22180 using output attributes. 22181 22182 commit 6bd510a211f25d52e74791e4a429cd2218ced541 22183 Author: Alex Deucher <alex@samba.(none)> 22184 Date: Wed Jan 16 18:09:49 2008 -0500 22185 22186 RADEON: add a message about render accel on newer cards 22187 22188 commit 85bf3439fe2579aec48f5cd8d65a9d51b1ae8535 22189 Author: Alex Deucher <alex@samba.(none)> 22190 Date: Wed Jan 16 17:52:06 2008 -0500 22191 22192 R300: only init3d on r3xx 22193 22194 commit dbb2ca471dfbff245b30c5055871dee0dc0e3d15 22195 Author: Alex Deucher <alex@samba.(none)> 22196 Date: Wed Jan 16 17:10:02 2008 -0500 22197 22198 R300: only enable render accel on non-IGP r3xx/r4xx chips for now 22199 22200 commit 3c72b100bcfacee600644669b586e86cfd32754e 22201 Author: Alex Deucher <alex@samba.(none)> 22202 Date: Wed Jan 16 16:55:42 2008 -0500 22203 22204 R300: First pass at render accel 22205 22206 This first pass is pretty limited. All it currently supports 22207 is transforms for rotation. No blending yet. 22208 22209 Based on inital implementation from Wolke Liu with 22210 additional lock-up fixes by Dave Airlie. 22211 22212 commit 2ba3562d2af911fdd90881049599e239d27260bc 22213 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22214 Date: Sat Jan 12 17:11:59 2008 +0200 22215 22216 ati wrapper: xf86PciInfo.h is enough 22217 22218 commit c2caeb11a97dad5379d70881c5c0fd834a8c3d54 22219 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22220 Date: Sat Jan 12 16:18:34 2008 +0200 22221 22222 ati wrapper: add DriverRec's and use them 22223 22224 commit 19e1b180fec6f83a474e125465bc60111c0f43e0 22225 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22226 Date: Sat Jan 12 16:18:07 2008 +0200 22227 22228 mach64: load for both "ati" and "mach64" as driver names 22229 22230 similar to r128/radeon 22231 22232 commit 92f54400d5450b29b3a76d5ecc927cf0d73e156e 22233 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22234 Date: Sat Jan 12 16:17:47 2008 +0200 22235 22236 mach64: add version (need not match with ati) 22237 22238 similar to r128/radeon 22239 22240 commit 311ec7b6c54a50a4b8a5a445f7283da2b0b2e0f5 22241 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22242 Date: Sat Jan 12 16:17:21 2008 +0200 22243 22244 atimisc: rename to mach64 22245 22246 commit 5244e235262290eab8a3546f449295c12ed8f6c7 22247 Author: Alex Deucher <alex@samba.(none)> 22248 Date: Mon Jan 14 16:11:09 2008 -0500 22249 22250 ATOMBIOS: disable the scaler by default on avivo cards 22251 22252 Fixes bug 14001 22253 22254 commit 729da30c80d6545b75c8faea44754634f477cc09 22255 Merge: 000741e25 625a885a9 22256 Author: Alex Deucher <alex@samba.(none)> 22257 Date: Mon Jan 14 10:05:01 2008 -0500 22258 22259 Merge branch 'master' of ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati 22260 22261 commit 625a885a964488da2a35065eb597a17ee57b22a9 22262 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22263 Date: Wed Jan 9 17:39:56 2008 +0200 22264 22265 ati wrapper: use pci ids 22266 22267 commit 48865200ca9f1daf87e52620b9f8175f88dd886f 22268 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22269 Date: Wed Jan 9 19:01:04 2008 +0200 22270 22271 radeon: remove stray include 22272 22273 commit a5a6b873353611cb7e46e5e375f039957c7051a7 22274 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22275 Date: Wed Jan 9 13:03:32 2008 +0200 22276 22277 radeon: remove stray _X_EXPORT 22278 22279 commit 80c9974b6cdc0147d777df6990b3f3aacd87969d 22280 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22281 Date: Tue Jan 8 18:15:47 2008 +0200 22282 22283 r128: pci-rework conversion 22284 22285 compile-tested only 22286 22287 commit cd4b39403d74f314e1c2cfa4cf0e4fe542891dc3 22288 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22289 Date: Sat Mar 10 23:12:15 2007 +0200 22290 22291 Drop symbol lists from r128, radeon, theatre. 22292 22293 compile-tested only 22294 22295 commit 63b4b8213fabc5a57f897b60b6eaa9f78d86e6ff 22296 Author: George Sapountzis <gsap7 (a] yahoo.gr> 22297 Date: Wed Mar 21 20:24:22 2007 +0200 22298 22299 [mach64] Drop symbol lists. 22300 22301 egrep LOADER\|SymLists\|Symbols src/*.[hc] 22302 22303 not needed after server commit bca9364f3f4a2376edbcf57a34f704ce28be21ba, i.e. 22304 xserver 1.2 22305 22306 compile-tested only 22307 22308 commit 000741e250e54122b0adc91694eb4bfa320a70fb 22309 Author: Alex Deucher <alex@samba.(none)> 22310 Date: Thu Jan 10 14:49:48 2008 -0500 22311 22312 RADEON: clean up output handling 22313 22314 commit 10e7636c02478b8ffe183bb0c46229ca0d6584e1 22315 Author: Kristian Hgsberg <krh (a] bitplanet.net> 22316 Date: Wed Jan 9 12:47:39 2008 -0500 22317 22318 RADEON: fix crtc routing on r4xx cards when using atom to init DVO chip 22319 22320 commit 3af671f5963810dbfd63abc9889b1d46b68f404c 22321 Author: Alex Deucher <alex@botch2.(none)> 22322 Date: Wed Jan 9 11:30:25 2008 -0500 22323 22324 RADEON: restore FP2 regs before external encoders 22325 22326 This may fix krh's dvi problem 22327 22328 commit 2a54c6bb09ade2ec8f998dfc1624017029d47fa3 22329 Author: Alex Deucher <alex@botch2.(none)> 22330 Date: Tue Jan 8 18:43:54 2008 -0500 22331 22332 RADEON: Make default output actually work... 22333 22334 commit fa3e2055225c27e25465fc46786da1b7574fd3cc 22335 Author: Alex Deucher <alex@botch2.(none)> 22336 Date: Mon Jan 7 01:13:09 2008 -0500 22337 22338 RADEON: add default outputs if no connected devices 22339 22340 If no connected devices found at server startup, default 22341 to something so the server comes up. LVDS on mobility chips, 22342 DAC or TMDS on others. 22343 22344 commit d972cc9237eb90b49b11f8d2bdc5b61f628911dc 22345 Author: Alex Deucher <alex@botch2.(none)> 22346 Date: Sat Jan 5 17:19:06 2008 -0500 22347 22348 RADEON: Fix TVStandard option 22349 22350 commit 45656b9d5a426053da2a094de39c2690c0c6f799 22351 Author: Alex Deucher <alex@botch2.(none)> 22352 Date: Sat Jan 5 12:00:55 2008 -0500 22353 22354 R128: Like powerpc, don't use VGA by default on sparc 22355 22356 commit b8e8db4675d07e45782de0d7c67ee0fd85eaedb3 22357 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22358 Date: Fri Jan 4 20:16:19 2008 -0500 22359 22360 RADEON: fix tvdac load detection at server start up 22361 22362 commit 3ba7f393d0669df36848715799de8affc10a5534 22363 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22364 Date: Fri Jan 4 20:00:48 2008 -0500 22365 22366 RADEON: further fixup for pScrn->pScreen issue 22367 22368 the previous fix seems to cause the driver to hang on 22369 some cards. 22370 22371 commit a0de9c0844f9e066e0f02e8cd8045bdd278e6494 22372 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22373 Date: Fri Jan 4 19:48:30 2008 -0500 22374 22375 RADEON: improve ntsc image centering 22376 22377 commit 03d2f25801c8a8ec15030f06008df112d07c1a2d 22378 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22379 Date: Fri Jan 4 18:19:54 2008 -0500 22380 22381 RADEON: fix compile 22382 22383 commit 308848783ed9ae27aed7c7de6ee813d375ef495e 22384 Author: Dave Airlie <airlied (a] linux.ie> 22385 Date: Sat Jan 5 09:11:55 2008 +1000 22386 22387 i2c: a line mux of 0 is valid 22388 22389 commit 94d5a432f72801f821d1c4ce952baba17552659d 22390 Author: Dave Airlie <airlied (a] linux.ie> 22391 Date: Fri Jan 4 17:16:54 2008 +1000 22392 22393 atombios: i2c lines are all done with sw so we can accept any of them. 22394 22395 hch reported this working on his laptop with 0x7e60 as EDID for LVDS 22396 22397 commit 2e4473b63d65801ae8ac5a8391de232b2201d958 22398 Author: Alex Deucher <alex@samba.(none)> 22399 Date: Thu Jan 3 15:47:50 2008 -0500 22400 22401 RADEON: fix crash when setting rotation in the config file 22402 22403 xf86CrtcRotate() accesses pScrn->pScreen which is not set 22404 during ScreenInit(). This should also be fixed in the server. 22405 See bug 12129 22406 22407 commit c652208861bffca94f06b7f67688ce220e050bfb 22408 Author: Michel Dnzer <michel (a] tungstengraphics.com> 22409 Date: Thu Jan 3 17:54:58 2008 +0100 22410 22411 radeon: Adapt manpage to reality wrt AGP options. 22412 22413 commit ab451e4b7a5423d61b57cf0646599267d8504af4 22414 Author: Michel Dnzer <michel (a] tungstengraphics.com> 22415 Date: Thu Jan 3 17:52:39 2008 +0100 22416 22417 radeon: Miscellaneous warning fixes. 22418 22419 commit 394c52273328e90518568b694ee79dc1a8dab651 22420 Author: Dave Airlie <airlied (a] linux.ie> 22421 Date: Thu Jan 3 18:56:16 2008 +1000 22422 22423 r500: tvout avoid doing dpms here it makes my tv mode go all crappy 22424 22425 need to investigate further 22426 22427 commit 1c647279f021d01e110980727b7c7dd7efae1642 22428 Author: Dave Airlie <airlied (a] linux.ie> 22429 Date: Thu Jan 3 11:55:28 2008 +1000 22430 22431 r600: change devices list depending on connector 22432 22433 commit f36db6e10d32a68b32d20ae4ad02cfc0bfd1c9c3 22434 Author: Dave Airlie <airlied (a] linux.ie> 22435 Date: Thu Jan 3 11:27:47 2008 +1000 22436 22437 r600: fixup crash on unknown output type 22438 22439 not sure this is 100% the correct approach 22440 22441 commit 1accfdd590828e95e0d68a576c8ee05a06a86e43 22442 Author: Alex Deucher <alex@samba.(none)> 22443 Date: Wed Jan 2 19:48:28 2008 -0500 22444 22445 RADEON: various avivo tv-out related clean-ups 22446 22447 commit ce34090c758ac91171cb6adb9b8a36e4adbf99cf 22448 Merge: 2180f04b6 30cab1dbe 22449 Author: Alex Deucher <alex@samba.(none)> 22450 Date: Wed Jan 2 16:41:36 2008 -0500 22451 22452 Merge branch 'master' of git+ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati 22453 22454 commit 2180f04b6fb94a638f6274cb4455d5688b324dbc 22455 Author: Alex Deucher <alex@samba.(none)> 22456 Date: Wed Jan 2 16:41:06 2008 -0500 22457 22458 RADEON: preliminary support for ATOM RMX 22459 22460 not functional yet. 22461 22462 commit dab4dc285154d40303aadaa849b85f8e251e578e 22463 Author: Alex Deucher <alex@samba.(none)> 22464 Date: Wed Jan 2 16:27:19 2008 -0500 22465 22466 RADEON: add support for ATOM component video table 22467 22468 Component output is still not working. 22469 22470 commit 30cab1dbebb7bdb925f2fe9f91b84183312bbbfd 22471 Author: Alex Deucher <alex@botch2.(none)> 22472 Date: Wed Jan 2 15:41:01 2008 -0500 22473 22474 RADEON: Make sure all old IGP chips have HasSingleDac set 22475 22476 fix the csv file and re-gen the headers. 22477 22478 commit e8e585651215b011e3ad07c59d0eab9107ccd8c6 22479 Author: Dave Airlie <airlied (a] redhat.com> 22480 Date: Wed Jan 2 10:14:46 2008 +1000 22481 22482 PLL/r600: tweak pll to pick first one found instead of keeping going 22483 22484 commit 14aa4060ad27ecb3d40b2b17ee4cf7cc55a121dd 22485 Author: Dave Airlie <airlied (a] redhat.com> 22486 Date: Wed Jan 2 09:49:44 2008 +1000 22487 22488 r600: fix tv-out output naming 22489 22490 commit f65374f5e15bfd391a1838a686cd87d3bab8043d 22491 Author: Maciej Cencora <m.cencora (a] gmail.com> 22492 Date: Mon Dec 31 09:44:34 2007 +1000 22493 22494 atombios: initial rs690 patches 22495 22496 commit a674f683e6699c30664d9cd468a64de991c3fd7e 22497 Author: Dave Airlie <airlied (a] linux.ie> 22498 Date: Sun Dec 30 17:43:17 2007 +1000 22499 22500 atombios: enable TV-out detection code - tv out works for me with this 22501 22502 commit aa7c28cbd943bb525698515d444cb5097880e364 22503 Author: Dave Airlie <airlied (a] linux.ie> 22504 Date: Sun Dec 30 17:40:37 2007 +1000 22505 22506 atombios: enable support for using tv timings 22507 22508 enable support for the atombios tv timings selection by programming the crtc 22509 with the tv timings if a tv is detected on the output 22510 22511 commit 0bc3fd595a73e12a424571697d164a09a6a4c072 22512 Author: Dave Airlie <airlied (a] linux.ie> 22513 Date: Sun Dec 30 16:39:58 2007 +1000 22514 22515 atombios: add support for reading tv standard from atombios 22516 22517 fix typo in atombios header file 22518 22519 commit bfa22d676a6f333503104041f62222f4de9bb7d8 22520 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22521 Date: Wed Dec 26 12:29:47 2007 -0500 22522 22523 RADEON: fix typo that broke tv load detection 22524 22525 commit 0c99554c6fab1192f5e8595469c21b5f33e1eb4f 22526 Author: David Miller <davem (a] davemloft.net> 22527 Date: Wed Dec 26 02:19:12 2007 -0500 22528 22529 [RADEON]: Add missing break in SCLK calculation. 22530 22531 commit 6e0d5cc1c62fbfc1962fa0d6333f0c0a8c6327bd 22532 Author: David Miller <davem (a] davemloft.net> 22533 Date: Wed Dec 26 02:17:34 2007 -0500 22534 22535 [RADEON]: Like powerpc, don't use VGA by default on sparc. 22536 22537 commit 2b1fae668ddabbc72e5fc31365302ea722174df1 22538 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22539 Date: Wed Dec 26 02:04:06 2007 -0500 22540 22541 RADEON: fix PAL tv-out 22542 22543 Many thanks to Andrew Randrianasulu for providing me with 22544 pll reg dumps 22545 22546 commit d736eb5732da573162c70712dc4e8b0114986702 22547 Author: Alex Deucher <alex@botch2.(none)> 22548 Date: Mon Dec 24 12:34:15 2007 -0500 22549 22550 RADEON: default "IgnoreLidStatus" to TRUE 22551 22552 Seems there are lots of busted ACPI lid status and people 22553 starting X with the lid closed. 22554 22555 commit 653da558148cc601bc1f80253e92ef98c75ef37a 22556 Author: Alex Deucher <alex@botch2.(none)> 22557 Date: Mon Dec 24 01:11:56 2007 -0500 22558 22559 RADEON: restore crtc regs before VGA regs. 22560 22561 It seems some radeons don't restore text console properly if 22562 the crtc regs are restored after the VGA regs. 22563 Thanks to Sverre Froyen for helping me track this down 22564 22565 commit ad3325f672a241449ca239c8ee3a24b6d7703d76 22566 Author: Alex Deucher <alex@botch2.(none)> 22567 Date: Sun Dec 23 17:18:42 2007 -0500 22568 22569 RADEON: Add "IgnoreLidStatus" option 22570 22571 Generally, users that start X with the laptop lid closed 22572 want to use one or more external monitors rather than the 22573 internal panel and do not want the internal panel to be on 22574 by default. Others, it seems, want to always have the 22575 internal panel on, regardless of the lid. I can't win. 22576 Enable this option to force the latter. 22577 22578 commit 20eedf348a527e1e0a5450bc22d7564895034a66 22579 Author: Alex Deucher <alex@botch2.(none)> 22580 Date: Sun Dec 23 11:27:55 2007 -0500 22581 22582 RADEON: fix pll input setup on mac cards 22583 22584 the function was exiting before the complete setup was finished. 22585 22586 commit 4f2e833e8ebaba3ad85ec5314fff8fa05395b679 22587 Author: Arkadiusz Miskiewicz <arekm (a] maven.pl> 22588 Date: Fri Dec 21 18:56:34 2007 -0500 22589 22590 configure.ac fixes 22591 22592 commit 2b6e8e2b8f74e94560de89693ecbc7260536591e 22593 Author: Arkadiusz Miskiewicz <arekm (a] maven.pl> 22594 Date: Fri Dec 21 18:10:17 2007 -0500 22595 22596 RADEON: various cleanups 22597 22598 commit 4c6f60e3b19ac55ab1255c79df03b1df5950864e 22599 Author: Alex Deucher <alex@botch2.(none)> 22600 Date: Fri Dec 21 17:33:04 2007 -0500 22601 22602 RADEON: clean up prototypes 22603 22604 commit 8c761afdcb9baf1649b93449692fb9ab67bc2c80 22605 Author: Alex Deucher <alex@botch2.(none)> 22606 Date: Fri Dec 21 16:24:49 2007 -0500 22607 22608 RADEON: more re-org 22609 22610 move save/restore routines into legacy_crtc/output 22611 22612 commit 910773d3a6c717f9d4762ea7b9ee6c3ae061781e 22613 Author: Alex Deucher <alex@botch2.(none)> 22614 Date: Fri Dec 21 15:53:15 2007 -0500 22615 22616 RADEON: more re-org. move XAA Mem init to radeon_accel.c 22617 22618 commit 0631a23bd103f9b74e525da2c41304eab60c6f17 22619 Author: Alex Deucher <alex@botch2.(none)> 22620 Date: Fri Dec 21 15:40:18 2007 -0500 22621 22622 RADEON: fix rn50 reversion from last merge 22623 22624 commit 5b917797a13c6caa80028d1842a284598e874288 22625 Author: Alex Deucher <alex@botch2.(none)> 22626 Date: Fri Dec 21 15:30:20 2007 -0500 22627 22628 RADEON: remove no longer used radeon_display.c 22629 22630 commit bf14aa5f88fc3b4e69d71db5b23248b8bb2018d0 22631 Author: Alex Deucher <alex@botch2.(none)> 22632 Date: Fri Dec 21 15:29:47 2007 -0500 22633 22634 RADEON: re-org legacy crtc/output code into separate files 22635 22636 commit be7f8fd338f5af8b632f16a83db41e15d00af469 22637 Author: Dave Airlie <airlied (a] linux.ie> 22638 Date: Fri Dec 21 10:03:49 2007 +1000 22639 22640 fix mode bandwidth configure check 22641 22642 commit 3f9b597dedc45379b0bc0b631f3f924c403bca48 22643 Author: Dave Airlie <airlied (a] linux.ie> 22644 Date: Fri Dec 21 09:55:42 2007 +1000 22645 22646 fixup clip test include handling 22647 22648 commit 9a5b501332c0a1f10af20845af48c9ddd2ce26a0 22649 Author: Dave Airlie <airlied (a] linux.ie> 22650 Date: Fri Dec 21 09:45:55 2007 +1000 22651 22652 set ddc line correctly post-merge 22653 22654 commit 3c31b96afa20913ad947e68fe0c3a662e5eafbdd 22655 Merge: eb99c3c5c f5e8c1850 22656 Author: Dave Airlie <airlied (a] linux.ie> 22657 Date: Fri Dec 21 09:36:22 2007 +1000 22658 22659 Merge remote branch 'origin/atombios-support' 22660 22661 Conflicts: 22662 22663 src/radeon_display.c 22664 src/radeon_driver.c 22665 22666 commit eb99c3c5c9a2249cb84920f0f225e525fc3a4144 22667 Author: Alex Deucher <alex@botch2.(none)> 22668 Date: Thu Dec 20 18:14:38 2007 -0500 22669 22670 Bump for RC release 22671 22672 commit 8d49ff1da917b7f8240267953ef6ce4ff04daecb 22673 Author: Alex Deucher <alex@botch2.(none)> 22674 Date: Thu Dec 20 01:09:57 2007 -0500 22675 22676 RADEON: check for xf86_crtc_clip_video_helper() in configure.ac 22677 22678 use xf86_crtc_clip_video_helper() from the server if available. 22679 22680 commit fb7a4e24f2da3561ef81371ca4013a4f13806e91 22681 Author: Adam Jackson <ajax (a] redhat.com> 22682 Date: Wed Dec 19 19:15:19 2007 -0500 22683 22684 Fix RN50 mode filtering. 22685 22686 The old code would attempt to limit the maximum pixel size of the screen 22687 by limiting the maximum PLL frequency. This ends up confusing the PLL 22688 computation code since sometimes your maximum freq can be lower than your 22689 minimum freq. More to the point it's just wrong, maximum PLL frequency 22690 isn't the same thing as maximum pixel clock, and even that isn't the same 22691 thing as maximum scanout pixels per second. 22692 22693 The correct thing to do is filter by the mode's effective memory bandwidth. 22694 22695 commit c1b0b69cc50516c3b7e881b0eb46cb3cd2e9dce6 22696 Merge: 0e6634870 ce4fa1ced 22697 Author: Alex Deucher <alex@botch2.(none)> 22698 Date: Wed Dec 19 11:54:50 2007 -0500 22699 22700 Merge branch 'master' of ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati 22701 22702 commit 0e6634870d1ab38ee8c83f6bda1ba60364997853 22703 Author: Alex Deucher <alex@botch2.(none)> 22704 Date: Wed Dec 19 11:54:27 2007 -0500 22705 22706 RADEON: skip empty connectors when creating outputs 22707 22708 commit 6afbf718c151dc3c5c59bd3136b58a93a114d798 22709 Author: Alex Deucher <alex@botch2.(none)> 22710 Date: Wed Dec 19 11:48:38 2007 -0500 22711 22712 RADEON: add support for legacy radeons with DVI and no connector table 22713 22714 commit f5e8c185001e62e744310667c2d1bd3fe6542a62 22715 Author: Dave Airlie <airlied (a] clockmaker.usersys.redhat.com> 22716 Date: Wed Dec 19 10:38:58 2007 +1000 22717 22718 more endian related fixage 22719 22720 commit 98b247066d00db66abe91f518cd93b5c4da4cfb4 22721 Author: Dave Airlie <airlied (a] clockmaker.usersys.redhat.com> 22722 Date: Wed Dec 19 10:25:41 2007 +1000 22723 22724 fix big endian build since zaphod fixups 22725 22726 commit ce4fa1cedec0cf56b9979dfaa12a8d3a7c643df4 22727 Author: Arkadiusz Miskiewicz <arekm (a] maven.pl> 22728 Date: Tue Dec 18 15:34:14 2007 -0500 22729 22730 RADEON: fix fd leak in lid detect code 22731 22732 commit 20568f66f9a9a60a33bd9a69ccc14a891c656836 22733 Author: Arkadiusz Miskiewicz <arekm (a] maven.pl> 22734 Date: Tue Dec 18 15:32:10 2007 -0500 22735 22736 RADEON: more cleanups and warning fixes 22737 22738 commit 1496194200adbcb044ec3977367a0908262e389c 22739 Author: Arkadiusz Miskiewicz <arekm (a] maven.pl> 22740 Date: Tue Dec 18 15:29:53 2007 -0500 22741 22742 RADEON: driver cleanups, warning fixes 22743 22744 commit bd7206fa120495037e3fea0c920d0031d7715bf6 22745 Author: Alex Deucher <alex@botch2.(none)> 22746 Date: Tue Dec 18 03:03:11 2007 -0500 22747 22748 RADEON: fix another merge error 22749 22750 this broken legacy radeons 22751 22752 commit 65a3ac7530e11bb7d818a988fd0cf1dde7688fa4 22753 Author: Alex Deucher <alex@samba.(none)> 22754 Date: Tue Dec 18 00:15:38 2007 -0500 22755 22756 RADEON: more PLL tweaks 22757 22758 commit d93a0e10b8bc6e3797a3cf6c1e28ca413a7c38e4 22759 Author: Alex Deucher <alex@samba.(none)> 22760 Date: Mon Dec 17 20:32:45 2007 -0500 22761 22762 RADEON: post div tweaks for legacy radeon 22763 22764 commit 03b8b49f6f502c45552b018fd8c44d366b2d576f 22765 Author: Alex Deucher <alex@samba.(none)> 22766 Date: Mon Dec 17 20:20:04 2007 -0500 22767 22768 RADEON: fix typo from merge 22769 22770 commit 19b9d3708852b7efe2b05249c8359dadb924dd94 22771 Merge: cf685f37e 29706ca58 22772 Author: Alex Deucher <alex@samba.(none)> 22773 Date: Mon Dec 17 20:07:32 2007 -0500 22774 22775 Merge branch 'atombios-support' of git+ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 22776 22777 commit cf685f37ec874f0aacd09e7c4eb0402c6daec1b0 22778 Merge: 2a134af01 44d07c4cc 22779 Author: Alex Deucher <alex@samba.(none)> 22780 Date: Mon Dec 17 20:07:07 2007 -0500 22781 22782 Merge branch 'master' of git+ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 22783 22784 merge master and fix conflicts 22785 22786 commit 29706ca585ebd9b7b3521521a17016322e9ecccb 22787 Author: Dave Airlie <airlied (a] linux.ie> 22788 Date: Tue Dec 18 10:55:38 2007 +1000 22789 22790 fixup shadow setup on !r600 22791 22792 commit 44d07c4ccce9acb5bd21a17acb082e91f7225764 22793 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22794 Date: Mon Dec 17 18:56:12 2007 -0500 22795 22796 RADEON: typo from last commit 22797 22798 commit 4da3782239921eb377216d4de4a9cc5bb55e0e8a 22799 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22800 Date: Mon Dec 17 18:51:31 2007 -0500 22801 22802 RADEON: add output enable masks 22803 22804 add output enable masks for outputs that drive 22805 more than one connector. Make sure we don't turn off 22806 an output that's driving another connector. 22807 22808 commit 5c5d2d19b2b032a06dd333b4ecc029aac342fb93 22809 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22810 Date: Mon Dec 17 18:15:55 2007 -0500 22811 22812 RADEON: whitespace clean-ups 22813 22814 commit 9f1d8220315c8894a17f2cc328025dc682b0c6e0 22815 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22816 Date: Mon Dec 17 18:04:05 2007 -0500 22817 22818 RADEON: more PLL fixes 22819 22820 - reduce the calculation accuracy 22821 - certain LVDS panels seem to only like certain ref_divs 22822 - add pll flags to handle special cases 22823 - adjust the pll limits on legacy cards 22824 22825 commit 2a134af01bc85de758ad83a85b23d1bba4a1a0f5 22826 Author: Dave Airlie <airlied (a] redhat.com> 22827 Date: Mon Dec 17 15:00:36 2007 +1000 22828 22829 r600: add shadow support to r600 driver to at least make 2d useable 22830 22831 commit 614414611a9f246cbc74f579a79987fff97cf571 22832 Author: Dave Airlie <airlied (a] redhat.com> 22833 Date: Mon Dec 17 11:10:14 2007 +1000 22834 22835 radeon: cleanup pitch calculation and make r600 work again 22836 22837 commit 79a375dbc7f323e2f551490a35f44ec36bed877c 22838 Author: George Wu <geo (a] flood.OCF.Berkeley.EDU> 22839 Date: Mon Dec 17 10:55:36 2007 +1000 22840 22841 r600: might as well fix VT for R600 22842 22843 commit bc213ee723a45f2c396b4ed211a50f7642349973 22844 Author: Alex Deucher <alex@samba.(none)> 22845 Date: Sun Dec 16 14:54:00 2007 -0500 22846 22847 RADEON: fix sclock printout 22848 22849 commit 4747c1f3cd4167b6a51d4864a297719ea48b9346 22850 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22851 Date: Sun Dec 16 14:07:29 2007 -0500 22852 22853 RADEON: Make sure LVDS_EN bit is set when enabling LVDS 22854 22855 commit a9817b2cf436a536dbc43ad77abc3bdcc53d346d 22856 Author: Alex Deucher <alex@samba.(none)> 22857 Date: Sat Dec 15 20:51:53 2007 -0500 22858 22859 RADEON: clean up units in PLL calculation 22860 22861 commit b653e5a628bfa4dfb168e96f93f41eb910f409fb 22862 Author: Michel Dnzer <michel (a] tungstengraphics.com> 22863 Date: Sat Dec 15 00:50:10 2007 +0100 22864 22865 radeon: Default to 1x again with non-v3 AGP cards. 22866 22867 Seems more reliable in general than what was set up by firmware - fingers 22868 crossed... 22869 22870 commit 6229825fa5d6715569098afbb21a40f7a2e7e6be 22871 Author: Michel Dnzer <michel (a] tungstengraphics.com> 22872 Date: Sat Dec 15 00:48:26 2007 +0100 22873 22874 radeon: Warning fixes. 22875 22876 The lid detection code probably wouldn't work on other non-x86 platforms 22877 though... 22878 22879 commit 818ccf0fd4b5879171c5f20526d5a58638f8fde5 22880 Author: Fredrik Hglund <fredrik (a] kde.org> 22881 Date: Fri Dec 14 23:56:12 2007 +0100 22882 22883 RADEON: Fix the vertex coordinates for transformed pictures 22884 22885 This partially fixes transformed pictures on R100/R200 based 22886 cards. The texture still doesn't appear to be clamped correctly, 22887 but since that doesn't matter for rotations at perpendicular 22888 angles, I'm committing this now so randr rotation and reflection 22889 will work properly. 22890 22891 commit 3cfbcf4cafbdfdb33411d16e51fb1f77cd0f93dd 22892 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22893 Date: Fri Dec 14 17:11:00 2007 -0500 22894 22895 RADEON: Fix PLL set up on certain notebooks 22896 22897 Some LVDS panels require specific PLL dividers as 22898 specified in the bios tables. Make sure to use them 22899 if the output is LVDS. 22900 22901 commit a84d446fd301d456bcea8f7abdc52e5a30776412 22902 Author: Alex Deucher <alex@botch2.(none)> 22903 Date: Fri Dec 14 02:17:14 2007 -0500 22904 22905 RADEON: select fb_div0 for LVDS on RV410 (x700) mobility 22906 22907 Fixes bug 8038 22908 I wonder if desktop RV410 need a similar fix. 22909 If your x700 laptop panel has problems after this let me know. 22910 22911 commit b3eed3d87f76779b5a62a3115f99a31484dc38e0 22912 Author: Alex Deucher <alex@samba.(none)> 22913 Date: Fri Dec 14 00:20:10 2007 -0500 22914 22915 RADEON: fix typo in previous cursor fix 22916 22917 commit 814c6c48aebba2e45ce257289b922cd7e92caf2a 22918 Author: Alex Deucher <alex@samba.(none)> 22919 Date: Thu Dec 13 18:45:09 2007 -0500 22920 22921 RADEON: rework PLL calculation 22922 22923 - Take into account the limits from the bios tables 22924 - Unify the PLL calculation between legacy and avivo chips 22925 22926 commit f5ac34983411e4c4f41ab1817dce582830f398fd 22927 Merge: f2b2e0804 6ccf5b33d 22928 Author: Alex Deucher <alex@samba.(none)> 22929 Date: Wed Dec 12 22:37:44 2007 -0500 22930 22931 Merge branch 'master' of git+ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 22932 22933 merge and fix conflicts 22934 22935 commit f2b2e0804183b52d9d3f56ad85b3552ece76c544 22936 Author: Alex Deucher <alex@samba.(none)> 22937 Date: Wed Dec 12 22:18:37 2007 -0500 22938 22939 RADEON: fix rotation on avivo chips 22940 22941 There are still some issues, but this is better than before. 22942 22943 commit 6ccf5b33d27218ae1c45ab93c122438ed536d8ba 22944 Author: Alex Deucher <alex@botch2.(none)> 22945 Date: Wed Dec 12 20:12:06 2007 -0500 22946 22947 RADEON: only enable vblanks if we want them 22948 22949 should fix bug 13610 22950 22951 commit 1668f2056f56370f1b5681c13f1e14904e301216 22952 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22953 Date: Wed Dec 12 19:39:08 2007 -0500 22954 22955 RADEON: use /proc/acpi to determine lid status 22956 22957 Linux only 22958 22959 commit 33a39947f7f79533cd90007a17d57b20126642c6 22960 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 22961 Date: Wed Dec 12 18:50:18 2007 -0500 22962 22963 RADEON: fix cursors when using rotation 22964 22965 allocate separate cursor buffers for each crtc 22966 22967 commit 9e5efdecd12092031a4aebce58747cb4a6f48f28 22968 Author: Arkadiusz Miskiewicz <arekm (a] maven.pl> 22969 Date: Tue Dec 11 23:53:03 2007 -0500 22970 22971 sparse fixes and cleanups from arekm 22972 22973 commit 372bf41818fdafc6a9d2914aee3a8e359f668f02 22974 Author: Alex Deucher <alex@botch2.(none)> 22975 Date: Tue Dec 11 14:04:58 2007 -0500 22976 22977 RADEON: handle HMDI properly (untested) and fix some merge leftovers 22978 22979 commit 3c22ad977c25d5ca2811821fcac6bb8ecd79994a 22980 Merge: c9a0cee97 f3d2ec3a5 22981 Author: Alex Deucher <alex@botch2.(none)> 22982 Date: Tue Dec 11 13:11:15 2007 -0500 22983 22984 Merge branch 'master' of ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 22985 22986 fix conflicts 22987 22988 commit f3d2ec3a5ae61215c792018320158750e7aa937c 22989 Author: Alex Deucher <alex@botch2.(none)> 22990 Date: Tue Dec 11 11:57:27 2007 -0500 22991 22992 RADEON: rewrite PLL computation 22993 22994 Algorithm adapted from BeOS radeon driver with some tweaks by me. 22995 Actually calulate and use the reference divider rather than using the bios default. 22996 Also, always calculate the PLL, rather than falling back to bios dividers. 22997 This should fix bugs 12913, 13590, 13533, possibly others. 22998 22999 commit c9a0cee97ca69e8fe1e1937c7670fa903214cded 23000 Author: Dave Airlie <airlied (a] linux.ie> 23001 Date: Tue Dec 11 06:03:46 2007 +1000 23002 23003 more zaphod fixes - some other work maybe needed 23004 23005 commit 9b125312ab6edc585e4f5931a6a6de81e13b6acc 23006 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 23007 Date: Mon Dec 10 13:53:15 2007 -0500 23008 23009 RADEON: only update crtc values when RMX is active 23010 23011 commit 3a161e1b5d80361e318ced8da5c19e797749d693 23012 Author: Alex Deucher <alex@botch2.(none)> 23013 Date: Mon Dec 10 00:57:26 2007 -0500 23014 23015 RADEON: bios PLL cleanup 23016 23017 commit 731830297f2fc4a416882aacfb0d9b5f8ed32520 23018 Author: Dave Airlie <airlied (a] linux.ie> 23019 Date: Mon Dec 10 15:50:38 2007 +1000 23020 23021 fixup some warnings 23022 23023 commit 2818e2b02ca90c9dfa50905b5311b2ae83ac3b0c 23024 Author: Dave Airlie <airlied (a] linux.ie> 23025 Date: Mon Dec 10 15:43:52 2007 +1000 23026 23027 add more to configure.ac for using out-of-tree mode src 23028 23029 commit cc3c36100986f9d8060bc2d433373d4806f8e730 23030 Author: Dave Airlie <airlied (a] linux.ie> 23031 Date: Mon Dec 10 15:25:56 2007 +1000 23032 23033 add support for building against legacy servers similiar to Intel codepaths 23034 23035 commit 9c278cb7fa7f18d13bde053fd75221cfba9da377 23036 Merge: 6451ea2dc cc167b9bb 23037 Author: Dave Airlie <airlied (a] linux.ie> 23038 Date: Mon Dec 10 15:18:03 2007 +1000 23039 23040 Merge branch 'zaphod-lolz' of git://git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 23041 23042 Conflicts: 23043 23044 src/radeon.h 23045 src/radeon_crtc.c 23046 src/radeon_driver.c 23047 src/radeon_output.c 23048 23049 commit 6451ea2dcc4fac762442f699935864f4a8d445f7 23050 Merge: 0d89556bf 5896ca409 23051 Author: Dave Airlie <airlied (a] linux.ie> 23052 Date: Mon Dec 10 15:08:42 2007 +1000 23053 23054 Merge branch 'master' into atombios-support 23055 23056 commit 5896ca4097d439f59f90f397939132c061c3c364 23057 Author: LisaWu <liswu (a] ati.com> 23058 Date: Fri Dec 7 09:45:05 2007 +0100 23059 23060 radeon: Use %u instead of %d for unsigned value. 23061 23062 commit df44f8380268c27d3978c4e91d736f093322b8b8 23063 Author: Michel Dnzer <michel (a] tungstengraphics.com> 23064 Date: Fri Dec 7 09:41:47 2007 +0100 23065 23066 radeon: Use gettimeofday instead of xf86getsecs. 23067 23068 commit cc167b9bb7f1c3b8579e51e7bc2fca2f8eba6bd1 23069 Author: Dave Airlie <airlied (a] redhat.com> 23070 Date: Fri Dec 7 15:41:36 2007 +1000 23071 23072 disable tiling for zaphod heads 23073 23074 commit 2ce8d192533a8c795714c5a9fb308ec74db40287 23075 Author: Dave Airlie <airlied (a] redhat.com> 23076 Date: Fri Dec 7 15:35:21 2007 +1000 23077 23078 don't add fboffset to info->FB it already is mapped at the offset 23079 23080 commit 0dcd926d3092100854b3e362d6659d4950508aeb 23081 Author: Dave Airlie <airlied (a] redhat.com> 23082 Date: Fri Dec 7 14:45:04 2007 +1000 23083 23084 radeon: bring back zaphod all is forgiven. 23085 23086 You've whined, you've cried, you've nagged, and you're guilt trippin has 23087 made me do it... It actually wasn't as hard as I thought it would be. 23088 23089 Still not perfect, couple of things to fix yet 23090 23091 commit bb5ede557bf32a42eef158ff0fbcfe1c6ede098a 23092 Author: Dave Airlie <airlied (a] redhat.com> 23093 Date: Fri Dec 7 14:30:32 2007 +1000 23094 23095 radeon: move savedreg/modereg into entity instead of info 23096 23097 commit 64ab1cdf343a9a69e7e9e64f0bba77c54a94e9d0 23098 Author: James Cloos <cloos (a] jhcloos.com> 23099 Date: Thu Dec 6 15:51:12 2007 -0500 23100 23101 Add missing PHONY line for automatic ChangeLog generation 23102 23103 commit 0d89556bfa41a3acbd6afe85b062e3a21f2ca057 23104 Author: Dave Airlie <airlied (a] ppcg5.localdomain> 23105 Date: Thu Dec 6 19:23:06 2007 +1100 23106 23107 powerpc: build fixes from last merge 23108 23109 commit dbe3d2608ecc9896db9c23b3a347b50748c51e13 23110 Merge: 48e31cdaa 21ed43539 23111 Author: Dave Airlie <airlied (a] redhat.com> 23112 Date: Thu Dec 6 14:22:03 2007 +1000 23113 23114 Merge branch 'master' into atombios-support 23115 23116 Conflicts: 23117 23118 src/radeon_output.c 23119 23120 commit 21ed435398e4a398dd8a0a5d7c1d4cc45e916332 23121 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 23122 Date: Tue Dec 4 17:08:58 2007 -0500 23123 23124 RADEON: add MacModel imac-g5-isight for iMac G5 iSight 23125 23126 Thanks to tienne Bersac for helping to figure this out. 23127 23128 commit 54bfd522405d9fdfb69d3a59e111ac3d63483dbb 23129 Author: tienne Bersac <bersace03 (a] laposte.net> 23130 Date: Tue Dec 4 14:22:42 2007 -0500 23131 23132 RADEON: fix typo 23133 23134 commit 5022d006cfc06ca0395981526b2c2c94c6878567 23135 Author: Michel Dnzer <michel (a] tungstengraphics.com> 23136 Date: Sun Dec 2 17:27:33 2007 +0100 23137 23138 radeon: Further XVideo fixes. 23139 23140 * Make sure pitch constraints are always met for DMA upload blits. 23141 * RGB24 is not affected by endianness. 23142 23143 commit 6ed55b70b23dfdc7b41103ea59c1df2bda5e41e6 23144 Author: Kusanagi Kouichi <slash (a] ma.neweb.ne.jp> 23145 Date: Sun Dec 2 17:18:46 2007 +0100 23146 23147 radeon: Fix crash with XVideo 24bit RGB images. 23148 23149 See https://bugs.freedesktop.org/show_bug.cgi?id=13274 . 23150 23151 commit a697b590899bb7704ec4d7ae9a9c3cbbfcaef382 23152 Author: Michel Dnzer <michel (a] tungstengraphics.com> 23153 Date: Sun Dec 2 17:11:20 2007 +0100 23154 23155 Fix build against xserver master. 23156 23157 (DE)ALLOCATE_LOCAL are gone. 23158 23159 commit 48e31cdaa0caa21573879af5b9267773fe89176a 23160 Author: George Wu <geo (a] ocf.berkeley.edu> 23161 Date: Sun Dec 2 15:25:09 2007 +1000 23162 23163 RADEON/R600: small code cleanup 23164 23165 commit 00b4480aa2c5d7f751e34fc964f431b90b14c8d2 23166 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 23167 Date: Sat Dec 1 14:18:40 2007 -0500 23168 23169 RADEON: add options for force TV out as detected and to set TV standard 23170 23171 Also fix a typo in internal tv-out parsing 23172 23173 commit 1e029fef5fe264f2ced445b80bf6070abcb84b82 23174 Author: Alex Deucher <alex@samba.(none)> 23175 Date: Sat Dec 1 00:58:51 2007 -0500 23176 23177 RADEON: move GPIO lookup to a separate function 23178 23179 commit dcbef1ba9dfcf35c28e058832a55adf00afb472e 23180 Author: Alex Deucher <alex@samba.(none)> 23181 Date: Sat Dec 1 00:35:25 2007 -0500 23182 23183 RADEON: fix typo in previous commit 23184 23185 check gpio table revision before connector table revision 23186 23187 commit fdce0598a2228c48c84deae1d7bebb2d7b3e979b 23188 Author: Alex Deucher <alex@samba.(none)> 23189 Date: Sat Dec 1 00:15:34 2007 -0500 23190 23191 RADEON: convert atombios connector table parsing to use ATOM structs 23192 23193 convert and add hpd info 23194 23195 commit e3d7de9cc956aec5f940ad6db09e826b3a69523a 23196 Author: Alex Deucher <alex@botch2.(none)> 23197 Date: Fri Nov 30 20:14:42 2007 -0500 23198 23199 RADEON: remove unused cruft from last atom import 23200 23201 commit d5d83411e8a884154d671aad440524507cce313e 23202 Author: Alex Deucher <alex@botch2.(none)> 23203 Date: Fri Nov 30 20:11:42 2007 -0500 23204 23205 RADEON: save/restore avivo crtc cursor control 23206 23207 this should prevent the cursor from showing up on in text 23208 mode or vesafb etc. after running the driver. 23209 23210 commit af0196f7bf0d1d5d211391149c18935d64ed2b06 23211 Merge: d9858a2b3 0175b7998 23212 Author: Alex Deucher <alex@botch2.(none)> 23213 Date: Fri Nov 30 16:40:28 2007 -0500 23214 23215 Merge branch 'master' of ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 23216 23217 merge master and fix conflicts 23218 23219 commit d9858a2b3744b99003cfb9f31b743a2d31b322e9 23220 Author: Dave Airlie <airlied (a] linux.ie> 23221 Date: Sat Dec 1 06:49:59 2007 +1000 23222 23223 radeon: add in pll spread spectrum workaround 23224 23225 commit 0175b79987ef4d7b0ce8238c3bdde989e504516a 23226 Author: Alex Deucher <alex@botch2.(none)> 23227 Date: Fri Nov 30 15:37:42 2007 -0500 23228 23229 RADEON: rework MacModel option 23230 23231 this brings in some previous research from Michel Dnzer, 23232 Sjoerd Simons, and myself. Hopefully, the driver will pick 23233 the correct MacModel in more cases. This also changes the 23234 default connector table for desktop Macs to dual DVI rather 23235 than DVI+VGA as that seems to be the case more often than not. 23236 External TMDS chips are handled separately now as well. 23237 Eventually we should add an option to allow the user to specify 23238 what external TMDS chip they need, but we don't have enough info 23239 yet, so we'll rely on OF to init the external chip in most cases 23240 for now. 23241 23242 commit e1945f1f25a34310bd58ce128c8ff27ecc985618 23243 Merge: b368b0f22 df7777bff 23244 Author: Alex Deucher <alex@botch2.(none)> 23245 Date: Fri Nov 30 14:30:55 2007 -0500 23246 23247 Merge branch 'atombios-support' of ssh://agd5f (a] git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 23248 23249 commit b368b0f22cd1d7ef9b4c65d82929c76f3b82d573 23250 Author: Alex Deucher <alex@botch2.(none)> 23251 Date: Fri Nov 30 14:29:27 2007 -0500 23252 23253 RADEON: disable atom pll set for r4xx cards 23254 23255 the clocks do not get set correctly in all cases. this needs 23256 further investigation. 23257 23258 commit 5af15739571c09260750bcfd3620e16fd7fec862 23259 Author: Alex Deucher <alex@botch2.(none)> 23260 Date: Fri Nov 30 14:24:30 2007 -0500 23261 23262 RADEON: small cleanup of pll code 23263 23264 commit df7777bff40c1feabcc12d2148ad6ac5213efbb3 23265 Author: George Wu <geo (a] ocf.berkeley.edu> 23266 Date: Fri Nov 30 17:49:33 2007 +1000 23267 23268 Add LVTMA PWRSEQ registers to fix VT switching for LVDS 23269 23270 commit 9840a0fd4fc8c980533fcd4a02c55cd0d5634b6d 23271 Author: Alex Deucher <alex@botch2.(none)> 23272 Date: Thu Nov 29 13:27:37 2007 -0500 23273 23274 RADEON: add MacModel "mini-internal" for minis with internal TMDS 23275 23276 Some macs (minis and powerbooks) use internal tmds, others use external tmds 23277 and not just for dual-link TMDS, it shows up with single-link as well. 23278 Unforunately, there doesn't seem to be any good way to figure it out. 23279 23280 commit 017c939cf0a2b12fbdc1681cc70c28b23ae3b397 23281 Author: Alex Deucher <alex@samba.(none)> 23282 Date: Thu Nov 29 02:52:14 2007 -0500 23283 23284 RADEON: implement CLUT adjust support 23285 23286 commit 9963b0fe01feb6dd0cb555b874a48f6fa3b255cb 23287 Author: Alex Deucher <alex@samba.(none)> 23288 Date: Thu Nov 29 00:46:23 2007 -0500 23289 23290 RADEON: fix cursor offset on avivo chips 23291 23292 commit 6c56e3d7655b17e93e8823aefe34b05291104695 23293 Author: Alex Deucher <alex@botch2.(none)> 23294 Date: Tue Nov 27 15:27:36 2007 -0500 23295 23296 RADEON: switch r4xx to atombios load detection 23297 23298 works great 23299 23300 commit 7561242e5b79bc2798ca3aace2b79e1a36949488 23301 Author: Alex Deucher <alex@botch2.(none)> 23302 Date: Tue Nov 27 14:50:36 2007 -0500 23303 23304 RADEON: re-org load detection for legacy chips 23305 23306 commit febdcc2dccd42acbcd68ae630b7811cae5c58e8a 23307 Author: Dave Airlie <airlied (a] linux.ie> 23308 Date: Wed Nov 28 05:10:57 2007 +1000 23309 23310 legacy: fix fb/agp read/writes 23311 23312 commit bb8545146959b748994be055d5b3de66ec66c8b2 23313 Author: Alex Deucher <alex@samba.(none)> 23314 Date: Mon Nov 26 17:34:51 2007 -0500 23315 23316 RADEON: first pass at TV/Component video 23317 23318 Untested and not likely to work just yet. 23319 23320 commit e2bde646b864dca9056d9ecfe23a0d905647ea9a 23321 Author: Alex Deucher <alex@samba.(none)> 23322 Date: Mon Nov 26 14:35:57 2007 -0500 23323 23324 RADEON: move crtc output source selection into atombios_output.c 23325 23326 The function fits better as an output function and should now 23327 work with clones as well. 23328 23329 commit 16c9be4107678a2a58d3418b7f1cc94d695ca8d6 23330 Author: Alex Deucher <alex@samba.(none)> 23331 Date: Mon Nov 26 14:20:54 2007 -0500 23332 23333 RADEON: add default connector table for avivo chips 23334 23335 commit 6f080d00e6f4f84d5e0d6b4eff302bf42c230e81 23336 Author: Arkadiusz Miskiewicz <arekm (a] maven.pl> 23337 Date: Mon Nov 26 12:43:30 2007 -0500 23338 23339 RADEON: fix backlight control on some laptops 23340 23341 It seems the bios scratch regs are involved in backlight control 23342 on some laptops. This patch fixes the problematic laptops and doesn't 23343 seem to break the previous bios lid and output control fixes. 23344 23345 commit dcf22aed87366f4625fb5042cb84fecccd9ceece 23346 Author: Alex Deucher <alex@botch2.(none)> 23347 Date: Mon Nov 26 11:10:03 2007 -0500 23348 23349 RADEON: only return status unknown for XPRESS chips 23350 23351 this seems to cause more issues than it attempted to fix 23352 so limit it to XPRESS chips for now. 23353 23354 commit 206e280f02324641b4fe5a1986e26adf0e021fd4 23355 Author: Alex Deucher <alex@botch2.(none)> 23356 Date: Mon Nov 26 09:39:27 2007 -0500 23357 23358 RADEON: fix typo in man page 23359 23360 commit 4e792db655dc92d2864db36b4d8f6714908de8e8 23361 Author: Dave Airlie <airlied (a] redhat.com> 23362 Date: Fri Nov 23 15:44:44 2007 +1000 23363 23364 r500: set default minimum pll freq 23365 23366 commit a13b4c69c105c096dd05e6de2c5c154c9a8bcc71 23367 Author: Dave Airlie <airlied (a] redhat.com> 23368 Date: Fri Nov 23 15:25:06 2007 +1000 23369 23370 r5xx: cleanup pll code.. 23371 23372 Clean the PLL code to use the radeon pll structs. 23373 23374 commit 5d792b5673bbf4784eb0ec059221e9b57232a122 23375 Author: Dave Airlie <airlied (a] redhat.com> 23376 Date: Fri Nov 23 15:03:13 2007 +1000 23377 23378 radeon: fix up memory mapping issues for vt switch 23379 23380 commit 558a2ef266c1ca517c7fb464b0ccfef83238c913 23381 Author: Dave Airlie <airlied (a] redhat.com> 23382 Date: Fri Nov 23 14:39:32 2007 +1000 23383 23384 fix silly spaces 23385 23386 commit 1bda4424a7031de437acfca9c41d4b3668e36051 23387 Author: Dave Airlie <airlied (a] redhat.com> 23388 Date: Fri Nov 23 14:39:19 2007 +1000 23389 23390 r600: add hi agp address for mc 23391 23392 commit 133234c31a294f24a3968a576aad2bb8b89d0f6a 23393 Author: Dave Airlie <airlied (a] redhat.com> 23394 Date: Fri Nov 23 14:15:18 2007 +1000 23395 23396 atombios: use values from object header 23397 23398 commit dbf6eae7e7a4bd1bc60fefdc7ab6276ed3f097c4 23399 Author: Dave Airlie <airlied (a] redhat.com> 23400 Date: Fri Nov 23 11:55:05 2007 +1000 23401 23402 atombios: add initial object header parsing for r600 cards 23403 23404 commit 6b103915c11fc79d2efc43c44fc2a00c3bc64ede 23405 Author: Dave Airlie <airlied (a] redhat.com> 23406 Date: Fri Nov 23 09:24:20 2007 +1000 23407 23408 r500: make vt switch work again for me 23409 23410 commit d24208276aad7669feeed527dced60c76d39eae6 23411 Author: Dave Airlie <airlied (a] redhat.com> 23412 Date: Fri Nov 23 08:50:25 2007 +1000 23413 23414 avivo: fix typo in register saving 23415 23416 commit 197a62704742a4a19736c2637ac92d1dc5ab34ed 23417 Author: Adam Jackson <ajackson (a] redhat.com> 23418 Date: Thu Nov 22 20:26:23 2007 +1000 23419 23420 radeon: fix openoffice/render bug on r100 chips 23421 23422 commit 64010fc4eae8359c01e430f64252931242c91435 23423 Author: Dave Airlie <airlied (a] linux.ie> 23424 Date: Thu Nov 22 20:25:31 2007 +1000 23425 23426 Revert "Disable RENDER acceleration by default on some RV200 chips." 23427 23428 This reverts commit 145da701bf4fb9c0ad9f95620b20914ae0126852. 23429 23430 pull in fix from ajax next commit 23431 23432 commit 145da701bf4fb9c0ad9f95620b20914ae0126852 23433 Author: Stefan Dirsch <sndirsch (a] suse.de> 23434 Date: Thu Nov 22 08:38:09 2007 +0100 23435 23436 Disable RENDER acceleration by default on some RV200 chips. 23437 23438 Novell Bug #341699: Render acceleration is known to be broken 23439 on at least "Radeon 7500 QW" and "Radeon Mobility M7 LW". 23440 23441 commit e810c3ae9908cd57e95b1b091cded87cbfc12fdc 23442 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 23443 Date: Thu Nov 22 02:37:55 2007 +0100 23444 23445 really do not set up surface regs for depth buf on r100-class igps (bug #13080) 23446 23447 fix the if condition testing for these chips... 23448 23449 commit 0cc7e94849c1525750fabd04cf58f2dee88372e0 23450 Author: Alex Deucher <alex@botch2.(none)> 23451 Date: Wed Nov 21 17:06:17 2007 -0500 23452 23453 RADEON: reorder crtc/pll setup 23454 23455 commit d56bde98efceaa8344e62f8e98db90c4bb642331 23456 Author: Alex Deucher <alex@botch2.(none)> 23457 Date: Wed Nov 21 17:03:39 2007 -0500 23458 23459 RADEON: fix crtc to output routing 23460 23461 Thanks to AMD for the information 23462 23463 commit a12390c832abe423def60e39cd5a9118e5910339 23464 Merge: d5317922f e74dca194 23465 Author: Alex Deucher <alex@botch2.(none)> 23466 Date: Wed Nov 21 02:24:48 2007 -0500 23467 23468 Merge branch 'atombios-support' of git://git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 23469 23470 commit d5317922f29a57b6c4127826a2fc126c5fd7c117 23471 Author: Alex Deucher <alex@botch2.(none)> 23472 Date: Wed Nov 21 02:23:37 2007 -0500 23473 23474 RADEON: attempt to fix crtc to output routing 23475 23476 The output routing seems to be based on the output ids from the bios 23477 connector tables and the connected status in the bios scratch regs. 23478 I don't fully understand this yet, but this seems to work 23479 for the most part, however changing modes can sometimes 23480 lead to a blanked head. This can be remedied by forcing 23481 a dpms off cycle. 23482 23483 commit 81ce299bffd75540925b4c8234adf11226147165 23484 Author: Alex Deucher <alex@botch2.(none)> 23485 Date: Wed Nov 21 01:35:44 2007 -0500 23486 23487 RADEON: provide clearer debugging info 23488 23489 commit e74dca19416b13f97db9d1fc06299b988057d6a4 23490 Author: Dave Airlie <airlied (a] redhat.com> 23491 Date: Wed Nov 21 16:24:25 2007 +1000 23492 23493 re-enable mobility chips 23494 23495 commit 9c5b813dd6b3492cbc9833bc59792a5cec457e51 23496 Author: Alex Deucher <alex@botch2.(none)> 23497 Date: Wed Nov 21 01:22:42 2007 -0500 23498 23499 RADEON: major re-org and clean up of atom output control 23500 23501 - use radeon_output->devices to determine output 23502 - clean up and simplify dpms and mode set 23503 23504 commit 7634cb6b96f938bc6615eb2c49ae75aaefd04cce 23505 Author: Alex Deucher <alex@botch2.(none)> 23506 Date: Wed Nov 21 00:10:14 2007 -0500 23507 23508 RADEON: make naming consistent and remove some cruft 23509 23510 commit 908748343fc9a6cdc38af0fc028c63a82766da3f 23511 Author: Alex Deucher <alex@botch2.(none)> 23512 Date: Wed Nov 21 00:05:42 2007 -0500 23513 23514 RADEON: store devices ids from bios for each driver output 23515 23516 commit 3975da2ea8cb628f7f66c3f26c5dfa181cd1c532 23517 Merge: e283aa332 295ce277b 23518 Author: Alex Deucher <alex@botch2.(none)> 23519 Date: Tue Nov 20 23:52:29 2007 -0500 23520 23521 Merge branch 'atombios-support' of git://git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 23522 23523 commit e283aa332adf0134243a4fa3d14263719cd8a3fd 23524 Author: Alex Deucher <alex@botch2.(none)> 23525 Date: Tue Nov 20 23:35:46 2007 -0500 23526 23527 RADEON: add LVDS atom support 23528 23529 commit e4bc3e1e7bb45571367d41b5328ff2590810b0f9 23530 Author: Alex Deucher <alex@botch2.(none)> 23531 Date: Tue Nov 20 18:09:29 2007 -0500 23532 23533 RADEON: enable/disable the right TMDS controller 23534 23535 commit 3e47683ffaa44a89cda9bcddf530643befb27efa 23536 Author: Alex Deucher <alex@botch2.(none)> 23537 Date: Tue Nov 20 18:01:15 2007 -0500 23538 23539 RADEON: fixup for bios tables with wrong connector types 23540 23541 commit 7412952eb1d1e9857cdab8417f7305f676900827 23542 Author: Alex Deucher <alex@botch2.(none)> 23543 Date: Tue Nov 20 18:00:12 2007 -0500 23544 23545 RADEON: switch to using ATOM defines for bios device table 23546 23547 commit 295ce277bb0a44b9539b3dba575e7aff279dc2d0 23548 Author: Dave Airlie <airlied (a] redhat.com> 23549 Date: Wed Nov 21 08:56:40 2007 +1000 23550 23551 add missing files for make dist 23552 23553 noted by ndim on #radeonhd 23554 23555 commit 20083b0695987b25e442ecbdec24f3cb6f1ac2ae 23556 Author: Dave Airlie <airlied (a] redhat.com> 23557 Date: Wed Nov 21 08:53:44 2007 +1000 23558 23559 LVDS on r500/r600 needs some work disable mobile chips for now 23560 23561 commit e4b8a4479ddea9b083b3a763dc0b9302e7b9a82a 23562 Author: Dave Airlie <airlied (a] redhat.com> 23563 Date: Wed Nov 21 08:06:12 2007 +1000 23564 23565 r600: add memory controller regs from AMD 23566 23567 commit aa88da974b97ea1e9bbb47b3494543575c09d912 23568 Author: Dave Airlie <airlied (a] redhat.com> 23569 Date: Wed Nov 21 08:01:35 2007 +1000 23570 23571 rs690 is !> r600 23572 23573 commit c8872603454e6a4ffed9fc7d9adc2c364a429608 23574 Author: Dave Airlie <airlied (a] clockmaker.usersys.redhat.com> 23575 Date: Tue Nov 20 22:33:39 2007 +1000 23576 23577 radeon: restructure pci ids to avoid effort later 23578 23579 This uses a single file with all the pciids and parameters for radeon 23580 family and setup in it. I don't run the perl script at build time to avoid 23581 a perl dependency on build but adding pci ids should be done via the csv file 23582 with openoffice or gnumeric if possible. 23583 23584 commit a5b34c2f1f7d5346c4489cb30e07291d1217026c 23585 Author: Dave Airlie <airlied (a] redhat.com> 23586 Date: Tue Nov 20 16:37:29 2007 +1000 23587 23588 r600: use standard memory controller setup paths 23589 23590 commit 4a523da5221d53f2efa49da2326500e9b0b9f14d 23591 Author: Dave Airlie <airlied (a] redhat.com> 23592 Date: Tue Nov 20 15:31:11 2007 +1000 23593 23594 r600: get r600 to work non-accelerated. 23595 23596 DDC still not working yet 23597 23598 commit 80023441ba46882bc810ff3790c7148059f155f5 23599 Author: Dave Airlie <airlied (a] redhat.com> 23600 Date: Tue Nov 20 14:10:23 2007 +1000 23601 23602 r600: block r600 startup due to lack of memory controller info 23603 23604 commit f6fbbacc17bf9b1073d3e993b225987fd9173182 23605 Author: Dave Airlie <airlied (a] redhat.com> 23606 Date: Tue Nov 20 13:41:55 2007 +1000 23607 23608 atombios: add warnings for r500 and r600 23609 23610 commit 45a8b083c123b820c008f04ab857a64a8facec14 23611 Author: Dave Airlie <airlied (a] redhat.com> 23612 Date: Tue Nov 20 13:37:00 2007 +1000 23613 23614 atombios: add all r5xx and r6xx pci ids 23615 23616 commit 5d023e2c3c2ab44ea57ffadc9607025d602c376c 23617 Merge: 0d1e0c780 c88726034 23618 Author: Dave Airlie <airlied (a] redhat.com> 23619 Date: Tue Nov 20 13:02:43 2007 +1000 23620 23621 Merge branch 'master' into atombios-support 23622 23623 Conflicts: 23624 23625 src/radeon_chipset.h 23626 src/radeon_driver.c 23627 src/radeon_probe.c 23628 23629 commit 0d1e0c7805b3d8e56ccb49465e6b144afb7bdc51 23630 Author: Dave Airlie <airlied (a] redhat.com> 23631 Date: Tue Nov 20 09:08:04 2007 +1000 23632 23633 r5xx: add 71c5 for macbook pro 23634 23635 commit d5909b30595c103bb5f42cd1704330f944bba49c 23636 Author: Dave Airlie <airlied (a] redhat.com> 23637 Date: Tue Nov 20 08:15:58 2007 +1000 23638 23639 r5xx: cleanups after last merge 23640 23641 commit fe2f7a09050fb7a345a1f52239f8f3c4f1053891 23642 Merge: 744c8cb6c 49055d8af 23643 Author: Dave Airlie <airlied (a] redhat.com> 23644 Date: Tue Nov 20 08:04:32 2007 +1000 23645 23646 Merge branch 'master' into agd-atom-merge 23647 23648 Conflicts: 23649 23650 src/radeon_cursor.c 23651 src/radeon_output.c 23652 23653 commit 744c8cb6c293fcaa687566f52901644e699baace 23654 Merge: e258fbe41 e530af79a 23655 Author: Dave Airlie <airlied (a] redhat.com> 23656 Date: Tue Nov 20 07:56:33 2007 +1000 23657 23658 Merge branch 'agd-atom' of ssh://people.freedesktop.org/~agd5f/xf86-video-ati-atom into agd-atom 23659 23660 commit e258fbe411d255a1044b61d7ff738aee3fb5b7f4 23661 Author: Dave Airlie <airlied (a] redhat.com> 23662 Date: Mon Nov 19 16:35:05 2007 +1000 23663 23664 makes 2-headed cursor work 23665 23666 commit e530af79adf51b3e95a0eca676c915a34dcbf4a7 23667 Merge: 69e197f2c 52aba8d73 23668 Author: Alex Deucher <alex@botch2.(none)> 23669 Date: Mon Nov 19 00:59:30 2007 -0500 23670 23671 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 23672 23673 commit 69e197f2c8002aacf2587754c8d3bd63c88f85b1 23674 Merge: 5e8940fa6 862dcabfe 23675 Author: Alex Deucher <alex@botch2.(none)> 23676 Date: Mon Nov 19 00:57:34 2007 -0500 23677 23678 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 23679 23680 commit fca47ad083449f4cf9063dd970cdcebea6a7f110 23681 Author: Dave Airlie <airlied (a] redhat.com> 23682 Date: Mon Nov 19 15:53:40 2007 +1000 23683 23684 add z3ro's pciids 23685 23686 commit 5e8940fa6e33d09091aa4bcf04b0f9e79596e1b8 23687 Author: Alex Deucher <alex@botch2.(none)> 23688 Date: Mon Nov 19 00:52:38 2007 -0500 23689 23690 fix logic in connector table check for TVs and switch counter to symbolic names 23691 23692 commit c19123fd9483758eb6b286c3dffcb6d79d5b1ee5 23693 Author: Dave Airlie <airlied (a] redhat.com> 23694 Date: Mon Nov 19 15:46:58 2007 +1000 23695 23696 add firegl card on ajaxs machine 23697 23698 commit f02f340e466a415b4e01648ca1e323f4ce125885 23699 Author: Alex Deucher <alex@botch2.(none)> 23700 Date: Mon Nov 19 00:39:19 2007 -0500 23701 23702 Don't assign a gpio for TV 23703 23704 commit 52aba8d73189ba959f19c0437499d5e7a8829827 23705 Merge: 862dcabfe 5e8940fa6 23706 Author: Dave Airlie <airlied (a] redhat.com> 23707 Date: Mon Nov 19 15:30:46 2007 +1000 23708 23709 Merge branch 'agd-atom' of ssh://people.freedesktop.org/~agd5f/xf86-video-ati-atom into agd-atom 23710 23711 commit 862dcabfe0c10751d815e5cdd7436c10c2c2db10 23712 Author: Dave Airlie <airlied (a] redhat.com> 23713 Date: Mon Nov 19 15:30:08 2007 +1000 23714 23715 r520: nail i2c enable/disable issue 23716 23717 commit f3dd7f413b670eeb6b8639f6677d72050ad5fe04 23718 Author: Alex Deucher <alex@botch2.(none)> 23719 Date: Mon Nov 19 00:19:39 2007 -0500 23720 23721 Don't detect TV out for now 23722 23723 commit 94de0e22d7229ca71e18e1e849d8545d9ca7bafe 23724 Author: Dave Airlie <airlied (a] redhat.com> 23725 Date: Mon Nov 19 14:47:33 2007 +1000 23726 23727 i2c: fix bus enable stuff 23728 23729 commit 3f1fc7eef13ea02fa5119e9b51d499841b801f2d 23730 Author: Alex Deucher <alex@botch2.(none)> 23731 Date: Mon Nov 19 00:02:14 2007 -0500 23732 23733 CRTs/DFPs may share a DVI port, but TV and CV don't 23734 23735 commit 8f84c5ad4c4af14612ea68fe6f24d0d527f00acc 23736 Author: Alex Deucher <alex@botch2.(none)> 23737 Date: Sun Nov 18 23:43:06 2007 -0500 23738 23739 fix typo in loop 23740 23741 commit 384cd8f52c89d089c6559e2eedbae45641fcd14e 23742 Merge: f3f0e4ec9 234b60730 23743 Author: Dave Airlie <airlied (a] redhat.com> 23744 Date: Mon Nov 19 14:02:55 2007 +1000 23745 23746 Merge branch 'agd-atom' of ../xf86-video-ati into agd-atom 23747 23748 commit f3f0e4ec92c935c89ddb2f4241fe4335a521b439 23749 Author: Alex Deucher <alex@botch2.(none)> 23750 Date: Sun Nov 18 23:14:01 2007 -0500 23751 23752 RADEON: unify connectortype handling 23753 23754 commit 234b6073054ac7630e82781683e666b94b2f12de 23755 Author: Dave Airlie <airlied (a] redhat.com> 23756 Date: Mon Nov 19 14:02:09 2007 +1000 23757 23758 restore avivo memory map registers at correct places 23759 23760 commit 459a30ba511fe2fa8051380a9741fcfd9bb401ef 23761 Author: Dave Airlie <airlied (a] redhat.com> 23762 Date: Mon Nov 19 13:44:38 2007 +1000 23763 23764 fix type for r520 agp code 23765 23766 commit 760af92412ef0d5cc44e52e7cec11fd80c4aaaeb 23767 Author: Alex Deucher <alex@botch2.(none)> 23768 Date: Sun Nov 18 22:34:59 2007 -0500 23769 23770 RADEON: unify DDC line handling 23771 23772 commit e73bf6290da20dd61798ace775999ce1cb550934 23773 Author: Dave Airlie <airlied (a] redhat.com> 23774 Date: Mon Nov 19 13:32:16 2007 +1000 23775 23776 add x1900xt support 23777 23778 commit 2e37937bacd624d616b91c41006c113791ebe98d 23779 Author: Alex Deucher <alex@botch2.(none)> 23780 Date: Sun Nov 18 20:18:50 2007 -0500 23781 23782 RADEON: step one in output rework 23783 23784 re-organize the output type 23785 23786 commit 679e7a2e0d1b213524b8109193483bc9840fb116 23787 Author: Alex Deucher <alex@botch2.(none)> 23788 Date: Sun Nov 18 17:56:51 2007 -0500 23789 23790 Few fixes from the last commit. 23791 23792 Update parser works fine on r4xx. 23793 23794 commit 1cd7cc3e6758ab1012f3ced6e958a1517f45557f 23795 Author: Alex Deucher <alex@botch2.(none)> 23796 Date: Sun Nov 18 17:44:36 2007 -0500 23797 23798 WIP: new atom code comples. 23799 23800 I commented out the object parsing for the time being as 23801 using it will require some thought as to new output 23802 related data structures. 23803 23804 commit b155fa872ee4ca5d801e942aee6e619cef104f35 23805 Author: Alex Deucher <alex@botch2.(none)> 23806 Date: Sat Nov 17 00:34:56 2007 -0500 23807 23808 WIP: more new ATOM integration work 23809 23810 commit 67db114d97abed7a607467e5d67c7b4ffa2c347e 23811 Merge: 7d06a8791 ea1534659 23812 Author: Alex Deucher <alex@botch2.(none)> 23813 Date: Fri Nov 16 14:29:53 2007 -0500 23814 23815 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 23816 23817 commit ea1534659de87d3d75eb20d808d039cff22cb537 23818 Author: Dave Airlie <airlied (a] linux.ie> 23819 Date: Fri Nov 16 18:46:02 2007 +1000 23820 23821 avivo: fixup some i2c stuff 23822 23823 commit 7d06a8791839ce6b22e2449646832b79cebf1b21 23824 Author: Alex Deucher <alex@botch2.(none)> 23825 Date: Fri Nov 16 02:43:00 2007 -0500 23826 23827 WIP: sync up with the latest ATOM bios code in rhd 23828 23829 doesn't compile ATM 23830 23831 commit 3614d80ceb9a7b3615b0baab3cf2dd34ed4ab464 23832 Author: Dave Airlie <airlied (a] linux.ie> 23833 Date: Fri Nov 16 17:22:39 2007 +1000 23834 23835 add missing hpd register 23836 23837 commit 9a2715fda97ac0ebcb45650a416e0652aab575b8 23838 Author: Dave Airlie <airlied (a] linux.ie> 23839 Date: Fri Nov 16 17:22:25 2007 +1000 23840 23841 make i2c unlock/lock registers for gpios 23842 23843 commit 20dc549fff9a4137c93ebed449d05e0c437b6bc1 23844 Author: Dave Airlie <airlied (a] linux.ie> 23845 Date: Fri Nov 16 17:01:33 2007 +1000 23846 23847 avivo i2c: consolidate the avivo i2c code 23848 23849 commit 3e62730f79a13883a65a568bc821bc56055a4ab7 23850 Author: Dave Airlie <airlied (a] linux.ie> 23851 Date: Fri Nov 16 15:19:00 2007 +1000 23852 23853 atombios: fixup warnings in atombios files 23854 23855 commit cca7af3c4910983f7f090792986fcbfa0dc97cfb 23856 Author: Dave Airlie <airlied (a] linux.ie> 23857 Date: Fri Nov 16 15:04:01 2007 +1000 23858 23859 remove avivo_reg.h 23860 23861 commit d39eb2077c6b2fc094ccd952772528eb9428c587 23862 Author: Dave Airlie <airlied (a] linux.ie> 23863 Date: Fri Nov 16 15:00:50 2007 +1000 23864 23865 radeon: rename a large section of avivo regs to documented names 23866 23867 commit b7774c28dde72a205a40be78003df72eabfb9b1f 23868 Author: Dave Airlie <airlied (a] linux.ie> 23869 Date: Fri Nov 16 14:48:36 2007 +1000 23870 23871 Add copyright headers 23872 23873 commit 3cfcd2164b400bd0d1cb4ede8eeb01abba9d75c8 23874 Merge: efac14e66 718bfd3b6 23875 Author: Alex Deucher <alex@botch2.(none)> 23876 Date: Thu Nov 15 23:25:39 2007 -0500 23877 23878 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 23879 23880 commit efac14e669a0c6184f8848191eb49ffb21934ee1 23881 Author: Dave Airlie <airlied (a] linux.ie> 23882 Date: Thu Nov 15 23:17:25 2007 -0500 23883 23884 r5xx: fix typo for crtc offset 23885 23886 commit e6db621c37ff615be286462f000d67a662c5c331 23887 Author: Alex Deucher <alex@botch2.(none)> 23888 Date: Thu Nov 15 23:15:56 2007 -0500 23889 23890 fix INMC() and OUTMC() on !AVIVO chips 23891 23892 WR_EN is bit 8 so don't use OUTREG8. 23893 23894 commit 52ba3fdd1ce05983fabedff234cfaf4c60fba38d 23895 Author: Alex Deucher <alex@botch2.(none)> 23896 Date: Thu Nov 15 23:12:30 2007 -0500 23897 23898 atombios_dac_detect() takes care of primary vs tv dac itself 23899 23900 commit 49055d8aff91ff12186feaf5343c8fd2f96bcba0 23901 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 23902 Date: Thu Nov 15 22:56:09 2007 -0500 23903 23904 RADEON: set proper defaults for tv dac BGADJ/DACADJ 23905 23906 we should get these values from the bios tables, but for now use 23907 some reasonable defaults. This should fix the washed out color 23908 problems on bugs 1082 and 12844. 23909 23910 commit 821acf38b716ab87c3d07263d6e4a139fe54803f 23911 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 23912 Date: Thu Nov 15 22:28:42 2007 -0500 23913 23914 RADEON: Make sure we set the MT properly for connected status unknown 23915 23916 commit a94123f33ec6584fbdfc4b9ecd543d1357de8814 23917 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 23918 Date: Thu Nov 15 22:19:54 2007 -0500 23919 23920 Revert "Portability fix from netbsd" 23921 23922 This reverts commit c9264aa53bf1470ad9104d1e7c4a8ce13c49c270. 23923 This breaks damage support. See bug 13244 23924 23925 commit 718bfd3b61879172eee819fdab7080d5d4c0a756 23926 Author: Dave Airlie <airlied (a] linux.ie> 23927 Date: Fri Nov 16 10:37:04 2007 +1000 23928 23929 r5xx: fix typo for crtc offset 23930 23931 commit e9d721c31372db045550f9562534b28f16121bb9 23932 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 23933 Date: Tue Nov 13 23:42:42 2007 +0100 23934 23935 ignore sometime bogus agp_mode bit from chip (bug #13190) 23936 23937 bit is wrong on at least X700 cards with rialto pcie-agp bridge chip, 23938 should be safe to use just the bit from the bridge hopefully to make 23939 agp setup work on these cards and not adversely affect others. 23940 23941 commit b865886d00907899297ae864358eb26d9980975d 23942 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 23943 Date: Sun Nov 11 20:38:41 2007 -0500 23944 23945 bump for RC release 23946 23947 commit 7aeb35e5ad1aed6e78a3d8565fbfbfe66232ab45 23948 Author: Alex Deucher <alex@botch2.(none)> 23949 Date: Sun Nov 11 19:29:30 2007 -0500 23950 23951 fix from last commit 23952 23953 commit 2ea95900547165e86ad3f8a41ce3331a05bad60e 23954 Author: Alex Deucher <alex@botch2.(none)> 23955 Date: Sun Nov 11 18:43:03 2007 -0500 23956 23957 Add full parsing support for atom bios connector table 23958 23959 commit 7ce730828c293f0810dfdc554df48dfd76e35c49 23960 Author: Alex Deucher <alex@botch2.(none)> 23961 Date: Sun Nov 11 12:37:01 2007 -0500 23962 23963 rework crtc output source setup 23964 23965 commit d61b6c78aa7810a2f9b9e2d9d95aab4295de80ce 23966 Author: Alex Deucher <alex@botch2.(none)> 23967 Date: Sun Nov 11 11:58:17 2007 -0500 23968 23969 make sure i2c bus exists before using it 23970 23971 commit 342e3e207efda42ba679731c30dfb9d5e9d5643f 23972 Author: Alex Deucher <alex@botch2.(none)> 23973 Date: Fri Nov 9 17:11:43 2007 -0500 23974 23975 combine outputs based on id 23976 23977 commit 8078c299d5941460243944d55051547c1a4d3791 23978 Author: Alex Deucher <alex@botch2.(none)> 23979 Date: Fri Nov 9 16:35:08 2007 -0500 23980 23981 use atom to program plls on r4xx 23982 23983 commit 5febe2c96642f61d006abe6e8081e69d5b95adc0 23984 Author: Alex Deucher <alex@botch2.(none)> 23985 Date: Fri Nov 9 16:24:56 2007 -0500 23986 23987 turn off vga control when using ext modes 23988 23989 commit 83f170c32c08c74a9e3466ffa0e0a0606c74427b 23990 Author: Alex Deucher <alex@botch2.(none)> 23991 Date: Thu Nov 8 19:33:13 2007 -0500 23992 23993 add pci ids for X1550PRO 23994 23995 commit 5cdcaba0f6e9de2d15cfcc109ab97d9fd423e3bf 23996 Author: Alex Deucher <alex@botch2.(none)> 23997 Date: Thu Nov 8 19:28:03 2007 -0500 23998 23999 make sure to assign gpio 24000 24001 commit 2dcb852778301b9284a2b4906dcf64f95ed638b7 24002 Author: Alex Deucher <alex@botch2.(none)> 24003 Date: Thu Nov 8 18:39:23 2007 -0500 24004 24005 pull in another of Dave's fixes 24006 24007 commit 96273016a0bbdfa4d3a4e6275a3b09eeeadaa534 24008 Author: Alex Deucher <alex@botch2.(none)> 24009 Date: Wed Nov 7 01:22:30 2007 -0500 24010 24011 fix and move crtc source set up atombios_output.c 24012 24013 it's really more output related. 24014 24015 commit 5c495c81cc3bcd4a38d06954243ed3bdc85bdc07 24016 Author: Alex Deucher <alex@botch2.(none)> 24017 Date: Wed Nov 7 01:04:11 2007 -0500 24018 24019 add support for initing external tmds via ATOM 24020 24021 commit 5c13d9355280e6de44ebbf8de7ea89a6b91c7388 24022 Author: Alex Deucher <alex@botch2.(none)> 24023 Date: Wed Nov 7 00:17:28 2007 -0500 24024 24025 add avivo output stuff 24026 24027 commit 68e7f5c67e2e9d2162b469ce31f452f3f89756b5 24028 Author: Alex Deucher <alex@botch2.(none)> 24029 Date: Tue Nov 6 23:43:29 2007 -0500 24030 24031 more avivo updates 24032 24033 commit 303562dfb57e13c027b2aa9289d54e547c829ff1 24034 Author: Alex Deucher <alex@botch2.(none)> 24035 Date: Tue Nov 6 23:06:46 2007 -0500 24036 24037 add additional connector types 24038 24039 commit 0d3e0735f710cb7b9505e4330997aa332f73c102 24040 Author: Alex Deucher <alex@botch2.(none)> 24041 Date: Tue Nov 6 22:59:25 2007 -0500 24042 24043 First round of avivo support 24044 24045 commit 0abfe3150ce3eed4db93ccc2975bd4622dfa54a7 24046 Author: Alex Deucher <alex@botch2.(none)> 24047 Date: Tue Nov 6 18:47:00 2007 -0500 24048 24049 Add atombios files 24050 24051 commit 20f01950e42babc308b4470df6a3c6628c932003 24052 Author: Alex Deucher <alex@botch2.(none)> 24053 Date: Tue Nov 6 18:04:43 2007 -0500 24054 24055 for r4xx ATOM cards, just use ATOM for PLL 24056 24057 while crtc timing and pll seem to work fine, output setup 24058 and routing don't seem to work too reliably with atom. 24059 AMD claims ATOM was still pretty new with r4xx so 24060 it's probably better to stick with direct programming for 24061 some things. 24062 24063 commit 37af2f6b52c8f217b205486742a6b46a864da8ed 24064 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24065 Date: Mon Nov 5 10:58:25 2007 +0100 24066 24067 radeon: Use indirect register access for (un)hiding the cursor. 24068 24069 This avoids hangs that occur when touching these registers while the card is 24070 busy with some cards. 24071 24072 commit 5659e2f1b08e44d0c4f21a403e91b4801ef78f62 24073 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24074 Date: Mon Nov 5 10:32:04 2007 +0100 24075 24076 radeon: Fix build warning on powerpc. 24077 24078 commit d41d73f5f9d8baee7137e94da2c7852ea0412180 24079 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24080 Date: Mon Nov 5 10:30:23 2007 +0100 24081 24082 Revert "radeon: Reinstate sync in radeon_crtc_show/hide_cursor." 24083 24084 This reverts commit 0241cac643fa1c08a45ea44f5c670b290e760ad8. 24085 24086 Taking the DRI lock in paths called from the SIGIO handler isn't safe. For 24087 example, the SIGIO handler may be called while the lock is held by an AIGLX 24088 context, resulting in deadlock. See e.g. 24089 https://bugs.freedesktop.org/show_bug.cgi?id=13005 . 24090 24091 commit 78a3eabff382e8ebe33df2039076fb083bcc361b 24092 Author: Alex Deucher <alex@botch2.(none)> 24093 Date: Sun Nov 4 14:11:26 2007 -0500 24094 24095 WIP: get ATOM crtc stuff working on r4xx 24096 24097 commit 5e9ebd8e496b72b051053d637c63b2956b7861d3 24098 Author: Alex Deucher <alex@botch2.(none)> 24099 Date: Sat Nov 3 20:46:17 2007 -0400 24100 24101 Initial support for r4xx 24102 24103 - add r3xx/r4xx MC access macros and functions 24104 24105 commit c106075ccb81ca2ee4894743e676fd37653c8dce 24106 Author: Alex Deucher <alex@botch2.(none)> 24107 Date: Sat Nov 3 18:29:20 2007 -0400 24108 24109 More of Dave's ATOM init code. 24110 24111 commit e18f5d61806b445ad77d93e258fbce9422b52bb6 24112 Author: Alex Deucher <alex@botch2.(none)> 24113 Date: Sat Nov 3 18:20:55 2007 -0400 24114 24115 Initial integration of Atom code and some of Dave's code. 24116 24117 ATOM builds, but it's not hooked up yet. 24118 24119 commit aaafc6a8f33d4736c37c43d3fd5e43085a2590fb 24120 Author: Brice Goglin <bgoglin (a] debian.org> 24121 Date: Sun Oct 28 21:10:32 2007 +0100 24122 24123 Fix typo in 'disposing cursor' 24124 24125 Bug #12984. 24126 24127 commit a5c55c418319c1315ad73c305ef48b44fe1d9580 24128 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24129 Date: Sat Oct 27 19:46:11 2007 -0400 24130 24131 RADEON: disable FP DETECT for DVI. 24132 24133 These bits don't seem to be reliable or routed right 24134 in all cases. 24135 24136 commit 49cf7cb3db36ce7734f7e314a040a240191d6477 24137 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24138 Date: Thu Oct 25 21:23:09 2007 -0400 24139 24140 RADEON: Fix IgnoreEDID option 24141 24142 commit 70bbc0cfe699588f8ae4123efca0e1baffc91000 24143 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24144 Date: Thu Oct 25 20:56:11 2007 -0400 24145 24146 RADEON: clean up mode handling and improve reliability of DDC 24147 24148 - Reduce the number of times we do DDC and don't 24149 skip the magical GPIO dance when doing DDC. 24150 - If no DDC modes are found, add the screen modes 24151 24152 commit 1a4f590c2ac071f0c348f38e9f26fa6093722453 24153 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24154 Date: Thu Oct 25 19:25:36 2007 -0400 24155 24156 RADEON: document DefaultTMDSPLL option 24157 24158 commit 2d26e0d0bcf25e210dc7771e76638dcfafcc2356 24159 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24160 Date: Thu Oct 25 19:18:18 2007 -0400 24161 24162 RADEON: Add driver option to force TV DAC load detection on 24163 24164 commit a9306b7986467c0794c91625ad2ece597b718462 24165 Author: Alex Deucher <alex@botch2.(none)> 24166 Date: Thu Oct 18 20:06:58 2007 -0400 24167 24168 RADEON: fix possible crash in radeon_crtc_mode_set() 24169 24170 when radeon_crtc_mode_set() is called during ScreenInit(), 24171 we don't have pScrn->pScreen yet. 24172 24173 commit 115b4b65ef679a46d90b7fc8ac22ace37038b3fc 24174 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24175 Date: Wed Oct 17 20:25:47 2007 -0400 24176 24177 RADEON: Remove LVDSBiosNativeMode Option 24178 24179 Shouldn't be needed any longer as I've finally sorted out 24180 the LVDS issues due to the crtc setup. 24181 24182 commit f2eb5c1cc69a4f7b0754ec6b2efde4fa1650ab6d 24183 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24184 Date: Tue Oct 16 18:54:28 2007 -0400 24185 24186 RADEON: finish fixing LVDS/RMX 24187 24188 For panels, both the mode and crtc values need to be updated based 24189 on the native mode timing. This shoudl fix any remaining problems 24190 with RMX. 24191 24192 commit 93561688087493fccb7ac387d728feec61212180 24193 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24194 Date: Tue Oct 16 10:57:06 2007 +0200 24195 24196 radeon: Handle 64 bit PCI resource bases better. 24197 24198 commit 08945fc8a14ba57e32350e85daa3354619e9e72e 24199 Author: Alex Deucher <alex@botch2.(none)> 24200 Date: Mon Oct 15 21:12:46 2007 -0400 24201 24202 RADEON: allow the user to set any tv standard 24203 24204 allow the user to set any tv standard rather than 24205 limiting it to what the bios says is supported. 24206 24207 commit 571548b2b95ce289f26a77c11deb6266f0b18516 24208 Author: Alex Deucher <alex@botch2.(none)> 24209 Date: Mon Oct 15 20:33:56 2007 -0400 24210 24211 RADEON: use native timing for RMX 24212 24213 RMX should work again with native timings after the previous 24214 patch. 24215 24216 commit 2f87bff293a343b40c1be096933a5ae126632468 24217 Author: Alex Deucher <alex@botch2.(none)> 24218 Date: Mon Oct 15 20:06:28 2007 -0400 24219 24220 RADEON: Fix subtle change in crtc reg init 24221 24222 At some point we changed how hsync_wid and vsync_wid were clipped. 24223 Previously we used the field size as a mask when building the 24224 register. This got changed to setting the value to the field 24225 size if it was greater. this probably explains number stange 24226 mode bugs. 24227 24228 commit fd5bb7bb5e968127b87102320eccc1222f205e5a 24229 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24230 Date: Sat Oct 13 18:56:43 2007 +0200 24231 24232 radeon: Improve detection of default value for Option "MacModel" on Linux. 24233 24234 * Detect all PowerBooks with dual link DVI (according to developer.apple.com) 24235 and both Mac Mini G4 models. Other PowerBooks and iBooks should be covered 24236 by the 'detected as' line in /proc/cpuinfo. 24237 * Give specific instructions for overriding and reporting incorrect detection. 24238 * Only perform detection when Option "MacModel" isn't present with a valid 24239 value. 24240 * Close /proc/cpuinfo file handle when done with it. 24241 * Coding style cleanups. 24242 24243 commit fbf121add5584049627f07345fc502b2aefc88ce 24244 Author: Eugen Dedu <Eugen.Dedu (a] pu-pm.univ-fcomte.fr> 24245 Date: Sat Oct 13 16:51:58 2007 +0200 24246 24247 radeon: Derive default value of Option "MacModel" from /proc/cpuinfo on Linux. 24248 24249 commit 5db3afaa1fdb69d382ac769ef40191a4b964d28e 24250 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24251 Date: Thu Oct 11 20:25:20 2007 -0400 24252 24253 RADEON: return status unknown for flaky chip/connector combinations 24254 24255 This should at least get something on the screen. 24256 XPRESS chips, I'm looking in your general direction... 24257 24258 commit 1148d332f52f3780897aae3fcd5d6a67687c42cd 24259 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24260 Date: Thu Oct 11 19:09:45 2007 -0400 24261 24262 RADEON: fix potential crash in mode handling 24263 24264 commit 905b19df33d6f912336b65895e97775ffec07728 24265 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24266 Date: Thu Oct 11 18:55:08 2007 -0400 24267 24268 RADEON: small cleanup from last commit 24269 24270 commit 7afd04c1e4ffa6e4e5ba08ae90ba002237dc282b 24271 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24272 Date: Thu Oct 11 00:16:45 2007 -0400 24273 24274 RADEON: tell the bios not to muck with the hardware while the driver is active 24275 24276 by toggling the appropriate bios scratch regs you can tell 24277 the bios not the touch the hw while the driver is active. 24278 This should prevent the bios from scrambling the hardware 24279 when users open the lid or toggle bios hotkeys. 24280 24281 fixes bug 12567 24282 24283 commit 1b231d28fdda5cdc44bb9d2075d4edfd8f17e21f 24284 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24285 Date: Wed Oct 10 21:54:46 2007 -0400 24286 24287 RADEON: attempt to detect lid status on legacy bios 24288 24289 I'm not sure how reliable this is. The lid closed 24290 bit only seems to be set if the lid is closed at boot 24291 so LVDS will default to on if you restart X with the 24292 lid closed after boot. 24293 24294 commit 47ef583b993607bc3a372dd125c6e5228c4ec7bf 24295 Author: Alex Deucher <alex@botch2.(none)> 24296 Date: Wed Oct 10 19:35:31 2007 -0400 24297 24298 RADEON: more XPRESS connector table hacks 24299 24300 Seems XPRESS desktop chips with DVI-D list a proprietary 24301 connector type rather than DVI-D. try and do the right thing. 24302 Also, XPRESS chips seem to include connector type 8 and ddc 24303 type 8 which are unknown at this time. 24304 24305 commit c9264aa53bf1470ad9104d1e7c4a8ce13c49c270 24306 Author: Joerg Sonnenberger <joerg (a] britannica.bec.de> 24307 Date: Tue Oct 9 23:18:29 2007 -0400 24308 24309 Portability fix from netbsd 24310 24311 commit 80eee856938756e1222526b6c39cee8b5252b409 24312 Author: Matthieu Herb <> 24313 Date: Tue Oct 9 16:17:50 2007 -0400 24314 24315 RADEON: fix console restore on netbsd 24316 24317 Include the mode restore bugfix from monolithic Xorg, that is derived 24318 from the version in xsrc which in turn was provided by Matthieu Herb 24319 over 3 years ago on the XFree86 lists. Suggested by various 24320 developers, hold-back due to the working state in xorg-server 1.1.1. 24321 Tracing down the exact change showed that the changed default color 24322 depth made this issue a lot more prominent again. Discussed with Eric 24323 Anholt. 24324 24325 commit 547fef4c6382f8a4951c086ee531b804cba31075 24326 Author: Lars Weiler <pylon (a] gentoo.org> 24327 Date: Tue Oct 9 16:10:34 2007 -0400 24328 24329 include stdint.h to fix the build on some platforms 24330 24331 fixes bug 12733 24332 24333 commit ddec5783103045084a66a34d71326cc6e4a3f598 24334 Author: George Sapountzis <gsap7 (a] yahoo.gr> 24335 Date: Tue Oct 9 13:27:27 2007 +0300 24336 24337 theatre compiles with pci-rework. 24338 24339 commit 051435610a66735fd455bbb526fa294fcfe8c0b6 24340 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24341 Date: Sun Oct 7 19:39:47 2007 -0400 24342 24343 RADEON: still more LVDS fixes 24344 24345 Seems some laptops need the native mode from the bios for 24346 LVDS while others seem to prefer a CVT mode. Add an option 24347 to pick the preferred mode. The default it to use the bios 24348 table timing. 24349 24350 commit 0b03a73b7dcb4aa192c42f2a4c842d324c358122 24351 Author: Dave Airlie <airlied (a] linux.ie> 24352 Date: Sat Oct 6 20:11:19 2007 +1000 24353 24354 radeon: add support for DDC type 6 24355 24356 thanks to arekm on xorg-devel for testing/feedback 24357 24358 commit 83d7506a603889b6c18736cff9b2cd3296c95c84 24359 Author: Alex Deucher <alex@botch2.(none)> 24360 Date: Fri Oct 5 20:25:04 2007 -0400 24361 24362 Bump for RC release 24363 24364 commit fe66cc4e02a52e1d014ae5de192c65e8abd7f58a 24365 Author: Tormod Volden <bugzi06.fdo.tormod (a] xoxy.net> 24366 Date: Fri Oct 5 20:10:55 2007 -0400 24367 24368 RADEON: rename to PCI_CHIP_RS485_5975 everywhere 24369 24370 RS482_5975 is actually RS485. see bug 12048 24371 24372 commit 314e8fdbe5369e4c1786d176d03d0d2a046f25d5 24373 Author: Tormod Volden <bugzi06.fdo.tormod (a] xoxy.net> 24374 Date: Fri Oct 5 20:05:54 2007 -0400 24375 24376 RADEON: Disable DRI by default for RS482 cards 24377 24378 fixes bug 12048 24379 24380 commit abbafe6db7c0d24efc61a05950d4214c00d3f043 24381 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24382 Date: Thu Oct 4 11:33:09 2007 +0200 24383 24384 radeon: Fix compile warnings. 24385 24386 commit 1110957ba703475847b8e72023c6edcbd1068aef 24387 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24388 Date: Thu Oct 4 11:32:48 2007 +0200 24389 24390 radeon: Fix for pci-rework. 24391 24392 Surprisingly easy, thanks to George's pci-rework changes. 24393 24394 commit 7addf41885ec5658f531624a9c24ea5bd7d22d19 24395 Merge: 3a958ba13 78c6bd305 24396 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24397 Date: Thu Oct 4 11:06:08 2007 +0200 24398 24399 Merge remote branch 'origin/pci-rework' 24400 24401 commit 3a958ba136c3fae5a4ddd56373ac7cd47046f10e 24402 Author: Alex Deucher <alex@botch2.(none)> 24403 Date: Wed Oct 3 22:59:14 2007 -0400 24404 24405 RADEON: remove some cruft 24406 24407 commit b6bda79f72df5e5bf9c6b71fa3298e765da506bd 24408 Author: Alex Deucher <alex@botch2.(none)> 24409 Date: Wed Oct 3 22:56:27 2007 -0400 24410 24411 RADEON: more fixes for single crtc chips 24412 24413 commit 5f5c4e6ad61c45c24f1443b91b4bc5375efdebc0 24414 Author: Alex Deucher <alex@botch2.(none)> 24415 Date: Wed Oct 3 22:46:21 2007 -0400 24416 24417 RADEON: remove RADEONSaveMode() 24418 24419 It's only called by RADEONSave(), so just move its 24420 contents into RADEONSave(). this mirrors RADEONRestore(). 24421 24422 commit bfede412b3a3cd11769a580b167c528734146096 24423 Author: Alex Deucher <alex@botch2.(none)> 24424 Date: Wed Oct 3 22:19:16 2007 -0400 24425 24426 RADEON: minor fixes for external TMDS 24427 24428 commit 597dffce9bdc200003d0be880235258386a0bdd7 24429 Author: Alex Deucher <alex@botch2.(none)> 24430 Date: Wed Oct 3 21:43:08 2007 -0400 24431 24432 RADEON: final fix for RMX/LVDS 24433 24434 It seems that on quite a few laptops the LVDS H/V timing 24435 from the bios tables are way off or just doesn't work period. 24436 Either we are using it wrong or we need some additional checks 24437 when we parse it. Only the dot clock seems to really matter, 24438 so use cvt modes and update the dotclock. This seems to work 24439 correctly in all cases. 24440 24441 commit cc0c2d8e61600652b1f9cb3dc49db2ef62b1e40d 24442 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24443 Date: Wed Oct 3 00:59:37 2007 -0400 24444 24445 RADEON: Finally sort out LVDS modes 24446 24447 - The panel timing from the bios tables is for the crtc, 24448 not the native panel mode, so add cvt modes for the panel 24449 and fix up the crtc values in mode_fixup(). 24450 - rename and reconfigure RADEONValidateFPModes() to what it 24451 really does: add screen modes. I suppose for backwards 24452 compatibility we ought to add the screen modes to every 24453 reasonable output, but everyone should really use xrandr 24454 or an output monitor section to add custom modes. Hopefully 24455 this will go away at some point. 24456 24457 commit 0ca184c3c35032df39ea7ce5d2d4aba1a97b6426 24458 Author: Alex Deucher <alex@botch2.(none)> 24459 Date: Tue Oct 2 19:38:18 2007 -0400 24460 24461 RADEON: fix RADEONInitDispBandwidth() on single crtc cards 24462 24463 Should fix bug 12490 24464 24465 commit a5fb2a5d54cdd562a97881cdea1a4af35b483aab 24466 Author: Dave Airlie <airlied (a] linux.ie> 24467 Date: Mon Oct 1 08:48:27 2007 +1000 24468 24469 radeon: fix powerpc build 24470 24471 commit 22519fde1e002f28d6036d448fcd18452d00f1bb 24472 Author: Alex Deucher <alex@botch2.(none)> 24473 Date: Sun Sep 30 13:11:20 2007 -0400 24474 24475 RADEON: add support for ext tmds table and ext tmds chip init 24476 24477 This probably won't work on all chips as the various gpio lines 24478 seem to need special magic to to actually talk to the i2c slave 24479 chips. 24480 24481 commit dcc376e2d2a13329dd03f1bc4b471329757a6f5f 24482 Author: Alex Deucher <alex@botch2.(none)> 24483 Date: Wed Sep 26 23:27:38 2007 -0400 24484 24485 RADEON: remove cruft 24486 24487 commit 99ceaefa18c6e07b55106cca0ea8996fa73667be 24488 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24489 Date: Sun Sep 30 12:02:54 2007 -0400 24490 24491 RADEON: more work on LVDS mode validation and fixups 24492 24493 Hopefully this will fix up the weird mode issues with LVDS and 24494 native and RMX modes, but I can't seem to get quite the right 24495 combo to fix everyone. 24496 24497 commit d808781d48adf01e80b5bb476bae2d2f599030f1 24498 Author: Dave Airlie <airlied (a] redhat.com> 24499 Date: Fri Sep 28 14:50:42 2007 +1000 24500 24501 rc410: disable DRI by default due to it not working 24502 24503 commit 4d9f944e40c6b9ac7e80428dff49c6e83fda1d8f 24504 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24505 Date: Thu Sep 27 17:50:15 2007 +0200 24506 24507 radeon: Fix build on little endian platforms. 24508 24509 This was accidentally broken when moving away from using pixmap data pointers 24510 directly. 24511 24512 commit 5902e913010a2fb0c17efe01f9aade73f9646f16 24513 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24514 Date: Thu Sep 27 12:53:34 2007 +0200 24515 24516 radeon: EXA pitch fixes. 24517 24518 The combined pitch/offset registers only support pitches up to 16320 bytes. 24519 24520 With EXA >= 2.3, set the maxPitchBytes field accordingly. With older versions, 24521 limit maxX such that the pitch of 32bpp pixmaps doesn't exceed the limit. 24522 24523 Also check the limit in RADEONGetOffsetPitch just in case. 24524 24525 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=12551 . 24526 24527 commit ca728973bce504e03bf6228d18287906c35eaa3f 24528 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24529 Date: Thu Sep 27 11:14:21 2007 +0200 24530 24531 radeon: Don't use pixmap data pointer directly with EXA. 24532 24533 Fixes segfault with current xserver master without AccelDFS. 24534 24535 commit 21593d04d222b05dbba9abd31eaa3bfb91d999b6 24536 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24537 Date: Mon Sep 24 23:01:20 2007 -0400 24538 24539 RADEON: more clean up of mode code 24540 24541 also print out LVDS mode info from bios for both 24542 legacy and atom bioses 24543 24544 commit 22cccb99d762567757d3fd38795f71e943f1faf9 24545 Author: Alex Deucher <alex@botch2.(none)> 24546 Date: Mon Sep 24 00:10:01 2007 -0400 24547 24548 bump for RC release 24549 24550 commit fad9e7b00de051f632000eb2304ac23a524f7c8e 24551 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24552 Date: Sun Sep 23 23:43:26 2007 -0400 24553 24554 RADEON: Rework LVDS mode validation 24555 24556 The old validation code wasn't really well suited to randr. 24557 This fixes several issues: 24558 - missing display size for panels with edid 24559 - broken duplicate modes 24560 24561 commit bf9674b9df351108e069d037ec10869a4154a881 24562 Author: Alex Deucher <alex@botch2.(none)> 24563 Date: Sun Sep 23 15:01:34 2007 -0400 24564 24565 RADEON: properly set default RMX types 24566 24567 commit b27135bce8b41d69290613b440a338b0a7fe0200 24568 Author: Alex Deucher <alex@botch2.(none)> 24569 Date: Sun Sep 23 14:44:38 2007 -0400 24570 24571 RADEON: fix up LVDS handling for r3xx and newer 24572 24573 commit d5cf268be2c9b3e320d7101f5213f7d5609b6308 24574 Author: Alex Deucher <alex@botch2.(none)> 24575 Date: Sat Sep 22 18:31:51 2007 -0400 24576 24577 RADEON: Fix crash when Xv window is outside of either crtc 24578 24579 commit 8aca0ca6be7122e0ece20985a1862553a1b225c0 24580 Author: Alex Deucher <alex@botch2.(none)> 24581 Date: Sat Sep 22 17:56:50 2007 -0400 24582 24583 Revert "RADEON: fix crash when Xv window is outside of either crtc" 24584 24585 This reverts commit 9109e62e3be7f96b41b534ab517fdf1baf458806. 24586 24587 This breaks ABI. better fix to come. 24588 24589 commit d6e1d9be1cd5786a24ed4bb8e45fa7e04df56622 24590 Author: Alex Deucher <alex@botch2.(none)> 24591 Date: Sat Sep 22 16:14:06 2007 -0400 24592 24593 RADEON: Don't make the entity as shareable 24594 24595 This should prevent people from mistakenly trying to run 24596 zaphod mode 24597 24598 commit 81114af1cdddc0b10d076f2e38c7a00c1223cc48 24599 Author: Alex Deucher <alex@botch2.(none)> 24600 Date: Sat Sep 22 15:51:23 2007 -0400 24601 24602 RADEON: preliminary support for mac mini 24603 24604 Option "MacModel" "mini" 24605 may not be 100% correct yet 24606 24607 commit 6c482e453bc8156886294d0c1b8f3f1b3dcf4b36 24608 Author: Alex Deucher <alex@botch2.(none)> 24609 Date: Sat Sep 22 15:11:20 2007 -0400 24610 24611 RADEON: Fix RMX on LVDS 24612 24613 LVDS + RMX doesn't seem to like having the crtc values tweaked. 24614 24615 commit 9109e62e3be7f96b41b534ab517fdf1baf458806 24616 Author: Michel Dnzer <daenzer (a] debian.org> 24617 Date: Sat Sep 22 14:19:10 2007 -0400 24618 24619 RADEON: fix crash when Xv window is outside of either crtc 24620 24621 Should fix bug 12175 24622 24623 commit d2ce4a5003ce1291ea2327b2c00a0b24408fe26c 24624 Author: Alex Deucher <alex@botch2.(none)> 24625 Date: Sat Sep 22 10:41:55 2007 -0400 24626 24627 RADEON: RMX updates 24628 24629 - add option to turn off RMX 24630 - turn off RMX by default on DVI 24631 - add infrastructure to support more RMX modes 24632 24633 commit f95b9ab729376083bf0d12987ee260ec2aba721e 24634 Author: Alex Deucher <alex@botch2.(none)> 24635 Date: Fri Sep 21 00:12:37 2007 -0400 24636 24637 RADEON: remove more old cruft 24638 24639 commit 5a6f74103f0ec0d451d0e2573442efe5922848af 24640 Author: Maciej Cencora <m.cencora (a] gmail.com> 24641 Date: Thu Sep 20 23:56:08 2007 -0400 24642 24643 RADEON: fix video in on RV380 (tested on X600 VIVO) 24644 24645 commit c72a365386e19f9257db041d44b09ad499cc9f6a 24646 Author: Alex Deucher <alex@botch2.(none)> 24647 Date: Thu Sep 20 23:49:57 2007 -0400 24648 24649 RADEON: fix up dvo support (still no external chip init) 24650 24651 commit 5e4d98470b6412a686883c554e7eb7badbe78c4d 24652 Author: Alex Deucher <alex@botch2.(none)> 24653 Date: Thu Sep 20 23:22:48 2007 -0400 24654 24655 RADEON: round 3 on the PLLs. should fix the LVDS issues 24656 24657 commit c5e2a2f09af807006c7ea493a8e90ff77abe207c 24658 Author: Alex Deucher <alex@botch2.(none)> 24659 Date: Wed Sep 19 19:58:28 2007 -0400 24660 24661 bump for RC release 24662 24663 commit 46ff78b9f010ca24178d4363761be00eb3ecb632 24664 Author: Alex Deucher <alex@botch2.(none)> 24665 Date: Wed Sep 19 19:49:58 2007 -0400 24666 24667 RADEON: add default connector setup for single crtc chips 24668 24669 commit 509ca0cb0e1cde905b47db2bbac6f2a58523b279 24670 Author: Alex Deucher <alex@botch2.(none)> 24671 Date: Wed Sep 19 19:41:17 2007 -0400 24672 24673 RADEON: fix external TMDS parsing in legacy bios connector table 24674 24675 commit 5a59547ef7e986c9613023d941c3354cd476faf1 24676 Author: Alex Deucher <alex@botch2.(none)> 24677 Date: Wed Sep 19 19:28:09 2007 -0400 24678 24679 Revert "RADEON: initial pass at external TMDS support" 24680 24681 This reverts commit 4000a710c93dd2d82891e4082bc7fa922ba9c5f4. 24682 24683 This needs to be reworked and needs more soak time. so revert 24684 for now. 24685 24686 commit dc333884c540d536bffe51a5ebfaf1822af6a91c 24687 Author: Alex Deucher <alex@botch2.(none)> 24688 Date: Tue Sep 18 22:23:12 2007 -0400 24689 24690 RADEON: fix plls again 24691 24692 previous fix seemed to break other chips. Lets see how 24693 this goes. 24694 24695 commit 271e541088f455d1cfedff87e88c9a4fdbdbe424 24696 Author: Alex Deucher <alex@botch2.(none)> 24697 Date: Tue Sep 18 22:12:31 2007 -0400 24698 24699 RADEON: Remove more mergedfb cruft 24700 24701 commit 8ae69c496eba701e744cca0605f73242673f7b3f 24702 Author: Alex Deucher <alex@botch2.(none)> 24703 Date: Tue Sep 18 20:12:21 2007 -0400 24704 24705 RADEON: adjust pll restore 24706 24707 some chips seem to be pickier than others. fixes bug 12467 24708 24709 commit 8a84dc8ee260716026fa661a6dced2000621efa2 24710 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24711 Date: Tue Sep 18 11:53:39 2007 +0200 24712 24713 radeon: Remove declaration of unused variable. 24714 24715 commit 2fbbd9c5f0ec27e72780e8b054bc59502f1e5651 24716 Author: Alex Deucher <alex@botch2.(none)> 24717 Date: Tue Sep 18 00:45:56 2007 -0400 24718 24719 RADEON: note that MergedFB support has been replaced with randr 1.2 24720 24721 commit 81b708dbd35c29bf2d67e8c655af97b60e61498b 24722 Author: Alex Deucher <alex@botch2.(none)> 24723 Date: Mon Sep 17 23:54:03 2007 -0400 24724 24725 RADEON: turn off load detection on tv dac by default 24726 24727 it's just too unreliable and even when it is most people 24728 don't want it. enable it via output attribute or force 24729 the outputs on with xrandr 24730 24731 commit c3ab53159d42157985a25e9125eaa5362273b73d 24732 Author: Alex Deucher <alex@botch2.(none)> 24733 Date: Mon Sep 17 23:28:51 2007 -0400 24734 24735 RADEON: XPRESS connector tables are crap 24736 24737 besides specifying the pimary DAC (which they don't have), 24738 most seem to specify DDC_CRT2 for DDC, which does not seem 24739 to ever work. Some have reported success with DDC_MONID so 24740 lets give that a try. 24741 24742 commit 1fafa970fa78ddea29a8804b6cd0aea52e4fa2e1 24743 Author: Alex Deucher <alex@botch2.(none)> 24744 Date: Mon Sep 17 23:10:13 2007 -0400 24745 24746 RADEON: ignore edid digital bit on analog ports 24747 24748 Some monitors have this bit set even on analog ports 24749 24750 commit 4000a710c93dd2d82891e4082bc7fa922ba9c5f4 24751 Author: Alex Deucher <alex@botch2.(none)> 24752 Date: Mon Sep 17 23:04:17 2007 -0400 24753 24754 RADEON: initial pass at external TMDS support 24755 24756 - Based on Dave Airlie's initial work and the dvo support 24757 in the intel driver. 24758 - Only sil164 is supported at the moment. 24759 - Once we get some testing and such, we ought to move the dvo 24760 drivers out of the drivers so they can be shared among all drivers 24761 - Doesn't seem to work on my card (r4xx ATOM) card ATM 24762 - Legacy bios table programming sequence not implemented yet 24763 24764 commit 38515d402555eaa61c686d42973e59f659b07466 24765 Author: Alex Deucher <alex@botch2.(none)> 24766 Date: Sat Sep 15 11:39:47 2007 -0400 24767 24768 RADEON: don't restore bios scratch regs 24769 24770 Leave them as programmed by the bios 24771 fixes bug 12424 24772 24773 commit 78c6bd305f2ea70629d1dda3c0c48c067451f5fb 24774 Author: George Sapountzis <gsap7 (a] yahoo.gr> 24775 Date: Mon Sep 10 19:00:16 2007 +0300 24776 24777 [mach64] Revert to multiple mappings per BAR. 24778 24779 libpciaccess added subrange mappings and mtrr handling. 24780 24781 compile-tested only. 24782 24783 commit 0241cac643fa1c08a45ea44f5c670b290e760ad8 24784 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24785 Date: Sat Sep 8 00:22:40 2007 +0200 24786 24787 radeon: Reinstate sync in radeon_crtc_show/hide_cursor. 24788 24789 Make sure the DRI lock is held though, as these can be called asynchronously. 24790 24791 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=12245 . 24792 24793 commit 49933e2f7d590811f2bc8c0d51a09f3b7f14845e 24794 Author: Alex Deucher <alex@botch2.(none)> 24795 Date: Sat Sep 1 12:23:07 2007 -0400 24796 24797 RADEON: automatically disable tiling if requested virtual desktop exceeds surface limits 24798 24799 Also, tweak default desktop sizes and add some informational messages 24800 24801 commit d49a60bc939c9f3e9d57c23e44263f5bb52f09fb 24802 Author: Sjoerd Simons <sjoerd (a] luon.net> 24803 Date: Thu Aug 30 11:40:13 2007 +0200 24804 24805 radeon: Fix Option "MacModel". 24806 24807 commit 2b860726fff058623e0e8e1b0aca092c246875ce 24808 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24809 Date: Thu Aug 30 11:38:28 2007 +0200 24810 24811 radeon: Don't build currently unused RADEONSavePalette. 24812 24813 commit 12187a6aa93049c002a4171344d03c713f7f3c5d 24814 Author: Alex Deucher <alex@botch2.(none)> 24815 Date: Wed Aug 29 23:11:30 2007 -0400 24816 24817 RADEON: Add quirk and connector tables for apple laptops 24818 24819 As far as I can tell there are three apple laptop variants: 24820 ibook - LVDS, TVDAC drives TV or VGA via dongle 24821 powerbook-duallink - LVDS, TV, External TMDS/Primary DAC 24822 powerbook - LVDS, TV, Internal TMDS/Primary DAC 24823 use Option "MacModel" "<string>" 24824 to enable the appropriate quirks where string is one of the above 24825 24826 We can't yet init the external TMDS directly, but if OF inits it, 24827 it should work. This should also fix bug 9955. 24828 24829 Please test! 24830 24831 commit 61c1fdaa8553581944f78a11e6f9aa76163a468a 24832 Author: Alex Deucher <alex@botch2.(none)> 24833 Date: Tue Aug 28 23:47:19 2007 -0400 24834 24835 RADEON: add option to force tmds pll to default table 24836 24837 Also rework the tmds pll output attribute handling a bit 24838 24839 commit 0d9087bc60bb95c770b899cfed29699c02bdac49 24840 Author: Alex Deucher <alex@botch2.(none)> 24841 Date: Tue Aug 28 23:08:20 2007 -0400 24842 24843 RADEON: remove some cruft forgotten in a previous commit 24844 24845 commit 6ff0645ecfe65727e8ef5d5e6215b4e03078e1a5 24846 Author: George Sapountzis <gsap7 (a] yahoo.gr> 24847 Date: Tue Aug 28 15:37:11 2007 +0300 24848 24849 [mach64] Convert to pci-rework, keeping source-code compatibility. 24850 24851 It still uses the old probe method though, this is due to the ati wrapper. 24852 24853 commit 7b38d9a1209f87255e5bb0aefe46a363ce4fb6ef 24854 Merge: 2e3d43af1 673f79972 24855 Author: George Sapountzis <gsap7 (a] yahoo.gr> 24856 Date: Tue Aug 28 16:01:12 2007 +0300 24857 24858 Merge branch 'master' into pci-rework 24859 24860 Conflicts: 24861 24862 src/ati.c 24863 24864 commit 673f799729824f4439dd5f681f75dd5aab50947f 24865 Author: Alex Deucher <alex@botch2.(none)> 24866 Date: Tue Aug 28 00:42:30 2007 -0400 24867 24868 RADEON: Update tv attributes immediately 24869 24870 commit ad6f7ad1b2ccae0bc0a416b9b0ca22709c9d5199 24871 Author: Alex Deucher <alex@botch2.(none)> 24872 Date: Tue Aug 28 00:08:41 2007 -0400 24873 24874 RADEON: remove the "default" tv_standard option 24875 24876 commit 17e0f9e6cbfdb115034d327bd34d46339fd632b7 24877 Author: Alex Deucher <alex@botch2.(none)> 24878 Date: Mon Aug 27 23:59:03 2007 -0400 24879 24880 RADEON: enable load detection for tvdac if output count for tvdac < 2 24881 24882 commit 42839fb5a8584196e7b18375bff6c426ed0347d9 24883 Author: Alex Deucher <alex@botch2.(none)> 24884 Date: Mon Aug 27 23:44:13 2007 -0400 24885 24886 RADEON: make load detection an output attribute for analog outputs 24887 24888 Since TV/VGA/DVI-I can share the TV DAC, we often get false detection 24889 of all inputs that share that DAC. Make load detection an output 24890 attribute. Enabled by default on primary dac and on cards where 24891 tv dac is (usually) dedicated to tv (non-IGP mobilities). 24892 24893 commit a5a1055d64ab4fa16bfb03a412ae6c4fe69ff65d 24894 Author: Alex Deucher <alex@botch2.(none)> 24895 Date: Mon Aug 27 22:42:22 2007 -0400 24896 24897 RADEON: make tmds pll an output attribute 24898 24899 sometimes the bios tmds plls are busted for certain 24900 monitors. sometimes the dirver tables are. Let the user 24901 pick at run time. 24902 24903 commit a12e4aa01bf1c5723c3c791ff9bdc26eef21d5ea 24904 Author: Alex Deucher <alex@botch2.(none)> 24905 Date: Sun Aug 26 18:51:29 2007 -0400 24906 24907 Bump for new release 24908 24909 commit d43596e5f5d7c60f96b57bc3e743a9b40eb7109d 24910 Author: Alex Deucher <alex@botch2.(none)> 24911 Date: Sun Aug 26 18:07:50 2007 -0400 24912 24913 RADEON: Fix rotation. works now 24914 24915 commit 47eb3327c258bb0cfd9a1d5677624b9988a39057 24916 Author: Alex Deucher <alex@botch2.(none)> 24917 Date: Sun Aug 26 15:43:22 2007 -0400 24918 24919 RADEON: minor tweak to tv out 24920 24921 commit f36720377737210c985b196d9a988efdd767f1c7 24922 Author: Alex Deucher <alex@samba.(none)> 24923 Date: Sun Aug 26 14:13:06 2007 -0400 24924 24925 RADEON: fix depth 16 palette for real this time 24926 24927 commit f2b13f1457bf860b075310d3962254be0ed7bea3 24928 Author: Alex Deucher <alex@botch2.(none)> 24929 Date: Sun Aug 26 13:27:19 2007 -0400 24930 24931 RADEON: Only update pixclks_cntl when updating tv routing 24932 24933 No need to re-set all of the pll2 stuff 24934 24935 commit 5d044b9f74c7aa7e12f2822896fed881e2ca9d19 24936 Author: Alex Deucher <alex@botch2.(none)> 24937 Date: Sat Aug 25 21:03:08 2007 -0400 24938 24939 RADEON: fix crtc clipping for Xv 24940 24941 commit 3fd2d22a02812d5f86cdc1c9503f48362b0c362b 24942 Author: Alex Deucher <alex@botch2.(none)> 24943 Date: Sat Aug 25 17:37:35 2007 -0400 24944 24945 RADEON: remove fbdev option 24946 24947 FBDev support is currently broken, and it not really compatible 24948 with randr 24949 24950 commit 3469e1aa08792890fa6a5c72da52a1992a0b382c 24951 Author: Alex Deucher <alex@botch2.(none)> 24952 Date: Fri Aug 24 20:42:13 2007 -0400 24953 24954 RADEON: add extra green data in depth 16 24955 24956 Apparently some radeons need this? 24957 24958 commit 71f650d1bc432514516f7ac64a5e8a54c5227881 24959 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24960 Date: Fri Aug 24 09:21:39 2007 +0200 24961 24962 Require xorg-server >= 1.3 for RandR 1.2. 24963 24964 commit d7230939f523610c57f92bdfc72966bdbc6f1070 24965 Author: Michel Dnzer <michel (a] tungstengraphics.com> 24966 Date: Fri Aug 24 09:21:14 2007 +0200 24967 24968 64 bit warning fixes. 24969 24970 For printf vs. CARD32, use %u or %x and and a cast to unsigned. 24971 24972 commit 91c45fedfd155a153dcd2c3f3e30986bfbd44e6f 24973 Author: Dave Airlie <airlied@nx6125b.(none)> 24974 Date: Fri Aug 24 15:05:01 2007 +1000 24975 24976 radeon: don't disable dac if either tv or vga is using it 24977 24978 On my rs480 I had to vt switch to get hotplug VGA working due to the tv-out 24979 code turning off the dac when the vga code was actually using it. 24980 24981 commit 056ca6bb5adf974290693b55de6cd6880d2132d1 24982 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24983 Date: Thu Aug 23 20:19:47 2007 -0400 24984 24985 RADEON: NONE to None to match other port info 24986 24987 commit 13fd53286bdda2c55683bdb5f63e7d345f6c63ef 24988 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24989 Date: Thu Aug 23 20:17:51 2007 -0400 24990 24991 RADEON: set (hopefully) reasonable default max desktop sizes 24992 24993 Based on the amount of vram. We really need ttm... 24994 24995 commit 53bad86ca48a9b6529c1f0989ee568d9d48841c6 24996 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 24997 Date: Thu Aug 23 19:24:59 2007 -0400 24998 24999 RADEON: clean up the logic in crtc_mode_set() 25000 25001 commit 4712dedea225e9e07177aebda2ffc6290d1f53c7 25002 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25003 Date: Thu Aug 23 18:51:34 2007 -0400 25004 25005 RADEON: Always assume LVDS is connected 25006 25007 Not all bioses seem to set the right scratch bits. If we have 25008 LVDS (via bios table or otherwise) assume it's connected. 25009 25010 commit b0f170c5f736ecba1a5899d602c4173fe9b9b1fa 25011 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25012 Date: Thu Aug 23 12:55:40 2007 +0200 25013 25014 radeon: Remove unnecessary #include <time.h>. 25015 25016 commit ac54c0e4360099697755d14b1030def73d8235b0 25017 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25018 Date: Wed Aug 22 14:33:59 2007 +0200 25019 25020 radeon: Warning fixes. 25021 25022 commit de26e406f52b3b13f03eee2b8023924ec6406f0a 25023 Author: Alon Ziv <alonz (a] nolaviz.org> 25024 Date: Mon Jul 30 22:47:59 2007 +0300 25025 25026 radeon: Sane handling of timeouts in WaitForVerticalSync(2). 25027 25028 RADEONWaitForVerticalSync() and RADEONWaitForVerticalSync2() need to wait 25029 for a timeout specified in milliseconds; looping around usleep() causes 25030 the timeout to be unnecessarily long, as the OS may sleep longer than 25031 requested (on Linux the minimum actual sleep value may be several ms). 25032 25033 The new logic uses gettimeofday() in the loop to see when the (absolute) 25034 timeout has arrived. 25035 25036 Signed-off-by: Alon Ziv <alonz (a] nolaviz.org> 25037 25038 commit c66e5de26ae93caa368213f3cce139aacec955d2 25039 Author: Sascha Sommer <saschasommer (a] freenet.de> 25040 Date: Thu Aug 23 12:11:51 2007 +0200 25041 25042 radeon: Round down RMX stretch ratios. 25043 25044 Fixes issues with RMX scaling, see 25045 https://bugs.freedesktop.org/show_bug.cgi?id=8983 . 25046 25047 commit 633c1fff10a3be4c9f48c1995e330d60bf6abbb2 25048 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25049 Date: Thu Aug 23 12:11:41 2007 +0200 25050 25051 radeon: Sync pages when enabling page flipping with EXA as well. 25052 25053 Exclude the DRI window(s) though to avoid scribbling over 3D rendering. 25054 25055 commit 4f8010ce22043c0f8d60c0f49d270ce98c9d2466 25056 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25057 Date: Thu Aug 23 12:11:12 2007 +0200 25058 25059 radeon: Don't synchronize DRI windows between pages when possible. 25060 25061 commit 7b527054a7c81d1d1dbc79d41b9e53064dab68cb 25062 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25063 Date: Thu Aug 23 12:10:33 2007 +0200 25064 25065 radeon: Restore memmap registers even if only AGP location changed. 25066 25067 commit d7ba9f001c0ab645984526afd0e64d1c6a6d654a 25068 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25069 Date: Thu Aug 23 11:39:54 2007 +0200 25070 25071 radeon: Change a test to info->IsIGP from several IGP families. 25072 25073 commit 8c7c22e22b6076abc80e4e1aaa8d1f4cf2f3ed14 25074 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25075 Date: Thu Aug 23 11:38:17 2007 +0200 25076 25077 radeon: Wait for pending overlay flip to finish before emitting new one. 25078 25079 commit 92fa7cc00688d7bfc1fb72e645ac30c6d92669c6 25080 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25081 Date: Thu Aug 23 11:38:16 2007 +0200 25082 25083 radeon: Don't call RADEONDRIRefreshArea when the damaged region is empty. 25084 25085 commit 5cb20c2dc5eca9d7d7d78e9924ea1b90076e7253 25086 Author: Dave Airlie <airlied (a] linux.ie> 25087 Date: Thu Aug 23 18:46:41 2007 +1000 25088 25089 radeon: bug 11899 + debian 435040 - test for usefbdev before calling int10 25090 (cherry picked from commit 3334c247540b113f9c4a5ce1879d8648a45a959a) 25091 25092 commit 9d6261c024c697a211da4b6300420153eb72a264 25093 Author: Roland Br <roland (a] verifysoft.de> 25094 Date: Thu Aug 23 18:33:34 2007 +1000 25095 25096 radeon: bug 11860 fix some mem leaks 25097 25098 commit 7bc1f862bc5f992f213143fbafef52459ba7db4a 25099 Author: Roland Br <roland (a] verifysoft.de> 25100 Date: Thu Aug 23 18:37:35 2007 +1000 25101 25102 radeon: bug 11861 - dead code removal in radeon_video.c 25103 (cherry picked from commit 8e3a6f83016cd8c4cfd43ceee4cbf0a8dc018b2a) 25104 25105 commit 9d38c8aa1a7d6fb1af41ee8abdb4a95f94843538 25106 Author: Dave Airlie <airlied (a] clockmaker.usersys.redhat.com> 25107 Date: Thu Aug 23 20:10:24 2007 +1000 25108 25109 radeon: cleanup some warnings 25110 25111 commit e4c8969b48a6c8dcc4e7f9852479d24a0204fc0d 25112 Author: Dave Airlie <airlied (a] redhat.com> 25113 Date: Thu Aug 23 19:56:21 2007 +1000 25114 25115 updated release numbering for randr 1.2 25116 25117 commit 81a8093f2ffdce59d4e8a44f65f3d5c771fd6425 25118 Merge: c08e6ec9e 53a67e319 25119 Author: Dave Airlie <airlied (a] redhat.com> 25120 Date: Thu Aug 23 19:51:47 2007 +1000 25121 25122 Merge remote branch 'origin/randr-1.2' 25123 25124 commit c08e6ec9e7ac06caee53689b0ec50ef6a7a0ba37 25125 Merge: 5793e8753 a0d0fcd3b 25126 Author: Dave Airlie <airlied (a] clockmaker.usersys.redhat.com> 25127 Date: Thu Aug 23 19:42:19 2007 +1000 25128 25129 Merge branch 'master' into randr-merge 25130 25131 Conflicts: 25132 25133 src/radeon.h 25134 src/radeon_bios.c 25135 src/radeon_display.c 25136 src/radeon_dri.c 25137 src/radeon_driver.c 25138 src/radeon_modes.c 25139 src/radeon_probe.h 25140 src/radeon_video.c 25141 25142 commit 53a67e31904bec9a3aa1bd24de8034dcafea1d2a 25143 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25144 Date: Tue Aug 21 21:20:41 2007 -0400 25145 25146 RADEON: Fix color problem on pre-R3xx chips tv-out 25147 25148 commit e9719e8e02eef46717ae9b4d8c7998466dac30cb 25149 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25150 Date: Tue Aug 21 21:17:20 2007 -0400 25151 25152 RADEON: more tv out fixes and clean up 25153 25154 commit 36c22a49580d86a6518b67f31a78bd53d39491af 25155 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25156 Date: Tue Aug 21 20:28:39 2007 -0400 25157 25158 RADEON: fix tv-out on R3xx 25159 25160 R3xx apparently needs the tv clock forced on. 25161 25162 commit 9470bd67731059f26859ed5f0bea3ade09e2c80c 25163 Author: Alex Deucher <alex (a] botch2.com> 25164 Date: Mon Aug 20 20:54:06 2007 -0400 25165 25166 RADEON: Add DefaultConnectorTable option 25167 25168 This option skips the parsing the BIOS connector table 25169 and falls back to chip specific defaults. 25170 Also remove man page section for the now gone bioshotkeys 25171 option. 25172 25173 commit a0d0fcd3bab765b4db25e04884fd8a342abb9c66 25174 Author: iLisa Wu <liswu (a] ati.com> 25175 Date: Mon Aug 20 15:44:07 2007 +0200 25176 25177 Fix crash with no valid mode in xorg.conf's modelist and empty Virtual 25178 25179 If the resolution defined in xorg.conf failed to find a matching mode in the 25180 supported modelist, and no virtual desktop dimensions are defined the xorg.conf 25181 either, virtual X and Y dimension will be set to 0 which will cause Xserver 25182 crash. 25183 25184 (Novell bugzilla #296856, closed) 25185 25186 commit b275febdb0918e8cebdffbb433b0eeb3ff4d3746 25187 Author: Alex Deucher <alex@samba.(none)> 25188 Date: Sun Aug 19 20:55:32 2007 -0400 25189 25190 RADEON: turn off TVCLK when blanking tv encoder 25191 25192 commit a90d675832ddb02c81ace010ccbf02619b70edac 25193 Author: Alex Deucher <alex (a] botch2.com> 25194 Date: Thu Aug 16 21:55:14 2007 -0400 25195 25196 RADEON: fix Xv clipping and overlay sourcing 25197 25198 - Basically just copied from the intel driver. I'm planning to push 25199 this to the server soon, but add it now to get things working 25200 and to provide compat for older servers. 25201 25202 - Overlay crtc source control attribute now called XV_CRTC 25203 The old attribute XV_SWITCHCRT has been removed. If anyone cares, 25204 we can add it back as an alias to XV_CRTC 25205 XV_CRTC: -1 auto, 0 crtc0, 1 crtc1 25206 25207 commit 5793e8753d11432bf95c7c6dd80c811e16aba058 25208 Author: Alex Deucher <alex (a] botch2.com> 25209 Date: Wed Aug 15 19:26:36 2007 -0400 25210 25211 RADEON: Remove RADEONRestoreMode() 25212 25213 Since we no longer use it to write modes all it did 25214 was restore some of the regs for a console restore. 25215 Just move the relevant bits into RADEONRestore() and 25216 remove it. 25217 25218 commit 6f011aaabaf18d66ffc255ad76aaf938b2396302 25219 Author: Lisa Wu <Lisa.Wu (a] amd.com> 25220 Date: Wed Aug 15 19:17:51 2007 -0400 25221 25222 RADEON: fix console restore on r3xx and r4xx 25223 25224 When restoring the console we need to: 25225 1) we need to restore DAC registers after all other registers are 25226 restored and CRTCs are enabled. 25227 2) we need to enable CRTC2 registers before CRTC1 registers 25228 25229 commit c8dad98abb042c6abbbee18f9ae4db72084bc513 25230 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25231 Date: Wed Aug 15 00:55:24 2007 -0400 25232 25233 RADEON: make sure crtc routing is correct in r300_detect_tv() 25234 25235 Also remove some unused variable 25236 25237 commit f8cd74435f0072dbf5f6e83d67d2d5e1f4e82c91 25238 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25239 Date: Tue Aug 14 23:50:57 2007 -0400 25240 25241 RADEON: minor cleanup 25242 25243 commit 6f398cd07ea734dd66a8eac71b629e59123d75b8 25244 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25245 Date: Tue Aug 14 23:42:32 2007 -0400 25246 25247 RADEON: Implement improved tv load detection for r300 25248 25249 The previous implementation resulted in false positives 25250 on occasion. This method works much more reliably. 25251 Based on beos code by Thomas Kurschel 25252 25253 commit 366a1d4c240ac93622caff97b652696db99bf2e6 25254 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25255 Date: Tue Aug 14 23:25:37 2007 -0400 25256 25257 RADEON: Turn off tv encoder in disableoutputs() 25258 25259 commit 5b4a04c23e3f1ec2490418b111f417c16463c709 25260 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25261 Date: Tue Aug 14 23:11:52 2007 -0400 25262 25263 RADEON: Make sure RMX is always programmed for crtc1 25264 25265 If the RMX registers are left as programmed by the bios 25266 this can lead to a blank screen when crtc1 is feeding a 25267 DAC. Fix found and reported by Lisa Wu <liswu (a] ati.com> 25268 Fixes bug 11985 25269 25270 commit db2a828b2f21b92cd654b309d137204334975b89 25271 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25272 Date: Tue Aug 14 22:47:29 2007 -0400 25273 25274 RADEON: Add load detection 25275 25276 Based on the beos driver by Thomas Kurschel and the 25277 existing load detection code in this driver. 25278 25279 commit 6fc3ddbbb6fbbee1f6076c776e5b46c0c772b6d4 25280 Author: Alex Deucher <alex (a] botch2.com> 25281 Date: Sun Aug 12 12:00:20 2007 -0400 25282 25283 RADEON: order the VGA and DVI ports correctly 25284 25285 commit b7738d2ef82e1759adf78e5db1291f8739b4166f 25286 Author: Alex Deucher <alex (a] botch2.com> 25287 Date: Sat Aug 11 17:50:42 2007 -0400 25288 25289 RADEON: fix connector setup when there's no bios tables 25290 25291 commit d0895f67e327bb268fd59fcfd8fc22678d804f57 25292 Author: Luc Verhaegen <libv (a] skynet.be> 25293 Date: Thu Aug 9 12:11:40 2007 +0200 25294 25295 AGPFastWrite risk reduction. 25296 25297 Actively warn the user in the log about the effects of AGPFastWrite and 25298 sanitise AGPFastWrite handling while we're here. 25299 25300 commit c01000bc684f6a23a38a52f0808182cefaa686bb 25301 Author: Alex Deucher <alex (a] botch2.com> 25302 Date: Wed Aug 8 00:07:16 2007 -0400 25303 25304 RADEON: fix typo in comment 25305 25306 commit 5c3f49e651c36f3bd14fa29359e24825d8f7f77f 25307 Author: Lisa Wu <liswu (a] ati.com> 25308 Date: Wed Aug 8 00:05:47 2007 -0400 25309 25310 RADEON: make sure RADEON_CRTC2_CRT2_ON bit state is properly accounted for 25311 25312 - when the TV DAC is used for crtc1 make sure to keep this bit set when 25313 initializing crtc2. 25314 - fixes bug 11894 25315 25316 commit 08fe7ad00fba523775e95b5e6295fe23a5119d60 25317 Author: Alex Deucher <alex (a] botch2.com> 25318 Date: Tue Aug 7 23:16:05 2007 -0400 25319 25320 RADEON: fix ConnectorTable option after connector table re-work 25321 25322 We may want to add the possiblity of more connectors to this option 25323 25324 commit f56b90a60393ed187c0e39e149b2a3a32331ea52 25325 Author: Alex Deucher <alex (a] botch2.com> 25326 Date: Tue Aug 7 23:08:02 2007 -0400 25327 25328 RADEON: Make sure the default TV standard is supported 25329 25330 commit a88a0e77eec514325b4d07bf7be9fb04e6f8e244 25331 Author: Alex Deucher <alex (a] botch2.com> 25332 Date: Tue Aug 7 00:43:20 2007 -0400 25333 25334 RADEON: minor fix to legacy bios connector table 25335 25336 commit 3752808d6c08a9727370ef8d79088e787791e131 25337 Author: Alex Deucher <alex (a] botch2.com> 25338 Date: Sun Aug 5 16:13:39 2007 -0400 25339 25340 RADEON: Always set MonType in radeon_detect() 25341 25342 - Always set the the MonType in radeon detect even if no 25343 monitor is detected so users can still force disconnected 25344 outputs on. For DVI-I users will have to set the 25345 dvi_monitor_type attribute to force digital or analog, 25346 everything else should just work. 25347 25348 commit b4ec3e436afb5bc99ec755cbd96eee9a1ec492a5 25349 Author: Alex Deucher <alex (a] botch2.com> 25350 Date: Sun Aug 5 15:37:04 2007 -0400 25351 25352 RADEON: add tv out properties 25353 25354 commit 1fc2a1120e7c05938e2bd72d3c7837ecff8bc9da 25355 Author: Alex Deucher <alex (a] botch2.com> 25356 Date: Sun Aug 5 14:41:57 2007 -0400 25357 25358 RADEON: switch DVI-I monitor type attribute to string 25359 25360 commit 7d8eb3751d74bd8a1fb9fa2d2fcb9c4c895f6ba4 25361 Author: Alex Deucher <alex (a] botch2.com> 25362 Date: Sun Aug 5 14:31:18 2007 -0400 25363 25364 RADEON: make backlight attribute lower case 25365 25366 commit e71bb88bc9052af8866fb6945dbc06dbb0e6d1c0 25367 Author: Alex Deucher <alex (a] botch2.com> 25368 Date: Sun Aug 5 14:26:15 2007 -0400 25369 25370 RADEON: switch RMX attribute to string type 25371 25372 Still not actually hooked up. 25373 25374 commit 284323135ec04635dfa9cabd5790a35b953abca2 25375 Author: Alex Deucher <alex (a] botch2.com> 25376 Date: Sun Aug 5 04:35:24 2007 -0400 25377 25378 RADEON: typo 25379 25380 commit d1abdad167aa24ac970c69422435df443c82ebd6 25381 Author: Alex Deucher <alex (a] botch2.com> 25382 Date: Sun Aug 5 03:45:02 2007 -0400 25383 25384 RADEON: fixes 25385 25386 - fix output ordering 25387 - set tv-out to return un-connected for now in radeon_detect() 25388 25389 commit 0cca25d8d6a0cb0d29b68e6cd9c699d1390aede1 25390 Author: Alex Deucher <alex (a] botch2.com> 25391 Date: Sun Aug 5 03:19:24 2007 -0400 25392 25393 RADEON: Fix DVI-I support in ATOM bios connector table parsing 25394 25395 commit ba5496ae7973786802962bf649dd91c219531749 25396 Author: Alex Deucher <alex (a] botch2.com> 25397 Date: Sun Aug 5 02:27:32 2007 -0400 25398 25399 RADEON: refactor output init to handle multiple DVI or VGA 25400 25401 - refactor output init to handle multiple DVI or VGA with the new 25402 bios table parsing 25403 25404 commit 2ec22783ddf4c522df9e5fd1b2003854486d7a2b 25405 Author: Alex Deucher <alex (a] botch2.com> 25406 Date: Sun Aug 5 01:39:35 2007 -0400 25407 25408 RADEON: attempt to do the right thing for standards other than PAL or NTSC 25409 25410 commit cc8e1d95f1b90a259beea4e8cc4d7e29af660919 25411 Author: Alex Deucher <alex (a] botch2.com> 25412 Date: Sun Aug 5 01:14:36 2007 -0400 25413 25414 RADEON: Major rework of BIOS table parsing 25415 25416 - greatly simplify ATOM and legacy connector table parsing 25417 - use bios tables to detect LVDS and TV outputs 25418 - add support for TV table parsing (legacy only) 25419 25420 commit 1de52d91ff3a04b9b587b858e1e5be40d3a7fd0a 25421 Author: Dave Airlie <airlied (a] linux.ie> 25422 Date: Sat Aug 4 17:58:58 2007 +1000 25423 25424 update configure.ac for 6.6.193 release 25425 25426 commit 371001c0433db1d17e468f3ea99ea57f922145a3 25427 Author: Dave Airlie <airlied (a] linux.ie> 25428 Date: Sat Aug 4 17:51:19 2007 +1000 25429 25430 radeon: remove unused variables 25431 25432 commit 90946c53d7a4b23d03270ad6da0450759a11de3d 25433 Author: Dave Airlie <airlied (a] linux.ie> 25434 Date: Sat Aug 4 17:51:06 2007 +1000 25435 25436 ati: add -Wall for gcc 25437 25438 As per luc's "suggestion" in Novell bug. 25439 25440 This may generate warnings on 64-bit until we get rid of CARD32 25441 25442 commit e30a145934df8f6a7f71290d6c75e4239f9d52f7 25443 Author: Dave Airlie <airlied (a] linux.ie> 25444 Date: Sat Aug 4 17:44:46 2007 +1000 25445 25446 radeon: cleanup some pieces of the dpms/blank register programming 25447 25448 commit a156db5e8b037ed12a448f70045453baf9d0c504 25449 Author: Luc Verhaegen <libv (a] skynet.be> 25450 Date: Sat Aug 4 17:37:18 2007 +1000 25451 25452 Clean up PortInfo to CRTC mapping 25453 25454 Also sanitise blanking and DPMS functions 25455 25456 Fixes from Novell Bug 264720, and fd.o 10772 25457 25458 commit b61a49f2a5401560f85e11bcdd005287433cad12 25459 Author: Alex Deucher <alex (a] botch2.com> 25460 Date: Fri Aug 3 19:27:59 2007 -0400 25461 25462 RADEON: Change indexing of TV constants table in preparation for standard re-work 25463 25464 commit 288fa627274cb399059262d4f8bd844fc220a042 25465 Author: Alex Deucher <alex (a] botch2.com> 25466 Date: Thu Aug 2 02:37:16 2007 -0400 25467 25468 RADEON: avoid a divide by 0 and only save tv out regs if the chip has them 25469 25470 commit b66a1bc7994b33d349c1519761e431959311c85f 25471 Author: Alex Deucher <alex (a] botch2.com> 25472 Date: Thu Aug 2 02:11:20 2007 -0400 25473 25474 RADEON: fix tv-out enable/disable 25475 25476 commit d86592c8d5ce45d81d8a726c263e870e94fbcf11 25477 Author: Alex Deucher <alex (a] botch2.com> 25478 Date: Thu Aug 2 00:50:51 2007 -0400 25479 25480 RADEON: add missing break 25481 25482 commit 971feb34843225030fff05b3f9d3801534fbf2d4 25483 Author: Alex Deucher <alex (a] botch2.com> 25484 Date: Thu Aug 2 00:50:04 2007 -0400 25485 25486 RADEON: move tv dac enable to enabledisplay() 25487 25488 commit 98d7e00437bea78e03180eb30ff30de3455d9d1a 25489 Author: Alex Deucher <alex (a] botch2.com> 25490 Date: Thu Aug 2 00:20:50 2007 -0400 25491 25492 RADEON: limit tv modes to the only one we can program at the moment 25493 25494 commit 5c549c1d42f7bbc556942af13aff2661fae856f2 25495 Author: Alex Deucher <alex (a] botch2.com> 25496 Date: Wed Aug 1 23:45:07 2007 -0400 25497 25498 RADEON: remove unused elements 25499 25500 commit b03978028fd975eb6946503d3a56a49c5a67f339 25501 Author: Alex Deucher <alex (a] botch2.com> 25502 Date: Wed Aug 1 22:39:16 2007 -0400 25503 25504 RADEON: convert hard coded tv out values to calculations 25505 25506 commit cf54222f1fa37366b2c2b39c82f8afc02f32e63c 25507 Author: Alex Deucher <alex (a] botch2.com> 25508 Date: Tue Jul 31 02:01:49 2007 -0400 25509 25510 RADEON: more fixes... 25511 25512 commit 22d460d3ad991223aa1fbd7e5edeb45e36c65dc0 25513 Author: Alex Deucher <alex (a] botch2.com> 25514 Date: Tue Jul 31 01:28:05 2007 -0400 25515 25516 RADEON: fix name of tv output 25517 25518 commit 4822a2b837334f408f962646ab5ea4f8b0335ac9 25519 Author: Alex Deucher <alex (a] botch2.com> 25520 Date: Tue Jul 31 01:18:40 2007 -0400 25521 25522 RADEON: tv-out fixes. works now. tested on rv350. 25523 25524 VT siwtch is busted, and xrandr doesn't play nice yet. 25525 uncomment code in radeon_output.c to test. 25526 25527 commit fe494c9db2995bb8ce7a028ecf9626e0cb0cf506 25528 Author: Alex Deucher <alex (a] botch2.com> 25529 Date: Sun Jul 29 15:26:34 2007 -0400 25530 25531 RADEON: add info about tv out code and authorship 25532 25533 commit 6b9b7a7bdc290d07de9b226691ec8025af8db896 25534 Author: Alex Deucher <alex (a] botch2.com> 25535 Date: Sun Jul 29 15:23:14 2007 -0400 25536 25537 RADEON: Initial pass at integrated tv out support 25538 25539 Based on the GATOS tv-out support by Federico Ulivi <fulivi (a] lycos.com> 25540 and information from ati with substantial rework by myself. 25541 25542 Code is not actually hooked up yet. 25543 25544 commit 8d043db1817d94edeb72ab208dfea60026715d48 25545 Author: Alex Deucher <alex (a] botch2.com> 25546 Date: Wed Jul 25 20:37:58 2007 -0400 25547 25548 RADEON: Compute PLL VCO gain 25549 25550 commit 62f06d89da3f7160d5e4df8d7ce6fe1a94e9d07c 25551 Author: Alex Deucher <alex (a] botch2.com> 25552 Date: Wed Jul 25 20:22:25 2007 -0400 25553 25554 RADEON: write out saved vclk and pixclk values 25555 25556 commit 165a07cbbfcd94e3d1fac434b8fada8d29428a09 25557 Author: Brice Goglin <Brice.Goglin (a] ens-lyon.org> 25558 Date: Wed Jul 25 19:01:02 2007 +0200 25559 25560 Minor fixes in the manpages 25561 25562 - ati.man 25563 . add references to radeon and r128 manpages in SEE ALSO 25564 (Xavier Bestel in Debian bug #386001). 25565 . remove the manpage suffix from atimisc reference in the text 25566 since there is no such manpage for now. 25567 - radeon.man 25568 . typo in "specifying" (reported by A. Costa in Debian bug #432059). 25569 . replace a non-ascii character with the corresponding groff escape 25570 sequence so that it works whatever the locale (Julien Cristau). 25571 25572 commit 9cc3ab8320162f371bba15dc131f23c5de2013fc 25573 Author: Alex Deucher <alex (a] botch2.com> 25574 Date: Wed Jul 25 02:24:14 2007 -0400 25575 25576 RADEON: hack around console restore hang on r4xx 25577 25578 r4xx seems to hang when unblanking the crtc(s) when 25579 restoring the console. This needs to be investigated 25580 further. 25581 25582 commit 193d58b0ea9de3e326017e50b88e2f3726854506 25583 Author: Alex Deucher <alex (a] botch2.com> 25584 Date: Wed Jul 25 01:27:58 2007 -0400 25585 25586 RADEON: fix randr on r4xx 25587 25588 Seems the way we parse the ATOM connector table results 25589 in reversed connectors. 25590 25591 commit 262e32bd9ea4123b116362b33b5798753b2568fe 25592 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25593 Date: Thu Jul 19 23:48:57 2007 -0400 25594 25595 RADEON: fix palette when depth != 24 25596 25597 commit 18b00b47a483e7854727e99126808ddf361e7a4a 25598 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25599 Date: Thu Jul 19 19:01:42 2007 -0400 25600 25601 R128: don't clip modes to panel on laptops when only using the CRT port 25602 25603 fixes bug 5832 25604 25605 commit 882fe7631586b0a7919f808588a2ea4fb555f7e8 25606 Author: Brice Goglin <brice.goglin (a] ens-lyon.org> 25607 Date: Thu Jul 19 18:44:27 2007 -0400 25608 25609 MACH64: avoid crash in Xv code 25610 25611 fixes bug 11054 25612 25613 commit 7b3e22e96cfc98621bd20fa76317e6d8f7242165 25614 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25615 Date: Thu Jul 5 20:37:10 2007 -0400 25616 25617 RADEON: set default values for crtc_offset_cntls 25618 25619 commit 2a35ccfb8ffee9f0ef6f8d4f0eeb80a471543caa 25620 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25621 Date: Thu Jul 5 20:04:04 2007 -0400 25622 25623 RADEON: Fix tiling on r1xx and r2xx 25624 25625 spotted by Andrew Randrianasulu <randrik (a] mail.ru> 25626 fixes bug 11357 25627 25628 commit ac9cbaf32176cf144bc694ac879e7c9e2920f762 25629 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25630 Date: Mon Jul 2 10:06:02 2007 +0200 25631 25632 radeon: Fully zero-initialize info->CRT2pScrn->monitor. 25633 25634 This prevents modes from getting rejected due to the uninitialized maxPixClock 25635 value. 25636 25637 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=9816 . 25638 25639 commit bdcae622100c81a4d9a53938542b64908bacd195 25640 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25641 Date: Fri Jun 29 01:15:57 2007 -0400 25642 25643 RADEON: save crtc on/off state for console restore 25644 25645 - prevents a possible hang if console is only using 1 crtc 25646 25647 commit 21be0d3a1f4700572c6425800596785d6850626f 25648 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25649 Date: Fri Jun 29 00:38:50 2007 -0400 25650 25651 RADEON: simplify console restore 25652 25653 - still need to track crtcs for blank/unblank when restoring text console 25654 25655 commit d9bf28b55d98fea2d285c9c46362aaf2175d0a46 25656 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25657 Date: Thu Jun 28 23:52:28 2007 -0400 25658 25659 RADEON: factor out surface_cntl init into one function 25660 25661 commit 9f193985627be8e6ea1418a424e825ddbc4957b2 25662 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25663 Date: Thu Jun 28 23:43:13 2007 -0400 25664 25665 RADEON: move crtc base setups to new functions 25666 25667 commit 0f361e9e80a29d287fa42436c32c657e3c102539 25668 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25669 Date: Thu Jun 28 23:08:07 2007 -0400 25670 25671 RADEON: fix corruption after 3D apps run 25672 25673 commit 1d4630067d293d26284d5fe230debef3913ec6cf 25674 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25675 Date: Thu Jun 28 22:24:04 2007 -0400 25676 25677 RADEON: set info->CurrentLayout.displayWidth in RADEONSetPitch() 25678 25679 commit 2754d1ba01fc3367019487e0c0f59d74c950aaa7 25680 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25681 Date: Thu Jun 28 22:15:22 2007 -0400 25682 25683 RADEON: Switch cursor back to memcpy() 25684 25685 commit 7901bcafa92dccd319ddb5de4627d806a39f15f9 25686 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25687 Date: Mon Jun 25 18:39:14 2007 -0400 25688 25689 RADEON: make sure we unblank in RADEONRestore() 25690 25691 reported by Andrew Randrianasulu (bugs 11357, 11146) 25692 25693 commit 6a3599d6155d073a3116c4b13bdf04b44bb9e087 25694 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25695 Date: Mon Jun 25 18:20:54 2007 -0400 25696 25697 RADEON: make sure we set the stride properly 25698 25699 commit 771c37fcac8104b894120cc3fc0154b31b0318d1 25700 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25701 Date: Mon Jun 25 17:53:26 2007 -0400 25702 25703 RADEON: attempt to fix cursor on big endian (only tested on x86) 25704 25705 commit 5b5b90c2cea7e36895354f5872acd3fc769653f9 25706 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25707 Date: Mon Jun 25 16:27:54 2007 -0400 25708 25709 RADEON: fix possible segfault on mobility chips with MM tables 25710 25711 spotted by Stefan Buehler 25712 25713 commit 01bd5eb713500d3e3d4351865d460c8d1e476454 25714 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25715 Date: Mon Jun 25 16:16:18 2007 -0400 25716 25717 RADEON: fix possible segfault on mobility chips with MM tables 25718 25719 reported by Stefan Buehler 25720 25721 commit 52cc1dc1491559a9055f3ba6dd54064bb382ad86 25722 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25723 Date: Mon Jun 25 15:49:19 2007 -0400 25724 25725 RADEON: tiling and FB size fixes 25726 25727 - fix tiling on older radeons 25728 - allow the user to specify FB size using Virtual line in config 25729 Current default is 1600x1200 25730 - reset ecp_div for Xv in randr crtc mode set 25731 25732 commit 3bb46c03fddd3bd79bdab887366aeeced0290a3a 25733 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25734 Date: Fri Jun 22 01:12:38 2007 -0400 25735 25736 RADEON: bump crtc max to 1200 so 1600x1200 will work out of the box 25737 25738 - a better fix will be forthcoming... 25739 25740 commit b9fac5695bf8f123b391e3b162c4865258a685dd 25741 Author: Andrew Randrianasulu <randrik (a] mail.ru> 25742 Date: Fri Jun 22 01:01:14 2007 -0400 25743 25744 RADEON: Fix VT switch hangs 25745 25746 - re-order DRI resume functions in EnterVT() 25747 - fixes bug 11287 25748 25749 commit 92e65d5e0d6817ff4c9a08020a0a9b3a8c3c98b0 25750 Author: Andrew Randrianasulu <randrik (a] mail.ru> 25751 Date: Fri Jun 22 00:55:00 2007 -0400 25752 25753 RADEON: Fix VT switch hangs 25754 25755 - reorder RADEONDRISetVBlankInterrupt() and RADEONDRIResume() 25756 - see bug 11287 25757 25758 commit ec1fd2c388ec2c4033644776ee8588405a3d25f1 25759 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25760 Date: Tue Jun 19 00:39:59 2007 -0400 25761 25762 RADEON: Split FP reg restore to only touch regs relevant to the output 25763 25764 - only restore output specific regs 25765 - restore fp2_gen_cntl on R200 DAC restore as R200 DAC2 is connected via 25766 DVO 25767 25768 commit 0f5a92667281ca3f80218bc67d4031125a5e4793 25769 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25770 Date: Mon Jun 18 23:35:21 2007 -0400 25771 25772 RADEON: only touch LVDS on mobility chips 25773 25774 commit b72ff160f908bf3aa9f64705377e92d80360a4f7 25775 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25776 Date: Mon Jun 18 21:51:45 2007 -0400 25777 25778 RADEON: only touch LVDS on mobility chips 25779 25780 commit 80313621ffa8c409ae63fc1b28c15fc4abdc3a7b 25781 Author: Henry Zhao <henryz (a] localhost.localdomain> 25782 Date: Fri Jun 15 17:00:05 2007 -0700 25783 25784 Update CRT2pScrn->monitor->Last to reflect the 25785 last mode of CRT2pScrn. See bug 11278. 25786 25787 commit 5868af53204e8fc210735e68e891e15085285629 25788 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25789 Date: Tue Jun 12 23:53:48 2007 -0400 25790 25791 RADEON: yet another LVDS... 25792 25793 commit f19a6f7ee5bf4ec632e7813359f167599c08e823 25794 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25795 Date: Tue Jun 12 23:44:41 2007 -0400 25796 25797 RADEON: additional LVDS off fix (missed in previous commit) 25798 25799 commit 4f2a1ba9c2fc20042d67132f986a86a9783a245e 25800 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25801 Date: Tue Jun 12 23:41:24 2007 -0400 25802 25803 RADEON: additional LVDS off fix (missed in previous commit) 25804 25805 commit 9fc416ce08cc5d74faa7b184e80ce43ed2f4fbee 25806 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25807 Date: Tue Jun 12 23:35:09 2007 -0400 25808 25809 radeon: Fix panel size detection from registers with stretched mode programmed. 25810 25811 commit c4045cbc5e6975520efef422ce9df2a7d99a7ff4 25812 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25813 Date: Tue Jun 12 23:32:18 2007 -0400 25814 25815 RADEON: make sure to clear RADEON_LVDS_ON when turning LVDS off 25816 25817 - fixes bug 3483 25818 25819 commit 090888d3baf61a70640d28e99afdce20952cdea0 25820 Author: Tormod Volden <bugzi06.fdo.tormod (a] xoxy.net> 25821 Date: Tue Jun 12 23:30:18 2007 -0400 25822 25823 RADEON: limit PanelPwrDly to 2000 ms on ATOM bios as per legacy bios 25824 25825 - fixes bug 11238 25826 25827 commit 44748a826b52924a6e050215d8d62755e0ac7fb2 25828 Author: Tormod Volden <bugzi06.fdo.tormod (a] xoxy.net> 25829 Date: Tue Jun 12 23:23:49 2007 -0400 25830 25831 RADEON: limit PanelPwrDly to 2000 ms on ATOM bios as per legacy bios 25832 25833 - fixes bug 11238 25834 25835 commit c292fc64499ff4cc135c07deda99cf4169f8fef4 25836 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25837 Date: Tue Jun 12 23:20:18 2007 -0400 25838 25839 RADEON: make sure RADEON_LVDS_ON is cleared when turning LVDS off 25840 25841 - fixes bug 3483 25842 25843 commit c6a3286d6f6ddda89115d98d45665dadd78bf41d 25844 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25845 Date: Mon Jun 11 09:39:38 2007 +0200 25846 25847 radeon: Fix panel size detection from registers with stretched mode programmed. 25848 25849 commit 7a6b3c0f77c69019268f585c51c3dcbcc99014d4 25850 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25851 Date: Mon Jun 11 09:39:38 2007 +0200 25852 25853 radeon: Don't loop indefinitely if no mode matches detected panel size. 25854 25855 commit f31fd9ce598841c505a0b5ed32bf124f49fea332 25856 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25857 Date: Mon Jun 11 09:39:38 2007 +0200 25858 25859 radeon: Fix some more ErrorFs when setting mode. 25860 25861 commit 9dd9f09a9021653e5590d27bd7aa0e7af85416ad 25862 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25863 Date: Thu Jun 7 15:32:31 2007 -0400 25864 25865 RADEON: fix some issues with bios table init code 25866 25867 - fix some issues with bios table init code 25868 - re-org preinit to better handle bios table init of cards 25869 25870 commit 9ad311eeb0ac2b70a862dbb9de278154ba9142ec 25871 Author: Tilman Sauerbeck <tilman (a] code-monkey.de> 25872 Date: Thu Jun 7 11:03:02 2007 +0200 25873 25874 Fixed 'make dist'. 25875 25876 commit 2527f2b69aa7dffa3ba4359c45955e3185bdf0e6 25877 Author: Alex Deucher <alex (a] botch2.com> 25878 Date: Wed Jun 6 00:08:45 2007 -0400 25879 25880 RADEON: implement support for posting cards based on x86 bios tables (untested) 25881 25882 This is based on the netbsd radeonfb driver by Garrett D'Amore. 25883 The code is not hooked up yet, but should allow you to post cards 25884 without needing to execute bios code. This should be useful on 25885 non-x86 platforms, for posting secndary cards, and for suspend/resume. 25886 Works on legacy bioses only (no ATOM support). 25887 25888 commit e67d1420bf65055ecb6fdfe6b1b1f53aae83854a 25889 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25890 Date: Tue Jun 5 22:31:50 2007 -0400 25891 25892 RADEON: fixup RADEONDisplayVideo() to better handle the crtc being used 25893 25894 commit 7886405308e3288d5c86b6f2c7dbfa8ff865139c 25895 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25896 Date: Mon Jun 4 21:16:56 2007 -0400 25897 25898 RADEON: fix indenting 25899 25900 commit f54ad565a69deb52547fd04e123f56fc1294cd0a 25901 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25902 Date: Mon Jun 4 21:13:16 2007 -0400 25903 25904 RADEON: no need to go through all the crtcs, we've got what we need 25905 25906 commit 687879bff716ad01f9f158860deb8ba770faab99 25907 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25908 Date: Mon Jun 4 21:08:40 2007 -0400 25909 25910 RADEON: add support for RMX on DVI, make sure RMX is only enabled on crtc1 25911 25912 commit e6161e472ff266f69547704a61040228a8704b06 25913 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25914 Date: Mon Jun 4 20:32:55 2007 -0400 25915 25916 RADEON: remove some cruft 25917 25918 commit be0ce38232ea4f5679c5829ab663939144c6e617 25919 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 25920 Date: Mon Jun 4 20:22:15 2007 -0400 25921 25922 RADEON: number the outputs if there are more than one of the same 25923 25924 commit 8ec617f6493dd0aea5d11f92e3d58c3feef8c8fd 25925 Author: Dave Airlie <airlied (a] linux.ie> 25926 Date: Sun Jun 3 17:32:43 2007 +1000 25927 25928 radeon: disable irqs at server start until 3D app starts 25929 25930 commit 2d40fa55e8d7a1cfb204d66ca4a4d95a3b13d5b5 25931 Author: Dave Airlie <airlied (a] linux.ie> 25932 Date: Sun Jun 3 17:32:43 2007 +1000 25933 25934 radeon: disable irqs at server start until 3D app starts 25935 25936 commit dcb64a4d3947e5a9fbda4b72e29a5b6102370f07 25937 Author: Dave Airlie <airlied (a] linux.ie> 25938 Date: Sun Jun 3 17:10:49 2007 +1000 25939 25940 radeon: disable vbl interrupts when no 3d is running on a new enough drm 25941 25942 commit 73d8e3ec8536b4777490b7ba457566f02233811f 25943 Author: Dave Airlie <airlied (a] linux.ie> 25944 Date: Sun Jun 3 17:07:59 2007 +1000 25945 25946 randr-1.2 increase dri minor version for mesa to know we can do vbl on both 25947 25948 commit 63f0d4ed0a98830ecbe18c6e4174689111a59b68 25949 Author: Dave Airlie <airlied (a] linux.ie> 25950 Date: Sun Jun 3 16:49:14 2007 +1000 25951 25952 randr-1.2: add support for vblank on both heads 25953 25954 Also disable vbl when 3D isn't running 25955 25956 commit d7775c1b38b5bdc439a27ec2c3c3a03a5b24cf57 25957 Author: Dave Airlie <airlied (a] linux.ie> 25958 Date: Sun Jun 3 12:46:31 2007 +1000 25959 25960 randr-1.2: make native mode preferred 25961 25962 If we have a native mode from the BIOS make it preferred mode 25963 25964 commit 5a9516fe4f17854acd4fbf3a8eadf5139081dbd4 25965 Author: Paul TBBle Hampson <Paul.Hampson (a] pobox.com> 25966 Date: Sat Jun 2 14:28:18 2007 -0400 25967 25968 RADEONProbePLLParameters sets pll->reference_div, and 25969 RADEONGetPanelInfoFromReg uses it. 25970 25971 This ensures that the former is called before the latter. 25972 (this should fix randr on ppc) 25973 25974 commit 802804461ebdca9a951e7e562ec68fd08d8eae01 25975 Author: Matthieu Herrb <matthieu (a] deville.herrb.com> 25976 Date: Thu May 31 22:23:57 2007 -0600 25977 25978 Fix build whithout XF86DRI 25979 25980 commit 0e0946e0aa2527794e07473199e851bbfbc47cb1 25981 Author: Michel Dnzer <michel (a] tungstengraphics.com> 25982 Date: Wed May 30 18:56:53 2007 +0200 25983 25984 radeon: Lots of warning fixes. 25985 25986 Move code where it's used, remove unused variables, etc. 25987 25988 commit 31c1be420d5277dd15505bd73e6144827a0580cd 25989 Author: Dave Airlie <airlied@nx6125b.(none)> 25990 Date: Wed May 30 17:49:01 2007 +1000 25991 25992 remove these syncs, at least on rs480 it doesn't break 25993 25994 probably requires testing on other r300 based cards, with the syncs in 25995 we hang when moving the cursor into the second CRTC. 25996 25997 commit 7fc02657c4d740941fbda5a8823cf45de3eca3f8 25998 Merge: 800bf5327 4c61c0ee9 25999 Author: Dave Airlie <airlied@nx6125b.(none)> 26000 Date: Wed May 30 17:27:22 2007 +1000 26001 26002 Merge branch 'origin' into randr-1.2-test 26003 26004 Conflicts: 26005 26006 src/radeon_cursor.c 26007 src/radeon_display.c 26008 src/radeon_driver.c 26009 26010 commit 800bf53279e2c2bf854682bbfd6fa16d03afed00 26011 Author: Matthieu Herrb <matthieu (a] deville.herrb.com> 26012 Date: Tue May 29 23:31:13 2007 -0600 26013 26014 Fix build without XF86DRI 26015 26016 commit 4c61c0ee91a2ffeefce30972a584486f1df1d1ae 26017 Author: Matthieu Herrb <matthieu (a] deville.herrb.com> 26018 Date: Tue May 29 21:35:35 2007 -0600 26019 26020 Fix build without XF86DRI 26021 26022 commit 5337e7bd0069a3f2c4ab22b21a19471427ad3d81 26023 Author: Dave Airlie <airlied@nx6125b.(none)> 26024 Date: Wed May 30 08:10:44 2007 +1000 26025 26026 radeon: add bios quirk for nx6125 monid 26027 26028 commit bff809dc8ed07ac39e9b576a87916486a5e37156 26029 Author: Dave Airlie <airlied@nx6125b.(none)> 26030 Date: Wed May 30 08:02:26 2007 +1000 26031 26032 rs480: more unknown regs 26033 26034 Hardcode the values from a working fglrx run, this works for me now 26035 26036 I've no idea what it might do for anyone else 26037 26038 commit 104105fee5c3945d3f210e6a4cb73ab492c61543 26039 Author: Dave Airlie <airlied@nx6125b.(none)> 26040 Date: Tue May 29 19:09:33 2007 +1000 26041 26042 rs480: make second crtc work with magic number in magic register. 26043 26044 I've no idea why or what this does. 26045 26046 commit 5aa603bcabbb077dec169c48438c2e2ebe1195d7 26047 Author: Dave Airlie <airlied@nx6125b.(none)> 26048 Date: Tue May 29 07:23:24 2007 +1000 26049 26050 rs480: only has single dac 26051 26052 commit c52322354fe64725733842b3356798c50e7735d5 26053 Merge: dd6a966e8 975da595f 26054 Author: Dave Airlie <airlied@nx6125b.(none)> 26055 Date: Tue May 29 07:21:48 2007 +1000 26056 26057 Merge branch 'origin' 26058 26059 commit c2637a01f9aca4032262c66ade305f5fe2c54294 26060 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26061 Date: Sat May 26 18:26:35 2007 -0400 26062 26063 RADEON: remove some dead code from the last commit 26064 26065 commit bbb769c4107bfcae682e46d026e54cbfb67d62cd 26066 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26067 Date: Sat May 26 18:25:06 2007 -0400 26068 26069 RADEON: more rotation work (still not there) 26070 26071 - once again borrowed heavily from intel 26072 26073 commit a69f90be9384244744fabfa76469ede9cd26ac98 26074 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26075 Date: Sat May 26 17:54:48 2007 -0400 26076 26077 RADEON: first pass at rotation (not working yet) 26078 26079 - based heavily on intel 26080 26081 commit 3b619d88a65a5801c85d120f2d39704194433f10 26082 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26083 Date: Sat May 26 17:06:18 2007 -0400 26084 26085 RADEON: clean up and add comment regarding clones 26086 26087 commit 089ae4afd23322bf602d340c531c36a4c8b461ed 26088 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26089 Date: Sat May 26 15:49:11 2007 -0400 26090 26091 RADEON: Switch SetOutputType() to use names rather than numbers 26092 26093 commit c4cef0c9481257a744ac99dfc7beb988ce51a8ee 26094 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26095 Date: Sat May 26 13:12:37 2007 -0400 26096 26097 RADEON: make sure we have the lock when SYNCing 26098 26099 commit 899c62e987d5c7524817ba85675e39a8a28e0232 26100 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26101 Date: Sat May 26 12:54:01 2007 -0400 26102 26103 RADEON: Make sure LVDS and FP2 routing info gets written 26104 26105 commit 18857184ffa6847815d349c020b003f8401e36ee 26106 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26107 Date: Sat May 26 12:27:00 2007 -0400 26108 26109 RADEON: fix handling of DRI lock 26110 26111 commit 3a61453efb4f04492cef823b6dd1273b55c6a785 26112 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26113 Date: Sat May 26 01:38:09 2007 -0400 26114 26115 RADEON: implement backlight control for LVDS 26116 26117 This code is currently disabled as I'm not sure which 26118 laptops actually use this method for backlight control. 26119 My laptop seems to use another method as adjusting the 26120 backlight level doesn't seem to touch LVDS_GEN_CNTL. 26121 Maybe just macs? 26122 26123 commit ceec3f62257bafe4771e75d3c4f1d2a517d7acf8 26124 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26125 Date: Fri May 25 23:29:51 2007 -0400 26126 26127 RADEON: implement auto/analog/digital output property for DVI-I (untested) 26128 26129 commit e3e9c608651e7cbb9851489274815f47d2a8dbbb 26130 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26131 Date: Fri May 25 19:26:24 2007 -0400 26132 26133 RADEON: more clean of last commits 26134 26135 commit a7d2c9c0f6155c4e65a2e73f3832c0d8ca7af5fc 26136 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26137 Date: Fri May 25 19:09:11 2007 -0400 26138 26139 RADEON: fixup last commit and add support for ext TMDS detect 26140 26141 commit e50ca35ce2ea79dadb38ce14e459eed836452ff7 26142 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26143 Date: Fri May 25 18:56:32 2007 -0400 26144 26145 RADEON: clean up non-DDC probe, add detect for DVI 26146 26147 commit abb9b57e7b4162ab6c21fd1e809d24f13f7e1ea1 26148 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26149 Date: Wed May 23 21:37:58 2007 -0400 26150 26151 RADEON: clean up some logic 26152 26153 commit 679236defd28bea8874004377436347950ff7bec 26154 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26155 Date: Wed May 23 21:18:40 2007 -0400 26156 26157 RADEON: add output properties (not functional yet) 26158 26159 commit 975da595f032c145ad74079ff8edeaead779dc7b 26160 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26161 Date: Tue May 22 10:56:47 2007 +0200 26162 26163 radeon: Provide new DRI texOffsetStart hook when available with EXA. 26164 26165 commit e38cad5633c974467f6417051eda8d487ac3686e 26166 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26167 Date: Tue May 22 00:09:46 2007 -0400 26168 26169 RADEON: Only add valid connectors 26170 26171 commit 8275151baac22c34149cef0b7d922771d24abc3e 26172 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26173 Date: Mon May 21 10:25:48 2007 +0200 26174 26175 radeon: HW cursor cleanup. 26176 26177 Don't needlessly turn the HW cursor on/off in RADEONLoadCursor*(). 26178 26179 Besides cleaning up the code, this semms to avoid some HW cursor related 3D 26180 lockups, see https://bugs.freedesktop.org/show_bug.cgi?id=10815 . My best 26181 guess is that this is because the engine is now always idled before touching 26182 the CRTC registers. 26183 26184 commit 137e3fc1899078af0f72303ab0a4e6cf35804a7b 26185 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26186 Date: Mon May 21 10:25:48 2007 +0200 26187 26188 radeon: Suppress debugging output by default. 26189 26190 It can be enabled at runtime by increasing the log verbosity level. 26191 26192 Also change the prefix from (**) to (II) to make grepping the log file for 26193 defaults overridden by xorg.conf more useful again. 26194 26195 Turn some MC related debugging output into normal informational output as it's 26196 useful for recognizing corner cases that can cause stability issues. 26197 26198 commit aed193a47a939451d9a6d05b02653b1d73e1d523 26199 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26200 Date: Sun May 20 21:13:59 2007 -0400 26201 26202 RADEON: fix crtc1 (un)blanking code after last commit 26203 26204 - crtc1 MC control is on CRTC_GEN_CNTL 26205 - fix indenting 26206 26207 commit 870c8043068a0f44b53d1148371b2cc1e3970a7b 26208 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26209 Date: Sun May 20 21:03:10 2007 -0400 26210 26211 RADEON: several updates 26212 26213 - move output init from InitCrtc() functions to the output mode_set() 26214 - take the crtc off the MC when blanking 26215 - move EnableDisplay() to output commit() 26216 - clean up some dead code 26217 - don't enable dacs in initcrtc() functions, this is taken care of in EnableDisplay() 26218 26219 commit 09bfc8ed000f95ede5b73f2bad69edc1a4d9bac6 26220 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26221 Date: Sun May 20 18:06:22 2007 -0400 26222 26223 update to 6.6.192 for rc release 26224 26225 commit 764cb73e8dec4040cdd418d249fc504399fca3ee 26226 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26227 Date: Sun May 20 17:26:26 2007 -0400 26228 26229 Fix regular/"xinerama"/zaphod dualhead mode 26230 26231 - logic in RADEONUnblank() was wrong 26232 - Calling RADEONSetupConnectors() on second instance screwed up the port info 26233 - still seem to be HW cursor issues with zaphod mode 26234 26235 commit 9a147fef8e0e2ede2a0008c4ecfbd9b00c8dc5f6 26236 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26237 Date: Fri May 18 23:45:11 2007 -0400 26238 26239 RADEON: if connector is VGA set TMDS to none 26240 26241 commit bbd6faff4c6acb48970d774375c8a61861405f96 26242 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26243 Date: Fri May 18 23:30:19 2007 -0400 26244 26245 RADEON: Change default LVDS i2c line for powerpc 26246 26247 commit 7940ec364f3cbf02ba64b92c74cbaad4555baf38 26248 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26249 Date: Fri May 18 00:23:08 2007 -0400 26250 26251 RADEON: cleanup 26252 26253 commit f71bfde7352ef858c1041037d7dc77c237e315a4 26254 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26255 Date: Fri May 18 00:12:03 2007 -0400 26256 26257 RADEON: more cleanup 26258 26259 commit f711e266d6927dec648f6ff26f15c6f48643f78c 26260 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26261 Date: Thu May 17 23:59:07 2007 -0400 26262 26263 RADEON: turn off all outputs in screeninit(). We'll turn on the ones we want later 26264 26265 commit bfc1c372d7475b7fa6bffb8a79dec1dc2f11ec59 26266 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26267 Date: Thu May 17 23:45:29 2007 -0400 26268 26269 RADEON: More cleanup 26270 26271 commit c60b3bc9e3e7463bdb42e2478be8cc3f22c63c68 26272 Author: Paul TBBle Hampson <Paul.Hampson (a] pobox.com> 26273 Date: Thu May 17 23:26:54 2007 -0400 26274 26275 Fix for infinite loop in RADEONGetLVDSInfo 26276 26277 commit 11289f8206cab1a94bd64a3938cf9af50f19497e 26278 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26279 Date: Tue May 15 00:48:31 2007 -0400 26280 26281 RADEON: remove some debugging code 26282 26283 commit fba1a11e287ebf04cf311645d31299896bad7283 26284 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26285 Date: Tue May 15 00:44:29 2007 -0400 26286 26287 RADEON: Move DAC regs to their own Save() function 26288 26289 commit 9ed00e959fcdab2739a7e64ca1e303a6faf28f15 26290 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26291 Date: Tue May 15 00:36:05 2007 -0400 26292 26293 RADEON: Lots of small fixes 26294 26295 - remove extra crtc2_base set 26296 - remove some debugging code 26297 - clean up RestoreMode() 26298 - make sure a DDC reg gets assigned for LCD_DDC 26299 - make sure we adjust the right frame in AdjustFrame() 26300 - rename RADEONMapControllers() to RADEONPrintPortMap() to 26301 reflect what it actually does now 26302 - make i2c bus name match the DDC port 26303 - remove or comment out un-needed code 26304 26305 commit 5106c8fe5f74865cac8b00937739a4efde9fc254 26306 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26307 Date: Sun May 13 21:48:26 2007 -0400 26308 26309 RADEON: Further cleanup 26310 26311 commit 224a73e41e9be344d5644203e7ebd5a3a8272604 26312 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26313 Date: Sun May 13 21:41:59 2007 -0400 26314 26315 RADEON: remove remnants of "old" multi-head support 26316 26317 The old screen based multi-head code is broken at this point 26318 and I have no intention of fixing it, so to clean things up 26319 I've removed it. 26320 26321 commit 9c2f20a83a5fed14225f4c3ebcd8ca41e9d4bd48 26322 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26323 Date: Sun May 13 21:13:23 2007 -0400 26324 26325 RADEON: additional cleanups 26326 26327 commit e187321ab8dd58d2b2fe92c062d070ba4820a2bf 26328 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26329 Date: Sun May 13 19:49:10 2007 -0400 26330 26331 RADEON: add new ConnectorTable option and re-add PanelSize option 26332 26333 commit e60a7bcaf5611ad4706b1d1442ce4cae49145f42 26334 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26335 Date: Sun May 13 18:14:29 2007 -0400 26336 26337 RADEON: remove mergedfb, etc. cruft from man page 26338 26339 commit 51caa5dccc4a2e14d1b41bb7868ad20468dd4cfe 26340 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26341 Date: Sun May 13 18:10:02 2007 -0400 26342 26343 RADEON: remove old mergedfb and dualhead options that are no longer used 26344 26345 commit c5da9d4040cb08598d171d20f84d3f6c20a033e0 26346 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26347 Date: Sun May 13 18:05:38 2007 -0400 26348 26349 RADEON: remove more dead code 26350 26351 commit e776fa9ecc53d63f916a5447a76fb2bb3b824167 26352 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26353 Date: Sun May 13 15:43:39 2007 -0400 26354 26355 RADEON: re-org randr code 26356 26357 - move crtc stuff to radeon_crtc.c 26358 - move output stuff to radeon_output.c 26359 26360 commit 58ce388452b7bc790c438d75c9cf4a0f69f0d7b2 26361 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26362 Date: Sun May 13 15:05:01 2007 -0400 26363 26364 RADEON: Remove dead code 26365 26366 commit 66e8e6c8348d007930730e90295588efe8108844 26367 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26368 Date: Sun May 13 14:25:03 2007 -0400 26369 26370 RADEON: several fixes 26371 26372 - give better DDC names 26373 - disable DAC if connector is DVI-D 26374 - fix indentation in ProbeDDCModes() 26375 26376 commit 1c16c2ce9c5b02b03d23da965127d82eea4c4039 26377 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26378 Date: Sun May 13 13:54:04 2007 -0400 26379 26380 RADEON: fix VT switch 26381 26382 commit 7c66e903368f77ecc4d3bd1c9f08d2adbd85e83c 26383 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26384 Date: Sun May 13 13:37:59 2007 -0400 26385 26386 RADEON: move crtc offset handling into init/save/restore functions 26387 26388 commit aec078eb0740651fba8ec602e8239bd679efc8ad 26389 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26390 Date: Sun May 13 11:57:57 2007 -0400 26391 26392 RADEON: fix up DDCConnected() 26393 26394 commit 0550c37ecc434b8075fb3c367d100ff27625bb64 26395 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26396 Date: Sun May 13 11:44:50 2007 -0400 26397 26398 RADEON: don't need to pass pScrn to EnableDisplay() 26399 26400 commit 117220527de9fd3158f600645bcfcaf46847f45f 26401 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26402 Date: Sun May 13 11:37:35 2007 -0400 26403 26404 RADEON: remove hardcoded output limit and PortInfo stuff 26405 26406 commit 7e5c29961ac2a9e9dbe5d6d2d73d11cd018d62b5 26407 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26408 Date: Fri May 11 18:00:40 2007 +0200 26409 26410 RADEON: Fix RMX after the last commit 26411 26412 commit ab5603edd8fc3ef0560bdfb6a6d9c6af2a49d1e5 26413 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26414 Date: Fri May 11 17:34:35 2007 +0200 26415 26416 RADEON: Move LVDS, TMDS, DAC properties to the output rec 26417 26418 commit 94eb0681de0641e490f06486468617a727fefe86 26419 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26420 Date: Fri May 11 14:39:48 2007 +0200 26421 26422 RADEON: switch output dpms to use RADEONEnableDisplay() 26423 26424 commit 673ede5578d5d9caf2adf0445fe1e684b034eea5 26425 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26426 Date: Fri May 11 14:28:17 2007 +0200 26427 26428 RADEON: re-arrange output mode setting 26429 26430 commit 0070a7d787adaae99f7bc2659be4b0f49f439db5 26431 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26432 Date: Fri May 11 13:06:15 2007 +0200 26433 26434 RADEON: more re-org 26435 26436 - move crtc mode setting around 26437 - add dri lock/unlock to crtc lock/unlock calls 26438 26439 commit bba456232ac9a6218aa7fbd504d6093fa72860cf 26440 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26441 Date: Thu May 10 15:33:51 2007 +0200 26442 26443 RADEON: fix VT switch 26444 26445 commit 1779a12a947401e5c6bcf784b47e9b3c80d37204 26446 Author: Jesse Barnes <jbarnes (a] jbarnes-mobile.amr.corp.intel.com> 26447 Date: Thu May 10 06:22:35 2007 -0700 26448 26449 Add cscope files to .gitignore 26450 26451 commit 0cb23277666db3b30438c6f88840d861e04df414 26452 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26453 Date: Thu May 10 15:20:56 2007 +0200 26454 26455 RADEON: randr driver re-org checkpoint 26456 26457 - split the mode setting per-crtc 26458 - reduce start up flicker 26459 26460 commit 33c370b1d8350945f80ac12097d3e91243a400f2 26461 Author: Jesse Barnes <jbarnes (a] jbarnes-mobile.amr.corp.intel.com> 26462 Date: Wed May 9 16:16:39 2007 -0700 26463 26464 RADEON: 26465 - fix an ugly modesetting bug: if we happened to set the mode on 26466 CRTC1 before CRTC2, CRTC2's RestoreMode function would clobber 26467 CRTC1's CRTC_OFFSET register since we never updated 26468 ModeReg.crtc_offset... so make AdjustFrame use ModeReg and pull 26469 the call to it up before RestoreMode, seems to work ok here. 26470 26471 commit 6263248a0044777a352e4ee7380b4b8f9afd091b 26472 Author: Jesse Barnes <jbarnes (a] jbarnes-mobile.amr.corp.intel.com> 26473 Date: Wed May 9 14:52:00 2007 -0700 26474 26475 RADEON: 26476 - use fixup_mode hook to set RADEON_USE_RMX flag so panel scaling 26477 works 26478 - use valid_mode hook to prune invalid default modes from list 26479 - use adjusted_mode in crtc_mode_set (using adjusted_mode from 26480 fixup hook) 26481 26482 commit d2497009e395800fbde5777465f3087a54b94418 26483 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26484 Date: Wed May 9 22:23:45 2007 +0200 26485 26486 randr checkpoint 26487 26488 - server still hangs if you start with external monitor connected 26489 - RMX not working 26490 - more than 2 outputs now possible (untested) 26491 26492 commit 4488f0737d5268168eab41440b7a3b5732efb15e 26493 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26494 Date: Wed May 9 16:41:13 2007 +0200 26495 26496 more randr re-work 26497 26498 - remove RADEONQueryConnectedDisplays(); randr takes care of this now 26499 - print edid after randr detection 26500 26501 commit 4d992386e2ab9d8c50f0484445564325dfb42930 26502 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26503 Date: Wed May 9 16:12:21 2007 +0200 26504 26505 enable all outputs on each crtc on mode restore 26506 26507 commit 2618cf2aa8ed76411b943eb90c95869814c2f151 26508 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26509 Date: Wed May 9 15:48:40 2007 +0200 26510 26511 More re-org to allow more than 2 outputs 26512 26513 - Move radeon output crtc map into crtc_set_mode 26514 - in modeinit, set up all outputs attached to crtc 26515 26516 commit 61b9e79cbeee6f735a4c82ec8a802aee85d8b890 26517 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26518 Date: Wed May 9 15:11:23 2007 +0200 26519 26520 start to clean up MapControllers 26521 26522 commit 6a724dd798c1a7b461672993c02be83a7bccded6 26523 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26524 Date: Wed May 9 14:48:17 2007 +0200 26525 26526 minor cleanup 26527 26528 commit 03860fed24b4f76cc7f1f4210ec8f8040fa04777 26529 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26530 Date: Tue May 8 19:27:48 2007 +0200 26531 26532 remove some old mergedfb cruft 26533 26534 commit 76670f665ebec7cdf40a04bf9379cb3ad4417507 26535 Merge: 83f81ed5e a3ee42207 26536 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26537 Date: Tue May 8 18:41:25 2007 +0200 26538 26539 Merge branch 'master' into randr-1.2 and fix conflicts 26540 26541 commit 83f81ed5e3c33c94c80500316c37a7cbfc51f41f 26542 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26543 Date: Mon May 7 22:54:16 2007 +0200 26544 26545 RADEON: Fix cursor handling for randr 26546 26547 commit 5fc21cd34436919300018d8d4850fc67db284eb3 26548 Author: Jesse Barnes <jbarnes (a] jbarnes-mobile.amr.corp.intel.com> 26549 Date: Mon May 7 13:31:20 2007 -0700 26550 26551 - fix randr 1.2 on pre-RV350 chips by enabling DAC_MACRO_CNTL writes 26552 - set dac_cntl on non-primary crtcs 26553 - set XCRT_CNT_EN in CRTC_EXT_CNTL just because 26554 - fix warnings in calls to xf86PrintModeline (wants screen index not 26555 pScrnInfo) 26556 26557 commit 99b3df154317f0209618e532282a3e7ad091c00f 26558 Author: Alex Deucher <alex (a] t41p.hsd1.va.comcast.net> 26559 Date: Mon May 7 19:02:35 2007 +0200 26560 26561 RADEON - update randr cursor handling, LVDS setup fix 26562 26563 - quick fix for the cursor handling to update to the latest 26564 server bits 26565 - make sure connector type is CONNECTOR_PROPRIETARY for LVDS. 26566 26567 commit a3ee42207aab77d93655a82fdcb32be38268b85f 26568 Author: Dave Airlie <airlied (a] linux.ie> 26569 Date: Wed Apr 25 09:52:22 2007 +1000 26570 26571 radeon: another attempt at fixing the mergedfb refresh rate 26572 26573 This attempts to keep it inside the 32-bit limit when multiplying things out 26574 later in the randr tree. 26575 26576 Let me know if I screwed this up.. 26577 26578 commit 3828237200fc16d4d32664fb8358950c213d4897 26579 Author: Dave Airlie <airlied@nx6125b.(none)> 26580 Date: Sun Apr 22 11:36:00 2007 +1000 26581 26582 radeon: add support for DDC on some laptop chipsets 26583 26584 I noticed fglrx has DDC for the panel in the rs480 laptop, however radeon 26585 didn't pick it up, so I valgrinded fglrx and spotted 0x1a0/0x1a4 accesses 26586 I actually noticed this before from the BIOS but never figured it out. 26587 26588 So now I get DDC from the LCD on this laptop. 26589 26590 commit dd6a966e862b774a8e8b9e1a085309219673efad 26591 Author: Dave Airlie <airlied@nx6125b.(none)> 26592 Date: Sun Apr 22 11:36:00 2007 +1000 26593 26594 radeon: add support for DDC on some laptop chipsets 26595 26596 I noticed fglrx has DDC for the panel in the rs480 laptop, however radeon 26597 didn't pick it up, so I valgrinded fglrx and spotted 0x1a0/0x1a4 accesses 26598 I actually noticed this before from the BIOS but never figured it out. 26599 26600 So now I get DDC from the LCD on this laptop. 26601 26602 commit c81ed9bd7b37c9d02141d10f6c7bad3d0c57032f 26603 Author: Dave Airlie <airlied (a] linux.ie> 26604 Date: Sat Apr 21 18:58:40 2007 +1000 26605 26606 radeon: fix build on older server 26607 26608 commit 406eec71116a58d42288a7f1c809a92d5bda7350 26609 Author: Dave Airlie <airlied (a] linux.ie> 26610 Date: Sat Apr 21 18:56:28 2007 +1000 26611 26612 radeon: fix build since patches for IBM don't actually build 26613 26614 commit ad119960095b4b64f4c6793f65950c9967ce4989 26615 Author: Matthias Hopf <mhopf (a] suse.de> 26616 Date: Thu Apr 19 11:54:46 2007 +0200 26617 26618 Disable RMX for IBM Lewis server. 26619 26620 Due to the hardware layout RMX ddc_mode has to be set. 26621 If ddc_mode is set, RADEONValdiateFPModes() shouldn't be called. 26622 Bugzilla #10620 (3). 26623 26624 commit 16ef77df4ebaf5ea13baa82972aaf98e71ac32ee 26625 Author: Matthias Hopf <mhopf (a] suse.de> 26626 Date: Wed Apr 18 17:36:15 2007 +0200 26627 26628 Set sync polarity restriction flags even for non-"digital separate" monitors. 26629 26630 According to Lisa Wu, this is correct regarding the VESA EEDID standard. 26631 Bugzilla #10620 (2), original patch by Lisa Wu @ATI 26632 26633 commit 0abce69f0d826a7ca1a41d963cd4730b6e01c145 26634 Author: Matthias Hopf <mhopf (a] suse.de> 26635 Date: Wed Apr 18 17:32:52 2007 +0200 26636 26637 Fix inconsistent use of Mode lists. 26638 26639 Some scans used to only check every second entry, some stopped at the entry 26640 before the last entry. 26641 Bugzilla #10620 (1), original patch by Lisa Wu @ATI 26642 26643 commit aea801cf9a5ce519a53d6fffd9a3a2e526ec79ea 26644 Author: Matthias Hopf <mhopf (a] suse.de> 26645 Date: Fri Apr 13 16:16:05 2007 +0200 26646 26647 Fix crash if MergedFB and secondary head not found 26648 26649 If the secondary head isn't found (Monitor unplugged etc.) but MergedFB 26650 is configured, the driver segfaults because it tries to access the mode 26651 list private structures, which are not filled in. 26652 26653 commit 07ddffb32e6293c77b32c94b87ec468caef3d6f5 26654 Author: Matthias Hopf <mhopf (a] suse.de> 26655 Date: Wed Apr 11 14:36:51 2007 +0200 26656 26657 Fixed typo in mode list generation. 26658 26659 commit 4effa67ea75736a31b9e78a7b35acf74b067c43e 26660 Author: Dave Airlie <airlied (a] linux.ie> 26661 Date: Mon Apr 9 22:08:31 2007 +1000 26662 26663 radeon: add support for enabling direct rendering on RS480 26664 26665 Thanks to Matthew Garrett and Ubuntu for the hw loan to get this working. 26666 26667 Still no 3D driver support but at least you should get CP acceleration for 26668 2D now. 26669 26670 commit 6b25a4c48796e022a093f3072574ffe9709ecaf4 26671 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26672 Date: Wed Apr 4 10:47:29 2007 +0200 26673 26674 radeon: Link nearest modes by default for clone mode. 26675 26676 This makes sure all modes of both CRTCs will be available by default with 26677 MergedFB. 26678 26679 commit 3c892f163ec1fa9be6e733aab091c9b718f41efc 26680 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26681 Date: Wed Apr 4 09:52:37 2007 +0200 26682 26683 radeon: Always sort modes when adding to list. 26684 26685 This makes sure mode lists will always be sorted from larger to smaller. 26686 26687 commit 3a8190ccc79969925257e7b980b78d79053d208d 26688 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26689 Date: Wed Apr 4 09:47:07 2007 +0200 26690 26691 radeon: Don't shrink virtual size based on secondary modes. 26692 26693 commit 9b1e97284ce185d358ca756a235d2cee346fa53f 26694 Author: Henry Zhao <henryz (a] localhost.localdomain> 26695 Date: Sat Mar 31 23:01:52 2007 -0800 26696 26697 10205: Radeon driver's own mode validation code does not work properly 26698 26699 commit 9c2dcd19be8fc2cc29e637d1e9748e66196e3900 26700 Author: Henry Zhao <henryz (a] localhost.localdomain> 26701 Date: Sat Mar 31 20:10:03 2007 -0800 26702 26703 9337: EDID modes do not participate in validation for CRT monitor 26704 26705 commit 1acd6d6fa42acec07fb11aeb189f492ddb021cb4 26706 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26707 Date: Fri Mar 30 11:06:10 2007 +0200 26708 26709 radeon: Guard some MergedFB specific code with info->MergedFB tests. 26710 26711 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=10442 . 26712 26713 commit 1a71106c0e4fe5f650239dc694163fdf52d33663 26714 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26715 Date: Tue Mar 27 10:13:21 2007 +0200 26716 26717 radeon: Fix typo. 26718 26719 commit f87e57d4d773a019d1cc8a10425c57480430f6a4 26720 Author: Dave Airlie <airlied (a] linux.ie> 26721 Date: Tue Mar 27 18:08:54 2007 +1000 26722 26723 radeon: fix up crtc debug dereference problem 26724 26725 commit 66b4a571a4e7960da6807d3f30955aa08e89ccc6 26726 Author: Dave Airlie <airlied (a] linux.ie> 26727 Date: Tue Mar 27 17:00:37 2007 +1000 26728 26729 update number to 6.6.191 for rc release 26730 26731 commit fca30a6b581cb6c1466ab1bc316df8fed5d82b60 26732 Author: Alex Deucher <alex (a] botch2.com> 26733 Date: Mon Mar 26 23:26:51 2007 -0400 26734 26735 fix NULL dereference 26736 26737 when IsSecondary is true, crtc1 is NULL 26738 Noticed by Sverre Froyen. 26739 26740 commit 97d8d1ed10d069343f2b3172ba64ca421821a602 26741 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26742 Date: Sat Mar 24 20:02:12 2007 +0200 26743 26744 Move atichip.c from ati to atimisc. 26745 26746 commit aa8f5b02ebc9be60df48722588261627d6a457e8 26747 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26748 Date: Sat Mar 24 19:53:02 2007 +0200 26749 26750 [mach64] Use Mach64Chipsets[] instead of ATIChipNames[]. 26751 26752 atimisc is PCI-only now, we can get the chip name with xf86TokenToString(). 26753 26754 commit 39e896a1e688ea2d2d21f88c1c5d34c5810aac1c 26755 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26756 Date: Sat Mar 24 19:47:18 2007 +0200 26757 26758 [mach64] Set pATI->ChipRevision correctly, instead of overriding. 26759 26760 There is no need to override pATI->ChipRevision for GX/CX, as it is only 26761 reported with a printf. 26762 26763 commit f046a910ca117279fbabc6281b2e23439ec9ea4e 26764 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26765 Date: Fri Mar 23 22:19:17 2007 +0200 26766 26767 Drop probing by driver name from "Device" section. 26768 26769 atimisc is PCI-only now, so we only need to scan the PCI space. 26770 26771 commit 9cd175d9cd4ed710fccb303664c77519ecaf1e21 26772 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26773 Date: Fri Mar 23 22:12:48 2007 +0200 26774 26775 Fold FillIn() back to Probe(). 26776 26777 commit d7a8cd0e476034796fc38e25a28cd28d05ea4a13 26778 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26779 Date: Fri Mar 23 22:10:03 2007 +0200 26780 26781 Drop now unneeded _X_EXPORT's. 26782 26783 commit 58626d8b78f26f0d9c480033d3c3a12e541342b1 26784 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26785 Date: Fri Mar 23 21:30:19 2007 +0200 26786 26787 Move {atimach64,r128,radeon}_probe.c from ati to subdrivers. 26788 26789 Subdrivers are now loaded from the wrapper Probe function rather than at screen 26790 creation time. 26791 26792 The wrapper Identify callback only prints chip families now, chip lists are 26793 printed when a subdriver is loaded. This also avoids duplication of subdriver 26794 Identify callbacks. 26795 26796 Unknown radeons should still get a list of known radeon and then probe fails... 26797 26798 Probe for atimisc last to avoid needless loading in most cases (r128, radeon). 26799 26800 commit 2e3d43af1e5077cc61dd8668551a6291368d9ed2 26801 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26802 Date: Mon Mar 19 19:05:04 2007 +0200 26803 26804 [mach64] Compute MMIO address once for each case (PIO, MMIO). 26805 26806 For an MMIO-only driver, the extra MMIO address probing is not needed since we 26807 already compute the MMIO address during probe and do not make any BIOS calls. 26808 26809 For a PIO driver, this extra probing could even result in a wrong MMIO address 26810 since the checking is done by reading through PIO, not MMIO. The MMIO address 26811 is later corrected but having the extra probe there is useless. 26812 26813 commit e7defc0e930f1b2d83623a769b2bfdb05c66a5fe 26814 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26815 Date: Mon Mar 19 19:04:10 2007 +0200 26816 26817 [mach64] Move setting of PIO and MMIO addresses up, after BIOS init. 26818 26819 Partially revert "Minor refactoring of pATI->Block0Base computation." 26820 26821 This parially reverts commit 2c8b33b761e4411451aea1eea3a89d629aeea40d. 26822 26823 commit 3dfdba195f1ff794ab00585ab80e7ea6ac92fcf4 26824 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26825 Date: Sun Mar 18 16:40:07 2007 +0200 26826 26827 [mach64] Make mapping code friendlier to pci-rework source-code compatibility. 26828 26829 commit 321db6d87c9110654ab386e5ec270116eca1e04f 26830 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26831 Date: Sun Mar 18 16:15:20 2007 +0200 26832 26833 [mach64] Add macros for pci-rework source-code compatibilty. 26834 26835 commit dfd07b6e99020d1db43d7ce0cae4423d8c6b1f05 26836 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26837 Date: Sun Mar 18 16:38:26 2007 +0200 26838 26839 [mach64] Drop checking that BAR base + size fits in 32 bit. 26840 26841 I can't see how these tests could fail on 32-bit. OTOH they are potentially 26842 problematic with 64-bit PCI addresses. 26843 26844 commit e674338a98c50800637b8ebc01adf3aec2a3eb38 26845 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26846 Date: Sun Mar 18 15:50:41 2007 +0200 26847 26848 [mach64] Single mapping per BAR. 26849 26850 - We reset write-combining of the FB when MMIO registers fall in the linear 26851 aperture. 26852 26853 - We set MMIOInLinear in all three aperture mappings since the mapping code 26854 depends on it. 26855 26856 - We map the linear aperture in its entirety for MMIOInLinear also. 26857 This simplifies mapping of HW cursor and the LE aperture on BE archs, as 26858 they always fall in the linear aperture. 26859 26860 - I assumed that BAR[2] is page-aligned (we know it's 4K-aligned for sure). 26861 26862 commit e7969de8ec572d8801806f9525a9a081a58a03f1 26863 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26864 Date: Sun Mar 18 15:35:25 2007 +0200 26865 26866 [mach64] Drop poking the VGA MMIO aperture. 26867 26868 We will not be able to handle this when we do mappings for PCI BARs only and 26869 it should be handled with a CPIO driver anyway. 26870 26871 commit 166c760a86165330175023e07c4b2bd6891633c5 26872 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26873 Date: Tue Mar 20 09:16:02 2007 +0100 26874 26875 Fix advertised minimum minor version of the DRI module. 26876 26877 We don't automatically require bumped minor versions. 26878 26879 commit 1bdd376dbd57de8925244f0808f974d6d8cff39d 26880 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26881 Date: Tue Mar 20 09:14:39 2007 +0100 26882 26883 radeon: Only sync to hardware when really necessary with EXA. 26884 26885 In particular, don't sync again after accelerated DownloadFromScreen, which 26886 syncs implicitly. This avoids calling into the kernel when it's not necessary, 26887 which can be relevant in some situations. 26888 26889 commit 3cfa3a5c8daf03aaad6fc30d275709f6eb717d29 26890 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26891 Date: Tue Mar 20 09:13:24 2007 +0100 26892 26893 radeon: Unify code to release the CP. 26894 26895 commit e174d8df8c801fad95e5f79cff69187c200bee6e 26896 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26897 Date: Tue Mar 20 09:12:51 2007 +0100 26898 26899 radeon: Minor BlockHandler cleanups. 26900 26901 Don't flush indirect buffer in BlockHandler; it's done in LeaveServer. 26902 26903 Also set the EXA engine mode to unknown only at the end. 26904 26905 commit 113fb4b61e709a9b54fc2ef73efce06011e771c1 26906 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26907 Date: Tue Mar 20 09:12:09 2007 +0100 26908 26909 radeon: Disable CP line acceleration on RV280s. 26910 26911 There have been several reports of stability issues with things like the 26912 xscreensaver hack hypercube. 26913 26914 commit 4651d00b183cb498879d605c4b93cd3a0c63cb33 26915 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26916 Date: Tue Mar 20 09:10:19 2007 +0100 26917 26918 radeon: Make sure 3D clients will re-upload textures to video RAM after LeaveVT. 26919 26920 Walk the SAREA texList and bump the age of every active object, so their owners 26921 will consider them kicked out when they grab the HW lock next time. 26922 26923 commit 2d2fb54ba370c1df9ef5102e83c17a7ff5c55403 26924 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26925 Date: Thu Mar 15 17:26:56 2007 +0100 26926 26927 atimisc: Fix strict aliasing violation flagged by -Wall. 26928 26929 commit cfb82631fb683a0dc9bbc8f2d214646a8f1bca0b 26930 Author: Michel Dnzer <michel (a] tungstengraphics.com> 26931 Date: Thu Mar 15 17:12:43 2007 +0100 26932 26933 radeon: Fix build warnings. 26934 26935 commit 50a524445ed1ea3c4fab56de772921ba86a21d35 26936 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26937 Date: Thu Mar 8 17:43:48 2007 +0200 26938 26939 Move atioption.c from ati to atimisc. 26940 26941 commit 52684241ba1ffb393cb655ecc09cbd0cef5b1717 26942 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26943 Date: Thu Mar 8 07:07:02 2007 +0200 26944 26945 Move atiprobe.c from ati to atimisc. 26946 26947 commit 2a1cd107a593630001799d6cd9e72c64222553b2 26948 Author: Jesse Barnes <jbarnes (a] jbarnes-mobile.amr.corp.intel.com> 26949 Date: Tue Mar 6 14:35:50 2007 -0800 26950 26951 Add prepare/commit hooks to output and crtc func table 26952 26953 commit 703c6fc0142ffc600285c13fe6dafecf988c0a1d 26954 Author: Jesse Barnes <jbarnes (a] jbarnes-mobile.amr.corp.intel.com> 26955 Date: Tue Mar 6 10:12:42 2007 -0800 26956 26957 Update for new CRTC resize hooks. 26958 26959 commit 1e270b411cf691c69194932b864ee2c9db292263 26960 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26961 Date: Sun Mar 4 18:21:49 2007 +0200 26962 26963 [mach64] Bug 5586: overlay scaler limited to 720 pixels on ATI Rage Pro. 26964 26965 commit 54f9b8c1d01ecb97e2d4d247918f0efc86ed6037 26966 Author: George Sapountzis <gsap7 (a] yahoo.gr> 26967 Date: Sun Mar 4 18:16:40 2007 +0200 26968 26969 [mach64] Xv: use single surface and encoding structs. 26970 26971 commit e1e55b533d3ae528c8da37dcb77ed906d05697fb 26972 Author: Dave Airlie <airlied (a] linux.ie> 26973 Date: Sun Mar 4 19:11:34 2007 +1100 26974 26975 radeon: add option to set pci aperture size from config file 26976 26977 This requires a drm > 1.26 to work 26978 26979 commit 58ee31d015cf8bec0edca62a46faec0b3505be8c 26980 Author: Dave Airlie <airlied (a] linux.ie> 26981 Date: Sun Feb 25 23:29:09 2007 +1100 26982 26983 fix typo 26984 26985 commit 3cfe94d5438961b869766dacbcd13fde8b770ca2 26986 Merge: 31c018ca4 3b43788c4 26987 Author: Dave Airlie <airlied (a] linux.ie> 26988 Date: Sun Feb 25 23:27:19 2007 +1100 26989 26990 Merge branch 'radeon-randrv12-v4' 26991 26992 Conflicts: 26993 26994 src/radeon_display.c 26995 src/radeon_driver.c 26996 26997 commit 3b43788c45f51ad2d3e8e64383c412f4ddd97207 26998 Author: Dave Airlie <airlied (a] linux.ie> 26999 Date: Sun Feb 25 23:17:31 2007 +1100 27000 27001 cleanup radeon code against master server 27002 27003 commit 24c6fa7cfac5602ba9e6e2f331bcac52fab258e5 27004 Author: Dave Airlie <airlied (a] linux.ie> 27005 Date: Sun Feb 25 21:43:54 2007 +1100 27006 27007 make radeon randr build against master server 27008 27009 commit 4374895d393bf170134ccffc188521fe515d5a77 27010 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27011 Date: Sat Feb 24 21:25:13 2007 +0200 27012 27013 [mach64] Consolidate adjustments of mode timings, part 3. 27014 27015 Cosmetic. 27016 27017 commit 8b9a4ec8af90b2f9f0d0a3a7f6627061df7e01b3 27018 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27019 Date: Sat Feb 24 21:22:49 2007 +0200 27020 27021 [mach64] Consolidate adjustments of mode timings, part 2. 27022 27023 Factor out to seperate function. 27024 27025 commit bb226c9e7218d2a65c056fe74cb7eece0550670f 27026 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27027 Date: Sat Feb 24 21:20:53 2007 +0200 27028 27029 [mach64] Consolidate adjustments of mode timings, part 1. 27030 27031 Move LCD block from atimode.c to atimach64.c . 27032 27033 commit 31c018ca4a18ce426b29006f103f56eee7f985fa 27034 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 27035 Date: Wed Feb 21 13:47:37 2007 +0100 27036 27037 add missing part of last commit... 27038 27039 commit df07fa14da73e92d1a6ee0173468ab5c075d1cbb 27040 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 27041 Date: Wed Feb 21 13:05:42 2007 +0100 27042 27043 fix alignment issues with planar yuv and a bug with packed uyvy 27044 27045 respect that all source planar yuv planes are already dword aligned. 27046 Some attempts to fix up odd widths and odd heights (which are a bit strange 27047 for 4:2:0 formats). They still don't quite work 100% correctly (at the borders) 27048 but neither do they with packed yuv formats. 27049 While here, fix totally broken packed UYVY format by inserting missing break... 27050 Both bugs reported by Felipe Contreras. 27051 27052 commit dfcb431adfbbaaee0d262d32735585555a0cbde4 27053 Author: Dave Airlie <airlied (a] linux.ie> 27054 Date: Tue Feb 20 21:20:28 2007 +1100 27055 27056 radeon: move some i2c registers out into common register file 27057 27058 These regs are more useful than just multimedia 27059 27060 commit 7a25512415bb40b772491bb4a773dfe02ac8b71a 27061 Author: Alex Deucher <alex (a] botch2.com> 27062 Date: Fri Feb 16 00:44:11 2007 -0500 27063 27064 Add "DRI" option 27065 27066 This option allows you to disable the DRI per card. It also 27067 removes the "RN50Force3D" option as it is now covered by this 27068 option. RN50 users should set this to TRUE if they want to force 27069 the DRI on. 27070 27071 commit fa30ec6d5cd9bf4eb1a960592ca7311175219e4b 27072 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27073 Date: Tue Feb 13 15:35:32 2007 +0200 27074 27075 [mach64] PreInit: maxPitch, minor cosmetic. 27076 27077 commit 1777dcc1956d910073e13322767d9a3de41a949b 27078 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27079 Date: Tue Feb 13 15:03:40 2007 +0200 27080 27081 [mach64] PreInit: pitchInc is local. 27082 27083 commit 9cdf4fcc3a93a4e20244286392dc31a0b8e6a10a 27084 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27085 Date: Tue Feb 13 11:53:42 2007 +0200 27086 27087 [mach64] PreInit: Use goto bail. 27088 27089 Match ATILock - ATIUnlock. 27090 27091 commit 482b85144fee42b6543dcc97a75899e363e05e9e 27092 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27093 Date: Mon Feb 12 14:27:37 2007 +0200 27094 27095 [mach64] Enable RENDER acceleration on the Pro variants. 27096 27097 commit 63248f0b4308a4487cda3aa22daa36e3e0d38d14 27098 Author: Dave Airlie <airlied (a] linux.ie> 27099 Date: Mon Feb 12 19:37:36 2007 +1100 27100 27101 fix LVDS by moving bios reading around 27102 27103 commit 6a25f620d40cbb063de94aa6b5267296cd5a0670 27104 Author: David Airlie <airlied (a] asimov.stargames.com.au> 27105 Date: Mon Feb 12 15:59:43 2007 +1100 27106 27107 remove assignments to pOutput before it is created 27108 27109 commit 9234d8045c5fefbd1a781209409e55a13e3e5370 27110 Author: David Airlie <airlied (a] asimov.stargames.com.au> 27111 Date: Mon Feb 5 11:53:18 2007 +1100 27112 27113 remove all the locking glxgears runs fine over screen changes without it 27114 27115 commit eda9fb15e8bc042661a91a3d8c921006dfb3ddd9 27116 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27117 Date: Tue Nov 21 04:02:41 2006 +0200 27118 27119 Consolidate atixv.c w/ atimach64xv.c, part 2. 27120 27121 Move {Initialize,Close}XVideo from atixv.c to atimach64xv.c . 27122 27123 commit 67b240ddd4a6549a23885afc723b6efa7c52db4c 27124 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27125 Date: Tue Oct 31 16:44:10 2006 +0200 27126 27127 Consolidate atixv.c w/ atimach64xv.c, part 1. 27128 27129 Merge ATIXVInitializeAdaptor() in ATIMach64XVInitialiseAdaptor(). 27130 27131 commit 4dac7896549392f7378cad9620618cec28f41a16 27132 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27133 Date: Sun Oct 29 03:55:39 2006 +0200 27134 27135 Consolidate aticursor.c w/ atimach64cursor.c 27136 27137 commit 23b12b9ff55224e8c3c45eb58ea5fbf2f747d82e 27138 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27139 Date: Tue Oct 31 16:10:07 2006 +0200 27140 27141 Consolidate atiaccel.c w/ atimach64accel.c, part 2. 27142 27143 Merge ATIInitializeAcceleration() in ATIMach64AccelInit(). 27144 27145 commit bc3b222b57d7edf7bdce00a53e8abd9047c314a4 27146 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27147 Date: Sat Oct 28 19:29:12 2006 +0300 27148 27149 Consolidate atiaccel.c w/ atimach64accel.c, part 1. 27150 27151 Move ATIResizeOffscreenLinear() from atiaccel.c to atimach64xv.c and make it 27152 static. It is only used by XV memory management functions. 27153 27154 commit d876889add41c8052d4f3feaa21ddf384f5a7054 27155 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27156 Date: Tue Nov 21 04:47:06 2006 +0200 27157 27158 Drop non-PCI support from atividmem.c 27159 27160 - we require a PCI device (pVideo != NULL) 27161 - map the VGA framebuffer with xf86MapDomainMemory() 27162 27163 commit 2c8b33b761e4411451aea1eea3a89d629aeea40d 27164 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27165 Date: Tue Nov 21 04:51:02 2006 +0200 27166 27167 Minor refactoring of pATI->Block0Base computation. 27168 27169 commit a659b7235332e19659c56cc9d7af362bd8885a29 27170 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27171 Date: Sun Feb 4 11:28:54 2007 +0200 27172 27173 Cleanup checks for pATI->PCIInfo. 27174 27175 Fixes a bug where shadowfb would not be enabled by default in the non-accel 27176 case because pATI->PCIInfo is now set after ATIProcessOptions() is called. 27177 27178 commit 9d77aabdff919360f0c9333105436c31f1f5749a 27179 Merge: ff8ea19fc 57822be75 27180 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27181 Date: Mon Feb 5 19:16:51 2007 +0200 27182 27183 Merge branch 'mach64-pci-1' 27184 27185 commit d7317604c843e21c1df048a9253ed55a1957cccd 27186 Author: David Airlie <airlied (a] linux.ie> 27187 Date: Fri Feb 2 14:54:00 2007 +1100 27188 27189 get rid of mergetype and displaytype from radeon.h 27190 27191 commit 15c68ac19ff9971ae02cf52036ba36ddea1a6759 27192 Author: David Airlie <airlied (a] linux.ie> 27193 Date: Fri Feb 2 14:30:16 2007 +1100 27194 27195 rename some variables to output from connector 27196 27197 commit 72ef23ef7dd159d827e8e122482a58928d532268 27198 Author: David Airlie <airlied (a] linux.ie> 27199 Date: Fri Feb 2 14:26:29 2007 +1100 27200 27201 fixup mode probing a bit neater 27202 27203 commit 96acf6b2b242454345cc4b9cfc7ca07e0b597b43 27204 Author: David Airlie <airlied (a] linux.ie> 27205 Date: Fri Feb 2 14:26:19 2007 +1100 27206 27207 bring over setproperty from intel code 27208 27209 commit 76bc53f9b153880730ab61dcd2b6e4e7717e4058 27210 Author: David Airlie <airlied (a] linux.ie> 27211 Date: Fri Feb 2 14:15:03 2007 +1100 27212 27213 move i2c buses into outputs 27214 27215 commit 47fb9ce657f018177a35b449a4d716dc03be9327 27216 Author: David Airlie <airlied (a] linux.ie> 27217 Date: Fri Feb 2 11:45:40 2007 +1100 27218 27219 fixup callers to crtcsetmode fixes my cursors 27220 27221 commit 720730b1b0de632488d3b9818210ec5e9c7f07ae 27222 Author: David Airlie <airlied (a] linux.ie> 27223 Date: Thu Feb 1 16:43:38 2007 +1100 27224 27225 add locks for crtcs and some missing lines from intel update 27226 27227 commit 6748732658850ea506f623a3622aa7135513ffd0 27228 Author: David Airlie <airlied (a] linux.ie> 27229 Date: Thu Feb 1 16:26:31 2007 +1100 27230 27231 update to latest intel codebase modulo using their mode set 27232 27233 and it breaks my cursor 27234 27235 commit a77f08298dc7e097025e3f7f92e3665c0ef30095 27236 Author: David Airlie <airlied (a] linux.ie> 27237 Date: Mon Jan 29 16:12:24 2007 +1100 27238 27239 radeon: always reset cursor regs 27240 27241 commit ff8ea19fcdce099732f9359e53cd62b9a04bfa6d 27242 Author: Alex Deucher <alex (a] botch2.hsd1.va.comcast.net> 27243 Date: Tue Jan 30 21:11:09 2007 -0500 27244 27245 add option to force the DRI on for RN50 chips 27246 27247 Allow user to force the DRI on for RN50 chips. 27248 3D is not guaranteed to work on these chips, 27249 however in some cases it does. 27250 fixes bug 9802. 27251 27252 commit 57822be75740f339445f2375d44632560f4bbe57 27253 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27254 Date: Tue Nov 21 03:08:26 2006 +0200 27255 27256 Drop pATI->Chipset, no longer used. 27257 27258 commit 8b7c88e9ed549777413136c0f2ccc61747b010d8 27259 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27260 Date: Thu Dec 28 14:46:10 2006 +0200 27261 27262 Add option "probe_sparse" to force probing sparse I/O bases. 27263 27264 commit 3654b786f1ad1f4e6bfb2fff5ff057254958f7c0 27265 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27266 Date: Wed Dec 27 17:59:14 2006 +0200 27267 27268 Sane probe, based on work by Luc Verhaegen <libv (a] skynet.be>. 27269 27270 commit 0117c9bde0d0bbb487bc259cf55e8066ebd63f6a 27271 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27272 Date: Wed Dec 27 17:49:45 2006 +0200 27273 27274 Drop checking of sparse I/O bases conflicts 27275 27276 This will be handled with an option somehow. 27277 27278 commit 87cf2f1874ca8a7cc492d2e2b6918ba69a8aa713 27279 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27280 Date: Mon Aug 14 17:55:14 2006 +0300 27281 27282 Minor refactoring of module (un)loading for TV_OUT. 27283 27284 commit 0b31333c9147af21c50493872779da12a3668e5d 27285 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27286 Date: Mon Nov 20 22:08:49 2006 +0200 27287 27288 Drop support for non-programmable clocks. 27289 27290 From README.ati: 27291 27292 Clocks for supported programmable clock generators: 27293 27294 The driver currently supports all programmable clock generators known 27295 to exist on Mach64 adapters. 27296 27297 Clocks for unsupported programmable clock generators: 27298 27299 This case is unlikely to occur, but is documented for the sake of 27300 completeness. 27301 27302 Thus: 27303 27304 - check for (pATI->ProgrammableClock > ATI_CLOCK_FIXED) && 27305 (pATI->ProgrammableClock < ATI_CLOCK_MAX) 27306 27307 - drop "probe_clocks" option 27308 27309 - pATIHW->ClockUnmap is no longer used 27310 27311 - pATIHW->ClockMap is only used with NewHW.crtc which is always ATI_CRTC_MACH64 27312 and has the identity map, so drop it 27313 27314 - (pATI->ProgrammableClock != ATI_CLOCK_INTERNAL) => (pATI->depth <= 8) 27315 27316 commit c9a9ea5ebd1c27881c1fac19842cfa3af55c126e 27317 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27318 Date: Mon Nov 20 15:09:41 2006 +0200 27319 27320 Banked memory is no longer needed. 27321 27322 - (pATI->BankInfo.BankSize = 0) in all cases, cull pATI->BankInfo 27323 27324 - only keep the minimal pATIHW.SetBank interface for save/restore 27325 27326 - clean ATISwap() a little, (NewHW.crtc != ATI_CRTC_VGA) 27327 27328 - (UseSmallApertures == TRUE) <=> pATI->VGAAdapter 27329 27330 commit 383e963e275d351ea3631c352f5795340162d69f 27331 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27332 Date: Sun Jan 28 00:44:31 2007 +0200 27333 27334 Always use the linear aperture. 27335 27336 - drop (pATI->OptionLinear == FALSE) 27337 27338 - AcceleratorVideoRAM is always set, i.e. VGAVideoRAM is not used 27339 27340 - pATI->LinearBase is always set 27341 27342 - xf86LinearVidMem() is now checked in atipreinit() for both CPIO and MMIO 27343 27344 commit 9b126f45db27496c1ad16db65a61fe641a018983 27345 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27346 Date: Mon Nov 20 20:38:59 2006 +0200 27347 27348 We always use the accelerator CRTC within the xserver. 27349 27350 - cull (pATI->NewHW.crtc != ATI_CRTC_MACH64). 27351 27352 commit 0abcb2b2c84a50e64bb6aa9e94760659ecf33add 27353 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27354 Date: Sun Jan 28 00:19:15 2007 +0200 27355 27356 Clean pATI->Adapter and pATI->VGAAdapter. 27357 27358 - cull (pATI->Adapter != ATI_ADAPTER_MACH64) 27359 27360 - treat pATI->VGAAdapter as Bool 27361 27362 commit 817b4b5338dba4d8eb862f1f79ef0edb75aad7f4 27363 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27364 Date: Wed Dec 27 21:16:18 2006 +0200 27365 27366 Keep PCI mach64 only, drop: 27367 27368 - Chip < ATI_CHIP_88800GXC 27369 - Chipset != ATI_CHIPSET_ATI 27370 - Adapter != ATI_ADAPTER_MACH64 27371 27372 - depth < 8 27373 27374 commit d350860e29f043e98bfb1da74b26280f1755ab6f 27375 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27376 Date: Wed Dec 27 20:56:45 2006 +0200 27377 27378 Keep PCI mach64 only (atipreinit.c, atimode.c), drop: 27379 27380 - Chip < ATI_CHIP_88800GXC 27381 - Chipset != ATI_CHIPSET_ATI 27382 - Adapter != ATI_ADAPTER_MACH64 27383 27384 - depth < 8 27385 27386 atimode.c only: 27387 27388 - NewHW.crtc != ATI_CRTC_MACH64 27389 27390 This allows to drop VGACalculate(), VGAWonderCalculate() cruft early. 27391 27392 commit 5fa77f2d122e7267911a15235338d8d3f9eece2e 27393 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27394 Date: Wed Dec 27 14:50:48 2006 +0200 27395 27396 Drop probing for VGA, VGAWonder, 8514, Mach32. 27397 27398 - ChipHasSUBSYS_CNTL 27399 - Coprocessor 27400 - SharedAccelerator 27401 - SharedVGA <=> (VGAAdapter != ATI_ADAPTER_NONE) 27402 27403 commit 1b2a9e1735da63bbb8cb41939668f789cee22579 27404 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27405 Date: Mon Nov 20 02:35:56 2006 +0200 27406 27407 Drop AVOID_NON_PCI. 27408 27409 This was not set anyway, because configure.ac would compute ATIMISC_NON_PCI and 27410 then test ATI_AVOID_NON_PCI to set AVOID_NON_PCI... 27411 27412 commit 2fb84dd2fc1a663c7b9d77e706189439ffe08381 27413 Author: Alex Deucher <alex (a] botch2.hsd1.va.comcast.net> 27414 Date: Thu Jan 25 02:01:42 2007 -0500 27415 27416 I think we can say r300 support is not experimental 27417 27418 commit e155659f4b006edff31f1fda88b32bf608dbeece 27419 Author: Alex Deucher <alex (a] botch2.hsd1.va.comcast.net> 27420 Date: Thu Jan 25 01:59:27 2007 -0500 27421 27422 update man page. fixes bug 6457 27423 27424 Most cards support 2D and 3D now 27425 27426 commit fa12fe1cc90dd745f3eea35a07d4f3efd652e7b4 27427 Author: Dave Airlie <airlied (a] linux.ie> 27428 Date: Tue Jan 23 19:37:51 2007 +1100 27429 27430 remove unused restore palette function 27431 27432 commit 4d3649b84a3325043c38cece4499de0095cebd71 27433 Author: Dave Airlie <airlied (a] linux.ie> 27434 Date: Tue Jan 23 19:37:21 2007 +1100 27435 27436 remove isused and isactive 27437 27438 commit 9149e763865598c307cbefc753ff6ebdeeaf32ae 27439 Author: David Airlie <airlied (a] asimov.stargames.com.au> 27440 Date: Tue Jan 23 15:20:11 2007 +1100 27441 27442 radeon: fixup problem with cursor not going off 27443 27444 Also fixup secondary dac detect to return unknown for now 27445 27446 commit 2a13a3b641d9acf4f50472e1c8ba07633c3b78d6 27447 Author: David Airlie <airlied (a] asimov.stargames.com.au> 27448 Date: Tue Jan 23 10:36:06 2007 +1100 27449 27450 fixup init disp bandwidth 27451 27452 This probably break old-style dualhead badly 27453 27454 commit cbd84bed13582e82f8b2e84aa152602474c09cd4 27455 Author: Dave Airlie <airlied (a] linux.ie> 27456 Date: Mon Jan 22 22:19:46 2007 +1100 27457 27458 removed even more mergedfb remains 27459 27460 commit 6296882135b183425a219efb75374d3b172de370 27461 Author: Dave Airlie <airlied (a] linux.ie> 27462 Date: Mon Jan 22 22:14:48 2007 +1100 27463 27464 remove more mergedfb relics 27465 27466 commit 26e1bab2c7a4ad52710746265b3da495b54dab36 27467 Author: Dave Airlie <airlied (a] linux.ie> 27468 Date: Mon Jan 22 22:11:55 2007 +1100 27469 27470 radeon: fixup some of the video code after randr 27471 27472 This may not be complete, I need to test on dual-head system later 27473 27474 commit 9008e1caa45e4a18e6f3289c7b17036730fe578a 27475 Author: Dave Airlie <airlied (a] linux.ie> 27476 Date: Mon Jan 22 21:16:47 2007 +1100 27477 27478 remove unused merged code 27479 27480 commit e29a32d711553fcb4ca9928122ac285fe0b0c1a7 27481 Author: Dave Airlie <airlied (a] linux.ie> 27482 Date: Mon Jan 22 21:09:53 2007 +1100 27483 27484 add back overlay CRTC chooser this may not work yet 27485 27486 commit aa9d04ba94a3cd60b248231da517e2817591fc69 27487 Author: Dave Airlie <airlied (a] linux.ie> 27488 Date: Sun Jan 21 20:29:56 2007 +1100 27489 27490 remove alot more mergedfb support 27491 27492 commit e8be0056e8ff666f63a294691661f5dab289203e 27493 Author: Dave Airlie <airlied (a] linux.ie> 27494 Date: Thu Jan 18 16:41:25 2007 +1100 27495 27496 remove mergedfb 27497 27498 commit a648050a3cc60f92b1ca0b3d707aadf93d076d91 27499 Author: Dave Airlie <airlied (a] linux.ie> 27500 Date: Wed Jan 17 17:28:07 2007 +1100 27501 27502 move some cursor code around for show/hide 27503 27504 commit d2ecfb507282726122bb8b0d17fd3637d0ae7d46 27505 Author: Dave Airlie <airlied (a] linux.ie> 27506 Date: Wed Jan 17 17:16:27 2007 +1100 27507 27508 make dri work again 27509 27510 commit 369f7c85ceff983defb7657b80ec9cd3e5440b07 27511 Author: Dave Airlie <airlied (a] linux.ie> 27512 Date: Wed Jan 17 11:14:09 2007 +1100 27513 27514 comment out unblanking on VT switch 27515 27516 commit bdb66a2042f02c4b57bd3c0181a00b39fcbdb232 27517 Author: Dave Airlie <airlied (a] linux.ie> 27518 Date: Sun Jan 14 21:07:05 2007 +1100 27519 27520 radeon: get randr-1.2 mode switching mostly working on my laptop 27521 27522 The main change is to fix the dac macro + cntl register writes for rv350 and 27523 above, this still has a problem with resetting the same mode after connect 27524 disconnect cycle, need to talk to keithp 27525 27526 commit aa0e7337815d4daca4df0671768621330b759011 27527 Author: Dave Airlie <airlied (a] linux.ie> 27528 Date: Sun Jan 14 19:17:32 2007 +1100 27529 27530 use more randr-ish names for variables 27531 27532 commit 9d37f23aea43b74a7ec640e6b03617c8392e2572 27533 Author: Dave Airlie <airlied (a] linux.ie> 27534 Date: Sun Jan 14 19:10:11 2007 +1100 27535 27536 radeon: move blank/unblank to use randr 27537 27538 commit 6bd4fe42789c38e7e804826715214ce6badcca6a 27539 Author: Dave Airlie <airlied (a] linux.ie> 27540 Date: Sun Jan 14 18:38:59 2007 +1100 27541 27542 disable dpms on/off as it was writing regs before we wanted to 27543 27544 commit 3949288ed26a91e180b178f4796f2f7e1bdc8ed6 27545 Author: Dave Airlie <airlied (a] linux.ie> 27546 Date: Sun Jan 14 16:26:39 2007 +1100 27547 27548 radeon: destroy and fixup LVDS crtc 27549 27550 commit f5ffb9fb3bdd7cec020240c5831ae92ee744972c 27551 Author: Timo Jyrinki <timo.jyrinki (a] hut.fi> 27552 Date: Fri Jan 12 23:08:50 2007 +1100 27553 27554 ati: fix detection of some newer radeons using ati wrapper 27555 27556 This is mainly due to the cards having a different resource 1. 27557 Fixes 6796 27558 27559 commit 1e690c57ad1dfaeb760ef6e799faa6398e9a0551 27560 Author: Dave Airlie <airlied (a] linux.ie> 27561 Date: Fri Jan 12 23:03:10 2007 +1100 27562 27563 always create a second CRTC I think this should solve any problems 27564 27565 commit 25b36a4c56422ad8e25a4a1c55055c2f062213ca 27566 Author: Dave Airlie <airlied (a] linux.ie> 27567 Date: Fri Jan 5 16:13:16 2007 +1100 27568 27569 minor randr1.2 fixups 27570 27571 commit abe8791e4f9fa3e88273897b351387cd33822734 27572 Author: Dave Airlie <airlied (a] linux.ie> 27573 Date: Fri Jan 5 15:09:43 2007 +1100 27574 27575 hook up DPMS through xf86 27576 27577 commit 8a9c68c2234b91ed38555f892afdad30b9e4b455 27578 Author: Dave Airlie <airlied (a] linux.ie> 27579 Date: Fri Jan 5 15:09:15 2007 +1100 27580 27581 set binding for new heads 27582 27583 commit d6a2b8aeb5a97ee907fd45b574bc6e4ab4b3aede 27584 Author: Dave Airlie <airlied (a] linux.ie> 27585 Date: Fri Jan 5 09:33:54 2007 +1100 27586 27587 fix LUT 27588 27589 commit ef1d36e56dec1fec37cee2dfd9cb5bf8ce2c485c 27590 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 27591 Date: Wed Jan 3 15:56:23 2007 +0100 27592 27593 fix uninitialized mode regs (bug 9495) 27594 27595 Since the reorganization of the mode setting code, the mode registers relying 27596 on state already set (by bios) were not read, thus clearing out all bits the 27597 driver does not touch. At the very least, this could lead to completely 27598 nonfunctional to misbehaving dvi output (see bug 9495). Fix this by using the 27599 SavedReg values, which also makes it more obvious that those are bits which 27600 were not set by the driver previously, but come from register readback. 27601 (cherry picked from 9506f7015a1e442f4ca0bd3bfae555ec7e8a5f37 commit) 27602 27603 commit 55aa832157bdebcba2d58896777942d108c352b0 27604 Author: Dave Airlie <airlied (a] linux.ie> 27605 Date: Thu Jan 4 17:57:31 2007 +1100 27606 27607 fix cursor handling 27608 27609 commit 4f8a7cafdc77e98dc44f9eced876560b1ee01117 27610 Author: Dave Airlie <airlied (a] linux.ie> 27611 Date: Thu Jan 4 16:23:39 2007 +1100 27612 27613 hook up crtc modesetting 27614 27615 commit 52f749c8a613ee316044abe82156ee270412ced8 27616 Author: Dave Airlie <airlied (a] linux.ie> 27617 Date: Thu Jan 4 16:23:26 2007 +1100 27618 27619 fix typo 27620 27621 commit d7ff61c6822cbede7f5b59b411048d33dbae9ee4 27622 Author: Dave Airlie <airlied (a] linux.ie> 27623 Date: Thu Jan 4 13:38:26 2007 +1100 27624 27625 hook up detection of output 27626 27627 commit 50439d7e9f4f6f9933bacd59f8bb3e655a35dbc2 27628 Author: Dave Airlie <airlied (a] linux.ie> 27629 Date: Thu Jan 4 13:21:57 2007 +1100 27630 27631 fixup some function calls 27632 27633 commit e067bfee92f2b0877108355619b2fb9188a9d15a 27634 Author: Dave Airlie <airlied (a] linux.ie> 27635 Date: Thu Jan 4 13:21:45 2007 +1100 27636 27637 add more files for back compat 27638 27639 commit f1f34627ffbe2136ac3e023c01c0430412919ded 27640 Author: Dave Airlie <airlied (a] linux.ie> 27641 Date: Thu Jan 4 12:50:03 2007 +1100 27642 27643 make restore mode non-static 27644 27645 commit 249ec67c296b34d0c7c1d78602628d2f7ce6a045 27646 Author: Dave Airlie <airlied (a] linux.ie> 27647 Date: Thu Jan 4 12:43:03 2007 +1100 27648 27649 more minor fixes to get radeon up to speed 27650 27651 commit a43c1d55f5f855d9e6ae939dd4eec1c607b6d514 27652 Author: Dave Airlie <airlied (a] linux.ie> 27653 Date: Thu Jan 4 12:29:51 2007 +1100 27654 27655 bring radeon randr code inline with intel randr code 27656 27657 commit c54a31565e3ea50f1af7b81b1bf0c0e437099f47 27658 Author: Dave Airlie <airlied (a] linux.ie> 27659 Date: Thu Jan 4 10:39:01 2007 +1100 27660 27661 radeon: fixup fastwrites to be always off 27662 27663 My 8500 in i845 doesn't work with fastwrites even setup by the firmware. 27664 27665 commit dda2bbfdc2e0da6e8b9cb8d1bf78b91f1bb8131f 27666 Author: Dave Airlie <airlied (a] linux.ie> 27667 Date: Thu Jan 4 10:29:54 2007 +1100 27668 27669 radeon: don't touch Controller[1] on single CRTC cards. 27670 27671 This might fix bug 9371 27672 27673 commit 9506f7015a1e442f4ca0bd3bfae555ec7e8a5f37 27674 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 27675 Date: Wed Jan 3 15:56:23 2007 +0100 27676 27677 fix uninitialized mode regs (bug 9495) 27678 27679 Since the reorganization of the mode setting code, the mode registers relying 27680 on state already set (by bios) were not read, thus clearing out all bits the 27681 driver does not touch. At the very least, this could lead to completely 27682 nonfunctional to misbehaving dvi output (see bug 9495). Fix this by using the 27683 SavedReg values, which also makes it more obvious that those are bits which 27684 were not set by the driver previously, but come from register readback. 27685 27686 commit d5c6306385cbc7b13a132ae72ed04e04e24d9886 27687 Author: George Sapountzis <gsap7 (a] yahoo.gr> 27688 Date: Tue Dec 19 09:45:24 2006 +0200 27689 27690 Use portable XORG_RELEASE_VERSION macro. 27691 27692 commit 97df0a32b14e34f22dce9cd3b375c1ac5fb1c5eb 27693 Author: Michel Dnzer <michel (a] tungstengraphics.com> 27694 Date: Tue Dec 19 11:28:21 2006 +0100 27695 27696 radeon: Advertise 2D engine limits to EXA when we're not using the 3D engine. 27697 27698 Also round up to the maximum width and height, as that's what EXA compares. 27699 27700 commit 295823d0879a5b574bb79843a6acd43adb9259e5 27701 Author: Michel Dnzer <michel (a] tungstengraphics.com> 27702 Date: Tue Dec 19 11:19:20 2006 +0100 27703 27704 radeon: Default to leaving AGP transfer mode and fast writes unchanged. 27705 27706 Based on the assumption that firmware should have set up the card and host 27707 bridge appropriately for these settings, this may actually be safer, at least 27708 for the transfer rate; leaving fast writes enabled is hopefully safe as well, 27709 it certainly is on my sytem. 27710 27711 See https://bugs.freedesktop.org/show_bug.cgi?id=9284 . 27712 27713 commit 81f3b4070b70483d6da4f7587e52a64ac69c8ca0 27714 Author: Eamon Walsh <ewalsh (a] tycho.nsa.gov> 27715 Date: Mon Dec 18 14:06:15 2006 -0500 27716 27717 Revert "Convert callers of LookupWindow() to dixLookupWindow()." 27718 27719 This reverts commit 48ff33a1770f3684cd50184db8f1944a456d9974. 27720 27721 commit 87592ffb717da1f0a1767a38918d16d60953599c 27722 Author: Dave Airlie <airlied (a] linux.ie> 27723 Date: Sun Dec 17 20:10:58 2006 +1100 27724 27725 add setdpi call 27726 27727 commit 41cd50487bc85708a3f791dfa70bf85223d91a65 27728 Author: Dave Airlie <airlied (a] linux.ie> 27729 Date: Sun Dec 17 20:08:07 2006 +1100 27730 27731 add displayWidth set 27732 27733 commit 555e0f988c571aa47bc62ec6d9612a71f1a1f59b 27734 Author: Dave Airlie <airlied (a] linux.ie> 27735 Date: Sun Dec 17 19:59:19 2006 +1100 27736 27737 add randr pre init 27738 27739 commit 3e2c72d7ac80ed874bcf2887a5253dac1c9bb02c 27740 Author: Dave Airlie <airlied (a] linux.ie> 27741 Date: Sun Dec 17 19:49:08 2006 +1100 27742 27743 fixup even more remnants of old code 27744 27745 commit 4356b031886e00b3ed5ac1dbceeadebd7fc29fc2 27746 Author: Dave Airlie <airlied (a] linux.ie> 27747 Date: Sun Dec 17 19:47:45 2006 +1100 27748 27749 fix remnant of old code 27750 27751 commit 17d39502a8dbea395051007aebffdf23644e9ee1 27752 Author: Dave Airlie <airlied (a] linux.ie> 27753 Date: Sun Dec 17 19:44:39 2006 +1100 27754 27755 use getddcmodes properly 27756 27757 commit bf0b364cbd8682f297e2d110ece1f72e7151340d 27758 Author: Dave Airlie <airlied (a] linux.ie> 27759 Date: Sun Dec 17 17:31:08 2006 +1100 27760 27761 fix up output namings 27762 27763 commit 4962a5430844114de864418d803f3182b90792ff 27764 Author: Dave Airlie <airlied (a] linux.ie> 27765 Date: Sun Dec 17 16:58:07 2006 +1100 27766 27767 remove some code that moved into server 27768 27769 commit b47013f41249516ae82e22bd75caf573da2cc13c 27770 Author: Dave Airlie <airlied (a] linux.ie> 27771 Date: Sat Dec 16 12:02:37 2006 +1100 27772 27773 hook up mode detect to new randr code 27774 27775 commit 48ff33a1770f3684cd50184db8f1944a456d9974 27776 Author: Eamon Walsh <ewalsh (a] tycho.nsa.gov> 27777 Date: Fri Dec 15 17:43:39 2006 -0500 27778 27779 Convert callers of LookupWindow() to dixLookupWindow(). 27780 27781 commit 62a5e7565b66d7b7d65609c034f34b55bd266617 27782 Author: Dave Airlie <airlied (a] linux.ie> 27783 Date: Fri Dec 15 19:48:42 2006 +1100 27784 27785 bring over latest code from Intel git repo. 27786 27787 this code doesn't work yet just realigns with intel work 27788 27789 commit 29124400c7f193317d41d8cfd748371a239cfea1 27790 Author: Dave Airlie <airlied (a] linux.ie> 27791 Date: Sun Dec 10 18:00:17 2006 +1100 27792 27793 hook up randr crtc setting 27794 27795 commit 120c854f185c1e7711cf0dee19303fdb8894d49d 27796 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 27797 Date: Fri Dec 8 02:51:52 2006 +0100 27798 27799 radeon: avoid unnecessary OUTPLL/INPLL calls when displaying video 27800 27801 It is not necessary to always emit a OUTPLL/INPLL pair when we display 27802 a video frame. On some chips there are erratas for which the workarounds 27803 cause a 10ms delay by those calls. This is related to #5876 though those 27804 affected may suffer from other slowness issues too. 27805 27806 commit ccd37b3e88cdcfe03b7c707b6082339eb8f11e6b 27807 Author: Michel Dnzer <michel (a] tungstengraphics.com> 27808 Date: Thu Dec 7 18:41:48 2006 +0100 27809 27810 radeon: Move Xv option handling code into RADEONPreInitXv(). 27811 27812 This unclutters RADEONPreInit() somewhat, but more importantly moves comparison 27813 against info->ChipFamily after that's initialized. 27814 27815 commit 02be8c04a268eaa71bdfe891331758715fef4e71 27816 Author: Michel Dnzer <michel (a] tungstengraphics.com> 27817 Date: Thu Dec 7 18:31:58 2006 +0100 27818 27819 radeon: Make cache flushing for R300 series less invasive. 27820 27821 Instead of calling the DRM CP idle ioctl, just emit the cache flush commands 27822 into the CP stream. 27823 27824 commit be46f7b6f7a729695a0affe2328077666283af12 27825 Author: Michel Dnzer <michel (a] tungstengraphics.com> 27826 Date: Thu Dec 7 16:27:10 2006 +0100 27827 27828 Fix radeon compile warnings. 27829 27830 commit 59c5b4f69c97fb233144677ab18095f88aed006b 27831 Author: Daniel Stone <daniel (a] fooishbar.org> 27832 Date: Wed Dec 6 18:48:55 2006 +0200 27833 27834 Makefile.am: re-add ChangeLog to CLEANFILES 27835 27836 commit 7611f14dc697b75cd7bb6493bb65b3a7ead379b8 27837 Author: Daniel Stone <daniel (a] fooishbar.org> 27838 Date: Wed Dec 6 18:47:33 2006 +0200 27839 27840 Makefile.am: fix changelog hook 27841 Make changelog hook safe for all situations. 27842 27843 commit f2a9d2d61addc4e3d5e128d8078a04c9a136be6a 27844 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 27845 Date: Wed Dec 6 00:16:41 2006 +0100 27846 27847 radeon: fix up packed yuv broken by planar yuv patch 27848 27849 fix the forgotten leftuv value for packed yuv which is 27850 needed to get correct uv starting pixel (fixes broken clipping / 27851 non-null src start pixel as tvtime uses) 27852 27853 commit d2aa57a4bdd4a0deb2c659bb07f3f57d1116f0d1 27854 Author: Roland Scheidegger <sroland (a] tungstengraphics.com> 27855 Date: Tue Dec 5 01:30:13 2006 +0100 27856 27857 radeon: use overlay scaler native planar yuv capability. 27858 27859 Radeons can do planar yuv just fine, there is no need to convert all data 27860 to packed yuv manually. This saves some cpu cycles as well as some 27861 (graphic card) ram. 27862 27863 commit 5d5fa1b86e5179b061f0db47fe0227d1b84c37f8 27864 Author: Dave Airlie <airlied (a] linux.ie> 27865 Date: Mon Dec 4 18:53:33 2006 +1100 27866 27867 update radeon driver to Intel driver xf86Crtc interfaces 27868 27869 commit 51d1cf19e71dd5de47f2c6467f4a1685eefd9e1e 27870 Author: Dave Airlie <airlied (a] linux.ie> 27871 Date: Sun Dec 3 16:30:01 2006 +1100 27872 27873 Add radeon randr 1.2 initial attempt at support 27874 27875 This doesn't do a huge amount yet 27876 27877 commit eb17c9aed2144701ad7bd1042b2905446e4d708a 27878 Author: airlied <airlied (a] optimus.localdomain> 27879 Date: Sat Nov 18 15:48:02 2006 +1100 27880 27881 move radeon over to not using X's mode management like the intel driver. 27882 27883 this builds and at least seems to pick a mode.. might be near to hooking 27884 up randr1.2 soon 27885 27886 commit 941b5120916ebff69ec0ee4ca7a46105d5306e5d 27887 Author: airlied <airlied (a] optimus.localdomain> 27888 Date: Sat Nov 18 10:41:12 2006 +1100 27889 27890 make radeon modes use generic code 27891 27892 commit 608427283ea59cdfe3d85e1cf32719bc3809dc16 27893 Author: airlied <airlied (a] optimus.localdomain> 27894 Date: Sat Nov 18 10:35:31 2006 +1100 27895 27896 add more functions from i830 driver 27897 27898 This adds more generic functions from the i830 driver to the radeon 27899 27900 commit d5b8cafc185b7d9fb909cb18a08615f81d89eaf8 27901 Author: airlied <airlied (a] optimus.localdomain> 27902 Date: Sat Nov 18 10:22:38 2006 +1100 27903 27904 add generic mode handling code from Intel driver 27905 27906 commit 9cfa82e1670ad85746926995972a535ddf03ee07 27907 Author: Dave Airlie <airlied (a] linux.ie> 27908 Date: Sun Dec 3 15:56:38 2006 +1100 27909 27910 move monitor detection into single function 27911 27912 commit 032a00d155ecdcdcab3ca0d047f552fe6b338a97 27913 Author: airlied <airlied (a] optimus.localdomain> 27914 Date: Sat Nov 18 10:23:12 2006 +1100 27915 27916 make DDC function more generic 27917 27918 commit 2632ae24b3b50c69567e92611ce7c2cd2ee98baa 27919 Author: Dave Airlie <airlied (a] linux.ie> 27920 Date: Sun Dec 3 15:22:38 2006 +1100 27921 27922 re-program bios 4 scratch so we know if the LCD panel is there post startup 27923 27924 commit ddec94e8f2f5746bad0e998d15303e771803d22b 27925 Author: Dave Airlie <airlied (a] linux.ie> 27926 Date: Sun Dec 3 11:07:23 2006 +1100 27927 27928 move crtc to connector mapping to a connector to crtc mapping. 27929 27930 Still not allowing any clone modes, but heading in the correct direction 27931 I hope... there is a chance this will regress something from superpatch.. 27932 27933 commit 218cb7f8db35d0ce8f9fc4ece72bd5ec3f52106f 27934 Author: Dave Airlie <airlied (a] linux.ie> 27935 Date: Sun Dec 3 09:53:35 2006 +1100 27936 27937 radeon: dynamically allocate connectors 27938 27939 commit f22fd02bc7f7ef67e33c9bd93eae2d7488669b5e 27940 Author: Dave Airlie <airlied (a] linux.ie> 27941 Date: Sat Dec 2 19:48:00 2006 +1100 27942 27943 make CRTCs malloc'ed rather than part of Ent 27944 27945 This is leading towards randr-1.2 believe me :-) 27946 27947 commit 13efd1f75fbd29a4d1f69d7d5a5bb2b8d90ad820 27948 Author: Dave Airlie <airlied (a] linux.ie> 27949 Date: Sat Dec 2 19:16:41 2006 +1100 27950 27951 move HasCRTC2 to pRADEONEnt from info 27952 27953 commit 3933b86390eb2f4198d490acd4914fb470c40f47 27954 Author: Michel Dnzer <michel (a] tungstengraphics.com> 27955 Date: Fri Dec 1 18:00:17 2006 +0100 27956 27957 Make log file output clear about origin of used overlay scaler buffer width. 27958 27959 commit 563dfb52f5b24d2f765fae0394e599c11dcf0b5b 27960 Author: Michel Dnzer <michel (a] tungstengraphics.com> 27961 Date: Fri Dec 1 17:56:19 2006 +0100 27962 27963 Increase default overlay scaler buffer width for RV350. 27964 27965 Works with 1920x1080 video on my M10. 27966 27967 commit 984ea3d9b7c27d8183ca94ad9d7905f03613fe10 27968 Author: Michel Dnzer <michel (a] tungstengraphics.com> 27969 Date: Fri Dec 1 17:42:23 2006 +0100 27970 27971 Fix build warning. 27972 27973 commit 9f5ea3981449f29ff204eb154166e8fc813205fa 27974 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 27975 Date: Wed Nov 29 16:46:57 2006 +0100 27976 27977 radeon: bug #1462, predownscale to make HD video work. 27978 27979 Use the overlay scaler's predownscale capability to make videos with large 27980 horizontal resolution work if it exceeds the scaler buffer width. Make the 27981 scaler buffer width user-configurable since we don't know it for all chips, 27982 and using predownscaling may otherwise reduce quality even if it wouldn't 27983 be needed. This should fix bug #1462. 27984 27985 commit 35a5eaf5adf103ef57f8d4590dae25b0c50bd780 27986 Author: airlied <airlied (a] optimus.localdomain> 27987 Date: Tue Nov 28 17:31:05 2006 +1100 27988 27989 fixup EXA after previous commit 27990 27991 commit ba047ae863aac869595a31ea94f102084636fd29 27992 Author: airlied <airlied (a] optimus.localdomain> 27993 Date: Tue Nov 28 17:26:10 2006 +1100 27994 27995 radeon: initial move framebuffer away from zero prep. 27996 27997 In theory the driver should be able to handle the front buffer not at VRAM 0 27998 In practice it didn't.. this is cleanup for at least XAA parts of the driver 27999 to allow for the frontbuffer to move. It has to re-organise a large part of ScreenInit so things happen in the correct order otherwise some things get setup in-correctly. (not sure EXA with fb not at 0 works yet...) 28000 28001 commit 0d42cab2d889070ab6de05990e2b19724595f7af 28002 Author: Alex Deucher <alex@t41p.(none)> 28003 Date: Wed Nov 22 13:44:49 2006 -0500 28004 28005 clean up previous VIP commit. 28006 28007 commit 8e0f57571228729cc40ed59266484ecc43819f44 28008 Author: Alex Deucher <alex@t41p.(none)> 28009 Date: Wed Nov 22 13:38:01 2006 -0500 28010 28011 Don't mess with the VIP bus on mobility chips. Seems to cause fan 28012 problems on asus laptops (see bug 7463). We don't support video-in 28013 on any laptops at the moment anyway. 28014 28015 commit e4850610596a08082a528c38186268b72012e78c 28016 Author: Alain Pteut <alain.peteut (a] gmx.ch> 28017 Date: Wed Nov 22 12:54:55 2006 -0500 28018 28019 enable i2c for PCI_CHIP_RV250_If. allows AIW fuctionality to work. 28020 patch and testing from Alain Pteut 28021 28022 commit ff38c2c9cf0c6acbdaf53b681ec721487cf411f8 28023 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28024 Date: Sun Nov 19 17:11:46 2006 +0100 28025 28026 radeon: Bug #8989: Fix page flipping artifacts with XAA. 28027 28028 We still have to force-sync the pages on enabling page flipping with XAA as the 28029 second page may have been clobbered by the offscreen pixmap cache since they 28030 were last synced. 28031 28032 commit 2dd0892817090709b2fb7ffba920595ada00fcad 28033 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28034 Date: Sun Nov 19 17:09:11 2006 +0100 28035 28036 Fix compile warnings. 28037 28038 commit 8c1b9269aaac746989bdb9be02915722bcf6fec0 28039 Author: airlied <airlied (a] optimus.localdomain> 28040 Date: Fri Nov 17 18:47:50 2006 +1100 28041 28042 forgot SetPitch function is called 28043 28044 commit 3a0338bb487739a1740ddfb4d36d53cfa7f798c6 28045 Author: airlied <airlied (a] optimus.localdomain> 28046 Date: Fri Nov 17 18:37:00 2006 +1100 28047 28048 split mode handling code into separate modes source file 28049 28050 commit 13e0a95564220d42ad8060b43b0aaaba3002157a 28051 Author: David Airlie <airlied (a] asimov.stargames.com.au> 28052 Date: Mon Nov 13 09:35:18 2006 +1100 28053 28054 fix up issue with bindings and mergedfb 28055 28056 commit 63444770ff30afeb134befed5bb3a5c28b7bc4e9 28057 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28058 Date: Thu Nov 9 22:01:28 2006 +0200 28059 28060 [mach64] Prefer double-buffered visuals for GLX. 28061 28062 So that 3D apps apropriately fail in DRI... 28063 28064 commit 0411f68d71c75cda38ad46ad5c0f5cc3cda64464 28065 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28066 Date: Thu Nov 9 22:00:56 2006 +0200 28067 28068 [mach64] Use asynchronous DMA mode by default. 28069 28070 commit 8dd668db05e8f1282fa0e29a91e38c7cbdc97aa1 28071 Author: Dave Airlie <airlied (a] linux.ie> 28072 Date: Sun Nov 12 16:33:21 2006 +1100 28073 28074 radeon: abstract init display bandwidth from mergedfb 28075 28076 commit 45a9f434b77ad892efcdba267a935135b972dd42 28077 Author: Dave Airlie <airlied (a] linux.ie> 28078 Date: Sun Nov 12 16:12:56 2006 +1100 28079 28080 add radeon Init2 function, to abstract some mergedfb mode setting 28081 28082 commit d0828804aec0852afaca44ee41c4792edc40dbb4 28083 Author: Dave Airlie <airlied (a] linux.ie> 28084 Date: Sun Nov 12 15:59:02 2006 +1100 28085 28086 add initial controller binding, remove uses of mergedfb test 28087 28088 commit 78061ca460c9d6fd338ccf273bbfa152f471622e 28089 Author: Dave Airlie <airlied (a] linux.ie> 28090 Date: Mon Oct 23 16:49:59 2006 +1000 28091 28092 radeon: separate static controller setup from non-static 28093 28094 commit f917f65e8d2b13d8b9bfc4320f4a935f5a5c4bff 28095 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28096 Date: Thu Nov 2 21:16:53 2006 +0200 28097 28098 Officially require libdrm 2.2.0 28099 28100 Note that this is only for the headers at compile-time. The driver still 28101 uses libdrm 1.2.0 from Xserver at run-time. 28102 28103 commit ca8c148942371c9445f32985625ce38eabc016e1 28104 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28105 Date: Thu Oct 26 15:40:46 2006 +0300 28106 28107 Handle .99 in version patch level. 28108 28109 commit 23cec754158c94b43225b2070fc0a0b003e7bd25 28110 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28111 Date: Thu Oct 26 15:39:49 2006 +0300 28112 28113 [mach64] EXA: reserve DRI buffers only when DRI is active. 28114 28115 commit 9cd78af1d06f753e7bd7e1acaff25398bde45f6e 28116 Author: Dave Airlie <airlied (a] linux.ie> 28117 Date: Mon Oct 23 16:52:07 2006 +1000 28118 28119 radeon: set preferred modes 28120 28121 commit 914a5f0117ccb76292bdd81ea4a2a79d33319550 28122 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28123 Date: Mon Oct 23 10:13:24 2006 +0200 28124 28125 radeon: Pitch doesn't matter when height == 1. 28126 28127 commit 2bcb51d66edaa944379cf8c8ca1ba91fffdc20a8 28128 Author: Michel Daenzer <michel (a] tungstengraphics.com> 28129 Date: Sun Oct 22 17:26:28 2006 +0200 28130 28131 Bug #6756: Attempt to fix repeat picture acceleration. 28132 28133 Always use normalized texture coordinates on R200 and fall back if a POT 28134 texture pitch doesn't match the HW's implicit pitch. 28135 28136 commit dabffb8335027b60ca1fc554423e196dfb9acd6d 28137 Author: Henry Zhao <henry.zhao (a] sun.com> 28138 Date: Tue Oct 17 21:17:47 2006 -0400 28139 28140 improved results for mergedfb xrandr refresh rates - bug 6966 28141 28142 commit 831fc4a009841f68fe161b6e0db5b212d7d8079b 28143 Author: Henry Zhao <henry.zhao (a] sun.com> 28144 Date: Sun Oct 15 17:12:06 2006 -0400 28145 28146 fix bug 6966 (Henry Zhao, Shem Multinymous) 28147 28148 commit 5425be996692425e57884e5a4c159fc79510e89f 28149 Author: Alex Deucher <alex@t41p.(none)> 28150 Date: Sun Oct 15 16:49:24 2006 -0400 28151 28152 clean up magic number 28153 28154 commit 457178f418983041ea59df703a17513a47d17044 28155 Author: Alex Deucher <alex@t41p.(none)> 28156 Date: Sun Oct 15 16:40:08 2006 -0400 28157 28158 Disable all ouputs on initial mode set. The ones we want will get 28159 enabled by RADEONEnableDisplay(). 28160 28161 commit 3fb3f410931a13542fdd4c51c4c9ecada7b37d76 28162 Author: Alex Deucher <alex@t41p.(none)> 28163 Date: Sun Oct 15 13:29:48 2006 -0400 28164 28165 Rework the output init code. Divide into separate functions 28166 for each output. 28167 28168 commit 45e13f3c8786e75276cce85de4af653713d9b4d8 28169 Author: Alex Deucher <alex@t41p.(none)> 28170 Date: Sun Oct 15 13:18:01 2006 -0400 28171 28172 Remove paneloff option. This is already covered by monitorlayout and 28173 shouldn't be needed with xrandr++. 28174 Also remove commented out Bioshotkey code. 28175 28176 commit 4abb5b7afea8d37e95461335e52b16168e555dab 28177 Author: Alex Deucher <alex@t41p.(none)> 28178 Date: Sun Oct 15 13:10:22 2006 -0400 28179 28180 Remove Bioshotkeys option. Leave the bios as configured at boot. 28181 Most distros enable this option by defalt anyway, and with xrandr++ 28182 it won't be an issue anyway. 28183 28184 commit 2933140b5f1f8a6e270460d8918a14943bd0f174 28185 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28186 Date: Sun Oct 15 17:05:02 2006 +0200 28187 28188 radeon: Use xf86XVFillKeyHelperDrawable() when available. 28189 28190 commit 4edbfb9eb952f9b3f5f597749192c17baf15cdce 28191 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28192 Date: Sun Oct 15 17:03:13 2006 +0200 28193 28194 radeon: Track screen damage all the time for page flipping. 28195 28196 This way we don't ever have to forcedly synchronize the pages so we hopefully 28197 never clobber one page's contents when we shouldn't. 28198 28199 commit c9c2ac2b42398f5bfad909f3289e2545bb7b3888 28200 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28201 Date: Sat Oct 14 15:49:22 2006 +0200 28202 28203 radeon: Don't try to synchronize pages when the CP isn't running. 28204 28205 Also move the DamageEmpty() call into RADEONDRIRefreshArea(). 28206 28207 commit c19915a401ccf69046bed03c8d5067f2e72a094e 28208 Author: Dave Airlie <airlied (a] linux.ie> 28209 Date: Fri Oct 6 09:55:40 2006 +1000 28210 28211 always turn on DAC for scanning the CRT connect. 28212 28213 Ubuntu 63994 - myself and BenH discussed this as a good idea on irc 28214 28215 commit b8547aa89f93a7fefbefb1c0d2743999930ac785 28216 Author: Dave Airlie <airlied (a] linux.ie> 28217 Date: Wed Oct 4 01:44:24 2006 +1000 28218 28219 update version number to 6.6.99 in preparation for a 6.7 branch at some stage 28220 28221 commit 23a6f97e08fd49e1cae03cd97cae67a5f06b7634 28222 Author: Alex Deucher <alex@samba.(none)> 28223 Date: Tue Oct 3 13:48:09 2006 -0400 28224 28225 FP timing regs required for both internal and external TMDS 28226 28227 commit 829be884702043fe4c9e349488d2585bf0561ae9 28228 Author: Alex Deucher <alex@samba.(none)> 28229 Date: Tue Oct 3 13:41:21 2006 -0400 28230 28231 remove bogus registers 28232 28233 commit cd900580dbc8928ebc28514162b4c1dbfe9fd5ae 28234 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28235 Date: Tue Oct 3 00:32:22 2006 +0300 28236 28237 [mach64] Fix build for !defined(USE_EXA). 28238 28239 commit 22860a3c0685a25f982983654303ad02aedc02ec 28240 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28241 Date: Sun Jul 16 19:23:56 2006 +0300 28242 28243 Bug 6242: [mach64] Use private DMA buffers. 28244 28245 Map the DMA buffers read-only. This eliminates a security problem where a 28246 client can alter the contents of the DMA buffer after submission to the DRM. 28247 28248 commit cb53fe7c1e45937746e43437ae6adb0355306ae9 28249 Author: Alex Deucher <alex@t41p.(none)> 28250 Date: Sun Oct 1 18:46:05 2006 -0400 28251 28252 fix LVDS plus internal TMDS 28253 28254 commit 145c716de6d42778e447581578f3e4db13c9a653 28255 Author: David Woodhouse <dwmw2 (a] infradead.org> 28256 Date: Sun Oct 1 18:22:14 2006 +0200 28257 28258 Fix radeon vs fbdev pitch calculation. 28259 28260 Fetch the correct displayWidth from fbdevhw after setting the mode. 28261 28262 commit d43ad88fa3913437f6987ab5ab46a38f0cb555a8 28263 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28264 Date: Sat Sep 30 19:25:20 2006 +0200 28265 28266 radeon: Add support for page flipping with EXA. 28267 28268 Also use the damage layer directly instead of via shadowfb and blit dirty 28269 rectangles to the second page in LeaveServer in order to try and improve the 28270 tradeoff between performance and correctness. 28271 28272 commit 146ea328b7962f56c6ab47fbe2adc7561b36189c 28273 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28274 Date: Sat Sep 30 19:22:26 2006 +0200 28275 28276 radeon: Fix function name typo which could probably lead to an unresolved symbol. 28277 28278 Also remove declaration of an unused variable. 28279 28280 commit 6671c1b01bf29d8f1cacf9306ef658b967d8a3cf 28281 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28282 Date: Sat Sep 30 18:29:45 2006 +0200 28283 28284 radeon: Always call RADEONGetAccessibleVRAM(). 28285 28286 This fixes issues with Option "UseFBDev" and the new memmap code. 28287 28288 OTOH, there is potential for this to break IGPs, but that's what Bugzilla and 28289 git-bisect are for... 28290 28291 commit c979c7f87fc449cc9631820101edff82c3013e66 28292 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 28293 Date: Sat Sep 30 15:14:42 2006 +0200 28294 28295 radeon: Check ROM more thoroughly before treating it as an x86 BIOS. 28296 28297 This prevents the ROMs on some Mac cards getting treated as x86 BIOSen, 28298 resulting in garbage values. 28299 28300 commit 37c114acbabe35881c39642e293519e2635907aa 28301 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28302 Date: Sat Sep 30 15:11:42 2006 +0200 28303 28304 Remove some dead code. 28305 28306 Fix two compile warnings. 28307 28308 commit 6d7b0763173f56050411d7bdf9a583b9bfc050c4 28309 Author: Adam Jackson <ajax (a] benzedrine.nwnk.net> 28310 Date: Fri Sep 29 18:50:44 2006 -0400 28311 28312 Undo unintended configure.ac change. 28313 28314 Dammit, git, grow a sensible UI already. 28315 28316 commit f96ddd32d3a9e64dc0adec5c50386d6db5b182ee 28317 Author: Ronald Wahl <rwahl (a] gmx.de> 28318 Date: Fri Sep 29 17:51:42 2006 -0400 28319 28320 Hook up DFP power to DPMS events. 28321 28322 commit f6b42149c5a3fee663575fe6bbe78b62695bdbc8 28323 Author: Alex Deucher <alex@samba.(none)> 28324 Date: Fri Sep 29 00:44:36 2006 -0400 28325 28326 tvdac handling was reversed between r200 and other chips 28327 28328 commit c0660495080719c052d6393ede707755929102cd 28329 Author: Dave Airlie <airlied (a] linux.ie> 28330 Date: Wed Sep 27 18:12:52 2006 +1000 28331 28332 radeon: check DDC info for both connectors not just the pScrn one. 28333 28334 Otherwise we disable DDC before we've checked the CRT if LVDS is in use 28335 28336 commit 5b98545618141e86164f3ce66469f9e16b1cf6d4 28337 Author: Dave Airlie <airlied (a] linux.ie> 28338 Date: Wed Sep 27 18:10:30 2006 +1000 28339 28340 radeon: don't move connectors around if LVDS connector is connector 1 28341 28342 On my laptop this allow me to plug or not plug a monitor without affecting 28343 my lvds 28344 28345 commit 74c725bd86348b5bf9bba0c50e8a6a7885d7ff7b 28346 Author: Dave Airlie <airlied (a] linux.ie> 28347 Date: Mon Sep 25 11:35:32 2006 +1000 28348 28349 radeon: cleanup some bits from superpatch 28350 28351 commit 9827afaa40720a58e2fc029c0bf92ad2fe223d11 28352 Merge: 10b4b46c1 ad8259c81 28353 Author: Dave Airlie <airlied (a] linux.ie> 28354 Date: Mon Sep 25 11:25:17 2006 +1000 28355 28356 Merge branch 'radeon-sp' 28357 28358 commit ad8259c814629e741ed5567923f40879cc0c7051 28359 Author: Dave Airlie <airlied (a] linux.ie> 28360 Date: Mon Sep 25 11:25:03 2006 +1000 28361 28362 radeon : add back tv-dac register write 28363 28364 commit 10b4b46c16ff3748856b732f9a39de40ba197112 28365 Author: Dave Airlie <airlied (a] linux.ie> 28366 Date: Mon Sep 25 09:08:47 2006 +1000 28367 28368 remove all CVS annotations from the ATI driver files 28369 28370 commit 976dec04786e9b050945c290840b4cdc978c0b23 28371 Author: Dave Airlie <airlied (a] linux.ie> 28372 Date: Mon Sep 25 09:00:21 2006 +1000 28373 28374 radeon: put back in reverseddc code 28375 28376 BenH claims it is needed, agd5f can fix it up later if it isn't. 28377 28378 commit 980fb2f6bd7641c8c57769b0c67e3561903e89a9 28379 Author: Dave Airlie <airlied (a] linux.ie> 28380 Date: Sat Sep 23 08:31:56 2006 +1000 28381 28382 radeon: add some missing bits of superpatch for crtc registers 28383 28384 commit 1ba4f36821e60289cad937abbb0edb273c88436c 28385 Author: Dave Airlie <airlied (a] linux.ie> 28386 Date: Sat Sep 23 08:25:42 2006 +1000 28387 28388 radeon: finish radeon_driver.c import for register changes 28389 28390 commit 547543bbefe605a453bfa5ae6d063ae02c5f040e 28391 Author: Alex Deucher <alexdeucher (a] gmail.com> 28392 Date: Sat Sep 23 08:21:59 2006 +1000 28393 28394 radeon: re-organise FP and CRTC register setting routines 28395 28396 commit fd978140bcb7670f28c684c06c2b6c611c26bef4 28397 Author: Dave Airlie <airlied (a] linux.ie> 28398 Date: Sat Sep 23 08:09:57 2006 +1000 28399 28400 radeon: initial CRTC register setting alignment 28401 28402 commit 406f4911e7ce821002c111d6bcdec35f5b56c943 28403 Author: Dave Airlie <airlied (a] linux.ie> 28404 Date: Sat Sep 23 07:59:24 2006 +1000 28405 28406 radeon: setup the radeon init function to use secondary 28407 28408 commit 452f0f8079d65679905ed5178a256534ef0db0e4 28409 Author: Dave Airlie <airlied (a] linux.ie> 28410 Date: Sat Sep 23 07:41:57 2006 +1000 28411 28412 radeon: bring over final set of changes for radeon_display.c from susperpatch 28413 28414 This brings over Alex's final set of changes... 28415 28416 i'm unsure about the disappearance of reverseddc - alex to review 28417 28418 commit a8b9aa69259a0706d0688d2656ece5a292799242 28419 Author: Dave Airlie <airlied (a] linux.ie> 28420 Date: Sat Sep 23 07:38:25 2006 +1000 28421 28422 radeon: align tabs and make display detection like superpatch 28423 28424 commit 8d4ae98c528a2309dc3c545893dbba2e7f76916c 28425 Author: Dave Airlie <airlied (a] linux.ie> 28426 Date: Sat Sep 23 07:26:44 2006 +1000 28427 28428 radeon: align some output with superpatch 28429 28430 commit 65b54847c5e79b2ec3cb1a08b265229443c9deb9 28431 Author: Dave Airlie <airlied (a] linux.ie> 28432 Date: Sat Sep 23 07:22:31 2006 +1000 28433 28434 radeon: change function name to Displays rather than Monitors 28435 28436 commit 7e16d9fff96dcdff8c25dbbe2a51be0023b7985a 28437 Author: Dave Airlie <airlied (a] linux.ie> 28438 Date: Sat Sep 23 07:21:15 2006 +1000 28439 28440 radeon: make GetConnectedMonitors a static void 28441 28442 commit 6f63b3dc8b3e2299f5f02abc1c565176d2d46cee 28443 Author: Dave Airlie <airlied (a] linux.ie> 28444 Date: Sat Sep 23 07:19:43 2006 +1000 28445 28446 radeon: some minor alignments and cleanups for info-> 28447 28448 commit c9f3ced3b1c194cacf1ab9436ee6c2d4efd3a8e9 28449 Author: Alex Deucher <alexdeucher (a] gmail.com> 28450 Date: Sat Sep 23 07:16:10 2006 +1000 28451 28452 radeon: add reverse display option to switch displays on non-mobility cards 28453 28454 commit e752443c540d9a94522050c61ef7c33d912fece5 28455 Author: Dave Airlie <airlied (a] linux.ie> 28456 Date: Fri Sep 22 07:10:10 2006 +1000 28457 28458 radeon: flesh out mapcontrollers 28459 28460 commit 7a0e4eb47655c4701fc3a4b33b6c8e9f378ccc9e 28461 Author: Dave Airlie <airlied (a] linux.ie> 28462 Date: Fri Sep 22 07:05:56 2006 +1000 28463 28464 radeon: add some info to the map controllers 28465 28466 commit 75bf69c868be6164d0041a32f52244f1ef83e050 28467 Author: Dave Airlie <airlied (a] linux.ie> 28468 Date: Fri Sep 22 07:03:52 2006 +1000 28469 28470 radeon: add simple map controllers call 28471 28472 commit 8ba5abd1329347c7fbf4b4d09c95da0d23b4884a 28473 Author: Dave Airlie <airlied (a] linux.ie> 28474 Date: Fri Sep 22 06:58:41 2006 +1000 28475 28476 radeon: move comment 28477 28478 commit e8d77b7c36fa791a2350a6f4b15728b15d8f3de0 28479 Author: Dave Airlie <airlied (a] linux.ie> 28480 Date: Fri Sep 22 06:55:51 2006 +1000 28481 28482 radeon: refactor blank/unblank 28483 28484 commit 8534723f5710e2916b4a22efe22a6b7365cc8db6 28485 Author: Alex Deucher <alexdeucher (a] gmail.com> 28486 Date: Fri Sep 22 06:36:03 2006 +1000 28487 28488 radeon: make DVI_I default connector 28489 28490 commit d7fc79fd9f094ac462de0883af7e6030cb6195a6 28491 Author: Dave Airlie <airlied (a] linux.ie> 28492 Date: Fri Sep 22 06:35:34 2006 +1000 28493 28494 radeon: add enable display function 28495 28496 commit 4a54886d510f26b29d27e5c9a73647554291b1a6 28497 Merge: 70264d645 9fa176c79 28498 Author: Dave Airlie <airlied (a] linux.ie> 28499 Date: Fri Sep 22 06:00:04 2006 +1000 28500 28501 Merge branch 'master' into radeon-spm1 28502 28503 Conflicts: 28504 28505 src/radeon_driver.c 28506 28507 commit 9fa176c7989030e7340cc9b85d0b6a7b34303091 28508 Author: Alex Deucher <alex@samba.(none)> 28509 Date: Tue Sep 19 13:58:17 2006 -0400 28510 28511 Provide all resolutions (instead only the largest one) in MergedFB clone 28512 mode without config file (Henry Zhao) 28513 28514 commit e742aeb28c7d9d6e75932c408bcc7c44af52e303 28515 Author: Alex Deucher <alex@samba.(none)> 28516 Date: Tue Sep 19 13:45:33 2006 -0400 28517 28518 fix large negative refresh rates with mergedfb (Henry Zhao) 28519 fixes bug 6966 28520 28521 commit 9c0c805704f1df1b66adc2a6b9313597ae5b86cc 28522 Author: Dave Airlie <airlied (a] linux.ie> 28523 Date: Mon Sep 18 19:41:21 2006 +1000 28524 28525 radeon: fix rn50 memmap 28526 28527 This should fixup the rn50/m6/m7 memory map problem that has been reported 28528 since BenH's memory mapping changes 28529 28530 commit 70264d645724718a48ca229fedc04e33681b7d31 28531 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28532 Date: Mon Sep 18 09:56:25 2006 +1000 28533 28534 radeon; put back old TV DAC code for now 28535 28536 commit 53eb843ad8ec5a292dff0844330ecfa9b7b2a0cb 28537 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28538 Date: Mon Sep 18 07:55:21 2006 +1000 28539 28540 radeon: move code around in InitCrtcRegisters 28541 28542 commit bcae4352079109a9dd16117bf9ec19dd215180bf 28543 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28544 Date: Mon Sep 18 07:33:48 2006 +1000 28545 28546 radeon: cleanup DPMS set on/off code to use new controller structure 28547 28548 commit 87622f56b3dd7e2d2f00ad71e5b569eac31fb2b1 28549 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28550 Date: Mon Sep 18 07:15:54 2006 +1000 28551 28552 radeon: cleanup DacPowerSet 28553 28554 commit a417fa081926b2f84e2cbdc355cc7462d1ebdfad 28555 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28556 Date: Mon Sep 18 07:15:41 2006 +1000 28557 28558 radeon: fixup InitFPRegisters function call to not pass info 28559 28560 commit 8bff57a938e92dfc8729e40f7aa9ad9a96262131 28561 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28562 Date: Mon Sep 18 07:02:35 2006 +1000 28563 28564 radeon: copy of overupdated bios hotkeys code from Alex 28565 28566 commit 76352c684d5e840c9f7ac0781e3ecb32951b32db 28567 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28568 Date: Mon Sep 18 06:54:28 2006 +1000 28569 28570 radeon: minor cleanup of masking for crtc registers 28571 28572 commit a7b97697ed13ac25844890439ffba31524e03a16 28573 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28574 Date: Mon Sep 18 06:53:23 2006 +1000 28575 28576 radeon: initialise controller structures 28577 28578 commit 362ee3778764633cb4c9388aa329c5b715885038 28579 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28580 Date: Mon Sep 18 06:41:14 2006 +1000 28581 28582 radeon: move InitFPRegisters up to align with superpatch 28583 28584 commit 11fb9a14741746c752b0001c1b1d39a54e69657a 28585 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28586 Date: Mon Sep 18 06:35:50 2006 +1000 28587 28588 radeon: add tv dac adj lookup 28589 28590 commit a15ac59c4d1a15380910be2a7438111b09161c81 28591 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28592 Date: Mon Sep 18 06:35:28 2006 +1000 28593 28594 radeon: re-add ATI patch for rn50 modes 28595 28596 commit d80ea2d67b63ea2f2ce0277952af186b69b19e7d 28597 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28598 Date: Mon Sep 18 06:30:30 2006 +1000 28599 28600 radeon: re-add DAC power on/off code around detect 28601 28602 commit 0a93936f9f048b8837f1c765fea7ccb61f072697 28603 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28604 Date: Mon Sep 18 06:25:15 2006 +1000 28605 28606 radeon: restore secondary head behaviour 28607 28608 commit 6da5a0bfdedb2062ec835f45514c22917abdcf4c 28609 Author: Alex Deucher <alexdeucher (a] gmail.com> 28610 Date: Mon Sep 18 06:16:58 2006 +1000 28611 28612 radeon: move to using Alex's controller structure 28613 28614 commit cd1bc1d1febf0a79d835c7dc8b353123278dec51 28615 Author: Alex Deucher <alexdeucher (a] gmail.com> 28616 Date: Mon Sep 18 02:44:09 2006 +1000 28617 28618 radeon: add save/restore of some new registers 28619 28620 move some registers about 28621 28622 commit 9b5c4cf8a98b0c20bb821f5003495481fbd0c10c 28623 Author: Alex Deucher <alexdeucher (a] gmail.com> 28624 Date: Mon Sep 18 02:28:32 2006 +1000 28625 28626 radeon: add pre init controller function 28627 28628 commit 311e776c4429485d29aaf11d350cab8e6fe7564b 28629 Author: Alex Deucher <alexdeucher (a] gmail.com> 28630 Date: Mon Sep 18 02:21:49 2006 +1000 28631 28632 radeon: add missing tv-out registers to register list 28633 28634 commit 65cb77c79c2c5cb536494200d4fba6815d7e771c 28635 Author: Alex Deucher <alexdeucher (a] gmail.com> 28636 Date: Mon Sep 18 02:21:29 2006 +1000 28637 28638 radeon: restructure files along lines of Alex's superpatch 28639 28640 commit 25fa71ca0a24190cc8be2e80769443e0e09303b3 28641 Author: Henry Zhao <henryz (a] ati.com> 28642 Date: Mon Sep 18 03:15:54 2006 +1000 28643 28644 Power up DAC around check for CRT connectedness 28645 28646 commit 39fcac431eba60adf2d4e357cfc811dbae36a4e1 28647 Author: Henry Zhao <henryz (a] ati.com> 28648 Date: Mon Sep 18 03:09:43 2006 +1000 28649 28650 Bump up clock rate for ES1000 to allow higher resolutions 28651 28652 commit 7d9e2b450f0fa42553156f57572a6dab70a0ea13 28653 Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 28654 Date: Tue Sep 12 07:04:23 2006 +1000 28655 28656 radeon: this stops mergedfb working on my ATI M7 development board 28657 28658 I've no idea what this is for, it came from an ATI codedrop and it may 28659 fix some laptop or production system, but it looks to me more like a dirty 28660 hack than a true fix, as I have an M7 which I need to specify CRT for the 28661 second head and it breaks mergedfb for me. 28662 28663 commit ad1778eb1eb59de14ec309b99131931cbccf64e9 28664 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28665 Date: Sun Sep 10 23:32:44 2006 +0200 28666 28667 Revert "* [ Bug 6966 ] large negative refresh rates reported" 28668 28669 This reverts caaed927a07ffbac68b08246185ef93c1e7bb98c commit. 28670 28671 See https://bugs.freedesktop.org/show_bug.cgi?id=8137 . 28672 28673 commit 6e92981b024b59c2237f949321f6b2591dada519 28674 Author: Tilman Sauerbeck <tilman (a] code-monkey.de> 28675 Date: Sun Sep 3 21:11:10 2006 +0200 28676 28677 Removed the warning re. the r300 DRI driver status. 28678 28679 commit caaed927a07ffbac68b08246185ef93c1e7bb98c 28680 Author: Henry Zhao <henryz (a] localhost.localdomain> 28681 Date: Thu Aug 31 13:27:45 2006 -0700 28682 28683 * [ Bug 6966 ] large negative refresh rates reported 28684 * [ Bug 6309 ] radeon/r128 fails to read hsync/vsync rates when rang 28685 * descriptor is missing 28686 * [ Bug 5473 ] Blank screen with Radeon Mobility X700 (Acer Ferrari 4005) 28687 * Reboot should not be needed for external CRT to function when connected 28688 later on Ferrari 4000 28689 * Modes need to be sorted, and sorted correctly 28690 * Provide all resolutions (instead only the largest one) in MergedFB clone 28691 mode without config file 28692 * Bump up clock rate for ES1000 to allow higher resolution 28693 28694 commit 6d2028b84ff2b82c849b74776258f8e52c952e36 28695 Author: Dave Airlie <airlied (a] linux.ie> 28696 Date: Fri Aug 25 10:05:44 2006 +1000 28697 28698 and of course it takes two commits to do a simple commit 28699 28700 commit e90c6feddbcd719fda3bbc3d48fd55045e9d5adc 28701 Author: Dave Airlie <airlied (a] linux.ie> 28702 Date: Fri Aug 25 09:59:58 2006 +1000 28703 28704 add changelog hook from nv driver 28705 28706 commit 1e06212f792d1ed1f7545e16824b86fbd34fa284 28707 Author: Adam Jackson <ajax (a] benzedrine.nwnk.net> 28708 Date: Thu Aug 24 19:52:30 2006 -0400 28709 28710 Enable CPIO in the atimisc driver on ia64 platforms. 28711 28712 This fixes many mach64s, but potentially breaks mach32. Mach64 is widespread 28713 on ia64 machines but PCI mach32's are quite rare on any platform. 28714 28715 commit dfa47b09b5b8d7eeaa3f8ae629ebb3c7efb51511 28716 Author: Adam Jackson <ajax (a] benzedrine.nwnk.net> 28717 Date: Thu Aug 24 19:50:46 2006 -0400 28718 28719 Disable DRI on xpress 200{,m} since we know it's broken. 28720 28721 commit 1dcd9b1f75ba2307beb42556aa7f2f945171d5fb 28722 Author: Adam Jackson <ajax (a] benzedrine.nwnk.net> 28723 Date: Thu Aug 24 19:42:26 2006 -0400 28724 28725 Disable DRI on RN50 in a more pleasant way. 28726 28727 Rather than publish a non-existant driver name for RN50, just don't bother 28728 initializing it in the first place. 28729 28730 commit 64b18e246586b1043a12f381a9d4c834464ed9b5 28731 Author: Dave Airlie <airlied (a] linux.ie> 28732 Date: Fri Aug 25 09:40:04 2006 +1000 28733 28734 update version numbers for release 28735 28736 commit be229d97e1f0282f2099c7ace7cb257a2738a58f 28737 Author: Dave Airlie <airlied (a] linux.ie> 28738 Date: Fri Aug 25 09:39:46 2006 +1000 28739 28740 make colortiling default to on all cards 28741 28742 commit 750ed51bc3d5119e2d2a91a05ae5a971dcd3b88a 28743 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28744 Date: Mon Aug 7 17:09:02 2006 +0300 28745 28746 [mach64] RENDER support: drop caching DP_WRITE_MASK. 28747 28748 Caching DP_WRITE_MASK causes artifacts, e.g. with evas_xrender_x11_test. 28749 28750 commit b2beea2fa8949874d3d57fb9b43fe85cc08a8bff 28751 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28752 Date: Sat Aug 5 18:36:24 2006 +0300 28753 28754 [mach64] RENDER support: save/restore, cache texture registers. 28755 28756 commit 43aaed99950640c3695b3c2b91faabf00c6338a7 28757 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28758 Date: Sat Aug 12 22:12:07 2006 +0300 28759 28760 [mach64] RENDER support (bug #6877). 28761 28762 RENDER accleration is disabled by default. It is strongly recommended that the 28763 patch from bug #6772 and the 'exa-damagetrack branch' are merged in the xserver 28764 before enabling RENDER acceleration. To enable RENDER acceleration, add the 28765 following in xorg.conf: 28766 28767 Section "Device" 28768 [...] 28769 28770 Option "AccelMethod" "exa" 28771 Option "RenderAccel" "true" 28772 EndSection 28773 28774 commit d113f40fd97408661193895ff3aa82262f100143 28775 Author: Dave Airlie <airlied (a] linux.ie> 28776 Date: Sat Aug 12 17:16:33 2006 +1000 28777 28778 radeon: make some local data static 28779 28780 This is inspired by a patch in XFree86 from David Dawes, it isn't the 28781 same as we have different gamma curves. 28782 28783 commit a1da2b5e0a38feae01107676a76bdc5c185ad34b 28784 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28785 Date: Sat Aug 5 03:18:34 2006 +0300 28786 28787 [mach64] Halve max w/h for hardware-accelerated operations. 28788 28789 EXA hits these limits for some operations on offscreen pixmaps and GTPRO seems 28790 to support the reduced limits. 28791 28792 commit a66c8d2f9855caa10142b39f155dd8232b1f2389 28793 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28794 Date: Sat Aug 5 03:16:56 2006 +0300 28795 28796 [mach64] EXA: add memcpy-based UTS/DFS. 28797 28798 EXA hits more optimized paths when it does not have to fallback because of 28799 missing UTS/DFS. 28800 28801 commit 0fb8503d60cf646c91a806f9ed322f9ceff18baa 28802 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28803 Date: Sat Aug 5 03:16:15 2006 +0300 28804 28805 [mach64] XAA: minor refactoring of memory manager setup. 28806 28807 - DRI: factor out to separate function and add comments on layout 28808 - no DRI: factor out to separate function, maxScanlines is always equal to 28809 ATIMach64MaxY 28810 28811 commit e203d86643d5d70bf18248712d05b72b79aee705 28812 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28813 Date: Sat Aug 5 03:15:42 2006 +0300 28814 28815 [mach64] EXA support. 28816 28817 commit dc1e289a611a17090e6dc7ae8a8d3f26d20df4eb 28818 Author: Samuel Thibault <samuel.thibault (a] ens-lyon.org> 28819 Date: Sat Jul 29 01:05:35 2006 +0300 28820 28821 Bug #6623: Fix I2C bus Mach64 initialisation failure. 28822 28823 commit b377f403e0cd5253402ef3945e5944d1a2763a94 28824 Author: Marc Aurele La France <tsi (a] xfree86.org> 28825 Date: Sat Jul 29 01:03:34 2006 +0300 28826 28827 Fix warnings for Option:1 bit fields. 28828 28829 commit 91b8b0ca41ad0b9659f0982a05148cab8558e9d5 28830 Author: Marc Aurele La France <tsi (a] xfree86.org> 28831 Date: Sat Jul 29 01:02:48 2006 +0300 28832 28833 Bug #807: copy throttling. 28834 28835 Fix corruption associated with the engine randomly not waiting for a copy 28836 operation to commit its results. 28837 28838 commit 7800407a2db4b7307192bf3b43fffb9c4748b9c0 28839 Author: Marc Aurele La France <tsi (a] xfree86.org> 28840 Date: Sat Jul 29 01:00:56 2006 +0300 28841 28842 Bug #807: invalidate read-back cache. 28843 28844 Fix corruption when scaling 1-pixel wide pixmaps, in which case the first CPU 28845 read of the framebuffer will return stale data. 28846 28847 This bug always manifests itself in mach64 EXA (to be committed soon), with 28848 this patch mach64 EXA passes rendercheck. 28849 28850 commit 49b11c540d37152eee4907ab30353ede01e4fb56 28851 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28852 Date: Sat Jul 29 00:52:28 2006 +0300 28853 28854 Bug #807: split out TestRegisterCaching block for DP regs. 28855 28856 commit 290c5aaa580428bc4748a47e4b9bcc22b90fc8d4 28857 Author: George Sapountzis <gsap7 (a] yahoo.gr> 28858 Date: Sat Jul 29 00:51:53 2006 +0300 28859 28860 Bug #807: split out TestRegisterCaching block for DP regs. 28861 28862 commit fba8c839b47e2c8d6a6d65950c3431ff5b870aa0 28863 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28864 Date: Sun Jul 23 18:51:47 2006 +0200 28865 28866 Bug #6819: Fix broken rendering with *8b8g8r8 source formats. 28867 28868 Use the corresponding texture formats with R200 family. 28869 28870 Unfortunately, it looks like the R100 family can't accelerate these source 28871 formats, so we have to fall back for them. 28872 28873 commit 60ebe97e2d5d27c2bf4213e201cef74503809cb6 28874 Author: Tilman Sauerbeck <tilman (a] code-monkey.de> 28875 Date: Mon Jul 3 19:37:28 2006 +0200 28876 28877 Move variable declarations before statements to be C89 compatible 28878 28879 commit 6900ff3c5e574881b1470e2363b200e99ab8bf3e 28880 Author: Alex Williamson <alex.williamson (a] hp.com> 28881 Date: Mon Jul 3 13:09:47 2006 +0200 28882 28883 Bug #6872: Disable digital output while reading ROM with Radeon 7000. 28884 28885 This is a workaround for ATI Radeon 7000 erratum F1. 28886 28887 commit 920ac1f56fad6f48a30e65836ce054815f270738 28888 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28889 Date: Mon Jul 3 01:21:18 2006 +0200 28890 28891 Do even less cache flushing when the previous engine mode is known. 28892 28893 OTOH, flush everything when the previous engine mode is unknown, and mark the 28894 engine mode as unknown in a couple more cases. 28895 28896 commit dd0dd4bab76907b4b1b6d814f24e73e57d71f587 28897 Author: Eric Anholt <anholt (a] freebsd.org> 28898 Date: Mon Jul 3 01:10:10 2006 +0200 28899 28900 Bug #6762: Remove gratuitous cache flushing. 28901 28902 commit 22f165f290f57e638d198f24a37f62fc16b72bae 28903 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28904 Date: Mon Jul 3 00:39:45 2006 +0200 28905 28906 Build and warning fixes for !defined(USE_XAA). 28907 28908 commit ea5105230c7c3519730b8d3c250e9ba9a4e4374a 28909 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28910 Date: Mon Jul 3 00:03:22 2006 +0200 28911 28912 Reserve HW cursor memory statically instead of from EXA offscreen. 28913 28914 As a result, quite a bit of code can be removed that attempted to deal with the 28915 EXA offscreen memory disappearing, with varying success. Because the HW cursor 28916 memory is now always immediately after the front buffer and before the back 28917 buffer, this also fixes bug #6808. 28918 28919 As the HW cursor memory is now reserved statically with EXA and XAA, change 28920 RADEONUseHWCursor() to check info->cursor instead of info->cursor_offset, for 28921 which 0 might become valid with HW cursor enabled in the future. 28922 28923 Change info->cursor_offset from unsigned long to CARD32 as the corresponding 28924 register is 32 bits wide, and to avoid issues with printf. 28925 28926 commit 66586f905857ea5bf37cb5d34949cd450a9ed067 28927 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28928 Date: Sun Jul 2 23:38:20 2006 +0200 28929 28930 Clean up some redundant code. 28931 28932 commit 1159c85861dc507ebf323aca82e32fcc67726420 28933 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28934 Date: Sun Jul 2 18:24:28 2006 +0200 28935 28936 Bug #6548: Fix playback of interlaced video with more than 1023 lines. 28937 28938 This will still break with 2048 lines... 28939 28940 commit 11bedb7c845fc6074383890d5525315aebe05f2c 28941 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28942 Date: Sun Jul 2 17:03:23 2006 +0200 28943 28944 Bug #6939: Disable RENDER acceleration by default on RN50. 28945 28946 Based on patch from Weixing Zhang, but still allows enabling it with Option 28947 "RenderAccel". 28948 28949 commit 4a0c22e32901a510c7aad070be25c31519c284e3 28950 Author: Weixing Zhang <anzhang (a] ati.com> 28951 Date: Sun Jul 2 16:55:25 2006 +0200 28952 28953 Bug #6939: Disable 3D acceleration on RN50 by setting invalid ClientDriverName. 28954 28955 commit 9c4a1e3300d30b23806988aa9328fc6ef92e2663 28956 Author: Frank Gevaerts <frank (a] gevaerts.be> 28957 Date: Thu Jun 29 18:21:22 2006 +0200 28958 28959 Fix build for !defined(USE_EXA). 28960 28961 commit ee0544b670567491b7770f469a64a6cd85519ba1 28962 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28963 Date: Thu Jun 29 18:19:58 2006 +0200 28964 28965 Remove unused variables. 28966 28967 commit 15c556f5a8c9069d3dd00ce139d46c39b7298951 28968 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28969 Date: Thu Jun 29 18:18:21 2006 +0200 28970 28971 Don't declare unused variables on __powerpc__ and __alpha__. 28972 28973 commit 055ac2836ec0afd51cf554d043a59068c36a07d3 28974 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28975 Date: Thu Jun 29 18:14:45 2006 +0200 28976 28977 Fix strict aliasing violation flagged by gcc -Wall. 28978 28979 commit 4b1904017caa976c138594a86e75feaf470e72b5 28980 Author: Dave Airlie <airlied (a] linux.ie> 28981 Date: Mon Jun 26 20:34:34 2006 +1000 28982 28983 radeon: force CP and VIP clocks on some r300 and rv100 chips. 28984 28985 fglrx appears to do this on r300 and Jerome Glisse has spent most of his 28986 life tracking down what caused the 9800 to lockup. This is my attempt to 28987 fix this. Please report if it works for you. 28988 28989 Signed-off-by: Dave Airlie <airlied (a] linux.ie> 28990 28991 commit c5d8c8db3e75976687bfb364f0a7b235c7ae8793 28992 Author: Michel Dnzer <michel (a] tungstengraphics.com> 28993 Date: Sat Jun 24 16:01:19 2006 +0200 28994 28995 Bug #7274: Fix corruption with 'small' accelerated DownloadFromScreen transfers. 28996 28997 commit f25ba3dc0e7850f0b1b78f54c0e7211bf2e81b3a 28998 Author: Tilman Sauerbeck <tilman (a] code-monkey.de> 28999 Date: Fri Jun 16 11:38:18 2006 +0200 29000 29001 Remove obsolete FIXME. 29002 29003 commit 764013ec61b9043f1487513658e374569f7154a9 29004 Author: Tilman Sauerbeck <tilman (a] code-monkey.de> 29005 Date: Fri Jun 16 10:56:02 2006 +0200 29006 29007 Use AC_CHECK_HEADER() to check for exa.h 29008 29009 commit 3c701d49b3eb7adb82267cec53163e475913aca3 29010 Author: Tilman Sauerbeck <tilman (a] code-monkey.de> 29011 Date: Fri Jun 16 09:59:29 2006 +0200 29012 29013 Use AC_MSG_RESULT() instead of echo for check results. 29014 29015 commit dbef62615f38aebf4cdc579e3b26a7e7c719abc4 29016 Author: Tilman Sauerbeck <tilman (a] code-monkey.de> 29017 Date: Fri Jun 16 08:50:49 2006 +0200 29018 29019 conditionally declare pPCI and PciReg, too 29020 29021 commit a01c01787cfbe3706ae7747cde1657abd79ecdaa 29022 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29023 Date: Wed Jun 21 09:38:47 2006 +0200 29024 29025 Bug #7283: Build fixes for !defined(XF86DRI). 29026 29027 Thanks to Matthieu Herrb for pointing out some of these. 29028 29029 commit 79d1c573133e64ddf4446b6bb0eaab4a5a6c73ab 29030 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29031 Date: Sun Jun 18 17:39:09 2006 +0200 29032 29033 Implement accelerated EXA DownloadFromScreen hook. 29034 29035 x11perf -getimage numbers only increase by about 20-30% on my PowerBook with an 29036 M9, but by about 100 times(!) with a PCIe X550. I suspect the former could 29037 perform better with PCI as opposed to AGP transfers, which would also remove 29038 the need to disable this by default with AGP. 29039 29040 commit 2bcdc7cfd013bd457d35f927390e5cc3544f30f1 29041 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29042 Date: Sun Jun 18 16:01:00 2006 +0200 29043 29044 Remove #include <fbdevhw.h>, not used here. 29045 29046 commit 25a5ccb63a9e3074b4c8eaf48844877d84ff76bb 29047 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29048 Date: Sun Jun 18 15:26:55 2006 +0200 29049 29050 Prefer visuals with stencil when it's hardware acclerated. 29051 29052 Also make the formatting of the if statements consistent. 29053 29054 commit 8fcc436f19f9c1371295af2d0c45608fb43c584b 29055 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29056 Date: Sun Jun 18 15:16:08 2006 +0200 29057 29058 Take part of the blame for the radeon driver. 29059 29060 commit c8366d7b457565a06acdf6b40c5448c7f13f1641 29061 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29062 Date: Sun Jun 18 14:24:14 2006 +0200 29063 29064 Add Option "DepthBits". 29065 29066 This allows getting higher depth buffer precision and stencil hardware 29067 acceleration in depth 16 or saving some video RAM at the cost of these features 29068 in depth 24. 29069 29070 commit 23a21d00db6f3b88b54c44648f968018a03677d3 29071 Author: Tilman Sauerbeck <tilman (a] code-monkey.de> 29072 Date: Fri Jun 16 16:37:07 2006 +1000 29073 29074 [PATCH] move the declaration of pMach64 into the correct ifndef statement 29075 Signed-off-by: Tilman Sauerbeck <tilman (a] code-monkey.de> 29076 Signed-off-by: Dave Airlie <airlied (a] linux.ie> 29077 29078 commit c6b33db9d2c751e34c2152b0bb7e0e84a812589f 29079 Author: Dave Airlie <airlied (a] linux.ie> 29080 Date: Fri Jun 16 16:22:03 2006 +1000 29081 29082 ati: prepare for release 29083 29084 Update some versions for release 29085 29086 commit 9c534b5330534fc6cf392936a9b200a5128c787c 29087 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29088 Date: Thu Jun 15 16:09:55 2006 +0200 29089 29090 Some more ignores. 29091 29092 commit 3d68698647fb17a6f5ae9a225f552cb0130fa946 29093 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29094 Date: Thu Jun 15 16:07:15 2006 +0200 29095 29096 Fix semantic glitches pointed out by gcc -Wall. 29097 29098 The radeon driver builds warning-free with gcc -Wall here now. 29099 29100 commit ab2b48e0df328fd08a2412911126b3184a601aea 29101 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29102 Date: Thu Jun 15 16:06:13 2006 +0200 29103 29104 Remove another unused variable. 29105 29106 commit 56bd5a370472663af5607cd327f757367b238db0 29107 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29108 Date: Thu Jun 15 15:35:05 2006 +0200 29109 29110 Fix some more cosmetic warnings. 29111 29112 commit ec93e08b24f7c0e583970d08bfe18ad142e6b047 29113 Merge: 051c49d0f 6ca986242 29114 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29115 Date: Thu Jun 15 15:03:27 2006 +0200 29116 29117 Merge branch 'master' of ssh+git://git.freedesktop.org/git/xorg/driver/xf86-video-ati 29118 29119 commit 6ca986242c80bcfe7ca2d435a772bbdf034892c4 29120 Author: Dave Airlie <airlied (a] linux.ie> 29121 Date: Thu Jun 15 21:24:46 2006 +1000 29122 29123 remove unused variable 29124 29125 commit 9a5f5bc0d3f5e62b79355993ca6246382394daee 29126 Author: Dave Airlie <airlied (a] linux.ie> 29127 Date: Thu Jun 15 21:24:16 2006 +1000 29128 29129 remove all printf specifier warnings on Linux 29130 29131 commit 051c49d0f002c72d1d8f15f6df264d8b093d4e95 29132 Merge: a840afc75 dfac8191d 29133 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29134 Date: Thu Jun 15 13:10:44 2006 +0200 29135 29136 Merge branch 'master' of ssh+git://git.freedesktop.org/git/xorg/driver/xf86-video-ati 29137 29138 commit dfac8191dbda29ef7f6ce33d7356fee89c17d59b 29139 Author: Dave Airlie <airlied (a] linux.ie> 29140 Date: Thu Jun 15 21:05:16 2006 +1000 29141 29142 fix last ati*.[ch] format specifiers 29143 29144 commit 33525540f3a9d6760c6edcfe765d2a004a6b894f 29145 Author: Dave Airlie <airlied (a] linux.ie> 29146 Date: Thu Jun 15 21:03:01 2006 +1000 29147 29148 fix more printf formats 29149 29150 commit 6bcf6f29af24b6725e097dd72af8e8b9f6d0676c 29151 Author: Dave Airlie <airlied (a] linux.ie> 29152 Date: Thu Jun 15 20:59:33 2006 +1000 29153 29154 some more unused variable cleanups 29155 29156 commit a840afc75a0453b4694e3f987ebdbddc6844a169 29157 Merge: c90c49da3 d511ef486 29158 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29159 Date: Thu Jun 15 12:46:58 2006 +0200 29160 29161 Merge branch 'master' of ssh+git://git.freedesktop.org/git/xorg/driver/xf86-video-ati 29162 29163 commit d511ef486a15b582e5d1708a17a00701f5cf1ccd 29164 Author: Dave Airlie <airlied (a] linux.ie> 29165 Date: Thu Jun 15 20:39:22 2006 +1000 29166 29167 ati/r128: add includes to get C function decls 29168 29169 Also fix a void pointer cast 29170 29171 commit f46c3bebe821ee0031770d98d04bd406d1abb9ed 29172 Author: Dave Airlie <airlied (a] linux.ie> 29173 Date: Thu Jun 15 20:34:56 2006 +1000 29174 29175 fix some more unused variables and bad parameter passing 29176 29177 commit 6c3f81683abab8ff6b8f2fcaf01a2a1a63af5ef6 29178 Author: Dave Airlie <airlied (a] linux.ie> 29179 Date: Thu Jun 15 20:25:00 2006 +1000 29180 29181 ati: fix -Wall in atidri.c 29182 29183 atidri.c has some parameter passing bugs, fixed those and cleaned other 29184 warnings. 29185 29186 commit 58c6aac0669101481ef6b8217e5cede4573ece42 29187 Author: Dave Airlie <airlied (a] linux.ie> 29188 Date: Thu Jun 15 19:58:11 2006 +1000 29189 29190 theatre: fix all -Wall warnings in theatre* 29191 29192 This fixes all the warnings, it doesn't clean up the horrible mess that is 29193 theatre*.[ch]. 29194 29195 commit c90c49da3844f93d2d29070b15bb7def72b98b0f 29196 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29197 Date: Thu Jun 15 10:29:46 2006 +0200 29198 29199 Collapse .gitignore files into single toplevel file, sorted with LC_ALL=C. 29200 29201 Suggested by Keith Packard. 29202 29203 commit b7a3e4c9b33f72ae447f99c0c765f24fce988142 29204 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29205 Date: Thu Jun 15 09:45:12 2006 +0200 29206 29207 Fix some implicit function declarations. 29208 29209 commit 556ddc9c41020271e92389a3900d097ec443c625 29210 Author: Michel Dnzer <michel (a] tungstengraphics.com> 29211 Date: Thu Jun 15 09:30:43 2006 +0200 29212 29213 .cvsignore -> .gitignore 29214 29215 commit 4708327969a21fc171dcd80a1b506028cc8e4f90 29216 Author: Dave Airlie <airlied (a] linux.ie> 29217 Date: Thu Jun 15 07:37:03 2006 +1000 29218 29219 ati: remove Changelog from ATI driver. 29220 29221 commit ecbec8379c24bd6c16ced674d5e6affcd3985e46 29222 Author: Michel Daenzer <michel (a] daenzer.net> 29223 Date: Thu Jun 8 17:29:12 2006 +0000 29224 29225 Bug #5678: Fix setting of transfer mode for AGPv3 devices (Tilman 29226 Sauerbeck). 29227 29228 commit 874ff105bc63c4fc242fbabaec04fde2cb81fdd4 29229 Author: Daniel Stone <daniel (a] fooishbar.org> 29230 Date: Sat Jun 3 10:50:33 2006 +0000 29231 29232 Bug #1889: Stop the CCE before an engine reset. (Conn) 29233 29234 commit f9db5595fe4d62c54b3a501b2af64b4e2b7035cd 29235 Author: Michel Daenzer <michel (a] daenzer.net> 29236 Date: Sun May 21 14:33:34 2006 +0000 29237 29238 Derive ATI_VERSION_{MAJOR,MINOR,PATCH} from package version in order to 29239 hopefully confuse bug triagers a little less. 29240 29241 commit 25ba85cab8b3d831da646fef19f01769410fc958 29242 Author: Alex Deucher <agd5f (a] yahoo.com> 29243 Date: Wed May 10 02:46:37 2006 +0000 29244 29245 - remove obsolete option "OverlayOnCRT2" (bug 6488) 29246 29247 commit ede75fe2b1850a4d9f03c724b7c1d1ed65e60f2f 29248 Author: Michel Daenzer <michel (a] daenzer.net> 29249 Date: Tue May 9 14:40:23 2006 +0000 29250 29251 Drop support for radeon DRM <= 1.7. The 2.4 kernel has version 1.7, and 29252 various problems up to system hangs on X server startup have been 29253 reported with that and DRI enabled. 29254 29255 commit bc93a46c4fe499f23b9c0ff52e2e36b4c587f7d6 29256 Author: Michel Daenzer <michel (a] daenzer.net> 29257 Date: Mon May 8 09:29:34 2006 +0000 29258 29259 Bugzilla #4640 <https://bugs.freedesktop.org/show_bug.cgi?id=4640> Patch 29260 #5273 <https://bugs.freedesktop.org/attachment.cgi?id=5273>: 29261 Add cases for FOURCC_RGBA32 and FOURCC_RGB24. This prevents clients from 29262 receiving incorrect data sizes with these formats. (Jan Schmidt) 29263 29264 commit 2b0cdd9448a24ea067b0d78f319b99c1041df2e0 29265 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29266 Date: Wed May 3 04:25:27 2006 +0000 29267 29268 Enable decoding of version 4 TMDS PLL config tables, they actually work, 29269 the problem in the comment only concerned pre-production cards that Hui 29270 tested apparently. Also set back the default value to what it was 29271 initially, fixes some more cards that don't have a proper table (Mac 29272 Mini for example) 29273 29274 commit 36c04b88416905b9ae305583e8e34fefdbfafc50 29275 Author: Michel Daenzer <michel (a] daenzer.net> 29276 Date: Mon May 1 17:49:41 2006 +0000 29277 29278 Bugzilla #6755 <https://bugs.freedesktop.org/show_bug.cgi?id=6755> Patch 29279 #5536 <https://bugs.freedesktop.org/attachment.cgi?id=5536>: 29280 Change HostDataBlit interface to take dst_offset_pitch and coordinates 29281 instead of just a destination pointer, as the latter is not sufficient 29282 with tiling. Also, use HW clipping to avoid overwriting destination 29283 data outside of the specified width. 29284 Adapt to new HostDataBlit interface. This fixes corruption with 29285 UploadToScreen to the front buffer (from exaPutImage). 29286 29287 commit b90d3710e67caad0686efab491c2b76ae877920c 29288 Author: Michel Daenzer <michel (a] daenzer.net> 29289 Date: Sat Apr 29 21:30:23 2006 +0000 29290 29291 Bugzilla #1992 <https://bugs.freedesktop.org/show_bug.cgi?id=1992> Patch 29292 #4988 <https://bugs.freedesktop.org/attachment.cgi?id=4988>: Fix Option 29293 "DisplayPriority" "HIGH" for R300 family. 29294 29295 commit 30f3ce5c5ba205a23748def11de019ee727897f3 29296 Author: Michel Daenzer <michel (a] daenzer.net> 29297 Date: Sat Apr 29 20:34:57 2006 +0000 29298 29299 Properly de-initialize EXA driver data, so EXA can be initialized again in 29300 the next generation. 29301 29302 commit d705429c230c9cc1645ae540a51dba62cae9f23e 29303 Author: Michel Daenzer <michel (a] daenzer.net> 29304 Date: Sat Apr 29 20:18:48 2006 +0000 29305 29306 Fix tests for (un)forcing HW cursor. The broken tests caused problems when 29307 switching to console with 3D clients running. 29308 29309 commit 6142cb079d4a5844e4863c800baf4ae19c636af5 29310 Author: Michel Daenzer <michel (a] daenzer.net> 29311 Date: Sat Apr 29 20:12:38 2006 +0000 29312 29313 Prefer double-buffered visuals. As a consequence, the root window's visual 29314 is double-buffered by default, which is desirable in a number of 29315 situations, in particular when running a GLX compositing manager with 29316 AIGLX. 29317 29318 commit eb841974d041b19461b7fb726687df8938b2e39b 29319 Author: Eric Anholt <anholt (a] freebsd.org> 29320 Date: Fri Apr 28 02:55:08 2006 +0000 29321 29322 Bug #6761: Fix font rendering in non-DRI mode by replacing RB2D_DSTCACHE_* 29323 with RB3D_DSTCACHE_*. The RB2D versions are read-only mirrors of the 29324 RB3D. This is masked when DRI is enabled because the DRM uses the right 29325 registers in its flushing. 29326 29327 commit 85e2ee6de6fd347aad66785ec7119f92013cd994 29328 Author: Dave Airlie <airlied (a] linux.ie> 29329 Date: Wed Apr 26 08:38:47 2006 +0000 29330 29331 Make the ATI driver use the PCI BAR size to determine the MMIO register 29332 aperture size. Otherwise it tries to overmap things if they are up near 29333 the end of memory. 29334 29335 commit 13b19edf5accb7427436d3f0b1c9a9efb4fe98fa 29336 Author: Eric Anholt <anholt (a] freebsd.org> 29337 Date: Wed Apr 26 07:26:15 2006 +0000 29338 29339 Add support for some component-alpha operations (InReverse, OutReverse) 29340 necessary for testing upcoming general component alpha acceleration 29341 assistance in EXA. Tested with rendercheck on an r100. 29342 29343 commit 44a45bcb82d49349f4fdfae3e4b8f24f98d99b46 29344 Author: Eric Anholt <anholt (a] freebsd.org> 29345 Date: Tue Apr 25 23:31:56 2006 +0000 29346 29347 Use a CPPFLAGS with XORG_CFLAGS so that the exa test can actually find its 29348 headers and not fail. Also correct some nearby spelling. 29349 29350 commit cc1825b503ce7e785cbb4611fd177477a1081b6e 29351 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29352 Date: Mon Apr 24 07:44:52 2006 +0000 29353 29354 Fix enabling of external VGA output on some cards with dual entity setup 29355 instead of MergedFB. This was broken due to bogus manipulations of a 29356 register between the 2 entities. This fix is a big "violent" in that it 29357 always enables the analog output regardless of what is plugged. It 29358 should work though (at the expense of some additional power 29359 consumption). A proper fix should be easy to do once Alex gets his 29360 output mapping rework in. 29361 29362 commit a66145bc9d487f7e19237cf292f12750cb959f61 29363 Author: Dave Airlie <airlied (a] linux.ie> 29364 Date: Fri Apr 21 00:38:44 2006 +0000 29365 29366 Bugzilla 6672: fix interrupts on radeon since memmap changes 29367 29368 commit f1f057df9aaba42986d9e20afe6f609a90318333 29369 Author: Dave Airlie <airlied (a] linux.ie> 29370 Date: Fri Apr 21 00:34:43 2006 +0000 29371 29372 Add support for backing up the PCIE GART table on VT switch. This makes 29373 suspend/resume work a lot better on PCIE cards. 29374 29375 commit 31396df3b0101571bb85a4a66b67f5a8b0c992a9 29376 Author: Adam Jackson <ajax (a] nwnk.net> 29377 Date: Sat Apr 8 01:19:17 2006 +0000 29378 29379 Bump to 6.6.0 and release. This module's versioning is completely screwed, 29380 so until we split out mach64/r128/radeon please just increment the 29381 package number whenever you make significant changes. 29382 29383 commit d706f494ed22afd369991e8c75a2d4f5aed15305 29384 Author: Aaron Plattner <aplattner (a] nvidia.com> 29385 Date: Fri Apr 7 23:06:02 2006 +0000 29386 29387 Add a DrawablePtr argument to the XV functions to pave the way for 29388 redirected video. 29389 29390 commit f94685aec26b65f60c00a4b1bcd8f26bd80eb6e4 29391 Author: Eric Anholt <anholt (a] freebsd.org> 29392 Date: Tue Apr 4 07:42:36 2006 +0000 29393 29394 Bug #5632: Fix up ati.4 to explain that it is a wrapper driver for the 29395 other three. While here, touch up radeon.4, and add an authors list 29396 gleaned from checking copyright headers. 29397 29398 commit 58c9b1d84f514755945e7a0d25fd5530af34ce95 29399 Author: Eric Anholt <anholt (a] freebsd.org> 29400 Date: Sun Apr 2 21:45:12 2006 +0000 29401 29402 Add missing .cvsignore lines. 29403 29404 commit f1ce6170ef21e1f9769337c3138cd8d65c33e40a 29405 Author: Alex Deucher <agd5f (a] yahoo.com> 29406 Date: Sat Apr 1 23:02:40 2006 +0000 29407 29408 - Fix dpi when switching from clone to dualhead with MergedFB. 29409 - Add ConstantDPI option to force a particlar dpi across mode changes Both 29410 based on Thomas Winischhofer's sis code. 29411 29412 commit 607f18cfbfc2a7b5509f8b567bf510fce31b361e 29413 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29414 Date: Mon Mar 27 06:12:57 2006 +0000 29415 29416 Fix a regression of my latest fix that broke color tiling on some setups. 29417 RADEONRestoreMemMapRegisters() must not clear CRTC_OFFSET_CNTL and 29418 friends unconditionally but only when there is a map change or it will 29419 clear them after the DRI updates the AGP location, thus turning tiling 29420 back off on the CRTC. Also remove a duplicate call to 29421 RADEONInitCommonRegisters(). 29422 29423 commit bf1c4471137f10b7de9d79519ca2240058997e18 29424 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 29425 Date: Thu Mar 23 18:04:17 2006 +0000 29426 29427 commit the actual changes instead of just the Changelog... 29428 29429 commit aa5fff9bae4818bbf99e8d9caa8f2ce9905294ca 29430 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 29431 Date: Thu Mar 23 17:59:25 2006 +0000 29432 29433 Use programmable tap coefficients for (hopefully) better video image 29434 quality when downscaling. Remove unnecessary code programming the tap 29435 coefficients twice. Disable some code which is supposed to improve 29436 filtering quality but causes very visible image quality problems at 29437 least with rv250 when upscaling. Fix potentially broken FOURCC_RGB16T 29438 format (untested if it was broken, untested if it works now...). Change 29439 some magic numbers to (new) macro defines. 29440 29441 commit 17c52ff15ebe2cf220d5988ee13b599c41799886 29442 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29443 Date: Thu Mar 23 01:37:15 2006 +0000 29444 29445 Call RADEONRestoreCommonRegisters() before RADEONRestoreCrtc2Registers() 29446 instead of after. I can't find any good reason why we did it backward 29447 until now and it definitely causes problems as the workaround for VT 29448 switch in RADEONRestoreCommonRegisters() will break output to the 29449 secondary head in various situations. Also does an unrelated minor 29450 update to a comment. 29451 29452 commit 673e4d7501861066fdd0874919a4524b820b944b 29453 Author: Kristian Hgsberg <krh (a] redhat.com> 29454 Date: Thu Mar 23 00:06:56 2006 +0000 29455 29456 Add atipciids.h. 29457 29458 commit 5dcb4152dca6ac1fe9342f44eb48c0198b15772e 29459 Author: Kristian Hgsberg <krh (a] redhat.com> 29460 Date: Wed Mar 22 22:30:14 2006 +0000 29461 29462 Convert use of xf86fopen() and other xf86 wrapped libc symbols to use libc 29463 symbols directly. The xf86* versions aren't supposed to be used 29464 directly. 29465 Drop libc wrapper; don't include xf86_ansic.h and add includes now missing. 29466 29467 commit 21acdf371e3cf913c5affbd3e86641cfb115311c 29468 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29469 Date: Sat Mar 18 00:08:24 2006 +0000 29470 29471 Clear the offsets when updating the memory map instead of when enabling the 29472 CRTCs and try to make that code more reliable. Doesn't disable the 29473 CRTCs beforehand. 29474 29475 commit 323ecb92e40d71c5ef994b41b6d8dedba6dd6203 29476 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 29477 Date: Fri Mar 17 03:00:53 2006 +0000 29478 29479 Fix various small cosmetic issues. Change a driver message, get the order 29480 right for requesting drm versions, replace the use of some numbers with 29481 the respective macro defines in radeon_video.c, and add some more macro 29482 defines. None of that really matters. 29483 29484 commit dc552edf191e14249421f8b27146874f16b80c1a 29485 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29486 Date: Thu Mar 16 21:53:58 2006 +0000 29487 29488 Add missing return statement (the driver would not start randomly), add 29489 RV410 to the list of new generation cards for the memory map setup and 29490 fix a bogus message. 29491 29492 commit 901409f47cc990721c1648b1c08e23450e6a4452 29493 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29494 Date: Thu Mar 16 04:32:22 2006 +0000 29495 29496 Fix incorrect CRTC2_OFFSET values when using old-style dual head (not 29497 MergedFB). Fixes Xinerama for me. 29498 29499 commit 4d6656062129da0489eb4bc898871379ba891d8f 29500 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29501 Date: Thu Mar 16 03:09:34 2006 +0000 29502 29503 Add various workarounds that seem to fix some remaning lockup scenarios I'm 29504 experiencing with the driver when setting the memory map. Some of the 29505 magic delays are a bit dodgy but they seem to work, I suppose I can't 29506 do better now without help from ATI. Also removed some really too noisy 29507 debug messages. 29508 29509 commit 36799e3ea7ddde5fce528e8bc17ea942933a9111 29510 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29511 Date: Wed Mar 15 04:03:37 2006 +0000 29512 29513 Fix various issues with DRI & server recycle by putting DRICloseSreen in 29514 the normal CloseScreen callback chain at a spot that matches the call 29515 to DRIFinishScreenInit in order to unwrap at the right time. 29516 29517 commit 50923b6a9325592edc8b02b17646f96b9b8c4a01 29518 Author: Luc Verhaegen <libv (a] skynet.be> 29519 Date: Wed Mar 15 00:46:33 2006 +0000 29520 29521 Add some autoconf to keep this driver from building EXA when a slightly 29522 older server with older exa is installed. 29523 The way in which this problem was treated by the actually related 29524 developers explains a lot about this driver. 29525 29526 commit 74f4ed4275866475b3c182fae962df2fc0946b85 29527 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29528 Date: Mon Mar 13 02:57:17 2006 +0000 29529 29530 Fix typo in the function setting the color space transformation factors for 29531 old radeon 29532 29533 commit 65625c6c2053792cd14b54c904bf46f360ef1565 29534 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29535 Date: Mon Mar 13 00:09:44 2006 +0000 29536 29537 Earliest fix introduced a bug due to another chicken & egg problem in card 29538 detection. Now split PreInitConfig into PreInitChipType and PreInitVRAM 29539 so that PreInitDRI can be done just in between. Restores proper AGPx4 29540 functionality (when enabled) 29541 29542 commit 3a290115af41b0ecb94eb201d7e7abb42ebbfec6 29543 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29544 Date: Sun Mar 12 06:53:27 2006 +0000 29545 29546 Stop using xf86PciInfo.h, instead use a local copy of the PCI IDs we need 29547 in atipciids.h so we can update the ATI driver independently of the 29548 server when new chips are added 29549 29550 commit d6b8f0a5eeb332563ec3ec7082f03be1797d26d8 29551 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 29552 Date: Sat Mar 11 04:19:47 2006 +0000 29553 29554 Add #ifdef XF86DRI guards around references to info->allowPageFlip and 29555 info->depthMoves, since they are only defined if building with DRI. 29556 29557 commit 69fe90bc501df9bd9e74b894bfca79a4025725bd 29558 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 29559 Date: Fri Mar 10 13:00:49 2006 +0000 29560 29561 Use "FBTexPercent" option for XAA too (only for increasing memory reserved 29562 for textures). 29563 29564 commit 8033d7423267fc1b3725df6a077dd910b586daaf 29565 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29566 Date: Thu Mar 9 23:26:27 2006 +0000 29567 29568 More memory map fixes !!! This time, we add a way to get the DRM kernel 29569 module version early, in order to properly limit the framebuffer size 29570 when a module that doesn't support the new map is loaded. That will 29571 improve backward compatibility. I also made the minimum required kernel 29572 module version 1.3 which hopefully shouldn't be a problem for anybody, 29573 and cleaned up the ordering in which we test for DRM version to enable 29574 features 29575 29576 commit c637939d3c4ce5cdddbc5200d5ffbca4c3927050 29577 Author: Michel Daenzer <michel (a] daenzer.net> 29578 Date: Thu Mar 9 15:41:16 2006 +0000 29579 29580 RN50: Skip modes that exceed memory bandwidth. 29581 Bugzilla #5766 <https://bugs.freedesktop.org/show_bug.cgi?id=5766> Patch 29582 #4636 <https://bugs.freedesktop.org/attachment.cgi?id=4636> 29583 - Acknowledge that RN50 only has one CRTC, and use this to distinguish it 29584 from RV100. 29585 - Fix detection of RN50 memory type and bus width. 29586 - Model RN50 memory bandwidth limits by capping the pixel clock range based 29587 on memory clock, bpp and memory bus width. (ATI Technologies Inc.) 29588 29589 commit 890b95a3ddad7634f3aea635cb236b221e902acc 29590 Author: Eric Anholt <anholt (a] freebsd.org> 29591 Date: Thu Mar 9 06:06:25 2006 +0000 29592 29593 Update drivers that support EXA for the new EXA ABI. This consists of 29594 moving all the accel and card members into the driver ptr, filling in 29595 the exa_major/ exa_minor fields, and always using LoadSubModule so we 29596 can check the module version up front. Only tested on ATI. 29597 29598 commit e19ccfa18fee046c54d7a9401e4e0098b2300598 29599 Author: Michel Daenzer <michel (a] daenzer.net> 29600 Date: Fri Mar 3 16:41:41 2006 +0000 29601 29602 Add Option "FBTexPercent" to override the amount of video RAM reserved for 29603 OpenGL textures with EXA. 29604 29605 commit ca00035eccb6e3495b27f8e64f2d016d8d2a9ce2 29606 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29607 Date: Thu Mar 2 02:05:17 2006 +0000 29608 29609 Fix the fix ... I didn't interpret PciInfo->size properly and forgot that 29610 the function works in Kb not bytes... Ooops. 29611 29612 commit b2145aea36bb035bff048366c607b967d70fff49 29613 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29614 Date: Wed Mar 1 21:35:14 2006 +0000 29615 29616 Extend the alignement workaround to post-rv280 chips as well (thanks Hui) 29617 and also limit the CPU accessible memory to the size of the PCI BAR 29618 size (yeah, it was passed to the driver, I just got blind for a while 29619 it seems) 29620 29621 commit 163581a82fe9395672c31b41cd67aa44158f5201 29622 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29623 Date: Tue Feb 28 23:34:03 2006 +0000 29624 29625 Workaround for rv280 bug which needs the framebuffer memory mapping to be 29626 aligned to its own size. Also add back limit of 128Mb of mapped fb 29627 space for now until I fully sort out what's up with dual function 29628 cards. I would really need to access the PCI BAR sizes here but that 29629 information seem to be missing from the PCI info passed to the driver. 29630 29631 commit 5ff8fb2bea1fe7d14db53c19ff3a3a56668ec417 29632 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 29633 Date: Tue Feb 28 00:52:53 2006 +0000 29634 29635 Add the newly added (and tons of previously forgotten ones) pci ids to the 29636 ati wrapper too (probably not required for it to work as everything 29637 should default to radeon anyway there). 29638 29639 commit 1574b41a0f894f9b6e30d3ee94e596695a4c64f0 29640 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29641 Date: Sun Feb 26 10:01:32 2006 +0000 29642 29643 Fix page flipping with XAA. The new code ended up calling ShadowFBInit() 29644 before XAAInit(), which breaks ShadowFB. 29645 29646 commit 1f43a584a684af8078631425886cce236500e0e6 29647 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 29648 Date: Sat Feb 25 01:57:05 2006 +0000 29649 29650 Add pci ids known to exist (see #4284 for instance). There are still 29651 entries which probably don't really exist (cancelled cards and such), 29652 leave them as-is. Fix the name of some entries, mostly based on the 29653 .inf file of the newest catalyst driver. Use own family id for rv410 29654 and rs400, though there is no different code (yet?). 29655 29656 commit 5c141bb15d1163e04c012a0cdf0699d534f0be37 29657 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29658 Date: Thu Feb 16 23:27:44 2006 +0000 29659 29660 Memory map and misc fixes. This completely reworks the way the driver 29661 detects available & accessible video memory and initializes the card 29662 internal memory map. Along the way, I re-ordered some of the stuffs in 29663 pre-init in order to remove some dodgy codepath, fixed various init, 29664 exit and vt switch issues, and overall made the driver more robust. The 29665 full memory map fix requires an updated DRM. The X driver no longer 29666 tries to position the AGP aperture, this becomes entirely DRM 29667 responsibility. 29668 29669 commit 8c6e5a6eb50d371fd4ec700c26aa628cb5f45731 29670 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29671 Date: Wed Feb 15 23:36:55 2006 +0000 29672 29673 The engine setup would trigger a bogus line write before the proper 29674 addresses are setup, thus causing the card to try to bus master over 29675 the system bus to some stale location (usually 0 but could depend on 29676 whatever was used before X). Remove that useless init bit, and always 29677 setup some good enough engine source & destination pointers in case we 29678 have some other spurrious engine activity going on. 29679 29680 commit d4ec0223029d8fe03a836a5a6e9a7cc8897b4c90 29681 Author: Benjamin Herrenschmidt <benh (a] kernel.crashing.org> 29682 Date: Mon Feb 13 05:30:13 2006 +0000 29683 29684 Fix EXA acceleration of textures with byteswap or tiling From John Clemens 29685 <john (a] deater.net> 29686 29687 commit 79c60ebbc34aaf0ac0b0723898d5b63015547178 29688 Author: Eric Anholt <anholt (a] freebsd.org> 29689 Date: Tue Feb 7 04:39:59 2006 +0000 29690 29691 Fix a copy'n'paste-o in the EXA render implementation that resulted in 29692 [ax]8b8g8r8 pict formats not being supported on r200, but no other 29693 issues. 29694 29695 commit 8d74f3d49e73626f5d658ba6586a7c1487dae141 29696 Author: Daniel Stone <daniel (a] fooishbar.org> 29697 Date: Thu Jan 19 14:20:37 2006 +0000 29698 29699 Bug #5656: Set all Radeon XPRESS 200 R[CS]4xx-based chips to have IsIGP = 29700 1, fixing modesetting issues. 29701 29702 commit e490f4b86e6bba70f7c884fca3cf41d159323689 29703 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 29704 Date: Fri Jan 13 01:37:43 2006 +0000 29705 29706 XF86_VERSION_CURRENT -> XORG_VERSION_CURRENT 29707 29708 commit ac296d9137b02753bc79a9930d168855adea9c95 29709 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 29710 Date: Fri Jan 13 01:02:01 2006 +0000 29711 29712 When failing to parse MetaModes, reset CRT2pScrn to NULL after free'ing it 29713 so we don't crash when trying to use it later (such as in 29714 RADEONFreeRec). 29715 29716 commit dd98ce78407c9eddd7adaad1d73d6910e6949b87 29717 Author: Adam Jackson <ajax (a] nwnk.net> 29718 Date: Sun Jan 8 19:02:16 2006 +0000 29719 29720 Bug #5523: Make radeon compile without USE_XAA. (Tilman Sauerbeck) 29721 29722 commit a99429b2bc6407ff2a094e6ee30ce2032f734966 29723 Author: Kevin E Martin <kem (a] kem.org> 29724 Date: Wed Dec 21 02:29:59 2005 +0000 29725 29726 Update package version for X11R7 release. 29727 29728 commit 96e225b950c74e64f3f788be8a8489c5075ea55a 29729 Author: Adam Jackson <ajax (a] nwnk.net> 29730 Date: Mon Dec 19 16:25:52 2005 +0000 29731 29732 Stub COPYING files 29733 29734 commit 6bea785ea0bdf01673a6ddc45765ae7372fdd371 29735 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 29736 Date: Mon Dec 19 09:07:50 2005 +0000 29737 29738 Typo fixes, mailing list & url updates, and other changes to prepare for 29739 X11R6.9 & 7.0 releases. 29740 29741 commit 07d5cf8fc37eb72dbb344081ff07d59dbf568e9c 29742 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 29743 Date: Thu Dec 15 23:15:37 2005 +0000 29744 29745 Bugzilla #5238 <https://bugs.freedesktop.org/show_bug.cgi?id=5238> Patch 29746 #3978 <https://bugs.freedesktop.org/attachment.cgi?id=3978> r128 man 29747 page dualhead note update (Alex Deucher) 29748 29749 commit 6db5cbfbf6d405d9871892d1255dd7442e4539e9 29750 Author: Kevin E Martin <kem (a] kem.org> 29751 Date: Thu Dec 15 00:21:32 2005 +0000 29752 29753 Update package version number for final X11R7 release candidate. Bump 29754 driver version number. 29755 29756 commit 726fd3a81dfbd1b0c8e4ede104a48523daecab92 29757 Author: Adam Jackson <ajax (a] nwnk.net> 29758 Date: Wed Dec 14 19:41:28 2005 +0000 29759 29760 Bug #1760: Fix ati wrapper multihead to be dlloader friendly. 29761 29762 commit 328a2d74dabe32ce35e879799412666feedb0eef 29763 Author: Adam Jackson <ajax (a] nwnk.net> 29764 Date: Fri Dec 9 18:25:02 2005 +0000 29765 29766 Bug #3869: Don\'t walk off the end of an array in connector detection. 29767 (Eric Anholt) 29768 29769 commit 03961a05438e916b6725f4e547f23a84ead13c5e 29770 Author: Adam Jackson <ajax (a] nwnk.net> 29771 Date: Fri Dec 9 18:15:31 2005 +0000 29772 29773 Bug #1106: Fix memory allocation for Render. (Michel Daenzer) 29774 29775 commit 091e5c584124f9b849c0f46802b2f0d8b0cc2388 29776 Author: Kevin E Martin <kem (a] kem.org> 29777 Date: Thu Dec 8 17:55:18 2005 +0000 29778 29779 Add configure options to allow hard-coded paths to be changed. 29780 29781 commit e38e2df15c9a4b08dee6386a225e98827bb35aad 29782 Author: Kevin E Martin <kem (a] kem.org> 29783 Date: Thu Dec 8 17:54:40 2005 +0000 29784 29785 Allow hard-coded paths to be configurable. 29786 29787 commit b8203604bf9bd3813668313a1c8c17040b9d93ce 29788 Author: Kevin E Martin <kem (a] kem.org> 29789 Date: Tue Dec 6 22:48:33 2005 +0000 29790 29791 Change *man_SOURCES ==> *man_PRE to fix autotools warnings. 29792 29793 commit a86920e812bfd40ec9facfc234220e83066dd263 29794 Author: Kevin E Martin <kem (a] kem.org> 29795 Date: Sat Dec 3 05:49:32 2005 +0000 29796 29797 Update package version number for X11R7 RC3 release. 29798 29799 commit 010adc8988846e6f76f111a9fd32eca021d12fbc 29800 Author: Kevin E Martin <kem (a] kem.org> 29801 Date: Sat Dec 3 05:45:17 2005 +0000 29802 29803 Bump version numbers for X11R6.9/X11R7 RC3. 29804 29805 commit a5efa6724e7fd8318e262060b084630632dfcb44 29806 Author: Kevin E Martin <kem (a] kem.org> 29807 Date: Fri Dec 2 02:16:07 2005 +0000 29808 29809 Remove extraneous AC_MSG_RESULT. 29810 29811 commit feeef3cba1f4736a96daf3002ec902ced2874198 29812 Author: Adam Jackson <ajax (a] nwnk.net> 29813 Date: Wed Nov 30 18:51:22 2005 +0000 29814 29815 Bump libdrm dep to 2.0. 29816 29817 commit 7ca2b3fa30fedd07951f7c6e73ed4d2a6f6606ea 29818 Author: Adam Jackson <ajax (a] nwnk.net> 29819 Date: Tue Nov 29 23:29:58 2005 +0000 29820 29821 Only build dlloader modules by default. 29822 29823 commit e167e5151e32d54919e96cbfdcf932e16b1cab2a 29824 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 29825 Date: Mon Nov 28 22:04:08 2005 +0000 29826 29827 Change *mandir targets to use new *_MAN_DIR variables set by xorg-macros.m4 29828 update to fix bug #5167 (Linux prefers *.1x man pages in man1 subdir) 29829 29830 commit 0015b1a007bed61aaf389f8f927e4ae0568d7da7 29831 Author: Eric Anholt <anholt (a] freebsd.org> 29832 Date: Mon Nov 21 10:49:09 2005 +0000 29833 29834 Add .cvsignores for drivers. 29835 29836 commit 00b8f6ccfbb5163c47e9def0e4ecc6ac7be7e8af 29837 Author: Eric Anholt <anholt (a] freebsd.org> 29838 Date: Fri Nov 11 11:05:42 2005 +0000 29839 29840 Bug #1028: Don't (incorrectly) mark the 24/8 z/stencil case as slow, which 29841 resulted in whining at runtime about the visual not being supported. 29842 29843 commit 33326847ccc9cf4cd6a23a257e5392e348e6dd22 29844 Author: Kevin E Martin <kem (a] kem.org> 29845 Date: Wed Nov 9 21:31:20 2005 +0000 29846 29847 Bump version numbers for X11R6.9/X11R7 RC2. 29848 29849 commit 68e11246bd6f6a641807230834dce4e7fd6911af 29850 Author: Kevin E Martin <kem (a] kem.org> 29851 Date: Wed Nov 9 21:15:14 2005 +0000 29852 29853 Update package version number for X11R7 RC2 release. 29854 29855 commit 1582d3b34643babcbb794d2df8d7c9a755d569c9 29856 Author: Matthias Hopf <mhopf (a] suse.de> 29857 Date: Wed Nov 9 17:42:57 2005 +0000 29858 29859 Bug #4998: Add ';' and ',' as valid separators for MetaModes as well. 29860 29861 commit 2d22d78d157e07e6ec6a3e2d03fc17f102f6e00e 29862 Author: Kevin E Martin <kem (a] kem.org> 29863 Date: Wed Nov 9 07:30:03 2005 +0000 29864 29865 Fix DoLoadableServer NO build. 29866 29867 commit 999f67763a5ea9fef7b7b576262f0da079efcfee 29868 Author: Michel Daenzer <michel (a] daenzer.net> 29869 Date: Tue Nov 8 16:30:48 2005 +0000 29870 29871 bugzilla #988 (https://bugs.freedesktop.org/show_bug.cgi?id=988) Fix typo 29872 which may or may not have had a negative impact on stability with R300 29873 class cards. 29874 29875 commit 5aa401efa0b8d81da418eafeb857142a33f85da9 29876 Author: Michel Daenzer <michel (a] daenzer.net> 29877 Date: Tue Nov 8 14:05:00 2005 +0000 29878 29879 bugzilla #4494 (https://bugs.freedesktop.org/show_bug.cgi?id=4494) Call 29880 RADEONDGAInit() even if EXA is enabled; the problematic DGA 29881 acceleration hooks are already disabled in that case. 29882 This fixes DGA issues such as DGA mouse (typically used by games in 29883 fullscreen mode) not working. 29884 Fix suggested by Eric Anholt. 29885 29886 commit 345180131a7b2752773dc2a3ee1221ad1e05c815 29887 Author: Michel Daenzer <michel (a] daenzer.net> 29888 Date: Tue Nov 8 13:35:23 2005 +0000 29889 29890 bugzilla #4916 (https://bugs.freedesktop.org/show_bug.cgi?id=4916) Fix 29891 segfault with MergedFB and EXA enabled and HW cursor temporarily 29892 disabled. 29893 29894 commit 23911421ffa7bdd2e191a72c41bf1f26c92c1a67 29895 Author: Michel Daenzer <michel (a] daenzer.net> 29896 Date: Tue Nov 8 09:44:41 2005 +0000 29897 29898 bugzilla #4951 (https://bugs.freedesktop.org/show_bug.cgi?id=4951) 29899 attachment #3689 (http://bugs.freedesktop.org/attachment.cgi?id=3689) 29900 Fix HW cursor getting silently disabled with EXA. 29901 29902 commit e1259c0117a190cf4c6e7be07d5aa83e03afb3a0 29903 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 29904 Date: Mon Nov 7 19:28:40 2005 +0000 29905 29906 Replaced the variadic macros(gcc) by macros according to C99 standard 29907 29908 commit 67bdaae04ab37f38a6d0c8dc4b6328c3ba3b3393 29909 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 29910 Date: Sun Nov 6 07:41:29 2005 +0000 29911 29912 Sun bug #6325220: Xorg dumps core in dual monitor configuration with radeon 29913 driver if CRT2Position not specified in xorg.conf 29914 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6325220> 29915 (Stuart Krietman - Sun Microsystems) 29916 29917 commit 733ebf6e4a7c7b2f1a861077d43da2a853e4c0d7 29918 Author: Kevin E Martin <kem (a] kem.org> 29919 Date: Tue Nov 1 15:08:52 2005 +0000 29920 29921 Update pkgcheck depedencies to work with separate build roots. 29922 29923 commit dd0eb76337bfdaec8f456380a6ced9e0487e190f 29924 Author: Adam Jackson <ajax (a] nwnk.net> 29925 Date: Fri Oct 21 18:40:18 2005 +0000 29926 29927 Bug #4829: Remove some useless DRI symbol checks. 29928 29929 commit e6c1859e3a7c84b882e0c30d0783255ed1258d9e 29930 Author: Donnie Berkholz <spyderous (a] gentoo.org> 29931 Date: Thu Oct 20 23:01:58 2005 +0000 29932 29933 Bug #4831 <https://bugs.freedesktop.org/show_bug.cgi?id=4831>. Check for 29934 x86_64 in addition to amd64 in host_cpu. 29935 29936 commit 33460fb393c9591babcc43b30a73eef3971765c2 29937 Author: Kevin E Martin <kem (a] kem.org> 29938 Date: Wed Oct 19 02:48:02 2005 +0000 29939 29940 Update package version number for RC1 release. 29941 29942 commit 77d5c3549db4f6a7ff9f969f885d7a4e58ab2b16 29943 Author: Kevin E Martin <kem (a] kem.org> 29944 Date: Tue Oct 18 15:46:37 2005 +0000 29945 29946 Do the following to make the drivers pass distcheck: 29947 - Only define EXTRA_DIST when there are actually man pages to be processed 29948 - Add missing man pages to keyboard and ati drivers EXTRA_DIST 29949 29950 commit 35e717078c840414b0d6a0e37b3ffed0a7b102aa 29951 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 29952 Date: Tue Oct 18 00:01:52 2005 +0000 29953 29954 Use @DRIVER_MAN_SUFFIX@ instead of $(DRIVER_MAN_SUFFIX) in macro 29955 substitutions to work better with BSD make 29956 29957 commit ff6e5b0412fe347575992c83bb7cb47b6f96f390 29958 Author: Adam Jackson <ajax (a] nwnk.net> 29959 Date: Mon Oct 17 22:57:32 2005 +0000 29960 29961 More 1.7 braindamage: define EXTRA_DIST in terms of @DRIVER_NAME@ instead 29962 of indirectly 29963 29964 commit a2e7c55850d383e3af948d445b1645d8a47811cf 29965 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 29966 Date: Mon Oct 17 00:09:01 2005 +0000 29967 29968 Use sed & cpp to substitute variables in driver man pages 29969 29970 commit 524fb1440760a14c53fb3f238aad6c3cdb7e6a08 29971 Author: Dave Airlie <airlied (a] linux.ie> 29972 Date: Sun Oct 9 10:36:50 2005 +0000 29973 29974 Secure the Radeon PCIE GART tables in framebuffer memory by not allowing 29975 the clients to get access to a secure area of framebuffer. 29976 29977 commit e526211b86ec5343002978debfcf78f74a10c69e 29978 Author: Sren Sandmann Pedersen <sandmann (a] daimi.au.dk> 29979 Date: Tue Oct 4 20:01:03 2005 +0000 29980 29981 Check in generated README files 29982 29983 commit 431d0e144ae4c294e1f6110fc3db497ce9bd9cfd 29984 Author: Sren Sandmann Pedersen <sandmann (a] daimi.au.dk> 29985 Date: Mon Oct 3 21:22:07 2005 +0000 29986 29987 - For all drivers that have a <driver>.sgml file, add code in their build 29988 system to build the README file at make dist time 29989 - in util/macros/xorg-macros.m4, add a new XORG_CHECK_LINUXDOC macro that 29990 will check if the required tools and files exist, and if so set a 29991 conditional. 29992 - util/modular/symlink.sh 29993 - Link all the <driver>.sgml to xf86-video-<driver>/README.sgml 29994 - Add all the README.<driver> to the list of excluded files 29995 - xc/programs/Xserver/hw/xfree86/doc/sgml/SiS.sgml: Various changes to make 29996 it spew less warnings when the text file is built. 29997 29998 commit 9147c20e5ff662ac511a19e7623b406dffee2242 29999 Author: Adam Jackson <ajax (a] nwnk.net> 30000 Date: Mon Oct 3 06:14:58 2005 +0000 30001 30002 Bug #4001: Don't crash when PanelSize is unset. (Alan Coopersmith) 30003 30004 commit 7e2090e3ebf0bd699d66d336353c097d658a86ae 30005 Author: Matthieu Herrb <matthieu.herrb (a] laas.fr> 30006 Date: Sat Oct 1 20:42:06 2005 +0000 30007 30008 missed one missing return value. 30009 30010 commit a843e4281134de1545d538cb8970179b1d6977f6 30011 Author: Matthieu Herrb <matthieu.herrb (a] laas.fr> 30012 Date: Sat Oct 1 17:09:48 2005 +0000 30013 30014 add missing return value on success in RADEONSetupMemXAA(). 30015 30016 commit 98f7d043f894c73fecaf14440c1a24b6c2fc4b06 30017 Author: Sren Sandmann Pedersen <sandmann (a] daimi.au.dk> 30018 Date: Wed Sep 28 15:10:38 2005 +0000 30019 30020 Make ati driver distcheck 30021 radeon_commonfuncs.c \ 30022 radeon_exa_render.c \ 30023 radeon_exa_funcs.c \ 30024 30025 commit e7b1af5d1677e53820dd1c0926cc8b5a4284901e 30026 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 30027 Date: Tue Sep 27 23:55:15 2005 +0000 30028 30029 Move the RADEONEnablePageFlip/RADEONChangeSurfaces calls back to the end of 30030 the function (the code above will cause pixmap cache to be freed and 30031 copied around, and these functions must be called after that otherwise 30032 pixmap damage will occur). May be related to #4440, #4456 (unclear if 30033 these setups involve 3d apps) 30034 30035 commit bc3c26fe365347c427e04385c1df95a5eec14f28 30036 Author: Eric Anholt <anholt (a] freebsd.org> 30037 Date: Tue Sep 27 23:34:11 2005 +0000 30038 30039 Fix Radeon MMIO Render acceleration with EXA by writing out floating- point 30040 coordinates rather than integers. 30041 30042 commit 58dfb95c446b73fbc34784f0894cf89152dab043 30043 Author: Adam Jackson <ajax (a] nwnk.net> 30044 Date: Sat Sep 24 18:45:45 2005 +0000 30045 30046 Disable the {Open,Close}FullScreen DRI protocol. Remove empty FullScreen 30047 stubs from drivers, comment the non-empty ones. 30048 30049 commit 65b5f0017e29903565aff0b7527d955ed016c9a6 30050 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 30051 Date: Tue Sep 20 16:53:34 2005 +0000 30052 30053 Only include radeon_dri.h & radeon_sarea.h if XF86DRI is defined. 30054 Include "compiler.h" to define __inline__ on non-gcc compilers 30055 30056 commit 0b7441b19a1a5df91aca6b1a02944afb636bc523 30057 Author: Eric Anholt <anholt (a] freebsd.org> 30058 Date: Sun Sep 18 02:32:23 2005 +0000 30059 30060 Break EXA ABI while we still can. Add coordinates to the UploadToScreen 30061 hook so we can upload a subset of a pixmap, and convert the current 30062 drivers to respect that. Use this support to directly UploadToScreen in 30063 exaGlyphs, providing a 47.4% +/-2.4% decrease in wall time for ls -lR 30064 programs/Xserver in an antialiased gnome-terminal on an M6 (n=3, caches 30065 hot). I would have bumped major version, only I can't tell what the 30066 EXA_VERSION_* is supposed to be doing as opposed to the module version. 30067 30068 commit 64e7db2698ee48af92594c16d72e33c16372c9d7 30069 Author: Eric Anholt <anholt (a] freebsd.org> 30070 Date: Sat Sep 17 20:02:01 2005 +0000 30071 30072 - Don't try to upload 0 byte-per-pixel (PICT_a1) data using 30073 RADEONHostDataBlit. 30074 - Disable the shortcut for switching from 3d to 3d in radeon_exa.c. It 30075 appears that we do need the cache flush here, thought it's not clear 30076 why. Disable the 2d to 2d shortcut while here, since I'm unsure of what 30077 we're doing. Exposed by the following bit: 30078 - Bug #4485: Add a new routine, exaGlyphs, to handle font drawing. Glyphs 30079 were being accumulated in from non-migratable scratch pixmaps, causing 30080 the destination pixmap to move towards screen but the migration 30081 necessary for source never to happen, leading to abysmal performance. 30082 Instead, copy the scratch glyph data into a real pixmap first, then 30083 composite from that into the destination, allowing for migration. time 30084 ls -lR from programs/Xserver showed 26.9% (+/- 6.3%) decrease in wall 30085 time (n=3). 30086 - Create exaDrawableUse* wrapping exaPixmapUse*, but which are aware of 30087 windows needing backing store. Makes migration code prettier, and 30088 ensures that composited windows will be migrated as normal when we turn 30089 off cw for EXA. (issue brought up by keithp) 30090 30091 commit 3c71198a088e71a105e274f176c8ceb503615d46 30092 Author: Dave Airlie <airlied (a] linux.ie> 30093 Date: Sat Sep 17 07:47:51 2005 +0000 30094 30095 Switch radeon driver to using a card type rather than having a IsPCI flag, 30096 allow for PCI/AGP/PCIE cards. Set PCI gart allocation to only happen on 30097 PCIE cards. Add Radeon PCIE card detection using capability bits. 30098 30099 commit 10cf3523f236a017a1ca7e5ee57e02176b924059 30100 Author: Eric Anholt <anholt (a] freebsd.org> 30101 Date: Wed Sep 14 22:06:41 2005 +0000 30102 30103 Bug #2573: Remove the disabling of automatic cache flushing on non-r300 30104 cards. It has been implicated in hanging at least r300s and some r100s, 30105 and seemed to be the cause of some bad rendering for me. 30106 30107 commit d68894a463a3bb08c97e91ff6fc4b65b29183ff6 30108 Author: Eric Anholt <anholt (a] freebsd.org> 30109 Date: Tue Sep 13 22:04:12 2005 +0000 30110 30111 Remove the hardcoded USE_EXA/USE_XAA defines now that they've been 30112 hardcoded in configure.ac (config.h). 30113 30114 commit baa9b5b6e19cf9794bf810e3c582bf1d7f808663 30115 Author: Dave Airlie <airlied (a] linux.ie> 30116 Date: Mon Sep 12 09:27:07 2005 +0000 30117 30118 Fix Radeon EXA for when RENDER isn't defined 30119 30120 commit 8c89610389197dc94d3bceaf44d9d2ea91848f1b 30121 Author: Dave Airlie <airlied (a] linux.ie> 30122 Date: Mon Sep 12 09:17:42 2005 +0000 30123 30124 Move GART offset/size outside XAA 30125 30126 commit eba44d9747adde8385f2a242dee5a8e73ce5a205 30127 Author: Daniel Stone <daniel (a] fooishbar.org> 30128 Date: Mon Sep 12 09:04:36 2005 +0000 30129 30130 Define USE_XAA and USE_EXA for the time being. 30131 30132 commit a255e7b0984890daad98b74ace0c572965fe61f5 30133 Author: Eric Anholt <anholt (a] freebsd.org> 30134 Date: Sun Sep 11 21:00:44 2005 +0000 30135 30136 Attempt to fix up the modular pieces for EXA support. (untested) 30137 30138 commit 3e0d9c945a6a71cc476d27341ef18618529a91a8 30139 Author: Eric Anholt <anholt (a] freebsd.org> 30140 Date: Sun Sep 11 20:58:53 2005 +0000 30141 30142 Add support for EXA to the radeon driver. Building EXA and XAA support is 30143 controlled at compile time, plus the runtime option of Option 30144 "AccelMethod" "EXA" or "XAA". The XAA support appears to remain as 30145 before, while the EXA pieces need just a little more polishing. Notable 30146 features: 30147 - Render acceleration working on Radeon 100 and 200-series with DRI on. 30148 - DRI works with EXA Notable issues: 30149 - DGA disabled in the EXA case. 30150 - Backbuffer moves disabled in the EXA case. 30151 - No textured XVideo. 30152 - MMIO render acceleration is close but still has some issues. 30153 - Memory pressure while using Composite is really troublesome with DRI on. 30154 This patch is based on an initial patch by Zack Rusin, with significant 30155 work by Benjamin Herrenschmidt and myself. 30156 30157 commit fd62082b68ac3aadd8ffc441352d75d88334904e 30158 Author: Dave Airlie <airlied (a] linux.ie> 30159 Date: Sun Sep 11 08:51:38 2005 +0000 30160 30161 Add support for allocating PCI GART table in framebuffer memory. This is 30162 needed to support PCIE Radeons using a new DRM. 30163 30164 commit cb63f8d9c0563fb0eff28e2be6d4adf5666540d2 30165 Author: Dave Airlie <airlied (a] linux.ie> 30166 Date: Tue Sep 6 10:03:19 2005 +0000 30167 30168 Handle VIP timeouts more gracefully -- impose a hard limit of 10ms on 30169 waiting for VIP_BUSY so we don't lock up hard, spinning. 30170 30171 commit dbd83cf18e1bf5e7e31e1ac150227bd27e8b1739 30172 Author: Adam Jackson <ajax (a] nwnk.net> 30173 Date: Sun Aug 28 20:51:14 2005 +0000 30174 30175 Bug #4148: Bump required DRM version for r300 to 1.17. 30176 30177 commit ce0589ff3b9fcf8c6bab64da4dba40ba540a6a03 30178 Author: Adam Jackson <ajax (a] nwnk.net> 30179 Date: Sun Aug 28 18:10:34 2005 +0000 30180 30181 Bug #1109: Fix VGA init on Rage Mobility 7500 (Marc Le France) 30182 30183 commit 250c8732a54924eeb459880f2905f8e59d038e60 30184 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 30185 Date: Sun Aug 28 18:00:23 2005 +0000 30186 30187 Modified the licens type from GPL to a X/MIT one 30188 30189 commit 94c0196652c3cfa0558f61b55f5df9bf50f9ac49 30190 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 30191 Date: Fri Aug 26 20:04:01 2005 +0000 30192 30193 #include "extensions/dpms.h" -> <X11/extensions/dpms.h> 30194 30195 commit 107e42cabe1cc41fe86b0218c3229b5f34b2df84 30196 Author: Daniel Stone <daniel (a] fooishbar.org> 30197 Date: Fri Aug 26 07:00:24 2005 +0000 30198 30199 Fix 'ugly' warning in r128. (Branden Robinson, Debian #069) 30200 30201 commit d3f5911c85ccfaaf50c05ce80d56dac13fd15701 30202 Author: Daniel Stone <daniel (a] fooishbar.org> 30203 Date: Fri Aug 26 05:30:52 2005 +0000 30204 30205 Only use vgahw symbols if we're building vgahw for this architecture. TODO: 30206 Fix this for the modular tree (maybe an unconditional yes?). Define 30207 that we're using a DPMS server and include DPMS headers. (Debian #024b) 30208 30209 commit 879213455740d2d653d38c712275f07610e4dc9c 30210 Author: Alex Deucher <agd5f (a] yahoo.com> 30211 Date: Fri Aug 26 02:11:24 2005 +0000 30212 30213 - fix pitch increment (Aapo Tahkola, Roland Scheidegger) 30214 30215 commit 6519cc71e4387ef979de90be517145611511ff1a 30216 Author: Daniel Stone <daniel (a] fooishbar.org> 30217 Date: Thu Aug 25 03:34:58 2005 +0000 30218 30219 Fix bug in Radeon connector table parsing -- parse all entries, not just 30220 the first. (Ben Herrenschmidt) 30221 30222 commit 8734a5d0bafb60f5e38b2cee10587f01fc016379 30223 Author: Daniel Stone <daniel (a] fooishbar.org> 30224 Date: Thu Aug 25 03:07:50 2005 +0000 30225 30226 Blacklist Radeon M300 from multimedia I2C code. Handle I2C timeouts more 30227 gracefully -- impose a hard limit of 10ms on waiting for I2C_GO so we 30228 don't lock up hard, spinning. 30229 30230 commit 578564b4f7fce1f02d056c2b12ffb7803d268829 30231 Author: Adam Jackson <ajax (a] nwnk.net> 30232 Date: Mon Aug 22 03:05:52 2005 +0000 30233 30234 Explicit license statement. 30235 30236 commit c9005c91bf6aaf4728a6606284f5d24e0b4a8282 30237 Author: Daniel Stone <daniel (a] fooishbar.org> 30238 Date: Thu Aug 18 09:03:46 2005 +0000 30239 30240 Update autogen.sh to one that does objdir != srcdir. 30241 30242 commit 508a7421a1386b5b783a867d860b858a55f216d3 30243 Author: Alan Hourihane <alanh (a] fairlite.demon.co.uk> 30244 Date: Mon Aug 15 07:30:05 2005 +0000 30245 30246 Egbert's 64bit fixes for mixed 32/64bit clients 30247 30248 commit 204290412fbe8b4ac1a0f4f97bc7aec2847a1df7 30249 Author: Daniel Stone <daniel (a] fooishbar.org> 30250 Date: Sun Aug 14 16:14:51 2005 +0000 30251 30252 Make DGA, CPIO, non-PCI, and TV-Out support detected per-platform, using 30253 the same semantics as in drivers/ati/Imakefile. 30254 30255 commit adb093bdf3cea8108a40bcd72f09924b416de0c1 30256 Author: Alex Deucher <agd5f (a] yahoo.com> 30257 Date: Mon Aug 8 23:42:36 2005 +0000 30258 30259 - radeon XPRESS chips are actually RV3xx based 30260 30261 commit 9c11018beb186b4b303524685fb62629b9ca863e 30262 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30263 Date: Sun Aug 7 14:46:55 2005 +0000 30264 30265 Modified: 30266 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 30267 Update warning about DRI support on R300 and later cards. 30268 30269 commit a28b2c76c7df2fe65a93eed34bc85c2cf5afb314 30270 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30271 Date: Sun Aug 7 14:40:58 2005 +0000 30272 30273 Modified: 30274 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 30275 Properly configure fbLocation and related registers for R300 cards. (Dave 30276 Airlie) 30277 30278 commit 692fb0f8b51abeccaf70b51fc1defc62132b15e9 30279 Author: Alex Deucher <agd5f (a] yahoo.com> 30280 Date: Sat Aug 6 17:23:16 2005 +0000 30281 30282 - missing line in mergedfb option parsing. Spotted by Christoph Bauer 30283 30284 commit 14e77a39320a7a074b5c2eeeea9750b469abc978 30285 Author: Ian Romanick <idr (a] us.ibm.com> 30286 Date: Fri Aug 5 23:04:32 2005 +0000 30287 30288 I have resolved this by making all the "Direct rendering disabled" messages 30289 X_WARNING and all the "Direct rendering enabled" messages X_INFO. 30290 Ideally, we need to factor some of this stuff out to common code. There 30291 is a *LOT* of duplication among the *_dri.c files under 30292 programs/Xserver/hw/xfree86/drivers. 30293 I'm trying to clean up as much low hanging fruit as I can and got tired of 30294 waiting for ajax. :P 30295 30296 commit 8e88d36f78c13025847310b411402f1d46a2e0cd 30297 Author: Adam Jackson <ajax (a] nwnk.net> 30298 Date: Sun Jul 31 17:19:27 2005 +0000 30299 30300 Post-merge fixups: 30301 - Disable building the gamma driver 30302 - Fix numerous Imakefiles to get header paths and subdirs right 30303 - Symlink client-side GLX lib source from Mesa 30304 - Add new shader subdirs to the Mesa core build 30305 - Tweak the DDX drivers to not ask for "xf86dri.h"; they don't need it, 30306 since they don't generate protocol, and it was breaking the build. 30307 30308 commit fb9f712fd1044feb8ce6a99aabaeb5966123e268 30309 Author: Kevin E Martin <kem (a] kem.org> 30310 Date: Fri Jul 29 21:22:42 2005 +0000 30311 30312 Various changes preparing packages for RC0: 30313 - Verify and update package version numbers as needed 30314 - Implement versioning scheme 30315 - Change bug address to point to bugzilla bug entry form 30316 - Disable loadable i18n in libX11 by default (use --enable-loadable-i18n to 30317 reenable it) 30318 - Fix makedepend to use pkgconfig and pass distcheck 30319 - Update build script to build macros first 30320 - Update modular Xorg version 30321 30322 commit ee45e1fb5da793fa14770b28fd662e69d52b2c6a 30323 Author: Michel Daenzer <michel (a] daenzer.net> 30324 Date: Fri Jul 29 19:45:14 2005 +0000 30325 30326 bugzilla #3911 (https://bugs.freedesktop.org/show_bug.cgi?id=3911) 30327 attachment #3191 (http://bugs.freedesktop.org/attachment.cgi?id=3191) 30328 Disable bus mastering while updating MC_FB_LOCATION and friends to 30329 prevent the X server from hanging on startup every now and then under 30330 some circumstances. (ATI Technologies Inc.) 30331 30332 commit 44a66ec5a01c3b4dfa046a66b80a3b7e9dfff537 30333 Author: Sren Sandmann Pedersen <sandmann (a] daimi.au.dk> 30334 Date: Fri Jul 15 15:02:47 2005 +0000 30335 30336 Install theathre drivers in @moduledir@/multimedia 30337 30338 commit dbcf104689ca368523715eb5945bcd800985799b 30339 Author: Kevin E Martin <kem (a] kem.org> 30340 Date: Fri Jul 15 03:39:58 2005 +0000 30341 30342 Fix filename typo 30343 30344 commit 2256fabf1cd69c950639f236f61e59e68cc8e6fa 30345 Author: Kevin E Martin <kem (a] kem.org> 30346 Date: Thu Jul 14 22:30:47 2005 +0000 30347 30348 Fix check for headers. Use {XORG,DRI}_CFLAGS subst instead of replacing 30349 CFLAGS. 30350 30351 commit 520317863ffb365eca17c96b4d545adb554bbbd6 30352 Author: Kevin E Martin <kem (a] kem.org> 30353 Date: Thu Jul 14 19:35:48 2005 +0000 30354 30355 Make compiling with DRI support configurable/autodetecable. Autodetect DRI 30356 sdk headers. Expand help string for xorg-module-dir option. 30357 30358 commit ab5264838ae34bb29791bc08d5a4f853a995faf1 30359 Author: Sren Sandmann Pedersen <sandmann (a] daimi.au.dk> 30360 Date: Thu Jul 14 17:10:01 2005 +0000 30361 30362 fuck you cvs 30363 30364 commit a608c4422665ce1e933483b7c2a25ab2daae5930 30365 Author: Sren Sandmann Pedersen <sandmann (a] daimi.au.dk> 30366 Date: Thu Jul 14 16:03:36 2005 +0000 30367 30368 Try adding build system for ATI driver again. Who knows, maybe CVS works 30369 today. 30370 30371 commit d715561d28d82591adf5c7504068c1818eb2245c 30372 Author: Adam Jackson <ajax (a] nwnk.net> 30373 Date: Mon Jul 11 02:29:45 2005 +0000 30374 30375 Prep for modular builds by adding guarded #include "config.h" everywhere. 30376 30377 commit 5b37b695019b7a4522a96a63f1f6b035160e6a25 30378 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 30379 Date: Mon Jul 4 23:44:59 2005 +0000 30380 30381 Bug #2604: Disable workaround for horrible flickering when scrolling 30382 vertically in a virtual screen when color tiling and page flip are both 30383 enabled, as it potentially does more harm than good. The flickering 30384 still needs fixing probably though. 30385 30386 commit 08b886850f115f404187acd619aa2c2ca18ae3c7 30387 Author: Daniel Stone <daniel (a] fooishbar.org> 30388 Date: Fri Jul 1 22:43:11 2005 +0000 30389 30390 Change all misc.h and os.h references to <X11/foo.h>. 30391 30392 commit 1c7c3372776d6d116facabbf055f27af3e96f59f 30393 Author: Alex Deucher <agd5f (a] yahoo.com> 30394 Date: Fri Jul 1 04:16:42 2005 +0000 30395 30396 - fix depth tiling for r3/4xx chips (Aapo Tahkola) 30397 - adjust limits for tiled surfaces on r3/4xx 30398 30399 commit e8c2574557950bfa150806432d566cd38c47158f 30400 Author: Alex Deucher <agd5f (a] yahoo.com> 30401 Date: Fri Jul 1 02:56:04 2005 +0000 30402 30403 - Fix Support for Philips FM1236/F tuner on ATI AIW 9600 XT (Jeff Smith) 30404 Bug 3401 30405 30406 commit d258e1c1566d005693260230712c9d78f1447dba 30407 Author: Alex Deucher <agd5f (a] yahoo.com> 30408 Date: Thu Jun 30 04:46:11 2005 +0000 30409 30410 - change back part of a comment I seem to have accidently changed. 30411 30412 commit 6d9c02d58eae1966e8fa8d7d305849c6e57434ad 30413 Author: Alex Deucher <agd5f (a] yahoo.com> 30414 Date: Thu Jun 30 02:02:10 2005 +0000 30415 30416 - small mergedfb fix from Thomas Winischhofer 30417 30418 commit 85365b791542cab221c37c9b3b4593ea0c6c9e2d 30419 Author: Alex Deucher <agd5f (a] yahoo.com> 30420 Date: Thu Jun 30 01:40:17 2005 +0000 30421 30422 - add support for colortiling on r3/4xx hardware (disabled by default). In 30423 order for pageflipping to work on r3/4xx we will need to update the the 30424 sarea and drm so that the crtc1/2 xytiling regs get updated. Initially 30425 worked out by Aapo Tahkola. 30426 30427 commit 3ec99c9cd821aeacdd57a9bbc20c4e4ec4711790 30428 Author: Alex Deucher <agd5f (a] yahoo.com> 30429 Date: Thu Jun 30 00:56:01 2005 +0000 30430 30431 - Increase the priority of display 0/1 in the memory controller for r3/4xx 30432 hardware when displaypriority is set to HIGH. Fixes display problems in 30433 high res modes. Originally reported by Aapo Tahkola. 30434 30435 commit a7523e6c0be7a600963e9de10cf0060d16117bf6 30436 Author: Alex Deucher <agd5f (a] yahoo.com> 30437 Date: Thu Jun 30 00:33:36 2005 +0000 30438 30439 - fix some OUTREGs that should be OUTPLLs in dynamicclocks code 30440 30441 commit c7e1d31c6aec3a99061c77c4d6f57e2e874e37f1 30442 Author: Adam Jackson <ajax (a] nwnk.net> 30443 Date: Sun Jun 26 04:38:12 2005 +0000 30444 30445 Bug #3628: Fix video driver submodule visibility. 30446 30447 commit 1beca962ecee5b5c51eb2fb8451fcc2a2ed4b97c 30448 Author: Adam Jackson <ajax (a] nwnk.net> 30449 Date: Sat Jun 25 21:16:53 2005 +0000 30450 30451 Bug #3626: _X_EXPORT tags for video and input drivers. 30452 30453 commit 9a4895c4c2fd9614d83adfe3967f4f46dc4dfff6 30454 Author: Eric Anholt <anholt (a] freebsd.org> 30455 Date: Wed Jun 22 22:50:49 2005 +0000 30456 30457 Add missing detection of the PCI ID for 0x5b62, Radeon X600 (Vladimir 30458 Kushnir). 30459 30460 commit 43e3cc7c21160f4536ed860f97d02545c267dfbc 30461 Author: Alex Deucher <agd5f (a] yahoo.com> 30462 Date: Wed Jun 15 23:40:50 2005 +0000 30463 30464 Sync up radeon mergedfb with Thomas' latest sis mergedfb changes: 30465 - non-rectangular mergedfb desktops 30466 - screen offsets 30467 30468 commit b67f9f5c875efaa41bded7ed2f38b2a5fbd17597 30469 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 30470 Date: Sat Jun 4 20:26:28 2005 +0000 30471 30472 Sync with 6.8.2 branch: Bug #591 30473 (https://bugs.freedesktop.org/show_bug.cgi?id=591) attachment #308 30474 (https://bugs.freedesktop.org/attachment.cgi?id=308): Fix video timing 30475 problems with Sony PCG-C1VN Vaio Picturebook notebook && co. Patch by 30476 Marc Aurele La France 30477 30478 commit daae70d9083e9c15358f160f0df3176e344f1cb9 30479 Author: Alan Coopersmith <Alan.Coopersmith (a] sun.com> 30480 Date: Sat Jun 4 20:04:36 2005 +0000 30481 30482 Sync with 6.8.2 branch: Bugzilla #1306 30483 (https://bugs.freedesktop.org/show_bug.cgi?id=1306) attachment #1291 30484 (https://bugs.freedesktop.org/attachment.cgi?id=1291): Fix SEGV in 30485 "radeon" driver. Patch by Kevin E. Martin <kem (a] freedesktop.org> 30486 30487 commit c27c193f192b117232dbe39575d3d5ab7c6a1b1e 30488 Author: Felix Kuehling <fxkuehl (a] gmx.de> 30489 Date: Fri May 27 22:33:23 2005 +0000 30490 30491 Fix for bugzilla https://bugs.freedesktop.org/show_bug.cgi?id=3414 (patch 30492 https://bugs.freedesktop.org/attachment.cgi?id=2782): Don't use RMX if 30493 we are Dell Server. 30494 30495 commit 3b674815e0a6d19b0719e71261c0909b92173aed 30496 Author: Adam Jackson <ajax (a] nwnk.net> 30497 Date: Tue May 24 23:26:50 2005 +0000 30498 30499 Typo fix from bug #3163 commit 30500 30501 commit 0845d1db801ae74688464bd4c41a7e40e4b50186 30502 Author: Egbert Eich <eich (a] suse.de> 30503 Date: Mon May 23 09:50:20 2005 +0000 30504 30505 Explicitely disable DPMS power states during close screen. Fix macro that 30506 takes a value and a mask of bits to modify to be behave evquvalently if 30507 the bits that should remain unchanged are set or unset in the value 30508 (Bugzilla #3369). 30509 30510 commit 74b86b036def72e5180a46871b8452ba9eea9bb1 30511 Author: Adam Jackson <ajax (a] nwnk.net> 30512 Date: Wed May 18 17:47:35 2005 +0000 30513 30514 Bug #3163: Create new DRIINFO_*_VERSION macros to indicate the version 30515 number of the DRIInfoRec, disambiguating it from the XF86DRI protocol 30516 version number. Modify DRIQueryVersion to return the libdri version 30517 number, which all DDXes were requesting implicitly. Fix the DDXes to 30518 check for the DRIINFO version they were compiled against. 30519 30520 commit 0f4a8fc6aa5157a8b48beb49f30511665cfa4b63 30521 Author: Matthieu Herrb <matthieu.herrb (a] laas.fr> 30522 Date: Wed Apr 20 21:07:44 2005 +0000 30523 30524 programs/Xserver/hw/xfree86/drivers/ati/theatre.h Build fix for 30525 DoLoadableServer = NO systems. 30526 30527 commit 3355d12f832909b05e8b022fad784c9f7af7341f 30528 Author: Daniel Stone <daniel (a] fooishbar.org> 30529 Date: Wed Apr 20 12:25:22 2005 +0000 30530 30531 Fix includes right throughout the Xserver tree: 30532 change "foo.h" to <X11/foo.h> for core headers, e.g. X.h, Xpoll.h; 30533 change "foo.h", "extensions/foo.h" and "X11/foo.h" to 30534 <X11/extensions/foo.h> for extension headers, e.g. Xv.h; 30535 change "foo.[ch]" to <X11/Xtrans/foo.[ch]> for Xtrans files. 30536 30537 commit 31cb4758eae4703d4de31a0b291847cefa855fc3 30538 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 30539 Date: Sun Apr 17 23:28:09 2005 +0000 30540 30541 Added fifo_read and fifo_write in tge generic bus definition 30542 30543 commit 2b363e02c7c122f07e6fc406fb8ee52d97b55cad 30544 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 30545 Date: Sun Apr 17 23:26:36 2005 +0000 30546 30547 Added new options for the ATI driver: RageTheatreMicrocPath and 30548 RageTheatreMicrocType 30549 30550 commit 6286efeba564c63ec198c091e4bba22027586901 30551 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 30552 Date: Sun Apr 17 23:23:45 2005 +0000 30553 30554 Addeded support for FI1216ME (PAL) version of the tuner, detection of 30555 UDA1380 sound processor 30556 30557 commit 9a63c216e90622cd96d4555155df06dd66c3235b 30558 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 30559 Date: Sun Apr 17 23:18:26 2005 +0000 30560 30561 Use the theatre_detect to determine the chip ID and then load the 30562 coresponding theatre module 30563 30564 commit 247057762813f12f171b1fa05f31c5ace4c245a7 30565 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 30566 Date: Sun Apr 17 23:15:31 2005 +0000 30567 30568 Added functions that access the FIFO on RT200 30569 30570 commit bcccf9259784f21d84d52c526aa68df0787118b4 30571 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 30572 Date: Sun Apr 17 23:13:45 2005 +0000 30573 30574 Added RT200 registers 30575 30576 commit da9d27f8c8cccf2f40a244b25ffe0d3162c41a63 30577 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 30578 Date: Sun Apr 17 23:09:28 2005 +0000 30579 30580 This is the theatre chip detection module 30581 30582 commit e4623bbce161c41ccac65210fd2539211e8af715 30583 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 30584 Date: Sun Apr 17 23:07:54 2005 +0000 30585 30586 Now the detection of the theatre chip is in other module called 30587 theatre_detect 30588 30589 commit 405b4dcb067346fce7fd9f6d8a6507a58c740ac7 30590 Author: Bogdan Diaconescu <b_diaconescu (a] yahoo.com> 30591 Date: Sun Apr 17 23:06:17 2005 +0000 30592 30593 Added the RageTheatre200 video demodulator support 30594 30595 commit 310eb24a45b145efbd47196a1c3ec750cf004154 30596 Author: Daniel Stone <daniel (a] fooishbar.org> 30597 Date: Mon Apr 4 23:07:08 2005 +0000 30598 30599 Release SDA/SDL I2C lines from an asserted-low state after a DDC probe, 30600 which caused Apple Studio Display monitors to shut off after a second 30601 or so (Ben Herrenschmidt). 30602 30603 commit a416667f980a8679352be897af2fb3215da2d000 30604 Author: Alex Deucher <agd5f (a] yahoo.com> 30605 Date: Thu Mar 31 23:18:10 2005 +0000 30606 30607 - Add new Radeon pci ids (ATI devrel), bug 2827 30608 30609 commit 2e7365ffb767099faf3f191dc65341dbb3ef6247 30610 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30611 Date: Thu Mar 31 18:26:05 2005 +0000 30612 30613 Modified: 30614 programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 30615 Patch from Antti Ajanki - provide VBI capture support for PAL and SECAM 30616 (You need GATOS km module to actually get at the data, the change 30617 merely configures the hardware to handle it correctly). 30618 30619 commit 9a3c1c8e4cba2885b707db0f855e11dc6e640b49 30620 Author: Michel Daenzer <michel (a] daenzer.net> 30621 Date: Tue Mar 29 06:14:34 2005 +0000 30622 30623 bugzilla #2844 (https://bugs.freedesktop.org/show_bug.cgi?id=2844) 30624 attachment #2230 (http://bugs.freedesktop.org/attachment.cgi?id=2230) 30625 Wait for idle before uploading HW cursor data to the framebuffer to 30626 avoid lockups when moving the mouse on some machines. 30627 30628 commit 97863f112850be5206ea555297354c1c104fcdbe 30629 Author: Kevin E Martin <kem (a] kem.org> 30630 Date: Tue Mar 29 03:49:04 2005 +0000 30631 30632 bugzilla #2840 (https://bugs.freedesktop.org/show_bug.cgi?id=2840) 30633 attachment #2227 (https://bugs.freedesktop.org/attachment.cgi?id=2227) 30634 Fix build problems introduced by previous Radeon PLL errata 30635 workarounds. 30636 30637 commit 04cbb8d174cc5e3c7ecdd4e171170213ecb3c7ed 30638 Author: Dave Airlie <airlied (a] linux.ie> 30639 Date: Sat Mar 26 00:53:01 2005 +0000 30640 30641 bugzilla #2057 (https://bugs.freedesktop.org/show_bug.cgi?id=2057) 30642 attachment #1516 (https://bugs.freedesktop.org/attachment.cgi?id=1516) 30643 Add TVOUT Support for Mach64 (Leif Delgass, fixed up for Xorg by me). 30644 30645 commit 5be4bf9000bdf58584a10a6b8e285d0f173304fa 30646 Author: Alex Deucher <agd5f (a] yahoo.com> 30647 Date: Thu Mar 24 06:45:52 2005 +0000 30648 30649 - radeon PLL access errata workarounds (Benjamin Herrenschmidt, acked by 30650 Hui Yu) 30651 30652 commit 34af3da80304a13b6407dcc0f4db54d6c8bb7a41 30653 Author: Michel Daenzer <michel (a] daenzer.net> 30654 Date: Fri Mar 18 22:36:47 2005 +0000 30655 30656 Add support for production version of ATI RN50/ES1000. (ATI Technologies 30657 Inc.) 30658 30659 commit b56a5f96048b920a7751318dde211e881f53b2bb 30660 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30661 Date: Thu Mar 17 03:36:25 2005 +0000 30662 30663 Modified: 30664 programs/Xserver/hw/xfree86/drivers/ati/radeon_mergedfb.c Back out 30665 RadeonWaitForIdleMMIO that I put in earlier. 30666 From discussion on dri-devel we do not need to be as careful with INREG() 30667 access on Radeons as we are on mach64 chips. 30668 The fix is backed out due to: 30669 it breaks DRI on SMP systems (due to absense of the lock in mouse handling 30670 functions) 30671 I am no longer able to reproduce the lockup I saw before 30672 30673 commit 01a9c1157cdfbc5f74b035383e060b8419979563 30674 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 30675 Date: Tue Mar 8 23:41:37 2005 +0000 30676 30677 Fix accidentally removed depth pitch adjustment if displayWidth is not a 30678 multiple of 32 pixels (bug #788) 30679 30680 commit 9b994b40710103b263364153429ebdb9ea3f58ed 30681 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30682 Date: Fri Mar 4 22:21:40 2005 +0000 30683 30684 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c Let through all chips 30685 R300 and above - presumably the filtering should take place at the DRM 30686 driver level. Correct cut'n'paste position misprint. Big thanks for 30687 Michel Danzer for pointing out both improvements. 30688 30689 commit 7d37a9f1f9fffe98067c3b5bc4749cc799fb0c77 30690 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30691 Date: Fri Mar 4 05:48:31 2005 +0000 30692 30693 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c Turns out Radeon R420 30694 chips work fine with R300 microcode. Thanks go to Rune Petersen for 30695 careful investigation. 30696 30697 commit f1edbb7852d85f660188498f578a0a697c2094fe 30698 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 30699 Date: Tue Feb 22 00:39:15 2005 +0000 30700 30701 Fix building radeon driver with BuildXF86DRI set to NO, the breakage was 30702 caused by the tiling support (bug #2481) 30703 30704 commit b853dbb50fb7c46fb5a70f349a3f264f0bb94e8c 30705 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30706 Date: Sat Feb 19 16:00:12 2005 +0000 30707 30708 Modified: 30709 programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 30710 Fix an INREG without WaitForIdleMMIO. This one should be unlikely to 30711 trigger, but we should still do this in the name of correctness. 30712 30713 commit 84290849b4aee336e9db0779d93fc2d7ed8881b1 30714 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30715 Date: Sat Feb 19 00:57:23 2005 +0000 30716 30717 Modified: 30718 programs/Xserver/hw/xfree86/drivers/ati/radeon_mergedfb.c 30719 Add a call to RADEONWaitForIdleMMIO() within RADEONChooseCursorCRTC() as 30720 this function uses OUTREGP() which, internally, calls INREG(). 30721 Thus (due to SilkenMouse) INREG() could be called while CP engine was busy. 30722 I have observed this lockup on RV350 + R300 3d driver and have no idea 30723 why no issues were reported for R200 cards. Perhaps, few of them have 30724 enough memory to run merged fb and 3d simultaneously ? 30725 This fix is the right thing to do, however, we might also need to put fifo 30726 management calls into other cursor handling functions (in particular 30727 those that write cursor image directly to video memory). 30728 30729 commit 60070063e2130af0b9d5d34eed7adbae0c0fa67b 30730 Author: Michel Daenzer <michel (a] daenzer.net> 30731 Date: Fri Feb 18 19:55:35 2005 +0000 30732 30733 Bug #2576: Add support for ATI RN50/ES1000. (ATI Technologies Inc.) 30734 30735 commit 32032cdc78097840b7e2685fdec11bc0816aa0ac 30736 Author: Adam Jackson <ajax (a] nwnk.net> 30737 Date: Fri Feb 18 16:42:11 2005 +0000 30738 30739 Bug #2567: Fix reversed memset arguments. (David Krause) 30740 30741 commit d5fadd6d86ae2a2f8c1bb55e36ab3d6d49649b46 30742 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 30743 Date: Wed Feb 9 02:22:39 2005 +0000 30744 30745 do not enable color tiling on r300 chips and newer, since it doesn't work 30746 (yet) (Bugzilla #2504) 30747 30748 commit 5ff35300e66a81ecc17c4e9396eb6d158f5c8e1b 30749 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 30750 Date: Wed Feb 2 21:35:30 2005 +0000 30751 30752 fix error with color tiling and hw scrolling, it looks like rv100 doesn't 30753 need the special treatment, after all (reported by Stephane Marchesin) 30754 30755 commit 7a9031683914bb1d17d223634e5dbf2e18955651 30756 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 30757 Date: Mon Jan 31 20:10:28 2005 +0000 30758 30759 change option EnableColorTiling to ColorTiling 30760 30761 commit a2b123f7672ea678290798786ecb408a86d8c683 30762 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 30763 Date: Wed Jan 26 18:23:41 2005 +0000 30764 30765 (Stephane Marchesin, me) Add support for color (framebuffer) tiling. 30766 Convert all XAA accel code to use explicit dst and src offset pitch 30767 control to make it work. Use the new drm surface ioctls to set up the 30768 tiled surfaces (if dri is enabled). 30769 30770 commit e0df1e41724cd009fe7c833d2f9555f5d87e3202 30771 Author: Roland Scheidegger <rscheidegger_lists (a] hispeed.ch> 30772 Date: Wed Jan 26 17:39:56 2005 +0000 30773 30774 fix issues with mergedfb + pageflip (fix up sarea frame values) 30775 30776 commit aeb01807263c47f49abaf00031d8505e8ebda82c 30777 Author: Egbert Eich <eich (a] suse.de> 30778 Date: Mon Jan 24 15:40:21 2005 +0000 30779 30780 Don't exclude 'RADEONInfoPtr info = ...' declaration even when not building 30781 with DRI support (Bugzilla #2370) 30782 30783 commit c6c30cfa354eb456638f10771a7cb02271696af2 30784 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30785 Date: Thu Jan 13 17:35:49 2005 +0000 30786 30787 Modified: 30788 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 30789 Some of the newer Radeons (R420 for example) have MM_TABLE's with size 0x33 30790 (instead of 0x0c that we know). Furthermore, they lockup during access 30791 of I2C bus ! 30792 Add a check to validate MM_TABLE. Note: the check is not more stringent as 30793 other card manufacturers (like Asus or Tekram) might make cards with 30794 slightly different MM_TABLE. 30795 30796 commit 414c035a965b9db15656058c0e9cfeb3acf0158f 30797 Author: Alex Deucher <agd5f (a] yahoo.com> 30798 Date: Fri Dec 24 18:26:19 2004 +0000 30799 30800 - Fix MCLK / SCLK calculation (Benjamin Herrenschmidt) bug 2144 30801 - formatting cleanup 30802 30803 commit 7c7b278883676412f5709cf812587b1408628133 30804 Author: Alex Deucher <agd5f (a] yahoo.com> 30805 Date: Sun Dec 19 19:49:33 2004 +0000 30806 30807 - Add support for Dualhead on M3/M4 (bug 1760) 30808 30809 commit 08a45118b53129f3b9c30004404b906c7d7690be 30810 Author: Alex Deucher <agd5f (a] yahoo.com> 30811 Date: Sun Dec 19 19:29:10 2004 +0000 30812 30813 - Make dualhead ram allocation cleaner and related messages clearer 30814 30815 commit 97cfd7b1489a9510cd233d6dc7154c210f32b9f5 30816 Author: Alex Deucher <agd5f (a] yahoo.com> 30817 Date: Sun Dec 19 06:56:37 2004 +0000 30818 30819 - Add DRI resume support (Micha Feigin) bug 2063 30820 30821 commit 5b8577ae3200d967b4316156f16028cbcfd21983 30822 Author: Alex Deucher <agd5f (a] yahoo.com> 30823 Date: Sun Dec 19 06:11:06 2004 +0000 30824 30825 - fix warnings (Ronny V. Vindenes) 30826 30827 commit 447124a663a24a7fffb0cd241d18dd896a43098b 30828 Author: Alex Deucher <agd5f (a] yahoo.com> 30829 Date: Sat Dec 18 21:37:48 2004 +0000 30830 30831 - fix warning and potential error in radeon_driver.c 30832 30833 commit ac6166158a58b377e7383950c169956f78b62093 30834 Author: Alex Deucher <agd5f (a] yahoo.com> 30835 Date: Fri Dec 17 19:19:32 2004 +0000 30836 30837 - Add VGAAccess option to r128 driver. Based on Benh's radeon patch. r128 30838 should now work on ppcs without usefbdev. To use this on ppc make sure 30839 you set "usefbdev" to false. (bug 2089) 30840 - Apply Benjamin Herrenschmidt's radeon patches (bug 2064) 30841 30842 commit ac860ed6bbed84f6fc39a9acf53ee87646cfc8dc 30843 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30844 Date: Fri Dec 17 16:50:36 2004 +0000 30845 30846 Modified: 30847 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 30848 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c 30849 Move DMA robustness fix into radeon_dri.c::RADEONEnterServer() as per 30850 suggestion by Michel Dnzer. 30851 I could not trigger a lockup, even with r300_demo (possibly it has code 30852 that flushes cache inside ?), so this must be good enough.. 30853 30854 commit 3ba5b525c4d6390184fdf79b28c954d5caadc6a4 30855 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30856 Date: Thu Dec 16 16:26:56 2004 +0000 30857 30858 Modified: 30859 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 30860 Add DMA robustness fix from Nicolai Haehnle. 30861 TODO: move this fix into DRM driver or other parts of Xserver when R300 30862 support is more stable. 30863 This fix affects only R300 and later variants and results in small slowdown 30864 when using CP acceleration in comparison to the previous version of the 30865 driver. 30866 We are still a lot faster in CP mode than using MMIO register access. 30867 30868 commit c7b5f4a9eb49c0edd690c972be1015c98408ffaf 30869 Author: Adam Jackson <ajax (a] nwnk.net> 30870 Date: Wed Dec 15 19:35:33 2004 +0000 30871 30872 Bug #2008: Fix gcc4 compile problem. 30873 30874 commit 8c1969f22fba2e1c9ed34d5ca77bee3a9f577640 30875 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30876 Date: Sun Dec 12 20:42:29 2004 +0000 30877 30878 Modified: 30879 programs/Xserver/hw/xfree86/drivers/ati/radeon_render.c 30880 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 30881 Make DMA blitting code more robust. 30882 30883 commit 8a9077bc70fd800983f138a224b95301992a2296 30884 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30885 Date: Sun Dec 12 19:28:37 2004 +0000 30886 30887 Modified: 30888 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 30889 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c 30890 Enable DRM support permanently, allowing its use for any CHIP_FAMILY > 30891 R300. Fix bug in naming Mesa3d driver (it was defaulting to r200) 30892 30893 commit 845df6dd96785b59c44c670009c4fcdb20e147ab 30894 Author: Alex Deucher <agd5f (a] yahoo.com> 30895 Date: Sun Dec 12 17:29:45 2004 +0000 30896 30897 - missing diff from agp 8x support 30898 30899 commit 36b92808e4ff1311f91543388c18cbe8e9eb4aab 30900 Author: Vladimir Dergachev <volodya (a] mindspring.com> 30901 Date: Sun Dec 12 16:05:35 2004 +0000 30902 30903 Modified: 30904 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 30905 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 30906 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c 30907 programs/Xserver/hw/xfree86/drivers/ati/radeon_common.h 30908 programs/Xserver/hw/xfree86/drivers/ati/radeon_version.h Add R300 (and 30909 later) DRM support, protected by X_R300_DRM option. 30910 Note: at the moment this is for 2d support only, if enabled running 30911 glxgears locks up my machine, but DMAForXv=yes works fine. 30912 30913 commit 68ea67ce98b8f3ad4c17f987e0e2b92f79a96fbf 30914 Author: Alex Deucher <agd5f (a] yahoo.com> 30915 Date: Sun Dec 12 02:00:48 2004 +0000 30916 30917 - add AGP 8x support (Hui Yu, Michel Daenzer) bug 1819 30918 30919 commit 93675d288d5ebb35d4724ebbfc455a703e25a311 30920 Author: Alex Deucher <agd5f (a] yahoo.com> 30921 Date: Sun Dec 12 00:14:32 2004 +0000 30922 30923 - only mess with Dynamicclocks on mobility chips (Benh, me) bug 1912 30924 - fix display detection (Hui Yu) bug 1559 30925 - Add new option "BIOSHotkeys" to allow the bios to control output 30926 switching (off by default) bug 2000 30927 - update the savage man page to document the "CrtOnly" option 30928 30929 commit ed217003127028483fd2956851d438977169d154 30930 Author: Daniel Stone <daniel (a] fooishbar.org> 30931 Date: Mon Dec 6 13:27:56 2004 +0000 30932 30933 Apply Debian patch to re-POST via VBE if driver thinks it has no video 30934 memory (a symptom of incompletely coming up from D3 power state), which 30935 fixes some Dell laptops (closes: #1890). 30936 30937 commit 2b8ab42b05b32710fa4f71a25e2f84192e637fbd 30938 Author: Michel Daenzer <michel (a] daenzer.net> 30939 Date: Mon Dec 6 06:27:53 2004 +0000 30940 30941 When direct rendering is enabled, use hostdata blits to transfer data from 30942 system memory to video RAM, which should reduce CPU usage especially 30943 with larger videos. Can be disabled via Option "DMAForXv" if there 30944 should be any stability issues, but it's been stable for me during a 30945 week of testing. Based on a patch by Nikolaus Meine 30946 <meine (a] tnt.uni-hannover.de>. Probably fix endianness issues in some 30947 newer XVideo code, untested. 30948 Also use hostdata blits to transfer RENDER image data to video RAM to avoid 30949 idling the accelerator engine. Increases RENDER performance 30950 significantly for me. 30951 These changes were only tested on an M9 in a Titanium PowerBook but should 30952 work with all Radeons where direct rendering is supported. 30953 30954 commit 960d15ae3a44efe7c02a3d6d0acbabe63e07be74 30955 Author: Daniel Stone <daniel (a] fooishbar.org> 30956 Date: Sat Dec 4 22:34:54 2004 +0000 30957 30958 Disable interrupt handling after the DRM handler has been removed, 30959 preventing lockups (closes: #1886). 30960 Allow pitches with a granularity of 32, rather than 64, because some 30961 clamshell iBooks have 800-byte pitches. Change Xv allocation 30962 granularity to 8 bytes, rather than 16 bytes, also (closes: #1888). 30963 30964 commit 4706dfb7804a46e3cc5aafc3644f5a1134ab3d19 30965 Author: Daniel Stone <daniel (a] fooishbar.org> 30966 Date: Sat Dec 4 22:24:06 2004 +0000 30967 30968 Retry CCE idle commands until the (higher) timeout is reached, and properly 30969 stop the engine if we totally fail -- gets rid of serious performance 30970 issues and lockups with r128 DRI. 30971 30972 commit bce508a288d6b95708dbbfc6b235e2d95b6b96ad 30973 Author: Markus Kuhn <Markus.Kuhn (a] cl.cam.ac.uk> 30974 Date: Sat Dec 4 00:42:59 2004 +0000 30975 30976 Encoding of numerous files changed to UTF-8 30977 30978 commit e9c9fd279ed43cc0d779266efcea69a67541c794 30979 Author: Egbert Eich <eich (a] suse.de> 30980 Date: Tue Nov 30 08:38:44 2004 +0000 30981 30982 Make Xorg the default server to install. 30983 Avoid PIO access on IA64. Some IA64 machine check if legacy ports outside 30984 the VGA range are accessed. The ATi driver however does this to probe 30985 for ISA Mach8/32/64. Since no IA64 has ISA slots this restriction 30986 should not be relevant to the user. 30987 Avoid recursive calls of xf86scanpci(). This function normally detects that 30988 it has been called before by checking if the PCI structure is filled 30989 out. So far if this was not the case (because PCI probing has failed 30990 for some reason) the function is traversed again. With the chipset 30991 specific PCI bus probing this can lead to an endless recursive loop as 30992 the post-probing code calls xf86scanpci() from within this function. 30993 The OS specific PCI code for Linux worked only if bus 0 was populated as it 30994 checked for the presence of /proc/bus/pci/00. Fixed to check for 30995 /proc/bus/pci/<bus_to_look_for> instead. 30996 30997 commit fffbcb761c178e025e3d55d428a803b1284481e0 30998 Author: Matthieu Herrb <matthieu.herrb (a] laas.fr> 30999 Date: Sat Nov 27 15:14:15 2004 +0000 31000 31001 xc/programs/Xserver/hw/xfree86/driver/ati/radeon.man 31002 xc/programs/Xserver/hw/xfree86/driver/voodoo/voodo.man 31003 xc/programs/Xserver/hw/xfree86/driver/wsfb/wsfb.man XF86Config -> 31004 __xconfigfile__ 31005 31006 commit ad6347d187f9afa053c755aa3a2da9d821916cb2 31007 Author: Daniel Stone <daniel (a] fooishbar.org> 31008 Date: Tue Nov 23 21:27:43 2004 +0000 31009 31010 Use fp_horz_stretch instead of fp_vert_stretch for horizontal stretch 31011 factor calculation in panel size derivation in the Radeon driver 31012 (closes: #1881). 31013 31014 commit ac054e427c6ce8cd5b89459dd3f16ca20e6b2d6a 31015 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31016 Date: Sun Oct 24 18:17:36 2004 +0000 31017 31018 Add registers to facilitate work with stereo 3d. 31019 31020 commit 445ed4b9b61347c866ee67dc0cb110904d667b84 31021 Author: Alex Deucher <agd5f (a] yahoo.com> 31022 Date: Fri Oct 22 03:51:58 2004 +0000 31023 31024 - make the overlay work on crtc2 again 31025 - make PutVideo() MergedFB aware 31026 31027 commit 54cffc3fe273cf3c9ed2a8aef833d4bd337a8a09 31028 Author: Egbert Eich <eich (a] suse.de> 31029 Date: Mon Oct 18 15:19:44 2004 +0000 31030 31031 Fixing one of the infamous 'control reaches end of non void function 31032 problems'. 31033 31034 commit 7a97d363327e0b2c8646d961e00d05f99766b54f 31035 Author: Egbert Eich <eich (a] suse.de> 31036 Date: Mon Oct 18 14:21:45 2004 +0000 31037 31038 Made handling of DevelDrivers for x86-64 more conformant to other 31039 platforms. 31040 Compress all font encodings (Stefan Dirsch). 31041 Fixed warnings. 31042 Turn on forwarding XNSpotLocation event to XIM server in OffTheSpot and 31043 Root mode (bugzilla #1580, James Su). 31044 Added another compose key combination for the Euro symbol (Stefan Dirsch). 31045 Added support for UTF-8 in ja_JP, ko_KR and zh_TW locales (Mike Fabian). 31046 Changed default encoding for ru from KOI8-R to ISO8859-5 (Mike Fabian). 31047 This is the encoding that is also used by glibc. We may break other 31048 libcs - lets see who complains. 31049 Added explanation for DESTDIR to install to a different directory than /. 31050 Added some early bailouts to atiprobe if PCI structure pointer is NULL to 31051 prevent sig11. 31052 XV support doesn't depend on 2D acceleration any more. This patch removes 31053 this limitation from the NSC driver. This is a patch that I have 31054 committed to XFree86 a while ago but never ported over to X.Org. 31055 Matthieu Herrb contributed some build fixes. 31056 Fixing SetDPMSTimers() so that DPMS*Time == 0 disables the appropriate 31057 timer. This takes advantage of the fact that TimerSet() with a timeout 31058 argument 0 behaves like TimerCanel(). 31059 Use /dev/xconsole (named pipe) or devpts for system logger (Werner Fink). 31060 Create missing links for backward compatibility to XFree86 (Stefan Dirsch). 31061 Changed comment to mention xorg. 31062 Changed cursor for the 'kill' action from XC_draped_box to XC_Pirate. If 31063 you don't like it we can change it back (original author unkown). 31064 Added 'pic' to the man page rendering command pipeline (Werner Fink). 31065 Added missing return value (Stefan Dirsch, Roland Mainz) 31066 31067 commit c7ab340fecb94a89c25d6bc7f686b7c97cd3067f 31068 Author: Alex Deucher <agd5f (a] yahoo.com> 31069 Date: Sun Oct 17 22:45:57 2004 +0000 31070 31071 - Fix DFP detection in r128 driver (bug 1386, Klaus Kusche) 31072 - Fix EmulateWheelInertia defaults (bug 1115, Andrew Pimlott) 31073 31074 commit ace6e1b7933cb415b8efc41dffa335938f9c89b5 31075 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31076 Date: Sun Oct 17 17:57:33 2004 +0000 31077 31078 Modified: 31079 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c On my Inspiron 31080 8600 both internal and external ports are marked DAC_PRIMARY in BIOS. 31081 So be extra careful - only swap when the first port is not DAC_PRIMARY 31082 31083 commit 22405e81b812c239b5d8653b402b5f846e6e10b1 31084 Author: Alex Deucher <agd5f (a] yahoo.com> 31085 Date: Sun Oct 17 16:52:02 2004 +0000 31086 31087 - Add Thomas' latest sis MergedFB enhancements to the radeon driver 31088 31089 commit 186efa31b2cb7285717678bb1c4df4cd9e182395 31090 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31091 Date: Wed Oct 13 23:02:42 2004 +0000 31092 31093 Modified: 31094 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 31095 xc/programs/Xserver/hw/xfree86/drivers/i2c/fi1236.h Squash annoying warning 31096 about fi1236_dump_status 31097 31098 commit 43b24861e36023613a99627c57ac44721ed1a548 31099 Author: Egbert Eich <eich (a] suse.de> 31100 Date: Tue Oct 12 19:13:43 2004 +0000 31101 31102 Set fbdev mode as the default mode on PPC (Olaf Hering). 31103 Added support for IBM space saver keyboard (Stefan Dirsch). Added support 31104 for Cherry CyMotion Master XPress (Marcus Schaefer). 31105 Change order of SetDisplayDevice(), HWRestore(), UnbindGART() and 31106 RestoreBIOSMemSize() to be exactly opposite to the Save procedure in 31107 EnterVT() (Matthias Hopf, Alan Hourihane). 31108 Fix text mode restauration by removing the assumption that the register 31109 which determines which head is programmed is set. to the active head by 31110 the BIOS (Mark Vojkovich). 31111 When I wrote the resource code 5 years ago I made some assumptions which 31112 turned out to be false: I've assumed that the bus number of the PCI 31113 hostbridge would be the PCI bus the bridge links to. This is not 31114 correct. Fixing this assumption is not easy. However I hope that the 31115 attached patch will make the system work regardless as it 'ignores' 31116 host bridges from which the target bus is not known. This should not 31117 matter at all as we really don't care about host bridges (unless we 31118 have bridge specific code which retrieves information about the 31119 bridge). 31120 Fixed server crash on reset when a structure allocated in PreInit() was 31121 freed on CloseScreen(). 31122 Fixed ring buffer lock ups that happened because the structure that 31123 contained ringbuffer data was not zeroed after allocation. 31124 Fixed numerous warnings due to signed unsigned comparisons. 31125 programs/Xserver/hw/xfree86/drivers/nv/nv_driver.c: 31126 (NVBacklightEnable): Changed the order in which the sequencer registers 31127 and the backlight control registers are written. The sequencer control 31128 register need to be written first otherwise DPMS screen blanking 31129 produces vertical bars on a mobile device. lib/Xau/Imakefile: Build 31130 libXau static library PIC so it can be linked into toolkits that would 31131 like to wrap its functionality. 31132 31133 commit 515b7c4b748f70c17aee149a33ee1b41632daa1a 31134 Author: Alex Deucher <agd5f (a] yahoo.com> 31135 Date: Mon Oct 11 23:20:26 2004 +0000 31136 31137 - fix resume on radeons (bug 1220) 31138 - clean up some old cruft in the savage i2c code 31139 31140 commit 07e1aaee862c4835601c704e14fcb12e28f4e78e 31141 Author: Egbert Eich <eich (a] suse.de> 31142 Date: Mon Oct 11 09:58:04 2004 +0000 31143 31144 Improving DPMS handling on VT swich and server termination/abort: previous 31145 version called the driver directly and too late. 31146 Unblank secondary screen explicitely. Don't rely on the value read during 31147 register save as the BIOS have blanked the secondary head. 31148 Checking if server isn't switched away before calling sync. Sanity check 31149 for possible bugs in aother areas of the code. 31150 Fixing default amount of of allocated video memory from AGP for i810: Use 31151 16MB if less than 192MB are installed else use 24MB (Matthias Hopf). 31152 31153 commit a8c77480e7f4ddf8c45bea3ecec7b6564c0d6452 31154 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31155 Date: Fri Oct 8 00:56:05 2004 +0000 31156 31157 Modified: 31158 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c Remove redundant 31159 and incorrect line. 31160 31161 commit ec831d4fba23dc5f4d0a1450aa63efe5ff5937a2 31162 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31163 Date: Tue Oct 5 01:16:01 2004 +0000 31164 31165 Modified: 31166 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre.h Patch from Matthieu 31167 Herrb - fix compilation with DoLoadableServer=NO 31168 31169 commit f1dadde8f642b332a2dcd2e2a5779f25fec7a3de 31170 Author: Adam Jackson <ajax (a] nwnk.net> 31171 Date: Mon Oct 4 21:52:47 2004 +0000 31172 31173 Correct radeon man page, IGP is supported for 3D but R400 is not. 31174 31175 commit e3dba9841521466629828501c4fdccbba3d6af89 31176 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31177 Date: Mon Oct 4 15:24:34 2004 +0000 31178 31179 Modified: 31180 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c Do not probe for 31181 Rage Theatre if no multimedia table is present and no parameters have 31182 been specified. 31183 31184 commit 4aadb784928ebbd60dce172c9a9a80e8d84e943f 31185 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31186 Date: Mon Oct 4 14:53:11 2004 +0000 31187 31188 Modified: 31189 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c This is really 31190 embarassing - I found a part of code that had <cr> at the end of the 31191 lines. I am certain I never saw it before - talk about code rot ! 31192 31193 commit db5a27081d9d3ab4cc1481a7d110d11103b3de19 31194 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31195 Date: Sun Oct 3 23:54:40 2004 +0000 31196 31197 Modified: 31198 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_mm_i2c.c Remove debugging 31199 statement that cause loading of DLL modules to fail. 31200 31201 commit 37a8d8a53f12f829783011c1e077fefb5729e618 31202 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31203 Date: Sun Oct 3 14:41:45 2004 +0000 31204 31205 Modified: 31206 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c Fix compilation 31207 with gcc 3.4.x (patch by Ronny V. Vindenes) 31208 31209 commit 6dbcd503fe21c6dec27c2b97adc0f39ba0a7ab2a 31210 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31211 Date: Sun Oct 3 03:07:17 2004 +0000 31212 31213 Modified: 31214 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.[c,h] 31215 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre.c 31216 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_reg.h Port the rest of 31217 GATOS Radeon-specific code. Remove "checkpoint" noise from logs. Test 31218 with AIW Radeon 7500 31219 31220 commit 7cb911291d459f08f54444d09081d3378fe73b8b 31221 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31222 Date: Sun Oct 3 00:01:14 2004 +0000 31223 31224 Modified: 31225 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon.h 31226 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 31227 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.[c,h] 31228 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre.c Detect and initialize 31229 Rage Theatre (RT100) 31230 31231 commit 6c1996174b9f7afc8f4e33c35e4864b31696d4d6 31232 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31233 Date: Sat Oct 2 20:55:42 2004 +0000 31234 31235 Modified: 31236 xc/programs/Xserver/hw/xfree86/drivers/ati/Imakefile 31237 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_reg.h 31238 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.[c,h] 31239 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_vip.c Add VIP bus access 31240 code, hook it into initialization system. 31241 31242 commit 136649a1c2065624a9cf17386054c6595aa6d85e 31243 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31244 Date: Sat Oct 2 20:35:04 2004 +0000 31245 31246 Modified: 31247 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre* 31248 xc/programs/Xserver/hw/xfree86/drivers/ati/generic_bus.h Add Rage Theatre 31249 (RT100) module code and hook it into the build system 31250 31251 commit 7a17b1c294fa8c7af0b1cdea0618cb6060479356 31252 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31253 Date: Sat Oct 2 01:59:13 2004 +0000 31254 31255 Modified: 31256 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon.h 31257 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 31258 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.h 31259 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_mm_i2c.c Add code to 31260 access and initialize multimedia i2c bus. Hook it up. 31261 31262 commit e0123e24b6895dde7b385528da24c4e6e9d5a4cb 31263 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31264 Date: Sat Oct 2 01:36:43 2004 +0000 31265 31266 Add more multimedia registers. 31267 31268 commit 576f53be4f0fa54bde7167872b7508b98e5c9454 31269 Author: Vladimir Dergachev <volodya (a] mindspring.com> 31270 Date: Sat Oct 2 00:43:08 2004 +0000 31271 31272 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 31273 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon.h Add TV-in specific 31274 options. Add TV-in specific fields to radeon per-screen structure, add 31275 code to initialize them. Preserve MM_TABLE (instead of whole VBIOS as 31276 was done in GATOS drivers) 31277 31278 commit 9a9ba82172428b09312fdd6e11da04b21c8529fc 31279 Author: Adam Jackson <ajax (a] nwnk.net> 31280 Date: Wed Sep 22 04:38:03 2004 +0000 31281 31282 Bug #1252, #1253, #1255, #1256: Various typo fixes from Dave Jones. 31283 31284 commit 700b0dc5148d0170c9223765d208ccef447242df 31285 Author: Adam Jackson <ajax (a] nwnk.net> 31286 Date: Tue Sep 21 18:10:34 2004 +0000 31287 31288 Convert the ATI drivers to ANSI function prototypes. 31289 31290 commit 17e5f3c1169066be1004306d12b5398cf01cbd35 31291 Author: Eric Anholt <anholt (a] freebsd.org> 31292 Date: Fri Sep 17 05:16:45 2004 +0000 31293 31294 Bugzilla #1170: Fix an overcounting of DMA in BEGIN_RING that wasn't 31295 harmful. Bugzilla #1169: Add counting for 31296 BEGIN_RING/OUT_RING/ADVANCE_RING to avoid these sorts of problems in 31297 the future. 31298 31299 commit 29bbee6b8a8dd1b883e07547691ff8e1289459ce 31300 Author: Adam Jackson <ajax (a] nwnk.net> 31301 Date: Fri Sep 17 03:04:50 2004 +0000 31302 31303 Bug #1192: Remove cfb support from drivers where its use is an option. 31304 Delete xf24_32bpp, as s3virge was the last user. Fix up some comments 31305 to refer to fb rather than cfb. 31306 31307 commit cec06d4126f5f31d14ace73798c71ec75e544c45 31308 Author: Alex Deucher <agd5f (a] yahoo.com> 31309 Date: Thu Sep 16 22:15:05 2004 +0000 31310 31311 Update comment 31312 31313 commit dbeb8d28ac3469f4a7d7d8dae979ae876d4cdcba 31314 Author: Kevin E Martin <kem (a] kem.org> 31315 Date: Fri Sep 3 23:26:19 2004 +0000 31316 31317 Updated release notes for bug #999. 31318 Update Radeon man page to note that RenderAccel is now enabled by default 31319 on certain chips. 31320 31321 commit c829f667c8a79bbdd40e1f924d57638e7bd425dc 31322 Author: Michel Daenzer <michel (a] daenzer.net> 31323 Date: Mon Aug 30 04:31:50 2004 +0000 31324 31325 Hotfix for Eric Anholt's fix for bug #1156 : Clear the 31326 RADEON_NONSURF_AP0_SWP_16BPP bit for 4 bpp data, otherwise it breaks at 31327 depth 16. 31328 31329 commit 677246740b5b538e73a7fbde661ec6a97319430b 31330 Author: Adam Jackson <ajax (a] nwnk.net> 31331 Date: Fri Aug 27 20:31:26 2004 +0000 31332 31333 Back out my last change, dupe. 31334 31335 commit a8eaa15264ff0db1596d653e91260c9a39a8b5ee 31336 Author: Adam Jackson <ajax (a] nwnk.net> 31337 Date: Fri Aug 27 18:32:40 2004 +0000 31338 31339 Bug #1061 (documentation-only update): Document the RenderAccel option in 31340 the radeon manpage. 31341 31342 commit 360dab062f895c19e1784146e6684f75a20c5b24 31343 Author: Eric Anholt <anholt (a] freebsd.org> 31344 Date: Fri Aug 27 03:36:28 2004 +0000 31345 31346 Bug #1156: Fix the build on PPC by splitting the byte-order-setting code 31347 into a separate routine. While here, fix the byte swapping in the 32bpp 31348 case. (Michel Daenzer) 31349 31350 commit 9b19bbb4d37b79c97365aa492cd93931d492b3a8 31351 Author: Eric Anholt <anholt (a] freebsd.org> 31352 Date: Fri Aug 27 03:23:47 2004 +0000 31353 31354 Bug #1157: Fix Radeon line acceleration in the DMA case (Michel Daenzer). 31355 31356 commit 838488b6b46359334e5e781589dbee3832909628 31357 Author: Kevin E Martin <kem (a] kem.org> 31358 Date: Fri Aug 27 02:38:56 2004 +0000 31359 31360 Fix Radeon compilation without Xinerama (Bug #1179, Simon Schubert). 31361 31362 commit 5f54d4f45e2adc6b15e98152db88743c243aa128 31363 Author: Kevin E Martin <kem (a] kem.org> 31364 Date: Wed Aug 25 00:30:41 2004 +0000 31365 31366 Fix hangs during startup with Radeon driver (Bug #1007, Sjoerd Simons). 31367 Fix i810 driver startup on Sony VAIO laptop (Bug #1084, Dave Airlie). 31368 Fix several library build problem (Bug #1147). 31369 Fix build problems (Bugs #1148 and #1149). 31370 31371 commit dde880e876f455f1fd66efbcc055e177886397c9 31372 Author: Eric Anholt <anholt (a] freebsd.org> 31373 Date: Tue Aug 17 20:22:55 2004 +0000 31374 31375 Reenable Radeon Render acceleration now that interactions with 3D clients 31376 should be resolved. 31377 31378 commit 279efef503fdbd442a9bd00f0073d9f6006d9e61 31379 Author: Eric Anholt <anholt (a] freebsd.org> 31380 Date: Sun Aug 15 00:44:57 2004 +0000 31381 31382 Fix unresolved symbol warnings in r128 driver (Adam Jackson) 31383 31384 commit 955f2b89f94aaa20b5eafd045b01e3eeef3d29ad 31385 Author: Kevin E Martin <kem (a] kem.org> 31386 Date: Thu Aug 12 05:00:22 2004 +0000 31387 31388 Fix context switching on Radeons and build problem when XF86DRI is not 31389 defined (Bug #922, Eric Anholt). 31390 31391 commit 8f3ea53f26493a5f02410297a587d1d95c067094 31392 Author: Adam Jackson <ajax (a] nwnk.net> 31393 Date: Thu Aug 12 02:23:46 2004 +0000 31394 31395 oops, these didn't added somehow. 31396 31397 commit 6ec74414bd33e9d8dfd5119058d15f3a6206e9d9 31398 Author: Adam Jackson <ajax (a] nwnk.net> 31399 Date: Thu Aug 12 01:03:06 2004 +0000 31400 31401 Factor out R{128,ADEON}Chipsets into their own headers so they can be 31402 sanely included in both the ati core and the r128/radeon submodules; 31403 r128 and radeon can now be directly loaded under dlloader. Teach imstt 31404 about the dlloader- frendly way to talk to fbdevhw. (Partial for bugs 31405 #393 and #400.) 31406 31407 commit a5b1e4dd7adacd17b63582e477df3a585efb249a 31408 Author: Adam Jackson <ajax (a] nwnk.net> 31409 Date: Wed Aug 11 01:20:28 2004 +0000 31410 31411 Bug #978: Silence more unresolved symbol warnings from the ati driver. 31412 Patch from Stefan Dirsch. 31413 31414 commit 1356884132491010a2e41eb2b5be4a227b9b8138 31415 Author: Kevin E Martin <kem (a] kem.org> 31416 Date: Wed Aug 11 01:14:14 2004 +0000 31417 31418 Fix Radeon build problem with #define DoLoadableServer NO 31419 31420 commit f5c93cb015e334975c2797a23b68798372a284c9 31421 Author: Adam Jackson <ajax (a] nwnk.net> 31422 Date: Tue Aug 10 04:01:42 2004 +0000 31423 31424 Bug #978: Silence unresolved symbol warnings from the ati driver. Reported 31425 by Michel Daenzer. 31426 31427 commit a2da1caf5848a78b10afd99811fd1a5677af6786 31428 Author: Kevin E Martin <kem (a] kem.org> 31429 Date: Mon Aug 9 22:37:22 2004 +0000 31430 31431 Fix install problem on platforms not using xorg.cf/xfree86.cf (Bug #339, 31432 Harold L. Hunt II, Alexander Gottwald). 31433 Fix crash when using X core font in zh_CN.UTF-8 locale (Bug #368, Yu Shao, 31434 David Dawes). 31435 Fix glXMakeCurrent(Dpy, None, NULL) crash (Bug #719, Adam Jackson). 31436 HP-PA build fix (Bug #828, Guy Martin, Paul Anderson). 31437 Fix SDK build for GATOS and Wacom driver (Bug #829, Bryan Stine). 31438 Fix attempt to read video ROM before enabling it (Bug #843, Ivan 31439 Kokshaysky, Mike A. Harris). 31440 Fix detection of primary adapter (Bug #843, Ivan Kokshaysky, Mike A. 31441 Harris). 31442 Clarify xset man page description of how to use the keyboard repeat rate 31443 settings (Bug #846, Mike A. Harris). 31444 Fix problem where print-screen key would get remapped to sys-req in certain 31445 keymaps, which broke GNOME printscreen functionality (Bug #847, Owen 31446 Taylor). 31447 Fix several render problems: 31448 - MMIO mode support 31449 - Hang on IGP chips 31450 - VT switching hang 31451 - 3D render corruption (Bug #922, Hui Yu). 31452 31453 commit 3171664c6a02424736864204b731249097fa7aae 31454 Author: Kevin E Martin <kem (a] kem.org> 31455 Date: Fri Aug 6 20:07:35 2004 +0000 31456 31457 Disable Render acceleration in Radeon driver temporarily until it no longer 31458 conflicts with the 3D code. 31459 31460 commit 27ef7adf3c1f6c3ca79b42b468d08b8542348cdd 31461 Author: Michel Daenzer <michel (a] daenzer.net> 31462 Date: Wed Aug 4 13:17:31 2004 +0000 31463 31464 Use info->ModeReg.surface_cntl to restore byte swapping for the framebuffer 31465 aperture on big endian machines, remove superfluous local variables and 31466 register reads. 31467 Adapt framebuffer aperture byte swapping to texture format before copying 31468 data to offscreen area and restore it afterwards on big endian 31469 machines, fixes Render acceleration there. 31470 reviewed by: Hui Yu <hyu (a] ati.com>, Kevin E. Martin <kem (a] freedesktop.org> 31471 31472 commit 2efdc9f847fa9ff3096667b16df84817fc65605d 31473 Author: Eric Anholt <anholt (a] freebsd.org> 31474 Date: Wed Aug 4 10:05:37 2004 +0000 31475 31476 - Add two new XAA hooks, SetupForCPUToScreenTexture2 and 31477 SetupForCPUToScreenAlphaTexture2. These add a dstFormat argument after 31478 the previous format argument, which the driver needs to use to properly 31479 set up the destination format. Two new arrays are added for the list of 31480 destination formats supported that correspond to the previous format 31481 arrays for sources. 31482 - Make Render acceleration only occur when the new hook for that 31483 acceleration type is supplied and the dst format list is set, along 31484 with the src format list being set. Without knowing the destination 31485 format, the Render acceleration couldn't properly support all the 31486 destinations it might encounter. 31487 - Bump XAA module minor version. 31488 - Update the Radeon Render acceleration to use the new hooks when the XAA 31489 module is sufficiently new. Fix a bug in the src/dst alpha booleans for 31490 ops, and use them to set blend_cntl to support destinations without 31491 alpha. Add missing PICT_a1r5g5b5 texture format, and add list 31492 terminator. (!) 31493 31494 commit 1e16c66551eed750330bb74a92208c45a7112cd0 31495 Author: Matthieu Herrb <matthieu.herrb (a] laas.fr> 31496 Date: Tue Aug 3 09:42:20 2004 +0000 31497 31498 programs/Xserver/hw/xfree86/drivers/ati/radeon_merged_fb.c Add missing 31499 argument to xf86DrvMsg(). 31500 31501 commit fd440a1e2110e05d116207800c123c281fbf5e19 31502 Author: Matthieu Herrb <matthieu.herrb (a] laas.fr> 31503 Date: Tue Aug 3 09:38:03 2004 +0000 31504 31505 programs/Xserver/hw/xfree86/drivers/ati/r128_driver.c 31506 programs/Xserver/hw/xfree86/drivers/ati/r128_probe.h 31507 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 31508 programs/Xserver/hw/xfree86/drivers/ati/radeon_probe.h Fix Constification 31509 problems. 31510 31511 commit dd0f86c4cb046c79a65144f767e97fd99e7f61c9 31512 Author: Kevin E Martin <kem (a] kem.org> 31513 Date: Sun Aug 1 23:22:01 2004 +0000 31514 31515 Fix Radeon crash on server exit or generation 31516 31517 commit c64f90b673f5c487dfe059f1636b04a1c2ab7dc2 31518 Author: Adam Jackson <ajax (a] nwnk.net> 31519 Date: Sat Jul 31 01:21:18 2004 +0000 31520 31521 Change several LoaderSymbol calls introduced by the bug #400 patch to 31522 *Weak() resolver functions. 31523 31524 commit b091b4b074cd61c9067ba95ef016f7a29a38c38d 31525 Author: Hui YU <hyu (a] ati.com> 31526 Date: Fri Jul 30 22:20:21 2004 +0000 31527 31528 Support for New radeon chips: R420/M18, R423, RV370/M22, RV380/M24, RS300. 31529 Add special handlings for DELL triple-head server (RV100). Misc. bug 31530 fixes for flat panel, host aperture, etc (Bug #946) 31531 31532 commit d04f61bfaf6952e832d52e1cd6014e86435aebd9 31533 Author: Adam Jackson <ajax (a] nwnk.net> 31534 Date: Fri Jul 30 20:30:51 2004 +0000 31535 31536 Bug #400 (partial): Driver fixes for the dlloader. When using dlloader, all 31537 framebuffer formats except cfb and the overlay modes should work, and 31538 r128 and radeon need to be loaded from the ati driver (both issues to 31539 be fixed soon). Tested on i740, s3virge, mach64, tdfx, vesa, and vga 31540 drivers. elfloader users shouldn't be affected. 31541 31542 commit 6dcb78bb818c3c1c7fe406bee5379f9bef023f4b 31543 Author: Alex Deucher <agd5f (a] yahoo.com> 31544 Date: Wed Jul 28 02:35:30 2004 +0000 31545 31546 - add gamma correction support to the radeon video overlay 31547 31548 commit 26029dc9378bbd95835d397402bfc9e6b4f1a478 31549 Author: Alex Deucher <agd5f (a] yahoo.com> 31550 Date: Mon Jul 26 23:14:37 2004 +0000 31551 31552 - Add Radeon DynamicClocks option 31553 - Add small fixes and clean ups from ati's last code drop (typo_fixes, 31554 remove_fudge, laptop, xvfix) 31555 - fix possible segfault in mga_dri.c (Ryan Underwood) 31556 - Add Xv support to pre-nm2160 neomagic chipsets 31557 31558 commit 9f251b16e702da486f824307e72a6f80c6987b5d 31559 Author: Adam Jackson <ajax (a] nwnk.net> 31560 Date: Mon Jul 26 19:14:58 2004 +0000 31561 31562 Bug #848: Clean up gatos notice messages. 31563 31564 commit 6f1e01ecdb11913fd768b9918f516b8379b7858f 31565 Author: Eric Anholt <anholt (a] freebsd.org> 31566 Date: Fri Jul 23 05:36:14 2004 +0000 31567 31568 - Add a new option, BuildDevelDRIDrivers, which controls whether 31569 DevelDRIDrivers is non-empty. DevelDRIDrivers contains those drivers 31570 aren't built by default either due to being unusable on that 31571 architecture (but buildable for testing) or due to security concerns, 31572 as in the mach64 and savage cases. 31573 - Merge the Mach64 DDX's DRI support, under the XF86DRI_DEVEL define, which 31574 is set only when BuildDevelDRIDrivers is set. 31575 - Fix up of the sets of DRI drivers built for various architectures. 31576 - Fix build of unichrome DRI driver, which is now enabled on x86. 31577 31578 commit b68322f467b7e524092af15e33ab872abb99b6a2 31579 Author: Eric Anholt <anholt (a] freebsd.org> 31580 Date: Thu Jul 22 07:03:04 2004 +0000 31581 31582 Merge and resolve conflicts from DRI/DRM/Mesa 20040721 imports. 31583 31584 commit 57de5e285d6271fc4b2bc8a951a98be6308d649d 31585 Author: Eric Anholt <anholt (a] freebsd.org> 31586 Date: Fri Jul 16 04:38:54 2004 +0000 31587 31588 Bugzilla #803,804: Fix build in the !XF86DRI case after DRI and Radeon 31589 Render acceleration merges. 31590 31591 commit 9b0ec803bd588d9af778bfd307ee6d958e4a1e73 31592 Author: Adam Jackson <ajax (a] nwnk.net> 31593 Date: Sat Jul 10 14:22:35 2004 +0000 31594 31595 Bugzilla #849: Fix DRI pagesize assumptions in radeon and r128 drivers. 31596 from mharris@, reviewed and applied to DRI by airlied@. 31597 31598 commit ccd95bfb291a0c70272fab6822ef379b83dc298d 31599 Author: Egbert Eich <eich (a] suse.de> 31600 Date: Tue Jul 6 14:37:47 2004 +0000 31601 31602 Separated Intel drivers from default DriDrivers to avoid building them on 31603 IA64 (Egbert Eich). 31604 Fixed wrong function prototype (Egbert Eich). 31605 Don't test for generic VGA on IA64 (Egbert Eich). 31606 Fixed a segfault when accessing a structure before verifying the pointer 31607 exists (Egbert Eich). 31608 Added a showcache option for debugging (Egbert Eich). 31609 Increase default video RAM size to 16MB when DRI is enabled and more than 31610 128MB are available (Egbert Eich). Fixed lockups during mode switch. 31611 Problem was introduced when attempting to copy the behavior during 31612 LeaveVT()/EnterVT() but but forgetting to call I810DRILeave() before 31613 I810DRIEnter(). The entire DRILeave()/Enter() scenario has been 31614 commented out as it didn't seem to be necessary (Egbert Eich). 31615 Fix TweakMemorySize() (tested with i855/i865) (Egbert Eich). 31616 increased MAX_DEVICES to 128 (Egbert Eich). 31617 Use OS provided PCI config space access as default method (Egbert Eich). 31618 Added support for Linux 2.6 proc file format. 31619 Fixed unaligned accesses to pieces of the VBE info block. VESA did not 31620 align elements to size (Egbert Eich). 31621 31622 commit 6bdf3d36ea3086432af5157540fde4e0aa0c44dc 31623 Author: Eric Anholt <anholt (a] freebsd.org> 31624 Date: Sun Jun 20 07:34:55 2004 +0000 31625 31626 Add file missed in last commit: Bug #748: 31627 - Add Render acceleration for Radeon 100 and 200-series cards, enabled by 31628 default. Notable performance gains include 27fps in cairogears instead 31629 of 6fps (compared to 50 with glitz), and my disks are now the 31630 bottleneck in an ls -lR in gnome-terminal. Only supported in DRI mode 31631 because the MMIO submission hangs the card so far, but the code is left 31632 in because it may be supportable soon. 31633 - Add xorg.conf options to disable Render acceleration and to force the 31634 subpixel order in the server (Hui Yu). 31635 Many thanks to Hui Yu at ATI for the code this was based off of, Michel 31636 Daenzer for bugfixes and suggestions, and LinuxFund for sponsoring the 31637 work. 31638 31639 commit 8c1b37420ea890f76edb97aed61ee427a871cd3f 31640 Author: Eric Anholt <anholt (a] freebsd.org> 31641 Date: Sun Jun 20 07:26:08 2004 +0000 31642 31643 Bug #748: 31644 - Add Render acceleration for Radeon 100 and 200-series cards, enabled by 31645 default. Notable performance gains include 27fps in cairogears instead 31646 of 6fps (compared to 50 with glitz), and my disks are now the 31647 bottleneck in an ls -lR in gnome-terminal. Only supported in DRI mode 31648 because the MMIO submission hangs the card so far, but the code is left 31649 in because it may be supportable soon. 31650 - Add xorg.conf options to disable Render acceleration and to force the 31651 subpixel order in the server (Hui Yu). 31652 Many thanks to Hui Yu at ATI for the code this was based off of, Michel 31653 Daenzer for bugfixes and suggestions, and LinuxFund for sponsoring the 31654 work. 31655 31656 commit 6ecf374d500afe6da494dfdd6566396ec65b6d6a 31657 Author: Eric Anholt <anholt (a] freebsd.org> 31658 Date: Wed Jun 16 09:43:59 2004 +0000 31659 31660 Merge DRI-trunk-20040613 changes in programs/Xserver/hw/xfree86/drivers, 31661 with the following notes: 31662 - Savage and Mach64 (= ati/ati*.[ch] changes) DRI not merged due to 31663 insecurity. 31664 - VIA driver converted to new drmContext and drmHandle names. 31665 - Radeon driver merge conflicted in many places, and MergedFB at least 31666 could probably use some checking at this point. 31667 31668 commit bea8085e04136b0ef513c17bb65c54069ec531e1 31669 Author: Eric Anholt <anholt (a] freebsd.org> 31670 Date: Wed Jun 16 09:25:58 2004 +0000 31671 31672 DRI trunk-20040613 import 31673 31674 commit 7b588cda922992a8ee2d04853ba1533cf43592a3 31675 Author: Eric Anholt <anholt (a] freebsd.org> 31676 Date: Wed Jun 16 09:25:58 2004 +0000 31677 31678 Initial revision 31679 31680 commit 6cbb5bbff43d65a762d432659793333868837f0b 31681 Author: Eric Anholt <anholt (a] freebsd.org> 31682 Date: Wed Jun 16 09:22:56 2004 +0000 31683 31684 DRI XFree86-4_3_99_12-merge import 31685 31686 commit 3586c3c7955365a2af0b8a286ea0d0488ddfd137 31687 Author: Egbert Eich <eich (a] suse.de> 31688 Date: Fri Apr 23 19:26:46 2004 +0000 31689 31690 Merging XORG-CURRENT into trunk 31691 31692 commit 538768e0f472026325be9dd37dc4bfe7ea7d3b75 31693 Author: Egbert Eich <eich (a] suse.de> 31694 Date: Sun Mar 14 08:33:20 2004 +0000 31695 31696 Importing vendor version xf86-4_4_99_1 on Sun Mar 14 00:26:39 PST 2004 31697 31698 commit 5f179e796d1cd310790d06eebb474ad67114ec84 31699 Author: Egbert Eich <eich (a] suse.de> 31700 Date: Wed Mar 3 12:12:18 2004 +0000 31701 31702 Importing vendor version xf86-4_4_0 on Wed Mar 3 04:09:24 PST 2004 31703 31704 commit 493c44580191beff63575c4032f772fbbf1b06eb 31705 Author: Egbert Eich <eich (a] suse.de> 31706 Date: Thu Feb 26 13:35:52 2004 +0000 31707 31708 readding XFree86's cvs IDs 31709 31710 commit 942a47fd3c5e8df85642672c01072bf97285bc30 31711 Author: Egbert Eich <eich (a] suse.de> 31712 Date: Thu Feb 26 09:23:18 2004 +0000 31713 31714 Importing vendor version xf86-4_3_99_903 on Wed Feb 26 01:21:00 PST 2004 31715 31716 commit dda79b5ee3eb49f924a3213592d93ba981c99394 31717 Author: Kaleb Keithley <kaleb (a] freedesktop.org> 31718 Date: Mon Feb 23 20:35:02 2004 +0000 31719 31720 Import most of XFree86 4.4RC3. This import excludes files which have the 31721 new license. If we want to, later we can import 4.4RC3 again and pick 31722 up the files that have the new license, but for now the vendor branch 31723 is "pure." 31724 31725 commit 08104f18ac497561c969694ff709da6c361ccea7 31726 Author: Egbert Eich <eich (a] suse.de> 31727 Date: Thu Jan 29 08:08:34 2004 +0000 31728 31729 Importing vendor version xf86-012804-2330 on Thu Jan 29 00:06:33 PST 2004 31730 31731 commit 162d76b9c48ed17d6e645057db58c9f13436f25f 31732 Author: Kaleb Keithley <kaleb (a] freedesktop.org> 31733 Date: Fri Dec 19 20:55:06 2003 +0000 31734 31735 XFree86 4.3.99.902 (RC 2) 31736 31737 commit 195772ec74f7a9b7d5bf294b3bc2c3b960e08a31 31738 Author: Kaleb Keithley <kaleb (a] freedesktop.org> 31739 Date: Thu Dec 4 22:03:16 2003 +0000 31740 31741 XFree86 4.3.99.901 (RC 1) 31742 31743 commit 73105e438e90f950249cb265ba99001d4db3dc60 31744 Author: Kaleb Keithley <kaleb (a] freedesktop.org> 31745 Date: Tue Nov 25 19:28:36 2003 +0000 31746 31747 XFree86 4.3.99.16 Bring the tree up to date for the Cygwin folks 31748 31749 commit 0a53864f9581196604d04fd28b4e94a9b0b73d18 31750 Author: Kaleb Keithley <kaleb (a] freedesktop.org> 31751 Date: Tue Nov 25 19:28:36 2003 +0000 31752 31753 Initial revision 31754 31755 commit c56aaef79028261e0a3be0c86ffd3f16b6bdd9ba 31756 Author: Kaleb Keithley <kaleb (a] freedesktop.org> 31757 Date: Fri Nov 14 16:48:55 2003 +0000 31758 31759 XFree86 4.3.0.1 31760 31761 commit d09f463a5d1ce73e0b65d5276fbcca393fa2da46 31762 Author: Kaleb Keithley <kaleb (a] freedesktop.org> 31763 Date: Fri Nov 14 16:48:55 2003 +0000 31764 31765 Initial revision 31766 31767 commit d9af6dc32652502d84ea8da5d57a5ab45429c4ad 31768 Author: Kaleb Keithley <kaleb (a] freedesktop.org> 31769 Date: Fri Nov 14 15:54:49 2003 +0000 31770 31771 R6.6 is the Xorg base-line 31772
Indexes created Wed Oct 28 00:26:01 UTC 2026