ChangeLog revision 6322c902
1commit df6662864b56d1aeeea4c67d9b668e64197afb20 2Author: Michel Dänzer <michel.daenzer@amd.com> 3Date: Thu Apr 7 11:44:05 2016 +0900 4 5 Bump version for 7.7.0 release 6 7commit c12614cfdc5729041c204e3ead1dbfde38011f0c 8Author: Michel Dänzer <michel.daenzer@amd.com> 9Date: Mon Apr 4 14:59:41 2016 +0900 10 11 Update manpage entry for Option "TearFree" 12 13 It's now effective for rotation as well. 14 15 (Ported from amdgpu commit faf9d720b7d650f5f1ea657a874d08eac3972e60) 16 17 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 18 19commit 4693b1bd5b5c381e8b7b68a6f7f0c6696d6a68df 20Author: Michel Dänzer <michel.daenzer@amd.com> 21Date: Thu Mar 31 17:02:55 2016 +0900 22 23 Identify DRM event queue entries by sequence number instead of by pointer 24 25 If the memory for an entry was allocated at the same address as that for 26 a previously cancelled entry, the handler could theoretically be called 27 prematurely, triggered by the DRM event which was submitted for the 28 cancelled entry. 29 30 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 31 32commit 83734317e6bdaeebb4462a63f541e73a1d7c2f77 33Author: Michel Dänzer <michel.daenzer@amd.com> 34Date: Wed Mar 30 11:44:09 2016 +0900 35 36 Update pixmap pitch in radeon_set_pixmap_bo 37 38 Stop second guessing it in drmmode_crtc_scanout_create. 39 40 Fixes display corruption in some cases with TearFree enabled. 41 42 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94751 43 44 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 45 46commit 0b3aac1de9db42bfca545fa331e4985836682ec7 47Author: Michel Dänzer <michel.daenzer@amd.com> 48Date: Mon Mar 28 18:43:36 2016 +0900 49 50 DRI3: Refuse to open DRM file descriptor for ssh clients (v2) 51 52 Fixes hangs when attempting to use DRI3 on display connections forwarded 53 via SSH. 54 55 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93261 56 57 v2: Don't do this for Xorg > 1.18.99.1 since the corresponding xserver 58 change has landed in Git master. 59 60 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v1) 61 62commit 8a6cd4bda05b9569b3dd0a5a75b2cc385b9ecba9 63Author: Michel Dänzer <michel.daenzer@amd.com> 64Date: Fri Mar 25 11:50:15 2016 +0900 65 66 Revert "Use render node for DRI3 if available" 67 68 This reverts commit cd94248ffa7d8fe0b57476f79e7e860dee66d1b0. 69 70 It broke VDPAU<->GL interop with DRI3 enabled, because the Gallium VDPAU 71 code doesn't support DRI3 yet. We can consider re-enabling this once 72 there is a Mesa release where the Gallium VDPAU code supports DRI3. 73 74 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94675 75 76 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 77 78commit ed31cffba0d1bd4b14e5348a1456e4377277059c 79Author: Michel Dänzer <michel.daenzer@amd.com> 80Date: Thu Mar 24 18:08:22 2016 +0900 81 82 Call RADEONBlockHandler_KMS before setting initial modes 83 84 Doing it the other way around meant that there was still a possibility 85 for the front buffer contents to be uninitialized when they start being 86 scanned out. 87 88 (Ported from amdgpu commit 4a60b4b1851a3cbc2d8ad9048d68eeb6947cf132) 89 90 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 91 92commit e592f32f8b5f5873fcc18b10a69dd5e4ccf11073 93Author: Michel Dänzer <michel.daenzer@amd.com> 94Date: Wed Mar 23 18:25:48 2016 +0900 95 96 Require xserver 1.9 or newer 97 98 1.9.0 was released in August 2010. 99 100 We were already unintentionally relying on things not available in 1.8 101 for at least a year, and nobody has complained. 102 103 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 104 105commit bde466e5d44cad64b4e4eceaa5de80fdbf86356e 106Author: Michel Dänzer <michel.daenzer@amd.com> 107Date: Wed Mar 23 18:16:42 2016 +0900 108 109 Fix build against older versions of xserver 110 111 Also slightly clean up the error handling in radeon_scanout_do_update. 112 113 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94614 114 115 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 116 117commit b2b079be0e7ef2704c3df9ee25fe46166e450e83 118Author: Michel Dänzer <michel.daenzer@amd.com> 119Date: Wed Mar 23 17:29:47 2016 +0900 120 121 Use radeon_alloc_pixmap_bo for allocating scanout BOs 122 123 This enables tiling for scanout BOs used for TearFree / ShadowPrimary, 124 and simplifies the code a little. 125 126 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 127 128commit d21ac4669a8b2cdd4eec5e5a94d1950b7423b8b5 129Author: Michel Dänzer <michel.daenzer@amd.com> 130Date: Wed Mar 23 17:27:27 2016 +0900 131 132 DRI3 only works with acceleration 133 134 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94214 135 136 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 137 138commit eb1a2c57e6c45d7ca57b2e2be643194243b7aae1 139Author: Michel Dänzer <michel.daenzer@amd.com> 140Date: Wed Mar 23 17:23:31 2016 +0900 141 142 TearFree can only work with acceleration 143 144 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 145 146commit d670c5c9851b4eff21c845d26c7d7e4eb5ee0fa9 147Author: Michel Dänzer <michel.daenzer@amd.com> 148Date: Fri Mar 18 16:46:08 2016 +0900 149 150 Check for xf86CursorResetCursor 151 152 If it's available, Xorg calls it on each mode configuration change. It 153 does what xf86_reload_cursors does (and more), so we don't need to call 154 the latter anymore. 155 156 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 157 158commit 7f3d0780ca65a90117c2a61362dbc0899bd9c0b0 159Author: Michel Dänzer <michel.daenzer@amd.com> 160Date: Wed Mar 9 18:28:47 2016 +0900 161 162 Don't try DRI2/Present flipping while the HW cursor can't be used 163 164 Flipping doesn't interact correctly with SW cursor: A flip makes the SW 165 cursor disappear. It will only appear again when the cursor is moved, 166 but it will be surrounded by corruption, because the SW cursor code 167 will restore stale screen contents at the old cursor location before 168 drawing the cursor at the new location. 169 170 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 171 172commit 3de480e83c0a1824838d662d6d67c9fe85277298 173Author: Michel Dänzer <michel.daenzer@amd.com> 174Date: Wed Mar 9 16:44:32 2016 +0900 175 176 Factor out HW cursor checking code into drmmode_can_use_hw_cursor 177 178 And add a check for RandR 1.4 multihead. 179 180 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 181 182commit 2e0b458e747c7df81dd5f9579a762262af3350e1 183Author: Ryan Kennedy <Ryan.Kennedy@amd.com> 184Date: Sun Mar 20 17:57:56 2016 -0400 185 186 Use correct tiling in drmmode_create_bo_pixmap 187 188 The callers were using 0 for the tiling flags, causing surfaces to be 189 incorrectly treated as linear. This issue was causing corruption on Cayman 190 just after login with GNOME3. The kernel was rejecting the cs in 191 drmmode_copy_fb due to invalid texture pitch. 192 193 Signed-off-by: Ryan Kennedy <Ryan.Kennedy@amd.com> 194 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 195 196commit a03271de5ecdaa7790d1316e993c4450b91fe936 197Author: Michel Dänzer <michel.daenzer@amd.com> 198Date: Wed Mar 9 18:23:10 2016 +0900 199 200 present: Return rotated CRTCs from radeon_present_get_crtc 201 202 Sync-to-vblank works fine with rotation. We're still checking for 203 rotation in radeon_present_check_flip. 204 205 Returning NULL from here resulted in the xserver present code falling 206 back to the fake CRTC running at 1 fps. 207 208 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 209 210commit 40191d82370eb7e58bd34c44966cbf44c3703229 211Author: Michel Dänzer <michel.daenzer@amd.com> 212Date: Wed Mar 9 18:15:57 2016 +0900 213 214 present: Clear drmmode->fb_id before calling set_mode_major for unflip 215 216 Without this, drmmode_set_mode_major may just re-set the FB for the 217 last flipped BO, in which case the display will probably freeze. 218 219 Reproduction recipe: Enable rotation while a fullscreen client is 220 flipping. 221 222 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 223 224commit 798c4fd16d339b1ad5fd729cc884be084c60e38b 225Author: Michel Dänzer <michel.daenzer@amd.com> 226Date: Fri Dec 25 18:57:42 2015 +0900 227 228 Make Option "TearFree" effective for rotated/reflected outputs as well (v2) 229 230 Support varies by xserver version: 231 232 < 1.12: No support for the driver handling rotation/reflection 233 1.12-1.15: Support for driver handling rotation/reflection, but there's 234 a bug preventing the HW cursor from being visible everywhere 235 it should be on rotated outputs, so we can only support 236 TearFree for reflection. 237 >= 1.16: While the bug above is still there (fixes pending review), 238 the driver can force SW cursor for rotated outputs, so we 239 can support TearFree for rotation as well. 240 241 v2: Don't set crtc->driverIsPerformingTransform after xf86CrtcRotate if 242 it wasn't set before. Fixes breaking rotation with TearFree disabled. 243 244 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v1) 245 246commit eb611a2e4ecce7a1ab85fd72b9b78e3269311dd5 247Author: Michel Dänzer <michel.daenzer@amd.com> 248Date: Tue Feb 2 18:04:11 2016 +0900 249 250 Consolidate pScreen usage in drmmode_set_mode_major 251 252 We were already relying on pScrn->pScreen being non-NULL in some cases, 253 which is supposedly always true ever since this function is no longer 254 getting called from ScreenInit. 255 256 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 257 258commit 06602171386e538081c298645fb7ca1a70fe80cc 259Author: Michel Dänzer <michel.daenzer@amd.com> 260Date: Thu Feb 4 17:13:57 2016 +0900 261 262 Remove check for XF86_CRTC_VERSION 3 263 264 We require xserver >= 1.8, which was already at version 3. 265 266 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 267 268commit a88985f5d1e39caca49ceb65678aaa9cb622a0d2 269Author: Michel Dänzer <michel.daenzer@amd.com> 270Date: Mon Feb 29 18:12:43 2016 +0900 271 272 Deal with modesets and page flips crossing on a CRTC 273 274 If we set a mode while a flip is pending, the kernel driver may program 275 the flip to the hardware after the modeset. If that happens, the hardware 276 will display the BO from the flip, whereas we will assume it displays the 277 BO from the modeset. In other words, the display will most likely freeze, 278 at least until another modeset. 279 280 Prevent this condition by waiting for a pending flip to finish before 281 setting a mode. 282 283 Fixes display freezing when setting rotation or a transform with 284 TearFree enabled. 285 286 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 287 288commit f5d968cbba3c9b7ec202161f2157d8d64778c817 289Author: Michel Dänzer <michel.daenzer@amd.com> 290Date: Thu May 21 12:54:31 2015 +0900 291 292 Make DRM event queue xf86CrtcPtr based instead of ScrnInfoPtr based 293 294 This allows for a minor simplification of the code. 295 296 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 297 298commit e87365117acbd80b7d80fbb5eb30890ef7153291 299Author: Michel Dänzer <michel.daenzer@amd.com> 300Date: Thu Feb 25 16:49:16 2016 +0900 301 302 DRI2: Also clear dri2_flipping when client disconnects before event 303 304 Fixes the following problem: 305 306 With DRI3 enabled, run glxgears with LIBGL_DRI3_DISABLE=1, make it 307 fullscreen and press Escape while it's still fullscreen. This could 308 result in dri2_flipping not getting cleared, spuriously preventing apps 309 using DRI3 from flipping. 310 311 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 312 313commit d5dbb07db22d5420c81dfebc060f0dd86e7b8a20 314Author: Michel Dänzer <michel.daenzer@amd.com> 315Date: Thu Feb 25 18:02:12 2016 +0900 316 317 Remove radeon_scanout_flip_handler 318 319 No longer necessary now that radeon_drm_queue_handler can handle 320 e->handler == NULL. 321 322 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 323 324commit 3989766edde85d1abe7024577b98fc9b007bc02a 325Author: Michel Dänzer <michel.daenzer@amd.com> 326Date: Thu Feb 25 16:43:48 2016 +0900 327 328 drm_queue: Don't abort events immediately from radeon_drm_abort_client 329 330 Keep them around until the DRM event arrives, but then call the abort 331 functions instead of the handler functions. 332 333 This is a prerequisite for the following fix. 334 335 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 336 337commit 7bc089ee9740a3ef91f16e36786d8ed64e165acd 338Author: jimqu <Jim.Qu@amd.com> 339Date: Tue Feb 16 17:05:48 2016 +0900 340 341 glamor: Return NullPixmap on failure to create shareable pixmap 342 343 If we were asked to create a shareable pixmap, it doesn't make sense 344 to return a pixmap which isn't shareable. Doing so caused trouble down 345 the line such as a crash with older versions of glamor when trying to 346 use GLX pixmaps of bpp < 32 via DRI2. 347 348 Signed-off-by: JimQu <jim.qu@amd.com> 349 (ported from amdgpu commit 2fcb7dadd3c71cd405cbbaafc777697538ca9c29) 350 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 351 352commit 4cc32031467157ab8788f5c684fb1ac67ae96ff9 353Author: jimqu <Jim.Qu@amd.com> 354Date: Tue Feb 16 17:03:24 2016 +0900 355 356 Move radeon_glamor_destroy_pixmap before radeon_glamor_create_pixmap 357 358 The next commit will call the former from the latter. No functional 359 change. 360 361 Signed-off-by: JimQu <jim.qu@amd.com> 362 (ported from amdgpu commit 5269a2228bff6023c1a7f3e8534027e1d7addc25) 363 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 364 365commit a9b181083d56b5de9b6822d0ed89fcc86de777a6 366Author: Michel Dänzer <michel.daenzer@amd.com> 367Date: Tue Feb 16 16:59:40 2016 +0900 368 369 Remove pci_dev test from radeon_get_scrninfo 370 371 The pci_dev parameter can never be NULL since we only support KMS. 372 373 Reported-by: Tom St Denis <tom.stdenis@amd.com> 374 (ported from amdgpu commit 6e42c58375a4c3229da93c27bbd104af145c6163) 375 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 376 377commit 2c7f068b0a0493daae86e5eacdf8d9812fcd28dc 378Author: Mykola Lysenko <Mykola.Lysenko@amd.com> 379Date: Mon Feb 15 18:21:39 2016 +0900 380 381 Initialize drmmode_crtc dpms_mode to DPMSModeOff 382 383 This disables query of disabled pipes for drmWaitVBlank on X start 384 385 Signed-off-by: Mykola Lysenko <Mykola.Lysenko@amd.com> 386 (ported from amdgpu commit 4eb9cedca080b30c57ded349a397620ee7d0cd46) 387 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 388 389commit 9358e44f03c95396b7f7a4f7398f5b24d2559bb2 390Author: Michel Dänzer <michel.daenzer@amd.com> 391Date: Mon Feb 15 18:20:31 2016 +0900 392 393 sync: Check if miSyncShmScreenInit symbol is resolved at runtime 394 395 It may be disabled in the Xorg build, either explicitly or because the 396 xshmfence library isn't available. 397 398 (ported from amdgpu commit 1d0b0c1794e65e581a48aa9fb19679d928d82a5d) 399 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 400 401commit cd94248ffa7d8fe0b57476f79e7e860dee66d1b0 402Author: Jammy Zhou <Jammy.Zhou@amd.com> 403Date: Tue Feb 16 16:36:33 2016 +0900 404 405 Use render node for DRI3 if available 406 407 Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com> 408 (ported from amdgpu commit ea558e645786b08d75307716036045170e97b43e) 409 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 410 411commit d41fcccf2f11d0393d252e76bbbe6d233c4ac443 412Author: Michel Dänzer <michel.daenzer@amd.com> 413Date: Wed Jan 20 11:20:30 2016 +0900 414 415 Set RADEON_GEM_NO_CPU_ACCESS flag for BOs which don't need CPU access 416 417 Failing to do this was resulting in the kernel driver unnecessarily 418 leaving open the possibility of CPU access to those BOs. 419 420 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 421 422commit 9483a3d777919b224f70c3b4d01e4b320a57db31 423Author: Michel Dänzer <michel.daenzer@amd.com> 424Date: Fri Jan 15 16:48:46 2016 +0900 425 426 Fix RandR CRTC transforms 427 428 Currently, Xorg will only transform the cursor as of the first time the 429 cursor image changes after a transform is set. 430 431 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=80678 432 433 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 434 435commit b16856b25086ffb27365ac2249b8da921066ce62 436Author: Michel Dänzer <michel.daenzer@amd.com> 437Date: Thu Jan 14 17:32:54 2016 +0900 438 439 Set the RandR primary output on startup if Xorg hasn't 440 441 Fixes xrandr (XRRGetOutputPrimary) not reporting any output as primary 442 after startup. 443 444 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 445 446commit c7cf00487cd6d4a5d0f39d5b92ff04f6420d6a32 447Author: Michel Dänzer <michel.daenzer@amd.com> 448Date: Fri Nov 27 12:52:28 2015 +0900 449 450 Build RandR 1.4 provider name from chipset name and bus ID 451 452 Instead of just "radeon", it's now e.g. "KAVERI @ pci:0000:00:01.0". 453 454 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 455 456commit ff9a6b6f079a8419f4e6fadfee778060618bf735 457Author: Michel Dänzer <michel.daenzer@amd.com> 458Date: Fri Dec 11 17:21:27 2015 +0900 459 460 HAS_DIRTYTRACKING_ROTATION also supports multiple CRTCs 461 462 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 463 464commit 78fbca095ae9887a2d3de48bb07975e2d1126e68 465Author: Michel Dänzer <michel.daenzer@amd.com> 466Date: Tue Dec 1 15:38:13 2015 +0900 467 468 Load fb module before glamoregl/shadow modules 469 470 Fixes unresolved symbols on some systems. 471 472 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93105 473 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 474 475commit b19417e2fddf4df725951aea5ad5e9558338f59e 476Author: Michel Dänzer <michel.daenzer@amd.com> 477Date: Fri Nov 20 18:48:56 2015 +0900 478 479 Don't advertise any PRIME offloading capabilities without acceleration 480 481 Acceleration is required even for display offloading. Trying to enable 482 display offloading without acceleration resulted in a crash. 483 484 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 485 486commit 10b7c3def58bb34acc38f076bc230e25b454ab79 487Author: Michel Dänzer <michel.daenzer@amd.com> 488Date: Tue Nov 10 16:31:09 2015 +0900 489 490 glamor: Deal with glamor_egl_destroy_textured_pixmap being removed 491 492 When it's not available, it's safe to call down to the glamor 493 DestroyPixmap hook instead. 494 495 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 496 497commit 535e5438b2c32f774b9c8c27ee0289b4749548ef 498Author: Michel Dänzer <michel.daenzer@amd.com> 499Date: Tue Nov 10 17:31:52 2015 +0900 500 501 glamor: Restore all ScreenRec hooks during CloseScreen 502 503 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 504 505commit dfb52777e43f3855476cff812d64de687a199674 506Author: Michel Dänzer <michel.daenzer@amd.com> 507Date: Thu Nov 12 13:49:04 2015 +0900 508 509 Post 7.6.1 release version bump 510 511commit 99cb8c3faf1a4ce368b7500f17a2a7868c15e8e8 512Author: Michel Dänzer <michel.daenzer@amd.com> 513Date: Thu Nov 12 13:38:14 2015 +0900 514 515 Set version for 7.6.1 release 516 517commit 4a2114a717ec8d78cacd8fe158e132d62013faac 518Author: Michel Dänzer <michel.daenzer@amd.com> 519Date: Thu Nov 12 13:32:25 2015 +0900 520 521 Remove duplicate free(output_ids) call 522 523 Fixes double-free regression introduced by commit 875ad48e ("Simplify 524 drmmode_set_mode_major() and avoid leaking memory."). 525 526commit 960a9f37096b53ec9603a99c7b1f3679f915e03f 527Author: Michel Dänzer <michel.daenzer@amd.com> 528Date: Thu Nov 12 13:20:18 2015 +0900 529 530 Post 7.6.0 release version bump 531 532commit 9aae4b71fb402841439e3267fcb7f410d8429c3a 533Author: Michel Dänzer <michel.daenzer@amd.com> 534Date: Thu Nov 12 11:56:48 2015 +0900 535 536 Bump version for 7.6.0 release 537 538commit 98291869ac4a542a0b478920586407ff9d2c8ef0 539Author: Tom St Denis <tom.stdenis@amd.com> 540Date: Wed Nov 11 16:01:41 2015 +0900 541 542 Clean up radeon_dri2_create_buffer2() 543 544 Remove the depth_pixmap variable from the function and clear 545 out any dead/odd behaviour that results. 546 547 Signed-off-by: Tom St Denis <tom.stdenis@amd.com> 548 (ported from amdgpu commit 6000aef4e2f0a121b94023484406fb6f04688f74) 549 550 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 551 552commit c6fc7e309a8a922f94a1f5f3e8bfb9058cff7ad1 553Author: Michel Dänzer <michel.daenzer@amd.com> 554Date: Wed Nov 11 15:59:06 2015 +0900 555 556 Properly handle drmModeAddFB failure in drmmode_crtc_scanout_allocate 557 558 We were printing an error message, but not propagating the failure. That 559 would probably lead to trouble down the road. 560 561 (ported from amdgpu commit 21e72fb2418b5cc7fc849a9cf951186e209036b0) 562 563 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 564 565commit 4e4f4d53da0539ef9feb8766230a6e9927ae005b 566Author: Michel Dänzer <michel.daenzer@amd.com> 567Date: Wed Nov 11 15:57:21 2015 +0900 568 569 Eliminate redundant data parameter from drmmode_crtc_scanout_create 570 571 drmmode_crtc_scanout_create just needs to call 572 drmmode_crtc_scanout_allocate when scanout->bo is NULL. 573 574 This makes it clearer to the reader / compiler that 575 drmmode_crtc_scanout_create doesn't dereference scanout->bo when it's 576 NULL. 577 578 (ported from amdgpu commit 8da1d0c870e1081d77925807d6e3bbc61a23f54f) 579 580 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 581 582commit 421a7e797bdd58d83e81af7a6512cc715a3df514 583Author: Michel Dänzer <michel.daenzer@amd.com> 584Date: Wed Nov 11 15:51:58 2015 +0900 585 586 Don't advertise rotation support without hardware acceleration v2 587 588 Rotation currently doesn't work without acceleration (doesn't actually 589 rotate with Option "NoAccel", crashes with Option "AccelMethod" "none" 590 or when glamor fails to initialize) and would probably be too slow 591 anyway. 592 593 v2: Also remove now dead code checking for ShadowFB from 594 drmmode_crtc_scanout_allocate(). 595 596 (ported from amdgpu commit dc40582d5ff94d812cbc08f95cf14b80cd0f410d) 597 598 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 599 600commit 875ad48e7b5cdb7beefbf18dddcbee3ed22b5446 601Author: Tom St Denis <tom.stdenis@amd.com> 602Date: Wed Nov 11 15:51:19 2015 +0900 603 604 Simplify drmmode_set_mode_major() and avoid leaking memory. 605 606 The function would leak the memory allocated for output_ids. This 607 patch addresses that as well as simplifies the logic somewhat. 608 609 Signed-off-by: Tom St Denis <tom.stdenis@amd.com> 610 (ported from amdgpu commit 460560502a1bdf26d06f3c30df46fa9f28ffb9e5) 611 612 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 613 614commit 789d7d6a04cca6b36fb088a074027807ccb8dd61 615Author: Tom St Denis <tom.stdenis@amd.com> 616Date: Wed Nov 11 15:48:51 2015 +0900 617 618 Clean up allocation in RADEONInitVideo() 619 620 The allocation of the adapters should use the correct sizeof (even if 621 allocating an array of pointers). 622 623 Signed-off-by: Tom St Denis <tom.stdenis@amd.com> 624 (ported from amdgpu commit db3bb2061b9ac16b0922d9afae99874820356a04) 625 626 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 627 628commit d88fa0dd5d37604de8efb05853738cfaca6a3166 629Author: Tom St Denis <tom.stdenis@amd.com> 630Date: Wed Nov 11 15:46:50 2015 +0900 631 632 Simplify pick best crtc to fold two loops into one 633 634 This patch folds the two for loops from radeon_pick_best_crtc() into 635 one to reduce the LOC and make the routine easier to read. 636 637 Signed-off-by: Tom St Denis <tom.stdenis@amd.com> 638 (ported from amdgpu commit 3055724aef76a624718f26d5f0f9e9d567ffbcfb) 639 640 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 641 642commit dbbcd75b3c80aba77673904d46bca97779fd8a8d 643Author: Tom St Denis <tom.stdenis@amd.com> 644Date: Wed Nov 11 12:54:54 2015 +0900 645 646 dri2: Avoid calculation with undefined msc value 647 648 If the get_msc() call fails for any reason we should avoid updating the 649 vblank counter delta with undefined data. 650 651 Signed-off-by: Tom St Denis <tom.stdenis@amd.com> 652 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> (minor fixups) 653 (ported from amdgpu commit 8823c3d4c6db70cff7699b31088f2d92db8faaf4) 654 655 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 656 657commit 7186a8713ba004de4991f21c1a9fc4abc62aeff4 658Author: Stephen Chandler Paul <cpaul@redhat.com> 659Date: Fri Oct 23 09:59:36 2015 -0400 660 661 Handle failures in setting a CRTC to a DRM mode properly 662 663 This fixes a bug where running the card out of PPLL's when hotplugging 664 another monitor would result in all of the displays going blank and 665 failing to work properly until X was restarted or the user switched to 666 another VT. 667 668 [Michel Dänzer: Pass errno instead of -ret to strerror()] 669 670 Signed-off-by: Stephen Chandler Paul <cpaul@redhat.com> 671 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 672 673commit 548e97b3b7d1e94075a54ca2bb4eb683025098a7 674Author: Michel Dänzer <michel.daenzer@amd.com> 675Date: Fri Oct 16 16:26:58 2015 +0900 676 677 Call xf86CrtcRotate from initial drmmode_set_desired_modes call 678 679 Fixes various problems when rotation is specified in xorg.conf. 680 681 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92475 682 683 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 684 685commit ce9914af8d8d5243977023ec7b09c605f9cca8b4 686Author: Michel Dänzer <michel.daenzer@amd.com> 687Date: Fri Oct 2 18:07:18 2015 +0900 688 689 Only align screen / scanout pixmap height where necessary 690 691 When using glamor acceleration, the pixmap's header has to have a height 692 that matches exactly what the actual height is minus the GPU memory 693 alignment. Otherwise CRTCs scanning out from the main scanout buffer 694 (e.g. every CRTC that isn't rotated or transformed in some way) won't 695 always work. This results in a bug where rotating one monitor in a 696 multi-monitor setup won't always work properly. Easiest way to reproduce 697 this: 698 699 - Have two monitors (I've gotten this working with a 1920x1080 and 700 1280x1024, along with two 1920x1080s) 701 - Rotate one of them from 0° to 90°, then rotate the same monitor from 702 90° to 180°. The monitor that hasn't been rotated won't properly 703 update, and will stay on a blank screen 704 705 This doesn't seem to make any difference when using EXA for 706 acceleration. 707 708 Compared to Stephen Chandler's patch, this drops the height alignment 709 in most places and only keeps it where it's really necessary. 710 711 Reported-and-Tested-by: Stephen Chandler Paul <cpaul@redhat.com> 712 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 713 714commit 0288a4b87b65ba54f37fbeeea3cb32238deee92e 715Author: Michel Dänzer <michel.daenzer@amd.com> 716Date: Wed Aug 19 16:54:14 2015 +0900 717 718 DRI2: Keep MSC monotonic when moving window between CRTCs 719 720 This mirrors the DRI3 implementation in xserver. Fixes VDPAU video 721 playback hanging when moving the window between CRTCs. 722 723 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=66384 724 725 (Ported from amdgpu commit 63948ea091a9b324327ade7ec4fc5d67ca7e6f6f) 726 727 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 728 729commit 1d886b526dc49f32dc6744b7a882894bdac4e846 730Author: Michel Dänzer <michel.daenzer@amd.com> 731Date: Wed Aug 19 18:11:45 2015 +0900 732 733 DRI2: Don't ignore rotated CRTCs in amdgpu_dri2_drawable_crtc 734 735 Waiting for vblank interrupts works fine with rotated CRTCs. The only 736 case we can't handle with rotation is page flipping, which is handled 737 in can_exchange(). 738 739 This fixes gnome-shell hanging on rotation, probably because 740 amdgpu_dri2_get_msc returned MSC/UST 0 for rotated CRTCs. 741 742 (Ported from amdgpu commit 7b3212e33cd36fb6f122774df27b56ec4e1a22b8) 743 744 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 745 746commit 270da55340766074cabff8af4258e29fe2f0fc81 747Author: Alex Deucher <alexander.deucher@amd.com> 748Date: Mon Aug 10 15:32:34 2015 -0400 749 750 add new OLAND pci id 751 752 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 753 754commit f123096877f0a0044f3d0315285441f2b5790c29 755Author: Mario Kleiner <mario.kleiner.de@gmail.com> 756Date: Mon Aug 10 23:34:39 2015 +0200 757 758 Make selection between DRI2 and DRI3 consistent with other drivers. (v2) 759 760 Add Option "DRI" to allow selection of maximum DRI level. 761 762 This allows the user to select the maximum level of DRI 763 implementation to use, DRI2 or DRI3. It replaces the old 764 option "DRI3" which had exactly the same purpose, but 765 differs from the method used in both intel ddx and nouveau ddx. 766 Make this consistent before a new stable driver is released. 767 768 v2: Retain handling of old Option "DRI3" for backwards 769 compatibility, but Option "DRI" will take precedence 770 over "DRI3" if both are provided. 771 772 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 773 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 774 775commit 3791fceabf2cb037467dc41c15364e9f9ec1e47e 776Author: Michel Dänzer <michel.daenzer@amd.com> 777Date: Thu Aug 6 16:27:01 2015 +0900 778 779 Wait for scanout BO initialization to finish before setting mode 780 781 This should avoid intermittent artifacts which could sometimes be visible 782 when setting a new scanout pixmap, e.g. on server startup or when 783 changing resolutions. 784 785 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 786 787commit 4e3dfa69e4630df2e0ec0f5b81d61159757c4664 788Author: Michel Dänzer <michel.daenzer@amd.com> 789Date: Thu Aug 6 16:16:38 2015 +0900 790 791 Only call drmmode_copy_fb (at most) once on server startup 792 793 It doesn't make sense to copy the screen contents from console when VT 794 switching back to Xorg or when Xorg resets. 795 796 Fixes intermittent artifacts when VT switching back from console to the 797 gdm login screen. 798 799 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 800 801commit 09c7cdb923965f9a1ea11d2f449bc02114408938 802Author: Michel Dänzer <michel.daenzer@amd.com> 803Date: Thu Jul 30 15:58:54 2015 +0900 804 805 glamor: Move declaration of struct radeon_pixmap out of #if/#else blocks 806 807 Reviewed-by: Dave Airlie <airlied@redhat.com> 808 809commit 936582fde0db461c5c3d78ce4f5f4c93a88a489d 810Author: Michel Dänzer <michel.daenzer@amd.com> 811Date: Thu Jul 30 15:56:36 2015 +0900 812 813 Remove unused local variable pRADEONEnt 814 815 ../../src/radeon_probe.c: In function 'radeon_get_scrninfo': 816 ../../src/radeon_probe.c:157:22: warning: variable 'pRADEONEnt' set but not used [-Wunused-but-set-variable] 817 RADEONEntPtr pRADEONEnt; 818 ^ 819 820 Reviewed-by: Dave Airlie <airlied@redhat.com> 821 822commit 5510cd6027d2387efdf33575e3bfc424cb11bfd8 823Author: Dave Airlie <airlied@redhat.com> 824Date: Mon Jul 27 09:22:57 2015 +1000 825 826 radeon: move radeon_pixmap forward declaration into other block 827 828 There is already a radeon_pixmap forward decl here, the #else 829 block is missing one. 830 831 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 832 Signed-off-by: Dave Airlie <airlied@redhat.com> 833 834commit b32a0a3de84a44b9af4f1ca8be19f10d7fa31b12 835Author: Dave Airlie <airlied@redhat.com> 836Date: Wed Jul 22 13:35:07 2015 +1000 837 838 radeon: cleanup the entity rec 839 840 Some of these were set, some of them were 841 always opposites, so clean things up. 842 843 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 844 Signed-off-by: Dave Airlie <airlied@redhat.com> 845 846commit fcb32231a38f9461d12720cbf72f63502197a711 847Author: Emil Velikov <emil.l.velikov@gmail.com> 848Date: Tue Jul 21 23:55:20 2015 +0100 849 850 Do not link radeon_drv.so against libpciaccess 851 852 It's only used/needed by ati_drv.so 853 854 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 855 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 856 857commit b6d871bf299c7d0f106c07ee4d8bd3b2337f53cc 858Author: Dave Airlie <airlied@gmail.com> 859Date: Thu Jul 9 19:25:36 2015 +1000 860 861 radeon: adopt for new X server dirty tracking APIs. 862 863 Signed-off-by: Dave Airlie <airlied@redhat.com> 864 865commit 95f5d09e3667ded027ae648c97eb4737d8bf67c5 866Author: Michel Dänzer <michel.daenzer@amd.com> 867Date: Thu May 28 11:24:42 2015 +0900 868 869 present: Handle DPMS off in radeon_present_get_ust_msc 870 871 The DRM_IOCTL_WAIT_VBLANK ioctl may return an error during DPMS off, 872 which would trigger an error message in drmmode_crtc_get_ust_msc. 873 874 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 875 876commit 211862b777d0be251a4662f5dd24f2d400544c09 877Author: Michel Dänzer <michel.daenzer@amd.com> 878Date: Fri May 1 18:20:01 2015 +0900 879 880 present: Look at all CRTCs to determine if we can flip 881 882 Inspired by modesetting driver change by Kenneth Graunke. 883 884 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 885 886commit 802d33e474a82262d9cdf11b03568b0c4929cd0d 887Author: Michel Dänzer <michel.daenzer@amd.com> 888Date: Fri May 1 18:54:16 2015 +0900 889 890 present: Fall back to modeset for unflip operation 891 892 It's not always possible to use the page flip ioctl for this, e.g. 893 during DPMS off. We were previously just skipping the unflip in that 894 case, which could result in hangs when setting DPMS off while a 895 fullscreen Present app is running, e.g. at the GNOME3 lock screen. 896 897 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 898 899commit a8ed62010d5012dfb27773595c446b217f3c00c5 900Author: Piotr Redlewski <predlewski@gmail.com> 901Date: Mon Jul 6 19:57:07 2015 +0200 902 903 Do not try to enable already enabled CRTCs in DPMS hook 904 905 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91227 906 907 Signed-off-by: Piotr Redlewski <predlewski@gmail.com> 908 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 909 910commit fc07c370dac4649511eb7322d05a61605a539f39 911Author: Mario Kleiner <mario.kleiner.de@gmail.com> 912Date: Sun Jun 28 02:45:58 2015 +0200 913 914 Fix inconsistent default eg_tile_split in evergreen_accel.c 915 916 eg_tile_split() maps the default: switch-case to tile split 917 flag 6, like for a tile split size of 4096 Bytes. All other 918 instances of tile split mappings or reverse mappings in both 919 the ddx and in Mesa's radeon gallium drivers assign the default: 920 case to tile split flag 4, consistent with a default of 1024 921 Bytes. 922 923 Make this one outlier consistent with other instances of 924 mappings in the code to avoid surprises. 925 926 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 927 928commit ddaba449e8d6fe9fc0d97085e4045843fd8d7af9 929Author: Piotr Redlewski <predlewski@gmail.com> 930Date: Wed Jun 24 18:44:39 2015 +0200 931 932 Disable CRTCs when disabling the outputs 933 934 When turning the outputs off (DPMSModeOff), CRTCs stayed enabled. This led 935 to higher gpu temperatures than with fglrx driver. 936 937 v2: when entering DPMS also disable active CRTCs 938 v3: use drmmode_set_mode_major() for enabling CRTCs when leaving DPMS 939 940 Signed-off-by: Piotr Redlewski <predlewski@gmail.com> 941 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 942 943commit acc11877423ecd81a6e0a7f38466f80e43efee20 944Author: Michel Dänzer <michel.daenzer@amd.com> 945Date: Tue Jun 16 17:34:57 2015 +0900 946 947 Skip disabled CRTCs in radeon_scanout_(do_)update 948 949 The vblank / page flip ioctls don't work as expected for a disabled CRTC. 950 951 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 952 953commit fc9fadaebbc8aead6e030d93a9ccd84561f8f59e 954Author: Mario Kleiner <mario.kleiner.de@gmail.com> 955Date: Mon Jun 22 00:09:17 2015 +0200 956 957 Don't set TILE_SPLIT flags if surface.tile_split == 0. 958 959 On pre-Evergreen hw, libdrm's r6_surface_best() helper 960 for the surface managers radeon_surface_best() routine 961 is a no-op and therefore doesn't assign any tile_split 962 settings to created surfaces, so it leaves 963 surface.tile_split on its "undefined" value of 0. 964 965 Mesa's DRI3/Present backend creates DRI3 Pixmaps via 966 the DRIImage extension and the radeon gallium driver 967 implementation of that extension uses the libdrm 968 surface manager for backing bo creation and treats 969 an undefined surface.tile_split==0, as returned by 970 the surface manager for pre-evergreen, as a signal 971 to not assign any tile_split flags to the DRI3 Pixmaps 972 bo. 973 974 The ddx also uses libdrm surface manager to create the 975 x-screen pixmap, but so far treated the returned undefined 976 surface.tile_split==0 by mapping it to eg_tile_split()'s 977 default tile_split flags, which are different from Mesa's 978 tiling flags for DRI3 pixmaps. Under DRI3/Present this 979 causes a mismatch of src pixmap and destination root 980 pixmaps tiling flags and thereby prevents page flipping 981 for pixmap presents. 982 983 Change the ddx code to treat surface.tile_split==0 the 984 same way as the radeon gallium driver to avoid mismatched 985 tiling flags and thereby allow DRI3/Present page-flip to 986 work on pre-Evergreen hw. 987 988 Tested on RV730 and Evergreen "Juniper". 989 990 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 991 992commit 49f5b0bc301414df049e00d226034e3d6e56421b 993Author: Michel Dänzer <michel.daenzer@amd.com> 994Date: Fri May 1 19:02:36 2015 +0900 995 996 Don't attempt a DRI2/Present page flip while the other one is flipping 997 998 Fixes corrupted display and hangs when switching between DRI2 and DRI3 999 fullscreen apps, e.g. a compositor using DRI3 and a fullscreen app using 1000 DRI2 or vice versa. 1001 1002 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1003 1004commit afab7839fc15722dbaa7203d00fe7f6ce5336b9d 1005Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1006Date: Wed Jun 10 20:55:08 2015 +0200 1007 1008 Allow/Fix use of multiple ZaphodHead outputs per x-screen. (v2) 1009 1010 Defining multiple ZaphodHead outputs per x-screen in a 1011 multiple x-screen's per gpu configuration caused all 1012 outputs except one per x-screen to go dark, because 1013 there was a fixed mapping x-screen number -> crtc number, 1014 limiting the number of crtc's per x-screen to one. 1015 1016 On a ZaphodHead's setup, be more clever and assign 1017 as many crtc's to a given x-screen as there are 1018 ZaphodHeads defined for that screen, assuming 1019 there are enough unused crtc's available. 1020 1021 Tested on a triple display setup with different combos 1022 of one, two or three ZaphodHeads per one, two or three 1023 x-screens. 1024 1025 This is a port of similar code from xf86-video-nouveau. 1026 1027 v2: Implement suggestions by Michel Dänzer: Less verbose 1028 debug output, more clear warning message on crtc allocation 1029 failure. Move clearing of per gpu assigned_crtc mask to 1030 CloseScreen, indeed testing shows no need for the more 1031 complex new server generation check from v1. 1032 1033 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1034 (v1) Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1035 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 1036 1037commit d7c82731a8bf3d381bc571b94d80d9bb2dd6e40d 1038Author: Michel Dänzer <michel.daenzer@amd.com> 1039Date: Thu May 21 16:01:16 2015 +0900 1040 1041 DRI2: Fix handling of drmmode_crtc_get_ust_msc return code 1042 1043 Fixes regression introduced in commit 1044 76c2923ac5c7230a8b2f9f8329c308d28b44d9c0 ("DRI2: Split out helper for 1045 getting UST and MSC of a specific CRTC"). 1046 1047 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1048 1049commit e58fc380ccf2a581d28f041fd74b963626ca5404 1050Author: Michel Dänzer <michel.daenzer@amd.com> 1051Date: Thu May 28 10:59:22 2015 +0900 1052 1053 glamor: Only wait for GPU writes in radeon_glamor_prepare_access_cpu_ro 1054 1055 We don't need to wait for GPU reads to finish before reading with the 1056 CPU. 1057 1058 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1059 1060commit 5f82a720374c9c1caebb42bfbeea1f0cf8847d28 1061Author: Michel Dänzer <michel.daenzer@amd.com> 1062Date: Thu May 21 15:46:32 2015 +0900 1063 1064 present: Remove get_drmmode_crtc helper 1065 1066 It was getting the drmmode_crtc_private_ptr in a roundabout way. 1067 1068 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1069 1070commit d64a13ebe0ecd241ee3260dbffd8f4a01e254183 1071Author: Michel Dänzer <michel.daenzer@amd.com> 1072Date: Wed May 20 09:58:38 2015 +0900 1073 1074 Replace SyncFence typedef with declaration of struct _SyncFence 1075 1076 Fixes build failure in some environments: 1077 1078 CC radeon_sync.lo 1079 In file included from radeon_sync.c:28: 1080 /home/tinderbox/xorg-build/include/xorg/misync.h:31: error: redefinition of typedef 'SyncFence' 1081 radeon.h:93: note: previous declaration of 'SyncFence' was here 1082 make[2]: *** [radeon_sync.lo] Error 1 1083 1084 Reported-Tested-and-Acked-by: David Airlie <airlied@redhat.com> 1085 1086commit 818c180c8932233b214a35ba0647af82f7bcec3d 1087Author: Michel Dänzer <michel.daenzer@amd.com> 1088Date: Tue May 12 17:43:57 2015 +0900 1089 1090 glamor: Deal with glamor_glyphs_init being removed from xserver 1091 1092 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1093 1094commit 7c4b78ab10b82c6dba9f72034ff7583859cca63d 1095Author: Alex Deucher <alexander.deucher@amd.com> 1096Date: Tue May 12 13:21:24 2015 -0400 1097 1098 add new bonaire pci id 1099 1100 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 1101 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1102 1103commit 80eede245d1eda27eaba108b0761a24bfd69aff6 1104Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1105Date: Sat May 9 07:24:36 2015 +0200 1106 1107 present: Fix present notify timestamps and counts. 1108 1109 Invalid (msc,ust)=(0,0) were returned to the server 1110 because a wrong crtc_id was used to specify which 1111 crtc should deliver pageflip completion data. Fix 1112 it in accordance with the dri2 implementation. 1113 1114 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1115 1116commit e3be8b0a8cf484ff16597413a6172788178e80c8 1117Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1118Date: Sat May 9 07:24:35 2015 +0200 1119 1120 present: Move check for async flips 1121 1122 Check for and reject currently unsupported async_flip 1123 inside radeon_present_check_flip() instead of inside 1124 radeon_present_flip(). 1125 1126 This way the server can detect early that async flips 1127 aren't supported by the ddx and can deal with this 1128 correctly by using its non-vsync'ed CopyRegion fallback. 1129 1130 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1131 1132commit 1584dc545c78e0bce8d4b4b9f26b568e2c211453 1133Author: Michel Dänzer <michel.daenzer@amd.com> 1134Date: Tue Apr 28 17:14:11 2015 +0900 1135 1136 Skip disabled CRTCs in drmmode_set_desired_modes() even if set_hw == FALSE 1137 1138 Not skipping a disabled CRTC results in a crash. 1139 1140 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90187 1141 1142 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1143 1144commit 340baf3804b7d9de8082badfc715a9bee7856e8f 1145Author: Michel Dänzer <michel.daenzer@amd.com> 1146Date: Mon Apr 27 15:07:06 2015 +0900 1147 1148 Don't call radeon_get_pixmap_private() when using EXA 1149 1150 It only works when using glamor. 1151 1152 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90169 1153 1154 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1155 1156commit 80f3d727f93cb6efedd2b39338d2301035965fe2 1157Author: Michel Dänzer <michel.daenzer@amd.com> 1158Date: Wed Apr 22 13:33:15 2015 +0900 1159 1160 On screen resize, clear the new buffer before displaying it 1161 1162 Fixes garbage being intermittently visible during a screen resize. 1163 1164 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=27757#c7 1165 1166 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1167 1168commit 3999bf88cdb192fe2f30b03bd2ed6f6a3f9f9057 1169Author: Michel Dänzer <michel.daenzer@amd.com> 1170Date: Thu Apr 2 18:29:38 2015 +0900 1171 1172 Make drmmode_copy_fb() work with glamor as well 1173 1174 Needed for Xorg -background none. 1175 1176 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1177 1178commit a4a8cdbcc10c1c5f07485a2af9e9e81e490c3e1d 1179Author: Michel Dänzer <michel.daenzer@amd.com> 1180Date: Thu Apr 2 17:46:34 2015 +0900 1181 1182 Update scanout pixmap contents before setting a mode with it 1183 1184 This ensures the scanout pixmaps used for Option "TearFree" and Option 1185 "ShadowPrimary" have been initialized when their initial mode is set. 1186 1187 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1188 1189commit 673e1c7637687c74fc9bdeeeffb7ace0d04b734f 1190Author: Michel Dänzer <michel.daenzer@amd.com> 1191Date: Thu Apr 2 17:54:33 2015 +0900 1192 1193 Defer initial modeset until the first BlockHandler invocation 1194 1195 This ensures that the screen pixmap contents have been initialized when 1196 the initial modes are set. 1197 1198 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=27757 1199 1200 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1201 1202commit 37874a4eeace5df04b02c8fc28f67b824e3f0f5f 1203Author: Michel Dänzer <michel.daenzer@amd.com> 1204Date: Tue Apr 21 17:19:15 2015 +0900 1205 1206 Defer initial drmmode_copy_fb call until root window creation 1207 1208 That's late enough for acceleration to be fully initialized, but still 1209 early enough to set pScreen->canDoBGNoneRoot. 1210 1211 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1212 1213commit 39c497f3efca5ca08343b884f44c93215dcdef31 1214Author: Michel Dänzer <michel.daenzer@amd.com> 1215Date: Thu Apr 2 18:10:42 2015 +0900 1216 1217 Only copy fbcon BO contents if bgNoneRoot is TRUE 1218 1219 Otherwise, the X server will initialize the screen pixmap contents 1220 anyway. 1221 1222 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1223 1224commit 1af044d7eee211fd4b248c236280274a68334da5 1225Author: Michel Dänzer <michel.daenzer@amd.com> 1226Date: Wed Apr 22 18:43:02 2015 +0900 1227 1228 DRI2: Clear old->devPrivate.ptr in fixup_glamor 1229 1230 It doesn't point to the memory of the newly allocated BO. Fixes crash 1231 running piglit with Option "ShadowPrimary" enabled. 1232 1233 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1234 1235commit 43159ef400c3b18b9f4d3e6fa1c4aef2d60d38fe 1236Author: Michel Dänzer <michel.daenzer@amd.com> 1237Date: Mon Apr 20 18:44:36 2015 +0900 1238 1239 Add Option "TearFree" v4 1240 1241 Avoids tearing by flipping between two scanout BOs per (non-rotated) CRTC 1242 1243 v2: 1244 * Fix condition for TearFree log message (Richard Wilbur) 1245 * Log warning message about DRI page flipping being enabled because of 1246 TearFree (or ShadowPrimary) also when building without glamor support 1247 1248 v3: 1249 * Only override fb_id/x/y if all scanout pixmaps have been successfully 1250 allocated 1251 1252 v4: 1253 * Make log warning clearer if drmModePageFlip returns an error 1254 1255 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v1) 1256 1257commit ed401f5b4f07375db17ff05e294907ec95fc946d 1258Author: Michel Dänzer <michel.daenzer@amd.com> 1259Date: Fri Apr 3 12:11:00 2015 +0900 1260 1261 glamor: Remove the stride member of struct radeon_pixmap 1262 1263 Its value was always the same as that of the PixmapRec devKind member. 1264 1265 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1266 1267commit ae92d1765fa370a8d94c2856ad6c45d273ec3c69 1268Author: Michel Dänzer <michel.daenzer@amd.com> 1269Date: Wed Mar 18 16:23:24 2015 +0900 1270 1271 glamor: Add Option "ShadowPrimary" v2 1272 1273 When this option is enabled, most pixmaps (including the screen pixmap) 1274 are allocated in system RAM and mostly accessed by the CPU. Changed areas 1275 of the screen pixmap are copied to dedicated per-CRTC scanout pixmaps 1276 regularly, triggered by the vblank interrupt. 1277 1278 v2: 1279 * Set region data pointer to NULL for keeping only the extents 1280 * Move pRegion and pBox local variable declarations closer to their uses 1281 in drmmode_set_mode_major() 1282 1283commit eea79472a84672ee4dc7adc4487cec6a4037048a 1284Author: Michel Dänzer <michel.daenzer@amd.com> 1285Date: Wed Apr 1 15:51:52 2015 +0900 1286 1287 glamor: Add wrappers for the X server rendering hooks 1288 1289 They can choose between using the GPU or CPU for the operation. 1290 1291commit 051d46382656ffc3e6cac1aab3aee7efdf5b623a 1292Author: Michel Dänzer <michel.daenzer@amd.com> 1293Date: Thu Mar 19 17:34:27 2015 +0900 1294 1295 glamor: Add radeon_pixmap parameter to radeon_glamor_create_textured_pixmap 1296 1297commit 2fa021f77372ca93375a3d13a0c43a9089674899 1298Author: Michel Dänzer <michel.daenzer@amd.com> 1299Date: Thu Mar 19 17:38:47 2015 +0900 1300 1301 glamor: Remove unused function radeon_glamor_pixmap_is_offscreen 1302 1303commit e96349ba6281fd18b8bf9c76629128276b065e6c 1304Author: Michel Dänzer <michel.daenzer@amd.com> 1305Date: Wed Apr 1 16:04:13 2015 +0900 1306 1307 Add RADEON_CREATE_PIXMAP_SCANOUT flag 1308 1309 It means that the pixmap is used for scanout exclusively. 1310 1311commit 9be7dd382e86d2b804de81d4e2af7431b2e16843 1312Author: Michel Dänzer <michel.daenzer@amd.com> 1313Date: Wed Apr 1 12:40:16 2015 +0900 1314 1315 Split out struct drmmode_scanout for rotation shadow buffer information 1316 1317 Will be used for other kinds of dedicated scanout buffers as well. 1318 1319commit c32b0530302739f6512755bccf281c2300617376 1320Author: Michel Dänzer <michel.daenzer@amd.com> 1321Date: Thu Mar 19 17:46:48 2015 +0900 1322 1323 Rename scanout_pixmap_x field to prime_pixmap_x 1324 1325 To avoid confusion with upcoming changes. 1326 1327commit 5921ba4ca705a0d919515626088f3948cc4848c1 1328Author: Michel Dänzer <michel.daenzer@amd.com> 1329Date: Tue Mar 31 15:14:52 2015 +0900 1330 1331 present: Don't flip between BOs with different tiling parameters 1332 1333 The kernel driver doesn't handle that correctly yet. 1334 1335 Fixes or at least avoids issues with OpenGL fullscreen apps with DRI3 1336 enabled and using PRIME or with (2D) tiling disabled. 1337 1338 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89720 1339 1340 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1341 1342commit 428e416e7cb04a1e0527da39cfebf70218879a77 1343Author: Michel Dänzer <michel.daenzer@amd.com> 1344Date: Thu Apr 2 10:34:03 2015 +0900 1345 1346 Add radeon_get_pixmap_tiling_flags helper 1347 1348 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1349 1350commit 98fb4199e63fedd4607cddee64bf602d6398df81 1351Author: Michel Dänzer <michel.daenzer@amd.com> 1352Date: Tue Mar 31 12:25:18 2015 +0900 1353 1354 Only enable SYNC extension fences and the Present extension along with DRI3 1355 1356 This avoids some trouble with the Gallium nine state tracker, which uses 1357 the Present extension even when DRI3 is disabled. 1358 1359 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89772 1360 1361 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1362 1363commit f8b0f23e9f4af9f9097ee5e72d53b45173163c41 1364Author: Michel Dänzer <michel.daenzer@amd.com> 1365Date: Fri Mar 27 12:34:55 2015 +0900 1366 1367 DRI2: Use radeon_get_pixmap_handle 1368 1369 Now we can share pixmaps with no struct radeon_bo via DRI2. 1370 1371 Fixes VDPAU video playback freezing when using an OpenGL compositor with 1372 DRI3 enabled. 1373 1374 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89755 1375 1376 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1377 Tested-by: Nick Sarnie <commendsarnex@gmail.com> 1378 1379commit ccbda955ebae1d457d35293833f12791e0f9fb0b 1380Author: Michel Dänzer <michel.daenzer@amd.com> 1381Date: Fri Mar 27 12:16:44 2015 +0900 1382 1383 Move get_pixmap_handle helper to radeon_bo_helper.c 1384 1385 No functional change. 1386 1387 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1388 Tested-by: Nick Sarnie <commendsarnex@gmail.com> 1389 1390commit de5ddd09db82141b263338dcf0c28e01f58268ee 1391Author: Michel Dänzer <michel.daenzer@amd.com> 1392Date: Thu Mar 26 16:33:02 2015 +0900 1393 1394 Move radeon_drm_handler/abort_proc fields to drmmode_flipdata_rec 1395 1396 Their values are the same for all DRM flip ioctl calls within a single 1397 radeon_do_pageflip() call. 1398 1399 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1400 1401commit e8c0f6319fbf4c3ea11e22ab1a68837031bdec8c 1402Author: Michel Dänzer <michel.daenzer@amd.com> 1403Date: Thu Mar 26 16:27:35 2015 +0900 1404 1405 Simplify radeon_do_pageflip() error handling slightly more 1406 1407 We don't need the local variable old_fb_id. 1408 1409 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1410 1411commit 8fc22360d5520469c82092ccb0fcf2af330c573f 1412Author: Michel Dänzer <michel.daenzer@amd.com> 1413Date: Thu Mar 26 15:58:01 2015 +0900 1414 1415 Increase robustness against DRM page flip ioctl failures v3 1416 1417 Centralize cleanup, only clean up things that have been allocated for 1418 the failed ioctl call. 1419 1420 Fixes double-free after a flip ioctl failure. 1421 1422 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89681 1423 1424 v2: Only call drmModeRmFB for flipdata->old_fb_id on receipt of last DRM 1425 page flip event. Fixes Black screen on making glxgears fullscreen with 1426 DRI3 enabled. 1427 v3: Avoid double-free of flipdata in the unlikely case that calloc fails 1428 for flipcarrier, but only for the second or later CRTC. 1429 1430 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v2) 1431 1432commit b16609b453bb1a181198cf27778f205dc23fb642 1433Author: Michel Dänzer <michel.daenzer@amd.com> 1434Date: Thu Mar 26 17:15:21 2015 +0900 1435 1436 glamor: Handle GLAMOR_* flags removed from xserver 1437 1438 The behaviour is the same as when the removed flags were passed in. 1439 1440 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1441 1442commit 391900a670addec39515f924265bfa9f8bfa9ec0 1443Author: Michel Dänzer <michel.daenzer@amd.com> 1444Date: Wed Mar 18 12:56:07 2015 +0900 1445 1446 glamor: Avoid generating GEM flink names for BOs shared via DRI3 v2 1447 1448 We can't create our own struct radeon_bo representation in this case 1449 because destroying that would make the GEM handle inaccessible to glamor 1450 as well. So just get the handle directly via dma-buf. 1451 1452 v2: Close dma-buf file descriptor, pointed out by Axel Davy. 1453 1454 Reviewed-by: Axel Davy <axel.davy@ens.fr> 1455 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1456 1457commit dfad91fffb5bd013785223b42d78886df839eacf 1458Author: Michel Dänzer <michel.daenzer@amd.com> 1459Date: Thu Mar 19 12:37:01 2015 +0900 1460 1461 Present: Add radeon_present_get_pixmap_handle helper 1462 1463 Reviewed-by: Axel Davy <axel.davy@ens.fr> 1464 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1465 1466commit 7b4fc4a677d252d01c2bf80d162bc35814059eaa 1467Author: Michel Dänzer <michel.daenzer@amd.com> 1468Date: Thu Mar 19 12:15:52 2015 +0900 1469 1470 Make radeon_do_pageflip take a BO handle directly 1471 1472 Reviewed-by: Axel Davy <axel.davy@ens.fr> 1473 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1474 1475commit af6076241c0d322b295a4e898407ae2472bd8eb4 1476Author: Michel Dänzer <michel.daenzer@amd.com> 1477Date: Tue Mar 24 13:37:01 2015 +0900 1478 1479 Adapt radeon_sync.c for misyncshm.h vs misync.h as well 1480 1481 This should have been part of the previous commit. Without this, SYNC 1482 extension fences (and by extension DRI3) were disabled because it checked 1483 for HAVE_MISYNC_H, but configure now only defines HAVE_MISYNCSHM_H. 1484 1485commit aa7825eb29cdf6ac9d7b28ad18186807ff384687 1486Author: Michel Dänzer <michel.daenzer@amd.com> 1487Date: Mon Mar 23 10:12:17 2015 +0900 1488 1489 configure.ac: Check for misyncshm.h again 1490 1491 We need this header, which was added after the other misync*.h headers, 1492 along with DRI3 support. 1493 1494commit 6291baaed261e36a63dc001307427fe00ba82259 1495Author: Michel Dänzer <michel.daenzer@amd.com> 1496Date: Thu Mar 19 00:12:06 2015 +0900 1497 1498 EXA: Return NULL from radeon_get_pixmap_bo if there is no driver private 1499 1500 This mirrors what the glamor part of the function does. Fixes a crash 1501 running glxgears_pixmap with DRI3 enabled, reported by "marvin24" on IRC. 1502 1503commit f68d9b5ba0c91a725b5eec9386c61bea8824c299 1504Author: Michel Dänzer <michel.daenzer@amd.com> 1505Date: Wed Mar 18 18:29:27 2015 +0900 1506 1507 dri3: Use screen->CreatePixmap instead of fbCreatePixmap directly 1508 1509 Fixes crash with EXA reported by "marvin24" on IRC. 1510 1511commit 4b0997e56dec0053cb2cb793e0f4ae35055ff7e6 1512Author: Michel Dänzer <michel.daenzer@amd.com> 1513Date: Wed Mar 18 12:30:09 2015 +0900 1514 1515 glamor: Add glamor_fd_from_pixmap define for standalone glamor tree 1516 1517commit af1862a37570fa512a525ab47d72b30400d2e2d6 1518Author: Michel Dänzer <michel.daenzer@amd.com> 1519Date: Wed Mar 18 11:05:40 2015 +0900 1520 1521 Always include misync.h before other misync headers 1522 1523 Older versions of xserver didn't include misync.h from other misync 1524 headers as needed. 1525 1526commit fcd37f65f485291084c174666bd605e215bf1398 1527Author: Michel Dänzer <michel.daenzer@amd.com> 1528Date: Wed Mar 18 10:39:03 2015 +0900 1529 1530 DRI3: Use open hook instead of open_client 1531 1532 We don't need the client pointer, and this allows it to work with older 1533 versions of xserver as well. 1534 1535commit f940fd741b15f03393037c5bb904cd74f012de9d 1536Author: Michel Dänzer <michel.daenzer@amd.com> 1537Date: Wed Mar 18 10:06:47 2015 +0900 1538 1539 Initialize boolean variable before calling xf86GetOptValBool 1540 1541 We were just lucky that it contained 0 in most cases. 1542 1543commit 4a98f60117c387a228d5cbaadb6e298fb4e865df 1544Author: Michel Dänzer <michel.daenzer@amd.com> 1545Date: Tue Mar 17 16:47:11 2015 +0900 1546 1547 Add xorg_list_for_each_entry_safe fallback in radeon_list.h 1548 1549 Fixes another build failure against older versions of xserver reported 1550 by "Pali" on IRC. 1551 1552commit 694e04720b886060fe3eefdce59741f218c8269f 1553Author: Michel Dänzer <michel.daenzer@amd.com> 1554Date: Tue Mar 17 16:32:09 2015 +0900 1555 1556 Simplify includes in radeon_{dri3,present}.c 1557 1558 Just include radeon.h first. Fixes build failures against older versions 1559 of xserver because compat-api.h wasn't getting picked up. Reported by 1560 "Pali" on IRC. 1561 1562commit 64e1e4dbdd3caee6f5d8f6b6c094b4533fa94953 1563Author: Michel Dänzer <michel.daenzer@amd.com> 1564Date: Tue Mar 10 16:53:11 2015 +0900 1565 1566 Add DRI3 support v2 1567 1568 Must be enabled with 1569 1570 Option "DRI3" 1571 1572 in xorg.conf. 1573 1574 v2: Adapt to v2 of patches 11/12. 1575 1576 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1577 1578commit 69ae0194778fe4276895839db92383f63f0b5de4 1579Author: David Heidelberger <david.heidelberger@ixit.cz> 1580Date: Fri Mar 6 17:57:22 2015 +0900 1581 1582 Handle tiling in radeon_set_shared_pixmap_backing 1583 1584 [ Michel Dänzer: Fixups for glamor ] 1585 Signed-off-by: David Heidelberger <david.heidelberger@ixit.cz> 1586 Signed-off-by: Axel Davy <axel.davy@ens.fr> 1587 1588 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1589 1590commit 3c65fb849e1ba9fb6454bcaa55b696548902f3fc 1591Author: Michel Dänzer <michel.daenzer@amd.com> 1592Date: Tue Mar 10 16:52:54 2015 +0900 1593 1594 Add support for the Present extension v2 1595 1596 v2: Fix up for struct radeon_drm_queue -> radeon_drm_queue_entry. 1597 Swapped order of patches 11 & 12 because the Present extension uses 1598 SYNC fences. 1599 1600 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1601 1602commit 8fc9a241ab59ffbcdc178d6415332c88a54e85fe 1603Author: Michel Dänzer <michel.daenzer@amd.com> 1604Date: Thu Mar 5 18:18:56 2015 +0900 1605 1606 Add support for SYNC extension fences v2 1607 1608 v2: Swapped order of patches 11 & 12 because the Present extension uses 1609 SYNC fences. 1610 1611 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1612 1613commit 4a35e2f33d9cdfb608423046391311109f96fb6b 1614Author: Michel Dänzer <michel.daenzer@amd.com> 1615Date: Thu Mar 5 18:34:07 2015 +0900 1616 1617 Fold radeon_glamor_flush into radeon_cs_flush_indirect 1618 1619 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1620 1621commit 4b8adebb80158bcf81ada83bb88517febe931b12 1622Author: Michel Dänzer <michel.daenzer@amd.com> 1623Date: Thu Mar 5 18:40:23 2015 +0900 1624 1625 Move #include "radeon_glamor.h" from radeon.h to where it's needed 1626 1627 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1628 1629commit 76c2923ac5c7230a8b2f9f8329c308d28b44d9c0 1630Author: Michel Dänzer <michel.daenzer@amd.com> 1631Date: Wed Mar 4 10:30:19 2015 +0900 1632 1633 DRI2: Split out helper for getting UST and MSC of a specific CRTC 1634 1635 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1636 1637commit 6c3a721cde9317233072b573f9502348dcd21b16 1638Author: Michel Dänzer <michel.daenzer@amd.com> 1639Date: Tue Mar 10 16:52:18 2015 +0900 1640 1641 DRI2: Use helper functions for DRM event queue management v3 1642 1643 This is mostly in preparation for Present support, but it also simplifies 1644 the DRI2 specific code a little. 1645 1646 v2: Fix up for struct radeon_drm_queue -> radeon_drm_queue_entry. 1647 v3: Removed excess 0s from conversion from microseconds to seconds, 1648 thanks to Richard Wilbur <richard.wilbur@gmail.com> for the catch! 1649 1650 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1651 1652commit c3fa22a479e61d1899fa9d327d9c4e2a7f64b0c1 1653Author: Michel Dänzer <michel.daenzer@amd.com> 1654Date: Wed Mar 11 17:47:59 2015 +0900 1655 1656 DRI2: Move radeon_dri2_flip_event_handler 1657 1658 In preparation for the next change, which will modify it to a static 1659 function which needs to be in the new place. No functional change. 1660 1661 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1662 1663commit 65045112fdc8a9fa36e0e00f46739a6152b775ff 1664Author: Michel Dänzer <michel.daenzer@amd.com> 1665Date: Wed Mar 11 17:33:54 2015 +0900 1666 1667 DRI2: Remove superfluous assignments to *_info->frame 1668 1669 That field is only used for page flipping. 1670 1671 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1672 1673commit ad27f16f308079d06a2b1c788b3cb0947531253a 1674Author: Michel Dänzer <michel.daenzer@amd.com> 1675Date: Wed Mar 11 17:30:11 2015 +0900 1676 1677 DRI2: Simplify blit fallback handling for scheduled swaps 1678 1679 Also use radeon_dri2_schedule_event when possible. 1680 1681 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1682 1683commit b4af8a327ed8420f0ff4ea0f113f4a59406ed4d3 1684Author: Michel Dänzer <michel.daenzer@amd.com> 1685Date: Mon Mar 2 18:59:54 2015 +0900 1686 1687 Add DRM event queue helpers v2 1688 1689 v2: Rename struct radeon_drm_queue to struct radeon_drm_queue_event, 1690 thanks to Richard Wilbur <richard.wilbur@gmail.com> for the suggestion. 1691 Also changed the corresponding parameter and local variable names from 1692 'q' to 'e'. 1693 1694 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1695 1696commit 7c3470f4b659206ed23f761948936ede3a2dba3d 1697Author: Michel Dänzer <michel.daenzer@amd.com> 1698Date: Fri Mar 6 18:51:29 2015 +0900 1699 1700 Move xorg_list backwards compatibility to new radeon_list.h header 1701 1702 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1703 1704commit 7388d0b6c54b9d536fdb161e3aa61b326627b939 1705Author: Michel Dänzer <michel.daenzer@amd.com> 1706Date: Thu Mar 5 17:35:06 2015 +0900 1707 1708 Require at least xserver 1.8 1709 1710 So we can rely on the list.h header. 1711 1712 xserver 1.8 was released in April 2010. 1713 1714 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1715 1716commit b8ec9ed4fe86952763b963c86f0af0dcae69aa6c 1717Author: Jerome Glisse <jglisse@redhat.com> 1718Date: Thu Feb 19 14:47:41 2015 -0500 1719 1720 Proper leak fix, previous leak fix was bogus. 1721 1722 Signed-off-by: Jérôme Glisse <jglisse@redhat.com> 1723 1724commit 63dc36dc49f93cb00111b497ab6805194bc9d240 1725Author: Jerome Glisse <jglisse@redhat.com> 1726Date: Thu Feb 19 11:43:08 2015 -0500 1727 1728 Avoid leaking memory on output. 1729 1730 Signed-off-by: Jérôme Glisse <jglisse@redhat.com> 1731 1732commit c80ea1e3e8c5f155cedadbe4588870fb1f410ab4 1733Author: Rolf Eike Beer <eb@emlix.com> 1734Date: Tue Jan 13 12:11:26 2015 +0100 1735 1736 radeon: remove unneeded inclusion of sarea.h 1737 1738 None of the structs or defines from that header is used. 1739 1740 Signed-off-by: Rolf Eike Beer <eb@emlix.com> 1741 1742commit 04da199231bb3f11cf17f94574a8df05855a7b82 1743Author: Adam Jackson <ajax@redhat.com> 1744Date: Wed Dec 17 14:03:58 2014 -0500 1745 1746 Remove dead accelDFS flag 1747 1748 Signed-off-by: Adam Jackson <ajax@redhat.com> 1749 1750commit c9f8f642fd495937400618a4fc25ecae3f8888fc 1751Author: Michel Dänzer <michel.daenzer@amd.com> 1752Date: Mon Nov 17 12:32:29 2014 +0900 1753 1754 Prefer drmModeSetCursor2 over drmModeSetCursor 1755 1756 The former includes information about the position of the hotspot within 1757 the cursor image. 1758 1759 Copied from xf86-video-modesetting. 1760 1761 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1762 1763commit 2f11dcd43966cf2ee26e61960fd72e6644f5e037 1764Author: Dave Airlie <airlied@redhat.com> 1765Date: Mon Nov 10 13:49:29 2014 +1000 1766 1767 radeon: add support for DP 1.2 display hotplug (v2) 1768 1769 This allows for dynamic creation of conneectors when the 1770 kernel tells us. 1771 1772 v2: fix dpms off crash 1773 1774 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1775 Signed-off-by: Dave Airlie <airlied@redhat.com> 1776 1777commit c88424d1f4aaa78b569e5d44f0b4a47de2f422f4 1778Author: Dave Airlie <airlied@redhat.com> 1779Date: Mon Nov 10 14:17:54 2014 +1000 1780 1781 radeon: move output name creation to its own function 1782 1783 The secondary indent is deliberate to make the next patch more 1784 parseable for mst support. 1785 1786 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1787 Signed-off-by: Dave Airlie <airlied@redhat.com> 1788 1789commit 32b003cb7657e07d5af6338ad44d768eda87fd33 1790Author: Dave Airlie <airlied@redhat.com> 1791Date: Mon Nov 10 14:12:34 2014 +1000 1792 1793 radeon: stop caching mode resources 1794 1795 This is step one towards MST connector hotplug support, 1796 it stop caching the mode resources structure, and 1797 just passes a pointer to it around. 1798 1799 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1800 Signed-off-by: Dave Airlie <airlied@redhat.com> 1801 1802commit 7c7b38e0b375b6e8853ad2d1092302ea83f6f570 1803Author: David Heidelberger <david.heidelberger@ixit.cz> 1804Date: Sun Oct 12 16:34:21 2014 +0200 1805 1806 radeon/vdpau: don't report VDPAU for < r300 1807 1808 Signed-off-by: David Heidelberger <david.heidelberger@ixit.cz> 1809 1810commit c74de9fec13fac2c836bb2a07ae6f90e1d61e667 1811Author: Michel Dänzer <michel.daenzer@amd.com> 1812Date: Wed Aug 6 11:08:00 2014 +0900 1813 1814 PRIME: Don't advertise offload capabilities when acceleration is disabled 1815 1816 Xorg tends to crash if the user tries to actually use the offload 1817 capabilities with acceleration disabled. 1818 1819 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57200 1820 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1821 1822commit 636a6382f80412ae4e28abe0cd9f98f84cd52291 1823Author: Michel Dänzer <michel.daenzer@amd.com> 1824Date: Wed Oct 15 12:02:17 2014 +0900 1825 1826 Remove duplicate OPTION_PAGE_FLIP entry 1827 1828 Reported by 'amanual' on IRC 1829 1830commit c854b4479ec5122390ede81b655b259efd9b1800 1831Author: Alex Deucher <alexdeucher@gmail.com> 1832Date: Thu Oct 2 10:22:22 2014 -0400 1833 1834 radeon: bump version post release 1835 1836 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1837 1838commit 068a59e010ce6bfcd54f5a18cc08c55c54b8618d 1839Author: Alex Deucher <alexander.deucher@amd.com> 1840Date: Wed Oct 1 23:38:52 2014 -0400 1841 1842 radeon: bump version for release 1843 1844 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1845 1846commit 5fee565e420d3efbf72fdf4f63c3d5d93d8ceddb 1847Author: Alex Deucher <alexander.deucher@amd.com> 1848Date: Wed Oct 1 23:36:31 2014 -0400 1849 1850 radeon: require libdrm_radeon 2.4.58 for latest SI/CI pci ids 1851 1852 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1853 1854commit 56c7fb8ada4cd9cf096c6b06a8d368d286d74e68 1855Author: Thomas DeRensis <tderensis@gmail.com> 1856Date: Wed Sep 10 21:20:32 2014 -0400 1857 1858 radeon: fix build warnings regarding const qualifier 1859 1860 Signed-off-by: Thomas DeRensis <tderensis@gmail.com> 1861 1862commit b9a9b244336ed50d7df4f36135f875a3bb6ca948 1863Author: Thomas DeRensis <tderensis@gmail.com> 1864Date: Thu Sep 4 21:45:19 2014 -0400 1865 1866 radeon: fix a leak in radeon_vbo_get_bo() 1867 1868 The dma_bo struct was never freed in the error path. Reported by cppcheck. 1869 1870 Signed-off-by: Thomas DeRensis <tderensis@gmail.com> 1871 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 1872 1873commit ba5fcc4d6b6eae3f2c55c6229a53d24d160789c1 1874Author: Alex Deucher <alexander.deucher@amd.com> 1875Date: Thu Aug 21 11:32:45 2014 -0400 1876 1877 radeon/kms: fix mullins pci id 1878 1879 strange errant change from: 1880 39fef269f521c92a2a31c80447e9401bacb3797a 1881 1882 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1883 1884commit efef34930650d6d80f7b527f4cee76d9e5954ace 1885Author: Alex Deucher <alexander.deucher@amd.com> 1886Date: Thu Aug 21 11:30:17 2014 -0400 1887 1888 radeon/kms: add new SI pci ids 1889 1890 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1891 1892commit 39fef269f521c92a2a31c80447e9401bacb3797a 1893Author: Alex Deucher <alexander.deucher@amd.com> 1894Date: Thu Aug 21 11:27:49 2014 -0400 1895 1896 radeon/kms: add new CIK pci ids 1897 1898 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1899 1900commit 7dab6b2fe9cf690b0dae1c127d0374322bd8278e 1901Author: Michel Dänzer <michel.daenzer@amd.com> 1902Date: Wed Aug 20 17:51:35 2014 +0900 1903 1904 Revert "glamor: Set environment variable RADEON_THREAD=0" 1905 1906 This reverts commit 4b5060f357a3cb248c9359c92c1e9c42ef6434c8. 1907 1908 Further testing shows that disabling the thread is only a moderate win in 1909 some cases, but a much bigger loss in some other cases. 1910 1911 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1912 1913commit 1ff383360886c5cee1bb75abcc20675aca0336a6 1914Author: Alex Deucher <alexander.deucher@amd.com> 1915Date: Tue Aug 12 12:27:12 2014 -0400 1916 1917 radeon: fix warnings when building against older xservers 1918 1919 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1920 1921commit 94202cbfbca05a503acdc1cca2f8409d141173af 1922Author: Alex Deucher <alexdeucher@gmail.com> 1923Date: Fri Aug 1 21:55:40 2014 +0200 1924 1925 radeon: enable hawaii accel conditionally (v3) 1926 1927 Only if the kernel has the new CP firmware. 1928 1929 v2: check value of ACCEL_WORKING2 1930 v3 (Andreas Boll): 1931 - check for value 2 or 3 of ACCEL_WORKING2 1932 - update man page 1933 1934 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> (v2) 1935 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1936 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 1937 1938commit 03930edd49f6b8a8d79910c7be5408b47db9649b 1939Author: Andreas Boll <andreas.boll.dev@gmail.com> 1940Date: Mon Aug 4 16:23:13 2014 +0200 1941 1942 radeon: remove definitions already present in radeon_drm.h 1943 1944 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 1945 1946commit 91849fba0742ef61ba327e71fc3ce8f754af0a6f 1947Author: Andreas Boll <andreas.boll.dev@gmail.com> 1948Date: Mon Aug 4 16:23:12 2014 +0200 1949 1950 radeon: drop radeon_drm.h 1951 1952 Now we use libdrm's radeon_drm.h. 1953 1954 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 1955 1956commit b4658901bf1d619f8ff9b5f94344894f935ee6c8 1957Author: Andreas Boll <andreas.boll.dev@gmail.com> 1958Date: Mon Aug 4 16:23:11 2014 +0200 1959 1960 radeon: move RADEON_TILING_{MASK, LINEAR} from radeon_drm.h to radeon.h 1961 1962 This allows us to drop radeon_drm.h from xf86-video-ati and use instead 1963 radeon_drm.h from libdrm. 1964 1965 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 1966 1967commit 656bae7361c1e018553ef6d6d8c9efad616a4513 1968Author: Andreas Boll <andreas.boll.dev@gmail.com> 1969Date: Mon Aug 4 16:23:10 2014 +0200 1970 1971 radeon: drop redundant radeon_drm.h includes 1972 1973 Already included via radeon.h. 1974 1975 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 1976 1977commit fbf575cb010e558a87fad0de45660738c8180896 1978Author: Michel Dänzer <michel.daenzer@amd.com> 1979Date: Wed Jul 30 17:03:24 2014 +0900 1980 1981 Add Emacs .dir-local.el file 1982 1983 Based on the one from the Gallium radeon winsys, but enabling tabs for 1984 indentation. 1985 1986commit 4b5060f357a3cb248c9359c92c1e9c42ef6434c8 1987Author: Michel Dänzer <michel.daenzer@amd.com> 1988Date: Thu Jun 19 18:27:59 2014 +0900 1989 1990 glamor: Set environment variable RADEON_THREAD=0 1991 1992 Reviewed-by: Marek Olšák <marek.olsak@amd.com> 1993 1994commit 9b54caf6509a9c02dd17c9c43d6be8f7ddc98054 1995Author: Hans de Goede <hdegoede@redhat.com> 1996Date: Mon Jul 28 15:55:12 2014 +0200 1997 1998 configure: Include xorg-server.h before 1999 glamor.h 2000 2001 glamor.h cannot be included without first including xorg-server.h, this also 2002 applies to including it from configure snippets. 2003 2004 Without this the configure glamor checks fail on systems with the latest 2005 glibc, throwing this error: 2006 2007 In file included from /usr/include/xorg/misc.h:115:0, 2008 from /usr/include/xorg/screenint.h:50, 2009 from /usr/include/xorg/scrnintstr.h:50, 2010 from /usr/include/xorg/glamor.h:32, 2011 from conftest.c:61: 2012 /usr/include/xorg/os.h:579:2: error: expected identifier or '(' before 2013 '__exten 2014 strndup(const char *str, size_t n); 2015 ^ 2016 2017 This is caused by HAVE_STRNDUP not being set (it is set from xorg-server.h), 2018 causing os.h to redefine it. 2019 2020 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 2021 2022commit c4ae0e2cbcc0e2ebf9f13ee92d59b5120254a1dc 2023Author: Michel Dänzer <michel.daenzer@amd.com> 2024Date: Mon Jun 30 10:20:12 2014 +0900 2025 2026 Handle CRTC DPMS from output DPMS hooks 2027 2028 This fixes at least two issues: 2029 2030 The CRTC DPMS hook isn't called after a modeset, so the vertical blank 2031 interrupt emulation code considered the CRTC disabled after a modeset. As 2032 a side effect, page flipping was no longer used after a modeset. 2033 2034 This change also makes sure the vertical blank interrupt emulation code 2035 runs before the hardware CRTC is disabled and after it's enabled from the 2036 output DPMS hook. The wrong order could cause gnome-shell to hang after 2037 a suspend/resume and/or DPMS off/on cycle. 2038 2039 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2040 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2041 2042commit c0c3cac8613e31e310242695d0955b452f116e25 2043Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2044Date: Wed Jun 25 16:21:57 2014 +0200 2045 2046 bump version post release 2047 2048commit 906a0ec9224146098bb4581486129d2934d36495 2049Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2050Date: Wed Jun 25 16:10:21 2014 +0200 2051 2052 bump version for release 2053 2054commit cc615d06db0332fc6e673b55632bcc7bf957b44b 2055Author: Michel Dänzer <michel.daenzer@amd.com> 2056Date: Wed Jun 4 16:58:53 2014 +0900 2057 2058 Rename Option "NoAccel" to "Accel" 2059 2060 Removes the need for a double negation when forcing acceleration on. 2061 2062 Note that this change is backwards compatible, as the option parser 2063 automagically handles the 'No' prefix. 2064 2065 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2066 2067commit 851b2cf8714618843725f6d067915375485ade9d 2068Author: Adam Jackson <ajax@redhat.com> 2069Date: Wed May 21 09:34:32 2014 -0400 2070 2071 kms: Use own thunk function instead of shadowUpdatePackedWeak 2072 2073 I plan to delete the Weak functions from a future server. 2074 2075 Signed-off-by: Adam Jackson <ajax@redhat.com> 2076 2077commit b2dba2906f0b2284f17f53fd5251ba0f03d52a8b 2078Author: Michel Dänzer <michel.daenzer@amd.com> 2079Date: Thu May 15 16:07:53 2014 +0900 2080 2081 Don't disable acceleration on >= SI on attempts to force EXA 2082 2083 Also make this case clear in the log file: 2084 2085 (WW) RADEON(0): EXA not supported, using glamor 2086 2087 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2088 2089commit be1469cc23aba46daf3293b3d09c5f2e792e7f42 2090Author: Alex Deucher <alexander.deucher@amd.com> 2091Date: Fri May 2 17:48:07 2014 -0400 2092 2093 radeon: enable tiling for mullins 2094 2095 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2096 2097commit 2ae8e4b8d12f5c9bab6655eb8cd3c5c1d5cfb10e 2098Author: Samuel Li <samuel.li@amd.com> 2099Date: Tue Nov 12 15:30:42 2013 -0500 2100 2101 radeon: add Mullins pci ids. 2102 2103 Signed-off-by: Samuel Li <samuel.li@amd.com> 2104 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2105 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2106 2107commit 5c86a3461597647224c33d5190b4531aeeb2655f 2108Author: Samuel Li <samuel.li@amd.com> 2109Date: Thu Apr 17 15:17:28 2014 -0400 2110 2111 radeon: add support for Mullins. 2112 2113 Signed-off-by: Samuel Li <samuel.li@amd.com> 2114 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2115 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2116 2117commit fdb90ffc50acbb7d5ba0598470f9feeac6ce55fc 2118Author: Alex Deucher <alexander.deucher@amd.com> 2119Date: Fri May 2 17:45:45 2014 -0400 2120 2121 radeon: require libdrm_radeon 2.4.54 for mullins support 2122 2123 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2124 2125commit 06e3c8c53ef4bd159f5864eabf726438d008b49a 2126Author: Dave Airlie <airlied@redhat.com> 2127Date: Wed Apr 23 13:39:42 2014 +1000 2128 2129 radeon: fix use-after-free in modesetting cleanup 2130 2131 noticed while looking at something else. 2132 2133 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2134 Signed-off-by: Dave Airlie <airlied@redhat.com> 2135 2136commit dbac18c361f9e514ecb40d0617f9d68b65a542e0 2137Author: Michel Dänzer <michel.daenzer@amd.com> 2138Date: Mon Apr 28 17:51:56 2014 +0900 2139 2140 Revert "Adapt to load_cursor_argb signature change in xserver 1.15.99.902" 2141 2142 This reverts commit 48d3dbc8a0d3bfde88f46e402e530438f9317715. 2143 2144 xserver Git has been updated to be backwards compatible with the 2145 previous API. 2146 2147commit c84230d686c078aac1dc98d82153f8b02521b2e1 2148Author: Michel Dänzer <michel.daenzer@amd.com> 2149Date: Fri Apr 25 09:17:51 2014 +0900 2150 2151 dri2: Handle PRIME for source buffer as well in radeon_dri2_copy_region2 2152 2153 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=77810 2154 2155 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2156 2157commit 48d3dbc8a0d3bfde88f46e402e530438f9317715 2158Author: Michel Dänzer <michel.daenzer@amd.com> 2159Date: Tue Apr 15 17:45:35 2014 +0900 2160 2161 Adapt to load_cursor_argb signature change in xserver 1.15.99.902 2162 2163 Apart from the compiler warning below, not doing this may result in 2164 accidentally using software cursors. 2165 2166 ../../src/drmmode_display.c:808:5: warning: initialization from incompatible pointer type [enabled by default] 2167 .load_cursor_argb = drmmode_load_cursor_argb, 2168 ^ 2169 ../../src/drmmode_display.c:808:5: warning: (near initialization for 'drmmode_crtc_funcs.load_cursor_argb') [enabled by default] 2170 2171 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2172 2173commit 409786a8f780d78a83bf0bddea5d37117ff6fa39 2174Author: Michel Dänzer <michel.daenzer@amd.com> 2175Date: Thu Apr 10 11:43:04 2014 +0900 2176 2177 glamor: Fix test for creating shared pixmaps 2178 2179 The pixmap usage hint is not a bitmask in general. The test for 2180 CREATE_PIXMAP_USAGE_SHARED was incorrectly triggering for a glamor internal 2181 usage hint being added in the xserver tree. 2182 2183 Tested-by: Ed Tomlinson <edtoml@gmail.com> 2184 2185commit aecf1c4e5f4718adcfb85836830d065d3f4f97a5 2186Author: Michel Dänzer <michel.daenzer@amd.com> 2187Date: Thu Apr 10 15:10:56 2014 +0900 2188 2189 dri2: Fix conflicting CreatePixmap usage flag definitions 2190 2191 RADEON_CREATE_PIXMAP_TILING_MICRO_SQUARE was the same as 2192 RADEON_CREATE_PIXMAP_DRI2. 2193 2194 Disambiguate the definitions and rearrange them to try and prevent this 2195 from happening again. 2196 2197 Tested-by: Ed Tomlinson <edtoml@gmail.com> 2198 2199commit b50da3b96c212086cb58501dbe988d64f1f35b6d 2200Author: Hans de Goede <hdegoede@redhat.com> 2201Date: Fri Apr 11 09:44:37 2014 +0200 2202 2203 Fix building on older servers without xf86platformBus.h 2204 2205 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 2206 2207commit ed0cfbb4fe77146b0b38f777bc28f3a4ea6da07f 2208Author: Hans de Goede <hdegoede@redhat.com> 2209Date: Fri Mar 7 13:27:30 2014 +0100 2210 2211 Add support for server managed fds 2212 2213 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 2214 2215commit 3d7861fe112f25874319d4cdc12b745fbcd359cf 2216Author: Hans de Goede <hdegoede@redhat.com> 2217Date: Mon Mar 17 10:38:13 2014 +0100 2218 2219 Add radeon_get_drm_master_fd helper function 2220 2221 This is a preparation patch for adding server-managed-fd support without it 2222 turning into a goto fest. 2223 2224 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 2225 2226commit a63342ad15408071437c80b411d14196f3288aed 2227Author: Hans de Goede <hdegoede@redhat.com> 2228Date: Mon Mar 17 10:36:55 2014 +0100 2229 2230 radeon_open_drm_master get rid of unnecessary goto 2231 2232 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 2233 2234commit bdc412044f6ced056cd57320d1b2ee0d967c2191 2235Author: Michel Dänzer <michel.daenzer@amd.com> 2236Date: Thu Mar 13 16:40:19 2014 +0900 2237 2238 Build against glamor in the xserver tree if available 2239 2240commit 921a153f9964ca452e1241f76c7f7d653f42ceaf 2241Author: Alex Deucher <alexander.deucher@amd.com> 2242Date: Tue Mar 11 12:24:13 2014 -0400 2243 2244 update man page to reflect tiling changes for CI parts 2245 2246 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2247 2248commit 5eee8a4d5c86bb1cc34d8caf2f2b64b53c241fa5 2249Author: Alex Deucher <alexander.deucher@amd.com> 2250Date: Mon Mar 10 16:20:09 2014 -0400 2251 2252 radeon: enable tiling by default on CIK 2253 2254 Now that mesa 10.1 is released, we can enable this by 2255 default for CIK parts. Tiling improves memory bandwidth 2256 utilization. 2257 2258 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2259 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2260 2261commit 691ec3d99c30111a4789830dfccb6eb5d3c40187 2262Author: Alex Deucher <alexander.deucher@amd.com> 2263Date: Mon Mar 10 16:17:34 2014 -0400 2264 2265 radeon: require libdrm 2.4.51 2266 2267 Required for proper tiling support on CIK parts. 2268 2269 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2270 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2271 2272commit 515bcf14d514f9dcaaf30fd0bf1ef6dd6ba9a0cd 2273Author: Michel Dänzer <michel.daenzer@amd.com> 2274Date: Tue Mar 4 12:34:26 2014 +0900 2275 2276 Allow enabling glamor on R500 (and R300) class 3D engines as well. 2277 2278 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=75709 2279 2280 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2281 2282commit ea6d0affe52d82becadab6fb1c87f9261b0605a2 2283Author: Michel Dänzer <michel.daenzer@amd.com> 2284Date: Mon Feb 24 13:12:21 2014 +0900 2285 2286 Only log debugging output about initializing colormaps when we're doing so 2287 2288 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2289 2290commit 8da17f30c70f4494ce22ad781a1cee17041812f3 2291Author: Jérôme Glisse <jglisse@redhat.com> 2292Date: Mon Feb 24 14:50:25 2014 -0500 2293 2294 evergreen: fix shader constant upload on ppc 2295 2296 The number of dword we have to swap is (16*4) ie 16 vectors of 2297 4 floats each not 16 floats. Never hit this issue before because 2298 we never had more than 4 constant vector. 2299 2300 Signed-off-by: Jérôme Glisse <jglisse@redhat.com> 2301 2302commit cadb6b493942a84bfeb298751dce0dee39257a06 2303Author: Alex Deucher <alexander.deucher@amd.com> 2304Date: Fri Feb 21 08:33:21 2014 -0500 2305 2306 radeon: don't install colormap handling if there are no crtcs 2307 2308 Fixes a crash on cards with 0 crtcs. 2309 2310 Discussion: 2311 http://lists.freedesktop.org/archives/dri-devel/2014-February/054186.html 2312 2313 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2314 2315commit 7cd972a85705341dd8306eefc558ed9e5def05d7 2316Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2317Date: Thu Feb 20 11:14:03 2014 +0100 2318 2319 return immediately in preinit when called with PROBE_DETECT 2320 2321 This fixes a crash with Xorg -configure. 2322 2323 Bug: 2324 https://bugs.launchpad.net/ubuntu/+source/xorg/+bug/1278046 2325 2326 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2327 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2328 2329commit 8de6f7b2f476d3baa9c7e2bb3544e4bafaad46b7 2330Author: Alex Deucher <alexander.deucher@amd.com> 2331Date: Fri Jan 24 11:04:30 2014 -0500 2332 2333 bump version post release 2334 2335 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2336 2337commit 0333f5bda27dc0ec2edc180c7a4dc9a432f13f97 2338Author: Alex Deucher <alexander.deucher@amd.com> 2339Date: Fri Jan 24 10:19:49 2014 -0500 2340 2341 radeon: bump version for release 2342 2343 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2344 2345commit fc4167f2a85d9cba65078d8fc6f08c7a619ad66e 2346Author: Alex Deucher <alexander.deucher@amd.com> 2347Date: Fri Jan 24 10:17:08 2014 -0500 2348 2349 Require glamor 0.6.0 2350 2351 This is required for Xv support and a number of important 2352 performance improvements. 2353 2354 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2355 2356commit f2a0a5cf6c5a21e2a02280e110a4eb8e6609dace 2357Author: Michel Dänzer <michel.daenzer@amd.com> 2358Date: Wed Jan 22 11:04:42 2014 +0900 2359 2360 Don't require the glamoregl module to be pre-loaded with xserver >= 1.15 2361 2362 The issues with loading it on demand have been fixed in xserver 1.15. 2363 2364 Inspired by Jérôme Glisse on IRC. 2365 2366commit 3213df16d61302148be0088c8f93c6a5a88558f1 2367Author: Michel Dänzer <michel.daenzer@amd.com> 2368Date: Wed Jan 8 11:30:59 2014 +0900 2369 2370 dri2: Make last_vblank_seq local unsigned to match dpms_last_seq 2371 2372 Without this, I was occasionally running into gnome-shell hangs due to 2373 wildly off vblank sequence values. Doesn't seem to happen anymore with 2374 this change. 2375 2376 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2377 2378commit bcc454ea2fb239e13942270faec7801270615b9c 2379Author: Alex Deucher <alexander.deucher@amd.com> 2380Date: Mon Jan 6 09:52:50 2014 -0500 2381 2382 radeon/exa: Always use a scratch surface for UTS to vram 2383 2384 If we don't, we may hit a buffer that crosses the 2385 visible vram boundary resulting in a sigbus when the 2386 CPU accesses the buffer beyond the PCI aperture. 2387 This will introduce an extra copy in certain cases. 2388 2389 This is based on Michel's patch from bug 44099 updated 2390 for all asic families. 2391 2392 Bug: 2393 https://bugs.freedesktop.org/show_bug.cgi?id=44099 2394 2395 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2396 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2397 2398commit 04ef035c9315b4a6fbf1b14720be87cee4099a9f 2399Author: Alex Deucher <alexander.deucher@amd.com> 2400Date: Tue Dec 24 15:14:35 2013 -0500 2401 2402 drm/radeon: fix SUMO2 pci id 2403 2404 0x9649 is sumo2, not sumo. 2405 2406 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2407 2408commit bfbff3b246db509c820df17b8fcf5899882ffcfa 2409Author: Robert Millan <rmh@freebsd.org> 2410Date: Fri Dec 20 11:03:14 2013 +0000 2411 2412 radeon: Restore kernel module load on FreeBSD. 2413 2414 Since the introduction of a call to drmCheckModesettingSupported() 2415 in radeon_kernel_mode_enabled(), with abort condition if such call 2416 fails, the drmOpen() call in radeon_open_drm_master() no longer 2417 takes the responsibility of loading the radeon kernel module. 2418 2419 However at least on FreeBSD (and GNU/kFreeBSD), X is still relied 2420 on to load the modules it needs. This commit restores the old 2421 behaviour of loading kernel modules on these systems. 2422 2423 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=72852 2424 Signed-off-by: Robert Millan <rmh@freebsd.org> 2425 2426commit 796c9a0cb587f528326bede11fa3f3eb7d3edaf1 2427Author: Robert Millan <rmh@freebsd.org> 2428Date: Fri Dec 20 10:55:09 2013 +0000 2429 2430 radeon: Set first parameter of drmOpen() to NULL 2431 2432 Since the introduction of a call to drmCheckModesettingSupported() 2433 in radeon_kernel_mode_enabled(), with abort condition if such call 2434 fails, the first argument to drmOpen() call in radeon_open_drm_master() 2435 has become a no-op red herring. 2436 2437 Such argument (a kernel module name) is supposed to result in load 2438 of specified kernel module. However, this will never happen. The 2439 problem is that if the code containing drmOpen() call is reached, it 2440 means that drmCheckModesettingSupported() check has previously 2441 succeeded, which implies the module is already loaded. 2442 2443 So, drmOpen() will never load a kernel module. But it gives the 2444 impression that it will. 2445 2446 In order to avoid this confusion, this commit replaces it with NULL, 2447 like xf86-video-intel driver does. 2448 2449 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=72849 2450 Signed-off-by: Robert Millan <rmh@freebsd.org> 2451 2452commit d571d6af70ef27efd1ed6420eb892bdde963ed7a 2453Author: Alex Deucher <alexander.deucher@amd.com> 2454Date: Tue Sep 24 11:39:10 2013 -0400 2455 2456 radeon/kms: add Hawaii pci ids 2457 2458 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2459 2460commit e38a92e00d015a6b80a1f3a16d58c61f084b066f 2461Author: Alex Deucher <alexander.deucher@amd.com> 2462Date: Tue Sep 24 11:35:00 2013 -0400 2463 2464 radeon: add support for Hawaii 2465 2466 Disabled by default until the acceleration code stablizes. 2467 2468 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2469 2470commit e4cd0f4392ea11c93088ad429f36eaaf9bcbf505 2471Author: Alex Deucher <alexander.deucher@amd.com> 2472Date: Wed Oct 30 15:21:26 2013 -0400 2473 2474 radeon: enable tiling on SI by default (v2) 2475 2476 Now that mesa 9.2 is out with support for tiling 2477 on SI asics, we can enable it here. Tiling improves 2478 memory bandwidth utilization. 2479 2480 V2: update man page 2481 2482 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2483 2484commit 3b38701a72fa1cad1e4610a2f4330b3da4cc6391 2485Author: Vadim Girlin <vadimgirlin@gmail.com> 2486Date: Fri Nov 1 10:36:39 2013 -0400 2487 2488 radeon: disable 2D tiling on buffers < 128 pixels 2489 2490 Seems to run into alignment problems with certain 2491 card configurations. 2492 2493 bug: 2494 https://bugs.freedesktop.org/show_bug.cgi?id=70675 2495 2496 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2497 2498commit 0c921edf0162fed616cea9d02e168b719243bcd2 2499Author: Jerome Glisse <jglisse@redhat.com> 2500Date: Wed Oct 30 13:01:02 2013 -0400 2501 2502 radeon/glamor: with new pixmap for dri2 drawable no need to create new texture 2503 2504 When creating a new pixmap/bo for dri2 drawable there is no need to create a 2505 new texture, instead the texture associated with the new pixmap should be use 2506 otherwise there is a missmatch between the bo backing the texture for the 2507 drawable and the bo used by dri2 client. 2508 2509 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 2510 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2511 2512commit f1dc677e79cd7a88d7379a934ebc7d87a3b18805 2513Author: Christopher James Halse Rogers <raof@ubuntu.com> 2514Date: Mon Sep 23 12:25:29 2013 -0700 2515 2516 EXA/evergreen: Paranoia around linear tiling. (v2) 2517 2518 The last two bytes of tiling_mode contain the actual tiling mode; the rest are 2519 extra tiling configuration bits. These configuration bits are not necessarily 2520 zero for a linear buffer, so mask them out before checking for linearity 2521 2522 v2: Also fix up evergreen_textured_videofuncs.c 2523 2524 agd5f: remove trailing whitespace 2525 2526 Signed-off-by: Christopher James Halse Rogers <raof@ubuntu.com> 2527 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2528 2529commit 67fb82a3f0759b171fea21b475a70fa825693570 2530Author: Alex Deucher <alexander.deucher@amd.com> 2531Date: Tue Oct 1 09:35:30 2013 -0400 2532 2533 radeon: fix the non-glamor build harder... 2534 2535 I need to stop pushing patches first thing in the morning. 2536 2537 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2538 2539commit afc0374fdec3e24ece25805724459463e9a19f5e 2540Author: Alex Deucher <alexander.deucher@amd.com> 2541Date: Tue Oct 1 09:32:02 2013 -0400 2542 2543 drm/radeon: fix non-glamor build 2544 2545 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2546 2547commit 2d791370dfc5570eb74d7a1fb3baf4d4c8ecf243 2548Author: Dave Airlie <airlied@redhat.com> 2549Date: Mon Sep 23 07:57:15 2013 +0100 2550 2551 radeon: use glamor Xv support if present. 2552 2553 This creates adaptors using glamor if possible. 2554 2555 Signed-off-by: Dave Airlie <airlied@redhat.com> 2556 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2557 2558commit cbb99f659ee7b18ded0008a606e41ded38c1a194 2559Author: Alex Deucher <alexander.deucher@amd.com> 2560Date: Mon Sep 30 08:55:16 2013 -0400 2561 2562 Revert "radeon: add glamor Xv support (v2)" 2563 2564 This causes problems if glamor doesn't have Xv support 2565 enabled. I just noticed that Dave has a better version, 2566 so use that instead. 2567 2568 This reverts commit 4fc1fa920584ace2c84d75af82d06962d0c84ec8. 2569 2570commit 4fc1fa920584ace2c84d75af82d06962d0c84ec8 2571Author: Alex Deucher <alexander.deucher@amd.com> 2572Date: Thu Sep 26 11:02:55 2013 -0400 2573 2574 radeon: add glamor Xv support (v2) 2575 2576 v2: guard new glamor Xv bits with USE_GLAMOR 2577 2578 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2579 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2580 2581commit 41dfe327ac8740ac2cd84def96b5947224e422e7 2582Author: Alex Deucher <alexander.deucher@amd.com> 2583Date: Thu Sep 26 11:00:49 2013 -0400 2584 2585 radeon: fix limit handling for cards with >4G of ram 2586 2587 We can overflow the 32-bit limit. 2588 2589 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2590 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2591 2592commit c45e728107269c6f51599dad4f6a02ccfef703f1 2593Author: Michel Dänzer <michel.daenzer@amd.com> 2594Date: Wed Sep 18 10:57:52 2013 +0200 2595 2596 DRI2: Install client callback only once 2597 2598 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=60182 2599 Acked-by: Alex Deucher <alexander.deucher@amd.com> 2600 2601commit b955ddd0d41801e4ca0c30a70a5d0b27c3f366c8 2602Author: Alex Deucher <alexander.deucher@amd.com> 2603Date: Mon Sep 16 10:58:22 2013 -0400 2604 2605 glamor: require 0.5.1 or newer 2606 2607 0.5.1 fixes a number of issues. 2608 2609 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2610 2611commit 282587cd0709850e7bacb1d8307065d95dc2c97d 2612Author: Alex Deucher <alexander.deucher@amd.com> 2613Date: Mon Sep 16 09:41:13 2013 -0400 2614 2615 radeon: enable glamor by default (v4) 2616 2617 It's required for newer asics, so enable it by 2618 default. 2619 2620 v2: update the autoconf help. Require --disable-glamor 2621 to build without glamor support. 2622 v3: default to yes rather than auto 2623 v4: fix help text 2624 2625 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2626 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2627 2628commit fdb7563a5cbc736b09c2864b67a93b475c98b2bd 2629Author: Alex Deucher <alexander.deucher@amd.com> 2630Date: Thu Jan 24 21:17:11 2013 -0500 2631 2632 radeon/kms: add berlin pci ids 2633 2634 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2635 2636commit 8927d33f76ee12bc618fecfc59fc7ff1fcedcd5e 2637Author: Mark Kettenis <kettenis@openbsd.org> 2638Date: Thu Aug 22 10:32:46 2013 -0400 2639 2640 Fix shadowfb on big-endian machines 2641 2642 For shadowfb, the framebuffer needs to have the RADEON_TILING_SURFACE 2643 flag set, otherwise the appropriate byte swapping won't happen. 2644 2645 See https://bugs.freedesktop.org/show_bug.cgi?id=66663 2646 2647 Signed-off-by: Mark Kettenis <kettenis@openbsd.org> 2648 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2649 2650commit fa83d3d1636c315bc43dd622d407eb058e3ef976 2651Author: Alex Deucher <alexander.deucher@amd.com> 2652Date: Tue Aug 20 09:29:23 2013 -0400 2653 2654 radeon: disallow glamor on pre-R600 asics 2655 2656 I'm not sure they can handle the shaders properly, especially 2657 only older parts like r300. 2658 2659 This will avoid display corruption problems reported by people 2660 using glamor on older asics by falling back to EXA if they try 2661 and enable glamor. 2662 2663 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2664 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2665 2666commit d0323622ee9b97a5f246baffbb2c65930a78ed14 2667Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2668Date: Wed Aug 7 11:27:07 2013 +0200 2669 2670 bump version post release 2671 2672 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2673 2674commit 9c97cca5c24409ca8447c99f051a12fd2d494e79 2675Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2676Date: Wed Aug 7 10:48:17 2013 +0200 2677 2678 radeon: bump version for release 2679 2680 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2681 2682commit 16270cfb202ab67dd152644ef019b2f1ee4d0341 2683Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2684Date: Wed Aug 7 10:29:33 2013 +0200 2685 2686 add bicubic_table.py to EXTRA_DIST 2687 2688 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2689 2690commit 2cb9197ca7a337c911f38b5de562a2364b922b86 2691Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2692Date: Wed Aug 7 10:28:52 2013 +0200 2693 2694 kill unused radeon_driver.c 2695 2696 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2697 2698commit c5cbfcf575b0b4aea6f797558ae974c1453c8e07 2699Author: Alex Deucher <alexander.deucher@amd.com> 2700Date: Tue Jul 30 10:08:25 2013 -0400 2701 2702 drmmode: add support for multi-screen reverse optimus 2703 2704 Initial reverse optimus didn't consider multiple screens, so 2705 this overhauls the code to use the new X server interface, 2706 and allows for multiple outputs on the dGPU to be used with 2707 the iGPU doing the rendering. Ported from Dave's nouveau 2708 patch. 2709 2710 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2711 2712commit 429d5b797769895eb4f5fef816ce4e2f3a342031 2713Author: Dave Airlie <airlied@redhat.com> 2714Date: Tue Jan 8 15:56:37 2013 +1000 2715 2716 radeon: add support for reverse prime (v2) 2717 2718 This adds support for reverse prime configurations 2719 2720 v2: fix compilation with older xservers 2721 2722 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2723 2724commit 4de9356a2900ae0fb380a2350791ef045629cd05 2725Author: Alex Deucher <alexander.deucher@amd.com> 2726Date: Mon Aug 5 17:57:16 2013 -0400 2727 2728 radeon: fix naming clashes with multiple GPUs (v3) 2729 2730 The compat naming code for UMS causes problems 2731 with multiple GPU as you may end up with the same 2732 output name on multiple GPUs. Adjust the naming on 2733 secondary GPUs to avoid conflicts. 2734 2735 v2: integrate Dave's fixes for nouveau 2736 v3: keep compat with existing naming on primary GPU 2737 2738 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2739 2740commit 2ae6bb18fefddb309920fa69c9b56c3a7f3db7b4 2741Author: Grigori Goronzy <greg@chown.ath.cx> 2742Date: Wed Jul 31 12:01:20 2013 +0200 2743 2744 EXA/evergreen/ni: replace magic number 2745 2746 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2747 2748commit 6a278369c05a298a4367306d986467a9ceacae8c 2749Author: Raul Fernandes <rgfernandes@gmail.com> 2750Date: Tue Jul 30 09:26:05 2013 -0400 2751 2752 EXA/6xx/7xx: optimize non-overlapping Copy 2753 2754 In case dst and src rectangles of a Copy operation in the same surface 2755 don't overlap, it is safe to skip the scratch surface. This is a 2756 common case. 2757 2758 Based on evergreen/ni patch from Grigori Goronzy. 2759 2760 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2761 2762commit 4375a6e75e5d41139be7031a0dee58c057ecbd07 2763Author: Grigori Goronzy <greg@chown.ath.cx> 2764Date: Mon Jul 22 02:30:28 2013 +0200 2765 2766 EXA/evergreen/ni: accelerate PictOpOver with component alpha 2767 2768 Subpixel text rendering is typically done with a solid src and a 2769 pixmap mask. Traditionally, this cannot be accelerated in a single 2770 pass and requires two passes [1]. However, we can cheat a little 2771 with a constant blend color. 2772 2773 We can use: 2774 2775 const.A = src.A / src.A 2776 const.R = src.R / src.A 2777 const.G = src.G / src.A 2778 const.B = src.B / src.A 2779 2780 dst.A = const.A * (src.A * mask.A) + (1 - (src.A * mask.A)) * dst.A 2781 dst.R = const.R * (src.A * mask.R) + (1 - (src.A * mask.R)) * dst.R 2782 dst.G = const.G * (src.A * mask.G) + (1 - (src.A * mask.G)) * dst.G 2783 dst.B = const.B * (src.A * mask.B) + (1 - (src.A * mask.B)) * dst.B 2784 2785 This only needs a single source value. src.A is cancelled down in 2786 the right places. 2787 2788 [1] http://anholt.livejournal.com/32058.html 2789 2790commit 94d0d14914a025525a0766669b556eaa6681def7 2791Author: Grigori Goronzy <greg@chown.ath.cx> 2792Date: Thu Jul 18 16:06:23 2013 +0200 2793 2794 EXA/evergreen/ni: fast solid pixmap support 2795 2796 Solid pixmaps are currently implemented with scratch pixmaps, which 2797 is slow. This replaces the hack with a proper implementation. The 2798 Composite shader can now either sample a src/mask or use a constant 2799 value. 2800 2801commit 5bb04351c43a91a1d60348b7293544da05d75e72 2802Author: Grigori Goronzy <greg@chown.ath.cx> 2803Date: Fri Jul 27 17:31:53 2012 +0200 2804 2805 EXA/evergreen/ni: optimize non-overlapping Copy 2806 2807 In case dst and src rectangles of a Copy operation in the same surface 2808 don't overlap, it is safe to skip the scratch surface. This is a 2809 common case. 2810 2811commit c08e09b7bec441c4bf93b4cae4de1260754bf940 2812Author: Grigori Goronzy <greg@chown.ath.cx> 2813Date: Sat May 18 13:46:03 2013 +0200 2814 2815 Fix RADEON_FALLBACK logging 2816 2817commit c16c59f8f9b6aa7a4a6a6465582ad98f02a3606a 2818Author: Mark Kettenis <kettenis@openbsd.org> 2819Date: Sun Jul 7 13:44:13 2013 +0200 2820 2821 Always go through DFS/UTS when byte swapping is needed. 2822 2823 Before commit ef9bfb262db7004bef3704e5d914687e50d3fca4 and 2824 e5bd99faa3b6629a55168386d5dfa936ee4e97ae, byte swapping for the front buffer 2825 used to be done in hardware. Now that this no longer happens we need to let 2826 DFS/UTS ihandle the byte swapping. 2827 2828 See https://bugs.freedesktop.org/show_bug.cgi?id=66663 . 2829 2830 Signed-off-by: Mark Kettenis <kettenis@openbsd.org> 2831 2832commit 1239dbbd8c8d9b55756c1de52cad353171a06522 2833Author: Alex Deucher <alexander.deucher@amd.com> 2834Date: Sun Jul 7 20:22:10 2013 -0400 2835 2836 radeon: bump libdrm requirement to 2.4.46 for CIK support 2837 2838 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2839 2840commit 869e0b187b2d07e637c4482fe50d9f1b0c09f4e4 2841Author: Alex Deucher <alexander.deucher@amd.com> 2842Date: Thu Jan 24 21:12:28 2013 -0500 2843 2844 radeon/kms: add kabini pci ids 2845 2846 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2847 2848commit 0edcefc4c97e4572431c38fe170032470e22e0b7 2849Author: Alex Deucher <alexander.deucher@amd.com> 2850Date: Fri Jun 7 14:56:26 2013 -0400 2851 2852 radeon/kms: add bonaire pci ids 2853 2854 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2855 2856commit ebc32b27af25b23604e725eb50d844a8d26116bb 2857Author: Alex Deucher <alexander.deucher@amd.com> 2858Date: Fri Aug 31 18:17:24 2012 -0400 2859 2860 radeon: update cursor handling for CIK 2861 2862 CIK asics have 128x128 hw cursors 2863 2864 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2865 2866commit 4adaea996454b1ad5185f9c0f37667dbfc266495 2867Author: Alex Deucher <alexander.deucher@amd.com> 2868Date: Fri Jun 7 14:47:36 2013 -0400 2869 2870 radeon: add family ids for CIK 2871 2872 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2873 2874commit 3626ab147b6749a355e7d21710447f13b9796c77 2875Author: Leo Liu <leo.liu@amd.com> 2876Date: Fri Jun 14 10:27:20 2013 -0400 2877 2878 radeon: fix a memory leak in get_modes() callback. 2879 2880 leak happens when looping xrandr prop. 2881 2882 Signed-off-by: Leo Liu <leo.liu@amd.com> 2883 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2884 2885commit c0f7d03cb29166c6c00dd6fb698ba3f70c59dbdf 2886Author: Alex Deucher <alexdeucher@gmail.com> 2887Date: Mon Jun 17 11:38:27 2013 -0400 2888 2889 man: fix up previous commit 2890 2891 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2892 2893commit 03b5edc9dd3859eed2504155970613aaf88df852 2894Author: Eric S. Raymond <esr@thyrsus.com> 2895Date: Fri Jun 7 13:23:14 2013 -0400 2896 2897 Fix some incomplete list entries. 2898 2899 This necessarily involved adding some content, which I was able to 2900 Google for. Please review, correct, and make more specific. 2901 2902 Bodiless .TP entries screw up doclifter's page parsing. Please 2903 don't do that anymore. 2904 2905commit 7ce43639a8e53eefb94e5255a39c1232b6bdbf66 2906Author: Michel Dänzer <michel.daenzer@amd.com> 2907Date: Thu Jun 6 09:58:27 2013 +0200 2908 2909 autogen.sh: Restore passing --enable-maintainer-mode to configure 2910 2911 Looks like this was accidentally dropped when the GNOME Build API was 2912 implemented. 2913 2914commit 365e571d9de0b5979425c34210789afca4ea5f68 2915Author: Samuel Li <samuel.li@amd.com> 2916Date: Wed Mar 20 11:59:58 2013 -0400 2917 2918 radeon: use direct mapping for fast fb access. 2919 2920 Signed-off-by: Samuel Li <samuel.li@amd.com> 2921 2922commit bd2557ea5ef84b975060e929d5ece53ec464336f 2923Author: Ilija Hadzic <ilijahadzic@gmail.com> 2924Date: Wed May 8 22:39:48 2013 -0400 2925 2926 DRI2: add interpolated blanks to frame number in event handlers 2927 2928 The 'frame' argument passed to event handlers is the plain 2929 CRTC vblank counter that stops progressing when the 2930 associated display is in DPMS-off mode. If we have a 2931 DPMS-off period the frame counter and MSC will 2932 diverge, which can cause some higher-level functions 2933 to return incorrect values. 2934 2935 This patch fixes the problem by adding interpolated 2936 vblanks to the frame counter before using it in handler 2937 functions. 2938 2939 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 2940 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2941 2942commit 385a92b4fbe8f53b359ef6c463704414d00476fa 2943Author: Ilija Hadzic <ilijahadzic@gmail.com> 2944Date: Wed May 8 22:39:47 2013 -0400 2945 2946 DRI2: support scheduling emulated events with zero delay 2947 2948 Now that we fully emulating a running CRTC through DPMS-off state 2949 it is possible to come up with a zero delay when scheduling 2950 a swap or MSC-wait (e.g., if a call into respective wait function 2951 was entered very late). This patch wraps the TimerSet function 2952 into our own radeon_dri2_schedule_event such that the latter 2953 calls the event right away if zero delay is specified. 2954 2955 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 2956 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2957 2958commit e87b52e6ad41ad7a87e43b818d80e7d522d9c68d 2959Author: Ilija Hadzic <ilijahadzic@gmail.com> 2960Date: Wed May 8 22:39:46 2013 -0400 2961 2962 DRI2: make wait_msc and get_msc work on disabled CRTCs 2963 2964 Now that the running CRTC is emulated through DPMS-off 2965 states, wait_msc and get_msc must also work on disabled 2966 CRTCs. When CRTC is disabled, we must extrapolate the 2967 MSC count from present time and last MSC when CRTC was 2968 running and also use timers facility in wait_msc. 2969 2970 v2: CRTC-private now stores frame rate instead of nominal 2971 vblank period. 2972 2973 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 2974 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2975 2976commit f1584152bbeda98cc1907359f1c7954a84c8837d 2977Author: Ilija Hadzic <ilijahadzic@gmail.com> 2978Date: Wed May 8 22:39:45 2013 -0400 2979 2980 DRI2: move rename and rework radeon_dri2_deferred_swap 2981 2982 radeon_dri2_deferred_swap will be used to generate 2983 real events (not just fallbacks) so now it needs to 2984 generate real timestamp and frame counter. Also 2985 this function will be used both by schedule_swap 2986 and wait_msc, so give it a more generic name: 2987 radeon_dri2_deferred_event 2988 2989 v2: - Extrapolate the frame number from the time 2990 of actual execution of the function instead 2991 of using the MSC calculated (extrapolated) 2992 at event scheduling time. 2993 - CRTC-private now stores frame rate instead of 2994 nominal vblank period. 2995 2996 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 2997 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2998 2999commit 84bce7943b735aee19e26be4d47fdf140564f3f5 3000Author: Ilija Hadzic <ilijahadzic@gmail.com> 3001Date: Wed May 8 22:39:44 2013 -0400 3002 3003 DRI2: track CRTC in event record 3004 3005 Tracking the CRTC associated with an event will save us 3006 some lookups later in event handlers. 3007 3008 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3009 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3010 3011commit e40d5390b3efdea3e02267413350410d8e6a2970 3012Author: Ilija Hadzic <ilijahadzic@gmail.com> 3013Date: Wed May 8 22:39:43 2013 -0400 3014 3015 DRI2: hook up vblank extrapolation to schedule_swap 3016 3017 This patch hooks up swap-scheduling function with 3018 vblank-extrapolation function. Rather than waiting for 3019 fixed time, we calculate exactly how much we should wait 3020 and what we should update target_msc to using 3021 radeon_dri2_extrapolate_msc_delay helper function 3022 and schedule the swap completion using DIX's timer facility. 3023 3024 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3025 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3026 3027commit ffaa5abf207415159cdb28e90da49b95f497ef61 3028Author: Ilija Hadzic <ilijahadzic@gmail.com> 3029Date: Wed May 8 22:39:42 2013 -0400 3030 3031 DRI2: add vblank extrapolation function 3032 3033 Implement a helper function that will be called when emulating 3034 the running CRTC. The function should be called only when CRTC 3035 is in DPMS-off state. It will look at the vblank count and the 3036 time that was recorded last time the CRTC was running and 3037 calculate how long one must wait (from present time) until 3038 the target_msc is reached if the CRTC were running. 3039 3040 v2: - CRTC-private now stores frame rate instead of nominal 3041 vblank period. 3042 - DIX's timer facility can sometimes wake up the scheduled 3043 functions more than a millisecond earlier. To avoid 3044 generating an old MSC, we have to add more margin when 3045 converting the delay in microseconds to milliseconds. 3046 3047 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3048 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3049 3050commit 80ae2291d082b57c70d27a80182f00d760fb3d3a 3051Author: Ilija Hadzic <ilijahadzic@gmail.com> 3052Date: Wed May 8 22:39:41 2013 -0400 3053 3054 DRI2: adjust MSC by the number of interpolated vblanks 3055 3056 Normally, MSC value equals the vblank count of the CRTC 3057 on which the drawable resides. However, we are now interpolating 3058 vblanks through DPMS-off state, so if the CRTC is turned 3059 off at some point, the vlbank count does not change, but 3060 MSC does change by the amount that equals the cumulative 3061 number of interpolated vblanks. 3062 3063 So each time we interact with the CRTC's vblank facility, 3064 we have to adjust the vblank counter: in the request, we 3065 have to subtract the number of interpolated vblanks from 3066 MSC and in the reply we have to add the number of interpolated 3067 vblanks to the MSC. 3068 3069 This patch only makes the MSC correct when CRTC is on. 3070 The subsequent patch will construct the MSC when CRTC is not 3071 running. 3072 3073 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3074 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3075 3076commit 5fd2eb5d12cea32927a9e6c6ce4afd18aa7d046a 3077Author: Ilija Hadzic <ilijahadzic@gmail.com> 3078Date: Wed May 8 22:39:40 2013 -0400 3079 3080 drmmode: calculate interpolated vblanks while in dpms-off state 3081 3082 This adds provisions for interpolating vblanks while the CRTC 3083 is in DPMS-off state. When entering DPMS-off state, we 3084 record the last vblank time, sequence number and frame rate 3085 in CRTC-private structure. 3086 3087 When going back to DPMS-on state we read the current 3088 time and calculate how long we have been off. Then we derive 3089 how many vblanks that would have been had the CRTC remained 3090 running. These are the interpolated vblanks. 3091 3092 Finally, we accumulate the number of interpolated vblanks 3093 in CRTC-private structure to get the number of interpolated 3094 vblanks over the system lifetime. 3095 3096 v2: Track frame rate instead of vblank period. The former 3097 eliminates some roundoff errors. 3098 3099 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3100 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3101 3102commit e41ad30d09be4962cfb1942b0b9f63875dbb2d2a 3103Author: Ilija Hadzic <ilijahadzic@gmail.com> 3104Date: Wed May 8 22:39:39 2013 -0400 3105 3106 drmmode: remove dead code from drmmode_crtc_dpms 3107 3108 We are about to modify this function, so let's clean it 3109 up first. 3110 3111 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3112 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3113 3114commit 373671d2eed30e30b3bdee7e40426cf58c127234 3115Author: Ilija Hadzic <ilijahadzic@gmail.com> 3116Date: Wed May 8 22:39:38 2013 -0400 3117 3118 drmmode: add drmmode_get_current_ust function 3119 3120 The new helper function retrieves current time in the format 3121 that is compatible with vblank timestamps. 3122 3123 v2: - fix an incorrect statement in a comment 3124 - add a #define so that don't depend on libdrm patches that 3125 have not yet been accepted upstream 3126 3127 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3128 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3129 3130commit 006fbbd1d38a089b50ab3197d32815689ed249fa 3131Author: Ilija Hadzic <ilijahadzic@gmail.com> 3132Date: Wed May 8 22:39:37 2013 -0400 3133 3134 DRI2: make populate_vbl_request_type external 3135 3136 We'll need to use this function in other .c files so 3137 remove static annotation from it. Also, add radeon_ 3138 prefix because this function is now part of global namespace, 3139 so it would be nice to know where it is coming from. 3140 3141 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3142 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3143 3144commit 34660d8757ff16acd1686ff2872f4600d92a68bf 3145Author: Ilija Hadzic <ilijahadzic@gmail.com> 3146Date: Wed May 8 22:39:36 2013 -0400 3147 3148 DRI2: change signature of populate_vbl_request_type 3149 3150 This function no longer uses info argument, so nuke it. 3151 3152 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3153 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3154 3155commit ab6e85d1f5dd4657591b314b85fd2453f99c3712 3156Author: Michel Dänzer <michel.daenzer@amd.com> 3157Date: Fri Mar 22 10:51:05 2013 +0100 3158 3159 glamor: Keep old private in radeon_pixmap_set_bo 3160 3161commit 0a84f2760eccf07781ae6c3f55d2d246c155c9ce 3162Author: Alex Deucher <alexander.deucher@amd.com> 3163Date: Mon May 13 17:06:00 2013 -0400 3164 3165 radeon: add Hainan pci ids 3166 3167 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3168 3169commit 6cee1eaebfcef2b2d2c51ec30c31260fd98b564e 3170Author: Alex Deucher <alexander.deucher@amd.com> 3171Date: Mon May 13 17:01:30 2013 -0400 3172 3173 radeon: add hainan chip family 3174 3175 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3176 3177commit 8a9e9b21cbb44c23d37bf18c983e2c3f0fe33e87 3178Author: Michel Dänzer <michel.daenzer@amd.com> 3179Date: Thu Apr 25 17:10:12 2013 +0200 3180 3181 glamor: Fix build against xserver < 1.13 3182 3183 Those didn't define CREATE_PIXMAP_USAGE_SHARED. 3184 3185commit 6216b5aa3793abe71b864170bc9d4d745f027328 3186Author: Alex Deucher <alexander.deucher@amd.com> 3187Date: Thu Apr 25 14:19:37 2013 -0400 3188 3189 radeon: add new richland pci ids 3190 3191 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3192 3193commit fd5903f15b36555bf1ccc1c3865c9fffd499128a 3194Author: Alex Deucher <alexander.deucher@amd.com> 3195Date: Thu Apr 25 14:17:51 2013 -0400 3196 3197 radeon: add new SI pci ids 3198 3199 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3200 3201commit b2826580e4d246d45871b413e4deecb070162577 3202Author: Jerome Glisse <jglisse@redhat.com> 3203Date: Mon Apr 8 13:45:18 2013 -0400 3204 3205 radeonsi: add 2d tiling support 3206 3207 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 3208 3209commit 6e74aacc5e5da3b51744153dad1645caa6ea4ce3 3210Author: Michel Dänzer <michel.daenzer@amd.com> 3211Date: Wed Mar 27 11:57:29 2013 +0100 3212 3213 glamor: Only enable for depth >= 24 3214 3215 It can't handle lower depths yet. 3216 3217 Reported-by: Igor Vagulin <igor.vagulin@gmail.com> 3218 3219commit 33d8408eec806355c2e55726679ec50ef3b769f1 3220Author: Michel Dänzer <michel.daenzer@amd.com> 3221Date: Tue Mar 12 19:03:06 2013 +0100 3222 3223 glamor: Enable by default on SI 3224 3225 We should now handle its major initialization failure cases gracefully. 3226 3227 Reviewed-by: Jerome Glisse <jglisse@redhat.com> 3228 3229commit ad1d8874fab30257ba8ec8bee8ffa42a8fa9b5c0 3230Author: Michel Dänzer <michel.daenzer@amd.com> 3231Date: Fri Mar 15 18:28:27 2013 +0100 3232 3233 glamor: Initial PRIME pixmap sharing hooks. 3234 3235 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57200 3236 3237 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3238 Tested-by: Christoph Haag <haagch.christoph@googlemail.com> 3239 Tested-by: linedot <linedot@xcpp.org> 3240 3241commit 35b384a00a0c76ea556f55787fccc95ecd51bc0c 3242Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3243Date: Thu Mar 21 15:34:21 2013 +0100 3244 3245 drmmode: handle failure of drmModeGetConnector 3246 3247 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3248 3249commit 751c7259cf23a29fdd2653881e2db63ce3cf1ca5 3250Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3251Date: Thu Mar 21 10:40:42 2013 +0100 3252 3253 Move fd cleanup to correct place. 3254 3255 Oops, just after previous commit I realized that it 3256 needs to be done from RADEONFreeRec, because when preinit fails 3257 info pointer can be null in FreeScreen. 3258 3259 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3260 3261commit 1643b8cd2dc53ed36916e11003590c7037b4ddd6 3262Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3263Date: Thu Mar 21 10:33:19 2013 +0100 3264 3265 radeon: add refcounts to fix up zaphod open/close. 3266 3267 Oops, turns out my previous commits were buggy. 3268 Adding proper refcounts will handle this correctly. 3269 3270 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3271 3272commit a4d1bc43d4d10e2ca1e512c1010e962c587d6c16 3273Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3274Date: Wed Mar 20 17:13:44 2013 +0100 3275 3276 Cleanup drmmode in CloseScreen. 3277 3278 Prevents a use-after-free. 3279 3280 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3281 3282commit 3cb3ab8704813d7c14b9e547a30e37598d70c85a 3283Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3284Date: Wed Mar 20 17:12:29 2013 +0100 3285 3286 Close drm_fd in CloseScreen. 3287 3288 This prevents a small fd leak. 3289 3290 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3291 3292commit 462d8d3b86038438ab854072e546cc7a9efacdd7 3293Author: Michel Dänzer <michel.daenzer@amd.com> 3294Date: Tue Mar 12 18:59:19 2013 +0100 3295 3296 glamor: Bail if the glamoregl module wasn't loaded early 3297 3298 Trying to initialize glamor in that case crashes. 3299 3300 Reviewed-by: Jerome Glisse <jglisse@redhat.com> 3301 3302commit 3e5350bedf50a1a140e0eba08c39ed2451fb9115 3303Author: Alex Deucher <alexander.deucher@amd.com> 3304Date: Fri Mar 8 14:07:42 2013 -0500 3305 3306 radeon: add Richland pci ids 3307 3308 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3309 3310commit bdbf766d8170f6ea6c16b9780ef5a4007730588b 3311Author: Alex Deucher <alexander.deucher@amd.com> 3312Date: Fri Mar 8 14:00:09 2013 -0500 3313 3314 radeon: mark trinity APUs with IGP flag 3315 3316 Was missing on trinity. 3317 3318 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3319 3320commit 5ac335eaae135c0e66cc4643908f41174e2f1123 3321Author: Michel Dänzer <michel.daenzer@amd.com> 3322Date: Fri Mar 8 17:37:41 2013 +0100 3323 3324 Only parse Option "AccelMethod" if glamor was enabled at build time 3325 3326 Will hopefully make it more obvious when it wasn't. 3327 3328commit f1059e3f0934ccf1ae183099dbd4de700a7c81ac 3329Author: Colin Walters <walters@verbum.org> 3330Date: Wed Jan 4 17:37:06 2012 -0500 3331 3332 autogen.sh: Implement GNOME Build API 3333 3334 http://people.gnome.org/~walters/docs/build-api.txt 3335 3336 Signed-off-by: Adam Jackson <ajax@redhat.com> 3337 3338commit 343b01c9bf35b125cd0c3df8db7c01a5fb227bda 3339Author: Alex Deucher <alexander.deucher@amd.com> 3340Date: Thu Jan 24 19:57:23 2013 -0500 3341 3342 radeon/kms: add oland pci ids 3343 3344 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3345 3346commit 9980f1e0e0a5df5fb0c791687e5c98c13089d7d1 3347Author: Alex Deucher <alexander.deucher@amd.com> 3348Date: Thu Jan 24 19:52:04 2013 -0500 3349 3350 radeon: add oland chip family 3351 3352 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3353 3354commit d89e545826969910226cb96e73b64f18e92d2064 3355Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3356Date: Wed Jan 30 17:21:05 2013 +0100 3357 3358 bump version post release 3359 3360 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3361 3362commit 21a4f3eaca98c1c54bffcdf3926c631cec9c409d 3363Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3364Date: Wed Jan 30 16:55:27 2013 +0100 3365 3366 radeon: bump version for release 3367 3368 And fixup distcheck.. 3369 3370 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 3371 3372commit 102ee4a24e4c9ba0c0b65d3ca8599dc7e0216c1e 3373Author: Alex Deucher <alexdeucher@gmail.com> 3374Date: Fri Jan 11 09:52:32 2013 -0500 3375 3376 radeon/r200: program RE_WIDTH_HEIGHT properly 3377 3378 Values are inclusive. Mesa already sets these 3379 correctly. Also bump EXA/Xv limits from 2047 to 3380 2048. 3381 3382 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3383 3384commit 47689ef8b878a56ea094e21a6090b228e67d605e 3385Author: Michel Dänzer <michel.daenzer@amd.com> 3386Date: Fri Jan 11 15:41:48 2013 +0100 3387 3388 Make radeon_glamor_create_textured_pixmap() return TRUE without USE_GLAMOR. 3389 3390 Otherwise rotation won't work when glamor support isn't built in. 3391 3392 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3393 3394commit 1a8c05fd5559be842d8a038df73de34d4b261c15 3395Author: Michel Dänzer <michel.daenzer@amd.com> 3396Date: Fri Jan 11 13:22:24 2013 +0100 3397 3398 Make CRTC shadow pixmaps usable by glamor. 3399 3400 Fixes rotation with glamor. 3401 3402 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3403 3404commit e5e22d788b103c04d3460a221a3b9dc3be69e0a4 3405Author: Michel Dänzer <michel.daenzer@amd.com> 3406Date: Thu Jan 10 17:00:16 2013 +0100 3407 3408 DRI2/r[345]xx: 16 bit depth buffers use square micro tiling. 3409 3410 Fixes piglit sanity.tests (with 3D drivers which allocate depth buffers from 3411 the X server). 3412 3413 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3414 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3415 3416commit f5d2682584c74587890a9795ece7a8917214a83f 3417Author: Michel Dänzer <michel.daenzer@amd.com> 3418Date: Thu Jan 10 17:00:15 2013 +0100 3419 3420 DRI2: Properly use buffer depth passed in by client. 3421 3422 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3423 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3424 3425commit 4e35b2f530e2ca8c7b7220cacd05c661de43d20d 3426Author: Michel Dänzer <michel.daenzer@amd.com> 3427Date: Thu Jan 10 12:10:52 2013 +0100 3428 3429 Drop support for X servers older than 1.7. 3430 3431 The main purpose is to drop DRI2 compatibility code paths which are getting 3432 awkward to deal with and at this point are probably only tested lightly if at 3433 all. xserver 1.7 was released more than 3 years ago, and 1.6.2 was already 3434 required before. 3435 3436 Mostly mechanical, there's probably potential for more cleanups. 3437 3438 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3439 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3440 3441commit 857d729eb3672443f89c0516f8226eacb37dad2b 3442Author: Dave Airlie <airlied@redhat.com> 3443Date: Mon Jan 7 14:13:33 2013 +1000 3444 3445 radeon: fix damage reporting for slave pixmaps 3446 3447 This is definitely more correct, not sure if it'll fix any bugs. 3448 3449 Signed-off-by: Dave Airlie <airlied@redhat.com> 3450 3451commit 6981a5c087165b126c15ba0025cffdba218ab652 3452Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 3453Date: Wed Dec 19 10:35:43 2012 -0500 3454 3455 DRI2: limit the swap rate when CRTC is in DPMS-off state 3456 3457 If drawable is displayed on a CRTC and relevant CRTC is in 3458 DPMS off state, defer the swap by a fixed (hard-coded) time. 3459 3460 This patch fixes a bug that caused an application to render 3461 at uncontrolled rate when CRTC goes into DPMS "off" state, 3462 thus thrashing the GPU and CPU and likely offsetting the 3463 power savings achieved by shutting off the display. 3464 3465 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3466 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 3467 3468commit 3657672207322be651cdb94a811337b7c5668c84 3469Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 3470Date: Wed Dec 19 10:35:42 2012 -0500 3471 3472 DRI2: change signature of radeon_dri2_drawable_crtc 3473 3474 Return pointer to the CRTC instead of CRTC ID and expose 3475 consider_disabled option in arguments of this function. 3476 3477 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3478 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 3479 3480commit 61d0aec40e2521488c2fe43e7a6823e5c87d94d7 3481Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 3482Date: Wed Dec 19 10:35:41 2012 -0500 3483 3484 video: add option to include disabled CRTCs in best CRTC search 3485 3486 This patch adds an option called consider_disabled to 3487 radeon_pick_best_crtc function. If this option is set 3488 and searching for best-fit CRTC yields nothing, the search 3489 will be widened to include CRTCs in DPMS "off" state. 3490 The new option is not used yet; it will be in the patches 3491 to follow. 3492 3493 v2: Do not look at the CRTC mode, we only care about its box. 3494 3495 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3496 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 3497 3498commit a60d2152e928a7011fc7c44a885a34c3cdd4f0fe 3499Author: Michel Dänzer <michel.daenzer@amd.com> 3500Date: Thu Dec 27 12:13:57 2012 +0100 3501 3502 glamor: Enable full 2D acceleration for SI as well. 3503 3504 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3505 3506commit 04ebe14fe775d18b534e89b50b90ee293ae1910e 3507Author: Michel Dänzer <michel.daenzer@amd.com> 3508Date: Thu Dec 27 11:54:27 2012 +0100 3509 3510 glamor: Release the drawable after passing to glamor_push_pixels 3511 3512 Ported from xf86-video-intel commit ca5c028c2b4d9bf02002acd484054fe427ea8d09 . 3513 3514 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3515 3516commit cee7d61b144162974238b282d76a70aa2f14cc0d 3517Author: Alex Deucher <alexdeucher@gmail.com> 3518Date: Tue Dec 18 15:49:47 2012 -0500 3519 3520 exa/Xv: fix coordinate limits on AVIVO IGPs 3521 3522 RS6xx asics are r4xx derived, but seem to have r3xx 3523 limitations as far as clipping is concerned. Spotted 3524 by Michel on IRC. 3525 3526 Fixes: 3527 https://bugs.freedesktop.org/show_bug.cgi?id=58469 3528 3529 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3530 3531commit 3c7d024c2671dac541b8f2daed55040f4fd5d62d 3532Author: Michel Dänzer <michel.daenzer@amd.com> 3533Date: Wed Dec 12 10:43:55 2012 +0100 3534 3535 Remove dead option definitions. 3536 3537 These were only used with UMS. 3538 3539 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3540 3541commit bdbd84a6c99943f4e012c92ad9cb920cbd676621 3542Author: Michel Dänzer <michel.daenzer@amd.com> 3543Date: Mon Dec 10 23:35:41 2012 +0100 3544 3545 Remove local variables total_size_bytes. 3546 3547 Not actually used for anything. Refactoring gone wrong? 3548 3549 Also remove a comment that doesn't seem to make sense anymore. 3550 3551 Stumbled over this due to a warning by clang: 3552 3553 ../../src/radeon_kms.c:1562:6: warning: variable 'total_size_bytes' is uninitialized when used here [-Wuninitialized] 3554 total_size_bytes += (64 * 4 * 64); 3555 ^~~~~~~~~~~~~~~~ 3556 ../../src/radeon_kms.c:1557:25: note: initialize the variable 'total_size_bytes' to silence this warning 3557 int total_size_bytes; 3558 ^ 3559 = 0 3560 3561 Reviewed-by: Alex Deucher <aleander.deucher@amd.com> 3562 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3563 3564commit cf0a81547bcde32bdd2b080c0a546a49dd2eb628 3565Author: Michel Dänzer <michel.daenzer@amd.com> 3566Date: Tue Dec 11 11:48:33 2012 +0100 3567 3568 Remove unused local variable 'path'. 3569 3570 ../../src/radeon_probe.c: In function 'radeon_platform_probe': 3571 ../../src/radeon_probe.c:270:11: error: unused variable 'path' [-Werror=unused-variable] 3572 3573 Reviewed-by: Alex Deucher <aleander.deucher@amd.com> 3574 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3575 3576commit d268038a5bdd1655f7533c8dd2edf74fbc82bbee 3577Author: Michel Dänzer <michel.daenzer@amd.com> 3578Date: Mon Jul 2 09:07:14 2012 +0200 3579 3580 Remove dead code flagged by gcc -Wunused-but-set-variable. 3581 3582 Reviewed-by: Alex Deucher <aleander.deucher@amd.com> 3583 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3584 3585commit 793e1b0ea579cce7489aee57a95f20384d6ecce8 3586Author: Alex Deucher <alexdeucher@gmail.com> 3587Date: Thu Dec 6 09:21:19 2012 -0500 3588 3589 radeon: fix ring count in R300PrepareTexturedVideo 3590 3591 Fixes: 3592 https://bugs.freedesktop.org/show_bug.cgi?id=31364 3593 3594 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3595 3596commit f402805b22e4f4f735924d59c69474b24e623a25 3597Author: Jerome Glisse <jglisse@redhat.com> 3598Date: Fri Nov 30 10:07:51 2012 -0500 3599 3600 radeon: avoid copying over itself ddx fb 3601 3602 In some rare case it seems that the old fb could already be the 3603 ddx fb. As copying from the same bo will trigger a ttm deadlck 3604 detection which will result in infinite kernel/userspace loop. 3605 Avoid doing any copy as anyway it's useless. 3606 3607 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 3608 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 3609 3610commit 60cd6ceaf44b506433aebf6b3a639a17604dfddd 3611Author: Alex Deucher <alexdeucher@gmail.com> 3612Date: Wed Nov 21 18:42:56 2012 -0500 3613 3614 radeon: add new SI pci id 3615 3616 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3617 3618commit 53af6aa99dfb67b5223c28d5129a79694165048d 3619Author: Jerome Glisse <jglisse@redhat.com> 3620Date: Wed Nov 14 12:15:47 2012 -0500 3621 3622 radeon: avoid segfault when pixmap exceed GPU capabilities 3623 3624 We might get a request for a pixmap/drawable that is too big 3625 for GPU capabilities in the dri2 get buffer path. In face of 3626 such things just return NULL to dri2 get buffer request. The 3627 GL driver should then use something like transparent black to 3628 avoid something too ugly on the screen. 3629 3630 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 3631 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 3632 3633commit 71eeb0833489ece745f1e5dc41379ec63161ee81 3634Author: Alex Deucher <alexdeucher@gmail.com> 3635Date: Tue Nov 6 09:43:55 2012 -0500 3636 3637 bump verion post release 3638 3639 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3640 3641commit cf377c991f4c1cd4eee1a8530ba9fdf6f9f63301 3642Author: Alex Deucher <alexdeucher@gmail.com> 3643Date: Tue Nov 6 09:12:39 2012 -0500 3644 3645 radeon: bump version for release 3646 3647 7.0.0, RIP UMS 3648 3649 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3650 3651commit 20bfc652ce40008ea561db2984bccf137409c7fd 3652Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 3653Date: Mon Sep 24 17:19:09 2012 -0400 3654 3655 radeon/radeon_platform_probe: fix Zaphod mode breakage 3656 3657 Using radeon_platform_probe function breaks the Zaphod mode because 3658 it attempts to call xf86AddEntityToScreen multiple times, but nobody 3659 calls xf86SetEntityShared prior to that. Consequently, calls for all 3660 but first device instance fail. 3661 3662 Prior to introduction of platform bus, the logic was that the Probe 3663 function would make the entity sharable, which would cause Xserver 3664 to later make it shared prior to adding it to screen. With the 3665 platform bus loading, add to screen happens in the probe 3666 function so we have to make it shared there. 3667 3668 v2: do not make the entity shared if it was not previously marked 3669 sharable. Should fix: 3670 3671 https://bugs.freedesktop.org/show_bug.cgi?id=56680 3672 https://bugs.freedesktop.org/show_bug.cgi?id=56663 3673 3674 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3675 3676commit b2807b53083af5db0b85645666b0127be16623c7 3677Author: Alex Deucher <alexdeucher@gmail.com> 3678Date: Fri Nov 2 17:57:18 2012 -0400 3679 3680 man: note that 2D tiling requires Mesa 9.0 or newer 3681 3682 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3683 3684commit 1038e29162d869e2ce37f9b8f5f7c8ea37c3e53d 3685Author: Alex Deucher <alexdeucher@gmail.com> 3686Date: Fri Nov 2 10:04:46 2012 -0400 3687 3688 Revert "radeon/radeon_platform_probe: fix Zaphod mode breakage" 3689 3690 This reverts commit 39eac3104c2f08b4d78aab3f88fd104301eb4711. 3691 3692 This breaks non-zaphod configurations with xserver 1.13. 3693 See: 3694 https://bugs.freedesktop.org/show_bug.cgi?id=56680 3695 https://bugs.freedesktop.org/show_bug.cgi?id=56663 3696 3697commit d748894c84e891662cde6d7a5748f74a5aa5175e 3698Author: Alex Deucher <alexdeucher@gmail.com> 3699Date: Fri Nov 2 08:50:16 2012 -0400 3700 3701 man: update man pages for trinity (ARUBA) 3702 3703 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3704 3705commit 6740e8e90b934611c3bb0f9e8732d66e25b8c727 3706Author: Andreas Boll <andreas.boll.dev@gmail.com> 3707Date: Fri Nov 2 13:30:44 2012 +0100 3708 3709 radeon/man: ColorTiling2D is by default on for r6xx-cayman 3710 3711 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 3712 3713commit 39eac3104c2f08b4d78aab3f88fd104301eb4711 3714Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 3715Date: Mon Sep 24 17:41:32 2012 -0400 3716 3717 radeon/radeon_platform_probe: fix Zaphod mode breakage 3718 3719 Using radeon_platform_probe function breaks the Zaphod mode because 3720 it attempts to call xf86AddEntityToScreen multiple times, but nobody 3721 calls xf86SetEntityShared prior to that. Consequently, calls for all 3722 but first device instance fail. 3723 3724 Prior to introduction of platform bus, the logic was that the Probe 3725 function would make the entity sharable, which would cause Xserver 3726 to later make it shared prior to adding it to screen. With the 3727 platform bus loading, add to screen happens in the probe 3728 function so we have to make it shared there. 3729 3730 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3731 3732commit 2e96aa641bc8a5c263a5b643961383cdf530d044 3733Author: Marek Olšák <maraeo@gmail.com> 3734Date: Sun Oct 28 12:50:49 2012 -0400 3735 3736 radeon: turn on 2D tiling by default on r6xx-cayman 3737 3738 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3739 3740commit e830c039479539735d0836082abef24f6841c803 3741Author: Marek Olšák <maraeo@gmail.com> 3742Date: Sun Oct 28 12:49:15 2012 -0400 3743 3744 radeon: log if 2D tiling is enabled 3745 3746 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3747 3748commit ce1b745dcb60dc516ad999756240b78e72a7aa54 3749Author: Alex Deucher <alexander.deucher@amd.com> 3750Date: Tue Oct 16 13:05:46 2012 -0400 3751 3752 radeon: add new pci ids for SI 3753 3754 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3755 3756commit e3519be285bdab50ef2193f98826a351a4414fae 3757Author: Dave Airlie <airlied@redhat.com> 3758Date: Tue Oct 16 16:09:23 2012 +1000 3759 3760 radeon/dri2: fix the window->pixmap math properly. 3761 3762 Okay kwin did things different, and I really hope this time this is the 3763 correct transformation operation. It works here with gears under kwin 3764 and under mutter and under xcompmgr. 3765 3766 Signed-off-by: Dave Airlie <airlied@redhat.com> 3767 3768commit bd9e2c064d93afbe4ab84febb2a43ae6a95ab17b 3769Author: Dave Airlie <airlied@redhat.com> 3770Date: Thu Oct 11 14:36:11 2012 +1000 3771 3772 radeon/dri2: fix damage reporting on pageflip 3773 3774 when you use MPX or slave usb devices, the damage posting was incorrect, 3775 and resulted in misrendered boxes on the screen. 3776 3777 Fix it by reporting damage correctly. 3778 3779 Signed-off-by: Dave Airlie <airlied@redhat.com> 3780 3781commit e8cb0b721e6ea251f85c799ca0563bfa59a2d37c 3782Author: Adam Jackson <ajax@redhat.com> 3783Date: Wed Sep 26 09:27:54 2012 -0400 3784 3785 Remove mibstore.h 3786 3787 Signed-off-by: Adam Jackson <ajax@redhat.com> 3788 3789commit 8637f772347c958fa3beea6c0dc9c22255db70e5 3790Author: Michel Dänzer <michel.daenzer@amd.com> 3791Date: Thu Sep 13 19:50:35 2012 +0200 3792 3793 glamor: Force acceleration for DRI2 copies. 3794 3795 Improves performance of non-pageflipped 3D apps on SI. 3796 3797 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3798 3799commit 40b87f0e9db2642ca8a70e994fb4d3742e199640 3800Author: Michel Dänzer <michel.daenzer@amd.com> 3801Date: Wed Sep 12 18:44:10 2012 +0200 3802 3803 glamor: Fix DRI2 crash introduced by PRIME changes. 3804 3805 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3806 3807commit 646ffd49d2cf0d3130301e8ad2f49f620bd9ef1b 3808Author: Dave Airlie <airlied@redhat.com> 3809Date: Mon Sep 10 15:31:06 2012 +1000 3810 3811 radeon: fix build against old servers 3812 3813 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=54711 3814 3815 Reported-by: wolput@onsneteindhoven.nl 3816 Signed-off-by: Dave Airlie <airlied@redhat.com> 3817 3818commit 7c7f2775604da61537c24c249884aaa220e06e13 3819Author: Dave Airlie <airlied@redhat.com> 3820Date: Mon Sep 10 11:27:45 2012 +1000 3821 3822 radeon: don't fail to load if we have no modes and gpu driver. 3823 3824 This ensures radeon loads on output less GPUs as a GPU driver. 3825 3826 There are server bugs that also need to be fixed. 3827 3828 Signed-off-by: Dave Airlie <airlied@redhat.com> 3829 3830commit 7fe16dd037ac688726869816888bd77fe4356f8e 3831Author: Dave Airlie <airlied@gmail.com> 3832Date: Wed Sep 5 10:41:51 2012 +1000 3833 3834 radeon: fix typo in glamor paths 3835 3836 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=54519 3837 3838 Reported-By: scaty@dcinformatique.com 3839 Signed-off-by: Dave Airlie <airlied@redhat.com> 3840 3841commit cdc985e9190ce54dc6274ebc0e87ab954b15bb0d 3842Author: Dave Airlie <airlied@gmail.com> 3843Date: Tue Sep 4 18:18:50 2012 +1000 3844 3845 radeon: fix build against older servers 3846 3847 Another silly one. 3848 3849 Signed-off-by: Dave Airlie <airlied@redhat.com> 3850 3851commit bd0e9c51407cb99f1fe4f0acfedc37ee0c354040 3852Author: Dave Airlie <airlied@gmail.com> 3853Date: Tue Sep 4 18:17:06 2012 +1000 3854 3855 radeon/dri2: fix build against older servers. 3856 3857 Older servers won't have DRI2UpdatePrime, so fix build. 3858 3859 bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=54433 3860 Signed-off-by: Dave Airlie <airlied@redhat.com> 3861 3862commit 75c1f62f44d7b3709c1fa103d59733866967b841 3863Author: Dave Airlie <airlied@redhat.com> 3864Date: Mon Sep 3 13:09:02 2012 +1000 3865 3866 radeon: add capability and provider support 3867 3868 Signed-off-by: Dave Airlie <airlied@redhat.com> 3869 3870commit ca8b61e916998458203dfa4b9d0c997fa7d90ba9 3871Author: Dave Airlie <airlied@redhat.com> 3872Date: Mon Sep 3 13:08:37 2012 +1000 3873 3874 radeon: allocated shared pixmaps as untiled gtt. 3875 3876 Signed-off-by: Dave Airlie <airlied@redhat.com> 3877 3878commit 8c1bf9d8fe3948b72795984e625ef46b2f0bf654 3879Author: Dave Airlie <airlied@redhat.com> 3880Date: Mon Sep 3 13:08:05 2012 +1000 3881 3882 radeon: add pixmap sharing hooks. 3883 3884 This hooks into EXA and the dirty tracking to add sharing and output 3885 offload support. 3886 3887 Signed-off-by: Dave Airlie <airlied@redhat.com> 3888 3889commit 3add8df8122697acfe126d4857f3946ce44b8305 3890Author: Dave Airlie <airlied@redhat.com> 3891Date: Mon Sep 3 13:07:07 2012 +1000 3892 3893 radeon: add dri2 offload support. 3894 3895 This adds the new dri2 interfaces required for radeon to be offloaded to. 3896 3897 Signed-off-by: Dave Airlie <airlied@redhat.com> 3898 3899commit f7502a11c8ef9c453ceb40d26109977116df88c2 3900Author: Dave Airlie <airlied@redhat.com> 3901Date: Mon Sep 3 12:56:56 2012 +1000 3902 3903 radeon: add shared support to pixmaps. 3904 3905 this just adds the interface and shared support to the pixmap. 3906 3907 Signed-off-by: Dave Airlie <airlied@redhat.com> 3908 3909commit f71139a2afe8fffb628331402bf829a6d67c9fff 3910Author: Dave Airlie <airlied@redhat.com> 3911Date: Mon Sep 3 12:37:22 2012 +1000 3912 3913 radeon: add platform bus loading support. 3914 3915 This allows the radeon driver to be loaded via the platform bus mechanism. 3916 3917 Signed-off-by: Dave Airlie <airlied@redhat.com> 3918 3919commit 9911e72ea79e000ee0492d19a434aa98b83b5066 3920Author: Dave Airlie <airlied@redhat.com> 3921Date: Fri Aug 17 17:02:51 2012 +1000 3922 3923 radeon: damage full screen after pageflip 3924 3925 in order for offload to work we need to cause a full 3926 screen damage to be registered after pageflip. 3927 3928 Signed-off-by: Dave Airlie <airlied@redhat.com> 3929 3930commit 0873b8cef21abf1e7286731758f0db1e7f2e7c97 3931Author: Dave Airlie <airlied@redhat.com> 3932Date: Fri Aug 3 16:45:50 2012 +1000 3933 3934 ati: fix stub driver loader to load all drivers properly 3935 3936 Signed-off-by: Dave Airlie <airlied@redhat.com> 3937 3938commit cdf522f5992adf7b76f9153c777e6dddcf7758a6 3939Author: Adam Jackson <ajax@redhat.com> 3940Date: Wed Aug 15 13:38:28 2012 -0400 3941 3942 Implement ->driverFunc 3943 3944 Copied from fbdev, makes it so we can run without iopl. 3945 3946 Signed-off-by: Adam Jackson <ajax@redhat.com> 3947 3948commit 763f387ccac2399f1d58ec112c8bed45196bea5e 3949Author: Adam Jackson <ajax@redhat.com> 3950Date: Wed Aug 15 12:49:10 2012 -0400 3951 3952 Don't bother to include vbe.h 3953 3954 The only VBE code here went away when we dropped UMS. 3955 3956 Signed-off-by: Adam Jackson <ajax@redhat.com> 3957 3958commit aabcf5ea30195274c79ae94c0fe6edc965f5c123 3959Author: Tomas Chvatal <tchvatal@suse.cz> 3960Date: Mon Aug 13 22:58:18 2012 +0200 3961 3962 Add switch for udev to disable automagic detection. 3963 3964 Signed-off-by: Tomas Chvatal <tchvatal@suse.cz> 3965 3966commit aef90993621331f01d17b95a751d4924d77eed0c 3967Author: Alex Deucher <alexdeucher@gmail.com> 3968Date: Mon Aug 6 11:00:45 2012 -0400 3969 3970 radeon: add some new SI pci ids 3971 3972 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3973 3974commit 6ef1ad6a46348d3aecd8d1f5e94431ca2298853c 3975Author: Michel Dänzer <michel.daenzer@amd.com> 3976Date: Fri Jul 13 11:15:25 2012 +0200 3977 3978 Deal more gracefully with DRI2 being unavailable at build or run time. 3979 3980 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3981 3982commit ef8a404391036d8aa814dbda2407c789b8a64b92 3983Author: Michel Dänzer <michel.daenzer@amd.com> 3984Date: Thu Jul 5 20:14:48 2012 +0200 3985 3986 Initial SI support. 3987 3988 Defaults to shadowfb. 3D acceleration is available with glamor. 2D 3989 acceleration is disabled until the radeonsi driver can handle glamor's 3990 shaders. 3991 3992 v2: add chip flags (Alex Deucher) 3993 3994 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3995 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3996 3997commit e9edd2f5002c642b59f028b3ec076d604ae8ce9d 3998Author: Michel Dänzer <michel.daenzer@amd.com> 3999Date: Wed Jun 20 08:40:07 2012 +0200 4000 4001 Initial glamor support. 4002 4003 Enable at build time with --enable-glamor and runtime with 4004 4005 Option "AccelMethod" "glamor" 4006 4007 The most notable lack of functionality is XVideo. Use something like VDPAU for 4008 now. 4009 4010 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4011 4012commit 9eac8021f3d33a63156f9f5d43a220e88bb3f8db 4013Author: Michel Dänzer <michel.daenzer@amd.com> 4014Date: Wed Jun 27 14:48:47 2012 +0200 4015 4016 EXA: Factor out pixmap BO allocation into a helper function. 4017 4018 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4019 4020commit 060c7836e7f7777bacca4e23f57c5985beab33bc 4021Author: Michel Dänzer <michel.daenzer@amd.com> 4022Date: Wed Jul 4 11:49:18 2012 +0200 4023 4024 Some cosmetic fixups for the radeon manpage. 4025 4026 ATI -> ATI/AMD 4027 PCIE -> PCIe 4028 4029 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4030 4031commit c3a9f64fe40c78331b9340c3832d344f38b11564 4032Author: Michel Dänzer <michel.daenzer@amd.com> 4033Date: Wed Jul 4 11:46:30 2012 +0200 4034 4035 Drop UMS specific parts from the radeon manpage. 4036 4037 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4038 4039commit 5ebebfdb4fadada157d1369f35a8610c0fa72e02 4040Author: Dave Airlie <airlied@redhat.com> 4041Date: Fri Jun 29 15:03:43 2012 +0100 4042 4043 radeon: fix conflict with fix in master. 4044 4045 Signed-off-by: Dave Airlie <airlied@redhat.com> 4046 4047commit ff12ddf613ad42e4566d6e10fc6cf33a4e445736 4048Author: Dave Airlie <airlied@redhat.com> 4049Date: Fri Jun 29 15:00:44 2012 +0100 4050 4051 configure: bump to 6.99.99 4052 4053 this is in prep for 7.0 release. 4054 4055 Signed-off-by: Dave Airlie <airlied@redhat.com> 4056 4057commit 50689ec8dbd4a68527b2ac16cecac298b8d441d0 4058Merge: 6a60fcd ae682a6 4059Author: Dave Airlie <airlied@redhat.com> 4060Date: Fri Jun 29 14:59:47 2012 +0100 4061 4062 Merge remote-tracking branch 'origin/kms-only' 4063 4064 This merges the removal of the UMS code and subsequent simplifications. 4065 4066 Hell yes. 4067 4068 * origin/kms-only: (24 commits) 4069 radeon: fix radeonchipsets properly 4070 radeon: fix RADEONChipsets 4071 radeon: migrate remainder of radeon_driver.c to rest of driver 4072 radeon: avoid including large static struct twice. 4073 radeon: drop some more unused macros/inlines 4074 radeon: drop another unused struct member. 4075 radeon: more unused stuff 4076 radeon: drop cardType shouldn't matter to userspace. 4077 radeon: drop radeonGetPixmapOffset 4078 radeon: move more functions into texture video 4079 radeon: drop legacy memory, merge into radeon video 4080 radeon: drop unused txoffset vars. 4081 radeon: drop vb_mc_addr, not needed anymore 4082 radeon: drop offset member of r600 accel object 4083 radeon: move macros into radeon.h 4084 radeon: drop more wrapper macros. 4085 radeon: further macro cleanups. 4086 radeon: make exa copy funcs static. 4087 radeon: drop FUNC_NAME macro and ONCE_ONLY stuff 4088 radeon: drop radeon_commonfuncs. 4089 ... 4090 4091 Conflicts: 4092 configure.ac 4093 4094commit 6a60fcdf060a0f553d8d4f7939c2a05aa04bedfa 4095Author: Dave Airlie <airlied@redhat.com> 4096Date: Fri Jun 29 14:57:58 2012 +0100 4097 4098 configure: bump version after release. 4099 4100 Signed-off-by: Dave Airlie <airlied@redhat.com> 4101 4102commit 008eb209d506ed7195fd544dfa91be9783946fd5 4103Author: Dave Airlie <airlied@redhat.com> 4104Date: Fri Jun 29 14:46:55 2012 +0100 4105 4106 configure: bump version number to 6.14.6 pre-release 4107 4108 Signed-off-by: Dave Airlie <airlied@redhat.com> 4109 4110commit 72fea2635a56c4f572c07fc50fc58f703a21eb4f 4111Author: Dave Airlie <airlied@redhat.com> 4112Date: Fri Jun 29 14:46:17 2012 +0100 4113 4114 radeon: require libdrm 2.4.36 for KMS support. 4115 4116 This is due to some commits to the surface manager that fix bugs. 4117 4118 Signed-off-by: Dave Airlie <airlied@redhat.com> 4119 4120commit 179b035835bfa99283d44ba2d5ee17e686196535 4121Author: Michel Dänzer <michel.daenzer@amd.com> 4122Date: Wed Jun 27 19:26:51 2012 +0200 4123 4124 Fix up displayWidth vs. virtualX confusion in radeon_setup_kernel_mem(). 4125 4126 It was using the pitch (displayWidth) for the virtual width (virtualX). This 4127 prevented using page flipping in some cases, as displayWidth was already 4128 overaligned for virtualX, so the DRI2 front and back buffers ended up having 4129 a different pitch. 4130 4131 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4132 Reviewed-by: Jerome Glisse <jglisse@redhat.com> 4133 4134commit 5dc5012320536f7b1e86a81ed3b823e3626828e1 4135Author: Dave Airlie <airlied@redhat.com> 4136Date: Mon Jun 25 09:23:04 2012 +0100 4137 4138 radeon: dixPrivateKeyRegistered is only in server 1.9 4139 4140 just fallback to the old behaviour on older servers. 4141 4142 should fix: https://bugs.freedesktop.org/show_bug.cgi?id=51388 4143 Review-over-irc: Michel Dänzer <michel.daenzer@amd.com> 4144 Signed-off-by: Dave Airlie <airlied@redhat.com> 4145 4146commit ae682a6a1781ab023ab36e3154d6b3e88da63d64 4147Author: Dave Airlie <airlied@redhat.com> 4148Date: Mon Jun 18 13:11:31 2012 +0100 4149 4150 radeon: fix radeonchipsets properly 4151 4152commit 6514bcdc0fc199b4f3a8b4273bd29e0fe255d232 4153Author: Dave Airlie <airlied@redhat.com> 4154Date: Mon Jun 18 13:02:03 2012 +0100 4155 4156 radeon: fix RADEONChipsets 4157 4158 reported on irc by okias. 4159 4160 Signed-off-by: Dave Airlie <airlied@redhat.com> 4161 4162commit dd4ab5aba9047d522dbcfe8a341c368312e1a471 4163Author: Dave Airlie <airlied@redhat.com> 4164Date: Sat Jun 16 08:47:32 2012 +0100 4165 4166 radeon: migrate remainder of radeon_driver.c to rest of driver 4167 4168 This helps make a few more things static and the driver generally 4169 smaller. 4170 4171 Signed-off-by: Dave Airlie <airlied@redhat.com> 4172 4173commit a19237fad2492d70777167c631f4e6e1fae0e908 4174Author: Dave Airlie <airlied@redhat.com> 4175Date: Sat Jun 16 08:19:35 2012 +0100 4176 4177 radeon: avoid including large static struct twice. 4178 4179 Just extern this in the one file that needs access. 4180 4181 Signed-off-by: Dave Airlie <airlied@redhat.com> 4182 4183commit c8a5c5a0dd9adc46b03b898c9c1719dbbde05ba1 4184Author: Dave Airlie <airlied@redhat.com> 4185Date: Fri Jun 15 19:52:14 2012 +0100 4186 4187 radeon: drop some more unused macros/inlines 4188 4189 These are no longer used anywhere. 4190 4191 Signed-off-by: Dave Airlie <airlied@redhat.com> 4192 4193commit 3097f273443b5b8fce097e2b1e09a17b02d06a40 4194Author: Dave Airlie <airlied@redhat.com> 4195Date: Fri Jun 15 19:45:15 2012 +0100 4196 4197 radeon: drop another unused struct member. 4198 4199 Signed-off-by: Dave Airlie <airlied@redhat.com> 4200 4201commit cb97e75acab84b67b7b7358860788638efc9b344 4202Author: Dave Airlie <airlied@redhat.com> 4203Date: Fri Jun 15 19:38:45 2012 +0100 4204 4205 radeon: more unused stuff 4206 4207 Remove all CurrentLayout stuff. 4208 4209 Signed-off-by: Dave Airlie <airlied@redhat.com> 4210 4211commit 81593e7deb688fa3108a0589c1418459ec0df4de 4212Author: Dave Airlie <airlied@redhat.com> 4213Date: Fri Jun 15 18:52:52 2012 +0100 4214 4215 radeon: drop cardType shouldn't matter to userspace. 4216 4217 Signed-off-by: Dave Airlie <airlied@redhat.com> 4218 4219commit 997b0dddfeb180043ece5eff0ac0d6a1a751f9b5 4220Author: Dave Airlie <airlied@redhat.com> 4221Date: Fri Jun 15 18:43:01 2012 +0100 4222 4223 radeon: drop radeonGetPixmapOffset 4224 4225 no longer needed. 4226 4227 Signed-off-by: Dave Airlie <airlied@redhat.com> 4228 4229commit 639061edb3ac35a24e4d6633c9b6384434dcac64 4230Author: Dave Airlie <airlied@redhat.com> 4231Date: Fri Jun 15 18:38:48 2012 +0100 4232 4233 radeon: move more functions into texture video 4234 4235 make more thing static. 4236 4237 Signed-off-by: Dave Airlie <airlied@redhat.com> 4238 4239commit 85711a0a6a553232b603dbbb2b2d234c35391578 4240Author: Dave Airlie <airlied@redhat.com> 4241Date: Fri Jun 15 17:26:50 2012 +0100 4242 4243 radeon: drop legacy memory, merge into radeon video 4244 4245 This merges these two functions into textured video code. 4246 4247 Signed-off-by: Dave Airlie <airlied@redhat.com> 4248 4249commit 445ca1d66ce7b37762731fe67407cc1b978bb129 4250Author: Dave Airlie <airlied@redhat.com> 4251Date: Fri Jun 15 17:13:48 2012 +0100 4252 4253 radeon: drop unused txoffset vars. 4254 4255 These vars aren't being used anymore. 4256 4257 Signed-off-by: Dave Airlie <airlied@redhat.com> 4258 4259commit 832244de9e6d7aa1b8fbeb29c9a6a86923744d60 4260Author: Dave Airlie <airlied@redhat.com> 4261Date: Fri Jun 15 17:09:00 2012 +0100 4262 4263 radeon: drop vb_mc_addr, not needed anymore 4264 4265 This field is totally unused now, so drop it. 4266 4267 Signed-off-by: Dave Airlie <airlied@redhat.com> 4268 4269commit 6559b6e28bf111b64c3aeb5ec1260acfeb3e0b7f 4270Author: Dave Airlie <airlied@redhat.com> 4271Date: Fri Jun 15 17:05:54 2012 +0100 4272 4273 radeon: drop offset member of r600 accel object 4274 4275 This is always 0 now. 4276 4277 Signed-off-by: Dave Airlie <airlied@redhat.com> 4278 4279commit 857179e538c9752fab239a21312d0fcc80022151 4280Author: Dave Airlie <airlied@redhat.com> 4281Date: Fri Jun 15 16:55:23 2012 +0100 4282 4283 radeon: move macros into radeon.h 4284 4285 Drop separate header file, there isn't many separate macros now. 4286 4287 Signed-off-by: Dave Airlie <airlied@redhat.com> 4288 4289commit 148e889149f5ecc3dc2473f2a9e01572eeb037d3 4290Author: Dave Airlie <airlied@redhat.com> 4291Date: Fri Jun 15 16:52:24 2012 +0100 4292 4293 radeon: drop more wrapper macros. 4294 4295 Signed-off-by: Dave Airlie <airlied@redhat.com> 4296 4297commit ad8629e3323efe484dd1421d94dd934e1bfafe56 4298Author: Dave Airlie <airlied@redhat.com> 4299Date: Fri Jun 15 16:49:13 2012 +0100 4300 4301 radeon: further macro cleanups. 4302 4303 drop some more of the wrapper macros. 4304 4305 Signed-off-by: Dave Airlie <airlied@redhat.com> 4306 4307commit 96644f96f1a2df1d7d1ca8cbdc9a105affdb4f32 4308Author: Dave Airlie <airlied@redhat.com> 4309Date: Fri Jun 15 16:34:53 2012 +0100 4310 4311 radeon: make exa copy funcs static. 4312 4313 These are no longer used outside this area. 4314 4315 Signed-off-by: Dave Airlie <airlied@redhat.com> 4316 4317commit 83904b820b81d226940ec6d663246f243f92e7ba 4318Author: Dave Airlie <airlied@redhat.com> 4319Date: Fri Jun 15 16:30:41 2012 +0100 4320 4321 radeon: drop FUNC_NAME macro and ONCE_ONLY stuff 4322 4323 Drop all the CP vs MMIO macros that name functions. 4324 4325 Signed-off-by: Dave Airlie <airlied@redhat.com> 4326 4327commit b422d8085aab4e01a1ac4ffb247c73fe51aafa0d 4328Author: Dave Airlie <airlied@redhat.com> 4329Date: Fri Jun 15 16:14:54 2012 +0100 4330 4331 radeon: drop radeon_commonfuncs. 4332 4333 modify all the macros and drop it into radeon_accel.c 4334 4335 Signed-off-by: Dave Airlie <airlied@redhat.com> 4336 4337commit 29dd3f84eecd2892a438f004403521de8986a91d 4338Author: Dave Airlie <airlied@redhat.com> 4339Date: Fri Jun 15 16:11:03 2012 +0100 4340 4341 radeon: misc cleanups. 4342 4343 Signed-off-by: Dave Airlie <airlied@redhat.com> 4344 4345commit 678c1f3425c0fa7e98795c8e351da77d04652e02 4346Author: Dave Airlie <airlied@redhat.com> 4347Date: Fri Jun 15 16:06:27 2012 +0100 4348 4349 make pci chipsets static 4350 4351 Signed-off-by: Dave Airlie <airlied@redhat.com> 4352 4353commit 95cbbe999418c72dc407069d31e759d70f49ab16 4354Author: Dave Airlie <airlied@redhat.com> 4355Date: Fri Jun 15 16:01:53 2012 +0100 4356 4357 drop RING_LOCALS/ACCEL_PREAMBLE 4358 4359 these were unused now. 4360 4361 Signed-off-by: Dave Airlie <airlied@redhat.com> 4362 4363commit 18d5ae3bd9075ac1a2ee21b071ac133e2e634b62 4364Author: Dave Airlie <airlied@redhat.com> 4365Date: Fri Jun 15 10:05:03 2012 +0100 4366 4367 radeon: drop all UMS/DRI1/XAA/overlay support. 4368 4369 This overhauls the radeon driver and removes all the old UMS-only code, 4370 it drops all the UMS, DRI1, XAA, overlay Xv, video capture, tv tuners 4371 4372 There are probably a lot more cleanups that will fall out of this afterwards. 4373 4374 So far this is compile/build tested. 4375 4376 Signed-off-by: Dave Airlie <airlied@redhat.com> 4377 4378commit 248e912c487636d7352cfad43c03fc9f19fc2215 4379Author: Michel Dänzer <michel.daenzer@amd.com> 4380Date: Fri Jun 8 12:18:48 2012 +0200 4381 4382 UMS: Fix CRTC DPMS state check. 4383 4384 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4385 4386commit 58883711136fa6671d4f3250d4ee973e2953ebb4 4387Author: Alex Deucher <alexdeucher@gmail.com> 4388Date: Thu Jun 7 11:47:21 2012 -0400 4389 4390 configure: bump version post release 4391 4392 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4393 4394commit 3a9103f855c67bbcd7b97ee2e96836353e0d789d 4395Author: Alex Deucher <alexdeucher@gmail.com> 4396Date: Thu Jun 7 11:27:08 2012 -0400 4397 4398 configure: bump for release 4399 4400 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4401 4402commit f1693ee09e9b251eb75e3e5fe0ee7699fec6fb98 4403Author: Alex Deucher <alexdeucher@gmail.com> 4404Date: Thu Jun 7 11:23:31 2012 -0400 4405 4406 configure: bump libdrm_radeon requirement 4407 4408 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4409 4410commit 77a056c1725c9c44a328ed324687bdf35144e9e2 4411Author: Michel Dänzer <michel.daenzer@amd.com> 4412Date: Wed Jun 6 12:29:37 2012 +0200 4413 4414 Don't page-flip or wait on a CRTC while we're VT-switched away. 4415 4416 We don't know what the CRTC state is. 4417 4418 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4419 4420commit 1d9f1eb7e65c57c52759ccbfa51e8bdd5a10acac 4421Author: Michel Dänzer <michel.daenzer@amd.com> 4422Date: Wed Jun 6 12:16:08 2012 +0200 4423 4424 Don't wait on a CRTC which has been disabled via DPMS (bug #49761). 4425 4426 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=49761 . 4427 4428 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4429 4430commit 4603285aa8efaf2614f15a38254465cec2075f11 4431Author: Alex Deucher <alexdeucher@gmail.com> 4432Date: Tue Jun 5 09:48:59 2012 -0400 4433 4434 radeon: add new PCI ids 4435 4436 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4437 4438commit 9307609420b4b209767d2057b4803bcb16cc1455 4439Author: Dave Airlie <airlied@redhat.com> 4440Date: Sat Jun 2 17:16:34 2012 +0100 4441 4442 radeon: improve smooth startup fallbacks. 4443 4444 If we can't handover the framebuffer, memset it to black. 4445 4446 mostly ported from nouveau. 4447 4448 Signed-off-by: Dave Airlie <airlied@redhat.com> 4449 4450commit 89a4c79a43a2b6963e41d7812e8fe5beedb556af 4451Author: Dave Airlie <airlied@redhat.com> 4452Date: Sat Jun 2 17:14:14 2012 +0100 4453 4454 radeon: fix smooth startup with tiling enabled. 4455 4456 We need to use the surface we worked out when we allocated the front bo, 4457 not work out a new surface from scratch. 4458 4459 This fixes smooth handover from plymouth to gdm on F17. 4460 4461 Signed-off-by: Dave Airlie <airlied@redhat.com> 4462 4463commit c1b9b2c9d71a87a0b98ccdb542da8a30c8c4d99a 4464Author: Dave Airlie <airlied@redhat.com> 4465Date: Fri Jun 1 12:22:06 2012 +0100 4466 4467 radeon: fix compat api for newest planned API. 4468 4469 Signed-off-by: Dave Airlie <airlied@redhat.com> 4470 4471commit 68c51d148b22ee8cbe98b5d5441e3494804ee808 4472Author: Dave Airlie <airlied@redhat.com> 4473Date: Fri Jun 1 12:21:34 2012 +0100 4474 4475 drmmode: drop flags arg to adjust frame. 4476 4477 Signed-off-by: Dave Airlie <airlied@redhat.com> 4478 4479commit b1e9c3081d2b975682a62389aa29063b79f9655c 4480Author: Dave Airlie <airlied@redhat.com> 4481Date: Thu May 24 17:54:41 2012 +0100 4482 4483 radeon: move include down further 4484 4485 fixes tinderbox build. 4486 4487 Signed-off-by: Dave Airlie <airlied@redhat.com> 4488 4489commit 5ed959585f35a4eff2cd803ae222b46d164f210d 4490Author: Dave Airlie <airlied@redhat.com> 4491Date: Thu May 24 08:17:11 2012 +0100 4492 4493 radeon: update compat-api.h for block handler + enable/disable fb. 4494 4495 This updates the compat stuff for the latest block handler code, 4496 and the enable/disable interface. 4497 4498 Signed-off-by: Dave Airlie <airlied@redhat.com> 4499 4500commit 2de17efdcb502f91fb145130c12be3cb164f03c5 4501Author: Dave Airlie <airlied@redhat.com> 4502Date: Thu May 24 08:16:39 2012 +0100 4503 4504 radeon/xvmc: fix build with new API. 4505 4506 This was missing the compat include. 4507 4508 Signed-off-by: Dave Airlie <airlied@redhat.com> 4509 4510commit a1f47768309aa8b14b13f05ba01bb490024c4cbc 4511Author: Dave Airlie <airlied@redhat.com> 4512Date: Thu May 24 07:55:14 2012 +0100 4513 4514 radeon: fix a few more RHDAtomBiosFunc uses 4515 4516 I missed these in my initial search/replace for some reason. 4517 4518 Signed-off-by: Dave Airlie <airlied@redhat.com> 4519 4520commit c26e40cf60eb6a2bffb5ec7cba1f1bbffcbf1d53 4521Author: Dave Airlie <airlied@redhat.com> 4522Date: Wed May 23 12:12:54 2012 +0100 4523 4524 radeon/atombios: rewrite atombios parser code to not use xf86Screens. 4525 4526 Just pass the ScrnInfoPtr around instead. 4527 4528 Signed-off-by: Dave Airlie <airlied@redhat.com> 4529 4530commit 14122214edabd2a574b66577289a7c8732f4e953 4531Author: Dave Airlie <airlied@redhat.com> 4532Date: Wed May 23 11:57:35 2012 +0100 4533 4534 radeon/generic_bus: stop passing scrnIndex 4535 4536 Just pass a pointer to the screen, removes usage of xf86Screens lookup 4537 4538 Signed-off-by: Dave Airlie <airlied@redhat.com> 4539 4540commit 2101e485480e3ffd6a17ed12f5287b8b1e4e0223 4541Author: Dave Airlie <airlied@redhat.com> 4542Date: Tue May 22 16:17:55 2012 +0100 4543 4544 compat for new server API 4545 4546commit 9d2549057ffced996bfaeb8df4901ef57da636c7 4547Author: Dave Airlie <airlied@redhat.com> 4548Date: Wed May 16 15:45:53 2012 +0100 4549 4550 ati: convert to new screen conversion APIs 4551 4552 The compat header takes care of the old server vs new server. 4553 4554 this commit was autogenerated from util/modular/x-driver-screen-scrn-conv.sh 4555 4556 Signed-off-by: Dave Airlie <airlied@redhat.com> 4557 4558commit 80a45bfc26e20d1d481f13cb15c0720d165a8306 4559Author: Dave Airlie <airlied@redhat.com> 4560Date: Wed May 16 15:42:59 2012 +0100 4561 4562 radeon: add compat-api.h 4563 4564commit 84c3900b5179a8d2cc5f925e94ed5b2bbdee1ad9 4565Author: Dave Airlie <airlied@redhat.com> 4566Date: Wed May 23 08:49:45 2012 +0100 4567 4568 radeon/xaa: drop scrnIndex parameter to some functions 4569 4570 This isn't needed, and makes api changes later easier. 4571 4572 Signed-off-by: Dave Airlie <airlied@redhat.com> 4573 4574commit 3fb694b308ebadd1b849836059b6b56bb19385f7 4575Author: Thierry Vignaud <thierry.vignaud@gmail.com> 4576Date: Wed May 16 14:43:53 2012 +0200 4577 4578 UMS/EXA: Add reminder for potential solid picture performance issue. 4579 4580commit 4b9bad959438725e4434e2aa4f142542d8dcc260 4581Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 4582Date: Fri May 11 05:25:32 2012 +0200 4583 4584 radeon: use GB_GR and BG_RG formats for packed yuv video for r600+ 4585 4586 Those formats were invented for exactly that purpose so use them. 4587 This saves some code and also some hw resources (only need one 4588 sampler instead of two for packed yuv). 4589 Only tested on EG. 4590 4591commit 37786e9027b8c8d1f9ec9928915784dd28853766 4592Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 4593Date: Fri May 11 05:25:33 2012 +0200 4594 4595 radeon: avoid rounding errors in texture coords for textured xv on EG+ 4596 4597 make sure the division is done with floats, otherwise the coordinate 4598 can be wrong up to 1 texel. 4599 Particularly visible with clipping and small source scaled up (since one 4600 texel can be a shift of several pixels) but could be seen even unscaled. 4601 Should provide more accurate coords without clipping too depending on the 4602 scale factor probably. 4603 This is a straight port of 688c8a54a00b01e73a11970ad2abe858f8c7c5c4 4604 when I apparently forgot the eg code... 4605 4606commit faea3aafa8c9a7c1bc866ffcd847972c5569c8bc 4607Author: Anisse Astier <anisse@astier.eu> 4608Date: Thu May 10 17:22:59 2012 +0200 4609 4610 Fail more gracefully when drm surface manager can't be initialized 4611 4612 Should make bugs like https://bugs.freedesktop.org/show_bug.cgi?id=48138 4613 easier to diagnose. 4614 4615 [ Michel Dänzer: Appended newline to error message. ] 4616 4617 Signed-off-by: Anisse Astier <anisse@astier.eu> 4618 Singed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4619 4620commit b0b7d8d26fd107df342b5c87b0a38e5bb08101a9 4621Author: Michel Dänzer <michel.daenzer@amd.com> 4622Date: Wed May 9 11:08:49 2012 +0200 4623 4624 EXA/UMS: Synchronize to the GPU before writing solid colour to scratch pixmap. 4625 4626 UMS doesn't do this automagically. It's a big hammer that will probably suck 4627 for performance, but I don't have any better ideas right now. 4628 4629 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4630 4631commit 6bda7ceda645e838723883d133d614def1511d16 4632Author: Michel Dänzer <michel.daenzer@amd.com> 4633Date: Thu May 3 15:07:30 2012 +0200 4634 4635 EXA: Attempt to fix solid picture acceleration with UMS. 4636 4637 Only compile tested, but should fix 4638 https://bugs.freedesktop.org/show_bug.cgi?id=49182 . 4639 4640 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4641 4642commit b33cf9cf9dce5add8aaccab10ba5aaa64247c8ce 4643Author: Alex Deucher <alexdeucher@gmail.com> 4644Date: Wed May 2 09:55:16 2012 -0400 4645 4646 radeon: add MacModel entry for SAM440ep embedded board 4647 4648 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4649 4650commit 0bda305f7ab2a4720b3fea3f318ab2a73be151e5 4651Author: Alex Deucher <alexander.deucher@amd.com> 4652Date: Fri Apr 13 13:28:10 2012 -0400 4653 4654 EXA: Support acceleration of solid pictures on R2xx. 4655 4656 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 4657 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 4658 4659commit a7754b076e355fef3aea082b6e2d3aefbb8e7a9b 4660Author: Alex Deucher <alexander.deucher@amd.com> 4661Date: Fri Apr 13 13:24:46 2012 -0400 4662 4663 EXA: Support acceleration of solid pictures on R1xx. 4664 4665 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 4666 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 4667 4668commit cac2d7ed54918579418fc762558497d3ec547fad 4669Author: Alex Deucher <alexander.deucher@amd.com> 4670Date: Sat Apr 14 08:53:39 2012 -0400 4671 4672 EXA: Support acceleration of solid pictures on Evergreen/NI. 4673 4674 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 4675 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 4676 4677commit d88b9700137ee407c483f263bb55c77cd6f92fef 4678Author: Michel Dänzer <michel.daenzer@amd.com> 4679Date: Fri Apr 13 09:35:38 2012 +0200 4680 4681 EXA: Support acceleration of solid pictures on R3xx-R7xx. 4682 4683 Allocate 1x1 scratch pixmaps to hold the solid picture colours. 4684 4685 This works around https://bugs.freedesktop.org/show_bug.cgi?id=47266 and might 4686 improve performance in other cases as well. 4687 4688 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4689 4690commit 66b586b9b9cdaf70f0fcd547b5a04f044d848d44 4691Author: Michel Dänzer <michel.daenzer@amd.com> 4692Date: Mon Apr 16 15:33:36 2012 +0200 4693 4694 RADEONCopySwap: Fix RADEON_HOST_DATA_SWAP_16BIT case. 4695 4696 It was the same code as for RADEON_HOST_DATA_SWAP_32BIT. This caused bus errors 4697 on FreeBSD/PPC, but I'm not sure how it could not cause problems anywhere... 4698 4699 Reported-by: Andreas Tobler <andreast@fgznet.ch> 4700 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4701 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 4702 4703commit 1e656cd5973bbcf85b63406638ede0bccd65e28f 4704Author: Michel Dänzer <michel.daenzer@amd.com> 4705Date: Mon Apr 16 15:44:01 2012 +0200 4706 4707 Make radeon_setup_kernel_mem failures more graceful and verbose. 4708 4709 So that bugs like https://bugs.freedesktop.org/show_bug.cgi?id=48138 can be 4710 diagnosed more easily. 4711 4712 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4713 4714commit d282719a9c2fb0ee32830aa75b8dfbb9392954ed 4715Author: Jerome Glisse <jglisse@redhat.com> 4716Date: Wed Apr 4 17:08:30 2012 -0400 4717 4718 r6xx-r9xx: force 1D tiling for buffer with height < 64 4719 4720 Due to some old kernel issue, height is 8 aligned insided the ddx 4721 For buffer with height btw 57 & 63 this lead ddx to believe it can 4722 allocate a 2D tiled surface while mesa will not align height and 4723 will assume 1D tiled leading to disagreement and rendering issue. 4724 This patch force buffer with height < 64 to be 1D tiled. 4725 4726 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 4727 4728commit de2419e5509398328903ca61f4ea80852cba6bc4 4729Author: Alex Deucher <alexdeucher@gmail.com> 4730Date: Thu Mar 29 00:28:39 2012 -0400 4731 4732 configure: bump version post release 4733 4734 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4735 4736commit 9425c50e93903fb64d9e569cfdc1e2c35d16ce25 4737Author: Alex Deucher <alexdeucher@gmail.com> 4738Date: Thu Mar 29 00:19:12 2012 -0400 4739 4740 configure: bump version for release 4741 4742 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4743 4744commit dc18d771713ecd893c7d5833da6e0661093161dc 4745Author: Alex Deucher <alexdeucher@gmail.com> 4746Date: Wed Mar 28 17:32:53 2012 -0400 4747 4748 configure: bump libdrm requirement for TN support 4749 4750 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4751 4752commit 03535904a3e1542b3924d0a062c4b022ca196888 4753Author: Alex Deucher <alexdeucher@gmail.com> 4754Date: Tue Mar 27 09:48:28 2012 -0400 4755 4756 radeon: man page updates 4757 4758 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4759 4760commit 6ed191c33caa33c12c2c6dafcba3a5ab1bf4a02f 4761Author: Alex Deucher <alexdeucher@gmail.com> 4762Date: Tue Mar 20 19:57:53 2012 -0400 4763 4764 radeon/kms: add TN pci ids 4765 4766 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4767 4768commit 55d65fcf33eb383e3fbc7a1d469ab68a70a7ab37 4769Author: Alex Deucher <alexdeucher@gmail.com> 4770Date: Tue Mar 20 19:54:57 2012 -0400 4771 4772 radeon/kms: add support for TN (trinity) APUs 4773 4774 - KMS only 4775 - Includes full EXA/Xv support 4776 4777 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4778 4779commit b5cf9bd693cf4090956add4c33c4fae9c3069a03 4780Author: Marek Olšák <maraeo@gmail.com> 4781Date: Wed Mar 7 11:01:40 2012 -0500 4782 4783 r6xx: initialize SX_MISC 4784 4785 If Mesa set it to 1, the DDX would not render anything = the monitor would 4786 basically freeze. 4787 4788 agd5f: update emit count as well. 4789 4790 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4791 4792commit 355dc4295912c153f5333421594fa90aa119a056 4793Author: Michel Dänzer <michel.daenzer@amd.com> 4794Date: Tue Mar 6 15:52:40 2012 +0100 4795 4796 DRI2: Unreference buffers immediately when event wait info is invalidated. 4797 4798 Deferring this could result in trying to unreference buffers from a previous 4799 server generation, i.e. accessing freed memory. 4800 4801 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4802 Tested-by: Christian König <Christian.koenig@amd.com> 4803 4804commit fe51469b2e02e4d565050bab077985270fb58a9b 4805Author: Michel Dänzer <michel.daenzer@amd.com> 4806Date: Tue Mar 6 15:52:40 2012 +0100 4807 4808 Re-register DRM FD wakeup handler for each server generation. 4809 4810 Fixes hang when trying to use DRI2 swap scheduling after a server reset. 4811 4812 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4813 Tested-by: Christian König <Christian.koenig@amd.com> 4814 4815commit 878454ae8d8e96dd27a19d0b30940d014c4cd7e2 4816Author: Hans Verkuil <hverkuil@xs4all.nl> 4817Date: Fri Feb 24 09:35:39 2012 -0500 4818 4819 Fix ConnectorTable crash in radeon_output.c 4820 4821 The sam440ep PPC board requires a ConnectorTable xorg.conf option, but putting 4822 in that option causes the radeon driver to crash. I finally traced it to a 4823 copy-and-paste bug in radeon_output.c as a result of a major rework in commit 4824 82f12e5a40c1fbcb91910a0f8b725c34fff02aae. 4825 4826 The actual crash occurred in RADEONPrintPortMap(). 4827 4828 Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> 4829 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4830 4831commit 688c8a54a00b01e73a11970ad2abe858f8c7c5c4 4832Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 4833Date: Sat Feb 18 21:12:34 2012 +0100 4834 4835 radeon: avoid rounding errors in texture coords for textured xv 4836 4837 make sure the division is done with floats, otherwise the coordinate 4838 can be wrong up to 1 texel. 4839 Particularly visible with clipping and small source scaled up (since one 4840 texel can be a shift of several pixels) but could be seen even unscaled. 4841 Should provide more accurate coords without clipping too depending on the 4842 scale factor probably. 4843 Changed for r100-r600, though only tested on r300. 4844 4845commit 2778b56252124ef6f636a493d2e1457b43911c37 4846Author: Jerome Glisse <jglisse@redhat.com> 4847Date: Mon Feb 13 20:42:57 2012 -0500 4848 4849 radeon: r6xx-eg use linear general when using scratch bo 4850 4851 In path where we need to use scratch bo as temporary area, 4852 consider it as linear buffer. Not linear aligned. Fix some 4853 case such as in bugs: 4854 4855 https://bugs.freedesktop.org/show_bug.cgi?id=45827 4856 4857 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 4858 4859commit c66ae235700f5efe64eb168327551b8f1d153c9c 4860Author: Michel Dänzer <michel.daenzer@amd.com> 4861Date: Mon Feb 13 10:43:58 2012 +0100 4862 4863 Handle new xorg_list API. 4864 4865 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=45937 4866 4867 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4868 4869commit 7ff277e22c629308915307bbee96eb25ff77f8b9 4870Author: Alex Deucher <alexdeucher@gmail.com> 4871Date: Fri Feb 10 13:04:59 2012 -0500 4872 4873 radeon: fix crash in drmmode_create_bo_pixmap() 4874 4875 Only init surface on r6xx+. Return NULL rather than 4876 FALSE. 4877 4878 Fixes: 4879 https://bugs.freedesktop.org/show_bug.cgi?id=45829 4880 4881 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4882 4883commit 60b949f34df5db05e0e102cc3daa33469aa50cfc 4884Author: Alex Deucher <alexdeucher@gmail.com> 4885Date: Fri Feb 10 13:11:26 2012 -0500 4886 4887 radeon/kms: reusing fd message is not an error 4888 4889 It's standard behavior. 4890 4891 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4892 4893commit e20284409937d784847339b5d466a95012d85940 4894Author: Michel Dänzer <michel.daenzer@amd.com> 4895Date: Fri Feb 3 12:21:59 2012 +0100 4896 4897 EXA/r6xx+: Only set write domain or read domains, not both. (Bug #43893) 4898 4899 Avoids an accounting bug in libdrm_radeon 2.4.31 or older. 4900 4901 See https://bugs.freedesktop.org/show_bug.cgi?id=43893 4902 4903 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4904 4905commit 5a7f64c5170ca424c9dca739662018e30df13413 4906Author: Michel Dänzer <michel.daenzer@amd.com> 4907Date: Wed Feb 8 11:35:35 2012 +0100 4908 4909 Fix UMS build failure. 4910 4911 And some UMS specific warnings. 4912 4913 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4914 4915commit 87f776b966f4200c97a989536d4b71822ae4c0b3 4916Author: Michel Dänzer <michel.daenzer@amd.com> 4917Date: Wed Feb 8 11:10:27 2012 +0100 4918 4919 Remove unused local variable 'height'. 4920 4921 Pointed out by gcc -Wunused-variable. 4922 4923 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4924 4925commit f63262e2f112a348c45f0dcecd891c8b6d9c5ee8 4926Author: Michel Dänzer <michel.daenzer@amd.com> 4927Date: Wed Feb 8 10:28:45 2012 +0100 4928 4929 evergreen: Initialize source surface member for textured video. 4930 4931 Fixes crash reported by Ole Salscheider on IRC. 4932 4933 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4934 4935commit c8f104d38870f14049402bbc14f662c151caeeef 4936Author: Jerome Glisse <jglisse@redhat.com> 4937Date: Tue Feb 7 15:04:37 2012 -0500 4938 4939 radeon: fix tiling for weird resolution 4940 4941 Should also fix xv for some case. 4942 4943 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 4944 4945commit 615033f2b5e3817e335e9d022fc9fdcf8ac8b11a 4946Author: Jerome Glisse <jglisse@redhat.com> 4947Date: Tue Dec 13 11:08:19 2011 -0500 4948 4949 r600-evergreen: use common surface allocator for tiling v11 4950 4951 Use libdrm common surface code so mesa,ddx have same idea 4952 about tiling surface and what their pitch should be and 4953 the alignment constraint. 4954 4955 v2 fix remaining issue add new option to conditionaly enable 4956 v3 fix fbcon copy and r600 exa copy path 4957 v4 fix non tiled path 2D tiling on GPU >= R600, set it to false 4958 as default 4959 v5 adapt to pixel/element size split of libdrm/radeon 4960 v6 update to properly handle falling back to 1d tiled 4961 v6 final fix to tile split value on evergreen and newer 4962 v7 fix default array mode on r6xx, fix height alignment issue 4963 on evergreen 4964 v8 fix tile split value 4965 v9 add stencil tile split support, simplify dri2 for stencil 4966 with evergreen 4967 v10 Try to fix xv path regarding tiling. Adapt to libdrm API 4968 change. Try to fix case where there is no surface which 4969 means non tiled bo. 4970 v11 check for proper libdrm 4971 4972 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 4973 4974commit 36c190671081967bac6fff48aaf66d67b639a48c 4975Author: Michel Dänzer <michel.daenzer@amd.com> 4976Date: Wed Feb 1 13:21:02 2012 +0100 4977 4978 Fix vline range calculations. 4979 4980 The range passed in is in pixmap coordinates, so the CRTC offset needs to be 4981 added to the clamping limits and subtracted from the clamped range for 4982 pre-AVIVO display engines. 4983 4984 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4985 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 4986 4987commit bb0e1531ac6949d38025d7dcb19234fee33b2acf 4988Author: Michel Dänzer <michel.daenzer@amd.com> 4989Date: Wed Feb 1 13:07:11 2012 +0100 4990 4991 Check for empty vline ranges after clamping. 4992 4993 The clamping could turn a previously non-empty range into an empty one. 4994 4995 Also, start == stop means the range is empty. 4996 4997 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 4998 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 4999 5000commit 13b3aed4ef9afbcbaea1dcf0ed1acb162b240a3f 5001Author: Alex Deucher <alexdeucher@gmail.com> 5002Date: Tue Jan 10 09:35:09 2012 -0500 5003 5004 EXA/r6xx+: fix rop setting for overlapping copies 5005 5006 Need to use GXCopy for the src to temp copy, then 5007 the original rop for the temp to dest copy. 5008 5009 Noticed by: Frank Huang 5010 5011 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5012 5013commit eb6d769a087b2ed5952f477fc3f0b0625810a287 5014Author: Egbert Eich <eich@suse.de> 5015Date: Tue Nov 15 18:50:56 2011 +0100 5016 5017 DPMS: Split non-modeset CRTC DPMS function. 5018 5019 RADEONRestore() calls crtc->funcs->dpms() after most of the mode setting 5020 subsystems have been restored. This function enables the CRTCs but does 5021 more: it calls DRM pre- and post-modeset ioctls and sets up the palettes 5022 (LUTs). 5023 None of these two things are needed. Accessing the palette registers after 5024 restoring the PLLs can even lead to lockups. 5025 Thus the CRTC DPMS function is split into two parts: one that just enables 5026 /disables the CRTC and one which wraps this function and does the rest. 5027 Now the inner function can be called directly from RADEONRestore() as 5028 there is no need to go thru the RandR hooks in this function while the 5029 RandR hook uses the wrappering function so the full functionality is 5030 preserved from an RandR point of view. 5031 5032 Signed-off-by: Egbert Eich <eich@freedesktop.org> 5033 Reviewed-by: Alex Deucher <alexdeucher@gmail.com> 5034 5035commit ac51e331895b216d288bc7bd108a38b362214668 5036Author: Egbert Eich <eich@suse.de> 5037Date: Mon Nov 14 19:10:01 2011 +0100 5038 5039 UMS: Fix lockups in palette save/restore on pre-AVIVO chips. 5040 5041 The reintroduction of palette save/restore in 5efdf514 causes some 5042 pre-AVIVO chips to lock up. An investigation revealed that accessing 5043 palette registers when the associated PLL is not running is causing 5044 this. With UMS the PLL setup that is saved has been done by the BIOS 5045 typically. 5046 A similar issue was observed when VGA palette save/restore had 5047 been reinitroduced with 80eee856: 5048 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=480312 5049 and has been worked around for Linux without further investigation 5050 by 87e66ce7. 5051 To fix the issue we now 5052 a. introduce 'on-demand' palette saving (ie the palette is 5053 saved before it is first altered). This guarantees that 5054 the palette register are only associated when the associated 5055 CRTC is active and thus the PLLs are powered up and running. 5056 b. move palette restore before PLL restore. 5057 c. eliminate generic VGA palette save/restore which seems to be 5058 unneeded when the palette is restored natively. 5059 It is believed that this caused the behavior described in 5060 https://bugs.freedesktop.org/show_bug.cgi?id=18407#c27 5061 5062 Signed-off-by: Egbert Eich <eich@freedesktop.org> 5063 Reviewed-by: Alex Deucher <alexdeucher@gmail.com> 5064 5065commit 0a8d04eeac95f4db9d03ee31070bd825a7feb0b2 5066Author: Matthieu Herrb <matthieu.herrb@laas.fr> 5067Date: Sun Jan 1 18:27:54 2012 +0100 5068 5069 Update for new vgaHW API. 5070 5071 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr> 5072 Reviewed-by: Mark Kettenis <kettenis@openbsd.org> 5073 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5074 5075commit ae45d7e6d8e6844cd4586c9ee97c21b257fa788f 5076Author: Michel Dänzer <michel.daenzer@amd.com> 5077Date: Wed Dec 28 11:48:36 2011 +0100 5078 5079 DRI2: Can't use page flipping for pixmaps. (bug #42913) 5080 5081 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=42913 . 5082 5083 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5084 5085commit aacbd629b02cbee3f9e6a0ee452b4e3f21376bd3 5086Author: Alex Deucher <alexdeucher@gmail.com> 5087Date: Mon Dec 12 09:32:30 2011 -0500 5088 5089 radeon: add some new pci ids 5090 5091 fixes: 5092 https://bugs.freedesktop.org/show_bug.cgi?id=43739 5093 5094 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5095 5096commit bc54e415e2fd344939c5c788ea0686133a7e2c69 5097Author: Dave Airlie <airlied@redhat.com> 5098Date: Tue Dec 6 15:47:45 2011 +0000 5099 5100 radeon: add original radeon to always tiled. 5101 5102 and actually enable it for M7, previous commit only did one function. 5103 5104 Signed-off-by: Dave Airlie <airlied@redhat.com> 5105 5106commit ba46c7b0cf72d157748981eb3224d5eefb6200aa 5107Author: Dave Airlie <airlied@redhat.com> 5108Date: Tue Dec 6 13:42:49 2011 +0000 5109 5110 radeon: refine always tiled depth check 5111 5112 So it appears the M7 family always tiles its depth buffer also. 5113 5114 Signed-off-by: Dave Airlie <airlied@redhat.com> 5115 5116commit 98b2d5fe1722a43c4bbe7711ed7180a3fb65305f 5117Author: Dave Airlie <airlied@redhat.com> 5118Date: Mon Dec 5 18:44:28 2011 +0000 5119 5120 radeon: r200 depth buffers are always tiled 5121 5122 When we do the allocations we need to make sure the always tiled 5123 nature is taken into account. 5124 5125 Signed-off-by: Dave Airlie <airlied@redhat.com> 5126 5127commit 7dcefc69d9fbceae27cd03083c815e01a19b527e 5128Author: Alex Deucher <alexdeucher@gmail.com> 5129Date: Mon Dec 5 09:21:48 2011 -0500 5130 5131 Xv: Evergreen+ asics support 16k surfaces 5132 5133 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5134 5135commit 72e386d42516e7cd3c2cbf2fffc9174cd3ec8451 5136Author: Alex Deucher <alexander.deucher@amd.com> 5137Date: Wed Nov 30 19:38:35 2011 -0500 5138 5139 radeon: add some new pci ids 5140 5141 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 5142 5143commit 3853c3020d05175ae180b9a188dec7c425bdd0b8 5144Author: Dave Airlie <airlied@redhat.com> 5145Date: Mon Nov 28 18:38:30 2011 +0000 5146 5147 fixup xinerama since 9151f3b1c2ebcc34e63195888ba696f2183ba5e2 5148 5149 since the driver would call RRFirstOutput without checking if randr has 5150 been enabled, and it would crash in privates code. 5151 5152 reported by vereteran on #radeon 5153 5154 Signed-off-by: Dave Airlie <airlied@redhat.com> 5155 Acked-on-irc-by: Michel Dänzer <michel.daenzer@amd.com> 5156 5157commit d669c34f140c000f88c4b4e464e44e6c8694f581 5158Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 5159Date: Mon Nov 21 11:35:40 2011 +1100 5160 5161 ddx/evergreen: Fix endian of ALU constants 5162 5163 The constants are written directly into a buffer object shared with the 5164 card and we "forget" to swap them. This patch fixes it by doing the swap 5165 in evergreen_set_alu_consts() in-place (ie, it modifies the buffer), 5166 which should be fine with the way we use it in the ddx. 5167 5168 This makes everything work fine on my caicos card on a G5 including some 5169 quik tests with Xv, gnome3 shell, etc... 5170 5171 Thanks a lot to Jerome Glisse for holding my hand through debugging that 5172 (and finding the actual bug). 5173 5174 Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> 5175 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5176 5177commit 422bdd4fe6cb728e1dd08a56f6ee2d0f009cbfcb 5178Author: Alex Deucher <alexdeucher@gmail.com> 5179Date: Mon Nov 14 09:39:16 2011 -0500 5180 5181 radeon: add missing FireMV pci id 5182 5183 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5184 5185commit 534fb6e413a909a9d1afd57d1c711844b7c5ddf4 5186Author: Dave Airlie <airlied@redhat.com> 5187Date: Fri Nov 11 10:26:51 2011 +0000 5188 5189 ati: enable bg none when fbcon succeeds and we are built against ABI after 10. 5190 5191 One less patch to keep carrying in Fedora. 5192 5193 Signed-off-by: Dave Airlie <airlied@redhat.com> 5194 5195commit 89452c08048c98fb5cc3dc551b3824be40d52cf2 5196Author: Michel Dänzer <michel.daenzer@amd.com> 5197Date: Tue Nov 8 11:23:11 2011 +0100 5198 5199 UMS: Guard references to PCITAG / pciTag with XSERVER_LIBPCIACCESS (bug #42690) 5200 5201 Should fix https://bugs.freedesktop.org/show_bug.cgi?id=42690 . 5202 5203 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5204 5205commit 5ec34ed95948f7164184551615c1fc4c3eef3b98 5206Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 5207Date: Thu Nov 3 20:16:47 2011 -0400 5208 5209 DRI/DRI2: remove hard-coded limitation to 6 crtcs 5210 5211 DRM's hard limit to the number of CRTCs is 32. ATI DDX unnecessarily 5212 clips this limit to 6 by hard coding initial assumption for 5213 output->possible_crtcs mask to 0x7f (before it gets trimmed down to 5214 what's really possible for a given output) and by allocating only 6 5215 entries for for cursor_bo[] array in RADEONInfoRec. 5216 5217 Fix this and thus allow the ATI DDX to deal with as many CRTCs 5218 as the DRM allows (32), so it is ready if anything with >6 CRTCs 5219 comes out. 5220 5221 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 5222 5223commit 4853ab2cdc3b97948c7cd69eaf4fff54f59774fc 5224Author: Michel Dänzer <michel.daenzer@amd.com> 5225Date: Fri Nov 4 12:15:53 2011 +0100 5226 5227 Turn compile time check into runtime check. 5228 5229 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5230 5231commit bcdb54fe16ebf2e239b84eebf20e8adfe5094bff 5232Author: Alex Deucher <alexdeucher@gmail.com> 5233Date: Thu Oct 20 11:11:35 2011 -0400 5234 5235 check for xserver 1.9.4.901 to enable tiling by default 5236 5237 Previous xservers had a bug in the EXA code which caused 5238 display corruption in some cases. 5239 5240 See: 5241 https://bugs.freedesktop.org/show_bug.cgi?id=33929 5242 5243 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5244 5245commit edde94cba5321e6e51e7fd4d79dde5abc4944495 5246Author: Michel Dänzer <michel.daenzer@amd.com> 5247Date: Wed Nov 2 13:40:05 2011 +0100 5248 5249 Bump version post release. 5250 5251commit 93459f842c2d8dc178a1954b8e05150fcb96ac9a 5252Author: Michel Dänzer <michel.daenzer@amd.com> 5253Date: Wed Nov 2 12:51:15 2011 +0100 5254 5255 Bump version for 6.14.3 release. 5256 5257commit fe3ce559b7f07d39ded39abb38576846d6eb185b 5258Author: Jeremy Huddleston <jeremyhu@apple.com> 5259Date: Tue Nov 1 20:27:03 2011 -0700 5260 5261 Build fix for -Werror=int-to-pointer-cast 5262 5263 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 5264 5265commit d3d6a5da07aaec5961e51c9a8f90c1490ee101b6 5266Author: Tormod Volden <debian.tormod@gmail.com> 5267Date: Sat Oct 29 23:08:46 2011 +0200 5268 5269 radeon: do not include xf86PciInfo.h 5270 5271 We already use atipciids.h instead most places. 5272 5273 Signed-off-by: Tormod Volden <debian.tormod@gmail.com> 5274 5275commit 70da7001e81363ed6ef2c4727c512daf53ae29fe 5276Author: Jeremy Huddleston <jeremyhu@apple.com> 5277Date: Sat Oct 29 20:15:09 2011 -0700 5278 5279 Use malloc/calloc/realloc/free directly 5280 5281 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 5282 5283commit a3bb07efb1757c33d70e2e1928219d12a4dd6498 5284Author: Michel Dänzer <michel.daenzer@amd.com> 5285Date: Tue Oct 25 17:43:58 2011 +0200 5286 5287 EXA >= R6xx / KMS: Avoid running out of CS space at inconvenient times. 5288 5289 Otherwise we may end up with things not properly set up at the beginning of the 5290 next CS. 5291 5292 Fixes http://bugs.debian.org/645007 . 5293 5294 In contrast to the Composite code for < R6xx, this isn't necessary with UMS, 5295 as the draw packet only uses constant space in the indirect buffer, and nothing 5296 else can mess with the 3D state between indirect buffers. 5297 5298 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5299 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 5300 5301commit 23788c4a5e3b6affb9b183e1393edd0e5ca4550e 5302Author: Michel Dänzer <michel.daenzer@amd.com> 5303Date: Thu Oct 27 19:12:22 2011 +0200 5304 5305 EXA < R6xx: Make sure 2D state is re-emitted after running out of CS space. 5306 5307 Otherwise it's basically luck what the 2D state ends up being at the beginning 5308 of the next CS. 5309 5310 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5311 5312commit 6e0e1a821accc6ca95f4134e49b66a6b168c1934 5313Author: Michel Dänzer <michel.daenzer@amd.com> 5314Date: Tue Oct 25 17:39:57 2011 +0200 5315 5316 Make radeon_dri2_create_buffer(s) more robust. (Bug #30047) 5317 5318 In particular, handle and propagate failure to allocate GPU accessible memory, 5319 instead of crashing. Fixes https://bugs.freedesktop.org/show_bug.cgi?id=30047 . 5320 5321 Also take care not to leak resources in error paths. 5322 5323 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5324 5325commit 856583dbca9319c77fed40daa9956e81a0068f9e 5326Author: Jerome Glisse <jglisse@redhat.com> 5327Date: Thu Oct 20 14:17:14 2011 -0400 5328 5329 radeon/kms: fallback to vesa if GPU is not supported by UMS 5330 5331 For GPU not supported by UMS, test in probe so that we properly 5332 fallback to vesa. 5333 5334 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 5335 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 5336 5337commit f6f1b4f7789cfef763c063e671b76b60be8bdd8e 5338Author: Maarten Lankhorst <m.b.lankhorst@gmail.com> 5339Date: Tue Oct 18 14:30:39 2011 +0200 5340 5341 dri2: Add vdpau driver name entry 5342 5343 libvdpau has a driver loading mechanism that looks for a dri2 driver 5344 first before falling back to nvidia, so lets use that. 5345 5346 Allows use of libvdpau_rx00 without having to set things up separately, 5347 similar to the patch to xf86-video-nouveau. 5348 5349 Signed-off-by: Maarten Lankhorst <m.b.lankhorst@gmail.com> 5350 Reviewed-by: Christian König <deathsimple@vodafone.de> 5351 Reviewed-by: Michel Dänzer <michel@daenzer.net> 5352 Tested-by: Michel Dänzer <michel@daenzer.net> 5353 5354commit a6b2bd2d184f10d4c56c4ee17186aedb238a36ec 5355Author: Alex Deucher <alexdeucher@gmail.com> 5356Date: Wed Sep 21 17:00:16 2011 -0400 5357 5358 UMS: fix DDIA enable on some rs690 systems 5359 5360 DVOOutputControl checks the value of of bios scratch reg 3 5361 on some tables and assumes the encoder is already enabled 5362 if the DFP2_ACTIVE bit is set. Clear that bit so the table 5363 sets the DDIA enable bit properly. 5364 5365 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5366 5367commit d78860ba53d9bfcf6c28e1cfd2d970709b5e20fa 5368Author: Michel Dänzer <michel.daenzer@amd.com> 5369Date: Fri Aug 12 11:21:33 2011 +0200 5370 5371 Only call radeon_dri2_close_screen() if DRI2 was enabled. 5372 5373 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5374 5375commit 11330ca5dc61a70fe4507e63230f9133ca22d891 5376Author: Michel Dänzer <michel.daenzer@amd.com> 5377Date: Fri Aug 12 11:21:32 2011 +0200 5378 5379 Remove dead variable remain_size_bytes. 5380 5381 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5382 5383commit fcf0cca9c0ab0f692b222f619aee8f1cdad3b519 5384Author: Michel Dänzer <michel.daenzer@amd.com> 5385Date: Tue Sep 20 12:34:05 2011 +0200 5386 5387 KMS Color Tiling requires xserver which supports EXA_MIXED_PIXMAPS. 5388 5389 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5390 5391commit c96e6fb8a5f5be2319fdb4c431c1ba5279997fe8 5392Author: Alex Deucher <alexdeucher@gmail.com> 5393Date: Sat Sep 17 08:26:12 2011 -0400 5394 5395 man: note that the list of marketing names is non-exhaustive 5396 5397 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5398 5399commit 1a51fce6f6ab169c882a86b936909c0820f27a68 5400Author: Alex Deucher <alexdeucher@gmail.com> 5401Date: Thu Sep 15 17:52:25 2011 -0400 5402 5403 update man page with new marking names 5404 5405 Note that the driver support all currently 5406 shipping asics and the names in the man page 5407 are just a sampling. 5408 5409 Fixes: 5410 https://bugs.freedesktop.org/show_bug.cgi?id=40808 5411 5412 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5413 5414commit 64f237a4cf3ef5bcb3163e88e1447ff275a1eefa 5415Author: Michel Dänzer <michel.daenzer@amd.com> 5416Date: Thu Aug 18 19:11:08 2011 +0200 5417 5418 Convert register ranges for >= r6xx from enums to defines. 5419 5420 Avoids lots of "comparison between 'enum <anonymous>' and 'enum <anonymous>'" 5421 warnings with newer versions of gcc. See 5422 https://bugs.freedesktop.org/show_bug.cgi?id=38238 . 5423 5424 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5425 5426commit 95991fcce45f0dd904f76b9f98f9c7ed5708e4fa 5427Author: Michel Dänzer <michel.daenzer@amd.com> 5428Date: Wed Aug 17 11:10:34 2011 +0200 5429 5430 Bail if we're trying to start up in UMS mode on KMS. 5431 5432 Ideally, the display manager will start the X server again, and everything 5433 will be fine and dandy. But in the worst case, at least we won't hit the 5434 hardware behind the KMS driver's back. 5435 5436 (This change intentionally makes (ab)use of the fact that Bool is defined as 5437 int). 5438 5439 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5440 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 5441 5442commit f95a41b7851565c282d22f8d679db1377428f165 5443Author: Michel Dänzer <michel.daenzer@amd.com> 5444Date: Thu Aug 11 11:22:57 2011 +0200 5445 5446 video: Don't round up bottom/right edge for clipping source width/height. 5447 5448 It's not necessary: If the top/left edge was rounded down, this will be 5449 compensated by the subtraction. 5450 5451 Worse, if the original source width/height is odd, rounding up may result in 5452 reading past the end of the source data. 5453 5454 Fixes http://bugs.debian.org/637258 . 5455 5456 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5457 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 5458 5459commit 93fc0843a1e31dc9237433bc2bf17df79e956d26 5460Author: Michel Dänzer <michel.daenzer@amd.com> 5461Date: Wed Aug 10 17:44:37 2011 +0200 5462 5463 Change my e-mail address to something that still works, and always will, I hope. 5464 5465commit 9151f3b1c2ebcc34e63195888ba696f2183ba5e2 5466Author: Michel Dänzer <michel.daenzer@amd.com> 5467Date: Tue Aug 9 19:13:26 2011 +0200 5468 5469 Prefer the CRTC of the primary output for synchronization. 5470 5471 See https://bugs.freedesktop.org/show_bug.cgi?id=39696 . 5472 5473 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 5474 5475commit 3b9fdc807dd7e52af0576299cefba596040f6f2f 5476Author: Alex Deucher <alexdeucher@gmail.com> 5477Date: Wed Aug 3 16:20:13 2011 -0400 5478 5479 r5xx+: Fix vline setup with crtc offsets 5480 5481 On r5xx+, vline is relative to to the viewport, not 5482 the scanlines. Based on initial patch and investigation 5483 from Herbert Pötzl (Bertl) on IRC. 5484 5485 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5486 5487commit 104b2d7c071f29266b1bc4184a74e9714d14febc 5488Author: Alex Deucher <alexander.deucher@amd.com> 5489Date: Mon Aug 1 10:05:30 2011 -0400 5490 5491 kms: fix possible leak in pageflip code 5492 5493 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 5494 5495commit d29bab632e9ecccba518d4107d52620bf75eb1cf 5496Author: Ville Syrjala <syrjala@sci.fi> 5497Date: Wed May 4 23:51:27 2011 +0300 5498 5499 kms: Move flip_count and co. to a per swap structure 5500 5501 If multiple drawables are doing page flipping, the global drmmode 5502 structure can't be used to keep per swap information. For example 5503 flip_count can increase prematurely due to another swap request, 5504 and then the previous swap request never gets completed, leading to a 5505 stuck client. Move the relevant pieces of data to a strucuture that 5506 gets allocated once per swap request and shared by all involved CRTCs. 5507 5508 Signed-off-by: Ville Syrjala <syrjala@sci.fi> 5509 5510commit 9493563c1ef4b51af0ee8a44cb4e7c5bb280347e 5511Author: Ville Syrjala <syrjala@sci.fi> 5512Date: Wed May 4 23:51:26 2011 +0300 5513 5514 dri2: Update front buffer pixmap and name before exchanging buffers 5515 5516 Buffer exchange assumes that the front buffer pixmap and name 5517 information is accurate. That may not be the case eg. if the window 5518 has been (un)redirected since the buffer was created. 5519 5520 Signed-off-by: Ville Syrjala <syrjala@sci.fi> 5521 5522commit 8c9266ed2da22a510243f9a952c14d4423f48a2b 5523Author: Alex Deucher <alexander.deucher@amd.com> 5524Date: Fri Jul 15 10:44:57 2011 -0400 5525 5526 radeon: add some new NI pci ids 5527 5528 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 5529 5530commit f59c3b294b0f715fc96e2bbe25893f2b31aa488b 5531Author: Christian König <deathsimple@vodafone.de> 5532Date: Thu Jul 14 11:49:06 2011 +0200 5533 5534 Register XvMC video decoding acceleration 5535 5536commit e8d0d437957b15252dfad775796a3949ed50dbcf 5537Author: Dave Airlie <airlied@redhat.com> 5538Date: Tue Jul 12 11:43:25 2011 -0400 5539 5540 evergreen: Emit SQ_LDS_RESOURCE_MGMT 5541 5542 Avoids rendering problems when compute changes this reg. 5543 5544 Fixes: 5545 https://bugs.freedesktop.org/show_bug.cgi?id=39119 5546 5547 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5548 5549commit 9bb31158466e6168116d841d12c8b4303f11c4a6 5550Author: Alex Deucher <alexdeucher@gmail.com> 5551Date: Wed Jun 22 12:24:28 2011 -0400 5552 5553 evergreen: fix num_banks for 2D tiling config 5554 5555 The field is encoded. 5556 5557 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5558 5559commit 122b471f734aa07427b01d4bec35ff1ac28290b5 5560Author: Alex Deucher <alexdeucher@gmail.com> 5561Date: Fri Jun 17 04:02:16 2011 -0400 5562 5563 dri2: fix copy pasto in a6154c00c64932332e8f6e334661ffd579cfd894 5564 5565 Reported-by: Nils Wallménius <nils.wallmenius@gmail.com> 5566 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5567 5568commit 122bedcbcf45cb583cf51b2fd04ed2805e0ca60b 5569Author: Alex Deucher <alexdeucher@gmail.com> 5570Date: Thu Jun 16 12:57:11 2011 -0400 5571 5572 dri2: missing bit from a6154c00c64932332e8f6e334661ffd579cfd894 5573 5574 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5575 5576commit a6154c00c64932332e8f6e334661ffd579cfd894 5577Author: Alex Deucher <alexdeucher@gmail.com> 5578Date: Thu Jun 16 12:48:43 2011 -0400 5579 5580 dri2/eg+: fix size and alignment of depth/stencil buffers 5581 5582 Base alignment may be 256B or 512B depending on the group 5583 size. Also need to check against front size for virtualX. 5584 5585 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5586 5587commit 41eb1fbb3d9da64feb4a96df7c575e44136b1538 5588Author: Alex Deucher <alexdeucher@gmail.com> 5589Date: Mon Jun 13 12:50:39 2011 -0400 5590 5591 kms/man: update ColorTiling info 5592 5593 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5594 5595commit 481baa5c74271cd7ce38bae3965d2bc4b8809058 5596Author: Alex Deucher <alexdeucher@gmail.com> 5597Date: Mon Jun 13 12:44:07 2011 -0400 5598 5599 kms: enable ColorTiling by default on r6xx-cayman asics 5600 5601 Currently only 1D tiling as 2D tiling still has some corner 5602 cases to fix up. 5603 5604 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5605 5606commit cbcc57b0fa6f581be777bef648f2bf3efe7443ee 5607Author: Alex Deucher <alexdeucher@gmail.com> 5608Date: Mon Apr 4 12:52:00 2011 -0400 5609 5610 radeon: add llano pci ids 5611 5612 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5613 5614commit 36afd1e1055eeadb2396dadcc68b214655bd90a9 5615Author: Alex Deucher <alexdeucher@gmail.com> 5616Date: Tue May 31 16:03:36 2011 -0400 5617 5618 radeon: add support for llano APUs 5619 5620 - KMS only 5621 - Includes full EXA/Xv support 5622 5623 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5624 5625commit 557f46dc2f18734ecf1f18dee7e951e0bf062e63 5626Author: Dave Airlie <airlied@redhat.com> 5627Date: Fri May 27 07:22:08 2011 +1000 5628 5629 bump version after release 5630 5631commit 2fca40ea65d9f2a6f8451c324bb4b82786f34f76 5632Author: Dave Airlie <airlied@redhat.com> 5633Date: Thu May 26 12:52:21 2011 +1000 5634 5635 radeon: bump version for release of 6.14.2 5636 5637commit fab868c5f7d8cafdb0176d2751f216819a5ba66a 5638Author: Alex Deucher <alexdeucher@gmail.com> 5639Date: Wed May 25 01:09:12 2011 -0400 5640 5641 EXA: make evergreen_fix_scissor_coordinates static 5642 5643 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5644 5645commit fe5c42f5155361006b687da824181418f688809f 5646Author: Alex Deucher <alexdeucher@gmail.com> 5647Date: Tue May 24 22:32:01 2011 -0400 5648 5649 cayman: endian fixes for shaders 5650 5651 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5652 5653commit 470ecd02347c32e79316046d01a7d5dad0e2fe99 5654Author: Alex Deucher <alexdeucher@gmail.com> 5655Date: Tue May 24 18:31:47 2011 -0400 5656 5657 EXA/Xv: add workarounds for eg/cayman scissors bugs 5658 5659 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5660 5661commit b913e7ba2b60d47a6660699210cc3cf6f5dc52c2 5662Author: Alex Deucher <alexdeucher@gmail.com> 5663Date: Tue May 24 18:02:52 2011 -0400 5664 5665 Revert "cayman: add scissors workaround." 5666 5667 Needs a proper workaround for a hw bug. 5668 5669 This reverts commit b77d374b0d11f48c33cfffdb4157c4ec4b05ea15. 5670 5671commit 82cb33c3f0e1ba802d7a94f3159b3c5c86cd4043 5672Author: Dave Airlie <airlied@redhat.com> 5673Date: Wed May 18 14:49:17 2011 +1000 5674 5675 cayman: enable all accel 5676 5677commit b77d374b0d11f48c33cfffdb4157c4ec4b05ea15 5678Author: Dave Airlie <airlied@redhat.com> 5679Date: Tue May 24 15:53:58 2011 +1000 5680 5681 cayman: add scissors workaround. 5682 5683 wasted a lot of time getting to this. 5684 5685commit ffeab7a7058298e15294a3b2c740c731e36dda1d 5686Author: Alex Deucher <alexdeucher@gmail.com> 5687Date: Mon Apr 18 18:16:51 2011 -0400 5688 5689 cayman: fix dword counts default state 5690 5691 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5692 5693commit 3cbfae361bf5e779d3364f0f31cfd25bd0f59e65 5694Author: Alex Deucher <alexdeucher@gmail.com> 5695Date: Wed Mar 2 20:48:19 2011 -0500 5696 5697 cayman: add spi state to default state 5698 5699 changed in e3145801b80fd4be4cf770128876e86e89bda66f 5700 5701 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5702 5703commit b8ade97c9d0fa5aacb0e3166868bb72e9bc679a6 5704Author: Alex Deucher <alexdeucher@gmail.com> 5705Date: Wed Mar 2 20:44:19 2011 -0500 5706 5707 cayman: first pass at exa/Xv shaders 5708 5709 Main differences with evergreen: 5710 - 4-way rather than 5-way 5711 - END_OF_PROGRAM bit removed from CF istructions, use 5712 CF_INST_END instead. 5713 - MEGA_FETCH* fields removed from VTX commands 5714 - no more VC, all fetches go through the TC 5715 5716 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5717 5718commit 01b646ed800732985c1638b147716641a99082f9 5719Author: Alex Deucher <alexdeucher@gmail.com> 5720Date: Wed Mar 2 20:39:38 2011 -0500 5721 5722 cayman: add a default state function 5723 5724 The rest of the state functions can be shared 5725 with evergreen. I've noted where there are 5726 differences. 5727 5728 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5729 5730commit 42eecc6f4fb1570769490bdaeac06817c6c36a7e 5731Author: Alex Deucher <alexdeucher@gmail.com> 5732Date: Wed Mar 2 20:20:56 2011 -0500 5733 5734 cayman: add 3D register headers 5735 5736 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5737 5738commit e1d28e011f4a5139cbc778973c63158ed2746716 5739Author: Alex Deucher <alexdeucher@gmail.com> 5740Date: Wed Mar 2 20:13:50 2011 -0500 5741 5742 kms/cayman: stub out exa support 5743 5744 Just fallbacks for now. 5745 5746 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5747 5748commit 21e44a20b8b1b64079ee77f45aaa5010206ed7b6 5749Author: Michel Dänzer <daenzer@vmware.com> 5750Date: Thu May 12 09:23:38 2011 +0200 5751 5752 UMS: Fix comparison of unsigned variable against < 0. 5753 5754 Pointed out by clang: 5755 5756 ../../src/radeon_crtc.c:242:18: error: comparison of unsigned expression < 0 is always false [-Werror,-Wtautological-compare] 5757 error = error < 0 ? 0xffffffff : error; 5758 ~~~~~ ^ ~ 5759 5760 If a UMS regression is bisected to this commit, the assignment should probably 5761 just be removed, as it's a no-op in the current form. 5762 5763commit 3b893d81982c9381393c92625e308541e0071b05 5764Author: Michel Dänzer <daenzer@vmware.com> 5765Date: Fri May 13 09:10:02 2011 +0200 5766 5767 KMS: Fix output properties logic error. 5768 5769 Pointed out by clang: 5770 5771 ../../src/drmmode_display.c:1023:30: error: use of logical && with constant operand; switch to bitwise & or remove constant [-Werror,-Wconstant-logical-operand] 5772 if (props && (props->flags && DRM_MODE_PROP_ENUM)) { 5773 ^ ~~~~~~~~~~~~~~~~~~ 5774 5775 Reviewed-by: Alex Deucher <alexdeucher@gmail.com> 5776 5777commit f83d58cf5b33686139067f8f898b8e566ba5c253 5778Author: Nicolas Kaiser <nikai@nikai.net> 5779Date: Fri May 13 00:56:31 2011 +0200 5780 5781 man: fix typos 5782 5783 Signed-off-by: Nicolas Kaiser <nikai@nikai.net> 5784 5785commit 90abffbd30f44b9cf76a6e28103ddcb5419b4522 5786Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 5787Date: Fri May 6 09:45:23 2011 -0400 5788 5789 DRI2: fix high-crtc/vblank oversight/bug 5790 5791 improvements to high-crtc handling done in 5792 f0b7d7b449cc77bb2b281d81108507f8bc2e6018 introduced a bug that caused 5793 the populate_vbl_request_type to never use the high-crtc field even 5794 when it should. The reason is that the offending patch put the code 5795 under #ifdef DRM_VBLANK_HIGH_CRTC_MASK which is not visible outside the 5796 enum type, so #else was always taken in compilation type. This patch 5797 fixes it by basing #ifdef on (pre-processor visible) 5798 DRM_VBLANK_HIGH_CRTC_SHIFT constant 5799 5800 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 5801 5802commit 62a4cd180fe884dca24586d453395472516e6496 5803Author: Alex Deucher <alexdeucher@gmail.com> 5804Date: Wed May 4 01:13:55 2011 -0400 5805 5806 fusion: fix tiling enable logic 5807 5808 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5809 5810commit 76638ca687b02d3b1494b9868f817fd4fd892c64 5811Author: Alex Deucher <alexdeucher@gmail.com> 5812Date: Wed May 4 01:06:22 2011 -0400 5813 5814 fusion: enable tiling if DFS works 5815 5816 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5817 5818commit a6d2dba6573a3512d550d7e442bf42ea03012bbc 5819Author: Dave Airlie <airlied@redhat.com> 5820Date: Wed May 4 10:44:43 2011 +1000 5821 5822 radeon: add add hw DFS support for fusion 5823 5824 Fusion had a bug setting up the VM on earlier kernels so we need to work 5825 around that and only enable accel on a new enough kernel. 5826 5827 Signed-off-by: Dave Airlie <airlied@redhat.com> 5828 5829commit 859e052af49e68a826b77a9135c7f067dc331a06 5830Author: Alex Deucher <alexdeucher@gmail.com> 5831Date: Tue May 3 15:15:04 2011 -0400 5832 5833 radeon: add some new pci ids 5834 5835 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5836 5837commit 8f8bbf628c6eed037f57bc8c155f0ecdacbebad1 5838Author: Alex Deucher <alexdeucher@gmail.com> 5839Date: Sun May 1 13:19:15 2011 -0400 5840 5841 man: add cayman to man page 5842 5843 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5844 5845commit 903e90c31cf0319be9297529aa7b8daa1756cf63 5846Author: Alex Deucher <alexdeucher@gmail.com> 5847Date: Wed Apr 20 03:10:08 2011 -0400 5848 5849 EXA/Xv: used cached bo tiling flags for accel setup on 6xx+ 5850 5851 This avoids calling into the kernel for each bo in the accel 5852 code. This is a follow on to: 5853 cc7d1fa39da40a532fcdbe6c7924ca47a879e66a 5854 5855 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5856 5857commit 982c22f16c8eeee9be81779fbfe17d8d3f9b6897 5858Author: Adam Jackson <ajax@redhat.com> 5859Date: Thu Apr 14 16:04:50 2011 -0400 5860 5861 R520: Fix textures larger than 2k 5862 5863 Ported from the equivalent fix in Mesa. 5864 5865commit cc7d1fa39da40a532fcdbe6c7924ca47a879e66a 5866Author: Michel Dänzer <daenzer@vmware.com> 5867Date: Mon Apr 4 17:37:12 2011 +0200 5868 5869 EXA: Cache BO tiling flags. 5870 5871 Calling into the kernel every time is quite expensive, and nobody else should 5872 ever change the tiling flags. 5873 5874 There's still more to do along the same lines for >= R6xx. 5875 5876commit f0b7d7b449cc77bb2b281d81108507f8bc2e6018 5877Author: Michel Dänzer <daenzer@vmware.com> 5878Date: Tue Apr 5 13:36:01 2011 +0200 5879 5880 DRI2: Some cleanups for the scheduling mess. 5881 5882 * Fix build against libdrm that doesn't define *_VBLANK_HIGH_CRTC*. 5883 * If we have more than two CRTCs but can't use DRM_VBLANK_HIGH_CRTC_MASK, don't 5884 enable scheduling in the first place rather than relying on 5885 DRM_VBLANK_SECONDARY magically doing something sensible for higher CRTCs. 5886 * Only set up client state tracking when scheduling is enabled. 5887 * Only declare pRADEONEnt when it's needed, and break long lines. 5888 5889commit 7acf9bc833de539fa2259a051c66a99445a54bc4 5890Author: Alex Deucher <alexdeucher@gmail.com> 5891Date: Mon Apr 4 11:08:37 2011 -0400 5892 5893 radeon: add some additional ontario pci ids 5894 5895 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5896 5897commit 0af6386f85a0f1ba14864a3334164733a10a6cb8 5898Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 5899Date: Thu Mar 24 13:33:27 2011 -0400 5900 5901 xf86-video-ati: (revised #2) add support for vblank on crtc > 1 5902 5903 Hi Alex, 5904 5905 Enclosed is a revised version of two patches sent on Mar 18 and Mar 22, 5906 respectively. Details summarized in these two threads: 5907 http://lists.freedesktop.org/archives/dri-devel/2011-March/009463.html 5908 http://lists.freedesktop.org/archives/dri-devel/2011-March/009582.html 5909 5910 This patch reconciles the DDX with the change in libdrm sent to this list 5911 earlier today. Specifically, it refers to a symbol that has been renamed 5912 from DRM_CAP_HIGH_CRTC to DRM_CAP_VBLANK_HIGH_CRTC. It *supersedes* the 5913 previous patch (i.e. apply it to the master branch as it exists at the 5914 time of this writing, not as an incremental patch to the one sent previously). 5915 5916 Regards, 5917 5918 Ilija 5919 5920 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 5921 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5922 5923commit fe2e0ad3ffa58f40311319c950b842e2928a5740 5924Author: matthew green <mrg@eterna.com.au> 5925Date: Mon Mar 21 12:17:58 2011 -0400 5926 5927 bug fix for r6xx/r7xx UMS 5928 5929 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5930 5931commit 83978ad5fe37581e2b2f3fbd9c073d91b2ae1d50 5932Author: Alex Deucher <alexdeucher@gmail.com> 5933Date: Thu Mar 17 19:07:43 2011 -0400 5934 5935 bump version post release 5936 5937 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5938 5939commit 38d9368e59b2990bf32a028ece2132451b402350 5940Author: Alex Deucher <alexdeucher@gmail.com> 5941Date: Thu Mar 17 18:59:55 2011 -0400 5942 5943 bump for release 5944 5945 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5946 5947commit cdfc007eccc9b130fc08012deef304488eb6e469 5948Author: Dave Airlie <airlied@redhat.com> 5949Date: Wed Mar 16 10:40:42 2011 +1000 5950 5951 radeon/exa: correct function name 5952 5953 this corrects the function name so it matches the contents. 5954 5955 Signed-off-by: Dave Airlie <airlied@redhat.com> 5956 5957commit b93d7658f669f6dc1cfacebcfe955a1e113a537c 5958Author: Dave Airlie <airlied@redhat.com> 5959Date: Wed Mar 16 10:37:22 2011 +1000 5960 5961 radeon/exa: fix scaling check 5962 5963 This caused a regression in firefox, as these numbers are xFixed 5964 values hence 1 is actually 65536. 5965 5966 Should fix: https://bugzilla.redhat.com/show_bug.cgi?id=685084 5967 5968 Signed-off-by: Dave Airlie <airlied@redhat.com> 5969 5970commit 8567b8cd9e136ef69906e02286b3752db2404741 5971Author: Alex Deucher <alexdeucher@gmail.com> 5972Date: Tue Mar 15 13:13:06 2011 -0400 5973 5974 APU: no tiling yet 5975 5976 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5977 5978commit 4d350497012fa31a417ada662006e2d64db2a4b5 5979Author: Dave Airlie <airlied@redhat.com> 5980Date: Tue Mar 15 10:32:19 2011 +1000 5981 5982 radeon: exa shaders don't handle scaling either. 5983 5984 rendercheck tsrccoords test fails. 5985 5986 Signed-off-by: Dave Airlie <airlied@redhat.com> 5987 5988commit 6319a33cb7bd8dba1eaf47bee1e2608cee5d7600 5989Author: Alex Deucher <alexdeucher@gmail.com> 5990Date: Wed Mar 2 19:38:04 2011 -0500 5991 5992 kms: add cayman pci ids 5993 5994 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5995 5996commit fd65def45a57e3242596274d270791df8abc8e29 5997Author: Alex Deucher <alexdeucher@gmail.com> 5998Date: Thu Jan 13 17:02:31 2011 -0500 5999 6000 kms: no accel yet for cayman 6001 6002 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6003 6004commit 585e2a3b803b6d3e070f2427e9cb49bab0949239 6005Author: Alex Deucher <alexdeucher@gmail.com> 6006Date: Thu Jan 13 16:58:01 2011 -0500 6007 6008 kms: add cayman chip family 6009 6010 KMS only. No UMS support for cayman. 6011 6012 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6013 6014commit 6835e22256ba7a74c02504383f63baaba7e87156 6015Author: Alex Deucher <alexdeucher@gmail.com> 6016Date: Wed Mar 2 12:20:45 2011 -0500 6017 6018 r6xx+: truncate point sampled coordinates 6019 6020 By default the hardware rounds texcoords. However, 6021 for point sampled textures, the expected behavior is 6022 to truncate. When we have point sampled textures, 6023 set the truncate bit in the sampler. 6024 6025 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6026 6027commit 8a8862e4165dc5be779bce5e327a31da4826db3b 6028Author: Sedat Dilek <sedat.dilek@googlemail.com> 6029Date: Sat Feb 26 21:04:22 2011 +0100 6030 6031 kms: Fix warning XNFprintf is deprecated (v2) 6032 6033 The first take of my patch missed the breakage against pre-1.10 6034 xservers and now I know XNFasprintf() is not a 1:1 replacement 6035 for XNFprintf(). 6036 The idea for the fix seemed to be so obvious, but as usual "really" 6037 test the patch you are sending to any ML. 6038 I am awfully sorry for the overhasty bad hack. 6039 Thanks to Alan and Alex for the help to fix it correctly, now. 6040 6041 Tested-By: Sedat Dilek <sedat.dilek@gmail.com> 6042 Signed-off-by: Sedat Dilek <sedat.dilek@gmail.com> 6043 6044commit 88304fe9c0ce69012d44a77a5368c35236511dac 6045Author: Alex Deucher <alexdeucher@gmail.com> 6046Date: Mon Feb 28 11:29:19 2011 -0500 6047 6048 kms: don't rotate if acceleration is not enabled 6049 6050 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6051 6052commit 99141761e947e2153ed1a3bc485dc5716a22acb1 6053Author: Alex Deucher <alexdeucher@gmail.com> 6054Date: Mon Feb 28 11:20:44 2011 -0500 6055 6056 kms: don't enable tiling if accel is off 6057 6058 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6059 6060commit 7928c120fb629a782846bcd5734d9fa14cbb099d 6061Author: Dave Airlie <airlied@redhat.com> 6062Date: Mon Feb 28 14:11:57 2011 +1000 6063 6064 xv: fix height alignments for U/V planes 6065 6066 The kernel CS checks showed we were incorrectly aligning the 6067 U/V allocations for when we used them as texture sources. 6068 6069 This should fix 6070 https://bugs.freedesktop.org/show_bug.cgi?id=34567 6071 6072 Signed-off-by: Dave Airlie <airlied@redhat.com> 6073 6074commit e55520c0e582f2a500b3b01c2cb78750e15256ca 6075Author: Michel Dänzer <daenzer@vmware.com> 6076Date: Sat Feb 26 15:27:59 2011 +0100 6077 6078 Revert "kms: Fix warning XNFprintf is deprecated" 6079 6080 This reverts commit cb4dc7ab66016d0c980800f8dbf71bbe993889d3. 6081 6082 Looks like this caused https://bugs.freedesktop.org/attachment.cgi?id=43852 , 6083 but even if it didn't, the following needs to be addressed first: 6084 6085 cc1: warnings being treated as errors 6086 ../../src/radeon_kms.c: In function ‘radeon_open_drm_master’: 6087 ../../src/radeon_kms.c:413: error: passing argument 1 of ‘XNFasprintf’ from incompatible pointer type 6088 /usr/local/include/xorg/Xprintf.h:57: note: expected ‘char **’ but argument is of type ‘char *’ 6089 ../../src/radeon_kms.c:413: error: passing argument 2 of ‘XNFasprintf’ makes pointer from integer without a cast 6090 /usr/local/include/xorg/Xprintf.h:57: note: expected ‘const char * restrict’ but argument is of type ‘uint16_t’ 6091 ../../src/radeon_kms.c:413: error: assignment makes pointer from integer without a cast 6092 make[2]: *** [radeon_kms.lo] Error 1 6093 6094commit cb4dc7ab66016d0c980800f8dbf71bbe993889d3 6095Author: Sedat Dilek <sedat.dilek@gmail.com> 6096Date: Fri Feb 25 17:21:52 2011 -0500 6097 6098 kms: Fix warning XNFprintf is deprecated 6099 6100 Noticed when building ati/radeon ddx against xserver 1.10-rc3. 6101 6102 [ build.log ] 6103 ../../src/radeon_kms.c:412:5: warning: 'XNFprintf' is deprecated (declared at /usr/include/xorg/os.h:273) 6104 6105 agd5f: don't break the build against pre-1.10 xservers. 6106 6107 Signed-off-by: Sedat Dilek <sedat.dilek@gmail.com> 6108 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6109 6110commit ecfdb209afe2aafc378baab8c511f5df7b000270 6111Author: Sedat Dilek <sedat.dilek@googlemail.com> 6112Date: Fri Feb 25 21:48:14 2011 +0100 6113 6114 UMS: Fix build against xserver 1.10-rc3 6115 6116 This issue was introduced due to last minute backout of RandR-1.4 6117 in xserver 1.10-rc3. 6118 6119 Switch to "#ifdef RANDR_14_INTERFACE" as suggested by Keith Packard. 6120 See also <http://lists.x.org/archives/xorg-devel/2011-February/019643.html>. 6121 6122 Note: 6123 The ddx needs a rebuild as the X video driver ABI changed to version 10.0. 6124 6125 Reported-by: Alex Deucher <alexdeucher@gmail.com> 6126 CC: Keith Packard <keithp@keithp.com> 6127 Signed-off-by: Sedat Dilek <sedat.dilek@gmail.com> 6128 6129commit 04293a3a84acd96ccdf8e5e865cf973325ca43a1 6130Author: Alex Deucher <alexdeucher@gmail.com> 6131Date: Thu Feb 24 02:33:41 2011 -0500 6132 6133 dri2: disable pageflipping for transformed displays 6134 6135 Based on Ben's nouveau patch. 6136 6137 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6138 6139commit 91070cfd75d5607c4a72ace780f830f0ddb40e84 6140Author: Alex Deucher <alexdeucher@gmail.com> 6141Date: Wed Feb 23 00:55:28 2011 -0500 6142 6143 Xv: use aligned height to calculate Xv buffer size 6144 6145 Make sure it's large enough for the tiling mode so the 6146 CS checker doesn't complain. Should fix: 6147 https://bugs.freedesktop.org/show_bug.cgi?id=34567 6148 6149 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6150 6151commit acd54a48a6680956ed903fc890204c0c910b6712 6152Author: Mark Kettenis <kettenis@openbsd.org> 6153Date: Sun Feb 20 22:51:04 2011 +0100 6154 6155 Properly restore CLOCK_CNTL_INDEX register 6156 6157 RADEONRestore() restores the CLOCK_CNTL_INDEX register before calling 6158 RADEONRestorePLLRegisters(), which modifies the CLOCK_CNTL_INDEX register 6159 to access the PLL registers. As a result we may end up with the wrong clock 6160 being selected when exiting X. This happens on platforms where the driver 6161 doesn't save and restore the VGA state. 6162 6163 Signed-off-by: Mark Kettenis <kettenis@openbsd.org> 6164 6165commit d7c2ba4286ea32401c0026cb92ec167f7b5753d8 6166Author: Alex Deucher <alexdeucher@gmail.com> 6167Date: Wed Feb 16 18:30:34 2011 -0500 6168 6169 man: add btc chips to the list 6170 6171 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6172 6173commit d0f8bffbbc26818eccdd305ef142387dc4975bba 6174Author: Alex Deucher <alexdeucher@gmail.com> 6175Date: Wed Feb 16 15:09:20 2011 -0500 6176 6177 kms: add tiling support for evergreen/NI 6178 6179 - requires 2.6.37 or newer 6180 - 1D only at the moment 6181 - disabled by default 6182 6183 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6184 6185commit 0471d8412acd82e281a35fc4c6bb2d53b1ff5802 6186Author: Alex Deucher <alexdeucher@gmail.com> 6187Date: Wed Feb 16 15:07:35 2011 -0500 6188 6189 kms: EXA/Xv tiling fixes 6190 6191 - properly set tiling flags for temp surfaces 6192 - fix CB non_disp_tiling bits on evergreen 6193 6194 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6195 6196commit 4d7e1498f7d9eb50e2eddabca193fc27bde24f0e 6197Author: Alex Deucher <alexdeucher@gmail.com> 6198Date: Tue Feb 15 11:24:23 2011 -0500 6199 6200 fix Xv on pre-r6xx asics 6201 6202 broken by: 6203 a9c1358099:"Xv: fix textured video alignment" 6204 6205 Reported by: 6206 Artur Skawina <art.08.09@gmail.com> 6207 6208 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6209 6210commit a9a59717d11af37a2dda5555f6a83c5b65449527 6211Author: Alex Deucher <alexdeucher@gmail.com> 6212Date: Sat Feb 12 05:34:50 2011 -0500 6213 6214 kms: use worst case base/pitch align if we don't have drm tiling info 6215 6216 To avoid CS rejection. 6217 6218 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6219 6220commit 5f9b7fdb4ce27f8c6a64134bb4d5ebeb1c958cca 6221Author: Alex Deucher <alexdeucher@gmail.com> 6222Date: Sat Feb 12 05:19:12 2011 -0500 6223 6224 kms: always get the tiling info even if we aren't tiling 6225 6226 We need the proper group size to calculate alignment so the 6227 ddx and drm agree. 6228 6229 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6230 6231commit 355f0f7a67ce239b7fd2f2e81bcaaa88b2a09f03 6232Author: Alex Deucher <alexdeucher@gmail.com> 6233Date: Sat Feb 12 04:52:52 2011 -0500 6234 6235 kms: fix rotate pitch align 6236 6237 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6238 6239commit f1dc419c989addc4737aed06ec8b8acdb4d40063 6240Author: Alex Deucher <alexdeucher@gmail.com> 6241Date: Fri Feb 11 17:21:10 2011 -0500 6242 6243 kms: evergreen/ni big endian accel support 6244 6245 Based on 6xx/7xx patches from Cédric Cano. 6246 6247 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6248 6249commit e8dc728a549323f1babe337b9d42ad504af1ca39 6250Author: Cédric Cano <ccano@interfaceconcept.com> 6251Date: Fri Feb 11 17:06:47 2011 -0500 6252 6253 ums: atombios endian fixes 6254 6255 agd5f: fix a few more cases. 6256 6257 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6258 6259commit 48ffad957f1dbca909515ffa00629f4caa68706b 6260Author: Cédric Cano <ccano@interfaceconcept.com> 6261Date: Fri Feb 11 17:00:31 2011 -0500 6262 6263 kms: 6xx/7xx big endian accel support 6264 6265 agd5f: minor cleanups 6266 6267 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6268 6269commit 151b22bd7c3b1002a7261538611fb2b468815c86 6270Author: Cédric Cano <ccano@interfaceconcept.com> 6271Date: Fri Feb 11 16:33:16 2011 -0500 6272 6273 kms/6xx+: endian swap cursor uploads 6274 6275 agd5f: remove kms surface for cursor since 6276 we now do the swap in the cursor load to avoid 6277 breaking pre-6xx cards with BE. 6278 6279 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6280 6281commit 309fac041f998e20338d320aabaeda5cc1bb6348 6282Author: Alex Deucher <alexdeucher@gmail.com> 6283Date: Fri Feb 11 14:32:45 2011 -0500 6284 6285 kms/exa/xv: fix alignment checking in accel state setup 6286 6287 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6288 6289commit 25fe3ec4c6dfcdf6ba72a032bc9cb2c84d84029a 6290Author: Alex Deucher <alexdeucher@gmail.com> 6291Date: Fri Feb 11 14:03:56 2011 -0500 6292 6293 kms/exa: UTS/DFS base alignment fixes 6294 6295 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6296 6297commit 3d10278ce511f5dabb68ed86ee43eaaf43585983 6298Author: Alex Deucher <alexdeucher@gmail.com> 6299Date: Fri Feb 11 13:51:11 2011 -0500 6300 6301 Xv: fix textured video alignment harder 6302 6303 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6304 6305commit a9c135809945fb372ce390f06918b9fd8e455b3f 6306Author: Alex Deucher <alexdeucher@gmail.com> 6307Date: Fri Feb 11 13:21:43 2011 -0500 6308 6309 Xv: fix textured video alignment 6310 6311 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6312 6313commit e5bd99faa3b6629a55168386d5dfa936ee4e97ae 6314Author: Alex Deucher <alexdeucher@gmail.com> 6315Date: Fri Feb 11 12:14:55 2011 -0500 6316 6317 kms: remove RADEON_TILING_SURFACE flag on front bo 6318 6319 Should have been removed in: 6320 ef9bfb262db7004bef3704e5d914687e50d3fca4 6321 6322 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6323 6324commit 92ffce1892d24a98d789aad57a4230cadb0c812a 6325Author: Alex Deucher <alexdeucher@gmail.com> 6326Date: Fri Feb 11 03:12:43 2011 -0500 6327 6328 EXA/r6xx+: properly account for height alignment in copy temp buffer 6329 6330 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6331 6332commit 39104c6e8461cf49c1bb03a18858ad75a9d98b46 6333Author: Alex Deucher <alexdeucher@gmail.com> 6334Date: Thu Feb 10 14:31:43 2011 -0500 6335 6336 remove EVERGREENSetAccelState() 6337 6338 It was a duplicate of the R600 variant. 6339 6340 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6341 6342commit 7ac3a2e0bcdadff7c7172a9f833f526b526da16b 6343Author: Alex Deucher <alexdeucher@gmail.com> 6344Date: Thu Feb 10 14:24:50 2011 -0500 6345 6346 6xx+: switch to linear aligned rather than linear general 6347 6348 linear aligned is supposedly more performant, but more 6349 importantly, linear general only works on the CB without 6350 slices. The texture blocks technically don't support 6351 linear general although, I think linear general gets 6352 upgraded to linear aligned in the hw which is why it 6353 currently works. 6354 6355 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6356 6357commit e3145801b80fd4be4cf770128876e86e89bda66f 6358Author: Alex Deucher <alexdeucher@gmail.com> 6359Date: Thu Feb 10 14:14:55 2011 -0500 6360 6361 evergreen/NI: consolidate spi setup 6362 6363 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6364 6365commit be67ded05621aff9c85525372fd119071d3278ec 6366Author: Alex Deucher <alexdeucher@gmail.com> 6367Date: Thu Feb 10 14:06:38 2011 -0500 6368 6369 6xx/7xx: consolidate spi setup 6370 6371 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6372 6373commit 2c5ae1724307e0dba5d0306fe27c1e15a7390a2f 6374Author: Alex Deucher <alexdeucher@gmail.com> 6375Date: Thu Feb 10 13:47:53 2011 -0500 6376 6377 evergreen/ni: consolidate CB state handling 6378 6379 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6380 6381commit c52d817b51e13447802fe338be2d247ffba2a669 6382Author: Alex Deucher <alexdeucher@gmail.com> 6383Date: Thu Feb 10 13:35:02 2011 -0500 6384 6385 6xx/7xx: consolidate remaining CB state 6386 6387 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6388 6389commit 5f9bc127d302404432b631e3e774192950f57b42 6390Author: Alex Deucher <alexdeucher@gmail.com> 6391Date: Thu Feb 10 03:28:01 2011 -0500 6392 6393 kms/r6xx+: clean up pitch/height alignment in EXA UTS/DFS 6394 6395commit a4899db96029acde6cd400fc0541693a487898e7 6396Author: Alex Deucher <alexdeucher@gmail.com> 6397Date: Sat Feb 5 14:11:42 2011 -0500 6398 6399 UMS: fix spelling in error message 6400 6401 Noticed by: Warren Block <wblock@wonkity.com> 6402 6403 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6404 6405commit 84450ea9221f32296c1f786a6aa126a08bb9756f 6406Author: Alex Deucher <alexdeucher@gmail.com> 6407Date: Sat Feb 5 14:05:48 2011 -0500 6408 6409 UMS/DCE3.2: fix segfault 6410 6411 atombios_pick_dig_encoder() needs to be called 6412 with a valid crtc assigned. 6413 6414 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6415 6416commit a27b5dbd9fba689f1814c0888d1b7459b6c6b1a1 6417Author: Alex Deucher <alexdeucher@gmail.com> 6418Date: Thu Feb 3 19:06:22 2011 -0500 6419 6420 Bump version post release 6421 6422 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6423 6424commit 0a1a0513a61f392580bde39cca4880f2c19abc8d 6425Author: Alex Deucher <alexdeucher@gmail.com> 6426Date: Thu Feb 3 18:09:17 2011 -0500 6427 6428 bump version for release 6429 6430 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6431 6432commit ef9bfb262db7004bef3704e5d914687e50d3fca4 6433Author: Alex Deucher <alexdeucher@gmail.com> 6434Date: Tue Feb 1 19:24:44 2011 -0500 6435 6436 kms/pre-6xx: fix pageflipping with tiling 6437 6438 Use UTS/DFS to tile/untile as appropriate for sw access on 6439 pre-6xx. 6440 6441 Fixes: 6442 https://bugs.freedesktop.org/show_bug.cgi?id=33738 6443 6444 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6445 6446commit bb16dd77321e5a64a3cb2d2ca9982117799ac1a8 6447Author: Alex Deucher <alexdeucher@gmail.com> 6448Date: Tue Feb 1 19:20:00 2011 -0500 6449 6450 kms/r6xx+: fix tiling and pageflipping harder 6451 6452 Thanks for Michel for final fix. Fixes: 6453 https://bugs.freedesktop.org/show_bug.cgi?id=33738 6454 for r6xx+ asics. A similar approach for pre-r6xx 6455 asics is pending. 6456 6457 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6458 6459commit 9c124f781049ef11a8b19894a29b7f62975b3011 6460Author: Alex Deucher <alexdeucher@gmail.com> 6461Date: Tue Feb 1 18:59:40 2011 -0500 6462 6463 radeon: remove device 0x4243 6464 6465 0x4243 is a PCI bridge, not a gpu. 6466 6467 See: 6468 https://bugs.freedesktop.org/show_bug.cgi?id=33815 6469 6470 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6471 6472commit 66eb81b62e5ae8e1d7bd44ed8a179e5ec1ca69af 6473Author: Michel Dänzer <daenzer@vmware.com> 6474Date: Mon Jan 31 12:26:20 2011 +0100 6475 6476 UMS: Slightly improve xserver version check. 6477 6478 This will ensure the xserver definitions necessary to compile the new code are 6479 there. xf86CrtcSetMode will be unresolved at runtime with a few xserver 6480 commits, but I'm afraid this is the best we can do. 6481 6482 See https://bugs.freedesktop.org/show_bug.cgi?id=33719 . 6483 6484commit 3dc28c86eb57cac819e4ed650acfe1f7df6ef565 6485Author: Alex Deucher <alexdeucher@gmail.com> 6486Date: Thu Jan 27 13:09:26 2011 -0500 6487 6488 ums: fix build against xserver >= 1.10 6489 6490 xf86CrtcSetMode was removed in: 6491 http://cgit.freedesktop.org/xorg/xserver/commit/?id=8b35118c03590a7ad3786d3284bafb3f40fcb8cc 6492 6493commit f58e1354b78bf6b70120bddfe1566da3b0723f72 6494Author: Jerome Glisse <jglisse@redhat.com> 6495Date: Wed Jan 26 16:13:30 2011 -0500 6496 6497 dp: fix displayport support by syncing with KMS code 6498 6499 Warning the dp clock value are divided by 10 in ddx (10 times 6500 bigger than kernel value) this is somethings very picky. 6501 6502 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 6503 6504commit 0f90133ac61c5e1d80b6fa9256e079349940029f 6505Author: Alex Deucher <alexdeucher@gmail.com> 6506Date: Mon Jan 24 12:40:22 2011 -0500 6507 6508 vbo: warn if we add a bo with a pointer to the wait list 6509 6510commit 42529603ecf86fcfd0a8f3495d9db9ce1ee9b4c3 6511Author: Alex Deucher <alexdeucher@gmail.com> 6512Date: Mon Jan 24 12:30:56 2011 -0500 6513 6514 vbo: always flush the cbuf bo in flush_indirect() 6515 6516 Always flush the cbuf in case we end up with a cbuf mapped in 6517 Prepare*(), but never end up issuing a draw call since the cbuf 6518 may be in use by multiple ops. 6519 6520 The CS check for the cbuf bo is no longer necessay because 6521 the cbuf bo is NULL at this point due to the radeon_vbo_put() 6522 just prior. 6523 6524 Fixes: 6525 https://bugs.freedesktop.org/show_bug.cgi?id=32188 6526 6527commit 9b0e03e98ab739efb4031f81cc4a1a50b3d87a42 6528Author: Alex Deucher <alexdeucher@gmail.com> 6529Date: Mon Jan 24 00:13:25 2011 -0500 6530 6531 vbo: avoid extra cs revalidation in vbo alloc 6532 6533 Make sure we have a valid bo before revalidating. 6534 6535commit fadee0409a8e13b78bbccb83dd70f590fee23d57 6536Author: Alex Deucher <alexdeucher@gmail.com> 6537Date: Fri Jan 21 17:30:28 2011 -0500 6538 6539 vbo: balance radeon_vbo_get() and radeon_vbo_put() 6540 6541 bo_ref() and bo_map() in radeon_vbo_get() 6542 bo_unmap() and bo_unref() in radeon_vbo_put() 6543 6544 rather than doing the bo_map() separately in 6545 radeon_vbo_space(). 6546 6547commit af7d81625a8cf873e6efc881489b3eda9861bd03 6548Author: Alex Deucher <alexdeucher@gmail.com> 6549Date: Fri Jan 21 17:01:25 2011 -0500 6550 6551 evergreen: use vb_offset rather than vb_start_op for cbuf offset 6552 6553commit 4817fac5f728b777939e2e2bee16b842c9e1367b 6554Author: Alex Deucher <alexdeucher@gmail.com> 6555Date: Fri Jan 21 15:59:32 2011 -0500 6556 6557 vbo: remove useless radeon_vb_discard() 6558 6559 We already reset vb_start_op to -1 in the 6560 UMS/KMS ib discard functions. 6561 6562commit 0a03f03a65aad925ba2d9c76b1d3356184607bf9 6563Author: Alex Deucher <alexdeucher@gmail.com> 6564Date: Thu Jan 20 17:07:32 2011 -0500 6565 6566 evergreen/ni: fix Xv artifacts 6567 6568 Port of the 6xx/7xx fix to evergreen. Bad texture size 6569 for texture cache flush. 6570 6571commit f41cf83750ba9a2e0797fdb21ea9104b85ce53aa 6572Author: Marton Balint <cus@fazekas.hu> 6573Date: Thu Jan 20 17:04:56 2011 -0500 6574 6575 r6xx/7xx: fix Xv artifacts 6576 6577 bad textures size for cache flushes. 6578 6579 Fixes: 6580 https://bugs.freedesktop.org/show_bug.cgi?id=22007 6581 6582commit 57fbddfc21d8c6794f378489b764cc2a0ad4a48c 6583Author: Michel Dänzer <daenzer@vmware.com> 6584Date: Tue Jan 18 16:07:52 2011 +0100 6585 6586 Fix crash freeing KMS video memory. 6587 6588 Where's that brown paper bag? :} 6589 6590commit 6548bb9836253c586023ffe5ad1497ddabaa50fc 6591Author: Michel Dänzer <daenzer@vmware.com> 6592Date: Tue Jan 18 10:23:41 2011 +0100 6593 6594 Fix KMS textured video leaks (bug #33193). 6595 6596 v2: Fix radeon_legacy_free_memory() argument type error pointed out by 6597 Marton Balint, refactor video memory freeing logic into helper function. 6598 6599 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=33193 . 6600 6601commit edc3496b55577ee8509ddd9188e6f2bcdf7169a1 6602Author: Dave Airlie <airlied@redhat.com> 6603Date: Mon Jan 17 07:55:29 2011 +1000 6604 6605 radeon: add back flush in block handler. 6606 6607 this is also needed for certain things like stipple rendering. 6608 6609 Signed-off-by: Dave Airlie <airlied@redhat.com> 6610 6611commit 63d7dece3f4be8b14012b3a2bedd850831c0437b 6612Author: Alex Deucher <alexdeucher@gmail.com> 6613Date: Thu Jan 13 11:24:26 2011 -0500 6614 6615 rs880 fix typo in HD 4250 string 6616 6617 Noticed by Nigel Taylor 6618 6619 Fixes: 6620 https://bugs.freedesktop.org/show_bug.cgi?id=33057 6621 6622commit c5b3db18d888552328e9718ea022794fc5bde352 6623Author: Alex Deucher <alexdeucher@gmail.com> 6624Date: Tue Jan 11 16:21:18 2011 -0500 6625 6626 kms: fix pitch aligment for scanout 6627 6628 Display has slightly stricter pitch alignment requirements 6629 than other blocks. Factor that in when aligning pitch. 6630 6631 Fixes: 6632 https://bugs.freedesktop.org/show_bug.cgi?id=32997 6633 6634commit bbd7adce889359b5eb3239b73e904b3ede283e12 6635Author: Alex Deucher <alexdeucher@gmail.com> 6636Date: Tue Jan 11 15:41:03 2011 -0500 6637 6638 radeon: fix yet another pitch align 6639 6640commit af2e6d7d2f1b3d8f8f6b0acfb2b7b0cfaff7bcdb 6641Author: Alex Deucher <alexdeucher@gmail.com> 6642Date: Tue Jan 11 14:42:17 2011 -0500 6643 6644 radeon: fix pitch align in pageflip code 6645 6646commit e5d0a400d08da2358fac9c2ad12042f125525736 6647Merge: 0e432df e27e9b4 6648Author: Alex Deucher <alexdeucher@gmail.com> 6649Date: Mon Jan 10 18:16:49 2011 -0500 6650 6651 Merge branch 'kms-pflip' of git+ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 6652 6653commit 0e432dff9e06a183acaeb20db29cbd03ff0f4b82 6654Author: Alex Deucher <alexdeucher@gmail.com> 6655Date: Thu Jan 6 20:56:45 2011 -0500 6656 6657 NI: add pci ids 6658 6659commit 34d61785b9bc13287bd7ab8bdd8a1b99a3df4440 6660Author: Alex Deucher <alexdeucher@gmail.com> 6661Date: Tue Nov 23 20:54:57 2010 -0500 6662 6663 Add EXA/Xv acceleration support for NI chips 6664 6665commit 97322c00faf08461b941edf97efe86d8b082a0ce 6666Author: Alex Deucher <alexdeucher@gmail.com> 6667Date: Wed Nov 24 12:18:04 2010 -0500 6668 6669 Bail on NI cards with UMS 6670 6671 NI cards require KMS. 6672 6673commit c678b79f99238473df6ff1bedbae840950cdf88c 6674Author: Alex Deucher <alexdeucher@gmail.com> 6675Date: Tue Nov 23 20:45:25 2010 -0500 6676 6677 add NI family ids 6678 6679commit e27e9b4e50ad42885ad2e25be897cdf29aa59712 6680Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 6681Date: Sun Dec 19 03:56:17 2010 +0100 6682 6683 ddx/ati: Bugfix for pageflip consistency check. 6684 6685 target_msc wasn't passed from vblank event handler 6686 to pageflip routine, due to a missing initalization. 6687 Now fixed. 6688 6689 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 6690 6691commit 30c5f217f91ed1ee15b5fff0002521d98682efb7 6692Author: Alex Deucher <alexdeucher@gmail.com> 6693Date: Fri Dec 17 13:40:40 2010 -0500 6694 6695 exa/dfs/ontario: always use the CPU for DFS for now 6696 6697 There are issues with rendering to gart memory that haven't 6698 been sorted out completely yet. 6699 6700commit e97079c1b621433b4aa3ecfdd2aa0884aac5df76 6701Author: Alex Deucher <alexdeucher@gmail.com> 6702Date: Thu Dec 16 18:49:55 2010 -0500 6703 6704 DCE4.1: fix DIGx routing 6705 6706 Works more like DCE4.0. With only two DIGx blocks 6707 the routing is hardcoded based on link. 6708 This fixes blank screen issues when changing crtc 6709 routing due to incorrect crtc to dig mapping. 6710 6711 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6712 6713commit 3b1fb12dec7df4e9434b2c51568ac8d997638922 6714Author: Alex Deucher <alexdeucher@gmail.com> 6715Date: Sun Dec 12 19:16:47 2010 -0500 6716 6717 man: minor formatting fixes 6718 6719commit b1af52e522158788a9613d6f1a2309a3987ad59f 6720Author: Alex Deucher <alexdeucher@gmail.com> 6721Date: Tue Dec 7 19:17:09 2010 -0500 6722 6723 man: add rs880 info 6724 6725commit f9bbb26dd97254b66de11bb2abd821aa293ecba5 6726Author: Matt Turner <mattst88@gmail.com> 6727Date: Thu Dec 2 17:00:11 2010 -0500 6728 6729 Add missing call to vbeFree (bug 4417) 6730 6731 Also kill unneeded vbe.h include from radeon_bios.c. 6732 6733 Signed-off-by: Matt Turner <mattst88@gmail.com> 6734 6735commit 9fc16ff3f8806971cb495e8838af1e5f9149d34f 6736Author: Alex Deucher <alexdeucher@gmail.com> 6737Date: Fri Dec 3 02:04:28 2010 -0500 6738 6739 radeon: remove duplicated define 6740 6741 xfixedtofloat macro was defined several times 6742 6743commit f074762fff4253a457cb48519bb33a2e2c90f8b9 6744Author: Alex Deucher <alexdeucher@gmail.com> 6745Date: Fri Dec 3 01:25:41 2010 -0500 6746 6747 radeon/exa: fallback for non-affine transforms 6748 6749 We need to implement a texture lookup with perspective 6750 divide for non-affine transforms. For now just fallback. 6751 Should fix: 6752 https://bugs.freedesktop.org/show_bug.cgi?id=31799 6753 although it appears either EXA or the xserver gets this 6754 wrong too. 6755 6756commit 774b114baa932078f8bf91ea302850a8c54c2c07 6757Author: Nicolas Joly <njoly@pasteur.fr> 6758Date: Thu Dec 2 16:02:44 2010 -0500 6759 6760 ati: remove leftover word in ati.man 6761 6762 fixes: 6763 https://bugs.freedesktop.org/show_bug.cgi?id=28297 6764 6765commit 69639ef377a9d6701cdef902f8a1c5e0b58cf833 6766Author: Alex Deucher <alexdeucher@gmail.com> 6767Date: Thu Dec 2 02:12:24 2010 -0500 6768 6769 radeon/kms: pageflipping man page updates 6770 6771commit f48af8a6cfa1ac665f07b8f9712e94b77bc4f5e9 6772Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 6773Date: Tue Nov 23 04:41:53 2010 +0100 6774 6775 ddx/ati: Fix reporting of pageflip completion events on multi-head. 6776 6777 When a drawable is page-flipped on multiple crtc's (fullscreen 6778 drawable on mirror-mode or multi-head x-screen), only one pageflip 6779 event is finally delivered, after the last participating crtc signals 6780 flip completion, this to avoid visual corruption. 6781 6782 Old code returned vblank count and timestamps of flip completion 6783 of this last crtc, instead of the values of the "master crtc", the 6784 one that was used for initially scheduling/triggering the pagflip 6785 via vblank events. (master = radeon_dri2_drawable_crtc(drawable)) 6786 6787 This patch makes sure that the pageflip completion values of the 6788 "master" crtc are returned, otherwise client applications will 6789 get confused by the random (msc, ust) values returned by whichever 6790 crtc was the last to complete its flip. Without this, the returned 6791 values change randomly and jump forward and backward in time and 6792 count. 6793 6794 The patch also implements a consistency check on returned vblank 6795 count values of pageflip completion. Impossible values are detected, 6796 a x-warning is logged and returned (msc,ust) values are marked invalid, 6797 so clients could perform error handling. Such a warning would indicate 6798 bugs in the pageflip completion routine of future kms drivers or the 6799 ddx to aid driver debugging. 6800 6801 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 6802 6803commit 122536ee0aeb1eef1a9d80d5e464dcb423dc2837 6804Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 6805Date: Mon Nov 22 04:11:07 2010 +0100 6806 6807 ddx/ati: Add option "SwapbuffersWait" to control vsync of DRI2 swaps. 6808 6809 A new optional kms driver option "SwapbuffersWait" is defined 6810 for xorg.conf, which defaults to "on". If "on", DRI2 bufferswaps 6811 will be synchronized to vsync, otherwise not. 6812 6813 This currently only affects copy-swaps, not pageflipped swaps. 6814 It also requires a swap_interval setting of zero by the OpenGL 6815 client. 6816 6817 Ideally, we'd provide a way for dri2 to pass the current swap 6818 interval to the ddx so we could change this dynamically. 6819 6820 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 6821 6822commit 0de680730294bd623f6b3e189faa7b88a09d3a2a 6823Author: Alex Deucher <alexdeucher@gmail.com> 6824Date: Wed Nov 17 17:39:39 2010 -0500 6825 6826 radeon/kms: add pageflip support 6827 6828 requires radeon drm 2.8.0 or higher 6829 6830 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6831 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 6832 6833commit fccdca8db34010f566bd068c74cdef0f4a8cb7f5 6834Author: Alex Deucher <alexdeucher@gmail.com> 6835Date: Wed Nov 17 17:37:25 2010 -0500 6836 6837 radeon/kms: allow tiled front buffer on 6xx/7xx 6838 6839 Use UTS/DFS to tile/untile as appropriate for sw access. 6840 Also enables pageflipping with tiling enabled. 6841 6842commit 035f7f3ab529ca19b853066792af8a23d08a0f53 6843Author: Dave Airlie <airlied@redhat.com> 6844Date: Wed Dec 1 10:23:14 2010 +1000 6845 6846 radeon: add some missing math.h and assert.h includes 6847 6848commit 0f924e83d7be6fca7e024acdf973089efdf3b627 6849Author: Dave Airlie <airlied@redhat.com> 6850Date: Wed Dec 1 10:23:33 2010 +1000 6851 6852 radeon: make sure EXA is off when EXA is disabled with --disable-exa 6853 6854commit e142e55c56d8440d7185b594c696ec9eeb699cda 6855Author: Alex Deucher <alexdeucher@gmail.com> 6856Date: Tue Nov 30 15:51:21 2010 -0500 6857 6858 evergreen: set default group_bytes to 256 6859 6860 512 seems to cause issues on certain cards. 6861 6862 Fixes: 6863 https://bugs.freedesktop.org/show_bug.cgi?id=31724 6864 6865commit 90f831361844f1b80b3f6bb718ff5ac584d73d48 6866Author: Alex Deucher <alexdeucher@gmail.com> 6867Date: Mon Nov 29 18:09:05 2010 -0500 6868 6869 evergreen: use dot4 for transforms 6870 6871commit 3cae361d0448b6e231c80f53d64bdbbdd74dc4cf 6872Author: Alex Deucher <alexdeucher@gmail.com> 6873Date: Mon Nov 29 17:44:47 2010 -0500 6874 6875 6xx/7xx: clean up gpr/const handling in shaders 6876 6877commit d9bcac516f2a810acb300b29169e56a2df0b47ac 6878Author: Alex Deucher <alexdeucher@gmail.com> 6879Date: Mon Nov 29 17:23:30 2010 -0500 6880 6881 r6xx/r7xx use dot4 for transforms 6882 6883commit 5d3f33729be0639cef17372345b2dab6127e39d9 6884Author: Alex Deucher <alexdeucher@gmail.com> 6885Date: Wed Nov 24 10:59:09 2010 -0500 6886 6887 radeon/kms: fix output enumeration 6888 6889 If a system has a mix of DVI-I, and DVI-D connectors, or a mix 6890 of HDMI-A and HDMI-B connectors, the enumeration was per 6891 connector type, but the naming was the same for multiple 6892 connectors (DVI-[0-n] for DVI-I, -D, -A, and HDMI-[0-n] 6893 for HDMI-A, and -B), you'd end up with multiple connectors 6894 with the same name. Fix this by tracking the number of 6895 connectors across similar connector types. 6896 6897commit dc470b5382ffa27bd4a673b21fb6b978d3716ca0 6898Author: Alex Deucher <alexdeucher@gmail.com> 6899Date: Tue Nov 23 01:46:15 2010 -0500 6900 6901 radeon/man: divide driver options into UMS and KMS 6902 6903commit 4cbb7100df4b983d77c20a77431cf2f5101bde36 6904Author: Alex Deucher <alexdeucher@gmail.com> 6905Date: Tue Nov 23 01:22:04 2010 -0500 6906 6907 update man page to include ontario 6908 6909commit 718dfd443897d70d97fc19afddfbcfe1e39a9c59 6910Author: Alex Deucher <alexdeucher@gmail.com> 6911Date: Mon Nov 22 13:11:50 2010 -0500 6912 6913 Ontario: add workaround for small pixmap issues with DFS 6914 6915 On past asics these kind a problems were usually due to a bad 6916 backend map config. I suspect in this case, and the previous 6917 cases it may be related to using linear buffers rather than 6918 tiled. 6919 6920 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6921 6922commit 61c97b0812c820564c7bc9d64a998c109bcd46b3 6923Author: Alex Deucher <alexdeucher@gmail.com> 6924Date: Mon Nov 22 13:10:16 2010 -0500 6925 6926 Add EXA/Xv acceleration support for Ontario Fusion APUs 6927 6928 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6929 6930commit 0659f35dfbeda7dd3ff58da1deb6484fabadc332 6931Author: Alex Deucher <alexdeucher@gmail.com> 6932Date: Mon Nov 22 13:07:00 2010 -0500 6933 6934 Add Ontario fusion APU pci ids 6935 6936 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6937 6938commit fa0cec331e6c94cebf82270dcd6c179b15851c5e 6939Author: Alex Deucher <alexdeucher@gmail.com> 6940Date: Mon Nov 22 12:56:45 2010 -0500 6941 6942 ontario: add UMS modesetting support 6943 6944 Accel not enabled yet. 6945 6946 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6947 6948commit 7ad1f01573e930aa6238eaeb0e11708ae548dfce 6949Author: Jerome Glisse <jglisse@redhat.com> 6950Date: Fri Nov 19 10:27:40 2010 -0500 6951 6952 evergreen: fix typo for fetch resource offset 6953 6954 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 6955 6956commit 3455a3b58532ea3ad901a317126968ab6cbb21b7 6957Author: Alex Deucher <alexdeucher@gmail.com> 6958Date: Wed Nov 17 17:32:41 2010 -0500 6959 6960 radeon/kms: fix buffer base alignment for tiling 6961 6962 On r6xx+, 2D tiling can require larger than 4k base alignment. 6963 6964commit beb7fecd0191e38fb238134ba612985062cf9770 6965Author: Dave Airlie <airlied@redhat.com> 6966Date: Mon Nov 15 11:59:58 2010 +1000 6967 6968 dce4: improve the VT switching code. 6969 6970 add in lots more blocks of regs to save/restore 6971 6972commit 51e51f864a691e51b5193d102398ced667e0beb1 6973Author: Dave Airlie <airlied@redhat.com> 6974Date: Thu Nov 11 18:48:40 2010 +1000 6975 6976 evergreen: remove warnings since last commit 6977 6978commit abbf73ee990512ac16ca77e8bb23288495e1f9f4 6979Author: Dave Airlie <airlied@redhat.com> 6980Date: Thu Nov 11 09:38:15 2010 +1000 6981 6982 evergreen: add UMS VT switch support. 6983 6984 This isn't perfect, but it brings back text VTs here on the 6985 DAC and DVI outputs. 6986 6987commit e1dfaf93d06bc5eafdbc2e1823d19204ce8f242c 6988Author: Dave Airlie <airlied@redhat.com> 6989Date: Thu Nov 11 09:15:27 2010 +1000 6990 6991 avivo: use arrays to store the crtc/pll info for save/restore 6992 6993 this is just prep work for evergreen VT save/restore 6994 6995commit 0c2834e67df6e143a7d15f373faaddca1dda6b18 6996Author: Michel Dänzer <daenzer@vmware.com> 6997Date: Fri Nov 5 09:30:40 2010 +0100 6998 6999 Fix incorrect CS size in one textured video path. (Bug #31364) 7000 7001 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=31364 . 7002 7003commit 2a7b080727f61559cf8444c93a5088f4155a8fdf 7004Author: Alex Deucher <alexdeucher@gmail.com> 7005Date: Fri Oct 29 16:56:45 2010 -0400 7006 7007 radeon/kms: more alignment fixes 7008 7009commit bf60af579382a0d48f7a65a4cec88759cc8b683c 7010Author: Dave Airlie <airlied@redhat.com> 7011Date: Thu Oct 28 13:06:54 2010 +1000 7012 7013 dri2: reference count the client privates key/callback 7014 7015 This lets multi-screen work better, but still having issues after server 7016 recycle, but it doesn't crash at least. 7017 7018commit fb22d0c06a7dc42216230e198ff443d8035e9d21 7019Author: Dave Airlie <airlied@redhat.com> 7020Date: Thu Oct 28 11:10:03 2010 +1000 7021 7022 radeon: only register block handler once per fd. 7023 7024 should fix https://bugs.freedesktop.org/show_bug.cgi?id=29726 7025 7026 the problem is of course the second head instance tries to access the 7027 fd and fails, however I think this might break syncing on the second 7028 head but not sure, but its better than just hanging up the X server 7029 7030commit 040140b5d7cbeaf5d83687bc87e4a236ff9f0cbd 7031Author: Alex Deucher <alexdeucher@gmail.com> 7032Date: Wed Oct 27 20:16:17 2010 -0400 7033 7034 drm/radeon: unify fb bo alignment harder 7035 7036 More duplicated paths discoved... 7037 7038commit f07f9b7b61c05f2de1d61bb0e2f71bd017c8d36a 7039Author: Alex Deucher <alexdeucher@gmail.com> 7040Date: Wed Oct 27 12:37:42 2010 -0400 7041 7042 kms/radeon: unify fb bo alignment handling 7043 7044 Previously there were 3 different paths with what should 7045 have had duplicated code: 7046 - EXACreatePixmap2 7047 - Initial front buffer creation 7048 - Randr resize 7049 7050 This patch attempts to unify the alignment across all 3. 7051 7052 This may fix tiling issues in some cases and should make 7053 buffer pitches match for pageflipping. 7054 7055commit d31046ba6c8eee9b7decc3875697d37c38bc38f3 7056Author: Michel Dänzer <daenzer@vmware.com> 7057Date: Wed Oct 27 12:12:04 2010 +0200 7058 7059 Re-use result of GetScreenPixmap call. 7060 7061 Fixes compile warning due to local variable ppix being unused when building 7062 against current xserver Git. 7063 7064commit 4dd7f835c17b00707f1d8d4e36a24380ba52761e 7065Author: Michel Dänzer <daenzer@vmware.com> 7066Date: Wed Oct 27 12:09:03 2010 +0200 7067 7068 Fix another stray xfree() call. 7069 7070commit f815e9c3c61b1ed73e7dd9383587efcdd8dfc07f 7071Author: Michel Dänzer <daenzer@vmware.com> 7072Date: Wed Oct 27 11:29:12 2010 +0200 7073 7074 Don't try to delete DRI2 event list entries that were never added. (Bug #31086) 7075 7076 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=31086 . 7077 7078commit df92245e1fd0ead2581b675cd96ac8bcc8cb65d7 7079Author: Alex Deucher <alexdeucher@gmail.com> 7080Date: Sun Oct 24 14:27:27 2010 -0400 7081 7082 r100: add some missing bits in SE_COORD_FMT 7083 7084 Texture coordinates work fine with or without these, 7085 but this should be more correct I think although 7086 I don't think it matters since we aren't sending w 7087 anyway. 7088 7089commit fc7aa6e3bbb86eaa2f34b827629f7e82cde65ac7 7090Author: Alex Deucher <alexdeucher@gmail.com> 7091Date: Sun Oct 24 11:56:48 2010 -0400 7092 7093 UMS/radeon: fix typo in restore palette 7094 7095commit 5efdf514ab0e5439114e8dd7a49105812155eb69 7096Author: Alex Deucher <alexdeucher@gmail.com> 7097Date: Sun Oct 24 11:45:19 2010 -0400 7098 7099 UMS/radeon: fix screen dimming on VT switch 7100 7101 Save and restore the palettes on VT switch. The restore 7102 has to be done after the vga restore to work properly as 7103 determined by Jonathan Kollasch. 7104 7105 Fixes: 7106 https://bugs.freedesktop.org/show_bug.cgi?id=18407 7107 7108commit 1e3ede62c39307d0210cb2044badd619a4f44fec 7109Author: Alex Deucher <alexdeucher@gmail.com> 7110Date: Sun Oct 24 11:24:07 2010 -0400 7111 7112 UMS/atom: default to DVI mode if we don't have monitor hdmi info 7113 7114 Fixes: 7115 https://bugs.freedesktop.org/show_bug.cgi?id=30330 7116 7117commit 51f26a7f01b68e373867c03881868ca8830664d6 7118Author: Mark Schreiber <mark7@alumni.cmu.edu> 7119Date: Sat Apr 10 10:55:08 2010 -0700 7120 7121 Correct spelling and grammar 7122 7123commit 0f575984594e6977b4a7cb794dad8c6e56c850f1 7124Author: Alex Deucher <alexdeucher@gmail.com> 7125Date: Tue Oct 19 19:21:39 2010 -0400 7126 7127 Xv: limit overlay to 2047x2047 7128 7129 Fixes: 7130 https://bugs.freedesktop.org/show_bug.cgi?id=15391 7131 7132commit 0f184630cd4760199430320215e4280438fc4ce5 7133Author: Alex Deucher <alexdeucher@gmail.com> 7134Date: Tue Oct 19 17:13:06 2010 -0400 7135 7136 radeon: fix compilation on xserver >= 1.10 7137 7138 Fixes: 7139 https://bugs.freedesktop.org/show_bug.cgi?id=30451 7140 7141 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7142 7143commit 179b0e15924ae2757efaece6488ae835ba6436e3 7144Author: Dave Airlie <airlied@redhat.com> 7145Date: Tue Oct 19 13:37:22 2010 +1000 7146 7147 r600: for 2d tiles we need to align to channels no banks. 7148 7149commit 809cacbc23ab37aa7952665f9b2e37c03a34f173 7150Author: Adam Jackson <ajax@redhat.com> 7151Date: Mon Oct 18 12:32:11 2010 -0400 7152 7153 kms: Let the server know if we got more than 1 EDID block 7154 7155 Otherwise things like xf86MonitorIsHDMI() won't work right. 7156 7157 Signed-off-by: Adam Jackson <ajax@redhat.com> 7158 7159commit 042037e485396981cf4a420c247104ba5d016ca0 7160Author: Alex Deucher <alexdeucher@gmail.com> 7161Date: Thu Oct 14 17:00:51 2010 -0400 7162 7163 evergreen: requires same cursor fix as avivo chips 7164 7165commit 0f37458c253aec46d98db6661c4def711df8fc17 7166Author: Adam Jackson <ajax@redhat.com> 7167Date: Thu Oct 14 14:37:34 2010 -0400 7168 7169 Don't allow the config file to override BIOS location 7170 7171 Signed-off-by: Adam Jackson <ajax@redhat.com> 7172 7173commit 2918df3a312fc600e83f84c298b9653d2eb3a8d5 7174Author: Alex Deucher <alexdeucher@gmail.com> 7175Date: Wed Oct 13 17:16:04 2010 -0400 7176 7177 evergreen: increase XV_CRTC Xv attribute range 7178 7179 Cover all 6 possible crtcs. 7180 7181commit 0d1f9fd03d0196dda5c1ce34b2e68b007c1874f2 7182Author: Alex Deucher <alexdeucher@gmail.com> 7183Date: Sun Oct 10 12:06:22 2010 -0400 7184 7185 DCE3+: UMS modesetting fixes 7186 7187 - add mising LVTMA case statement for DCE3.0 dig encoder 7188 - some DCE4 systems have EN/DISABLE_OUTPUT actions 7189 7190commit acf4024aa0ef192355b2bd8281507b1c2e6b7de4 7191Author: Alex Deucher <alexdeucher@gmail.com> 7192Date: Fri Oct 8 16:30:36 2010 -0400 7193 7194 ATOM DDC fixes for UMS 7195 7196 - Header size was already subtraced from table size 7197 - Only hw capable ddc pads are shared with aux 7198 7199commit 74fd2b91477106a26a2d9fb4b11c885910996041 7200Author: Alex Deucher <alexdeucher@gmail.com> 7201Date: Thu Oct 7 17:33:00 2010 -0400 7202 7203 DCE3.2 UMS: fix duallink 7204 7205 Should fix: 7206 https://bugs.freedesktop.org/show_bug.cgi?id=30686 7207 7208commit 8c544272f218fd8bd93a898b96afedc2498b27bb 7209Author: Alex Deucher <alexdeucher@gmail.com> 7210Date: Thu Oct 7 17:25:47 2010 -0400 7211 7212 fix the non-kms build 7213 7214 Should fix: 7215 https://bugs.freedesktop.org/show_bug.cgi?id=30685 7216 7217commit 81360adffb2a66b9a95a38671f9227a9718c9841 7218Author: Oldřich Jedlička <oldium.pro@seznam.cz> 7219Date: Mon Sep 6 21:14:52 2010 +0200 7220 7221 radeon: proper DRI2 pending events handling when client gone. (v6) 7222 7223 Properly handle asynchronous DRI2 events for disconnected clients. 7224 Track client's pending requests and mark them as invalid when the 7225 client disconnects. 7226 7227 This is based on the version from Alban Browaeys in bug #29065. 7228 7229 v1 (Alban Browaeys): Based upon a detailed explanation from Oldřich 7230 Jedlička and comments from Christopher James Halse Rogers. 7231 on http://lists.x.org/archives/xorg-driver-ati/2010-August/016780.html . 7232 7233 v2: Updated version to apply on master. Removed unnecessary 7234 client_index field from _DRI2FrameEvent. Added freeing/removing from 7235 list to failed paths of radeon_dri2_schedule_wait_msc and 7236 radeon_dri2_schedule_swap. 7237 7238 v3: Adopt to older xorg-server that doesn't have dixRegisterPrivateKey. 7239 7240 v4: Conditional include of list.h, unreachable return removed. 7241 7242 v5: Distribute list.h as xorg_list.h, remove xorg-server version check. 7243 Use the version from xorg-server when available (checked in 7244 configure.ac). 7245 7246 v6: Removed xorg_list.h, made DRI2 scheduling features dependent on 7247 list.h presence. 7248 7249commit adee138f007e26307f1aab7f8fe066150c12e55d 7250Author: Jesse Adkins <jesserayadkins@gmail.com> 7251Date: Tue Sep 28 13:29:50 2010 -0700 7252 7253 Purge cvs tags. 7254 7255 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com> 7256 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 7257 7258commit f1bf9cbb471d60ee4b0422bc4e896e3b86f4d75f 7259Author: Alex Deucher <alexdeucher@gmail.com> 7260Date: Wed Oct 6 11:44:42 2010 -0400 7261 7262 radeon: man page updates 7263 7264 - tv-out is not supported on carding using 7265 the rage theatre chip for tv-out 7266 - add evergreen cards to the list 7267 7268commit 0921ecc1c751df0dd56e0b1d0b78ab53d7164904 7269Author: Alex Deucher <alexdeucher@gmail.com> 7270Date: Mon Oct 4 12:34:47 2010 -0400 7271 7272 evergreen: use EXACreatePixmap2 if available 7273 7274commit 5bdb6434975584eef90eb9e5955b9c2a14b7f327 7275Author: Alex Deucher <alexdeucher@gmail.com> 7276Date: Mon Oct 4 12:31:51 2010 -0400 7277 7278 evergreen: port Karl's UTS/DFS changes to evergreen 7279 7280commit be8f45cbd313b68ad663f303c64edb4525b8f981 7281Merge: e843faf ff5f466 7282Author: Alex Deucher <alexdeucher@gmail.com> 7283Date: Mon Oct 4 12:36:56 2010 -0400 7284 7285 Merge branch 'evergreen_accel' of git+ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 7286 7287commit ff5f466e6ca0ee541a255facff6c9198976dfd52 7288Author: Alex Deucher <alexdeucher@gmail.com> 7289Date: Fri Oct 1 15:46:43 2010 -0400 7290 7291 evergreen/dri2: fix depth allocation for depth+stencil 7292 7293 evergreen uses separate allocations for depth and stencil, 7294 so to handle that, create a depth buffer large enough to 7295 handle both. This is required for using the stencil 7296 buffer in mesa. 7297 7298commit e843faf355c864beab81e74f0e39f8ad53d4c2bf 7299Author: Alex Deucher <alexdeucher@gmail.com> 7300Date: Thu Sep 30 19:30:35 2010 -0400 7301 7302 radeon: fix fbLocation for >32 bit MC addresses 7303 7304 If the fbLocation was at an address >32 bits, we'd fail. 7305 Change fbLocation to uint64_t and properly cast when needed. 7306 7307commit 886febc882053e09294225e85b102f965041b62b 7308Author: Alex Deucher <alexdeucher@gmail.com> 7309Date: Thu Sep 30 19:20:17 2010 -0400 7310 7311 r6xx: fix bad mask when setting up HDP_NONSURFACE_BASE 7312 7313 This fails for MC addresses >32 bits 7314 7315commit 16589b862feb796e15dc6e471d25604a4f5e7c1c 7316Author: Alex Deucher <alexdeucher@gmail.com> 7317Date: Mon Sep 27 18:28:43 2010 -0400 7318 7319 bump version post release 7320 7321commit cc5005af61f45a3552f7358dc5aa711e42f5af54 7322Author: Alex Deucher <alexdeucher@gmail.com> 7323Date: Mon Sep 27 18:20:53 2010 -0400 7324 7325 bump version for release 7326 7327commit 7f8820fcec8c90bf2f823170bd08a23e8b4ff7af 7328Author: Michael Cree <mcree@orcon.net.nz> 7329Date: Mon Sep 27 13:39:12 2010 -0400 7330 7331 Fix some unaligned 32bit accesses in the AtomBios code. 7332 7333 On the Alpha architecture unaligned 32bit accesses incur a software 7334 trap to the kernel and pollute the kernel logs. Fixed by use of the 7335 ldl_u() interface. 7336 7337 Signed-off-by: Michael Cree <mcree@orcon.net.nz> 7338 7339commit c4f834cdfbe96aa47ac5fb039f9dd7aa9730c8a3 7340Author: Nicolas Reinecke <nr@das-labor.org> 7341Date: Mon Sep 27 13:33:55 2010 -0400 7342 7343 radeon: Convert remaining x(c)alloc/xfree to m/calloc/free. 7344 7345 Fixes deprecation warnings missed out by 7346 f7a91ece264af9f3fd2fc18e99aefcda93ce9f5c 7347 7348commit f8fb9312d791af1f77020e8c2d35bb30841ed9aa 7349Author: Karl Tomlinson <karlt+@karlt.net> 7350Date: Sun Aug 22 22:46:33 2010 +1200 7351 7352 RADEONPrepareAccess_CS: fallback to DFS when pixmap is in VRAM 7353 7354 This avoids costly CPU VRAM reads and lets EXA manage a system memory cache 7355 of the portions of pixmaps needed for unaccelerated operations. 7356 7357 https://bugs.freedesktop.org/show_bug.cgi?id=27139 7358 7359commit 35c4ff936601ee083f51510a5192fb97d622a483 7360Author: Karl Tomlinson <karlt+@karlt.net> 7361Date: Sun Aug 22 22:28:06 2010 +1200 7362 7363 radeon: complete UTS and DFS even when a scratch BO is not necessary 7364 7365 Turns on the big-endian paths even for little-endian systems, and adds 7366 similar paths to the r6xx/r7xx functions. 7367 7368 This makes UTS and DFS reliable, which will let PrepareAccess (with 7369 mixed pixmaps) choose to fail based on whether the pixmap is in VRAM 7370 (to avoid CPU reads). 7371 7372commit d46381a3a6bf10903803f5acaa7aa0ce06373b96 7373Author: Karl Tomlinson <karlt+@karlt.net> 7374Date: Sun Aug 22 21:02:45 2010 +1200 7375 7376 radeon: complete big endian UTS and DFS even when scratch allocation fails. 7377 7378 On big endian systems, PrepareAccess will fail when byte-swapping is 7379 required so UploadToScreen and DownloadFromScreen cannot rely on 7380 fallback to PrepareAccess. 7381 7382 When scratch BO space allocation fails, this patch merely adds simple 7383 fallback to direct CPU access without any GPU blit. This sometimes 7384 requires a CS flush even in UploadToScreen. 7385 (No allocation retry after a flush is added here.) 7386 7387commit 4ced4e1eff67946e306c0c67c9ed59dd5f3c4ba9 7388Author: Karl Tomlinson <karlt+@karlt.net> 7389Date: Sun Aug 22 20:04:42 2010 +1200 7390 7391 RADEONDownloadFromScreenCS: flush CS writes before mapping BO for read 7392 7393 If unflushed CS operations write to the pixmap BO, then these need to be 7394 flushed before mapping the BO for read. This currently only affects big 7395 endian systems and only when the operation writes to the GTT domain. 7396 7397commit a4eef8faffbb1ea2f742273ee855f4e6f992e5c8 7398Author: Karl Tomlinson <karlt+@karlt.net> 7399Date: Sat Aug 21 22:29:34 2010 +1200 7400 7401 FinishAccess_CS: set bo_mapped to FALSE on unmap 7402 7403 This is actually only necessary when PrepareAccess may behave differently on 7404 different calls with the same pixmap, which currently doesn't happen. 7405 7406 However resetting bo_mapped is necessary to let PrepareAccess (with mixed 7407 pixmaps) choose to fail based on whether the pixmap is in VRAM (to avoid CPU 7408 reads). 7409 7410commit bfebe039af0c0282d04eb6234b6e6d1e02097146 7411Author: Karl Tomlinson <karlt+@karlt.net> 7412Date: Sat Aug 21 21:44:39 2010 +1200 7413 7414 DownloadFromScreenCS: download via a scratch BO if pixmap domain is unknown 7415 7416 radeon_bo_is_busy() may return without setting the domain out-parameter. 7417 If this happens, then download via a scratch GTT BO to avoid CPU VRAM read. 7418 7419commit e26a59e9db8067882327f872e3d2d760ce4c66f3 7420Author: Mathieu Bérard <mathieu@mberard.eu> 7421Date: Thu Sep 9 19:32:38 2010 -0400 7422 7423 Fix NULL possible deref in evergreen_cp_wait_vline_sync 7424 7425commit b90cb61ccb0f4f80e0627141f223354a9371d47d 7426Author: Alex Deucher <alexdeucher@gmail.com> 7427Date: Tue Sep 7 11:51:16 2010 -0400 7428 7429 radeon: set interlaced and doublescan enabled for randr outputs 7430 7431 interlaced used to work without setting these parameters. Changes 7432 in the xserver seem to require them now. 7433 7434 Should fix: 7435 https://bugs.freedesktop.org/show_bug.cgi?id=29591 7436 7437commit 7cc0d3311f23ad569af004e0e4a0e2efbac107e5 7438Author: Alex Deucher <alexdeucher@gmail.com> 7439Date: Fri Sep 3 00:32:27 2010 -0400 7440 7441 evergreen: fix Xv 7442 7443 VS const buffer offset was wrong. 7444 7445 fixes: 7446 https://bugs.freedesktop.org/show_bug.cgi?id=29788 7447 7448commit 7c6ce4e62693f446d7d3c8a86502ccc03c0e55b0 7449Author: Alex Deucher <alexdeucher@gmail.com> 7450Date: Thu Sep 2 20:34:56 2010 -0400 7451 7452 evergreen: add additional default state 7453 7454 Note, you also need a drm patch to fix the GPU hangs: 7455 drm/radeon/kms/evergreen: fix gpu hangs in userspace accel code 7456 7457 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7458 7459commit 2b98ec1f7e931019a4ab699a56d5dfaa395946fb 7460Author: Alex Deucher <alexdeucher@gmail.com> 7461Date: Wed Sep 1 13:24:19 2010 -0400 7462 7463 r6xx/r7xx: fix typos 7464 7465 some stray - signs 7466 7467commit 966ac1be81da76c8aa4ea46b63f3ca5358a2c021 7468Author: Alex Deucher <alexdeucher@gmail.com> 7469Date: Fri Aug 27 18:22:21 2010 -0400 7470 7471 evergreen: work around bad data in some i2c tables 7472 7473 The 7th entry in a lot of evergreen i2c gpio tables is partially 7474 zeroed. Fix the entry. 7475 7476 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7477 7478commit 91f707d308d4bbf16c3d62d046cf280fef5a8f4c 7479Author: Heikki Lindholm <holin@iki.fi> 7480Date: Fri Aug 27 02:26:24 2010 -0400 7481 7482 xv: fix non-kms/non-dri Xv column ordering on big endian systems 7483 7484 Column order is wrong on big endian systems, primarly because of a 7485 bits / bytes mix up with the bpp variable. Fix tested with r100 and 7486 r300, screen depth 16 and 32 with YV12 and YUY2 (overlay, textured video), 7487 RGBA and RGBT (overlay). 7488 7489 Should fix: https://bugs.freedesktop.org/show_bug.cgi?id=29041 7490 7491 Signed-off-by: Heikki Lindholm <holin@iki.fi> 7492 7493commit 6a2c8587a4e05a8be2a2e975a6660942cfe115d6 7494Author: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com> 7495Date: Fri Aug 27 13:14:33 2010 +1000 7496 7497 dri2: Reference count DRI2 buffers 7498 7499 When a client calls ScheduleSwap we set up a kernel callback when the 7500 relevent vblank event occurs. However, it's possible for the client 7501 to go away between calling ScheduleSwap and the vblank event, 7502 resulting in the buffers being destroyed before they're passed to 7503 radeon_dri2_frame_event_handler. 7504 7505 Add reference-counting to the buffers and take a reference in 7506 radeon_dri2_schedule_swap to ensure the buffers won't be destroyed 7507 before the vblank event is dealt with. 7508 7509 This parallels the approach taken by the Intel DDX in commit 7510 0d2392d44aae95d6b571d98f7ec323cf672a687f. 7511 7512 Fixes: http://bugs.freedesktop.org/show_bug.cgi?id=29065 7513 7514 v2: Don't write completion events to the client if it has quit. 7515 v3: Don't try to unref the NULL buffers from a DRI2_WAITMSC event. 7516 Take a ref in schedule_swap earlier, so the offscreen fallback 7517 doesn't incorrectly destroy the buffers. 7518 7519 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com> 7520 Signed-off-by: Dave Airlie <airlied@redhat.com> 7521 7522commit e9928fe036e9382fd7bc353f3f05531445f08977 7523Author: Dave Airlie <airlied@redhat.com> 7524Date: Wed Aug 25 10:42:39 2010 +1000 7525 7526 remove explicit buffer submit from copy region 7527 7528 port of 0be3e95c844247746742805830860ace9f546d99 from intel driver. 7529 7530 Remove explicit batchbuffer submit in DRI2 copyregion 7531 7532 Now that we submit from the flush callback chain, we know we'll always 7533 submit before the client receives the reply or event that blocks it from 7534 rendering the next frame. 7535 7536 Signed-off-by: Dave Airlie <airlied@redhat.com> 7537 7538commit 9f13049ddf06f6f2138851a548cfb82f12a52f42 7539Author: Dave Airlie <airlied@redhat.com> 7540Date: Wed Aug 25 08:56:37 2010 +1000 7541 7542 radeon: add correct flushing for direct rendered 7543 7544 this is a port of 69d65f9184006eac790efcff78a0e425160e95aa from the Intel 7545 driver. 7546 7547 Submit batch buffers from flush callback chain 7548 7549 There are a few cases where the server will flush client output buffers 7550 but our block handler only catches the most common (before going into select 7551 If the server flushes client buffers before we submit our batch buffer, 7552 the client may receive a damage event for rendering that hasn't happened yet 7553 7554 Instead, we can hook into the flush callback chain, which the server will 7555 invoke just before flushing output. This lets us submit batch buffers 7556 before sending out events, preserving ordering. 7557 7558 Fixes 28438: [bisected] incorrect character in gnome-terminal under compiz 7559 https://bugs.freedesktop.org/show_bug.cgi?id=28438 7560 7561 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net> 7562 7563 Signed-off-by: Dave Airlie <airlied@redhat.com> 7564 7565commit 141cbc283fddeb67c2a6c47a0f0f5c2aa2bfb430 7566Author: Alex Deucher <alexdeucher@gmail.com> 7567Date: Tue Aug 24 18:01:05 2010 -0400 7568 7569 evergreen Xv: remove stray line that caused a segfault 7570 7571 fixes: 7572 https://bugs.freedesktop.org/show_bug.cgi?id=29788 7573 7574commit f170dddd3538a587f2363ef5fa10a4b484e762da 7575Author: Alex Deucher <alexdeucher@gmail.com> 7576Date: Mon Aug 23 18:58:12 2010 -0400 7577 7578 evergreen: use vbo pool for constant buffers 7579 7580commit eede93b057bbdddcde5da3220a3f8c6f73784a7e 7581Author: Alex Deucher <alexdeucher@gmail.com> 7582Date: Mon Aug 23 13:16:27 2010 -0400 7583 7584 radeon: move vbo data to a separate struct 7585 7586 this way we can share the vbo code with const buffers 7587 7588commit d8abf27dbd14f5eb746c5e8b8b1436ad292d8ec6 7589Author: Alex Deucher <alexdeucher@gmail.com> 7590Date: Mon Aug 23 01:31:27 2010 -0400 7591 7592 evergreen: fix dword counts in default state 7593 7594commit 5a7c9d94733a0db1d3565447acc9f0e751db5950 7595Author: Alex Deucher <alexdeucher@gmail.com> 7596Date: Mon Aug 23 00:43:48 2010 -0400 7597 7598 radeon: fix legacy lvds dpms sequence 7599 7600 Take from my kms commit. 7601 7602 Should fix: 7603 https://bugs.freedesktop.org/show_bug.cgi?id=19459 7604 7605 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7606 7607commit 6930d2c981221757b1e11ef194809f085753a611 7608Author: Alex Deucher <alexdeucher@gmail.com> 7609Date: Fri Aug 20 16:55:21 2010 -0400 7610 7611 Add initial EXA and Xv support for evergreen 7612 7613 Based on the r6xx/r7xx code updated for evergreen. 7614 Still causes GPU hangs in some cases. We haven't 7615 tracked down why yet. Might be related to constant 7616 buffer persistence. 7617 7618 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7619 7620commit bdd41fecdb19c83c6c7b793016b61d38065dcd13 7621Author: Alex Deucher <alexdeucher@gmail.com> 7622Date: Fri Aug 20 01:17:13 2010 -0400 7623 7624 evergreen: set encoder type to DVI for HDMI 7625 7626 Fixes: 7627 http://bugs.freedesktop.org/show_bug.cgi?id=27452 7628 7629 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7630 7631commit ad8ea1f6e5fcb4f163622cf5eab953ea812b5829 7632Author: Alex Deucher <alexdeucher@gmail.com> 7633Date: Thu Aug 19 17:04:35 2010 -0400 7634 7635 DCE3+: switch pads to ddc mode when doing i2c 7636 7637 The pins for ddc and aux are shared so you need to switch the 7638 mode when doing ddc. The ProcessAuxChannel table already sets 7639 the pin mode to DP. This should fix unreliable ddc issues 7640 on DP ports using non-DP monitors. 7641 7642 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7643 7644commit fd686668289258ffaf6b81057545e50612aac6a8 7645Author: Dave Airlie <airlied@redhat.com> 7646Date: Thu Aug 12 12:59:18 2010 +1000 7647 7648 radeon: fixup non-kms build 7649 7650commit 5a9865d90c23c4ce0f46d380ea9119eac87a99eb 7651Author: Dave Airlie <airlied@redhat.com> 7652Date: Thu Aug 12 11:18:06 2010 +1000 7653 7654 evergreen: fix multi-head cursor support 7655 7656commit 8351bb9f085fde5dc47b115375efcc61adc23859 7657Author: Dave Airlie <airlied@redhat.com> 7658Date: Tue Aug 10 15:05:45 2010 +1000 7659 7660 radeon: take 8/10 encoding into a/c on displayport link 7661 7662commit b91e0efcb24eea32e6978c146c5409eeeeac0a62 7663Author: Dave Airlie <airlied@redhat.com> 7664Date: Tue Aug 10 15:05:25 2010 +1000 7665 7666 dce32: remove rmx workaround 7667 7668commit 94bc1b7156cd0866566dc44a823c7e051bb45175 7669Author: Dave Airlie <airlied@redhat.com> 7670Date: Tue Aug 10 13:24:52 2010 +1000 7671 7672 atombios: fixup set crtc source like KMS 7673 7674 This removes a bunch of strict aliasing warnings and fixes the 7675 codepaths up like the latest KMS code, including a workaround for a bug 7676 on evergreen. 7677 7678commit 9bc716eb62d4e0eed2902b92437a42634eef6ba1 7679Author: Dave Airlie <airlied@redhat.com> 7680Date: Tue Aug 10 13:23:21 2010 +1000 7681 7682 atombios: move adjust pixel clock around to follow KMS code flow 7683 7684 This reworks the pixel clock adjusting code to follow the KMS style, 7685 also fixes warnings in this code. 7686 7687commit 31de43bf9d9eb93cc2b2150474ea7404beabe49d 7688Author: Dave Airlie <airlied@redhat.com> 7689Date: Tue Aug 10 13:21:39 2010 +1000 7690 7691 displayport: retry on timeout 7692 7693 this is ported from KMS 7694 7695commit 9ef67335583d36080d227e8bce1966afe08e0486 7696Author: Dave Airlie <airlied@redhat.com> 7697Date: Tue Aug 10 13:21:01 2010 +1000 7698 7699 evergreen: don't call YUV table on evergreen 7700 7701commit 1cce55e8ba43e7958cb67147aeaeb068826ab99f 7702Author: Dave Airlie <airlied@redhat.com> 7703Date: Tue Aug 10 13:20:13 2010 +1000 7704 7705 evergreen: add support to parse firmware info for ext dp clk 7706 7707commit bbffd67d3296344e8735b007cdee83146d38369c 7708Author: Dave Airlie <airlied@redhat.com> 7709Date: Tue Aug 10 13:14:54 2010 +1000 7710 7711 atombios: realign digital transmitter/encoder setup with kms 7712 7713commit 6244153467665f5007e2fc7786b4bcc4b0b96030 7714Author: Dave Airlie <airlied@redhat.com> 7715Date: Tue Aug 10 12:57:22 2010 +1000 7716 7717 update atombios.h to latest from kernel 7718 7719commit bb7c77ca75e857f90791b0dd1c04c8e2f19d0e3c 7720Author: Alex Deucher <alexdeucher@gmail.com> 7721Date: Mon Aug 9 22:18:53 2010 -0400 7722 7723 atom: upstream parser update 7724 7725 fixes digital output problems on evergreen asics 7726 7727commit cc0a167ff2db9b097891883ded147af1d67e4407 7728Author: Alex Deucher <alexdeucher@gmail.com> 7729Date: Tue May 25 18:17:15 2010 -0400 7730 7731 r6xx/r7xx: default to micro (1D) tiling for now 7732 7733 SW de-tiling used in the r600c 3D driver has issues with 7734 certain configurations. 7735 7736 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7737 7738commit b5bfdbd70d9671250957ccd41dfc8818850d257e 7739Author: Alex Deucher <alexdeucher@gmail.com> 7740Date: Thu Aug 5 17:26:28 2010 -0400 7741 7742 r6xx/r7xx: add support for tiling with kms (v3) 7743 7744 Requires radeon drm 2.6.0 and updated mesa. 7745 7746 v2: - fix lockup due to not emiting DB_DEPTH_INFO 7747 https://bugs.freedesktop.org/show_bug.cgi?id=28342 7748 - fix drm minor version to deal with evergreen accel 7749 v3: rebase on current ddx git 7750 7751 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7752 7753commit a3c59c6f6be7067421e348142da0ca13428dcd57 7754Author: Alex Deucher <alexdeucher@gmail.com> 7755Date: Wed Aug 4 12:07:36 2010 -0400 7756 7757 radeon: add new pci ids 7758 7759commit a456587b77ae357750179a50f8db2a17c0f2738e 7760Author: Alex Deucher <alexdeucher@gmail.com> 7761Date: Mon Aug 2 14:24:41 2010 -0400 7762 7763 r6xx/r7xx: move syrface sync emit to the functions that emit surface info 7764 7765 reduces code duplication. 7766 7767commit 8eba977cab1878ba247da8160771d41194d8014f 7768Author: Alex Deucher <alexdeucher@gmail.com> 7769Date: Thu Jun 10 13:34:59 2010 -0400 7770 7771 EXA: move more common helpers to exa_shared 7772 7773commit 71c1a2704af23b61439cee5ce784f7fe267a8a26 7774Author: Alex Deucher <alexdeucher@gmail.com> 7775Date: Thu Jun 10 11:52:43 2010 -0400 7776 7777 remove unused reg headers from radeon_exa_shared.c 7778 7779commit d73aef78919005369af1b60df138439b4b6105c3 7780Author: Alex Deucher <alexdeucher@gmail.com> 7781Date: Mon Aug 2 13:27:12 2010 -0400 7782 7783 r6xx/r7xx: set VGT_MAX_VTX_INDX to a larger value 7784 7785commit dacaf5d827b58c39f9e5a7ac0530f9ea6e257347 7786Author: Dave Airlie <airlied@redhat.com> 7787Date: Mon Aug 2 08:33:51 2010 +1000 7788 7789 fix make distcheck 7790 7791commit c79ce215a01b45fc63b483da167ae37ec7aefad6 7792Author: Dave Airlie <airlied@redhat.com> 7793Date: Sun Aug 1 16:51:48 2010 +1000 7794 7795 radeon/r600: restructure exa + vbo to provide more sharing 7796 7797 This is a precursor for r300/500 vbo support. 7798 7799 Signed-off-by: Dave Airlie <airlied@redhat.com> 7800 7801commit 82254b59268140c4102ae3cd713743ae2be15c00 7802Author: Alex Deucher <alexdeucher@gmail.com> 7803Date: Fri Jul 30 17:15:05 2010 -0400 7804 7805 r6xx/r7xx: unify composite mask and non-mask pixel shader 7806 7807commit 1c17f3a192f644e8e38b5cfb1470f49434bfba27 7808Author: Alex Deucher <alexdeucher@gmail.com> 7809Date: Fri Jul 30 16:34:54 2010 -0400 7810 7811 r6xx/r7xx: clean up composite vertex shader 7812 7813 keep CF, ALU, Fetch instructions in separate groups 7814 7815commit f9d6c0de231357f96e2e0de71e6c9221bcb36bd4 7816Author: Gaetan Nadon <memsize@videotron.ca> 7817Date: Fri Jul 23 13:28:42 2010 -0400 7818 7819 The local copy of the modes code is no longer required. 7820 7821 The server 1.2 as shipped in the tarball on the web does not contain the 7822 modes code. It was added just after and found in git branch server-1.2-branch. 7823 7824 The modes code was initially included in version ati 6.8.0 and fails to compile 7825 with server 1.2 as it requires randr 1.2. The modes code is included in server 7826 versions 1.3 and later, so there is no need to provide an unknown version of 7827 the modes code in the ati driver tarball. It will never be used. 7828 7829 This patch makes the ati driver requiring server 1.3 or later. 7830 Version 6.8.0 configures and builds ok on server 1.3 7831 Master branch post 6.13.1 configures and builds ok on server 1.3 7832 7833 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7834 7835commit 0028419acb0762eeb950de5fe702c93e70301612 7836Author: Gaetan Nadon <memsize@videotron.ca> 7837Date: Sat Jul 24 10:51:18 2010 -0400 7838 7839 config: add AM_PROG_CC_C_O for per-target compilation flags 7840 7841 Per-target compilation flags (theatre200_drv_la_CFLAGS) are required 7842 when multiple targets which require different compiler flags, 7843 are build in the same makefile. 7844 7845 Automake issues a command with -c and -o flags which not all compilers 7846 support. The object fles are prefixed with theatre200_drv_la. 7847 The macro AM_PROG_CC_C_O must then be used to provide this feature 7848 on compilers that do not have it. If not, a warning is issued at make time. 7849 7850 This macros checks for compiler support and if missing, uses a "compile" 7851 script it generates in the package root directory. 7852 7853 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7854 7855commit aca0a8669b538d58f018f95c9b22e6b3ec1ffe50 7856Author: Gaetan Nadon <memsize@videotron.ca> 7857Date: Wed Jul 21 16:49:04 2010 -0400 7858 7859 config: add comments for main statements 7860 7861commit 66e614f7115efeec237b3b916d9637e8b3e8985c 7862Author: Gaetan Nadon <memsize@videotron.ca> 7863Date: Wed Jul 21 16:07:00 2010 -0400 7864 7865 config: replace deprecated use of AC_OUTPUT with AC_CONFIG_FILES 7866 7867 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7868 7869commit 16e5510c90ef1ba2bbaab78d18943f080b86d809 7870Author: Gaetan Nadon <memsize@videotron.ca> 7871Date: Wed Jul 21 14:37:41 2010 -0400 7872 7873 config: replace deprecated AC_HELP_STRING with AS_HELP_STRING 7874 7875 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7876 7877commit b36d8c09e91382f4cfa71635374ec88f5b676d1c 7878Author: Gaetan Nadon <memsize@videotron.ca> 7879Date: Wed Jul 21 14:05:22 2010 -0400 7880 7881 config: replace deprecated AM_CONFIG_HEADER with AC_CONFIG_HEADERS 7882 7883 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7884 7885commit cd9351b04c2d6982b28c647a63d550eb3e1937eb 7886Author: Alex Deucher <alexdeucher@gmail.com> 7887Date: Wed Jul 21 13:48:24 2010 -0400 7888 7889 r6xx/r7xx: group op variable state 7890 7891 Group the op variable state into one emit block, re-order 7892 to reduce dwords emitted. 7893 7894commit 5f838c664e8010f4e51afecd4100d73a96fe1209 7895Author: Gaetan Nadon <memsize@videotron.ca> 7896Date: Wed Jul 21 09:27:42 2010 -0400 7897 7898 config: complete AC_INIT m4 quoting 7899 7900 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7901 7902commit 48ec2e65c268b426ab9a3e214d174447cf5b5936 7903Author: Gaetan Nadon <memsize@videotron.ca> 7904Date: Wed Jul 21 08:33:38 2010 -0400 7905 7906 config: remove unrequired AC_SUBST for LIBDRM and LIBUDEV 7907 7908 These macros are called by the PKG_CHECK_MODULES macro. 7909 7910 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7911 7912commit 1aabb7eb7d8f06c7481151145db3b9a722ce4ef0 7913Author: Gaetan Nadon <memsize@videotron.ca> 7914Date: Tue Jul 20 21:54:11 2010 -0400 7915 7916 config: remove unrequired AC_SUBST([DRI_CFLAGS]) 7917 7918 This macro is called by PKG_CHECK_MODULES 7919 7920 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7921 7922commit 8f92b349821a3ee5ed8df55273d905b9605385aa 7923Author: Gaetan Nadon <memsize@videotron.ca> 7924Date: Tue Jul 20 21:44:57 2010 -0400 7925 7926 config: remove unrequired AC_SUBST([XORG_CFLAGS]) 7927 7928 This macro is called by PKG_CHECK_MODULES 7929 7930 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7931 7932commit 5b483b832f9c4a5b92ffb7f72a470669201f1fba 7933Author: Gaetan Nadon <memsize@videotron.ca> 7934Date: Tue Jul 20 20:24:42 2010 -0400 7935 7936 config: remove unrequired AC_HEADER_STDC 7937 7938 Autoconf says: 7939 "This macro is obsolescent, as current systems have conforming 7940 header files. New programs need not use this macro". 7941 7942 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7943 7944commit 6574e3a16eb3631ee7e00ee60a8c9ba95c8b84ef 7945Author: Gaetan Nadon <memsize@videotron.ca> 7946Date: Tue Jul 20 19:41:30 2010 -0400 7947 7948 config: remove AC_PROG_CC as it overrides AC_PROG_C_C99 7949 7950 XORG_STRICT_OPTION from XORG_DEFAULT_OPTIONS calls 7951 AC_PROG_C_C99. This sets gcc with -std=gnu99. 7952 If AC_PROG_CC macro is called afterwards, it resets CC to gcc. 7953 7954 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7955 7956commit 593eff2924c2ad161d8b490fbf6d7e433fbe2a80 7957Author: Gaetan Nadon <memsize@videotron.ca> 7958Date: Tue Jul 20 18:45:18 2010 -0400 7959 7960 config: update AC_PREREQ statement to 2.60 7961 7962 Unrelated to the previous patches, the new value simply reflects 7963 the reality that the minimum level for autoconf to configure 7964 all x.org modules is 2.60 dated June 2006. 7965 7966 ftp://ftp.gnu.org/gnu/autoconf/autoconf-2.60.tar.gz 7967 7968 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7969 7970commit c2ab6ffc25aa6759cbbb4c1fbbd4a136b38983bf 7971Author: Gaetan Nadon <memsize@videotron.ca> 7972Date: Tue Jul 20 16:15:29 2010 -0400 7973 7974 config: upgrade to util-macros 1.8 for additional man page support 7975 7976 Use MAN_SUBST now supplied in XORG_MANPAGE_SECTIONS 7977 The value of MAN_SUBST is the same for all X.Org packages. 7978 7979commit cdeb1949c820242f05a8897d3ddd0718f204dacf 7980Author: Jerome Glisse <jglisse@redhat.com> 7981Date: Thu Jul 15 16:21:41 2010 -0400 7982 7983 kms: don't call cursor helper if using software cursor 7984 7985 Fix : 7986 https://bugzilla.redhat.com/show_bug.cgi?id=601713 7987 https://bugzilla.redhat.com/show_bug.cgi?id=598358 7988 7989 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 7990 7991commit 06691376b1ee963c711420edaf5a03eab6f5658f 7992Author: Dave Airlie <airlied@redhat.com> 7993Date: Wed Jul 7 13:15:03 2010 +1000 7994 7995 fix build on non-kms 7996 7997commit 052cf0169ae70d5448af6dc4db840b2fc195569b 7998Author: Dave Airlie <airlied@redhat.com> 7999Date: Wed Jul 7 11:10:46 2010 +1000 8000 8001 configure.ac: bump version post release 8002 8003commit ad999e633ff41d27eed9d2c6535e163a7181b0bd 8004Author: Dave Airlie <airlied@redhat.com> 8005Date: Wed Jul 7 10:49:22 2010 +1000 8006 8007 set version for release 8008 8009commit 37b348059b1c15d7b381cd3df3db52bd9ee6613e 8010Author: Alex Deucher <alexdeucher@gmail.com> 8011Date: Wed Jun 30 12:56:48 2010 -0400 8012 8013 remove rv100 quirk 8014 8015 Some RV100 cards with 2 VGA ports show up with DVI+VGA, however 8016 some boards with DVI+VGA have the same subsystem ids. Better 8017 to have a VGA port show up as DVI than having a non-useable 8018 DVI port. 8019 8020 reported by DHR in irc. 8021 8022 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8023 8024commit c8ea942bd0e9d1c055d50b94440aa4dae425f20b 8025Author: Cooper Yuan <cooperyuan@gmail.com> 8026Date: Tue Jun 29 20:34:57 2010 +0800 8027 8028 Remove HDP_SOFT_RESET function, there is no need to reset HDP block. 8029 8030 This commit can fix an issue reported on DELL server, system gets hang 8031 during soft resetting while another application tries to access PCI 8032 configuration space. 8033 8034commit 139b38bf67ec10d876cc56df833541d497ae4fa4 8035Author: Alex Deucher <alexdeucher@gmail.com> 8036Date: Thu Jun 24 14:55:09 2010 -0400 8037 8038 r6xx shader: use ADDR() for CF_DWORD0 8039 8040 no change in functionality 8041 8042commit 801e83227a59a29eea425ea612083bbf2b536c30 8043Author: Wolfram <bugzilla1@malloc.de> 8044Date: Mon Jun 21 18:59:19 2010 -0400 8045 8046 r6xx/r7xx: fix ums cmd buffer leak 8047 8048 Fixes: 8049 https://bugs.freedesktop.org/show_bug.cgi?id=27957 8050 8051commit b13d719080b75fc6db4d15d2d323b8fce8b7ad06 8052Author: Alex Deucher <alexdeucher@gmail.com> 8053Date: Mon Jun 21 18:49:44 2010 -0400 8054 8055 r6xx/r7xx: fix miscount in state emit 8056 8057commit c3c5c8e2cc91b51a24effdffb85281216eed731d 8058Author: Alex Deucher <alexdeucher@gmail.com> 8059Date: Mon Jun 21 14:30:17 2010 -0400 8060 8061 r6xx/r7xx accel: add back some additional default state 8062 8063 This adds back everything removed in c29157bbf5b0dd26857675282ab094082fbaed0d 8064 except CB_FOG_*, CB_CLEAR_* and the VPORT transforms. Those shouldn't 8065 be needed as we aren't using fog or viewport transforms. We probably don't 8066 need all the state that was added back either but I can't reproduce any 8067 problems here, so it's hard to say which parts are problematic. 8068 8069 Should fix: 8070 https://bugs.freedesktop.org/show_bug.cgi?id=28629 8071 and several corruption reports on #radeon. 8072 8073 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8074 8075commit 800cb2088fec698d0626063a9ab198ff534938c0 8076Author: Michel Dänzer <daenzer@vmware.com> 8077Date: Mon Jun 21 08:15:14 2010 +0200 8078 8079 DRI2: Fix up confusion between windows and pixmaps. 8080 8081 Fixes crashes with xserver master, where looking up a pixmap private on a 8082 window no longer happens to work. 8083 8084commit f7a91ece264af9f3fd2fc18e99aefcda93ce9f5c 8085Author: Michel Dänzer <daenzer@vmware.com> 8086Date: Mon Jun 14 08:52:16 2010 +0200 8087 8088 Convert x(c)alloc/xfree to m/calloc/free. 8089 8090 Fixes deprecation warnings with xserver master and should also work with older 8091 xservers. 8092 8093commit ea37d24b1b6d4cbcf73e680846de25b72af216e3 8094Author: Dave Airlie <airlied@redhat.com> 8095Date: Mon Jun 21 13:55:15 2010 +1000 8096 8097 radeon: fix support for 1.9 server master. 8098 8099 This moves pixmap and mode set into CSR where its allowed. Should work fine on 8100 old servers also. 8101 8102 Signed-off-by: Dave Airlie <airlied@redhat.com> 8103 8104commit fdd8ecafd054f65842351aee6ee6fba7af6613b2 8105Author: Alex Deucher <alexdeucher@gmail.com> 8106Date: Wed Jun 16 19:02:10 2010 -0400 8107 8108 r6xx/r7xx: macro safety fixes 8109 8110commit 4651d77211b508cb6b76931807780e317f232220 8111Author: Alex Deucher <alexdeucher@gmail.com> 8112Date: Wed Jun 16 12:28:36 2010 -0400 8113 8114 radeon: fix depth 16 with ums 8115 8116 Fixes: 8117 https://bugs.freedesktop.org/show_bug.cgi?id=28494 8118 https://bugzilla.redhat.com/show_bug.cgi?id=554967 8119 8120 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8121 8122commit 1e1d6a515428b6884fea586d180346fc74ef75a1 8123Author: Alex Deucher <alexdeucher@gmail.com> 8124Date: Wed Jun 16 12:20:03 2010 -0400 8125 8126 r3xx-r5xx Xv: disable bicubic filtering by default 8127 8128 - makes Xv more consistent with r1xx/r2xx/r6xx/r7xx 8129 - Xv attributes like brightness, contrast, hue, etc. only work 8130 when bicubic is disabled. 8131 - avoids performance issues on some systems when sampling from textures 8132 in GART with kms. 8133 - It can be re-enabled with using xvattr 8134 8135 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8136 8137commit c29157bbf5b0dd26857675282ab094082fbaed0d 8138Author: Alex Deucher <alexdeucher@gmail.com> 8139Date: Tue Jun 15 19:39:42 2010 -0400 8140 8141 r6xx/r7xx: remove unnecessary state emit 8142 8143 No need to emit state that doesn't impact that we use for EXA/Xv. 8144 8145 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8146 8147commit 35280545fcfb911f189d7657bd1040a28450fe7b 8148Author: Alex Deucher <alexdeucher@gmail.com> 8149Date: Tue Jun 15 19:21:32 2010 -0400 8150 8151 r6xx/r7xx: reorg default and clipping state emit 8152 8153 Group ordered registers to save command buffer space. 8154 Reduces the default and clipping state from 256 to 160 dwords. 8155 8156 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8157 8158commit a2528a734c1d4e8639c49e5d222e3630a93ffbfd 8159Author: Alex Deucher <alexdeucher@gmail.com> 8160Date: Tue Jun 15 17:00:34 2010 -0400 8161 8162 r6xx/r7xx accel: remove some duplicate emits and minor clean up 8163 8164 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8165 8166commit 5f093357f18eb9bea641394ab86a92a1766d8f2e 8167Author: Gaetan Nadon <memsize@videotron.ca> 8168Date: Sat Jun 12 15:11:46 2010 -0400 8169 8170 COPYING: replace stub file with actual copyright notices 8171 8172 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 8173 8174commit 0c2118586d3edeecc2473b5d685472df4b5e70fa 8175Author: Dave Airlie <airlied@redhat.com> 8176Date: Fri Jun 11 15:53:18 2010 +1000 8177 8178 radeon: fixup last fix, use CURSOR_WIDTH not RADEON_CURSOR_WIDTH. 8179 8180commit 9a117f768cb8261327fd3c324da9c98875785cc1 8181Author: Dave Airlie <airlied@redhat.com> 8182Date: Fri Jun 11 14:28:24 2010 +1000 8183 8184 radeon: set proper stride for cursor in tiling flags. 8185 8186 definitely incorrect, but hope it doesn't break anything. 8187 8188 Signed-off-by: Dave Airlie <airlied@redhat.com> 8189 8190commit b6346ede94f0d0b11ee04770cf52508cb0a5e6c6 8191Author: Dave Airlie <airlied@redhat.com> 8192Date: Thu Jun 10 20:13:58 2010 -0400 8193 8194 drmmode: fix big endian issue with properties. 8195 8196 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. 8197 8198 Signed-off-by: Dave Airlie <airlied@redhat.com> 8199 8200commit 1971dc6d758dea13b9fb6d1c8f516c165628b5e5 8201Author: Alex Deucher <alexdeucher@gmail.com> 8202Date: Tue Jun 8 11:18:28 2010 -0400 8203 8204 radeon: fix rn50 cloning with kms 8205 8206 Since they only have one crtc sometimes the xserver doesn't assign 8207 a crtc to one of the outputs even though both outputs have common modes 8208 which results in only one monitor being enabled. Assign a crtc in 8209 preinit so that both outputs light up. 8210 8211 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8212 8213commit 426114b4a99d37b394efe3336968bb0ab9b6e9ae 8214Author: Dave Airlie <airlied@redhat.com> 8215Date: Tue Jun 8 11:34:35 2010 +1000 8216 8217 xv: fix Xv on M6/RV100 under KMS. 8218 8219 pRADEONEnt->HasCRTC2 wasn't setup under KMS. 8220 8221 Signed-off-by: Dave Airlie <airlied@redhat.com> 8222 8223commit f64bf0de8e2de7c1bf9cc0c614603dd23c9060ad 8224Author: Alex Deucher <alexdeucher@gmail.com> 8225Date: Thu Jun 3 14:48:21 2010 -0400 8226 8227 evergreen: explicitly disable accel on evergreen 8228 8229 Previously we relied on the drm not having accel enabled 8230 to make sure evergreen used shadowfb, now we when we enable 8231 accel in the drm, we need to make sure the ddx doesn't try 8232 and use it until it's implemented. 8233 8234 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8235 8236commit 428125c095b18c2a2864c1aef24ac0f384b6be54 8237Author: Marek Olšák <maraeo@gmail.com> 8238Date: Sun May 16 20:11:17 2010 +0000 8239 8240 r3xx-r5xx: enable color tiling by default on KMS 8241 8242 Signed-off-by: Marek Olšák <maraeo@gmail.com> 8243 8244commit 6712ce4f4715c8ce7c2fcddb52ca9b103bdd634c 8245Author: Marek Olšák <maraeo@gmail.com> 8246Date: Sun May 16 20:09:31 2010 +0000 8247 8248 rs740: follow macro-switch too 8249 8250 Signed-off-by: Marek Olšák <maraeo@gmail.com> 8251 8252commit 842fa162e9d3bbad2bc44d3732bbc8e5a54402ea 8253Author: Alex Deucher <alexdeucher@gmail.com> 8254Date: Tue May 18 11:21:01 2010 -0400 8255 8256 dri2: use radeon_pick_best_crtc() 8257 8258 It's already exposed and used by exa and Xv and has 8259 the same functionality. radeon_covering_crtc() can 8260 be dropped eventually. 8261 8262commit 64e6858aaf5d0e39ecc9f9804689012f3a38660a 8263Author: Alex Deucher <alexdeucher@gmail.com> 8264Date: Mon May 17 12:50:44 2010 -0400 8265 8266 r2xx exa: fix typo in 27c3326863deae36bc35e2c3b73fffd400208ff1 8267 8268 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8269 8270commit 27c3326863deae36bc35e2c3b73fffd400208ff1 8271Author: Alex Deucher <alexdeucher@gmail.com> 8272Date: Mon May 17 09:53:32 2010 -0400 8273 8274 r1xx/r2xx EXA: handle RepeatNone properly with transforms 8275 8276 Fallback with unsupported ops like r3xx, r6xx. 8277 8278 Should fix fdo bug 28142. 8279 8280 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8281 8282commit 30591320ec46e491ba20904cc64f3405b51c6505 8283Author: Jerome Glisse <jglisse@redhat.com> 8284Date: Fri May 7 17:35:09 2010 +0200 8285 8286 kms: add support for the MSC swap & sync API 8287 8288 This patch is mostly a port over of Intel ddx code for 8289 MSC support. It needs a radeon KMS module with version 8290 2.4 which has the query for hw crtc id. 8291 8292 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> 8293 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 8294 8295commit 766024dcc61c83490540910ce752f9bfe6dddba4 8296Author: Marek Olšák <maraeo@gmail.com> 8297Date: Mon May 3 22:56:27 2010 +0200 8298 8299 r3xx-r5xx: fix texturing with small macrotiled pixmaps 8300 8301 Pixmaps smaller than a macrotile cannot be used as textures because hardware 8302 automatically switches to macro-linear and therefore sampled pixels are 8303 messed up. This behavior is known as MACRO_SWITCH. 8304 8305 The only sane workaround seems to be not using macrotiling for small pixmaps. 8306 8307 The function RADEONMacroSwitch has been ported from r300g and implements 8308 MACRO_SWITCH the same way it's implemented in hardware. It's been well tested 8309 in r300g. 8310 8311 This commit also fixes blit-based framebuffer reads, which are used for tiled 8312 surfaces in r300g, when ColorTiling is enabled. 8313 8314 Signed-off-by: Marek Olšák <maraeo@gmail.com> 8315 8316commit 6aedd57f6c2f08d0151a8bd1c5893d40d3db709a 8317Author: Alex Deucher <alexdeucher@gmail.com> 8318Date: Tue May 4 14:30:52 2010 -0400 8319 8320 radeon: fix pll_out_min default for pre-avivo 8321 8322 Fixes fdo bug 27960 8323 8324commit b2ecd1bb507ed4e0d9ba6ebd498c51da2e8136fa 8325Author: Michel Dänzer <daenzer@vmware.com> 8326Date: Fri Apr 23 18:28:22 2010 +0200 8327 8328 FB/MMIO_cnt should be ints, not pointers. 8329 8330 Pointed out by compiler warnings. 8331 8332 https://bugs.freedesktop.org/show_bug.cgi?id=27817 8333 8334commit 1ad41054989f7c3edec373ccf09dceda6d7c94b4 8335Author: Alex Deucher <alexdeucher@gmail.com> 8336Date: Thu Apr 22 16:48:21 2010 -0400 8337 8338 r3xx: fix gb pipe setup for SE cards 8339 8340 Noticed by Tormod Volden 8341 8342commit 59e287d7c484b4addd4a06d013670577639c7ae2 8343Author: Alex Deucher <alexdeucher@gmail.com> 8344Date: Thu Apr 22 02:05:40 2010 -0400 8345 8346 r3xx-r5xx: fix vertex units 8347 8348 Noticed by Tormod Volden. 8349 8350 RV3xx is 2, RV560,RV570 is 8 8351 8352commit 2a6b409496f26da0436972b5feae6ea035dde08d 8353Author: Owain Ainsworth <zerooa@googlemail.com> 8354Date: Mon Apr 19 14:37:33 2010 +0100 8355 8356 Reference count shared driver mappings. 8357 8358 With MMIO it wasn't *such* a bit deal if we leaked the smallish mapping. 8359 with FB it could be a larger deal. So instead of worrying about this, 8360 reference count the mappings in the entity structure and unmap them when 8361 no one cares anymore. 8362 8363 Prompted by a discussion with airlied 8364 8365 Signed-off-by: Dave Airlie <airlied@redhat.com> 8366 8367commit 761f0de5556e46f166280476185977f720efe586 8368Author: Owain G. Ainsworth <zerooa@googlemail.com> 8369Date: Mon Apr 19 04:05:36 2010 +0100 8370 8371 Make consistent use of fbOffset and share fb mappings. 8372 8373 What we were doing previously was mapping the framebuffer for zaphod for 8374 only this driver instances chunk, however, fbOffset was (rightly) set to 8375 the offset into the whole framebuffer we were using. 8376 8377 Since in some cases we did operations on the FB virtual address + 8378 fbOffset (for example zeroing the framebuffer on entervt) we were 8379 actually pissing all over ourselves in those cases. 8380 8381 Fix this by implementing shared fb mappings like we do for MMIO already, 8382 and whenever we wish to refer to our area of FB space we always use 8383 fbOffset. Fixes zaphod for some users on r600 chipsets, my 4870 is still 8384 behaving strangely on screen 0, but I suspect that is another bug. 8385 8386 Once calculation (in PreInitAccel) is now wrong because of this, however 8387 dri on zaphod does now happen so this is irrelavent, add a comment to 8388 that effect. 8389 8390 Signed-off-by: Dave Airlie <airlied@redhat.com> 8391 8392commit 2059d628c5fb03222a62502cc1b5724bf296a7b3 8393Author: Owain G. Ainsworth <zerooa@googlemail.com> 8394Date: Mon Apr 19 04:05:35 2010 +0100 8395 8396 Kill per-context SAREA support. 8397 8398 It has never been used, and since the world is changing it almost 8399 certainly never will be. Good riddance. 8400 8401 Signed-off-by: Dave Airlie <airlied@redhat.com> 8402 8403commit c4b46f56279cc18c8114235ca41cfb41ffede67f 8404Author: Owain G. Ainsworth <zerooa@googlemail.com> 8405Date: Mon Apr 19 04:05:34 2010 +0100 8406 8407 Remove remnants of the ShowCache option. 8408 8409 The rest of it died a long time ago. 8410 8411 Signed-off-by: Dave Airlie <airlied@redhat.com> 8412 8413commit 03b34c5fb12c4df180d4df113d2b654f6afd7230 8414Author: Dave Airlie <airlied@redhat.com> 8415Date: Mon Apr 19 17:58:07 2010 +1000 8416 8417 radeon: add FireMV 2400 pci id. 8418 8419 From RH#581927 - thanks to Carsten Clasohm 8420 8421 Signed-off-by: Dave Airlie <airlied@redhat.com> 8422 8423commit 6c07816025f70e522986614c29c300ea13e5d932 8424Author: Alex Deucher <alexdeucher@gmail.com> 8425Date: Fri Apr 16 18:50:17 2010 -0400 8426 8427 r5xx texvid: deal with large numbers of verts 8428 8429 should fix fdo bug 25884 8430 8431commit 5ee320a6b1b4c65fe592c8ac4d1981799242d59e 8432Author: Alex Deucher <alexdeucher@gmail.com> 8433Date: Fri Apr 16 18:42:56 2010 -0400 8434 8435 r3xx texvid: deal with large numbers of verts 8436 8437 should fix fdo bug 25884 8438 8439commit 47af3f4f266232517486238917d82fc5ca9c82e6 8440Author: Alex Deucher <alexdeucher@gmail.com> 8441Date: Fri Apr 16 20:50:18 2010 -0400 8442 8443 r2xx texvid: deal with large numbers of verts 8444 8445 should fix fdo bug 25884 8446 8447commit 57577d5cd0641b7cad02242478699bcfece59227 8448Author: Alex Deucher <alexdeucher@gmail.com> 8449Date: Fri Apr 16 20:40:56 2010 -0400 8450 8451 r1xx texvid: deal with large numbers of verts 8452 8453 should fix fdo bug 25884 8454 8455commit 31a888e78fa403f2bddacee098a29f36eaa969cb 8456Author: Alex Deucher <alexdeucher@gmail.com> 8457Date: Fri Apr 16 15:52:59 2010 -0400 8458 8459 r5xx textured video: split into prepare and draw functions 8460 8461commit f7ebe4a127965bb9e46d20c60331bcc4dab76fc5 8462Author: Alex Deucher <alexdeucher@gmail.com> 8463Date: Fri Apr 16 15:46:24 2010 -0400 8464 8465 r3xx textured video: split into prepare and draw functions 8466 8467commit 10799b4ce0c18c5a7d92d688549e8e777344f111 8468Author: Alex Deucher <alexdeucher@gmail.com> 8469Date: Fri Apr 16 15:38:49 2010 -0400 8470 8471 r2xx textured video: split into prepare and draw functions 8472 8473commit bd1cf42201a1f918080bf34786a6de327cb31b5f 8474Author: Alex Deucher <alexdeucher@gmail.com> 8475Date: Fri Apr 16 15:31:40 2010 -0400 8476 8477 r1xx textured video: split into prepare and draw functions 8478 8479commit c1136f94b80346065893f8a43c0fbf60c8c9b057 8480Author: Alex Deucher <alexdeucher@gmail.com> 8481Date: Fri Apr 16 15:23:14 2010 -0400 8482 8483 Xv: track vtx_count, is_planar in port priv 8484 8485commit 2f680b631357661511456f07b7b4ba1aa2cbf30c 8486Author: Alex Deucher <alexdeucher@gmail.com> 8487Date: Fri Apr 16 13:43:41 2010 -0400 8488 8489 radeon/kms: fix possible crtc mask for evergreen 8490 8491commit 4656f5dff1ed72fa2c7a1484305f2aef7b65ff2b 8492Author: Alex Deucher <alexdeucher@gmail.com> 8493Date: Fri Apr 16 13:36:25 2010 -0400 8494 8495 radeon/kms: fix crash when using more than two heads 8496 8497 fixes fdo bug 27692 8498 8499commit a69e749d0562887af6bd236c38802472e54640c4 8500Author: Dave Airlie <airlied@redhat.com> 8501Date: Fri Apr 16 13:58:54 2010 +1000 8502 8503 kms: add uevent support. 8504 8505 When we get a hotplug event from the kernel we should notify the client side to reconfigure displays. 8506 8507 based on work by ajax in intel driver. 8508 8509 Signed-off-by: Dave Airlie <airlied@redhat.com> 8510 8511commit 22a46dddd375b2b9399e12fdf168fa5292ff17a4 8512Author: Alex Deucher <alexdeucher@gmail.com> 8513Date: Tue Apr 13 11:08:59 2010 -0400 8514 8515 radeon: add support for pll algo selection 8516 8517 tv-out on atom systems is very particular about it's 8518 dividers. force it to use the old algo. 8519 8520 Should fix fdo bug 27593. 8521 8522commit c1b817c45477c21234abaaebe78feb7ea4fd92b1 8523Author: Alex Deucher <alexdeucher@gmail.com> 8524Date: Mon Apr 12 16:32:04 2010 -0400 8525 8526 atom: disable TV encoder when VGA is in use 8527 8528 Switching between TV and VGA caused problems on some 8529 systems since the TV encoder was left enabled. 8530 8531 Should fix fdo bug 25520. 8532 8533commit 878814aeab7031ff6169fcaf2be869b7595c8699 8534Author: Alex Deucher <alexdeucher@gmail.com> 8535Date: Mon Apr 12 11:16:14 2010 -0400 8536 8537 Xv overlay: set scalerwidth to 1920 on r3xx/r4xx chips 8538 8539 reported on phoronix forums. 8540 8541commit 276fe739f1caa95778213054375a65288b80f320 8542Author: Alex Deucher <alexdeucher@gmail.com> 8543Date: Fri Apr 9 20:03:09 2010 -0400 8544 8545 radeon: add some new pci ids 8546 8547commit eb5665688ef9b52f03f61546351d0848cab54740 8548Author: Alex Deucher <alexdeucher@gmail.com> 8549Date: Thu Apr 8 00:31:52 2010 -0400 8550 8551 radeon: don't setup Xv on rn50 8552 8553 It has no overlay or 3D engine. 8554 8555 fixes fdo bug 27528 8556 8557commit 3c6c6afcbfc6f2c582e13a7c37fe2014b0875b22 8558Author: Alex Deucher <alexdeucher@gmail.com> 8559Date: Mon Apr 5 12:12:56 2010 -0400 8560 8561 change version post release 8562 8563commit fb7911912e60b2cdbc2152b96847775b767ca3ef 8564Author: Alex Deucher <alexdeucher@gmail.com> 8565Date: Mon Apr 5 11:59:24 2010 -0400 8566 8567 bump version for release 8568 8569commit 476a1c6e8b42807b897b8c6f8550ec42b2d5c10e 8570Author: Alex Deucher <alexdeucher@gmail.com> 8571Date: Thu Apr 1 12:42:56 2010 -0400 8572 8573 radeon: fix possible gpio i2c table overrun 8574 8575 GPIO_I2C_INFO does not always have ATOM_MAX_SUPPORTED_DEVICE 8576 entries. Should fix Novell bug 589022. 8577 8578commit 6baa96c44ca93b88acf5233335cee233e59d5af4 8579Author: Dave Airlie <airlied@redhat.com> 8580Date: Thu Apr 1 11:24:32 2010 +1000 8581 8582 r600/exa: further cleanup, use the object struct in the accel state. 8583 8584 This cleans up the accel state objects as well. 8585 8586 Signed-off-by: Dave Airlie <airlied@redhat.com> 8587 8588commit 87f49a24019967552c5478d6af89ba438b4261c8 8589Author: Dave Airlie <airlied@redhat.com> 8590Date: Thu Apr 1 10:52:58 2010 +1000 8591 8592 r600: cleanup wasteful variables. 8593 8594 Just assign directly to the structs. 8595 8596 Signed-off-by: Dave Airlie <airlied@redhat.com> 8597 8598commit 478319e55d52fce241d040912fea952071e3b742 8599Author: Dave Airlie <airlied@redhat.com> 8600Date: Thu Apr 1 10:45:52 2010 +1000 8601 8602 r600: reduce function call overhead. 8603 8604 Create a small accel object that can be used to reduce the amount 8605 of parameters passed to SetAccelState. This can be cleaned up a lot more. 8606 8607 Signed-off-by: Dave Airlie <airlied@redhat.com> 8608 8609commit bc93395b3eb5e3511c1b62af90693269f4fa6e13 8610Author: Alex Deucher <alexdeucher@gmail.com> 8611Date: Wed Mar 31 22:30:24 2010 -0400 8612 8613 r6xx+ EXA: fix swapped domains in kms UTS 8614 8615 Noticed by Dave. 8616 8617 Should fix fdo bug 27284 8618 8619commit 3a33b1a92c328b690bdc1aafa0011e4326109815 8620Author: Owain G. Ainsworth <zerooa@googlemail.com> 8621Date: Tue Mar 30 17:14:02 2010 +0100 8622 8623 Add support for MacModel autodetection on OpenBSD. 8624 8625 This allows radeon-equipped macs to work without an xorg.conf on openbsd. 8626 Patch originally written by Mark Kettenis (kettenis@openbsd.org) 8627 8628commit c7eeda8c3f3514ba95ebf2893fbe124bf526b3df 8629Author: Alex Deucher <alexdeucher@gmail.com> 8630Date: Fri Mar 26 01:38:36 2010 -0400 8631 8632 radeon: add support eDP connectors with and LCD device tag 8633 8634 should fix fdo bug 27322 8635 8636commit ff323e11485fa5030ac278c836bf80a535c22249 8637Author: Alex Deucher <alexdeucher@gmail.com> 8638Date: Fri Mar 26 01:24:16 2010 -0400 8639 8640 radeon: fix name string for eDP 8641 8642commit 53ac06161eb2b8cffb1b88e24a9a21cfd12e5883 8643Author: Alex Deucher <alexdeucher@gmail.com> 8644Date: Tue Mar 23 13:34:38 2010 -0400 8645 8646 r6xx+ EXA/Xv: add a R600SetAccelState function 8647 8648 This moves CS bo checking and alignment checks into 8649 a central location. It also cleans up the code. 8650 8651commit 7a044472dfea7cf05ba4c2b87be30e94e2ae0b62 8652Author: Alex Deucher <alexdeucher@gmail.com> 8653Date: Tue Mar 23 01:27:22 2010 -0400 8654 8655 r6xx+ EXA: always use a temp surface for overlapping copies 8656 8657 The scanline based fallback code was complex and added a lot of 8658 overhead. It also didn't work with kms. 8659 8660commit d33bddaedae81126ab7f0023af1c6443bb0b4c4f 8661Author: Alex Deucher <alexdeucher@gmail.com> 8662Date: Mon Mar 22 18:15:23 2010 -0400 8663 8664 r6xx+ EXA: always use the accel_state state in DoPrepareCopy 8665 8666commit dda3f5a99e7a2dc5d57860f4d07df3498e1e21df 8667Author: Alex Deucher <alexdeucher@gmail.com> 8668Date: Mon Mar 22 18:05:36 2010 -0400 8669 8670 r6xx EXA/Xv: track src/dst domains 8671 8672 Much of the code is shared, so track the src/dst 8673 domains so we make sure the uses consistent domains 8674 for each bo. 8675 8676commit 5c256808cb5fea955eea96ffe9196473715156aa 8677Author: Alex Deucher <alexdeucher@gmail.com> 8678Date: Wed Mar 17 23:47:50 2010 -0400 8679 8680 XAA: disable render accel 8681 8682 It's been reported broken for a while. Should fix 8683 fdo bug 27151, others. 8684 8685commit a0683be5cc082bdbdd3bc4e9b52f39f423650946 8686Author: Dave Airlie <airlied@redhat.com> 8687Date: Thu Mar 18 12:36:25 2010 +1000 8688 8689 radeon: avoid using DRI1 init path on DRI2 driver. 8690 8691 I was playing with multi-seat and found this code, fixed 8692 it up to be sane and more DRI2 like. 8693 8694 Signed-off-by: Dave Airlie <airlied@redhat.com> 8695 8696commit 080a5414593e9b59ed923f26aa6057747b0c868f 8697Author: Alex Deucher <alexdeucher@gmail.com> 8698Date: Tue Mar 16 12:33:39 2010 -0400 8699 8700 kms: fix ums naming compat for DisplayPort 8701 8702commit 819b4015349b5d8c5ffa5f979097599774fce5bb 8703Author: Alex Deucher <alexdeucher@gmail.com> 8704Date: Mon Mar 15 13:47:29 2010 -0400 8705 8706 dump version for rc release 8707 8708commit d3482a947e3731be4ed0b00b4e3079470700dc4c 8709Author: Michael Cree <mcree@orcon.net.nz> 8710Date: Fri Mar 12 22:23:31 2010 +1300 8711 8712 Fix some word accesses in AtomBios to work on all architectures. 8713 8714 The UINT16LE_TO_CPU(), etc., macros are used in the AtomBios code to 8715 fix up endian issues but they do not address bad alignment or assist 8716 architectures that cannot perform hardware byte or word accesses. 8717 This patch inserts use of the ldw_u(), etc., interface of the Xserver 8718 into certain AtomBios accesses to address alignment issues. 8719 8720 This resolves Debian bug 572311, namely that the driver when compiled 8721 for generic Alpha architecture (i.e. doesn't use the byte-word extension) 8722 resulted in no display output on certain Radeon cards. 8723 8724 Signed-off-by: Michael Cree <mcree@orcon.net.nz> 8725 8726commit 488c9fd8300505cc6c0c2f8f0f00849f27cc5d63 8727Author: Alex Deucher <alexdeucher@gmail.com> 8728Date: Mon Mar 15 12:25:57 2010 -0400 8729 8730 r6xx/r7xx: fix domain handling in accel code 8731 8732 Noticed by Pauli and Michel on IRC. 8733 8734 Improves GetImage performace by a factor of ~10. 8735 8736commit 2ace2591d92fb6d3ce7a6453edb04b36a6c49a32 8737Author: Alex Deucher <alexdeucher@gmail.com> 8738Date: Mon Mar 15 12:03:58 2010 -0400 8739 8740 radeon: remove some leftover debugging output 8741 8742commit 67e81c8f17ddde6eba633d2a5aef528e1d598d89 8743Author: Andrzej Hajda <andrzej.hajda@wp.pl> 8744Date: Wed Mar 10 18:19:35 2010 -0500 8745 8746 radeon: add support for pal on legacy IGP chips 8747 8748 Based on my initial non-working patch. 8749 8750 Fixes some element of fdo bug 12007 8751 8752commit 3a44f1cb0d2bb748692b1024003de8ee88ca77a5 8753Author: Alex Deucher <alexdeucher@gmail.com> 8754Date: Tue Mar 9 09:44:01 2010 -0500 8755 8756 atom: i2c gpio fixes 8757 8758 Basically a port of my kms patch. This allows us 8759 to remove some quirks. 8760 8761commit d85cb40f516c67305e818302bec7ee817df4144c 8762Author: Matt Turner <mattst88@gmail.com> 8763Date: Sun Mar 7 14:24:35 2010 -0500 8764 8765 Don't check for Xinerama. 8766 8767 It doesn't seem to be used anywhere, so don't require it. 8768 8769 CC: Jerome Glisse <jglisse@redhat.com> 8770 CC: Alex Deucher <alexdeucher@gmail.com> 8771 CC: Dave Airlie <airlied@redhat.com> 8772 Signed-off-by: Matt Turner <mattst88@gmail.com> 8773 8774commit e7b41f8cb082ed462d29bf3fc440072424cbd852 8775Author: Alex Deucher <alexdeucher@gmail.com> 8776Date: Fri Mar 5 19:16:11 2010 -0500 8777 8778 radeon: disable frac fb div with new pll code 8779 8780 fixes fdo bug 26897 8781 8782commit 14aff767490c253cbcdd411f812e50b91673119e 8783Author: Alex Deucher <alexdeucher@gmail.com> 8784Date: Wed Mar 3 13:31:19 2010 -0500 8785 8786 radeon: add new RS880 pci id 8787 8788commit e6dc886634b38e4a36af7b5f0b23299d5acd7244 8789Author: Dave Airlie <airlied@redhat.com> 8790Date: Tue Mar 2 10:25:15 2010 +1000 8791 8792 radeon: bump configure.ac 8793 8794commit 4975658f05c387b39b3e96a292a7683f17645c2c 8795Author: Dave Airlie <airlied@redhat.com> 8796Date: Sat Feb 27 16:47:19 2010 +1000 8797 8798 pciids: hopefully fix HP 8799 8800commit e76b90b399c3cc0f0998c0209300c46f97505498 8801Author: Alex Deucher <alexdeucher@gmail.com> 8802Date: Fri Feb 26 15:01:28 2010 -0500 8803 8804 rv740: disable dfs workaround for drm 1.32+ 8805 8806 rv740 pipe setup was fixed in 2.6.33 8807 8808commit 2de0af5f5b806f9dbfdb1e9b6a5cf96d9433961d 8809Author: Matt Turner <mattst88@gmail.com> 8810Date: Wed Feb 24 22:46:28 2010 -0500 8811 8812 Use RADEON_ALIGN instead of open coding it. 8813 8814 Also fix some RADEON_ALIGN(x, 63), which would return incorrect results 8815 for odd x. Though this shouldn't happen, it's still not right. You 8816 wouldn't ever write (x + 62) & ~62 which is clearly wrong (and what it 8817 expands to). 8818 8819 CC: Jerome Glisse <jglisse@redhat.com> 8820 CC: Alex Deucher <alexdeucher@gmail.com> 8821 CC: Dave Airlie <airlied@redhat.com> 8822 Signed-off-by: Matt Turner <mattst88@gmail.com> 8823 8824commit c7e81d2f3a372e0d5f751dd0c5091aec2b56d936 8825Author: Matt Turner <mattst88@gmail.com> 8826Date: Wed Feb 24 22:46:27 2010 -0500 8827 8828 Use/define RADEON_GPU_PAGE_SIZE instead of sprinkling 4096 everywhere. 8829 8830 Also, define RADEON_BUFFER_ALIGN in terms of it, and replace some 8831 RADEON_ALIGN(x, RADEON_BUFFER_ALIGN) with RADEON_ALIGN(x, 8832 RADEON_GPU_PAGE_SIZE) since this is really what was intended. 8833 8834 CC: Jerome Glisse <jglisse@redhat.com> 8835 CC: Alex Deucher <alexdeucher@gmail.com> 8836 CC: Dave Airlie <airlied@redhat.com> 8837 Signed-off-by: Matt Turner <mattst88@gmail.com> 8838 8839commit c0a5c9403dff254e1669df606a4193794270aeff 8840Author: Dave Airlie <airlied@redhat.com> 8841Date: Fri Feb 26 19:17:28 2010 +1000 8842 8843 pci: add HP vendor id 8844 8845commit 3970cc5563900f08ba5236b23f765c926616c345 8846Author: Cooper Yuan <cooperyuan@gmail.com> 8847Date: Fri Feb 26 16:14:12 2010 +0800 8848 8849 add more triple-head server ID 8850 8851commit b499eee7506f7478649e0000e9f8fcd00ac9d88d 8852Author: James Le Cuirot <chewi@aura-online.co.uk> 8853Date: Thu Feb 25 21:13:18 2010 -0500 8854 8855 clean up kms zaphod handling 8856 8857 Taken from James patch on bug 24523 8858 8859 agd5f: adapt to the current code 8860 8861commit 433c8617341f5768255826435a2b09afba684f02 8862Author: Alex Deucher <alexdeucher@gmail.com> 8863Date: Thu Feb 25 21:03:13 2010 -0500 8864 8865 update ZaphodHeads option 8866 8867 Now that the screen section mapping is fixed, 8868 Make the option per-instance and allow multiple 8869 outputs to be specified; e.g., DVI-0 and HDMI-0 8870 associated with instance 0 and LVDS and VGA-0 8871 associated with instance 1. 8872 8873commit f3e20b01c77f02bdf3483ceee27a1f00d240c33d 8874Author: James Le Cuirot <chewi@aura-online.co.uk> 8875Date: Thu Feb 25 19:10:19 2010 -0500 8876 8877 zaphod: Fix entity instance assignment. 8878 8879 Each screen needs to have a pointer back to its 8880 device's entity instance, Currently Screen1 is 8881 pointing to Screen0's instance so when it comes 8882 to match up the screen sections with the detected 8883 screens, Screen0 gets chosen in both cases. 8884 8885 See bug 24523 8886 8887commit 1b7e9a2e50f77819b3aff4e37ba39eaec69ff932 8888Author: Alex Deucher <alexdeucher@gmail.com> 8889Date: Tue Feb 23 11:39:55 2010 -0500 8890 8891 radeon: fixes for zaphodheads option 8892 8893 Needed for systems with more than two outputs. 8894 Both KMS and non-KMS zaphod work on systems with 8895 more than two outputs now. 8896 8897commit bd83e5f1c75b8c00b43846c7225568917fc9a8ce 8898Author: Dave Airlie <airlied@redhat.com> 8899Date: Tue Feb 23 19:27:24 2010 +1000 8900 8901 zaphod: hopefully fix kms + zaphod 8902 8903commit 0c3468d812e3790ce03d9e76779ae81e7b7b82d5 8904Author: Alex Deucher <alexdeucher@gmail.com> 8905Date: Mon Feb 22 17:34:51 2010 -0500 8906 8907 AVIVO: set frac fb divider pll flag 8908 8909 matches previous behavior 8910 8911commit 212e152536c4325e6799018891d9aee132681f48 8912Author: Alex Deucher <alexdeucher@gmail.com> 8913Date: Mon Feb 22 17:27:24 2010 -0500 8914 8915 radeon: update new pll algo 8916 8917 - add support for pre-avivo chips 8918 - add support for fixed post/ref dividers 8919 - add support for non-fractional fb dividers 8920 8921 By default avivo chips use the new algo and 8922 pre-avivo chips use the old algo. Use the 8923 "NewPLL" option to toggle between them (set to 8924 TRUE for the new algo, FALSE for the old). 8925 8926commit e68d3a3890fc81c51f2006b5548da1e8756ad2fd 8927Author: Alex Deucher <alexdeucher@gmail.com> 8928Date: Thu Feb 18 14:02:00 2010 -0500 8929 8930 RS600: add connector quirk 8931 8932 System lists DVI port as HDMI. 8933 8934 fixes fdo bug 26605 8935 8936commit a3b730eceb522c7ac1ef3dd6f6c7d773118d03f7 8937Author: Jerome Glisse <jglisse@redhat.com> 8938Date: Thu Feb 18 14:13:48 2010 +0100 8939 8940 r6xx/kms: when reseting BO force default state emission 8941 8942 In KMS world each cs need to fully initialize the 3D engine 8943 when we were erasing cs in IBDiscard for r6xx we weren't 8944 forcing reemission of default state. This patch force this 8945 emission. Fix : 8946 8947 http://bugs.freedesktop.org/show_bug.cgi?id=26603 8948 http://bugzilla.kernel.org/show_bug.cgi?id=15284 8949 8950commit 579cdcf9b4e38c791a497b747a055fc0a07d8dd6 8951Author: Alex Deucher <alexdeucher@gmail.com> 8952Date: Wed Feb 17 12:22:48 2010 -0500 8953 8954 radeon: add ZaphodHeads option 8955 8956 Allows users that want to use zaphod mode to select 8957 which xrandr outputs are assigned to which head. E.g., 8958 8959 Option "ZaphodHeads" "LVDS,VGA-0" 8960 will assign LVDS to the first zaphod driver instance 8961 and VGA-0 to the second instance. 8962 8963commit 47136fa347d1756523239746b4c74cd5278a1118 8964Author: Michel Dänzer <daenzer@vmware.com> 8965Date: Mon Feb 15 16:44:39 2010 +0100 8966 8967 EXA: Fix order of coordinates passed to radeon_pick_best_crtc(). 8968 8969 The wrong order prevented tearing avoidance from working with EXA/DRI2. 8970 8971commit 6e95905f759b844373be62ec513c78153ed00222 8972Author: Michel Dänzer <daenzer@vmware.com> 8973Date: Mon Feb 15 17:02:16 2010 +0100 8974 8975 Allocate BOs for 'video offscreen surfaces' and 'TV-in' in VRAM. 8976 8977 I think these can only really work in VRAM, though right now there's probably 8978 nothing to prevent them from getting evicted... 8979 8980commit 78e7047c5235b09858b66dd3688d39aaa27d7589 8981Author: Pauli Nieminen <suokkos@gmail.com> 8982Date: Mon Feb 15 13:40:37 2010 +0200 8983 8984 Allocate Xv buffers to GTT. 8985 8986 KMS doesn't have acceleration for upload to vram. memcpy/memmove to VRAM 8987 directly is very slow (40M/s in benchmark) which causes visible problems 8988 to video. 8989 8990 Allocating video buffer in GTT will give good performance (350-450M/s) 8991 for memmove operation. This is nice performance boost for Xv under KMS. 8992 8993 Signed-off-by: Pauli Nieminen <suokkos@gmail.com> 8994 8995commit 3ec25e59854b6b03ad763bc374d3475a50f562d8 8996Author: Pauli Nieminen <suokkos@gmail.com> 8997Date: Mon Feb 15 13:45:28 2010 +0200 8998 8999 Use memcpy instead of memove for RADEONSwapCopy. 9000 9001 memcpy has about 25% better performance than memmove when 9002 destination is GTT (wc caching). Changing memmove to memcpy 9003 in SwapCopy will reduce CPU time spent moving data between 9004 host and GPU. 9005 9006 memcpy will specially help Xv when playing high resolution 9007 videos. 9008 9009 The swap path doesn't support overlapping memory copies either 9010 so extra safety is not worth the performance hit in commonly 9011 used code path. 9012 9013 Signed-off-by: Pauli Nieminen <suokkos@gmail.com> 9014 9015 [ Michel Dänzer: Fixed up whitespace ] 9016 9017commit 221ef11b31756deb7134801730e76c040e841f5c 9018Author: Alex Deucher <alexdeucher@gmail.com> 9019Date: Fri Feb 12 14:15:10 2010 -0500 9020 9021 r6xx/r7xx: implement EXA vline support 9022 9023commit 8ad4025affe1fb2f417e3a3031d74f83be5df253 9024Author: Alex Deucher <alexdeucher@gmail.com> 9025Date: Fri Feb 12 11:34:47 2010 -0500 9026 9027 radeon: consolidate crtc selector for vline wait 9028 9029 Use the Xv version as it takes into account the area 9030 covered by the op rather than just picking the largest 9031 crtc area. 9032 9033commit 4f9d1714a7382594b834d446bbe502663f6a2778 9034Author: Alex Deucher <alexdeucher@gmail.com> 9035Date: Fri Feb 5 12:27:51 2010 -0500 9036 9037 r500: fix relocs for Xv planar video 9038 9039 fixes fdo bug 26445 9040 9041commit b7ca1ab11ac0e4e4be5a1e7789ae8633b460a2d4 9042Author: Dave Airlie <airlied@redhat.com> 9043Date: Fri Feb 5 13:37:20 2010 +1000 9044 9045 rs400/rs480: mc idle bit is bit 2 like original radeon 9046 9047 backport from a KMS fix, the rs400/480 mc idle is bit 2 not bit 4. 9048 9049 Signed-off-by: Dave Airlie <airlied@redhat.com> 9050 9051commit 97c387e269672b146a95b45fbef5c8c18e473e36 9052Author: Gaetan Nadon <memsize@videotron.ca> 9053Date: Fri Jan 29 16:08:17 2010 -0500 9054 9055 config: remove dead LINUXDOC macro usage 9056 9057 This module does not generate LINUXDOC documentation. 9058 9059 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 9060 9061commit 05fa2dee0c63d56fd523ee1ebab479bcb9f5e89d 9062Author: Alex Deucher <alexdeucher@gmail.com> 9063Date: Thu Feb 4 13:13:58 2010 -0500 9064 9065 evergreen: minor clean up in transmitter setup 9066 9067 both link and encoder must be set for linkb 9068 9069commit 76eea5e02776f7beeb8f4491a374bd36792eda92 9070Author: Alex Deucher <alexdeucher@gmail.com> 9071Date: Thu Feb 4 10:52:14 2010 -0500 9072 9073 r600: reduce number of cache flushes 9074 9075 We don't need to flush so often. Next step 9076 would be to move the flushing to the drm and only 9077 flush after each command buffer rather than each 9078 draw. 9079 9080commit 8d63d70f7ebaf9d250f0449d3720ef47516c05df 9081Author: Alex Deucher <alexdeucher@gmail.com> 9082Date: Wed Feb 3 15:22:22 2010 -0500 9083 9084 evergreen: use external clock source for DP PHY 9085 9086 DP CRTC clock always comes from DCPLL. This frees 9087 up PPll1/2 for non-DP-mode PHYs and CRTCs 9088 9089commit ecbc26431914216a8b207e81451282ea07c8b92f 9090Author: Alex Deucher <alexdeucher@gmail.com> 9091Date: Wed Feb 3 03:52:07 2010 -0500 9092 9093 evergreen: blank/unblank DP in dpms calls 9094 9095commit 710a2fd07235349f084ec40626cc28fbae523d27 9096Author: Alex Deucher <alexdeucher@gmail.com> 9097Date: Wed Feb 3 03:00:24 2010 -0500 9098 9099 evergreen: pll fixes 9100 9101 SetPixelClock needs to be called for DP outputs with a 9102 valid PLL. 9103 9104commit 9ab2377ee23c5dc360ba135a77aa6b181a1862e6 9105Author: Alex Deucher <alexdeucher@gmail.com> 9106Date: Tue Feb 2 15:15:36 2010 -0500 9107 9108 evergreen: DP requires coherent mode 9109 9110commit 634da80056525a48ec17ffb81e2fb59b168bea9e 9111Author: Alex Deucher <alexdeucher@gmail.com> 9112Date: Tue Feb 2 15:09:57 2010 -0500 9113 9114 evergreen: fix units on frac_fb_div 9115 9116commit a887818f491f6c7315c56c4e0d0b702c4c6aa4ac 9117Author: Alex Deucher <alexdeucher@gmail.com> 9118Date: Mon Feb 1 11:01:47 2010 -0500 9119 9120 evergreen: add pci ids 9121 9122commit cb2772b69480268c059766c4f6b209ce590ede0e 9123Author: Alex Deucher <alexdeucher@gmail.com> 9124Date: Fri Jan 29 12:59:46 2010 -0500 9125 9126 evergreen: add atombios crtc/pll functions 9127 9128commit bd8e04cb7b39f38b6958273582a9b324a9f0759a 9129Author: Alex Deucher <alexdeucher@gmail.com> 9130Date: Mon Feb 1 10:07:43 2010 -0500 9131 9132 evergreen: add atom support for digital outputs 9133 9134 analog is already supported by the existing code. 9135 9136commit 8ad40d3c32ad5b91725bd37fcade6bed504df421 9137Author: Alex Deucher <alexdeucher@gmail.com> 9138Date: Tue Jan 26 15:39:44 2010 -0500 9139 9140 evergreen: add crtc set base/format support 9141 9142commit 2c9cf0a07ff9e5a4989861bc2fdfd71d841013a8 9143Author: Alex Deucher <alexdeucher@gmail.com> 9144Date: Thu Dec 10 15:29:14 2009 -0500 9145 9146 evergreen: add lut support 9147 9148commit 65246545c3dd9bfef73e8a0f200bd5909b89a167 9149Author: Alex Deucher <alexdeucher@gmail.com> 9150Date: Tue Jan 26 12:10:02 2010 -0500 9151 9152 evergreen: add hw cursor support 9153 9154commit c05cad56b69d239fa2e69905d15f4f08b9db4c55 9155Author: Alex Deucher <alexdeucher@gmail.com> 9156Date: Thu Jan 28 01:28:52 2010 -0500 9157 9158 evergreen: add base asic support 9159 9160commit ed63e1b1abe8810b5da6b4140892337eef08a9ea 9161Author: Alex Deucher <alexdeucher@gmail.com> 9162Date: Fri Jan 29 12:42:33 2010 -0500 9163 9164 evergreen: add register and utility defines 9165 9166commit 019260ec4a9f9fbf2ac63a2ca3314aa308471f7e 9167Author: Alex Deucher <alexdeucher@gmail.com> 9168Date: Thu Dec 10 14:57:39 2009 -0500 9169 9170 evergreen: add chip enums 9171 9172commit e86ae564d8b94d2cc8d3c7cceb452a197ab6387a 9173Author: Alex Deucher <alexdeucher@gmail.com> 9174Date: Mon Feb 1 12:03:23 2010 -0500 9175 9176 atom: update to the latest upstream atombios.h changes 9177 9178commit 77b13a02c70842a58e0590d0243f0ae016c5a640 9179Author: Alex Deucher <alexdeucher@gmail.com> 9180Date: Mon Feb 1 12:16:20 2010 -0500 9181 9182 r500: fix CS section size missmatch in Xv code 9183 9184commit 5c11264517d1b9808415d47e9dedc345610f0f86 9185Author: Dave Airlie <airlied@linux.ie> 9186Date: Fri Jan 29 15:10:33 2010 +1000 9187 9188 radeon: further digital encoder cleanups. 9189 9190 this just cleans up a few more bits of the digital encoder setup. 9191 9192 Signed-off-by: Dave Airlie <airlied@redhat.com> 9193 9194commit 872a11fa9efb0b76cef83089ea1bb80792b94930 9195Author: Alex Deucher <alexdeucher@gmail.com> 9196Date: Thu Jan 28 01:19:23 2010 -0500 9197 9198 radeon: clean up dig encoder handling 9199 9200 - switch the var name to dig_encoder 9201 - quiet coherent messages 9202 - clean up dig encoder selection 9203 9204 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9205 9206commit c1a73ff51b371ecc6f57a81a33f644344b4e3f70 9207Author: Dave Airlie <airlied@linux.ie> 9208Date: Thu Jan 28 12:58:29 2010 +1000 9209 9210 radeon: add support for picking the digitial encoder 9211 9212 Testing on a W500 laptop found the LVDS + DisplayPort would conflict 9213 over the choice of digital encoder, and only one would work. 9214 9215 This patch fixes the DCE3 case where LVTMA requires the second digital 9216 encoder and lets DP/DVI use the first in most cases. 9217 9218 Signed-off-by: Dave Airlie <airlied@redhat.com> 9219 9220commit 95b56275d80dcee48a7927df124602c34cd72235 9221Author: Alex Deucher <alexdeucher@gmail.com> 9222Date: Wed Jan 27 15:48:25 2010 -0500 9223 9224 avivo: add some hotplug detect regs 9225 9226commit 5a4327f7784361933484895c9af751ccfa242d48 9227Author: Alex Deucher <alexdeucher@gmail.com> 9228Date: Tue Jan 26 16:06:00 2010 -0500 9229 9230 radeon: minor pll updates 9231 9232 add new fixed post divider option 9233 9234commit cbd642af7c76469d701471daea0c8d167567ccf8 9235Author: Alex Deucher <alexdeucher@gmail.com> 9236Date: Tue Jan 26 12:03:45 2010 -0500 9237 9238 radeon: use mmio bar size rather than hardcoded number for register ops 9239 9240 newer asics have larger mmio bars 9241 9242commit e5933fd74923a07aa401bdb4ecd8d1d641148440 9243Author: Alex Deucher <alexdeucher@gmail.com> 9244Date: Mon Jan 25 13:18:47 2010 -0500 9245 9246 KMS: fix no accel option 9247 9248 NO_ACCEL wasn't handled correctly with KMS. 9249 9250 fixes fdo bug 26198 9251 9252commit ba99a82ced1a6101365876723d22bcfec4607537 9253Author: Dave Airlie <airlied@redhat.com> 9254Date: Mon Jan 25 13:50:13 2010 +1000 9255 9256 radeon/kms: only enable EXA pixmaps under kms if we support render accel 9257 9258 I've seen RN50s with 64MB of RAM that are slow as molasses, this 9259 should fix them. 9260 9261 RH bug 556400 9262 9263 Signed-off-by: Dave Airlie <airlied@redhat.com> 9264 9265commit de40d36264364434541377c7f65f3836dc514fc0 9266Author: Kusanagi Kouichi <slash@ac.auone-net.jp> 9267Date: Fri Jan 22 17:18:51 2010 +0900 9268 9269 Remove unnecessary checks. 9270 9271 Signed-off-by: Kusanagi Kouichi <slash@ac.auone-net.jp> 9272 9273commit 30a19b75cc82b8e04c45e6684b84f9a4ccc0505b 9274Author: Alan Coopersmith <alan.coopersmith@sun.com> 9275Date: Tue Jan 19 14:54:04 2010 -0800 9276 9277 Fix configure with --disable-dri 9278 9279 AM_CONDITIONAL calls can't be made inside if statements, since 9280 automake won't know what to do when the if isn't true. 9281 9282 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 9283 9284commit bd011784c0609f1fd03dcb784744c8f5cdef33da 9285Author: Alex Deucher <alexdeucher@gmail.com> 9286Date: Tue Jan 19 11:59:38 2010 -0500 9287 9288 ATOM: Upstream parser updates 9289 9290commit 95d63e408cc88b6934bec84a0b1ef94dfe8bee7b 9291Author: Jerome Glisse <jglisse@redhat.com> 9292Date: Mon Jan 18 10:46:50 2010 +0100 9293 9294 r6xx/r7xx: emit relocation for FRAG & TILE buffer 9295 9296 FRAG & TILE buffer are unused but still they need 9297 to be associated with a valid relocation so that 9298 userspace can't try to abuse them to overwritte 9299 GART and then try to write anywhere in system 9300 memory. 9301 9302commit bbaf71fb46e3e0acdadcb91e9b1c73396bd9aa8e 9303Author: Alan Coopersmith <alan.coopersmith@sun.com> 9304Date: Fri Jan 15 15:09:27 2010 -0800 9305 9306 Update Sun license notices to current X.Org standard form 9307 9308 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 9309 9310commit 3d158716a60d5a113a541ea2f680b81a1be41ad2 9311Author: Dave Airlie <airlied@redhat.com> 9312Date: Thu Jan 14 11:21:33 2010 +1000 9313 9314 displayport: fix DDC on DVI->DP convertors. 9315 9316 Seems we have to not do auxch DDC if we aren't talking to a DP 9317 sink. 9318 9319commit 6a363f68415d37c302151581f2a86855dba39b67 9320Author: Alex Deucher <alexdeucher@gmail.com> 9321Date: Mon Jan 11 15:13:45 2010 -0500 9322 9323 rv100: fix typo in fix for bug 25992 9324 9325 Noticed by Maarten Maathuis. 9326 9327commit 7b01e1ee29f681bf1735ecded6445d12beeb52d8 9328Author: Alex Deucher <alexdeucher@gmail.com> 9329Date: Mon Jan 11 12:40:06 2010 -0500 9330 9331 rv100: reject modes >135 Mhz with DVI 9332 9333 Due to heat issues. fixes bug 25992 9334 9335 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9336 9337commit 74da9a66a420a40bb5d562f07e97ce6ac2eb010f 9338Author: Alex Deucher <alexdeucher@gmail.com> 9339Date: Fri Jan 8 16:51:43 2010 -0500 9340 9341 radeon/kms: add new strings for eDP and TV 9342 9343commit 61977f61f516b54d1fd9df34d0f0892fbf5227d7 9344Author: Alex Deucher <alexdeucher@gmail.com> 9345Date: Thu Jan 7 02:15:31 2010 -0500 9346 9347 radeon: fallback to i2c for edid if aux fails 9348 9349 In case of a DP->DVI adapter for example. 9350 9351commit 958d073869404f60e56dc0cc70b3e7de85904694 9352Author: Alex Deucher <alexdeucher@gmail.com> 9353Date: Thu Jan 7 02:12:00 2010 -0500 9354 9355 radeon: add initial support for eDP (embedded DisplayPort) 9356 9357 Should fix fdo bug 25931 9358 9359commit 944ae73f7cd5f9a1622b38e06a58565812cb102f 9360Author: Alex Deucher <alexdeucher@gmail.com> 9361Date: Thu Jan 7 01:52:35 2010 -0500 9362 9363 radeon: updated ObjectID.h 9364 9365commit 48aa5064aff4b9adf768e480df2312d4375e9c40 9366Author: Alex Deucher <alexdeucher@gmail.com> 9367Date: Mon Jan 4 11:05:44 2010 -0500 9368 9369 radeon: fix LVDS power sequence on Mac cards 9370 9371 Noticed by John R. Dunning. Fix taken from radeonfb. 9372 9373 I'm not sure if this sequence would be useful on any PC 9374 laptops or not so make it mac specific for now. 9375 9376commit 36bd69affc996c92c40b7360a7fbaa1a3a46abfd 9377Author: Gaetan Nadon <memsize@videotron.ca> 9378Date: Wed Dec 16 15:52:18 2009 -0500 9379 9380 configure.ac: remove unused sdkdir=$(pkg-config...) statement 9381 9382 The sdkdir variable isn't use, so remove the statement. 9383 9384 Acked-by: Dan Nicholson <dbn.lists@gmail.com> 9385 9386 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 9387 9388commit 9d0f3af7278dc939fd4e6f3ea69d9f488a9fbed7 9389Author: Alex Deucher <alexdeucher@gmail.com> 9390Date: Tue Dec 22 08:45:27 2009 -0500 9391 9392 radeon: add cvt timing if we only have panel w/h 9393 9394 fixes mac laptops without an edid 9395 9396 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9397 9398commit 6e1f5553c6d7e3b5d089af2e3d587efe95936855 9399Author: Alex Deucher <alexdeucher@gmail.com> 9400Date: Mon Dec 21 17:32:40 2009 -0500 9401 9402 ATOM: add new power table defs 9403 9404commit 4b05c47ac657f9a93d76221269761ed64c81f716 9405Author: Dave Airlie <airlied@redhat.com> 9406Date: Thu Dec 17 14:30:23 2009 +1000 9407 9408 radeon: use new libdrm api 9409 9410commit 1d620800bd79290967fa487decf798e318c45f25 9411Author: Ingmar Vanhassel <ingmar@exherbo.org> 9412Date: Sun Dec 20 21:04:56 2009 +0100 9413 9414 Replace shave with automake silent-rules 9415 9416 Which are enabled by default in Xorg's util-macros 1.3 if automake-1.11 9417 is used. 9418 9419 Signed-off-by: Ingmar Vanhassel <ingmar@exherbo.org> 9420 9421commit 52f6a2ff682ad1f8e26ef1cd7d380493dcd57752 9422Author: Gaetan Nadon <memsize@videotron.ca> 9423Date: Wed Dec 16 15:29:53 2009 -0500 9424 9425 src/Makefile.am: missing 3 header files in tarball 9426 9427 radeon_dummy_bufmgr.h 9428 radeon_vbo.h 9429 simple_list.h 9430 Replace leading spaces with tab for radeon_atomwrapper.h 9431 9432 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 9433 9434commit 3a30210d50b27f8772fc5045133940246764fce9 9435Author: Matthias Hopf <mhopf@suse.de> 9436Date: Tue Dec 15 10:53:48 2009 -0500 9437 9438 fix 200M freezes on VT switch if CRTC is disabled 9439 9440 It appears that RS4xx chips need to have the crtc 9441 enabled when the timing is programmed. 9442 9443 agd5f: minor fixes/cleanup of the original patch 9444 9445commit 299d395bd3f294239dee58ab7d607d7d2c657f61 9446Author: Corbin Simpson <MostAwesomeDude@gmail.com> 9447Date: Sun Dec 13 14:51:17 2009 -0800 9448 9449 dri: be more useful if disabling DRI due to KMS/UMS mismatch. 9450 9451 It's only fair. 9452 9453 Compile-tested only. 9454 9455 Signed-off-by: Dave Airlie <airlied@redhat.com> 9456 9457commit 0e5c9d87b5d7e0751df71cc8958ca5ccaed25104 9458Author: Alex Deucher <alexdeucher@gmail.com> 9459Date: Thu Dec 10 14:27:43 2009 -0500 9460 9461 ATOM: split set base/format into separate functions 9462 9463 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9464 9465commit 46630da5fd6f45bb8ea150b870162997480d69c7 9466Author: Alex Deucher <alexdeucher@gmail.com> 9467Date: Wed Dec 9 12:55:25 2009 -0500 9468 9469 AVIVO: add new PLL code 9470 9471 This should hopefully help the problems with flickering 9472 and blinking monitors reported on some systems. If there 9473 are problems, the old PLL algorithm can be selected with: 9474 Option "NewPLL" "FALSE" 9475 in the device section of your X config. 9476 9477 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9478 9479commit f082b1693d6f7f763ccf5a8436a89890ca2c6129 9480Author: Matthijs Kooijman <matthijs@stdin.nl> 9481Date: Thu Dec 3 12:10:14 2009 -0500 9482 9483 radeon: fix crtc2 dpms 9484 9485 noticed by Matthijs Kooijman on fdo bug 22140 9486 9487commit 3a96fbf3b84522cf1ba1b176e82b662222331c14 9488Author: Alex Deucher <alexdeucher@gmail.com> 9489Date: Wed Dec 2 18:02:33 2009 -0500 9490 9491 radeon: only read RADEON_PPLL_REF_DIV for ref div on pre-avivo 9492 9493 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9494 9495commit f03450796d2e9247a1228c4e2abb1dfad7aecddf 9496Author: Kusanagi Kouichi <slash@ac.auone-net.jp> 9497Date: Wed Dec 2 01:36:37 2009 -0500 9498 9499 radeon: Lift hardcoded limit from RADEONQueryImageAttributes 9500 9501 The dimension of an XvImage is limited to 2048 x 2048 even if an adaptor 9502 supports larger image. 9503 XvCreateImage and XvShmCreateImage lower the width or height of an image. 9504 XvPutImage and XvShmPutImage return BadValue. 9505 The cause is that 2048 is hardcoded in RADEONQueryImageAttributes. 9506 9507commit efbc2c80ab02879edf3b7b3d65b16c45ddce5017 9508Merge: 88a50a3 a612813 9509Author: Dave Airlie <airlied@redhat.com> 9510Date: Wed Dec 2 14:30:36 2009 +1000 9511 9512 Merge remote branch 'origin/displayport' 9513 9514commit 88a50a30df11a06263209340a42251851f8e2334 9515Author: Dave Airlie <airlied@redhat.com> 9516Date: Mon Nov 30 15:32:12 2009 +1000 9517 9518 r600: fix multi-operation in single batch support. 9519 9520 This ports the mesa DMA buffer handling with the 3 lists, 9521 9522 Signed-off-by: Dave Airlie <airlied@redhat.com> 9523 9524commit b2597deea3a3953ff50d54ff37e3c043eac409f4 9525Author: Kusanagi Kouichi <slash@ac.auone-net.jp> 9526Date: Mon Nov 30 11:10:50 2009 -0500 9527 9528 R600: fix some warnings 9529 9530commit 4d90dc3cb248e9d61c0c490bba80f6727fafd318 9531Author: Alex Deucher <alexdeucher@gmail.com> 9532Date: Mon Nov 30 11:05:38 2009 -0500 9533 9534 atom: pull misc mode info for lvds panel mode from bios tables 9535 9536 sync polarity, etc. This will likely fix LVDS problems 9537 on some laptops. 9538 9539 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9540 9541commit a612813c5f329f00a9271e7bb69abd2630a49403 9542Author: Alex Deucher <alexdeucher@gmail.com> 9543Date: Fri Nov 27 17:20:30 2009 -0500 9544 9545 radeon: clean up DP code 9546 9547 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9548 9549commit a8dbf7c23481501987971a9c0b6cb0760f86127f 9550Author: Dave Airlie <airlied@redhat.com> 9551Date: Fri Nov 27 10:16:00 2009 +1000 9552 9553 r600/accel: cleanup and merge a lot of the accel functions 9554 9555 All of the drawing ops were the exact same modulo the vtx size, 9556 this along with the vertex buffer wrapping code could all be consolidated 9557 into a smaller set of functions. 9558 9559 This also adds 2 VBO which we switch between, and merges a #define to 9560 enable the multiple operations in one CS under KMS mode. 9561 9562 Multi-operation still isn't working though. 9563 9564 Signed-off-by: Dave Airlie <airlied@redhat.com> 9565 9566commit 971e463b2cfdc1ba1c78a24fb439d33dd9155dfc 9567Author: Alex Deucher <alexdeucher@gmail.com> 9568Date: Thu Nov 26 12:45:15 2009 -0500 9569 9570 dce3: CV1/TV1OutputControl tables didn't go away until dce3.2 9571 9572 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9573 9574commit 95385f0906f371a1ed6e60eb3e597e699c7a3222 9575Author: Jerome Glisse <jglisse@redhat.com> 9576Date: Thu Nov 26 12:21:21 2009 +0100 9577 9578 kms: Fix resizing when acceleration is disabled 9579 9580 When acceleration is disabled we need to reallocate 9581 a new shadow framebuffer and we should also avoid 9582 calling any EXA function as EXA is disabled in such 9583 case. 9584 9585commit 8195385d3cedf70d89ca576371af6bca5045bbd2 9586Author: Dave Airlie <airlied@redhat.com> 9587Date: Thu Nov 26 15:43:54 2009 +1000 9588 9589 dce3: no CV1 or TV1 Output Control tables anymore. 9590 9591 According to the DCE3 docs we should only use DAC1/2 not 9592 CV1/TV OutputControl, also my rv730 bios doesn't have any 9593 CV1/TV tables. 9594 9595 Signed-off-by: Dave Airlie <airlied@redhat.com> 9596 9597commit 381b14d5422131ea8ff942ab06a328f6beca0d3f 9598Author: Dave Airlie <airlied@redhat.com> 9599Date: Thu Nov 26 15:28:15 2009 +1000 9600 9601 atombios: CV mode retrival was broken. 9602 9603 a) crev and frev are reversed 9604 b) my rv730 bios only has one mode in it, so 9605 bounds check the table size. 9606 9607commit a67a13c315a9235be052662e2a3270686cccc0f0 9608Author: Dave Airlie <airlied@redhat.com> 9609Date: Thu Nov 26 13:41:36 2009 +1000 9610 9611 radeon: fix typo in unused as of yet CV code 9612 9613commit cd9bfb5892eebef6dff14ac414dff074db5ddee4 9614Author: Dave Airlie <airlied@redhat.com> 9615Date: Thu Nov 26 06:43:13 2009 +1000 9616 9617 fixes displayport for me here 9618 9619commit 256de862a3c017dcb4ed0d8689294a552eda0d18 9620Author: Dave Airlie <airlied@redhat.com> 9621Date: Thu Nov 26 06:35:39 2009 +1000 9622 9623 displayport: cleanup call dp from dpms on instead. 9624 9625commit 05551295c5e0946745163f17e5c1d3d41b94bcbf 9626Author: Kristian Høgsberg <krh@bitplanet.net> 9627Date: Tue Nov 24 21:09:03 2009 -0500 9628 9629 dri2: Use drmGetDeviceNameFromFd() instead of open coded loop. 9630 9631commit 0061c4db1d3aecdca13efb5133ab8784dc37df95 9632Author: Dave Airlie <airlied@redhat.com> 9633Date: Wed Nov 25 15:39:10 2009 +1000 9634 9635 Revert "r600: enable multiple operations in one CS" 9636 9637 turn this off for now, on my rv635 desktop, I started getting blanks 9638 in places in firefox and the odd bit of font corruption, need to 9639 track that down. 9640 9641 This reverts commit 985a065518b1d33599de33f7fe082d3302db58a6. 9642 9643commit 8b28534bcf877557a5681fa3b4f107c74615d3de 9644Author: Dave Airlie <airlied@redhat.com> 9645Date: Wed Nov 25 13:24:44 2009 +1000 9646 9647 radeon/exa: change option to determine exa pixmap usage. 9648 9649 This moves to a boolean instead of using VRAM sizing. 9650 9651 as per Michel's suggestions on list. 9652 9653 Signed-off-by: Dave Airlie <airlied@redhat.com> 9654 9655commit f0acb16fa19844453adc6db3399977fba7c8a0db 9656Author: Dave Airlie <airlied@redhat.com> 9657Date: Wed Nov 25 12:04:47 2009 +1000 9658 9659 r600/xv: drop inited 3d false in xv code 9660 9661commit 985a065518b1d33599de33f7fe082d3302db58a6 9662Author: Dave Airlie <airlied@redhat.com> 9663Date: Wed Nov 25 11:54:08 2009 +1000 9664 9665 r600: enable multiple operations in one CS 9666 9667 This switches on multiple ops in a single CS under KMS/DRI2. 9668 9669 It gets for on a Pentium D 3 + rv740 from 330,000 to 500,000 9670 with x11perf -aa10text. 9671 9672 It also knocks a couple of seconds of gtkperf -a 9673 9674 Signed-off-by: Dave Airlie <airlied@redhat.com> 9675 9676commit af816ac752820255f245793b53a7cca5a4a49cd4 9677Author: Dave Airlie <airlied@redhat.com> 9678Date: Wed Nov 25 11:53:07 2009 +1000 9679 9680 r600: fixup problems with EXA operation reset for multiple ops 9681 9682 To put multiple ops into one CS, you can't just discard the whole 9683 IB. This add supports for reset the CS cdw to the correct place 9684 after an op discards. 9685 9686 Still doesn't enable the final accel bits. 9687 9688commit 3d8dcbc29323a3c644100bec13aa93f024653bd3 9689Author: Dave Airlie <airlied@redhat.com> 9690Date: Wed Nov 25 11:43:57 2009 +1000 9691 9692 kms: flush cs on close screen. 9693 9694 this is needed for server recycle. 9695 9696 Signed-off-by: Dave Airlie <airlied@redhat.com> 9697 9698commit 19f1a357944f9d8f4567a2691a68067ec033ccb7 9699Author: Dave Airlie <airlied@redhat.com> 9700Date: Wed Nov 25 11:41:52 2009 +1000 9701 9702 radeon: fix check for no work in operation 9703 9704commit 3a460a14b9603159f10d89da27b559c36a184e27 9705Author: Dave Airlie <airlied@redhat.com> 9706Date: Wed Nov 25 10:33:17 2009 +1000 9707 9708 r600: refactor code to help future acceleration speedups. 9709 9710 This changes the vertex buffer index to be an offset, and 9711 records the start of the vb for each operation and uses 9712 that to set the operations up. 9713 9714 This still flushes after each operation to make sure we have 9715 no regressions in non-kms/kms cases. 9716 9717 Signed-off-by: Dave Airlie <airlied@redhat.com> 9718 9719commit 797a3f0c71c94477eec565ea2c95553c6f66d9fd 9720Author: Dave Airlie <airlied@redhat.com> 9721Date: Wed Nov 25 10:07:59 2009 +1000 9722 9723 kms: allow prepare copy to fail without dying 9724 9725commit ba76acc76e0d0a51a1c488e4d5494a71e155a05b 9726Author: Dave Airlie <airlied@redhat.com> 9727Date: Tue Nov 24 15:38:47 2009 +1000 9728 9729 radeon/kms: drop special r600 indirect flush 9730 9731 This just merges it with the main kms cs flush 9732 9733commit 3c527d752d3cced219c6113acb932511b16e35fb 9734Author: Dave Airlie <airlied@redhat.com> 9735Date: Tue Nov 24 15:11:12 2009 +1000 9736 9737 r600: enable mixed pixmaps 9738 9739commit b9eb2380df5145c85b2ef694a04628c055c47429 9740Author: Alex Deucher <alexdeucher@gmail.com> 9741Date: Mon Nov 23 12:12:02 2009 -0500 9742 9743 DCE3.2: fix uniphy2 dvi issues 9744 9745 In some cases the atom transmitter table sets the 9746 golden value of this reg differently which some monitors 9747 don't like. I haven't had time to dig further, so this 9748 works around it for now. 9749 9750 Fixes fdo bug 24313 9751 9752 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9753 9754commit d0dd5122d8d75b7c4a72df07dc7f562b2e75dd5e 9755Author: Gaetan Nadon <memsize@videotron.ca> 9756Date: Mon Nov 23 09:25:05 2009 -0500 9757 9758 Makefile.am: add ChangeLog and INSTALL on MAINTAINERCLEANFILES 9759 9760 Now that the INSTALL file is generated. 9761 Allows running make maintainer-clean. 9762 9763commit 3d7888aa0493f46d826f4d80dea36f1f642f3c6b 9764Author: Dave Airlie <airlied@redhat.com> 9765Date: Mon Nov 23 15:26:15 2009 +1000 9766 9767 radeon: fix warning in printf 9768 9769commit 3a1a8b7b73424a769baf917bba2223ea6515b596 9770Author: Dave Airlie <airlied@redhat.com> 9771Date: Mon Nov 23 15:07:10 2009 +1000 9772 9773 radeon: surround mixed pixmaps usage with ifdef check for old servers 9774 9775commit e20af9c94982ec6487dae42c141d33cf6a7a2eb4 9776Author: Dave Airlie <airlied@redhat.com> 9777Date: Fri Nov 20 12:15:02 2009 +1000 9778 9779 radeon: avoid using hw pixmaps when we have little VRAM. 9780 9781 This patch returns NULL for pixmap creation when we are using 9782 mixed pixmaps and the pixmap has a size. 9783 9784 The size check is necessary for the front buffer. 9785 9786 We add a flag to force pixmap creation for certain pixmaps 9787 that need to be hw, like the DRI2 and Xv ones. 9788 9789 Idea from Michel and workarounds from Ben Skeggs. 9790 9791 v2: add Option "EXALowVRAM" to allow configuring this, value in MBs. 9792 9793 Signed-off-by: Dave Airlie <airlied@redhat.com> 9794 9795commit 543338842ba7fe7da4902947d386ea1cabd1cdf0 9796Author: Alex Deucher <alexdeucher@gmail.com> 9797Date: Fri Nov 20 16:38:43 2009 -0500 9798 9799 dp fix timing 9800 9801 aux channel atom command table delay is 10 usec units 9802 this gets the link training working, but alas, no image. 9803 9804commit dd3eab848cf352bb96c3d01fe6028d8a4a8e451e 9805Author: Alex Deucher <alexdeucher@gmail.com> 9806Date: Fri Nov 20 01:06:49 2009 -0500 9807 9808 r300: render target limit is 2560 9809 9810 limit was increased to 4021 in r4xx. 9811 9812 fixes fdo bug 25191. 9813 9814 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9815 9816commit df35d40d4ec260cdab57a9b36e4e96b2a6c2a827 9817Author: Alex Deucher <alexdeucher@gmail.com> 9818Date: Thu Nov 19 20:01:42 2009 -0500 9819 9820 DP: hack that makes DP work. 9821 9822 Something's wrong with the status check. 9823 9824commit f42412ec4dbd21f562a7fe4239c0a1f17ef1b8f2 9825Author: Alex Deucher <alexdeucher@gmail.com> 9826Date: Thu Nov 19 19:40:54 2009 -0500 9827 9828 wip 9829 9830commit eb9bc133fc426e67b397e661bfd22bf62009d9d3 9831Author: Dave Airlie <airlied@redhat.com> 9832Date: Fri Nov 20 09:23:31 2009 +1000 9833 9834 kms: recalculate the flush limits after screen resize. 9835 9836 When we resize the front buffer we need to reduce the flush limits 9837 appropriately. 9838 9839 Signed-off-by: Dave Airlie <airlied@redhat.com> 9840 9841commit f7f58ef4c042e492618665a6c5555e8e67387ab3 9842Author: Dave Airlie <airlied@redhat.com> 9843Date: Fri Nov 20 09:22:39 2009 +1000 9844 9845 radeon: r100/r200 have a 2047 scissor limit 9846 9847 We were overflowing this in my case with a 2704 width desktop, 9848 so videos were stopping around 700 bytes across the screen. 9849 9850 can I haz shatter already? 9851 9852 Signed-off-by: Dave Airlie <airlied@redhat.com> 9853 9854commit 54c09778d79de1516ee511d6d7d5b54dd4bf9130 9855Author: Alex Deucher <alexdeucher@gmail.com> 9856Date: Thu Nov 19 13:32:30 2009 -0500 9857 9858 DP wip 9859 9860 - use coherent 9861 - fix up lane/clock numbers 9862 9863commit 02d017a0552db2c8a48506c5b26c48a1a2ab75b6 9864Author: Gaetan Nadon <memsize@videotron.ca> 9865Date: Wed Oct 28 14:41:41 2009 -0400 9866 9867 INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206 9868 9869 Automake 'foreign' option is specified in configure.ac. 9870 Remove from Makefile.am 9871 9872commit 396ee9b7e7e3b1dd64532270c4e32cdb33857596 9873Author: Gaetan Nadon <memsize@videotron.ca> 9874Date: Wed Oct 28 14:09:09 2009 -0400 9875 9876 INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206 9877 9878 Add missing INSTALL file. Use standard GNU file on building tarball 9879 README may have been updated 9880 Remove AUTHORS file as it is empty and no content available yet. 9881 Remove NEWS file as it is empty and no content available yet. 9882 9883commit 1dcda0c4d5d506e18a7989ff05f2ad8569a73f23 9884Author: Gaetan Nadon <memsize@videotron.ca> 9885Date: Mon Oct 26 12:54:21 2009 -0400 9886 9887 Several driver modules do not have a ChangeLog target in Makefile.am #23814 9888 9889 The git generated ChangeLog replaces the hand written one. 9890 Update configure.ac to xorg-macros level 1.3. 9891 Use XORG_DEFAULT_OPTIONS which replaces four XORG_* macros 9892 Update Makefile.am to add ChangeLog target if missing 9893 Remove ChangeLog from EXTRA_DIST or *CLEAN variables 9894 This is a pre-req for the INSTALL_CMD 9895 9896commit 5b517bcc353e7111d2365a213505f27cfae30924 9897Author: Gaetan Nadon <memsize@videotron.ca> 9898Date: Thu Oct 22 12:34:16 2009 -0400 9899 9900 .gitignore: use common defaults with custom section # 24239 9901 9902 Using common defaults will reduce errors and maintenance. 9903 Only the very small or inexistent custom section need periodic maintenance 9904 when the structure of the component changes. Do not edit defaults. 9905 9906commit 605c869076985b69350c893cf2bbcc90de8dc6bf 9907Author: Alex Deucher <alexdeucher@gmail.com> 9908Date: Wed Nov 18 01:51:12 2009 -0500 9909 9910 more dp cleanup 9911 9912 - clean up handling of linkb, etc. 9913 - add support for ATOM_TRANSMITTER_ACTION_SETUP_VSEMPH 9914 to existing transmitter control function 9915 - fix lots of warnings 9916 9917commit bce1911f66f446b793125ac8cd5463bf6292629e 9918Author: Alex Deucher <alexdeucher@gmail.com> 9919Date: Wed Nov 18 00:04:33 2009 -0500 9920 9921 radeon: pull previous displayport work into master 9922 9923commit 437113124bbd6fb166825169eabec4dfde900dd9 9924Author: Alex Deucher <alexdeucher@gmail.com> 9925Date: Tue Nov 17 14:16:22 2009 -0500 9926 9927 radeon: deal with connectors sourced to the same encoder 9928 9929 Some systems have multiple connectors connected to the 9930 same encoder; e.g., DVI and HDMI connected to the same 9931 encoder with the same ddc line. Since we expose 9932 connectors as xrandr outputs, randr treats them separately 9933 which results in it trying to source the same encoder to 9934 different crtcs. If we have an HDMI and DVI-D port on the 9935 same encoder, pick the one to be considered connected 9936 based on the edid (HDMI if edid indicates HDMI, DVI 9937 otherwise). Also, don't turn off (dpms) encoders that are 9938 shared. 9939 9940 Fixes fdo bug 21767. 9941 9942 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9943 9944commit 9b8605d4e2bf7703c7848f3e2022a8c7328a3acd 9945Author: Alex Deucher <alexdeucher@gmail.com> 9946Date: Tue Nov 17 11:45:09 2009 -0500 9947 9948 radeon: fix potential memory leak in ddc code 9949 9950 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9951 9952commit ce8299962003de572122561a6f7f61eaccf633b2 9953Author: Dave Airlie <airlied@redhat.com> 9954Date: Tue Nov 17 15:23:30 2009 +1000 9955 9956 kms: check for 0 mask for cloned outputs. 9957 9958 Signed-off-by: Dave Airlie <airlied@redhat.com> 9959 9960commit 06d68f7fa3870f9601e837e1834d33e8395008b1 9961Author: Dave Airlie <airlied@redhat.com> 9962Date: Tue Nov 17 15:10:04 2009 +1000 9963 9964 kms: add cloning support. 9965 9966 We have to translate from the kernel encoder cloning to the randr 9967 connector cloning, we do this by constructing an encoder mask per output 9968 and an encoder cloning mask per output then comparing them. 9969 9970 Signed-off-by: Dave Airlie <airlied@redhat.com> 9971 9972commit 7587ce16ab0380337fe801f457c1d1d9b4141cc5 9973Author: Alex Deucher <alexdeucher@gmail.com> 9974Date: Mon Nov 16 19:16:03 2009 -0500 9975 9976 r600: remove un-needed format conversions 9977 9978 we do the normalization and xforms in the vertex 9979 shader. 9980 9981 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9982 9983commit ba605b5c9ff48765f2b66960aa4cc81b6235d012 9984Author: Alex Deucher <alexdeucher@gmail.com> 9985Date: Mon Nov 16 19:00:52 2009 -0500 9986 9987 r600: fix num format in vtx fetch 9988 9989 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 9990 9991commit 0c4710c67a2fee2061fc3da43c9f908585693cfa 9992Author: Alex Deucher <alexdeucher@gmail.com> 9993Date: Thu Nov 12 11:11:17 2009 -0500 9994 9995 radeon: man page updates 9996 9997commit 2af2744c80f3307e6d84edba9479be5bd36e1d03 9998Author: Alex Deucher <alexdeucher@gmail.com> 9999Date: Wed Nov 11 13:12:24 2009 -0500 10000 10001 radeon: fix support for external tmds on some legacy chips 10002 10003 Should fix bug 11801. 10004 10005 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10006 10007commit f17657fdc83d8f4c0386d2c7dade98de5b94acbe 10008Author: Alex Deucher <alexdeucher@gmail.com> 10009Date: Mon Nov 9 16:59:11 2009 -0500 10010 10011 r600/r700: typo, fix mask of DB_ALPHA_TO_MASK 10012 10013 noticed by glisse. 10014 10015commit 0ee7763fb33bf7fbc4ae1e8372cbc74578bfb720 10016Author: Alex Deucher <alexdeucher@gmail.com> 10017Date: Mon Nov 9 15:09:34 2009 -0500 10018 10019 IGP: some IGP chips report as AGP 10020 10021 Set bus type appropriately. fixes bug 25002 10022 10023 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10024 10025commit da1fcddaade58e371c7bee555ce32944c689f810 10026Author: Alex Deucher <alexdeucher@gmail.com> 10027Date: Thu Nov 5 14:13:49 2009 -0500 10028 10029 radeon: properly fix unused label warnings 10030 10031commit 6e496f46fdc90d0a6cef41e9d45a6458aaed1eca 10032Author: Alex Deucher <alexdeucher@gmail.com> 10033Date: Thu Nov 5 14:11:55 2009 -0500 10034 10035 Revert "Fix warning about unused defined labels in radeon_exa_funcs.c." 10036 10037 This reverts commit b6ac42cce2156880ac18b32e15acbff40b2f53e9. 10038 10039 This patch is very broken. 10040 10041commit b6ac42cce2156880ac18b32e15acbff40b2f53e9 10042Author: Tomáš Chvátal <scarabeus@gentoo.org> 10043Date: Sat Oct 31 23:07:07 2009 +0100 10044 10045 Fix warning about unused defined labels in radeon_exa_funcs.c. 10046 10047 Signed-off-by: Tomáš Chvátal <scarabeus@gentoo.org> 10048 10049commit 87d7235790866f9c19ef08972d5237f09d940cd9 10050Author: Alex Deucher <alexdeucher@gmail.com> 10051Date: Thu Nov 5 10:23:03 2009 -0500 10052 10053 EXA: fallback if no pMaskPicture->pDrawable 10054 10055 A solid or gradient mask could be used for blending 10056 the source picture onto the destination picture. 10057 10058 Fixes fdo bug 24838 10059 10060 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10061 10062commit 873897c54429e230a23011a9dd2f1069f4eb7752 10063Author: Alex Deucher <alexdeucher@gmail.com> 10064Date: Thu Nov 5 00:51:11 2009 -0500 10065 10066 DCE3+: call transmitter init on mode set 10067 10068 Generally this is done at post, but might not always 10069 be done with softboot or for connectors on docking 10070 stations. 10071 10072 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10073 10074commit 3028374488cc0f34942ac372c8d05cf15898a613 10075Author: Alex Deucher <alexdeucher@gmail.com> 10076Date: Wed Nov 4 18:48:03 2009 -0500 10077 10078 ATOM/DCE3+: fix up usPixelClock calculation for Transmitter tables 10079 10080 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10081 10082commit 52973e8f2c4104de60d00c0e5c29a31f77bf3de7 10083Author: Alex Deucher <alexdeucher@gmail.com> 10084Date: Sun Nov 1 13:58:03 2009 -0500 10085 10086 radeon/atom: fix tv-out on r5xx 10087 10088 fixes bug 24796 10089 10090 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10091 10092commit 19b4f5f9b647cd470616b8ec1d34f1e4afa357ad 10093Author: Alex Deucher <alexdeucher@gmail.com> 10094Date: Tue Oct 27 11:55:20 2009 -0400 10095 10096 ATOM: fix up DVO for DCE 3.x 10097 10098 AdjustDisplayPll takes care of this for us. 10099 10100commit 5a0019126a57138ee506d9a66738c9e8b75cbb96 10101Author: Alex Deucher <alexdeucher@gmail.com> 10102Date: Tue Oct 27 11:39:06 2009 -0400 10103 10104 radeon: fix DVO on AVIVO chips 10105 10106 DVO generally requires 2x ppll on AVIVO cards due to the 10107 way the tmds chip is wired up. Because of this, disable 10108 cloning of DVO outputs. 10109 10110 fixes fdo bug 21857. 10111 10112 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10113 10114commit 2a6eeec72d6bb04a32225883f431c1d2e8cff123 10115Author: Alex Deucher <alexdeucher@gmail.com> 10116Date: Tue Oct 27 11:18:44 2009 -0400 10117 10118 atom: loosen pll min output limits 10119 10120 Limiting the pll output range is a good thing generally as 10121 it limits the number of possible pll combinations for a given 10122 frequency presumably to the ones that work best on each card. 10123 That's why the limits are in the bios tables. However, certain 10124 duallink DVI monitors seem to like pll combinations that would 10125 be limited by this at least on pre-DCE 3.0 r6xx hardware. This 10126 might need to be adjusted per family or per clock range in the 10127 future. 10128 10129 Fixes fdo bug 24727. 10130 10131 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10132 10133commit f0d9d80fee4176eaba9435a9539f29d0eefe2a87 10134Author: Luca Tettamanti <kronos.it@gmail.com> 10135Date: Sat Oct 24 16:45:23 2009 -0400 10136 10137 Disable color tiling on r600/r700 10138 10139 Not supported yet. 10140 10141commit ac499c3c0ea92f83ebd8127bb8cb5de625de6c94 10142Author: Alex Deucher <alexdeucher@gmail.com> 10143Date: Fri Oct 23 10:48:55 2009 -0400 10144 10145 radeon: clamp the internal FB map to the aperture size v2 10146 10147 We don't use the invisible memory yet and on cards with 10148 large amounts of vram this can cause the top of GART 10149 calculation to overflow. 10150 10151 Fixes bug fdo bug 24301: 10152 http://bugs.freedesktop.org/show_bug.cgi?id=24301 10153 10154 v2: only clamp cards with more than 512 MB. This seems 10155 to cause problems on some older cards due to the way the 10156 drm and ddx set up the internal memory map. 10157 10158commit e57b54daf318ff59315a1c9ed21934f288c5c782 10159Author: Alex Deucher <alexdeucher@gmail.com> 10160Date: Thu Oct 22 16:07:58 2009 -0400 10161 10162 R7xx: program additional CUR/GRPH regs for 40 bit addresses 10163 10164 The *_HIGH regs are reversed. The secondary ones are in the 10165 primary block and vice versa. 10166 10167commit 4cf06dfba617529291ce4b4c306c4fc1bba110ee 10168Author: Alex Deucher <alexdeucher@gmail.com> 10169Date: Thu Oct 22 02:08:45 2009 -0400 10170 10171 ATOM: use indirect addressing for regs >= 0x10000 10172 10173commit 66b194a78c470cb3978f310828dd96c3f3e96944 10174Author: Alex Deucher <alexdeucher@gmail.com> 10175Date: Wed Oct 21 15:58:18 2009 -0400 10176 10177 ATOM: fix up get clock info 10178 10179 Newer revisions of the atom firmware table have 10180 changed. This helps select better pll dividers 10181 in some cases. Noticed by Mathias Froehlich. 10182 10183commit bd89b7501f294ac645390ef144df569953c81dc4 10184Author: Alex Deucher <alexdeucher@gmail.com> 10185Date: Sun Oct 18 16:21:36 2009 -0400 10186 10187 radeon: fixup CustomEDID option 10188 10189 - add support for analog option to force edid to analog 10190 mode 10191 - fix compilation on older xservers 10192 - updated man page 10193 10194commit 255ac8d54e3dc12e579e05edb41b40946a00eab8 10195Author: Alex Deucher <alexdeucher@gmail.com> 10196Date: Tue Oct 13 12:27:52 2009 -0400 10197 10198 radeon: add XV_CRTC attribute for textured video 10199 10200 This attribute allows the user to override which 10201 crtc is synced with when XV_VSYNC is enabled. This 10202 is useful for clone modes where the user can might want 10203 to override the default. 10204 10205commit 801d900add4731f3c424ebb780ad2dbd7a1c6a25 10206Author: Alex Deucher <alexdeucher@gmail.com> 10207Date: Mon Oct 12 10:00:57 2009 -0400 10208 10209 ATOM: minor cleanups to pll setup 10210 10211commit f4407962cd7b272e0860319f11f6a6583ef226c2 10212Author: Csillag Kristof <csillag.kristof@united-consult.hu> 10213Date: Fri Oct 9 18:00:09 2009 -0400 10214 10215 Fix some issues with CustomEDID support 10216 10217 - Fixes a memory allocation problem introduced with the previous 10218 version of the CustomEDID patch (now in GIT), and 10219 - Enhances the syntax of the CustomEDID option so that I can force 10220 digital output, even if the EDID was acquired over an analog link. 10221 10222commit de55995e82c3875f70b6394fff440d695d062113 10223Author: Alex Deucher <alexdeucher@gmail.com> 10224Date: Fri Oct 9 11:07:30 2009 -0400 10225 10226 r600 EXA: fix up mask reg mixup 10227 10228commit a168caf5349b757873ad89d977772feaa7fb6e58 10229Author: Alex Deucher <alexdeucher@gmail.com> 10230Date: Fri Oct 9 10:32:26 2009 -0400 10231 10232 r600 EXA: fix some compile warnings 10233 10234commit fb8dda2928bd0def2b34c8c8e414129e650d6eb5 10235Author: Alex Deucher <alexdeucher@gmail.com> 10236Date: Fri Oct 9 10:30:21 2009 -0400 10237 10238 r600 EXA: properly check planemask 10239 10240 We only support per-component masks. 10241 10242commit 1b25bad2d8ebe481ffb4b0fd3a85ab636bd4fec2 10243Author: Dave Airlie <airlied@linux.ie> 10244Date: Fri Oct 9 20:09:52 2009 +1000 10245 10246 Revert "radeon: clamp the internal FB map to the aperture size" 10247 10248 This reverts commit 5f846360c46f5a989f5d0fde6d251cdbd61d4968. 10249 10250 Numerous reports of system hangs since this, I'm guessing 10251 some sort of conflict with the drm memory setup. This code 10252 has always been fragile between kernel/userspace drivers. 10253 10254commit a956d478043b44c49aed39c7aa3d576ef32cadc5 10255Author: Dave Airlie <airlied@linux.ie> 10256Date: Fri Oct 9 20:08:31 2009 +1000 10257 10258 Revert "kms: setup colormap and gamma correctly." 10259 10260 This reverts commit 60d9685abddccec17c1a9a5ec48cbe9c92543e0f. 10261 10262 which breaks colormap on 1.6 X server 10263 10264 Conflicts: 10265 10266 src/drmmode_display.c 10267 10268commit a493feb94c0d84aaf5bfd84c6940d7f03974dd86 10269Author: Dave Airlie <airlied@redhat.com> 10270Date: Fri Oct 9 15:05:25 2009 +1000 10271 10272 kms: don't use scratch pixmaps at all 10273 10274 use our own wrapper to allocate a pixmap for wrapping a bo in. 10275 10276commit 0caffbd6fd6b356b9b7d715c015285eaa91e3e36 10277Author: Dave Airlie <airlied@redhat.com> 10278Date: Fri Oct 9 15:04:56 2009 +1000 10279 10280 drmmode: reload cursors on modeset 10281 10282commit 02e12ae6be7bc1976f63848fa2854d320d5ab36e 10283Author: Alex Deucher <alexdeucher@gmail.com> 10284Date: Wed Oct 7 19:56:58 2009 -0400 10285 10286 ATOM: rework crtc modeset 10287 10288 - clean up tv timing handling 10289 - unify SetCRTCTiming and SetCRTCDTDTiming interfaces 10290 10291commit d499eeaf22f77a1294e99aa38a50aa6810bb684a 10292Author: Alex Deucher <alexdeucher@gmail.com> 10293Date: Wed Oct 7 17:43:11 2009 -0400 10294 10295 ATOM: reorder crtc dpms based on bios recommendations 10296 10297commit 4ccd2a21f9c50515b4246f35454f76ef49006c76 10298Author: Alex Deucher <alexdeucher@gmail.com> 10299Date: Wed Oct 7 16:30:03 2009 -0400 10300 10301 ATOM: add support for AdjustDisplayPll table 10302 10303 Depending on the output and clock, this table will 10304 adjust the pixelclock accordingly. 10305 10306commit 5f846360c46f5a989f5d0fde6d251cdbd61d4968 10307Author: Alex Deucher <alexdeucher@gmail.com> 10308Date: Wed Oct 7 17:01:55 2009 -0400 10309 10310 radeon: clamp the internal FB map to the aperture size 10311 10312 We don't use the invisible memory yet and on cards with 10313 large amounts of vram this can cause the top of GART 10314 calculation to overflow. 10315 10316 Fixes bug fdo bug 24301: 10317 http://bugs.freedesktop.org/show_bug.cgi?id=24301 10318 10319commit 186bf719c394116341abff471fb4b4803650da9a 10320Author: Dave Airlie <airlied@redhat.com> 10321Date: Wed Oct 7 17:23:36 2009 +1000 10322 10323 radeon: fix rotation since mixed pixmaps. 10324 10325 Passing null in here stops the mixed pixmap code doing it wrong, 10326 but I'm not 100% sure how it is meant to work, we have the 10327 same issue with the real front bo by the looks of it. 10328 10329commit e08411af1aa8c7d4233ba593b84360397cdbb307 10330Author: Roel Kluin <roel.kluin@gmail.com> 10331Date: Tue Oct 6 18:31:17 2009 -0400 10332 10333 radeon: Fix duplicated bit settings 10334 10335 [agd5f: adapted from kms patch] 10336 10337 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10338 10339commit e59ae08270711512e64b70f79b6476cc2c52d230 10340Author: Michel Dänzer <daenzer@vmware.com> 10341Date: Tue Oct 6 00:13:09 2009 +0200 10342 10343 Don't call radeon_cs_flush_indirect() without KMS. 10344 10345 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=24316 . 10346 10347commit 457646d734558672402b6bcd15cadb66741edbed 10348Author: Jerome Glisse <jglisse@redhat.com> 10349Date: Mon Oct 5 15:52:42 2009 +0200 10350 10351 kms: assume accel is working is info ioctl return -EINVAL 10352 10353 -EINVAL should only happen if the info ioctl doesn't support 10354 the get accel query. This patch assume that on such case accel 10355 is working. 10356 10357commit 9d596562496863d65850306d2126d8df98464de4 10358Author: Dave Airlie <airlied@linux.ie> 10359Date: Mon Oct 5 18:35:20 2009 +1000 10360 10361 kms: only touch gamma stuff on 1.7 10362 10363 for correct colormap operation you probably want 1.7 10364 10365commit 876f43d4682df4d47d3e158f75e0ca6040f41c70 10366Author: Alex Deucher <alexdeucher@gmail.com> 10367Date: Mon Oct 5 02:23:03 2009 -0400 10368 10369 R1/2xx EXA: Minimise number of draw primitives used for Composite operations. 10370 10371commit 86cafb8affea448bdf58300044e755201b822d2a 10372Author: Dave Airlie <airlied@redhat.com> 10373Date: Mon Oct 5 14:17:25 2009 +1000 10374 10375 kms: don't use scratch pixmaps when copying fbcon. 10376 10377 scratch pixmaps seem to interact badly with mixed pixmaps, it appears 10378 some state may be getting left around in the privates somewhere, since 10379 scratch pixmap headers don't get destroyed. 10380 10381commit f8471512ea9f1d38140dfe98a0f832e9f935f51b 10382Author: Csillag Kristof <csillag.kristof@united-consult.hu> 10383Date: Sun Oct 4 18:18:56 2009 -0400 10384 10385 radeon: add support for Custom EDID 10386 10387 Allows you to specify an edid per output from a file 10388 to override what is detected by DDC. Useful for 10389 problematic monitors or KVM switches that block 10390 DDC. Specifying an EDID that is not compatible with 10391 your monitor could damage your monitor so use with 10392 caution. 10393 10394 agd5f: cache the custom edid at startup so we don't 10395 have to read it from file every time the output is 10396 queried. 10397 10398commit 60d9685abddccec17c1a9a5ec48cbe9c92543e0f 10399Author: Dave Airlie <airlied@redhat.com> 10400Date: Mon Oct 5 12:41:30 2009 +1000 10401 10402 kms: setup colormap and gamma correctly. 10403 10404 We need to program the gamma at modeset. no need to use 10405 old cmap interface at all. 10406 10407commit b0b10e548fd69b3ff67a0f9b71456b89c61cb4d6 10408Author: Alex Deucher <alexdeucher@gmail.com> 10409Date: Sun Oct 4 17:07:32 2009 -0400 10410 10411 R6/7xx EXA: Add support for BGRA picture formats. 10412 10413 Also add mysteriously missing case statements for ABGR destination formats. 10414 10415commit 8404c37b2ff95aba31db97676f106ec2f30fda7f 10416Author: Dave Airlie <airlied@linux.ie> 10417Date: Sun Oct 4 18:18:51 2009 +1000 10418 10419 radeon: fix build without kms 10420 10421commit 9460ea864b12ec1fbd11c5d9a20bb5a4279d9d3d 10422Author: Michel Dänzer <daenzer@vmware.com> 10423Date: Sat Oct 3 16:33:33 2009 +0200 10424 10425 Fix KMS on big endian machines. 10426 10427 Requires at least xserver 1.7 to work properly. 10428 10429 Also make sure the front buffer is and stays tiled if colour tiling is enabled. 10430 10431commit aee6b6f2c120baf477b4845ddc1a94637c31db2f 10432Author: Michel Dänzer <daenzer@vmware.com> 10433Date: Sat Oct 3 16:33:33 2009 +0200 10434 10435 EXA: Bail earlier from PrepareCopy hook on trivially unsupportable conditions. 10436 10437commit 30031b169def5caadb394225919a6079a0957142 10438Author: Michel Dänzer <daenzer@vmware.com> 10439Date: Sat Oct 3 16:33:33 2009 +0200 10440 10441 R3/5xx EXA: Add support for BGRA picture formats. 10442 10443 Also add mysteriously missing case statements for ABGR destination formats. 10444 10445commit 4b4ce36081ca151c24e028c54b59986f41731a73 10446Author: Michel Dänzer <daenzer@vmware.com> 10447Date: Sat Oct 3 16:33:32 2009 +0200 10448 10449 R3/5xx EXA: Minimise number of draw primitives used for Composite operations. 10450 10451 This should reduce the kernel CS checker overhead, if nothing else. 10452 10453 I'll leave porting this to other chipset families to others who can test it. 10454 10455commit eade1e5be159c9f2965d611925596d33cab11d6d 10456Author: Michel Dänzer <daenzer@vmware.com> 10457Date: Sat Oct 3 16:33:32 2009 +0200 10458 10459 EXA: Don't always flush when switching between 2D/3D engines with KMS. 10460 10461 This seems to work fine here now and help 2D performance quite a bit, let's see 10462 if it breaks anybody else's setup... 10463 10464commit 926e414fc835ace141c066830d11d8ce32dbb06c 10465Author: Michel Dänzer <daenzer@vmware.com> 10466Date: Sat Oct 3 16:33:32 2009 +0200 10467 10468 KMS: Double-buffer textured video source image upload. 10469 10470 In order to avoid stalling on previous frame. 10471 10472 OTOH without KMS we can't do this but have to wait for the previous frame to 10473 finish rendering. 10474 10475commit 6a3b75fae4147244212298ff55c7ab36c3d80d30 10476Author: Michel Dänzer <daenzer@vmware.com> 10477Date: Sat Oct 3 16:33:31 2009 +0200 10478 10479 KMS: Don't hide HW cursor when it's updated. 10480 10481 Inspired by the intel driver; might avoid/reduce HW cursor flicker in some 10482 cases. 10483 10484commit c4ba15b4871c70cced7f1761e24be692cdb28d2b 10485Author: Michel Dänzer <daenzer@vmware.com> 10486Date: Sat Oct 3 16:33:31 2009 +0200 10487 10488 Fix some compiler warnings. 10489 10490commit cc45856a18dd3e6f7e44d9eb507b31419da70977 10491Author: Kenneth Graunke <kenny@whitecape.org> 10492Date: Fri Oct 2 14:31:36 2009 -0400 10493 10494 radeon: Remove gatos message from driver startup 10495 10496 Remove a message on driver startup which directs people to the old GATOS 10497 website; said site contains highly out of date software which is no longer 10498 necessary. 10499 10500commit d302481493ff61eb7f8648b7afe7407b6b653cf7 10501Author: Jerome Glisse <jglisse@redhat.com> 10502Date: Wed Sep 16 15:18:40 2009 +0200 10503 10504 radeon/kms: fallback to shadowfb if kernel report acceleration is off 10505 10506 This will fallback to shadowfb is GPU accel failed somewhere during 10507 KMS kernel init. 10508 10509commit 7968e1fb89f6b59d1654df48249bf4b81990c008 10510Author: Alex Deucher <alexdeucher@gmail.com> 10511Date: Fri Sep 25 19:05:38 2009 -0400 10512 10513 radeon: fix segfault in MMIO path in RADEONInit3DEngine 10514 10515 fixes fdo bug 24158. 10516 10517 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10518 10519commit 0380a9cf98d34e88dd81370ade5525680ec89c02 10520Author: Alex Deucher <alexdeucher@gmail.com> 10521Date: Fri Sep 25 16:46:43 2009 -0400 10522 10523 kms/r600: add support for vline relocs 10524 10525 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10526 10527commit 3efecebb10de7f7bacf9f8c57ae20fd508097294 10528Author: Alex Deucher <alexdeucher@gmail.com> 10529Date: Fri Sep 25 14:44:41 2009 -0400 10530 10531 radeon: fix vline handling for kms 10532 10533 drm crtc ids do not correspond to actual hw crtcs, 10534 as such the vline stuff was never enabled for Xv. 10535 10536 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10537 10538commit 9733dcde0a21b7503aa20254724f2910b541b990 10539Author: Dave Airlie <airlied@linux.ie> 10540Date: Sat Sep 26 06:15:58 2009 +1000 10541 10542 r600: fix build 10543 10544commit ff18595eace42ddfc931f4f948cb5adf59ebcf52 10545Author: Dave Airlie <airlied@linux.ie> 10546Date: Sat Sep 26 06:04:51 2009 +1000 10547 10548 r600/xv: fix dst bo write domain 10549 10550commit 718a611f3cc7f5eea95cd73a9dcc7913c922fcbc 10551Author: Dave Airlie <airlied@linux.ie> 10552Date: Sat Sep 26 06:03:51 2009 +1000 10553 10554 radeon: fix zaphod 10555 10556commit 86c367be517e992aabc1bcfea2a2dde1a1ae520c 10557Author: Alex Deucher <alexdeucher@gmail.com> 10558Date: Fri Sep 25 13:03:44 2009 -0400 10559 10560 r600: fix bo accounting for Xv 10561 10562 fixes Xv after changes in: 10563 8f80e37eed3ec028718b4e71bbb9b598847fd94e 10564 10565 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10566 10567commit 8f80e37eed3ec028718b4e71bbb9b598847fd94e 10568Author: Dave Airlie <airlied@linux.ie> 10569Date: Fri Sep 25 11:37:38 2009 +1000 10570 10571 r600: space check wasn't correct. 10572 10573 the r6xx space check code was incorrect for the DFS case, since 10574 the dst bo was in GTT but the hardcoded cp_start function 10575 assumed the bos were where it thought. 10576 10577 Ripped out assumptions and replaced with code more like other radeons. 10578 10579commit 90669f6cb5ada9067d6fccc7f54c69b367862f42 10580Author: Alex Deucher <alexdeucher@gmail.com> 10581Date: Thu Sep 24 10:40:40 2009 -0400 10582 10583 radeon: remove old pre-randr rotation cruft 10584 10585 leftover from the randr 1.2 conversion. 10586 10587 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 10588 10589commit 97a4e747bfac14f34646c55ddf639e8fe22f2f55 10590Author: Michael Olbrich <m.olbrich@pengutronix.de> 10591Date: Sun Sep 20 14:19:41 2009 +0200 10592 10593 use AC_CHECK_HEADER instead of AC_CHECK_FILE 10594 10595 AC_CHECK_FILE is not possible when cross-compiling. Use AC_CHECK_HEADER / 10596 AC_PREPROC_IFELSE instead. 10597 10598 [ Michel Dänzer: Shuffled things around slightly to make it work on my setup ] 10599 10600 Signed-off-by: Michael Olbrich <m.olbrich@pengutronix.de> 10601 Signed-off-by: Michel Dänzer <daenzer@vmware.com> 10602 10603commit 579070ed9f0218a975e840331b6fe9fa8fd05789 10604Author: Alex Deucher <alexdeucher@gmail.com> 10605Date: Sat Sep 19 19:15:14 2009 -0400 10606 10607 kms: disable dga 10608 10609 Can be dangerous with kms: 10610 http://lists.freedesktop.org/archives/xorg-devel/2009-September/002113.html 10611 10612commit da7487f6ac7b32ed7866af53da2925aa52ed7aae 10613Author: Alex Deucher <alexdeucher@gmail.com> 10614Date: Sat Sep 19 13:08:02 2009 -0400 10615 10616 kms: init dga using xf86DiDGAInit 10617 10618 fixes cursor problems in games like UT2004 and Penumbra Overture 10619 10620 see fdo bug 24034 10621 10622commit 52279251fae9df99c569c16e2522bbd346d8ec38 10623Author: Dave Airlie <airlied@redhat.com> 10624Date: Fri Sep 18 14:29:32 2009 +1000 10625 10626 avivo: disable VGA rendering core when starting X. 10627 10628 this thing can be rendering to VRAM when we don't expect it. 10629 turn it off. 10630 10631 Signed-off-by: Dave Airlie <airlied@redhat.com> 10632 10633commit 2c46bafcb77c4125a27c18ad6ca2f6de5f143a2d 10634Author: Pauli Nieminen <suokkos@gmail.com> 10635Date: Thu Sep 17 15:13:51 2009 +0300 10636 10637 Fix configuration to disable KMS if not finding libdrm_radeon. 10638 10639 Most of KMS building was disabled if LIBDRM_RADEON was not detected. But at 10640 least configure was reporting KMS enabled even tough it was realy disabled. 10641 10642commit ee9f6d802988a4d7f7985687036db854f4a14fe5 10643Author: Pauli Nieminen <suokkos@gmail.com> 10644Date: Mon Sep 14 16:36:48 2009 +0300 10645 10646 radeon: Add missing libdrm_radeon cflags to compiler paramaters. 10647 10648 This fixes compilation if libdrm_radeon is installed to non-standard location like /opt. 10649 10650commit ac853ca0a950d6147cd6a0b4a1b089f5d2f080ea 10651Author: Alex Deucher <alexdeucher@gmail.com> 10652Date: Fri Sep 11 14:25:31 2009 -0400 10653 10654 atom: fix typo in asus quirks 10655 10656 Should be DVI-I, not DVI-D 10657 10658commit a88f60e7b789cd23a79e6e18a7e3fa008d15aa77 10659Author: Alex Deucher <alexdeucher@gmail.com> 10660Date: Thu Sep 10 13:42:31 2009 -0400 10661 10662 r600: add CS UTS and DFS hooks 10663 10664commit 0bb0ff0e55aaf8e64017d204e376a14f0a7384a3 10665Author: Dave Airlie <airlied@linux.ie> 10666Date: Thu Sep 10 11:48:00 2009 +1000 10667 10668 r6xx: fix EXA crash with gtkperf -a 10669 10670 unmapping was getting unbalanced. 10671 10672commit 10a58d54857484b6e89763ecf463f54dc290c702 10673Author: Dave Airlie <airlied@redhat.com> 10674Date: Wed Sep 9 14:24:49 2009 +1000 10675 10676 kms: add property support. 10677 10678 this adds support to the userspace DDX to get properties from the kernel 10679 and expose them over randr. Its most liberated from the Intel driver. 10680 10681commit fad9ee6cdae5ae45e6406e3071a84276f1b59a85 10682Author: Alex Deucher <alexdeucher@gmail.com> 10683Date: Tue Sep 8 17:50:50 2009 -0400 10684 10685 radeon: replace dga code with DiDGA 10686 10687commit 651fe5a47435c6a5cc1cdfb33c5c6601d692adcb 10688Author: Adam Jackson <ajax@redhat.com> 10689Date: Tue Sep 8 13:21:07 2009 +1000 10690 10691 radeon: fix dri2 struct zeroing 10692 10693commit 6990f2ac6478bf92929a4400ef84fb2142699204 10694Merge: 917f2d7 853f4c3 10695Author: Dave Airlie <airlied@redhat.com> 10696Date: Tue Sep 8 11:26:32 2009 +1000 10697 10698 Merge branch 'r6xx-cs' 10699 10700commit 853f4c3d1ea8f975ab2855f18d3ae336a4095091 10701Author: Dave Airlie <airlied@redhat.com> 10702Date: Tue Sep 8 11:25:39 2009 +1000 10703 10704 r600: more alignment fixups + vb map/unmap 10705 10706 I'm not so sure the vb map/unmap is a good idea, I think 10707 it pretty much locksteps the cpu/gpu, so we should really 10708 work out if we really need to flush this often, since 10709 mesa doesn't have to and we are just doing 3D ops. 10710 10711commit 917f2d7cd1815279710939edc27133ca6609460f 10712Author: Alex Deucher <alexdeucher@gmail.com> 10713Date: Sun Sep 6 10:53:29 2009 -0400 10714 10715 RV280: add agp quirk 10716 10717 deb bug 545040 10718 10719commit 78fcbf577ad6eba6399cc39f74b7ce5f9c8e265e 10720Author: Michel Dänzer <daenzer@vmware.com> 10721Date: Thu Sep 3 14:55:05 2009 +0200 10722 10723 EXA: Check for solid/gradient pictures the same way for all generations. 10724 10725 In particular, also catch them for >= R300. 10726 10727commit 794ae743c305331741e7cf45dd71c755fe325ed5 10728Author: Michel Dänzer <daenzer@vmware.com> 10729Date: Wed Sep 2 09:19:17 2009 +0200 10730 10731 EXA: Fix Composite restart dst/src switcheroo. 10732 10733 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=22636 . 10734 10735 Now, where's that brown paper bag? :} 10736 10737commit c4ab50c5eafca3d04129a72453496eb8afb26b34 10738Author: Alex Deucher <alexdeucher@gmail.com> 10739Date: Tue Sep 1 13:43:00 2009 -0400 10740 10741 radeon: add some notes about DVO setup 10742 10743commit 22fbd6476f4b3513378e848affc4dd269e9d4498 10744Author: Dave Airlie <airlied@itt42.(none)> 10745Date: Tue Sep 1 15:22:57 2009 +1000 10746 10747 r100/r200: seriously WRAP is illegal for rectangular textures 10748 10749 unless we scale all the bos up to POT 10750 10751commit 8f4196e88855f10762254fca9e0a0988e7b5562f 10752Author: Alex Deucher <alexdeucher@gmail.com> 10753Date: Mon Aug 31 19:41:59 2009 -0400 10754 10755 r6xx/r7xx: various CS fixes from Dave 10756 10757commit a32384f9b9ed73c14484a380dfa14652ed478ad6 10758Author: Alex Deucher <alexdeucher@gmail.com> 10759Date: Mon Aug 31 13:26:05 2009 -0400 10760 10761 r1xx: fix textured video 10762 10763 Copy/paste bug from when kms support was added. 10764 10765commit 3d6bf0f8cf128ee67d448491be44d345547b75c5 10766Author: Pauli Nieminen <suokkos@gmail.com> 10767Date: Fri Aug 28 17:05:44 2009 +0300 10768 10769 radeon: Fix DRI2BufferPtr to be DRI2Buffer2Ptr for xserver 1.6. 10770 10771 Bump requirement for xserver in KMS mode to 1.6.2 for DRI2Buffer2Ptr. 10772 10773commit 6521dbf3884185ca5692d35d879d251890cc4f08 10774Author: Roland Scheidegger <sroland@tungstengraphics.com> 10775Date: Sat Aug 29 12:11:03 2009 +0200 10776 10777 r100/r200: fix section size mismatch for textured video 10778 10779 this fixes the driver complaining though reportedly it still does 10780 not work (on r100) 10781 10782commit 39dfac41eebbcdacb1ae0c17cea325ea2966c4a5 10783Author: Michel Dänzer <daenzer@vmware.com> 10784Date: Fri Aug 28 12:20:04 2009 +0200 10785 10786 EXA: RENDER repeat fix and more cleanup. 10787 10788 Apparently changing pPict->repeatType causes badness, see 10789 http://bugs.freedesktop.org/show_bug.cgi?id=23560 . 10790 10791 Now we just use a normalized repeatType value derived from pPict->repeat and 10792 pPict->repeatType everywhere. 10793 10794commit 7d36f13afa5da742aa6c989a2652e92c96e5f80e 10795Author: Michel Dänzer <daenzer@vmware.com> 10796Date: Thu Aug 27 09:52:11 2009 +0200 10797 10798 KMS: Remove explicit radeon_bo_wait calls before radeon_bo_map. 10799 10800 libdrm_radeon always takes care of this implicitly now. 10801 10802commit 066c90e8f5b93c80195dc43486ad3efa083a1f3a 10803Author: Michel Dänzer <daenzer@vmware.com> 10804Date: Thu Aug 27 09:52:11 2009 +0200 10805 10806 KMS: Don't use a blit for UploadToScreen if the pixmap BO will be idle anyway. 10807 10808 Direct CPU writes should be at least as fast in that case. 10809 10810commit 174b61bb786a841cebd354e23b4b0caa85b541aa 10811Author: Michel Dänzer <daenzer@vmware.com> 10812Date: Thu Aug 27 09:52:11 2009 +0200 10813 10814 KMS: DownloadFromScreen improvements. 10815 10816 * Drop superfluous RADEONDownloadFromScreenGTT function, EXA does the same 10817 thing when we return FALSE. 10818 * Take unflushed operations into account for determining which GEM domain the 10819 pixmap BO will end up in. 10820 * Only use a blit if it ends up in VRAM. 10821 10822commit 7623e169e1f7d5afbd4108de03f28098bca519db 10823Author: Michel Dänzer <daenzer@vmware.com> 10824Date: Thu Aug 27 09:52:11 2009 +0200 10825 10826 EXA: Allocate pixmap BOs in VRAM. 10827 10828 This is the intention with EXA in xserver Git, and IME it improves performance 10829 significantly even with older versions. 10830 10831commit b90b823790657707b5bf7249095833bef959b0e8 10832Author: Michel Dänzer <daenzer@vmware.com> 10833Date: Thu Aug 27 09:52:11 2009 +0200 10834 10835 DRI2: Set tiling for depth/stencil buffers with all DRI2 interface versions. 10836 10837commit 27d9664f8862d1948dd32021f8e9bcb94f23e4ef 10838Author: Michel Dänzer <daenzer@vmware.com> 10839Date: Thu Aug 27 09:52:10 2009 +0200 10840 10841 EXA: RENDER repeat cleanups. 10842 10843 We can't rely on the server setting pPict->repeatType to something sensible 10844 when pPict->repeat is FALSE, but we can do it ourselves. 10845 10846commit 9b5517fcf7f2397a1a814d396fc9dc24769a6d9f 10847Author: Michel Dänzer <daenzer@vmware.com> 10848Date: Thu Aug 27 09:52:10 2009 +0200 10849 10850 Throttle DRI2 buffer swaps / frontbuffer flushes. 10851 10852commit e87f0f50f31a59ca1f60d4582d4a57ed00854fb7 10853Author: Alex Deucher <alexdeucher@gmail.com> 10854Date: Wed Aug 26 02:13:38 2009 -0400 10855 10856 r6xx/r7xx: set EXA_HANDLES_PIXMAPS 10857 10858commit 3212c26b90c0f6f1a7248b4da3ed985a9c2e9381 10859Author: Alex Deucher <alexdeucher@gmail.com> 10860Date: Wed Aug 26 01:42:10 2009 -0400 10861 10862 r6xx/r7xx: more WIP 10863 10864commit 5a08e68cc254fb255e631b456e331c32456ef0e7 10865Author: Alex Deucher <alexdeucher@gmail.com> 10866Date: Tue Aug 25 19:24:41 2009 -0400 10867 10868 r6xx/r7xx: fix some define problems in Xv code 10869 10870commit 9aa214e125b7927d62b9fe124a851d0373c24d7e 10871Author: Alex Deucher <alexdeucher@gmail.com> 10872Date: Tue Aug 25 18:45:49 2009 -0400 10873 10874 r6xx/r7xx: fix reloc for vtx buffer 10875 10876commit bba51187055932ecd466f5f817428d6c773747b9 10877Author: Alex Deucher <alexdeucher@gmail.com> 10878Date: Tue Aug 25 18:37:15 2009 -0400 10879 10880 R6xx/r7xx: unmap vb bo when done 10881 10882commit 599adfc1f5e6d708be7ad30f4871de3046775727 10883Author: Alex Deucher <alexdeucher@gmail.com> 10884Date: Tue Aug 25 18:13:14 2009 -0400 10885 10886 r6xx/r7xx: fix flipped domains 10887 10888commit 2e83cca8d7efaf1a6836cfb9ea5893fd9d70175f 10889Author: Alex Deucher <alexdeucher@gmail.com> 10890Date: Tue Aug 25 18:05:43 2009 -0400 10891 10892 r6xx/r7xx: more cs exa wip 10893 10894commit 65852de027989c105246fa4e4eed432f29525a22 10895Author: Alex Deucher <alexdeucher@gmail.com> 10896Date: Tue Aug 25 17:29:42 2009 -0400 10897 10898 r6xx/r7xx EXA: WIP 10899 10900commit 69ec7a35e2a0a3d802ec093a6aab2d7ed2cc88be 10901Author: Alex Deucher <alexdeucher@gmail.com> 10902Date: Tue Aug 25 16:14:02 2009 -0400 10903 10904 r6xx/r7xx: first pass at kms accel support 10905 10906 Adapted from various patches from Dave and Jerome. 10907 10908commit 04692f80678ffea15ac33820381ce45865e697b8 10909Author: Dave Airlie <airlied@linux.ie> 10910Date: Wed Aug 26 05:31:56 2009 +1000 10911 10912 radeon: afaics these aren't radeon regs but r128 regs. 10913 10914 so remove them from radeon use. 10915 10916commit 9cf965bbc977f0523437c0ecf1d7363b17de2468 10917Author: Alex Deucher <alexdeucher@gmail.com> 10918Date: Tue Aug 25 12:47:34 2009 -0400 10919 10920 R6xx/r7xx: add begin/end batch macros 10921 10922commit b6368cc572c79bce9a9366242c727c13cab3f006 10923Author: Alex Deucher <alexdeucher@gmail.com> 10924Date: Tue Aug 25 12:14:33 2009 -0400 10925 10926 r6xx/r7xx: move more common state to default state setup 10927 10928commit fc74e1194c980d978667e02c60a29a761a694bde 10929Author: Alex Deucher <alexdeucher@gmail.com> 10930Date: Tue Aug 25 11:39:56 2009 -0400 10931 10932 RV740: disable small DFS transfers 10933 10934 Seems problematic on all rv740 chips, so until we 10935 can find a solution, disable them. 10936 10937commit 1c909ecd88e75c6e16af7aa068e32a7a60b0af37 10938Author: Jerome Glisse <jglisse@redhat.com> 10939Date: Tue Aug 25 11:21:41 2009 +0200 10940 10941 radeon: Fix X recycling with KMS 10942 10943 On X recycling CloseScreen drop master so grab it again 10944 in InitScreen and avoid to reallocate buffer & various 10945 others cs/bo utilities on X recycling. 10946 10947commit 39aaf9addc923852598ba6fb914459671b78e492 10948Author: Dave Airlie <airlied@linux.ie> 10949Date: Mon Aug 24 18:59:22 2009 +1000 10950 10951 r600: use exa fallback code for solid/copy codepaths. 10952 10953 Solid seem to try and handle 1bpp which really no. 10954 10955commit e4b2d57ddca8739ec0b452565713398d036be0d5 10956Author: Dave Airlie <airlied@redhat.com> 10957Date: Mon Aug 24 14:11:24 2009 +1000 10958 10959 r600: fix bad unit in texture setup 10960 10961commit bac224912c750dc1c85ff2d9b8526dad6c23b572 10962Author: Dave Airlie <airlied@redhat.com> 10963Date: Sat Aug 22 21:17:59 2009 +1000 10964 10965 radeon: don't spec any initial placement for pixmaps. 10966 10967 allow the first use to decide placement. 10968 10969commit 77f98717d825162da106c6898cdbcbdf5c984ae6 10970Author: Dave Airlie <airlied@redhat.com> 10971Date: Sat Aug 22 21:16:25 2009 +1000 10972 10973 exa/cs: add DFS from GTT optimisation 10974 10975 This uses the new libdrm busy interface, once I had this in place 10976 I added a error if this happened and it does on my desktop here, 10977 so may as well add the optimisation that used to be in my old KMS tree. 10978 10979 Signed-off-by: Dave Airlie <airlied@redhat.com> 10980 10981commit b1b77a4d6fb7404af9568644e1a8e050fdfa956e 10982Author: Alex Deucher <alexdeucher@gmail.com> 10983Date: Wed Aug 19 13:29:04 2009 -0400 10984 10985 rs600: add support for DisplayPriority HIGH 10986 10987 Still haven't implemented the actual watermark 10988 calculation, but this forces the display requests 10989 to urgent if the user specifies DisplayPriority HIGH. 10990 10991commit 72e0d1b2cb11a67b8e4be4c74913ee44dc051c5b 10992Author: Dave Airlie <airlied@redhat.com> 10993Date: Wed Aug 19 19:01:30 2009 +1000 10994 10995 radeon/kms: add initial colortiling support (disabled by default). 10996 10997 This requires an X server from git with createpixmap2 support fixed up in it. 10998 10999 On 1.6 and previous it won't do any tiling, if you are running git server 11000 please upgrade to latest git. 11001 11002 Option "AllowColorTiling" "true" to enable and do some testing 11003 11004commit 479a6daefe46f985c415b0d000b1b1b820f3924e 11005Author: Alex Deucher <alexdeucher@gmail.com> 11006Date: Tue Aug 18 02:53:29 2009 -0400 11007 11008 AVIVO: add support for DisplayPriority HIGH 11009 11010 If the DisplayPriority option is set to HIGH, 11011 force the diplay mem requests to urgent. 11012 11013commit 8fd6f69c84103b277de70181988e843da23c85b0 11014Author: Dave Airlie <airlied@redhat.com> 11015Date: Mon Aug 17 15:50:21 2009 +1000 11016 11017 radeon: move detected monitor type debugging. 11018 11019 We should print this after DAC detection to be useful for TV 11020 11021commit d25185f33d06a8162469e18688e95ee19a5d94d8 11022Author: Dave Airlie <airlied@redhat.com> 11023Date: Mon Aug 17 15:31:16 2009 +1000 11024 11025 atom: recall crtc source script after tv dpms on 11026 11027 For some reason we lost the DACB source when we dpms tv back on. 11028 11029commit a971f1cfbbda1e3d16fb41f4a07f3b43fd59bc0d 11030Author: Dave Airlie <airlied@redhat.com> 11031Date: Mon Aug 17 15:25:17 2009 +1000 11032 11033 atombios: store dac detect result into save register. 11034 11035 We end up writing back an empty save register over the dac 11036 detection results so the atombios table can no longer 11037 figure out the S-video vs composite so sets the wrong one up. 11038 11039 this gets tv-out on my r580 with a composite connector working. 11040 11041commit efef7c01ad38e078de2fa3f9e528e4ef7d05d00a 11042Author: Dave Airlie <airlied@redhat.com> 11043Date: Mon Aug 17 13:44:20 2009 +1000 11044 11045 atom/tvout: set the correct scaler up for tv out 11046 11047 If we need to use the second scaler actually set the second scaler 11048 up. 11049 11050commit 032a8a06315d12e66240e7ef74b5b165f6d0d1a9 11051Author: Dave Airlie <airlied@redhat.com> 11052Date: Mon Aug 17 12:18:23 2009 +1000 11053 11054 kms: make tv out match the connector name 11055 11056commit 36e51c4d63836863dd7a17cbf6e5a9b7c1bbd31a 11057Author: Dave Airlie <airlied@redhat.com> 11058Date: Mon Aug 17 12:17:31 2009 +1000 11059 11060 r100/r200: dont emit wrong clamp modes. 11061 11062 for rect textures you don't want to use wrap clamping which 11063 is the default. 11064 11065commit 595f7c3a60ca982ca2836a0174e1f36f9d1a4fec 11066Author: Dave Airlie <airlied@itt42.(none)> 11067Date: Mon Aug 17 09:38:09 2009 +1000 11068 11069 radeon/tv: fix typo in crtc register modifications 11070 11071commit 92baeafde8f0c1eed2b6769e96f8093e3d6d7478 11072Author: Alex Deucher <alexdeucher@gmail.com> 11073Date: Tue Aug 11 15:29:17 2009 -0400 11074 11075 RV280: add agp quirk 11076 11077 from Hartmut Niemann 11078 11079commit 04288fdb8b8c62fc2eb32fd467fdf05151ad121a 11080Author: Jerome Glisse <jglisse@redhat.com> 11081Date: Tue Aug 11 19:32:52 2009 +0200 11082 11083 radeon/r600: abstract vertex buffer for r600 & r700 hardware 11084 11085 Abstract vertex buffer, preliminary work before others change 11086 to bring cs support along IB support. 11087 11088commit 72daee2a18bfb809260d1bcbe80cc14afea3b0fa 11089Author: Adam Jackson <ajax@redhat.com> 11090Date: Mon Aug 10 13:59:49 2009 -0400 11091 11092 Fix misleading filenames in DRI setup failure messages 11093 11094commit cd99d9f0d715f1f74de4fe22f2fd30046f2c7568 11095Author: Michel Dänzer <daenzer@vmware.com> 11096Date: Fri Aug 7 11:56:50 2009 +0200 11097 11098 Properly let radeon_cs_flush_indirect() re-emit 2D state if necessary. 11099 11100 Also remove a superfluous assignment. 11101 11102commit 577ff3ce922e457cc32f80d4365cb1da81552e72 11103Author: Michel Dänzer <daenzer@vmware.com> 11104Date: Fri Aug 7 11:56:49 2009 +0200 11105 11106 Add support for EXA_MIXED_PIXMAPS in xserver master. 11107 11108commit e755fa56d03aa338d0c6345ed41e32aa3115ad4b 11109Author: Michel Dänzer <daenzer@vmware.com> 11110Date: Fri Aug 7 11:56:49 2009 +0200 11111 11112 Minor fixes for KMS EXA DownloadFromScreen hook. 11113 11114 Check we can handle the bpp, and remove superfluous flush. 11115 11116commit 4cebafae81c1738a1c330d52a2c5248869f06411 11117Author: Alex Deucher <alexdeucher@gmail.com> 11118Date: Fri Aug 7 11:56:40 2009 +0200 11119 11120 Add KMS EXA UploadToScreen hook. 11121 11122 Fixups by Michel Dänzer: Doesn't seem to be a win in terms of raw numbers, but 11123 should allow more pipelining and avoiding BO memory waste for small glyph 11124 pixmaps. 11125 11126commit 9243791322e36b9231e6a3f04024ad66325385e3 11127Author: Dave Airlie <airlied@redhat.com> 11128Date: Fri Aug 7 14:07:32 2009 +1000 11129 11130 r200: fixup scissors for DDX. 11131 11132 a) turn of R200_RE_CNTL - SCISSOR_ENABLE - this save us emitting R200_RE_TOP_LEFT, note scissor is still enabled. 11133 b) disable aux scissors. 11134 11135commit bd03977e320591ca55b1a2fbb32414c53cb3f72e 11136Author: Alex Deucher <alexdeucher@gmail.com> 11137Date: Tue Aug 4 15:21:44 2009 -0400 11138 11139 radeon: fix compile with !kms 11140 11141commit 4fc1e67e606daf9227e74518c54315b60cd38301 11142Author: Alex Deucher <alexdeucher@gmail.com> 11143Date: Tue Aug 4 15:17:39 2009 -0400 11144 11145 radeon: re-add asus connector quirks 11146 11147 fixes bug 19943 11148 11149commit 22074cf0e58fddba743924532625e6fca49b6bdc 11150Author: Jerome Glisse <jglisse@redhat.com> 11151Date: Tue Aug 4 21:09:25 2009 +0200 11152 11153 radeon/kms: add simple DownloadFromScreen implementation 11154 11155 What we want to do is add userspace object support to radeon 11156 kernel modesetting. Also this DFS is dumb and might endup doing 11157 blit from GTT to GTT. 11158 11159commit fce31b61a88522733863a9b4e9f1c935c439cb4e 11160Author: Alex Deucher <alexdeucher@gmail.com> 11161Date: Tue Aug 4 11:16:05 2009 -0400 11162 11163 RS880: enable accel 11164 11165commit 6b1b4b9214d58175727f7992fc42bc7244c42bf2 11166Author: Alex Deucher <alexdeucher@gmail.com> 11167Date: Mon Aug 3 15:55:04 2009 -0400 11168 11169 radeon: add some new r7xx pci ids 11170 11171commit f564460e94c9d0f1cf3ff4b8535481b2b8b4e9c1 11172Author: Alex Deucher <alexdeucher@gmail.com> 11173Date: Sat Aug 1 16:53:47 2009 -0400 11174 11175 radeon: use XAA in some cases 11176 11177 Use XAA in low memory situations or when the DRI is disabled. 11178 Using shadowfb might also be a viable option, maybe even a better option... 11179 fixes bug 21683 11180 11181commit c71b2f050e8996787eae463eddbfdb5864ffa65a 11182Author: Bryce Harrington <bryce@canonical.com> 11183Date: Sat Aug 1 16:50:43 2009 -0400 11184 11185 radeon: AGPMode quirk needed for SiS 11186 11187 fixes bug 23065 11188 11189commit e3659ed06fc5bb8817f1dbd7c2d6bc94c67b30f7 11190Author: Bryce Harrington <bryce@canonical.com> 11191Date: Sat Aug 1 16:48:48 2009 -0400 11192 11193 radeon: AGPMode quirk needed for IBM Thinkpad T40 with Mobility M7 LW 11194 11195 fixes bug 23064 11196 11197commit 2391531ed6b7c11ddd5ab91b2369821cc5f8b8a7 11198Author: Bryce Harrington <bryce@canonical.com> 11199Date: Sat Aug 1 16:41:41 2009 -0400 11200 11201 radeon: AGPMode quirk needed for HP Omnibook 6200 11202 11203 fixed bug 23063 11204 11205commit 474eda02257152ced52364f38cbad24c20aebbc0 11206Author: Alex Deucher <alexdeucher@gmail.com> 11207Date: Sat Aug 1 16:06:52 2009 -0400 11208 11209 radeon: reload bicubic Xv texture on VT switch 11210 11211 fixed bicubic filtering after VT switch or suspend/resume 11212 11213commit 447a2ce1b88aa2d6d5713e93174c4002617720f7 11214Author: Dave Airlie <airlied@redhat.com> 11215Date: Thu Jul 30 09:54:01 2009 +1000 11216 11217 radeon: fix r600 shadow fb mode with cursor allocation in wrong place 11218 11219 From RHEL QE testing, we could end up with the cursor at 0 since 11220 we think EXA is in use when really it isn't. the info->useEXA = FALSe 11221 might be unnecessary but better to be explicit 11222 11223 Signed-off-by: Dave Airlie <airlied@redhat.com> 11224 11225commit 2b9a5be3bb22fad1d52ec9eea7733c5e956250b7 11226Author: Alex Deucher <alexdeucher@gmail.com> 11227Date: Wed Jul 29 03:34:11 2009 -0400 11228 11229 radeon: fix typo in object header to connector type conversion 11230 11231 Should fix bug 19943 11232 11233commit 63c873cbd4d1d21d9f688028c0900c79fadc42c1 11234Author: Dave Airlie <airlied@redhat.com> 11235Date: Tue Jul 28 15:22:40 2009 +1000 11236 11237 ati: change to using ABI version check 11238 11239commit 21a621c297ac71c65c239ea960c38706e718b91c 11240Author: Dave Airlie <airlied@redhat.com> 11241Date: Tue Jul 28 13:32:28 2009 +1000 11242 11243 ati: update for resources/RAC API removal 11244 11245commit 5ad2519c2a7c0df389b2cd7cf7151c7e4b7252a8 11246Author: Michel Dänzer <daenzer@vmware.com> 11247Date: Mon Jul 27 14:22:02 2009 +0200 11248 11249 KMS: Explicitly specify VRAM and GTT domains when allocating BOs for pixmaps. 11250 11251 Without this, apparently they are initially allocated in system RAM (or 11252 possibly GTT), so any GPU rendering to them first triggers a copy of the whole 11253 uninitialized data to VRAM... 11254 11255 We may want to explore more sophisticated schemes in the future, but for now 11256 this seems to improve KMS 2D performance quite a bit. 11257 11258commit 57f2c83a22f27567506c555af431f89e6031204c 11259Author: Alex Deucher <alexdeucher@gmail.com> 11260Date: Sat Jul 25 14:19:38 2009 -0400 11261 11262 radeon: Set PCI/PCIE bus type properly per asic 11263 11264 - r1xx-rv350 chips have the old pci gart 11265 - rv380+ chips have newer pcie gart 11266 11267 Select the right kind regardless of whether the user selects 11268 PCI or PCIE. 11269 11270commit 2afc46fa74ce7730f766a3370d323c6b59694186 11271Author: Jerome Glisse <jglisse@redhat.com> 11272Date: Thu Jul 23 20:13:15 2009 +0200 11273 11274 radeon: fix KMS shadowfb for r6xx and newer hw. 11275 11276 We need to map front buffer and also to avoid EXA initialization 11277 when falling back to shadowfb on newer r6xx/r7xx when KMS is enabled. 11278 11279commit 328e4c816348b6f867df30b39856dbe78c8e0dcc 11280Author: Alex Deucher <alexdeucher@gmail.com> 11281Date: Thu Jul 23 13:28:36 2009 -0400 11282 11283 RS780: fill in MC access functions 11284 11285 Also, r6xx/r7xx don't have indirect MC space, so 11286 don't try to access it. 11287 11288commit f96d5b255425fbd02be2cad26edb590d474a5640 11289Author: Alex Deucher <alexdeucher@gmail.com> 11290Date: Thu Jul 23 13:25:04 2009 -0400 11291 11292 R6xx/R7xx: clip rendering to destination surface dimensions 11293 11294commit e372f845b0defaf2d2c9ef3cbbf7498e09d9372e 11295Author: Michel Dänzer <daenzer@vmware.com> 11296Date: Sun Jul 19 16:05:29 2009 +0200 11297 11298 Guard reference to CRTC active field. 11299 11300 Fixes build against older xserver. 11301 11302commit e38305aebdc95f80f5b4b3e5ba541ea67dc05f01 11303Author: Michel Dänzer <daenzer@vmware.com> 11304Date: Sat Jul 18 22:23:45 2009 +0200 11305 11306 Also drop DRM master in KMS CloseScreen. 11307 11308 The LeaveVT hook isn't always called when the server dies, e.g. when quitting 11309 from the GDM greeter. This may cause existing servers to fall over if the dying 11310 server process still exists when they try to re-acquire master and set a mode. 11311 11312 Also use drmSet/DropMaster() rather than ioctl() directly. 11313 11314commit a43c660a00147bfae5ca601f4720b2680b75211f 11315Author: Michel Dänzer <daenzer@vmware.com> 11316Date: Sat Jul 18 22:21:59 2009 +0200 11317 11318 Set CRTC active flag to TRUE after a successful KMS mode set. 11319 11320 Otherwise some things like changing the colour map won't work properly. 11321 11322commit f32069f42b3e4643f15148d0b96164def00dcc74 11323Author: Michel Dänzer <daenzer@vmware.com> 11324Date: Sat Jul 18 22:20:08 2009 +0200 11325 11326 Set the STALL bit in the CRTC_GUI_TRIG_VLINE registers. 11327 11328 As recommended by the register reference when using the WAIT_CRTC_VLINE bit in 11329 the WAIT_UNTIL register, as we are. 11330 11331commit 27bd9fc9b2c0eeff488c1f26f1355a7fa3655520 11332Author: Michel Dänzer <daenzer@vmware.com> 11333Date: Sat Jul 18 22:14:27 2009 +0200 11334 11335 Enable vsync for DRI2 region copies. 11336 11337commit 4b1b28865ffb17f58a70d955212fd51cb1fb55ab 11338Author: Alex Deucher <alexdeucher@gmail.com> 11339Date: Thu Jul 16 17:09:33 2009 -0400 11340 11341 R6xx/r7xx: reduce the amount of default state setup 11342 11343 Most of the default state setup was extraneous or duplicated 11344 in the accel ops. There's no need to reset all the consts 11345 everytime you reset the default state for example. This leads 11346 to a 10x reduction in default state. 11347 11348commit 7e6557b81026d8a4d6c837839e68c28b151f8a97 11349Author: Peter Hutterer <peter.hutterer@who-t.net> 11350Date: Thu Jul 16 11:33:57 2009 +1000 11351 11352 Update to xextproto 7.1 support. 11353 11354 DPMS header was split into dpms.h (client) and dpmsconst.h (server). Drivers 11355 need to include dpmsconst.h if xextproto 7.1 is available. 11356 11357 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 11358 11359commit 95d431e8b260dd43cf6d93b90eecab1dc0ec7c26 11360Author: Dave Airlie <airlied@redhat.com> 11361Date: Wed Jul 15 12:07:01 2009 +1000 11362 11363 radeon: emit colorpitch relocs. 11364 11365 This causes relocs to be emitted for the colorpitch. 11366 11367 Shouldn't have no effect on current kernels, but will keep DDX 11368 going on kms kernels when tiling lands 11369 11370 Also contains a missing reloc space for r200 11371 11372commit 8312763977b22e108b7d003cbf3256e0165e4439 11373Author: Alex Deucher <alexdeucher@gmail.com> 11374Date: Tue Jul 14 18:13:27 2009 -0400 11375 11376 Add an r420 AGP quirk 11377 11378 fixes bug 22726 11379 11380commit 5473eeeb45468fd200fec49a8506281628a5e4b5 11381Author: Alex Deucher <alexdeucher@gmail.com> 11382Date: Tue Jul 14 14:05:13 2009 -0400 11383 11384 R3/4/5xx: only upload the bicubic texture once 11385 11386 Upload the bicubic texture once during textured video init 11387 rather than once per frame. Suggested by Michel Daenzer on 11388 IRC. 11389 11390commit 0485f27bc3d75cb6ab320e8164dbe6ea2713c78e 11391Author: Alex Deucher <alexdeucher@gmail.com> 11392Date: Tue Jul 14 13:11:12 2009 -0400 11393 11394 R3/4/5xx: fix bicubic Xv filtering with KMS changes 11395 11396 fixes bug 22730 11397 11398commit ca4a9efef987108266ae59afbb5142d4d2000c8e 11399Author: Alex Deucher <alexdeucher@gmail.com> 11400Date: Mon Jul 13 11:28:41 2009 -0400 11401 11402 Add configure option to disable kms support 11403 11404 Useful to building with mixed kms/non-kms setups 11405 11406commit 035e8d1d5593c12828bb079de4e663cf1b1f1674 11407Author: Edward O'Callaghan <eocallaghan@auroraux.org> 11408Date: Sun Jul 12 13:45:29 2009 +0200 11409 11410 Fix a warning 11411 11412 Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com> 11413 11414commit 76af48c43f829e7aebacc9f2a623823fa26ee22b 11415Author: Gaetan Nadon <memsize@videotron.ca> 11416Date: Wed Jul 8 09:58:56 2009 -0400 11417 11418 xf86-video-ati: use XORG_CHANGELOG macro to create ChangeLog. #22611 11419 11420 Build break: Makefile.am: command not found: git-log 11421 Adding the macro in configure.ac and use it in Makefile.am 11422 Refer to: https://bugs.freedesktop.org/show_bug.cgi?id=22611 11423 Tested: running autogen.sh, make and 'make dist' 11424 11425 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 11426 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 11427 11428commit 43db263d301082e84e9bc304816bcbb206fe280e 11429Author: Alex Deucher <alexdeucher@gmail.com> 11430Date: Thu Jul 9 12:48:06 2009 -0400 11431 11432 RV280: another AGP quirk 11433 11434 fdo bug 12544 11435 11436commit 69b5e5496f10a9f566d2e563862c96cb41952eb6 11437Author: Alex Deucher <alexdeucher@gmail.com> 11438Date: Thu Jul 9 12:42:45 2009 -0400 11439 11440 RV280: Add an AGP quirk 11441 11442 lp bug 370205 via fdo bug 12544 11443 11444commit a6d6c900e60a3685ee9e93c34eb7f6d237c45fd8 11445Author: Alex Deucher <alexdeucher@gmail.com> 11446Date: Thu Jul 9 11:30:57 2009 -0400 11447 11448 RS740: fixup display latency setting as per rs690 11449 11450 Both chips program the same in this regard 11451 11452commit ea407570d39ace3162d372eda56bf791bfd80c24 11453Author: Dave Airlie <airlied@redhat.com> 11454Date: Thu Jul 9 10:46:23 2009 +1000 11455 11456 radeon: add shave support. 11457 11458 clean up output a lot 11459 11460commit 8c03c1fdb5ea35570064946557050c87ca30582a 11461Author: Alex Deucher <alexdeucher@gmail.com> 11462Date: Wed Jul 8 18:34:57 2009 -0400 11463 11464 R6xx/R7xx: fix hangs on x2 cards with PM options 11465 11466 Changing the PCIE lanes on x2 cards results in a hang, 11467 so for now, disable it. 11468 11469 Fixes fdo bug 22669 11470 11471commit 9645838c57f6b40837fdce23ce7f9faefb3d9966 11472Author: Michel Dänzer <daenzer@vmware.com> 11473Date: Wed Jul 8 19:18:02 2009 +0200 11474 11475 Further non-KMS fixes / cleanups. 11476 11477 Only compile tested. 11478 11479commit 2d86ec4ddb1fb83be95ed9cdcbabf4d568a6d7b6 11480Author: Alex Deucher <alexdeucher@gmail.com> 11481Date: Wed Jul 8 12:33:34 2009 -0400 11482 11483 Fix the build for the !XF86DRM_MODE case 11484 11485 compile tested only. 11486 11487commit a68afc2a80b8581cc3d67e9b5a95a10b0d4ead0e 11488Author: Michel Dänzer <daenzer@vmware.com> 11489Date: Wed Jul 8 09:15:28 2009 +0200 11490 11491 Fix RADEON_TRACE_FALL build. 11492 11493commit 9ca0aeaad01b7c682396b30af48668f287a59d6e 11494Author: Michel Dänzer <daenzer@vmware.com> 11495Date: Wed Jul 8 09:13:36 2009 +0200 11496 11497 Pass on CS relocation offset values. 11498 11499 Not actually used ATM, but just in case... 11500 11501commit f718dd794477fbdd5eeb4183f43ff25517c63034 11502Author: Michel Dänzer <daenzer@vmware.com> 11503Date: Wed Jul 8 09:11:18 2009 +0200 11504 11505 Don't set EXA UploadTo/DownloadFromScreen hooks when they'd always return FALSE. 11506 11507commit 42e76d0b24b76fbcc6b08a4ea5003dbec368645c 11508Author: Michel Dänzer <daenzer@vmware.com> 11509Date: Wed Jul 8 08:51:10 2009 +0200 11510 11511 Don't always flush CS when switching from 2D to 3D. 11512 11513 This works fine here and doubles x11perf -aa10text scores, but may cause 11514 problems on other setups according to Dave. Let's see... 11515 11516commit 94fe049c35187ddc3ec0472e9309c3a22879f415 11517Author: Michel Dänzer <daenzer@vmware.com> 11518Date: Wed Jul 8 08:36:22 2009 +0200 11519 11520 KMS CS fixes, take two. 11521 11522 Avoid flushing CS in the middle of (setting up state for) an operation, 11523 properly finish/restart the operation. 11524 11525commit 0519f15af2fe5e7e5a6c94e7203a96fe363c643e 11526Author: Michel Dänzer <daenzer@vmware.com> 11527Date: Tue Jul 7 10:04:09 2009 +0200 11528 11529 Fix DRI2 with current xserver Git. 11530 11531commit 7ad11fec2579807371360cb680e593f723ebaa1a 11532Author: Dave Airlie <airlied@redhat.com> 11533Date: Mon Jul 6 18:43:08 2009 +1000 11534 11535 radeon: add make dist headers 11536 11537commit e68fbf743ec88e0fa3a9f1ffe83daf251e250b21 11538Author: Dave Airlie <airlied@redhat.com> 11539Date: Mon Jul 6 18:32:32 2009 +1000 11540 11541 git: fix git-log to git log 11542 11543commit 5eeb1fd19c6797cc76ce308570dffe34cb82b24d 11544Author: Dave Airlie <airlied@linux.ie> 11545Date: Mon Jul 6 17:45:50 2009 +1000 11546 11547 radeon: oops fix compile without libdrm 11548 11549commit 98824f37fc74dc1fec2677515d0dc3cfc263f00a 11550Author: Dave Airlie <airlied@redhat.com> 11551Date: Mon Jul 6 15:29:53 2009 +1000 11552 11553 radeon: move state changes into flush function 11554 11555commit 503b51343267248b4c6994b88238dd5384528f03 11556Author: Dave Airlie <airlied@redhat.com> 11557Date: Mon Jul 6 15:20:13 2009 +1000 11558 11559 radeon: taken from kms-support + CS fixes commit. 11560 11561 re-emit 2D state and reset 3D state emit on flush 11562 11563commit d78d50591851d14543f0935d051a59ef29751bbc 11564Author: Dave Airlie <airlied@redhat.com> 11565Date: Mon Jul 6 14:45:07 2009 +1000 11566 11567 radeon: port to new space checking in libdrm interface 11568 11569 This uses a new libdrm interface which shares code with mesa. 11570 11571 It also fixes the bo to flush when full instead of never flushing. 11572 11573 It survives gtkperf -a here which the driver didn't before now 11574 11575commit e1200cb89218930d01330ba0114e013438655cce 11576Author: Dave Airlie <airlied@redhat.com> 11577Date: Mon Jul 6 15:10:29 2009 +1000 11578 11579 Revert "KMS CS fixes." 11580 11581 This reverts commit 11cf8a28eb46f48d101888552266921bee8b63fc. 11582 11583 I'll bring this back in a few minutes, I've got a big change introducing 11584 a new cs space accounting scheme which this messes up and I think 11585 probably conflicts with what I've done. 11586 11587 I promise i'll go over this and pick the bits that are still useful :) 11588 11589commit 11cf8a28eb46f48d101888552266921bee8b63fc 11590Author: Michel Dänzer <daenzer@vmware.com> 11591Date: Sun Jul 5 15:14:59 2009 +0200 11592 11593 KMS CS fixes. 11594 11595 * Flush CS before running out of space. 11596 * Don't flush CS every time when switching between 2D and 3D engine, but make 11597 sure to re-initialize 2D/3D state after flushing. 11598 11599 Doesn't seem quite perfect yet (some rendering seems to get dropped on the floor 11600 sometimes), but this should fix a lot of CS failures people have been seeing. 11601 11602commit 8d2f712eaf1e569fd92bbe2db5aceb43b7b367d1 11603Author: Dave Airlie <airlied@linux.ie> 11604Date: Fri Jul 3 10:40:59 2009 +1000 11605 11606 radeon: don't set up overlay under KMS until we figure out how. 11607 11608commit 2a622168606d98c404708b879e6417c8f7d7305c 11609Author: Michel Dänzer <daenzer@vmware.com> 11610Date: Thu Jul 2 15:03:53 2009 +0200 11611 11612 Test BO pointer before passing it to radeon_bo_unref(). 11613 11614commit 36a4ea5a9c7062f4c8aa7480bb7488ce2062390f 11615Author: Michel Dänzer <daenzer@vmware.com> 11616Date: Thu Jul 2 14:52:44 2009 +0200 11617 11618 Warning fixes. 11619 11620commit 65465d56c1bf943113518a5b2f57b3170213bab9 11621Author: Dave Airlie <airlied@linux.ie> 11622Date: Thu Jul 2 18:38:38 2009 +1000 11623 11624 radeon: fix up configure check better 11625 11626commit ad17108b72c9adf773a122b4c054c7b92e6a8b9d 11627Author: Dave Airlie <airlied@linux.ie> 11628Date: Thu Jul 2 17:25:57 2009 +1000 11629 11630 radeon: fix it properly. 11631 11632 I blame the flu 11633 11634commit e1b9cb9ea665eaa84e64a386299efa0819d55df4 11635Author: Dave Airlie <airlied@redhat.com> 11636Date: Thu Jul 2 17:18:47 2009 +1000 11637 11638 radeon: fix compile without kms 11639 11640commit de427fac6912a625f7e4d8b982a1f7a1ba80dfbe 11641Author: Dave Airlie <airlied@redhat.com> 11642Date: Thu Jul 2 15:41:43 2009 +1000 11643 11644 radeon/dri2: fix dri2 - needs newest libdrm_radeon 11645 11646commit 2b671adba54ebff1756dd990123f4bade6e35d61 11647Author: Dave Airlie <airlied@redhat.com> 11648Date: Thu Jul 2 15:25:19 2009 +1000 11649 11650 radeon/dri2: call proper gem bo function 11651 11652commit 851b2b8cab883e4422745cca1827aa8ac9f193a3 11653Author: Dave Airlie <airlied@redhat.com> 11654Date: Thu Jul 2 15:17:13 2009 +1000 11655 11656 radeon/kms: add textured video BO size checks 11657 11658commit 9d092060236e8587697fdd57a7a01f3c0dddd680 11659Author: Dave Airlie <airlied@redhat.com> 11660Date: Thu Jul 2 15:17:03 2009 +1000 11661 11662 kms: drop DGA 11663 11664commit 207174ea964c561076bdd5297a2fb6f136b4b645 11665Author: Dave Airlie <airlied@redhat.com> 11666Date: Thu Jul 2 15:16:47 2009 +1000 11667 11668 radeon: fix CS imbalance 11669 11670commit 4d24a9d2c0be0c025266cc8cdf7e8d7bb065cdb1 11671Author: Dave Airlie <airlied@redhat.com> 11672Date: Thu Jul 2 14:55:45 2009 +1000 11673 11674 radeon: enable KMS build 11675 11676commit 9bcd3ae7966eb7a17b922bcd33e0d6f016245614 11677Author: Dave Airlie <airlied@redhat.com> 11678Date: Thu Jul 2 14:55:18 2009 +1000 11679 11680 radeon/kms: fixup 3d state emission for KMS 11681 11682 This fixes font rendering under KMS (should have looked at dmesg) 11683 11684commit ac1607bb23f0c7584648407a3ac3fb27142e186e 11685Author: Dave Airlie <airlied@redhat.com> 11686Date: Thu Jul 2 14:55:05 2009 +1000 11687 11688 kms: no need to reference fifo slots 11689 11690commit 1782ce28953184776c90eb1255208a3e0ad245f0 11691Author: Dave Airlie <airlied@redhat.com> 11692Date: Tue Jun 30 16:55:26 2009 +1000 11693 11694 radeon: add KMS support (still disabled) 11695 11696 This adds DRI2 + KMS + driver pixmaps support to the driver. 11697 11698 I've decided to just do a completely separate KMS driver file 11699 instead of hacking the crap out of radeon_driver.c. So now 11700 I do the KMS check in radeon_probe.c time and set the DDX 11701 pointed up to a completely different set at this stage. 11702 11703 This avoids a lot of if (kms) type crap in the code at 11704 the expense of making sure we make changes to both files 11705 if necessary. 11706 11707 This code is still disabled in configure.ac as I broke EXA composite 11708 rendering somehow in KMS mode 11709 11710commit ac4bd24a66c1bdda0293f770a3f891e2b88cc8ee 11711Author: Alex Deucher <alexdeucher@gmail.com> 11712Date: Tue Jun 30 16:00:06 2009 -0400 11713 11714 EXA: fix the build on servers without EXA_HANDLES_PIXMAPS 11715 11716commit 49a0ca19ee4109cdd39a43fbf4bedf3f0557d6ec 11717Author: Michel Dänzer <daenzer@vmware.com> 11718Date: Tue Jun 30 17:22:50 2009 +0200 11719 11720 Warning fixes. 11721 11722commit e932836691aeaec37794fdaed2dabb22710fd171 11723Author: Dave Airlie <airlied@redhat.com> 11724Date: Tue Jun 30 16:24:37 2009 +1000 11725 11726 radeon: initial preparation for kms patch. 11727 11728 This patch contains most of the changes to the EXA and texture video 11729 accel code. 11730 11731 It adds a few bits of pixmap support but doesn't actually do anything 11732 useful KMS yet. 11733 11734 Testing this should not have any regressions over what we have already, 11735 biggest worries are r6xx, I've fixed a textured video one, but no idea 11736 what other might lurk 11737 11738 It won't build against libdrm radeon yet either 11739 11740commit bb04b450ed00ca4b1aa44c33085567d47b33b547 11741Author: Fredrik Höglund <fredrik@kde.org> 11742Date: Mon Jun 29 22:38:26 2009 +0200 11743 11744 Check if the composite op is supported in R200CheckComposite. 11745 11746commit f39cafc5c4a93a16ac93756473ebb1f33c7881db 11747Author: Alex Deucher <alexdeucher@gmail.com> 11748Date: Sun Jun 28 20:42:49 2009 -0400 11749 11750 radeon: call hdmi-b connectors dvi in randr 11751 11752 Technically HDMI-B is a glorfied DL DVI, but this is confusing 11753 to users, so call it DVI when reporting randr outputs. This 11754 allows us to remove two quirks where the bios was actually 11755 correct. 11756 11757commit 407e676b04117c7ca3dcc0cc072f3552b03c64e5 11758Author: Roland Scheidegger <sroland@tungstengraphics.com> 11759Date: Sun Jun 28 16:54:32 2009 -0400 11760 11761 R6xx/R7xx: fix pixel centers 11762 11763 Make sure we are using GL mode (centers at 0.5) rather 11764 than D3D mode (centers at 0.0). This also fixes fdo 11765 bug 21963. 11766 11767commit 68001981f22173ff949720055dba89291f284474 11768Author: Alex Deucher <alexdeucher@gmail.com> 11769Date: Sun Jun 28 16:20:10 2009 -0400 11770 11771 Add some missing M96 pci ids 11772 11773 fixes fdo bug 22404 11774 11775commit 06d5d8d7fd92b58b2467c782e57a80a3520a2c90 11776Author: Adam Jackson <ajax@redhat.com> 11777Date: Thu Jun 25 13:51:36 2009 -0400 11778 11779 Add a (harmlessly) missing break statement. 11780 11781commit 7d1d968fdc8de1e0ab0be916a1f3a4a398af47aa 11782Author: Adam Jackson <ajax@redhat.com> 11783Date: Thu Jun 25 13:49:18 2009 -0400 11784 11785 Cosmetic cleanup to digital output detection 11786 11787 No functional change, but this logic would need extending anyway for 11788 DisplayID, so unify it now. 11789 11790commit 96581746eea2f4e29a04f9266ec9649f25480424 11791Author: Adam Jackson <ajax@redhat.com> 11792Date: Wed Jun 24 17:05:47 2009 -0400 11793 11794 atom: Fix phantom VGA connector when HDMI-B present. 11795 11796 HDMI-B is really a glorified DL-DVI connector, the analog pins work fine. 11797 11798commit 496adc4dc7cf53fc6acfecb1158f1491d5f36168 11799Author: Michel Dänzer <daenzer@vmware.com> 11800Date: Fri Jun 19 14:37:01 2009 +0200 11801 11802 Revert "Build fixes for -Werror that aren't ready for prime time yet." 11803 11804 This reverts commit ab20ea1530ec8a3a9fd4089775c4e157e50fe1bf. 11805 11806 Whoops, pushed wrong branch... 11807 11808commit ab20ea1530ec8a3a9fd4089775c4e157e50fe1bf 11809Author: Michel Dänzer <daenzer@vmware.com> 11810Date: Sun Jun 14 16:50:07 2009 +0200 11811 11812 Build fixes for -Werror that aren't ready for prime time yet. 11813 11814commit 77e3537d312175a25f0e21cc07c3a96f78c3b35a 11815Author: Markus Gapp <markus.gapp@gmx.net> 11816Date: Sat Jun 13 11:51:29 2009 -0400 11817 11818 Add quirk for asus hd3450 board 11819 11820 Fixes bug 22266 11821 11822commit e8e0bb1ba29761c94d473e42ca6ada161cfff557 11823Author: Alex Deucher <alexdeucher@gmail.com> 11824Date: Fri Jun 12 11:07:43 2009 -0400 11825 11826 Fix count in 3d state setup after 43374c7420e378918bec062f4cbd581f16adb6f0 11827 11828commit 808c90a24c48da7fa97e15e2f12be5bb8fd8cc96 11829Author: Alex Deucher <alexdeucher@gmail.com> 11830Date: Thu Jun 11 16:09:40 2009 -0400 11831 11832 Only enable frac fb divs on rs600/rs690/rs740 for now 11833 11834 fixes bug 21413 11835 11836commit d095bb3157faaee10b0130ea9b1032e0b533aaee 11837Author: Alex Deucher <alexdeucher@gmail.com> 11838Date: Thu Jun 11 12:34:15 2009 -0400 11839 11840 Re-enable frac feedback dividers on AVIVO hardware 11841 11842 Preferring a freq <= target frequency seems to fix the 11843 previous issues with frac fb divs. Appears to work fine 11844 on all hardware I've tested on. See fdo bug 22229 for 11845 details. Should fix bugs: 22229, 21553, 21413, and 11846 possibly other ones related to unstable images at certain 11847 resolutions. 11848 11849commit 55fbdbae83d1563b472f49d0436c9298e390be66 11850Author: Alex Deucher <alexdeucher@gmail.com> 11851Date: Thu Jun 11 12:31:53 2009 -0400 11852 11853 Add PLL flag to prefer frequencies <= the target freq 11854 11855 This appears to be needed when using fractional feedback 11856 dividers. Based on a patch from Tom Hirst. See fdo 11857 bug 22229 for more details. 11858 11859commit 43374c7420e378918bec062f4cbd581f16adb6f0 11860Author: Alex Deucher <alexdeucher@gmail.com> 11861Date: Tue Jun 9 14:39:54 2009 -0400 11862 11863 r1xx/r2xx: clip 3D rendering to destination buffer 11864 11865commit 74b0a38c75c1d6a6344015d32d84b43d9e80df1d 11866Author: Alex Deucher <alexdeucher@gmail.com> 11867Date: Tue Jun 9 14:26:21 2009 -0400 11868 11869 Consolidate indentical RADEONDoneSolid/RADEONDoneCopy functions 11870 11871commit f6dff8836729cc9d29264080c9cfa0e1df6182cf 11872Author: Alex Deucher <alexdeucher@gmail.com> 11873Date: Tue Jun 9 13:31:15 2009 -0400 11874 11875 r3xx/r4xx/r5xx: clip 3D rendering to destination buffer 11876 11877commit 48dbf09e1309c815ac8b733145a0310083a29933 11878Author: Alex Deucher <alexdeucher@gmail.com> 11879Date: Tue Jun 9 13:12:40 2009 -0400 11880 11881 r3xx/r4xx/r5xx: refactor exa prepare composite 11882 11883 - consolidate code duplicated between r3xx/r4xx and r5xx 11884 - make pixel shader setup easier to read and less 11885 redundant 11886 11887commit ec1a3ccdf61c2226617b571cc4a209a542e9d7e3 11888Author: Alex Deucher <alexdeucher@gmail.com> 11889Date: Mon Jun 8 15:31:57 2009 -0400 11890 11891 R6xx/r7xx: don't clamp tex coords in composite VS 11892 11893 Fixes broken repeat modes. see fdo bugs 21818, 22139 11894 11895commit 32c218c5633fb7ffeb19e717bf645c1b25dd4ef5 11896Author: Alan Coopersmith <alan.coopersmith@sun.com> 11897Date: Thu Jun 4 07:23:03 2009 -0700 11898 11899 Add link to patch submission instructions to README 11900 11901 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 11902 11903commit 39f736d6cb65fc7c3fd769478023fc87c1af5979 11904Author: Alex Deucher <alexdeucher@gmail.com> 11905Date: Thu Jun 4 09:57:22 2009 -0400 11906 11907 atom: connector quirk 11908 11909 Some r6xx cards have a VGA and an HDMI port with 11910 a shared ddc line listed as an HDMI port with both 11911 analog and digital encoders. 11912 11913 Fixes bug 19943 11914 11915commit 7599dc40855c33a5fbd8e9bbc2b4cd62752fb7df 11916Author: Alex Deucher <alexdeucher@gmail.com> 11917Date: Fri May 29 16:29:50 2009 -0400 11918 11919 Add RV740 (HD4770) support 11920 11921 Verified by ernstp on IRC 11922 11923commit ee0424078a3ea26f58c2998c9e58e065fb63f1a0 11924Author: Alex Deucher <alexdeucher@gmail.com> 11925Date: Wed May 27 18:37:58 2009 -0400 11926 11927 radeon: remove dead code 11928 11929commit b34df233115c0d82d7bcf82e041afbc55981ce82 11930Author: Alex Deucher <alexdeucher@gmail.com> 11931Date: Tue May 26 17:47:01 2009 -0400 11932 11933 R3xx/R4xx: increase EXA render target limits 11934 11935 now that we are using 1/12 subpixel mode 11936 11937commit 1457bc089d1395d18cf582a4d52444005f013753 11938Author: Matthieu Herrb <matthieu.herrb@laas.fr> 11939Date: Tue May 26 17:21:07 2009 -0400 11940 11941 radeon: fix segfault if Option "NoAccel" is set 11942 11943commit 4079f97958cf1d5fd76a83aa61b93d6289bb9b1e 11944Author: Michel Dänzer <daenzer@vmware.com> 11945Date: Sat May 23 13:51:00 2009 +0200 11946 11947 Bump version to mark Git snapshot. 11948 11949 The fourth version component isn't visible in the log file. 11950 11951commit 88a9e98341d96e5e7f48b69aed597d1bada6313a 11952Author: Alex Deucher <alexdeucher@gmail.com> 11953Date: Tue May 19 17:11:34 2009 -0400 11954 11955 Pre-atom: Improve engine clock setup code 11956 11957 Also remove SetMemoryClk() for pre-atom cards for now 11958 as it requires quite a bit more asic specific work. 11959 To set the mclk we'll need to use the mem reset/dll tables 11960 in the bios. 11961 11962commit 1429133ffdbfe046249fcaaa4cbee432a1012e55 11963Author: Michel Dänzer <daenzer@vmware.com> 11964Date: Mon May 18 17:59:00 2009 +0200 11965 11966 EXA: Set new EXA_SUPPORTS_OFFSCREEN_OVERLAPS flag when we have a 2D engine. 11967 11968commit e9cca1e7dc521d3756bb2d97f1d703b9ad65cb47 11969Author: Alex Deucher <alexdeucher@gmail.com> 11970Date: Mon May 18 10:48:25 2009 -0400 11971 11972 r4xx: Add some missing pci ids 11973 11974 Noticed by Romain in bug 21798 11975 11976commit 76fffd51027820b3a6ea10de2be67d5fd4f3e843 11977Author: Alex Deucher <alexdeucher@gmail.com> 11978Date: Fri May 15 16:35:13 2009 -0400 11979 11980 R100: switch to normalized texture coordinates 11981 11982 for consistency with the other asics 11983 11984commit 1a1eef31e36f98b7a30e27e567c1b66e0e4cf7dc 11985Author: Alex Deucher <alexdeucher@gmail.com> 11986Date: Thu May 14 15:48:54 2009 -0400 11987 11988 radeon: try harder when checking if a card is posted 11989 11990 fixes debian bug 524280, possibly others. 11991 11992commit 22e39392297fa11003df90c175db3c449d8f9853 11993Author: Michel Dänzer <daenzer@vmware.com> 11994Date: Thu May 14 10:24:07 2009 +0200 11995 11996 Pre-R600 EXA: Fix coordinate limits off-by-one error. 11997 11998 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=21598 . 11999 12000commit fa09b058c7a17689989e600ffd465856a058579d 12001Author: Alex Deucher <alexdeucher@gmail.com> 12002Date: Wed May 13 16:10:40 2009 -0400 12003 12004 R6xx/R7xx Xv: normalize texture coordinates in the vertex shader 12005 12006commit 026b6f820d6caea17d2a082193e850713d5770a8 12007Author: Alex Deucher <alexdeucher@gmail.com> 12008Date: Wed May 13 15:48:32 2009 -0400 12009 12010 R6xx/R7xx: do EXA transforms in the vertex shader 12011 12012commit cd89241396d1931b04cfbdd8d553be16dbf9c360 12013Author: Alex Deucher <alexdeucher@gmail.com> 12014Date: Tue May 12 17:30:02 2009 -0400 12015 12016 R3xx-R5xx: do EXA transforms in the vertex shader 12017 12018commit 15ee78d37f8b64c3e6d234e7ab37a31e82327b6b 12019Author: Alex Deucher <alexdeucher@gmail.com> 12020Date: Mon May 11 02:22:13 2009 -0400 12021 12022 RV770: add missing pci id 12023 12024 Fixes bug 21209 12025 12026commit 5ff51ea83f9fdd836e830c77e3561eaef4b83183 12027Author: Alex Deucher <alexdeucher@gmail.com> 12028Date: Fri May 8 13:19:26 2009 -0400 12029 12030 R7xx: add support for reading disabled bios images 12031 12032commit 76fdae9018330aa5af0d57339baff3ca4834e1b5 12033Author: Alex Deucher <alexdeucher@gmail.com> 12034Date: Thu May 7 02:59:32 2009 -0400 12035 12036 R3xx/R4xx tex vid: no need to check if IS_R300_3D 12037 12038 function is r3xx/r4xx specific already 12039 12040commit 23a8ada06a5b1ccc975e89f129c00382ecbf29df 12041Author: Alex Deucher <alexdeucher@gmail.com> 12042Date: Thu May 7 02:34:22 2009 -0400 12043 12044 R3xx/R4xx tex vid: increase the guardband limit for rendering with a tri 12045 12046 limit goes from 2880 to 4021 now that we've switched to 1/12 subpixel mode. 12047 12048 This allows us to render with a clipped tri at higher resolutions preventing 12049 diagonal tearing. 12050 12051commit 0c3541f7013c53fb8096ada63cc7b3577524aa4c 12052Author: Alex Deucher <alexdeucher@gmail.com> 12053Date: Wed May 6 12:24:35 2009 -0400 12054 12055 AVIVO: set default cursor mode in cursor init 12056 12057 Should fix bug 21557 12058 12059commit da58e351b6398fa18b1d9c1a57a6e49b285f610f 12060Author: Alex Deucher <alexdeucher@gmail.com> 12061Date: Mon May 4 16:03:53 2009 -0400 12062 12063 AVIVO: fix cursoe corruption bands for real 12064 12065 Don't have to leave both cursors enabled, just have to use 12066 the same mode for both cursors whether or not they are enabled. 12067 12068commit 2c8e130f73c680d4a7381b2ef37982b82c6ee478 12069Author: Alex Deucher <alexdeucher@gmail.com> 12070Date: Mon May 4 13:40:17 2009 -0400 12071 12072 radeon: switch to EXA by default 12073 12074commit 8155a62df6817358a12cc2d613c1470ca55996ef 12075Author: Alex Deucher <alexdeucher@gmail.com> 12076Date: Mon May 4 12:57:36 2009 -0400 12077 12078 AVIVO: move cusor offscreen when disabling 12079 12080 Adapted from Yang's patch. Setting size to zero is 12081 actually a 1x1 cursor. 12082 12083commit a1c64ea5224009779ccad66b0f84d861eae966ac 12084Author: Alex Deucher <alexdeucher@gmail.com> 12085Date: Sun May 3 19:42:44 2009 -0400 12086 12087 r3xx-r5xx: switch to 1/12 subpixel precision 12088 12089 - based on similar patch in Jerome's cs ddx tree 12090 - also fix clipping offsets 12091 - should eventually allow for 4k render targets 12092 - mesa driver uses 1/12 mode, this avoids changing the 12093 subpixel mode when switching between ddx and mesa 12094 12095commit 3880bc3c95764acccd62cfa438932458df95daf6 12096Author: Alex Deucher <alexdeucher@gmail.com> 12097Date: Sun May 3 19:01:28 2009 -0400 12098 12099 Remove some old debugging info 12100 12101commit 3362381aa3a97c1da564f8a479ed4e7b0d6368b8 12102Author: Alex Deucher <alexdeucher@gmail.com> 12103Date: Sun May 3 18:52:00 2009 -0400 12104 12105 r2xx-r4xx: fix typo in last i2c fix and clarify hw i2c pin sel 12106 12107commit e01b8dc827ce1913927e1a4fad05a99a2dd9e805 12108Author: Alex Deucher <alexdeucher@gmail.com> 12109Date: Sun May 3 13:09:39 2009 -0400 12110 12111 RS690: gpio/connector fixes 12112 12113 Fixes bug 21521 12114 12115commit 1f70c9f05df9017d87b37f887e1eccd6d0568a02 12116Author: Alex Deucher <alexdeucher@gmail.com> 12117Date: Thu Apr 30 19:18:13 2009 -0400 12118 12119 R3xx-R5xx: don't set TX_OFFSET_RS in RS_INST_COUNT 12120 12121 Isn't necessary and seems to cause problems for RS690 users 12122 12123commit de17d36c1f710305870e3c636cae3f742f1cb6d6 12124Author: Alex Deucher <alexdeucher@gmail.com> 12125Date: Thu Apr 30 19:13:09 2009 -0400 12126 12127 AVIVO: first pass at display watermark setup 12128 12129 This helps with flickering and blanking when 12130 there is contention for MC bandwidth. 12131 12132commit 7d9f643ae3d07e51e644a5979ca90bc2c102bc89 12133Author: Alex Deucher <alexdeucher@gmail.com> 12134Date: Wed Apr 29 12:38:15 2009 -0400 12135 12136 radeon: fix several dpms issues 12137 12138 - atom dpms was unblanking for standby/suspend 12139 - return if r600+ in radeon_crtc_modeset_ioctl() 12140 - remove seprate standby/suspend handling in legacy 12141 crtc dpms. we turn the outputs off, so turn the 12142 crtcs off too. 12143 - disable the crtcs in legacy crtc dpms 12144 - move radeon_crtc_modeset_ioctl() calls to radeon_crtc_dpms() 12145 so they will get called for both atom and legacy paths 12146 12147 Should fix bug 21321 12148 12149commit 35183ef528411eb0122cf48550d93f921d291d12 12150Author: Alex Deucher <alexdeucher@gmail.com> 12151Date: Wed Apr 29 01:55:10 2009 -0400 12152 12153 AVIVO: better fix for cursor flickering/corruption issues 12154 12155 Should prevent ghost cursor from being left on screen. 12156 12157commit 53e75cbd0ba8f39799d7b776cdc59d60aa2a6916 12158Author: Alex Deucher <alexdeucher@gmail.com> 12159Date: Wed Apr 29 01:05:31 2009 -0400 12160 12161 R6xx/r7xx: fix CURSOR_SWAPPING_* macros 12162 12163 r6xx/r7xx have different swapper regs 12164 12165commit 4e948e2aaa59267a9aa797396add2c6b7b4c1e7c 12166Author: Alex Deucher <alexdeucher@gmail.com> 12167Date: Tue Apr 28 19:28:25 2009 -0400 12168 12169 AVIVO: fix cursor corruption/flickering issues 12170 12171 When both crtcs are enabled, both cursors need to be enabled 12172 or you get corruption bands. 12173 12174 Thanks to Yang Zhao for figuring this out. 12175 12176commit 4d0beaf02b28d5a5dac4ff6053593a2f979f92ec 12177Author: Alex Deucher <alexdeucher@gmail.com> 12178Date: Tue Apr 28 12:33:24 2009 -0400 12179 12180 RV410: SE variants only have 1 quad pipe 12181 12182 Should fix EXA corruption with the 3D engine. 12183 Mesa and drm patches forthcoming. 12184 12185 Reported by Kano on IRC. 12186 12187commit abd565ae71d9844d5afcd658628ca7778ce26ab1 12188Author: Alex Deucher <alexdeucher@gmail.com> 12189Date: Mon Apr 27 14:21:38 2009 -0400 12190 12191 AVIVO: disable fractional feedback divs 12192 12193 Doesn't seem to work well in all cases. Needs more investigation. 12194 fixes bug 21413 12195 12196commit 93626b1277f57bd559cf71b263221209b02e0046 12197Author: Alex Deucher <alexdeucher@gmail.com> 12198Date: Mon Apr 27 03:35:24 2009 -0400 12199 12200 AVIVO: make sure cursor width isn't negative 12201 12202commit 32af0f097bccc4499d67e969d46ba35e7487c5e2 12203Author: Dave Airlie <airlied@linux.ie> 12204Date: Sat Apr 25 19:39:34 2009 +1000 12205 12206 Revert "r300/r500: set R clamp mode." 12207 12208 This reverts commit dab3d61619d91ae6a3da3e9b29b8258313653813. 12209 12210commit b2c8f57bccbd46164907049a26ad2dd209d7b108 12211Author: Dave Airlie <airlied@linux.ie> 12212Date: Sat Apr 25 19:39:19 2009 +1000 12213 12214 Revert "r3xx/r5xx: fix typo in last commit" 12215 12216 This reverts commit 1a7b982457ead5f21c6e62a409ca0dbf86005607. 12217 12218commit 253731dce6ab25122f2eda559233c098536b7143 12219Author: Dave Airlie <airlied@linux.ie> 12220Date: Sat Apr 25 19:39:05 2009 +1000 12221 12222 Revert "radeon: attempt to fix R clamping for repeat again" 12223 12224 This reverts commit 5998f262a52cb85b334fcc1fe24c7b2ae474ce93. 12225 12226commit 5998f262a52cb85b334fcc1fe24c7b2ae474ce93 12227Author: Dave Airlie <airlied@linux.ie> 12228Date: Fri Apr 24 17:30:57 2009 +1000 12229 12230 radeon: attempt to fix R clamping for repeat again 12231 12232commit 7ed27b926f28a1f088eace5034a7d9985f13752d 12233Author: Alex Deucher <alexdeucher@gmail.com> 12234Date: Thu Apr 23 14:06:06 2009 -0400 12235 12236 r2xx/r3xx/r4xx: further i2c fixups 12237 12238 - hw i2c engine has pin selection on r2xx/r2xx/r3xx chips 12239 - also switch hw i2c pin sel for external tmds 12240 12241commit 0da80f34b5aa5999d2030ffbc5187328fa4e4ae5 12242Author: Alex Deucher <alexdeucher@gmail.com> 12243Date: Wed Apr 22 18:42:36 2009 -0400 12244 12245 radeon: minor cleanup and warning fix 12246 12247commit 1a7b982457ead5f21c6e62a409ca0dbf86005607 12248Author: Alex Deucher <alexdeucher@gmail.com> 12249Date: Thu Apr 23 10:09:22 2009 -0400 12250 12251 r3xx/r5xx: fix typo in last commit 12252 12253 Noticed by osiris on IRC 12254 12255commit dab3d61619d91ae6a3da3e9b29b8258313653813 12256Author: Dave Airlie <airlied@redhat.com> 12257Date: Thu Apr 23 15:46:44 2009 +1000 12258 12259 r300/r500: set R clamp mode. 12260 12261 On rs690 without this set, we appear to end up fetching from well 12262 outside the texture sometimes into memory that isn't mapped. 12263 12264 This "shouldn't" have any effect. 12265 12266commit bbfd4625478e37d10fd58b23f46153f349370a52 12267Author: Alex Deucher <alexdeucher@gmail.com> 12268Date: Wed Apr 22 14:47:16 2009 -0400 12269 12270 radeon PM: update quirks 12271 12272commit e3c564114eea4d032f920fcebd4040f1934c21b0 12273Author: Alex Deucher <alexdeucher@gmail.com> 12274Date: Wed Apr 22 14:31:44 2009 -0400 12275 12276 AVIVO: fix for cursor corruption issue 12277 12278commit 24e4b73b4fbbb2c790e6120ede3caaa4e7e58359 12279Author: Alex Deucher <alexdeucher@gmail.com> 12280Date: Wed Apr 22 11:42:18 2009 -0400 12281 12282 radeon pll: add support for fractional feedback divs 12283 12284 Allows us to hit dot clocks much closer, especially on 12285 chips with non-27 Mhz reference clocks like most IGP chips. 12286 12287 This should fix most flickering and blanking problems with 12288 non-exact dot clocks. 12289 12290commit efa0825a86a8dc0f03ebb42c576ed26189e9d4bb 12291Author: Alex Deucher <alexdeucher@gmail.com> 12292Date: Tue Apr 21 13:51:11 2009 -0400 12293 12294 Tex vid: fix top clipping 12295 12296commit aa59b5b55e6ac64940fb3f0fa61f7b8fe36ba745 12297Author: Alex Deucher <alexdeucher@gmail.com> 12298Date: Tue Apr 21 12:06:24 2009 -0400 12299 12300 RS690: add quirk for acer board 12301 12302 thanks to Gino Badouri for testing 12303 12304commit 0e5164d968316bbb59b4484a7df087854738068f 12305Author: Alex Deucher <alexdeucher@gmail.com> 12306Date: Tue Apr 21 04:47:43 2009 -0400 12307 12308 Tex vid: rework part 2 12309 12310 - use pPriv->w/h directly for tex coords 12311 - take src x/y offset into account when calculating tex coords 12312 - when copying data, only optimize for y clipping. In order 12313 to deal with the x clipping optimization, the copy routines 12314 or tex coords would need to be fixed. This should fix clipping 12315 problems with the current code. 12316 12317commit 06a1b4169b506f69dd807380d9ce7210b3a3bc23 12318Author: Alex Deucher <alexdeucher@gmail.com> 12319Date: Tue Apr 21 01:21:51 2009 -0400 12320 12321 R4xx: avoid a PM related hang 12322 12323commit 228b9ac87668241b09980fca17e9d866891285e6 12324Author: Alex Deucher <alexdeucher@gmail.com> 12325Date: Tue Apr 21 01:02:34 2009 -0400 12326 12327 radeon PM: add some quirks 12328 12329commit b6febf7368d3df3f1d3fee6a87aa71c54e77cdcc 12330Author: Alex Deucher <alexdeucher@gmail.com> 12331Date: Mon Apr 20 18:04:18 2009 -0400 12332 12333 Tex vid: cleanup and unify common code 12334 12335commit d096d1189c8dc5e19f68cc605ccffe1a8fef048b 12336Author: Alex Deucher <alexdeucher@gmail.com> 12337Date: Mon Apr 20 13:05:18 2009 -0400 12338 12339 Endian fixes for object table 12340 12341commit 101ce28ec4af696549b7d6defebb5b4ad412176c 12342Author: Alex Deucher <alexdeucher@gmail.com> 12343Date: Sun Apr 19 12:13:18 2009 -0400 12344 12345 Remove old messages 12346 12347 R5xx/r6xx support isn't really experimental anymore 12348 12349commit db177c70ce88af19c8e05eb916a15f3e124876eb 12350Author: Alex Deucher <alexdeucher@gmail.com> 12351Date: Fri Apr 17 01:05:15 2009 -0400 12352 12353 Update Xv info in man page 12354 12355commit 00266177bed2dc8693df497ca3ec19f2dc4adc05 12356Author: Alex Deucher <alexdeucher@gmail.com> 12357Date: Thu Apr 16 20:33:28 2009 -0400 12358 12359 R3xx/R5xx: only apply Xv attributes if bicubic is disabled 12360 12361 Provides consistent output 12362 12363commit 85323a7f84381fef7fad20c7f7ec601637af9aa7 12364Author: Alex Deucher <alexdeucher@gmail.com> 12365Date: Thu Apr 16 11:10:51 2009 -0400 12366 12367 R3xx/R4xx: set tex caching for Y texture when doing planar rendering 12368 12369 Doesn't affect performance, but docs indicate its the right 12370 thing to do. 12371 12372commit 5ea5df22c038fc8f00984acc760e9d8c962bf902 12373Author: Alex Deucher <alexdeucher@gmail.com> 12374Date: Wed Apr 15 20:26:34 2009 -0400 12375 12376 Tex vid: remove remnants of XV_HWPLANAR 12377 12378 no longer needed as bicubic is the only thing that uses 12379 the old csc code. 12380 12381commit 9091b3f5f13dbea83ffd89679dac600e9f280bb2 12382Author: Alex Deucher <alexdeucher@gmail.com> 12383Date: Wed Apr 15 20:07:42 2009 -0400 12384 12385 R3xx/R4xx: fix up planar shader 12386 12387 We were overwriting the coord fetch address with the first 12388 tex fetch. Seemed to work however, luck I guess. Reorder 12389 the fetches to write to temp0 last. 12390 12391commit 14c13faeb9f9b7717a25fcc1ca97d46cc6ee0031 12392Author: Alex Deucher <alexdeucher@gmail.com> 12393Date: Wed Apr 15 19:53:12 2009 -0400 12394 12395 R5xx: add shader-based csc 12396 12397 - native planar support 12398 - Xv attributes 12399 12400commit 832efc7b90f5eb2da99512fcb902ab4838d2dcd1 12401Author: Alex Deucher <alexdeucher@gmail.com> 12402Date: Tue Apr 14 17:56:49 2009 -0400 12403 12404 R3xx/R4xx: Implement shader-based csc for packed formats 12405 12406commit 32625118c27041265d25811c00d25ab7e82fb340 12407Author: Alex Deucher <alexdeucher@gmail.com> 12408Date: Tue Apr 14 16:38:40 2009 -0400 12409 12410 tex vid: fix attribute setup typo for XV_COLORSPACE 12411 12412commit adf0912006b4f1597784dbfcc563d5c6d1c5667d 12413Author: Alex Deucher <alexdeucher@gmail.com> 12414Date: Tue Apr 14 16:16:33 2009 -0400 12415 12416 R6xx/R7xx: implement Xv attributes 12417 12418 - brightness, contrast, hue, etc. 12419 - TODO: implement gamma 12420 12421commit 8810fe92b5aed08888584c6914482586b59f71ab 12422Author: Alex Deucher <alexdeucher@gmail.com> 12423Date: Tue Apr 14 11:26:35 2009 -0400 12424 12425 R200: clean code, always use shader based csc 12426 12427 - consolidate common r2xx csc shader code 12428 - always use shader based csc for both packed and planar 12429 formats 12430 - always use native planar csc on r1xx 12431 12432commit 17685fefba68d188c7c0fe7a079180ec0722c046 12433Author: Alex Deucher <alexdeucher@gmail.com> 12434Date: Mon Apr 13 20:06:11 2009 -0400 12435 12436 Tex vid: general cleanup 12437 12438 - convert macros to more meaningful VTX_OUT_4 and VTX_OUT_6 12439 names to reflect that they actually do 12440 - fix indenting 12441 12442commit 093ab4c9a33b0b396b78c061c3321dc044bdccdc 12443Author: Alex Deucher <alexdeucher@gmail.com> 12444Date: Mon Apr 13 19:48:35 2009 -0400 12445 12446 R1xx: add support for native planar textured Xv 12447 12448commit ec0cb51df81c6c9a1de640d227fa9c9c33161083 12449Author: Alex Deucher <alexdeucher@gmail.com> 12450Date: Mon Apr 13 17:21:20 2009 -0400 12451 12452 R2xx tex vid: append verts for clip boxes 12453 12454 rather than sending a new draw packet for each rect 12455 12456commit fde075a30a8ee2c333aa1bbe8fbd177258b085ba 12457Author: Alex Deucher <alexdeucher@gmail.com> 12458Date: Mon Apr 13 17:13:51 2009 -0400 12459 12460 R1xx tex vid: append verts for clip boxes 12461 12462 rather than sending a new draw packet for each rect 12463 12464commit 12839fc17a2cca4ac14b9757bdaa63ba4679f96f 12465Author: Alex Deucher <alexdeucher@gmail.com> 12466Date: Mon Apr 13 17:04:31 2009 -0400 12467 12468 Tex vid: split by family 12469 12470commit a30737b337edb31528174b483c9094941a5d41bb 12471Author: Roland Scheidegger <sroland@tungstengraphics.com> 12472Date: Mon Apr 13 15:36:07 2009 -0400 12473 12474 r200/r300: implement brightness/contrast/hue/saturation/gamma controls for textured video 12475 12476 This implements 12477 contrast/brightness/hue/saturation controls for r200/r300 plus gamma (same 12478 gamma value for all channels used though separate values would be trivial) 12479 control for r300. 12480 Some issues left: 12481 - only r200/r300 12482 - still can't be combined with bicubic 12483 - controls will silently cease to work if the format used is packed and not 12484 planar (except for rv250) 12485 - gamma range is from 100 to 10000 corresponding to 0.1 and 10.0 like used in 12486 overlay. However, usable range is far smaller. Over 2.0 picture gets dark 12487 pretty quickly, and below 0.6 or so black seems to turn into purple (I've 12488 verified that even with gamma 1.0 black actually often seems to be RGB 1/0/1 so 12489 this explains this since that gets amplified by low gamma values - not sure if 12490 this is a rounding problem somewhere, bogus reference values or is somehow 12491 expected). 12492 - gamma adds a bit too many instructions for my taste (7) though the 12493 alternative (3 texture lookups + some swizzling instructions) doesn't seem any 12494 better. 12495 12496commit 9dd33cc8346073300312cd5f68de72d9bea1dd6b 12497Author: Dave Airlie <airlied@redhat.com> 12498Date: Fri Apr 17 10:30:21 2009 +1000 12499 12500 radeon: tweak ddc fix for all r300s. 12501 12502 Alex seems to think this might affect some other chips 12503 12504commit a846ec14dd65bb12f314f83ffb45c4c4be564a5e 12505Author: Dave Airlie <airlied@redhat.com> 12506Date: Fri Apr 17 10:24:15 2009 +1000 12507 12508 radeon: fix DDC on rv410 VGA ports for non-atom use. 12509 12510 The rv410 hw i2c block appears to hold down whatever i2c lines 12511 it is connected to when in reset, so when doing DDC change hw i2c to 12512 point at different i2c lines. 12513 12514commit efa1476fc0765b53298624198137dc5cc71d37a8 12515Author: Alex Deucher <alexdeucher@gmail.com> 12516Date: Thu Apr 16 16:43:02 2009 -0400 12517 12518 PM: fix reversed logic in pm block handler 12519 12520commit 2b95de17781959457a809c8fecc6bbb08336c83f 12521Author: Alex Deucher <alexdeucher@gmail.com> 12522Date: Thu Apr 16 12:52:41 2009 -0400 12523 12524 RS400/RS480: attempt to fix ClockGating on RS4xx chips 12525 12526commit c0215d2f2a27755fb04d20aaa4320a5e7da7ee04 12527Author: Alex Deucher <alexdeucher@gmail.com> 12528Date: Thu Apr 16 11:58:19 2009 -0400 12529 12530 RS690/RS740: fix connector enumeration in some cases 12531 12532 Should fix Masta-G's issue reported on IRC. 12533 12534commit 573339114660f06e5f257648ba413bab0d6083ce 12535Author: Alex Deucher <alexdeucher@gmail.com> 12536Date: Thu Apr 16 11:19:01 2009 -0400 12537 12538 R3xx/R4xx EXA: set explicit tex cache regions 12539 12540 When using a mask, set explicit cache partitions for 12541 each texture. Gives 1% performance improvement in 12542 x11perf with anti-aliased text. 12543 12544commit eea9800890b56bac9c07b7bd9c9e33fae2938af3 12545Author: Alex Deucher <alexdeucher@gmail.com> 12546Date: Thu Apr 16 10:41:28 2009 -0400 12547 12548 Revert "DCE3/3.2: cleanup dpms after previous changes" 12549 12550 This reverts commit f8c7d6a6162196a743f6885ecaf63ba50de1722a. 12551 12552 This is apparently still needed for some setups, however, I can't 12553 reproduce this locally anymore. 12554 12555commit 099cb27d760af61b967239d960bc77f71e828655 12556Author: Alex Deucher <alexdeucher@gmail.com> 12557Date: Wed Apr 15 16:23:01 2009 -0400 12558 12559 Change ForceLowPowerMode to 4 pcie lanes 12560 12561 This fixes corruption for some users 12562 12563commit 5abcea88d1f3307a7d08ae7757c4b535df1503bf 12564Author: Alex Deucher <alexdeucher@gmail.com> 12565Date: Wed Apr 15 15:52:04 2009 -0400 12566 12567 PM: don't touch clock gating regs unless ClockGating option is set 12568 12569 Fixes an M26 hang reported by tormod on IRC 12570 12571commit bae88f81e2a01c0d547132b5b91e717ff9d0497d 12572Author: Alex Deucher <alexdeucher@gmail.com> 12573Date: Wed Apr 15 14:38:35 2009 -0400 12574 12575 man page: fix missing break 12576 12577commit 4b3a3785f7b06205eeec6ca6b8913ecc2b94bd1f 12578Author: Alex Deucher <alexdeucher@gmail.com> 12579Date: Sat Apr 4 11:39:35 2009 -0400 12580 12581 Unify ForceLowPowerMode and DynamicPM Options 12582 12583 Force lower power mode and switch to a lower mode 12584 when idle. 12585 12586commit 63cb57c5eb0d6bc083f54eb100cb972b0e7bca69 12587Author: Alex Deucher <alexdeucher@gmail.com> 12588Date: Wed Apr 1 16:22:02 2009 -0400 12589 12590 Add DynamicPM Option 12591 12592 Dyanmically switch between power states. Switch to a low 12593 power state when the system is idle (DPMS off). 12594 12595commit a34a8b37afbea6ed4bf8ca42364195e174250c48 12596Author: Alex Deucher <alexdeucher@gmail.com> 12597Date: Wed Apr 1 15:23:17 2009 -0400 12598 12599 Set default low power PCIE lanes to 2 12600 12601 1 lane seems to cause occasional corruption when 12602 blitting to/from gart memory. 12603 12604commit 1f0dc778dc25f4f85fedd73c55c847cab2c79fc5 12605Author: Alex Deucher <alexdeucher@gmail.com> 12606Date: Wed Apr 1 15:12:27 2009 -0400 12607 12608 Add support for setting the number PCIE lanes 12609 12610commit 7e10b6222e8f44a3ecc6aaea55a7a7680d133bb3 12611Author: Alex Deucher <alexdeucher@gmail.com> 12612Date: Wed Apr 1 15:05:38 2009 -0400 12613 12614 Add ForceLowPowerMode Option 12615 12616 Force the chip to a low power mode at the expense 12617 of performance. 12618 12619commit 533d01a5933cd491bbc09cd463ea62475abf4bf2 12620Author: Alex Deucher <alexdeucher@gmail.com> 12621Date: Tue Mar 31 17:24:29 2009 -0400 12622 12623 radeon: Add functions to set sclk/mclk using atombios 12624 12625commit adb099409768e695b9928fa6aa5760f93dadd9af 12626Author: Alex Deucher <alexdeucher@gmail.com> 12627Date: Tue Mar 31 17:13:11 2009 -0400 12628 12629 radeon: Add functions to set sclk/mclk on r1xx-r4xx 12630 12631commit 68e2a959ccc3d1a5d0731f1b55fdf1b2412635b2 12632Author: Alex Deucher <alexdeucher@gmail.com> 12633Date: Tue Mar 31 17:02:45 2009 -0400 12634 12635 Add unified ClockGating Option 12636 12637 - replaces DynamicClocks Option as the name was misleading 12638 - unified interface for atom and com based bioses 12639 - fix up clock gating code for newer r3xx asics 12640 12641commit 336cee0e6d19068582b189b2c747d92cb2139d2f 12642Author: Dave Airlie <airlied@redhat.com> 12643Date: Mon Apr 13 18:11:48 2009 +1000 12644 12645 radeon: legacy bandwidth calcs for later r3xx/r4xx cards. 12646 12647 This is an attempt to rationalise the code using the register info 12648 files available to me here. 12649 12650 Further info is required: 12651 r350 check for stop_req > 15 then subtract 10: what other chips need it? 12652 get rs480 working instead of using magic. 12653 12654commit 367ba7af76e7f63f3186a6e8d9bbe21aa31fc6d1 12655Author: Alex Deucher <alexdeucher@gmail.com> 12656Date: Fri Apr 10 09:37:15 2009 -0400 12657 12658 radeon: fix tmds_pll randr property 12659 12660commit 15f1a2f31a7bc33f276a65283dbad0c106bee73e 12661Author: Alex Deucher <alexdeucher@gmail.com> 12662Date: Thu Apr 9 20:17:03 2009 -0400 12663 12664 radeon: remove old crtc size heuristics 12665 12666 IIRC, the old randr code used to use this to use for front buffer sizing, 12667 but it has since been changed. 12668 12669commit ce9077b2f5a5e749c0078dfb79d26ef8711a823a 12670Author: Alex Deucher <alexdeucher@gmail.com> 12671Date: Thu Apr 9 19:59:19 2009 -0400 12672 12673 R6xx/R7xx: cleanup macro 12674 12675commit 6532aeddacdeda3aa534264741c8648e1c449e76 12676Author: Michel Dänzer <daenzer@vmware.com> 12677Date: Thu Apr 9 09:41:56 2009 +0200 12678 12679 Clear the framebuffer before initializing modes. 12680 12681 Fixes garbage being visible shortly on server startup or when VT switching back 12682 to X. 12683 12684commit 0c19053975925fed637aaa41d440e0cab8f6b596 12685Author: Alex Deucher <alexdeucher@gmail.com> 12686Date: Wed Apr 8 13:34:23 2009 -0400 12687 12688 radeon: chip family typo 12689 12690commit 2888dd9ae9689b1cd72115dc0ceea1f5957299b0 12691Author: Alex Deucher <alexdeucher@gmail.com> 12692Date: Wed Apr 8 13:10:01 2009 -0400 12693 12694 R6xx/R7xx: set proper 3D client driver name for r600 12695 12696commit 61a083116a2c89e145aa19da951243536e07542e 12697Author: Alex Deucher <alexdeucher@gmail.com> 12698Date: Wed Apr 8 10:44:30 2009 -0400 12699 12700 Bump version post release 12701 12702commit 1b02b93895c31a0c9d641e47b46dce43b40edd97 12703Author: Alex Deucher <alexdeucher@gmail.com> 12704Date: Wed Apr 8 10:34:44 2009 -0400 12705 12706 Bump for release 12707 12708commit da74b94a5da8f0bcf4ae99b20e594d2f0f4fc6f6 12709Author: Alex Deucher <alexdeucher@gmail.com> 12710Date: Tue Apr 7 18:59:06 2009 -0400 12711 12712 ATOM: code cleanup 12713 12714 - remove cases that aren't possible (e.g., no TMDSA on DCE3 cards) 12715 - condense duplicate cases 12716 12717commit a24aa7250d09ae68ceaaba987314388b5e52096c 12718Author: Alex Deucher <alexdeucher@gmail.com> 12719Date: Tue Apr 7 18:52:08 2009 -0400 12720 12721 DCE3/3.2: further fixes 12722 12723 disable the dig transmitter and encoder before setup and enable 12724 12725commit 284e1943f6726ed08e796f99e49fbd90ac7fa8a4 12726Author: Alex Deucher <alexdeucher@gmail.com> 12727Date: Wed Apr 1 15:07:01 2009 -0400 12728 12729 Cast info->sclk to int 12730 12731commit b580e388e0e31843308c783c1ca859672dc6ca0e 12732Author: Alex Deucher <alexdeucher@gmail.com> 12733Date: Wed Apr 1 14:59:43 2009 -0400 12734 12735 Fix missing ')' in PCIE macro 12736 12737commit 8ec27be3c8043a6ddc816b0b1480dd19f8d73832 12738Author: Alex Deucher <alexdeucher@gmail.com> 12739Date: Tue Apr 7 11:33:44 2009 -0400 12740 12741 radeon: fix Xv vsync for multi-head 12742 12743commit f8c7d6a6162196a743f6885ecaf63ba50de1722a 12744Author: Alex Deucher <alexdeucher@gmail.com> 12745Date: Sat Apr 4 16:05:20 2009 -0400 12746 12747 DCE3/3.2: cleanup dpms after previous changes 12748 12749commit a707d355c3c6ff92252c5a060a1fc32d97547552 12750Author: Alex Deucher <alexdeucher@gmail.com> 12751Date: Sat Apr 4 15:20:01 2009 -0400 12752 12753 DCE3/3.2: fix up transmitter/encoder setup 12754 12755 Should fix bug 21050 12756 12757commit 215e12f9c0e8ac62c23af1add776ef88f9a0dc54 12758Author: Eduard Fuchs <edfuchs@uni-kassel.de> 12759Date: Thu Apr 2 12:36:59 2009 -0400 12760 12761 Enable byte swapping for r6xx/r7xx Hardware 12762 12763 Patch from Eduard Fuchs with some cleanup from me. 12764 12765 Tested at 32 bpp on MPC8641HPCN board (PowerPC) with 12766 HD2400 PCIe card 12767 12768commit 0ded9fab6bc3fecc1976ae8369b788659e4d7793 12769Author: Alex Deucher <alexdeucher@gmail.com> 12770Date: Thu Apr 2 12:09:00 2009 -0400 12771 12772 Add RV790 (HD 4890) support 12773 12774commit 0c6f20a9404c6a4210c5180bdca61c13fc186754 12775Author: Alex Deucher <alexdeucher@gmail.com> 12776Date: Tue Mar 31 15:51:11 2009 -0400 12777 12778 Add PCIE register accessors 12779 12780commit 229d4e20e5adcdf9207944c19ca721477936ff02 12781Author: Alex Deucher <alexdeucher@gmail.com> 12782Date: Tue Mar 31 15:39:42 2009 -0400 12783 12784 Unify mc_idle() 12785 12786commit f44b486848094b75479abef61c1aa075aede835c 12787Author: Alex Deucher <alexdeucher@gmail.com> 12788Date: Tue Mar 31 15:33:11 2009 -0400 12789 12790 COMBIS: Clarify case 0 for LCD DDC table 12791 12792commit e6f3c48899a92174a55f6fe7327412bc88039d87 12793Author: Alex Deucher <alexdeucher@gmail.com> 12794Date: Tue Mar 31 15:30:40 2009 -0400 12795 12796 COMBIOS: fix default sclk/mclk from bios 12797 12798commit 75bf5272936093bc024559e5744d97d8e79943b1 12799Author: Alex Deucher <alexdeucher@gmail.com> 12800Date: Tue Mar 31 15:17:37 2009 -0400 12801 12802 mac mini: 3rd time's the charm 12803 12804 fixes 10fbbac4bed59e12ad794ac873dd415d9ee4e340 12805 fixes bugs 20796 and 20979 12806 12807commit 10fbbac4bed59e12ad794ac873dd415d9ee4e340 12808Author: Alex Deucher <alexdeucher@gmail.com> 12809Date: Tue Mar 31 15:14:27 2009 -0400 12810 12811 Mac mini: fix connector setup for real 12812 12813 fixes f061308e7abcf93d1612681b935387fee23e0a24 12814 fixes bugs 20796 and 20979 12815 12816commit 8f2a03bfc86422669a46c2853050979f099a17a8 12817Author: Alex Deucher <alexdeucher@gmail.com> 12818Date: Mon Mar 30 11:30:34 2009 -0400 12819 12820 r1xx-r3xx TV: fix white noise 12821 12822 should fix bug 13872 12823 12824commit f061308e7abcf93d1612681b935387fee23e0a24 12825Author: Alex Deucher <alexdeucher@gmail.com> 12826Date: Fri Mar 27 01:14:12 2009 -0400 12827 12828 mac mini: fix connector setup regression 12829 12830 fixes bug 20796 12831 12832commit 47134ca77988f8c3b42d55ac4fcbb5dc182878d4 12833Author: Alex Deucher <alexdeucher@gmail.com> 12834Date: Thu Mar 26 17:18:47 2009 -0400 12835 12836 radeon: re-enable r/g/b intensity attributes for overlay 12837 12838 Noticed by sroland on IRC. 12839 12840commit e2b5e48e2652daf03fcae852d8368f29ea4abcd9 12841Author: Alex Deucher <alexdeucher@gmail.com> 12842Date: Thu Mar 26 13:42:25 2009 -0400 12843 12844 R6xx/R7xx EXA: rework composite pixel shader 12845 12846 - move to vram storage 12847 - move swizzle logic to tex setup 12848 12849commit 24dbd3f93c640f819a9f5029f5a49e41ab875bd4 12850Author: Alex Deucher <alexdeucher@gmail.com> 12851Date: Thu Mar 26 12:01:07 2009 -0400 12852 12853 R6xx/R7xx: clean up logic in EXA Comp PS setup 12854 12855commit 076e4e3ac1c81086aecb4e70dd30fd5d26ec7bc8 12856Author: Alex Deucher <alexdeucher@gmail.com> 12857Date: Thu Mar 26 11:36:26 2009 -0400 12858 12859 R6xx/R7xx: clean up bool const code 12860 12861 3 regs: 1 bit per bool, 32 bools per ps/vs/gs 12862 12863commit d2c3964fe04be42fe538f36439ed5ffca96e436a 12864Author: Roland Scheidegger <sroland@tungstengraphics.com> 12865Date: Wed Mar 18 01:55:12 2009 +0100 12866 12867 fix textured video allocation bug 12868 12869 size needs to be calculated after dstPitch adjustments, got already fixed 12870 for earlier than R600 chips by planar textured yuv patches, clean this up. 12871 12872commit 18e56eb179fde28477487c63e6f9ebf7579e2cd5 12873Author: Roland Scheidegger <sroland@tungstengraphics.com> 12874Date: Thu Mar 5 02:07:46 2009 +0100 12875 12876 don't convert planar yuv to packed for r200 12877 12878 uses 3 textures for planar yuv and does yuv->rgb conversion in the shader. 12879 Similar to r300 code, but might have precision issues - hardware alu should 12880 have enough precision but hardware consts are only 8bit and we'd want 12881 at least 11. 12882 This also enables textured video on rv250 (and also supports packed yuv 12883 on that chip by using basically the same shader with packed data). 12884 12885commit 58530bf4912800f9e09ebaea42a13cff8a80c19e 12886Author: Roland Scheidegger <sroland@tungstengraphics.com> 12887Date: Sat Feb 21 04:46:31 2009 +0100 12888 12889 don't convert planar yuv to packed for r300 12890 12891 uses 3 textures for planar yuv and does yuv->rgb conversion in the shader. 12892 small performance advantage, but manual texture cache setting is necessary 12893 otherwise it may be measurably slower (but probably not relevant) in some 12894 cases. 12895 Unlike some other drivers, using MADs instead of DP3s, since this requires 12896 less instructions due to no MOVs are required, the end result is the same 12897 though the constants need to be different. 12898 Use of this is user settable for now (XV_HWPLANAR attrib). 12899 12900commit 97e19d96ba65a3df2fa3bbf73cfcc01b6dc3e796 12901Author: Roland Scheidegger <sroland@tungstengraphics.com> 12902Date: Tue Dec 30 22:23:39 2008 +0100 12903 12904 clip fixes 12905 12906 This fixes some oddities observed when the video is only partly visible. 12907 Instead of recalculating the geometry of the video, always use the same. 12908 Also fixes a assignment present twice, and another issue (bring in line with 12909 what the overlay code does). 12910 12911commit 7aaf5f08e4147b75c4975e6d680d3f5a93424164 12912Author: Alex Deucher <alexdeucher@gmail.com> 12913Date: Mon Mar 23 12:55:02 2009 -0400 12914 12915 R4xx ATOM: fix crtc routing for CRT1 12916 12917 should fix bug 20814 12918 12919commit 2fe6b393d9dbb193d4f0a3351ad76618cda07e2c 12920Author: Alex Deucher <alexdeucher@gmail.com> 12921Date: Sun Mar 22 01:31:30 2009 -0400 12922 12923 radeon: add new chip ids 12924 12925commit 1ddee7cd6fd267b2fc86f21af27c5425eb0835a4 12926Author: Alex Deucher <alexdeucher@gmail.com> 12927Date: Fri Mar 20 13:44:00 2009 -0400 12928 12929 DCE3.2: use RMX for for non-native modes on DVI 12930 12931 DVI seems to have issues with low dotclocks, so 12932 use the scaler instead. 12933 12934 Fixes bug 20754 12935 12936commit 44081e8fd2bf974272d5447b45dd7b206cf939c1 12937Author: Alex Deucher <alexdeucher@gmail.com> 12938Date: Thu Mar 19 20:48:16 2009 -0400 12939 12940 R6xx/R7xx EXA: fix maxPitchBytes 12941 12942 should now allow accel up the hw max of 8192x8192 12943 12944commit e8e2367b66d9a4f5977d95f52b37cdbad75e409a 12945Author: Alex Deucher <alexdeucher@gmail.com> 12946Date: Thu Mar 19 20:09:32 2009 -0400 12947 12948 Add new pci ids 12949 12950commit 68e83f10adb23fc359c1da881f5d666cd08b4067 12951Author: Alex Deucher <alexdeucher@gmail.com> 12952Date: Wed Mar 18 20:58:19 2009 -0400 12953 12954 bump version post release 12955 12956commit a196aae19603c8651dc502f5cb0df4d55befe163 12957Author: Alex Deucher <alexdeucher@gmail.com> 12958Date: Wed Mar 18 20:52:12 2009 -0400 12959 12960 bump for release 12961 12962commit 537da7018b0fd2ad263fc5cdc55b3b85cf8e9205 12963Author: Alex Deucher <alexdeucher@gmail.com> 12964Date: Wed Mar 18 19:22:55 2009 -0400 12965 12966 ATOM: fix up tv-out support 12967 12968 - fix purple hue when using ATOMTvOut option 12969 - fix TV load detection on newer r6xx/r7xx chips 12970 - disable CV for now 12971 12972commit c0e2513ab128ddd5be0ed626d9e31777a98983ef 12973Author: Alex Deucher <alexdeucher@gmail.com> 12974Date: Wed Mar 18 14:08:30 2009 -0400 12975 12976 R6xx/R7xx AGP: disable gart data transfers 12977 12978 UTS/DFS/Xv 12979 12980 Doesn't seem to be reliable on AGP. 12981 12982 fixes bugs: 20436, 20236, several reports on ML and IRC 12983 12984commit e3e1f0c7b892b094b292081021a0e97932b8e079 12985Author: Fabio <fabio.ped@libero.it> 12986Date: Wed Mar 18 12:13:53 2009 -0400 12987 12988 man page update 12989 12990commit 6ae54d3ea023968887bf0cb0ec1be2115621d554 12991Author: Alex Deucher <alexdeucher@gmail.com> 12992Date: Wed Mar 18 11:50:22 2009 -0400 12993 12994 AVIVO: Fix dac load detection 12995 12996commit 04e36b794c216e4fa77c780fd989df1ef1c7f0d9 12997Author: Fabio <fabio.ped@libero.it> 12998Date: Tue Mar 17 11:29:26 2009 -0400 12999 13000 man page updates 13001 13002 from bug 20648 13003 13004commit a6855c370194b6df307ea33724fe17a85d67607e 13005Author: Alex Deucher <alexdeucher@gmail.com> 13006Date: Mon Mar 16 13:52:42 2009 -0400 13007 13008 radeon: add another AGP quirk 13009 13010 bug 20647 13011 13012commit a55ced5ee20c07e743c7c0978803fd10589c1531 13013Author: Alex Deucher <alexdeucher@gmail.com> 13014Date: Sun Mar 15 12:08:04 2009 -0400 13015 13016 Fix fallout from r6xx/r7xx EXA merge in IB handling 13017 13018 Noticed Christiaan van Dijk. 13019 Should fix bug 20510 13020 13021commit a065eeb9b301ab452cc8fbd3cac1fd9a3841c4e1 13022Author: Alex Deucher <alexdeucher@gmail.com> 13023Date: Sat Mar 14 18:40:28 2009 -0400 13024 13025 atom: fix up bios scartch register usage 13026 13027 the bios scratch registers keep 3 types of state for each output: 13028 1. dpms 13029 2. active 13030 3. connected 13031 13032 some of the command tables use the active flags. We used to set the active 13033 flags based on dpms state, but instead set them based on whether they are 13034 connected or not. 13035 13036commit 5c4ef924d082fd3b279f62fda7a97f563b009d38 13037Author: Alex Deucher <alexdeucher@gmail.com> 13038Date: Sat Mar 14 18:16:49 2009 -0400 13039 13040 radeon: fix typo in bios scratch reg setup 13041 13042commit 96fce5a56269509a935e8ee621a5e17cc9005ddc 13043Author: Alex Deucher <alexdeucher@gmail.com> 13044Date: Fri Mar 13 19:29:58 2009 -0400 13045 13046 Bump post-release 13047 13048commit 8da1b82a70de7b3ef38905bd77c58a20fd44ddd6 13049Author: Alex Deucher <alexdeucher@gmail.com> 13050Date: Fri Mar 13 19:23:19 2009 -0400 13051 13052 bump for release 13053 13054commit 28d9d699dcdff394aa561505f7f54f08db800828 13055Author: Alex Deucher <alexdeucher@gmail.com> 13056Date: Fri Mar 13 18:18:31 2009 -0400 13057 13058 radeon: man page updates 13059 13060commit 06963796d00b5cb6704412a7d4417fcefae4aa4d 13061Author: Adam Jackson <ajax@redhat.com> 13062Date: Fri Mar 13 18:14:43 2009 -0400 13063 13064 uniphy: start adding DisplayPort setup 13065 13066commit 1db81afe4f8db55e02a65cd22ad84c6bbb27e8ef 13067Author: Alex Deucher <alexdeucher@gmail.com> 13068Date: Fri Mar 13 17:51:56 2009 -0400 13069 13070 radeon: add support for 30 bit LUTs 13071 13072 All radeons have them. Thanks to Yang Zhao for figuring 13073 this out. 13074 13075commit 5bb326c64ad3bf5c93943280dcbcb25c8910d1aa 13076Author: Alex Deucher <alexdeucher@gmail.com> 13077Date: Fri Mar 13 14:10:58 2009 -0400 13078 13079 R6xx/R7xx: EXA VSync Option not supported yet 13080 13081commit 8c2d40894448329e721360811c124998a17ba2ba 13082Author: Alex Deucher <alexdeucher@gmail.com> 13083Date: Fri Mar 13 14:03:43 2009 -0400 13084 13085 R6xx/R7xx: fix up vline stuff along the lines of previous chips 13086 13087 Avoids an additional function call. 13088 13089commit 176c1b43fc3ee48e87c134446754360b3683c24b 13090Author: Alex Deucher <alexdeucher@gmail.com> 13091Date: Fri Mar 13 13:28:42 2009 -0400 13092 13093 R6xx/R7xx: return in RADEONWaitForIdleMMIO() if accel is off 13094 13095 Don't wait for engine idle if we haven't initialized the engine. 13096 Behavior should be consistent with versions of the driver prior 13097 to accel support being added. 13098 13099 Should fix bug 20645 13100 13101commit d538a4508248adf2debaf0cc870a2d1c48abd986 13102Author: Alex Deucher <alexdeucher@gmail.com> 13103Date: Fri Mar 13 13:22:03 2009 -0400 13104 13105 R6xx/r7xx: clarify accel messages 13106 13107 We only support EXA and and only with DRI. 13108 13109commit 1b7090a4c0d6a38701fbfb928fac8e15d40fb81f 13110Author: Adam Jackson <ajax@redhat.com> 13111Date: Fri Mar 13 17:41:59 2009 -0400 13112 13113 Add some more cases to the DVI single-link checks. 13114 13115commit 104e13da74de37477e3a31a5de96c5ebb8fe888b 13116Author: Cooper Yuan <cooperyuan@gmail.com> 13117Date: Fri Mar 13 11:47:43 2009 -0400 13118 13119 radeon: save bios scratch registers in Preinit() 13120 13121 The display detect routines can change these which causes 13122 havok with some bioses. 13123 13124commit f70d7a0fd9ac3659d93511ab362eb80a7ff49777 13125Author: Dave Airlie <airlied@redhat.com> 13126Date: Fri Mar 13 12:56:29 2009 +1000 13127 13128 r600: reload shaders into VRAM on resume 13129 13130 As VRAM gets zeroed out over s/r, we need to reload the 13131 shaders. 13132 13133commit 1fe0dcfe98dfe30499ba53c1f208c4ef10bef001 13134Author: Yang Zhao <yang@yangman.ca> 13135Date: Thu Mar 12 18:50:32 2009 -0400 13136 13137 R6xx/R7xx shader: Fix OFFSET_[XYZ] macro for TEX_DWORD2 to accept floats 13138 13139 Values for OFFSET_[XYZ] are 5-bits two's-complement fixed-point with 13140 one-bit after decimal point. Values in [-8.0, 7.5] are valid. Inputs 13141 that do not exactly land on 0.5 increments are rounded towards 0 to the 13142 nearest increment. 13143 13144commit f356ca7795531e0c080775efb598141eb7db44ba 13145Author: Alex Deucher <alexdeucher@gmail.com> 13146Date: Thu Mar 12 03:42:40 2009 -0400 13147 13148 RS600: fix page table size for rs600 as well 13149 13150commit ef69336584c45e35aabb797a06c4030e75d66507 13151Author: Dave Airlie <airlied@redhat.com> 13152Date: Thu Mar 12 14:54:18 2009 +1000 13153 13154 r600: fix sizing of PCI GART table for r600 13155 13156commit 945ccbbd4fa2b65ccdfb23716c178c95b036734d 13157Author: Alex Deucher <alexdeucher@gmail.com> 13158Date: Tue Mar 10 09:53:14 2009 -0400 13159 13160 radeon: clean more thoroughly in RADEONFreeRec() 13161 13162commit 03ec2099fdc859b23e5fa56ac8696ec045f1b110 13163Author: 오유연(Yu-yeon Oh) <yuyeon_oh@tmax.co.kr> 13164Date: Tue Mar 10 09:51:08 2009 -0400 13165 13166 radeon_driver.c small memory bug 13167 13168commit 60dac878cc8626b0821c4c7a6a6dfa12a5597cfa 13169Author: Alex Deucher <alexdeucher@gmail.com> 13170Date: Mon Mar 9 20:48:39 2009 -0400 13171 13172 R4xx: add R4xxATOM option 13173 13174 If set, enables the use of atombios for modesetting 13175 on r4xx cards. 13176 13177commit 758fcc33c232c32e7b66261fca51de161d1a4800 13178Author: Dave Airlie <airlied@redhat.com> 13179Date: Tue Mar 10 10:34:13 2009 +1000 13180 13181 r600: enable DRI by default 13182 13183 fallback to shadowfb if DRI fails pre init 13184 13185commit 7735e411954fd2cb6105c30ebd5a1b209974ce22 13186Author: Alex Deucher <alexdeucher@gmail.com> 13187Date: Mon Mar 9 17:49:25 2009 -0400 13188 13189 R6xx/r7xx: clarify EXA message 13190 13191commit 1d355798f3d421cabf2a3a0ba32a580483476ab0 13192Author: Alex Deucher <alexdeucher@gmail.com> 13193Date: Mon Mar 9 15:13:01 2009 -0400 13194 13195 AVIVO: fix panning 13196 13197 low 2 bits of X need to be 0. 13198 13199commit d1add18f3d238c755f9875b5e5c211e4af742482 13200Author: root <root@benzylpiperazine.boston.devel.redhat.com> 13201Date: Fri Mar 6 17:53:59 2009 -0500 13202 13203 output: Filter out dual-link modes from DP->DVI connections 13204 13205 There's not enough pins on a DP->DVI connector for this to possibly work. 13206 13207commit 04481bf450650d1ea5aadabc9213b899b0fe8119 13208Author: root <root@benzylpiperazine.boston.devel.redhat.com> 13209Date: Fri Mar 6 17:32:15 2009 -0500 13210 13211 atom: Enable DisplayPort source to DVI/HDMI sink 13212 13213 DP source to DP sink still doesn't work yet, of course. 13214 13215commit 0ba73e040b94590867f8b1071a26da2526a3c375 13216Author: Adam Jackson <ajax@redhat.com> 13217Date: Fri Mar 6 15:38:00 2009 -0500 13218 13219 output: Allow for multiple DisplayPort outputs. 13220 13221commit 8f7c17a6b77ca368f067c2acb803cc8f9ef19351 13222Author: Tormod Volden <debian.tormod@gmail.com> 13223Date: Fri Mar 6 09:42:39 2009 -0500 13224 13225 RV350: Add AGPMode quirk for Thinkpad 13226 13227 fixes bug 15096 (and Ubuntu bug #195051) 13228 13229commit 5dc4b69fbdd0e8ca48cc00321f193260a8f10bd0 13230Author: Alex Deucher <alexdeucher@gmail.com> 13231Date: Thu Mar 5 12:26:04 2009 -0500 13232 13233 radeon: add a few more default common modes for lvds 13234 13235commit 3b0fc22ad9e5c0f120a74c4f3d8e48c270f0ff29 13236Author: Alex Deucher <alexdeucher@gmail.com> 13237Date: Wed Mar 4 19:04:34 2009 -0500 13238 13239 R6xx/R7xx: wire up DMAForXv option like older asics 13240 13241 Select between SW and HW-assisted uploads 13242 13243commit 52f06ace04ad8141effc45fb6a0107a05bb46a73 13244Author: Alex Deucher <alexdeucher@gmail.com> 13245Date: Wed Mar 4 18:51:44 2009 -0500 13246 13247 R6xx/R7xx: disable XV_BICUBIC attribute 13248 13249 It's not implemented yet. 13250 13251commit b1fd883b59b85fed8782e035890098908902f4ce 13252Author: Alex Deucher <alexdeucher@gmail.com> 13253Date: Wed Mar 4 11:17:16 2009 -0500 13254 13255 RN50: fix up cloning on servers 13256 13257 Since the rn50 only has one crtc, but often has multiple outputs, 13258 occasionally one of the outputs doesn't get a crtc assigned eventhough 13259 the outputs may have common modes. If we see an output with modes, but 13260 no crtc, assign it so all the attached outputs light up. 13261 13262 fixes bugs 19457 and similar ones 13263 13264commit 592bcac52f113a95923a8f1cb8427e7552d5670b 13265Author: Alex Deucher <alexdeucher@gmail.com> 13266Date: Wed Mar 4 10:53:20 2009 -0500 13267 13268 ATOM: don't use fixed ref div for LVDS 13269 13270 The bios doesn't provide one. If there are problems 13271 with certain panels, we may need to adjust the PLL flags 13272 in certain cases. 13273 13274 Should fix bug 20458 13275 13276commit e88b0eac8cab8b54b769fd7b4ad3b3b65a90de09 13277Author: Alex Deucher <alexdeucher@gmail.com> 13278Date: Wed Mar 4 02:52:36 2009 -0500 13279 13280 radeon: just add some common modes for LVDS 13281 13282 avoids needing to hack around with the edid 13283 13284commit 2bb319fd80d20a3d4e5cc9416891bc7e4c3e431e 13285Author: Alex Deucher <alexdeucher@gmail.com> 13286Date: Wed Mar 4 02:50:45 2009 -0500 13287 13288 Revert "radeon: adjust LVDS so that default modes get added" 13289 13290 This reverts commit 1a2b16561d19ec9c027c562902f5fc086c856994. 13291 13292commit 1a2b16561d19ec9c027c562902f5fc086c856994 13293Author: Alex Deucher <alexdeucher@gmail.com> 13294Date: Tue Mar 3 20:21:17 2009 -0500 13295 13296 radeon: adjust LVDS so that default modes get added 13297 13298 we can scale, so add the default modes. for panels 13299 with an EDID, set the continous freq bit. for panels without 13300 and EDID, add a FAKE edid with the continous freq bit set. 13301 13302 Based on similar code in the Intel driver. 13303 13304commit 71117970df36cbe689ef15e9a6cca24439b4cd62 13305Author: Alex Deucher <alexdeucher@gmail.com> 13306Date: Tue Mar 3 19:40:30 2009 -0500 13307 13308 AVIVO: add aspect scaling mode 13309 13310 No luck yet for aspect on pre-avivo chips 13311 13312commit d586a2c6f821c821a4a7708a3382acb63187534f 13313Author: Alex Deucher <alexdeucher@gmail.com> 13314Date: Tue Mar 3 12:17:30 2009 -0500 13315 13316 Rotation: don't rotate if acceleration is not active 13317 13318commit fdc64948119e2218921b4dc0a492522442668875 13319Author: Alex Deucher <alexdeucher@gmail.com> 13320Date: Mon Mar 2 16:33:46 2009 -0500 13321 13322 R6xx/r7xx: remove some unneeded code I missed in the last commit 13323 13324commit 4ad1c4decfee653dbbc1ea2ca4270487be622382 13325Author: Alex Deucher <alexdeucher@gmail.com> 13326Date: Mon Mar 2 15:38:35 2009 -0500 13327 13328 R6xx/R7xx EXA: cleanup composite texture setup 13329 13330commit b7164ac4ad55e5d0fc474df8ae762b469b91ba30 13331Author: Alex Deucher <alexdeucher@gmail.com> 13332Date: Mon Mar 2 04:08:09 2009 -0500 13333 13334 R6xx/R7xx EXA: combine composite mask/non-mask VS 13335 13336 Also fix set_bool_const() 13337 the CF bool consts are not contiguous by shader type 13338 There are 96 boolean constants (32 each for PS, VS, GS) and 13339 they are ordered as follows: 13340 ps, vs, gs ... ps, vs, gs 13341 13342commit fa98f424de739be2c6005b740a74bbf1ee968a8b 13343Author: Alex Deucher <alexdeucher@gmail.com> 13344Date: Mon Mar 2 02:28:57 2009 -0500 13345 13346 R6xx/R7xx Xv: combine packed and planar shaders 13347 13348 use a bool const to select the tex fetch routine 13349 13350commit ccde35c3eda3fff0de29eb8c6fdc392629724a34 13351Author: Christian Koenig <deathsimple@vodafone.de> 13352Date: Sun Mar 1 23:38:37 2009 -0500 13353 13354 R6xx/R7xx: move shaders to r600_shader.c and fixup Xv PS 13355 13356 patches from Christian Koenig with some adjustments from me 13357 13358commit a8e631c1b1c9b46602aeca66f8e7e68154d0bfc8 13359Author: Alex Deucher <alexdeucher@gmail.com> 13360Date: Sun Mar 1 22:30:31 2009 -0500 13361 13362 R6xx/R7xx: code cleanups 13363 13364commit 528061d51bb4248d6fabec9579dead32a730467a 13365Author: Alex Deucher <alexdeucher@gmail.com> 13366Date: Sun Mar 1 18:28:13 2009 -0500 13367 13368 R6xx/R7xx: write vertexes directly to the IB 13369 13370 Reduces the vertex buffer setup overhead 13371 13372commit 96a0bafa16dbb67a3d10fa6fe45be13930495ddc 13373Author: Alex Deucher <alexdeucher@gmail.com> 13374Date: Sun Mar 1 17:58:36 2009 -0500 13375 13376 R6xx/R7xx: switch emit functions to macros 13377 13378 This improves performance due to decreased function call 13379 overhead. 13380 13381commit cbd579c41a16bddcbb8793eac2a41f2e958bef4d 13382Author: Alex Deucher <alexdeucher@gmail.com> 13383Date: Sun Mar 1 13:16:20 2009 -0500 13384 13385 Print a message when we have a shared DDC line 13386 13387commit a7f465f73363fce409870f62173d518b1bc02ae6 13388Author: Alex Deucher <alexdeucher@gmail.com> 13389Date: Sun Mar 1 12:39:31 2009 -0500 13390 13391 R300: Add AGP quirk 13392 13393 bug 7770 13394 13395commit 017bc7f34de2e40919f0e711205b78ee4e3643f6 13396Author: Tormod Volden <debian.tormod@gmail.com> 13397Date: Sat Feb 28 10:33:35 2009 -0500 13398 13399 M9+: Add AGP quirk for Sony Vaio 13400 13401commit 49b57767d0d2c041517b0764c2ed2d2ba5a7092c 13402Author: Bryce Harrington <bryce@bryceharrington.org> 13403Date: Fri Feb 27 20:31:03 2009 -0500 13404 13405 Quirk for RV280 on 82865G/PE/P DRAM Controller/Host-Hub 13406 13407 Solves issue where display suffers from line corruption. 13408 (See https://edge.launchpad.net/ubuntu/+bug/300304) 13409 13410commit 150abd4123664eab65ea43585dbdb9b3151c46af 13411Author: Tormod Volden <lists.tormod@gmail.com> 13412Date: Fri Feb 27 11:23:32 2009 -0500 13413 13414 Janitor: cosmetic clean-up of AGPMode quirk table 13415 13416commit c7d344f0ad255761d5ae30da59cf0a8b532dfc68 13417Author: Michel Dänzer <daenzer@vmware.com> 13418Date: Fri Feb 27 17:05:01 2009 +0100 13419 13420 EXA: Make sure Prepare/FinishAccess hooks can handle EXA_PREPARE_AUX* indices. 13421 13422 And signal this to the EXA core using the new EXA_SUPPORTS_PREPARE_AUX flag. 13423 13424commit f4d57105fc869e9318f447a1514f1b284a0a4ec0 13425Author: Michel Dänzer <daenzer@vmware.com> 13426Date: Fri Feb 27 09:23:24 2009 +0100 13427 13428 Revert "EXA: Adapt to EXA changes in xserver Git." 13429 13430 This reverts commit de358736dc696559ba99c71cf5b2a97508201630. 13431 13432commit 3285a4af17b19f2d1c24eebfb6aa51cddcf3ca50 13433Author: Alex Deucher <alexdeucher@gmail.com> 13434Date: Thu Feb 26 17:31:13 2009 -0500 13435 13436 DCE3.2: fix up Save()/Restore() 13437 13438commit 4fc978f3d83ef6370c9715e7ca5350640ccbfe10 13439Author: Alex Deucher <alexdeucher@gmail.com> 13440Date: Thu Feb 26 12:10:08 2009 -0500 13441 13442 R6xx/R7xx: use shadowfb if DRI is disabled 13443 13444commit 000756e052a291230e5c95e48b69a5aa9c4fab0e 13445Merge: 22d7746 8373f43 13446Author: Alex Deucher <alexdeucher@gmail.com> 13447Date: Thu Feb 26 11:44:13 2009 -0500 13448 13449 Merge branch 'r6xx-r7xx-support' of git+ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 13450 13451commit 22d7746bfc1bcbb3269afac88971254088ea9ece 13452Author: Alex Deucher <alexdeucher@gmail.com> 13453Date: Thu Feb 26 11:00:46 2009 -0500 13454 13455 RS600: enable the DRI by default 13456 13457commit 8373f4399b03961f2c928a9275d47e9f41bd92bb 13458Author: Alex Deucher <alexdeucher@gmail.com> 13459Date: Wed Feb 25 09:16:38 2009 -0500 13460 13461 R6xx/R7xx EXA: same surface and same coords equals nop 13462 13463 should fix bug 20305 13464 13465commit c74727015453ff3c3d6d06b812ebca9eb19a9767 13466Author: Alex Deucher <alexdeucher@gmail.com> 13467Date: Tue Feb 24 11:08:46 2009 -0500 13468 13469 R6xx/R7xx EXA: init copy_area to NULL 13470 13471commit 95ce13572dc2d9f5dd6cf55c23411e275c0aadf1 13472Author: Alex Deucher <alexdeucher@gmail.com> 13473Date: Tue Feb 24 10:51:13 2009 -0500 13474 13475 R6xx/R7xx EXA: Optimize temp surface for overlapping copies 13476 13477 - allocate temp surface in PrepareCopy() 13478 - fall back to old OverlapCopy() path if we are not able 13479 to allocate a temp surface 13480 13481commit 1a7db3fc2a0277d724d60d028064d8ef75019c28 13482Author: Mark van Doesburg <mark.vandoesburg@hetnet.nl> 13483Date: Tue Feb 24 10:44:19 2009 -0500 13484 13485 R6xx/R7xx EXA: use a temp surface for overlapping copy 13486 13487commit 957c2ba3459af152ad16a3f6dd4631044f9df6f9 13488Author: Alex Deucher <alexdeucher@gmail.com> 13489Date: Tue Feb 24 09:43:58 2009 -0500 13490 13491 RS600: fix MC addr mask 13492 13493 Noticed by osiris on IRC 13494 13495commit de358736dc696559ba99c71cf5b2a97508201630 13496Author: Michel Dänzer <daenzer@vmware.com> 13497Date: Tue Feb 24 08:30:23 2009 +0100 13498 13499 EXA: Adapt to EXA changes in xserver Git. 13500 13501 The PrepareAccess hook can now get six different indices. 13502 13503 Also remove superfluous instances of #include "exa.h". 13504 13505 Note that you may need to run autoheader manually to get config.h.in updated. 13506 13507commit 752b0218af36a45b333a12f66dfdce8c1e662aa8 13508Author: Michel Dänzer <daenzer@vmware.com> 13509Date: Tue Feb 24 08:13:06 2009 +0100 13510 13511 EXA: Pass pScrn and info into RadeonCompositeTile. 13512 13513 Saves some cycles looking them up. 13514 13515commit e2e508bc08d993fc108297e9cd2fa0ce183282bc 13516Author: Michel Dänzer <daenzer@vmware.com> 13517Date: Tue Feb 24 08:12:42 2009 +0100 13518 13519 Only call RADEONWaitForVLine if it might actually do anything useful. 13520 13521 Otherwise it and RADEONBiggerCrtcArea show up pretty high in x11perf -aa*text 13522 profiles. 13523 13524commit 5ab8699c97c050424cf813b9a60550aba32446a2 13525Author: Michel Dänzer <daenzer@vmware.com> 13526Date: Tue Feb 24 08:12:23 2009 +0100 13527 13528 Fix compile warning when building without EXA. 13529 13530commit 132e61741185f4a1114e66a34bdbf21e08e4d5d8 13531Author: Alex Deucher <alexdeucher@gmail.com> 13532Date: Mon Feb 23 12:13:41 2009 -0500 13533 13534 radeon: re-enable load detection output attribute for TV/CV 13535 13536commit 918d511f6f26d2fdeb63d6e32cd348707b4a2140 13537Author: Alex Deucher <alexdeucher@gmail.com> 13538Date: Mon Feb 23 10:22:19 2009 -0500 13539 13540 RS600: fix up MC setup 13541 13542 Please test if you have an RS600 13543 13544commit 27f8ca2cce65be2bcb3375231886d5444d251808 13545Author: Alex Deucher <alexdeucher@gmail.com> 13546Date: Thu Feb 19 21:18:07 2009 -0500 13547 13548 R6xx/R7xx: add wait for idle MMIO path 13549 13550commit 1f783117f79f95f1744681ccc7d80dc0e7b53057 13551Author: Alex Deucher <alexdeucher@gmail.com> 13552Date: Thu Feb 19 21:07:06 2009 -0500 13553 13554 RBBM_GUICNTL is pre-r6xx only 13555 13556commit f3933ed200dbf61b05a114f7697f2bb68d92abe7 13557Author: Alex Deucher <alexdeucher@gmail.com> 13558Date: Thu Feb 19 20:12:33 2009 -0500 13559 13560 Don't write new HDP location until we've written the new FB location 13561 13562commit 7f6801abb4bbd79830999d99ccdbd672a9984df6 13563Author: Alex Deucher <alexdeucher@gmail.com> 13564Date: Thu Feb 19 19:02:52 2009 -0500 13565 13566 radeon: one more 32 -> 64 just to be safe 13567 13568 missed this in cadae2a8d9979b2d61eae7d1f7a9ab8f5eb28e83 13569 13570commit ca22390c8a5256bbc8eeff775d5eebf0e892d890 13571Author: Alex Deucher <alexdeucher@gmail.com> 13572Date: Thu Feb 19 18:58:40 2009 -0500 13573 13574 R6xx/R7xx: fix up a few more paths 13575 13576commit cadae2a8d9979b2d61eae7d1f7a9ab8f5eb28e83 13577Author: Alex Deucher <alexdeucher@gmail.com> 13578Date: Thu Feb 19 18:50:49 2009 -0500 13579 13580 Fix MC setup on systems with more than 512 MB of VRAM 13581 13582 many thanks to app4des on IRC for helping me sort this out. 13583 13584commit 42492c70c481c88911337eefb97ba9d030adf7b6 13585Author: Alex Deucher <alexdeucher@gmail.com> 13586Date: Thu Feb 19 10:20:28 2009 -0500 13587 13588 R6xx/R7xx: wait for MC idle when changing the MC 13589 13590 should fix bug 19984 13591 13592commit e6475282486f4895bc68f6b093ecbb1aa6d25f72 13593Author: Alex Deucher <alexdeucher@gmail.com> 13594Date: Wed Feb 18 20:21:13 2009 -0500 13595 13596 R6xx/R7xx Xv: fix some missing bits from last commit 13597 13598commit adff8906c9899dde7711382577a63f4a726437ca 13599Author: Alex Deucher <alexdeucher@gmail.com> 13600Date: Wed Feb 18 20:01:57 2009 -0500 13601 13602 R6xx/R7xx EXA/Xv: properly deal with running out of vertex buffer space 13603 13604 As noted by mhopf, if VGT_MAX/MIN_INDX, etc. regs change, you need to re-emit 13605 CB blocks to avoid a hang. So, just set the VGT_MAX_INDX to a reasonably large value 13606 in the default state and don't touch them when drawing. When we run out of VB space, 13607 just draw the current buffer, grab a new one, and continue. 13608 13609commit cf85d4a1d43a3209c7ca9307aede2c2c243f7130 13610Author: Alex Deucher <alexdeucher@gmail.com> 13611Date: Wed Feb 18 19:02:56 2009 -0500 13612 13613 R6xx/R7xx: reset 3D state after VT switch 13614 13615commit 6545f0a2967414fa518a78440060b8b69c0146ee 13616Author: Alex Deucher <alexdeucher@gmail.com> 13617Date: Wed Feb 18 18:15:18 2009 -0500 13618 13619 R6xx/R7xx: fixup accel paths 13620 13621commit 61311b1de019c5a2344928a47347a966c229b08a 13622Author: Alex Deucher <alexdeucher@gmail.com> 13623Date: Wed Feb 18 17:27:05 2009 -0500 13624 13625 Bump version post release 13626 13627commit 0d3d1f1f7b450dabd3ad7d2df26ad1c72709b29b 13628Author: Alex Deucher <alexdeucher@gmail.com> 13629Date: Wed Feb 18 17:24:57 2009 -0500 13630 13631 bump for release 13632 13633commit 76a32f8391317513538dafee49cbb61c0d756356 13634Author: Alex Deucher <alexdeucher@gmail.com> 13635Date: Wed Feb 18 16:31:44 2009 -0500 13636 13637 R6xx: Connector quirk for asus board 13638 13639 bug 19943 13640 13641commit 97b8482dba4e99088b59d3ab1396be98a70e823e 13642Author: Alex Deucher <alexdeucher@gmail.com> 13643Date: Wed Feb 18 13:27:15 2009 -0500 13644 13645 Set default RMX type to FULL on LVDS 13646 13647commit 7d22b1799b34010d34c3600d6cb02ffe839a0780 13648Author: Alex Deucher <alexdeucher@gmail.com> 13649Date: Wed Feb 18 12:55:40 2009 -0500 13650 13651 DCE3.2+: allow output cloning 13652 13653commit 1a237a40958c006c56b80850bd77b2ac6c17e030 13654Author: Alex Deucher <alexdeucher@gmail.com> 13655Date: Wed Feb 18 11:46:20 2009 -0500 13656 13657 ATOM: reset crtc initialized flag on CloseScreen() 13658 13659 additional fix needed for bug 16781 13660 13661commit 9a108f0a0b7203458673ce6221e747a166d39617 13662Author: Alex Deucher <alexdeucher@gmail.com> 13663Date: Tue Feb 17 19:52:27 2009 -0500 13664 13665 ATOM: don't unblank uninitialized crtcs 13666 13667 If the crtc timing isn't setup, you might get stuck in a loop 13668 in the BlankCRTC command table 13669 13670 fixes bug 16781 13671 13672commit 3dff20e276615e8b77177689a4a5f8d91b3e8eac 13673Author: Alex Deucher <alexdeucher@gmail.com> 13674Date: Tue Feb 17 13:14:38 2009 -0500 13675 13676 R6xx/R7xx EXA: switch to surface sync packet 13677 13678commit 31bd6d28dd35021e0e706d4f094989deb856c26a 13679Author: Alex Deucher <alexdeucher@gmail.com> 13680Date: Tue Feb 17 11:24:02 2009 -0500 13681 13682 Fix crtc routing on pre-DCE3.2 systems 13683 13684 This should fix peterz and bug 20074 13685 13686commit 3c9e00c7f2ed494976713e8f77b729e0d76409db 13687Author: Alex Deucher <alexdeucher@gmail.com> 13688Date: Mon Feb 16 16:18:42 2009 -0500 13689 13690 RV350: add another AGP quirk 13691 13692commit 036a17bec5dabbb91640b907811da060dac154d6 13693Author: Alex Deucher <alexdeucher@gmail.com> 13694Date: Mon Feb 16 13:10:45 2009 -0500 13695 13696 RV280: add another agp quirk 13697 13698commit 00ac9d2f8d58725d0d3e6b1bf7f728d9ae7e6109 13699Author: David Miller <davem@davemloft.net> 13700Date: Sun Feb 15 18:26:14 2009 +1000 13701 13702 GART: Save/restore GART table consistently. 13703 13704 Always save the GART table with the RADEON_SURFACE_CNTL register 13705 zero'd out to make sure we always use the same endinanness. 13706 13707 fixed up as per Michel's suggestion for endianness. 13708 13709 Signed-off-by: David S. Miller <davem@davemloft.net> 13710 Signed-off-by: Dave Airlie <airlied@linux.ie> 13711 13712commit 2de90c8349409eb892edc57fd24b61b679eb7719 13713Author: David Miller <davem@davemloft.net> 13714Date: Sun Feb 15 18:24:01 2009 +1000 13715 13716 DRI: Fix page size used in RADEONDRIGetPciAperTableSize(). 13717 13718 The ATI GART has a fixed size of 4096 bytes. So using 13719 the system page size here is wrong. 13720 13721 Signed-off-by: David S. Miller <davem@davemloft.net> 13722 Signed-off-by: Dave Airlie <airlied@linux.ie> 13723 13724commit 7cde00f49649e25fd5816927c7a5e28b608fabcd 13725Author: Alex Deucher <alexdeucher@gmail.com> 13726Date: Thu Feb 12 17:39:44 2009 -0500 13727 13728 R6xx/R7xx: switch to drm for wait for idle 13729 13730 THIS REQUIRES AN UPDATED DRM 13731 13732commit e22cd4011b9be437ba89bff568e7fb82b4907d99 13733Author: Yang Zhao <yang@yangman.ca> 13734Date: Thu Feb 12 14:46:53 2009 -0500 13735 13736 R6xx/R7xx EXA: Further optimizations to overlapping copy 13737 13738 Diagonal overlapping copies can be reduced to either horizontal- or 13739 vertical-only offset, and the one with fewer copies is picked. 13740 13741commit da08b760bcf3d04d775c4440fafec10657bb1863 13742Author: Alex Deucher <alexdeucher@gmail.com> 13743Date: Thu Feb 12 13:53:11 2009 -0500 13744 13745 R6xx/R7xx EXA: handle running out of vertex space in the copy path 13746 13747commit e3be312b0b73982c24f1f5d9cf76d7caafae0853 13748Author: Alex Deucher <alexdeucher@gmail.com> 13749Date: Thu Feb 12 13:48:36 2009 -0500 13750 13751 R6xx/R7xx EXA: properly handle non repeat cases in the texture setup 13752 13753commit 3ff1eb7e50fef22adb6280dd7de77c3ecafdb451 13754Author: Michel Dänzer <daenzer@vmware.com> 13755Date: Thu Feb 12 19:24:41 2009 +0100 13756 13757 EXA: If making a pixmap offscreen fails, return ~0ULL as texture offset. 13758 13759 This allows AIGLX to fall back to the non-zero-copy code rather than messing up 13760 the 3D driver. 13761 13762commit c19aa4fc8da7ac4745624098b146bcc42c0436dd 13763Author: Michel Dänzer <daenzer@vmware.com> 13764Date: Thu Feb 12 19:22:12 2009 +0100 13765 13766 EXA: The source tiling code can't handle RepeatReflect yet. 13767 13768commit adc35636aa1e936ec74cba8931ccffe2b5c11656 13769Author: Alex Deucher <alexdeucher@gmail.com> 13770Date: Thu Feb 12 13:10:53 2009 -0500 13771 13772 R1xx/R2xx EXA: fix non repeat texture setup 13773 13774commit c6e1bf75c399807716b99486ab3301aeba60fa79 13775Author: Alex Deucher <alexdeucher@gmail.com> 13776Date: Thu Feb 12 12:49:24 2009 -0500 13777 13778 R3xx-R5xx EXA: fix texture setup for non-repeat case 13779 13780 fixes bug 19923 13781 13782commit 0804dde9012cf33f614a1c2ee6d5d0def040a9af 13783Author: Alex Deucher <alexdeucher@gmail.com> 13784Date: Thu Feb 12 12:02:57 2009 -0500 13785 13786 AVIVO: fix dualhead/rotation for real 13787 13788commit 3e5ac32f747bef70903e9cdf41652a827eece707 13789Author: Alex Deucher <alexdeucher@gmail.com> 13790Date: Wed Feb 11 17:40:20 2009 -0500 13791 13792 Revert "R6xx/R7xx: handle running out of vertex buffer space" 13793 13794 This reverts commit 4fd7228de7ad88edf825dbc3039df877795a9479. 13795 13796 Causes lockups. 13797 13798commit bd141aa73a77f68301715fb3b5664e2082327a80 13799Author: Alex Deucher <alexdeucher@gmail.com> 13800Date: Wed Feb 11 15:31:47 2009 -0500 13801 13802 R6xx/R7xx Xv: switch packed over to Yang's new shader code 13803 13804commit cb4a1ceaa32d6847b146a31291772e1183972ee7 13805Author: Yang Zhao <yang@yangman.ca> 13806Date: Wed Feb 11 15:18:00 2009 -0500 13807 13808 R6xx/R7xx Xv: Planar - Properly scale Y'CbCr values before converting to RGB 13809 13810 According to MPEG-2 spec, Y' and Cb/Cr values are scaled to [16, 235] 13811 and [16, 240], respectively, when packed into bytes. Properly take care 13812 of the reverse scaling before translating to RGB. 13813 13814 Conversion matrix has been simplified to remove 3rd column, as the fitting 13815 to [-0.5, 0.5] can be done with scaling. 13816 13817 Redundant MOV instructions were also removed, and now only 3 GPRs are required. 13818 13819commit 4fd7228de7ad88edf825dbc3039df877795a9479 13820Author: Alex Deucher <alexdeucher@gmail.com> 13821Date: Wed Feb 11 14:46:02 2009 -0500 13822 13823 R6xx/R7xx: handle running out of vertex buffer space 13824 13825 draw current VB and start a new one. 13826 13827commit ec60ef094762901ede4df5cde55c3f162e8c667b 13828Author: Alex Deucher <alexdeucher@gmail.com> 13829Date: Wed Feb 11 13:05:27 2009 -0500 13830 13831 R6xx/R7xx: be more verbose about what function ran out of VB space 13832 13833commit e7dedbc355970407eddb38370b58a8c96d204c1d 13834Author: Alex Deucher <alexdeucher@gmail.com> 13835Date: Wed Feb 11 12:45:14 2009 -0500 13836 13837 R6xx/R7xx: Move engine idle to sync functions 13838 13839commit 4a759a907ecd571460c7ff64cd6288380359b04b 13840Author: Alex Deucher <alexdeucher@gmail.com> 13841Date: Wed Feb 11 12:29:50 2009 -0500 13842 13843 adjust alignment 13844 13845commit d98359b20e1eb8eae893581ab199c299c8352dc4 13846Author: Alex Deucher <alexdeucher@gmail.com> 13847Date: Wed Feb 11 12:03:56 2009 -0500 13848 13849 Revert "Radeon EXA: wait for the engine to be idle before sw access" 13850 13851 This reverts commit 0c4694c58f2ec152b19023626bc7ced891d9da65. 13852 13853 This patch just adds latency and thus isn't a real solution 13854 13855commit d6c50b221cbab6e726948e2310c1def9fc38da64 13856Author: Alex Deucher <alexdeucher@gmail.com> 13857Date: Wed Feb 11 11:05:02 2009 -0500 13858 13859 R6xx/R7xx Xv: add support for packed uploads 13860 13861commit 8e437e996cc3f2c424c342701f4aa6bcf72ad08e 13862Author: Alex Deucher <alexdeucher@gmail.com> 13863Date: Wed Feb 11 10:53:50 2009 -0500 13864 13865 R6xx/R7xx Xv: Add native support for packed formats 13866 13867commit 0c4694c58f2ec152b19023626bc7ced891d9da65 13868Author: Alex Deucher <alexdeucher@gmail.com> 13869Date: Tue Feb 10 17:08:37 2009 -0500 13870 13871 Radeon EXA: wait for the engine to be idle before sw access 13872 13873 Really we need proper fencing with timestamps but this should 13874 help in the short term. 13875 13876 This should help with minor corruption issues and possibly 13877 lockups. 13878 13879commit 13ab61d90bb58112bef4ce7267ef127935315f80 13880Author: Fabio <fabio.ped@libero.it> 13881Date: Tue Feb 10 10:50:30 2009 -0500 13882 13883 man page updates 13884 13885 bug 20030 13886 13887commit 615f2f202cee5f0dd2c6b29329f843bb1c9c70d5 13888Author: Alex Deucher <alexdeucher@gmail.com> 13889Date: Tue Feb 10 10:17:47 2009 -0500 13890 13891 Bump version post RC release 13892 13893commit 1b4afc1c9f8458f9ab3434418f4ccf959f532ac1 13894Author: Alex Deucher <alexdeucher@gmail.com> 13895Date: Mon Feb 9 19:50:37 2009 -0500 13896 13897 R6xx/R7xx: Add checks to make sure we don't overrun VB space 13898 13899commit ff6efe513e0c3148812d6a9578d2265522417053 13900Author: Alex Deucher <alexdeucher@gmail.com> 13901Date: Mon Feb 9 17:43:24 2009 -0500 13902 13903 ATOM: warning fixes 13904 13905commit 87ad9006f3dafb45e131c9a3d6df713321f39998 13906Author: Alex Deucher <alexdeucher@gmail.com> 13907Date: Mon Feb 9 17:40:41 2009 -0500 13908 13909 RV350: add AGP quirk 13910 13911 bug 19981 13912 13913commit 610b8b65fd7aa97f9c03b2ff6ae607aa959eab0b 13914Author: Alex Deucher <alexdeucher@gmail.com> 13915Date: Mon Feb 9 16:33:20 2009 -0500 13916 13917 Bump for rc release 13918 13919commit e85b7a1c5948b5123eddf9dfbb9e002d74fbe9fb 13920Author: Alex Deucher <alexdeucher@gmail.com> 13921Date: Mon Feb 9 14:13:16 2009 -0500 13922 13923 R6xx/R7xx Xv: fix cache flush buffer size for planar 13924 13925commit 2a893bac1faffd28dce6d9a9693108196c71e6d3 13926Author: Alex Deucher <alexdeucher@gmail.com> 13927Date: Mon Feb 9 13:18:17 2009 -0500 13928 13929 R6xx/R7xx Xv: add accelerated uploads for planar formats 13930 13931commit 231aee18a73805be2f6c962e94a8345dd89fd0df 13932Author: Alex Deucher <alexdeucher@gmail.com> 13933Date: Mon Feb 9 13:02:27 2009 -0500 13934 13935 R6xx/R7xx Xv: implement native shader for planar formats 13936 13937commit 6c76bfe8105e3cf4e7e6ea1bfe1235be2079110f 13938Author: Alex Deucher <alexdeucher@gmail.com> 13939Date: Mon Feb 9 09:47:22 2009 -0500 13940 13941 R6xx/R7xx UTS: move actual upload to separate function 13942 13943 So it can be shared with Xv 13944 13945commit 04859362ce64ed4ae48305a83c92059899c1db4b 13946Author: Alex Deucher <alexdeucher@gmail.com> 13947Date: Sun Feb 8 20:13:06 2009 -0500 13948 13949 Add some missing r6xx/r7xx pci ids 13950 13951commit 132e4c575dc4675f4995e45f08c53c26bffd999a 13952Author: Yang Zhao <yang@yangman.ca> 13953Date: Sat Feb 7 13:54:51 2009 -0500 13954 13955 R6xx/R7xx EXA: Optimize overlapping copy 13956 13957 Overlapping copy is now done in chunks proportional to the 13958 non-overlapping area. 13959 13960 Diagonal overlaps are also handled properly. 13961 13962commit 8e9ef8ff581892cbe1b7ea56d48b9a1abd70179d 13963Author: Pierre Ossman <pierre@ossman.eu> 13964Date: Sat Feb 7 18:57:47 2009 +0100 13965 13966 Xv vsync support on r6xx/r7xx cards. 13967 13968commit 2222f0fd700f100b2e91fac2babe7d1b53f56c3e 13969Author: Pierre Ossman <pierre@ossman.eu> 13970Date: Sat Feb 7 18:56:42 2009 +0100 13971 13972 Fix bad range adjustment in VLINE code. 13973 13974commit 8ce6c024e31f6a3f5ae6c882738b1e64ae2944a1 13975Author: Alex Deucher <alexdeucher@gmail.com> 13976Date: Sat Feb 7 10:32:01 2009 -0500 13977 13978 R6xx/R7xx Xv: fix typos in cache flushing commands 13979 13980commit bba5d806cc46328ad9e4c521bfeeb2a59f96862b 13981Author: Alex Deucher <alexdeucher@gmail.com> 13982Date: Fri Feb 6 19:44:17 2009 -0500 13983 13984 AVIVO: better fix for rotation 13985 13986 This should handle initial rotation as well. 13987 13988commit 28cf3492f431f325cb08d7c6aade43d8886df3f6 13989Author: Alex Deucher <alexdeucher@gmail.com> 13990Date: Fri Feb 6 19:20:18 2009 -0500 13991 13992 AVIVO: fix rotation 13993 13994 When rotation is active, virtualX and virtualY change. 13995 13996commit 1b45936ae614244aa49b1a5d3c7fc39773c4f9b6 13997Author: Alex Deucher <alexdeucher@gmail.com> 13998Date: Fri Feb 6 18:29:55 2009 -0500 13999 14000 R6xx/R7xx EXA: Fix typo in DFS 14001 14002 noticed by pzad in IRC 14003 14004commit c06d89e16d5b2553142e8641e66080e1770c1563 14005Author: Alex Deucher <alexdeucher@gmail.com> 14006Date: Fri Feb 6 18:07:22 2009 -0500 14007 14008 R6xx/R7xx EXA: fallback on overlapping blits for now 14009 14010 Leave this disabled until we get a proper solution. 14011 14012commit 1d5fc3febf3470b94c423a1eda5e0683856909df 14013Author: Alex Deucher <alexdeucher@gmail.com> 14014Date: Fri Feb 6 14:43:30 2009 -0500 14015 14016 Revert "r6xx/r7xx EXA: Optimize overlapping copy" 14017 14018 This reverts commit 0dfadc1843e0d14b9cc1ee19a72f4fd60a2c495b. 14019 14020 This doesn't always work properly. 14021 14022commit def317e22d072405cd95ddb19d17eacd784ffd9e 14023Author: Alex Deucher <alexdeucher@gmail.com> 14024Date: Fri Feb 6 11:04:59 2009 -0500 14025 14026 R6xx/R7xx EXA: add accelerated UTS/DFS hooks 14027 14028 I'm not sure how much of a win these are. I need to 14029 do some benchmarking. 14030 14031commit 0dfadc1843e0d14b9cc1ee19a72f4fd60a2c495b 14032Author: Yang Zhao <yang@yangman.ca> 14033Date: Fri Feb 6 10:29:39 2009 -0500 14034 14035 r6xx/r7xx EXA: Optimize overlapping copy 14036 14037 When source and destination blocks are only offset horizontally, it 14038 appears to be unnecessary to perform careful, segment-by-segment copy. 14039 The code path that does this is taken out completely. 14040 14041 For the case where offset is only vertical, copying is now done by 14042 height of the non-overlapping area each time, instead of always 14043 line-by-line. 14044 14045commit 3d17bd199423e92d201c20f047d5e699942af976 14046Author: Alex Deucher <alexdeucher@gmail.com> 14047Date: Thu Feb 5 19:25:03 2009 -0500 14048 14049 Revert "R6xx/R7xx EXA: improve overlapping copy performance" 14050 14051 This reverts commit b24827c9d211e8a35da53b665385a7733d19910d. 14052 14053 This seems to cause corruption in some cases. 14054 14055commit bf74055f543e7f0664741620fb1fe827ebc12711 14056Author: Alex Deucher <alexdeucher@gmail.com> 14057Date: Thu Feb 5 17:31:43 2009 -0500 14058 14059 r6xx/r7xx EXA: fix corruption when doing sw access 14060 14061 need to wait until the engine is idle. Ideally we wait 14062 on a timestamp shadowed in memory, but polling the 14063 GRBM_STATUS reg will do for now. 14064 14065commit b24827c9d211e8a35da53b665385a7733d19910d 14066Author: Alex Deucher <alexdeucher@gmail.com> 14067Date: Thu Feb 5 16:21:37 2009 -0500 14068 14069 R6xx/R7xx EXA: improve overlapping copy performance 14070 14071 send vertices for each line of the copy, but only draw once 14072 14073commit 729fe756f809a41494dc161252d82313956e4a4b 14074Author: Alex Deucher <alexdeucher@gmail.com> 14075Date: Thu Feb 5 16:08:35 2009 -0500 14076 14077 r6xx/r7xx EXA: cleanup overlapping copy 14078 14079commit 16e01a5796a8e8b86ad25d3aa45b9e7044dc72d1 14080Author: Alex Deucher <alexdeucher@gmail.com> 14081Date: Tue Feb 3 19:46:10 2009 -0500 14082 14083 Fix encoder accounting 14084 14085 Should fix bug 19924 14086 14087commit d7bf7b9c17add31dc9a115b56a563c7f9bf2cdb6 14088Author: Alex Deucher <alexdeucher@gmail.com> 14089Date: Tue Feb 3 18:28:38 2009 -0500 14090 14091 EXA: fix and re-enable Solid() on R7xx 14092 14093 missing last bit in alu clause in solid PS 14094 14095commit 9abb09d5f43320bd27b336612d39d264fe93549f 14096Author: Alex Deucher <alexdeucher@gmail.com> 14097Date: Tue Feb 3 11:17:13 2009 -0500 14098 14099 Fix bad rv710 pci id 14100 14101commit e5b916770946a9eebcb4bd1e6f698220db8c718a 14102Author: Alex Deucher <alexdeucher@gmail.com> 14103Date: Tue Feb 3 10:48:41 2009 -0500 14104 14105 Allow rotation on r6xx/r7xx 14106 14107commit d1f071c7f1dad6babfbcfcc2cb2b722a4987f372 14108Author: Alex Deucher <alexdeucher@gmail.com> 14109Date: Tue Feb 3 10:44:10 2009 -0500 14110 14111 Initial R6xx/R7xx EXA and textured video support 14112 14113commit c88c3ef6f3db266c1aacba5297b8dfc8b66bf00e 14114Author: Wolke Liu <wolke.liu@amd.com> 14115Date: Mon Feb 2 17:01:34 2009 -0500 14116 14117 AVIVO: Save/restore vga pll registers 14118 14119 This fixes some VT switch issues on some chips 14120 14121commit 6fac3cefd1f46161c1e276ba40e72da2823aa9f6 14122Author: Alex Deucher <alexdeucher@gmail.com> 14123Date: Sat Jan 31 14:21:02 2009 -0500 14124 14125 Return NULL for encoder if no active device is assigned 14126 14127 fixes bug 19855 14128 14129commit db178c852d00e9b85513d750b5f8f7495f9e3360 14130Author: Alan Coopersmith <alan.coopersmith@sun.com> 14131Date: Fri Jan 30 20:37:41 2009 -0800 14132 14133 Add README with pointers to mailing list, bugzilla & git repos 14134 14135commit fa8e5a4fc236f8f15f462cb0d6164b194a65a118 14136Author: Michel Dänzer <daenzer@vmware.com> 14137Date: Fri Jan 30 16:16:26 2009 +0100 14138 14139 EXA: Accelerate Composite of RepeatPad/Reflect pictures when possible. 14140 14141commit 3a6e958c59b4b44fe620c8abe760ed1abf0007b1 14142Author: airlied <airlied@roo.(none)> 14143Date: Fri Jan 30 07:56:14 2009 +1000 14144 14145 rs780: include RS780 in the InitMemory to leave alone 14146 14147 This stops the MC_FB_LOCATION getting reassigned to 0, which 14148 is really bad thing to happen. 14149 14150 I've had reported memory corruption on these cards so hopefully this 14151 fixes it. 14152 14153commit bd8021d46e9066e4cd116c03a7b7adcfe2557aff 14154Merge: 4f88dd8 20d5dd3 14155Author: Alex Deucher <alexdeucher@gmail.com> 14156Date: Wed Jan 28 14:33:14 2009 -0500 14157 14158 Merge branch 'atom-rework' and fix conflicts 14159 14160commit 20d5dd387da555e895e2b73fb53e2b026dd91003 14161Author: Alex Deucher <alexdeucher@gmail.com> 14162Date: Wed Jan 28 14:12:26 2009 -0500 14163 14164 Move encoder specific data to encoder dev_priv 14165 14166commit 4f88dd83f290fb38e41256a7f1804008c0f28139 14167Merge: 1c5678f 2ca970c 14168Author: Alex Deucher <alexdeucher@gmail.com> 14169Date: Wed Jan 28 13:13:13 2009 -0500 14170 14171 Merge branch 'atom-tvout' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 14172 14173 Conflicts: 14174 14175 src/radeon_atombios.c 14176 14177commit 1a62360e8260eaf5f98fa6be3f89f64c8cae7d32 14178Author: Alex Deucher <alexdeucher@gmail.com> 14179Date: Mon Jan 26 10:11:28 2009 -0500 14180 14181 Encoders not assigned yet, use supported devices 14182 14183 need to move this to encoder setup 14184 14185commit 44d9ceacb1fa6e1859d5d8bd1c55405aea24ad85 14186Author: Alex Deucher <alexdeucher@gmail.com> 14187Date: Mon Jan 26 03:47:32 2009 -0500 14188 14189 Fix legacy output setup 14190 14191 don't have assigned encoders at this point. Need to 14192 eventually move this stuff to encoder dev_priv. 14193 14194commit 0c24cbcea56c8f36374aca0f8ba849adfbf0ecfe 14195Author: Alex Deucher <alexdeucher@gmail.com> 14196Date: Mon Jan 26 03:24:19 2009 -0500 14197 14198 ATOM: print useful output info for DPMS events 14199 14200commit 2f92b885457c5cee2098ebbf8c6fe5a5eb98ec48 14201Author: Alex Deucher <alexdeucher@gmail.com> 14202Date: Mon Jan 26 03:12:44 2009 -0500 14203 14204 Warning fixes 14205 14206commit 4cbf6189f3a7945a1141b0451e4f7881ceaaadcd 14207Author: Alex Deucher <alexdeucher@gmail.com> 14208Date: Mon Jan 26 02:59:51 2009 -0500 14209 14210 Move legacy output setup functions to legacy_output.c 14211 14212commit 93ed767255b60fbdf5b416b9bd06c366036a5141 14213Author: Alex Deucher <alexdeucher@gmail.com> 14214Date: Mon Jan 26 02:40:41 2009 -0500 14215 14216 Fix off by one when printing encoder name 14217 14218 - also cleanup some debugging output 14219 14220commit b515eee4b1cafae78ed39cd91c18b4f5e671b20b 14221Author: Alex Deucher <alexdeucher@gmail.com> 14222Date: Mon Jan 26 01:50:18 2009 -0500 14223 14224 Additional output cleanup 14225 14226commit 1c5678fc452dbc93135e9ece50a4cff6583f6078 14227Author: Michel Dänzer <daenzer@vmware.com> 14228Date: Sat Jan 24 11:10:00 2009 +0100 14229 14230 Drop memcpy fallbacks from EXA UploadToScreen and DownloadFromScreen hooks. 14231 14232 While in theory it's possible for the PrepareAccess hook to fail on big endian 14233 platforms, in practice it isn't at this point because there's no other users of 14234 the surface registers. 14235 14236commit 5cbd6082b9f8d7009ee9ef5e036673f99e1bfb05 14237Author: Michel Dänzer <daenzer@vmware.com> 14238Date: Sat Jan 24 11:07:12 2009 +0100 14239 14240 Don't transform EXA Composite mask coordinates when there's no mask. 14241 14242commit 7e54c4b8acafead8de5d0dd6fdc1b619f832dd6f 14243Author: Thomas Jaeger <ThJaeger@gmail.com> 14244Date: Fri Jan 23 20:03:27 2009 -0500 14245 14246 Fall back to software for unsupported repeat modes 14247 14248commit 289b4a377cfff0c2a8ec938951cfbae1e270570e 14249Author: Alex Deucher <alexdeucher@gmail.com> 14250Date: Fri Jan 23 20:00:43 2009 -0500 14251 14252 Remove OutputType and other cruft 14253 14254commit 5f98508effbdc1fd62d0df15798722c147d88da1 14255Author: Maciej Cencora <m.cencora@gmail.com> 14256Date: Fri Jan 23 19:33:25 2009 -0500 14257 14258 Make sure gb_num_pipes is initialized when DRI is disabled 14259 14260 Found with valgrind 14261 14262commit 0faab02cc732414af3cfb2014d3ddc1139aff063 14263Author: Alex Deucher <alexdeucher@gmail.com> 14264Date: Fri Jan 23 17:35:01 2009 -0500 14265 14266 Remove some unused cruft 14267 14268commit 025082160c53c01a44f4d681cc63bf26ac4561d5 14269Author: Alex Deucher <alexdeucher@gmail.com> 14270Date: Fri Jan 23 17:31:18 2009 -0500 14271 14272 track encoder state 14273 14274commit 82f12e5a40c1fbcb91910a0f8b725c34fff02aae 14275Author: Alex Deucher <alexdeucher@gmail.com> 14276Date: Fri Jan 23 16:21:33 2009 -0500 14277 14278 Remove TMDSType, DACType, LVDSType from output rec 14279 14280commit 7b42b57d748f77ff6ef6f8ade7cbc983a98d7204 14281Author: Alex Deucher <alexdeucher@gmail.com> 14282Date: Wed Jan 21 16:28:00 2009 -0500 14283 14284 few more logic pasto's bits I missed 14285 14286commit fef1c12fbe4ceb4676258850d1fa5cac994f7c21 14287Author: Alex Deucher <alexdeucher@gmail.com> 14288Date: Wed Jan 21 15:34:01 2009 -0500 14289 14290 Fix compilation with RADEON_TRACE_FALL set 14291 14292 noticed by curaga on phoronix forum 14293 14294commit 1305376b009d17d65b4935896dc35cb7530f7c52 14295Author: Alex Deucher <alexdeucher@gmail.com> 14296Date: Wed Jan 21 15:20:47 2009 -0500 14297 14298 Move active_device setup to detect() 14299 14300 Also add encoder debugging printouts 14301 14302commit 9e0cc0ca5d7bc68756ae92ee3b5b2392366be71a 14303Author: Alex Deucher <alexdeucher@gmail.com> 14304Date: Wed Jan 21 14:47:08 2009 -0500 14305 14306 Fix logic cut and paste error 14307 14308commit 12010ae9296635ffa5f42ce0cd14482aaa290cf7 14309Author: Alex Deucher <alexdeucher@gmail.com> 14310Date: Wed Jan 21 13:52:11 2009 -0500 14311 14312 More legacy rework 14313 14314commit 3f8bbede8bf639050f773780888b89cbdcb425b8 14315Author: Alex Deucher <alexdeucher@gmail.com> 14316Date: Wed Jan 21 13:45:11 2009 -0500 14317 14318 fix legacy crtc routing and add some debugging info 14319 14320commit 80e0162ea2c1f793964731268251aba2d3243127 14321Author: Alex Deucher <alexdeucher@gmail.com> 14322Date: Wed Jan 21 13:28:14 2009 -0500 14323 14324 ATOM: fix encoder init 14325 14326commit aea27d9caabb0e54e868a21e6e547733a4ca2709 14327Author: Alex Deucher <alexdeucher@gmail.com> 14328Date: Wed Jan 21 12:33:49 2009 -0500 14329 14330 Switch legacy output code to use new encoder objects 14331 14332commit 9b44a4b57b00b150d041a5d82cb2df42159c062b 14333Author: Alex Deucher <alexdeucher@gmail.com> 14334Date: Wed Jan 21 11:32:24 2009 -0500 14335 14336 ATOM: more output cleanup 14337 14338commit 9e283fa9de34407fa6f0d5a8006a611b104f4bf2 14339Author: Alex Deucher <alexdeucher@gmail.com> 14340Date: Wed Jan 21 03:16:20 2009 -0500 14341 14342 Fixup encoder setup on pre-ATOM chips 14343 14344commit 5850abbdcd22b70bdbccd3642f8ba1a27e38e7cd 14345Author: Alex Deucher <alexdeucher@gmail.com> 14346Date: Wed Jan 21 02:28:47 2009 -0500 14347 14348 clean up encoder setup 14349 14350 - make sure not to duplicate encoders 14351 - use single encoder array 14352 14353commit 152d1d4c179c262be4d0a5618400aa13b4820342 14354Author: Alex Deucher <alexdeucher@gmail.com> 14355Date: Tue Jan 20 00:58:18 2009 -0500 14356 14357 First pass at converting legacy code to encoder objects 14358 14359commit 7cc7ba8bf9ce2f0eebfd9b48aa648c416c333ff4 14360Author: Alex Deucher <alexdeucher@gmail.com> 14361Date: Mon Jan 19 18:42:12 2009 -0500 14362 14363 ATOM: round 1 of output rework 14364 14365commit 25f3878858e6efb486888e5271b60d4ba08c73c6 14366Author: Alex Deucher <alexdeucher@gmail.com> 14367Date: Mon Jan 19 15:54:42 2009 -0500 14368 14369 start to re-org outputs 14370 14371commit cbeb310d35e5ea25ba7a45a2c555468391cdf29e 14372Author: Alex Deucher <alexdeucher@gmail.com> 14373Date: Fri Jan 16 19:53:49 2009 -0500 14374 14375 ATOM: switch to define for external tmds 14376 14377commit d0d157abced43f822885fa26f14a4924766594cd 14378Author: Alex Deucher <alexdeucher@gmail.com> 14379Date: Fri Jan 16 19:51:07 2009 -0500 14380 14381 ATOM: combine DAC setup functions 14382 14383 plus a few other clean-ups 14384 14385commit f6f969338f8187529c7ae9b2adc980b71a3f5837 14386Author: Alex Deucher <alexdeucher@gmail.com> 14387Date: Fri Jan 16 18:10:34 2009 -0500 14388 14389 DCE30: LVTMA requires DIG2 encoder 14390 14391 Fixes bug 19544 14392 14393commit 549ae63aecc2624a69b41ce4db75883d03d59c23 14394Author: Alex Deucher <alexdeucher@gmail.com> 14395Date: Thu Jan 15 19:26:40 2009 -0500 14396 14397 RV280 Add another AGP quirk 14398 14399commit 0e69068a8ed82e3fb7349d8224933451955e5f16 14400Author: Tormod Volden <debian.tormod@gmail.com> 14401Date: Tue Jan 13 20:31:17 2009 -0500 14402 14403 [PATCH] Add yet another AGP quirk for RV280 14404 14405commit b437da34cb4f7c637674fb432ad81b9bf32244a3 14406Author: Christiaan van Dijk <dvbmail@xs4all.nl> 14407Date: Mon Jan 12 11:54:15 2009 -0500 14408 14409 R3xx/R4xx: Maximize the use of clipped triangles for Xv rendering 14410 14411 While trying to get tear free video on my RS690 I found some problems in the code for rendering the 14412 video as a clipped triangle. The size limitations are quite strict and scaling the video output window 14413 creates all kinds of weird effects. This patch maximizes the use of this function and resolves the 14414 scaling issues. There could be a problem with zero target height/width, not sure if this is handled 14415 elsewhere. 14416 14417commit 3d106b990fa8daba96e590c57b164640a2faae7f 14418Author: Alan Coopersmith <alan.coopersmith@sun.com> 14419Date: Fri Jan 9 16:26:04 2009 -0800 14420 14421 Remove xorgconfig & xorgcfg from See Also list in man page 14422 14423commit 14d29a223028c183ba11e717a3737285d0ac8de1 14424Author: Alex Deucher <alexdeucher@gmail.com> 14425Date: Wed Jan 7 17:04:42 2009 -0500 14426 14427 RV280: add another AGP quirk 14428 14429commit 1795706a6bb13272a58e8fa1bd5ad2603d553dde 14430Author: Nicos Gollan <gtdev@spearhead.de> 14431Date: Wed Jan 7 14:25:24 2009 -0500 14432 14433 [PATCH] Fixed enumerations in radeon-output.c 14434 14435commit f69ad5aa17bb907105c1657803628d52b5ff3f11 14436Author: Alex Deucher <alexdeucher@gmail.com> 14437Date: Wed Jan 7 11:33:53 2009 -0500 14438 14439 Bump version post release 14440 14441commit a0dd5d7ee3f038a9bfe051db8dbfac4934a81581 14442Author: Alex Deucher <alexdeucher@gmail.com> 14443Date: Tue Jan 6 19:08:28 2009 -0500 14444 14445 Bump version for release 14446 14447commit c3fb8bb279959512a4ced644a64cc660c5cd97bd 14448Author: Alex Deucher <alexdeucher@gmail.com> 14449Date: Tue Jan 6 17:41:22 2009 -0500 14450 14451 ATOM: rework encoder/transmitter setup 14452 14453commit 9bb6b8ceaf772e1ce63610812a12260ee3860c27 14454Author: Alex Deucher <alexdeucher@gmail.com> 14455Date: Tue Jan 6 16:25:26 2009 -0500 14456 14457 ATOM: refactor output dpms 14458 14459commit 467995d960bdefb1911eabfc2a047cf816013101 14460Author: Alex Deucher <alexdeucher@gmail.com> 14461Date: Tue Jan 6 15:18:41 2009 -0500 14462 14463 ATOM: Adjust PLL setup for recent atom changes 14464 14465commit ee6310174c44809c761dbdb423e56e61392a0e10 14466Author: Alex Deucher <alexdeucher@gmail.com> 14467Date: Tue Jan 6 14:59:22 2009 -0500 14468 14469 ATOM: handle cases where TMDS uses linkb 14470 14471 fixes bug 18564 14472 14473commit 690999b682c0f954e27d9f1b3a3d8db4ef865c21 14474Author: Alex Deucher <alexdeucher@gmail.com> 14475Date: Tue Jan 6 14:55:49 2009 -0500 14476 14477 ATOM: rework object table parsing 14478 14479commit e23baa2fa8b25b68cd473aec7a3d8cf3ee7b0ea8 14480Author: Alex Deucher <alexdeucher@gmail.com> 14481Date: Tue Jan 6 14:38:06 2009 -0500 14482 14483 R100: Fix typo in e55598cc1b35d398c9eb2c3946df71456775f621 14484 14485 Noticed by Andreas Stenglein 14486 14487commit d5cb24075928690377fd2dfaa32abfebff9990df 14488Author: Alex Deucher <alexdeucher@gmail.com> 14489Date: Sun Jan 4 21:04:52 2009 -0500 14490 14491 ATOM: add format rev 2 for digitial command tables 14492 14493 should fix bug 18911 14494 14495commit 1813456af3def38c6f53103054976f3092491746 14496Author: Alex Deucher <alexdeucher@gmail.com> 14497Date: Sun Jan 4 21:01:05 2009 -0500 14498 14499 ATOM: fail on more tables with unhandled revisions 14500 14501commit 90ce00a3dbf37833a71db234372695460c8f0342 14502Author: Alex Deucher <alexdeucher@gmail.com> 14503Date: Sun Jan 4 19:13:53 2009 -0500 14504 14505 AGP: Print both device and subsytem pci ids for agp 14506 14507 helpful for adding quirks with libpciaccess 14508 14509commit 006d736d890a6b2f1d7eb8ee20ccf467a939516f 14510Author: Alex Deucher <alexdeucher@gmail.com> 14511Date: Sun Jan 4 19:08:37 2009 -0500 14512 14513 Another AGP quirk 14514 14515 debian bug 510208 14516 14517commit c39ca66a7e216b7492fc5e3680ac535831cfb9a1 14518Author: Alex Deucher <alexdeucher@gmail.com> 14519Date: Tue Dec 30 19:05:58 2008 -0500 14520 14521 R2xx: add another AGP quirk 14522 14523 from http://www.phoronix.com/forums/showthread.php?t=14626 14524 14525commit 877be1b268f275b8ad83d5e245a0f0d3b1a05091 14526Author: Fabio <fabio.ped@libero.it> 14527Date: Mon Dec 29 18:11:10 2008 -0500 14528 14529 man page updates 14530 14531 bug 19329 14532 14533commit 837f7e8a48ea35d21de24b620830a2c6db854861 14534Author: Dave Airlie <airlied@linux.ie> 14535Date: Tue Dec 30 06:50:54 2008 +1000 14536 14537 atombios: endian fix in r600 object table handling 14538 14539 This fixes the r600 object table on r600. 14540 14541commit 45f8b911659dba7f43b9b70bde4b9c3be51b0b5b 14542Author: Alex Deucher <alexdeucher@gmail.com> 14543Date: Mon Dec 29 10:25:33 2008 -0500 14544 14545 RS600: remove gpio quirk for ddc 14546 14547 Doesn't seem necessary on RS600. I'll try and find out more. 14548 14549commit e913990030c536427bfecdb7b3d1e18f1beb90d2 14550Author: Alex Deucher <alexdeucher@gmail.com> 14551Date: Mon Dec 29 10:22:20 2008 -0500 14552 14553 Revert "radeon: maybe special handling only needed for DVI port not HDMI one." 14554 14555 This reverts commit c877450a75d3fd4a2f200af197c691669966f6f4. 14556 14557 This breaks RS690 setups. I think for now we just apply this quirk to 14558 RS690/RS740. I'll follow up with the bios folks about RS600. 14559 14560commit c877450a75d3fd4a2f200af197c691669966f6f4 14561Author: Dave Airlie <airlied@linux.ie> 14562Date: Mon Dec 29 09:05:45 2008 +1000 14563 14564 radeon: maybe special handling only needed for DVI port not HDMI one. 14565 14566 fixes ciaranm's rs600 board on radeon DDC 14567 14568commit 2a9bc6be3246ec230b181391221e3a46beb94a23 14569Author: Dave Airlie <airlied@linux.ie> 14570Date: Mon Dec 29 09:05:16 2008 +1000 14571 14572 radeon: a-bit motherboard has no DVI output 14573 14574 thanks to ciaranm for finding this. 14575 14576commit eccf3fed2be54f5c02fefef32e39c4084acbbe22 14577Author: Alex Deucher <alexdeucher@gmail.com> 14578Date: Fri Dec 26 19:51:58 2008 -0500 14579 14580 Make sure to include radeon_drm.h in the tarball 14581 14582commit e55598cc1b35d398c9eb2c3946df71456775f621 14583Author: Alex Deucher <alexdeucher@gmail.com> 14584Date: Fri Dec 26 19:31:55 2008 -0500 14585 14586 R2xx EXA: limit composite to 2047 like r1xx 14587 14588 See bug 19269 14589 14590commit 631123d144d088d4f77a0599c34deaa31d551d71 14591Author: Dave Airlie <airlied@redhat.com> 14592Date: Tue Dec 23 10:08:46 2008 +1000 14593 14594 radeon: update to 6.9.0.91 14595 14596 do a prerelease for end of year. 14597 14598commit 047f7603174f6047090f4fafbe6488d697eb78bc 14599Author: Dave Airlie <airlied@linux.ie> 14600Date: Tue Dec 23 08:16:25 2008 +1000 14601 14602 radeon: pass distcheck 14603 14604commit c0c33dab44e6966b1702d4e8cfba3537fc6e2d5c 14605Author: Patrick Haller <patrick.haller@haller-consult.com> 14606Date: Mon Dec 22 03:06:23 2008 -0500 14607 14608 Fix off by one in EXA composite limit checking 14609 14610 Patch from Patrick, with some updates from me: 14611 - fix r200 limits 14612 - note about r100 limits 14613 14614commit d01a609a2b07da4ca0f182e79459432584ec61f6 14615Author: Alex Deucher <alexdeucher@gmail.com> 14616Date: Mon Dec 22 02:18:43 2008 -0500 14617 14618 R3xx-R5xx: better fix for xv primitive tearing issues 14619 14620 R5xx: always use single clipped triangle 14621 R3xx/R4xx: use single clipped triangle up to guardband limit, 14622 then use quad. 14623 14624commit 4d009d14c9c72f6e1bc8100ec8d9b8037ee4e514 14625Merge: d52882d 4e96278 14626Author: Dave Airlie <airlied@redhat.com> 14627Date: Mon Dec 22 16:00:01 2008 +1000 14628 14629 Merge branch 'atom-tvout' 14630 14631commit 4e96278b581e296c1203d97a6d7aa3bff3977222 14632Author: Dave Airlie <airlied@redhat.com> 14633Date: Mon Dec 22 15:58:35 2008 +1000 14634 14635 atombios/tv: add an option to enable atom tv-out for users. 14636 14637 This code is still experimental but we will allow users to enable it for experimental reasons 14638 14639commit d52882d8188830dd52fa112dadcf5ea7f3e5fd5e 14640Author: Dave Airlie <airlied@redhat.com> 14641Date: Mon Dec 22 09:27:47 2008 +1000 14642 14643 radeon: setup 3D engine even when no DRI. 14644 14645 This should fix some missing font issues in EXA without DRI cases. 14646 14647commit 2346fd7cbd90dcdce2b361d374a53da064fa6a12 14648Author: Dave Airlie <airlied@redhat.com> 14649Date: Thu Dec 18 16:35:13 2008 +1000 14650 14651 radeon: add all new pci ids for rv730/rv710 families 14652 14653commit 2ca970c7aa1f3f212f03bf3287d6b60eeafd68ae 14654Author: Alex Deucher <alexdeucher@gmail.com> 14655Date: Wed Dec 17 13:07:19 2008 -0500 14656 14657 Allow arbitrary tv-out modes 14658 14659commit 5da147eec70abb7c9125a8940089fd3ff8bdea95 14660Author: Alex Deucher <alexdeucher@gmail.com> 14661Date: Wed Dec 17 05:37:58 2008 -0500 14662 14663 missed one in last commit 14664 14665commit 6bfbb930c549e67f68969f7f0483380417f755b5 14666Author: Alex Deucher <alexdeucher@gmail.com> 14667Date: Wed Dec 17 05:36:09 2008 -0500 14668 14669 Make sure we hit the right bios reg 14670 14671commit 6ed0e68060fd2775e2d4027ff1533729e87848ba 14672Author: Alex Deucher <alexdeucher@gmail.com> 14673Date: Wed Dec 17 04:54:50 2008 -0500 14674 14675 properly handle EnableYUV 14676 14677 A bit of a hack. we really need to restructure the 14678 bios scratch reg handling. 14679 14680commit f21e673aced1bb5138bbb80e6d9724530eb95ebf 14681Author: Alex Deucher <alexdeucher@gmail.com> 14682Date: Wed Dec 17 03:17:45 2008 -0500 14683 14684 Fix colors on tv-out 14685 14686 Call EnableYUV to set the transform and enable/disable it. 14687 should also fix green tint with tv connected at boot issues. 14688 14689commit b2b43905a5385a8bb0b59b8e50952863d8dacb59 14690Author: Alex Deucher <alexdeucher@gmail.com> 14691Date: Wed Dec 17 00:59:52 2008 -0500 14692 14693 Pre-avivo: fix FP setup 14694 14695 - make sure to clear various shadow timing bits 14696 - crtc1 select bit was set wrong. 14697 - should fix bugs like 19100 14698 14699commit 3754cc23abac99880eda87d06aa42e16081b6c8a 14700Author: Alex Deucher <alexdeucher@gmail.com> 14701Date: Wed Dec 17 00:56:19 2008 -0500 14702 14703 R2xx: switch EXA composite to rects rather quads 14704 14705 For conistency with Xv path and to reduce potential diagonal tearing 14706 14707commit be12cf8fcc7a5a3c3ee2c630474e2f9d9368f4d4 14708Author: Dave Airlie <airlied@redhat.com> 14709Date: Wed Dec 17 15:28:25 2008 +1000 14710 14711 radeon: r500 tv-out force scaler values to nice set that looks correct 14712 14713commit 01f7a50d76f706a456de7a4eb8323939d905c82b 14714Author: Dave Airlie <airlied@redhat.com> 14715Date: Wed Dec 17 13:43:21 2008 +1000 14716 14717 r500: re-enable TV out 14718 14719commit 45b9ff5e250d69de0df41f68b33524f7e2ce13b7 14720Author: Dave Airlie <airlied@redhat.com> 14721Date: Wed Dec 17 13:42:10 2008 +1000 14722 14723 radeon: r500 PAL timings are slightly incorrect 14724 14725 this stablises the PAL image at least 14726 14727commit 435d996949185201a97493e131f9920a32863acd 14728Author: Dave Airlie <airlied@redhat.com> 14729Date: Wed Dec 17 11:50:48 2008 +1000 14730 14731 radeon: tv timings have crev/frev wrong way around for some reason. 14732 14733commit b9549e89cbdf8fff4aa7bd9f3adfb7d5732508cc 14734Author: Dave Airlie <airlied@redhat.com> 14735Date: Wed Dec 17 11:40:53 2008 +1000 14736 14737 atombios: re-enable TV dpms 14738 14739commit db85ff2e0c52a030546c0cc8ed0b851418f3e045 14740Author: Dave Airlie <airlied@redhat.com> 14741Date: Wed Dec 17 11:37:17 2008 +1000 14742 14743 atom/r600: re-enable TV outputs 14744 14745commit ff7b52c9a48f2e417acf6bec3b72546ad9b33471 14746Author: Dave Airlie <airlied@redhat.com> 14747Date: Wed Dec 17 11:36:38 2008 +1000 14748 14749 radeon: use latest atombios constants for TV encoders in set_crtc_source 14750 14751commit 1e0c7d3f0bf6be67b976fdea8fa2df0b4b16435f 14752Author: Dave Airlie <airlied@redhat.com> 14753Date: Wed Dec 17 11:35:06 2008 +1000 14754 14755 atom: spc3 only requires ucEncoderMode for DIG outputs 14756 14757commit 105cfd0b8696347ef71ad9daa941abb308bd6c2c 14758Author: Dave Airlie <airlied@redhat.com> 14759Date: Wed Dec 17 11:31:15 2008 +1000 14760 14761 atom/tv: enable scaler can set the TV output scaler up. 14762 14763 Add the tv standard to the scaler enable for TV use. 14764 14765commit 2a768abbd2f18abe203571213260bccc225f70fd 14766Author: Dave Airlie <airlied@redhat.com> 14767Date: Wed Dec 17 11:29:07 2008 +1000 14768 14769 atom/tv: fixup the tv out default if bios doesn't specify 14770 14771commit b7fdd5da08bfec705691363b8890699c794dce53 14772Author: Dave Airlie <airlied@redhat.com> 14773Date: Wed Dec 17 11:24:56 2008 +1000 14774 14775 atombios: fixup parsing of TV tables on r500 and r600 cards. 14776 14777 The TV tables changed between for later cards. 14778 14779commit e3b08cb59716b525b3063b184eab70e32b9d66f0 14780Author: Dave Airlie <airlied@redhat.com> 14781Date: Tue Dec 16 15:43:16 2008 +1000 14782 14783 atombios: correct fix for previous issue. 14784 14785 I committed an old patch. 14786 14787commit 1431bc6900b0e0d7f486b5b3ba281873472b3fdc 14788Author: Dave Airlie <airlied@redhat.com> 14789Date: Tue Dec 16 15:31:34 2008 +1000 14790 14791 atombios: fix issue with Clearing of MM_INDEX 0. 14792 14793 While playing with tv-out I discovered this issue with clearing 14794 the MM_INDEX register. 14795 14796commit f72367aff2d09161d30066a00673ba291b292a36 14797Author: Dave Airlie <airlied@redhat.com> 14798Date: Tue Dec 16 12:32:24 2008 +1000 14799 14800 radeon: add initial rv730 pciids. 14801 14802 Lots more to come, this just adds a few to start. 14803 14804commit 4d2429f3ec4d7d3e1512b268c7ff5cbff8d186b4 14805Author: Dave Airlie <airlied@redhat.com> 14806Date: Tue Dec 16 12:19:00 2008 +1000 14807 14808 DCE32: add support for DCE3.2 digital outputs. 14809 14810 This adds support for the digital outputs on the RV730. 14811 14812commit 4d3792fd1b9f733107582436540f483d8993199f 14813Author: Dave Airlie <airlied@redhat.com> 14814Date: Fri Dec 12 10:34:57 2008 +1000 14815 14816 radeon: fixes from Alex for some output engines 14817 (cherry picked from commit 1e8ac6ea7b35ccbc3649c96ead60f69158f66ebc) 14818 14819commit f892154dbd0d8b6dd8abdf2196ab0c7c18327412 14820Author: Alex Deucher <alexdeucher@gmail.com> 14821Date: Mon Dec 15 19:41:32 2008 -0500 14822 14823 R3xx-R5xx: switch back to quad rendering for Xv 14824 14825 The large clipped triangle trick hits the guardband limits on 14826 large displays. This gets textured video working again 14827 until we sort out how to deal with the situation better. 14828 14829 See bug 19046 14830 14831commit 62344e460f5d0cac81f94c535cda77d3d76528a9 14832Author: Alex Deucher <alexdeucher@gmail.com> 14833Date: Mon Dec 15 11:30:42 2008 -0500 14834 14835 Remove remnants of unused DDCMode option 14836 14837commit f1efee6a11dc9436276fb12cfb90aa79ada1192b 14838Author: Alex Deucher <alexdeucher@gmail.com> 14839Date: Mon Dec 15 11:27:17 2008 -0500 14840 14841 Remove mention of MergedFB from server log 14842 14843 noticed by Fabio, bug 19098 14844 14845commit 94bffac6968057cdedc792f8540fe44cb6ecfd7c 14846Author: Fabio <fabio.ped@libero.it> 14847Date: Mon Dec 15 11:22:56 2008 -0500 14848 14849 man page updates 14850 14851 from bug 19098 with some minor fixes from me 14852 14853commit c7899905ec824071f9cabbd1e271c4d76010b733 14854Author: Fabio <fabio.ped@libero.it> 14855Date: Mon Dec 15 10:58:23 2008 -0500 14856 14857 Remove duplicate ATI in some device names 14858 14859 from bug 19098 14860 14861commit 89a8fd5337c77abf1f3166e69ed061d85b5c20f0 14862Author: Dave Airlie <airlied@redhat.com> 14863Date: Mon Dec 15 14:41:10 2008 +1000 14864 14865 r600: don't idle engine 14866 (cherry picked from commit b77729448e487d37707ce028d8edc59352d1cc2d) 14867 14868commit 4ce255a9db8027f2471ba50e57724d3e4d5b16ba 14869Author: Fabio <fabio.ped@libero.it> 14870Date: Fri Dec 12 09:53:33 2008 -0500 14871 14872 Fix formatting in man page 14873 14874 fixes bug 19039 14875 14876commit 0e99017cb9a6d46b891705df73678c2705439532 14877Author: Joseph Adams <joeyadams3.14159@gmail.com> 14878Date: Thu Dec 11 00:57:03 2008 -0500 14879 14880 Add more eMac modes, fix crash in last patch 14881 14882commit fb46c30d316ab3ec54c54f3aec91d5164070a423 14883Author: Alex Deucher <alexdeucher@gmail.com> 14884Date: Wed Dec 10 17:16:56 2008 -0500 14885 14886 Add support for Apple eMacs 14887 14888 Thanks to Joseph Adams for helping me sort this out. 14889 Verified on eMac G4/1.0 with radeon 7500 14890 Later eMacs with radeon 9200 or 9600 chips may have 14891 different ddc setups. Need to verify. 14892 14893commit c0bcea9150ef215fa614733cef9a5b71a55a33bd 14894Author: Alex Deucher <alexdeucher@gmail.com> 14895Date: Wed Dec 10 11:40:47 2008 -0500 14896 14897 Add another AGP quirk 14898 14899 fixes bug 12544 14900 14901commit 607b02d6abf78147902a5944bd006897ad95e982 14902Author: Alex Deucher <alexdeucher@gmail.com> 14903Date: Wed Dec 10 03:54:35 2008 -0500 14904 14905 Add randr 1.3 panning support 14906 14907commit 24007acbf36bcfd09cb5e74ce4a8d918d393cf34 14908Author: Benjamin Close <Benjamin.Close@clearchain.com> 14909Date: Wed Dec 10 02:17:56 2008 -0500 14910 14911 [PATCH] xf86misc extension has been removed, update driver to reflect it 14912 14913commit 69e445f7f2ebd299070e44457b434bbc28b7e05c 14914Author: Dave Airlie <airlied@linux.ie> 14915Date: Tue Dec 9 18:59:11 2008 +1000 14916 14917 radeon: memset the pll setting function 14918 14919 Thanks to Peter Zijlstra for making me debug this and providing 14920 feedback. 14921 14922commit c72b732a2f5597cd72bd4ebe207799ae07f2681e 14923Author: Adrian Friedli <adi@koalatux.ch> 14924Date: Mon Dec 8 10:42:46 2008 -0500 14925 14926 Add another AGP quirk 14927 14928commit 979ad04d901676cb41aea5cf47a8a2e9c19064dd 14929Author: Alex Deucher <alexdeucher@gmail.com> 14930Date: Fri Dec 5 17:53:57 2008 -0500 14931 14932 Add missing .TP in man page 14933 14934commit b89d434011787fb71593395f462d142375783e35 14935Author: Pierre Ossman <pierre@ossman.eu> 14936Date: Fri Dec 5 17:52:02 2008 -0500 14937 14938 Fix scissor setup for Xv 14939 14940 We need to adjust the scissoring for each clip box, so move the setup 14941 to inside the while loop. 14942 14943commit 83c1115333cbe763f86422fa5aabf4ab21951e68 14944Author: Alex Deucher <alexdeucher@gmail.com> 14945Date: Fri Dec 5 11:52:18 2008 -0500 14946 14947 R200: Remove scissor setup from init3d() 14948 14949 We aren't using them, so no need to set them up. 14950 14951commit 7593d5399cdb2f42ecbcb5193580385b286a9db9 14952Author: Alex Deucher <alexdeucher@gmail.com> 14953Date: Fri Dec 5 11:18:15 2008 -0500 14954 14955 set coherent mode consistently (enabled for now) 14956 14957 It's set to enabled in the output setup, but then was set to 14958 disabled when teh output properties were setup. Set it up 14959 consistently. 14960 14961commit ad2579f8898251105a6b36b745afd1ce1dab103e 14962Author: Dave Airlie <airlied@linux.ie> 14963Date: Fri Dec 5 20:03:53 2008 +1000 14964 14965 radeon: damn you bios, HDMI cannot have DACs. 14966 14967commit bae30856f2cb3884387ef347327af6ff00b75854 14968Author: Alex Deucher <alexdeucher@gmail.com> 14969Date: Fri Dec 5 08:15:59 2008 +0100 14970 14971 Make VSync for EXA and Xv configurable 14972 14973commit 33638d9e388b330e2f4eb4debd05ba09924cf176 14974Author: Pierre Ossman <pierre@ossman.eu> 14975Date: Thu Dec 4 23:29:31 2008 +0100 14976 14977 Optimise RADEONWaitForVLine 14978 14979 Only avoid the vlines we are rendering to, instead of the entire 14980 screen. This way we don't stall the card for longer than we 14981 absolutely have to. 14982 14983 EXA calls fixed by Alex Deucher. 14984 14985commit d1690f5cc096e2f735c8b407c370a1c1cd7a8afc 14986Author: Pierre Ossman <pierre@ossman.eu> 14987Date: Wed Dec 3 21:49:37 2008 +0100 14988 14989 Improve tearing avoidance for Xvideo in two steps 14990 14991 - Fix up VLINE handling to trigger whenever scanout is outside the 14992 visible area. 14993 14994 - Render the video as a scissored triangle as R300+ cannot render a 14995 quad in a single pass. 14996 14997commit 4d98acbca2e630056bf56cdcd0e23007fded2ced 14998Author: Alex Deucher <alexdeucher@gmail.com> 14999Date: Wed Jul 16 03:14:06 2008 -0400 15000 15001 First pass at tear-free accel 15002 15003 if the dest pixmap is the front buffer, stall the pipe 15004 until the vline is outside the active area. 15005 For EXA, pick crtc based on the larger mode area; 15006 ideally we'd have one pixmap per crtc. 15007 For Xv, use dst window area to determine crtc. 15008 15009commit 82d67b70d0fb747cf7b1eb0b64244ae56cb84572 15010Author: Pierre Ossman <pierre@ossman.eu> 15011Date: Thu Dec 4 15:38:13 2008 -0500 15012 15013 Switch r200 Xv to use rect lists rather than quads to avoid tearing 15014 15015 Patch from Pierre with fixes from me. 15016 15017commit 9eccd0ad0827a1b8ec9abb2cebf5768d204a5ce9 15018Author: Alex Deucher <alexdeucher@gmail.com> 15019Date: Thu Dec 4 12:28:12 2008 -0500 15020 15021 Fix up US setup after r3xx bicubic merge 15022 15023commit e13fba853ba19e6b0f081b9b3d9fa76c38a0f82b 15024Merge: d296337 79bbdd9 15025Author: Alex Deucher <alexdeucher@gmail.com> 15026Date: Thu Dec 4 12:25:29 2008 -0500 15027 15028 Merge branch 'bicubic' of git://git.infradead.org/users/drzeus/xf86-video-ati 15029 15030commit d29633708b5451f5541e88371d831ae03019a7f5 15031Author: Dave Airlie <airlied@redhat.com> 15032Date: Thu Dec 4 09:08:48 2008 +1000 15033 15034 radeon: apply atom quirks to object table 15035 15036commit 941ac0d3ab839dc5aaaef2f3b7f5c7085b2894c5 15037Author: Alex Deucher <alexdeucher@gmail.com> 15038Date: Wed Dec 3 17:31:25 2008 -0500 15039 15040 Make sure 3D state is emited when switching to 3D IB 15041 15042 - fixes bug 18864 15043 15044commit 79bbdd984c925e37f5b3db2605339f1a21377fcf 15045Author: Pierre Ossman <pierre@ossman.eu> 15046Date: Wed Dec 3 20:33:36 2008 +0100 15047 15048 Change the XV_BICUBIC attribute to a tristate, where the third state 15049 automatically turns the filter on and off as appropriate. 15050 15051commit bbdf821c179242266fd86edf9e688a6be9cf985f 15052Author: Pierre Ossman <pierre@ossman.eu> 15053Date: Wed Dec 3 19:42:50 2008 +0100 15054 15055 Make the R300 bicubic shader program a bit easier to understand by 15056 having comments that describe the operation, not how it has to be 15057 implemented in the hardware. 15058 15059commit 8f4b22932c91f1d51fcbf1d1a99d852fcdaafdcc 15060Author: Pierre Ossman <pierre@ossman.eu> 15061Date: Wed Dec 3 19:31:31 2008 +0100 15062 15063 Now that we have a bicubic code path that messes up the pixel shader, 15064 we need to restore it in the non-bicubic code. 15065 15066commit 0851daa6e7db77e440188d774ef8b3236146bcb9 15067Author: Pierre Ossman <pierre@ossman.eu> 15068Date: Wed Dec 3 19:17:36 2008 +0100 15069 15070 Fix macros so that they are safe to use with expressions as parameters. 15071 15072commit e04bcb0066781f61e97787856261e3380036d304 15073Author: Pierre Ossman <pierre@ossman.eu> 15074Date: Wed Dec 3 19:16:03 2008 +0100 15075 15076 Fix node setup on R300 bicubic pixel shader. It was backwards and used 15077 the wrong base for sizes. 15078 15079commit 927aee181f3637bda1b761caf93fbc0652e1207b 15080Author: Alex Deucher <alexdeucher@gmail.com> 15081Date: Wed Dec 3 01:39:38 2008 -0500 15082 15083 Only flush IB when direct rendering is enabled 15084 15085commit ca74ebf8813b7fab8d08465ef9600f001c948e94 15086Author: Alex Deucher <alexdeucher@gmail.com> 15087Date: Tue Dec 2 22:26:21 2008 -0500 15088 15089 move 3d state init after RADEON_SWITCH_TO_3D() 15090 15091commit fc079c5267baf431bbecee7744e484783d393152 15092Author: Dave Airlie <airlied@redhat.com> 15093Date: Tue Dec 2 22:11:57 2008 -0500 15094 15095 Don't mix 2D and 3D in the same IB 15096 15097 fix from radeon-gem-cs 15098 15099commit 678693aabd0b4c1812ab44bf5abb4c8394b254bb 15100Author: Dave Airlie <airlied@redhat.com> 15101Date: Tue Dec 2 22:02:21 2008 -0500 15102 15103 Stability fixes from radeon-gem-cs 15104 15105commit f231e4d9b2c683dc16f42cf09484fae926a892c1 15106Author: Dave Airlie <airlied@redhat.com> 15107Date: Wed Dec 3 11:39:04 2008 +1000 15108 15109 radeon: fix HDMI having DACs in some BIOSes 15110 15111commit fe73d9a7dfe8ec5c8f1a8dc08e14b4e138aa9276 15112Author: Bryce Harrington <bryce@canonical.com> 15113Date: Tue Dec 2 19:47:45 2008 -0500 15114 15115 Add another AGP quirk 15116 15117commit 6e0c48958c6a50a639368e369ce72d813256aee4 15118Author: Alex Deucher <alexdeucher@gmail.com> 15119Date: Mon Dec 1 18:50:40 2008 -0500 15120 15121 Add quirk for r200 card with the primary dac wired to both ports 15122 15123 refactor load detection setup as a byproduct 15124 should fix bug 18719 15125 15126commit fa496d7b0397d9be57db90d0860928e9ced73cca 15127Author: Pierre Ossman <pierre@ossman.eu> 15128Date: Mon Dec 1 08:29:29 2008 +0100 15129 15130 Fix comments for R500 fragment shader to reflect the code. 15131 15132commit 0d95fca114194664d9ab612c90d3222912cf41ff 15133Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15134Date: Sat Aug 30 12:38:43 2008 -0700 15135 15136 Fix indentation on IS_R300_3D Xv code. 15137 15138commit 36a62b8c47def3f23a0b735875f59d23ed3c095a 15139Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15140Date: Sat Aug 30 11:15:20 2008 -0700 15141 15142 r3xx: Various Xv fixes. 15143 15144commit 01ed5462fad56f5b7bf867041d00443bdf81c653 15145Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15146Date: Thu Aug 7 18:08:54 2008 -0700 15147 15148 Switch to Mesa-style 24-bit float packing. 15149 Seems like Mesa's got it right, so we should follow suit. 15150 15151commit 85bbe889045aa434f71a3eb0ba8153b7d2cc5023 15152Author: Maciej Cencora <m.cencora@gmail.com> 15153Date: Thu Aug 7 07:06:46 2008 -0700 15154 15155 Enable bicubic Xv on r3xx+. Still some buggies in the actual display, but looks pretty good. 15156 15157commit bf6263cff50bb85c19ff6dfebb739333fa8aca18 15158Author: Maciej Cencora <m.cencora@gmail.com> 15159Date: Thu Aug 7 07:00:36 2008 -0700 15160 15161 Add r3xx-specific bicubic filtering code. FP, RS, mostly, everything else is shared with r5xx. 15162 15163commit da021c36bbdf3bca31ee50ebe01cdb9495c09b36 15164Author: Dave Airlie <airlied@linux.ie> 15165Date: Sun Nov 30 18:44:24 2008 +1000 15166 15167 radeon_drm.h: remove kernel defines 15168 15169 remove these we should update from an installed kernel copy not from the 15170 original kernel copy. 15171 15172commit 783cdb7374941bb1d2b63eea375fbf2f1b808cc3 15173Author: Dave Airlie <airlied@redhat.com> 15174Date: Thu Nov 27 15:29:06 2008 +1000 15175 15176 r100/r200: EXA misrenders dst == a8 + dst alpha use. 15177 15178 Fedora BZ#469556 15179 15180 I've played with various dst and texture formats and rewritten large 15181 parts of this to no avail, so I'm punting for now. 15182 15183commit 2d7cfc421713286edd5121b56999246f9ae70286 15184Author: Eygene Ryabinkin <rea-fbsd@codelabs.ru> 15185Date: Tue Nov 25 13:25:33 2008 -0500 15186 15187 Catch unsubstituted macro for PCI region base access 15188 15189 also fix up some warnings. 15190 15191commit af7690b9c300d37c3a0c51e27e0ad2ca009224a2 15192Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 15193Date: Tue Nov 25 02:57:30 2008 -0500 15194 15195 [PATCH] radeon: Fix PCI usage of 32-bit driver on 64-bit platform 15196 15197 The radeon driver is storing PCI addresses in unsigned long's which 15198 won't work well on 32-bit platforms with 64-bit physical address space 15199 such as PowerPC 4xx. This fixes it by using unsigned long long instead. 15200 15201commit 065938617c0feab17f4274a5350de02a692ba065 15202Author: Ramon van der Stelt <rvanderstelt@xs4all.nl> 15203Date: Tue Nov 25 02:23:46 2008 -0500 15204 15205 Interlaced mode fixes 15206 15207 see bug 12626 15208 15209commit 36a7dc6ea1e1929e986ab1159497c71521cb2f10 15210Author: Bryce Harrington <bryce@canonical.com> 15211Date: Tue Nov 25 01:04:37 2008 -0500 15212 15213 Additional AGP quirks 15214 15215 bug 18693 15216 15217commit 2acb8e4ed220fccb21b22a53c5142ab3a35d32ef 15218Author: Alex Deucher <alexdeucher@gmail.com> 15219Date: Tue Nov 25 15:07:09 2008 +1000 15220 15221 radeon: fixup shared DDC lines for some rv610 cards. 15222 15223 Some cards share DDC between a HDMI and VGA output, and expect the 15224 driver to use load detect or EDID to figure it out. 15225 15226 airlied- shipped in RHEL5 15227 15228 Signed-off-by: Dave Airlie <airlied@redhat.com> 15229 15230commit b01e35168de40d192fd7a9ce6884b9c7419afbd4 15231Author: Alex Deucher <alexdeucher@gmail.com> 15232Date: Mon Nov 24 21:06:42 2008 -0500 15233 15234 Fix up posted logic 15235 15236 noticed by benh on IRC 15237 15238commit d7a03e93372b17d642b42f5490464616ae7c99c1 15239Author: Alex Deucher <alexdeucher@gmail.com> 15240Date: Mon Nov 24 15:52:05 2008 -0500 15241 15242 Get hardcoded edid from the bios for servers that support it 15243 15244 should fix bug 11300 15245 15246commit 3858e31fa9d77d52a5cabb02f53f06385eab40f7 15247Author: Alex Deucher <alexdeucher@gmail.com> 15248Date: Mon Nov 24 14:39:41 2008 -0500 15249 15250 Don't support rotation is accel is not available 15251 15252 Fixes bug 18068 15253 15254commit 3d27876d91ef69d71f17b6cf8c07c26bf12857b4 15255Author: Dave Airlie <airlied@redhat.com> 15256Date: Mon Nov 10 15:47:13 2008 +1000 15257 15258 radeon: always align dstPitch to 64 bytes for textured video 15259 15260 bz# 18454 15261 15262commit 6c8b352264540ccc4622a7c0b704ce452dcd707d 15263Author: Dave Airlie <airlied@redhat.com> 15264Date: Wed Oct 29 15:53:57 2008 +1000 15265 15266 pciid: add HD3300 - 790GX chipset. 15267 15268 RH BZ (#466706) 15269 15270commit 902eaf768142c6c7dcc487e10775027b84cd1f9a 15271Author: Alex Deucher <alexdeucher@gmail.com> 15272Date: Thu Nov 6 15:46:43 2008 -0500 15273 15274 Check for LVDS on all IGP chips 15275 15276 - fixes bug 18395 15277 15278commit 44abaa0ea8e5e3bbe833fa73e6d011618bb6d7e7 15279Author: Adam Jackson <ajax@redhat.com> 15280Date: Mon Nov 3 15:09:28 2008 -0500 15281 15282 Fix HDMI output setup on DIG1/DIG2. 15283 15284 The only reliable way to detect an HDMI sink is to check for the 15285 appropriate EDID extension block, so this will only work right on 15286 servers with xf86DoEEDID(). RANDR 1.2 will call this internally for 15287 us if it exists, but pre-1.6 servers are out of luck. 15288 15289commit 937b7ac2a259cf504a19dcf62a58b1db1afb8eb9 15290Author: Bryce Harrington <bryce@canonical.com> 15291Date: Wed Oct 29 01:31:50 2008 -0400 15292 15293 Add AGP quirk table 15294 15295 - set known good default agp modes for problematic chip/bridge combos 15296 - should fix bug 17360, others 15297 15298commit 67bc2ae7f227a2afedc8e699cf0458fa7960f301 15299Author: Julien Cristau <jcristau@debian.org> 15300Date: Sun Oct 26 16:05:04 2008 +0100 15301 15302 Bug#7148: Fix manpage formatting 15303 15304commit 98fdd78a1229584b922f816f331a45bfc178c961 15305Author: Dave Airlie <airlied@linux.ie> 15306Date: Tue Oct 21 18:36:22 2008 +1000 15307 15308 atombios: oops typo 15309 15310commit 763dff6c117a10b6a57ba1854c5b798359789e44 15311Author: Dave Airlie <airlied@linux.ie> 15312Date: Tue Oct 21 18:35:44 2008 +1000 15313 15314 atombios: set missing pixel clock 15315 15316commit 435cf7da68186f2601c4b888296117d4f652c625 15317Author: Alex Deucher <alexdeucher@gmail.com> 15318Date: Wed Oct 15 10:52:14 2008 -0400 15319 15320 Add LCD dithering quirk for macbook pro 15321 15322 fixes bug 17897. 15323 15324commit c0e6cb6d0eeef8f2ea60d840e1cd668fa92cd7f9 15325Author: Alex Deucher <alexdeucher@gmail.com> 15326Date: Fri Oct 10 11:09:27 2008 -0400 15327 15328 Fix grey level dithering setup 15329 15330commit 9eb2b6cd28224f0cbbabf3df451696ae1a845b9c 15331Author: Dave Airlie <airlied@redhat.com> 15332Date: Fri Oct 10 16:18:05 2008 +1000 15333 15334 radeon: hopefully fix plls for 30" monitors. 15335 15336 Only do the HIGH trick on the DCE3 cards. Tested on rv610/630/770. 15337 15338commit 0975e007ec7933acf9cd8ad2f281313ef8d32ac9 15339Author: Dave Airlie <airlied@redhat.com> 15340Date: Thu Oct 9 17:16:28 2008 +1100 15341 15342 radeon: rename radeon_memory to radeon_legacy_memory. 15343 15344 This is a setup patch for the kms/memory management code. 15345 15346commit d744dc013cf5ca3141c0fdf55e36d713fe9f53ba 15347Author: Adam Jackson <ajax@redhat.com> 15348Date: Wed Oct 8 15:50:25 2008 -0400 15349 15350 Fix connector table mapping for lvtma/kaleidoscope/uniphy. 15351 15352 Fixes DPMS on LVDS. 15353 15354commit dc795ba66a8c0eb2c3fdb86bd42dfc17e0aa3b6e 15355Author: Alex Deucher <alexdeucher@gmail.com> 15356Date: Tue Oct 7 12:32:47 2008 -0400 15357 15358 Fixups from last DFP/LCD refactor 15359 15360 - fix DFP1 -> LCD1 15361 - use _INDEX rather than _SUPPORT defs 15362 15363commit 4a9e8d9fe24278b56aef51677fd35289d01584a7 15364Author: Alex Deucher <alexdeucher@gmail.com> 15365Date: Mon Oct 6 18:00:08 2008 -0400 15366 15367 set grey levels correctly for temporal dithering 15368 15369commit 5708624d1151877dd66a462d4a75fed6774604a7 15370Author: Alex Deucher <alexdeucher@gmail.com> 15371Date: Mon Oct 6 15:11:56 2008 -0400 15372 15373 Turn coherent mode off by default on DVI 15374 15375commit a5c5ce96279d01eb519bfb92b94c06a58acb7f07 15376Author: Alex Deucher <alexdeucher@gmail.com> 15377Date: Mon Oct 6 15:07:31 2008 -0400 15378 15379 Refactor atom LCD/DFP output setup 15380 15381 Simplify the code, properly handle dithering, coherent mode, encoding, etc. 15382 Should fix bug 17897. 15383 15384commit f9826a5694b7adb6920eb5bdf45d840d8fb14d53 15385Author: Alex Deucher <alexdeucher@gmail.com> 15386Date: Fri Oct 3 15:24:12 2008 -0400 15387 15388 Add support for DDC via atom commands for RV410 15389 15390 The atom calls use the hw i2c engine for DDC. For some 15391 reason, sw i2c doesn't seem to work on the VGA GPIO on 15392 RV410 chips, so we use atom in that case. 15393 15394 This fixes the longstanding VGA DDC problems on RV410/M26 15395 chips. 15396 15397commit 23c17c0121d43c2fd11bb5cfbaa4b02abac2b16d 15398Author: Alex Deucher <alexdeucher@gmail.com> 15399Date: Fri Oct 3 14:30:33 2008 -0400 15400 15401 man page updates 15402 15403 Update render accel and page flipping information. 15404 fixes bug 17887. 15405 15406commit 4ac6e499089aec3c79af5dbede5fe631b105ce3d 15407Author: Alex Deucher <alexdeucher@gmail.com> 15408Date: Wed Oct 1 18:53:15 2008 -0400 15409 15410 proper fix for initial rotation 15411 15412 The initial modeset was setting the MC regs, so just set them 15413 rather than the full modeset. 15414 15415commit 346228b20c69a965c9e7d67157f42c1d36a5b6c3 15416Author: Alex Deucher <alexdeucher@gmail.com> 15417Date: Wed Oct 1 14:38:27 2008 -0400 15418 15419 Workaround to make initial rotation work 15420 15421 Acceleration needs to be initialized before setting the mode 15422 for initial rotation to work. Changing the order in RADEONScreenInit() 15423 doesn't work because RADEONDRIKernelInit() hangs in the ioctl to 15424 initialize the CP if the mode is not set prior (even if the heads 15425 are turned off after setting the mode). The workaround is to set the 15426 modes twice, once before accel and once after. 15427 15428 This needs to be looked into further. 15429 15430commit c359c2a31caf9f75b9fc6b6bcbc3e9dc1fe404ba 15431Author: Michel Dänzer <michel@tungstengraphics.com> 15432Date: Tue Sep 30 11:04:54 2008 +0200 15433 15434 Pass base offset into RADEONDisplayVideo() explicitly. 15435 15436 This prevents the base offset from flipping back and forth if double buffering 15437 is enabled and the two buffers start on different sides of a 4 MB boundary. 15438 15439 Should fix http://bugs.freedesktop.org/show_bug.cgi?id=17254 ; please don't 15440 reopen that bug report without attaching a log file captured after reproducing 15441 the problem with this fix. 15442 15443commit ba21d5ae7fbbf5828c44c41655aff8b182453e2e 15444Author: Michel Dänzer <michel@tungstengraphics.com> 15445Date: Tue Sep 30 11:03:04 2008 +0200 15446 15447 Update GEN_INT_CNTL register value after calling DRM modeset ioctl. 15448 15449 Fixes sync-to-vblank hangs after mode switch with DRM vblank-rework. 15450 15451commit 3110646daf05fa09faaa70066cc8c3db1bb07cec 15452Author: Michel Dänzer <michel@tungstengraphics.com> 15453Date: Sun Sep 14 11:46:36 2008 +0200 15454 15455 Fix -Werror build. 15456 15457commit d82f2938f69402c641a1c8362fdc513419b27659 15458Author: Alex Deucher <alexdeucher@gmail.com> 15459Date: Fri Sep 26 13:51:24 2008 -0400 15460 15461 atombios updates from upstream 15462 15463commit d0d58b39e49c5ce00bc8bd12f721ad8c7f554b91 15464Author: Alex Deucher <alexdeucher@gmail.com> 15465Date: Wed Sep 24 11:54:34 2008 -0400 15466 15467 radeon: Further pll adjustments 15468 15469commit d100a6af8f828eb94f8ba6e8a96c24389b5cf46f 15470Author: Alex Deucher <alexdeucher@gmail.com> 15471Date: Fri Sep 19 14:04:59 2008 -0400 15472 15473 cleanup macbook quirk 15474 15475commit 1218e38c75351a20f7414ba6fcb3545689339442 15476Author: Alex Deucher <alexdeucher@gmail.com> 15477Date: Fri Sep 19 13:52:07 2008 -0400 15478 15479 more macbook fixes 15480 15481commit 83e1f132965deadedda7a622184eb400cd4c120c 15482Author: Alex Deucher <alexdeucher@gmail.com> 15483Date: Fri Sep 19 09:36:59 2008 -0400 15484 15485 Fix copy/paste error in the last commit 15486 15487 DVI-D -> DVI-I 15488 15489commit 4af825f7b13cba8c3b204e72a07cab5009331240 15490Author: Alex Deucher <alexdeucher@gmail.com> 15491Date: Fri Sep 19 00:29:20 2008 -0400 15492 15493 radeon: Add quirk for macbook 15494 15495 - should fix bug 17625 15496 15497commit 38aa450fce6a91b35c191fd07112490a62cc29c6 15498Author: Alex Deucher <alexdeucher@gmail.com> 15499Date: Tue Aug 26 21:50:23 2008 -0400 15500 15501 fix some fallout from the common allocator 15502 15503 should fix bug 17317 15504 15505commit a3cc1d7a421456186024c5c069e403d374a0f0b9 15506Author: Dave Airlie <airlied@linux.ie> 15507Date: Tue Aug 26 08:03:20 2008 +1000 15508 15509 radeon: fix powerpc build 15510 15511commit 6cebfe257f7ddad855ee743e4eb899bd6fac7f46 15512Author: Alex Deucher <alexdeucher@gmail.com> 15513Date: Fri Jul 11 19:32:06 2008 -0400 15514 15515 Switch EXA path back to static cursor allocation 15516 15517 pre-AVIVO cards have address limits for the cursor offset 15518 15519commit 4dff54a3c8d7c9f2d6ec50354ff0b92f1b7fcbdf 15520Author: Alex Deucher <alexdeucher@gmail.com> 15521Date: Fri Jul 11 17:28:03 2008 -0400 15522 15523 Switch cursors over to generic allocator 15524 15525commit 5b1978a4796bcc31ac2f01d303dc8f8f44323025 15526Author: Alex Deucher <alexdeucher@gmail.com> 15527Date: Mon Aug 25 10:05:28 2008 -0400 15528 15529 Bicubic fixes from the last cherry-pick 15530 15531commit 7daba77ded1c718e93ae8c372a39a6e85228d513 15532Author: Alex Deucher <alexdeucher@gmail.com> 15533Date: Fri Jul 11 02:02:38 2008 -0400 15534 15535 Convert randr, Xv to a common allocator 15536 15537commit 610fe1a937da78f4ac813ac919c158dde8f42442 15538Author: Alex Deucher <alexdeucher@gmail.com> 15539Date: Thu Jul 10 22:47:35 2008 -0400 15540 15541 Switch to using radeon_drm.h from the drm 15542 15543 modelled after Matthias' similar rhd change 15544 15545commit a6561f2ec673b38907f7181235386f32e60c32ba 15546Author: Alex Deucher <alexdeucher@gmail.com> 15547Date: Thu Jul 10 22:31:13 2008 -0400 15548 15549 Move DRI to a separate struct 15550 15551 based on Matthias' similar work in rhd 15552 15553commit 75ef8dc214715d3c5c50996b293933842903ba65 15554Author: Egbert Eich <eich@freedesktop.org> 15555Date: Thu Jul 10 21:49:55 2008 -0400 15556 15557 Cleanups from rhd port 15558 15559 - remove unused vars 15560 - remove static exa render vars 15561 15562commit 71ad140fa11f3a504c38d6bddf40e3a3c0a20e60 15563Author: Alex Deucher <alexdeucher@gmail.com> 15564Date: Thu Jul 10 21:24:16 2008 -0400 15565 15566 Move accel state to a separate struct 15567 15568commit 5b2e095c31b88d8495a4f86e6cb46b49fa4acd65 15569Author: Alex Deucher <alexdeucher@gmail.com> 15570Date: Thu Jul 10 20:07:44 2008 -0400 15571 15572 Move CP into a separate struct 15573 15574commit 2145309230e36aee7758bd244deb1e72ada0c065 15575Author: Alex Deucher <alexdeucher@gmail.com> 15576Date: Mon Aug 25 08:47:59 2008 -0400 15577 15578 match textured video macro names with other accel code 15579 15580 OUT_VIDEO_REG() -> OUT_ACCEL_REG() 15581 etc. 15582 15583commit 01daef0f095fbbaee701d5fe97f3dd7838b5f915 15584Author: Alex Deucher <alexdeucher@gmail.com> 15585Date: Mon Aug 25 08:26:16 2008 -0400 15586 15587 Additional cleanups and re-arragement following bicubic merge 15588 15589commit 0a51d08c24af040fe48690662b1a912acad51700 15590Author: Alex Deucher <alexdeucher@gmail.com> 15591Date: Mon Aug 25 06:43:14 2008 -0400 15592 15593 Whitespace cleanup from bucubic merge 15594 15595commit c0170fef510b53a6ca2c6ea7a99119235229c929 15596Author: Maciej Cencora <m.cencora@gmail.com> 15597Date: Thu Aug 7 06:53:39 2008 -0700 15598 15599 Add needed FP registers, etc. for r3xx bicubic Xv. 15600 15601commit d9c38326cf70f57ab777ffdf9520b8cdea9d9cb6 15602Author: Dennis Kasprzyk <onestone@opencompositing.org> 15603Date: Tue Aug 5 18:45:38 2008 +0200 15604 15605 Fix typos. 15606 15607commit d38ceba62aa5cee76baa342ce7719a983a79f09e 15608Author: Dennis Kasprzyk <onestone@opencompositing.org> 15609Date: Tue Aug 5 16:21:57 2008 +0200 15610 15611 Fix bicubic fp calculation. 15612 15613commit 6f9c7fde75edc0399559f975db9a6c4dc22714ae 15614Author: Dennis Kasprzyk <onestone@opencompositing.org> 15615Date: Tue Aug 5 15:38:42 2008 +0200 15616 15617 Fixed bicubic fragment program comments. 15618 15619commit 48b09ca40ccb28b5584069316fd38786a78c1dd3 15620Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15621Date: Sat Aug 2 02:36:21 2008 -0700 15622 15623 Switch from 32-bit floats to 16-bit half-floats. 15624 Massive bandwidth savings, or so I'm told. Yay? 15625 15626commit ebbb7fb634fcadf28ff99c1df2c3db89fd56932d 15627Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15628Date: Sat Aug 2 01:45:43 2008 -0700 15629 15630 Change floats to uint32_t hex. 15631 Useful for moving to 16-bit half-floats. 15632 15633commit 3b46162d8a90be3524cb513d42b9ad3d0bea45f5 15634Author: Dennis Kasprzyk <onestone@opencompositing.org> 15635Date: Fri Aug 1 15:24:01 2008 +0200 15636 15637 Move some ALU instructions after the TEX instructions, so we can do something usefull while we are waiting for the texture values. 15638 15639commit 2ecdec4bafc97212dde4d6908ee4ccf618adc0e1 15640Author: Dennis Kasprzyk <onestone@opencompositing.org> 15641Date: Fri Aug 1 04:03:12 2008 +0200 15642 15643 Another uneeded SEM_AQUIRE. 15644 15645commit e93b5d1b80d6203f63543b7b678e2f1d9221b5b3 15646Author: Dennis Kasprzyk <onestone@opencompositing.org> 15647Date: Fri Aug 1 03:00:26 2008 +0200 15648 15649 Smarter usage of the texture semaphore. 15650 15651commit 286f7aa18f19bd21b263701adab38b736dbeda0f 15652Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15653Date: Thu Jul 31 17:01:44 2008 -0700 15654 15655 Enable bicubic filtering for all r5xx HW. 15656 15657commit 242aa4f630b4c60aefa3c12dc459a4d4d0b334a0 15658Author: Dennis Kasprzyk <onestone@opencompositing.org> 15659Date: Fri Aug 1 01:11:28 2008 +0200 15660 15661 Remove one constant. 15662 15663commit 413eacb0538977b0b3c92df074d40510f4539abc 15664Author: Dennis Kasprzyk <onestone@opencompositing.org> 15665Date: Fri Aug 1 00:52:09 2008 +0200 15666 15667 Heavy optimizations. 15668 15669commit a0c4a949cb49e5ac1e857aef08a8742b9f7b49da 15670Author: Dennis Kasprzyk <onestone@opencompositing.org> 15671Date: Thu Jul 31 21:19:18 2008 +0200 15672 15673 Remove uneeded negations. 15674 15675commit abb2b2e757d92591ab9277824b9d9746ba98f875 15676Author: Dennis Kasprzyk <onestone@opencompositing.org> 15677Date: Thu Jul 31 20:31:03 2008 +0200 15678 15679 Implement LPR in one instruction. 15680 15681commit c370b74bec13194573348583c38adf710b880e79 15682Author: Dennis Kasprzyk <onestone@opencompositing.org> 15683Date: Thu Jul 31 19:50:49 2008 +0200 15684 15685 Set helper texture filter correctly. 15686 15687commit 5e85d5a5d0c50b29086ec0c219c8b52d25dbc2e9 15688Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15689Date: Wed Jul 30 19:45:15 2008 -0700 15690 15691 Update bicubic tables. 15692 Now including the 2048x1 texture as well. 15693 15694commit 17e5e9573e59c3d82d51c261b9c5005f6aec7d43 15695Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15696Date: Wed Jul 30 11:57:25 2008 -0700 15697 15698 Force R580-only for bicubic. 15699 Initial reading of docs suggest RV560 and RV570 can't handle it, 15700 but they're welcome to test. 15701 15702commit 83b52473d0e7102265918b07be053fcda17a14b4 15703Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15704Date: Wed Jul 30 01:03:57 2008 -0700 15705 15706 Bump bicubic cutoff to R580. 15707 RV530 chipsets can't handle fullscreen bicubic... 15708 15709commit 69a4998d1286bcdd7bfe874cd5628bc1cc232bae 15710Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15711Date: Wed Jul 30 01:03:46 2008 -0700 15712 15713 Oops, made a mistake with vertices. 15714 15715commit a4ec30a677906ec2ff9824c7ddca586655f6d1a8 15716Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15717Date: Wed Jul 30 00:54:32 2008 -0700 15718 15719 Merge upstream changes to vertices, and also add Xv attributes for textured video, including bicubic filtering. 15720 15721commit f3b81c7582aed307fa44e134ee161cd8a3158657 15722Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15723Date: Mon Jul 28 19:50:10 2008 -0700 15724 15725 Fix constants. 15726 15727commit 7dbb7023ba023ec1a38be63af9c9f49e40222b7b 15728Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15729Date: Mon Jul 28 17:30:41 2008 -0700 15730 15731 Finally got the fragment program fully working for bicubic filtering on r5xx. 15732 15733commit 0e4dd73b9ebc6f608eeff945b4d463a00c02e07c 15734Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15735Date: Mon Jul 28 12:16:50 2008 -0700 15736 15737 More bicubic FP buggies. 15738 15739commit a87647e0c27e0950f4d0d8203a1242a994ad3419 15740Author: Dennis Kasprzyk <onestone@compiz-fusion.org> 15741Date: Sun Jul 27 10:43:01 2008 -0700 15742 15743 Fix texture size, texture filter, vertex offsets, etc. 15744 15745commit 232aa3e943fef4c4037b255c3b64a0aaff90ab5c 15746Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15747Date: Sat Jul 26 10:56:56 2008 -0700 15748 15749 Make vertices emit properly. 15750 *bangs head against wall* 15751 15752commit 8c84f67b93d926095633830aa8d95930a48b1c7b 15753Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15754Date: Sun Jul 20 14:25:44 2008 -0700 15755 15756 Try to get tex coords from the VAP to RS properly when bicubic filtering is enabled. 15757 I'm soo bad at this... :c 15758 15759commit a4a7d5f5967c51c394229de5eccaec44cfec8f50 15760Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15761Date: Sat Jul 19 13:02:02 2008 -0700 15762 15763 Upload pixel shader to card for r5xx. This was ridiculous. Also it doesn't work yet... 15764 15765commit 20c1db2d7c110ab5c1117a57b169baa1ab070518 15766Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15767Date: Fri Jul 18 13:48:08 2008 -0700 15768 15769 Upload bicubic filter to card. This was a LOT easier than I had feared, to be honest. 15770 15771commit a760e628134c6d7d42ec3c98118b6e5f6fcd3e7f 15772Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15773Date: Fri Jul 18 12:21:20 2008 -0700 15774 15775 Allocate memory for the bicubic filter texture. 15776 15777commit b6c9e2bb5365de82315c6814f915e57b0c4fa444 15778Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15779Date: Fri Jul 18 11:24:14 2008 -0700 15780 15781 Fixed typos in the bicubic texture tables. Whoops, looks like I'm still asleep. 15782 15783commit ca51f4f37e1dbf53bf7ffc0e8f612e9609e11209 15784Author: Corbin Simpson <MostAwesomeDude@gmail.com> 15785Date: Fri Jul 18 11:06:34 2008 -0700 15786 15787 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". 15788 15789commit 1cf7a5494fa94e8d9f30f9b2905dfbe6d4faa445 15790Author: Bryce Harrington <bryce@bryceharrington.org> 15791Date: Wed Aug 20 09:46:59 2008 -0400 15792 15793 radeon: Fix pasto in connector table setup for vga powerbooks 15794 15795 fixes bug 17214 15796 15797commit a55e85f742d1334bf88e4681e553f025d2de38df 15798Author: Michel Dänzer <michel@tungstengraphics.com> 15799Date: Thu Aug 14 15:21:51 2008 -0400 15800 15801 Make sure video offerlay offsets don't exceed the hardware limit of 128 MB. 15802 15803 Always set the overlay base address such that the buffer offsets are as small 15804 as possible. This could still break in theory if the buffers were more than 15805 128 MB apart, but in reality this can't happen ATM because we always allocate 15806 a single memory area for all buffers. 15807 15808 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=16845 . 15809 15810commit 92ee21df344a989778e37369c7beb3904a00ead6 15811Author: Alex Deucher <alexdeucher@gmail.com> 15812Date: Thu Aug 14 14:49:45 2008 -0400 15813 15814 PLL adjustments 15815 15816 Seems higher dotclocks prefer a higher FB div. 15817 Someone with a lot of should try and find out where 15818 the div sweet spots are for various dotclock ranges. 15819 fixes bug 17125 15820 15821commit 0d5e0347af4322713075193154b8a348de4a0b52 15822Author: Alex Deucher <alexdeucher@gmail.com> 15823Date: Wed Aug 13 14:17:34 2008 -0400 15824 15825 Remove reset of 3D scissor registers when using the CP in the ddx 15826 15827 They should only affect 3D and init3d() should take care of that case 15828 noticed by libv on IRC. 15829 15830commit 7e456e9c427d48151b166bcac08c3e8a8b180f01 15831Author: Dave Airlie <airlied@linux.ie> 15832Date: Tue Aug 12 08:18:30 2008 +1000 15833 15834 radeon: add 0x9441 for hd 4870 x2 15835 15836 thanks to Michael Larabel for testing 15837 15838commit 4dbdeea7c9316575fba26b41fd347452e42cdcf2 15839Author: Alex Deucher <alexdeucher@gmail.com> 15840Date: Fri Aug 8 17:39:48 2008 -0400 15841 15842 Further cleanup and unification of i2c code 15843 15844 - unify the ddc and i2c code 15845 - add gpio mask support for legacy chips 15846 - remove the magic gpio dance for ancient monitors 15847 (if you have an ancient monitor that ddc stops 15848 working on let me know). This should speed up DDC 15849 on legacy chips. 15850 15851 -- radeon sw gpio i2c -- 15852 4 sets of gpio registers for clk and data and corresponding bit masks 15853 mask - locks the gpio for use by sw 15854 get - reads the value off the gpio pad 15855 put - sets the gpio direction to output 15856 a - "other stuff" On legacy chips you clear them if you want 15857 to use a gpio for i2c. In some cases they are used for the 15858 output value when the gpio in the output state. 15859 15860commit 268c848130ec1770bb645a74197b6aca7fc95abc 15861Author: Alex Deucher <alexdeucher@gmail.com> 15862Date: Fri Aug 8 15:50:07 2008 -0400 15863 15864 Fix VT switching on M6 chips 15865 15866 Some M6 chips have a faulty MEM_SIZE register that in 15867 some cases reports 0 on 8 MB cards. On EnterVT we check 15868 the MEM_SIZE reg as a check to see if the card is posted or 15869 not. Since this reg returns 0, the driver attempts to post 15870 the card which can lead to a hang. Switch this to check if 15871 either crtc is active as is done in the bios init code. 15872 fixes bug 13994 15873 15874commit 33f88f7fc90d9d93fdcbba9ad59dd70a6596bc3f 15875Author: Alex Deucher <alexdeucher@gmail.com> 15876Date: Thu Aug 7 17:57:07 2008 -0400 15877 15878 RS4xx DDC fixes take 3 15879 15880commit df0d1ef53100f0a19c5b5fdc349f5186c8d9bd87 15881Author: Alex Deucher <alexdeucher@gmail.com> 15882Date: Wed Aug 6 19:13:56 2008 -0400 15883 15884 RS4xx: Fix up ddc gpio 15885 15886 - I mixed up the regs previously 15887 15888commit 18429390440a829fb24ed3afd99ccf8278138496 15889Author: Alex Deucher <alexdeucher@gmail.com> 15890Date: Wed Aug 6 00:35:15 2008 -0400 15891 15892 Remove un-needed dac check for single crtc cards 15893 15894 Connector tables should be setup properly already and this 15895 check breaks systems with 2 DACs and 1 crtc like the es1000. 15896 15897commit 001c535687d8588873037ee5363d0a709f44b418 15898Author: Alex Deucher <alexdeucher@gmail.com> 15899Date: Tue Aug 5 17:14:48 2008 -0400 15900 15901 IGP: fix typo in IGP quirk handling 15902 15903commit eb65ddf70d182b6457e1ef5ebb820456039e8f6d 15904Author: Alex Deucher <alexdeucher@gmail.com> 15905Date: Tue Aug 5 15:01:33 2008 -0400 15906 15907 IGP: attempt to get external TMDS working 15908 15909 RS4xx chips have a tmds init table in the mobile info table 15910 15911commit 8b8990917809b9a35c6e9c1b9e3b12ff81c6dbb3 15912Author: Alex Deucher <alexdeucher@gmail.com> 15913Date: Tue Aug 5 12:16:06 2008 -0400 15914 15915 IGP: Updated quirk handling 15916 15917 - Hopefully finally fix DDC on RS4xx chips 15918 - RS3xx/RS4xx do not have internal TMDS 15919 - general quirk handling cleanup 15920 - we don't currently support mm gpio for external TMDS 15921 so return false. 15922 15923commit cb0deba5412a575d36f2f99377120b123506c946 15924Author: Calvin Fong <hoiwai930@gmail.com> 15925Date: Mon Aug 4 02:04:15 2008 -0400 15926 15927 IGP: add support for NTSC tv-out on legacy IGP chips 15928 15929 Patch from Calvin with some cleanups from me. 15930 No support for PAL yet. 15931 15932commit 942b18aca91819fa65d853cd15ffd1cd720cbd68 15933Author: Dave Airlie <airlied@redhat.com> 15934Date: Mon Aug 4 14:38:21 2008 +1000 15935 15936 radeon: make r600 use i2c table lookup for ddc. 15937 15938 This may fix DDC on rv770 cards. 15939 15940commit 1f3eee3682f3598a303c9c3accfbe01b245cacf9 15941Author: Alex Deucher <alexdeucher@gmail.com> 15942Date: Tue Jul 29 20:29:32 2008 -0400 15943 15944 Change prim types for exa and textured video to help avoid tearing 15945 15946 - r1xx - switch from tri fan to rect list 15947 - r2xx/r3xx/r4xx/r5xx - switch from tri fan to quad list 15948 15949commit 5b5441f8cc119db0d1e03dd35bd06015a26270dd 15950Author: Brice Goglin <bgoglin@debian.org> 15951Date: Tue Jul 29 23:31:14 2008 +0200 15952 15953 Increase default GART size to 32MB on >=R300 15954 15955commit 0a505297f09aefb1b4432176a263bfdf6f256f77 15956Author: Alex Deucher <alexdeucher@gmail.com> 15957Date: Mon Jul 28 17:12:41 2008 -0400 15958 15959 Fix error in driver connector table for powerbook w/ vga 15960 15961commit d5799ac53c3e1b2ca1da97780b171a44060c3aad 15962Author: Alex Deucher <alexdeucher@gmail.com> 15963Date: Mon Jul 28 11:09:10 2008 -0400 15964 15965 Add quirk for oem x300 card 15966 15967 - debian bug 492457 15968 15969commit 64dbadddcf6d069c0472f37afeab89d3e31e937d 15970Author: Owain Gordon Ainsworth <oga@openbsd.org> 15971Date: Mon Jul 28 10:40:58 2008 +1000 15972 15973 radeon: don't call ioctl unless DRI is initialised 15974 15975commit 1c5858484da4fb1c9bc3ac3b4d7a97863ab99730 15976Author: Alex Deucher <alexdeucher@gmail.com> 15977Date: Mon Jul 21 23:47:45 2008 -0400 15978 15979 First pass at InitDispBandwidth() for AVIVO chips 15980 15981 - support for LB allocation 15982 - MC priority bumps for display1/2 on RV515 variants and RS690 15983 If you are having display underflow problems (flickering on sides of 15984 screen in high res modes, etc.) on RV515 or RS690 boards, try setting: 15985 Option "DisplayPriority" "HIGH" in your config. 15986 - still no support for full display watermark programming yet 15987 15988 Something similar might be useful in rhd as well. 15989 15990commit b0378bb145c8a915c943bef7d17f2cdecfccc891 15991Author: Alex Deucher <alexdeucher@gmail.com> 15992Date: Mon Jul 21 13:47:09 2008 -0400 15993 15994 Interlaced mode fixups for AVIVO chips 15995 15996commit c18fad622a3c4f9572051120d83af68b625b5686 15997Author: Alex Deucher <alexdeucher@gmail.com> 15998Date: Mon Jul 21 10:36:48 2008 -0400 15999 16000 Clear display priority bits before resetting them 16001 16002commit dc231ff8e063313d2bcf5acccad67a9f8a7e3314 16003Author: Alex Deucher <alexdeucher@gmail.com> 16004Date: Mon Jul 21 10:30:41 2008 -0400 16005 16006 Clean up legacy display watermark setup 16007 16008 - makes crtc1 and crtc2 watermark setup independant. 16009 - fixes the case where only crtc2 is active 16010 16011commit f9034214f070fe3054cd9cefd7034252de234f38 16012Author: Michel Dänzer <michel@tungstengraphics.com> 16013Date: Mon Jul 21 09:09:02 2008 +0200 16014 16015 Call DRM modeset ioctl after the IRQ has been initialized. 16016 16017 This lets the DRM know it can safely disable the vblank interrupts. 16018 16019commit ecb6347a3d7071890600c98c2addef3a8ca260ee 16020Author: Alex Deucher <alexdeucher@gmail.com> 16021Date: Sat Jul 19 11:34:16 2008 -0400 16022 16023 Add oem quirk for external tmds setup on Dell Inspiron 8600 16024 16025 Noticed by fnord42 on IRC. 16026 16027commit df53d12a06fad41f00cff849458cb358ab5e2098 16028Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 16029Date: Thu Jul 17 14:37:05 2008 +1000 16030 16031 Fix console switch on R500 16032 16033 This patch fixes the console switch for me on R5xx. 16034 16035 There are two aspects to it: 16036 16037 - Fix the ordering of avivo_restore() to better match what's 16038 happening in the driver & ATOM, properly locking/unlocking and 16039 only enabling the CRTCs after everything has been properly 16040 programmed. 16041 16042 - Don't ASIC_INIT if the card has any CRTC enabled. This is the 16043 best I came up with for avoiding spurrious ASIC_INIT on cards that 16044 -are- POSTed but don't have the BIOS coming from c0000 on x86. The 16045 problem with spurrious ASIC_INIT is that we do it before we do 16046 RADEONSave(), so that screws up the console switch. 16047 16048 Note that I think we also should save/restore the palette, I don't think 16049 we do. right now, it's a minor issue for me because I fixed offb to be 16050 able to set it on AVIVO's but it might still have to be done in the long 16051 run. 16052 16053 Tested with a VGA analog setup on DACA and a DVI setup on TMDS 0. I 16054 haven't tested any other combo but that should get us going. 16055 16056 Cheers, 16057 Ben. 16058 16059 Signed-off-by: Dave Airlie <airlied@redhat.com> 16060 16061commit e8c07270e1fc2ee455320370692aaf90abac2773 16062Author: Alex Deucher <alexdeucher@gmail.com> 16063Date: Tue Jul 15 17:50:32 2008 -0400 16064 16065 ATOM: fix for r4xx cards with no VRAM_Usage tables 16066 16067 fixes bug 15954 16068 16069commit da41e71efd41907d9347a902720bce60b32550e5 16070Author: Alex Deucher <alexdeucher@gmail.com> 16071Date: Mon Jul 14 13:21:38 2008 -0400 16072 16073 ATOM: Add support for UpdateCRTC_DoubleBufferRegisters cmd table 16074 16075 locks/unlocks the crtc/grph/mode regs before updating the crtc 16076 16077commit a5e0cf13dc7ace6cf0e44e18b73b9a9266e740ab 16078Author: Julien Cristau <jcristau@debian.org> 16079Date: Sun Jul 6 12:17:28 2008 +0200 16080 16081 Link with -lpciaccess and -ldrm if needed 16082 16083 This makes sure the driver ends up with a DT_NEEDED reference to 16084 the libraries it's using. 16085 16086commit f8da849fd89322a54cc4d722767da755358eab70 16087Author: Michel Dänzer <michel@tungstengraphics.com> 16088Date: Sun Jul 13 11:31:03 2008 +0200 16089 16090 Only declare local variable when it's used. 16091 16092commit a6db4dc65aec560f3b1f37ff73ec6b0bd8207295 16093Author: Dave Airlie <airlied@linux.ie> 16094Date: Sat Jul 12 18:47:07 2008 +1000 16095 16096 atombios: fix typo in mode conversion for panel 16097 16098commit 19a0d5879851eff23a3e7d5cdea55bd784051115 16099Author: Dave Airlie <airlied@linux.ie> 16100Date: Sat Jul 12 17:04:25 2008 +1000 16101 16102 atombios: use macro to get offset into table 16103 16104commit a9746114369d516072d841ec356ec3ba3d0be71a 16105Author: Dave Airlie <airlied@linux.ie> 16106Date: Sat Jul 12 10:46:36 2008 +1000 16107 16108 pciid: add radeon HD3850. 16109 16110 pci id from legume on #radeon 16111 16112commit ab14f725676e4d4e45278c64b03fe2d328a3e439 16113Author: Alex Deucher <alexdeucher@gmail.com> 16114Date: Fri Jul 11 19:05:00 2008 -0400 16115 16116 R300: NUM_FPU adjustments for VAP_CNTL 16117 16118commit 810c28cc2660b73e03e4d27be97988fc5321a06f 16119Author: Alex Deucher <alexdeucher@gmail.com> 16120Date: Fri Jul 11 15:31:57 2008 -0400 16121 16122 Make sure cursor offsets are properly aligned when using EXA 16123 16124commit 7e67d0163579a44f104e8f354a929ac9b33e4c21 16125Author: Alex Deucher <alexdeucher@gmail.com> 16126Date: Fri Jul 11 14:30:17 2008 -0400 16127 16128 Fix cursor with multi-head and rotation 16129 16130commit 9086d008fb8c3cde0633b90ce19ffbf4eded388d 16131Author: Alex Deucher <alexdeucher@gmail.com> 16132Date: Tue Jul 8 09:57:04 2008 -0400 16133 16134 Add quirk for Falcon NW laptop 16135 16136commit 61f82ace0210251beb0bcc492218a75a193e1deb 16137Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 16138Date: Tue Jul 8 21:58:43 2008 +1000 16139 16140 atombios: add support for other endians. 16141 16142 This is a cleaned up (in as much as atombios can be..) of benh's patch. 16143 16144 airlied - removed benh's debugging for now, it might need to be put back later.. 16145 16146commit b4d1a47b5c6eafda5e274398eebe2701b030f22e 16147Author: Dave Airlie <airlied@linux.ie> 16148Date: Tue Jul 8 11:09:01 2008 +1000 16149 16150 rv770 initial support 16151 16152 adds pci ids and one register from AMD code 16153 16154commit 7ae4cec8cc8c90aee5dc4fa7abcce22321d4f4eb 16155Author: Roland Scheidegger <sroland@tungstengraphics.com> 16156Date: Mon Jul 7 14:39:47 2008 -0400 16157 16158 clamp tex coords (r100/r200) for textured video 16159 16160 fixes bug 14904 16161 16162commit 879f9a37856642b337e3c0233e38fd5443338eef 16163Author: Alex Deucher <alexdeucher@gmail.com> 16164Date: Fri Jul 4 13:54:49 2008 -0400 16165 16166 Revert "rotate: fix rotation in conf file." 16167 16168 This reverts commit 22c6ca0d5d226c725ba448e5db63fdba9e32332c. 16169 16170 There seems to be an ordering issue in that causes a nasty hang 16171 if the DRI is enabled. This needs more investigation. 16172 16173commit 22c6ca0d5d226c725ba448e5db63fdba9e32332c 16174Author: Dave Airlie <airlied@redhat.com> 16175Date: Thu Jul 3 20:14:59 2008 +1000 16176 16177 rotate: fix rotation in conf file. 16178 16179 need to set modes after init accel architecture 16180 16181 reported-by: Christoph Bartoschek <bartoschek@or.uni-bonn.de> 16182 16183commit ce1a3edd52b647744cfa4433301befb437d211bb 16184Author: Dave Airlie <airlied@redhat.com> 16185Date: Thu Jul 3 20:05:54 2008 +1000 16186 16187 radeon: drop all use of CPMode. 16188 16189 We never test the other codepath and I don't think I've ever recommended it 16190 for anyone. 16191 16192commit c037b4ce8769ad840a257e22b1e4ad58b8ed96fa 16193Author: Wolke Liu <wolke.lui@amd.com> 16194Date: Tue Jul 1 10:45:39 2008 -0400 16195 16196 Add pci id for FireMV 2400 16197 16198commit 72992668ca96717bf69bcea95a5deeaf7c57e826 16199Author: Alex Deucher <alex@botchco.com> 16200Date: Mon Jun 30 14:09:09 2008 -0400 16201 16202 RS4xx: enable the DRI by default on all variants 16203 16204 Some RS4xx chips had the DRI disabled by default. 16205 With the recent drm and ddx changes these chips have been 16206 reported to work now with the DRI. 16207 16208commit 0378c8ed88c829f09b5c6d51314325eaf2685fdb 16209Author: Michel Dänzer <michel@tungstengraphics.com> 16210Date: Sat Jun 28 14:06:45 2008 +0200 16211 16212 Restore versioning of interface for Mesa DRI driver. 16213 16214 This doesn't have anything to do with the package version. 16215 16216commit cfaa23d925e3c062cf87ea844566ac11ea02d69e 16217Author: Alex Deucher <alex@botchco.com> 16218Date: Fri Jun 27 20:36:54 2008 -0400 16219 16220 RADEON: IGP VGA fixes take 2 16221 16222 It seems only RS300/350/400/480 are missing the primary dac 16223 See bug 15708 16224 16225commit 31c27ffcb3c4c5334cf726ecd4e293a678b2a1ea 16226Author: Alex Deucher <alex@botchco.com> 16227Date: Fri Jun 27 20:29:04 2008 -0400 16228 16229 Revert "IGP: attempt to fix VGA on IGP chips" 16230 16231 This reverts commit e78e8a21b4040cd7f1983c241c860d9209398396. 16232 16233commit af79a77e6d8206b18bba97cf4e0770921dc439a7 16234Author: Brice Goglin <Brice.Goglin@ens-lyon.org> 16235Date: Fri Jun 27 17:09:13 2008 +0200 16236 16237 use PACKAGE_VERSION_* for radeon version 16238 16239 now radeon shows verion 6.9.0 instead of 4.3.0. 16240 16241 RADEON_VERSION_MAJOR_TILED is dropped from ddxDriverMajorVersion 16242 since RADEON_VERSION_MAJOR is bigger now. 16243 16244commit c83fbdfa076c107012b7dfbbfbbb2feede00542b 16245Author: Alex Deucher <alex@botchco.com> 16246Date: Thu Jun 26 19:48:45 2008 -0400 16247 16248 Bump for 6.9.0 release 16249 16250commit 32e1d1daf4f68ea37624afaa2bc1ea7742e1163b 16251Author: Brad Smith <brad@comstyle.com> 16252Date: Thu Jun 26 10:22:03 2008 -0400 16253 16254 RADEON: sys/endian.h needs sys/types.h on BSDs 16255 16256 See bug 16512 16257 16258commit dca522355a9039eca6efaba3b36397b246800f94 16259Author: Brad Smith <brad@comstyle.com> 16260Date: Thu Jun 26 10:13:59 2008 -0400 16261 16262 RADEON: fix copy/paste error in accel code 16263 16264commit f4292e110105910d0c19bf2db28e2682b27af2c7 16265Author: Dave Airlie <airlied@redhat.com> 16266Date: Fri Jun 20 15:31:43 2008 +1000 16267 16268 Revert "Revert "ATOM: disable TV-out for now"" 16269 16270 Oops I really didn't mean to do this, I was testing something and it slipped 16271 past. 16272 16273 This reverts commit dd18caa4b2efc430eaae0c4362b65641f9bef440. 16274 16275commit bb1cfcbfbc1ace1f257fc3bf6401501f1a6da827 16276Author: Alex Deucher <alex@botch2.com> 16277Date: Wed Jun 25 15:50:16 2008 -0400 16278 16279 RADEON: adjustments to Jerome's last commit 16280 16281 - Flush caches and wait for idle after drawing 16282 - Make sure 3D is idle too (after composite or textured video) 16283 16284commit 07be302f6dba61dd15918963fae66502230c74b7 16285Merge: 8c9b8de 5245974 16286Author: Jerome Glisse <glisse@freedesktop.org> 16287Date: Wed Jun 25 10:30:45 2008 +0200 16288 16289 Merge branch 'master' of git://anongit.freedesktop.org/xorg/driver/xf86-video-ati 16290 16291commit 8c9b8de0373797cba6f0a27e6b6f461e1070fef9 16292Author: Jerome Glisse <glisse@freedesktop.org> 16293Date: Wed Jun 25 10:28:26 2008 +0200 16294 16295 radeon: flush & wait for 2d & dma idle after 2d blit 16296 16297 This should help to avoid 2d & 3d engine to step on each 16298 other dma transaction. 16299 16300commit 52459745ec05de88adbc087e9566fe6d97ef424b 16301Author: Dave Airlie <airlied@linux.ie> 16302Date: Wed Jun 25 17:09:11 2008 +1000 16303 16304 ati: hopefully fix byteswap mess for those other OSes. 16305 16306commit 67a6ac0001bc9d062aa426384a11a41fa7a1c09a 16307Author: Alex Deucher <alex@botch2.com> 16308Date: Tue Jun 24 21:06:37 2008 -0400 16309 16310 bump for rc release 16311 16312commit bd68507d2d66e03d8bcde5f6e7ea9b2dbfe8b8a0 16313Author: Alex Deucher <alex@botch2.com> 16314Date: Tue Jun 24 20:59:58 2008 -0400 16315 16316 RADEON: warning fix 16317 16318commit 8e534d69c9b19fc085f7c5ca9e18f5ea04f6fc12 16319Author: Alex Deucher <alex@botch2.com> 16320Date: Tue Jun 24 20:08:35 2008 -0400 16321 16322 RADEON: cleanups 16323 16324 - fix some warnings 16325 - RS400 and RS480 are separate families now (update default tmds and dac2 16326 tables) 16327 16328commit e78e8a21b4040cd7f1983c241c860d9209398396 16329Author: Alex Deucher <alex@botch2.com> 16330Date: Tue Jun 24 19:53:28 2008 -0400 16331 16332 IGP: attempt to fix VGA on IGP chips 16333 16334 VGA has never worked on some IGP chips. While the chip only has 16335 one DAC, it appears to use a mix of Primary DAC and TVDAC controls. 16336 16337 See bug 15708 16338 16339commit faa4b4b8dbe9b8a452cfa60d53874350bb04e0cb 16340Author: Jiří Paleček <jpalecek@web.de> 16341Date: Tue Jun 24 19:20:50 2008 -0400 16342 16343 RADEON: remove extraneous line from new pll code 16344 16345commit 72feaa37ea07620f5f2ead438dbc72a1c8883cd3 16346Author: Jiří Paleček <jpalecek@web.de> 16347Date: Mon Jun 23 15:53:58 2008 -0400 16348 16349 RADEON: PLL tweaks 16350 16351 Patch from Jiří Paleček (see debian bug 465864) with some tweaks 16352 by me. 16353 16354 - abort rather than programming bad dividers if no pll dividers can be found 16355 - improve the pll selection algorithm 16356 - in general, prefer lower ref dividers 16357 16358 I've tested this patch on a wide variety of chips (r1xx-r6xx) and clocks. 16359 16360commit 9c2f909ea437a63a408d2398ecabe0b378dbb982 16361Author: Alex Deucher <alex@botch2.com> 16362Date: Mon Jun 23 10:38:15 2008 -0400 16363 16364 RADEON: adjust randr crtc/output prepare/commit ordering 16365 16366 This fixes some occasional mode change problems with multiple heads active. 16367 It seems radeons generally like to turn on the whole output/crtc setup 16368 in one shot. 16369 16370commit aea9bf75cf0774afd3e65fcf3fd3851f5fb21ca3 16371Author: Alex Deucher <alex@botch2.com> 16372Date: Sat Jun 21 10:57:05 2008 -0400 16373 16374 0x1002 0x5657 is actually RV410 16375 16376 See bug 14289 16377 16378 com_bios_scratch.diff 16379 16380commit 38ce8a984f96056b7345bcc06505ba27e0e6d5b4 16381Author: Dave Airlie <airlied@redhat.com> 16382Date: Fri Jun 20 14:16:22 2008 +1000 16383 16384 legacy: use entity MMIO for dpms as this can crossover between zaphod infos 16385 16386commit 1a7d9bc53512b0a5240176c949e6abf1ae2fb1fd 16387Author: Dave Airlie <airlied@redhat.com> 16388Date: Fri Jun 20 14:14:21 2008 +1000 16389 16390 atombios: use MMIO from the entity not the info 16391 16392commit ef624b88903b1a87ef5b6388e18291f75776b93d 16393Author: Alex Deucher <alex@botch2.com> 16394Date: Thu Jun 19 18:20:52 2008 -0400 16395 16396 RADEON: fix read past the end of an array 16397 16398commit cfe814a481d8cf2005d738a0ca9782f1ed4177f5 16399Author: Dave Airlie <airlied@linux.ie> 16400Date: Fri Jun 20 07:51:27 2008 +1000 16401 16402 r600: don't add fb offset here to make shadowfb work. 16403 16404 discovered on irc with wpwrak. 16405 16406commit 5b323a2cbbc412b6dd4e782438b64ee996558d05 16407Merge: dd18caa eed2415 16408Author: Dave Airlie <airlied@linux.ie> 16409Date: Fri Jun 20 07:49:16 2008 +1000 16410 16411 Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 16412 16413commit eed241553748125e902c44dcc8cf8e0605be071b 16414Author: Alex Deucher <alex@botch2.com> 16415Date: Tue Jun 17 17:42:31 2008 -0400 16416 16417 radeon: X_ERROR -> X_WARNING for num gb pipes 16418 16419 This just means your drm is old, not fatal or anything 16420 16421commit dd18caa4b2efc430eaae0c4362b65641f9bef440 16422Author: Dave Airlie <airlied@linux.ie> 16423Date: Sat Jun 14 18:37:56 2008 +1000 16424 16425 Revert "ATOM: disable TV-out for now" 16426 16427 This reverts commit effa245914823371e052cd9aa1143a02350891e7. 16428 16429commit b416e97a1f16ef700ba1aaca317dee82b7a3cf64 16430Author: Alex Deucher <alex@botch2.com> 16431Date: Fri Jun 13 02:21:31 2008 -0400 16432 16433 RADEON: man page updates 16434 16435commit 88ce87d79069662de28ecb3c10c140919927d584 16436Author: Alex Deucher <alex@botchco.com> 16437Date: Thu Jun 12 15:10:47 2008 -0400 16438 16439 Bump for 6.9.0rc1 release 16440 16441commit ada41723fdc8cbeeda12bb4ae2d16e8d1ba215de 16442Author: Alex Deucher <alex@botch2.com> 16443Date: Wed Jun 11 14:29:36 2008 -0400 16444 16445 R3xx: use DSTCACHE_CTLSTAT rather than RB2D_DSTCACHE_CTLSTAT 16446 16447 According to the hw guys, you should use DSTCACHE_CTLSTAT to 16448 flush the 2D dst cache rather than RB2D_DSTCACHE_CTLSTAT. 16449 16450commit 3d70c398afa0831387af81559453b77c218a54f9 16451Author: Dave Airlie <airlied@redhat.com> 16452Date: Thu Jun 12 06:18:26 2008 +1000 16453 16454 atom: fix another endian define 16455 16456commit f1794e4c34d801371461afed37e7bb7d1fc79305 16457Author: Dave Airlie <airlied@redhat.com> 16458Date: Thu Jun 12 06:10:30 2008 +1000 16459 16460 atom: fixup includes in cd operations to get correct endian bits in atombios.h 16461 16462commit f29976d12cc2020cd80c90fa3fd4b25c031559ba 16463Author: Dave Airlie <airlied@ppcg5.localdomain> 16464Date: Wed Jun 11 19:33:02 2008 +1000 16465 16466 radeon: add lots of endian bits into the atombios parser 16467 16468commit 09eb220971b5d2bfd7d1ff6f552060967a133152 16469Author: Dave Airlie <airlied@ppcg5.localdomain> 16470Date: Wed Jun 11 19:21:22 2008 +1000 16471 16472 atombios: fixup atombios bitfields for endianness 16473 16474 reverse the bits for big endian 16475 16476commit b80a3e85bf9dbfc8d96a745876d75d38bea97c1a 16477Author: Alex Deucher <alex@botchco.com> 16478Date: Mon Jun 9 20:29:47 2008 -0400 16479 16480 R5xx: fix RADEONSetAgpBase() for R5xx chips 16481 16482 Is there any reason to still do this in the ddx? 16483 Maybe real old drms? 16484 16485commit ab7936708dfbee4d3676262f8010e001b73a4a38 16486Author: Alex Deucher <alex@botchco.com> 16487Date: Mon Jun 9 17:05:04 2008 -0400 16488 16489 RS4xx: RC410 chips are actually RS400 based 16490 16491commit a54ca3c699c103e2e3df87101dca929a3eee4377 16492Author: Dave Airlie <airlied@linux.ie> 16493Date: Thu Jun 5 18:31:53 2008 +1000 16494 16495 radeon: fix typo in memory calcs 16496 16497 noticed by ghepeu on irc. 16498 16499commit 6e4e6d2a8f29f92efc219dca24ea31d1f37d5a0f 16500Author: Alex Deucher <alex@botchco.com> 16501Date: Tue Jun 3 20:32:20 2008 -0400 16502 16503 RADEON: minor cleanups 16504 16505commit 9d3afbf5fa4110928a9f965df2733c79db92ea99 16506Author: Michel Dänzer <michel@tungstengraphics.com> 16507Date: Tue Jun 3 11:40:49 2008 +0200 16508 16509 Call DRM_IOCTL_MODESET_CTL ioctl to avoid problems with DRM post vblank-rework. 16510 16511commit effa245914823371e052cd9aa1143a02350891e7 16512Author: Alex Deucher <alex@botchco.com> 16513Date: Sun Jun 1 20:47:50 2008 -0400 16514 16515 ATOM: disable TV-out for now 16516 16517 It only sort of works in NTSC mode and isn't handled correctly 16518 in all cases yet. Until we sort out the remaining details leave it 16519 disabled. 16520 See bugs: 16186, 16178, 16185 16521 16522commit 8504c6b0e40477ee544ad7f5366d569bdc53d6f0 16523Author: Alex Deucher <alex@botch2.com> 16524Date: Fri May 30 11:19:03 2008 -0400 16525 16526 RADEON: update RADEONGetVRAMType() for newer chips 16527 16528commit 7cb695a329a9b543ad61af08c2d7d6eaf56de35e 16529Author: Alex Deucher <alex@botch2.com> 16530Date: Fri May 30 10:10:56 2008 -0400 16531 16532 RADEON: add Int10 option 16533 16534 On some radeons you can't read the bios without initializing int10. 16535 On ATOM-based secondary cards, intitializing int10 tends to hang the card. 16536 16537commit 714b2c63ec1c4a1410a6d521c03e9d1f90937c01 16538Author: Alex Deucher <alex@botch2.com> 16539Date: Thu May 29 02:31:00 2008 -0400 16540 16541 RADEON: don't enable legacy bios table based init yet 16542 16543 Needs more work and we need to figure out how best to decide 16544 when to use it. 16545 16546commit e20b08525a64888287ec4a369d8f7dbde95c655d 16547Author: Alex Deucher <alex@botch2.com> 16548Date: Wed May 28 14:43:47 2008 -0400 16549 16550 ATOM: fail of we are not able to get a bios image 16551 16552commit d4fa17b426b0cafac3184b2ea9b675e2ff154ee9 16553Author: Alex Deucher <alex@botch2.com> 16554Date: Wed May 28 09:50:02 2008 -0400 16555 16556 R1xx: Gamma fixes for overlay 16557 16558commit c7e6a50f60a04bbbf1c1ddbf4738a5f75c94f950 16559Author: Alex Deucher <alex@botchco.com> 16560Date: Wed May 28 09:05:19 2008 -0400 16561 16562 RADEON: only attempt to read vbios from legacy ISA space if it's primary 16563 16564commit faea008806802ec0e045754ec1eca492ebae320e 16565Author: Alex Deucher <alex@botch2.com> 16566Date: Tue May 27 18:36:01 2008 -0400 16567 16568 R3/4/5xx: use get_param to get the num_gb_pipes from the drm 16569 16570commit 965a5dbcd9dc4bf1cdd7f2bbdec15e9733b2e090 16571Author: Alex Deucher <alex@botch2.com> 16572Date: Tue May 27 16:48:41 2008 -0400 16573 16574 RADEON: improve support for secondary cards 16575 16576 this should fix bugs 16115, 16035 16577 16578commit 5f951a5573f0c7572230c9aa4d3f75d67f91ed71 16579Author: Alex Deucher <alex@botch2.com> 16580Date: Wed May 21 00:35:14 2008 -0400 16581 16582 ATOM: remove duplicate code 16583 16584commit edce33e87fb79a540d8c285f205d4c3f2c2bc9f4 16585Author: Dave Airlie <airlied@redhat.com> 16586Date: Wed May 28 06:43:40 2008 +1000 16587 16588 radeon: fix surface access on avivo chips. 16589 16590 This should fix VT switch on vesafb 16591 16592commit 307bd65a25ee425d8359dd30572b002ce2338f91 16593Author: Michael Babej <mbabej@redhat.com> 16594Date: Tue May 27 19:13:49 2008 +1000 16595 16596 r600: fix vt switch issue 16597 16598 RH BZ 441492 16599 16600commit bf48c9eb8ec592515be9d1732d60283af715674d 16601Author: Dave Airlie <airlied@linux.ie> 16602Date: Sat May 24 11:51:20 2008 +1000 16603 16604 atombios: for LVDS set use ref div as per legacy 16605 16606commit 04500c8419b1aeaeac7968492b82e1d7cad1d05b 16607Author: Alan Coopersmith <alan.coopersmith@sun.com> 16608Date: Tue May 20 20:34:31 2008 -0700 16609 16610 Strip ^M's from src/AtomBios/includes/ObjectID.h 16611 16612commit 521a0488d1c414209c3534dc242000faa332a441 16613Author: Alan Coopersmith <alan.coopersmith@sun.com> 16614Date: Tue May 20 20:33:37 2008 -0700 16615 16616 On non-gcc compilers, use C99's __func__ instead of gcc's __FUNCTION__ 16617 16618commit b7c80d0c86646105d2bce5d4d59ba6c45aa7cafc 16619Author: Tobias Diedrich <ranma+freedesktop@tdiedrich.de> 16620Date: Tue May 20 19:16:54 2008 -0400 16621 16622 R200/R300: fix gamma setup for overlay 16623 16624 More pending. See bug 16001 16625 16626commit 69423482e2e94637142a9ba675589a1449a346a8 16627Author: Alex Deucher <alex@botch2.com> 16628Date: Tue May 20 18:57:13 2008 -0400 16629 16630 RADEON: cleanup connection detection and remove duplicate code 16631 16632commit 55e4469f59c82bb5762673de5f3f27d18b0bd9a3 16633Author: Alex Deucher <alex@botch2.com> 16634Date: Tue May 20 17:46:58 2008 -0400 16635 16636 RADEON: enable cloning on multi-crtc cards 16637 16638commit 12f185634071980041aaac6265d89708b181b215 16639Author: Alex Deucher <alex@botch2.com> 16640Date: Tue May 20 16:05:41 2008 -0400 16641 16642 ATOM: Ignore invalid connector entries 16643 16644commit 19e97f74e39fc2b35727708ac429de33d0b70162 16645Author: Egbert Eich <eich@freedesktop.org> 16646Date: Tue May 20 11:16:15 2008 +0200 16647 16648 Change RMX code to follow the programming algorithm suggested by ATI. 16649 16650 ATI provides the following algorithm to calculate the RMX scaling ratios 16651 in its programming specs: 16652 when RMX_AUTO_RATIO_HORZ_INC set to 1,Horizontal auto ratio result trucated, 16653 and then incremented by 1. 16654 Horz_Ratio = ( ((Active display width in characters (including overscan) + 1) 16655 / (Panel width in characters)) x 4096 + 1 ) 16656 else 16657 Horz_Ratio = ( ((Active display width in characters (including overscan)) 16658 / (Panel width in characters)) x 4096 + 1 ) 16659 16660 when RMX_AUTO_RATIO_VERT_INC set to 1, Vertical auto ratio result trucated, 16661 and then incremented by 1. 16662 Vert_Ration = ( ((Active display width in characters (including overscan) + 1) 16663 / (Panel width in characters)) x 4096 + 1) 16664 else 16665 Vert_Ration = ( ((Active display width in characters (including overscan)) 16666 / (Panel width in characters)) x 4096 + 1) 16667 16668 This patch implements this behavor. Additionally it avoids the use of floats. 16669 16670commit a4f3d0088ba763ed8eab1d331959b5ecde8262e8 16671Author: Alex Deucher <alex@botch2.com> 16672Date: Tue May 20 11:42:53 2008 -0400 16673 16674 R3/4/5xx: fixup logic from last commit 16675 16676 Spotted by otaylor and MrCooper 16677 16678commit 130e55738047f2a073bcc47e3e1400f7b694a81b 16679Author: Alex Deucher <alex@botch2.com> 16680Date: Tue May 20 11:32:42 2008 -0400 16681 16682 R3/4/5xx: fix EXA rotation 16683 16684 xrandr uses PictOpSrc for rotation which we were falling back on since 16685 render semanties require alpha=0 for REPEAT_NONE when there is no alpha 16686 channel and there is a transform. If the dst has no alpha channel we 16687 should be ok I think. 16688 16689 Otayler and I discussed this on IRC. the more general fix would be 16690 to clip the dst to the src and mask for bounded ops like in/add/over 16691 in the pixel exact transform case. 16692 16693commit cc9f510770700228e5d597c872e926a4e99bd950 16694Author: Corbin Simpson <MostAwesomeDude@gmail.com> 16695Date: Wed May 14 16:49:01 2008 -0400 16696 16697 RADEON: missing stdint.h includes 16698 16699commit 71fa57f871dba03260dba2180ce1dab44048ac1a 16700Author: Alex Deucher <alex@cube.(none)> 16701Date: Wed May 14 13:36:38 2008 -0400 16702 16703 Add RS600 support 16704 16705commit 582c1a1b2c7b1032e9f9f54ca36100c57f580c5c 16706Author: Alex Deucher <alex@cube.(none)> 16707Date: Tue May 13 20:50:25 2008 -0400 16708 16709 RS4xx: Split out RS400 and RS480 as separate families 16710 16711 RS400 (intel based IGP) and RS480 (AMD based IGP) have different 16712 MC setups and need to be handled differently 16713 16714commit 708e7c98f636738fbcc47a597bc94b309a4dc1c4 16715Author: Michel Dänzer <michel@tungstengraphics.com> 16716Date: Tue May 13 11:00:38 2008 +0200 16717 16718 Reinstate CARD* types that are part of external interfaces. 16719 16720 The compiler pointed this out... 16721 16722commit 908b7b940e3ef296836bc94680ebb8ab67650e31 16723Author: Matt Turner <mattst88@gmail.com> 16724Date: Mon May 12 12:06:33 2008 -0400 16725 16726 Replace CARD{8,16,32} with uint{8,16,32}_t 16727 16728 As has been done with xf86-video-intel, replace all CARD* datatypes with 16729 uint*_t datatypes available from stdint.h. 16730 16731commit 87e66ce76430890ab4939ffcd42f72b9288eb598 16732Author: Avi Rozen <avi.rozen@gmail.com> 16733Date: Mon May 12 08:33:38 2008 -0400 16734 16735 RADEON: fix lockup on start 16736 16737 see debian bug 480312 16738 16739commit 94bf8f01bd43cb103fffecfe04d04a214f892baf 16740Author: Dave Airlie <airlied@redhat.com> 16741Date: Mon May 12 20:02:51 2008 +1000 16742 16743 radeon: fix set_cursor_colours remove fb offset 16744 16745commit 1d0f1d31e2ed1d91ee87cb3e02ce48c8c07aa418 16746Author: Dave Airlie <airlied@redhat.com> 16747Date: Mon May 12 15:44:35 2008 +1000 16748 16749 radeon: rs485 vs rs485m.. mobile? non-mobile? bi? 16750 16751 So it appears we have in the wild two chipsets with the same PCI ID (0x1002:0x5974) 16752 that are mobile and non-mobile. 16753 16754 the RH bug references is a desktop system. 16755 The Dell Vostro 1100 also has this chipset with LVDS bits. 16756 16757commit 25e0c3945a51ae8c14b3a847ec75a256e1397f24 16758Author: Alex Deucher <alexdeucher@gmail.com> 16759Date: Mon May 12 10:28:33 2008 +1000 16760 16761 radeon: add initial support for cloning outputs. (single-CRTC cards only) 16762 16763 airlied - This code was originally written by Alex, and I've modified it to 16764 only run on single-CRTC cards for now until we can test it some more. 16765 16766commit 94405eb1c9e4e0ababc6aef03b753d6ed9eb7838 16767Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 16768Date: Fri May 9 05:28:44 2008 -0400 16769 16770 RADEON: fixup ifdef from last commit 16771 16772commit 18e20bc22a55ecfee9798c01079d7b24d19f0051 16773Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 16774Date: Thu May 8 08:45:51 2008 -0400 16775 16776 RADEON: disable MMIO paths for EXA composite/texvid on IGP/R5xx 16777 16778 The MMIO paths eventually lead to a hang on r5xx/IGP. I haven't 16779 been able to find out why yet. 16780 16781commit c3532268875fd24e6519bea2fb1b814d612bbdb4 16782Author: Dave Airlie <airlied@linux.ie> 16783Date: Wed May 7 02:37:18 2008 +1000 16784 16785 radeon: fix zaphod EXA with texture video 16786 16787commit ffc437f3606ab8ceba1ff152e4bb08988a58b54c 16788Author: Dave Airlie <airlied@linux.ie> 16789Date: Wed May 7 02:30:28 2008 +1000 16790 16791 avivo: fix zaphod cursor in theory 16792 16793commit e36ef14e3a1087e1fe41baa26ade2937f396001f 16794Author: Dave Airlie <airlied@linux.ie> 16795Date: Wed May 7 01:39:28 2008 +1000 16796 16797 radeon: fix textured-xv on zaphod 16798 16799commit 8fc19bee27c0f151d2ab3354f6ac0992b358436d 16800Author: Dave Airlie <airlied@linux.ie> 16801Date: Wed May 7 01:38:24 2008 +1000 16802 16803 radeon: zaphod: fix render accel for EXA 16804 16805commit fc41b9042a5220a8419cc7b69ca3850cae6b903c 16806Author: Dave Airlie <airlied@linux.ie> 16807Date: Wed May 7 01:32:23 2008 +1000 16808 16809 radeon: fix EXA pixmap offset on zaphod 16810 16811commit 4568cb820d567c6909a4be956d7e79b91232535e 16812Author: Dave Airlie <airlied@linux.ie> 16813Date: Wed May 7 01:19:39 2008 +1000 16814 16815 radeon: zaphod fixes for pciaccess not allowing multiple MMIO maps 16816 16817commit ca81fa086b21633a7fd926833fb6d1d4fa080646 16818Author: Dave Airlie <airlied@linux.ie> 16819Date: Wed May 7 01:12:01 2008 +1000 16820 16821 radeon: zaphod fix for cursor on second head 16822 16823 We don't need to add fbOffset here as the mmap we have of the framebuffer 16824 starts half way. 16825 16826commit 24b60c8965f6a0b3f0c2bb1e7236b4d6642c5918 16827Author: Julien Cristau <jcristau@debian.org> 16828Date: Fri May 2 15:30:45 2008 -0400 16829 16830 Add a test for __GLIBC__ to CD_Common_Types.h. 16831 16832 Atombios redefines the standard types but the definitions conflict 16833 with the ones from glibc (causes build failures on GNU/Hurd 16834 and GNU/kFreeBSD). 16835 16836commit f051359ac09c6b9416e39b9ca7d9dc0880aa1557 16837Author: thegraze <thegraze@googlemail.com> 16838Date: Fri May 2 14:02:16 2008 -0400 16839 16840 ATOM: add support for DragonFlyBSD 16841 16842commit 3d469cbc3225d890a895dac7cbc1ab7e08054b48 16843Author: Alex Deucher <alex@cube.(none)> 16844Date: Wed Apr 30 18:33:04 2008 -0400 16845 16846 RADEON: lock the cursors when updating 16847 16848 this should fix occasional corruption seen when updating 16849 the cursor. 16850 16851commit 445b71021843665ba32f37b2ce5c9d2857c07cc7 16852Author: Alex Deucher <alex@cube.(none)> 16853Date: Tue Apr 29 21:01:41 2008 -0400 16854 16855 RADEON: assorted fixes 16856 16857 - free rotate pixmaps on VT switch 16858 - save crtc/output status so we only turn on 16859 crtcs/outputs if they are off 16860 - show/hide cursors when changing modes 16861 16862commit 070cce5255a5c311f9d8b85ec54bd56655014933 16863Author: Stephan Wolf <stephan@letzte-bankreihe.de> 16864Date: Mon Apr 28 11:26:37 2008 -0400 16865 16866 R3xx+: further fix for IGP chips 16867 16868 see bug 15538 16869 16870commit 211e0041c7fc2df494b77428553943a2b526ee4e 16871Author: Alex Deucher <alex@cube.(none)> 16872Date: Sun Apr 27 21:08:00 2008 -0400 16873 16874 IGP: fix EXA composite corruption 16875 16876commit 656b06bdde129ca4fc370f5a2cf7311c9179b0ff 16877Author: Alex Deucher <alex@botch2.com> 16878Date: Sun Apr 27 20:20:49 2008 -0400 16879 16880 RADEON: remove duplicate register define 16881 16882 Also add more bit defs to wait_until register 16883 16884commit 8a9820a3aa49bc667f90ac291a27e4d7b4ae38b3 16885Author: Alex Deucher <alex@botch2.com> 16886Date: Sun Apr 27 19:02:22 2008 -0400 16887 16888 RADEON: decrease crtc/output verbosity 16889 16890commit c5d62fa0e8f52c3264ff9db3ff10cdf5a806bfc0 16891Author: Owen Taylor <otaylor@huygens.home.fishsoup.net> 16892Date: Thu Apr 17 13:14:53 2008 +0200 16893 16894 Emulate repeats by drawing in tiles 16895 16896 When we can't turn on hardware repeats, because the texture 16897 is non-power-of-two, or has padding at the ends of lines, 16898 try to draw the image in multiple tiles rather than falling 16899 back to software. (We can only do this when there is no 16900 transform.) 16901 16902commit eeb7b74bb6c813b0e3afa4b704f6ffb0d0aab92b 16903Author: Owen Taylor <otaylor@huygens.home.fishsoup.net> 16904Date: Thu Apr 17 13:14:25 2008 +0200 16905 16906 Turn on wrapping when repeating on R100 + R200 16907 16908 Actually enable repeats for R100 and R200. This corresponds 16909 to a R300 change made in the patch in: 16910 http://bugs.freedesktop.org/show_bug.cgi?id=15333 16911 16912commit e511f39dfef503006cf249b9f6934091eaade9b5 16913Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 16914Date: Thu Apr 17 05:04:34 2008 -0400 16915 16916 R300+: move more common code into init3d() 16917 16918 - pre-load r3xx tex instructions 16919 - setup RS instructions in init3d() 16920 16921commit 99435b7c18d931ea620044d0fdb4cc93dfcc6331 16922Author: Owen Taylor <otaylor@redhat.com> 16923Date: Thu Apr 17 02:46:11 2008 -0400 16924 16925 Radeon: Omit mask coordinates 16926 16927 Adapted from Owen's patch on bug 15546 16928 This fixes the slowness with aatext on r300 16929 and may speed up other chips marginally. 16930 16931commit 37614e1db9a595fbe8a21d7a045895e11d272db9 16932Author: Alex Deucher <alex@botch2.com> 16933Date: Tue Apr 15 09:48:16 2008 -0400 16934 16935 fix up some things from the last commit 16936 16937commit 1286fe5ce1c77453d57817b9b26b1bdb32ca7bc8 16938Author: Alex Deucher <alex@botch2.(none)> 16939Date: Mon Apr 14 20:02:14 2008 -0400 16940 16941 R300+: properly setup vap_cntl 16942 16943 this fixes tcl/pvs on RV515 among other things 16944 16945commit f72a4b805db26f10f69330b88459cbeae661189b 16946Author: Alex Deucher <alex@botch2.com> 16947Date: Mon Apr 14 14:10:40 2008 -0400 16948 16949 EXA: Don't wait for 3D idle after each Composite() 16950 16951 wait in CompositeDone() instead 16952 16953commit 4cd4acf1092aeb696b086a382a033aee471d2de9 16954Author: Alex Deucher <alex@botch2.com> 16955Date: Mon Apr 14 11:50:59 2008 -0400 16956 16957 R300: move more common code to init3d() 16958 16959commit 3c523c9a07402e17dff588fad842224c57e98223 16960Author: Alex Deucher <alex@botch2.com> 16961Date: Mon Apr 14 11:21:42 2008 -0400 16962 16963 R3xx+: 3D engine documentation and minor cleanups 16964 16965 - document the R300 exa/textured video code 16966 - minor cleanups of textured video code to clarify meaning 16967 16968commit ce025bbb2496d4de94b8d4ac450c64441b64ee04 16969Author: Alex Deucher <alex@botch2.com> 16970Date: Sat Apr 12 21:22:03 2008 -0400 16971 16972 R300+: consolidate some tcl/non-tcl paths 16973 16974 - Move more code to init3d() 16975 - MMIO textured video seems more reliable now on newer chips 16976 16977commit 11b54a319c7c9dd52e3fb13372697059dafe1cd3 16978Author: Alex Deucher <alex@botch2.com> 16979Date: Sat Apr 12 16:50:22 2008 -0400 16980 16981 R3xx+: fix XAA + textured video on non-TCL path 16982 16983commit dd15a2f5906725116b8cd9954243099055e88e37 16984Author: Alex Deucher <alex@botch2.com> 16985Date: Sat Apr 12 16:49:03 2008 -0400 16986 16987 R3xx+: more fixes to 2D/3D engine init 16988 16989commit f3e68d4b7afd2e23675bf6361c496814c9cb4b94 16990Author: Alex Deucher <alex@botch2.com> 16991Date: Fri Apr 11 10:59:07 2008 -0400 16992 16993 Fix exa glyph corruption on newer chips 16994 16995commit b59686d6427cbf8b35e36b020cbbc6a0c5149b22 16996Author: Alex Deucher <alex@botch2.com> 16997Date: Fri Apr 11 10:15:25 2008 -0400 16998 16999 R300+: pre-load vertex programs in init3D() 17000 17001commit acc5833a35ad6c29a57f659607afb27eebdc2ea5 17002Author: Alex Deucher <alex@botch2.com> 17003Date: Thu Apr 10 17:52:52 2008 -0400 17004 17005 R3xx+: consolidate more tcl code 17006 17007commit 6f8f75bd19ef1919c0291141675be2d0e29b3251 17008Author: Alex Deucher <alex@botch2.com> 17009Date: Thu Apr 10 17:08:50 2008 -0400 17010 17011 R3xx+: consolidate some common 3D code 17012 17013commit 4b9234e1c4f7c7f419cb4245d64f3f9756c98bb6 17014Author: Alex Deucher <alex@botch2.com> 17015Date: Thu Apr 10 16:58:22 2008 -0400 17016 17017 R3xx+: tcl wip 17018 17019commit 865c463e3afb4759758f569132be8bf1386da5cc 17020Author: Alex Deucher <alex@botch2.com> 17021Date: Thu Apr 10 16:51:04 2008 -0400 17022 17023 R300+: textured video tcl cleanup 17024 17025commit 79c8d4ca36a1c3e5fe759d4ccc379c36af8f1676 17026Author: Alex Deucher <alex@botch2.com> 17027Date: Thu Apr 10 16:28:18 2008 -0400 17028 17029 RADEON: cleanup 17030 17031commit c4821a287d29a65f3bcb7d60dc72ec13c0384008 17032Author: Alex Deucher <alex@botch2.com> 17033Date: Thu Apr 10 16:20:17 2008 -0400 17034 17035 Revert "R3xx/R5xx: move more VAP, etc. state setup into common init3d() function" 17036 17037 This reverts commit 305a3310963a5dd07b3495015b06aa8c7c4e6b02. 17038 17039 Conflicts: 17040 17041 src/radeon_commonfuncs.c 17042 src/radeon_exa_render.c 17043 src/radeon_textured_videofuncs.c 17044 17045commit 0032c80bf30bab189204e3e6929e18a19d753138 17046Author: Alex Deucher <alex@botch2.com> 17047Date: Thu Apr 10 14:35:00 2008 -0400 17048 17049 RADEON: store tcl status in driver rec 17050 17051commit 9e2ffe66d106abe34a670d2edc9905ed62c485e8 17052Author: Alex Deucher <alex@botch2.com> 17053Date: Thu Apr 10 14:24:04 2008 -0400 17054 17055 R3xx+: use the right register for engine flush 17056 17057commit e1a9f26c2d2cbca9ad159e723ec95b95be1ef349 17058Author: Alex Deucher <alex@botch2.com> 17059Date: Thu Apr 10 14:12:15 2008 -0400 17060 17061 R3xx+: minor textured video fixes 17062 17063 - set shader output swizzling correctly 17064 - flush the right cache register on r3xx+ 17065 17066commit d79040906cd25bd494feb5901f465bbd050aa923 17067Author: Alex Deucher <alex@botch2.com> 17068Date: Thu Apr 10 13:59:58 2008 -0400 17069 17070 R3xx+: EXA/textured video fixes 17071 17072 - get pipe config based on GB_PIPE_SELECT where applicable 17073 (adapted from a similar patch from Dave) 17074 - only flush the dst cache after submitting vertices, freeing 17075 the cache lines stalls the pipe 17076 - no need to wait for 3D idle after submitting vertices 17077 - fix PURGE_CACHE() and PURGE_ZCACHE() for r3xx+ 17078 - fix depth 16 with EXA composite 17079 17080commit 0a96173cc38e506728d4c3f2dd383ba56e856578 17081Author: Michel Dänzer <michel@tungstengraphics.com> 17082Date: Mon Apr 7 18:15:34 2008 +0200 17083 17084 Increase default CP timeout. 17085 17086 Helps avoid spurious timeouts causing problems, see 17087 http://bugs.freedesktop.org/show_bug.cgi?id=15203 . 17088 17089commit 255fbf465f5e7db2609a5a151bfa810249db52a0 17090Author: Owen W. Taylor <otaylor@fishsoup.net> 17091Date: Thu Apr 3 02:25:41 2008 -0400 17092 17093 Fix rendering of transformed sources for REPEAT_NONE with EXA on >= R300. 17094 17095 Use the border color when possible, otherwise fall back to software. 17096 17097commit bc0407e53237d7968808110bc0243076377acf6e 17098Author: Alex Deucher <alex@cube.(none)> 17099Date: Fri Apr 4 18:40:16 2008 -0400 17100 17101 ATOMBIOS: Add support for DynamicClocks option 17102 17103 This patch adds support for dynamic clock gating and static 17104 power management using the atom command tables. In some cases 17105 the bios may already set this up during post, so YMMV. 17106 17107 I was only able to test on desktop cards, so I haven't tested 17108 to see how much (if any) power this saves or how it affects the 17109 thermal footprint. 17110 17111commit 5f5e21bb50555c56bd371576074c28c929307ff1 17112Author: Alex Deucher <alex@cube.(none)> 17113Date: Fri Apr 4 14:29:45 2008 -0400 17114 17115 RADEON: warning fixes 17116 17117commit c8e9a973aaded24aad567a0e36d0c78a05d6b2fd 17118Author: Alex Deucher <alex@cube.(none)> 17119Date: Fri Apr 4 14:26:19 2008 -0400 17120 17121 RADEON: add some quirks 17122 17123commit 091963a635b79884afe77c026eabb48972fbe175 17124Author: Alex Deucher <alex@botch2.com> 17125Date: Thu Apr 3 22:35:16 2008 -0400 17126 17127 Minor cleanup 17128 17129commit 950e9860643c20acde0eca4e4ff26baacc1f2b69 17130Author: Alex Deucher <alex@botch2.com> 17131Date: Thu Apr 3 22:11:48 2008 -0400 17132 17133 Revert "RADEON: memmap rework 1" 17134 17135 This reverts commit dd8ee1b444f4b973a1e0fadca5f943f2162b5e94. 17136 17137 Conflicts: 17138 17139 src/radeon.h 17140 src/radeon_driver.c 17141 17142 This rework seems to have caused more trouble than it was worth. 17143 17144commit 88a1fe4a94c5d11aff22734b21c89890e4428cd5 17145Author: Alex Deucher <alex@botch2.com> 17146Date: Thu Apr 3 22:04:43 2008 -0400 17147 17148 Revert "RADEON: remove driver rec copies of mc info, use save rec directly" 17149 17150 This reverts commit be0858a84fbdf74c0b844f462933a221d48c707d. 17151 17152 Conflicts: 17153 17154 src/radeon_driver.c 17155 17156commit c40a7aa3989576a8144213e2f31b892d21df8686 17157Author: Owen W. Taylor <otaylor@fishsoup.net> 17158Date: Thu Apr 3 14:43:55 2008 -0400 17159 17160 R3xx/R5xx: Fix pitch and clamp mode for repeating textures 17161 17162 - We can always use TXPITCH on a R300 even when repeating, 17163 (previous check for pitch matching width was also wrong) 17164 - Fix clamp mode for repeating textures to be WRAP 17165 17166commit a8593482c1f2e0f2dbac06c2e5325ba8c83ed9ff 17167Author: Dave Airlie <airlied@redhat.com> 17168Date: Wed Apr 2 09:58:05 2008 +1000 17169 17170 atombios: fix the dual-head hopefully. 17171 17172 tested on r600 with DVI and VGA 17173 17174commit 61d883d116fab3e9b513432d65e705afc5bb39f1 17175Author: Dave Airlie <airlied@redhat.com> 17176Date: Wed Apr 2 09:57:38 2008 +1000 17177 17178 Revert "Revert "atombios: fixup the width/height to use the mode values not the scrn ones"" 17179 17180 This reverts commit fc9af578997b6f22ee8b17e83f37d98689291b0e. 17181 17182 I see your revert and raise you one... 17183 17184commit fc9af578997b6f22ee8b17e83f37d98689291b0e 17185Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17186Date: Tue Apr 1 09:25:45 2008 -0400 17187 17188 Revert "atombios: fixup the width/height to use the mode values not the scrn ones" 17189 17190 This reverts commit c2b1c8b706a6c7c1fd0af80091958473133d54e7. 17191 17192 These registers hold surface size. Using the mode values 17193 breaks dualhead. 17194 17195commit 959509dd54de053f526b534e379a46934127231f 17196Author: Dave Airlie <airlied@redhat.com> 17197Date: Mon Mar 31 14:29:44 2008 +1000 17198 17199 radeon: use correct DDC interfaces so quirks get applied 17200 17201 Radeon seemed to mess up applying certain quirks, hopefully this will fix it. 17202 17203commit 18f5f1cd2f52afed89fc11ade0920f3dfea87306 17204Author: Dave Airlie <airlied@redhat.com> 17205Date: Mon Mar 31 14:11:49 2008 +1000 17206 17207 radeon: split quirks into separate function and new quirk for IBM RN50 17208 17209 Add a connector table quirk for the IBM RN50. 17210 17211commit c2b1c8b706a6c7c1fd0af80091958473133d54e7 17212Author: Dave Airlie <airlied@redhat.com> 17213Date: Sun Mar 30 11:44:14 2008 +1000 17214 17215 atombios: fixup the width/height to use the mode values not the scrn ones 17216 17217 this fixes it properly, legacy appears to be okay. 17218 17219commit c5edea3d8c9254d3a21e390b8309e39e4c9635db 17220Author: Dave Airlie <airlied@redhat.com> 17221Date: Sun Mar 30 11:11:22 2008 +1000 17222 17223 r500/r600: fix rotation to fill screen 17224 17225 I'm not 100% sure this is the correct fix (maybe we shouldn't be using scrn 17226 virtualX/Y)... this will fix it for now until I get more time. 17227 17228commit 9c62c820ba45ebc14d5f36f5d7885863800b6adb 17229Author: Michel Dänzer <michel@tungstengraphics.com> 17230Date: Fri Mar 28 12:37:29 2008 +0100 17231 17232 Include config.h, so FGL_LINUX can actually be defined when it's tested... 17233 17234commit a00d9260a85b94a522c442aee24bc5ea4dc31c5c 17235Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17236Date: Thu Mar 27 20:03:13 2008 -0400 17237 17238 RADEON: fix lid issues on AVIVO chips for real this time :) 17239 17240commit f0e89c09074b2c7e641f73692bb39b0bf68eb49c 17241Author: Alex Deucher <alex@botch2.com> 17242Date: Thu Mar 27 19:15:18 2008 -0400 17243 17244 Revert "RADEON: attempt to fix lid issues" 17245 17246 This reverts commit 9b4473c1d830b88866dd22e8174a07195bd6fcf4. 17247 This doesn't help. 17248 17249commit 1442d396b938049b83f009a78ddabe2bf85641b6 17250Author: Dave Airlie <airlied@redhat.com> 17251Date: Thu Mar 27 14:02:51 2008 +1000 17252 17253 radeon: size bios to max of bar vs 64k. 17254 17255 reported by dwmw2: rhbz 438299 17256 17257commit de2f609ff0004ef8b74727bfebc2c74fb91205ea 17258Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17259Date: Wed Mar 26 18:35:21 2008 -0400 17260 17261 AVIVO: no need to call PreinitXv() on AVIVO chips as they have no overlay 17262 17263commit 75884c257bc2bcfa5b498a77d4c403f09face036 17264Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17265Date: Wed Mar 26 18:16:47 2008 -0400 17266 17267 XAA: update message about render so as to not confuse users 17268 17269commit 9b4473c1d830b88866dd22e8174a07195bd6fcf4 17270Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17271Date: Wed Mar 26 18:01:29 2008 -0400 17272 17273 RADEON: attempt to fix lid issues 17274 17275 On some laptops the bios attempts to re-program the chip 17276 when a lid event comes in. This should hopefully prevent 17277 the bios from doing that. 17278 17279commit 8b144830fe9b4a0cee4745023de5e7d387070f60 17280Author: Alex Deucher <alex@samba.(none)> 17281Date: Tue Mar 25 01:15:05 2008 -0400 17282 17283 RV250: disable textured video due to HW bug 17284 17285 The YUV->RGB conversion in the texture engine is broken 17286 on RV250 so the colors come out wrong. 17287 17288commit 1789f11ab91633d3928f8b71988d51ff44bda9d1 17289Author: Alex Deucher <alex@samba.(none)> 17290Date: Mon Mar 24 19:03:30 2008 -0400 17291 17292 R3xx/R5xx: flush PVS state before enabling pvs-bypass 17293 17294commit 305a3310963a5dd07b3495015b06aa8c7c4e6b02 17295Author: Alex Deucher <alex@samba.(none)> 17296Date: Mon Mar 24 14:25:03 2008 -0400 17297 17298 R3xx/R5xx: move more VAP, etc. state setup into common init3d() function 17299 17300 Also some minor code cleanups 17301 17302commit 399b1d405e602c62d6deebea6d7e1f38886cd8e2 17303Author: Alex Deucher <alex@samba.(none)> 17304Date: Mon Mar 24 13:04:57 2008 -0400 17305 17306 R3xx/R5xx: use non VAP/TCP for textured video 17307 17308 Just extra state to emit. 17309 17310commit cd77ec18f32a7b36acb655c927bbfd7044019f97 17311Author: Dave Airlie <airlied@redhat.com> 17312Date: Mon Mar 24 18:42:21 2008 +1000 17313 17314 r300: don't bother with VAP/TCL for render. 17315 17316 We just send more data to the card to process per transaction, without getting 17317 any actual gains, as we already pre-compute the vertices without needing 17318 any clipping or transforms from the card. 17319 17320 Perhaps some stuff could be done on-card, but so far the code is a lot 17321 faster if we avoid sending this extra info. 17322 17323 pre: 150000 glyphs/sec 17324 post: 185000 glyphs/sec 17325 17326commit 301c6739b88676a0c78fc72194e993f894b8dc28 17327Author: Alex Deucher <alex@botch2.com> 17328Date: Sun Mar 23 11:14:02 2008 -0400 17329 17330 RS4xx: Revert back to previous fifo settings for now 17331 17332 Setup of these registers needs more investigation. 17333 17334commit 9bea60b3eb378de5e1d44cc02a2763f4feae7882 17335Author: Alex Deucher <alex@botch2.com> 17336Date: Sat Mar 22 11:46:15 2008 -0400 17337 17338 RS4xx: more work on disp/disp2 fifo setup 17339 17340commit 90f11c3986c28daa7b600b9662da145af325d264 17341Author: Alex Deucher <alex@botch2.com> 17342Date: Sat Mar 22 11:29:51 2008 -0400 17343 17344 RS4xx: missed this on the last commit. 17345 17346commit 6d5066a451017a2683addc9e2496987626795dda 17347Author: Alex Deucher <alex@samba.(none)> 17348Date: Fri Mar 21 16:21:54 2008 -0400 17349 17350 RS4xx: attempt to set up disp/disp2 fifos correctly 17351 17352 If you have an XPRESS chip, please test!!! 17353 17354commit fb1cffac05ae20c8365b25a2042b0ae961880faf 17355Author: Alex Deucher <alex@samba.(none)> 17356Date: Fri Mar 21 15:24:36 2008 -0400 17357 17358 RS4xx: attempt to fix TMDS/DVO support 17359 17360 XPRESS chips added a second set of FP control registers. 17361 I don't have the hw to test however. 17362 17363commit 5e3b21284482df9974c9a58f248f0100def2bb0c 17364Author: Alex Deucher <alex@samba.(none)> 17365Date: Wed Mar 19 19:15:05 2008 -0400 17366 17367 Disable the setting of HARDWARE_CURSOR_BIT_ORDER_MSBFIRST 17368 17369 See bug 11796 17370 17371commit 17cd42ed31814ba329a6a68edd0d75390a7da40e 17372Author: Matt Turner <mattst88@gmail.com> 17373Date: Wed Mar 19 18:17:10 2008 -0400 17374 17375 Enable BSR in Log2 functions 17376 17377 This patch edits RADEONLog2 and ATILog2 to use the x86 BSR instruction instead 17378 of looping through bits. It should provide a somewhat of a speed increase in 17379 this function on x86 and AMD64 architectures. 17380 17381 Note: the BSR instruction was added with the 80386 CPU and is therefore not 17382 compatible with earlier CPUs, though I highly doubt it's even possible to use a 17383 286 in conjunction with a Radeon. 17384 17385 The inline assembly also works with Intel's compiler (icc). 17386 17387commit c83827b4d2b6f03c54429e757a756eb99ff8be6b 17388Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 17389Date: Wed Mar 19 17:58:34 2008 -0400 17390 17391 [PATCH] Compile warning fixes. 17392 17393 Minor changes to avoid declarations mixed with code. 17394 Ansified functions with empty prototype to specify they don't 17395 receive arguments. 17396 Added some prototypes to radeon.h, and major reorder on radeon.h 17397 adding prototypes in alphabetical order and specifying to file that 17398 defines it. 17399 17400commit bed9754ad21d6c0a7f61067b04ba31c430a7cecb 17401Merge: 55e446b f71ac0e 17402Author: Alex Deucher <alex@samba.(none)> 17403Date: Wed Mar 19 16:06:41 2008 -0400 17404 17405 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into r3xx-render 17406 17407commit 55e446b5bc091e6c7b3c2e9ae20b45130555c246 17408Author: Alex Deucher <alex@samba.(none)> 17409Date: Wed Mar 19 13:15:32 2008 -0400 17410 17411 R3xx/R5xx: Make sure to clamp the output of the FS 17412 17413commit b6aa4279cbe68cc8e4523795e9714fb798b62d98 17414Author: Alex Deucher <alex@samba.(none)> 17415Date: Wed Mar 19 12:45:01 2008 -0400 17416 17417 R5xx: bump textured video limits to 4096 17418 17419commit 4a445a3e8c4c5ecd9d4ef8daa26906c3ceaa94a1 17420Author: Alex Deucher <alex@samba.(none)> 17421Date: Wed Mar 19 12:31:51 2008 -0400 17422 17423 RADEON: add new macros to distinguish between R3xx and R5xx 3D 17424 17425commit 85d0c9e8d22ccc72bec87b3fd44da5d7609293e0 17426Author: Alex Deucher <alex@samba.(none)> 17427Date: Wed Mar 19 12:07:33 2008 -0400 17428 17429 RADEON: fixed textured video with XAA and tiling 17430 17431commit f5951db7b3522e0fe6af7f46a170c9c9a60a9bff 17432Author: Alex Deucher <alex@samba.(none)> 17433Date: Wed Mar 19 12:01:50 2008 -0400 17434 17435 RV515: fix textured video and EXA Composite 17436 17437 There seems to be an issue with the PVS setup on RV515, but 17438 bypassing it seems to work fine. 17439 17440commit 13573879fe56368ad06234712b677c23fabc56c6 17441Author: Dave Airlie <airlied@redhat.com> 17442Date: Wed Mar 19 15:06:47 2008 +1000 17443 17444 r500: make it work from startup. 17445 17446 I'm not sure why this worked or what is going wrong here, really the 17447 VAP internal architecture escapes me :) 17448 17449commit d331dd64d644a18ec99a2136cd0943b5edca1f03 17450Author: Alex Deucher <alex@samba.(none)> 17451Date: Tue Mar 18 19:44:26 2008 -0400 17452 17453 R3xx/R5xx: remove extra return after last commit 17454 17455commit bc34df7a9c35cdd38c49d5c22471f3f487a33d6e 17456Author: Alex Deucher <alex@samba.(none)> 17457Date: Tue Mar 18 19:39:47 2008 -0400 17458 17459 R3xx/R5xx: switch an ErrorF() to RADEONFALLBACK() 17460 17461commit 6f03f8fe0ecf4181dcf125049cf63bece0451fb2 17462Author: Alex Deucher <alex@samba.(none)> 17463Date: Tue Mar 18 19:36:05 2008 -0400 17464 17465 R3xx: we only use 2 temps, not 3 17466 17467commit 8bb71ab4a3eb4fb6ef7f709e87c8df387cb70ee3 17468Author: Tilman Sauerbeck <tilman@code-monkey.de> 17469Date: Tue Mar 18 14:36:08 2008 -0400 17470 17471 R3xx/R5xx: fix up a8-src-something_with_colors 17472 17473commit c362591d9b496df30668543158e4de44de742dc3 17474Author: Alex Deucher <alex@samba.(none)> 17475Date: Tue Mar 18 11:15:17 2008 -0400 17476 17477 R3xx/R5xx: remove some cruft 17478 17479commit 89fe6d2c7d7471e6088558130f6e49f46c31dd47 17480Author: Dave Airlie <airlied@linux.ie> 17481Date: Tue Mar 18 09:43:43 2008 -0400 17482 17483 R5xx: fix typ in r5xx render accel 17484 17485 This gets render working on r5xx 17486 17487commit 79b40ebcd8dedfc83e484c1024beeeaccc6124f3 17488Author: Alex Deucher <alex@samba.(none)> 17489Date: Tue Mar 18 02:46:49 2008 -0400 17490 17491 R5xx: first pass at render support (untested) 17492 17493commit 71292c8f193230255d1d980c2e996bb01d04fab6 17494Author: Alex Deucher <alex@samba.(none)> 17495Date: Tue Mar 18 00:45:37 2008 -0400 17496 17497 R5xx: bump tex/dst limits to 4096 17498 17499commit 30b52f8aa6a471455284f59b5b27252743892b13 17500Author: Alex Deucher <alex@samba.(none)> 17501Date: Mon Mar 17 23:20:10 2008 -0400 17502 17503 R3xx/R5xx: whitespace cleanup and cruft removal 17504 17505commit 9c9f1b538ed710c3066775fba0a8e936b63087b1 17506Author: Alex Deucher <alex@samba.(none)> 17507Date: Mon Mar 17 23:01:37 2008 -0400 17508 17509 R3xx: get masks working and cleanup 17510 17511 RS offset was wrong for mask texture 17512 17513commit ef94febd74f8ee63081b61e42f093a5a2b8fbf1e 17514Author: Alex Deucher <alex@samba.(none)> 17515Date: Mon Mar 17 22:27:19 2008 -0400 17516 17517 R3xx: minor adjustments 17518 17519commit f71ac0e40b9d950bcb3bba42a75d41f45b6ed1bf 17520Author: Alban Browaeys <prahal@yahoo.com> 17521Date: Mon Mar 17 20:48:48 2008 -0400 17522 17523 RADEON: Revert to old behavior when resetting the memmap on VT switch 17524 17525 Not sure why this needs to be done twice. Should fix bug 14980 17526 Probably needs more investigation. 17527 17528commit bedbbf196dc97ee5142e7dfae16fb6f317fca5a7 17529Author: Alex Deucher <alex@samba.(none)> 17530Date: Mon Mar 17 20:16:25 2008 -0400 17531 17532 R3xx: some progress 17533 17534commit af0e626c132de2dd9958fec657fcc85d4c0fe5e1 17535Author: Alex Deucher <alex@samba.(none)> 17536Date: Mon Mar 17 18:07:12 2008 -0400 17537 17538 R3xx: fix errant w 17539 17540commit 29ea5bfc0eb3194e2454fc3ee863df54f0300880 17541Author: Alex Deucher <alex@cube.(none)> 17542Date: Mon Mar 17 16:41:57 2008 -0400 17543 17544 RADEON: fix typo in RADEONAdjustMemMapRegisters() 17545 17546commit ab317e85c5ab1a249a510c34aeb3a908be1a66fc 17547Author: Alex Deucher <alex@cube.(none)> 17548Date: Mon Mar 17 15:28:09 2008 -0400 17549 17550 RADEON: make sure var is initialized properly in RADEONAdjustMemMapRegisters() 17551 17552commit 208d307227e15f37a6af5194398ed23266ff743a 17553Author: Dave Airlie <airlied@linux.ie> 17554Date: Sun Mar 16 19:39:23 2008 +1000 17555 17556 radeon: the 0x5974 appears to be a mobility chip... 17557 17558 After debugging with partymola on #radeon, adding this allowed his 17559 Dell Vostro 1000 to work properly 17560 17561commit 9bc7c2ec4048e1677547c1d60c51ccb954f7589a 17562Author: Alex Deucher <alex@samba.(none)> 17563Date: Fri Mar 14 20:12:22 2008 -0400 17564 17565 R3xx: odds and ends... 17566 17567 still not working. 17568 - swizzle US output for BGR formats 17569 - no need to write to temps in ALU ops, 17570 write to output only 17571 - flush the PVS before updating 17572 17573commit 96bea7906c4706fcd57a9cd8f1ce3feab6ac676d 17574Author: Alex Deucher <alex@samba.(none)> 17575Date: Fri Mar 14 15:59:36 2008 -0400 17576 17577 R3xx: theoretical support for component alpha 17578 17579 masks are still broken so... 17580 17581commit cffe3dcc8991cd7c457a9c1a9f41055aa9ea3436 17582Author: Alex Deucher <alex@samba.(none)> 17583Date: Fri Mar 14 14:37:43 2008 -0400 17584 17585 R3xx: VS WIP 17586 17587commit b73f52a50dfd6ff8d92f04d6b510c39582c6ac67 17588Author: Alex Deucher <alex@samba.(none)> 17589Date: Fri Mar 14 14:20:49 2008 -0400 17590 17591 R3xx/R5xx: enable VS for mask texture 17592 17593commit 569a14ca9be1e18fe9921edc816ac3dc32d6cca7 17594Author: Alex Deucher <alex@samba.(none)> 17595Date: Fri Mar 14 13:32:12 2008 -0400 17596 17597 R3xx/R5xx: Fix magic numbers in vertex shaders 17598 17599commit 4878997529601d62e257aa1c9112bd460561de73 17600Author: Alex Deucher <alex@samba.(none)> 17601Date: Thu Mar 13 21:23:40 2008 -0400 17602 17603 R3xx: make sure to set the FS code size correctly 17604 17605commit 22f46b88ef05afb6a6b6d70007ac4980a446430e 17606Author: Alex Deucher <alex@samba.(none)> 17607Date: Thu Mar 13 20:25:33 2008 -0400 17608 17609 R3xx: attempt to setup the rasterizer properly for mask texture 17610 17611 Not working yet 17612 17613commit 081fc9e892fa3d2e07b7db65b2e2719646255463 17614Author: Alex Deucher <alex@samba.(none)> 17615Date: Thu Mar 13 18:38:26 2008 -0400 17616 17617 R3xx: more mask work 17618 17619commit 2bf0236c03538ace3ce6d0e68f0829fc47d1385b 17620Author: Alex Deucher <alex@samba.(none)> 17621Date: Thu Mar 13 18:32:00 2008 -0400 17622 17623 R3xx: enable composite for non-mask cases 17624 17625commit 74286ba41302107d2fc626fee2181f7c4bc18164 17626Author: Alex Deucher <alex@samba.(none)> 17627Date: Thu Mar 13 18:25:32 2008 -0400 17628 17629 R3xx: add basic mask support 17630 17631commit a2bbe10d866567911b68f222b4758624bfe9bf84 17632Author: Alex Deucher <alex@samba.(none)> 17633Date: Thu Mar 13 18:16:53 2008 -0400 17634 17635 R300: setup source selects and output swizzling 17636 17637commit b9974ecce7d1932595226004858b08a7a6b188dc 17638Author: Alex Deucher <alex@samba.(none)> 17639Date: Thu Mar 13 17:35:38 2008 -0400 17640 17641 R3xx: set the texture id and add some register info 17642 17643commit 0ef700b7da5e554a0d0d166f3fde85ff45c9eb1f 17644Author: Alex Deucher <alex@samba.(none)> 17645Date: Thu Mar 13 17:02:25 2008 -0400 17646 17647 R3xx/R5xx: enable blending 17648 17649commit b35c09a597c93a1d9f06ef0091c96822b0653f98 17650Author: Dave Airlie <airlied@redhat.com> 17651Date: Thu Mar 13 18:42:29 2008 +1000 17652 17653 xv: fixup XAA on r500 textured video 17654 17655 the XAA area should never end up tiled. This may break with nooffscreen pixmaps 17656 17657commit d4446461c3630caff166456c351ace34f57cc119 17658Author: Matt Turner <mattst88@gmail.com> 17659Date: Tue Mar 11 21:20:53 2008 -0400 17660 17661 Properly fix uninitialized variables warnings 17662 17663 According to commit 9fd13e6773371c82b9799a5bda7c96ffa5cafe8c to 17664 xf86-video-intel by Kristian Høgsberg, there is a better way to fix the 17665 possibly initialized variables warnings. This patch will use Kristian's fix. 17666 17667commit 20adfd7390d9b1f100e0c4a14f175377b8335c82 17668Author: Alex Deucher <alex@cube.(none)> 17669Date: Tue Mar 11 20:09:35 2008 -0400 17670 17671 RADEON: enable output attributes that require a modeset immediately 17672 17673 This should fix bug 14915 17674 17675commit 53ba7f5771b0b53fb0d3bc29d64bdd3813756d10 17676Author: Alex Deucher <alex@cube.(none)> 17677Date: Tue Mar 11 19:12:40 2008 -0400 17678 17679 RADEON: fix vblank interrupts after VT switch or suspend/resume 17680 17681commit e946c097f0438afbea6f3dd37ee39d67d415708c 17682Author: Matt Turner <mattst88@gmail.com> 17683Date: Tue Mar 11 19:07:58 2008 -0400 17684 17685 [PATCH] Fix a few warnings 17686 17687commit 8e160508520c0a24ca90aad182f296461ca0d9b6 17688Author: Alex Deucher <alex@cube.(none)> 17689Date: Tue Mar 11 18:11:13 2008 -0400 17690 17691 DCE3: add support for PCIEPHY (untested) 17692 17693commit fbded88a2925f9f049936dad0736721e7b84a6ee 17694Author: Alex Deucher <alex@cube.(none)> 17695Date: Tue Mar 11 14:10:31 2008 -0400 17696 17697 ATOM: remove some cruft 17698 17699commit 3263f6e4a410281d620c288a92bb4521f7b6fc06 17700Author: Alex Deucher <alex@cube.(none)> 17701Date: Tue Mar 11 14:05:48 2008 -0400 17702 17703 DCE3: enable DPMS on DIG ports 17704 17705commit eb90e235b58c94f3d4d75394725ab2fe246a42ff 17706Author: Alex Deucher <alex@cube.(none)> 17707Date: Tue Mar 11 13:53:54 2008 -0400 17708 17709 DCE3: adjust PLL for DCE3 chips 17710 17711 this fixes stability issues on digital outputs and certain modes. 17712 17713commit 552615ccc5360baafb8bb41698c1ca27816fd4b2 17714Author: Alex Deucher <alex@cube.(none)> 17715Date: Tue Mar 11 13:38:29 2008 -0400 17716 17717 ATOMBIOS: enable load detection by default on both DACs 17718 17719 Load detection is reliable with atom, so enable it by default 17720 on both DACA and DACB, rather than just DACA. 17721 17722commit 78b10487cf222c96f8944ba25e2ea970506b3535 17723Author: Alex Deucher <alex@cube.(none)> 17724Date: Tue Mar 11 13:16:00 2008 -0400 17725 17726 DCE3: add output attribute to enable/disable coherent mode 17727 17728 Enabled by default. The TMDS transmitter can be programmed 17729 slightly differently depending on the chips in the panel. If you 17730 have problems with tmds on a particular panel, try disabling it. 17731 17732commit d20be31c46fbec623af4c3628a7c603ceacf500f 17733Author: Alex Deucher <alex@botch2.(none)> 17734Date: Mon Mar 10 21:05:43 2008 -0400 17735 17736 RV550: MC setup is like RV515 not RV530 17737 17738commit 38606b08b68842fbcc81c233009c1117269f3be9 17739Author: Matthieu Herrb <matthieu@bluenote.herrb.net> 17740Date: Sat Mar 8 23:22:59 2008 +0100 17741 17742 Makefile.am: nuke RCS Id 17743 17744commit 9d710ee1a44cf2f3a948fbdbe17ef09521cbe744 17745Author: Alex Deucher <alex@cube.(none)> 17746Date: Fri Mar 7 15:09:14 2008 -0500 17747 17748 AVIVO: clean up some unused variables 17749 17750commit c28c30c9f3d7bfebfd56a5c982c96f0090982054 17751Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17752Date: Fri Mar 7 14:10:49 2008 -0500 17753 17754 RADEON: Fix crash in last commit 17755 17756commit c3a3635865e380c784a226c8ead069d4716d6b75 17757Author: Dave Airlie <airlied@redhat.com> 17758Date: Thu Mar 6 20:17:45 2008 -0500 17759 17760 RADEON: fix tiling/interlaced interaction with randr 1.2 17761 17762commit df1b94dc4eb1f35b636dbf2ec0ab1c2da9937c0d 17763Author: Alex Deucher <alex@botch2.(none)> 17764Date: Thu Mar 6 19:22:08 2008 -0500 17765 17766 DCE3: Ignore outputs with DIN connectors for now 17767 17768commit cb2dc19387c7b6494c47c76d683cf38a48700768 17769Author: Alex Deucher <alex@cube.(none)> 17770Date: Thu Mar 6 18:33:12 2008 -0500 17771 17772 AVIVO: fix typo from a previous commit 17773 17774 Leave tv dpms hook disabled or you may get bad interactions 17775 with the shared DAC 17776 17777commit 77355de48057e5e7e0d5b3f3cf5a7a92220a53b1 17778Author: Alex Deucher <alex@cube.(none)> 17779Date: Thu Mar 6 17:46:00 2008 -0500 17780 17781 AVIVO: don't add outputs for invalid connectors 17782 17783commit 600dbe080997a01ceaf6be86723189d518bc1281 17784Merge: 594743a 5b7875d 17785Author: Alex Deucher <alex@cube.(none)> 17786Date: Thu Mar 6 17:31:37 2008 -0500 17787 17788 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 17789 17790commit 594743a99811a8b0f391412892414fcd158eeb56 17791Author: Alex Deucher <alex@cube.(none)> 17792Date: Thu Mar 6 17:30:21 2008 -0500 17793 17794 AVIVO: fix up memsize detection for IGP chips 17795 17796commit 5b7875d0cbfbdbcd1515c4e942d30de298b49dff 17797Author: Doug Chapman <doug.chapman@hp.com> 17798Date: Thu Mar 6 14:31:06 2008 -0500 17799 17800 Bug #14826: Fix a bogus check around xf86SetOperatingState. 17801 17802commit 651fe23f4c650ed91843dec48db24e18e8b91219 17803Merge: 3de2dc8 766f464 17804Author: Adam Jackson <ajax@redhat.com> 17805Date: Thu Mar 6 14:28:27 2008 -0500 17806 17807 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 17808 17809commit 41171c25cd235bafad26bcbabced16ead4b8c54b 17810Author: Alex Deucher <alex@cube.(none)> 17811Date: Thu Mar 6 14:05:18 2008 -0500 17812 17813 DCE3.0: add support for crtc memreq table 17814 17815commit 766f464dfdfccadef23e4232f2bce5db22195513 17816Author: Alex Deucher <alex@cube.(none)> 17817Date: Thu Mar 6 13:35:43 2008 -0500 17818 17819 RADEON: take 2 on proper pragma pack support for bsds 17820 17821 See bug 14594. Based on suggestion by Henry Zhao 17822 17823commit a842ce9ca6494e724a7828cead9b61c9ef02b6aa 17824Author: Alex Deucher <alex@cube.(none)> 17825Date: Thu Mar 6 12:32:18 2008 -0500 17826 17827 DCE3.0: Minor fixups 17828 17829commit 8a1ba374033591c725a78923aa30829e4de2a5ae 17830Author: Alex Deucher <alex@cube.(none)> 17831Date: Thu Mar 6 09:53:51 2008 -0500 17832 17833 RADEON: option to override TVDAC adj values from bios with driver defaults 17834 17835 If you have a washed out image on the tv dac, try this option. 17836 Option "DefaultTVDACAdj" "TRUE" 17837 17838commit 0ed48f8f651a28e189f9fee8c6b593da0178d21c 17839Author: Alex Deucher <alex@cube.(none)> 17840Date: Wed Mar 5 18:41:01 2008 -0500 17841 17842 AVIVO: Initial support for DCE 3.0 using atombios 17843 17844 DACs are working well, DIG support (DVI, HDMI, LVDS, etc.) 17845 still has some issues. 17846 17847commit 2901e99f1942842856cd39c1dcc8b22f3cf7d9e3 17848Author: Alex Deucher <alex@botch2.(none)> 17849Date: Wed Mar 5 10:40:06 2008 -0500 17850 17851 RADEON: fix fetching of dac2 adj values from newer bios tables 17852 17853commit 74eb981287d76836327830bd51272f605a07e0cc 17854Author: Alex Deucher <alex@botch2.(none)> 17855Date: Mon Mar 3 12:02:44 2008 -0500 17856 17857 ATOMBIOS: fix atombios parser support on *bsd 17858 17859 bsd requires a different pragma pack than Linux. 17860 See bug 14594. 17861 17862commit f7769ea86e265f347eb58c517ccb5ef8b35eec27 17863Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 17864Date: Sun Mar 2 14:49:21 2008 -0500 17865 17866 [PATCH] Ensure symbols used by other modules are visible. 17867 17868 The xf86-video-ati drivers are one of the cases where LoaderSymbol is 17869 widely used in some obscure ways. This patch fixes the problem, and 17870 allows compiling with -fvisibility=hidden. 17871 17872commit a4398ac3ad77216f2c8aa628425bef5f2912a0a9 17873Author: Alex Deucher <alex@cube.(none)> 17874Date: Sat Mar 1 18:52:26 2008 -0500 17875 17876 RS6xx: change isIGP checks to CHIP_FAMILY_RS690 17877 17878 these paths are only relevant on RS6xx chips 17879 17880commit 67d4d04836c05293b844bc505f303cfb04c0f8a4 17881Author: Alex Deucher <alex@cube.(none)> 17882Date: Sat Mar 1 18:33:18 2008 -0500 17883 17884 RADEON: use xf86SetDesiredModes() in screeninit and enterVT 17885 17886 this should restore the proper output state on VT switches 17887 17888commit be0858a84fbdf74c0b844f462933a221d48c707d 17889Author: Maciej Cencora <m.cencora@gmail.com> 17890Date: Sat Mar 1 18:11:58 2008 -0500 17891 17892 RADEON: remove driver rec copies of mc info, use save rec directly 17893 17894 info->mc_* were used and the immediately copied into info->ModeReg 17895 ones. Just use the ModeReg copies directly. 17896 17897commit dd8ee1b444f4b973a1e0fadca5f943f2162b5e94 17898Author: Alex Deucher <alex@cube.(none)> 17899Date: Sat Mar 1 16:23:51 2008 -0500 17900 17901 RADEON: memmap rework 1 17902 17903 Don't restore memmap regs on every mode switch. 17904 Just do memmap save/restore/setup on server start and VT switch. 17905 17906commit 1f6a23000001f3d1c21b5c04f94714a8caa7aa8b 17907Author: Alex Deucher <alex@cube.(none)> 17908Date: Sat Mar 1 15:53:42 2008 -0500 17909 17910 RADEON: only restore legacy dac regs on legacy radeons 17911 17912commit dee6cef8e62d0651c00319e03eea92940fd24aa4 17913Author: Alex Deucher <alex@botch2.(none)> 17914Date: Sat Mar 1 14:39:32 2008 -0500 17915 17916 RS4xx: enable exa render accel and textured video 17917 17918 RS6xx paths seem to work fine on RS4xx 17919 17920commit 129f737efe4e8d1a368e7db4b063bdcd9339cb09 17921Author: Alex Deucher <alex@cube.(none)> 17922Date: Sat Mar 1 14:32:30 2008 -0500 17923 17924 AVIVO: save/restore regs by block 17925 17926 Save/Restore the entire block for each output. 17927 This should fix VT switch problems. 17928 17929commit b069aadaa63a95d7a71b5cfbab83577b49501094 17930Author: Alex Deucher <alex@botch2.(none)> 17931Date: Fri Feb 29 22:36:02 2008 -0500 17932 17933 AVIVO: LVDS panels need dithering enabled 17934 17935 Fixes bug 14760 17936 17937commit fe87bdee815372b4b4d7d4c705e34681625b90f2 17938Author: Alex Deucher <alex@botch2.(none)> 17939Date: Fri Feb 29 13:10:13 2008 -0500 17940 17941 AVIVO: disable pageflipping on avivo chips until we have proper drm support 17942 17943commit fb3678c7f511d539a51cd090cb8b5041d7d2ba26 17944Author: Alex Deucher <alex@botch2.(none)> 17945Date: Fri Feb 29 13:01:21 2008 -0500 17946 17947 R5xx: fix register count when sending fragment program for textured video 17948 17949commit a66d37d1a896ec934989592457c2beff8e6f1639 17950Author: Alex Deucher <alex@botch2.(none)> 17951Date: Fri Feb 29 04:07:05 2008 -0500 17952 17953 fix off-by-one in last commit 17954 17955commit e56062960be0c8d3947861dd5e0691fce6516b99 17956Author: Alex Deucher <alex@cube.(none)> 17957Date: Thu Feb 28 19:16:39 2008 -0500 17958 17959 AVIVO: save/restore scaler regs 17960 17961commit ae1c39a9b3e666404d0931679c9078c2e125a8bc 17962Author: Alex Deucher <alex@cube.(none)> 17963Date: Thu Feb 28 18:53:55 2008 -0500 17964 17965 RS6xx: rework output parsing 17966 17967 Turns out it's not as complex as I originially thought. 17968 IGP chips just have non-standard GPIO entires for DDC. 17969 17970commit d8d6c9fe4ae7e1ab67dd041a251e901d97c29ed6 17971Author: Alex Deucher <alex@botch2.(none)> 17972Date: Thu Feb 28 17:01:14 2008 -0500 17973 17974 RS6xx: fix typos in previous commit 17975 17976 Noted by Maciej Cencora on IRC 17977 17978commit 46547ae8bdbc5c10f1fd028b95ec4c5c31a5b318 17979Author: Alex Deucher <alex@cube.(none)> 17980Date: Thu Feb 28 14:29:30 2008 -0500 17981 17982 AVIVO: disable dithering on DFPs 17983 17984 This should fix the color banding some people have noticed. 17985 Also save/restore DDIA regs on RS6xx 17986 17987commit 72a53d6f20ac29b3baddb7d8af04f19b76d2e04f 17988Author: Michel Dänzer <michel@tungstengraphics.com> 17989Date: Thu Feb 28 17:38:04 2008 +0100 17990 17991 Handle EXA coordinate limits more cleverly. 17992 17993 Generally set the 2D engine limits, and only enforce the 3D engine limits in the 17994 CheckComposite hook. This should still prevent useless migration of pixmaps the 17995 3D engine can't handle but allows for basic acceleration of bigger ones. 17996 17997 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=14708 . 17998 17999commit 5249f450a2487475a95531603cc8668db2c21c33 18000Author: Michel Dänzer <michel@tungstengraphics.com> 18001Date: Thu Feb 28 12:23:58 2008 +0100 18002 18003 Fix 16 bit packed YUV XVideo playback on big endian systems with DRI disabled. 18004 18005 http://bugs.freedesktop.org/show_bug.cgi?id=14668 18006 18007commit e40d75fd8b2aece9dae8076fac822a4a83025fb2 18008Author: Alex Deucher <alex@samba.(none)> 18009Date: Wed Feb 27 22:53:10 2008 -0500 18010 18011 R500: fragment program clean up and magic number conversion 18012 18013commit 140dadba36b2191f0e18e41dd987785abd5f55d2 18014Author: Alex Deucher <alex@samba.(none)> 18015Date: Wed Feb 27 22:21:12 2008 -0500 18016 18017 R300: fix up magic numbers in fragment program 18018 18019commit e521476bb5e2dfabc93747e43eb911a8a101357e 18020Author: Alex Deucher <alex@botch2.(none)> 18021Date: Wed Feb 27 21:26:55 2008 -0500 18022 18023 R300/R400: bump up the clip limits for textured video 18024 18025 This allows up to 2560x2560 (hw limit) 18026 18027commit 10db46f11d7e1c055c9ad6034c65ad163dad17dc 18028Author: Alex Deucher <alex@samba.(none)> 18029Date: Wed Feb 27 15:28:50 2008 -0500 18030 18031 AVIVO: make sure we select the right LUT for each crtc 18032 18033commit ea944f38dcfd871b27345698afea1cb986ecb049 18034Author: Alex Deucher <alex@samba.(none)> 18035Date: Wed Feb 27 14:37:52 2008 -0500 18036 18037 R300+: update RADEONCP_REFRESH() to reflect new location of scissor regs 18038 18039commit b865faf95666e2172c3eec143f77fe9c524e4983 18040Author: Alex Deucher <alex@samba.(none)> 18041Date: Wed Feb 27 14:05:44 2008 -0500 18042 18043 R100/R200: move r100/r200 specific 3D setup into appropriate blocks 18044 18045 R3xx+ doesn't have these regs. 18046 18047commit ee485ee0fc550414cd39cdb8fd2769fe0bc996a8 18048Author: George Sapountzis <gsap7@yahoo.gr> 18049Date: Wed Feb 27 18:41:31 2008 +0200 18050 18051 radeon: pass distcheck 18052 18053commit ff110c781299fb729e93572ba3d630ca59e8127d 18054Author: George Sapountzis <gsap7@yahoo.gr> 18055Date: Wed Feb 27 18:41:17 2008 +0200 18056 18057 add warning for r128 18058 18059commit b4a344f36332a12a51c9e896d93e9cb0672d31ef 18060Author: George Sapountzis <gsap7@yahoo.gr> 18061Date: Wed Feb 27 18:40:50 2008 +0200 18062 18063 add warning for mach64 18064 18065commit 81ce4097c12ab64a52d0992f981f588f5a297b7b 18066Author: George Sapountzis <gsap7@yahoo.gr> 18067Date: Wed Feb 27 18:40:23 2008 +0200 18068 18069 drop r128 18070 18071commit 873ff2033f1d0b84c91c48e829aa945f41b490b7 18072Author: George Sapountzis <gsap7@yahoo.gr> 18073Date: Wed Feb 27 18:40:09 2008 +0200 18074 18075 drop r128 (build system) 18076 18077commit 2b8b2676e7b2fd2952f5307a934ef0a11006f838 18078Author: George Sapountzis <gsap7@yahoo.gr> 18079Date: Wed Feb 27 18:39:49 2008 +0200 18080 18081 drop mach64 18082 18083commit 05bbbec11a04e7f6321b92c9dbec4ad4c0d130ff 18084Author: George Sapountzis <gsap7@yahoo.gr> 18085Date: Wed Feb 27 18:39:32 2008 +0200 18086 18087 drop mach64 (build system) 18088 18089commit 68888189cf8d460ef6f8f2f1431a6ffe9fcd8134 18090Author: Alex Deucher <alex@samba.(none)> 18091Date: Wed Feb 27 03:09:28 2008 -0500 18092 18093 RADEON: fix textured video uploads 18094 18095commit 7cfad216a3f1486be194c972b9b6bcf070358586 18096Author: Alex Deucher <alex@samba.(none)> 18097Date: Wed Feb 27 02:13:10 2008 -0500 18098 18099 RADEON: fix clipping for textured video 18100 18101commit a2dca1d68d751def34ef3c6f836574173737bf76 18102Author: Alex Deucher <alex@samba.(none)> 18103Date: Tue Feb 26 16:38:29 2008 -0500 18104 18105 RADEON: Convert textured video to use pipelined uploads 18106 18107commit b4fa1ce9d2da04d94521a82d3c2e95f0fe985ccc 18108Author: Alex Deucher <alex@samba.(none)> 18109Date: Tue Feb 26 14:45:25 2008 -0500 18110 18111 R300/R500: fix up clipping 18112 18113 Based on patch from Peter Zubaj. 18114 18115commit 44e527a117ab0a363135ff066c7f7e0c12e3dc89 18116Author: Alex Deucher <alex@samba.(none)> 18117Date: Tue Feb 26 14:33:24 2008 -0500 18118 18119 R300: fix cordinate clamping in render code 18120 18121 Based on Peter's fix for textured video 18122 18123commit 00ec17ad53d7ad43f19c9b723794ac1b8ef86826 18124Author: Peter Zubaj <pzubaj at marticonet.sk> 18125Date: Tue Feb 26 14:26:14 2008 -0500 18126 18127 R300: Fix clamping and pitch for textured video 18128 18129commit 4207793f9b6c891cb17ba715223d2002668885e3 18130Author: Alex Deucher <alex@samba.(none)> 18131Date: Tue Feb 26 13:16:22 2008 -0500 18132 18133 R300/R500: set the number graphics pipes properly 18134 18135 This should fix the checkerboard issues on r300/r350 cards. 18136 18137commit 823d8bf49e4c0a8c839354179942c5a1944c178d 18138Author: Alex Deucher <alex@samba.(none)> 18139Date: Tue Feb 26 12:25:14 2008 -0500 18140 18141 R300/R500: clean up magic numbers in textured video 18142 18143commit 8ea75b268f11794f4a7e7bac52cb256490ed3fd2 18144Author: Dave Airlie <airlied@linux.ie> 18145Date: Tue Feb 26 16:29:19 2008 +1000 18146 18147 regs: fix spelling properly 18148 18149commit f2816064a6c2c4c35ccba74b9aa80547e25c012e 18150Author: Dave Airlie <airlied@linux.ie> 18151Date: Tue Feb 26 16:28:24 2008 +1000 18152 18153 regs: fix spelling mistake 18154 18155 pointed out by plaes on irc 18156 18157commit 9d2ca30b90607085578dde1f314db663bd5f82ec 18158Author: Alex Deucher <alex@samba.(none)> 18159Date: Mon Feb 25 17:34:00 2008 -0500 18160 18161 R300/R500: clean up magic numbers in render code 18162 18163commit 153ad6fcf704cbf9f811d9986cd4baf04e82c9d2 18164Author: Dave Airlie <airlied@linux.ie> 18165Date: Mon Feb 25 07:10:48 2008 +1000 18166 18167 fixup check for EXA composite pointed out by Alan Swanson 18168 18169commit 85e470e64f629de72e361c77770e2e29998d1bf4 18170Merge: 27ddb39 1b84c76 18171Author: Alex Deucher <alex@samba.(none)> 18172Date: Sun Feb 24 05:37:22 2008 -0500 18173 18174 Merge master and fix conflicts 18175 18176 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 18177 18178 Conflicts: 18179 18180 src/radeon_commonfuncs.c 18181 18182commit 27ddb39b12a0b54e099fd5274c4c91f08e2d2822 18183Author: Alex Deucher <alex@samba.(none)> 18184Date: Sun Feb 24 05:30:11 2008 -0500 18185 18186 R300: clean up magic numbers in RADEONInit3DEngine 18187 18188commit d4c20f33ad6a1f88615cd7e09ad3638896873f9e 18189Author: Alex Deucher <alex@samba.(none)> 18190Date: Sun Feb 24 04:46:10 2008 -0500 18191 18192 R300: replace magic numbers in cache flush 18193 18194commit e52f1c8d2647b81d891ec0728dd582941a76c83f 18195Author: Alex Deucher <alex@samba.(none)> 18196Date: Sun Feb 24 04:43:18 2008 -0500 18197 18198 R300: fill in some more 3D bitfields 18199 18200commit 1b84c76f27c8d24cb42beae26abf000721901c1c 18201Author: Dave Airlie <airlied@redhat.com> 18202Date: Sun Feb 24 19:20:36 2008 +1100 18203 18204 rs690: initial textured video support 18205 18206commit 68158124366db883a901e960fe5f05f8df5efa42 18207Author: Dave Airlie <airlied@redhat.com> 18208Date: Tue Feb 19 19:51:18 2008 +1100 18209 18210 rs690: initial rs690 rotate 18211 18212 suffers same problem as r500 with clipping 18213 18214commit e614bb6965588bf09dcb87f5e08e67120ec9847f 18215Author: Dave Airlie <airlied@redhat.com> 18216Date: Sun Feb 24 01:46:05 2008 -0500 18217 18218 r500: convert fragprog to use register values 18219 18220commit 6ce9ee47c75620b2e5d211c5d59d17271a6a7b19 18221Author: Dave Airlie <airlied@redhat.com> 18222Date: Sat Feb 23 22:49:34 2008 -0500 18223 18224 r500: add textured video Xv adapter support 18225 18226commit 9aaf8b33b22b6ba112869558ae54e021b9487ad2 18227Author: Dave Airlie <airlied@redhat.com> 18228Date: Sat Feb 23 22:16:25 2008 -0500 18229 18230 r500: initial rotate support - not fully working yet. 18231 18232 Just an example of how to setup and run the r500 3D engine for rotation. 18233 this rotates for me but I get some strange clipping on the bottom of my screen 18234 18235commit 05dc3e4fc19d056ce99a7b110665adab2ca1ea21 18236Author: Adam Jackson <ajax@redhat.com> 18237Date: Sat Feb 23 20:29:51 2008 -0500 18238 18239 Clarify R500 US setup. 18240 18241commit 080606ad528972623e1ed1124d8efe7705a73446 18242Author: Adam Jackson <ajax@redhat.com> 18243Date: Sat Feb 23 20:21:17 2008 -0500 18244 18245 Add the R500 US index/data pair. 18246 18247commit edfb3b6bbf0ee17ace8e6ba704a6f54e249fec63 18248Author: Alex Deucher <alex@samba.(none)> 18249Date: Sat Feb 23 19:59:33 2008 -0500 18250 18251 RADEON: no textured video yet on XPRESS chips 18252 18253 Still need to sort out the VAP and PVS stuff 18254 18255commit 4146bfe5d00e40a86d17826fac50d04b2469621d 18256Author: Alex Deucher <alex@samba.(none)> 18257Date: Sat Feb 23 19:21:52 2008 -0500 18258 18259 R500: fix typo in new r5xx regs 18260 18261commit d9be9f34b0d3313e7b22b2a8bb0a8924ad3116bf 18262Author: Alex Deucher <alex@samba.(none)> 18263Date: Sat Feb 23 19:06:30 2008 -0500 18264 18265 RADEON: add textured video support for r1xx-r4xx radeons 18266 18267 Based on the kdrive ati video code by Eric Anholt. 18268 R3xx/R4xx still have some clipping issues in certain situations 18269 18270commit 9dc4acad79196e9d5d94dd710773bfa83456d47f 18271Author: Alex Deucher <alex@cube.(none)> 18272Date: Sat Feb 23 18:29:00 2008 -0500 18273 18274 RS6xx: gpio entry for DDIA varies depending on the number of DFP ports 18275 18276commit ed0a93edf28155308e7ab9d8705581bb38455ea0 18277Author: Adam Jackson <ajax@redhat.com> 18278Date: Sat Feb 23 15:02:17 2008 -0500 18279 18280 Fix R500_US_CONFIG. 18281 18282commit bc2bd6f841b51aeed3b6b4a47dbe758c200bc5a6 18283Author: Adam Jackson <ajax@redhat.com> 18284Date: Sat Feb 23 14:34:18 2008 -0500 18285 18286 Add R500 unified shader register block. 18287 18288commit a7b5c3bb74fc4de5e38a75ac31656445ce823464 18289Author: Alex Deucher <alex@cube.(none)> 18290Date: Fri Feb 22 19:35:11 2008 -0500 18291 18292 RS6xx: fix DDC on DDIA output (usually HDMI port) 18293 18294commit 3327a681e21101cc6f6e162f4e29f9937b69ccc3 18295Author: Alex Deucher <alex@cube.(none)> 18296Date: Fri Feb 22 17:05:56 2008 -0500 18297 18298 ATOM: properly set up DDIA output on RS6xx boards 18299 18300commit 1d0e9ab8b9451101b1b91943546f6c5833c21b3f 18301Author: Michel Dänzer <michel@tungstengraphics.com> 18302Date: Wed Feb 20 10:21:49 2008 +0100 18303 18304 radeon: Fix typo flagged by gcc -Wall. 18305 18306commit b5bd442b60dbc72fe4c1e928ab864aeb0fd7a3cb 18307Author: Alex Deucher <alex@botch2.(none)> 18308Date: Tue Feb 19 20:47:40 2008 -0500 18309 18310 R100: fix render accel for transforms 18311 18312 Not sure why we had a separate broken path for r100 vertex 18313 submission. 18314 18315commit a0a73208a21546ac120fb9a463261836c9ea7b55 18316Author: Alex Deucher <alex@botch2.(none)> 18317Date: Tue Feb 19 20:11:19 2008 -0500 18318 18319 RADEON: restore clock gating and CP clock errata on VT switch 18320 18321 This may help people with hangs on resume 18322 18323commit b77e2aff7453a9f370beba37ca3c25b92b3f97ff 18324Author: Alex Deucher <alex@botch2.(none)> 18325Date: Tue Feb 19 19:55:41 2008 -0500 18326 18327 RADEON: fix DDC types 5 and 6 18328 18329commit af82172a82f2bdf96e571def659a1c70f92dfdbf 18330Author: Alex Deucher <alex@botch2.(none)> 18331Date: Tue Feb 19 19:39:35 2008 -0500 18332 18333 RADEON: update man page with supported chips 18334 18335commit 1302d1b019c8091986aaf9c86b25c1f36a037788 18336Author: Alex Deucher <alex@botch2.(none)> 18337Date: Tue Feb 19 00:34:31 2008 -0500 18338 18339 Bump for 6.8 release 18340 18341commit 830d29e8b280c11e798fb5bf5278bf24ecbac7bf 18342Author: Alex Deucher <alex@cube.(none)> 18343Date: Mon Feb 18 21:17:12 2008 -0500 18344 18345 ATOM: fix RS6xx connector table parsing and fix some warnings 18346 18347commit 03aa4cc6d6e8c715a1c1d677cc1845223505b358 18348Author: Roland Scheidegger <sroland@tungstengraphics.com> 18349Date: Mon Feb 18 20:19:58 2008 -0500 18350 18351 bring back to life planar-to-packed conversion for rs4xx 18352 18353 Here's a patch to bring back the code for converting planar yuv to packed yuv, 18354 if a RS400 family chip is used (though I've no idea if they really all fail 18355 with planar yuv). 18356 18357 fixes bug 12744 18358 18359commit 66b0b3f43bb714bc007169d5ba6dabe34cfc008b 18360Author: George Sapountzis <gsap7@yahoo.gr> 18361Date: Mon Feb 18 10:33:59 2008 +0200 18362 18363 r128/radeon: probably correct fix for non pci access code 18364 18365commit d055b9e800ae50d08cca9db75fc666ce1da9ab52 18366Author: Dave Airlie <airlied@linux.ie> 18367Date: Sat Feb 16 09:13:43 2008 +1000 18368 18369 r128/radeon: hopeful fix for non pci access code 18370 18371commit c773bc6a314327da29e21e4ebac6fa7f3e98a0a7 18372Author: Dave Airlie <airlied@linux.ie> 18373Date: Sat Feb 16 09:05:07 2008 +1000 18374 18375 r128/radeon: fix build without pciaccess 18376 18377commit 690a52da5248f47a587a878d05fce9784957970b 18378Author: Dave Airlie <airlied@linux.ie> 18379Date: Sat Feb 16 08:33:36 2008 +1000 18380 18381 mach64: fix non pciaccess build 18382 18383commit fc85188fd95bf78b7f965cdde3e22b644c74ff81 18384Author: George Sapountzis <gsap7@yahoo.gr> 18385Date: Fri Feb 15 18:36:10 2008 +0200 18386 18387 ati: convert to pci probe 18388 18389 add pciids for each subdriver, make no use of the match_data functionality. 18390 18391 thanks to Alex Deucher for reviewing and testing. 18392 18393commit 665bd7e2f61cac3e029bbad5024034e5136deec1 18394Author: George Sapountzis <gsap7@yahoo.gr> 18395Date: Fri Feb 15 18:34:56 2008 +0200 18396 18397 ati wrapper: translate Device lines 18398 18399commit a596c1618f72179a45289a50a1f9e89462ce9667 18400Author: George Sapountzis <gsap7@yahoo.gr> 18401Date: Fri Feb 15 18:34:31 2008 +0200 18402 18403 r128: do not compile in PciChipsets twice 18404 18405commit 99cd8ff9a7e15fc2b4e55f8bc020f584173a8c2d 18406Author: George Sapountzis <gsap7@yahoo.gr> 18407Date: Fri Feb 15 18:33:51 2008 +0200 18408 18409 ati: drop duplicate pci-id defines 18410 18411commit 146b01b51069dc227d0b125babb3f6957c9b9de2 18412Author: George Sapountzis <gsap7@yahoo.gr> 18413Date: Fri Feb 15 18:33:29 2008 +0200 18414 18415 mach64: clean probe a little 18416 18417 do not report I/O ports now, they are reported later. 18418 18419commit f47d461331a032f9bdcf6f63336e848778cec6cc 18420Author: George Sapountzis <gsap7@yahoo.gr> 18421Date: Fri Feb 15 18:32:56 2008 +0200 18422 18423 mach64: minor cosmetic, I2C type 18424 18425commit 2c66f2e812195167df9ca113044d46deece776ac 18426Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 18427Date: Fri Feb 15 10:59:30 2008 +1000 18428 18429 make distcheck pass 18430 18431commit ed87f367ddab7366f84917b43b31d76df4ce1110 18432Author: Alex Deucher <alex@botch2.(none)> 18433Date: Wed Feb 13 12:53:46 2008 -0500 18434 18435 RADEON: disable LVDS if panel size is 0x0 18436 18437 if we can't get the panel size from the bios or edid 18438 or a user specified option, assume it's not connected. 18439 18440commit 422d7f441fdbb318d22d44db661ac9bd61387bd2 18441Author: Alex Deucher <alex@cube.(none)> 18442Date: Tue Feb 12 15:35:46 2008 -0500 18443 18444 R6xx: when both crtcs are in use make sure they are both enabled 18445 18446 sometimes setting a mode on one crtc can cuase a blank screen on the other. 18447 make sure they are both enabled if they should be. 18448 18449commit 860f5af75274cb236f536e1da09da6bd9a579530 18450Author: Alex Deucher <alex@cube.(none)> 18451Date: Tue Feb 12 14:46:49 2008 -0500 18452 18453 R6xx: fix up use of bios scratch regs to reflect the new offsets 18454 18455commit 32f2119b43a0faf6069d8cc0816f0d9f7914c07f 18456Author: Alex Deucher <alex@cube.(none)> 18457Date: Tue Feb 12 14:20:02 2008 -0500 18458 18459 R6xx: bios scratch regs moved. 18460 18461commit 8d64be6ebd7f50d4bcb587afeee8252c1367dc77 18462Author: Alex Deucher <alex@samba.(none)> 18463Date: Tue Feb 12 12:53:09 2008 -0500 18464 18465 RADEON: make sure we always set up the FP crtc shadow registers for crtc0 18466 18467 The behavior changed when I added rmx center mode support. In cases where 18468 crtc0 drives a DAC this can lead to a blank screen. 18469 18470commit 810d192ee046077a894e0fb5f2dfd6a7c0130766 18471Author: Alex Deucher <alex@cube.(none)> 18472Date: Mon Feb 11 19:23:01 2008 -0500 18473 18474 R6xx: add missing objects 18475 18476commit 85043439426e534e561259ce98bebdd8508b36a9 18477Author: Alex Deucher <alex@botch2.(none)> 18478Date: Mon Feb 11 16:36:58 2008 -0500 18479 18480 R6xx: make sure we set up the HDP base properly 18481 18482commit 9ab5d2ec7c583c74f364d7cfbb54bcd2cd8ae2f5 18483Author: Alex Deucher <alex@botch2.(none)> 18484Date: Mon Feb 11 15:26:51 2008 -0500 18485 18486 RADEON: always restore crtc1 before crtc0 when using both crtcs 18487 18488 In some rare cases restoring crtc0 first will result in a blank screen 18489 on crtc1. If you are having issues with a blank screen on crtc1 18490 that used to work on 6.6.3 or before, this should help. 18491 18492commit e33edca75bd9df0aa19a33e74c38a6d02610befd 18493Author: Alex Deucher <alex@botch2.(none)> 18494Date: Mon Feb 11 00:33:12 2008 -0500 18495 18496 RADEON: Implement proper (hopefully) BIOS/DRIVER interaction for ATOM bios 18497 18498 Tested on atom-based Desktop cards. It'd nice to get some testing 18499 on atom-based laptops. 18500 18501commit 6524e33435a786f7de0064cdd1b04c1120d21593 18502Author: Alex Deucher <alex@botch2.(none)> 18503Date: Sun Feb 10 18:52:52 2008 -0500 18504 18505 RADEON: Implement proper (hopefully) BIOS/DRIVER interaction for COM bios 18506 18507 Tested on my M10-based laptop. 18508 18509commit 8606c1bd175893c77e8c758246a1aed166c8be51 18510Author: George Sapountzis <gsap7@yahoo.gr> 18511Date: Fri Feb 8 19:07:03 2008 +0200 18512 18513 mach64: factor out BIOS panel info parsing 18514 18515commit 9f33218c80f5a6d6d9464aa3db8ae25a4759f351 18516Author: George Sapountzis <gsap7@yahoo.gr> 18517Date: Fri Feb 8 19:06:41 2008 +0200 18518 18519 mach64: minor cosmetic, LCD panel id 18520 18521commit 8cd5a465a03834b4b2f3a19f8d36fc9d2e18c6d4 18522Author: George Sapountzis <gsap7@yahoo.gr> 18523Date: Fri Feb 8 19:06:27 2008 +0200 18524 18525 mach64: minor cosmetic, DAC type 18526 18527commit 7f4db96123fdcba9874226352802d702c2a498bd 18528Author: Dave Airlie <airlied@redhat.com> 18529Date: Fri Feb 8 10:47:10 2008 +1000 18530 18531 add rv670 pciids 18532 18533commit e8899b9978291c62a65f468c92f340f65ad5479d 18534Author: Alex Deucher <alex@botch2.(none)> 18535Date: Thu Feb 7 19:27:38 2008 -0500 18536 18537 R6xx: fix ddc after my i2c rework 18538 18539 Seems r6xx does something different for its i2c table, 18540 revert to the old behavior for now. 18541 18542commit 435de6c4e46ff2bebd4cee58888a66b936cd3fdf 18543Author: Alex Deucher <alex@samba.(none)> 18544Date: Thu Feb 7 19:14:13 2008 -0500 18545 18546 RADEON: sync up with latest upstream versions 18547 18548 atombios.h 18549 ObjectID.h 18550 18551commit 692789a293970f70b88ccb6adcf0676d8b495ae2 18552Author: George Sapountzis <gsap7@yahoo.gr> 18553Date: Thu Feb 7 18:03:37 2008 +0200 18554 18555 mach64: factor out BIOS multimedia parsing 18556 18557commit 933328ffd6d1d872a18d3de8624c4df845a64588 18558Author: George Sapountzis <gsap7@yahoo.gr> 18559Date: Thu Feb 7 18:03:03 2008 +0200 18560 18561 mach64: complement hint for sparc 18562 18563commit 956c8c81f3ff434930a0cb17b027b2f8e4eeabb2 18564Author: George Sapountzis <gsap7@yahoo.gr> 18565Date: Thu Feb 7 18:02:31 2008 +0200 18566 18567 mach64: consolidate refclk #2 18568 18569commit dce4cc26a8e2bf53805ec63763243f3ff6b4a6d3 18570Author: George Sapountzis <gsap7@yahoo.gr> 18571Date: Thu Feb 7 18:02:17 2008 +0200 18572 18573 mach64: consolidate refclk #1 18574 18575commit f7ed807f0d82a7446ebc4acdd4e94df44a675f19 18576Author: George Sapountzis <gsap7@yahoo.gr> 18577Date: Thu Feb 7 18:01:59 2008 +0200 18578 18579 mach64: cosmetic 18580 18581commit cda1cd198f33c26ef1b51532a2126468369743b8 18582Author: George Sapountzis <gsap7@yahoo.gr> 18583Date: Thu Feb 7 18:01:33 2008 +0200 18584 18585 mach64: factor out BIOS clock parsing 18586 18587commit 73ff279469be9c7cbf9f533b85fcb553694ff413 18588Author: George Sapountzis <gsap7@yahoo.gr> 18589Date: Thu Feb 7 18:00:55 2008 +0200 18590 18591 mach64: BIOSBase is no longer used 18592 18593commit caea326cc6f1932bb299f451be013651a5749ea7 18594Author: Dave Airlie <airlied@linux.ie> 18595Date: Wed Feb 6 06:36:13 2008 +1000 18596 18597 r300: move fragprog setup code to prepare composite for now 18598 18599commit 470cd6a401c6a3e8fea981a8fe97c28be3cfb81d 18600Author: Dave Airlie <airlied@linux.ie> 18601Date: Wed Feb 6 06:04:13 2008 +1000 18602 18603 r300: remove r300 specific vertex emission 18604 18605 Set the vertex and fragment engine to expect the mask coords. 18606 18607commit f65e8dfac23adfa199026765fe3a1ea08cf4da67 18608Author: Alex Deucher <alex@cube.(none)> 18609Date: Sun Feb 3 00:09:59 2008 -0500 18610 18611 RADEON: rework i2c handling 18612 18613 Split out clk, data, and lock regs and masks. some cards use different 18614 regs and masks for each. For cards with ATOMBIOS, use the i2c bios 18615 table to grab the i2c data. 18616 18617commit a38a903debc0a50dbc73f59dc2741bbea76d2bd9 18618Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18619Date: Fri Feb 1 02:59:07 2008 -0500 18620 18621 RADEON: don't restore LVDS_PLL_CNTL for now 18622 18623 seems to cause problems with resume for some users. 18624 this needs further investigation. 18625 see bug 12596 18626 18627commit 5d7bea2b62c86868baf1960665a40249a43bafc5 18628Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18629Date: Fri Feb 1 02:53:22 2008 -0500 18630 18631 RADEON: remove redundant RADEONDisableDisplays() 18632 18633 use RADEONBlank() instead 18634 18635commit 73b437ce232c94c0067a0d2f70538b6e1e8c07a7 18636Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18637Date: Fri Feb 1 02:47:06 2008 -0500 18638 18639 RADEON: remove unused "aspect" scaler option 18640 18641commit f1fb9e4daa29bc379f653f847254db1496b625fd 18642Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18643Date: Fri Feb 1 02:39:18 2008 -0500 18644 18645 RADEON: Implement "center" mode for RMX on legacy radeons 18646 18647commit bcd590103e04bfdb4f12413beacebf344f07e88e 18648Author: Alex Deucher <alex@botch2.(none)> 18649Date: Tue Jan 29 12:12:54 2008 -0500 18650 18651 RADEON: update man page to reflect AGP 1x default revert 18652 18653commit 09348a83d06ba9d3129499c4daedd44a68771530 18654Author: Alex Deucher <alex@botch2.(none)> 18655Date: Tue Jan 29 12:09:24 2008 -0500 18656 18657 Seems the default is more reliable... we can't win. 18658 18659 Revert "radeon: Default to 1x again with non-v3 AGP cards." 18660 18661 This reverts commit b653e5a628bfa4dfb168e96f93f41eb910f409fb. 18662 18663commit 0c26806245381b925b73ff9f3b747e3dcf0ebd6f 18664Author: Alex Deucher <alex@botch2.(none)> 18665Date: Tue Jan 29 10:26:48 2008 -0500 18666 18667 RADEON: Add new RV380 pci id 18668 18669 bug 14289 18670 18671commit ce77ed78a877023da72dbe51609aef9a07e250b1 18672Author: Maciej Cencora <m.cencora@gmail.com> 18673Date: Mon Jan 28 19:02:56 2008 -0500 18674 18675 RS690: Implement MC idle check 18676 18677commit b7de4ff52cfbdd85ee65000613632e21b92af24e 18678Author: George Sapountzis <gsap7@yahoo.gr> 18679Date: Sat Jan 26 19:28:05 2008 +0200 18680 18681 mach64: add hint for sparc and minor cosmetic. 18682 18683commit 09d713aa0ed6367b4457420b3c2832fe8eca9b00 18684Author: Alex Deucher <alex@botch2.(none)> 18685Date: Wed Jan 23 18:05:41 2008 -0500 18686 18687 Revert "RADEON: adjust PAL mode hstart" 18688 18689 This reverts commit 719a9a376e34d99032af75e3f7b002670ccb816b. 18690 18691 This breaks TV out on some cards. 18692 18693commit 719a9a376e34d99032af75e3f7b002670ccb816b 18694Author: Alex Deucher <alex@botch2.(none)> 18695Date: Sun Jan 20 18:40:53 2008 -0500 18696 18697 RADEON: adjust PAL mode hstart 18698 18699 Most people seem to get a more aligned picture with this setting 18700 18701commit b2db8657fb888cff6d64c6dcb182caac389776ce 18702Author: Alex Deucher <alex@botch2.(none)> 18703Date: Sun Jan 20 18:33:22 2008 -0500 18704 18705 RADEON: re-work i2c for DDC 18706 18707 Unify the radeon/avivo paths and grab the data/clk masks from bios 18708 if available 18709 18710commit a43003b24022a833e604f41b2873c0350b34181c 18711Author: Alex Deucher <alex@botch2.(none)> 18712Date: Sat Jan 19 18:49:53 2008 -0500 18713 18714 RADEON: get dac2 adj values from the bios tables 18715 18716commit d4596c52ac9994be26e9ec2d7d57b3892c34abdb 18717Author: Alex Deucher <alex@botch2.(none)> 18718Date: Sat Jan 19 17:17:26 2008 -0500 18719 18720 RADEON: grab pll_in_min/pll_in_max from bios tables if available 18721 18722commit 9a0947c812d0d38d1bca6a91140ac210831a6cb4 18723Author: Alex Deucher <alex@botch2.(none)> 18724Date: Sat Jan 19 13:57:50 2008 -0500 18725 18726 ATOM: Use LVDS edid from bios if available 18727 18728commit 7238258c12def8ef273e5362f716d165f720c5a5 18729Author: Kusanagi Kouichi <slash@ma.neweb.ne.jp> 18730Date: Sat Jan 19 15:04:21 2008 +0100 18731 18732 radeon: Partial fix for XVideo RGB image distortions. 18733 18734commit 32be3cf9d6c34e60ff8c3d6cfe9f73f1869c50e4 18735Author: Brice Goglin <Brice.Goglin@ens-lyon.org> 18736Date: Fri Jan 18 14:42:14 2008 -0500 18737 18738 RADEON: print the name of the output when printing the EDID 18739 18740commit 4ba9430ee97dbce8f77db8de6ce9b753a75e453d 18741Author: George Sapountzis <gsap7@yahoo.gr> 18742Date: Fri Jan 18 18:18:18 2008 +0200 18743 18744 mach64: workaround for corruption at upper-left 18745 18746 commit possible workaround, it's reported multiple times ... 18747 18748commit 12c00111b68c9cf4872a424258c6f8b7247aac47 18749Author: George Sapountzis <gsap7@yahoo.gr> 18750Date: Fri Jan 18 18:17:40 2008 +0200 18751 18752 use stand-alone drivers for each chip family. 18753 18754 Do not load the ati wrapper when the user specifies the sub-driver name in 18755 the Driver line of xorg.conf. Also, for -configure cause the wrapper to fail 18756 and let each sub-driver speak for themselves. 18757 18758commit 24c7d134cd450f9e2cca85e4a2fc3253d250be04 18759Author: Alex Deucher <alex@botch2.(none)> 18760Date: Fri Jan 18 09:50:38 2008 -0500 18761 18762 RADEON: make sure EXA Composite is actually disabled on XPRESS chips. 18763 18764commit eaf425436008092abe81208321a2b3b6698a5d79 18765Author: Dave Airlie <airlied@redhat.com> 18766Date: Fri Jan 18 20:11:57 2008 +1000 18767 18768 fixup register 6594 save/restore 18769 18770commit 3de2dc88cf26ff5932f11cecdf975777b8aa2a4a 18771Author: Adam Jackson <ajax@redhat.com> 18772Date: Wed Jan 16 14:55:05 2008 -0500 18773 18774 Bump CRTC size limits on AVIVO chips so 30" displays work without tweaking. 18775 18776 Note that the CRTC size limits we're using right now are _not_ the 18777 hardware limits, they're just heuristics until we can resize the front 18778 buffer properly. 18779 18780commit 2a89a31481d71a56a9930073cf99d3ae7b4290e1 18781Author: Alex Deucher <alex@samba.(none)> 18782Date: Thu Jan 17 15:08:17 2008 -0500 18783 18784 RADEON: use radeon_output->Flags for tracking RMX rather than mode->Flags 18785 18786commit 495e3119250ffb48489debbaabe560d23753cc43 18787Author: Alex Deucher <alex@samba.(none)> 18788Date: Thu Jan 17 14:56:19 2008 -0500 18789 18790 AVIVO: Add support for RMX 18791 18792 Both centered and expansion modes are supported. Select 18793 using output attributes. 18794 18795commit 6bd510a211f25d52e74791e4a429cd2218ced541 18796Author: Alex Deucher <alex@samba.(none)> 18797Date: Wed Jan 16 18:09:49 2008 -0500 18798 18799 RADEON: add a message about render accel on newer cards 18800 18801commit 85bf3439fe2579aec48f5cd8d65a9d51b1ae8535 18802Author: Alex Deucher <alex@samba.(none)> 18803Date: Wed Jan 16 17:52:06 2008 -0500 18804 18805 R300: only init3d on r3xx 18806 18807commit dbb2ca471dfbff245b30c5055871dee0dc0e3d15 18808Author: Alex Deucher <alex@samba.(none)> 18809Date: Wed Jan 16 17:10:02 2008 -0500 18810 18811 R300: only enable render accel on non-IGP r3xx/r4xx chips for now 18812 18813commit 3c72b100bcfacee600644669b586e86cfd32754e 18814Author: Alex Deucher <alex@samba.(none)> 18815Date: Wed Jan 16 16:55:42 2008 -0500 18816 18817 R300: First pass at render accel 18818 18819 This first pass is pretty limited. All it currently supports 18820 is transforms for rotation. No blending yet. 18821 18822 Based on inital implementation from Wolke Liu with 18823 additional lock-up fixes by Dave Airlie. 18824 18825commit 2ba3562d2af911fdd90881049599e239d27260bc 18826Author: George Sapountzis <gsap7@yahoo.gr> 18827Date: Sat Jan 12 17:11:59 2008 +0200 18828 18829 ati wrapper: xf86PciInfo.h is enough 18830 18831commit c2caeb11a97dad5379d70881c5c0fd834a8c3d54 18832Author: George Sapountzis <gsap7@yahoo.gr> 18833Date: Sat Jan 12 16:18:34 2008 +0200 18834 18835 ati wrapper: add DriverRec's and use them 18836 18837commit 19e1b180fec6f83a474e125465bc60111c0f43e0 18838Author: George Sapountzis <gsap7@yahoo.gr> 18839Date: Sat Jan 12 16:18:07 2008 +0200 18840 18841 mach64: load for both "ati" and "mach64" as driver names 18842 18843 similar to r128/radeon 18844 18845commit 92f54400d5450b29b3a76d5ecc927cf0d73e156e 18846Author: George Sapountzis <gsap7@yahoo.gr> 18847Date: Sat Jan 12 16:17:47 2008 +0200 18848 18849 mach64: add version (need not match with ati) 18850 18851 similar to r128/radeon 18852 18853commit 311ec7b6c54a50a4b8a5a445f7283da2b0b2e0f5 18854Author: George Sapountzis <gsap7@yahoo.gr> 18855Date: Sat Jan 12 16:17:21 2008 +0200 18856 18857 atimisc: rename to mach64 18858 18859commit 5244e235262290eab8a3546f449295c12ed8f6c7 18860Author: Alex Deucher <alex@samba.(none)> 18861Date: Mon Jan 14 16:11:09 2008 -0500 18862 18863 ATOMBIOS: disable the scaler by default on avivo cards 18864 18865 Fixes bug 14001 18866 18867commit 729da30c80d6545b75c8faea44754634f477cc09 18868Merge: 000741e 625a885 18869Author: Alex Deucher <alex@samba.(none)> 18870Date: Mon Jan 14 10:05:01 2008 -0500 18871 18872 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 18873 18874commit 625a885a964488da2a35065eb597a17ee57b22a9 18875Author: George Sapountzis <gsap7@yahoo.gr> 18876Date: Wed Jan 9 17:39:56 2008 +0200 18877 18878 ati wrapper: use pci ids 18879 18880commit 48865200ca9f1daf87e52620b9f8175f88dd886f 18881Author: George Sapountzis <gsap7@yahoo.gr> 18882Date: Wed Jan 9 19:01:04 2008 +0200 18883 18884 radeon: remove stray include 18885 18886commit a5a6b873353611cb7e46e5e375f039957c7051a7 18887Author: George Sapountzis <gsap7@yahoo.gr> 18888Date: Wed Jan 9 13:03:32 2008 +0200 18889 18890 radeon: remove stray _X_EXPORT 18891 18892commit 80c9974b6cdc0147d777df6990b3f3aacd87969d 18893Author: George Sapountzis <gsap7@yahoo.gr> 18894Date: Tue Jan 8 18:15:47 2008 +0200 18895 18896 r128: pci-rework conversion 18897 18898 compile-tested only 18899 18900commit cd4b39403d74f314e1c2cfa4cf0e4fe542891dc3 18901Author: George Sapountzis <gsap7@yahoo.gr> 18902Date: Sat Mar 10 23:12:15 2007 +0200 18903 18904 Drop symbol lists from r128, radeon, theatre. 18905 18906 compile-tested only 18907 18908commit 63b4b8213fabc5a57f897b60b6eaa9f78d86e6ff 18909Author: George Sapountzis <gsap7@yahoo.gr> 18910Date: Wed Mar 21 20:24:22 2007 +0200 18911 18912 [mach64] Drop symbol lists. 18913 18914 egrep LOADER\|SymLists\|Symbols src/*.[hc] 18915 18916 not needed after server commit bca9364f3f4a2376edbcf57a34f704ce28be21ba, i.e. 18917 xserver 1.2 18918 18919 compile-tested only 18920 18921commit 000741e250e54122b0adc91694eb4bfa320a70fb 18922Author: Alex Deucher <alex@samba.(none)> 18923Date: Thu Jan 10 14:49:48 2008 -0500 18924 18925 RADEON: clean up output handling 18926 18927commit 10e7636c02478b8ffe183bb0c46229ca0d6584e1 18928Author: Kristian Høgsberg <krh@bitplanet.net> 18929Date: Wed Jan 9 12:47:39 2008 -0500 18930 18931 RADEON: fix crtc routing on r4xx cards when using atom to init DVO chip 18932 18933commit 3af671f5963810dbfd63abc9889b1d46b68f404c 18934Author: Alex Deucher <alex@botch2.(none)> 18935Date: Wed Jan 9 11:30:25 2008 -0500 18936 18937 RADEON: restore FP2 regs before external encoders 18938 18939 This may fix krh's dvi problem 18940 18941commit 2a54c6bb09ade2ec8f998dfc1624017029d47fa3 18942Author: Alex Deucher <alex@botch2.(none)> 18943Date: Tue Jan 8 18:43:54 2008 -0500 18944 18945 RADEON: Make default output actually work... 18946 18947commit fa3e2055225c27e25465fc46786da1b7574fd3cc 18948Author: Alex Deucher <alex@botch2.(none)> 18949Date: Mon Jan 7 01:13:09 2008 -0500 18950 18951 RADEON: add default outputs if no connected devices 18952 18953 If no connected devices found at server startup, default 18954 to something so the server comes up. LVDS on mobility chips, 18955 DAC or TMDS on others. 18956 18957commit d972cc9237eb90b49b11f8d2bdc5b61f628911dc 18958Author: Alex Deucher <alex@botch2.(none)> 18959Date: Sat Jan 5 17:19:06 2008 -0500 18960 18961 RADEON: Fix TVStandard option 18962 18963commit 45656b9d5a426053da2a094de39c2690c0c6f799 18964Author: Alex Deucher <alex@botch2.(none)> 18965Date: Sat Jan 5 12:00:55 2008 -0500 18966 18967 R128: Like powerpc, don't use VGA by default on sparc 18968 18969commit b8e8db4675d07e45782de0d7c67ee0fd85eaedb3 18970Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18971Date: Fri Jan 4 20:16:19 2008 -0500 18972 18973 RADEON: fix tvdac load detection at server start up 18974 18975commit 3ba7f393d0669df36848715799de8affc10a5534 18976Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18977Date: Fri Jan 4 20:00:48 2008 -0500 18978 18979 RADEON: further fixup for pScrn->pScreen issue 18980 18981 the previous fix seems to cause the driver to hang on 18982 some cards. 18983 18984commit a0de9c0844f9e066e0f02e8cd8045bdd278e6494 18985Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18986Date: Fri Jan 4 19:48:30 2008 -0500 18987 18988 RADEON: improve ntsc image centering 18989 18990commit 03d2f25801c8a8ec15030f06008df112d07c1a2d 18991Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18992Date: Fri Jan 4 18:19:54 2008 -0500 18993 18994 RADEON: fix compile 18995 18996commit 308848783ed9ae27aed7c7de6ee813d375ef495e 18997Author: Dave Airlie <airlied@linux.ie> 18998Date: Sat Jan 5 09:11:55 2008 +1000 18999 19000 i2c: a line mux of 0 is valid 19001 19002commit 94d5a432f72801f821d1c4ce952baba17552659d 19003Author: Dave Airlie <airlied@linux.ie> 19004Date: Fri Jan 4 17:16:54 2008 +1000 19005 19006 atombios: i2c lines are all done with sw so we can accept any of them. 19007 19008 hch reported this working on his laptop with 0x7e60 as EDID for LVDS 19009 19010commit 2e4473b63d65801ae8ac5a8391de232b2201d958 19011Author: Alex Deucher <alex@samba.(none)> 19012Date: Thu Jan 3 15:47:50 2008 -0500 19013 19014 RADEON: fix crash when setting rotation in the config file 19015 19016 xf86CrtcRotate() accesses pScrn->pScreen which is not set 19017 during ScreenInit(). This should also be fixed in the server. 19018 See bug 12129 19019 19020commit c652208861bffca94f06b7f67688ce220e050bfb 19021Author: Michel Dänzer <michel@tungstengraphics.com> 19022Date: Thu Jan 3 17:54:58 2008 +0100 19023 19024 radeon: Adapt manpage to reality wrt AGP options. 19025 19026commit ab451e4b7a5423d61b57cf0646599267d8504af4 19027Author: Michel Dänzer <michel@tungstengraphics.com> 19028Date: Thu Jan 3 17:52:39 2008 +0100 19029 19030 radeon: Miscellaneous warning fixes. 19031 19032commit 394c52273328e90518568b694ee79dc1a8dab651 19033Author: Dave Airlie <airlied@linux.ie> 19034Date: Thu Jan 3 18:56:16 2008 +1000 19035 19036 r500: tvout avoid doing dpms here it makes my tv mode go all crappy 19037 19038 need to investigate further 19039 19040commit 1c647279f021d01e110980727b7c7dd7efae1642 19041Author: Dave Airlie <airlied@linux.ie> 19042Date: Thu Jan 3 11:55:28 2008 +1000 19043 19044 r600: change devices list depending on connector 19045 19046commit f36db6e10d32a68b32d20ae4ad02cfc0bfd1c9c3 19047Author: Dave Airlie <airlied@linux.ie> 19048Date: Thu Jan 3 11:27:47 2008 +1000 19049 19050 r600: fixup crash on unknown output type 19051 19052 not sure this is 100% the correct approach 19053 19054commit 1accfdd590828e95e0d68a576c8ee05a06a86e43 19055Author: Alex Deucher <alex@samba.(none)> 19056Date: Wed Jan 2 19:48:28 2008 -0500 19057 19058 RADEON: various avivo tv-out related clean-ups 19059 19060commit ce34090c758ac91171cb6adb9b8a36e4adbf99cf 19061Merge: 2180f04 30cab1d 19062Author: Alex Deucher <alex@samba.(none)> 19063Date: Wed Jan 2 16:41:36 2008 -0500 19064 19065 Merge branch 'master' of git+ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 19066 19067commit 2180f04b6fb94a638f6274cb4455d5688b324dbc 19068Author: Alex Deucher <alex@samba.(none)> 19069Date: Wed Jan 2 16:41:06 2008 -0500 19070 19071 RADEON: preliminary support for ATOM RMX 19072 19073 not functional yet. 19074 19075commit dab4dc285154d40303aadaa849b85f8e251e578e 19076Author: Alex Deucher <alex@samba.(none)> 19077Date: Wed Jan 2 16:27:19 2008 -0500 19078 19079 RADEON: add support for ATOM component video table 19080 19081 Component output is still not working. 19082 19083commit 30cab1dbebb7bdb925f2fe9f91b84183312bbbfd 19084Author: Alex Deucher <alex@botch2.(none)> 19085Date: Wed Jan 2 15:41:01 2008 -0500 19086 19087 RADEON: Make sure all old IGP chips have HasSingleDac set 19088 19089 fix the csv file and re-gen the headers. 19090 19091commit e8e585651215b011e3ad07c59d0eab9107ccd8c6 19092Author: Dave Airlie <airlied@redhat.com> 19093Date: Wed Jan 2 10:14:46 2008 +1000 19094 19095 PLL/r600: tweak pll to pick first one found instead of keeping going 19096 19097commit 14aa4060ad27ecb3d40b2b17ee4cf7cc55a121dd 19098Author: Dave Airlie <airlied@redhat.com> 19099Date: Wed Jan 2 09:49:44 2008 +1000 19100 19101 r600: fix tv-out output naming 19102 19103commit f65374f5e15bfd391a1838a686cd87d3bab8043d 19104Author: Maciej Cencora <m.cencora@gmail.com> 19105Date: Mon Dec 31 09:44:34 2007 +1000 19106 19107 atombios: initial rs690 patches 19108 19109commit a674f683e6699c30664d9cd468a64de991c3fd7e 19110Author: Dave Airlie <airlied@linux.ie> 19111Date: Sun Dec 30 17:43:17 2007 +1000 19112 19113 atombios: enable TV-out detection code - tv out works for me with this 19114 19115commit aa7c28cbd943bb525698515d444cb5097880e364 19116Author: Dave Airlie <airlied@linux.ie> 19117Date: Sun Dec 30 17:40:37 2007 +1000 19118 19119 atombios: enable support for using tv timings 19120 19121 enable support for the atombios tv timings selection by programming the crtc 19122 with the tv timings if a tv is detected on the output 19123 19124commit 0bc3fd595a73e12a424571697d164a09a6a4c072 19125Author: Dave Airlie <airlied@linux.ie> 19126Date: Sun Dec 30 16:39:58 2007 +1000 19127 19128 atombios: add support for reading tv standard from atombios 19129 19130 fix typo in atombios header file 19131 19132commit bfa22d676a6f333503104041f62222f4de9bb7d8 19133Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19134Date: Wed Dec 26 12:29:47 2007 -0500 19135 19136 RADEON: fix typo that broke tv load detection 19137 19138commit 0c99554c6fab1192f5e8595469c21b5f33e1eb4f 19139Author: David Miller <davem@davemloft.net> 19140Date: Wed Dec 26 02:19:12 2007 -0500 19141 19142 [RADEON]: Add missing break in SCLK calculation. 19143 19144commit 6e0d5cc1c62fbfc1962fa0d6333f0c0a8c6327bd 19145Author: David Miller <davem@davemloft.net> 19146Date: Wed Dec 26 02:17:34 2007 -0500 19147 19148 [RADEON]: Like powerpc, don't use VGA by default on sparc. 19149 19150commit 2b1fae668ddabbc72e5fc31365302ea722174df1 19151Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19152Date: Wed Dec 26 02:04:06 2007 -0500 19153 19154 RADEON: fix PAL tv-out 19155 19156 Many thanks to Andrew Randrianasulu for providing me with 19157 pll reg dumps 19158 19159commit d736eb5732da573162c70712dc4e8b0114986702 19160Author: Alex Deucher <alex@botch2.(none)> 19161Date: Mon Dec 24 12:34:15 2007 -0500 19162 19163 RADEON: default "IgnoreLidStatus" to TRUE 19164 19165 Seems there are lots of busted ACPI lid status and people 19166 starting X with the lid closed. 19167 19168commit 653da558148cc601bc1f80253e92ef98c75ef37a 19169Author: Alex Deucher <alex@botch2.(none)> 19170Date: Mon Dec 24 01:11:56 2007 -0500 19171 19172 RADEON: restore crtc regs before VGA regs. 19173 19174 It seems some radeons don't restore text console properly if 19175 the crtc regs are restored after the VGA regs. 19176 Thanks to Sverre Froyen for helping me track this down 19177 19178commit ad3325f672a241449ca239c8ee3a24b6d7703d76 19179Author: Alex Deucher <alex@botch2.(none)> 19180Date: Sun Dec 23 17:18:42 2007 -0500 19181 19182 RADEON: Add "IgnoreLidStatus" option 19183 19184 Generally, users that start X with the laptop lid closed 19185 want to use one or more external monitors rather than the 19186 internal panel and do not want the internal panel to be on 19187 by default. Others, it seems, want to always have the 19188 internal panel on, regardless of the lid. I can't win. 19189 Enable this option to force the latter. 19190 19191commit 20eedf348a527e1e0a5450bc22d7564895034a66 19192Author: Alex Deucher <alex@botch2.(none)> 19193Date: Sun Dec 23 11:27:55 2007 -0500 19194 19195 RADEON: fix pll input setup on mac cards 19196 19197 the function was exiting before the complete setup was finished. 19198 19199commit 4f2e833e8ebaba3ad85ec5314fff8fa05395b679 19200Author: Arkadiusz Miskiewicz <arekm@maven.pl> 19201Date: Fri Dec 21 18:56:34 2007 -0500 19202 19203 configure.ac fixes 19204 19205commit 2b6e8e2b8f74e94560de89693ecbc7260536591e 19206Author: Arkadiusz Miskiewicz <arekm@maven.pl> 19207Date: Fri Dec 21 18:10:17 2007 -0500 19208 19209 RADEON: various cleanups 19210 19211commit 4c6f60e3b19ac55ab1255c79df03b1df5950864e 19212Author: Alex Deucher <alex@botch2.(none)> 19213Date: Fri Dec 21 17:33:04 2007 -0500 19214 19215 RADEON: clean up prototypes 19216 19217commit 8c761afdcb9baf1649b93449692fb9ab67bc2c80 19218Author: Alex Deucher <alex@botch2.(none)> 19219Date: Fri Dec 21 16:24:49 2007 -0500 19220 19221 RADEON: more re-org 19222 19223 move save/restore routines into legacy_crtc/output 19224 19225commit 910773d3a6c717f9d4762ea7b9ee6c3ae061781e 19226Author: Alex Deucher <alex@botch2.(none)> 19227Date: Fri Dec 21 15:53:15 2007 -0500 19228 19229 RADEON: more re-org. move XAA Mem init to radeon_accel.c 19230 19231commit 0631a23bd103f9b74e525da2c41304eab60c6f17 19232Author: Alex Deucher <alex@botch2.(none)> 19233Date: Fri Dec 21 15:40:18 2007 -0500 19234 19235 RADEON: fix rn50 reversion from last merge 19236 19237commit 5b917797a13c6caa80028d1842a284598e874288 19238Author: Alex Deucher <alex@botch2.(none)> 19239Date: Fri Dec 21 15:30:20 2007 -0500 19240 19241 RADEON: remove no longer used radeon_display.c 19242 19243commit bf14aa5f88fc3b4e69d71db5b23248b8bb2018d0 19244Author: Alex Deucher <alex@botch2.(none)> 19245Date: Fri Dec 21 15:29:47 2007 -0500 19246 19247 RADEON: re-org legacy crtc/output code into separate files 19248 19249commit be7f8fd338f5af8b632f16a83db41e15d00af469 19250Author: Dave Airlie <airlied@linux.ie> 19251Date: Fri Dec 21 10:03:49 2007 +1000 19252 19253 fix mode bandwidth configure check 19254 19255commit 3f9b597dedc45379b0bc0b631f3f924c403bca48 19256Author: Dave Airlie <airlied@linux.ie> 19257Date: Fri Dec 21 09:55:42 2007 +1000 19258 19259 fixup clip test include handling 19260 19261commit 9a5b501332c0a1f10af20845af48c9ddd2ce26a0 19262Author: Dave Airlie <airlied@linux.ie> 19263Date: Fri Dec 21 09:45:55 2007 +1000 19264 19265 set ddc line correctly post-merge 19266 19267commit 3c31b96afa20913ad947e68fe0c3a662e5eafbdd 19268Merge: eb99c3c f5e8c18 19269Author: Dave Airlie <airlied@linux.ie> 19270Date: Fri Dec 21 09:36:22 2007 +1000 19271 19272 Merge remote branch 'origin/atombios-support' 19273 19274 Conflicts: 19275 19276 src/radeon_display.c 19277 src/radeon_driver.c 19278 19279commit eb99c3c5c9a2249cb84920f0f225e525fc3a4144 19280Author: Alex Deucher <alex@botch2.(none)> 19281Date: Thu Dec 20 18:14:38 2007 -0500 19282 19283 Bump for RC release 19284 19285commit 8d49ff1da917b7f8240267953ef6ce4ff04daecb 19286Author: Alex Deucher <alex@botch2.(none)> 19287Date: Thu Dec 20 01:09:57 2007 -0500 19288 19289 RADEON: check for xf86_crtc_clip_video_helper() in configure.ac 19290 19291 use xf86_crtc_clip_video_helper() from the server if available. 19292 19293commit fb7a4e24f2da3561ef81371ca4013a4f13806e91 19294Author: Adam Jackson <ajax@redhat.com> 19295Date: Wed Dec 19 19:15:19 2007 -0500 19296 19297 Fix RN50 mode filtering. 19298 19299 The old code would attempt to limit the maximum pixel size of the screen 19300 by limiting the maximum PLL frequency. This ends up confusing the PLL 19301 computation code since sometimes your maximum freq can be lower than your 19302 minimum freq. More to the point it's just wrong, maximum PLL frequency 19303 isn't the same thing as maximum pixel clock, and even that isn't the same 19304 thing as maximum scanout pixels per second. 19305 19306 The correct thing to do is filter by the mode's effective memory bandwidth. 19307 19308commit c1b0b69cc50516c3b7e881b0eb46cb3cd2e9dce6 19309Merge: 0e66348 ce4fa1c 19310Author: Alex Deucher <alex@botch2.(none)> 19311Date: Wed Dec 19 11:54:50 2007 -0500 19312 19313 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 19314 19315commit 0e6634870d1ab38ee8c83f6bda1ba60364997853 19316Author: Alex Deucher <alex@botch2.(none)> 19317Date: Wed Dec 19 11:54:27 2007 -0500 19318 19319 RADEON: skip empty connectors when creating outputs 19320 19321commit 6afbf718c151dc3c5c59bd3136b58a93a114d798 19322Author: Alex Deucher <alex@botch2.(none)> 19323Date: Wed Dec 19 11:48:38 2007 -0500 19324 19325 RADEON: add support for legacy radeons with DVI and no connector table 19326 19327commit f5e8c185001e62e744310667c2d1bd3fe6542a62 19328Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 19329Date: Wed Dec 19 10:38:58 2007 +1000 19330 19331 more endian related fixage 19332 19333commit 98b247066d00db66abe91f518cd93b5c4da4cfb4 19334Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 19335Date: Wed Dec 19 10:25:41 2007 +1000 19336 19337 fix big endian build since zaphod fixups 19338 19339commit ce4fa1cedec0cf56b9979dfaa12a8d3a7c643df4 19340Author: Arkadiusz Miskiewicz <arekm@maven.pl> 19341Date: Tue Dec 18 15:34:14 2007 -0500 19342 19343 RADEON: fix fd leak in lid detect code 19344 19345commit 20568f66f9a9a60a33bd9a69ccc14a891c656836 19346Author: Arkadiusz Miskiewicz <arekm@maven.pl> 19347Date: Tue Dec 18 15:32:10 2007 -0500 19348 19349 RADEON: more cleanups and warning fixes 19350 19351commit 1496194200adbcb044ec3977367a0908262e389c 19352Author: Arkadiusz Miskiewicz <arekm@maven.pl> 19353Date: Tue Dec 18 15:29:53 2007 -0500 19354 19355 RADEON: driver cleanups, warning fixes 19356 19357commit bd7206fa120495037e3fea0c920d0031d7715bf6 19358Author: Alex Deucher <alex@botch2.(none)> 19359Date: Tue Dec 18 03:03:11 2007 -0500 19360 19361 RADEON: fix another merge error 19362 19363 this broken legacy radeons 19364 19365commit 65a3ac7530e11bb7d818a988fd0cf1dde7688fa4 19366Author: Alex Deucher <alex@samba.(none)> 19367Date: Tue Dec 18 00:15:38 2007 -0500 19368 19369 RADEON: more PLL tweaks 19370 19371commit d93a0e10b8bc6e3797a3cf6c1e28ca413a7c38e4 19372Author: Alex Deucher <alex@samba.(none)> 19373Date: Mon Dec 17 20:32:45 2007 -0500 19374 19375 RADEON: post div tweaks for legacy radeon 19376 19377commit 03b8b49f6f502c45552b018fd8c44d366b2d576f 19378Author: Alex Deucher <alex@samba.(none)> 19379Date: Mon Dec 17 20:20:04 2007 -0500 19380 19381 RADEON: fix typo from merge 19382 19383commit 19b9d3708852b7efe2b05249c8359dadb924dd94 19384Merge: cf685f3 29706ca 19385Author: Alex Deucher <alex@samba.(none)> 19386Date: Mon Dec 17 20:07:32 2007 -0500 19387 19388 Merge branch 'atombios-support' of git+ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 19389 19390commit cf685f37ec874f0aacd09e7c4eb0402c6daec1b0 19391Merge: 2a134af 44d07c4 19392Author: Alex Deucher <alex@samba.(none)> 19393Date: Mon Dec 17 20:07:07 2007 -0500 19394 19395 Merge branch 'master' of git+ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 19396 19397 merge master and fix conflicts 19398 19399commit 29706ca585ebd9b7b3521521a17016322e9ecccb 19400Author: Dave Airlie <airlied@linux.ie> 19401Date: Tue Dec 18 10:55:38 2007 +1000 19402 19403 fixup shadow setup on !r600 19404 19405commit 44d07c4ccce9acb5bd21a17acb082e91f7225764 19406Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19407Date: Mon Dec 17 18:56:12 2007 -0500 19408 19409 RADEON: typo from last commit 19410 19411commit 4da3782239921eb377216d4de4a9cc5bb55e0e8a 19412Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19413Date: Mon Dec 17 18:51:31 2007 -0500 19414 19415 RADEON: add output enable masks 19416 19417 add output enable masks for outputs that drive 19418 more than one connector. Make sure we don't turn off 19419 an output that's driving another connector. 19420 19421commit 5c5d2d19b2b032a06dd333b4ecc029aac342fb93 19422Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19423Date: Mon Dec 17 18:15:55 2007 -0500 19424 19425 RADEON: whitespace clean-ups 19426 19427commit 9f1d8220315c8894a17f2cc328025dc682b0c6e0 19428Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19429Date: Mon Dec 17 18:04:05 2007 -0500 19430 19431 RADEON: more PLL fixes 19432 19433 - reduce the calculation accuracy 19434 - certain LVDS panels seem to only like certain ref_divs 19435 - add pll flags to handle special cases 19436 - adjust the pll limits on legacy cards 19437 19438commit 2a134af01bc85de758ad83a85b23d1bba4a1a0f5 19439Author: Dave Airlie <airlied@redhat.com> 19440Date: Mon Dec 17 15:00:36 2007 +1000 19441 19442 r600: add shadow support to r600 driver to at least make 2d useable 19443 19444commit 614414611a9f246cbc74f579a79987fff97cf571 19445Author: Dave Airlie <airlied@redhat.com> 19446Date: Mon Dec 17 11:10:14 2007 +1000 19447 19448 radeon: cleanup pitch calculation and make r600 work again 19449 19450commit 79a375dbc7f323e2f551490a35f44ec36bed877c 19451Author: George Wu <geo@flood.OCF.Berkeley.EDU> 19452Date: Mon Dec 17 10:55:36 2007 +1000 19453 19454 r600: might as well fix VT for R600 19455 19456commit bc213ee723a45f2c396b4ed211a50f7642349973 19457Author: Alex Deucher <alex@samba.(none)> 19458Date: Sun Dec 16 14:54:00 2007 -0500 19459 19460 RADEON: fix sclock printout 19461 19462commit 4747c1f3cd4167b6a51d4864a297719ea48b9346 19463Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19464Date: Sun Dec 16 14:07:29 2007 -0500 19465 19466 RADEON: Make sure LVDS_EN bit is set when enabling LVDS 19467 19468commit a9817b2cf436a536dbc43ad77abc3bdcc53d346d 19469Author: Alex Deucher <alex@samba.(none)> 19470Date: Sat Dec 15 20:51:53 2007 -0500 19471 19472 RADEON: clean up units in PLL calculation 19473 19474commit b653e5a628bfa4dfb168e96f93f41eb910f409fb 19475Author: Michel Dänzer <michel@tungstengraphics.com> 19476Date: Sat Dec 15 00:50:10 2007 +0100 19477 19478 radeon: Default to 1x again with non-v3 AGP cards. 19479 19480 Seems more reliable in general than what was set up by firmware - fingers 19481 crossed... 19482 19483commit 6229825fa5d6715569098afbb21a40f7a2e7e6be 19484Author: Michel Dänzer <michel@tungstengraphics.com> 19485Date: Sat Dec 15 00:48:26 2007 +0100 19486 19487 radeon: Warning fixes. 19488 19489 The lid detection code probably wouldn't work on other non-x86 platforms 19490 though... 19491 19492commit 818ccf0fd4b5879171c5f20526d5a58638f8fde5 19493Author: Fredrik Höglund <fredrik@kde.org> 19494Date: Fri Dec 14 23:56:12 2007 +0100 19495 19496 RADEON: Fix the vertex coordinates for transformed pictures 19497 19498 This partially fixes transformed pictures on R100/R200 based 19499 cards. The texture still doesn't appear to be clamped correctly, 19500 but since that doesn't matter for rotations at perpendicular 19501 angles, I'm committing this now so randr rotation and reflection 19502 will work properly. 19503 19504commit 3cfbcf4cafbdfdb33411d16e51fb1f77cd0f93dd 19505Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19506Date: Fri Dec 14 17:11:00 2007 -0500 19507 19508 RADEON: Fix PLL set up on certain notebooks 19509 19510 Some LVDS panels require specific PLL dividers as 19511 specified in the bios tables. Make sure to use them 19512 if the output is LVDS. 19513 19514commit a84d446fd301d456bcea8f7abdc52e5a30776412 19515Author: Alex Deucher <alex@botch2.(none)> 19516Date: Fri Dec 14 02:17:14 2007 -0500 19517 19518 RADEON: select fb_div0 for LVDS on RV410 (x700) mobility 19519 19520 Fixes bug 8038 19521 I wonder if desktop RV410 need a similar fix. 19522 If your x700 laptop panel has problems after this let me know. 19523 19524commit b3eed3d87f76779b5a62a3115f99a31484dc38e0 19525Author: Alex Deucher <alex@samba.(none)> 19526Date: Fri Dec 14 00:20:10 2007 -0500 19527 19528 RADEON: fix typo in previous cursor fix 19529 19530commit 814c6c48aebba2e45ce257289b922cd7e92caf2a 19531Author: Alex Deucher <alex@samba.(none)> 19532Date: Thu Dec 13 18:45:09 2007 -0500 19533 19534 RADEON: rework PLL calculation 19535 19536 - Take into account the limits from the bios tables 19537 - Unify the PLL calculation between legacy and avivo chips 19538 19539commit f5ac34983411e4c4f41ab1817dce582830f398fd 19540Merge: f2b2e08 6ccf5b3 19541Author: Alex Deucher <alex@samba.(none)> 19542Date: Wed Dec 12 22:37:44 2007 -0500 19543 19544 Merge branch 'master' of git+ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 19545 19546 merge and fix conflicts 19547 19548commit f2b2e0804183b52d9d3f56ad85b3552ece76c544 19549Author: Alex Deucher <alex@samba.(none)> 19550Date: Wed Dec 12 22:18:37 2007 -0500 19551 19552 RADEON: fix rotation on avivo chips 19553 19554 There are still some issues, but this is better than before. 19555 19556commit 6ccf5b33d27218ae1c45ab93c122438ed536d8ba 19557Author: Alex Deucher <alex@botch2.(none)> 19558Date: Wed Dec 12 20:12:06 2007 -0500 19559 19560 RADEON: only enable vblanks if we want them 19561 19562 should fix bug 13610 19563 19564commit 1668f2056f56370f1b5681c13f1e14904e301216 19565Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19566Date: Wed Dec 12 19:39:08 2007 -0500 19567 19568 RADEON: use /proc/acpi to determine lid status 19569 19570 Linux only 19571 19572commit 33a39947f7f79533cd90007a17d57b20126642c6 19573Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19574Date: Wed Dec 12 18:50:18 2007 -0500 19575 19576 RADEON: fix cursors when using rotation 19577 19578 allocate separate cursor buffers for each crtc 19579 19580commit 9e5efdecd12092031a4aebce58747cb4a6f48f28 19581Author: Arkadiusz Miskiewicz <arekm@maven.pl> 19582Date: Tue Dec 11 23:53:03 2007 -0500 19583 19584 sparse fixes and cleanups from arekm 19585 19586commit 372bf41818fdafc6a9d2914aee3a8e359f668f02 19587Author: Alex Deucher <alex@botch2.(none)> 19588Date: Tue Dec 11 14:04:58 2007 -0500 19589 19590 RADEON: handle HMDI properly (untested) and fix some merge leftovers 19591 19592commit 3c22ad977c25d5ca2811821fcac6bb8ecd79994a 19593Merge: c9a0cee f3d2ec3 19594Author: Alex Deucher <alex@botch2.(none)> 19595Date: Tue Dec 11 13:11:15 2007 -0500 19596 19597 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 19598 19599 fix conflicts 19600 19601commit f3d2ec3a5ae61215c792018320158750e7aa937c 19602Author: Alex Deucher <alex@botch2.(none)> 19603Date: Tue Dec 11 11:57:27 2007 -0500 19604 19605 RADEON: rewrite PLL computation 19606 19607 Algorithm adapted from BeOS radeon driver with some tweaks by me. 19608 Actually calulate and use the reference divider rather than using the bios default. 19609 Also, always calculate the PLL, rather than falling back to bios dividers. 19610 This should fix bugs 12913, 13590, 13533, possibly others. 19611 19612commit c9a0cee97ca69e8fe1e1937c7670fa903214cded 19613Author: Dave Airlie <airlied@linux.ie> 19614Date: Tue Dec 11 06:03:46 2007 +1000 19615 19616 more zaphod fixes - some other work maybe needed 19617 19618commit 9b125312ab6edc585e4f5931a6a6de81e13b6acc 19619Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19620Date: Mon Dec 10 13:53:15 2007 -0500 19621 19622 RADEON: only update crtc values when RMX is active 19623 19624commit 3a161e1b5d80361e318ced8da5c19e797749d693 19625Author: Alex Deucher <alex@botch2.(none)> 19626Date: Mon Dec 10 00:57:26 2007 -0500 19627 19628 RADEON: bios PLL cleanup 19629 19630commit 731830297f2fc4a416882aacfb0d9b5f8ed32520 19631Author: Dave Airlie <airlied@linux.ie> 19632Date: Mon Dec 10 15:50:38 2007 +1000 19633 19634 fixup some warnings 19635 19636commit 2818e2b02ca90c9dfa50905b5311b2ae83ac3b0c 19637Author: Dave Airlie <airlied@linux.ie> 19638Date: Mon Dec 10 15:43:52 2007 +1000 19639 19640 add more to configure.ac for using out-of-tree mode src 19641 19642commit cc3c36100986f9d8060bc2d433373d4806f8e730 19643Author: Dave Airlie <airlied@linux.ie> 19644Date: Mon Dec 10 15:25:56 2007 +1000 19645 19646 add support for building against legacy servers similiar to Intel codepaths 19647 19648commit 9c278cb7fa7f18d13bde053fd75221cfba9da377 19649Merge: 6451ea2 cc167b9 19650Author: Dave Airlie <airlied@linux.ie> 19651Date: Mon Dec 10 15:18:03 2007 +1000 19652 19653 Merge branch 'zaphod-lolz' of git://git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 19654 19655 Conflicts: 19656 19657 src/radeon.h 19658 src/radeon_crtc.c 19659 src/radeon_driver.c 19660 src/radeon_output.c 19661 19662commit 6451ea2dcc4fac762442f699935864f4a8d445f7 19663Merge: 0d89556 5896ca4 19664Author: Dave Airlie <airlied@linux.ie> 19665Date: Mon Dec 10 15:08:42 2007 +1000 19666 19667 Merge branch 'master' into atombios-support 19668 19669commit 5896ca4097d439f59f90f397939132c061c3c364 19670Author: LisaWu <liswu@ati.com> 19671Date: Fri Dec 7 09:45:05 2007 +0100 19672 19673 radeon: Use %u instead of %d for unsigned value. 19674 19675commit df44f8380268c27d3978c4e91d736f093322b8b8 19676Author: Michel Dänzer <michel@tungstengraphics.com> 19677Date: Fri Dec 7 09:41:47 2007 +0100 19678 19679 radeon: Use gettimeofday instead of xf86getsecs. 19680 19681commit cc167b9bb7f1c3b8579e51e7bc2fca2f8eba6bd1 19682Author: Dave Airlie <airlied@redhat.com> 19683Date: Fri Dec 7 15:41:36 2007 +1000 19684 19685 disable tiling for zaphod heads 19686 19687commit 2ce8d192533a8c795714c5a9fb308ec74db40287 19688Author: Dave Airlie <airlied@redhat.com> 19689Date: Fri Dec 7 15:35:21 2007 +1000 19690 19691 don't add fboffset to info->FB it already is mapped at the offset 19692 19693commit 0dcd926d3092100854b3e362d6659d4950508aeb 19694Author: Dave Airlie <airlied@redhat.com> 19695Date: Fri Dec 7 14:45:04 2007 +1000 19696 19697 radeon: bring back zaphod all is forgiven. 19698 19699 You've whined, you've cried, you've nagged, and you're guilt trippin has 19700 made me do it... It actually wasn't as hard as I thought it would be. 19701 19702 Still not perfect, couple of things to fix yet 19703 19704commit bb5ede557bf32a42eef158ff0fbcfe1c6ede098a 19705Author: Dave Airlie <airlied@redhat.com> 19706Date: Fri Dec 7 14:30:32 2007 +1000 19707 19708 radeon: move savedreg/modereg into entity instead of info 19709 19710commit 64ab1cdf343a9a69e7e9e64f0bba77c54a94e9d0 19711Author: James Cloos <cloos@jhcloos.com> 19712Date: Thu Dec 6 15:51:12 2007 -0500 19713 19714 Add missing PHONY line for automatic ChangeLog generation 19715 19716commit 0d89556bfa41a3acbd6afe85b062e3a21f2ca057 19717Author: Dave Airlie <airlied@ppcg5.localdomain> 19718Date: Thu Dec 6 19:23:06 2007 +1100 19719 19720 powerpc: build fixes from last merge 19721 19722commit dbe3d2608ecc9896db9c23b3a347b50748c51e13 19723Merge: 48e31cd 21ed435 19724Author: Dave Airlie <airlied@redhat.com> 19725Date: Thu Dec 6 14:22:03 2007 +1000 19726 19727 Merge branch 'master' into atombios-support 19728 19729 Conflicts: 19730 19731 src/radeon_output.c 19732 19733commit 21ed435398e4a398dd8a0a5d7c1d4cc45e916332 19734Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19735Date: Tue Dec 4 17:08:58 2007 -0500 19736 19737 RADEON: add MacModel imac-g5-isight for iMac G5 iSight 19738 19739 Thanks to Étienne Bersac for helping to figure this out. 19740 19741commit 54bfd522405d9fdfb69d3a59e111ac3d63483dbb 19742Author: Étienne Bersac <bersace03@laposte.net> 19743Date: Tue Dec 4 14:22:42 2007 -0500 19744 19745 RADEON: fix typo 19746 19747commit 5022d006cfc06ca0395981526b2c2c94c6878567 19748Author: Michel Dänzer <michel@tungstengraphics.com> 19749Date: Sun Dec 2 17:27:33 2007 +0100 19750 19751 radeon: Further XVideo fixes. 19752 19753 * Make sure pitch constraints are always met for DMA upload blits. 19754 * RGB24 is not affected by endianness. 19755 19756commit 6ed55b70b23dfdc7b41103ea59c1df2bda5e41e6 19757Author: Kusanagi Kouichi <slash@ma.neweb.ne.jp> 19758Date: Sun Dec 2 17:18:46 2007 +0100 19759 19760 radeon: Fix crash with XVideo 24bit RGB images. 19761 19762 See https://bugs.freedesktop.org/show_bug.cgi?id=13274 . 19763 19764commit a697b590899bb7704ec4d7ae9a9c3cbbfcaef382 19765Author: Michel Dänzer <michel@tungstengraphics.com> 19766Date: Sun Dec 2 17:11:20 2007 +0100 19767 19768 Fix build against xserver master. 19769 19770 (DE)ALLOCATE_LOCAL are gone. 19771 19772commit 48e31cdaa0caa21573879af5b9267773fe89176a 19773Author: George Wu <geo@ocf.berkeley.edu> 19774Date: Sun Dec 2 15:25:09 2007 +1000 19775 19776 RADEON/R600: small code cleanup 19777 19778commit 00b4480aa2c5d7f751e34fc964f431b90b14c8d2 19779Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19780Date: Sat Dec 1 14:18:40 2007 -0500 19781 19782 RADEON: add options for force TV out as detected and to set TV standard 19783 19784 Also fix a typo in internal tv-out parsing 19785 19786commit 1e029fef5fe264f2ced445b80bf6070abcb84b82 19787Author: Alex Deucher <alex@samba.(none)> 19788Date: Sat Dec 1 00:58:51 2007 -0500 19789 19790 RADEON: move GPIO lookup to a separate function 19791 19792commit dcbef1ba9dfcf35c28e058832a55adf00afb472e 19793Author: Alex Deucher <alex@samba.(none)> 19794Date: Sat Dec 1 00:35:25 2007 -0500 19795 19796 RADEON: fix typo in previous commit 19797 19798 check gpio table revision before connector table revision 19799 19800commit fdce0598a2228c48c84deae1d7bebb2d7b3e979b 19801Author: Alex Deucher <alex@samba.(none)> 19802Date: Sat Dec 1 00:15:34 2007 -0500 19803 19804 RADEON: convert atombios connector table parsing to use ATOM structs 19805 19806 convert and add hpd info 19807 19808commit e3d7de9cc956aec5f940ad6db09e826b3a69523a 19809Author: Alex Deucher <alex@botch2.(none)> 19810Date: Fri Nov 30 20:14:42 2007 -0500 19811 19812 RADEON: remove unused cruft from last atom import 19813 19814commit d5d83411e8a884154d671aad440524507cce313e 19815Author: Alex Deucher <alex@botch2.(none)> 19816Date: Fri Nov 30 20:11:42 2007 -0500 19817 19818 RADEON: save/restore avivo crtc cursor control 19819 19820 this should prevent the cursor from showing up on in text 19821 mode or vesafb etc. after running the driver. 19822 19823commit af0196f7bf0d1d5d211391149c18935d64ed2b06 19824Merge: d9858a2 0175b79 19825Author: Alex Deucher <alex@botch2.(none)> 19826Date: Fri Nov 30 16:40:28 2007 -0500 19827 19828 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 19829 19830 merge master and fix conflicts 19831 19832commit d9858a2b3744b99003cfb9f31b743a2d31b322e9 19833Author: Dave Airlie <airlied@linux.ie> 19834Date: Sat Dec 1 06:49:59 2007 +1000 19835 19836 radeon: add in pll spread spectrum workaround 19837 19838commit 0175b79987ef4d7b0ce8238c3bdde989e504516a 19839Author: Alex Deucher <alex@botch2.(none)> 19840Date: Fri Nov 30 15:37:42 2007 -0500 19841 19842 RADEON: rework MacModel option 19843 19844 this brings in some previous research from Michel Dänzer, 19845 Sjoerd Simons, and myself. Hopefully, the driver will pick 19846 the correct MacModel in more cases. This also changes the 19847 default connector table for desktop Macs to dual DVI rather 19848 than DVI+VGA as that seems to be the case more often than not. 19849 External TMDS chips are handled separately now as well. 19850 Eventually we should add an option to allow the user to specify 19851 what external TMDS chip they need, but we don't have enough info 19852 yet, so we'll rely on OF to init the external chip in most cases 19853 for now. 19854 19855commit e1945f1f25a34310bd58ce128c8ff27ecc985618 19856Merge: b368b0f df7777b 19857Author: Alex Deucher <alex@botch2.(none)> 19858Date: Fri Nov 30 14:30:55 2007 -0500 19859 19860 Merge branch 'atombios-support' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 19861 19862commit b368b0f22cd1d7ef9b4c65d82929c76f3b82d573 19863Author: Alex Deucher <alex@botch2.(none)> 19864Date: Fri Nov 30 14:29:27 2007 -0500 19865 19866 RADEON: disable atom pll set for r4xx cards 19867 19868 the clocks do not get set correctly in all cases. this needs 19869 further investigation. 19870 19871commit 5af15739571c09260750bcfd3620e16fd7fec862 19872Author: Alex Deucher <alex@botch2.(none)> 19873Date: Fri Nov 30 14:24:30 2007 -0500 19874 19875 RADEON: small cleanup of pll code 19876 19877commit df7777bff40c1feabcc12d2148ad6ac5213efbb3 19878Author: George Wu <geo@ocf.berkeley.edu> 19879Date: Fri Nov 30 17:49:33 2007 +1000 19880 19881 Add LVTMA PWRSEQ registers to fix VT switching for LVDS 19882 19883commit 9840a0fd4fc8c980533fcd4a02c55cd0d5634b6d 19884Author: Alex Deucher <alex@botch2.(none)> 19885Date: Thu Nov 29 13:27:37 2007 -0500 19886 19887 RADEON: add MacModel "mini-internal" for minis with internal TMDS 19888 19889 Some macs (minis and powerbooks) use internal tmds, others use external tmds 19890 and not just for dual-link TMDS, it shows up with single-link as well. 19891 Unforunately, there doesn't seem to be any good way to figure it out. 19892 19893commit 017c939cf0a2b12fbdc1681cc70c28b23ae3b397 19894Author: Alex Deucher <alex@samba.(none)> 19895Date: Thu Nov 29 02:52:14 2007 -0500 19896 19897 RADEON: implement CLUT adjust support 19898 19899commit 9963b0fe01feb6dd0cb555b874a48f6fa3b255cb 19900Author: Alex Deucher <alex@samba.(none)> 19901Date: Thu Nov 29 00:46:23 2007 -0500 19902 19903 RADEON: fix cursor offset on avivo chips 19904 19905commit 6c56e3d7655b17e93e8823aefe34b05291104695 19906Author: Alex Deucher <alex@botch2.(none)> 19907Date: Tue Nov 27 15:27:36 2007 -0500 19908 19909 RADEON: switch r4xx to atombios load detection 19910 19911 works great 19912 19913commit 7561242e5b79bc2798ca3aace2b79e1a36949488 19914Author: Alex Deucher <alex@botch2.(none)> 19915Date: Tue Nov 27 14:50:36 2007 -0500 19916 19917 RADEON: re-org load detection for legacy chips 19918 19919commit febdcc2dccd42acbcd68ae630b7811cae5c58e8a 19920Author: Dave Airlie <airlied@linux.ie> 19921Date: Wed Nov 28 05:10:57 2007 +1000 19922 19923 legacy: fix fb/agp read/writes 19924 19925commit bb8545146959b748994be055d5b3de66ec66c8b2 19926Author: Alex Deucher <alex@samba.(none)> 19927Date: Mon Nov 26 17:34:51 2007 -0500 19928 19929 RADEON: first pass at TV/Component video 19930 19931 Untested and not likely to work just yet. 19932 19933commit e2bde646b864dca9056d9ecfe23a0d905647ea9a 19934Author: Alex Deucher <alex@samba.(none)> 19935Date: Mon Nov 26 14:35:57 2007 -0500 19936 19937 RADEON: move crtc output source selection into atombios_output.c 19938 19939 The function fits better as an output function and should now 19940 work with clones as well. 19941 19942commit 16c9be4107678a2a58d3418b7f1cc94d695ca8d6 19943Author: Alex Deucher <alex@samba.(none)> 19944Date: Mon Nov 26 14:20:54 2007 -0500 19945 19946 RADEON: add default connector table for avivo chips 19947 19948commit 6f080d00e6f4f84d5e0d6b4eff302bf42c230e81 19949Author: Arkadiusz Miskiewicz <arekm@maven.pl> 19950Date: Mon Nov 26 12:43:30 2007 -0500 19951 19952 RADEON: fix backlight control on some laptops 19953 19954 It seems the bios scratch regs are involved in backlight control 19955 on some laptops. This patch fixes the problematic laptops and doesn't 19956 seem to break the previous bios lid and output control fixes. 19957 19958commit dcf22aed87366f4625fb5042cb84fecccd9ceece 19959Author: Alex Deucher <alex@botch2.(none)> 19960Date: Mon Nov 26 11:10:03 2007 -0500 19961 19962 RADEON: only return status unknown for XPRESS chips 19963 19964 this seems to cause more issues than it attempted to fix 19965 so limit it to XPRESS chips for now. 19966 19967commit 206e280f02324641b4fe5a1986e26adf0e021fd4 19968Author: Alex Deucher <alex@botch2.(none)> 19969Date: Mon Nov 26 09:39:27 2007 -0500 19970 19971 RADEON: fix typo in man page 19972 19973commit 4e792db655dc92d2864db36b4d8f6714908de8e8 19974Author: Dave Airlie <airlied@redhat.com> 19975Date: Fri Nov 23 15:44:44 2007 +1000 19976 19977 r500: set default minimum pll freq 19978 19979commit a13b4c69c105c096dd05e6de2c5c154c9a8bcc71 19980Author: Dave Airlie <airlied@redhat.com> 19981Date: Fri Nov 23 15:25:06 2007 +1000 19982 19983 r5xx: cleanup pll code.. 19984 19985 Clean the PLL code to use the radeon pll structs. 19986 19987commit 5d792b5673bbf4784eb0ec059221e9b57232a122 19988Author: Dave Airlie <airlied@redhat.com> 19989Date: Fri Nov 23 15:03:13 2007 +1000 19990 19991 radeon: fix up memory mapping issues for vt switch 19992 19993commit 558a2ef266c1ca517c7fb464b0ccfef83238c913 19994Author: Dave Airlie <airlied@redhat.com> 19995Date: Fri Nov 23 14:39:32 2007 +1000 19996 19997 fix silly spaces 19998 19999commit 1bda4424a7031de437acfca9c41d4b3668e36051 20000Author: Dave Airlie <airlied@redhat.com> 20001Date: Fri Nov 23 14:39:19 2007 +1000 20002 20003 r600: add hi agp address for mc 20004 20005commit 133234c31a294f24a3968a576aad2bb8b89d0f6a 20006Author: Dave Airlie <airlied@redhat.com> 20007Date: Fri Nov 23 14:15:18 2007 +1000 20008 20009 atombios: use values from object header 20010 20011commit dbf6eae7e7a4bd1bc60fefdc7ab6276ed3f097c4 20012Author: Dave Airlie <airlied@redhat.com> 20013Date: Fri Nov 23 11:55:05 2007 +1000 20014 20015 atombios: add initial object header parsing for r600 cards 20016 20017commit 6b103915c11fc79d2efc43c44fc2a00c3bc64ede 20018Author: Dave Airlie <airlied@redhat.com> 20019Date: Fri Nov 23 09:24:20 2007 +1000 20020 20021 r500: make vt switch work again for me 20022 20023commit d24208276aad7669feeed527dced60c76d39eae6 20024Author: Dave Airlie <airlied@redhat.com> 20025Date: Fri Nov 23 08:50:25 2007 +1000 20026 20027 avivo: fix typo in register saving 20028 20029commit 197a62704742a4a19736c2637ac92d1dc5ab34ed 20030Author: Adam Jackson <ajackson@redhat.com> 20031Date: Thu Nov 22 20:26:23 2007 +1000 20032 20033 radeon: fix openoffice/render bug on r100 chips 20034 20035commit 64010fc4eae8359c01e430f64252931242c91435 20036Author: Dave Airlie <airlied@linux.ie> 20037Date: Thu Nov 22 20:25:31 2007 +1000 20038 20039 Revert "Disable RENDER acceleration by default on some RV200 chips." 20040 20041 This reverts commit 145da701bf4fb9c0ad9f95620b20914ae0126852. 20042 20043 pull in fix from ajax next commit 20044 20045commit 145da701bf4fb9c0ad9f95620b20914ae0126852 20046Author: Stefan Dirsch <sndirsch@suse.de> 20047Date: Thu Nov 22 08:38:09 2007 +0100 20048 20049 Disable RENDER acceleration by default on some RV200 chips. 20050 20051 Novell Bug #341699: Render acceleration is known to be broken 20052 on at least "Radeon 7500 QW" and "Radeon Mobility M7 LW". 20053 20054commit e810c3ae9908cd57e95b1b091cded87cbfc12fdc 20055Author: Roland Scheidegger <sroland@tungstengraphics.com> 20056Date: Thu Nov 22 02:37:55 2007 +0100 20057 20058 really do not set up surface regs for depth buf on r100-class igps (bug #13080) 20059 20060 fix the if condition testing for these chips... 20061 20062commit 0cc7e94849c1525750fabd04cf58f2dee88372e0 20063Author: Alex Deucher <alex@botch2.(none)> 20064Date: Wed Nov 21 17:06:17 2007 -0500 20065 20066 RADEON: reorder crtc/pll setup 20067 20068commit d56bde98efceaa8344e62f8e98db90c4bb642331 20069Author: Alex Deucher <alex@botch2.(none)> 20070Date: Wed Nov 21 17:03:39 2007 -0500 20071 20072 RADEON: fix crtc to output routing 20073 20074 Thanks to AMD for the information 20075 20076commit a12390c832abe423def60e39cd5a9118e5910339 20077Merge: d531792 e74dca1 20078Author: Alex Deucher <alex@botch2.(none)> 20079Date: Wed Nov 21 02:24:48 2007 -0500 20080 20081 Merge branch 'atombios-support' of git://git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 20082 20083commit d5317922f29a57b6c4127826a2fc126c5fd7c117 20084Author: Alex Deucher <alex@botch2.(none)> 20085Date: Wed Nov 21 02:23:37 2007 -0500 20086 20087 RADEON: attempt to fix crtc to output routing 20088 20089 The output routing seems to be based on the output ids from the bios 20090 connector tables and the connected status in the bios scratch regs. 20091 I don't fully understand this yet, but this seems to work 20092 for the most part, however changing modes can sometimes 20093 lead to a blanked head. This can be remedied by forcing 20094 a dpms off cycle. 20095 20096commit 81ce299bffd75540925b4c8234adf11226147165 20097Author: Alex Deucher <alex@botch2.(none)> 20098Date: Wed Nov 21 01:35:44 2007 -0500 20099 20100 RADEON: provide clearer debugging info 20101 20102commit e74dca19416b13f97db9d1fc06299b988057d6a4 20103Author: Dave Airlie <airlied@redhat.com> 20104Date: Wed Nov 21 16:24:25 2007 +1000 20105 20106 re-enable mobility chips 20107 20108commit 9c5b813dd6b3492cbc9833bc59792a5cec457e51 20109Author: Alex Deucher <alex@botch2.(none)> 20110Date: Wed Nov 21 01:22:42 2007 -0500 20111 20112 RADEON: major re-org and clean up of atom output control 20113 20114 - use radeon_output->devices to determine output 20115 - clean up and simplify dpms and mode set 20116 20117commit 7634cb6b96f938bc6615eb2c49ae75aaefd04cce 20118Author: Alex Deucher <alex@botch2.(none)> 20119Date: Wed Nov 21 00:10:14 2007 -0500 20120 20121 RADEON: make naming consistent and remove some cruft 20122 20123commit 908748343fc9a6cdc38af0fc028c63a82766da3f 20124Author: Alex Deucher <alex@botch2.(none)> 20125Date: Wed Nov 21 00:05:42 2007 -0500 20126 20127 RADEON: store devices ids from bios for each driver output 20128 20129commit 3975da2ea8cb628f7f66c3f26c5dfa181cd1c532 20130Merge: e283aa3 295ce27 20131Author: Alex Deucher <alex@botch2.(none)> 20132Date: Tue Nov 20 23:52:29 2007 -0500 20133 20134 Merge branch 'atombios-support' of git://git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 20135 20136commit e283aa332adf0134243a4fa3d14263719cd8a3fd 20137Author: Alex Deucher <alex@botch2.(none)> 20138Date: Tue Nov 20 23:35:46 2007 -0500 20139 20140 RADEON: add LVDS atom support 20141 20142commit e4bc3e1e7bb45571367d41b5328ff2590810b0f9 20143Author: Alex Deucher <alex@botch2.(none)> 20144Date: Tue Nov 20 18:09:29 2007 -0500 20145 20146 RADEON: enable/disable the right TMDS controller 20147 20148commit 3e47683ffaa44a89cda9bcddf530643befb27efa 20149Author: Alex Deucher <alex@botch2.(none)> 20150Date: Tue Nov 20 18:01:15 2007 -0500 20151 20152 RADEON: fixup for bios tables with wrong connector types 20153 20154commit 7412952eb1d1e9857cdab8417f7305f676900827 20155Author: Alex Deucher <alex@botch2.(none)> 20156Date: Tue Nov 20 18:00:12 2007 -0500 20157 20158 RADEON: switch to using ATOM defines for bios device table 20159 20160commit 295ce277bb0a44b9539b3dba575e7aff279dc2d0 20161Author: Dave Airlie <airlied@redhat.com> 20162Date: Wed Nov 21 08:56:40 2007 +1000 20163 20164 add missing files for make dist 20165 20166 noted by ndim on #radeonhd 20167 20168commit 20083b0695987b25e442ecbdec24f3cb6f1ac2ae 20169Author: Dave Airlie <airlied@redhat.com> 20170Date: Wed Nov 21 08:53:44 2007 +1000 20171 20172 LVDS on r500/r600 needs some work disable mobile chips for now 20173 20174commit e4b8a4479ddea9b083b3a763dc0b9302e7b9a82a 20175Author: Dave Airlie <airlied@redhat.com> 20176Date: Wed Nov 21 08:06:12 2007 +1000 20177 20178 r600: add memory controller regs from AMD 20179 20180commit aa88da974b97ea1e9bbb47b3494543575c09d912 20181Author: Dave Airlie <airlied@redhat.com> 20182Date: Wed Nov 21 08:01:35 2007 +1000 20183 20184 rs690 is !> r600 20185 20186commit c8872603454e6a4ffed9fc7d9adc2c364a429608 20187Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 20188Date: Tue Nov 20 22:33:39 2007 +1000 20189 20190 radeon: restructure pci ids to avoid effort later 20191 20192 This uses a single file with all the pciids and parameters for radeon 20193 family and setup in it. I don't run the perl script at build time to avoid 20194 a perl dependency on build but adding pci ids should be done via the csv file 20195 with openoffice or gnumeric if possible. 20196 20197commit a5b34c2f1f7d5346c4489cb30e07291d1217026c 20198Author: Dave Airlie <airlied@redhat.com> 20199Date: Tue Nov 20 16:37:29 2007 +1000 20200 20201 r600: use standard memory controller setup paths 20202 20203commit 4a523da5221d53f2efa49da2326500e9b0b9f14d 20204Author: Dave Airlie <airlied@redhat.com> 20205Date: Tue Nov 20 15:31:11 2007 +1000 20206 20207 r600: get r600 to work non-accelerated. 20208 20209 DDC still not working yet 20210 20211commit 80023441ba46882bc810ff3790c7148059f155f5 20212Author: Dave Airlie <airlied@redhat.com> 20213Date: Tue Nov 20 14:10:23 2007 +1000 20214 20215 r600: block r600 startup due to lack of memory controller info 20216 20217commit f6fbbacc17bf9b1073d3e993b225987fd9173182 20218Author: Dave Airlie <airlied@redhat.com> 20219Date: Tue Nov 20 13:41:55 2007 +1000 20220 20221 atombios: add warnings for r500 and r600 20222 20223commit 45a8b083c123b820c008f04ab857a64a8facec14 20224Author: Dave Airlie <airlied@redhat.com> 20225Date: Tue Nov 20 13:37:00 2007 +1000 20226 20227 atombios: add all r5xx and r6xx pci ids 20228 20229commit 5d023e2c3c2ab44ea57ffadc9607025d602c376c 20230Merge: 0d1e0c7 c887260 20231Author: Dave Airlie <airlied@redhat.com> 20232Date: Tue Nov 20 13:02:43 2007 +1000 20233 20234 Merge branch 'master' into atombios-support 20235 20236 Conflicts: 20237 20238 src/radeon_chipset.h 20239 src/radeon_driver.c 20240 src/radeon_probe.c 20241 20242commit 0d1e0c7805b3d8e56ccb49465e6b144afb7bdc51 20243Author: Dave Airlie <airlied@redhat.com> 20244Date: Tue Nov 20 09:08:04 2007 +1000 20245 20246 r5xx: add 71c5 for macbook pro 20247 20248commit d5909b30595c103bb5f42cd1704330f944bba49c 20249Author: Dave Airlie <airlied@redhat.com> 20250Date: Tue Nov 20 08:15:58 2007 +1000 20251 20252 r5xx: cleanups after last merge 20253 20254commit fe2f7a09050fb7a345a1f52239f8f3c4f1053891 20255Merge: 744c8cb 49055d8 20256Author: Dave Airlie <airlied@redhat.com> 20257Date: Tue Nov 20 08:04:32 2007 +1000 20258 20259 Merge branch 'master' into agd-atom-merge 20260 20261 Conflicts: 20262 20263 src/radeon_cursor.c 20264 src/radeon_output.c 20265 20266commit 744c8cb6c293fcaa687566f52901644e699baace 20267Merge: e258fbe e530af7 20268Author: Dave Airlie <airlied@redhat.com> 20269Date: Tue Nov 20 07:56:33 2007 +1000 20270 20271 Merge branch 'agd-atom' of ssh://people.freedesktop.org/~agd5f/xf86-video-ati-atom into agd-atom 20272 20273commit e258fbe411d255a1044b61d7ff738aee3fb5b7f4 20274Author: Dave Airlie <airlied@redhat.com> 20275Date: Mon Nov 19 16:35:05 2007 +1000 20276 20277 makes 2-headed cursor work 20278 20279commit e530af79adf51b3e95a0eca676c915a34dcbf4a7 20280Merge: 69e197f 52aba8d 20281Author: Alex Deucher <alex@botch2.(none)> 20282Date: Mon Nov 19 00:59:30 2007 -0500 20283 20284 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 20285 20286commit 69e197f2c8002aacf2587754c8d3bd63c88f85b1 20287Merge: 5e8940f 862dcab 20288Author: Alex Deucher <alex@botch2.(none)> 20289Date: Mon Nov 19 00:57:34 2007 -0500 20290 20291 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 20292 20293commit fca47ad083449f4cf9063dd970cdcebea6a7f110 20294Author: Dave Airlie <airlied@redhat.com> 20295Date: Mon Nov 19 15:53:40 2007 +1000 20296 20297 add z3ro's pciids 20298 20299commit 5e8940fa6e33d09091aa4bcf04b0f9e79596e1b8 20300Author: Alex Deucher <alex@botch2.(none)> 20301Date: Mon Nov 19 00:52:38 2007 -0500 20302 20303 fix logic in connector table check for TVs and switch counter to symbolic names 20304 20305commit c19123fd9483758eb6b286c3dffcb6d79d5b1ee5 20306Author: Dave Airlie <airlied@redhat.com> 20307Date: Mon Nov 19 15:46:58 2007 +1000 20308 20309 add firegl card on ajaxs machine 20310 20311commit f02f340e466a415b4e01648ca1e323f4ce125885 20312Author: Alex Deucher <alex@botch2.(none)> 20313Date: Mon Nov 19 00:39:19 2007 -0500 20314 20315 Don't assign a gpio for TV 20316 20317commit 52aba8d73189ba959f19c0437499d5e7a8829827 20318Merge: 862dcab 5e8940f 20319Author: Dave Airlie <airlied@redhat.com> 20320Date: Mon Nov 19 15:30:46 2007 +1000 20321 20322 Merge branch 'agd-atom' of ssh://people.freedesktop.org/~agd5f/xf86-video-ati-atom into agd-atom 20323 20324commit 862dcabfe0c10751d815e5cdd7436c10c2c2db10 20325Author: Dave Airlie <airlied@redhat.com> 20326Date: Mon Nov 19 15:30:08 2007 +1000 20327 20328 r520: nail i2c enable/disable issue 20329 20330commit f3dd7f413b670eeb6b8639f6677d72050ad5fe04 20331Author: Alex Deucher <alex@botch2.(none)> 20332Date: Mon Nov 19 00:19:39 2007 -0500 20333 20334 Don't detect TV out for now 20335 20336commit 94de0e22d7229ca71e18e1e849d8545d9ca7bafe 20337Author: Dave Airlie <airlied@redhat.com> 20338Date: Mon Nov 19 14:47:33 2007 +1000 20339 20340 i2c: fix bus enable stuff 20341 20342commit 3f1fc7eef13ea02fa5119e9b51d499841b801f2d 20343Author: Alex Deucher <alex@botch2.(none)> 20344Date: Mon Nov 19 00:02:14 2007 -0500 20345 20346 CRTs/DFPs may share a DVI port, but TV and CV don't 20347 20348commit 8f84c5ad4c4af14612ea68fe6f24d0d527f00acc 20349Author: Alex Deucher <alex@botch2.(none)> 20350Date: Sun Nov 18 23:43:06 2007 -0500 20351 20352 fix typo in loop 20353 20354commit 384cd8f52c89d089c6559e2eedbae45641fcd14e 20355Merge: f3f0e4e 234b607 20356Author: Dave Airlie <airlied@redhat.com> 20357Date: Mon Nov 19 14:02:55 2007 +1000 20358 20359 Merge branch 'agd-atom' of ../xf86-video-ati into agd-atom 20360 20361commit f3f0e4ec92c935c89ddb2f4241fe4335a521b439 20362Author: Alex Deucher <alex@botch2.(none)> 20363Date: Sun Nov 18 23:14:01 2007 -0500 20364 20365 RADEON: unify connectortype handling 20366 20367commit 234b6073054ac7630e82781683e666b94b2f12de 20368Author: Dave Airlie <airlied@redhat.com> 20369Date: Mon Nov 19 14:02:09 2007 +1000 20370 20371 restore avivo memory map registers at correct places 20372 20373commit 459a30ba511fe2fa8051380a9741fcfd9bb401ef 20374Author: Dave Airlie <airlied@redhat.com> 20375Date: Mon Nov 19 13:44:38 2007 +1000 20376 20377 fix type for r520 agp code 20378 20379commit 760af92412ef0d5cc44e52e7cec11fd80c4aaaeb 20380Author: Alex Deucher <alex@botch2.(none)> 20381Date: Sun Nov 18 22:34:59 2007 -0500 20382 20383 RADEON: unify DDC line handling 20384 20385commit e73bf6290da20dd61798ace775999ce1cb550934 20386Author: Dave Airlie <airlied@redhat.com> 20387Date: Mon Nov 19 13:32:16 2007 +1000 20388 20389 add x1900xt support 20390 20391commit 2e37937bacd624d616b91c41006c113791ebe98d 20392Author: Alex Deucher <alex@botch2.(none)> 20393Date: Sun Nov 18 20:18:50 2007 -0500 20394 20395 RADEON: step one in output rework 20396 20397 re-organize the output type 20398 20399commit 679e7a2e0d1b213524b8109193483bc9840fb116 20400Author: Alex Deucher <alex@botch2.(none)> 20401Date: Sun Nov 18 17:56:51 2007 -0500 20402 20403 Few fixes from the last commit. 20404 20405 Update parser works fine on r4xx. 20406 20407commit 1cd7cc3e6758ab1012f3ced6e958a1517f45557f 20408Author: Alex Deucher <alex@botch2.(none)> 20409Date: Sun Nov 18 17:44:36 2007 -0500 20410 20411 WIP: new atom code comples. 20412 20413 I commented out the object parsing for the time being as 20414 using it will require some thought as to new output 20415 related data structures. 20416 20417commit b155fa872ee4ca5d801e942aee6e619cef104f35 20418Author: Alex Deucher <alex@botch2.(none)> 20419Date: Sat Nov 17 00:34:56 2007 -0500 20420 20421 WIP: more new ATOM integration work 20422 20423commit 67db114d97abed7a607467e5d67c7b4ffa2c347e 20424Merge: 7d06a87 ea15346 20425Author: Alex Deucher <alex@botch2.(none)> 20426Date: Fri Nov 16 14:29:53 2007 -0500 20427 20428 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 20429 20430commit ea1534659de87d3d75eb20d808d039cff22cb537 20431Author: Dave Airlie <airlied@linux.ie> 20432Date: Fri Nov 16 18:46:02 2007 +1000 20433 20434 avivo: fixup some i2c stuff 20435 20436commit 7d06a8791839ce6b22e2449646832b79cebf1b21 20437Author: Alex Deucher <alex@botch2.(none)> 20438Date: Fri Nov 16 02:43:00 2007 -0500 20439 20440 WIP: sync up with the latest ATOM bios code in rhd 20441 20442 doesn't compile ATM 20443 20444commit 3614d80ceb9a7b3615b0baab3cf2dd34ed4ab464 20445Author: Dave Airlie <airlied@linux.ie> 20446Date: Fri Nov 16 17:22:39 2007 +1000 20447 20448 add missing hpd register 20449 20450commit 9a2715fda97ac0ebcb45650a416e0652aab575b8 20451Author: Dave Airlie <airlied@linux.ie> 20452Date: Fri Nov 16 17:22:25 2007 +1000 20453 20454 make i2c unlock/lock registers for gpios 20455 20456commit 20dc549fff9a4137c93ebed449d05e0c437b6bc1 20457Author: Dave Airlie <airlied@linux.ie> 20458Date: Fri Nov 16 17:01:33 2007 +1000 20459 20460 avivo i2c: consolidate the avivo i2c code 20461 20462commit 3e62730f79a13883a65a568bc821bc56055a4ab7 20463Author: Dave Airlie <airlied@linux.ie> 20464Date: Fri Nov 16 15:19:00 2007 +1000 20465 20466 atombios: fixup warnings in atombios files 20467 20468commit cca7af3c4910983f7f090792986fcbfa0dc97cfb 20469Author: Dave Airlie <airlied@linux.ie> 20470Date: Fri Nov 16 15:04:01 2007 +1000 20471 20472 remove avivo_reg.h 20473 20474commit d39eb2077c6b2fc094ccd952772528eb9428c587 20475Author: Dave Airlie <airlied@linux.ie> 20476Date: Fri Nov 16 15:00:50 2007 +1000 20477 20478 radeon: rename a large section of avivo regs to documented names 20479 20480commit b7774c28dde72a205a40be78003df72eabfb9b1f 20481Author: Dave Airlie <airlied@linux.ie> 20482Date: Fri Nov 16 14:48:36 2007 +1000 20483 20484 Add copyright headers 20485 20486commit 3cfcd2164b400bd0d1cb4ede8eeb01abba9d75c8 20487Merge: efac14e 718bfd3 20488Author: Alex Deucher <alex@botch2.(none)> 20489Date: Thu Nov 15 23:25:39 2007 -0500 20490 20491 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 20492 20493commit efac14e669a0c6184f8848191eb49ffb21934ee1 20494Author: Dave Airlie <airlied@linux.ie> 20495Date: Thu Nov 15 23:17:25 2007 -0500 20496 20497 r5xx: fix typo for crtc offset 20498 20499commit e6db621c37ff615be286462f000d67a662c5c331 20500Author: Alex Deucher <alex@botch2.(none)> 20501Date: Thu Nov 15 23:15:56 2007 -0500 20502 20503 fix INMC() and OUTMC() on !AVIVO chips 20504 20505 WR_EN is bit 8 so don't use OUTREG8. 20506 20507commit 52ba3fdd1ce05983fabedff234cfaf4c60fba38d 20508Author: Alex Deucher <alex@botch2.(none)> 20509Date: Thu Nov 15 23:12:30 2007 -0500 20510 20511 atombios_dac_detect() takes care of primary vs tv dac itself 20512 20513commit 49055d8aff91ff12186feaf5343c8fd2f96bcba0 20514Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20515Date: Thu Nov 15 22:56:09 2007 -0500 20516 20517 RADEON: set proper defaults for tv dac BGADJ/DACADJ 20518 20519 we should get these values from the bios tables, but for now use 20520 some reasonable defaults. This should fix the washed out color 20521 problems on bugs 1082 and 12844. 20522 20523commit 821acf38b716ab87c3d07263d6e4a139fe54803f 20524Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20525Date: Thu Nov 15 22:28:42 2007 -0500 20526 20527 RADEON: Make sure we set the MT properly for connected status unknown 20528 20529commit a94123f33ec6584fbdfc4b9ecd543d1357de8814 20530Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20531Date: Thu Nov 15 22:19:54 2007 -0500 20532 20533 Revert "Portability fix from netbsd" 20534 20535 This reverts commit c9264aa53bf1470ad9104d1e7c4a8ce13c49c270. 20536 This breaks damage support. See bug 13244 20537 20538commit 718bfd3b61879172eee819fdab7080d5d4c0a756 20539Author: Dave Airlie <airlied@linux.ie> 20540Date: Fri Nov 16 10:37:04 2007 +1000 20541 20542 r5xx: fix typo for crtc offset 20543 20544commit e9d721c31372db045550f9562534b28f16121bb9 20545Author: Roland Scheidegger <sroland@tungstengraphics.com> 20546Date: Tue Nov 13 23:42:42 2007 +0100 20547 20548 ignore sometime bogus agp_mode bit from chip (bug #13190) 20549 20550 bit is wrong on at least X700 cards with rialto pcie-agp bridge chip, 20551 should be safe to use just the bit from the bridge hopefully to make 20552 agp setup work on these cards and not adversely affect others. 20553 20554commit b865886d00907899297ae864358eb26d9980975d 20555Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20556Date: Sun Nov 11 20:38:41 2007 -0500 20557 20558 bump for RC release 20559 20560commit 7aeb35e5ad1aed6e78a3d8565fbfbfe66232ab45 20561Author: Alex Deucher <alex@botch2.(none)> 20562Date: Sun Nov 11 19:29:30 2007 -0500 20563 20564 fix from last commit 20565 20566commit 2ea95900547165e86ad3f8a41ce3331a05bad60e 20567Author: Alex Deucher <alex@botch2.(none)> 20568Date: Sun Nov 11 18:43:03 2007 -0500 20569 20570 Add full parsing support for atom bios connector table 20571 20572commit 7ce730828c293f0810dfdc554df48dfd76e35c49 20573Author: Alex Deucher <alex@botch2.(none)> 20574Date: Sun Nov 11 12:37:01 2007 -0500 20575 20576 rework crtc output source setup 20577 20578commit d61b6c78aa7810a2f9b9e2d9d95aab4295de80ce 20579Author: Alex Deucher <alex@botch2.(none)> 20580Date: Sun Nov 11 11:58:17 2007 -0500 20581 20582 make sure i2c bus exists before using it 20583 20584commit 342e3e207efda42ba679731c30dfb9d5e9d5643f 20585Author: Alex Deucher <alex@botch2.(none)> 20586Date: Fri Nov 9 17:11:43 2007 -0500 20587 20588 combine outputs based on id 20589 20590commit 8078c299d5941460243944d55051547c1a4d3791 20591Author: Alex Deucher <alex@botch2.(none)> 20592Date: Fri Nov 9 16:35:08 2007 -0500 20593 20594 use atom to program plls on r4xx 20595 20596commit 5febe2c96642f61d006abe6e8081e69d5b95adc0 20597Author: Alex Deucher <alex@botch2.(none)> 20598Date: Fri Nov 9 16:24:56 2007 -0500 20599 20600 turn off vga control when using ext modes 20601 20602commit 83f170c32c08c74a9e3466ffa0e0a0606c74427b 20603Author: Alex Deucher <alex@botch2.(none)> 20604Date: Thu Nov 8 19:33:13 2007 -0500 20605 20606 add pci ids for X1550PRO 20607 20608commit 5cdcaba0f6e9de2d15cfcc109ab97d9fd423e3bf 20609Author: Alex Deucher <alex@botch2.(none)> 20610Date: Thu Nov 8 19:28:03 2007 -0500 20611 20612 make sure to assign gpio 20613 20614commit 2dcb852778301b9284a2b4906dcf64f95ed638b7 20615Author: Alex Deucher <alex@botch2.(none)> 20616Date: Thu Nov 8 18:39:23 2007 -0500 20617 20618 pull in another of Dave's fixes 20619 20620commit 96273016a0bbdfa4d3a4e6275a3b09eeeadaa534 20621Author: Alex Deucher <alex@botch2.(none)> 20622Date: Wed Nov 7 01:22:30 2007 -0500 20623 20624 fix and move crtc source set up atombios_output.c 20625 20626 it's really more output related. 20627 20628commit 5c495c81cc3bcd4a38d06954243ed3bdc85bdc07 20629Author: Alex Deucher <alex@botch2.(none)> 20630Date: Wed Nov 7 01:04:11 2007 -0500 20631 20632 add support for initing external tmds via ATOM 20633 20634commit 5c13d9355280e6de44ebbf8de7ea89a6b91c7388 20635Author: Alex Deucher <alex@botch2.(none)> 20636Date: Wed Nov 7 00:17:28 2007 -0500 20637 20638 add avivo output stuff 20639 20640commit 68e7f5c67e2e9d2162b469ce31f452f3f89756b5 20641Author: Alex Deucher <alex@botch2.(none)> 20642Date: Tue Nov 6 23:43:29 2007 -0500 20643 20644 more avivo updates 20645 20646commit 303562dfb57e13c027b2aa9289d54e547c829ff1 20647Author: Alex Deucher <alex@botch2.(none)> 20648Date: Tue Nov 6 23:06:46 2007 -0500 20649 20650 add additional connector types 20651 20652commit 0d3e0735f710cb7b9505e4330997aa332f73c102 20653Author: Alex Deucher <alex@botch2.(none)> 20654Date: Tue Nov 6 22:59:25 2007 -0500 20655 20656 First round of avivo support 20657 20658commit 0abfe3150ce3eed4db93ccc2975bd4622dfa54a7 20659Author: Alex Deucher <alex@botch2.(none)> 20660Date: Tue Nov 6 18:47:00 2007 -0500 20661 20662 Add atombios files 20663 20664commit 20f01950e42babc308b4470df6a3c6628c932003 20665Author: Alex Deucher <alex@botch2.(none)> 20666Date: Tue Nov 6 18:04:43 2007 -0500 20667 20668 for r4xx ATOM cards, just use ATOM for PLL 20669 20670 while crtc timing and pll seem to work fine, output setup 20671 and routing don't seem to work too reliably with atom. 20672 AMD claims ATOM was still pretty new with r4xx so 20673 it's probably better to stick with direct programming for 20674 some things. 20675 20676commit 37af2f6b52c8f217b205486742a6b46a864da8ed 20677Author: Michel Dänzer <michel@tungstengraphics.com> 20678Date: Mon Nov 5 10:58:25 2007 +0100 20679 20680 radeon: Use indirect register access for (un)hiding the cursor. 20681 20682 This avoids hangs that occur when touching these registers while the card is 20683 busy with some cards. 20684 20685commit 5659e2f1b08e44d0c4f21a403e91b4801ef78f62 20686Author: Michel Dänzer <michel@tungstengraphics.com> 20687Date: Mon Nov 5 10:32:04 2007 +0100 20688 20689 radeon: Fix build warning on powerpc. 20690 20691commit d41d73f5f9d8baee7137e94da2c7852ea0412180 20692Author: Michel Dänzer <michel@tungstengraphics.com> 20693Date: Mon Nov 5 10:30:23 2007 +0100 20694 20695 Revert "radeon: Reinstate sync in radeon_crtc_show/hide_cursor." 20696 20697 This reverts commit 0241cac643fa1c08a45ea44f5c670b290e760ad8. 20698 20699 Taking the DRI lock in paths called from the SIGIO handler isn't safe. For 20700 example, the SIGIO handler may be called while the lock is held by an AIGLX 20701 context, resulting in deadlock. See e.g. 20702 https://bugs.freedesktop.org/show_bug.cgi?id=13005 . 20703 20704commit 78a3eabff382e8ebe33df2039076fb083bcc361b 20705Author: Alex Deucher <alex@botch2.(none)> 20706Date: Sun Nov 4 14:11:26 2007 -0500 20707 20708 WIP: get ATOM crtc stuff working on r4xx 20709 20710commit 5e9ebd8e496b72b051053d637c63b2956b7861d3 20711Author: Alex Deucher <alex@botch2.(none)> 20712Date: Sat Nov 3 20:46:17 2007 -0400 20713 20714 Initial support for r4xx 20715 20716 - add r3xx/r4xx MC access macros and functions 20717 20718commit c106075ccb81ca2ee4894743e676fd37653c8dce 20719Author: Alex Deucher <alex@botch2.(none)> 20720Date: Sat Nov 3 18:29:20 2007 -0400 20721 20722 More of Dave's ATOM init code. 20723 20724commit e18f5d61806b445ad77d93e258fbce9422b52bb6 20725Author: Alex Deucher <alex@botch2.(none)> 20726Date: Sat Nov 3 18:20:55 2007 -0400 20727 20728 Initial integration of Atom code and some of Dave's code. 20729 20730 ATOM builds, but it's not hooked up yet. 20731 20732commit aaafc6a8f33d4736c37c43d3fd5e43085a2590fb 20733Author: Brice Goglin <bgoglin@debian.org> 20734Date: Sun Oct 28 21:10:32 2007 +0100 20735 20736 Fix typo in 'disposing cursor' 20737 20738 Bug #12984. 20739 20740commit a5c55c418319c1315ad73c305ef48b44fe1d9580 20741Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20742Date: Sat Oct 27 19:46:11 2007 -0400 20743 20744 RADEON: disable FP DETECT for DVI. 20745 20746 These bits don't seem to be reliable or routed right 20747 in all cases. 20748 20749commit 49cf7cb3db36ce7734f7e314a040a240191d6477 20750Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20751Date: Thu Oct 25 21:23:09 2007 -0400 20752 20753 RADEON: Fix IgnoreEDID option 20754 20755commit 70bbc0cfe699588f8ae4123efca0e1baffc91000 20756Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20757Date: Thu Oct 25 20:56:11 2007 -0400 20758 20759 RADEON: clean up mode handling and improve reliability of DDC 20760 20761 - Reduce the number of times we do DDC and don't 20762 skip the magical GPIO dance when doing DDC. 20763 - If no DDC modes are found, add the screen modes 20764 20765commit 1a4f590c2ac071f0c348f38e9f26fa6093722453 20766Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20767Date: Thu Oct 25 19:25:36 2007 -0400 20768 20769 RADEON: document DefaultTMDSPLL option 20770 20771commit 2d26e0d0bcf25e210dc7771e76638dcfafcc2356 20772Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20773Date: Thu Oct 25 19:18:18 2007 -0400 20774 20775 RADEON: Add driver option to force TV DAC load detection on 20776 20777commit a9306b7986467c0794c91625ad2ece597b718462 20778Author: Alex Deucher <alex@botch2.(none)> 20779Date: Thu Oct 18 20:06:58 2007 -0400 20780 20781 RADEON: fix possible crash in radeon_crtc_mode_set() 20782 20783 when radeon_crtc_mode_set() is called during ScreenInit(), 20784 we don't have pScrn->pScreen yet. 20785 20786commit 115b4b65ef679a46d90b7fc8ac22ace37038b3fc 20787Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20788Date: Wed Oct 17 20:25:47 2007 -0400 20789 20790 RADEON: Remove LVDSBiosNativeMode Option 20791 20792 Shouldn't be needed any longer as I've finally sorted out 20793 the LVDS issues due to the crtc setup. 20794 20795commit f2eb5c1cc69a4f7b0754ec6b2efde4fa1650ab6d 20796Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20797Date: Tue Oct 16 18:54:28 2007 -0400 20798 20799 RADEON: finish fixing LVDS/RMX 20800 20801 For panels, both the mode and crtc values need to be updated based 20802 on the native mode timing. This shoudl fix any remaining problems 20803 with RMX. 20804 20805commit 93561688087493fccb7ac387d728feec61212180 20806Author: Michel Dänzer <michel@tungstengraphics.com> 20807Date: Tue Oct 16 10:57:06 2007 +0200 20808 20809 radeon: Handle 64 bit PCI resource bases better. 20810 20811commit 08945fc8a14ba57e32350e85daa3354619e9e72e 20812Author: Alex Deucher <alex@botch2.(none)> 20813Date: Mon Oct 15 21:12:46 2007 -0400 20814 20815 RADEON: allow the user to set any tv standard 20816 20817 allow the user to set any tv standard rather than 20818 limiting it to what the bios says is supported. 20819 20820commit 571548b2b95ce289f26a77c11deb6266f0b18516 20821Author: Alex Deucher <alex@botch2.(none)> 20822Date: Mon Oct 15 20:33:56 2007 -0400 20823 20824 RADEON: use native timing for RMX 20825 20826 RMX should work again with native timings after the previous 20827 patch. 20828 20829commit 2f87bff293a343b40c1be096933a5ae126632468 20830Author: Alex Deucher <alex@botch2.(none)> 20831Date: Mon Oct 15 20:06:28 2007 -0400 20832 20833 RADEON: Fix subtle change in crtc reg init 20834 20835 At some point we changed how hsync_wid and vsync_wid were clipped. 20836 Previously we used the field size as a mask when building the 20837 register. This got changed to setting the value to the field 20838 size if it was greater. this probably explains number stange 20839 mode bugs. 20840 20841commit fd5bb7bb5e968127b87102320eccc1222f205e5a 20842Author: Michel Dänzer <michel@tungstengraphics.com> 20843Date: Sat Oct 13 18:56:43 2007 +0200 20844 20845 radeon: Improve detection of default value for Option "MacModel" on Linux. 20846 20847 * Detect all PowerBooks with dual link DVI (according to developer.apple.com) 20848 and both Mac Mini G4 models. Other PowerBooks and iBooks should be covered 20849 by the 'detected as' line in /proc/cpuinfo. 20850 * Give specific instructions for overriding and reporting incorrect detection. 20851 * Only perform detection when Option "MacModel" isn't present with a valid 20852 value. 20853 * Close /proc/cpuinfo file handle when done with it. 20854 * Coding style cleanups. 20855 20856commit fbf121add5584049627f07345fc502b2aefc88ce 20857Author: Eugen Dedu <Eugen.Dedu@pu-pm.univ-fcomte.fr> 20858Date: Sat Oct 13 16:51:58 2007 +0200 20859 20860 radeon: Derive default value of Option "MacModel" from /proc/cpuinfo on Linux. 20861 20862commit 5db3afaa1fdb69d382ac769ef40191a4b964d28e 20863Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20864Date: Thu Oct 11 20:25:20 2007 -0400 20865 20866 RADEON: return status unknown for flaky chip/connector combinations 20867 20868 This should at least get something on the screen. 20869 XPRESS chips, I'm looking in your general direction... 20870 20871commit 1148d332f52f3780897aae3fcd5d6a67687c42cd 20872Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20873Date: Thu Oct 11 19:09:45 2007 -0400 20874 20875 RADEON: fix potential crash in mode handling 20876 20877commit 905b19df33d6f912336b65895e97775ffec07728 20878Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20879Date: Thu Oct 11 18:55:08 2007 -0400 20880 20881 RADEON: small cleanup from last commit 20882 20883commit 7afd04c1e4ffa6e4e5ba08ae90ba002237dc282b 20884Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20885Date: Thu Oct 11 00:16:45 2007 -0400 20886 20887 RADEON: tell the bios not to muck with the hardware while the driver is active 20888 20889 by toggling the appropriate bios scratch regs you can tell 20890 the bios not the touch the hw while the driver is active. 20891 This should prevent the bios from scrambling the hardware 20892 when users open the lid or toggle bios hotkeys. 20893 20894 fixes bug 12567 20895 20896commit 1b231d28fdda5cdc44bb9d2075d4edfd8f17e21f 20897Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20898Date: Wed Oct 10 21:54:46 2007 -0400 20899 20900 RADEON: attempt to detect lid status on legacy bios 20901 20902 I'm not sure how reliable this is. The lid closed 20903 bit only seems to be set if the lid is closed at boot 20904 so LVDS will default to on if you restart X with the 20905 lid closed after boot. 20906 20907commit 47ef583b993607bc3a372dd125c6e5228c4ec7bf 20908Author: Alex Deucher <alex@botch2.(none)> 20909Date: Wed Oct 10 19:35:31 2007 -0400 20910 20911 RADEON: more XPRESS connector table hacks 20912 20913 Seems XPRESS desktop chips with DVI-D list a proprietary 20914 connector type rather than DVI-D. try and do the right thing. 20915 Also, XPRESS chips seem to include connector type 8 and ddc 20916 type 8 which are unknown at this time. 20917 20918commit c9264aa53bf1470ad9104d1e7c4a8ce13c49c270 20919Author: Joerg Sonnenberger <joerg@britannica.bec.de> 20920Date: Tue Oct 9 23:18:29 2007 -0400 20921 20922 Portability fix from netbsd 20923 20924commit 80eee856938756e1222526b6c39cee8b5252b409 20925Author: Matthieu Herb <> 20926Date: Tue Oct 9 16:17:50 2007 -0400 20927 20928 RADEON: fix console restore on netbsd 20929 20930 Include the mode restore bugfix from monolithic Xorg, that is derived 20931 from the version in xsrc which in turn was provided by Matthieu Herb 20932 over 3 years ago on the XFree86 lists. Suggested by various 20933 developers, hold-back due to the working state in xorg-server 1.1.1. 20934 Tracing down the exact change showed that the changed default color 20935 depth made this issue a lot more prominent again. Discussed with Eric 20936 Anholt. 20937 20938commit 547fef4c6382f8a4951c086ee531b804cba31075 20939Author: Lars Weiler <pylon@gentoo.org> 20940Date: Tue Oct 9 16:10:34 2007 -0400 20941 20942 include stdint.h to fix the build on some platforms 20943 20944 fixes bug 12733 20945 20946commit ddec5783103045084a66a34d71326cc6e4a3f598 20947Author: George Sapountzis <gsap7@yahoo.gr> 20948Date: Tue Oct 9 13:27:27 2007 +0300 20949 20950 theatre compiles with pci-rework. 20951 20952commit 051435610a66735fd455bbb526fa294fcfe8c0b6 20953Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20954Date: Sun Oct 7 19:39:47 2007 -0400 20955 20956 RADEON: still more LVDS fixes 20957 20958 Seems some laptops need the native mode from the bios for 20959 LVDS while others seem to prefer a CVT mode. Add an option 20960 to pick the preferred mode. The default it to use the bios 20961 table timing. 20962 20963commit 0b03a73b7dcb4aa192c42f2a4c842d324c358122 20964Author: Dave Airlie <airlied@linux.ie> 20965Date: Sat Oct 6 20:11:19 2007 +1000 20966 20967 radeon: add support for DDC type 6 20968 20969 thanks to arekm on xorg-devel for testing/feedback 20970 20971commit 83d7506a603889b6c18736cff9b2cd3296c95c84 20972Author: Alex Deucher <alex@botch2.(none)> 20973Date: Fri Oct 5 20:25:04 2007 -0400 20974 20975 Bump for RC release 20976 20977commit fe66cc4e02a52e1d014ae5de192c65e8abd7f58a 20978Author: Tormod Volden <bugzi06.fdo.tormod@xoxy.net> 20979Date: Fri Oct 5 20:10:55 2007 -0400 20980 20981 RADEON: rename to PCI_CHIP_RS485_5975 everywhere 20982 20983 RS482_5975 is actually RS485. see bug 12048 20984 20985commit 314e8fdbe5369e4c1786d176d03d0d2a046f25d5 20986Author: Tormod Volden <bugzi06.fdo.tormod@xoxy.net> 20987Date: Fri Oct 5 20:05:54 2007 -0400 20988 20989 RADEON: Disable DRI by default for RS482 cards 20990 20991 fixes bug 12048 20992 20993commit abbafe6db7c0d24efc61a05950d4214c00d3f043 20994Author: Michel Dänzer <michel@tungstengraphics.com> 20995Date: Thu Oct 4 11:33:09 2007 +0200 20996 20997 radeon: Fix compile warnings. 20998 20999commit 1110957ba703475847b8e72023c6edcbd1068aef 21000Author: Michel Dänzer <michel@tungstengraphics.com> 21001Date: Thu Oct 4 11:32:48 2007 +0200 21002 21003 radeon: Fix for pci-rework. 21004 21005 Surprisingly easy, thanks to George's pci-rework changes. 21006 21007commit 7addf41885ec5658f531624a9c24ea5bd7d22d19 21008Merge: 3a958ba 78c6bd3 21009Author: Michel Dänzer <michel@tungstengraphics.com> 21010Date: Thu Oct 4 11:06:08 2007 +0200 21011 21012 Merge remote branch 'origin/pci-rework' 21013 21014commit 3a958ba136c3fae5a4ddd56373ac7cd47046f10e 21015Author: Alex Deucher <alex@botch2.(none)> 21016Date: Wed Oct 3 22:59:14 2007 -0400 21017 21018 RADEON: remove some cruft 21019 21020commit b6bda79f72df5e5bf9c6b71fa3298e765da506bd 21021Author: Alex Deucher <alex@botch2.(none)> 21022Date: Wed Oct 3 22:56:27 2007 -0400 21023 21024 RADEON: more fixes for single crtc chips 21025 21026commit 5f5c4e6ad61c45c24f1443b91b4bc5375efdebc0 21027Author: Alex Deucher <alex@botch2.(none)> 21028Date: Wed Oct 3 22:46:21 2007 -0400 21029 21030 RADEON: remove RADEONSaveMode() 21031 21032 It's only called by RADEONSave(), so just move its 21033 contents into RADEONSave(). this mirrors RADEONRestore(). 21034 21035commit bfede412b3a3cd11769a580b167c528734146096 21036Author: Alex Deucher <alex@botch2.(none)> 21037Date: Wed Oct 3 22:19:16 2007 -0400 21038 21039 RADEON: minor fixes for external TMDS 21040 21041commit 597dffce9bdc200003d0be880235258386a0bdd7 21042Author: Alex Deucher <alex@botch2.(none)> 21043Date: Wed Oct 3 21:43:08 2007 -0400 21044 21045 RADEON: final fix for RMX/LVDS 21046 21047 It seems that on quite a few laptops the LVDS H/V timing 21048 from the bios tables are way off or just doesn't work period. 21049 Either we are using it wrong or we need some additional checks 21050 when we parse it. Only the dot clock seems to really matter, 21051 so use cvt modes and update the dotclock. This seems to work 21052 correctly in all cases. 21053 21054commit cc0c2d8e61600652b1f9cb3dc49db2ef62b1e40d 21055Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21056Date: Wed Oct 3 00:59:37 2007 -0400 21057 21058 RADEON: Finally sort out LVDS modes 21059 21060 - The panel timing from the bios tables is for the crtc, 21061 not the native panel mode, so add cvt modes for the panel 21062 and fix up the crtc values in mode_fixup(). 21063 - rename and reconfigure RADEONValidateFPModes() to what it 21064 really does: add screen modes. I suppose for backwards 21065 compatibility we ought to add the screen modes to every 21066 reasonable output, but everyone should really use xrandr 21067 or an output monitor section to add custom modes. Hopefully 21068 this will go away at some point. 21069 21070commit 0ca184c3c35032df39ea7ce5d2d4aba1a97b6426 21071Author: Alex Deucher <alex@botch2.(none)> 21072Date: Tue Oct 2 19:38:18 2007 -0400 21073 21074 RADEON: fix RADEONInitDispBandwidth() on single crtc cards 21075 21076 Should fix bug 12490 21077 21078commit a5fb2a5d54cdd562a97881cdea1a4af35b483aab 21079Author: Dave Airlie <airlied@linux.ie> 21080Date: Mon Oct 1 08:48:27 2007 +1000 21081 21082 radeon: fix powerpc build 21083 21084commit 22519fde1e002f28d6036d448fcd18452d00f1bb 21085Author: Alex Deucher <alex@botch2.(none)> 21086Date: Sun Sep 30 13:11:20 2007 -0400 21087 21088 RADEON: add support for ext tmds table and ext tmds chip init 21089 21090 This probably won't work on all chips as the various gpio lines 21091 seem to need special magic to to actually talk to the i2c slave 21092 chips. 21093 21094commit dcc376e2d2a13329dd03f1bc4b471329757a6f5f 21095Author: Alex Deucher <alex@botch2.(none)> 21096Date: Wed Sep 26 23:27:38 2007 -0400 21097 21098 RADEON: remove cruft 21099 21100commit 99ceaefa18c6e07b55106cca0ea8996fa73667be 21101Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21102Date: Sun Sep 30 12:02:54 2007 -0400 21103 21104 RADEON: more work on LVDS mode validation and fixups 21105 21106 Hopefully this will fix up the weird mode issues with LVDS and 21107 native and RMX modes, but I can't seem to get quite the right 21108 combo to fix everyone. 21109 21110commit d808781d48adf01e80b5bb476bae2d2f599030f1 21111Author: Dave Airlie <airlied@redhat.com> 21112Date: Fri Sep 28 14:50:42 2007 +1000 21113 21114 rc410: disable DRI by default due to it not working 21115 21116commit 4d9f944e40c6b9ac7e80428dff49c6e83fda1d8f 21117Author: Michel Dänzer <michel@tungstengraphics.com> 21118Date: Thu Sep 27 17:50:15 2007 +0200 21119 21120 radeon: Fix build on little endian platforms. 21121 21122 This was accidentally broken when moving away from using pixmap data pointers 21123 directly. 21124 21125commit 5902e913010a2fb0c17efe01f9aade73f9646f16 21126Author: Michel Dänzer <michel@tungstengraphics.com> 21127Date: Thu Sep 27 12:53:34 2007 +0200 21128 21129 radeon: EXA pitch fixes. 21130 21131 The combined pitch/offset registers only support pitches up to 16320 bytes. 21132 21133 With EXA >= 2.3, set the maxPitchBytes field accordingly. With older versions, 21134 limit maxX such that the pitch of 32bpp pixmaps doesn't exceed the limit. 21135 21136 Also check the limit in RADEONGetOffsetPitch just in case. 21137 21138 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=12551 . 21139 21140commit ca728973bce504e03bf6228d18287906c35eaa3f 21141Author: Michel Dänzer <michel@tungstengraphics.com> 21142Date: Thu Sep 27 11:14:21 2007 +0200 21143 21144 radeon: Don't use pixmap data pointer directly with EXA. 21145 21146 Fixes segfault with current xserver master without AccelDFS. 21147 21148commit 21593d04d222b05dbba9abd31eaa3bfb91d999b6 21149Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21150Date: Mon Sep 24 23:01:20 2007 -0400 21151 21152 RADEON: more clean up of mode code 21153 21154 also print out LVDS mode info from bios for both 21155 legacy and atom bioses 21156 21157commit 22cccb99d762567757d3fd38795f71e943f1faf9 21158Author: Alex Deucher <alex@botch2.(none)> 21159Date: Mon Sep 24 00:10:01 2007 -0400 21160 21161 bump for RC release 21162 21163commit fad9e7b00de051f632000eb2304ac23a524f7c8e 21164Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21165Date: Sun Sep 23 23:43:26 2007 -0400 21166 21167 RADEON: Rework LVDS mode validation 21168 21169 The old validation code wasn't really well suited to randr. 21170 This fixes several issues: 21171 - missing display size for panels with edid 21172 - broken duplicate modes 21173 21174commit bf9674b9df351108e069d037ec10869a4154a881 21175Author: Alex Deucher <alex@botch2.(none)> 21176Date: Sun Sep 23 15:01:34 2007 -0400 21177 21178 RADEON: properly set default RMX types 21179 21180commit b27135bce8b41d69290613b440a338b0a7fe0200 21181Author: Alex Deucher <alex@botch2.(none)> 21182Date: Sun Sep 23 14:44:38 2007 -0400 21183 21184 RADEON: fix up LVDS handling for r3xx and newer 21185 21186commit d5cf268be2c9b3e320d7101f5213f7d5609b6308 21187Author: Alex Deucher <alex@botch2.(none)> 21188Date: Sat Sep 22 18:31:51 2007 -0400 21189 21190 RADEON: Fix crash when Xv window is outside of either crtc 21191 21192commit 8aca0ca6be7122e0ece20985a1862553a1b225c0 21193Author: Alex Deucher <alex@botch2.(none)> 21194Date: Sat Sep 22 17:56:50 2007 -0400 21195 21196 Revert "RADEON: fix crash when Xv window is outside of either crtc" 21197 21198 This reverts commit 9109e62e3be7f96b41b534ab517fdf1baf458806. 21199 21200 This breaks ABI. better fix to come. 21201 21202commit d6e1d9be1cd5786a24ed4bb8e45fa7e04df56622 21203Author: Alex Deucher <alex@botch2.(none)> 21204Date: Sat Sep 22 16:14:06 2007 -0400 21205 21206 RADEON: Don't make the entity as shareable 21207 21208 This should prevent people from mistakenly trying to run 21209 zaphod mode 21210 21211commit 81114af1cdddc0b10d076f2e38c7a00c1223cc48 21212Author: Alex Deucher <alex@botch2.(none)> 21213Date: Sat Sep 22 15:51:23 2007 -0400 21214 21215 RADEON: preliminary support for mac mini 21216 21217 Option "MacModel" "mini" 21218 may not be 100% correct yet 21219 21220commit 6c482e453bc8156886294d0c1b8f3f1b3dcf4b36 21221Author: Alex Deucher <alex@botch2.(none)> 21222Date: Sat Sep 22 15:11:20 2007 -0400 21223 21224 RADEON: Fix RMX on LVDS 21225 21226 LVDS + RMX doesn't seem to like having the crtc values tweaked. 21227 21228commit 9109e62e3be7f96b41b534ab517fdf1baf458806 21229Author: Michel Dänzer <daenzer@debian.org> 21230Date: Sat Sep 22 14:19:10 2007 -0400 21231 21232 RADEON: fix crash when Xv window is outside of either crtc 21233 21234 Should fix bug 12175 21235 21236commit d2ce4a5003ce1291ea2327b2c00a0b24408fe26c 21237Author: Alex Deucher <alex@botch2.(none)> 21238Date: Sat Sep 22 10:41:55 2007 -0400 21239 21240 RADEON: RMX updates 21241 21242 - add option to turn off RMX 21243 - turn off RMX by default on DVI 21244 - add infrastructure to support more RMX modes 21245 21246commit f95b9ab729376083bf0d12987ee260ec2aba721e 21247Author: Alex Deucher <alex@botch2.(none)> 21248Date: Fri Sep 21 00:12:37 2007 -0400 21249 21250 RADEON: remove more old cruft 21251 21252commit 5a6f74103f0ec0d451d0e2573442efe5922848af 21253Author: Maciej Cencora <m.cencora@gmail.com> 21254Date: Thu Sep 20 23:56:08 2007 -0400 21255 21256 RADEON: fix video in on RV380 (tested on X600 VIVO) 21257 21258commit c72a365386e19f9257db041d44b09ad499cc9f6a 21259Author: Alex Deucher <alex@botch2.(none)> 21260Date: Thu Sep 20 23:49:57 2007 -0400 21261 21262 RADEON: fix up dvo support (still no external chip init) 21263 21264commit 5e4d98470b6412a686883c554e7eb7badbe78c4d 21265Author: Alex Deucher <alex@botch2.(none)> 21266Date: Thu Sep 20 23:22:48 2007 -0400 21267 21268 RADEON: round 3 on the PLLs. should fix the LVDS issues 21269 21270commit c5e2a2f09af807006c7ea493a8e90ff77abe207c 21271Author: Alex Deucher <alex@botch2.(none)> 21272Date: Wed Sep 19 19:58:28 2007 -0400 21273 21274 bump for RC release 21275 21276commit 46ff78b9f010ca24178d4363761be00eb3ecb632 21277Author: Alex Deucher <alex@botch2.(none)> 21278Date: Wed Sep 19 19:49:58 2007 -0400 21279 21280 RADEON: add default connector setup for single crtc chips 21281 21282commit 509ca0cb0e1cde905b47db2bbac6f2a58523b279 21283Author: Alex Deucher <alex@botch2.(none)> 21284Date: Wed Sep 19 19:41:17 2007 -0400 21285 21286 RADEON: fix external TMDS parsing in legacy bios connector table 21287 21288commit 5a59547ef7e986c9613023d941c3354cd476faf1 21289Author: Alex Deucher <alex@botch2.(none)> 21290Date: Wed Sep 19 19:28:09 2007 -0400 21291 21292 Revert "RADEON: initial pass at external TMDS support" 21293 21294 This reverts commit 4000a710c93dd2d82891e4082bc7fa922ba9c5f4. 21295 21296 This needs to be reworked and needs more soak time. so revert 21297 for now. 21298 21299commit dc333884c540d536bffe51a5ebfaf1822af6a91c 21300Author: Alex Deucher <alex@botch2.(none)> 21301Date: Tue Sep 18 22:23:12 2007 -0400 21302 21303 RADEON: fix plls again 21304 21305 previous fix seemed to break other chips. Lets see how 21306 this goes. 21307 21308commit 271e541088f455d1cfedff87e88c9a4fdbdbe424 21309Author: Alex Deucher <alex@botch2.(none)> 21310Date: Tue Sep 18 22:12:31 2007 -0400 21311 21312 RADEON: Remove more mergedfb cruft 21313 21314commit 8ae69c496eba701e744cca0605f73242673f7b3f 21315Author: Alex Deucher <alex@botch2.(none)> 21316Date: Tue Sep 18 20:12:21 2007 -0400 21317 21318 RADEON: adjust pll restore 21319 21320 some chips seem to be pickier than others. fixes bug 12467 21321 21322commit 8a84dc8ee260716026fa661a6dced2000621efa2 21323Author: Michel Dänzer <michel@tungstengraphics.com> 21324Date: Tue Sep 18 11:53:39 2007 +0200 21325 21326 radeon: Remove declaration of unused variable. 21327 21328commit 2fbbd9c5f0ec27e72780e8b054bc59502f1e5651 21329Author: Alex Deucher <alex@botch2.(none)> 21330Date: Tue Sep 18 00:45:56 2007 -0400 21331 21332 RADEON: note that MergedFB support has been replaced with randr 1.2 21333 21334commit 81b708dbd35c29bf2d67e8c655af97b60e61498b 21335Author: Alex Deucher <alex@botch2.(none)> 21336Date: Mon Sep 17 23:54:03 2007 -0400 21337 21338 RADEON: turn off load detection on tv dac by default 21339 21340 it's just too unreliable and even when it is most people 21341 don't want it. enable it via output attribute or force 21342 the outputs on with xrandr 21343 21344commit c3ab53159d42157985a25e9125eaa5362273b73d 21345Author: Alex Deucher <alex@botch2.(none)> 21346Date: Mon Sep 17 23:28:51 2007 -0400 21347 21348 RADEON: XPRESS connector tables are crap 21349 21350 besides specifying the pimary DAC (which they don't have), 21351 most seem to specify DDC_CRT2 for DDC, which does not seem 21352 to ever work. Some have reported success with DDC_MONID so 21353 lets give that a try. 21354 21355commit 1fafa970fa78ddea29a8804b6cd0aea52e4fa2e1 21356Author: Alex Deucher <alex@botch2.(none)> 21357Date: Mon Sep 17 23:10:13 2007 -0400 21358 21359 RADEON: ignore edid digital bit on analog ports 21360 21361 Some monitors have this bit set even on analog ports 21362 21363commit 4000a710c93dd2d82891e4082bc7fa922ba9c5f4 21364Author: Alex Deucher <alex@botch2.(none)> 21365Date: Mon Sep 17 23:04:17 2007 -0400 21366 21367 RADEON: initial pass at external TMDS support 21368 21369 - Based on Dave Airlie's initial work and the dvo support 21370 in the intel driver. 21371 - Only sil164 is supported at the moment. 21372 - Once we get some testing and such, we ought to move the dvo 21373 drivers out of the drivers so they can be shared among all drivers 21374 - Doesn't seem to work on my card (r4xx ATOM) card ATM 21375 - Legacy bios table programming sequence not implemented yet 21376 21377commit 38515d402555eaa61c686d42973e59f659b07466 21378Author: Alex Deucher <alex@botch2.(none)> 21379Date: Sat Sep 15 11:39:47 2007 -0400 21380 21381 RADEON: don't restore bios scratch regs 21382 21383 Leave them as programmed by the bios 21384 fixes bug 12424 21385 21386commit 78c6bd305f2ea70629d1dda3c0c48c067451f5fb 21387Author: George Sapountzis <gsap7@yahoo.gr> 21388Date: Mon Sep 10 19:00:16 2007 +0300 21389 21390 [mach64] Revert to multiple mappings per BAR. 21391 21392 libpciaccess added subrange mappings and mtrr handling. 21393 21394 compile-tested only. 21395 21396commit 0241cac643fa1c08a45ea44f5c670b290e760ad8 21397Author: Michel Dänzer <michel@tungstengraphics.com> 21398Date: Sat Sep 8 00:22:40 2007 +0200 21399 21400 radeon: Reinstate sync in radeon_crtc_show/hide_cursor. 21401 21402 Make sure the DRI lock is held though, as these can be called asynchronously. 21403 21404 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=12245 . 21405 21406commit 49933e2f7d590811f2bc8c0d51a09f3b7f14845e 21407Author: Alex Deucher <alex@botch2.(none)> 21408Date: Sat Sep 1 12:23:07 2007 -0400 21409 21410 RADEON: automatically disable tiling if requested virtual desktop exceeds surface limits 21411 21412 Also, tweak default desktop sizes and add some informational messages 21413 21414commit d49a60bc939c9f3e9d57c23e44263f5bb52f09fb 21415Author: Sjoerd Simons <sjoerd@luon.net> 21416Date: Thu Aug 30 11:40:13 2007 +0200 21417 21418 radeon: Fix Option "MacModel". 21419 21420commit 2b860726fff058623e0e8e1b0aca092c246875ce 21421Author: Michel Dänzer <michel@tungstengraphics.com> 21422Date: Thu Aug 30 11:38:28 2007 +0200 21423 21424 radeon: Don't build currently unused RADEONSavePalette. 21425 21426commit 12187a6aa93049c002a4171344d03c713f7f3c5d 21427Author: Alex Deucher <alex@botch2.(none)> 21428Date: Wed Aug 29 23:11:30 2007 -0400 21429 21430 RADEON: Add quirk and connector tables for apple laptops 21431 21432 As far as I can tell there are three apple laptop variants: 21433 ibook - LVDS, TVDAC drives TV or VGA via dongle 21434 powerbook-duallink - LVDS, TV, External TMDS/Primary DAC 21435 powerbook - LVDS, TV, Internal TMDS/Primary DAC 21436 use Option "MacModel" "<string>" 21437 to enable the appropriate quirks where string is one of the above 21438 21439 We can't yet init the external TMDS directly, but if OF inits it, 21440 it should work. This should also fix bug 9955. 21441 21442 Please test! 21443 21444commit 61c1fdaa8553581944f78a11e6f9aa76163a468a 21445Author: Alex Deucher <alex@botch2.(none)> 21446Date: Tue Aug 28 23:47:19 2007 -0400 21447 21448 RADEON: add option to force tmds pll to default table 21449 21450 Also rework the tmds pll output attribute handling a bit 21451 21452commit 0d9087bc60bb95c770b899cfed29699c02bdac49 21453Author: Alex Deucher <alex@botch2.(none)> 21454Date: Tue Aug 28 23:08:20 2007 -0400 21455 21456 RADEON: remove some cruft forgotten in a previous commit 21457 21458commit 6ff0645ecfe65727e8ef5d5e6215b4e03078e1a5 21459Author: George Sapountzis <gsap7@yahoo.gr> 21460Date: Tue Aug 28 15:37:11 2007 +0300 21461 21462 [mach64] Convert to pci-rework, keeping source-code compatibility. 21463 21464 It still uses the old probe method though, this is due to the ati wrapper. 21465 21466commit 7b38d9a1209f87255e5bb0aefe46a363ce4fb6ef 21467Merge: 2e3d43a 673f799 21468Author: George Sapountzis <gsap7@yahoo.gr> 21469Date: Tue Aug 28 16:01:12 2007 +0300 21470 21471 Merge branch 'master' into pci-rework 21472 21473 Conflicts: 21474 21475 src/ati.c 21476 21477commit 673f799729824f4439dd5f681f75dd5aab50947f 21478Author: Alex Deucher <alex@botch2.(none)> 21479Date: Tue Aug 28 00:42:30 2007 -0400 21480 21481 RADEON: Update tv attributes immediately 21482 21483commit ad6f7ad1b2ccae0bc0a416b9b0ca22709c9d5199 21484Author: Alex Deucher <alex@botch2.(none)> 21485Date: Tue Aug 28 00:08:41 2007 -0400 21486 21487 RADEON: remove the "default" tv_standard option 21488 21489commit 17e0f9e6cbfdb115034d327bd34d46339fd632b7 21490Author: Alex Deucher <alex@botch2.(none)> 21491Date: Mon Aug 27 23:59:03 2007 -0400 21492 21493 RADEON: enable load detection for tvdac if output count for tvdac < 2 21494 21495commit 42839fb5a8584196e7b18375bff6c426ed0347d9 21496Author: Alex Deucher <alex@botch2.(none)> 21497Date: Mon Aug 27 23:44:13 2007 -0400 21498 21499 RADEON: make load detection an output attribute for analog outputs 21500 21501 Since TV/VGA/DVI-I can share the TV DAC, we often get false detection 21502 of all inputs that share that DAC. Make load detection an output 21503 attribute. Enabled by default on primary dac and on cards where 21504 tv dac is (usually) dedicated to tv (non-IGP mobilities). 21505 21506commit a5a1055d64ab4fa16bfb03a412ae6c4fe69ff65d 21507Author: Alex Deucher <alex@botch2.(none)> 21508Date: Mon Aug 27 22:42:22 2007 -0400 21509 21510 RADEON: make tmds pll an output attribute 21511 21512 sometimes the bios tmds plls are busted for certain 21513 monitors. sometimes the dirver tables are. Let the user 21514 pick at run time. 21515 21516commit a12e4aa01bf1c5723c3c791ff9bdc26eef21d5ea 21517Author: Alex Deucher <alex@botch2.(none)> 21518Date: Sun Aug 26 18:51:29 2007 -0400 21519 21520 Bump for new release 21521 21522commit d43596e5f5d7c60f96b57bc3e743a9b40eb7109d 21523Author: Alex Deucher <alex@botch2.(none)> 21524Date: Sun Aug 26 18:07:50 2007 -0400 21525 21526 RADEON: Fix rotation. works now 21527 21528commit 47eb3327c258bb0cfd9a1d5677624b9988a39057 21529Author: Alex Deucher <alex@botch2.(none)> 21530Date: Sun Aug 26 15:43:22 2007 -0400 21531 21532 RADEON: minor tweak to tv out 21533 21534commit f36720377737210c985b196d9a988efdd767f1c7 21535Author: Alex Deucher <alex@samba.(none)> 21536Date: Sun Aug 26 14:13:06 2007 -0400 21537 21538 RADEON: fix depth 16 palette for real this time 21539 21540commit f2b13f1457bf860b075310d3962254be0ed7bea3 21541Author: Alex Deucher <alex@botch2.(none)> 21542Date: Sun Aug 26 13:27:19 2007 -0400 21543 21544 RADEON: Only update pixclks_cntl when updating tv routing 21545 21546 No need to re-set all of the pll2 stuff 21547 21548commit 5d044b9f74c7aa7e12f2822896fed881e2ca9d19 21549Author: Alex Deucher <alex@botch2.(none)> 21550Date: Sat Aug 25 21:03:08 2007 -0400 21551 21552 RADEON: fix crtc clipping for Xv 21553 21554commit 3fd2d22a02812d5f86cdc1c9503f48362b0c362b 21555Author: Alex Deucher <alex@botch2.(none)> 21556Date: Sat Aug 25 17:37:35 2007 -0400 21557 21558 RADEON: remove fbdev option 21559 21560 FBDev support is currently broken, and it not really compatible 21561 with randr 21562 21563commit 3469e1aa08792890fa6a5c72da52a1992a0b382c 21564Author: Alex Deucher <alex@botch2.(none)> 21565Date: Fri Aug 24 20:42:13 2007 -0400 21566 21567 RADEON: add extra green data in depth 16 21568 21569 Apparently some radeons need this? 21570 21571commit 71f650d1bc432514516f7ac64a5e8a54c5227881 21572Author: Michel Dänzer <michel@tungstengraphics.com> 21573Date: Fri Aug 24 09:21:39 2007 +0200 21574 21575 Require xorg-server >= 1.3 for RandR 1.2. 21576 21577commit d7230939f523610c57f92bdfc72966bdbc6f1070 21578Author: Michel Dänzer <michel@tungstengraphics.com> 21579Date: Fri Aug 24 09:21:14 2007 +0200 21580 21581 64 bit warning fixes. 21582 21583 For printf vs. CARD32, use %u or %x and and a cast to unsigned. 21584 21585commit 91c45fedfd155a153dcd2c3f3e30986bfbd44e6f 21586Author: Dave Airlie <airlied@nx6125b.(none)> 21587Date: Fri Aug 24 15:05:01 2007 +1000 21588 21589 radeon: don't disable dac if either tv or vga is using it 21590 21591 On my rs480 I had to vt switch to get hotplug VGA working due to the tv-out 21592 code turning off the dac when the vga code was actually using it. 21593 21594commit 056ca6bb5adf974290693b55de6cd6880d2132d1 21595Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21596Date: Thu Aug 23 20:19:47 2007 -0400 21597 21598 RADEON: NONE to None to match other port info 21599 21600commit 13fd53286bdda2c55683bdb5f63e7d345f6c63ef 21601Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21602Date: Thu Aug 23 20:17:51 2007 -0400 21603 21604 RADEON: set (hopefully) reasonable default max desktop sizes 21605 21606 Based on the amount of vram. We really need ttm... 21607 21608commit 53bad86ca48a9b6529c1f0989ee568d9d48841c6 21609Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21610Date: Thu Aug 23 19:24:59 2007 -0400 21611 21612 RADEON: clean up the logic in crtc_mode_set() 21613 21614commit 4712dedea225e9e07177aebda2ffc6290d1f53c7 21615Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21616Date: Thu Aug 23 18:51:34 2007 -0400 21617 21618 RADEON: Always assume LVDS is connected 21619 21620 Not all bioses seem to set the right scratch bits. If we have 21621 LVDS (via bios table or otherwise) assume it's connected. 21622 21623commit b0f170c5f736ecba1a5899d602c4173fe9b9b1fa 21624Author: Michel Dänzer <michel@tungstengraphics.com> 21625Date: Thu Aug 23 12:55:40 2007 +0200 21626 21627 radeon: Remove unnecessary #include <time.h>. 21628 21629commit ac54c0e4360099697755d14b1030def73d8235b0 21630Author: Michel Dänzer <michel@tungstengraphics.com> 21631Date: Wed Aug 22 14:33:59 2007 +0200 21632 21633 radeon: Warning fixes. 21634 21635commit de26e406f52b3b13f03eee2b8023924ec6406f0a 21636Author: Alon Ziv <alonz@nolaviz.org> 21637Date: Mon Jul 30 22:47:59 2007 +0300 21638 21639 radeon: Sane handling of timeouts in WaitForVerticalSync(2). 21640 21641 RADEONWaitForVerticalSync() and RADEONWaitForVerticalSync2() need to wait 21642 for a timeout specified in milliseconds; looping around usleep() causes 21643 the timeout to be unnecessarily long, as the OS may sleep longer than 21644 requested (on Linux the minimum actual sleep value may be several ms). 21645 21646 The new logic uses gettimeofday() in the loop to see when the (absolute) 21647 timeout has arrived. 21648 21649 Signed-off-by: Alon Ziv <alonz@nolaviz.org> 21650 21651commit c66e5de26ae93caa368213f3cce139aacec955d2 21652Author: Sascha Sommer <saschasommer@freenet.de> 21653Date: Thu Aug 23 12:11:51 2007 +0200 21654 21655 radeon: Round down RMX stretch ratios. 21656 21657 Fixes issues with RMX scaling, see 21658 https://bugs.freedesktop.org/show_bug.cgi?id=8983 . 21659 21660commit 633c1fff10a3be4c9f48c1995e330d60bf6abbb2 21661Author: Michel Dänzer <michel@tungstengraphics.com> 21662Date: Thu Aug 23 12:11:41 2007 +0200 21663 21664 radeon: Sync pages when enabling page flipping with EXA as well. 21665 21666 Exclude the DRI window(s) though to avoid scribbling over 3D rendering. 21667 21668commit 4f8010ce22043c0f8d60c0f49d270ce98c9d2466 21669Author: Michel Dänzer <michel@tungstengraphics.com> 21670Date: Thu Aug 23 12:11:12 2007 +0200 21671 21672 radeon: Don't synchronize DRI windows between pages when possible. 21673 21674commit 7b527054a7c81d1d1dbc79d41b9e53064dab68cb 21675Author: Michel Dänzer <michel@tungstengraphics.com> 21676Date: Thu Aug 23 12:10:33 2007 +0200 21677 21678 radeon: Restore memmap registers even if only AGP location changed. 21679 21680commit d7ba9f001c0ab645984526afd0e64d1c6a6d654a 21681Author: Michel Dänzer <michel@tungstengraphics.com> 21682Date: Thu Aug 23 11:39:54 2007 +0200 21683 21684 radeon: Change a test to info->IsIGP from several IGP families. 21685 21686commit 8c7c22e22b6076abc80e4e1aaa8d1f4cf2f3ed14 21687Author: Michel Dänzer <michel@tungstengraphics.com> 21688Date: Thu Aug 23 11:38:17 2007 +0200 21689 21690 radeon: Wait for pending overlay flip to finish before emitting new one. 21691 21692commit 92fa7cc00688d7bfc1fb72e645ac30c6d92669c6 21693Author: Michel Dänzer <michel@tungstengraphics.com> 21694Date: Thu Aug 23 11:38:16 2007 +0200 21695 21696 radeon: Don't call RADEONDRIRefreshArea when the damaged region is empty. 21697 21698commit 5cb20c2dc5eca9d7d7d78e9924ea1b90076e7253 21699Author: Dave Airlie <airlied@linux.ie> 21700Date: Thu Aug 23 18:46:41 2007 +1000 21701 21702 radeon: bug 11899 + debian 435040 - test for usefbdev before calling int10 21703 (cherry picked from commit 3334c247540b113f9c4a5ce1879d8648a45a959a) 21704 21705commit 9d6261c024c697a211da4b6300420153eb72a264 21706Author: Roland Bär <roland@verifysoft.de> 21707Date: Thu Aug 23 18:33:34 2007 +1000 21708 21709 radeon: bug 11860 fix some mem leaks 21710 21711commit 7bc1f862bc5f992f213143fbafef52459ba7db4a 21712Author: Roland Bär <roland@verifysoft.de> 21713Date: Thu Aug 23 18:37:35 2007 +1000 21714 21715 radeon: bug 11861 - dead code removal in radeon_video.c 21716 (cherry picked from commit 8e3a6f83016cd8c4cfd43ceee4cbf0a8dc018b2a) 21717 21718commit 9d38c8aa1a7d6fb1af41ee8abdb4a95f94843538 21719Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 21720Date: Thu Aug 23 20:10:24 2007 +1000 21721 21722 radeon: cleanup some warnings 21723 21724commit e4c8969b48a6c8dcc4e7f9852479d24a0204fc0d 21725Author: Dave Airlie <airlied@redhat.com> 21726Date: Thu Aug 23 19:56:21 2007 +1000 21727 21728 updated release numbering for randr 1.2 21729 21730commit 81a8093f2ffdce59d4e8a44f65f3d5c771fd6425 21731Merge: c08e6ec 53a67e3 21732Author: Dave Airlie <airlied@redhat.com> 21733Date: Thu Aug 23 19:51:47 2007 +1000 21734 21735 Merge remote branch 'origin/randr-1.2' 21736 21737commit c08e6ec9e7ac06caee53689b0ec50ef6a7a0ba37 21738Merge: 5793e87 a0d0fcd 21739Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 21740Date: Thu Aug 23 19:42:19 2007 +1000 21741 21742 Merge branch 'master' into randr-merge 21743 21744 Conflicts: 21745 21746 src/radeon.h 21747 src/radeon_bios.c 21748 src/radeon_display.c 21749 src/radeon_dri.c 21750 src/radeon_driver.c 21751 src/radeon_modes.c 21752 src/radeon_probe.h 21753 src/radeon_video.c 21754 21755commit 53a67e31904bec9a3aa1bd24de8034dcafea1d2a 21756Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21757Date: Tue Aug 21 21:20:41 2007 -0400 21758 21759 RADEON: Fix color problem on pre-R3xx chips tv-out 21760 21761commit e9719e8e02eef46717ae9b4d8c7998466dac30cb 21762Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21763Date: Tue Aug 21 21:17:20 2007 -0400 21764 21765 RADEON: more tv out fixes and clean up 21766 21767commit 36c22a49580d86a6518b67f31a78bd53d39491af 21768Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21769Date: Tue Aug 21 20:28:39 2007 -0400 21770 21771 RADEON: fix tv-out on R3xx 21772 21773 R3xx apparently needs the tv clock forced on. 21774 21775commit 9470bd67731059f26859ed5f0bea3ade09e2c80c 21776Author: Alex Deucher <alex@botch2.com> 21777Date: Mon Aug 20 20:54:06 2007 -0400 21778 21779 RADEON: Add DefaultConnectorTable option 21780 21781 This option skips the parsing the BIOS connector table 21782 and falls back to chip specific defaults. 21783 Also remove man page section for the now gone bioshotkeys 21784 option. 21785 21786commit a0d0fcd3bab765b4db25e04884fd8a342abb9c66 21787Author: iLisa Wu <liswu@ati.com> 21788Date: Mon Aug 20 15:44:07 2007 +0200 21789 21790 Fix crash with no valid mode in xorg.conf's modelist and empty Virtual 21791 21792 If the resolution defined in xorg.conf failed to find a matching mode in the 21793 supported modelist, and no virtual desktop dimensions are defined the xorg.conf 21794 either, virtual X and Y dimension will be set to 0 which will cause Xserver 21795 crash. 21796 21797 (Novell bugzilla #296856, closed) 21798 21799commit b275febdb0918e8cebdffbb433b0eeb3ff4d3746 21800Author: Alex Deucher <alex@samba.(none)> 21801Date: Sun Aug 19 20:55:32 2007 -0400 21802 21803 RADEON: turn off TVCLK when blanking tv encoder 21804 21805commit a90d675832ddb02c81ace010ccbf02619b70edac 21806Author: Alex Deucher <alex@botch2.com> 21807Date: Thu Aug 16 21:55:14 2007 -0400 21808 21809 RADEON: fix Xv clipping and overlay sourcing 21810 21811 - Basically just copied from the intel driver. I'm planning to push 21812 this to the server soon, but add it now to get things working 21813 and to provide compat for older servers. 21814 21815 - Overlay crtc source control attribute now called XV_CRTC 21816 The old attribute XV_SWITCHCRT has been removed. If anyone cares, 21817 we can add it back as an alias to XV_CRTC 21818 XV_CRTC: -1 auto, 0 crtc0, 1 crtc1 21819 21820commit 5793e8753d11432bf95c7c6dd80c811e16aba058 21821Author: Alex Deucher <alex@botch2.com> 21822Date: Wed Aug 15 19:26:36 2007 -0400 21823 21824 RADEON: Remove RADEONRestoreMode() 21825 21826 Since we no longer use it to write modes all it did 21827 was restore some of the regs for a console restore. 21828 Just move the relevant bits into RADEONRestore() and 21829 remove it. 21830 21831commit 6f011aaabaf18d66ffc255ad76aaf938b2396302 21832Author: Lisa Wu <Lisa.Wu@amd.com> 21833Date: Wed Aug 15 19:17:51 2007 -0400 21834 21835 RADEON: fix console restore on r3xx and r4xx 21836 21837 When restoring the console we need to: 21838 1) we need to restore DAC registers after all other registers are 21839 restored and CRTCs are enabled. 21840 2) we need to enable CRTC2 registers before CRTC1 registers 21841 21842commit c8dad98abb042c6abbbee18f9ae4db72084bc513 21843Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21844Date: Wed Aug 15 00:55:24 2007 -0400 21845 21846 RADEON: make sure crtc routing is correct in r300_detect_tv() 21847 21848 Also remove some unused variable 21849 21850commit f8cd74435f0072dbf5f6e83d67d2d5e1f4e82c91 21851Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21852Date: Tue Aug 14 23:50:57 2007 -0400 21853 21854 RADEON: minor cleanup 21855 21856commit 6f398cd07ea734dd66a8eac71b629e59123d75b8 21857Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21858Date: Tue Aug 14 23:42:32 2007 -0400 21859 21860 RADEON: Implement improved tv load detection for r300 21861 21862 The previous implementation resulted in false positives 21863 on occasion. This method works much more reliably. 21864 Based on beos code by Thomas Kurschel 21865 21866commit 366a1d4c240ac93622caff97b652696db99bf2e6 21867Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21868Date: Tue Aug 14 23:25:37 2007 -0400 21869 21870 RADEON: Turn off tv encoder in disableoutputs() 21871 21872commit 5b4a04c23e3f1ec2490418b111f417c16463c709 21873Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21874Date: Tue Aug 14 23:11:52 2007 -0400 21875 21876 RADEON: Make sure RMX is always programmed for crtc1 21877 21878 If the RMX registers are left as programmed by the bios 21879 this can lead to a blank screen when crtc1 is feeding a 21880 DAC. Fix found and reported by Lisa Wu <liswu@ati.com> 21881 Fixes bug 11985 21882 21883commit db2a828b2f21b92cd654b309d137204334975b89 21884Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21885Date: Tue Aug 14 22:47:29 2007 -0400 21886 21887 RADEON: Add load detection 21888 21889 Based on the beos driver by Thomas Kurschel and the 21890 existing load detection code in this driver. 21891 21892commit 6fc3ddbbb6fbbee1f6076c776e5b46c0c772b6d4 21893Author: Alex Deucher <alex@botch2.com> 21894Date: Sun Aug 12 12:00:20 2007 -0400 21895 21896 RADEON: order the VGA and DVI ports correctly 21897 21898commit b7738d2ef82e1759adf78e5db1291f8739b4166f 21899Author: Alex Deucher <alex@botch2.com> 21900Date: Sat Aug 11 17:50:42 2007 -0400 21901 21902 RADEON: fix connector setup when there's no bios tables 21903 21904commit d0895f67e327bb268fd59fcfd8fc22678d804f57 21905Author: Luc Verhaegen <libv@skynet.be> 21906Date: Thu Aug 9 12:11:40 2007 +0200 21907 21908 AGPFastWrite risk reduction. 21909 21910 Actively warn the user in the log about the effects of AGPFastWrite and 21911 sanitise AGPFastWrite handling while we're here. 21912 21913commit c01000bc684f6a23a38a52f0808182cefaa686bb 21914Author: Alex Deucher <alex@botch2.com> 21915Date: Wed Aug 8 00:07:16 2007 -0400 21916 21917 RADEON: fix typo in comment 21918 21919commit 5c3f49e651c36f3bd14fa29359e24825d8f7f77f 21920Author: Lisa Wu <liswu@ati.com> 21921Date: Wed Aug 8 00:05:47 2007 -0400 21922 21923 RADEON: make sure RADEON_CRTC2_CRT2_ON bit state is properly accounted for 21924 21925 - when the TV DAC is used for crtc1 make sure to keep this bit set when 21926 initializing crtc2. 21927 - fixes bug 11894 21928 21929commit 08fe7ad00fba523775e95b5e6295fe23a5119d60 21930Author: Alex Deucher <alex@botch2.com> 21931Date: Tue Aug 7 23:16:05 2007 -0400 21932 21933 RADEON: fix ConnectorTable option after connector table re-work 21934 21935 We may want to add the possiblity of more connectors to this option 21936 21937commit f56b90a60393ed187c0e39e149b2a3a32331ea52 21938Author: Alex Deucher <alex@botch2.com> 21939Date: Tue Aug 7 23:08:02 2007 -0400 21940 21941 RADEON: Make sure the default TV standard is supported 21942 21943commit a88a0e77eec514325b4d07bf7be9fb04e6f8e244 21944Author: Alex Deucher <alex@botch2.com> 21945Date: Tue Aug 7 00:43:20 2007 -0400 21946 21947 RADEON: minor fix to legacy bios connector table 21948 21949commit 3752808d6c08a9727370ef8d79088e787791e131 21950Author: Alex Deucher <alex@botch2.com> 21951Date: Sun Aug 5 16:13:39 2007 -0400 21952 21953 RADEON: Always set MonType in radeon_detect() 21954 21955 - Always set the the MonType in radeon detect even if no 21956 monitor is detected so users can still force disconnected 21957 outputs on. For DVI-I users will have to set the 21958 dvi_monitor_type attribute to force digital or analog, 21959 everything else should just work. 21960 21961commit b4ec3e436afb5bc99ec755cbd96eee9a1ec492a5 21962Author: Alex Deucher <alex@botch2.com> 21963Date: Sun Aug 5 15:37:04 2007 -0400 21964 21965 RADEON: add tv out properties 21966 21967commit 1fc2a1120e7c05938e2bd72d3c7837ecff8bc9da 21968Author: Alex Deucher <alex@botch2.com> 21969Date: Sun Aug 5 14:41:57 2007 -0400 21970 21971 RADEON: switch DVI-I monitor type attribute to string 21972 21973commit 7d8eb3751d74bd8a1fb9fa2d2fcb9c4c895f6ba4 21974Author: Alex Deucher <alex@botch2.com> 21975Date: Sun Aug 5 14:31:18 2007 -0400 21976 21977 RADEON: make backlight attribute lower case 21978 21979commit e71bb88bc9052af8866fb6945dbc06dbb0e6d1c0 21980Author: Alex Deucher <alex@botch2.com> 21981Date: Sun Aug 5 14:26:15 2007 -0400 21982 21983 RADEON: switch RMX attribute to string type 21984 21985 Still not actually hooked up. 21986 21987commit 284323135ec04635dfa9cabd5790a35b953abca2 21988Author: Alex Deucher <alex@botch2.com> 21989Date: Sun Aug 5 04:35:24 2007 -0400 21990 21991 RADEON: typo 21992 21993commit d1abdad167aa24ac970c69422435df443c82ebd6 21994Author: Alex Deucher <alex@botch2.com> 21995Date: Sun Aug 5 03:45:02 2007 -0400 21996 21997 RADEON: fixes 21998 21999 - fix output ordering 22000 - set tv-out to return un-connected for now in radeon_detect() 22001 22002commit 0cca25d8d6a0cb0d29b68e6cd9c699d1390aede1 22003Author: Alex Deucher <alex@botch2.com> 22004Date: Sun Aug 5 03:19:24 2007 -0400 22005 22006 RADEON: Fix DVI-I support in ATOM bios connector table parsing 22007 22008commit ba5496ae7973786802962bf649dd91c219531749 22009Author: Alex Deucher <alex@botch2.com> 22010Date: Sun Aug 5 02:27:32 2007 -0400 22011 22012 RADEON: refactor output init to handle multiple DVI or VGA 22013 22014 - refactor output init to handle multiple DVI or VGA with the new 22015 bios table parsing 22016 22017commit 2ec22783ddf4c522df9e5fd1b2003854486d7a2b 22018Author: Alex Deucher <alex@botch2.com> 22019Date: Sun Aug 5 01:39:35 2007 -0400 22020 22021 RADEON: attempt to do the right thing for standards other than PAL or NTSC 22022 22023commit cc8e1d95f1b90a259beea4e8cc4d7e29af660919 22024Author: Alex Deucher <alex@botch2.com> 22025Date: Sun Aug 5 01:14:36 2007 -0400 22026 22027 RADEON: Major rework of BIOS table parsing 22028 22029 - greatly simplify ATOM and legacy connector table parsing 22030 - use bios tables to detect LVDS and TV outputs 22031 - add support for TV table parsing (legacy only) 22032 22033commit 1de52d91ff3a04b9b587b858e1e5be40d3a7fd0a 22034Author: Dave Airlie <airlied@linux.ie> 22035Date: Sat Aug 4 17:58:58 2007 +1000 22036 22037 update configure.ac for 6.6.193 release 22038 22039commit 371001c0433db1d17e468f3ea99ea57f922145a3 22040Author: Dave Airlie <airlied@linux.ie> 22041Date: Sat Aug 4 17:51:19 2007 +1000 22042 22043 radeon: remove unused variables 22044 22045commit 90946c53d7a4b23d03270ad6da0450759a11de3d 22046Author: Dave Airlie <airlied@linux.ie> 22047Date: Sat Aug 4 17:51:06 2007 +1000 22048 22049 ati: add -Wall for gcc 22050 22051 As per luc's "suggestion" in Novell bug. 22052 22053 This may generate warnings on 64-bit until we get rid of CARD32 22054 22055commit e30a145934df8f6a7f71290d6c75e4239f9d52f7 22056Author: Dave Airlie <airlied@linux.ie> 22057Date: Sat Aug 4 17:44:46 2007 +1000 22058 22059 radeon: cleanup some pieces of the dpms/blank register programming 22060 22061commit a156db5e8b037ed12a448f70045453baf9d0c504 22062Author: Luc Verhaegen <libv@skynet.be> 22063Date: Sat Aug 4 17:37:18 2007 +1000 22064 22065 Clean up PortInfo to CRTC mapping 22066 22067 Also sanitise blanking and DPMS functions 22068 22069 Fixes from Novell Bug 264720, and fd.o 10772 22070 22071commit b61a49f2a5401560f85e11bcdd005287433cad12 22072Author: Alex Deucher <alex@botch2.com> 22073Date: Fri Aug 3 19:27:59 2007 -0400 22074 22075 RADEON: Change indexing of TV constants table in preparation for standard re-work 22076 22077commit 288fa627274cb399059262d4f8bd844fc220a042 22078Author: Alex Deucher <alex@botch2.com> 22079Date: Thu Aug 2 02:37:16 2007 -0400 22080 22081 RADEON: avoid a divide by 0 and only save tv out regs if the chip has them 22082 22083commit b66a1bc7994b33d349c1519761e431959311c85f 22084Author: Alex Deucher <alex@botch2.com> 22085Date: Thu Aug 2 02:11:20 2007 -0400 22086 22087 RADEON: fix tv-out enable/disable 22088 22089commit d86592c8d5ce45d81d8a726c263e870e94fbcf11 22090Author: Alex Deucher <alex@botch2.com> 22091Date: Thu Aug 2 00:50:51 2007 -0400 22092 22093 RADEON: add missing break 22094 22095commit 971feb34843225030fff05b3f9d3801534fbf2d4 22096Author: Alex Deucher <alex@botch2.com> 22097Date: Thu Aug 2 00:50:04 2007 -0400 22098 22099 RADEON: move tv dac enable to enabledisplay() 22100 22101commit 98d7e00437bea78e03180eb30ff30de3455d9d1a 22102Author: Alex Deucher <alex@botch2.com> 22103Date: Thu Aug 2 00:20:50 2007 -0400 22104 22105 RADEON: limit tv modes to the only one we can program at the moment 22106 22107commit 5c549c1d42f7bbc556942af13aff2661fae856f2 22108Author: Alex Deucher <alex@botch2.com> 22109Date: Wed Aug 1 23:45:07 2007 -0400 22110 22111 RADEON: remove unused elements 22112 22113commit b03978028fd975eb6946503d3a56a49c5a67f339 22114Author: Alex Deucher <alex@botch2.com> 22115Date: Wed Aug 1 22:39:16 2007 -0400 22116 22117 RADEON: convert hard coded tv out values to calculations 22118 22119commit cf54222f1fa37366b2c2b39c82f8afc02f32e63c 22120Author: Alex Deucher <alex@botch2.com> 22121Date: Tue Jul 31 02:01:49 2007 -0400 22122 22123 RADEON: more fixes... 22124 22125commit 22d460d3ad991223aa1fbd7e5edeb45e36c65dc0 22126Author: Alex Deucher <alex@botch2.com> 22127Date: Tue Jul 31 01:28:05 2007 -0400 22128 22129 RADEON: fix name of tv output 22130 22131commit 4822a2b837334f408f962646ab5ea4f8b0335ac9 22132Author: Alex Deucher <alex@botch2.com> 22133Date: Tue Jul 31 01:18:40 2007 -0400 22134 22135 RADEON: tv-out fixes. works now. tested on rv350. 22136 22137 VT siwtch is busted, and xrandr doesn't play nice yet. 22138 uncomment code in radeon_output.c to test. 22139 22140commit fe494c9db2995bb8ce7a028ecf9626e0cb0cf506 22141Author: Alex Deucher <alex@botch2.com> 22142Date: Sun Jul 29 15:26:34 2007 -0400 22143 22144 RADEON: add info about tv out code and authorship 22145 22146commit 6b9b7a7bdc290d07de9b226691ec8025af8db896 22147Author: Alex Deucher <alex@botch2.com> 22148Date: Sun Jul 29 15:23:14 2007 -0400 22149 22150 RADEON: Initial pass at integrated tv out support 22151 22152 Based on the GATOS tv-out support by Federico Ulivi <fulivi@lycos.com> 22153 and information from ati with substantial rework by myself. 22154 22155 Code is not actually hooked up yet. 22156 22157commit 8d043db1817d94edeb72ab208dfea60026715d48 22158Author: Alex Deucher <alex@botch2.com> 22159Date: Wed Jul 25 20:37:58 2007 -0400 22160 22161 RADEON: Compute PLL VCO gain 22162 22163commit 62f06d89da3f7160d5e4df8d7ce6fe1a94e9d07c 22164Author: Alex Deucher <alex@botch2.com> 22165Date: Wed Jul 25 20:22:25 2007 -0400 22166 22167 RADEON: write out saved vclk and pixclk values 22168 22169commit 165a07cbbfcd94e3d1fac434b8fada8d29428a09 22170Author: Brice Goglin <Brice.Goglin@ens-lyon.org> 22171Date: Wed Jul 25 19:01:02 2007 +0200 22172 22173 Minor fixes in the manpages 22174 22175 - ati.man 22176 . add references to radeon and r128 manpages in SEE ALSO 22177 (Xavier Bestel in Debian bug #386001). 22178 . remove the manpage suffix from atimisc reference in the text 22179 since there is no such manpage for now. 22180 - radeon.man 22181 . typo in "specifying" (reported by A. Costa in Debian bug #432059). 22182 . replace a non-ascii character with the corresponding groff escape 22183 sequence so that it works whatever the locale (Julien Cristau). 22184 22185commit 9cc3ab8320162f371bba15dc131f23c5de2013fc 22186Author: Alex Deucher <alex@botch2.com> 22187Date: Wed Jul 25 02:24:14 2007 -0400 22188 22189 RADEON: hack around console restore hang on r4xx 22190 22191 r4xx seems to hang when unblanking the crtc(s) when 22192 restoring the console. This needs to be investigated 22193 further. 22194 22195commit 193d58b0ea9de3e326017e50b88e2f3726854506 22196Author: Alex Deucher <alex@botch2.com> 22197Date: Wed Jul 25 01:27:58 2007 -0400 22198 22199 RADEON: fix randr on r4xx 22200 22201 Seems the way we parse the ATOM connector table results 22202 in reversed connectors. 22203 22204commit 262e32bd9ea4123b116362b33b5798753b2568fe 22205Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22206Date: Thu Jul 19 23:48:57 2007 -0400 22207 22208 RADEON: fix palette when depth != 24 22209 22210commit 18b00b47a483e7854727e99126808ddf361e7a4a 22211Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22212Date: Thu Jul 19 19:01:42 2007 -0400 22213 22214 R128: don't clip modes to panel on laptops when only using the CRT port 22215 22216 fixes bug 5832 22217 22218commit 882fe7631586b0a7919f808588a2ea4fb555f7e8 22219Author: Brice Goglin <brice.goglin@ens-lyon.org> 22220Date: Thu Jul 19 18:44:27 2007 -0400 22221 22222 MACH64: avoid crash in Xv code 22223 22224 fixes bug 11054 22225 22226commit 7b3e22e96cfc98621bd20fa76317e6d8f7242165 22227Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22228Date: Thu Jul 5 20:37:10 2007 -0400 22229 22230 RADEON: set default values for crtc_offset_cntls 22231 22232commit 2a35ccfb8ffee9f0ef6f8d4f0eeb80a471543caa 22233Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22234Date: Thu Jul 5 20:04:04 2007 -0400 22235 22236 RADEON: Fix tiling on r1xx and r2xx 22237 22238 spotted by Andrew Randrianasulu <randrik@mail.ru> 22239 fixes bug 11357 22240 22241commit ac9cbaf32176cf144bc694ac879e7c9e2920f762 22242Author: Michel Dänzer <michel@tungstengraphics.com> 22243Date: Mon Jul 2 10:06:02 2007 +0200 22244 22245 radeon: Fully zero-initialize info->CRT2pScrn->monitor. 22246 22247 This prevents modes from getting rejected due to the uninitialized maxPixClock 22248 value. 22249 22250 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=9816 . 22251 22252commit bdcae622100c81a4d9a53938542b64908bacd195 22253Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22254Date: Fri Jun 29 01:15:57 2007 -0400 22255 22256 RADEON: save crtc on/off state for console restore 22257 22258 - prevents a possible hang if console is only using 1 crtc 22259 22260commit 21be0d3a1f4700572c6425800596785d6850626f 22261Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22262Date: Fri Jun 29 00:38:50 2007 -0400 22263 22264 RADEON: simplify console restore 22265 22266 - still need to track crtcs for blank/unblank when restoring text console 22267 22268commit d9bf28b55d98fea2d285c9c46362aaf2175d0a46 22269Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22270Date: Thu Jun 28 23:52:28 2007 -0400 22271 22272 RADEON: factor out surface_cntl init into one function 22273 22274commit 9f193985627be8e6ea1418a424e825ddbc4957b2 22275Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22276Date: Thu Jun 28 23:43:13 2007 -0400 22277 22278 RADEON: move crtc base setups to new functions 22279 22280commit 0f361e9e80a29d287fa42436c32c657e3c102539 22281Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22282Date: Thu Jun 28 23:08:07 2007 -0400 22283 22284 RADEON: fix corruption after 3D apps run 22285 22286commit 1d4630067d293d26284d5fe230debef3913ec6cf 22287Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22288Date: Thu Jun 28 22:24:04 2007 -0400 22289 22290 RADEON: set info->CurrentLayout.displayWidth in RADEONSetPitch() 22291 22292commit 2754d1ba01fc3367019487e0c0f59d74c950aaa7 22293Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22294Date: Thu Jun 28 22:15:22 2007 -0400 22295 22296 RADEON: Switch cursor back to memcpy() 22297 22298commit 7901bcafa92dccd319ddb5de4627d806a39f15f9 22299Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22300Date: Mon Jun 25 18:39:14 2007 -0400 22301 22302 RADEON: make sure we unblank in RADEONRestore() 22303 22304 reported by Andrew Randrianasulu (bugs 11357, 11146) 22305 22306commit 6a3599d6155d073a3116c4b13bdf04b44bb9e087 22307Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22308Date: Mon Jun 25 18:20:54 2007 -0400 22309 22310 RADEON: make sure we set the stride properly 22311 22312commit 771c37fcac8104b894120cc3fc0154b31b0318d1 22313Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22314Date: Mon Jun 25 17:53:26 2007 -0400 22315 22316 RADEON: attempt to fix cursor on big endian (only tested on x86) 22317 22318commit 5b5b90c2cea7e36895354f5872acd3fc769653f9 22319Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22320Date: Mon Jun 25 16:27:54 2007 -0400 22321 22322 RADEON: fix possible segfault on mobility chips with MM tables 22323 22324 spotted by Stefan Buehler 22325 22326commit 01bd5eb713500d3e3d4351865d460c8d1e476454 22327Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22328Date: Mon Jun 25 16:16:18 2007 -0400 22329 22330 RADEON: fix possible segfault on mobility chips with MM tables 22331 22332 reported by Stefan Buehler 22333 22334commit 52cc1dc1491559a9055f3ba6dd54064bb382ad86 22335Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22336Date: Mon Jun 25 15:49:19 2007 -0400 22337 22338 RADEON: tiling and FB size fixes 22339 22340 - fix tiling on older radeons 22341 - allow the user to specify FB size using Virtual line in config 22342 Current default is 1600x1200 22343 - reset ecp_div for Xv in randr crtc mode set 22344 22345commit 3bb46c03fddd3bd79bdab887366aeeced0290a3a 22346Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22347Date: Fri Jun 22 01:12:38 2007 -0400 22348 22349 RADEON: bump crtc max to 1200 so 1600x1200 will work out of the box 22350 22351 - a better fix will be forthcoming... 22352 22353commit b9fac5695bf8f123b391e3b162c4865258a685dd 22354Author: Andrew Randrianasulu <randrik@mail.ru> 22355Date: Fri Jun 22 01:01:14 2007 -0400 22356 22357 RADEON: Fix VT switch hangs 22358 22359 - re-order DRI resume functions in EnterVT() 22360 - fixes bug 11287 22361 22362commit 92e65d5e0d6817ff4c9a08020a0a9b3a8c3c98b0 22363Author: Andrew Randrianasulu <randrik@mail.ru> 22364Date: Fri Jun 22 00:55:00 2007 -0400 22365 22366 RADEON: Fix VT switch hangs 22367 22368 - reorder RADEONDRISetVBlankInterrupt() and RADEONDRIResume() 22369 - see bug 11287 22370 22371commit ec1fd2c388ec2c4033644776ee8588405a3d25f1 22372Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22373Date: Tue Jun 19 00:39:59 2007 -0400 22374 22375 RADEON: Split FP reg restore to only touch regs relevant to the output 22376 22377 - only restore output specific regs 22378 - restore fp2_gen_cntl on R200 DAC restore as R200 DAC2 is connected via 22379 DVO 22380 22381commit 0f5a92667281ca3f80218bc67d4031125a5e4793 22382Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22383Date: Mon Jun 18 23:35:21 2007 -0400 22384 22385 RADEON: only touch LVDS on mobility chips 22386 22387commit b72ff160f908bf3aa9f64705377e92d80360a4f7 22388Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22389Date: Mon Jun 18 21:51:45 2007 -0400 22390 22391 RADEON: only touch LVDS on mobility chips 22392 22393commit 80313621ffa8c409ae63fc1b28c15fc4abdc3a7b 22394Author: Henry Zhao <henryz@localhost.localdomain> 22395Date: Fri Jun 15 17:00:05 2007 -0700 22396 22397 Update CRT2pScrn->monitor->Last to reflect the 22398 last mode of CRT2pScrn. See bug 11278. 22399 22400commit 5868af53204e8fc210735e68e891e15085285629 22401Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22402Date: Tue Jun 12 23:53:48 2007 -0400 22403 22404 RADEON: yet another LVDS... 22405 22406commit f19a6f7ee5bf4ec632e7813359f167599c08e823 22407Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22408Date: Tue Jun 12 23:44:41 2007 -0400 22409 22410 RADEON: additional LVDS off fix (missed in previous commit) 22411 22412commit 4f2a1ba9c2fc20042d67132f986a86a9783a245e 22413Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22414Date: Tue Jun 12 23:41:24 2007 -0400 22415 22416 RADEON: additional LVDS off fix (missed in previous commit) 22417 22418commit 9fc416ce08cc5d74faa7b184e80ce43ed2f4fbee 22419Author: Michel Dänzer <michel@tungstengraphics.com> 22420Date: Tue Jun 12 23:35:09 2007 -0400 22421 22422 radeon: Fix panel size detection from registers with stretched mode programmed. 22423 22424commit c4045cbc5e6975520efef422ce9df2a7d99a7ff4 22425Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22426Date: Tue Jun 12 23:32:18 2007 -0400 22427 22428 RADEON: make sure to clear RADEON_LVDS_ON when turning LVDS off 22429 22430 - fixes bug 3483 22431 22432commit 090888d3baf61a70640d28e99afdce20952cdea0 22433Author: Tormod Volden <bugzi06.fdo.tormod@xoxy.net> 22434Date: Tue Jun 12 23:30:18 2007 -0400 22435 22436 RADEON: limit PanelPwrDly to 2000 ms on ATOM bios as per legacy bios 22437 22438 - fixes bug 11238 22439 22440commit 44748a826b52924a6e050215d8d62755e0ac7fb2 22441Author: Tormod Volden <bugzi06.fdo.tormod@xoxy.net> 22442Date: Tue Jun 12 23:23:49 2007 -0400 22443 22444 RADEON: limit PanelPwrDly to 2000 ms on ATOM bios as per legacy bios 22445 22446 - fixes bug 11238 22447 22448commit c292fc64499ff4cc135c07deda99cf4169f8fef4 22449Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22450Date: Tue Jun 12 23:20:18 2007 -0400 22451 22452 RADEON: make sure RADEON_LVDS_ON is cleared when turning LVDS off 22453 22454 - fixes bug 3483 22455 22456commit c6a3286d6f6ddda89115d98d45665dadd78bf41d 22457Author: Michel Dänzer <michel@tungstengraphics.com> 22458Date: Mon Jun 11 09:39:38 2007 +0200 22459 22460 radeon: Fix panel size detection from registers with stretched mode programmed. 22461 22462commit 7a6b3c0f77c69019268f585c51c3dcbcc99014d4 22463Author: Michel Dänzer <michel@tungstengraphics.com> 22464Date: Mon Jun 11 09:39:38 2007 +0200 22465 22466 radeon: Don't loop indefinitely if no mode matches detected panel size. 22467 22468commit f31fd9ce598841c505a0b5ed32bf124f49fea332 22469Author: Michel Dänzer <michel@tungstengraphics.com> 22470Date: Mon Jun 11 09:39:38 2007 +0200 22471 22472 radeon: Fix some more ErrorFs when setting mode. 22473 22474commit 9dd9f09a9021653e5590d27bd7aa0e7af85416ad 22475Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22476Date: Thu Jun 7 15:32:31 2007 -0400 22477 22478 RADEON: fix some issues with bios table init code 22479 22480 - fix some issues with bios table init code 22481 - re-org preinit to better handle bios table init of cards 22482 22483commit 9ad311eeb0ac2b70a862dbb9de278154ba9142ec 22484Author: Tilman Sauerbeck <tilman@code-monkey.de> 22485Date: Thu Jun 7 11:03:02 2007 +0200 22486 22487 Fixed 'make dist'. 22488 22489commit 2527f2b69aa7dffa3ba4359c45955e3185bdf0e6 22490Author: Alex Deucher <alex@botch2.com> 22491Date: Wed Jun 6 00:08:45 2007 -0400 22492 22493 RADEON: implement support for posting cards based on x86 bios tables (untested) 22494 22495 This is based on the netbsd radeonfb driver by Garrett D'Amore. 22496 The code is not hooked up yet, but should allow you to post cards 22497 without needing to execute bios code. This should be useful on 22498 non-x86 platforms, for posting secndary cards, and for suspend/resume. 22499 Works on legacy bioses only (no ATOM support). 22500 22501commit e67d1420bf65055ecb6fdfe6b1b1f53aae83854a 22502Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22503Date: Tue Jun 5 22:31:50 2007 -0400 22504 22505 RADEON: fixup RADEONDisplayVideo() to better handle the crtc being used 22506 22507commit 7886405308e3288d5c86b6f2c7dbfa8ff865139c 22508Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22509Date: Mon Jun 4 21:16:56 2007 -0400 22510 22511 RADEON: fix indenting 22512 22513commit f54ad565a69deb52547fd04e123f56fc1294cd0a 22514Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22515Date: Mon Jun 4 21:13:16 2007 -0400 22516 22517 RADEON: no need to go through all the crtcs, we've got what we need 22518 22519commit 687879bff716ad01f9f158860deb8ba770faab99 22520Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22521Date: Mon Jun 4 21:08:40 2007 -0400 22522 22523 RADEON: add support for RMX on DVI, make sure RMX is only enabled on crtc1 22524 22525commit e6161e472ff266f69547704a61040228a8704b06 22526Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22527Date: Mon Jun 4 20:32:55 2007 -0400 22528 22529 RADEON: remove some cruft 22530 22531commit be0ce38232ea4f5679c5829ab663939144c6e617 22532Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22533Date: Mon Jun 4 20:22:15 2007 -0400 22534 22535 RADEON: number the outputs if there are more than one of the same 22536 22537commit 8ec617f6493dd0aea5d11f92e3d58c3feef8c8fd 22538Author: Dave Airlie <airlied@linux.ie> 22539Date: Sun Jun 3 17:32:43 2007 +1000 22540 22541 radeon: disable irqs at server start until 3D app starts 22542 22543commit 2d40fa55e8d7a1cfb204d66ca4a4d95a3b13d5b5 22544Author: Dave Airlie <airlied@linux.ie> 22545Date: Sun Jun 3 17:32:43 2007 +1000 22546 22547 radeon: disable irqs at server start until 3D app starts 22548 22549commit dcb64a4d3947e5a9fbda4b72e29a5b6102370f07 22550Author: Dave Airlie <airlied@linux.ie> 22551Date: Sun Jun 3 17:10:49 2007 +1000 22552 22553 radeon: disable vbl interrupts when no 3d is running on a new enough drm 22554 22555commit 73d8e3ec8536b4777490b7ba457566f02233811f 22556Author: Dave Airlie <airlied@linux.ie> 22557Date: Sun Jun 3 17:07:59 2007 +1000 22558 22559 randr-1.2 increase dri minor version for mesa to know we can do vbl on both 22560 22561commit 63f0d4ed0a98830ecbe18c6e4174689111a59b68 22562Author: Dave Airlie <airlied@linux.ie> 22563Date: Sun Jun 3 16:49:14 2007 +1000 22564 22565 randr-1.2: add support for vblank on both heads 22566 22567 Also disable vbl when 3D isn't running 22568 22569commit d7775c1b38b5bdc439a27ec2c3c3a03a5b24cf57 22570Author: Dave Airlie <airlied@linux.ie> 22571Date: Sun Jun 3 12:46:31 2007 +1000 22572 22573 randr-1.2: make native mode preferred 22574 22575 If we have a native mode from the BIOS make it preferred mode 22576 22577commit 5a9516fe4f17854acd4fbf3a8eadf5139081dbd4 22578Author: Paul TBBle Hampson <Paul.Hampson@pobox.com> 22579Date: Sat Jun 2 14:28:18 2007 -0400 22580 22581 RADEONProbePLLParameters sets pll->reference_div, and 22582 RADEONGetPanelInfoFromReg uses it. 22583 22584 This ensures that the former is called before the latter. 22585 (this should fix randr on ppc) 22586 22587commit 802804461ebdca9a951e7e562ec68fd08d8eae01 22588Author: Matthieu Herrb <matthieu@deville.herrb.com> 22589Date: Thu May 31 22:23:57 2007 -0600 22590 22591 Fix build whithout XF86DRI 22592 22593commit 0e0946e0aa2527794e07473199e851bbfbc47cb1 22594Author: Michel Dänzer <michel@tungstengraphics.com> 22595Date: Wed May 30 18:56:53 2007 +0200 22596 22597 radeon: Lots of warning fixes. 22598 22599 Move code where it's used, remove unused variables, etc. 22600 22601commit 31c1be420d5277dd15505bd73e6144827a0580cd 22602Author: Dave Airlie <airlied@nx6125b.(none)> 22603Date: Wed May 30 17:49:01 2007 +1000 22604 22605 remove these syncs, at least on rs480 it doesn't break 22606 22607 probably requires testing on other r300 based cards, with the syncs in 22608 we hang when moving the cursor into the second CRTC. 22609 22610commit 7fc02657c4d740941fbda5a8823cf45de3eca3f8 22611Merge: 800bf53 4c61c0e 22612Author: Dave Airlie <airlied@nx6125b.(none)> 22613Date: Wed May 30 17:27:22 2007 +1000 22614 22615 Merge branch 'origin' into randr-1.2-test 22616 22617 Conflicts: 22618 22619 src/radeon_cursor.c 22620 src/radeon_display.c 22621 src/radeon_driver.c 22622 22623commit 800bf53279e2c2bf854682bbfd6fa16d03afed00 22624Author: Matthieu Herrb <matthieu@deville.herrb.com> 22625Date: Tue May 29 23:31:13 2007 -0600 22626 22627 Fix build without XF86DRI 22628 22629commit 4c61c0ee91a2ffeefce30972a584486f1df1d1ae 22630Author: Matthieu Herrb <matthieu@deville.herrb.com> 22631Date: Tue May 29 21:35:35 2007 -0600 22632 22633 Fix build without XF86DRI 22634 22635commit 5337e7bd0069a3f2c4ab22b21a19471427ad3d81 22636Author: Dave Airlie <airlied@nx6125b.(none)> 22637Date: Wed May 30 08:10:44 2007 +1000 22638 22639 radeon: add bios quirk for nx6125 monid 22640 22641commit bff809dc8ed07ac39e9b576a87916486a5e37156 22642Author: Dave Airlie <airlied@nx6125b.(none)> 22643Date: Wed May 30 08:02:26 2007 +1000 22644 22645 rs480: more unknown regs 22646 22647 Hardcode the values from a working fglrx run, this works for me now 22648 22649 I've no idea what it might do for anyone else 22650 22651commit 104105fee5c3945d3f210e6a4cb73ab492c61543 22652Author: Dave Airlie <airlied@nx6125b.(none)> 22653Date: Tue May 29 19:09:33 2007 +1000 22654 22655 rs480: make second crtc work with magic number in magic register. 22656 22657 I've no idea why or what this does. 22658 22659commit 5aa603bcabbb077dec169c48438c2e2ebe1195d7 22660Author: Dave Airlie <airlied@nx6125b.(none)> 22661Date: Tue May 29 07:23:24 2007 +1000 22662 22663 rs480: only has single dac 22664 22665commit c52322354fe64725733842b3356798c50e7735d5 22666Merge: dd6a966 975da59 22667Author: Dave Airlie <airlied@nx6125b.(none)> 22668Date: Tue May 29 07:21:48 2007 +1000 22669 22670 Merge branch 'origin' 22671 22672commit c2637a01f9aca4032262c66ade305f5fe2c54294 22673Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22674Date: Sat May 26 18:26:35 2007 -0400 22675 22676 RADEON: remove some dead code from the last commit 22677 22678commit bbb769c4107bfcae682e46d026e54cbfb67d62cd 22679Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22680Date: Sat May 26 18:25:06 2007 -0400 22681 22682 RADEON: more rotation work (still not there) 22683 22684 - once again borrowed heavily from intel 22685 22686commit a69f90be9384244744fabfa76469ede9cd26ac98 22687Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22688Date: Sat May 26 17:54:48 2007 -0400 22689 22690 RADEON: first pass at rotation (not working yet) 22691 22692 - based heavily on intel 22693 22694commit 3b619d88a65a5801c85d120f2d39704194433f10 22695Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22696Date: Sat May 26 17:06:18 2007 -0400 22697 22698 RADEON: clean up and add comment regarding clones 22699 22700commit 089ae4afd23322bf602d340c531c36a4c8b461ed 22701Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22702Date: Sat May 26 15:49:11 2007 -0400 22703 22704 RADEON: Switch SetOutputType() to use names rather than numbers 22705 22706commit c4cef0c9481257a744ac99dfc7beb988ce51a8ee 22707Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22708Date: Sat May 26 13:12:37 2007 -0400 22709 22710 RADEON: make sure we have the lock when SYNCing 22711 22712commit 899c62e987d5c7524817ba85675e39a8a28e0232 22713Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22714Date: Sat May 26 12:54:01 2007 -0400 22715 22716 RADEON: Make sure LVDS and FP2 routing info gets written 22717 22718commit 18857184ffa6847815d349c020b003f8401e36ee 22719Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22720Date: Sat May 26 12:27:00 2007 -0400 22721 22722 RADEON: fix handling of DRI lock 22723 22724commit 3a61453efb4f04492cef823b6dd1273b55c6a785 22725Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22726Date: Sat May 26 01:38:09 2007 -0400 22727 22728 RADEON: implement backlight control for LVDS 22729 22730 This code is currently disabled as I'm not sure which 22731 laptops actually use this method for backlight control. 22732 My laptop seems to use another method as adjusting the 22733 backlight level doesn't seem to touch LVDS_GEN_CNTL. 22734 Maybe just macs? 22735 22736commit ceec3f62257bafe4771e75d3c4f1d2a517d7acf8 22737Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22738Date: Fri May 25 23:29:51 2007 -0400 22739 22740 RADEON: implement auto/analog/digital output property for DVI-I (untested) 22741 22742commit e3e9c608651e7cbb9851489274815f47d2a8dbbb 22743Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22744Date: Fri May 25 19:26:24 2007 -0400 22745 22746 RADEON: more clean of last commits 22747 22748commit a7d2c9c0f6155c4e65a2e73f3832c0d8ca7af5fc 22749Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22750Date: Fri May 25 19:09:11 2007 -0400 22751 22752 RADEON: fixup last commit and add support for ext TMDS detect 22753 22754commit e50ca35ce2ea79dadb38ce14e459eed836452ff7 22755Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22756Date: Fri May 25 18:56:32 2007 -0400 22757 22758 RADEON: clean up non-DDC probe, add detect for DVI 22759 22760commit abb9b57e7b4162ab6c21fd1e809d24f13f7e1ea1 22761Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22762Date: Wed May 23 21:37:58 2007 -0400 22763 22764 RADEON: clean up some logic 22765 22766commit 679236defd28bea8874004377436347950ff7bec 22767Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22768Date: Wed May 23 21:18:40 2007 -0400 22769 22770 RADEON: add output properties (not functional yet) 22771 22772commit 975da595f032c145ad74079ff8edeaead779dc7b 22773Author: Michel Dänzer <michel@tungstengraphics.com> 22774Date: Tue May 22 10:56:47 2007 +0200 22775 22776 radeon: Provide new DRI texOffsetStart hook when available with EXA. 22777 22778commit e38cad5633c974467f6417051eda8d487ac3686e 22779Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22780Date: Tue May 22 00:09:46 2007 -0400 22781 22782 RADEON: Only add valid connectors 22783 22784commit 8275151baac22c34149cef0b7d922771d24abc3e 22785Author: Michel Dänzer <michel@tungstengraphics.com> 22786Date: Mon May 21 10:25:48 2007 +0200 22787 22788 radeon: HW cursor cleanup. 22789 22790 Don't needlessly turn the HW cursor on/off in RADEONLoadCursor*(). 22791 22792 Besides cleaning up the code, this semms to avoid some HW cursor related 3D 22793 lockups, see https://bugs.freedesktop.org/show_bug.cgi?id=10815 . My best 22794 guess is that this is because the engine is now always idled before touching 22795 the CRTC registers. 22796 22797commit 137e3fc1899078af0f72303ab0a4e6cf35804a7b 22798Author: Michel Dänzer <michel@tungstengraphics.com> 22799Date: Mon May 21 10:25:48 2007 +0200 22800 22801 radeon: Suppress debugging output by default. 22802 22803 It can be enabled at runtime by increasing the log verbosity level. 22804 22805 Also change the prefix from (**) to (II) to make grepping the log file for 22806 defaults overridden by xorg.conf more useful again. 22807 22808 Turn some MC related debugging output into normal informational output as it's 22809 useful for recognizing corner cases that can cause stability issues. 22810 22811commit aed193a47a939451d9a6d05b02653b1d73e1d523 22812Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22813Date: Sun May 20 21:13:59 2007 -0400 22814 22815 RADEON: fix crtc1 (un)blanking code after last commit 22816 22817 - crtc1 MC control is on CRTC_GEN_CNTL 22818 - fix indenting 22819 22820commit 870c8043068a0f44b53d1148371b2cc1e3970a7b 22821Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22822Date: Sun May 20 21:03:10 2007 -0400 22823 22824 RADEON: several updates 22825 22826 - move output init from InitCrtc() functions to the output mode_set() 22827 - take the crtc off the MC when blanking 22828 - move EnableDisplay() to output commit() 22829 - clean up some dead code 22830 - don't enable dacs in initcrtc() functions, this is taken care of in EnableDisplay() 22831 22832commit 09bfc8ed000f95ede5b73f2bad69edc1a4d9bac6 22833Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22834Date: Sun May 20 18:06:22 2007 -0400 22835 22836 update to 6.6.192 for rc release 22837 22838commit 764cb73e8dec4040cdd418d249fc504399fca3ee 22839Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22840Date: Sun May 20 17:26:26 2007 -0400 22841 22842 Fix regular/"xinerama"/zaphod dualhead mode 22843 22844 - logic in RADEONUnblank() was wrong 22845 - Calling RADEONSetupConnectors() on second instance screwed up the port info 22846 - still seem to be HW cursor issues with zaphod mode 22847 22848commit 9a147fef8e0e2ede2a0008c4ecfbd9b00c8dc5f6 22849Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22850Date: Fri May 18 23:45:11 2007 -0400 22851 22852 RADEON: if connector is VGA set TMDS to none 22853 22854commit bbd6faff4c6acb48970d774375c8a61861405f96 22855Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22856Date: Fri May 18 23:30:19 2007 -0400 22857 22858 RADEON: Change default LVDS i2c line for powerpc 22859 22860commit 7940ec364f3cbf02ba64b92c74cbaad4555baf38 22861Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22862Date: Fri May 18 00:23:08 2007 -0400 22863 22864 RADEON: cleanup 22865 22866commit f71bfde7352ef858c1041037d7dc77c237e315a4 22867Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22868Date: Fri May 18 00:12:03 2007 -0400 22869 22870 RADEON: more cleanup 22871 22872commit f711e266d6927dec648f6ff26f15c6f48643f78c 22873Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22874Date: Thu May 17 23:59:07 2007 -0400 22875 22876 RADEON: turn off all outputs in screeninit(). We'll turn on the ones we want later 22877 22878commit bfc1c372d7475b7fa6bffb8a79dec1dc2f11ec59 22879Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22880Date: Thu May 17 23:45:29 2007 -0400 22881 22882 RADEON: More cleanup 22883 22884commit c60b3bc9e3e7463bdb42e2478be8cc3f22c63c68 22885Author: Paul TBBle Hampson <Paul.Hampson@pobox.com> 22886Date: Thu May 17 23:26:54 2007 -0400 22887 22888 Fix for infinite loop in RADEONGetLVDSInfo 22889 22890commit 11289f8206cab1a94bd64a3938cf9af50f19497e 22891Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22892Date: Tue May 15 00:48:31 2007 -0400 22893 22894 RADEON: remove some debugging code 22895 22896commit fba1a11e287ebf04cf311645d31299896bad7283 22897Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22898Date: Tue May 15 00:44:29 2007 -0400 22899 22900 RADEON: Move DAC regs to their own Save() function 22901 22902commit 9ed00e959fcdab2739a7e64ca1e303a6faf28f15 22903Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22904Date: Tue May 15 00:36:05 2007 -0400 22905 22906 RADEON: Lots of small fixes 22907 22908 - remove extra crtc2_base set 22909 - remove some debugging code 22910 - clean up RestoreMode() 22911 - make sure a DDC reg gets assigned for LCD_DDC 22912 - make sure we adjust the right frame in AdjustFrame() 22913 - rename RADEONMapControllers() to RADEONPrintPortMap() to 22914 reflect what it actually does now 22915 - make i2c bus name match the DDC port 22916 - remove or comment out un-needed code 22917 22918commit 5106c8fe5f74865cac8b00937739a4efde9fc254 22919Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22920Date: Sun May 13 21:48:26 2007 -0400 22921 22922 RADEON: Further cleanup 22923 22924commit 224a73e41e9be344d5644203e7ebd5a3a8272604 22925Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22926Date: Sun May 13 21:41:59 2007 -0400 22927 22928 RADEON: remove remnants of "old" multi-head support 22929 22930 The old screen based multi-head code is broken at this point 22931 and I have no intention of fixing it, so to clean things up 22932 I've removed it. 22933 22934commit 9c2f20a83a5fed14225f4c3ebcd8ca41e9d4bd48 22935Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22936Date: Sun May 13 21:13:23 2007 -0400 22937 22938 RADEON: additional cleanups 22939 22940commit e187321ab8dd58d2b2fe92c062d070ba4820a2bf 22941Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22942Date: Sun May 13 19:49:10 2007 -0400 22943 22944 RADEON: add new ConnectorTable option and re-add PanelSize option 22945 22946commit e60a7bcaf5611ad4706b1d1442ce4cae49145f42 22947Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22948Date: Sun May 13 18:14:29 2007 -0400 22949 22950 RADEON: remove mergedfb, etc. cruft from man page 22951 22952commit 51caa5dccc4a2e14d1b41bb7868ad20468dd4cfe 22953Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22954Date: Sun May 13 18:10:02 2007 -0400 22955 22956 RADEON: remove old mergedfb and dualhead options that are no longer used 22957 22958commit c5da9d4040cb08598d171d20f84d3f6c20a033e0 22959Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22960Date: Sun May 13 18:05:38 2007 -0400 22961 22962 RADEON: remove more dead code 22963 22964commit e776fa9ecc53d63f916a5447a76fb2bb3b824167 22965Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22966Date: Sun May 13 15:43:39 2007 -0400 22967 22968 RADEON: re-org randr code 22969 22970 - move crtc stuff to radeon_crtc.c 22971 - move output stuff to radeon_output.c 22972 22973commit 58ce388452b7bc790c438d75c9cf4a0f69f0d7b2 22974Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22975Date: Sun May 13 15:05:01 2007 -0400 22976 22977 RADEON: Remove dead code 22978 22979commit 66e8e6c8348d007930730e90295588efe8108844 22980Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22981Date: Sun May 13 14:25:03 2007 -0400 22982 22983 RADEON: several fixes 22984 22985 - give better DDC names 22986 - disable DAC if connector is DVI-D 22987 - fix indentation in ProbeDDCModes() 22988 22989commit 1c16c2ce9c5b02b03d23da965127d82eea4c4039 22990Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22991Date: Sun May 13 13:54:04 2007 -0400 22992 22993 RADEON: fix VT switch 22994 22995commit 7c66e903368f77ecc4d3bd1c9f08d2adbd85e83c 22996Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 22997Date: Sun May 13 13:37:59 2007 -0400 22998 22999 RADEON: move crtc offset handling into init/save/restore functions 23000 23001commit aec078eb0740651fba8ec602e8239bd679efc8ad 23002Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23003Date: Sun May 13 11:57:57 2007 -0400 23004 23005 RADEON: fix up DDCConnected() 23006 23007commit 0550c37ecc434b8075fb3c367d100ff27625bb64 23008Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23009Date: Sun May 13 11:44:50 2007 -0400 23010 23011 RADEON: don't need to pass pScrn to EnableDisplay() 23012 23013commit 117220527de9fd3158f600645bcfcaf46847f45f 23014Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23015Date: Sun May 13 11:37:35 2007 -0400 23016 23017 RADEON: remove hardcoded output limit and PortInfo stuff 23018 23019commit 7e5c29961ac2a9e9dbe5d6d2d73d11cd018d62b5 23020Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23021Date: Fri May 11 18:00:40 2007 +0200 23022 23023 RADEON: Fix RMX after the last commit 23024 23025commit ab5603edd8fc3ef0560bdfb6a6d9c6af2a49d1e5 23026Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23027Date: Fri May 11 17:34:35 2007 +0200 23028 23029 RADEON: Move LVDS, TMDS, DAC properties to the output rec 23030 23031commit 94eb0681de0641e490f06486468617a727fefe86 23032Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23033Date: Fri May 11 14:39:48 2007 +0200 23034 23035 RADEON: switch output dpms to use RADEONEnableDisplay() 23036 23037commit 673ede5578d5d9caf2adf0445fe1e684b034eea5 23038Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23039Date: Fri May 11 14:28:17 2007 +0200 23040 23041 RADEON: re-arrange output mode setting 23042 23043commit 0070a7d787adaae99f7bc2659be4b0f49f439db5 23044Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23045Date: Fri May 11 13:06:15 2007 +0200 23046 23047 RADEON: more re-org 23048 23049 - move crtc mode setting around 23050 - add dri lock/unlock to crtc lock/unlock calls 23051 23052commit bba456232ac9a6218aa7fbd504d6093fa72860cf 23053Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23054Date: Thu May 10 15:33:51 2007 +0200 23055 23056 RADEON: fix VT switch 23057 23058commit 1779a12a947401e5c6bcf784b47e9b3c80d37204 23059Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 23060Date: Thu May 10 06:22:35 2007 -0700 23061 23062 Add cscope files to .gitignore 23063 23064commit 0cb23277666db3b30438c6f88840d861e04df414 23065Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23066Date: Thu May 10 15:20:56 2007 +0200 23067 23068 RADEON: randr driver re-org checkpoint 23069 23070 - split the mode setting per-crtc 23071 - reduce start up flicker 23072 23073commit 33c370b1d8350945f80ac12097d3e91243a400f2 23074Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 23075Date: Wed May 9 16:16:39 2007 -0700 23076 23077 RADEON: 23078 - fix an ugly modesetting bug: if we happened to set the mode on 23079 CRTC1 before CRTC2, CRTC2's RestoreMode function would clobber 23080 CRTC1's CRTC_OFFSET register since we never updated 23081 ModeReg.crtc_offset... so make AdjustFrame use ModeReg and pull 23082 the call to it up before RestoreMode, seems to work ok here. 23083 23084commit 6263248a0044777a352e4ee7380b4b8f9afd091b 23085Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 23086Date: Wed May 9 14:52:00 2007 -0700 23087 23088 RADEON: 23089 - use fixup_mode hook to set RADEON_USE_RMX flag so panel scaling 23090 works 23091 - use valid_mode hook to prune invalid default modes from list 23092 - use adjusted_mode in crtc_mode_set (using adjusted_mode from 23093 fixup hook) 23094 23095commit d2497009e395800fbde5777465f3087a54b94418 23096Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23097Date: Wed May 9 22:23:45 2007 +0200 23098 23099 randr checkpoint 23100 23101 - server still hangs if you start with external monitor connected 23102 - RMX not working 23103 - more than 2 outputs now possible (untested) 23104 23105commit 4488f0737d5268168eab41440b7a3b5732efb15e 23106Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23107Date: Wed May 9 16:41:13 2007 +0200 23108 23109 more randr re-work 23110 23111 - remove RADEONQueryConnectedDisplays(); randr takes care of this now 23112 - print edid after randr detection 23113 23114commit 4d992386e2ab9d8c50f0484445564325dfb42930 23115Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23116Date: Wed May 9 16:12:21 2007 +0200 23117 23118 enable all outputs on each crtc on mode restore 23119 23120commit 2618cf2aa8ed76411b943eb90c95869814c2f151 23121Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23122Date: Wed May 9 15:48:40 2007 +0200 23123 23124 More re-org to allow more than 2 outputs 23125 23126 - Move radeon output crtc map into crtc_set_mode 23127 - in modeinit, set up all outputs attached to crtc 23128 23129commit 61b9e79cbeee6f735a4c82ec8a802aee85d8b890 23130Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23131Date: Wed May 9 15:11:23 2007 +0200 23132 23133 start to clean up MapControllers 23134 23135commit 6a724dd798c1a7b461672993c02be83a7bccded6 23136Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23137Date: Wed May 9 14:48:17 2007 +0200 23138 23139 minor cleanup 23140 23141commit 03860fed24b4f76cc7f1f4210ec8f8040fa04777 23142Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23143Date: Tue May 8 19:27:48 2007 +0200 23144 23145 remove some old mergedfb cruft 23146 23147commit 76670f665ebec7cdf40a04bf9379cb3ad4417507 23148Merge: 83f81ed a3ee422 23149Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23150Date: Tue May 8 18:41:25 2007 +0200 23151 23152 Merge branch 'master' into randr-1.2 and fix conflicts 23153 23154commit 83f81ed5e3c33c94c80500316c37a7cbfc51f41f 23155Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23156Date: Mon May 7 22:54:16 2007 +0200 23157 23158 RADEON: Fix cursor handling for randr 23159 23160commit 5fc21cd34436919300018d8d4850fc67db284eb3 23161Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 23162Date: Mon May 7 13:31:20 2007 -0700 23163 23164 - fix randr 1.2 on pre-RV350 chips by enabling DAC_MACRO_CNTL writes 23165 - set dac_cntl on non-primary crtcs 23166 - set XCRT_CNT_EN in CRTC_EXT_CNTL just because 23167 - fix warnings in calls to xf86PrintModeline (wants screen index not 23168 pScrnInfo) 23169 23170commit 99b3df154317f0209618e532282a3e7ad091c00f 23171Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 23172Date: Mon May 7 19:02:35 2007 +0200 23173 23174 RADEON - update randr cursor handling, LVDS setup fix 23175 23176 - quick fix for the cursor handling to update to the latest 23177 server bits 23178 - make sure connector type is CONNECTOR_PROPRIETARY for LVDS. 23179 23180commit a3ee42207aab77d93655a82fdcb32be38268b85f 23181Author: Dave Airlie <airlied@linux.ie> 23182Date: Wed Apr 25 09:52:22 2007 +1000 23183 23184 radeon: another attempt at fixing the mergedfb refresh rate 23185 23186 This attempts to keep it inside the 32-bit limit when multiplying things out 23187 later in the randr tree. 23188 23189 Let me know if I screwed this up.. 23190 23191commit 3828237200fc16d4d32664fb8358950c213d4897 23192Author: Dave Airlie <airlied@nx6125b.(none)> 23193Date: Sun Apr 22 11:36:00 2007 +1000 23194 23195 radeon: add support for DDC on some laptop chipsets 23196 23197 I noticed fglrx has DDC for the panel in the rs480 laptop, however radeon 23198 didn't pick it up, so I valgrinded fglrx and spotted 0x1a0/0x1a4 accesses 23199 I actually noticed this before from the BIOS but never figured it out. 23200 23201 So now I get DDC from the LCD on this laptop. 23202 23203commit dd6a966e862b774a8e8b9e1a085309219673efad 23204Author: Dave Airlie <airlied@nx6125b.(none)> 23205Date: Sun Apr 22 11:36:00 2007 +1000 23206 23207 radeon: add support for DDC on some laptop chipsets 23208 23209 I noticed fglrx has DDC for the panel in the rs480 laptop, however radeon 23210 didn't pick it up, so I valgrinded fglrx and spotted 0x1a0/0x1a4 accesses 23211 I actually noticed this before from the BIOS but never figured it out. 23212 23213 So now I get DDC from the LCD on this laptop. 23214 23215commit c81ed9bd7b37c9d02141d10f6c7bad3d0c57032f 23216Author: Dave Airlie <airlied@linux.ie> 23217Date: Sat Apr 21 18:58:40 2007 +1000 23218 23219 radeon: fix build on older server 23220 23221commit 406eec71116a58d42288a7f1c809a92d5bda7350 23222Author: Dave Airlie <airlied@linux.ie> 23223Date: Sat Apr 21 18:56:28 2007 +1000 23224 23225 radeon: fix build since patches for IBM don't actually build 23226 23227commit ad119960095b4b64f4c6793f65950c9967ce4989 23228Author: Matthias Hopf <mhopf@suse.de> 23229Date: Thu Apr 19 11:54:46 2007 +0200 23230 23231 Disable RMX for IBM Lewis server. 23232 23233 Due to the hardware layout RMX ddc_mode has to be set. 23234 If ddc_mode is set, RADEONValdiateFPModes() shouldn't be called. 23235 Bugzilla #10620 (3). 23236 23237commit 16ef77df4ebaf5ea13baa82972aaf98e71ac32ee 23238Author: Matthias Hopf <mhopf@suse.de> 23239Date: Wed Apr 18 17:36:15 2007 +0200 23240 23241 Set sync polarity restriction flags even for non-"digital separate" monitors. 23242 23243 According to Lisa Wu, this is correct regarding the VESA EEDID standard. 23244 Bugzilla #10620 (2), original patch by Lisa Wu @ATI 23245 23246commit 0abce69f0d826a7ca1a41d963cd4730b6e01c145 23247Author: Matthias Hopf <mhopf@suse.de> 23248Date: Wed Apr 18 17:32:52 2007 +0200 23249 23250 Fix inconsistent use of Mode lists. 23251 23252 Some scans used to only check every second entry, some stopped at the entry 23253 before the last entry. 23254 Bugzilla #10620 (1), original patch by Lisa Wu @ATI 23255 23256commit aea801cf9a5ce519a53d6fffd9a3a2e526ec79ea 23257Author: Matthias Hopf <mhopf@suse.de> 23258Date: Fri Apr 13 16:16:05 2007 +0200 23259 23260 Fix crash if MergedFB and secondary head not found 23261 23262 If the secondary head isn't found (Monitor unplugged etc.) but MergedFB 23263 is configured, the driver segfaults because it tries to access the mode 23264 list private structures, which are not filled in. 23265 23266commit 07ddffb32e6293c77b32c94b87ec468caef3d6f5 23267Author: Matthias Hopf <mhopf@suse.de> 23268Date: Wed Apr 11 14:36:51 2007 +0200 23269 23270 Fixed typo in mode list generation. 23271 23272commit 4effa67ea75736a31b9e78a7b35acf74b067c43e 23273Author: Dave Airlie <airlied@linux.ie> 23274Date: Mon Apr 9 22:08:31 2007 +1000 23275 23276 radeon: add support for enabling direct rendering on RS480 23277 23278 Thanks to Matthew Garrett and Ubuntu for the hw loan to get this working. 23279 23280 Still no 3D driver support but at least you should get CP acceleration for 23281 2D now. 23282 23283commit 6b25a4c48796e022a093f3072574ffe9709ecaf4 23284Author: Michel Dänzer <michel@tungstengraphics.com> 23285Date: Wed Apr 4 10:47:29 2007 +0200 23286 23287 radeon: Link nearest modes by default for clone mode. 23288 23289 This makes sure all modes of both CRTCs will be available by default with 23290 MergedFB. 23291 23292commit 3c892f163ec1fa9be6e733aab091c9b718f41efc 23293Author: Michel Dänzer <michel@tungstengraphics.com> 23294Date: Wed Apr 4 09:52:37 2007 +0200 23295 23296 radeon: Always sort modes when adding to list. 23297 23298 This makes sure mode lists will always be sorted from larger to smaller. 23299 23300commit 3a8190ccc79969925257e7b980b78d79053d208d 23301Author: Michel Dänzer <michel@tungstengraphics.com> 23302Date: Wed Apr 4 09:47:07 2007 +0200 23303 23304 radeon: Don't shrink virtual size based on secondary modes. 23305 23306commit 9b1e97284ce185d358ca756a235d2cee346fa53f 23307Author: Henry Zhao <henryz@localhost.localdomain> 23308Date: Sat Mar 31 23:01:52 2007 -0800 23309 23310 10205: Radeon driver's own mode validation code does not work properly 23311 23312commit 9c2dcd19be8fc2cc29e637d1e9748e66196e3900 23313Author: Henry Zhao <henryz@localhost.localdomain> 23314Date: Sat Mar 31 20:10:03 2007 -0800 23315 23316 9337: EDID modes do not participate in validation for CRT monitor 23317 23318commit 1acd6d6fa42acec07fb11aeb189f492ddb021cb4 23319Author: Michel Dänzer <michel@tungstengraphics.com> 23320Date: Fri Mar 30 11:06:10 2007 +0200 23321 23322 radeon: Guard some MergedFB specific code with info->MergedFB tests. 23323 23324 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=10442 . 23325 23326commit 1a71106c0e4fe5f650239dc694163fdf52d33663 23327Author: Michel Dänzer <michel@tungstengraphics.com> 23328Date: Tue Mar 27 10:13:21 2007 +0200 23329 23330 radeon: Fix typo. 23331 23332commit f87e57d4d773a019d1cc8a10425c57480430f6a4 23333Author: Dave Airlie <airlied@linux.ie> 23334Date: Tue Mar 27 18:08:54 2007 +1000 23335 23336 radeon: fix up crtc debug dereference problem 23337 23338commit 66b4a571a4e7960da6807d3f30955aa08e89ccc6 23339Author: Dave Airlie <airlied@linux.ie> 23340Date: Tue Mar 27 17:00:37 2007 +1000 23341 23342 update number to 6.6.191 for rc release 23343 23344commit fca30a6b581cb6c1466ab1bc316df8fed5d82b60 23345Author: Alex Deucher <alex@botch2.com> 23346Date: Mon Mar 26 23:26:51 2007 -0400 23347 23348 fix NULL dereference 23349 23350 when IsSecondary is true, crtc1 is NULL 23351 Noticed by Sverre Froyen. 23352 23353commit 97d8d1ed10d069343f2b3172ba64ca421821a602 23354Author: George Sapountzis <gsap7@yahoo.gr> 23355Date: Sat Mar 24 20:02:12 2007 +0200 23356 23357 Move atichip.c from ati to atimisc. 23358 23359commit aa8f5b02ebc9be60df48722588261627d6a457e8 23360Author: George Sapountzis <gsap7@yahoo.gr> 23361Date: Sat Mar 24 19:53:02 2007 +0200 23362 23363 [mach64] Use Mach64Chipsets[] instead of ATIChipNames[]. 23364 23365 atimisc is PCI-only now, we can get the chip name with xf86TokenToString(). 23366 23367commit 39e896a1e688ea2d2d21f88c1c5d34c5810aac1c 23368Author: George Sapountzis <gsap7@yahoo.gr> 23369Date: Sat Mar 24 19:47:18 2007 +0200 23370 23371 [mach64] Set pATI->ChipRevision correctly, instead of overriding. 23372 23373 There is no need to override pATI->ChipRevision for GX/CX, as it is only 23374 reported with a printf. 23375 23376commit f046a910ca117279fbabc6281b2e23439ec9ea4e 23377Author: George Sapountzis <gsap7@yahoo.gr> 23378Date: Fri Mar 23 22:19:17 2007 +0200 23379 23380 Drop probing by driver name from "Device" section. 23381 23382 atimisc is PCI-only now, so we only need to scan the PCI space. 23383 23384commit 9cd175d9cd4ed710fccb303664c77519ecaf1e21 23385Author: George Sapountzis <gsap7@yahoo.gr> 23386Date: Fri Mar 23 22:12:48 2007 +0200 23387 23388 Fold FillIn() back to Probe(). 23389 23390commit d7a8cd0e476034796fc38e25a28cd28d05ea4a13 23391Author: George Sapountzis <gsap7@yahoo.gr> 23392Date: Fri Mar 23 22:10:03 2007 +0200 23393 23394 Drop now unneeded _X_EXPORT's. 23395 23396commit 58626d8b78f26f0d9c480033d3c3a12e541342b1 23397Author: George Sapountzis <gsap7@yahoo.gr> 23398Date: Fri Mar 23 21:30:19 2007 +0200 23399 23400 Move {atimach64,r128,radeon}_probe.c from ati to subdrivers. 23401 23402 Subdrivers are now loaded from the wrapper Probe function rather than at screen 23403 creation time. 23404 23405 The wrapper Identify callback only prints chip families now, chip lists are 23406 printed when a subdriver is loaded. This also avoids duplication of subdriver 23407 Identify callbacks. 23408 23409 Unknown radeons should still get a list of known radeon and then probe fails... 23410 23411 Probe for atimisc last to avoid needless loading in most cases (r128, radeon). 23412 23413commit 2e3d43af1e5077cc61dd8668551a6291368d9ed2 23414Author: George Sapountzis <gsap7@yahoo.gr> 23415Date: Mon Mar 19 19:05:04 2007 +0200 23416 23417 [mach64] Compute MMIO address once for each case (PIO, MMIO). 23418 23419 For an MMIO-only driver, the extra MMIO address probing is not needed since we 23420 already compute the MMIO address during probe and do not make any BIOS calls. 23421 23422 For a PIO driver, this extra probing could even result in a wrong MMIO address 23423 since the checking is done by reading through PIO, not MMIO. The MMIO address 23424 is later corrected but having the extra probe there is useless. 23425 23426commit e7defc0e930f1b2d83623a769b2bfdb05c66a5fe 23427Author: George Sapountzis <gsap7@yahoo.gr> 23428Date: Mon Mar 19 19:04:10 2007 +0200 23429 23430 [mach64] Move setting of PIO and MMIO addresses up, after BIOS init. 23431 23432 Partially revert "Minor refactoring of pATI->Block0Base computation." 23433 23434 This parially reverts commit 2c8b33b761e4411451aea1eea3a89d629aeea40d. 23435 23436commit 3dfdba195f1ff794ab00585ab80e7ea6ac92fcf4 23437Author: George Sapountzis <gsap7@yahoo.gr> 23438Date: Sun Mar 18 16:40:07 2007 +0200 23439 23440 [mach64] Make mapping code friendlier to pci-rework source-code compatibility. 23441 23442commit 321db6d87c9110654ab386e5ec270116eca1e04f 23443Author: George Sapountzis <gsap7@yahoo.gr> 23444Date: Sun Mar 18 16:15:20 2007 +0200 23445 23446 [mach64] Add macros for pci-rework source-code compatibilty. 23447 23448commit dfd07b6e99020d1db43d7ce0cae4423d8c6b1f05 23449Author: George Sapountzis <gsap7@yahoo.gr> 23450Date: Sun Mar 18 16:38:26 2007 +0200 23451 23452 [mach64] Drop checking that BAR base + size fits in 32 bit. 23453 23454 I can't see how these tests could fail on 32-bit. OTOH they are potentially 23455 problematic with 64-bit PCI addresses. 23456 23457commit e674338a98c50800637b8ebc01adf3aec2a3eb38 23458Author: George Sapountzis <gsap7@yahoo.gr> 23459Date: Sun Mar 18 15:50:41 2007 +0200 23460 23461 [mach64] Single mapping per BAR. 23462 23463 - We reset write-combining of the FB when MMIO registers fall in the linear 23464 aperture. 23465 23466 - We set MMIOInLinear in all three aperture mappings since the mapping code 23467 depends on it. 23468 23469 - We map the linear aperture in its entirety for MMIOInLinear also. 23470 This simplifies mapping of HW cursor and the LE aperture on BE archs, as 23471 they always fall in the linear aperture. 23472 23473 - I assumed that BAR[2] is page-aligned (we know it's 4K-aligned for sure). 23474 23475commit e7969de8ec572d8801806f9525a9a081a58a03f1 23476Author: George Sapountzis <gsap7@yahoo.gr> 23477Date: Sun Mar 18 15:35:25 2007 +0200 23478 23479 [mach64] Drop poking the VGA MMIO aperture. 23480 23481 We will not be able to handle this when we do mappings for PCI BARs only and 23482 it should be handled with a CPIO driver anyway. 23483 23484commit 166c760a86165330175023e07c4b2bd6891633c5 23485Author: Michel Dänzer <michel@tungstengraphics.com> 23486Date: Tue Mar 20 09:16:02 2007 +0100 23487 23488 Fix advertised minimum minor version of the DRI module. 23489 23490 We don't automatically require bumped minor versions. 23491 23492commit 1bdd376dbd57de8925244f0808f974d6d8cff39d 23493Author: Michel Dänzer <michel@tungstengraphics.com> 23494Date: Tue Mar 20 09:14:39 2007 +0100 23495 23496 radeon: Only sync to hardware when really necessary with EXA. 23497 23498 In particular, don't sync again after accelerated DownloadFromScreen, which 23499 syncs implicitly. This avoids calling into the kernel when it's not necessary, 23500 which can be relevant in some situations. 23501 23502commit 3cfa3a5c8daf03aaad6fc30d275709f6eb717d29 23503Author: Michel Dänzer <michel@tungstengraphics.com> 23504Date: Tue Mar 20 09:13:24 2007 +0100 23505 23506 radeon: Unify code to release the CP. 23507 23508commit e174d8df8c801fad95e5f79cff69187c200bee6e 23509Author: Michel Dänzer <michel@tungstengraphics.com> 23510Date: Tue Mar 20 09:12:51 2007 +0100 23511 23512 radeon: Minor BlockHandler cleanups. 23513 23514 Don't flush indirect buffer in BlockHandler; it's done in LeaveServer. 23515 23516 Also set the EXA engine mode to unknown only at the end. 23517 23518commit 113fb4b61e709a9b54fc2ef73efce06011e771c1 23519Author: Michel Dänzer <michel@tungstengraphics.com> 23520Date: Tue Mar 20 09:12:09 2007 +0100 23521 23522 radeon: Disable CP line acceleration on RV280s. 23523 23524 There have been several reports of stability issues with things like the 23525 xscreensaver hack hypercube. 23526 23527commit 4651d00b183cb498879d605c4b93cd3a0c63cb33 23528Author: Michel Dänzer <michel@tungstengraphics.com> 23529Date: Tue Mar 20 09:10:19 2007 +0100 23530 23531 radeon: Make sure 3D clients will re-upload textures to video RAM after LeaveVT. 23532 23533 Walk the SAREA texList and bump the age of every active object, so their owners 23534 will consider them kicked out when they grab the HW lock next time. 23535 23536commit 2d2fb54ba370c1df9ef5102e83c17a7ff5c55403 23537Author: Michel Dänzer <michel@tungstengraphics.com> 23538Date: Thu Mar 15 17:26:56 2007 +0100 23539 23540 atimisc: Fix strict aliasing violation flagged by -Wall. 23541 23542commit cfb82631fb683a0dc9bbc8f2d214646a8f1bca0b 23543Author: Michel Dänzer <michel@tungstengraphics.com> 23544Date: Thu Mar 15 17:12:43 2007 +0100 23545 23546 radeon: Fix build warnings. 23547 23548commit 50a524445ed1ea3c4fab56de772921ba86a21d35 23549Author: George Sapountzis <gsap7@yahoo.gr> 23550Date: Thu Mar 8 17:43:48 2007 +0200 23551 23552 Move atioption.c from ati to atimisc. 23553 23554commit 52684241ba1ffb393cb655ecc09cbd0cef5b1717 23555Author: George Sapountzis <gsap7@yahoo.gr> 23556Date: Thu Mar 8 07:07:02 2007 +0200 23557 23558 Move atiprobe.c from ati to atimisc. 23559 23560commit 2a1cd107a593630001799d6cd9e72c64222553b2 23561Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 23562Date: Tue Mar 6 14:35:50 2007 -0800 23563 23564 Add prepare/commit hooks to output and crtc func table 23565 23566commit 703c6fc0142ffc600285c13fe6dafecf988c0a1d 23567Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 23568Date: Tue Mar 6 10:12:42 2007 -0800 23569 23570 Update for new CRTC resize hooks. 23571 23572commit 1e270b411cf691c69194932b864ee2c9db292263 23573Author: George Sapountzis <gsap7@yahoo.gr> 23574Date: Sun Mar 4 18:21:49 2007 +0200 23575 23576 [mach64] Bug 5586: overlay scaler limited to 720 pixels on ATI Rage Pro. 23577 23578commit 54f9b8c1d01ecb97e2d4d247918f0efc86ed6037 23579Author: George Sapountzis <gsap7@yahoo.gr> 23580Date: Sun Mar 4 18:16:40 2007 +0200 23581 23582 [mach64] Xv: use single surface and encoding structs. 23583 23584commit e1e55b533d3ae528c8da37dcb77ed906d05697fb 23585Author: Dave Airlie <airlied@linux.ie> 23586Date: Sun Mar 4 19:11:34 2007 +1100 23587 23588 radeon: add option to set pci aperture size from config file 23589 23590 This requires a drm > 1.26 to work 23591 23592commit 58ee31d015cf8bec0edca62a46faec0b3505be8c 23593Author: Dave Airlie <airlied@linux.ie> 23594Date: Sun Feb 25 23:29:09 2007 +1100 23595 23596 fix typo 23597 23598commit 3cfe94d5438961b869766dacbcd13fde8b770ca2 23599Merge: 31c018c 3b43788 23600Author: Dave Airlie <airlied@linux.ie> 23601Date: Sun Feb 25 23:27:19 2007 +1100 23602 23603 Merge branch 'radeon-randrv12-v4' 23604 23605 Conflicts: 23606 23607 src/radeon_display.c 23608 src/radeon_driver.c 23609 23610commit 3b43788c45f51ad2d3e8e64383c412f4ddd97207 23611Author: Dave Airlie <airlied@linux.ie> 23612Date: Sun Feb 25 23:17:31 2007 +1100 23613 23614 cleanup radeon code against master server 23615 23616commit 24c6fa7cfac5602ba9e6e2f331bcac52fab258e5 23617Author: Dave Airlie <airlied@linux.ie> 23618Date: Sun Feb 25 21:43:54 2007 +1100 23619 23620 make radeon randr build against master server 23621 23622commit 4374895d393bf170134ccffc188521fe515d5a77 23623Author: George Sapountzis <gsap7@yahoo.gr> 23624Date: Sat Feb 24 21:25:13 2007 +0200 23625 23626 [mach64] Consolidate adjustments of mode timings, part 3. 23627 23628 Cosmetic. 23629 23630commit 8b9a4ec8af90b2f9f0d0a3a7f6627061df7e01b3 23631Author: George Sapountzis <gsap7@yahoo.gr> 23632Date: Sat Feb 24 21:22:49 2007 +0200 23633 23634 [mach64] Consolidate adjustments of mode timings, part 2. 23635 23636 Factor out to seperate function. 23637 23638commit bb226c9e7218d2a65c056fe74cb7eece0550670f 23639Author: George Sapountzis <gsap7@yahoo.gr> 23640Date: Sat Feb 24 21:20:53 2007 +0200 23641 23642 [mach64] Consolidate adjustments of mode timings, part 1. 23643 23644 Move LCD block from atimode.c to atimach64.c . 23645 23646commit 31c018ca4a18ce426b29006f103f56eee7f985fa 23647Author: Roland Scheidegger <sroland@tungstengraphics.com> 23648Date: Wed Feb 21 13:47:37 2007 +0100 23649 23650 add missing part of last commit... 23651 23652commit df07fa14da73e92d1a6ee0173468ab5c075d1cbb 23653Author: Roland Scheidegger <sroland@tungstengraphics.com> 23654Date: Wed Feb 21 13:05:42 2007 +0100 23655 23656 fix alignment issues with planar yuv and a bug with packed uyvy 23657 23658 respect that all source planar yuv planes are already dword aligned. 23659 Some attempts to fix up odd widths and odd heights (which are a bit strange 23660 for 4:2:0 formats). They still don't quite work 100% correctly (at the borders) 23661 but neither do they with packed yuv formats. 23662 While here, fix totally broken packed UYVY format by inserting missing break... 23663 Both bugs reported by Felipe Contreras. 23664 23665commit dfcb431adfbbaaee0d262d32735585555a0cbde4 23666Author: Dave Airlie <airlied@linux.ie> 23667Date: Tue Feb 20 21:20:28 2007 +1100 23668 23669 radeon: move some i2c registers out into common register file 23670 23671 These regs are more useful than just multimedia 23672 23673commit 7a25512415bb40b772491bb4a773dfe02ac8b71a 23674Author: Alex Deucher <alex@botch2.com> 23675Date: Fri Feb 16 00:44:11 2007 -0500 23676 23677 Add "DRI" option 23678 23679 This option allows you to disable the DRI per card. It also 23680 removes the "RN50Force3D" option as it is now covered by this 23681 option. RN50 users should set this to TRUE if they want to force 23682 the DRI on. 23683 23684commit fa30ec6d5cd9bf4eb1a960592ca7311175219e4b 23685Author: George Sapountzis <gsap7@yahoo.gr> 23686Date: Tue Feb 13 15:35:32 2007 +0200 23687 23688 [mach64] PreInit: maxPitch, minor cosmetic. 23689 23690commit 1777dcc1956d910073e13322767d9a3de41a949b 23691Author: George Sapountzis <gsap7@yahoo.gr> 23692Date: Tue Feb 13 15:03:40 2007 +0200 23693 23694 [mach64] PreInit: pitchInc is local. 23695 23696commit 9cdf4fcc3a93a4e20244286392dc31a0b8e6a10a 23697Author: George Sapountzis <gsap7@yahoo.gr> 23698Date: Tue Feb 13 11:53:42 2007 +0200 23699 23700 [mach64] PreInit: Use goto bail. 23701 23702 Match ATILock - ATIUnlock. 23703 23704commit 482b85144fee42b6543dcc97a75899e363e05e9e 23705Author: George Sapountzis <gsap7@yahoo.gr> 23706Date: Mon Feb 12 14:27:37 2007 +0200 23707 23708 [mach64] Enable RENDER acceleration on the Pro variants. 23709 23710commit 63248f0b4308a4487cda3aa22daa36e3e0d38d14 23711Author: Dave Airlie <airlied@linux.ie> 23712Date: Mon Feb 12 19:37:36 2007 +1100 23713 23714 fix LVDS by moving bios reading around 23715 23716commit 6a25f620d40cbb063de94aa6b5267296cd5a0670 23717Author: David Airlie <airlied@asimov.stargames.com.au> 23718Date: Mon Feb 12 15:59:43 2007 +1100 23719 23720 remove assignments to pOutput before it is created 23721 23722commit 9234d8045c5fefbd1a781209409e55a13e3e5370 23723Author: David Airlie <airlied@asimov.stargames.com.au> 23724Date: Mon Feb 5 11:53:18 2007 +1100 23725 23726 remove all the locking glxgears runs fine over screen changes without it 23727 23728commit eda9fb15e8bc042661a91a3d8c921006dfb3ddd9 23729Author: George Sapountzis <gsap7@yahoo.gr> 23730Date: Tue Nov 21 04:02:41 2006 +0200 23731 23732 Consolidate atixv.c w/ atimach64xv.c, part 2. 23733 23734 Move {Initialize,Close}XVideo from atixv.c to atimach64xv.c . 23735 23736commit 67b240ddd4a6549a23885afc723b6efa7c52db4c 23737Author: George Sapountzis <gsap7@yahoo.gr> 23738Date: Tue Oct 31 16:44:10 2006 +0200 23739 23740 Consolidate atixv.c w/ atimach64xv.c, part 1. 23741 23742 Merge ATIXVInitializeAdaptor() in ATIMach64XVInitialiseAdaptor(). 23743 23744commit 4dac7896549392f7378cad9620618cec28f41a16 23745Author: George Sapountzis <gsap7@yahoo.gr> 23746Date: Sun Oct 29 03:55:39 2006 +0200 23747 23748 Consolidate aticursor.c w/ atimach64cursor.c 23749 23750commit 23b12b9ff55224e8c3c45eb58ea5fbf2f747d82e 23751Author: George Sapountzis <gsap7@yahoo.gr> 23752Date: Tue Oct 31 16:10:07 2006 +0200 23753 23754 Consolidate atiaccel.c w/ atimach64accel.c, part 2. 23755 23756 Merge ATIInitializeAcceleration() in ATIMach64AccelInit(). 23757 23758commit bc3b222b57d7edf7bdce00a53e8abd9047c314a4 23759Author: George Sapountzis <gsap7@yahoo.gr> 23760Date: Sat Oct 28 19:29:12 2006 +0300 23761 23762 Consolidate atiaccel.c w/ atimach64accel.c, part 1. 23763 23764 Move ATIResizeOffscreenLinear() from atiaccel.c to atimach64xv.c and make it 23765 static. It is only used by XV memory management functions. 23766 23767commit d876889add41c8052d4f3feaa21ddf384f5a7054 23768Author: George Sapountzis <gsap7@yahoo.gr> 23769Date: Tue Nov 21 04:47:06 2006 +0200 23770 23771 Drop non-PCI support from atividmem.c 23772 23773 - we require a PCI device (pVideo != NULL) 23774 - map the VGA framebuffer with xf86MapDomainMemory() 23775 23776commit 2c8b33b761e4411451aea1eea3a89d629aeea40d 23777Author: George Sapountzis <gsap7@yahoo.gr> 23778Date: Tue Nov 21 04:51:02 2006 +0200 23779 23780 Minor refactoring of pATI->Block0Base computation. 23781 23782commit a659b7235332e19659c56cc9d7af362bd8885a29 23783Author: George Sapountzis <gsap7@yahoo.gr> 23784Date: Sun Feb 4 11:28:54 2007 +0200 23785 23786 Cleanup checks for pATI->PCIInfo. 23787 23788 Fixes a bug where shadowfb would not be enabled by default in the non-accel 23789 case because pATI->PCIInfo is now set after ATIProcessOptions() is called. 23790 23791commit 9d77aabdff919360f0c9333105436c31f1f5749a 23792Merge: ff8ea19 57822be 23793Author: George Sapountzis <gsap7@yahoo.gr> 23794Date: Mon Feb 5 19:16:51 2007 +0200 23795 23796 Merge branch 'mach64-pci-1' 23797 23798commit d7317604c843e21c1df048a9253ed55a1957cccd 23799Author: David Airlie <airlied@linux.ie> 23800Date: Fri Feb 2 14:54:00 2007 +1100 23801 23802 get rid of mergetype and displaytype from radeon.h 23803 23804commit 15c68ac19ff9971ae02cf52036ba36ddea1a6759 23805Author: David Airlie <airlied@linux.ie> 23806Date: Fri Feb 2 14:30:16 2007 +1100 23807 23808 rename some variables to output from connector 23809 23810commit 72ef23ef7dd159d827e8e122482a58928d532268 23811Author: David Airlie <airlied@linux.ie> 23812Date: Fri Feb 2 14:26:29 2007 +1100 23813 23814 fixup mode probing a bit neater 23815 23816commit 96acf6b2b242454345cc4b9cfc7ca07e0b597b43 23817Author: David Airlie <airlied@linux.ie> 23818Date: Fri Feb 2 14:26:19 2007 +1100 23819 23820 bring over setproperty from intel code 23821 23822commit 76bc53f9b153880730ab61dcd2b6e4e7717e4058 23823Author: David Airlie <airlied@linux.ie> 23824Date: Fri Feb 2 14:15:03 2007 +1100 23825 23826 move i2c buses into outputs 23827 23828commit 47fb9ce657f018177a35b449a4d716dc03be9327 23829Author: David Airlie <airlied@linux.ie> 23830Date: Fri Feb 2 11:45:40 2007 +1100 23831 23832 fixup callers to crtcsetmode fixes my cursors 23833 23834commit 720730b1b0de632488d3b9818210ec5e9c7f07ae 23835Author: David Airlie <airlied@linux.ie> 23836Date: Thu Feb 1 16:43:38 2007 +1100 23837 23838 add locks for crtcs and some missing lines from intel update 23839 23840commit 6748732658850ea506f623a3622aa7135513ffd0 23841Author: David Airlie <airlied@linux.ie> 23842Date: Thu Feb 1 16:26:31 2007 +1100 23843 23844 update to latest intel codebase modulo using their mode set 23845 23846 and it breaks my cursor 23847 23848commit a77f08298dc7e097025e3f7f92e3665c0ef30095 23849Author: David Airlie <airlied@linux.ie> 23850Date: Mon Jan 29 16:12:24 2007 +1100 23851 23852 radeon: always reset cursor regs 23853 23854commit ff8ea19fcdce099732f9359e53cd62b9a04bfa6d 23855Author: Alex Deucher <alex@botch2.hsd1.va.comcast.net> 23856Date: Tue Jan 30 21:11:09 2007 -0500 23857 23858 add option to force the DRI on for RN50 chips 23859 23860 Allow user to force the DRI on for RN50 chips. 23861 3D is not guaranteed to work on these chips, 23862 however in some cases it does. 23863 fixes bug 9802. 23864 23865commit 57822be75740f339445f2375d44632560f4bbe57 23866Author: George Sapountzis <gsap7@yahoo.gr> 23867Date: Tue Nov 21 03:08:26 2006 +0200 23868 23869 Drop pATI->Chipset, no longer used. 23870 23871commit 8b7c88e9ed549777413136c0f2ccc61747b010d8 23872Author: George Sapountzis <gsap7@yahoo.gr> 23873Date: Thu Dec 28 14:46:10 2006 +0200 23874 23875 Add option "probe_sparse" to force probing sparse I/O bases. 23876 23877commit 3654b786f1ad1f4e6bfb2fff5ff057254958f7c0 23878Author: George Sapountzis <gsap7@yahoo.gr> 23879Date: Wed Dec 27 17:59:14 2006 +0200 23880 23881 Sane probe, based on work by Luc Verhaegen <libv@skynet.be>. 23882 23883commit 0117c9bde0d0bbb487bc259cf55e8066ebd63f6a 23884Author: George Sapountzis <gsap7@yahoo.gr> 23885Date: Wed Dec 27 17:49:45 2006 +0200 23886 23887 Drop checking of sparse I/O bases conflicts 23888 23889 This will be handled with an option somehow. 23890 23891commit 87cf2f1874ca8a7cc492d2e2b6918ba69a8aa713 23892Author: George Sapountzis <gsap7@yahoo.gr> 23893Date: Mon Aug 14 17:55:14 2006 +0300 23894 23895 Minor refactoring of module (un)loading for TV_OUT. 23896 23897commit 0b31333c9147af21c50493872779da12a3668e5d 23898Author: George Sapountzis <gsap7@yahoo.gr> 23899Date: Mon Nov 20 22:08:49 2006 +0200 23900 23901 Drop support for non-programmable clocks. 23902 23903 From README.ati: 23904 23905 Clocks for supported programmable clock generators: 23906 23907 The driver currently supports all programmable clock generators known 23908 to exist on Mach64 adapters. 23909 23910 Clocks for unsupported programmable clock generators: 23911 23912 This case is unlikely to occur, but is documented for the sake of 23913 completeness. 23914 23915 Thus: 23916 23917 - check for (pATI->ProgrammableClock > ATI_CLOCK_FIXED) && 23918 (pATI->ProgrammableClock < ATI_CLOCK_MAX) 23919 23920 - drop "probe_clocks" option 23921 23922 - pATIHW->ClockUnmap is no longer used 23923 23924 - pATIHW->ClockMap is only used with NewHW.crtc which is always ATI_CRTC_MACH64 23925 and has the identity map, so drop it 23926 23927 - (pATI->ProgrammableClock != ATI_CLOCK_INTERNAL) => (pATI->depth <= 8) 23928 23929commit c9a9ea5ebd1c27881c1fac19842cfa3af55c126e 23930Author: George Sapountzis <gsap7@yahoo.gr> 23931Date: Mon Nov 20 15:09:41 2006 +0200 23932 23933 Banked memory is no longer needed. 23934 23935 - (pATI->BankInfo.BankSize = 0) in all cases, cull pATI->BankInfo 23936 23937 - only keep the minimal pATIHW.SetBank interface for save/restore 23938 23939 - clean ATISwap() a little, (NewHW.crtc != ATI_CRTC_VGA) 23940 23941 - (UseSmallApertures == TRUE) <=> pATI->VGAAdapter 23942 23943commit 383e963e275d351ea3631c352f5795340162d69f 23944Author: George Sapountzis <gsap7@yahoo.gr> 23945Date: Sun Jan 28 00:44:31 2007 +0200 23946 23947 Always use the linear aperture. 23948 23949 - drop (pATI->OptionLinear == FALSE) 23950 23951 - AcceleratorVideoRAM is always set, i.e. VGAVideoRAM is not used 23952 23953 - pATI->LinearBase is always set 23954 23955 - xf86LinearVidMem() is now checked in atipreinit() for both CPIO and MMIO 23956 23957commit 9b126f45db27496c1ad16db65a61fe641a018983 23958Author: George Sapountzis <gsap7@yahoo.gr> 23959Date: Mon Nov 20 20:38:59 2006 +0200 23960 23961 We always use the accelerator CRTC within the xserver. 23962 23963 - cull (pATI->NewHW.crtc != ATI_CRTC_MACH64). 23964 23965commit 0abcb2b2c84a50e64bb6aa9e94760659ecf33add 23966Author: George Sapountzis <gsap7@yahoo.gr> 23967Date: Sun Jan 28 00:19:15 2007 +0200 23968 23969 Clean pATI->Adapter and pATI->VGAAdapter. 23970 23971 - cull (pATI->Adapter != ATI_ADAPTER_MACH64) 23972 23973 - treat pATI->VGAAdapter as Bool 23974 23975commit 817b4b5338dba4d8eb862f1f79ef0edb75aad7f4 23976Author: George Sapountzis <gsap7@yahoo.gr> 23977Date: Wed Dec 27 21:16:18 2006 +0200 23978 23979 Keep PCI mach64 only, drop: 23980 23981 - Chip < ATI_CHIP_88800GXC 23982 - Chipset != ATI_CHIPSET_ATI 23983 - Adapter != ATI_ADAPTER_MACH64 23984 23985 - depth < 8 23986 23987commit d350860e29f043e98bfb1da74b26280f1755ab6f 23988Author: George Sapountzis <gsap7@yahoo.gr> 23989Date: Wed Dec 27 20:56:45 2006 +0200 23990 23991 Keep PCI mach64 only (atipreinit.c, atimode.c), drop: 23992 23993 - Chip < ATI_CHIP_88800GXC 23994 - Chipset != ATI_CHIPSET_ATI 23995 - Adapter != ATI_ADAPTER_MACH64 23996 23997 - depth < 8 23998 23999 atimode.c only: 24000 24001 - NewHW.crtc != ATI_CRTC_MACH64 24002 24003 This allows to drop VGACalculate(), VGAWonderCalculate() cruft early. 24004 24005commit 5fa77f2d122e7267911a15235338d8d3f9eece2e 24006Author: George Sapountzis <gsap7@yahoo.gr> 24007Date: Wed Dec 27 14:50:48 2006 +0200 24008 24009 Drop probing for VGA, VGAWonder, 8514, Mach32. 24010 24011 - ChipHasSUBSYS_CNTL 24012 - Coprocessor 24013 - SharedAccelerator 24014 - SharedVGA <=> (VGAAdapter != ATI_ADAPTER_NONE) 24015 24016commit 1b2a9e1735da63bbb8cb41939668f789cee22579 24017Author: George Sapountzis <gsap7@yahoo.gr> 24018Date: Mon Nov 20 02:35:56 2006 +0200 24019 24020 Drop AVOID_NON_PCI. 24021 24022 This was not set anyway, because configure.ac would compute ATIMISC_NON_PCI and 24023 then test ATI_AVOID_NON_PCI to set AVOID_NON_PCI... 24024 24025commit 2fb84dd2fc1a663c7b9d77e706189439ffe08381 24026Author: Alex Deucher <alex@botch2.hsd1.va.comcast.net> 24027Date: Thu Jan 25 02:01:42 2007 -0500 24028 24029 I think we can say r300 support is not experimental 24030 24031commit e155659f4b006edff31f1fda88b32bf608dbeece 24032Author: Alex Deucher <alex@botch2.hsd1.va.comcast.net> 24033Date: Thu Jan 25 01:59:27 2007 -0500 24034 24035 update man page. fixes bug 6457 24036 24037 Most cards support 2D and 3D now 24038 24039commit fa12fe1cc90dd745f3eea35a07d4f3efd652e7b4 24040Author: Dave Airlie <airlied@linux.ie> 24041Date: Tue Jan 23 19:37:51 2007 +1100 24042 24043 remove unused restore palette function 24044 24045commit 4d3649b84a3325043c38cece4499de0095cebd71 24046Author: Dave Airlie <airlied@linux.ie> 24047Date: Tue Jan 23 19:37:21 2007 +1100 24048 24049 remove isused and isactive 24050 24051commit 9149e763865598c307cbefc753ff6ebdeeaf32ae 24052Author: David Airlie <airlied@asimov.stargames.com.au> 24053Date: Tue Jan 23 15:20:11 2007 +1100 24054 24055 radeon: fixup problem with cursor not going off 24056 24057 Also fixup secondary dac detect to return unknown for now 24058 24059commit 2a13a3b641d9acf4f50472e1c8ba07633c3b78d6 24060Author: David Airlie <airlied@asimov.stargames.com.au> 24061Date: Tue Jan 23 10:36:06 2007 +1100 24062 24063 fixup init disp bandwidth 24064 24065 This probably break old-style dualhead badly 24066 24067commit cbd84bed13582e82f8b2e84aa152602474c09cd4 24068Author: Dave Airlie <airlied@linux.ie> 24069Date: Mon Jan 22 22:19:46 2007 +1100 24070 24071 removed even more mergedfb remains 24072 24073commit 6296882135b183425a219efb75374d3b172de370 24074Author: Dave Airlie <airlied@linux.ie> 24075Date: Mon Jan 22 22:14:48 2007 +1100 24076 24077 remove more mergedfb relics 24078 24079commit 26e1bab2c7a4ad52710746265b3da495b54dab36 24080Author: Dave Airlie <airlied@linux.ie> 24081Date: Mon Jan 22 22:11:55 2007 +1100 24082 24083 radeon: fixup some of the video code after randr 24084 24085 This may not be complete, I need to test on dual-head system later 24086 24087commit 9008e1caa45e4a18e6f3289c7b17036730fe578a 24088Author: Dave Airlie <airlied@linux.ie> 24089Date: Mon Jan 22 21:16:47 2007 +1100 24090 24091 remove unused merged code 24092 24093commit e29a32d711553fcb4ca9928122ac285fe0b0c1a7 24094Author: Dave Airlie <airlied@linux.ie> 24095Date: Mon Jan 22 21:09:53 2007 +1100 24096 24097 add back overlay CRTC chooser this may not work yet 24098 24099commit aa9d04ba94a3cd60b248231da517e2817591fc69 24100Author: Dave Airlie <airlied@linux.ie> 24101Date: Sun Jan 21 20:29:56 2007 +1100 24102 24103 remove alot more mergedfb support 24104 24105commit e8be0056e8ff666f63a294691661f5dab289203e 24106Author: Dave Airlie <airlied@linux.ie> 24107Date: Thu Jan 18 16:41:25 2007 +1100 24108 24109 remove mergedfb 24110 24111commit a648050a3cc60f92b1ca0b3d707aadf93d076d91 24112Author: Dave Airlie <airlied@linux.ie> 24113Date: Wed Jan 17 17:28:07 2007 +1100 24114 24115 move some cursor code around for show/hide 24116 24117commit d2ecfb507282726122bb8b0d17fd3637d0ae7d46 24118Author: Dave Airlie <airlied@linux.ie> 24119Date: Wed Jan 17 17:16:27 2007 +1100 24120 24121 make dri work again 24122 24123commit 369f7c85ceff983defb7657b80ec9cd3e5440b07 24124Author: Dave Airlie <airlied@linux.ie> 24125Date: Wed Jan 17 11:14:09 2007 +1100 24126 24127 comment out unblanking on VT switch 24128 24129commit bdb66a2042f02c4b57bd3c0181a00b39fcbdb232 24130Author: Dave Airlie <airlied@linux.ie> 24131Date: Sun Jan 14 21:07:05 2007 +1100 24132 24133 radeon: get randr-1.2 mode switching mostly working on my laptop 24134 24135 The main change is to fix the dac macro + cntl register writes for rv350 and 24136 above, this still has a problem with resetting the same mode after connect 24137 disconnect cycle, need to talk to keithp 24138 24139commit aa0e7337815d4daca4df0671768621330b759011 24140Author: Dave Airlie <airlied@linux.ie> 24141Date: Sun Jan 14 19:17:32 2007 +1100 24142 24143 use more randr-ish names for variables 24144 24145commit 9d37f23aea43b74a7ec640e6b03617c8392e2572 24146Author: Dave Airlie <airlied@linux.ie> 24147Date: Sun Jan 14 19:10:11 2007 +1100 24148 24149 radeon: move blank/unblank to use randr 24150 24151commit 6bd4fe42789c38e7e804826715214ce6badcca6a 24152Author: Dave Airlie <airlied@linux.ie> 24153Date: Sun Jan 14 18:38:59 2007 +1100 24154 24155 disable dpms on/off as it was writing regs before we wanted to 24156 24157commit 3949288ed26a91e180b178f4796f2f7e1bdc8ed6 24158Author: Dave Airlie <airlied@linux.ie> 24159Date: Sun Jan 14 16:26:39 2007 +1100 24160 24161 radeon: destroy and fixup LVDS crtc 24162 24163commit f5ffb9fb3bdd7cec020240c5831ae92ee744972c 24164Author: Timo Jyrinki <timo.jyrinki@hut.fi> 24165Date: Fri Jan 12 23:08:50 2007 +1100 24166 24167 ati: fix detection of some newer radeons using ati wrapper 24168 24169 This is mainly due to the cards having a different resource 1. 24170 Fixes 6796 24171 24172commit 1e690c57ad1dfaeb760ef6e799faa6398e9a0551 24173Author: Dave Airlie <airlied@linux.ie> 24174Date: Fri Jan 12 23:03:10 2007 +1100 24175 24176 always create a second CRTC I think this should solve any problems 24177 24178commit 25b36a4c56422ad8e25a4a1c55055c2f062213ca 24179Author: Dave Airlie <airlied@linux.ie> 24180Date: Fri Jan 5 16:13:16 2007 +1100 24181 24182 minor randr1.2 fixups 24183 24184commit abe8791e4f9fa3e88273897b351387cd33822734 24185Author: Dave Airlie <airlied@linux.ie> 24186Date: Fri Jan 5 15:09:43 2007 +1100 24187 24188 hook up DPMS through xf86 24189 24190commit 8a9c68c2234b91ed38555f892afdad30b9e4b455 24191Author: Dave Airlie <airlied@linux.ie> 24192Date: Fri Jan 5 15:09:15 2007 +1100 24193 24194 set binding for new heads 24195 24196commit d6a2b8aeb5a97ee907fd45b574bc6e4ab4b3aede 24197Author: Dave Airlie <airlied@linux.ie> 24198Date: Fri Jan 5 09:33:54 2007 +1100 24199 24200 fix LUT 24201 24202commit ef1d36e56dec1fec37cee2dfd9cb5bf8ce2c485c 24203Author: Roland Scheidegger <sroland@tungstengraphics.com> 24204Date: Wed Jan 3 15:56:23 2007 +0100 24205 24206 fix uninitialized mode regs (bug 9495) 24207 24208 Since the reorganization of the mode setting code, the mode registers relying 24209 on state already set (by bios) were not read, thus clearing out all bits the 24210 driver does not touch. At the very least, this could lead to completely 24211 nonfunctional to misbehaving dvi output (see bug 9495). Fix this by using the 24212 SavedReg values, which also makes it more obvious that those are bits which 24213 were not set by the driver previously, but come from register readback. 24214 (cherry picked from 9506f7015a1e442f4ca0bd3bfae555ec7e8a5f37 commit) 24215 24216commit 55aa832157bdebcba2d58896777942d108c352b0 24217Author: Dave Airlie <airlied@linux.ie> 24218Date: Thu Jan 4 17:57:31 2007 +1100 24219 24220 fix cursor handling 24221 24222commit 4f8a7cafdc77e98dc44f9eced876560b1ee01117 24223Author: Dave Airlie <airlied@linux.ie> 24224Date: Thu Jan 4 16:23:39 2007 +1100 24225 24226 hook up crtc modesetting 24227 24228commit 52f749c8a613ee316044abe82156ee270412ced8 24229Author: Dave Airlie <airlied@linux.ie> 24230Date: Thu Jan 4 16:23:26 2007 +1100 24231 24232 fix typo 24233 24234commit d7ff61c6822cbede7f5b59b411048d33dbae9ee4 24235Author: Dave Airlie <airlied@linux.ie> 24236Date: Thu Jan 4 13:38:26 2007 +1100 24237 24238 hook up detection of output 24239 24240commit 50439d7e9f4f6f9933bacd59f8bb3e655a35dbc2 24241Author: Dave Airlie <airlied@linux.ie> 24242Date: Thu Jan 4 13:21:57 2007 +1100 24243 24244 fixup some function calls 24245 24246commit e067bfee92f2b0877108355619b2fb9188a9d15a 24247Author: Dave Airlie <airlied@linux.ie> 24248Date: Thu Jan 4 13:21:45 2007 +1100 24249 24250 add more files for back compat 24251 24252commit f1f34627ffbe2136ac3e023c01c0430412919ded 24253Author: Dave Airlie <airlied@linux.ie> 24254Date: Thu Jan 4 12:50:03 2007 +1100 24255 24256 make restore mode non-static 24257 24258commit 249ec67c296b34d0c7c1d78602628d2f7ce6a045 24259Author: Dave Airlie <airlied@linux.ie> 24260Date: Thu Jan 4 12:43:03 2007 +1100 24261 24262 more minor fixes to get radeon up to speed 24263 24264commit a43c1d55f5f855d9e6ae939dd4eec1c607b6d514 24265Author: Dave Airlie <airlied@linux.ie> 24266Date: Thu Jan 4 12:29:51 2007 +1100 24267 24268 bring radeon randr code inline with intel randr code 24269 24270commit c54a31565e3ea50f1af7b81b1bf0c0e437099f47 24271Author: Dave Airlie <airlied@linux.ie> 24272Date: Thu Jan 4 10:39:01 2007 +1100 24273 24274 radeon: fixup fastwrites to be always off 24275 24276 My 8500 in i845 doesn't work with fastwrites even setup by the firmware. 24277 24278commit dda2bbfdc2e0da6e8b9cb8d1bf78b91f1bb8131f 24279Author: Dave Airlie <airlied@linux.ie> 24280Date: Thu Jan 4 10:29:54 2007 +1100 24281 24282 radeon: don't touch Controller[1] on single CRTC cards. 24283 24284 This might fix bug 9371 24285 24286commit 9506f7015a1e442f4ca0bd3bfae555ec7e8a5f37 24287Author: Roland Scheidegger <sroland@tungstengraphics.com> 24288Date: Wed Jan 3 15:56:23 2007 +0100 24289 24290 fix uninitialized mode regs (bug 9495) 24291 24292 Since the reorganization of the mode setting code, the mode registers relying 24293 on state already set (by bios) were not read, thus clearing out all bits the 24294 driver does not touch. At the very least, this could lead to completely 24295 nonfunctional to misbehaving dvi output (see bug 9495). Fix this by using the 24296 SavedReg values, which also makes it more obvious that those are bits which 24297 were not set by the driver previously, but come from register readback. 24298 24299commit d5c6306385cbc7b13a132ae72ed04e04e24d9886 24300Author: George Sapountzis <gsap7@yahoo.gr> 24301Date: Tue Dec 19 09:45:24 2006 +0200 24302 24303 Use portable XORG_RELEASE_VERSION macro. 24304 24305commit 97df0a32b14e34f22dce9cd3b375c1ac5fb1c5eb 24306Author: Michel Dänzer <michel@tungstengraphics.com> 24307Date: Tue Dec 19 11:28:21 2006 +0100 24308 24309 radeon: Advertise 2D engine limits to EXA when we're not using the 3D engine. 24310 24311 Also round up to the maximum width and height, as that's what EXA compares. 24312 24313commit 295823d0879a5b574bb79843a6acd43adb9259e5 24314Author: Michel Dänzer <michel@tungstengraphics.com> 24315Date: Tue Dec 19 11:19:20 2006 +0100 24316 24317 radeon: Default to leaving AGP transfer mode and fast writes unchanged. 24318 24319 Based on the assumption that firmware should have set up the card and host 24320 bridge appropriately for these settings, this may actually be safer, at least 24321 for the transfer rate; leaving fast writes enabled is hopefully safe as well, 24322 it certainly is on my sytem. 24323 24324 See https://bugs.freedesktop.org/show_bug.cgi?id=9284 . 24325 24326commit 81f3b4070b70483d6da4f7587e52a64ac69c8ca0 24327Author: Eamon Walsh <ewalsh@tycho.nsa.gov> 24328Date: Mon Dec 18 14:06:15 2006 -0500 24329 24330 Revert "Convert callers of LookupWindow() to dixLookupWindow()." 24331 24332 This reverts commit 48ff33a1770f3684cd50184db8f1944a456d9974. 24333 24334commit 87592ffb717da1f0a1767a38918d16d60953599c 24335Author: Dave Airlie <airlied@linux.ie> 24336Date: Sun Dec 17 20:10:58 2006 +1100 24337 24338 add setdpi call 24339 24340commit 41cd50487bc85708a3f791dfa70bf85223d91a65 24341Author: Dave Airlie <airlied@linux.ie> 24342Date: Sun Dec 17 20:08:07 2006 +1100 24343 24344 add displayWidth set 24345 24346commit 555e0f988c571aa47bc62ec6d9612a71f1a1f59b 24347Author: Dave Airlie <airlied@linux.ie> 24348Date: Sun Dec 17 19:59:19 2006 +1100 24349 24350 add randr pre init 24351 24352commit 3e2c72d7ac80ed874bcf2887a5253dac1c9bb02c 24353Author: Dave Airlie <airlied@linux.ie> 24354Date: Sun Dec 17 19:49:08 2006 +1100 24355 24356 fixup even more remnants of old code 24357 24358commit 4356b031886e00b3ed5ac1dbceeadebd7fc29fc2 24359Author: Dave Airlie <airlied@linux.ie> 24360Date: Sun Dec 17 19:47:45 2006 +1100 24361 24362 fix remnant of old code 24363 24364commit 17d39502a8dbea395051007aebffdf23644e9ee1 24365Author: Dave Airlie <airlied@linux.ie> 24366Date: Sun Dec 17 19:44:39 2006 +1100 24367 24368 use getddcmodes properly 24369 24370commit bf0b364cbd8682f297e2d110ece1f72e7151340d 24371Author: Dave Airlie <airlied@linux.ie> 24372Date: Sun Dec 17 17:31:08 2006 +1100 24373 24374 fix up output namings 24375 24376commit 4962a5430844114de864418d803f3182b90792ff 24377Author: Dave Airlie <airlied@linux.ie> 24378Date: Sun Dec 17 16:58:07 2006 +1100 24379 24380 remove some code that moved into server 24381 24382commit b47013f41249516ae82e22bd75caf573da2cc13c 24383Author: Dave Airlie <airlied@linux.ie> 24384Date: Sat Dec 16 12:02:37 2006 +1100 24385 24386 hook up mode detect to new randr code 24387 24388commit 48ff33a1770f3684cd50184db8f1944a456d9974 24389Author: Eamon Walsh <ewalsh@tycho.nsa.gov> 24390Date: Fri Dec 15 17:43:39 2006 -0500 24391 24392 Convert callers of LookupWindow() to dixLookupWindow(). 24393 24394commit 62a5e7565b66d7b7d65609c034f34b55bd266617 24395Author: Dave Airlie <airlied@linux.ie> 24396Date: Fri Dec 15 19:48:42 2006 +1100 24397 24398 bring over latest code from Intel git repo. 24399 24400 this code doesn't work yet just realigns with intel work 24401 24402commit 29124400c7f193317d41d8cfd748371a239cfea1 24403Author: Dave Airlie <airlied@linux.ie> 24404Date: Sun Dec 10 18:00:17 2006 +1100 24405 24406 hook up randr crtc setting 24407 24408commit 120c854f185c1e7711cf0dee19303fdb8894d49d 24409Author: Roland Scheidegger <sroland@tungstengraphics.com> 24410Date: Fri Dec 8 02:51:52 2006 +0100 24411 24412 radeon: avoid unnecessary OUTPLL/INPLL calls when displaying video 24413 24414 It is not necessary to always emit a OUTPLL/INPLL pair when we display 24415 a video frame. On some chips there are erratas for which the workarounds 24416 cause a 10ms delay by those calls. This is related to #5876 though those 24417 affected may suffer from other slowness issues too. 24418 24419commit ccd37b3e88cdcfe03b7c707b6082339eb8f11e6b 24420Author: Michel Dänzer <michel@tungstengraphics.com> 24421Date: Thu Dec 7 18:41:48 2006 +0100 24422 24423 radeon: Move Xv option handling code into RADEONPreInitXv(). 24424 24425 This unclutters RADEONPreInit() somewhat, but more importantly moves comparison 24426 against info->ChipFamily after that's initialized. 24427 24428commit 02be8c04a268eaa71bdfe891331758715fef4e71 24429Author: Michel Dänzer <michel@tungstengraphics.com> 24430Date: Thu Dec 7 18:31:58 2006 +0100 24431 24432 radeon: Make cache flushing for R300 series less invasive. 24433 24434 Instead of calling the DRM CP idle ioctl, just emit the cache flush commands 24435 into the CP stream. 24436 24437commit be46f7b6f7a729695a0affe2328077666283af12 24438Author: Michel Dänzer <michel@tungstengraphics.com> 24439Date: Thu Dec 7 16:27:10 2006 +0100 24440 24441 Fix radeon compile warnings. 24442 24443commit 59c5b4f69c97fb233144677ab18095f88aed006b 24444Author: Daniel Stone <daniel@fooishbar.org> 24445Date: Wed Dec 6 18:48:55 2006 +0200 24446 24447 Makefile.am: re-add ChangeLog to CLEANFILES 24448 24449commit 7611f14dc697b75cd7bb6493bb65b3a7ead379b8 24450Author: Daniel Stone <daniel@fooishbar.org> 24451Date: Wed Dec 6 18:47:33 2006 +0200 24452 24453 Makefile.am: fix changelog hook 24454 Make changelog hook safe for all situations. 24455 24456commit f2a9d2d61addc4e3d5e128d8078a04c9a136be6a 24457Author: Roland Scheidegger <sroland@tungstengraphics.com> 24458Date: Wed Dec 6 00:16:41 2006 +0100 24459 24460 radeon: fix up packed yuv broken by planar yuv patch 24461 24462 fix the forgotten leftuv value for packed yuv which is 24463 needed to get correct uv starting pixel (fixes broken clipping / 24464 non-null src start pixel as tvtime uses) 24465 24466commit d2aa57a4bdd4a0deb2c659bb07f3f57d1116f0d1 24467Author: Roland Scheidegger <sroland@tungstengraphics.com> 24468Date: Tue Dec 5 01:30:13 2006 +0100 24469 24470 radeon: use overlay scaler native planar yuv capability. 24471 24472 Radeons can do planar yuv just fine, there is no need to convert all data 24473 to packed yuv manually. This saves some cpu cycles as well as some 24474 (graphic card) ram. 24475 24476commit 5d5fa1b86e5179b061f0db47fe0227d1b84c37f8 24477Author: Dave Airlie <airlied@linux.ie> 24478Date: Mon Dec 4 18:53:33 2006 +1100 24479 24480 update radeon driver to Intel driver xf86Crtc interfaces 24481 24482commit 51d1cf19e71dd5de47f2c6467f4a1685eefd9e1e 24483Author: Dave Airlie <airlied@linux.ie> 24484Date: Sun Dec 3 16:30:01 2006 +1100 24485 24486 Add radeon randr 1.2 initial attempt at support 24487 24488 This doesn't do a huge amount yet 24489 24490commit eb17c9aed2144701ad7bd1042b2905446e4d708a 24491Author: airlied <airlied@optimus.localdomain> 24492Date: Sat Nov 18 15:48:02 2006 +1100 24493 24494 move radeon over to not using X's mode management like the intel driver. 24495 24496 this builds and at least seems to pick a mode.. might be near to hooking 24497 up randr1.2 soon 24498 24499commit 941b5120916ebff69ec0ee4ca7a46105d5306e5d 24500Author: airlied <airlied@optimus.localdomain> 24501Date: Sat Nov 18 10:41:12 2006 +1100 24502 24503 make radeon modes use generic code 24504 24505commit 608427283ea59cdfe3d85e1cf32719bc3809dc16 24506Author: airlied <airlied@optimus.localdomain> 24507Date: Sat Nov 18 10:35:31 2006 +1100 24508 24509 add more functions from i830 driver 24510 24511 This adds more generic functions from the i830 driver to the radeon 24512 24513commit d5b8cafc185b7d9fb909cb18a08615f81d89eaf8 24514Author: airlied <airlied@optimus.localdomain> 24515Date: Sat Nov 18 10:22:38 2006 +1100 24516 24517 add generic mode handling code from Intel driver 24518 24519commit 9cfa82e1670ad85746926995972a535ddf03ee07 24520Author: Dave Airlie <airlied@linux.ie> 24521Date: Sun Dec 3 15:56:38 2006 +1100 24522 24523 move monitor detection into single function 24524 24525commit 032a00d155ecdcdcab3ca0d047f552fe6b338a97 24526Author: airlied <airlied@optimus.localdomain> 24527Date: Sat Nov 18 10:23:12 2006 +1100 24528 24529 make DDC function more generic 24530 24531commit 2632ae24b3b50c69567e92611ce7c2cd2ee98baa 24532Author: Dave Airlie <airlied@linux.ie> 24533Date: Sun Dec 3 15:22:38 2006 +1100 24534 24535 re-program bios 4 scratch so we know if the LCD panel is there post startup 24536 24537commit ddec94e8f2f5746bad0e998d15303e771803d22b 24538Author: Dave Airlie <airlied@linux.ie> 24539Date: Sun Dec 3 11:07:23 2006 +1100 24540 24541 move crtc to connector mapping to a connector to crtc mapping. 24542 24543 Still not allowing any clone modes, but heading in the correct direction 24544 I hope... there is a chance this will regress something from superpatch.. 24545 24546commit 218cb7f8db35d0ce8f9fc4ece72bd5ec3f52106f 24547Author: Dave Airlie <airlied@linux.ie> 24548Date: Sun Dec 3 09:53:35 2006 +1100 24549 24550 radeon: dynamically allocate connectors 24551 24552commit f22fd02bc7f7ef67e33c9bd93eae2d7488669b5e 24553Author: Dave Airlie <airlied@linux.ie> 24554Date: Sat Dec 2 19:48:00 2006 +1100 24555 24556 make CRTCs malloc'ed rather than part of Ent 24557 24558 This is leading towards randr-1.2 believe me :-) 24559 24560commit 13efd1f75fbd29a4d1f69d7d5a5bb2b8d90ad820 24561Author: Dave Airlie <airlied@linux.ie> 24562Date: Sat Dec 2 19:16:41 2006 +1100 24563 24564 move HasCRTC2 to pRADEONEnt from info 24565 24566commit 3933b86390eb2f4198d490acd4914fb470c40f47 24567Author: Michel Dänzer <michel@tungstengraphics.com> 24568Date: Fri Dec 1 18:00:17 2006 +0100 24569 24570 Make log file output clear about origin of used overlay scaler buffer width. 24571 24572commit 563dfb52f5b24d2f765fae0394e599c11dcf0b5b 24573Author: Michel Dänzer <michel@tungstengraphics.com> 24574Date: Fri Dec 1 17:56:19 2006 +0100 24575 24576 Increase default overlay scaler buffer width for RV350. 24577 24578 Works with 1920x1080 video on my M10. 24579 24580commit 984ea3d9b7c27d8183ca94ad9d7905f03613fe10 24581Author: Michel Dänzer <michel@tungstengraphics.com> 24582Date: Fri Dec 1 17:42:23 2006 +0100 24583 24584 Fix build warning. 24585 24586commit 9f5ea3981449f29ff204eb154166e8fc813205fa 24587Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 24588Date: Wed Nov 29 16:46:57 2006 +0100 24589 24590 radeon: bug #1462, predownscale to make HD video work. 24591 24592 Use the overlay scaler's predownscale capability to make videos with large 24593 horizontal resolution work if it exceeds the scaler buffer width. Make the 24594 scaler buffer width user-configurable since we don't know it for all chips, 24595 and using predownscaling may otherwise reduce quality even if it wouldn't 24596 be needed. This should fix bug #1462. 24597 24598commit 35a5eaf5adf103ef57f8d4590dae25b0c50bd780 24599Author: airlied <airlied@optimus.localdomain> 24600Date: Tue Nov 28 17:31:05 2006 +1100 24601 24602 fixup EXA after previous commit 24603 24604commit ba047ae863aac869595a31ea94f102084636fd29 24605Author: airlied <airlied@optimus.localdomain> 24606Date: Tue Nov 28 17:26:10 2006 +1100 24607 24608 radeon: initial move framebuffer away from zero prep. 24609 24610 In theory the driver should be able to handle the front buffer not at VRAM 0 24611 In practice it didn't.. this is cleanup for at least XAA parts of the driver 24612 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...) 24613 24614commit 0d42cab2d889070ab6de05990e2b19724595f7af 24615Author: Alex Deucher <alex@t41p.(none)> 24616Date: Wed Nov 22 13:44:49 2006 -0500 24617 24618 clean up previous VIP commit. 24619 24620commit 8e0f57571228729cc40ed59266484ecc43819f44 24621Author: Alex Deucher <alex@t41p.(none)> 24622Date: Wed Nov 22 13:38:01 2006 -0500 24623 24624 Don't mess with the VIP bus on mobility chips. Seems to cause fan 24625 problems on asus laptops (see bug 7463). We don't support video-in 24626 on any laptops at the moment anyway. 24627 24628commit e4850610596a08082a528c38186268b72012e78c 24629Author: Alain Péteut <alain.peteut@gmx.ch> 24630Date: Wed Nov 22 12:54:55 2006 -0500 24631 24632 enable i2c for PCI_CHIP_RV250_If. allows AIW fuctionality to work. 24633 patch and testing from Alain Péteut 24634 24635commit ff38c2c9cf0c6acbdaf53b681ec721487cf411f8 24636Author: Michel Dänzer <michel@tungstengraphics.com> 24637Date: Sun Nov 19 17:11:46 2006 +0100 24638 24639 radeon: Bug #8989: Fix page flipping artifacts with XAA. 24640 24641 We still have to force-sync the pages on enabling page flipping with XAA as the 24642 second page may have been clobbered by the offscreen pixmap cache since they 24643 were last synced. 24644 24645commit 2dd0892817090709b2fb7ffba920595ada00fcad 24646Author: Michel Dänzer <michel@tungstengraphics.com> 24647Date: Sun Nov 19 17:09:11 2006 +0100 24648 24649 Fix compile warnings. 24650 24651commit 8c1b9269aaac746989bdb9be02915722bcf6fec0 24652Author: airlied <airlied@optimus.localdomain> 24653Date: Fri Nov 17 18:47:50 2006 +1100 24654 24655 forgot SetPitch function is called 24656 24657commit 3a0338bb487739a1740ddfb4d36d53cfa7f798c6 24658Author: airlied <airlied@optimus.localdomain> 24659Date: Fri Nov 17 18:37:00 2006 +1100 24660 24661 split mode handling code into separate modes source file 24662 24663commit 13e0a95564220d42ad8060b43b0aaaba3002157a 24664Author: David Airlie <airlied@asimov.stargames.com.au> 24665Date: Mon Nov 13 09:35:18 2006 +1100 24666 24667 fix up issue with bindings and mergedfb 24668 24669commit 63444770ff30afeb134befed5bb3a5c28b7bc4e9 24670Author: George Sapountzis <gsap7@yahoo.gr> 24671Date: Thu Nov 9 22:01:28 2006 +0200 24672 24673 [mach64] Prefer double-buffered visuals for GLX. 24674 24675 So that 3D apps apropriately fail in DRI... 24676 24677commit 0411f68d71c75cda38ad46ad5c0f5cc3cda64464 24678Author: George Sapountzis <gsap7@yahoo.gr> 24679Date: Thu Nov 9 22:00:56 2006 +0200 24680 24681 [mach64] Use asynchronous DMA mode by default. 24682 24683commit 8dd668db05e8f1282fa0e29a91e38c7cbdc97aa1 24684Author: Dave Airlie <airlied@linux.ie> 24685Date: Sun Nov 12 16:33:21 2006 +1100 24686 24687 radeon: abstract init display bandwidth from mergedfb 24688 24689commit 45a9f434b77ad892efcdba267a935135b972dd42 24690Author: Dave Airlie <airlied@linux.ie> 24691Date: Sun Nov 12 16:12:56 2006 +1100 24692 24693 add radeon Init2 function, to abstract some mergedfb mode setting 24694 24695commit d0828804aec0852afaca44ee41c4792edc40dbb4 24696Author: Dave Airlie <airlied@linux.ie> 24697Date: Sun Nov 12 15:59:02 2006 +1100 24698 24699 add initial controller binding, remove uses of mergedfb test 24700 24701commit 78061ca460c9d6fd338ccf273bbfa152f471622e 24702Author: Dave Airlie <airlied@linux.ie> 24703Date: Mon Oct 23 16:49:59 2006 +1000 24704 24705 radeon: separate static controller setup from non-static 24706 24707commit f917f65e8d2b13d8b9bfc4320f4a935f5a5c4bff 24708Author: George Sapountzis <gsap7@yahoo.gr> 24709Date: Thu Nov 2 21:16:53 2006 +0200 24710 24711 Officially require libdrm 2.2.0 24712 24713 Note that this is only for the headers at compile-time. The driver still 24714 uses libdrm 1.2.0 from Xserver at run-time. 24715 24716commit ca8c148942371c9445f32985625ce38eabc016e1 24717Author: George Sapountzis <gsap7@yahoo.gr> 24718Date: Thu Oct 26 15:40:46 2006 +0300 24719 24720 Handle .99 in version patch level. 24721 24722commit 23cec754158c94b43225b2070fc0a0b003e7bd25 24723Author: George Sapountzis <gsap7@yahoo.gr> 24724Date: Thu Oct 26 15:39:49 2006 +0300 24725 24726 [mach64] EXA: reserve DRI buffers only when DRI is active. 24727 24728commit 9cd78af1d06f753e7bd7e1acaff25398bde45f6e 24729Author: Dave Airlie <airlied@linux.ie> 24730Date: Mon Oct 23 16:52:07 2006 +1000 24731 24732 radeon: set preferred modes 24733 24734commit 914a5f0117ccb76292bdd81ea4a2a79d33319550 24735Author: Michel Dänzer <michel@tungstengraphics.com> 24736Date: Mon Oct 23 10:13:24 2006 +0200 24737 24738 radeon: Pitch doesn't matter when height == 1. 24739 24740commit 2bcb51d66edaa944379cf8c8ca1ba91fffdc20a8 24741Author: Michel Daenzer <michel@tungstengraphics.com> 24742Date: Sun Oct 22 17:26:28 2006 +0200 24743 24744 Bug #6756: Attempt to fix repeat picture acceleration. 24745 24746 Always use normalized texture coordinates on R200 and fall back if a POT 24747 texture pitch doesn't match the HW's implicit pitch. 24748 24749commit dabffb8335027b60ca1fc554423e196dfb9acd6d 24750Author: Henry Zhao <henry.zhao@sun.com> 24751Date: Tue Oct 17 21:17:47 2006 -0400 24752 24753 improved results for mergedfb xrandr refresh rates - bug 6966 24754 24755commit 831fc4a009841f68fe161b6e0db5b212d7d8079b 24756Author: Henry Zhao <henry.zhao@sun.com> 24757Date: Sun Oct 15 17:12:06 2006 -0400 24758 24759 fix bug 6966 (Henry Zhao, Shem Multinymous) 24760 24761commit 5425be996692425e57884e5a4c159fc79510e89f 24762Author: Alex Deucher <alex@t41p.(none)> 24763Date: Sun Oct 15 16:49:24 2006 -0400 24764 24765 clean up magic number 24766 24767commit 457178f418983041ea59df703a17513a47d17044 24768Author: Alex Deucher <alex@t41p.(none)> 24769Date: Sun Oct 15 16:40:08 2006 -0400 24770 24771 Disable all ouputs on initial mode set. The ones we want will get 24772 enabled by RADEONEnableDisplay(). 24773 24774commit 3fb3f410931a13542fdd4c51c4c9ecada7b37d76 24775Author: Alex Deucher <alex@t41p.(none)> 24776Date: Sun Oct 15 13:29:48 2006 -0400 24777 24778 Rework the output init code. Divide into separate functions 24779 for each output. 24780 24781commit 45e13f3c8786e75276cce85de4af653713d9b4d8 24782Author: Alex Deucher <alex@t41p.(none)> 24783Date: Sun Oct 15 13:18:01 2006 -0400 24784 24785 Remove paneloff option. This is already covered by monitorlayout and 24786 shouldn't be needed with xrandr++. 24787 Also remove commented out Bioshotkey code. 24788 24789commit 4abb5b7afea8d37e95461335e52b16168e555dab 24790Author: Alex Deucher <alex@t41p.(none)> 24791Date: Sun Oct 15 13:10:22 2006 -0400 24792 24793 Remove Bioshotkeys option. Leave the bios as configured at boot. 24794 Most distros enable this option by defalt anyway, and with xrandr++ 24795 it won't be an issue anyway. 24796 24797commit 2933140b5f1f8a6e270460d8918a14943bd0f174 24798Author: Michel Dänzer <michel@tungstengraphics.com> 24799Date: Sun Oct 15 17:05:02 2006 +0200 24800 24801 radeon: Use xf86XVFillKeyHelperDrawable() when available. 24802 24803commit 4edbfb9eb952f9b3f5f597749192c17baf15cdce 24804Author: Michel Dänzer <michel@tungstengraphics.com> 24805Date: Sun Oct 15 17:03:13 2006 +0200 24806 24807 radeon: Track screen damage all the time for page flipping. 24808 24809 This way we don't ever have to forcedly synchronize the pages so we hopefully 24810 never clobber one page's contents when we shouldn't. 24811 24812commit c9c2ac2b42398f5bfad909f3289e2545bb7b3888 24813Author: Michel Dänzer <michel@tungstengraphics.com> 24814Date: Sat Oct 14 15:49:22 2006 +0200 24815 24816 radeon: Don't try to synchronize pages when the CP isn't running. 24817 24818 Also move the DamageEmpty() call into RADEONDRIRefreshArea(). 24819 24820commit c19915a401ccf69046bed03c8d5067f2e72a094e 24821Author: Dave Airlie <airlied@linux.ie> 24822Date: Fri Oct 6 09:55:40 2006 +1000 24823 24824 always turn on DAC for scanning the CRT connect. 24825 24826 Ubuntu 63994 - myself and BenH discussed this as a good idea on irc 24827 24828commit b8547aa89f93a7fefbefb1c0d2743999930ac785 24829Author: Dave Airlie <airlied@linux.ie> 24830Date: Wed Oct 4 01:44:24 2006 +1000 24831 24832 update version number to 6.6.99 in preparation for a 6.7 branch at some stage 24833 24834commit 23a6f97e08fd49e1cae03cd97cae67a5f06b7634 24835Author: Alex Deucher <alex@samba.(none)> 24836Date: Tue Oct 3 13:48:09 2006 -0400 24837 24838 FP timing regs required for both internal and external TMDS 24839 24840commit 829be884702043fe4c9e349488d2585bf0561ae9 24841Author: Alex Deucher <alex@samba.(none)> 24842Date: Tue Oct 3 13:41:21 2006 -0400 24843 24844 remove bogus registers 24845 24846commit cd900580dbc8928ebc28514162b4c1dbfe9fd5ae 24847Author: George Sapountzis <gsap7@yahoo.gr> 24848Date: Tue Oct 3 00:32:22 2006 +0300 24849 24850 [mach64] Fix build for !defined(USE_EXA). 24851 24852commit 22860a3c0685a25f982983654303ad02aedc02ec 24853Author: George Sapountzis <gsap7@yahoo.gr> 24854Date: Sun Jul 16 19:23:56 2006 +0300 24855 24856 Bug 6242: [mach64] Use private DMA buffers. 24857 24858 Map the DMA buffers read-only. This eliminates a security problem where a 24859 client can alter the contents of the DMA buffer after submission to the DRM. 24860 24861commit cb53fe7c1e45937746e43437ae6adb0355306ae9 24862Author: Alex Deucher <alex@t41p.(none)> 24863Date: Sun Oct 1 18:46:05 2006 -0400 24864 24865 fix LVDS plus internal TMDS 24866 24867commit 145c716de6d42778e447581578f3e4db13c9a653 24868Author: David Woodhouse <dwmw2@infradead.org> 24869Date: Sun Oct 1 18:22:14 2006 +0200 24870 24871 Fix radeon vs fbdev pitch calculation. 24872 24873 Fetch the correct displayWidth from fbdevhw after setting the mode. 24874 24875commit d43ad88fa3913437f6987ab5ab46a38f0cb555a8 24876Author: Michel Dänzer <michel@tungstengraphics.com> 24877Date: Sat Sep 30 19:25:20 2006 +0200 24878 24879 radeon: Add support for page flipping with EXA. 24880 24881 Also use the damage layer directly instead of via shadowfb and blit dirty 24882 rectangles to the second page in LeaveServer in order to try and improve the 24883 tradeoff between performance and correctness. 24884 24885commit 146ea328b7962f56c6ab47fbe2adc7561b36189c 24886Author: Michel Dänzer <michel@tungstengraphics.com> 24887Date: Sat Sep 30 19:22:26 2006 +0200 24888 24889 radeon: Fix function name typo which could probably lead to an unresolved symbol. 24890 24891 Also remove declaration of an unused variable. 24892 24893commit 6671c1b01bf29d8f1cacf9306ef658b967d8a3cf 24894Author: Michel Dänzer <michel@tungstengraphics.com> 24895Date: Sat Sep 30 18:29:45 2006 +0200 24896 24897 radeon: Always call RADEONGetAccessibleVRAM(). 24898 24899 This fixes issues with Option "UseFBDev" and the new memmap code. 24900 24901 OTOH, there is potential for this to break IGPs, but that's what Bugzilla and 24902 git-bisect are for... 24903 24904commit c979c7f87fc449cc9631820101edff82c3013e66 24905Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 24906Date: Sat Sep 30 15:14:42 2006 +0200 24907 24908 radeon: Check ROM more thoroughly before treating it as an x86 BIOS. 24909 24910 This prevents the ROMs on some Mac cards getting treated as x86 BIOSen, 24911 resulting in garbage values. 24912 24913commit 37c114acbabe35881c39642e293519e2635907aa 24914Author: Michel Dänzer <michel@tungstengraphics.com> 24915Date: Sat Sep 30 15:11:42 2006 +0200 24916 24917 Remove some dead code. 24918 24919 Fix two compile warnings. 24920 24921commit 6d7b0763173f56050411d7bdf9a583b9bfc050c4 24922Author: Adam Jackson <ajax@benzedrine.nwnk.net> 24923Date: Fri Sep 29 18:50:44 2006 -0400 24924 24925 Undo unintended configure.ac change. 24926 24927 Dammit, git, grow a sensible UI already. 24928 24929commit f96ddd32d3a9e64dc0adec5c50386d6db5b182ee 24930Author: Ronald Wahl <rwahl@gmx.de> 24931Date: Fri Sep 29 17:51:42 2006 -0400 24932 24933 Hook up DFP power to DPMS events. 24934 24935commit f6b42149c5a3fee663575fe6bbe78b62695bdbc8 24936Author: Alex Deucher <alex@samba.(none)> 24937Date: Fri Sep 29 00:44:36 2006 -0400 24938 24939 tvdac handling was reversed between r200 and other chips 24940 24941commit c0660495080719c052d6393ede707755929102cd 24942Author: Dave Airlie <airlied@linux.ie> 24943Date: Wed Sep 27 18:12:52 2006 +1000 24944 24945 radeon: check DDC info for both connectors not just the pScrn one. 24946 24947 Otherwise we disable DDC before we've checked the CRT if LVDS is in use 24948 24949commit 5b98545618141e86164f3ce66469f9e16b1cf6d4 24950Author: Dave Airlie <airlied@linux.ie> 24951Date: Wed Sep 27 18:10:30 2006 +1000 24952 24953 radeon: don't move connectors around if LVDS connector is connector 1 24954 24955 On my laptop this allow me to plug or not plug a monitor without affecting 24956 my lvds 24957 24958commit 74c725bd86348b5bf9bba0c50e8a6a7885d7ff7b 24959Author: Dave Airlie <airlied@linux.ie> 24960Date: Mon Sep 25 11:35:32 2006 +1000 24961 24962 radeon: cleanup some bits from superpatch 24963 24964commit 9827afaa40720a58e2fc029c0bf92ad2fe223d11 24965Merge: 10b4b46 ad8259c 24966Author: Dave Airlie <airlied@linux.ie> 24967Date: Mon Sep 25 11:25:17 2006 +1000 24968 24969 Merge branch 'radeon-sp' 24970 24971commit ad8259c814629e741ed5567923f40879cc0c7051 24972Author: Dave Airlie <airlied@linux.ie> 24973Date: Mon Sep 25 11:25:03 2006 +1000 24974 24975 radeon : add back tv-dac register write 24976 24977commit 10b4b46c16ff3748856b732f9a39de40ba197112 24978Author: Dave Airlie <airlied@linux.ie> 24979Date: Mon Sep 25 09:08:47 2006 +1000 24980 24981 remove all CVS annotations from the ATI driver files 24982 24983commit 976dec04786e9b050945c290840b4cdc978c0b23 24984Author: Dave Airlie <airlied@linux.ie> 24985Date: Mon Sep 25 09:00:21 2006 +1000 24986 24987 radeon: put back in reverseddc code 24988 24989 BenH claims it is needed, agd5f can fix it up later if it isn't. 24990 24991commit 980fb2f6bd7641c8c57769b0c67e3561903e89a9 24992Author: Dave Airlie <airlied@linux.ie> 24993Date: Sat Sep 23 08:31:56 2006 +1000 24994 24995 radeon: add some missing bits of superpatch for crtc registers 24996 24997commit 1ba4f36821e60289cad937abbb0edb273c88436c 24998Author: Dave Airlie <airlied@linux.ie> 24999Date: Sat Sep 23 08:25:42 2006 +1000 25000 25001 radeon: finish radeon_driver.c import for register changes 25002 25003commit 547543bbefe605a453bfa5ae6d063ae02c5f040e 25004Author: Alex Deucher <alexdeucher@gmail.com> 25005Date: Sat Sep 23 08:21:59 2006 +1000 25006 25007 radeon: re-organise FP and CRTC register setting routines 25008 25009commit fd978140bcb7670f28c684c06c2b6c611c26bef4 25010Author: Dave Airlie <airlied@linux.ie> 25011Date: Sat Sep 23 08:09:57 2006 +1000 25012 25013 radeon: initial CRTC register setting alignment 25014 25015commit 406f4911e7ce821002c111d6bcdec35f5b56c943 25016Author: Dave Airlie <airlied@linux.ie> 25017Date: Sat Sep 23 07:59:24 2006 +1000 25018 25019 radeon: setup the radeon init function to use secondary 25020 25021commit 452f0f8079d65679905ed5178a256534ef0db0e4 25022Author: Dave Airlie <airlied@linux.ie> 25023Date: Sat Sep 23 07:41:57 2006 +1000 25024 25025 radeon: bring over final set of changes for radeon_display.c from susperpatch 25026 25027 This brings over Alex's final set of changes... 25028 25029 i'm unsure about the disappearance of reverseddc - alex to review 25030 25031commit a8b9aa69259a0706d0688d2656ece5a292799242 25032Author: Dave Airlie <airlied@linux.ie> 25033Date: Sat Sep 23 07:38:25 2006 +1000 25034 25035 radeon: align tabs and make display detection like superpatch 25036 25037commit 8d4ae98c528a2309dc3c545893dbba2e7f76916c 25038Author: Dave Airlie <airlied@linux.ie> 25039Date: Sat Sep 23 07:26:44 2006 +1000 25040 25041 radeon: align some output with superpatch 25042 25043commit 65b54847c5e79b2ec3cb1a08b265229443c9deb9 25044Author: Dave Airlie <airlied@linux.ie> 25045Date: Sat Sep 23 07:22:31 2006 +1000 25046 25047 radeon: change function name to Displays rather than Monitors 25048 25049commit 7e16d9fff96dcdff8c25dbbe2a51be0023b7985a 25050Author: Dave Airlie <airlied@linux.ie> 25051Date: Sat Sep 23 07:21:15 2006 +1000 25052 25053 radeon: make GetConnectedMonitors a static void 25054 25055commit 6f63b3dc8b3e2299f5f02abc1c565176d2d46cee 25056Author: Dave Airlie <airlied@linux.ie> 25057Date: Sat Sep 23 07:19:43 2006 +1000 25058 25059 radeon: some minor alignments and cleanups for info-> 25060 25061commit c9f3ced3b1c194cacf1ab9436ee6c2d4efd3a8e9 25062Author: Alex Deucher <alexdeucher@gmail.com> 25063Date: Sat Sep 23 07:16:10 2006 +1000 25064 25065 radeon: add reverse display option to switch displays on non-mobility cards 25066 25067commit e752443c540d9a94522050c61ef7c33d912fece5 25068Author: Dave Airlie <airlied@linux.ie> 25069Date: Fri Sep 22 07:10:10 2006 +1000 25070 25071 radeon: flesh out mapcontrollers 25072 25073commit 7a0e4eb47655c4701fc3a4b33b6c8e9f378ccc9e 25074Author: Dave Airlie <airlied@linux.ie> 25075Date: Fri Sep 22 07:05:56 2006 +1000 25076 25077 radeon: add some info to the map controllers 25078 25079commit 75bf69c868be6164d0041a32f52244f1ef83e050 25080Author: Dave Airlie <airlied@linux.ie> 25081Date: Fri Sep 22 07:03:52 2006 +1000 25082 25083 radeon: add simple map controllers call 25084 25085commit 8ba5abd1329347c7fbf4b4d09c95da0d23b4884a 25086Author: Dave Airlie <airlied@linux.ie> 25087Date: Fri Sep 22 06:58:41 2006 +1000 25088 25089 radeon: move comment 25090 25091commit e8d77b7c36fa791a2350a6f4b15728b15d8f3de0 25092Author: Dave Airlie <airlied@linux.ie> 25093Date: Fri Sep 22 06:55:51 2006 +1000 25094 25095 radeon: refactor blank/unblank 25096 25097commit 8534723f5710e2916b4a22efe22a6b7365cc8db6 25098Author: Alex Deucher <alexdeucher@gmail.com> 25099Date: Fri Sep 22 06:36:03 2006 +1000 25100 25101 radeon: make DVI_I default connector 25102 25103commit d7fc79fd9f094ac462de0883af7e6030cb6195a6 25104Author: Dave Airlie <airlied@linux.ie> 25105Date: Fri Sep 22 06:35:34 2006 +1000 25106 25107 radeon: add enable display function 25108 25109commit 4a54886d510f26b29d27e5c9a73647554291b1a6 25110Merge: 70264d6 9fa176c 25111Author: Dave Airlie <airlied@linux.ie> 25112Date: Fri Sep 22 06:00:04 2006 +1000 25113 25114 Merge branch 'master' into radeon-spm1 25115 25116 Conflicts: 25117 25118 src/radeon_driver.c 25119 25120commit 9fa176c7989030e7340cc9b85d0b6a7b34303091 25121Author: Alex Deucher <alex@samba.(none)> 25122Date: Tue Sep 19 13:58:17 2006 -0400 25123 25124 Provide all resolutions (instead only the largest one) in MergedFB clone 25125 mode without config file (Henry Zhao) 25126 25127commit e742aeb28c7d9d6e75932c408bcc7c44af52e303 25128Author: Alex Deucher <alex@samba.(none)> 25129Date: Tue Sep 19 13:45:33 2006 -0400 25130 25131 fix large negative refresh rates with mergedfb (Henry Zhao) 25132 fixes bug 6966 25133 25134commit 9c0c805704f1df1b66adc2a6b9313597ae5b86cc 25135Author: Dave Airlie <airlied@linux.ie> 25136Date: Mon Sep 18 19:41:21 2006 +1000 25137 25138 radeon: fix rn50 memmap 25139 25140 This should fixup the rn50/m6/m7 memory map problem that has been reported 25141 since BenH's memory mapping changes 25142 25143commit 70264d645724718a48ca229fedc04e33681b7d31 25144Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25145Date: Mon Sep 18 09:56:25 2006 +1000 25146 25147 radeon; put back old TV DAC code for now 25148 25149commit 53eb843ad8ec5a292dff0844330ecfa9b7b2a0cb 25150Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25151Date: Mon Sep 18 07:55:21 2006 +1000 25152 25153 radeon: move code around in InitCrtcRegisters 25154 25155commit bcae4352079109a9dd16117bf9ec19dd215180bf 25156Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25157Date: Mon Sep 18 07:33:48 2006 +1000 25158 25159 radeon: cleanup DPMS set on/off code to use new controller structure 25160 25161commit 87622f56b3dd7e2d2f00ad71e5b569eac31fb2b1 25162Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25163Date: Mon Sep 18 07:15:54 2006 +1000 25164 25165 radeon: cleanup DacPowerSet 25166 25167commit a417fa081926b2f84e2cbdc355cc7462d1ebdfad 25168Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25169Date: Mon Sep 18 07:15:41 2006 +1000 25170 25171 radeon: fixup InitFPRegisters function call to not pass info 25172 25173commit 8bff57a938e92dfc8729e40f7aa9ad9a96262131 25174Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25175Date: Mon Sep 18 07:02:35 2006 +1000 25176 25177 radeon: copy of overupdated bios hotkeys code from Alex 25178 25179commit 76352c684d5e840c9f7ac0781e3ecb32951b32db 25180Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25181Date: Mon Sep 18 06:54:28 2006 +1000 25182 25183 radeon: minor cleanup of masking for crtc registers 25184 25185commit a7b97697ed13ac25844890439ffba31524e03a16 25186Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25187Date: Mon Sep 18 06:53:23 2006 +1000 25188 25189 radeon: initialise controller structures 25190 25191commit 362ee3778764633cb4c9388aa329c5b715885038 25192Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25193Date: Mon Sep 18 06:41:14 2006 +1000 25194 25195 radeon: move InitFPRegisters up to align with superpatch 25196 25197commit 11fb9a14741746c752b0001c1b1d39a54e69657a 25198Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25199Date: Mon Sep 18 06:35:50 2006 +1000 25200 25201 radeon: add tv dac adj lookup 25202 25203commit a15ac59c4d1a15380910be2a7438111b09161c81 25204Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25205Date: Mon Sep 18 06:35:28 2006 +1000 25206 25207 radeon: re-add ATI patch for rn50 modes 25208 25209commit d80ea2d67b63ea2f2ce0277952af186b69b19e7d 25210Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25211Date: Mon Sep 18 06:30:30 2006 +1000 25212 25213 radeon: re-add DAC power on/off code around detect 25214 25215commit 0a93936f9f048b8837f1c765fea7ccb61f072697 25216Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25217Date: Mon Sep 18 06:25:15 2006 +1000 25218 25219 radeon: restore secondary head behaviour 25220 25221commit 6da5a0bfdedb2062ec835f45514c22917abdcf4c 25222Author: Alex Deucher <alexdeucher@gmail.com> 25223Date: Mon Sep 18 06:16:58 2006 +1000 25224 25225 radeon: move to using Alex's controller structure 25226 25227commit cd1bc1d1febf0a79d835c7dc8b353123278dec51 25228Author: Alex Deucher <alexdeucher@gmail.com> 25229Date: Mon Sep 18 02:44:09 2006 +1000 25230 25231 radeon: add save/restore of some new registers 25232 25233 move some registers about 25234 25235commit 9b5c4cf8a98b0c20bb821f5003495481fbd0c10c 25236Author: Alex Deucher <alexdeucher@gmail.com> 25237Date: Mon Sep 18 02:28:32 2006 +1000 25238 25239 radeon: add pre init controller function 25240 25241commit 311e776c4429485d29aaf11d350cab8e6fe7564b 25242Author: Alex Deucher <alexdeucher@gmail.com> 25243Date: Mon Sep 18 02:21:49 2006 +1000 25244 25245 radeon: add missing tv-out registers to register list 25246 25247commit 65cb77c79c2c5cb536494200d4fba6815d7e771c 25248Author: Alex Deucher <alexdeucher@gmail.com> 25249Date: Mon Sep 18 02:21:29 2006 +1000 25250 25251 radeon: restructure files along lines of Alex's superpatch 25252 25253commit 25fa71ca0a24190cc8be2e80769443e0e09303b3 25254Author: Henry Zhao <henryz@ati.com> 25255Date: Mon Sep 18 03:15:54 2006 +1000 25256 25257 Power up DAC around check for CRT connectedness 25258 25259commit 39fcac431eba60adf2d4e357cfc811dbae36a4e1 25260Author: Henry Zhao <henryz@ati.com> 25261Date: Mon Sep 18 03:09:43 2006 +1000 25262 25263 Bump up clock rate for ES1000 to allow higher resolutions 25264 25265commit 7d9e2b450f0fa42553156f57572a6dab70a0ea13 25266Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 25267Date: Tue Sep 12 07:04:23 2006 +1000 25268 25269 radeon: this stops mergedfb working on my ATI M7 development board 25270 25271 I've no idea what this is for, it came from an ATI codedrop and it may 25272 fix some laptop or production system, but it looks to me more like a dirty 25273 hack than a true fix, as I have an M7 which I need to specify CRT for the 25274 second head and it breaks mergedfb for me. 25275 25276commit ad1778eb1eb59de14ec309b99131931cbccf64e9 25277Author: Michel Dänzer <michel@tungstengraphics.com> 25278Date: Sun Sep 10 23:32:44 2006 +0200 25279 25280 Revert "* [ Bug 6966 ] large negative refresh rates reported" 25281 25282 This reverts caaed927a07ffbac68b08246185ef93c1e7bb98c commit. 25283 25284 See https://bugs.freedesktop.org/show_bug.cgi?id=8137 . 25285 25286commit 6e92981b024b59c2237f949321f6b2591dada519 25287Author: Tilman Sauerbeck <tilman@code-monkey.de> 25288Date: Sun Sep 3 21:11:10 2006 +0200 25289 25290 Removed the warning re. the r300 DRI driver status. 25291 25292commit caaed927a07ffbac68b08246185ef93c1e7bb98c 25293Author: Henry Zhao <henryz@localhost.localdomain> 25294Date: Thu Aug 31 13:27:45 2006 -0700 25295 25296 * [ Bug 6966 ] large negative refresh rates reported 25297 * [ Bug 6309 ] radeon/r128 fails to read hsync/vsync rates when rang 25298 * descriptor is missing 25299 * [ Bug 5473 ] Blank screen with Radeon Mobility X700 (Acer Ferrari 4005) 25300 * Reboot should not be needed for external CRT to function when connected 25301 later on Ferrari 4000 25302 * Modes need to be sorted, and sorted correctly 25303 * Provide all resolutions (instead only the largest one) in MergedFB clone 25304 mode without config file 25305 * Bump up clock rate for ES1000 to allow higher resolution 25306 25307commit 6d2028b84ff2b82c849b74776258f8e52c952e36 25308Author: Dave Airlie <airlied@linux.ie> 25309Date: Fri Aug 25 10:05:44 2006 +1000 25310 25311 and of course it takes two commits to do a simple commit 25312 25313commit e90c6feddbcd719fda3bbc3d48fd55045e9d5adc 25314Author: Dave Airlie <airlied@linux.ie> 25315Date: Fri Aug 25 09:59:58 2006 +1000 25316 25317 add changelog hook from nv driver 25318 25319commit 1e06212f792d1ed1f7545e16824b86fbd34fa284 25320Author: Adam Jackson <ajax@benzedrine.nwnk.net> 25321Date: Thu Aug 24 19:52:30 2006 -0400 25322 25323 Enable CPIO in the atimisc driver on ia64 platforms. 25324 25325 This fixes many mach64s, but potentially breaks mach32. Mach64 is widespread 25326 on ia64 machines but PCI mach32's are quite rare on any platform. 25327 25328commit dfa47b09b5b8d7eeaa3f8ae629ebb3c7efb51511 25329Author: Adam Jackson <ajax@benzedrine.nwnk.net> 25330Date: Thu Aug 24 19:50:46 2006 -0400 25331 25332 Disable DRI on xpress 200{,m} since we know it's broken. 25333 25334commit 1dcd9b1f75ba2307beb42556aa7f2f945171d5fb 25335Author: Adam Jackson <ajax@benzedrine.nwnk.net> 25336Date: Thu Aug 24 19:42:26 2006 -0400 25337 25338 Disable DRI on RN50 in a more pleasant way. 25339 25340 Rather than publish a non-existant driver name for RN50, just don't bother 25341 initializing it in the first place. 25342 25343commit 64b18e246586b1043a12f381a9d4c834464ed9b5 25344Author: Dave Airlie <airlied@linux.ie> 25345Date: Fri Aug 25 09:40:04 2006 +1000 25346 25347 update version numbers for release 25348 25349commit be229d97e1f0282f2099c7ace7cb257a2738a58f 25350Author: Dave Airlie <airlied@linux.ie> 25351Date: Fri Aug 25 09:39:46 2006 +1000 25352 25353 make colortiling default to on all cards 25354 25355commit 750ed51bc3d5119e2d2a91a05ae5a971dcd3b88a 25356Author: George Sapountzis <gsap7@yahoo.gr> 25357Date: Mon Aug 7 17:09:02 2006 +0300 25358 25359 [mach64] RENDER support: drop caching DP_WRITE_MASK. 25360 25361 Caching DP_WRITE_MASK causes artifacts, e.g. with evas_xrender_x11_test. 25362 25363commit b2beea2fa8949874d3d57fb9b43fe85cc08a8bff 25364Author: George Sapountzis <gsap7@yahoo.gr> 25365Date: Sat Aug 5 18:36:24 2006 +0300 25366 25367 [mach64] RENDER support: save/restore, cache texture registers. 25368 25369commit 43aaed99950640c3695b3c2b91faabf00c6338a7 25370Author: George Sapountzis <gsap7@yahoo.gr> 25371Date: Sat Aug 12 22:12:07 2006 +0300 25372 25373 [mach64] RENDER support (bug #6877). 25374 25375 RENDER accleration is disabled by default. It is strongly recommended that the 25376 patch from bug #6772 and the 'exa-damagetrack branch' are merged in the xserver 25377 before enabling RENDER acceleration. To enable RENDER acceleration, add the 25378 following in xorg.conf: 25379 25380 Section "Device" 25381 [...] 25382 25383 Option "AccelMethod" "exa" 25384 Option "RenderAccel" "true" 25385 EndSection 25386 25387commit d113f40fd97408661193895ff3aa82262f100143 25388Author: Dave Airlie <airlied@linux.ie> 25389Date: Sat Aug 12 17:16:33 2006 +1000 25390 25391 radeon: make some local data static 25392 25393 This is inspired by a patch in XFree86 from David Dawes, it isn't the 25394 same as we have different gamma curves. 25395 25396commit a1da2b5e0a38feae01107676a76bdc5c185ad34b 25397Author: George Sapountzis <gsap7@yahoo.gr> 25398Date: Sat Aug 5 03:18:34 2006 +0300 25399 25400 [mach64] Halve max w/h for hardware-accelerated operations. 25401 25402 EXA hits these limits for some operations on offscreen pixmaps and GTPRO seems 25403 to support the reduced limits. 25404 25405commit a66c8d2f9855caa10142b39f155dd8232b1f2389 25406Author: George Sapountzis <gsap7@yahoo.gr> 25407Date: Sat Aug 5 03:16:56 2006 +0300 25408 25409 [mach64] EXA: add memcpy-based UTS/DFS. 25410 25411 EXA hits more optimized paths when it does not have to fallback because of 25412 missing UTS/DFS. 25413 25414commit 0fb8503d60cf646c91a806f9ed322f9ceff18baa 25415Author: George Sapountzis <gsap7@yahoo.gr> 25416Date: Sat Aug 5 03:16:15 2006 +0300 25417 25418 [mach64] XAA: minor refactoring of memory manager setup. 25419 25420 - DRI: factor out to separate function and add comments on layout 25421 - no DRI: factor out to separate function, maxScanlines is always equal to 25422 ATIMach64MaxY 25423 25424commit e203d86643d5d70bf18248712d05b72b79aee705 25425Author: George Sapountzis <gsap7@yahoo.gr> 25426Date: Sat Aug 5 03:15:42 2006 +0300 25427 25428 [mach64] EXA support. 25429 25430commit dc1e289a611a17090e6dc7ae8a8d3f26d20df4eb 25431Author: Samuel Thibault <samuel.thibault@ens-lyon.org> 25432Date: Sat Jul 29 01:05:35 2006 +0300 25433 25434 Bug #6623: Fix I2C bus Mach64 initialisation failure. 25435 25436commit b377f403e0cd5253402ef3945e5944d1a2763a94 25437Author: Marc Aurele La France <tsi@xfree86.org> 25438Date: Sat Jul 29 01:03:34 2006 +0300 25439 25440 Fix warnings for Option:1 bit fields. 25441 25442commit 91b8b0ca41ad0b9659f0982a05148cab8558e9d5 25443Author: Marc Aurele La France <tsi@xfree86.org> 25444Date: Sat Jul 29 01:02:48 2006 +0300 25445 25446 Bug #807: copy throttling. 25447 25448 Fix corruption associated with the engine randomly not waiting for a copy 25449 operation to commit its results. 25450 25451commit 7800407a2db4b7307192bf3b43fffb9c4748b9c0 25452Author: Marc Aurele La France <tsi@xfree86.org> 25453Date: Sat Jul 29 01:00:56 2006 +0300 25454 25455 Bug #807: invalidate read-back cache. 25456 25457 Fix corruption when scaling 1-pixel wide pixmaps, in which case the first CPU 25458 read of the framebuffer will return stale data. 25459 25460 This bug always manifests itself in mach64 EXA (to be committed soon), with 25461 this patch mach64 EXA passes rendercheck. 25462 25463commit 49b11c540d37152eee4907ab30353ede01e4fb56 25464Author: George Sapountzis <gsap7@yahoo.gr> 25465Date: Sat Jul 29 00:52:28 2006 +0300 25466 25467 Bug #807: split out TestRegisterCaching block for DP regs. 25468 25469commit 290c5aaa580428bc4748a47e4b9bcc22b90fc8d4 25470Author: George Sapountzis <gsap7@yahoo.gr> 25471Date: Sat Jul 29 00:51:53 2006 +0300 25472 25473 Bug #807: split out TestRegisterCaching block for DP regs. 25474 25475commit fba8c839b47e2c8d6a6d65950c3431ff5b870aa0 25476Author: Michel Dänzer <michel@tungstengraphics.com> 25477Date: Sun Jul 23 18:51:47 2006 +0200 25478 25479 Bug #6819: Fix broken rendering with *8b8g8r8 source formats. 25480 25481 Use the corresponding texture formats with R200 family. 25482 25483 Unfortunately, it looks like the R100 family can't accelerate these source 25484 formats, so we have to fall back for them. 25485 25486commit 60ebe97e2d5d27c2bf4213e201cef74503809cb6 25487Author: Tilman Sauerbeck <tilman@code-monkey.de> 25488Date: Mon Jul 3 19:37:28 2006 +0200 25489 25490 Move variable declarations before statements to be C89 compatible 25491 25492commit 6900ff3c5e574881b1470e2363b200e99ab8bf3e 25493Author: Alex Williamson <alex.williamson@hp.com> 25494Date: Mon Jul 3 13:09:47 2006 +0200 25495 25496 Bug #6872: Disable digital output while reading ROM with Radeon 7000. 25497 25498 This is a workaround for ATI Radeon 7000 erratum F1. 25499 25500commit 920ac1f56fad6f48a30e65836ce054815f270738 25501Author: Michel Dänzer <michel@tungstengraphics.com> 25502Date: Mon Jul 3 01:21:18 2006 +0200 25503 25504 Do even less cache flushing when the previous engine mode is known. 25505 25506 OTOH, flush everything when the previous engine mode is unknown, and mark the 25507 engine mode as unknown in a couple more cases. 25508 25509commit dd0dd4bab76907b4b1b6d814f24e73e57d71f587 25510Author: Eric Anholt <anholt@freebsd.org> 25511Date: Mon Jul 3 01:10:10 2006 +0200 25512 25513 Bug #6762: Remove gratuitous cache flushing. 25514 25515commit 22f165f290f57e638d198f24a37f62fc16b72bae 25516Author: Michel Dänzer <michel@tungstengraphics.com> 25517Date: Mon Jul 3 00:39:45 2006 +0200 25518 25519 Build and warning fixes for !defined(USE_XAA). 25520 25521commit ea5105230c7c3519730b8d3c250e9ba9a4e4374a 25522Author: Michel Dänzer <michel@tungstengraphics.com> 25523Date: Mon Jul 3 00:03:22 2006 +0200 25524 25525 Reserve HW cursor memory statically instead of from EXA offscreen. 25526 25527 As a result, quite a bit of code can be removed that attempted to deal with the 25528 EXA offscreen memory disappearing, with varying success. Because the HW cursor 25529 memory is now always immediately after the front buffer and before the back 25530 buffer, this also fixes bug #6808. 25531 25532 As the HW cursor memory is now reserved statically with EXA and XAA, change 25533 RADEONUseHWCursor() to check info->cursor instead of info->cursor_offset, for 25534 which 0 might become valid with HW cursor enabled in the future. 25535 25536 Change info->cursor_offset from unsigned long to CARD32 as the corresponding 25537 register is 32 bits wide, and to avoid issues with printf. 25538 25539commit 66586f905857ea5bf37cb5d34949cd450a9ed067 25540Author: Michel Dänzer <michel@tungstengraphics.com> 25541Date: Sun Jul 2 23:38:20 2006 +0200 25542 25543 Clean up some redundant code. 25544 25545commit 1159c85861dc507ebf323aca82e32fcc67726420 25546Author: Michel Dänzer <michel@tungstengraphics.com> 25547Date: Sun Jul 2 18:24:28 2006 +0200 25548 25549 Bug #6548: Fix playback of interlaced video with more than 1023 lines. 25550 25551 This will still break with 2048 lines... 25552 25553commit 11bedb7c845fc6074383890d5525315aebe05f2c 25554Author: Michel Dänzer <michel@tungstengraphics.com> 25555Date: Sun Jul 2 17:03:23 2006 +0200 25556 25557 Bug #6939: Disable RENDER acceleration by default on RN50. 25558 25559 Based on patch from Weixing Zhang, but still allows enabling it with Option 25560 "RenderAccel". 25561 25562commit 4a0c22e32901a510c7aad070be25c31519c284e3 25563Author: Weixing Zhang <anzhang@ati.com> 25564Date: Sun Jul 2 16:55:25 2006 +0200 25565 25566 Bug #6939: Disable 3D acceleration on RN50 by setting invalid ClientDriverName. 25567 25568commit 9c4a1e3300d30b23806988aa9328fc6ef92e2663 25569Author: Frank Gevaerts <frank@gevaerts.be> 25570Date: Thu Jun 29 18:21:22 2006 +0200 25571 25572 Fix build for !defined(USE_EXA). 25573 25574commit ee0544b670567491b7770f469a64a6cd85519ba1 25575Author: Michel Dänzer <michel@tungstengraphics.com> 25576Date: Thu Jun 29 18:19:58 2006 +0200 25577 25578 Remove unused variables. 25579 25580commit 15c556f5a8c9069d3dd00ce139d46c39b7298951 25581Author: Michel Dänzer <michel@tungstengraphics.com> 25582Date: Thu Jun 29 18:18:21 2006 +0200 25583 25584 Don't declare unused variables on __powerpc__ and __alpha__. 25585 25586commit 055ac2836ec0afd51cf554d043a59068c36a07d3 25587Author: Michel Dänzer <michel@tungstengraphics.com> 25588Date: Thu Jun 29 18:14:45 2006 +0200 25589 25590 Fix strict aliasing violation flagged by gcc -Wall. 25591 25592commit 4b1904017caa976c138594a86e75feaf470e72b5 25593Author: Dave Airlie <airlied@linux.ie> 25594Date: Mon Jun 26 20:34:34 2006 +1000 25595 25596 radeon: force CP and VIP clocks on some r300 and rv100 chips. 25597 25598 fglrx appears to do this on r300 and Jerome Glisse has spent most of his 25599 life tracking down what caused the 9800 to lockup. This is my attempt to 25600 fix this. Please report if it works for you. 25601 25602 Signed-off-by: Dave Airlie <airlied@linux.ie> 25603 25604commit c5d8c8db3e75976687bfb364f0a7b235c7ae8793 25605Author: Michel Dänzer <michel@tungstengraphics.com> 25606Date: Sat Jun 24 16:01:19 2006 +0200 25607 25608 Bug #7274: Fix corruption with 'small' accelerated DownloadFromScreen transfers. 25609 25610commit f25ba3dc0e7850f0b1b78f54c0e7211bf2e81b3a 25611Author: Tilman Sauerbeck <tilman@code-monkey.de> 25612Date: Fri Jun 16 11:38:18 2006 +0200 25613 25614 Remove obsolete FIXME. 25615 25616commit 764013ec61b9043f1487513658e374569f7154a9 25617Author: Tilman Sauerbeck <tilman@code-monkey.de> 25618Date: Fri Jun 16 10:56:02 2006 +0200 25619 25620 Use AC_CHECK_HEADER() to check for exa.h 25621 25622commit 3c701d49b3eb7adb82267cec53163e475913aca3 25623Author: Tilman Sauerbeck <tilman@code-monkey.de> 25624Date: Fri Jun 16 09:59:29 2006 +0200 25625 25626 Use AC_MSG_RESULT() instead of echo for check results. 25627 25628commit dbef62615f38aebf4cdc579e3b26a7e7c719abc4 25629Author: Tilman Sauerbeck <tilman@code-monkey.de> 25630Date: Fri Jun 16 08:50:49 2006 +0200 25631 25632 conditionally declare pPCI and PciReg, too 25633 25634commit a01c01787cfbe3706ae7747cde1657abd79ecdaa 25635Author: Michel Dänzer <michel@tungstengraphics.com> 25636Date: Wed Jun 21 09:38:47 2006 +0200 25637 25638 Bug #7283: Build fixes for !defined(XF86DRI). 25639 25640 Thanks to Matthieu Herrb for pointing out some of these. 25641 25642commit 79d1c573133e64ddf4446b6bb0eaab4a5a6c73ab 25643Author: Michel Dänzer <michel@tungstengraphics.com> 25644Date: Sun Jun 18 17:39:09 2006 +0200 25645 25646 Implement accelerated EXA DownloadFromScreen hook. 25647 25648 x11perf -getimage numbers only increase by about 20-30% on my PowerBook with an 25649 M9, but by about 100 times(!) with a PCIe X550. I suspect the former could 25650 perform better with PCI as opposed to AGP transfers, which would also remove 25651 the need to disable this by default with AGP. 25652 25653commit 2bcdc7cfd013bd457d35f927390e5cc3544f30f1 25654Author: Michel Dänzer <michel@tungstengraphics.com> 25655Date: Sun Jun 18 16:01:00 2006 +0200 25656 25657 Remove #include <fbdevhw.h>, not used here. 25658 25659commit 25a5ccb63a9e3074b4c8eaf48844877d84ff76bb 25660Author: Michel Dänzer <michel@tungstengraphics.com> 25661Date: Sun Jun 18 15:26:55 2006 +0200 25662 25663 Prefer visuals with stencil when it's hardware acclerated. 25664 25665 Also make the formatting of the if statements consistent. 25666 25667commit 8fcc436f19f9c1371295af2d0c45608fb43c584b 25668Author: Michel Dänzer <michel@tungstengraphics.com> 25669Date: Sun Jun 18 15:16:08 2006 +0200 25670 25671 Take part of the blame for the radeon driver. 25672 25673commit c8366d7b457565a06acdf6b40c5448c7f13f1641 25674Author: Michel Dänzer <michel@tungstengraphics.com> 25675Date: Sun Jun 18 14:24:14 2006 +0200 25676 25677 Add Option "DepthBits". 25678 25679 This allows getting higher depth buffer precision and stencil hardware 25680 acceleration in depth 16 or saving some video RAM at the cost of these features 25681 in depth 24. 25682 25683commit 23a21d00db6f3b88b54c44648f968018a03677d3 25684Author: Tilman Sauerbeck <tilman@code-monkey.de> 25685Date: Fri Jun 16 16:37:07 2006 +1000 25686 25687 [PATCH] move the declaration of pMach64 into the correct ifndef statement 25688 Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de> 25689 Signed-off-by: Dave Airlie <airlied@linux.ie> 25690 25691commit c6b33db9d2c751e34c2152b0bb7e0e84a812589f 25692Author: Dave Airlie <airlied@linux.ie> 25693Date: Fri Jun 16 16:22:03 2006 +1000 25694 25695 ati: prepare for release 25696 25697 Update some versions for release 25698 25699commit 9c534b5330534fc6cf392936a9b200a5128c787c 25700Author: Michel Dänzer <michel@tungstengraphics.com> 25701Date: Thu Jun 15 16:09:55 2006 +0200 25702 25703 Some more ignores. 25704 25705commit 3d68698647fb17a6f5ae9a225f552cb0130fa946 25706Author: Michel Dänzer <michel@tungstengraphics.com> 25707Date: Thu Jun 15 16:07:15 2006 +0200 25708 25709 Fix semantic glitches pointed out by gcc -Wall. 25710 25711 The radeon driver builds warning-free with gcc -Wall here now. 25712 25713commit ab2b48e0df328fd08a2412911126b3184a601aea 25714Author: Michel Dänzer <michel@tungstengraphics.com> 25715Date: Thu Jun 15 16:06:13 2006 +0200 25716 25717 Remove another unused variable. 25718 25719commit 56bd5a370472663af5607cd327f757367b238db0 25720Author: Michel Dänzer <michel@tungstengraphics.com> 25721Date: Thu Jun 15 15:35:05 2006 +0200 25722 25723 Fix some more cosmetic warnings. 25724 25725commit ec93e08b24f7c0e583970d08bfe18ad142e6b047 25726Merge: 051c49d 6ca9862 25727Author: Michel Dänzer <michel@tungstengraphics.com> 25728Date: Thu Jun 15 15:03:27 2006 +0200 25729 25730 Merge branch 'master' of ssh+git://git.freedesktop.org/git/xorg/driver/xf86-video-ati 25731 25732commit 6ca986242c80bcfe7ca2d435a772bbdf034892c4 25733Author: Dave Airlie <airlied@linux.ie> 25734Date: Thu Jun 15 21:24:46 2006 +1000 25735 25736 remove unused variable 25737 25738commit 9a5f5bc0d3f5e62b79355993ca6246382394daee 25739Author: Dave Airlie <airlied@linux.ie> 25740Date: Thu Jun 15 21:24:16 2006 +1000 25741 25742 remove all printf specifier warnings on Linux 25743 25744commit 051c49d0f002c72d1d8f15f6df264d8b093d4e95 25745Merge: a840afc dfac819 25746Author: Michel Dänzer <michel@tungstengraphics.com> 25747Date: Thu Jun 15 13:10:44 2006 +0200 25748 25749 Merge branch 'master' of ssh+git://git.freedesktop.org/git/xorg/driver/xf86-video-ati 25750 25751commit dfac8191dbda29ef7f6ce33d7356fee89c17d59b 25752Author: Dave Airlie <airlied@linux.ie> 25753Date: Thu Jun 15 21:05:16 2006 +1000 25754 25755 fix last ati*.[ch] format specifiers 25756 25757commit 33525540f3a9d6760c6edcfe765d2a004a6b894f 25758Author: Dave Airlie <airlied@linux.ie> 25759Date: Thu Jun 15 21:03:01 2006 +1000 25760 25761 fix more printf formats 25762 25763commit 6bcf6f29af24b6725e097dd72af8e8b9f6d0676c 25764Author: Dave Airlie <airlied@linux.ie> 25765Date: Thu Jun 15 20:59:33 2006 +1000 25766 25767 some more unused variable cleanups 25768 25769commit a840afc75a0453b4694e3f987ebdbddc6844a169 25770Merge: c90c49d d511ef4 25771Author: Michel Dänzer <michel@tungstengraphics.com> 25772Date: Thu Jun 15 12:46:58 2006 +0200 25773 25774 Merge branch 'master' of ssh+git://git.freedesktop.org/git/xorg/driver/xf86-video-ati 25775 25776commit d511ef486a15b582e5d1708a17a00701f5cf1ccd 25777Author: Dave Airlie <airlied@linux.ie> 25778Date: Thu Jun 15 20:39:22 2006 +1000 25779 25780 ati/r128: add includes to get C function decls 25781 25782 Also fix a void pointer cast 25783 25784commit f46c3bebe821ee0031770d98d04bd406d1abb9ed 25785Author: Dave Airlie <airlied@linux.ie> 25786Date: Thu Jun 15 20:34:56 2006 +1000 25787 25788 fix some more unused variables and bad parameter passing 25789 25790commit 6c3f81683abab8ff6b8f2fcaf01a2a1a63af5ef6 25791Author: Dave Airlie <airlied@linux.ie> 25792Date: Thu Jun 15 20:25:00 2006 +1000 25793 25794 ati: fix -Wall in atidri.c 25795 25796 atidri.c has some parameter passing bugs, fixed those and cleaned other 25797 warnings. 25798 25799commit 58c6aac0669101481ef6b8217e5cede4573ece42 25800Author: Dave Airlie <airlied@linux.ie> 25801Date: Thu Jun 15 19:58:11 2006 +1000 25802 25803 theatre: fix all -Wall warnings in theatre* 25804 25805 This fixes all the warnings, it doesn't clean up the horrible mess that is 25806 theatre*.[ch]. 25807 25808commit c90c49da3844f93d2d29070b15bb7def72b98b0f 25809Author: Michel Dänzer <michel@tungstengraphics.com> 25810Date: Thu Jun 15 10:29:46 2006 +0200 25811 25812 Collapse .gitignore files into single toplevel file, sorted with LC_ALL=C. 25813 25814 Suggested by Keith Packard. 25815 25816commit b7a3e4c9b33f72ae447f99c0c765f24fce988142 25817Author: Michel Dänzer <michel@tungstengraphics.com> 25818Date: Thu Jun 15 09:45:12 2006 +0200 25819 25820 Fix some implicit function declarations. 25821 25822commit 556ddc9c41020271e92389a3900d097ec443c625 25823Author: Michel Dänzer <michel@tungstengraphics.com> 25824Date: Thu Jun 15 09:30:43 2006 +0200 25825 25826 .cvsignore -> .gitignore 25827 25828commit 4708327969a21fc171dcd80a1b506028cc8e4f90 25829Author: Dave Airlie <airlied@linux.ie> 25830Date: Thu Jun 15 07:37:03 2006 +1000 25831 25832 ati: remove Changelog from ATI driver. 25833 25834commit ecbec8379c24bd6c16ced674d5e6affcd3985e46 25835Author: Michel Daenzer <michel@daenzer.net> 25836Date: Thu Jun 8 17:29:12 2006 +0000 25837 25838 Bug #5678: Fix setting of transfer mode for AGPv3 devices (Tilman 25839 Sauerbeck). 25840 25841commit 874ff105bc63c4fc242fbabaec04fde2cb81fdd4 25842Author: Daniel Stone <daniel@fooishbar.org> 25843Date: Sat Jun 3 10:50:33 2006 +0000 25844 25845 Bug #1889: Stop the CCE before an engine reset. (Conn) 25846 25847commit f9db5595fe4d62c54b3a501b2af64b4e2b7035cd 25848Author: Michel Daenzer <michel@daenzer.net> 25849Date: Sun May 21 14:33:34 2006 +0000 25850 25851 Derive ATI_VERSION_{MAJOR,MINOR,PATCH} from package version in order to 25852 hopefully confuse bug triagers a little less. 25853 25854commit 25ba85cab8b3d831da646fef19f01769410fc958 25855Author: Alex Deucher <agd5f@yahoo.com> 25856Date: Wed May 10 02:46:37 2006 +0000 25857 25858 - remove obsolete option "OverlayOnCRT2" (bug 6488) 25859 25860commit ede75fe2b1850a4d9f03c724b7c1d1ed65e60f2f 25861Author: Michel Daenzer <michel@daenzer.net> 25862Date: Tue May 9 14:40:23 2006 +0000 25863 25864 Drop support for radeon DRM <= 1.7. The 2.4 kernel has version 1.7, and 25865 various problems up to system hangs on X server startup have been 25866 reported with that and DRI enabled. 25867 25868commit bc93a46c4fe499f23b9c0ff52e2e36b4c587f7d6 25869Author: Michel Daenzer <michel@daenzer.net> 25870Date: Mon May 8 09:29:34 2006 +0000 25871 25872 Bugzilla #4640 <https://bugs.freedesktop.org/show_bug.cgi?id=4640> Patch 25873 #5273 <https://bugs.freedesktop.org/attachment.cgi?id=5273>: 25874 Add cases for FOURCC_RGBA32 and FOURCC_RGB24. This prevents clients from 25875 receiving incorrect data sizes with these formats. (Jan Schmidt) 25876 25877commit 2b0cdd9448a24ea067b0d78f319b99c1041df2e0 25878Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 25879Date: Wed May 3 04:25:27 2006 +0000 25880 25881 Enable decoding of version 4 TMDS PLL config tables, they actually work, 25882 the problem in the comment only concerned pre-production cards that Hui 25883 tested apparently. Also set back the default value to what it was 25884 initially, fixes some more cards that don't have a proper table (Mac 25885 Mini for example) 25886 25887commit 36c04b88416905b9ae305583e8e34fefdbfafc50 25888Author: Michel Daenzer <michel@daenzer.net> 25889Date: Mon May 1 17:49:41 2006 +0000 25890 25891 Bugzilla #6755 <https://bugs.freedesktop.org/show_bug.cgi?id=6755> Patch 25892 #5536 <https://bugs.freedesktop.org/attachment.cgi?id=5536>: 25893 Change HostDataBlit interface to take dst_offset_pitch and coordinates 25894 instead of just a destination pointer, as the latter is not sufficient 25895 with tiling. Also, use HW clipping to avoid overwriting destination 25896 data outside of the specified width. 25897 Adapt to new HostDataBlit interface. This fixes corruption with 25898 UploadToScreen to the front buffer (from exaPutImage). 25899 25900commit b90d3710e67caad0686efab491c2b76ae877920c 25901Author: Michel Daenzer <michel@daenzer.net> 25902Date: Sat Apr 29 21:30:23 2006 +0000 25903 25904 Bugzilla #1992 <https://bugs.freedesktop.org/show_bug.cgi?id=1992> Patch 25905 #4988 <https://bugs.freedesktop.org/attachment.cgi?id=4988>: Fix Option 25906 "DisplayPriority" "HIGH" for R300 family. 25907 25908commit 30f3ce5c5ba205a23748def11de019ee727897f3 25909Author: Michel Daenzer <michel@daenzer.net> 25910Date: Sat Apr 29 20:34:57 2006 +0000 25911 25912 Properly de-initialize EXA driver data, so EXA can be initialized again in 25913 the next generation. 25914 25915commit d705429c230c9cc1645ae540a51dba62cae9f23e 25916Author: Michel Daenzer <michel@daenzer.net> 25917Date: Sat Apr 29 20:18:48 2006 +0000 25918 25919 Fix tests for (un)forcing HW cursor. The broken tests caused problems when 25920 switching to console with 3D clients running. 25921 25922commit 6142cb079d4a5844e4863c800baf4ae19c636af5 25923Author: Michel Daenzer <michel@daenzer.net> 25924Date: Sat Apr 29 20:12:38 2006 +0000 25925 25926 Prefer double-buffered visuals. As a consequence, the root window's visual 25927 is double-buffered by default, which is desirable in a number of 25928 situations, in particular when running a GLX compositing manager with 25929 AIGLX. 25930 25931commit eb841974d041b19461b7fb726687df8938b2e39b 25932Author: Eric Anholt <anholt@freebsd.org> 25933Date: Fri Apr 28 02:55:08 2006 +0000 25934 25935 Bug #6761: Fix font rendering in non-DRI mode by replacing RB2D_DSTCACHE_* 25936 with RB3D_DSTCACHE_*. The RB2D versions are read-only mirrors of the 25937 RB3D. This is masked when DRI is enabled because the DRM uses the right 25938 registers in its flushing. 25939 25940commit 85e2ee6de6fd347aad66785ec7119f92013cd994 25941Author: Dave Airlie <airlied@linux.ie> 25942Date: Wed Apr 26 08:38:47 2006 +0000 25943 25944 Make the ATI driver use the PCI BAR size to determine the MMIO register 25945 aperture size. Otherwise it tries to overmap things if they are up near 25946 the end of memory. 25947 25948commit 13b19edf5accb7427436d3f0b1c9a9efb4fe98fa 25949Author: Eric Anholt <anholt@freebsd.org> 25950Date: Wed Apr 26 07:26:15 2006 +0000 25951 25952 Add support for some component-alpha operations (InReverse, OutReverse) 25953 necessary for testing upcoming general component alpha acceleration 25954 assistance in EXA. Tested with rendercheck on an r100. 25955 25956commit 44a45bcb82d49349f4fdfae3e4b8f24f98d99b46 25957Author: Eric Anholt <anholt@freebsd.org> 25958Date: Tue Apr 25 23:31:56 2006 +0000 25959 25960 Use a CPPFLAGS with XORG_CFLAGS so that the exa test can actually find its 25961 headers and not fail. Also correct some nearby spelling. 25962 25963commit cc1825b503ce7e785cbb4611fd177477a1081b6e 25964Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 25965Date: Mon Apr 24 07:44:52 2006 +0000 25966 25967 Fix enabling of external VGA output on some cards with dual entity setup 25968 instead of MergedFB. This was broken due to bogus manipulations of a 25969 register between the 2 entities. This fix is a big "violent" in that it 25970 always enables the analog output regardless of what is plugged. It 25971 should work though (at the expense of some additional power 25972 consumption). A proper fix should be easy to do once Alex gets his 25973 output mapping rework in. 25974 25975commit a66145bc9d487f7e19237cf292f12750cb959f61 25976Author: Dave Airlie <airlied@linux.ie> 25977Date: Fri Apr 21 00:38:44 2006 +0000 25978 25979 Bugzilla 6672: fix interrupts on radeon since memmap changes 25980 25981commit f1f057df9aaba42986d9e20afe6f609a90318333 25982Author: Dave Airlie <airlied@linux.ie> 25983Date: Fri Apr 21 00:34:43 2006 +0000 25984 25985 Add support for backing up the PCIE GART table on VT switch. This makes 25986 suspend/resume work a lot better on PCIE cards. 25987 25988commit 31396df3b0101571bb85a4a66b67f5a8b0c992a9 25989Author: Adam Jackson <ajax@nwnk.net> 25990Date: Sat Apr 8 01:19:17 2006 +0000 25991 25992 Bump to 6.6.0 and release. This module's versioning is completely screwed, 25993 so until we split out mach64/r128/radeon please just increment the 25994 package number whenever you make significant changes. 25995 25996commit d706f494ed22afd369991e8c75a2d4f5aed15305 25997Author: Aaron Plattner <aplattner@nvidia.com> 25998Date: Fri Apr 7 23:06:02 2006 +0000 25999 26000 Add a DrawablePtr argument to the XV functions to pave the way for 26001 redirected video. 26002 26003commit f94685aec26b65f60c00a4b1bcd8f26bd80eb6e4 26004Author: Eric Anholt <anholt@freebsd.org> 26005Date: Tue Apr 4 07:42:36 2006 +0000 26006 26007 Bug #5632: Fix up ati.4 to explain that it is a wrapper driver for the 26008 other three. While here, touch up radeon.4, and add an authors list 26009 gleaned from checking copyright headers. 26010 26011commit 58c9b1d84f514755945e7a0d25fd5530af34ce95 26012Author: Eric Anholt <anholt@freebsd.org> 26013Date: Sun Apr 2 21:45:12 2006 +0000 26014 26015 Add missing .cvsignore lines. 26016 26017commit f1ce6170ef21e1f9769337c3138cd8d65c33e40a 26018Author: Alex Deucher <agd5f@yahoo.com> 26019Date: Sat Apr 1 23:02:40 2006 +0000 26020 26021 - Fix dpi when switching from clone to dualhead with MergedFB. 26022 - Add ConstantDPI option to force a particlar dpi across mode changes Both 26023 based on Thomas Winischhofer's sis code. 26024 26025commit 607f18cfbfc2a7b5509f8b567bf510fce31b361e 26026Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26027Date: Mon Mar 27 06:12:57 2006 +0000 26028 26029 Fix a regression of my latest fix that broke color tiling on some setups. 26030 RADEONRestoreMemMapRegisters() must not clear CRTC_OFFSET_CNTL and 26031 friends unconditionally but only when there is a map change or it will 26032 clear them after the DRI updates the AGP location, thus turning tiling 26033 back off on the CRTC. Also remove a duplicate call to 26034 RADEONInitCommonRegisters(). 26035 26036commit bf1c4471137f10b7de9d79519ca2240058997e18 26037Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 26038Date: Thu Mar 23 18:04:17 2006 +0000 26039 26040 commit the actual changes instead of just the Changelog... 26041 26042commit aa5fff9bae4818bbf99e8d9caa8f2ce9905294ca 26043Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 26044Date: Thu Mar 23 17:59:25 2006 +0000 26045 26046 Use programmable tap coefficients for (hopefully) better video image 26047 quality when downscaling. Remove unnecessary code programming the tap 26048 coefficients twice. Disable some code which is supposed to improve 26049 filtering quality but causes very visible image quality problems at 26050 least with rv250 when upscaling. Fix potentially broken FOURCC_RGB16T 26051 format (untested if it was broken, untested if it works now...). Change 26052 some magic numbers to (new) macro defines. 26053 26054commit 17c52ff15ebe2cf220d5988ee13b599c41799886 26055Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26056Date: Thu Mar 23 01:37:15 2006 +0000 26057 26058 Call RADEONRestoreCommonRegisters() before RADEONRestoreCrtc2Registers() 26059 instead of after. I can't find any good reason why we did it backward 26060 until now and it definitely causes problems as the workaround for VT 26061 switch in RADEONRestoreCommonRegisters() will break output to the 26062 secondary head in various situations. Also does an unrelated minor 26063 update to a comment. 26064 26065commit 673e4d7501861066fdd0874919a4524b820b944b 26066Author: Kristian Høgsberg <krh@redhat.com> 26067Date: Thu Mar 23 00:06:56 2006 +0000 26068 26069 Add atipciids.h. 26070 26071commit 5dcb4152dca6ac1fe9342f44eb48c0198b15772e 26072Author: Kristian Høgsberg <krh@redhat.com> 26073Date: Wed Mar 22 22:30:14 2006 +0000 26074 26075 Convert use of xf86fopen() and other xf86 wrapped libc symbols to use libc 26076 symbols directly. The xf86* versions aren't supposed to be used 26077 directly. 26078 Drop libc wrapper; don't include xf86_ansic.h and add includes now missing. 26079 26080commit 21acdf371e3cf913c5affbd3e86641cfb115311c 26081Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26082Date: Sat Mar 18 00:08:24 2006 +0000 26083 26084 Clear the offsets when updating the memory map instead of when enabling the 26085 CRTCs and try to make that code more reliable. Doesn't disable the 26086 CRTCs beforehand. 26087 26088commit 323ecb92e40d71c5ef994b41b6d8dedba6dd6203 26089Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 26090Date: Fri Mar 17 03:00:53 2006 +0000 26091 26092 Fix various small cosmetic issues. Change a driver message, get the order 26093 right for requesting drm versions, replace the use of some numbers with 26094 the respective macro defines in radeon_video.c, and add some more macro 26095 defines. None of that really matters. 26096 26097commit dc552edf191e14249421f8b27146874f16b80c1a 26098Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26099Date: Thu Mar 16 21:53:58 2006 +0000 26100 26101 Add missing return statement (the driver would not start randomly), add 26102 RV410 to the list of new generation cards for the memory map setup and 26103 fix a bogus message. 26104 26105commit 901409f47cc990721c1648b1c08e23450e6a4452 26106Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26107Date: Thu Mar 16 04:32:22 2006 +0000 26108 26109 Fix incorrect CRTC2_OFFSET values when using old-style dual head (not 26110 MergedFB). Fixes Xinerama for me. 26111 26112commit 4d6656062129da0489eb4bc898871379ba891d8f 26113Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26114Date: Thu Mar 16 03:09:34 2006 +0000 26115 26116 Add various workarounds that seem to fix some remaning lockup scenarios I'm 26117 experiencing with the driver when setting the memory map. Some of the 26118 magic delays are a bit dodgy but they seem to work, I suppose I can't 26119 do better now without help from ATI. Also removed some really too noisy 26120 debug messages. 26121 26122commit 36799e3ea7ddde5fce528e8bc17ea942933a9111 26123Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26124Date: Wed Mar 15 04:03:37 2006 +0000 26125 26126 Fix various issues with DRI & server recycle by putting DRICloseSreen in 26127 the normal CloseScreen callback chain at a spot that matches the call 26128 to DRIFinishScreenInit in order to unwrap at the right time. 26129 26130commit 50923b6a9325592edc8b02b17646f96b9b8c4a01 26131Author: Luc Verhaegen <libv@skynet.be> 26132Date: Wed Mar 15 00:46:33 2006 +0000 26133 26134 Add some autoconf to keep this driver from building EXA when a slightly 26135 older server with older exa is installed. 26136 The way in which this problem was treated by the actually related 26137 developers explains a lot about this driver. 26138 26139commit 74f4ed4275866475b3c182fae962df2fc0946b85 26140Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26141Date: Mon Mar 13 02:57:17 2006 +0000 26142 26143 Fix typo in the function setting the color space transformation factors for 26144 old radeon 26145 26146commit 65625c6c2053792cd14b54c904bf46f360ef1565 26147Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26148Date: Mon Mar 13 00:09:44 2006 +0000 26149 26150 Earliest fix introduced a bug due to another chicken & egg problem in card 26151 detection. Now split PreInitConfig into PreInitChipType and PreInitVRAM 26152 so that PreInitDRI can be done just in between. Restores proper AGPx4 26153 functionality (when enabled) 26154 26155commit 3a290115af41b0ecb94eb201d7e7abb42ebbfec6 26156Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26157Date: Sun Mar 12 06:53:27 2006 +0000 26158 26159 Stop using xf86PciInfo.h, instead use a local copy of the PCI IDs we need 26160 in atipciids.h so we can update the ATI driver independently of the 26161 server when new chips are added 26162 26163commit d6b8f0a5eeb332563ec3ec7082f03be1797d26d8 26164Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26165Date: Sat Mar 11 04:19:47 2006 +0000 26166 26167 Add #ifdef XF86DRI guards around references to info->allowPageFlip and 26168 info->depthMoves, since they are only defined if building with DRI. 26169 26170commit 69fe90bc501df9bd9e74b894bfca79a4025725bd 26171Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 26172Date: Fri Mar 10 13:00:49 2006 +0000 26173 26174 Use "FBTexPercent" option for XAA too (only for increasing memory reserved 26175 for textures). 26176 26177commit 8033d7423267fc1b3725df6a077dd910b586daaf 26178Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26179Date: Thu Mar 9 23:26:27 2006 +0000 26180 26181 More memory map fixes !!! This time, we add a way to get the DRM kernel 26182 module version early, in order to properly limit the framebuffer size 26183 when a module that doesn't support the new map is loaded. That will 26184 improve backward compatibility. I also made the minimum required kernel 26185 module version 1.3 which hopefully shouldn't be a problem for anybody, 26186 and cleaned up the ordering in which we test for DRM version to enable 26187 features 26188 26189commit c637939d3c4ce5cdddbc5200d5ffbca4c3927050 26190Author: Michel Daenzer <michel@daenzer.net> 26191Date: Thu Mar 9 15:41:16 2006 +0000 26192 26193 RN50: Skip modes that exceed memory bandwidth. 26194 Bugzilla #5766 <https://bugs.freedesktop.org/show_bug.cgi?id=5766> Patch 26195 #4636 <https://bugs.freedesktop.org/attachment.cgi?id=4636> 26196 - Acknowledge that RN50 only has one CRTC, and use this to distinguish it 26197 from RV100. 26198 - Fix detection of RN50 memory type and bus width. 26199 - Model RN50 memory bandwidth limits by capping the pixel clock range based 26200 on memory clock, bpp and memory bus width. (ATI Technologies Inc.) 26201 26202commit 890b95a3ddad7634f3aea635cb236b221e902acc 26203Author: Eric Anholt <anholt@freebsd.org> 26204Date: Thu Mar 9 06:06:25 2006 +0000 26205 26206 Update drivers that support EXA for the new EXA ABI. This consists of 26207 moving all the accel and card members into the driver ptr, filling in 26208 the exa_major/ exa_minor fields, and always using LoadSubModule so we 26209 can check the module version up front. Only tested on ATI. 26210 26211commit e19ccfa18fee046c54d7a9401e4e0098b2300598 26212Author: Michel Daenzer <michel@daenzer.net> 26213Date: Fri Mar 3 16:41:41 2006 +0000 26214 26215 Add Option "FBTexPercent" to override the amount of video RAM reserved for 26216 OpenGL textures with EXA. 26217 26218commit ca00035eccb6e3495b27f8e64f2d016d8d2a9ce2 26219Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26220Date: Thu Mar 2 02:05:17 2006 +0000 26221 26222 Fix the fix ... I didn't interpret PciInfo->size properly and forgot that 26223 the function works in Kb not bytes... Ooops. 26224 26225commit b2145aea36bb035bff048366c607b967d70fff49 26226Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26227Date: Wed Mar 1 21:35:14 2006 +0000 26228 26229 Extend the alignement workaround to post-rv280 chips as well (thanks Hui) 26230 and also limit the CPU accessible memory to the size of the PCI BAR 26231 size (yeah, it was passed to the driver, I just got blind for a while 26232 it seems) 26233 26234commit 163581a82fe9395672c31b41cd67aa44158f5201 26235Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26236Date: Tue Feb 28 23:34:03 2006 +0000 26237 26238 Workaround for rv280 bug which needs the framebuffer memory mapping to be 26239 aligned to its own size. Also add back limit of 128Mb of mapped fb 26240 space for now until I fully sort out what's up with dual function 26241 cards. I would really need to access the PCI BAR sizes here but that 26242 information seem to be missing from the PCI info passed to the driver. 26243 26244commit 5ff8fb2bea1fe7d14db53c19ff3a3a56668ec417 26245Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 26246Date: Tue Feb 28 00:52:53 2006 +0000 26247 26248 Add the newly added (and tons of previously forgotten ones) pci ids to the 26249 ati wrapper too (probably not required for it to work as everything 26250 should default to radeon anyway there). 26251 26252commit 1574b41a0f894f9b6e30d3ee94e596695a4c64f0 26253Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26254Date: Sun Feb 26 10:01:32 2006 +0000 26255 26256 Fix page flipping with XAA. The new code ended up calling ShadowFBInit() 26257 before XAAInit(), which breaks ShadowFB. 26258 26259commit 1f43a584a684af8078631425886cce236500e0e6 26260Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 26261Date: Sat Feb 25 01:57:05 2006 +0000 26262 26263 Add pci ids known to exist (see #4284 for instance). There are still 26264 entries which probably don't really exist (cancelled cards and such), 26265 leave them as-is. Fix the name of some entries, mostly based on the 26266 .inf file of the newest catalyst driver. Use own family id for rv410 26267 and rs400, though there is no different code (yet?). 26268 26269commit 5c141bb15d1163e04c012a0cdf0699d534f0be37 26270Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26271Date: Thu Feb 16 23:27:44 2006 +0000 26272 26273 Memory map and misc fixes. This completely reworks the way the driver 26274 detects available & accessible video memory and initializes the card 26275 internal memory map. Along the way, I re-ordered some of the stuffs in 26276 pre-init in order to remove some dodgy codepath, fixed various init, 26277 exit and vt switch issues, and overall made the driver more robust. The 26278 full memory map fix requires an updated DRM. The X driver no longer 26279 tries to position the AGP aperture, this becomes entirely DRM 26280 responsibility. 26281 26282commit 8c6e5a6eb50d371fd4ec700c26aa628cb5f45731 26283Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26284Date: Wed Feb 15 23:36:55 2006 +0000 26285 26286 The engine setup would trigger a bogus line write before the proper 26287 addresses are setup, thus causing the card to try to bus master over 26288 the system bus to some stale location (usually 0 but could depend on 26289 whatever was used before X). Remove that useless init bit, and always 26290 setup some good enough engine source & destination pointers in case we 26291 have some other spurrious engine activity going on. 26292 26293commit d4ec0223029d8fe03a836a5a6e9a7cc8897b4c90 26294Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 26295Date: Mon Feb 13 05:30:13 2006 +0000 26296 26297 Fix EXA acceleration of textures with byteswap or tiling From John Clemens 26298 <john@deater.net> 26299 26300commit 79c60ebbc34aaf0ac0b0723898d5b63015547178 26301Author: Eric Anholt <anholt@freebsd.org> 26302Date: Tue Feb 7 04:39:59 2006 +0000 26303 26304 Fix a copy'n'paste-o in the EXA render implementation that resulted in 26305 [ax]8b8g8r8 pict formats not being supported on r200, but no other 26306 issues. 26307 26308commit 8d74f3d49e73626f5d658ba6586a7c1487dae141 26309Author: Daniel Stone <daniel@fooishbar.org> 26310Date: Thu Jan 19 14:20:37 2006 +0000 26311 26312 Bug #5656: Set all Radeon XPRESS 200 R[CS]4xx-based chips to have IsIGP = 26313 1, fixing modesetting issues. 26314 26315commit e490f4b86e6bba70f7c884fca3cf41d159323689 26316Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26317Date: Fri Jan 13 01:37:43 2006 +0000 26318 26319 XF86_VERSION_CURRENT -> XORG_VERSION_CURRENT 26320 26321commit ac296d9137b02753bc79a9930d168855adea9c95 26322Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26323Date: Fri Jan 13 01:02:01 2006 +0000 26324 26325 When failing to parse MetaModes, reset CRT2pScrn to NULL after free'ing it 26326 so we don't crash when trying to use it later (such as in 26327 RADEONFreeRec). 26328 26329commit dd98ce78407c9eddd7adaad1d73d6910e6949b87 26330Author: Adam Jackson <ajax@nwnk.net> 26331Date: Sun Jan 8 19:02:16 2006 +0000 26332 26333 Bug #5523: Make radeon compile without USE_XAA. (Tilman Sauerbeck) 26334 26335commit a99429b2bc6407ff2a094e6ee30ce2032f734966 26336Author: Kevin E Martin <kem@kem.org> 26337Date: Wed Dec 21 02:29:59 2005 +0000 26338 26339 Update package version for X11R7 release. 26340 26341commit 96e225b950c74e64f3f788be8a8489c5075ea55a 26342Author: Adam Jackson <ajax@nwnk.net> 26343Date: Mon Dec 19 16:25:52 2005 +0000 26344 26345 Stub COPYING files 26346 26347commit 6bea785ea0bdf01673a6ddc45765ae7372fdd371 26348Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26349Date: Mon Dec 19 09:07:50 2005 +0000 26350 26351 Typo fixes, mailing list & url updates, and other changes to prepare for 26352 X11R6.9 & 7.0 releases. 26353 26354commit 07d5cf8fc37eb72dbb344081ff07d59dbf568e9c 26355Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26356Date: Thu Dec 15 23:15:37 2005 +0000 26357 26358 Bugzilla #5238 <https://bugs.freedesktop.org/show_bug.cgi?id=5238> Patch 26359 #3978 <https://bugs.freedesktop.org/attachment.cgi?id=3978> r128 man 26360 page dualhead note update (Alex Deucher) 26361 26362commit 6db5cbfbf6d405d9871892d1255dd7442e4539e9 26363Author: Kevin E Martin <kem@kem.org> 26364Date: Thu Dec 15 00:21:32 2005 +0000 26365 26366 Update package version number for final X11R7 release candidate. Bump 26367 driver version number. 26368 26369commit 726fd3a81dfbd1b0c8e4ede104a48523daecab92 26370Author: Adam Jackson <ajax@nwnk.net> 26371Date: Wed Dec 14 19:41:28 2005 +0000 26372 26373 Bug #1760: Fix ati wrapper multihead to be dlloader friendly. 26374 26375commit 328a2d74dabe32ce35e879799412666feedb0eef 26376Author: Adam Jackson <ajax@nwnk.net> 26377Date: Fri Dec 9 18:25:02 2005 +0000 26378 26379 Bug #3869: Don\'t walk off the end of an array in connector detection. 26380 (Eric Anholt) 26381 26382commit 03961a05438e916b6725f4e547f23a84ead13c5e 26383Author: Adam Jackson <ajax@nwnk.net> 26384Date: Fri Dec 9 18:15:31 2005 +0000 26385 26386 Bug #1106: Fix memory allocation for Render. (Michel Daenzer) 26387 26388commit 091e5c584124f9b849c0f46802b2f0d8b0cc2388 26389Author: Kevin E Martin <kem@kem.org> 26390Date: Thu Dec 8 17:55:18 2005 +0000 26391 26392 Add configure options to allow hard-coded paths to be changed. 26393 26394commit e38e2df15c9a4b08dee6386a225e98827bb35aad 26395Author: Kevin E Martin <kem@kem.org> 26396Date: Thu Dec 8 17:54:40 2005 +0000 26397 26398 Allow hard-coded paths to be configurable. 26399 26400commit b8203604bf9bd3813668313a1c8c17040b9d93ce 26401Author: Kevin E Martin <kem@kem.org> 26402Date: Tue Dec 6 22:48:33 2005 +0000 26403 26404 Change *man_SOURCES ==> *man_PRE to fix autotools warnings. 26405 26406commit a86920e812bfd40ec9facfc234220e83066dd263 26407Author: Kevin E Martin <kem@kem.org> 26408Date: Sat Dec 3 05:49:32 2005 +0000 26409 26410 Update package version number for X11R7 RC3 release. 26411 26412commit 010adc8988846e6f76f111a9fd32eca021d12fbc 26413Author: Kevin E Martin <kem@kem.org> 26414Date: Sat Dec 3 05:45:17 2005 +0000 26415 26416 Bump version numbers for X11R6.9/X11R7 RC3. 26417 26418commit a5efa6724e7fd8318e262060b084630632dfcb44 26419Author: Kevin E Martin <kem@kem.org> 26420Date: Fri Dec 2 02:16:07 2005 +0000 26421 26422 Remove extraneous AC_MSG_RESULT. 26423 26424commit feeef3cba1f4736a96daf3002ec902ced2874198 26425Author: Adam Jackson <ajax@nwnk.net> 26426Date: Wed Nov 30 18:51:22 2005 +0000 26427 26428 Bump libdrm dep to 2.0. 26429 26430commit 7ca2b3fa30fedd07951f7c6e73ed4d2a6f6606ea 26431Author: Adam Jackson <ajax@nwnk.net> 26432Date: Tue Nov 29 23:29:58 2005 +0000 26433 26434 Only build dlloader modules by default. 26435 26436commit e167e5151e32d54919e96cbfdcf932e16b1cab2a 26437Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26438Date: Mon Nov 28 22:04:08 2005 +0000 26439 26440 Change *mandir targets to use new *_MAN_DIR variables set by xorg-macros.m4 26441 update to fix bug #5167 (Linux prefers *.1x man pages in man1 subdir) 26442 26443commit 0015b1a007bed61aaf389f8f927e4ae0568d7da7 26444Author: Eric Anholt <anholt@freebsd.org> 26445Date: Mon Nov 21 10:49:09 2005 +0000 26446 26447 Add .cvsignores for drivers. 26448 26449commit 00b8f6ccfbb5163c47e9def0e4ecc6ac7be7e8af 26450Author: Eric Anholt <anholt@freebsd.org> 26451Date: Fri Nov 11 11:05:42 2005 +0000 26452 26453 Bug #1028: Don't (incorrectly) mark the 24/8 z/stencil case as slow, which 26454 resulted in whining at runtime about the visual not being supported. 26455 26456commit 33326847ccc9cf4cd6a23a257e5392e348e6dd22 26457Author: Kevin E Martin <kem@kem.org> 26458Date: Wed Nov 9 21:31:20 2005 +0000 26459 26460 Bump version numbers for X11R6.9/X11R7 RC2. 26461 26462commit 68e11246bd6f6a641807230834dce4e7fd6911af 26463Author: Kevin E Martin <kem@kem.org> 26464Date: Wed Nov 9 21:15:14 2005 +0000 26465 26466 Update package version number for X11R7 RC2 release. 26467 26468commit 1582d3b34643babcbb794d2df8d7c9a755d569c9 26469Author: Matthias Hopf <mhopf@suse.de> 26470Date: Wed Nov 9 17:42:57 2005 +0000 26471 26472 Bug #4998: Add ';' and ',' as valid separators for MetaModes as well. 26473 26474commit 2d22d78d157e07e6ec6a3e2d03fc17f102f6e00e 26475Author: Kevin E Martin <kem@kem.org> 26476Date: Wed Nov 9 07:30:03 2005 +0000 26477 26478 Fix DoLoadableServer NO build. 26479 26480commit 999f67763a5ea9fef7b7b576262f0da079efcfee 26481Author: Michel Daenzer <michel@daenzer.net> 26482Date: Tue Nov 8 16:30:48 2005 +0000 26483 26484 bugzilla #988 (https://bugs.freedesktop.org/show_bug.cgi?id=988) Fix typo 26485 which may or may not have had a negative impact on stability with R300 26486 class cards. 26487 26488commit 5aa401efa0b8d81da418eafeb857142a33f85da9 26489Author: Michel Daenzer <michel@daenzer.net> 26490Date: Tue Nov 8 14:05:00 2005 +0000 26491 26492 bugzilla #4494 (https://bugs.freedesktop.org/show_bug.cgi?id=4494) Call 26493 RADEONDGAInit() even if EXA is enabled; the problematic DGA 26494 acceleration hooks are already disabled in that case. 26495 This fixes DGA issues such as DGA mouse (typically used by games in 26496 fullscreen mode) not working. 26497 Fix suggested by Eric Anholt. 26498 26499commit 345180131a7b2752773dc2a3ee1221ad1e05c815 26500Author: Michel Daenzer <michel@daenzer.net> 26501Date: Tue Nov 8 13:35:23 2005 +0000 26502 26503 bugzilla #4916 (https://bugs.freedesktop.org/show_bug.cgi?id=4916) Fix 26504 segfault with MergedFB and EXA enabled and HW cursor temporarily 26505 disabled. 26506 26507commit 23911421ffa7bdd2e191a72c41bf1f26c92c1a67 26508Author: Michel Daenzer <michel@daenzer.net> 26509Date: Tue Nov 8 09:44:41 2005 +0000 26510 26511 bugzilla #4951 (https://bugs.freedesktop.org/show_bug.cgi?id=4951) 26512 attachment #3689 (http://bugs.freedesktop.org/attachment.cgi?id=3689) 26513 Fix HW cursor getting silently disabled with EXA. 26514 26515commit e1259c0117a190cf4c6e7be07d5aa83e03afb3a0 26516Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 26517Date: Mon Nov 7 19:28:40 2005 +0000 26518 26519 Replaced the variadic macros(gcc) by macros according to C99 standard 26520 26521commit 67bdaae04ab37f38a6d0c8dc4b6328c3ba3b3393 26522Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26523Date: Sun Nov 6 07:41:29 2005 +0000 26524 26525 Sun bug #6325220: Xorg dumps core in dual monitor configuration with radeon 26526 driver if CRT2Position not specified in xorg.conf 26527 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6325220> 26528 (Stuart Krietman - Sun Microsystems) 26529 26530commit 733ebf6e4a7c7b2f1a861077d43da2a853e4c0d7 26531Author: Kevin E Martin <kem@kem.org> 26532Date: Tue Nov 1 15:08:52 2005 +0000 26533 26534 Update pkgcheck depedencies to work with separate build roots. 26535 26536commit dd0eb76337bfdaec8f456380a6ced9e0487e190f 26537Author: Adam Jackson <ajax@nwnk.net> 26538Date: Fri Oct 21 18:40:18 2005 +0000 26539 26540 Bug #4829: Remove some useless DRI symbol checks. 26541 26542commit e6c1859e3a7c84b882e0c30d0783255ed1258d9e 26543Author: Donnie Berkholz <spyderous@gentoo.org> 26544Date: Thu Oct 20 23:01:58 2005 +0000 26545 26546 Bug #4831 <https://bugs.freedesktop.org/show_bug.cgi?id=4831>. Check for 26547 x86_64 in addition to amd64 in host_cpu. 26548 26549commit 33460fb393c9591babcc43b30a73eef3971765c2 26550Author: Kevin E Martin <kem@kem.org> 26551Date: Wed Oct 19 02:48:02 2005 +0000 26552 26553 Update package version number for RC1 release. 26554 26555commit 77d5c3549db4f6a7ff9f969f885d7a4e58ab2b16 26556Author: Kevin E Martin <kem@kem.org> 26557Date: Tue Oct 18 15:46:37 2005 +0000 26558 26559 Do the following to make the drivers pass distcheck: 26560 - Only define EXTRA_DIST when there are actually man pages to be processed 26561 - Add missing man pages to keyboard and ati drivers EXTRA_DIST 26562 26563commit 35e717078c840414b0d6a0e37b3ffed0a7b102aa 26564Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26565Date: Tue Oct 18 00:01:52 2005 +0000 26566 26567 Use @DRIVER_MAN_SUFFIX@ instead of $(DRIVER_MAN_SUFFIX) in macro 26568 substitutions to work better with BSD make 26569 26570commit ff6e5b0412fe347575992c83bb7cb47b6f96f390 26571Author: Adam Jackson <ajax@nwnk.net> 26572Date: Mon Oct 17 22:57:32 2005 +0000 26573 26574 More 1.7 braindamage: define EXTRA_DIST in terms of @DRIVER_NAME@ instead 26575 of indirectly 26576 26577commit a2e7c55850d383e3af948d445b1645d8a47811cf 26578Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26579Date: Mon Oct 17 00:09:01 2005 +0000 26580 26581 Use sed & cpp to substitute variables in driver man pages 26582 26583commit 524fb1440760a14c53fb3f238aad6c3cdb7e6a08 26584Author: Dave Airlie <airlied@linux.ie> 26585Date: Sun Oct 9 10:36:50 2005 +0000 26586 26587 Secure the Radeon PCIE GART tables in framebuffer memory by not allowing 26588 the clients to get access to a secure area of framebuffer. 26589 26590commit e526211b86ec5343002978debfcf78f74a10c69e 26591Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 26592Date: Tue Oct 4 20:01:03 2005 +0000 26593 26594 Check in generated README files 26595 26596commit 431d0e144ae4c294e1f6110fc3db497ce9bd9cfd 26597Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 26598Date: Mon Oct 3 21:22:07 2005 +0000 26599 26600 - For all drivers that have a <driver>.sgml file, add code in their build 26601 system to build the README file at make dist time 26602 - in util/macros/xorg-macros.m4, add a new XORG_CHECK_LINUXDOC macro that 26603 will check if the required tools and files exist, and if so set a 26604 conditional. 26605 - util/modular/symlink.sh 26606 - Link all the <driver>.sgml to xf86-video-<driver>/README.sgml 26607 - Add all the README.<driver> to the list of excluded files 26608 - xc/programs/Xserver/hw/xfree86/doc/sgml/SiS.sgml: Various changes to make 26609 it spew less warnings when the text file is built. 26610 26611commit 9147c20e5ff662ac511a19e7623b406dffee2242 26612Author: Adam Jackson <ajax@nwnk.net> 26613Date: Mon Oct 3 06:14:58 2005 +0000 26614 26615 Bug #4001: Don't crash when PanelSize is unset. (Alan Coopersmith) 26616 26617commit 7e2090e3ebf0bd699d66d336353c097d658a86ae 26618Author: Matthieu Herrb <matthieu.herrb@laas.fr> 26619Date: Sat Oct 1 20:42:06 2005 +0000 26620 26621 missed one missing return value. 26622 26623commit a843e4281134de1545d538cb8970179b1d6977f6 26624Author: Matthieu Herrb <matthieu.herrb@laas.fr> 26625Date: Sat Oct 1 17:09:48 2005 +0000 26626 26627 add missing return value on success in RADEONSetupMemXAA(). 26628 26629commit 98f7d043f894c73fecaf14440c1a24b6c2fc4b06 26630Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 26631Date: Wed Sep 28 15:10:38 2005 +0000 26632 26633 Make ati driver distcheck 26634 radeon_commonfuncs.c \ 26635 radeon_exa_render.c \ 26636 radeon_exa_funcs.c \ 26637 26638commit e7b1af5d1677e53820dd1c0926cc8b5a4284901e 26639Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 26640Date: Tue Sep 27 23:55:15 2005 +0000 26641 26642 Move the RADEONEnablePageFlip/RADEONChangeSurfaces calls back to the end of 26643 the function (the code above will cause pixmap cache to be freed and 26644 copied around, and these functions must be called after that otherwise 26645 pixmap damage will occur). May be related to #4440, #4456 (unclear if 26646 these setups involve 3d apps) 26647 26648commit bc3c26fe365347c427e04385c1df95a5eec14f28 26649Author: Eric Anholt <anholt@freebsd.org> 26650Date: Tue Sep 27 23:34:11 2005 +0000 26651 26652 Fix Radeon MMIO Render acceleration with EXA by writing out floating- point 26653 coordinates rather than integers. 26654 26655commit 58dfb95c446b73fbc34784f0894cf89152dab043 26656Author: Adam Jackson <ajax@nwnk.net> 26657Date: Sat Sep 24 18:45:45 2005 +0000 26658 26659 Disable the {Open,Close}FullScreen DRI protocol. Remove empty FullScreen 26660 stubs from drivers, comment the non-empty ones. 26661 26662commit 65b5f0017e29903565aff0b7527d955ed016c9a6 26663Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26664Date: Tue Sep 20 16:53:34 2005 +0000 26665 26666 Only include radeon_dri.h & radeon_sarea.h if XF86DRI is defined. 26667 Include "compiler.h" to define __inline__ on non-gcc compilers 26668 26669commit 0b7441b19a1a5df91aca6b1a02944afb636bc523 26670Author: Eric Anholt <anholt@freebsd.org> 26671Date: Sun Sep 18 02:32:23 2005 +0000 26672 26673 Break EXA ABI while we still can. Add coordinates to the UploadToScreen 26674 hook so we can upload a subset of a pixmap, and convert the current 26675 drivers to respect that. Use this support to directly UploadToScreen in 26676 exaGlyphs, providing a 47.4% +/-2.4% decrease in wall time for ls -lR 26677 programs/Xserver in an antialiased gnome-terminal on an M6 (n=3, caches 26678 hot). I would have bumped major version, only I can't tell what the 26679 EXA_VERSION_* is supposed to be doing as opposed to the module version. 26680 26681commit 64e7db2698ee48af92594c16d72e33c16372c9d7 26682Author: Eric Anholt <anholt@freebsd.org> 26683Date: Sat Sep 17 20:02:01 2005 +0000 26684 26685 - Don't try to upload 0 byte-per-pixel (PICT_a1) data using 26686 RADEONHostDataBlit. 26687 - Disable the shortcut for switching from 3d to 3d in radeon_exa.c. It 26688 appears that we do need the cache flush here, thought it's not clear 26689 why. Disable the 2d to 2d shortcut while here, since I'm unsure of what 26690 we're doing. Exposed by the following bit: 26691 - Bug #4485: Add a new routine, exaGlyphs, to handle font drawing. Glyphs 26692 were being accumulated in from non-migratable scratch pixmaps, causing 26693 the destination pixmap to move towards screen but the migration 26694 necessary for source never to happen, leading to abysmal performance. 26695 Instead, copy the scratch glyph data into a real pixmap first, then 26696 composite from that into the destination, allowing for migration. time 26697 ls -lR from programs/Xserver showed 26.9% (+/- 6.3%) decrease in wall 26698 time (n=3). 26699 - Create exaDrawableUse* wrapping exaPixmapUse*, but which are aware of 26700 windows needing backing store. Makes migration code prettier, and 26701 ensures that composited windows will be migrated as normal when we turn 26702 off cw for EXA. (issue brought up by keithp) 26703 26704commit 3c71198a088e71a105e274f176c8ceb503615d46 26705Author: Dave Airlie <airlied@linux.ie> 26706Date: Sat Sep 17 07:47:51 2005 +0000 26707 26708 Switch radeon driver to using a card type rather than having a IsPCI flag, 26709 allow for PCI/AGP/PCIE cards. Set PCI gart allocation to only happen on 26710 PCIE cards. Add Radeon PCIE card detection using capability bits. 26711 26712commit 10cf3523f236a017a1ca7e5ee57e02176b924059 26713Author: Eric Anholt <anholt@freebsd.org> 26714Date: Wed Sep 14 22:06:41 2005 +0000 26715 26716 Bug #2573: Remove the disabling of automatic cache flushing on non-r300 26717 cards. It has been implicated in hanging at least r300s and some r100s, 26718 and seemed to be the cause of some bad rendering for me. 26719 26720commit d68894a463a3bb08c97e91ff6fc4b65b29183ff6 26721Author: Eric Anholt <anholt@freebsd.org> 26722Date: Tue Sep 13 22:04:12 2005 +0000 26723 26724 Remove the hardcoded USE_EXA/USE_XAA defines now that they've been 26725 hardcoded in configure.ac (config.h). 26726 26727commit baa9b5b6e19cf9794bf810e3c582bf1d7f808663 26728Author: Dave Airlie <airlied@linux.ie> 26729Date: Mon Sep 12 09:27:07 2005 +0000 26730 26731 Fix Radeon EXA for when RENDER isn't defined 26732 26733commit 8c89610389197dc94d3bceaf44d9d2ea91848f1b 26734Author: Dave Airlie <airlied@linux.ie> 26735Date: Mon Sep 12 09:17:42 2005 +0000 26736 26737 Move GART offset/size outside XAA 26738 26739commit eba44d9747adde8385f2a242dee5a8e73ce5a205 26740Author: Daniel Stone <daniel@fooishbar.org> 26741Date: Mon Sep 12 09:04:36 2005 +0000 26742 26743 Define USE_XAA and USE_EXA for the time being. 26744 26745commit a255e7b0984890daad98b74ace0c572965fe61f5 26746Author: Eric Anholt <anholt@freebsd.org> 26747Date: Sun Sep 11 21:00:44 2005 +0000 26748 26749 Attempt to fix up the modular pieces for EXA support. (untested) 26750 26751commit 3e0d9c945a6a71cc476d27341ef18618529a91a8 26752Author: Eric Anholt <anholt@freebsd.org> 26753Date: Sun Sep 11 20:58:53 2005 +0000 26754 26755 Add support for EXA to the radeon driver. Building EXA and XAA support is 26756 controlled at compile time, plus the runtime option of Option 26757 "AccelMethod" "EXA" or "XAA". The XAA support appears to remain as 26758 before, while the EXA pieces need just a little more polishing. Notable 26759 features: 26760 - Render acceleration working on Radeon 100 and 200-series with DRI on. 26761 - DRI works with EXA Notable issues: 26762 - DGA disabled in the EXA case. 26763 - Backbuffer moves disabled in the EXA case. 26764 - No textured XVideo. 26765 - MMIO render acceleration is close but still has some issues. 26766 - Memory pressure while using Composite is really troublesome with DRI on. 26767 This patch is based on an initial patch by Zack Rusin, with significant 26768 work by Benjamin Herrenschmidt and myself. 26769 26770commit fd62082b68ac3aadd8ffc441352d75d88334904e 26771Author: Dave Airlie <airlied@linux.ie> 26772Date: Sun Sep 11 08:51:38 2005 +0000 26773 26774 Add support for allocating PCI GART table in framebuffer memory. This is 26775 needed to support PCIE Radeons using a new DRM. 26776 26777commit cb63f8d9c0563fb0eff28e2be6d4adf5666540d2 26778Author: Dave Airlie <airlied@linux.ie> 26779Date: Tue Sep 6 10:03:19 2005 +0000 26780 26781 Handle VIP timeouts more gracefully -- impose a hard limit of 10ms on 26782 waiting for VIP_BUSY so we don't lock up hard, spinning. 26783 26784commit dbd83cf18e1bf5e7e31e1ac150227bd27e8b1739 26785Author: Adam Jackson <ajax@nwnk.net> 26786Date: Sun Aug 28 20:51:14 2005 +0000 26787 26788 Bug #4148: Bump required DRM version for r300 to 1.17. 26789 26790commit ce0589ff3b9fcf8c6bab64da4dba40ba540a6a03 26791Author: Adam Jackson <ajax@nwnk.net> 26792Date: Sun Aug 28 18:10:34 2005 +0000 26793 26794 Bug #1109: Fix VGA init on Rage Mobility 7500 (Marc Le France) 26795 26796commit 250c8732a54924eeb459880f2905f8e59d038e60 26797Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 26798Date: Sun Aug 28 18:00:23 2005 +0000 26799 26800 Modified the licens type from GPL to a X/MIT one 26801 26802commit 94c0196652c3cfa0558f61b55f5df9bf50f9ac49 26803Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 26804Date: Fri Aug 26 20:04:01 2005 +0000 26805 26806 #include "extensions/dpms.h" -> <X11/extensions/dpms.h> 26807 26808commit 107e42cabe1cc41fe86b0218c3229b5f34b2df84 26809Author: Daniel Stone <daniel@fooishbar.org> 26810Date: Fri Aug 26 07:00:24 2005 +0000 26811 26812 Fix 'ugly' warning in r128. (Branden Robinson, Debian #069) 26813 26814commit d3f5911c85ccfaaf50c05ce80d56dac13fd15701 26815Author: Daniel Stone <daniel@fooishbar.org> 26816Date: Fri Aug 26 05:30:52 2005 +0000 26817 26818 Only use vgahw symbols if we're building vgahw for this architecture. TODO: 26819 Fix this for the modular tree (maybe an unconditional yes?). Define 26820 that we're using a DPMS server and include DPMS headers. (Debian #024b) 26821 26822commit 879213455740d2d653d38c712275f07610e4dc9c 26823Author: Alex Deucher <agd5f@yahoo.com> 26824Date: Fri Aug 26 02:11:24 2005 +0000 26825 26826 - fix pitch increment (Aapo Tahkola, Roland Scheidegger) 26827 26828commit 6519cc71e4387ef979de90be517145611511ff1a 26829Author: Daniel Stone <daniel@fooishbar.org> 26830Date: Thu Aug 25 03:34:58 2005 +0000 26831 26832 Fix bug in Radeon connector table parsing -- parse all entries, not just 26833 the first. (Ben Herrenschmidt) 26834 26835commit 8734a5d0bafb60f5e38b2cee10587f01fc016379 26836Author: Daniel Stone <daniel@fooishbar.org> 26837Date: Thu Aug 25 03:07:50 2005 +0000 26838 26839 Blacklist Radeon M300 from multimedia I2C code. Handle I2C timeouts more 26840 gracefully -- impose a hard limit of 10ms on waiting for I2C_GO so we 26841 don't lock up hard, spinning. 26842 26843commit 578564b4f7fce1f02d056c2b12ffb7803d268829 26844Author: Adam Jackson <ajax@nwnk.net> 26845Date: Mon Aug 22 03:05:52 2005 +0000 26846 26847 Explicit license statement. 26848 26849commit c9005c91bf6aaf4728a6606284f5d24e0b4a8282 26850Author: Daniel Stone <daniel@fooishbar.org> 26851Date: Thu Aug 18 09:03:46 2005 +0000 26852 26853 Update autogen.sh to one that does objdir != srcdir. 26854 26855commit 508a7421a1386b5b783a867d860b858a55f216d3 26856Author: Alan Hourihane <alanh@fairlite.demon.co.uk> 26857Date: Mon Aug 15 07:30:05 2005 +0000 26858 26859 Egbert's 64bit fixes for mixed 32/64bit clients 26860 26861commit 204290412fbe8b4ac1a0f4f97bc7aec2847a1df7 26862Author: Daniel Stone <daniel@fooishbar.org> 26863Date: Sun Aug 14 16:14:51 2005 +0000 26864 26865 Make DGA, CPIO, non-PCI, and TV-Out support detected per-platform, using 26866 the same semantics as in drivers/ati/Imakefile. 26867 26868commit adb093bdf3cea8108a40bcd72f09924b416de0c1 26869Author: Alex Deucher <agd5f@yahoo.com> 26870Date: Mon Aug 8 23:42:36 2005 +0000 26871 26872 - radeon XPRESS chips are actually RV3xx based 26873 26874commit 9c11018beb186b4b303524685fb62629b9ca863e 26875Author: Vladimir Dergachev <volodya@mindspring.com> 26876Date: Sun Aug 7 14:46:55 2005 +0000 26877 26878 Modified: 26879 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 26880 Update warning about DRI support on R300 and later cards. 26881 26882commit a28b2c76c7df2fe65a93eed34bc85c2cf5afb314 26883Author: Vladimir Dergachev <volodya@mindspring.com> 26884Date: Sun Aug 7 14:40:58 2005 +0000 26885 26886 Modified: 26887 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 26888 Properly configure fbLocation and related registers for R300 cards. (Dave 26889 Airlie) 26890 26891commit 692fb0f8b51abeccaf70b51fc1defc62132b15e9 26892Author: Alex Deucher <agd5f@yahoo.com> 26893Date: Sat Aug 6 17:23:16 2005 +0000 26894 26895 - missing line in mergedfb option parsing. Spotted by Christoph Bauer 26896 26897commit 14e77a39320a7a074b5c2eeeea9750b469abc978 26898Author: Ian Romanick <idr@us.ibm.com> 26899Date: Fri Aug 5 23:04:32 2005 +0000 26900 26901 I have resolved this by making all the "Direct rendering disabled" messages 26902 X_WARNING and all the "Direct rendering enabled" messages X_INFO. 26903 Ideally, we need to factor some of this stuff out to common code. There 26904 is a *LOT* of duplication among the *_dri.c files under 26905 programs/Xserver/hw/xfree86/drivers. 26906 I'm trying to clean up as much low hanging fruit as I can and got tired of 26907 waiting for ajax. :P 26908 26909commit 8e88d36f78c13025847310b411402f1d46a2e0cd 26910Author: Adam Jackson <ajax@nwnk.net> 26911Date: Sun Jul 31 17:19:27 2005 +0000 26912 26913 Post-merge fixups: 26914 - Disable building the gamma driver 26915 - Fix numerous Imakefiles to get header paths and subdirs right 26916 - Symlink client-side GLX lib source from Mesa 26917 - Add new shader subdirs to the Mesa core build 26918 - Tweak the DDX drivers to not ask for "xf86dri.h"; they don't need it, 26919 since they don't generate protocol, and it was breaking the build. 26920 26921commit fb9f712fd1044feb8ce6a99aabaeb5966123e268 26922Author: Kevin E Martin <kem@kem.org> 26923Date: Fri Jul 29 21:22:42 2005 +0000 26924 26925 Various changes preparing packages for RC0: 26926 - Verify and update package version numbers as needed 26927 - Implement versioning scheme 26928 - Change bug address to point to bugzilla bug entry form 26929 - Disable loadable i18n in libX11 by default (use --enable-loadable-i18n to 26930 reenable it) 26931 - Fix makedepend to use pkgconfig and pass distcheck 26932 - Update build script to build macros first 26933 - Update modular Xorg version 26934 26935commit ee45e1fb5da793fa14770b28fd662e69d52b2c6a 26936Author: Michel Daenzer <michel@daenzer.net> 26937Date: Fri Jul 29 19:45:14 2005 +0000 26938 26939 bugzilla #3911 (https://bugs.freedesktop.org/show_bug.cgi?id=3911) 26940 attachment #3191 (http://bugs.freedesktop.org/attachment.cgi?id=3191) 26941 Disable bus mastering while updating MC_FB_LOCATION and friends to 26942 prevent the X server from hanging on startup every now and then under 26943 some circumstances. (ATI Technologies Inc.) 26944 26945commit 44a66ec5a01c3b4dfa046a66b80a3b7e9dfff537 26946Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 26947Date: Fri Jul 15 15:02:47 2005 +0000 26948 26949 Install theathre drivers in @moduledir@/multimedia 26950 26951commit dbcf104689ca368523715eb5945bcd800985799b 26952Author: Kevin E Martin <kem@kem.org> 26953Date: Fri Jul 15 03:39:58 2005 +0000 26954 26955 Fix filename typo 26956 26957commit 2256fabf1cd69c950639f236f61e59e68cc8e6fa 26958Author: Kevin E Martin <kem@kem.org> 26959Date: Thu Jul 14 22:30:47 2005 +0000 26960 26961 Fix check for headers. Use {XORG,DRI}_CFLAGS subst instead of replacing 26962 CFLAGS. 26963 26964commit 520317863ffb365eca17c96b4d545adb554bbbd6 26965Author: Kevin E Martin <kem@kem.org> 26966Date: Thu Jul 14 19:35:48 2005 +0000 26967 26968 Make compiling with DRI support configurable/autodetecable. Autodetect DRI 26969 sdk headers. Expand help string for xorg-module-dir option. 26970 26971commit ab5264838ae34bb29791bc08d5a4f853a995faf1 26972Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 26973Date: Thu Jul 14 17:10:01 2005 +0000 26974 26975 fuck you cvs 26976 26977commit a608c4422665ce1e933483b7c2a25ab2daae5930 26978Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 26979Date: Thu Jul 14 16:03:36 2005 +0000 26980 26981 Try adding build system for ATI driver again. Who knows, maybe CVS works 26982 today. 26983 26984commit d715561d28d82591adf5c7504068c1818eb2245c 26985Author: Adam Jackson <ajax@nwnk.net> 26986Date: Mon Jul 11 02:29:45 2005 +0000 26987 26988 Prep for modular builds by adding guarded #include "config.h" everywhere. 26989 26990commit 5b37b695019b7a4522a96a63f1f6b035160e6a25 26991Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 26992Date: Mon Jul 4 23:44:59 2005 +0000 26993 26994 Bug #2604: Disable workaround for horrible flickering when scrolling 26995 vertically in a virtual screen when color tiling and page flip are both 26996 enabled, as it potentially does more harm than good. The flickering 26997 still needs fixing probably though. 26998 26999commit 08b886850f115f404187acd619aa2c2ca18ae3c7 27000Author: Daniel Stone <daniel@fooishbar.org> 27001Date: Fri Jul 1 22:43:11 2005 +0000 27002 27003 Change all misc.h and os.h references to <X11/foo.h>. 27004 27005commit 1c7c3372776d6d116facabbf055f27af3e96f59f 27006Author: Alex Deucher <agd5f@yahoo.com> 27007Date: Fri Jul 1 04:16:42 2005 +0000 27008 27009 - fix depth tiling for r3/4xx chips (Aapo Tahkola) 27010 - adjust limits for tiled surfaces on r3/4xx 27011 27012commit e8c2574557950bfa150806432d566cd38c47158f 27013Author: Alex Deucher <agd5f@yahoo.com> 27014Date: Fri Jul 1 02:56:04 2005 +0000 27015 27016 - Fix Support for Philips FM1236/F tuner on ATI AIW 9600 XT (Jeff Smith) 27017 Bug 3401 27018 27019commit d258e1c1566d005693260230712c9d78f1447dba 27020Author: Alex Deucher <agd5f@yahoo.com> 27021Date: Thu Jun 30 04:46:11 2005 +0000 27022 27023 - change back part of a comment I seem to have accidently changed. 27024 27025commit 6d9c02d58eae1966e8fa8d7d305849c6e57434ad 27026Author: Alex Deucher <agd5f@yahoo.com> 27027Date: Thu Jun 30 02:02:10 2005 +0000 27028 27029 - small mergedfb fix from Thomas Winischhofer 27030 27031commit 85365b791542cab221c37c9b3b4593ea0c6c9e2d 27032Author: Alex Deucher <agd5f@yahoo.com> 27033Date: Thu Jun 30 01:40:17 2005 +0000 27034 27035 - add support for colortiling on r3/4xx hardware (disabled by default). In 27036 order for pageflipping to work on r3/4xx we will need to update the the 27037 sarea and drm so that the crtc1/2 xytiling regs get updated. Initially 27038 worked out by Aapo Tahkola. 27039 27040commit 3ec99c9cd821aeacdd57a9bbc20c4e4ec4711790 27041Author: Alex Deucher <agd5f@yahoo.com> 27042Date: Thu Jun 30 00:56:01 2005 +0000 27043 27044 - Increase the priority of display 0/1 in the memory controller for r3/4xx 27045 hardware when displaypriority is set to HIGH. Fixes display problems in 27046 high res modes. Originally reported by Aapo Tahkola. 27047 27048commit a7523e6c0be7a600963e9de10cf0060d16117bf6 27049Author: Alex Deucher <agd5f@yahoo.com> 27050Date: Thu Jun 30 00:33:36 2005 +0000 27051 27052 - fix some OUTREGs that should be OUTPLLs in dynamicclocks code 27053 27054commit c7e1d31c6aec3a99061c77c4d6f57e2e874e37f1 27055Author: Adam Jackson <ajax@nwnk.net> 27056Date: Sun Jun 26 04:38:12 2005 +0000 27057 27058 Bug #3628: Fix video driver submodule visibility. 27059 27060commit 1beca962ecee5b5c51eb2fb8451fcc2a2ed4b97c 27061Author: Adam Jackson <ajax@nwnk.net> 27062Date: Sat Jun 25 21:16:53 2005 +0000 27063 27064 Bug #3626: _X_EXPORT tags for video and input drivers. 27065 27066commit 9a4895c4c2fd9614d83adfe3967f4f46dc4dfff6 27067Author: Eric Anholt <anholt@freebsd.org> 27068Date: Wed Jun 22 22:50:49 2005 +0000 27069 27070 Add missing detection of the PCI ID for 0x5b62, Radeon X600 (Vladimir 27071 Kushnir). 27072 27073commit 43e3cc7c21160f4536ed860f97d02545c267dfbc 27074Author: Alex Deucher <agd5f@yahoo.com> 27075Date: Wed Jun 15 23:40:50 2005 +0000 27076 27077 Sync up radeon mergedfb with Thomas' latest sis mergedfb changes: 27078 - non-rectangular mergedfb desktops 27079 - screen offsets 27080 27081commit b67f9f5c875efaa41bded7ed2f38b2a5fbd17597 27082Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 27083Date: Sat Jun 4 20:26:28 2005 +0000 27084 27085 Sync with 6.8.2 branch: Bug #591 27086 (https://bugs.freedesktop.org/show_bug.cgi?id=591) attachment #308 27087 (https://bugs.freedesktop.org/attachment.cgi?id=308): Fix video timing 27088 problems with Sony PCG-C1VN Vaio Picturebook notebook && co. Patch by 27089 Marc Aurele La France 27090 27091commit daae70d9083e9c15358f160f0df3176e344f1cb9 27092Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 27093Date: Sat Jun 4 20:04:36 2005 +0000 27094 27095 Sync with 6.8.2 branch: Bugzilla #1306 27096 (https://bugs.freedesktop.org/show_bug.cgi?id=1306) attachment #1291 27097 (https://bugs.freedesktop.org/attachment.cgi?id=1291): Fix SEGV in 27098 "radeon" driver. Patch by Kevin E. Martin <kem@freedesktop.org> 27099 27100commit c27c193f192b117232dbe39575d3d5ab7c6a1b1e 27101Author: Felix Kuehling <fxkuehl@gmx.de> 27102Date: Fri May 27 22:33:23 2005 +0000 27103 27104 Fix for bugzilla https://bugs.freedesktop.org/show_bug.cgi?id=3414 (patch 27105 https://bugs.freedesktop.org/attachment.cgi?id=2782): Don't use RMX if 27106 we are Dell Server. 27107 27108commit 3b674815e0a6d19b0719e71261c0909b92173aed 27109Author: Adam Jackson <ajax@nwnk.net> 27110Date: Tue May 24 23:26:50 2005 +0000 27111 27112 Typo fix from bug #3163 commit 27113 27114commit 0845d1db801ae74688464bd4c41a7e40e4b50186 27115Author: Egbert Eich <eich@suse.de> 27116Date: Mon May 23 09:50:20 2005 +0000 27117 27118 Explicitely disable DPMS power states during close screen. Fix macro that 27119 takes a value and a mask of bits to modify to be behave evquvalently if 27120 the bits that should remain unchanged are set or unset in the value 27121 (Bugzilla #3369). 27122 27123commit 74b86b036def72e5180a46871b8452ba9eea9bb1 27124Author: Adam Jackson <ajax@nwnk.net> 27125Date: Wed May 18 17:47:35 2005 +0000 27126 27127 Bug #3163: Create new DRIINFO_*_VERSION macros to indicate the version 27128 number of the DRIInfoRec, disambiguating it from the XF86DRI protocol 27129 version number. Modify DRIQueryVersion to return the libdri version 27130 number, which all DDXes were requesting implicitly. Fix the DDXes to 27131 check for the DRIINFO version they were compiled against. 27132 27133commit 0f4a8fc6aa5157a8b48beb49f30511665cfa4b63 27134Author: Matthieu Herrb <matthieu.herrb@laas.fr> 27135Date: Wed Apr 20 21:07:44 2005 +0000 27136 27137 programs/Xserver/hw/xfree86/drivers/ati/theatre.h Build fix for 27138 DoLoadableServer = NO systems. 27139 27140commit 3355d12f832909b05e8b022fad784c9f7af7341f 27141Author: Daniel Stone <daniel@fooishbar.org> 27142Date: Wed Apr 20 12:25:22 2005 +0000 27143 27144 Fix includes right throughout the Xserver tree: 27145 change "foo.h" to <X11/foo.h> for core headers, e.g. X.h, Xpoll.h; 27146 change "foo.h", "extensions/foo.h" and "X11/foo.h" to 27147 <X11/extensions/foo.h> for extension headers, e.g. Xv.h; 27148 change "foo.[ch]" to <X11/Xtrans/foo.[ch]> for Xtrans files. 27149 27150commit 31cb4758eae4703d4de31a0b291847cefa855fc3 27151Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 27152Date: Sun Apr 17 23:28:09 2005 +0000 27153 27154 Added fifo_read and fifo_write in tge generic bus definition 27155 27156commit 2b363e02c7c122f07e6fc406fb8ee52d97b55cad 27157Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 27158Date: Sun Apr 17 23:26:36 2005 +0000 27159 27160 Added new options for the ATI driver: RageTheatreMicrocPath and 27161 RageTheatreMicrocType 27162 27163commit 6286efeba564c63ec198c091e4bba22027586901 27164Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 27165Date: Sun Apr 17 23:23:45 2005 +0000 27166 27167 Addeded support for FI1216ME (PAL) version of the tuner, detection of 27168 UDA1380 sound processor 27169 27170commit 9a63c216e90622cd96d4555155df06dd66c3235b 27171Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 27172Date: Sun Apr 17 23:18:26 2005 +0000 27173 27174 Use the theatre_detect to determine the chip ID and then load the 27175 coresponding theatre module 27176 27177commit 247057762813f12f171b1fa05f31c5ace4c245a7 27178Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 27179Date: Sun Apr 17 23:15:31 2005 +0000 27180 27181 Added functions that access the FIFO on RT200 27182 27183commit bcccf9259784f21d84d52c526aa68df0787118b4 27184Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 27185Date: Sun Apr 17 23:13:45 2005 +0000 27186 27187 Added RT200 registers 27188 27189commit da9d27f8c8cccf2f40a244b25ffe0d3162c41a63 27190Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 27191Date: Sun Apr 17 23:09:28 2005 +0000 27192 27193 This is the theatre chip detection module 27194 27195commit e4623bbce161c41ccac65210fd2539211e8af715 27196Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 27197Date: Sun Apr 17 23:07:54 2005 +0000 27198 27199 Now the detection of the theatre chip is in other module called 27200 theatre_detect 27201 27202commit 405b4dcb067346fce7fd9f6d8a6507a58c740ac7 27203Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 27204Date: Sun Apr 17 23:06:17 2005 +0000 27205 27206 Added the RageTheatre200 video demodulator support 27207 27208commit 310eb24a45b145efbd47196a1c3ec750cf004154 27209Author: Daniel Stone <daniel@fooishbar.org> 27210Date: Mon Apr 4 23:07:08 2005 +0000 27211 27212 Release SDA/SDL I2C lines from an asserted-low state after a DDC probe, 27213 which caused Apple Studio Display monitors to shut off after a second 27214 or so (Ben Herrenschmidt). 27215 27216commit a416667f980a8679352be897af2fb3215da2d000 27217Author: Alex Deucher <agd5f@yahoo.com> 27218Date: Thu Mar 31 23:18:10 2005 +0000 27219 27220 - Add new Radeon pci ids (ATI devrel), bug 2827 27221 27222commit 2e7365ffb767099faf3f191dc65341dbb3ef6247 27223Author: Vladimir Dergachev <volodya@mindspring.com> 27224Date: Thu Mar 31 18:26:05 2005 +0000 27225 27226 Modified: 27227 programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 27228 Patch from Antti Ajanki - provide VBI capture support for PAL and SECAM 27229 (You need GATOS km module to actually get at the data, the change 27230 merely configures the hardware to handle it correctly). 27231 27232commit 9a3c1c8e4cba2885b707db0f855e11dc6e640b49 27233Author: Michel Daenzer <michel@daenzer.net> 27234Date: Tue Mar 29 06:14:34 2005 +0000 27235 27236 bugzilla #2844 (https://bugs.freedesktop.org/show_bug.cgi?id=2844) 27237 attachment #2230 (http://bugs.freedesktop.org/attachment.cgi?id=2230) 27238 Wait for idle before uploading HW cursor data to the framebuffer to 27239 avoid lockups when moving the mouse on some machines. 27240 27241commit 97863f112850be5206ea555297354c1c104fcdbe 27242Author: Kevin E Martin <kem@kem.org> 27243Date: Tue Mar 29 03:49:04 2005 +0000 27244 27245 bugzilla #2840 (https://bugs.freedesktop.org/show_bug.cgi?id=2840) 27246 attachment #2227 (https://bugs.freedesktop.org/attachment.cgi?id=2227) 27247 Fix build problems introduced by previous Radeon PLL errata 27248 workarounds. 27249 27250commit 04cbb8d174cc5e3c7ecdd4e171170213ecb3c7ed 27251Author: Dave Airlie <airlied@linux.ie> 27252Date: Sat Mar 26 00:53:01 2005 +0000 27253 27254 bugzilla #2057 (https://bugs.freedesktop.org/show_bug.cgi?id=2057) 27255 attachment #1516 (https://bugs.freedesktop.org/attachment.cgi?id=1516) 27256 Add TVOUT Support for Mach64 (Leif Delgass, fixed up for Xorg by me). 27257 27258commit 5be4bf9000bdf58584a10a6b8e285d0f173304fa 27259Author: Alex Deucher <agd5f@yahoo.com> 27260Date: Thu Mar 24 06:45:52 2005 +0000 27261 27262 - radeon PLL access errata workarounds (Benjamin Herrenschmidt, acked by 27263 Hui Yu) 27264 27265commit 34af3da80304a13b6407dcc0f4db54d6c8bb7a41 27266Author: Michel Daenzer <michel@daenzer.net> 27267Date: Fri Mar 18 22:36:47 2005 +0000 27268 27269 Add support for production version of ATI RN50/ES1000. (ATI Technologies 27270 Inc.) 27271 27272commit b56a5f96048b920a7751318dde211e881f53b2bb 27273Author: Vladimir Dergachev <volodya@mindspring.com> 27274Date: Thu Mar 17 03:36:25 2005 +0000 27275 27276 Modified: 27277 programs/Xserver/hw/xfree86/drivers/ati/radeon_mergedfb.c Back out 27278 RadeonWaitForIdleMMIO that I put in earlier. 27279 From discussion on dri-devel we do not need to be as careful with INREG() 27280 access on Radeons as we are on mach64 chips. 27281 The fix is backed out due to: 27282 it breaks DRI on SMP systems (due to absense of the lock in mouse handling 27283 functions) 27284 I am no longer able to reproduce the lockup I saw before 27285 27286commit 01a9c1157cdfbc5f74b035383e060b8419979563 27287Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 27288Date: Tue Mar 8 23:41:37 2005 +0000 27289 27290 Fix accidentally removed depth pitch adjustment if displayWidth is not a 27291 multiple of 32 pixels (bug #788) 27292 27293commit 9b994b40710103b263364153429ebdb9ea3f58ed 27294Author: Vladimir Dergachev <volodya@mindspring.com> 27295Date: Fri Mar 4 22:21:40 2005 +0000 27296 27297 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c Let through all chips 27298 R300 and above - presumably the filtering should take place at the DRM 27299 driver level. Correct cut'n'paste position misprint. Big thanks for 27300 Michel Danzer for pointing out both improvements. 27301 27302commit 7d37a9f1f9fffe98067c3b5bc4749cc799fb0c77 27303Author: Vladimir Dergachev <volodya@mindspring.com> 27304Date: Fri Mar 4 05:48:31 2005 +0000 27305 27306 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c Turns out Radeon R420 27307 chips work fine with R300 microcode. Thanks go to Rune Petersen for 27308 careful investigation. 27309 27310commit f1edbb7852d85f660188498f578a0a697c2094fe 27311Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 27312Date: Tue Feb 22 00:39:15 2005 +0000 27313 27314 Fix building radeon driver with BuildXF86DRI set to NO, the breakage was 27315 caused by the tiling support (bug #2481) 27316 27317commit b853dbb50fb7c46fb5a70f349a3f264f0bb94e8c 27318Author: Vladimir Dergachev <volodya@mindspring.com> 27319Date: Sat Feb 19 16:00:12 2005 +0000 27320 27321 Modified: 27322 programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 27323 Fix an INREG without WaitForIdleMMIO. This one should be unlikely to 27324 trigger, but we should still do this in the name of correctness. 27325 27326commit 84290849b4aee336e9db0779d93fc2d7ed8881b1 27327Author: Vladimir Dergachev <volodya@mindspring.com> 27328Date: Sat Feb 19 00:57:23 2005 +0000 27329 27330 Modified: 27331 programs/Xserver/hw/xfree86/drivers/ati/radeon_mergedfb.c 27332 Add a call to RADEONWaitForIdleMMIO() within RADEONChooseCursorCRTC() as 27333 this function uses OUTREGP() which, internally, calls INREG(). 27334 Thus (due to SilkenMouse) INREG() could be called while CP engine was busy. 27335 I have observed this lockup on RV350 + R300 3d driver and have no idea 27336 why no issues were reported for R200 cards. Perhaps, few of them have 27337 enough memory to run merged fb and 3d simultaneously ? 27338 This fix is the right thing to do, however, we might also need to put fifo 27339 management calls into other cursor handling functions (in particular 27340 those that write cursor image directly to video memory). 27341 27342commit 60070063e2130af0b9d5d34eed7adbae0c0fa67b 27343Author: Michel Daenzer <michel@daenzer.net> 27344Date: Fri Feb 18 19:55:35 2005 +0000 27345 27346 Bug #2576: Add support for ATI RN50/ES1000. (ATI Technologies Inc.) 27347 27348commit 32032cdc78097840b7e2685fdec11bc0816aa0ac 27349Author: Adam Jackson <ajax@nwnk.net> 27350Date: Fri Feb 18 16:42:11 2005 +0000 27351 27352 Bug #2567: Fix reversed memset arguments. (David Krause) 27353 27354commit d5fadd6d86ae2a2f8c1bb55e36ab3d6d49649b46 27355Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 27356Date: Wed Feb 9 02:22:39 2005 +0000 27357 27358 do not enable color tiling on r300 chips and newer, since it doesn't work 27359 (yet) (Bugzilla #2504) 27360 27361commit 5ff35300e66a81ecc17c4e9396eb6d158f5c8e1b 27362Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 27363Date: Wed Feb 2 21:35:30 2005 +0000 27364 27365 fix error with color tiling and hw scrolling, it looks like rv100 doesn't 27366 need the special treatment, after all (reported by Stephane Marchesin) 27367 27368commit 7a9031683914bb1d17d223634e5dbf2e18955651 27369Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 27370Date: Mon Jan 31 20:10:28 2005 +0000 27371 27372 change option EnableColorTiling to ColorTiling 27373 27374commit a2b123f7672ea678290798786ecb408a86d8c683 27375Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 27376Date: Wed Jan 26 18:23:41 2005 +0000 27377 27378 (Stephane Marchesin, me) Add support for color (framebuffer) tiling. 27379 Convert all XAA accel code to use explicit dst and src offset pitch 27380 control to make it work. Use the new drm surface ioctls to set up the 27381 tiled surfaces (if dri is enabled). 27382 27383commit e0df1e41724cd009fe7c833d2f9555f5d87e3202 27384Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 27385Date: Wed Jan 26 17:39:56 2005 +0000 27386 27387 fix issues with mergedfb + pageflip (fix up sarea frame values) 27388 27389commit aeb01807263c47f49abaf00031d8505e8ebda82c 27390Author: Egbert Eich <eich@suse.de> 27391Date: Mon Jan 24 15:40:21 2005 +0000 27392 27393 Don't exclude 'RADEONInfoPtr info = ...' declaration even when not building 27394 with DRI support (Bugzilla #2370) 27395 27396commit c6c30cfa354eb456638f10771a7cb02271696af2 27397Author: Vladimir Dergachev <volodya@mindspring.com> 27398Date: Thu Jan 13 17:35:49 2005 +0000 27399 27400 Modified: 27401 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 27402 Some of the newer Radeons (R420 for example) have MM_TABLE's with size 0x33 27403 (instead of 0x0c that we know). Furthermore, they lockup during access 27404 of I2C bus ! 27405 Add a check to validate MM_TABLE. Note: the check is not more stringent as 27406 other card manufacturers (like Asus or Tekram) might make cards with 27407 slightly different MM_TABLE. 27408 27409commit 414c035a965b9db15656058c0e9cfeb3acf0158f 27410Author: Alex Deucher <agd5f@yahoo.com> 27411Date: Fri Dec 24 18:26:19 2004 +0000 27412 27413 - Fix MCLK / SCLK calculation (Benjamin Herrenschmidt) bug 2144 27414 - formatting cleanup 27415 27416commit 7c7b278883676412f5709cf812587b1408628133 27417Author: Alex Deucher <agd5f@yahoo.com> 27418Date: Sun Dec 19 19:49:33 2004 +0000 27419 27420 - Add support for Dualhead on M3/M4 (bug 1760) 27421 27422commit 08a45118b53129f3b9c30004404b906c7d7690be 27423Author: Alex Deucher <agd5f@yahoo.com> 27424Date: Sun Dec 19 19:29:10 2004 +0000 27425 27426 - Make dualhead ram allocation cleaner and related messages clearer 27427 27428commit 97cfd7b1489a9510cd233d6dc7154c210f32b9f5 27429Author: Alex Deucher <agd5f@yahoo.com> 27430Date: Sun Dec 19 06:56:37 2004 +0000 27431 27432 - Add DRI resume support (Micha Feigin) bug 2063 27433 27434commit 5b8577ae3200d967b4316156f16028cbcfd21983 27435Author: Alex Deucher <agd5f@yahoo.com> 27436Date: Sun Dec 19 06:11:06 2004 +0000 27437 27438 - fix warnings (Ronny V. Vindenes) 27439 27440commit 447124a663a24a7fffb0cd241d18dd896a43098b 27441Author: Alex Deucher <agd5f@yahoo.com> 27442Date: Sat Dec 18 21:37:48 2004 +0000 27443 27444 - fix warning and potential error in radeon_driver.c 27445 27446commit ac6166158a58b377e7383950c169956f78b62093 27447Author: Alex Deucher <agd5f@yahoo.com> 27448Date: Fri Dec 17 19:19:32 2004 +0000 27449 27450 - Add VGAAccess option to r128 driver. Based on Benh's radeon patch. r128 27451 should now work on ppcs without usefbdev. To use this on ppc make sure 27452 you set "usefbdev" to false. (bug 2089) 27453 - Apply Benjamin Herrenschmidt's radeon patches (bug 2064) 27454 27455commit ac860ed6bbed84f6fc39a9acf53ee87646cfc8dc 27456Author: Vladimir Dergachev <volodya@mindspring.com> 27457Date: Fri Dec 17 16:50:36 2004 +0000 27458 27459 Modified: 27460 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 27461 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c 27462 Move DMA robustness fix into radeon_dri.c::RADEONEnterServer() as per 27463 suggestion by Michel D�nzer. 27464 I could not trigger a lockup, even with r300_demo (possibly it has code 27465 that flushes cache inside ?), so this must be good enough.. 27466 27467commit 3ba5b525c4d6390184fdf79b28c954d5caadc6a4 27468Author: Vladimir Dergachev <volodya@mindspring.com> 27469Date: Thu Dec 16 16:26:56 2004 +0000 27470 27471 Modified: 27472 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 27473 Add DMA robustness fix from Nicolai Haehnle. 27474 TODO: move this fix into DRM driver or other parts of Xserver when R300 27475 support is more stable. 27476 This fix affects only R300 and later variants and results in small slowdown 27477 when using CP acceleration in comparison to the previous version of the 27478 driver. 27479 We are still a lot faster in CP mode than using MMIO register access. 27480 27481commit c7b5f4a9eb49c0edd690c972be1015c98408ffaf 27482Author: Adam Jackson <ajax@nwnk.net> 27483Date: Wed Dec 15 19:35:33 2004 +0000 27484 27485 Bug #2008: Fix gcc4 compile problem. 27486 27487commit 8c1969f22fba2e1c9ed34d5ca77bee3a9f577640 27488Author: Vladimir Dergachev <volodya@mindspring.com> 27489Date: Sun Dec 12 20:42:29 2004 +0000 27490 27491 Modified: 27492 programs/Xserver/hw/xfree86/drivers/ati/radeon_render.c 27493 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 27494 Make DMA blitting code more robust. 27495 27496commit 8a9077bc70fd800983f138a224b95301992a2296 27497Author: Vladimir Dergachev <volodya@mindspring.com> 27498Date: Sun Dec 12 19:28:37 2004 +0000 27499 27500 Modified: 27501 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 27502 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c 27503 Enable DRM support permanently, allowing its use for any CHIP_FAMILY > 27504 R300. Fix bug in naming Mesa3d driver (it was defaulting to r200) 27505 27506commit 845df6dd96785b59c44c670009c4fcdb20e147ab 27507Author: Alex Deucher <agd5f@yahoo.com> 27508Date: Sun Dec 12 17:29:45 2004 +0000 27509 27510 - missing diff from agp 8x support 27511 27512commit 36b92808e4ff1311f91543388c18cbe8e9eb4aab 27513Author: Vladimir Dergachev <volodya@mindspring.com> 27514Date: Sun Dec 12 16:05:35 2004 +0000 27515 27516 Modified: 27517 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 27518 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 27519 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c 27520 programs/Xserver/hw/xfree86/drivers/ati/radeon_common.h 27521 programs/Xserver/hw/xfree86/drivers/ati/radeon_version.h Add R300 (and 27522 later) DRM support, protected by X_R300_DRM option. 27523 Note: at the moment this is for 2d support only, if enabled running 27524 glxgears locks up my machine, but DMAForXv=yes works fine. 27525 27526commit 68ea67ce98b8f3ad4c17f987e0e2b92f79a96fbf 27527Author: Alex Deucher <agd5f@yahoo.com> 27528Date: Sun Dec 12 02:00:48 2004 +0000 27529 27530 - add AGP 8x support (Hui Yu, Michel Daenzer) bug 1819 27531 27532commit 93675d288d5ebb35d4724ebbfc455a703e25a311 27533Author: Alex Deucher <agd5f@yahoo.com> 27534Date: Sun Dec 12 00:14:32 2004 +0000 27535 27536 - only mess with Dynamicclocks on mobility chips (Benh, me) bug 1912 27537 - fix display detection (Hui Yu) bug 1559 27538 - Add new option "BIOSHotkeys" to allow the bios to control output 27539 switching (off by default) bug 2000 27540 - update the savage man page to document the "CrtOnly" option 27541 27542commit ed217003127028483fd2956851d438977169d154 27543Author: Daniel Stone <daniel@fooishbar.org> 27544Date: Mon Dec 6 13:27:56 2004 +0000 27545 27546 Apply Debian patch to re-POST via VBE if driver thinks it has no video 27547 memory (a symptom of incompletely coming up from D3 power state), which 27548 fixes some Dell laptops (closes: #1890). 27549 27550commit 2b8ab42b05b32710fa4f71a25e2f84192e637fbd 27551Author: Michel Daenzer <michel@daenzer.net> 27552Date: Mon Dec 6 06:27:53 2004 +0000 27553 27554 When direct rendering is enabled, use hostdata blits to transfer data from 27555 system memory to video RAM, which should reduce CPU usage especially 27556 with larger videos. Can be disabled via Option "DMAForXv" if there 27557 should be any stability issues, but it's been stable for me during a 27558 week of testing. Based on a patch by Nikolaus Meine 27559 <meine@tnt.uni-hannover.de>. Probably fix endianness issues in some 27560 newer XVideo code, untested. 27561 Also use hostdata blits to transfer RENDER image data to video RAM to avoid 27562 idling the accelerator engine. Increases RENDER performance 27563 significantly for me. 27564 These changes were only tested on an M9 in a Titanium PowerBook but should 27565 work with all Radeons where direct rendering is supported. 27566 27567commit 960d15ae3a44efe7c02a3d6d0acbabe63e07be74 27568Author: Daniel Stone <daniel@fooishbar.org> 27569Date: Sat Dec 4 22:34:54 2004 +0000 27570 27571 Disable interrupt handling after the DRM handler has been removed, 27572 preventing lockups (closes: #1886). 27573 Allow pitches with a granularity of 32, rather than 64, because some 27574 clamshell iBooks have 800-byte pitches. Change Xv allocation 27575 granularity to 8 bytes, rather than 16 bytes, also (closes: #1888). 27576 27577commit 4706dfb7804a46e3cc5aafc3644f5a1134ab3d19 27578Author: Daniel Stone <daniel@fooishbar.org> 27579Date: Sat Dec 4 22:24:06 2004 +0000 27580 27581 Retry CCE idle commands until the (higher) timeout is reached, and properly 27582 stop the engine if we totally fail -- gets rid of serious performance 27583 issues and lockups with r128 DRI. 27584 27585commit bce508a288d6b95708dbbfc6b235e2d95b6b96ad 27586Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk> 27587Date: Sat Dec 4 00:42:59 2004 +0000 27588 27589 Encoding of numerous files changed to UTF-8 27590 27591commit e9c9fd279ed43cc0d779266efcea69a67541c794 27592Author: Egbert Eich <eich@suse.de> 27593Date: Tue Nov 30 08:38:44 2004 +0000 27594 27595 Make Xorg the default server to install. 27596 Avoid PIO access on IA64. Some IA64 machine check if legacy ports outside 27597 the VGA range are accessed. The ATi driver however does this to probe 27598 for ISA Mach8/32/64. Since no IA64 has ISA slots this restriction 27599 should not be relevant to the user. 27600 Avoid recursive calls of xf86scanpci(). This function normally detects that 27601 it has been called before by checking if the PCI structure is filled 27602 out. So far if this was not the case (because PCI probing has failed 27603 for some reason) the function is traversed again. With the chipset 27604 specific PCI bus probing this can lead to an endless recursive loop as 27605 the post-probing code calls xf86scanpci() from within this function. 27606 The OS specific PCI code for Linux worked only if bus 0 was populated as it 27607 checked for the presence of /proc/bus/pci/00. Fixed to check for 27608 /proc/bus/pci/<bus_to_look_for> instead. 27609 27610commit fffbcb761c178e025e3d55d428a803b1284481e0 27611Author: Matthieu Herrb <matthieu.herrb@laas.fr> 27612Date: Sat Nov 27 15:14:15 2004 +0000 27613 27614 xc/programs/Xserver/hw/xfree86/driver/ati/radeon.man 27615 xc/programs/Xserver/hw/xfree86/driver/voodoo/voodo.man 27616 xc/programs/Xserver/hw/xfree86/driver/wsfb/wsfb.man XF86Config -> 27617 __xconfigfile__ 27618 27619commit ad6347d187f9afa053c755aa3a2da9d821916cb2 27620Author: Daniel Stone <daniel@fooishbar.org> 27621Date: Tue Nov 23 21:27:43 2004 +0000 27622 27623 Use fp_horz_stretch instead of fp_vert_stretch for horizontal stretch 27624 factor calculation in panel size derivation in the Radeon driver 27625 (closes: #1881). 27626 27627commit ac054e427c6ce8cd5b89459dd3f16ca20e6b2d6a 27628Author: Vladimir Dergachev <volodya@mindspring.com> 27629Date: Sun Oct 24 18:17:36 2004 +0000 27630 27631 Add registers to facilitate work with stereo 3d. 27632 27633commit 445ed4b9b61347c866ee67dc0cb110904d667b84 27634Author: Alex Deucher <agd5f@yahoo.com> 27635Date: Fri Oct 22 03:51:58 2004 +0000 27636 27637 - make the overlay work on crtc2 again 27638 - make PutVideo() MergedFB aware 27639 27640commit 54cffc3fe273cf3c9ed2a8aef833d4bd337a8a09 27641Author: Egbert Eich <eich@suse.de> 27642Date: Mon Oct 18 15:19:44 2004 +0000 27643 27644 Fixing one of the infamous 'control reaches end of non void function 27645 problems'. 27646 27647commit 7a97d363327e0b2c8646d961e00d05f99766b54f 27648Author: Egbert Eich <eich@suse.de> 27649Date: Mon Oct 18 14:21:45 2004 +0000 27650 27651 Made handling of DevelDrivers for x86-64 more conformant to other 27652 platforms. 27653 Compress all font encodings (Stefan Dirsch). 27654 Fixed warnings. 27655 Turn on forwarding XNSpotLocation event to XIM server in OffTheSpot and 27656 Root mode (bugzilla #1580, James Su). 27657 Added another compose key combination for the Euro symbol (Stefan Dirsch). 27658 Added support for UTF-8 in ja_JP, ko_KR and zh_TW locales (Mike Fabian). 27659 Changed default encoding for ru from KOI8-R to ISO8859-5 (Mike Fabian). 27660 This is the encoding that is also used by glibc. We may break other 27661 libcs - lets see who complains. 27662 Added explanation for DESTDIR to install to a different directory than /. 27663 Added some early bailouts to atiprobe if PCI structure pointer is NULL to 27664 prevent sig11. 27665 XV support doesn't depend on 2D acceleration any more. This patch removes 27666 this limitation from the NSC driver. This is a patch that I have 27667 committed to XFree86 a while ago but never ported over to X.Org. 27668 Matthieu Herrb contributed some build fixes. 27669 Fixing SetDPMSTimers() so that DPMS*Time == 0 disables the appropriate 27670 timer. This takes advantage of the fact that TimerSet() with a timeout 27671 argument 0 behaves like TimerCanel(). 27672 Use /dev/xconsole (named pipe) or devpts for system logger (Werner Fink). 27673 Create missing links for backward compatibility to XFree86 (Stefan Dirsch). 27674 Changed comment to mention xorg. 27675 Changed cursor for the 'kill' action from XC_draped_box to XC_Pirate. If 27676 you don't like it we can change it back (original author unkown). 27677 Added 'pic' to the man page rendering command pipeline (Werner Fink). 27678 Added missing return value (Stefan Dirsch, Roland Mainz) 27679 27680commit c7ab340fecb94a89c25d6bc7f686b7c97cd3067f 27681Author: Alex Deucher <agd5f@yahoo.com> 27682Date: Sun Oct 17 22:45:57 2004 +0000 27683 27684 - Fix DFP detection in r128 driver (bug 1386, Klaus Kusche) 27685 - Fix EmulateWheelInertia defaults (bug 1115, Andrew Pimlott) 27686 27687commit ace6e1b7933cb415b8efc41dffa335938f9c89b5 27688Author: Vladimir Dergachev <volodya@mindspring.com> 27689Date: Sun Oct 17 17:57:33 2004 +0000 27690 27691 Modified: 27692 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c On my Inspiron 27693 8600 both internal and external ports are marked DAC_PRIMARY in BIOS. 27694 So be extra careful - only swap when the first port is not DAC_PRIMARY 27695 27696commit 22405e81b812c239b5d8653b402b5f846e6e10b1 27697Author: Alex Deucher <agd5f@yahoo.com> 27698Date: Sun Oct 17 16:52:02 2004 +0000 27699 27700 - Add Thomas' latest sis MergedFB enhancements to the radeon driver 27701 27702commit 186efa31b2cb7285717678bb1c4df4cd9e182395 27703Author: Vladimir Dergachev <volodya@mindspring.com> 27704Date: Wed Oct 13 23:02:42 2004 +0000 27705 27706 Modified: 27707 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 27708 xc/programs/Xserver/hw/xfree86/drivers/i2c/fi1236.h Squash annoying warning 27709 about fi1236_dump_status 27710 27711commit 43b24861e36023613a99627c57ac44721ed1a548 27712Author: Egbert Eich <eich@suse.de> 27713Date: Tue Oct 12 19:13:43 2004 +0000 27714 27715 Set fbdev mode as the default mode on PPC (Olaf Hering). 27716 Added support for IBM space saver keyboard (Stefan Dirsch). Added support 27717 for Cherry CyMotion Master XPress (Marcus Schaefer). 27718 Change order of SetDisplayDevice(), HWRestore(), UnbindGART() and 27719 RestoreBIOSMemSize() to be exactly opposite to the Save procedure in 27720 EnterVT() (Matthias Hopf, Alan Hourihane). 27721 Fix text mode restauration by removing the assumption that the register 27722 which determines which head is programmed is set. to the active head by 27723 the BIOS (Mark Vojkovich). 27724 When I wrote the resource code 5 years ago I made some assumptions which 27725 turned out to be false: I've assumed that the bus number of the PCI 27726 hostbridge would be the PCI bus the bridge links to. This is not 27727 correct. Fixing this assumption is not easy. However I hope that the 27728 attached patch will make the system work regardless as it 'ignores' 27729 host bridges from which the target bus is not known. This should not 27730 matter at all as we really don't care about host bridges (unless we 27731 have bridge specific code which retrieves information about the 27732 bridge). 27733 Fixed server crash on reset when a structure allocated in PreInit() was 27734 freed on CloseScreen(). 27735 Fixed ring buffer lock ups that happened because the structure that 27736 contained ringbuffer data was not zeroed after allocation. 27737 Fixed numerous warnings due to signed unsigned comparisons. 27738 programs/Xserver/hw/xfree86/drivers/nv/nv_driver.c: 27739 (NVBacklightEnable): Changed the order in which the sequencer registers 27740 and the backlight control registers are written. The sequencer control 27741 register need to be written first otherwise DPMS screen blanking 27742 produces vertical bars on a mobile device. lib/Xau/Imakefile: Build 27743 libXau static library PIC so it can be linked into toolkits that would 27744 like to wrap its functionality. 27745 27746commit 515b7c4b748f70c17aee149a33ee1b41632daa1a 27747Author: Alex Deucher <agd5f@yahoo.com> 27748Date: Mon Oct 11 23:20:26 2004 +0000 27749 27750 - fix resume on radeons (bug 1220) 27751 - clean up some old cruft in the savage i2c code 27752 27753commit 07e1aaee862c4835601c704e14fcb12e28f4e78e 27754Author: Egbert Eich <eich@suse.de> 27755Date: Mon Oct 11 09:58:04 2004 +0000 27756 27757 Improving DPMS handling on VT swich and server termination/abort: previous 27758 version called the driver directly and too late. 27759 Unblank secondary screen explicitely. Don't rely on the value read during 27760 register save as the BIOS have blanked the secondary head. 27761 Checking if server isn't switched away before calling sync. Sanity check 27762 for possible bugs in aother areas of the code. 27763 Fixing default amount of of allocated video memory from AGP for i810: Use 27764 16MB if less than 192MB are installed else use 24MB (Matthias Hopf). 27765 27766commit a8c77480e7f4ddf8c45bea3ecec7b6564c0d6452 27767Author: Vladimir Dergachev <volodya@mindspring.com> 27768Date: Fri Oct 8 00:56:05 2004 +0000 27769 27770 Modified: 27771 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c Remove redundant 27772 and incorrect line. 27773 27774commit ec831d4fba23dc5f4d0a1450aa63efe5ff5937a2 27775Author: Vladimir Dergachev <volodya@mindspring.com> 27776Date: Tue Oct 5 01:16:01 2004 +0000 27777 27778 Modified: 27779 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre.h Patch from Matthieu 27780 Herrb - fix compilation with DoLoadableServer=NO 27781 27782commit f1dadde8f642b332a2dcd2e2a5779f25fec7a3de 27783Author: Adam Jackson <ajax@nwnk.net> 27784Date: Mon Oct 4 21:52:47 2004 +0000 27785 27786 Correct radeon man page, IGP is supported for 3D but R400 is not. 27787 27788commit e3dba9841521466629828501c4fdccbba3d6af89 27789Author: Vladimir Dergachev <volodya@mindspring.com> 27790Date: Mon Oct 4 15:24:34 2004 +0000 27791 27792 Modified: 27793 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c Do not probe for 27794 Rage Theatre if no multimedia table is present and no parameters have 27795 been specified. 27796 27797commit 4aadb784928ebbd60dce172c9a9a80e8d84e943f 27798Author: Vladimir Dergachev <volodya@mindspring.com> 27799Date: Mon Oct 4 14:53:11 2004 +0000 27800 27801 Modified: 27802 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c This is really 27803 embarassing - I found a part of code that had <cr> at the end of the 27804 lines. I am certain I never saw it before - talk about code rot ! 27805 27806commit db5a27081d9d3ab4cc1481a7d110d11103b3de19 27807Author: Vladimir Dergachev <volodya@mindspring.com> 27808Date: Sun Oct 3 23:54:40 2004 +0000 27809 27810 Modified: 27811 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_mm_i2c.c Remove debugging 27812 statement that cause loading of DLL modules to fail. 27813 27814commit 37a8d8a53f12f829783011c1e077fefb5729e618 27815Author: Vladimir Dergachev <volodya@mindspring.com> 27816Date: Sun Oct 3 14:41:45 2004 +0000 27817 27818 Modified: 27819 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c Fix compilation 27820 with gcc 3.4.x (patch by Ronny V. Vindenes) 27821 27822commit 6dbcd503fe21c6dec27c2b97adc0f39ba0a7ab2a 27823Author: Vladimir Dergachev <volodya@mindspring.com> 27824Date: Sun Oct 3 03:07:17 2004 +0000 27825 27826 Modified: 27827 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.[c,h] 27828 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre.c 27829 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_reg.h Port the rest of 27830 GATOS Radeon-specific code. Remove "checkpoint" noise from logs. Test 27831 with AIW Radeon 7500 27832 27833commit 7cb911291d459f08f54444d09081d3378fe73b8b 27834Author: Vladimir Dergachev <volodya@mindspring.com> 27835Date: Sun Oct 3 00:01:14 2004 +0000 27836 27837 Modified: 27838 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon.h 27839 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 27840 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.[c,h] 27841 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre.c Detect and initialize 27842 Rage Theatre (RT100) 27843 27844commit 6c1996174b9f7afc8f4e33c35e4864b31696d4d6 27845Author: Vladimir Dergachev <volodya@mindspring.com> 27846Date: Sat Oct 2 20:55:42 2004 +0000 27847 27848 Modified: 27849 xc/programs/Xserver/hw/xfree86/drivers/ati/Imakefile 27850 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_reg.h 27851 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.[c,h] 27852 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_vip.c Add VIP bus access 27853 code, hook it into initialization system. 27854 27855commit 136649a1c2065624a9cf17386054c6595aa6d85e 27856Author: Vladimir Dergachev <volodya@mindspring.com> 27857Date: Sat Oct 2 20:35:04 2004 +0000 27858 27859 Modified: 27860 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre* 27861 xc/programs/Xserver/hw/xfree86/drivers/ati/generic_bus.h Add Rage Theatre 27862 (RT100) module code and hook it into the build system 27863 27864commit 7a17b1c294fa8c7af0b1cdea0618cb6060479356 27865Author: Vladimir Dergachev <volodya@mindspring.com> 27866Date: Sat Oct 2 01:59:13 2004 +0000 27867 27868 Modified: 27869 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon.h 27870 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 27871 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.h 27872 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_mm_i2c.c Add code to 27873 access and initialize multimedia i2c bus. Hook it up. 27874 27875commit e0123e24b6895dde7b385528da24c4e6e9d5a4cb 27876Author: Vladimir Dergachev <volodya@mindspring.com> 27877Date: Sat Oct 2 01:36:43 2004 +0000 27878 27879 Add more multimedia registers. 27880 27881commit 576f53be4f0fa54bde7167872b7508b98e5c9454 27882Author: Vladimir Dergachev <volodya@mindspring.com> 27883Date: Sat Oct 2 00:43:08 2004 +0000 27884 27885 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 27886 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon.h Add TV-in specific 27887 options. Add TV-in specific fields to radeon per-screen structure, add 27888 code to initialize them. Preserve MM_TABLE (instead of whole VBIOS as 27889 was done in GATOS drivers) 27890 27891commit 9a9ba82172428b09312fdd6e11da04b21c8529fc 27892Author: Adam Jackson <ajax@nwnk.net> 27893Date: Wed Sep 22 04:38:03 2004 +0000 27894 27895 Bug #1252, #1253, #1255, #1256: Various typo fixes from Dave Jones. 27896 27897commit 700b0dc5148d0170c9223765d208ccef447242df 27898Author: Adam Jackson <ajax@nwnk.net> 27899Date: Tue Sep 21 18:10:34 2004 +0000 27900 27901 Convert the ATI drivers to ANSI function prototypes. 27902 27903commit 17e5f3c1169066be1004306d12b5398cf01cbd35 27904Author: Eric Anholt <anholt@freebsd.org> 27905Date: Fri Sep 17 05:16:45 2004 +0000 27906 27907 Bugzilla #1170: Fix an overcounting of DMA in BEGIN_RING that wasn't 27908 harmful. Bugzilla #1169: Add counting for 27909 BEGIN_RING/OUT_RING/ADVANCE_RING to avoid these sorts of problems in 27910 the future. 27911 27912commit 29bbee6b8a8dd1b883e07547691ff8e1289459ce 27913Author: Adam Jackson <ajax@nwnk.net> 27914Date: Fri Sep 17 03:04:50 2004 +0000 27915 27916 Bug #1192: Remove cfb support from drivers where its use is an option. 27917 Delete xf24_32bpp, as s3virge was the last user. Fix up some comments 27918 to refer to fb rather than cfb. 27919 27920commit cec06d4126f5f31d14ace73798c71ec75e544c45 27921Author: Alex Deucher <agd5f@yahoo.com> 27922Date: Thu Sep 16 22:15:05 2004 +0000 27923 27924 Update comment 27925 27926commit dbeb8d28ac3469f4a7d7d8dae979ae876d4cdcba 27927Author: Kevin E Martin <kem@kem.org> 27928Date: Fri Sep 3 23:26:19 2004 +0000 27929 27930 Updated release notes for bug #999. 27931 Update Radeon man page to note that RenderAccel is now enabled by default 27932 on certain chips. 27933 27934commit c829f667c8a79bbdd40e1f924d57638e7bd425dc 27935Author: Michel Daenzer <michel@daenzer.net> 27936Date: Mon Aug 30 04:31:50 2004 +0000 27937 27938 Hotfix for Eric Anholt's fix for bug #1156 : Clear the 27939 RADEON_NONSURF_AP0_SWP_16BPP bit for 4 bpp data, otherwise it breaks at 27940 depth 16. 27941 27942commit 677246740b5b538e73a7fbde661ec6a97319430b 27943Author: Adam Jackson <ajax@nwnk.net> 27944Date: Fri Aug 27 20:31:26 2004 +0000 27945 27946 Back out my last change, dupe. 27947 27948commit a8eaa15264ff0db1596d653e91260c9a39a8b5ee 27949Author: Adam Jackson <ajax@nwnk.net> 27950Date: Fri Aug 27 18:32:40 2004 +0000 27951 27952 Bug #1061 (documentation-only update): Document the RenderAccel option in 27953 the radeon manpage. 27954 27955commit 360dab062f895c19e1784146e6684f75a20c5b24 27956Author: Eric Anholt <anholt@freebsd.org> 27957Date: Fri Aug 27 03:36:28 2004 +0000 27958 27959 Bug #1156: Fix the build on PPC by splitting the byte-order-setting code 27960 into a separate routine. While here, fix the byte swapping in the 32bpp 27961 case. (Michel Daenzer) 27962 27963commit 9b19bbb4d37b79c97365aa492cd93931d492b3a8 27964Author: Eric Anholt <anholt@freebsd.org> 27965Date: Fri Aug 27 03:23:47 2004 +0000 27966 27967 Bug #1157: Fix Radeon line acceleration in the DMA case (Michel Daenzer). 27968 27969commit 838488b6b46359334e5e781589dbee3832909628 27970Author: Kevin E Martin <kem@kem.org> 27971Date: Fri Aug 27 02:38:56 2004 +0000 27972 27973 Fix Radeon compilation without Xinerama (Bug #1179, Simon Schubert). 27974 27975commit 5f54d4f45e2adc6b15e98152db88743c243aa128 27976Author: Kevin E Martin <kem@kem.org> 27977Date: Wed Aug 25 00:30:41 2004 +0000 27978 27979 Fix hangs during startup with Radeon driver (Bug #1007, Sjoerd Simons). 27980 Fix i810 driver startup on Sony VAIO laptop (Bug #1084, Dave Airlie). 27981 Fix several library build problem (Bug #1147). 27982 Fix build problems (Bugs #1148 and #1149). 27983 27984commit dde880e876f455f1fd66efbcc055e177886397c9 27985Author: Eric Anholt <anholt@freebsd.org> 27986Date: Tue Aug 17 20:22:55 2004 +0000 27987 27988 Reenable Radeon Render acceleration now that interactions with 3D clients 27989 should be resolved. 27990 27991commit 279efef503fdbd442a9bd00f0073d9f6006d9e61 27992Author: Eric Anholt <anholt@freebsd.org> 27993Date: Sun Aug 15 00:44:57 2004 +0000 27994 27995 Fix unresolved symbol warnings in r128 driver (Adam Jackson) 27996 27997commit 955f2b89f94aaa20b5eafd045b01e3eeef3d29ad 27998Author: Kevin E Martin <kem@kem.org> 27999Date: Thu Aug 12 05:00:22 2004 +0000 28000 28001 Fix context switching on Radeons and build problem when XF86DRI is not 28002 defined (Bug #922, Eric Anholt). 28003 28004commit 8f3ea53f26493a5f02410297a587d1d95c067094 28005Author: Adam Jackson <ajax@nwnk.net> 28006Date: Thu Aug 12 02:23:46 2004 +0000 28007 28008 oops, these didn't added somehow. 28009 28010commit 6ec74414bd33e9d8dfd5119058d15f3a6206e9d9 28011Author: Adam Jackson <ajax@nwnk.net> 28012Date: Thu Aug 12 01:03:06 2004 +0000 28013 28014 Factor out R{128,ADEON}Chipsets into their own headers so they can be 28015 sanely included in both the ati core and the r128/radeon submodules; 28016 r128 and radeon can now be directly loaded under dlloader. Teach imstt 28017 about the dlloader- frendly way to talk to fbdevhw. (Partial for bugs 28018 #393 and #400.) 28019 28020commit a5b1e4dd7adacd17b63582e477df3a585efb249a 28021Author: Adam Jackson <ajax@nwnk.net> 28022Date: Wed Aug 11 01:20:28 2004 +0000 28023 28024 Bug #978: Silence more unresolved symbol warnings from the ati driver. 28025 Patch from Stefan Dirsch. 28026 28027commit 1356884132491010a2e41eb2b5be4a227b9b8138 28028Author: Kevin E Martin <kem@kem.org> 28029Date: Wed Aug 11 01:14:14 2004 +0000 28030 28031 Fix Radeon build problem with #define DoLoadableServer NO 28032 28033commit f5c93cb015e334975c2797a23b68798372a284c9 28034Author: Adam Jackson <ajax@nwnk.net> 28035Date: Tue Aug 10 04:01:42 2004 +0000 28036 28037 Bug #978: Silence unresolved symbol warnings from the ati driver. Reported 28038 by Michel Daenzer. 28039 28040commit a2da1caf5848a78b10afd99811fd1a5677af6786 28041Author: Kevin E Martin <kem@kem.org> 28042Date: Mon Aug 9 22:37:22 2004 +0000 28043 28044 Fix install problem on platforms not using xorg.cf/xfree86.cf (Bug #339, 28045 Harold L. Hunt II, Alexander Gottwald). 28046 Fix crash when using X core font in zh_CN.UTF-8 locale (Bug #368, Yu Shao, 28047 David Dawes). 28048 Fix glXMakeCurrent(Dpy, None, NULL) crash (Bug #719, Adam Jackson). 28049 HP-PA build fix (Bug #828, Guy Martin, Paul Anderson). 28050 Fix SDK build for GATOS and Wacom driver (Bug #829, Bryan Stine). 28051 Fix attempt to read video ROM before enabling it (Bug #843, Ivan 28052 Kokshaysky, Mike A. Harris). 28053 Fix detection of primary adapter (Bug #843, Ivan Kokshaysky, Mike A. 28054 Harris). 28055 Clarify xset man page description of how to use the keyboard repeat rate 28056 settings (Bug #846, Mike A. Harris). 28057 Fix problem where print-screen key would get remapped to sys-req in certain 28058 keymaps, which broke GNOME printscreen functionality (Bug #847, Owen 28059 Taylor). 28060 Fix several render problems: 28061 - MMIO mode support 28062 - Hang on IGP chips 28063 - VT switching hang 28064 - 3D render corruption (Bug #922, Hui Yu). 28065 28066commit 3171664c6a02424736864204b731249097fa7aae 28067Author: Kevin E Martin <kem@kem.org> 28068Date: Fri Aug 6 20:07:35 2004 +0000 28069 28070 Disable Render acceleration in Radeon driver temporarily until it no longer 28071 conflicts with the 3D code. 28072 28073commit 27ef7adf3c1f6c3ca79b42b468d08b8542348cdd 28074Author: Michel Daenzer <michel@daenzer.net> 28075Date: Wed Aug 4 13:17:31 2004 +0000 28076 28077 Use info->ModeReg.surface_cntl to restore byte swapping for the framebuffer 28078 aperture on big endian machines, remove superfluous local variables and 28079 register reads. 28080 Adapt framebuffer aperture byte swapping to texture format before copying 28081 data to offscreen area and restore it afterwards on big endian 28082 machines, fixes Render acceleration there. 28083 reviewed by: Hui Yu <hyu@ati.com>, Kevin E. Martin <kem@freedesktop.org> 28084 28085commit 2efdc9f847fa9ff3096667b16df84817fc65605d 28086Author: Eric Anholt <anholt@freebsd.org> 28087Date: Wed Aug 4 10:05:37 2004 +0000 28088 28089 - Add two new XAA hooks, SetupForCPUToScreenTexture2 and 28090 SetupForCPUToScreenAlphaTexture2. These add a dstFormat argument after 28091 the previous format argument, which the driver needs to use to properly 28092 set up the destination format. Two new arrays are added for the list of 28093 destination formats supported that correspond to the previous format 28094 arrays for sources. 28095 - Make Render acceleration only occur when the new hook for that 28096 acceleration type is supplied and the dst format list is set, along 28097 with the src format list being set. Without knowing the destination 28098 format, the Render acceleration couldn't properly support all the 28099 destinations it might encounter. 28100 - Bump XAA module minor version. 28101 - Update the Radeon Render acceleration to use the new hooks when the XAA 28102 module is sufficiently new. Fix a bug in the src/dst alpha booleans for 28103 ops, and use them to set blend_cntl to support destinations without 28104 alpha. Add missing PICT_a1r5g5b5 texture format, and add list 28105 terminator. (!) 28106 28107commit 1e16c66551eed750330bb74a92208c45a7112cd0 28108Author: Matthieu Herrb <matthieu.herrb@laas.fr> 28109Date: Tue Aug 3 09:42:20 2004 +0000 28110 28111 programs/Xserver/hw/xfree86/drivers/ati/radeon_merged_fb.c Add missing 28112 argument to xf86DrvMsg(). 28113 28114commit fd440a1e2110e05d116207800c123c281fbf5e19 28115Author: Matthieu Herrb <matthieu.herrb@laas.fr> 28116Date: Tue Aug 3 09:38:03 2004 +0000 28117 28118 programs/Xserver/hw/xfree86/drivers/ati/r128_driver.c 28119 programs/Xserver/hw/xfree86/drivers/ati/r128_probe.h 28120 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 28121 programs/Xserver/hw/xfree86/drivers/ati/radeon_probe.h Fix Constification 28122 problems. 28123 28124commit dd0f86c4cb046c79a65144f767e97fd99e7f61c9 28125Author: Kevin E Martin <kem@kem.org> 28126Date: Sun Aug 1 23:22:01 2004 +0000 28127 28128 Fix Radeon crash on server exit or generation 28129 28130commit c64f90b673f5c487dfe059f1636b04a1c2ab7dc2 28131Author: Adam Jackson <ajax@nwnk.net> 28132Date: Sat Jul 31 01:21:18 2004 +0000 28133 28134 Change several LoaderSymbol calls introduced by the bug #400 patch to 28135 *Weak() resolver functions. 28136 28137commit b091b4b074cd61c9067ba95ef016f7a29a38c38d 28138Author: Hui YU <hyu@ati.com> 28139Date: Fri Jul 30 22:20:21 2004 +0000 28140 28141 Support for New radeon chips: R420/M18, R423, RV370/M22, RV380/M24, RS300. 28142 Add special handlings for DELL triple-head server (RV100). Misc. bug 28143 fixes for flat panel, host aperture, etc (Bug #946) 28144 28145commit d04f61bfaf6952e832d52e1cd6014e86435aebd9 28146Author: Adam Jackson <ajax@nwnk.net> 28147Date: Fri Jul 30 20:30:51 2004 +0000 28148 28149 Bug #400 (partial): Driver fixes for the dlloader. When using dlloader, all 28150 framebuffer formats except cfb and the overlay modes should work, and 28151 r128 and radeon need to be loaded from the ati driver (both issues to 28152 be fixed soon). Tested on i740, s3virge, mach64, tdfx, vesa, and vga 28153 drivers. elfloader users shouldn't be affected. 28154 28155commit 6dcb78bb818c3c1c7fe406bee5379f9bef023f4b 28156Author: Alex Deucher <agd5f@yahoo.com> 28157Date: Wed Jul 28 02:35:30 2004 +0000 28158 28159 - add gamma correction support to the radeon video overlay 28160 28161commit 26029dc9378bbd95835d397402bfc9e6b4f1a478 28162Author: Alex Deucher <agd5f@yahoo.com> 28163Date: Mon Jul 26 23:14:37 2004 +0000 28164 28165 - Add Radeon DynamicClocks option 28166 - Add small fixes and clean ups from ati's last code drop (typo_fixes, 28167 remove_fudge, laptop, xvfix) 28168 - fix possible segfault in mga_dri.c (Ryan Underwood) 28169 - Add Xv support to pre-nm2160 neomagic chipsets 28170 28171commit 9f251b16e702da486f824307e72a6f80c6987b5d 28172Author: Adam Jackson <ajax@nwnk.net> 28173Date: Mon Jul 26 19:14:58 2004 +0000 28174 28175 Bug #848: Clean up gatos notice messages. 28176 28177commit 6f1e01ecdb11913fd768b9918f516b8379b7858f 28178Author: Eric Anholt <anholt@freebsd.org> 28179Date: Fri Jul 23 05:36:14 2004 +0000 28180 28181 - Add a new option, BuildDevelDRIDrivers, which controls whether 28182 DevelDRIDrivers is non-empty. DevelDRIDrivers contains those drivers 28183 aren't built by default either due to being unusable on that 28184 architecture (but buildable for testing) or due to security concerns, 28185 as in the mach64 and savage cases. 28186 - Merge the Mach64 DDX's DRI support, under the XF86DRI_DEVEL define, which 28187 is set only when BuildDevelDRIDrivers is set. 28188 - Fix up of the sets of DRI drivers built for various architectures. 28189 - Fix build of unichrome DRI driver, which is now enabled on x86. 28190 28191commit b68322f467b7e524092af15e33ab872abb99b6a2 28192Author: Eric Anholt <anholt@freebsd.org> 28193Date: Thu Jul 22 07:03:04 2004 +0000 28194 28195 Merge and resolve conflicts from DRI/DRM/Mesa 20040721 imports. 28196 28197commit 57de5e285d6271fc4b2bc8a951a98be6308d649d 28198Author: Eric Anholt <anholt@freebsd.org> 28199Date: Fri Jul 16 04:38:54 2004 +0000 28200 28201 Bugzilla #803,804: Fix build in the !XF86DRI case after DRI and Radeon 28202 Render acceleration merges. 28203 28204commit 9b0ec803bd588d9af778bfd307ee6d958e4a1e73 28205Author: Adam Jackson <ajax@nwnk.net> 28206Date: Sat Jul 10 14:22:35 2004 +0000 28207 28208 Bugzilla #849: Fix DRI pagesize assumptions in radeon and r128 drivers. 28209 from mharris@, reviewed and applied to DRI by airlied@. 28210 28211commit ccd95bfb291a0c70272fab6822ef379b83dc298d 28212Author: Egbert Eich <eich@suse.de> 28213Date: Tue Jul 6 14:37:47 2004 +0000 28214 28215 Separated Intel drivers from default DriDrivers to avoid building them on 28216 IA64 (Egbert Eich). 28217 Fixed wrong function prototype (Egbert Eich). 28218 Don't test for generic VGA on IA64 (Egbert Eich). 28219 Fixed a segfault when accessing a structure before verifying the pointer 28220 exists (Egbert Eich). 28221 Added a showcache option for debugging (Egbert Eich). 28222 Increase default video RAM size to 16MB when DRI is enabled and more than 28223 128MB are available (Egbert Eich). Fixed lockups during mode switch. 28224 Problem was introduced when attempting to copy the behavior during 28225 LeaveVT()/EnterVT() but but forgetting to call I810DRILeave() before 28226 I810DRIEnter(). The entire DRILeave()/Enter() scenario has been 28227 commented out as it didn't seem to be necessary (Egbert Eich). 28228 Fix TweakMemorySize() (tested with i855/i865) (Egbert Eich). 28229 increased MAX_DEVICES to 128 (Egbert Eich). 28230 Use OS provided PCI config space access as default method (Egbert Eich). 28231 Added support for Linux 2.6 proc file format. 28232 Fixed unaligned accesses to pieces of the VBE info block. VESA did not 28233 align elements to size (Egbert Eich). 28234 28235commit 6bdf3d36ea3086432af5157540fde4e0aa0c44dc 28236Author: Eric Anholt <anholt@freebsd.org> 28237Date: Sun Jun 20 07:34:55 2004 +0000 28238 28239 Add file missed in last commit: Bug #748: 28240 - Add Render acceleration for Radeon 100 and 200-series cards, enabled by 28241 default. Notable performance gains include 27fps in cairogears instead 28242 of 6fps (compared to 50 with glitz), and my disks are now the 28243 bottleneck in an ls -lR in gnome-terminal. Only supported in DRI mode 28244 because the MMIO submission hangs the card so far, but the code is left 28245 in because it may be supportable soon. 28246 - Add xorg.conf options to disable Render acceleration and to force the 28247 subpixel order in the server (Hui Yu). 28248 Many thanks to Hui Yu at ATI for the code this was based off of, Michel 28249 Daenzer for bugfixes and suggestions, and LinuxFund for sponsoring the 28250 work. 28251 28252commit 8c1b37420ea890f76edb97aed61ee427a871cd3f 28253Author: Eric Anholt <anholt@freebsd.org> 28254Date: Sun Jun 20 07:26:08 2004 +0000 28255 28256 Bug #748: 28257 - Add Render acceleration for Radeon 100 and 200-series cards, enabled by 28258 default. Notable performance gains include 27fps in cairogears instead 28259 of 6fps (compared to 50 with glitz), and my disks are now the 28260 bottleneck in an ls -lR in gnome-terminal. Only supported in DRI mode 28261 because the MMIO submission hangs the card so far, but the code is left 28262 in because it may be supportable soon. 28263 - Add xorg.conf options to disable Render acceleration and to force the 28264 subpixel order in the server (Hui Yu). 28265 Many thanks to Hui Yu at ATI for the code this was based off of, Michel 28266 Daenzer for bugfixes and suggestions, and LinuxFund for sponsoring the 28267 work. 28268 28269commit 6ecf374d500afe6da494dfdd6566396ec65b6d6a 28270Author: Eric Anholt <anholt@freebsd.org> 28271Date: Wed Jun 16 09:43:59 2004 +0000 28272 28273 Merge DRI-trunk-20040613 changes in programs/Xserver/hw/xfree86/drivers, 28274 with the following notes: 28275 - Savage and Mach64 (= ati/ati*.[ch] changes) DRI not merged due to 28276 insecurity. 28277 - VIA driver converted to new drmContext and drmHandle names. 28278 - Radeon driver merge conflicted in many places, and MergedFB at least 28279 could probably use some checking at this point. 28280 28281commit bea8085e04136b0ef513c17bb65c54069ec531e1 28282Author: Eric Anholt <anholt@freebsd.org> 28283Date: Wed Jun 16 09:25:58 2004 +0000 28284 28285 DRI trunk-20040613 import 28286 28287commit 7b588cda922992a8ee2d04853ba1533cf43592a3 28288Author: Eric Anholt <anholt@freebsd.org> 28289Date: Wed Jun 16 09:25:58 2004 +0000 28290 28291 Initial revision 28292 28293commit 6cbb5bbff43d65a762d432659793333868837f0b 28294Author: Eric Anholt <anholt@freebsd.org> 28295Date: Wed Jun 16 09:22:56 2004 +0000 28296 28297 DRI XFree86-4_3_99_12-merge import 28298 28299commit 3586c3c7955365a2af0b8a286ea0d0488ddfd137 28300Author: Egbert Eich <eich@suse.de> 28301Date: Fri Apr 23 19:26:46 2004 +0000 28302 28303 Merging XORG-CURRENT into trunk 28304 28305commit 538768e0f472026325be9dd37dc4bfe7ea7d3b75 28306Author: Egbert Eich <eich@suse.de> 28307Date: Sun Mar 14 08:33:20 2004 +0000 28308 28309 Importing vendor version xf86-4_4_99_1 on Sun Mar 14 00:26:39 PST 2004 28310 28311commit 5f179e796d1cd310790d06eebb474ad67114ec84 28312Author: Egbert Eich <eich@suse.de> 28313Date: Wed Mar 3 12:12:18 2004 +0000 28314 28315 Importing vendor version xf86-4_4_0 on Wed Mar 3 04:09:24 PST 2004 28316 28317commit 493c44580191beff63575c4032f772fbbf1b06eb 28318Author: Egbert Eich <eich@suse.de> 28319Date: Thu Feb 26 13:35:52 2004 +0000 28320 28321 readding XFree86's cvs IDs 28322 28323commit 942a47fd3c5e8df85642672c01072bf97285bc30 28324Author: Egbert Eich <eich@suse.de> 28325Date: Thu Feb 26 09:23:18 2004 +0000 28326 28327 Importing vendor version xf86-4_3_99_903 on Wed Feb 26 01:21:00 PST 2004 28328 28329commit dda79b5ee3eb49f924a3213592d93ba981c99394 28330Author: Kaleb Keithley <kaleb@freedesktop.org> 28331Date: Mon Feb 23 20:35:02 2004 +0000 28332 28333 Import most of XFree86 4.4RC3. This import excludes files which have the 28334 new license. If we want to, later we can import 4.4RC3 again and pick 28335 up the files that have the new license, but for now the vendor branch 28336 is "pure." 28337 28338commit 08104f18ac497561c969694ff709da6c361ccea7 28339Author: Egbert Eich <eich@suse.de> 28340Date: Thu Jan 29 08:08:34 2004 +0000 28341 28342 Importing vendor version xf86-012804-2330 on Thu Jan 29 00:06:33 PST 2004 28343 28344commit 162d76b9c48ed17d6e645057db58c9f13436f25f 28345Author: Kaleb Keithley <kaleb@freedesktop.org> 28346Date: Fri Dec 19 20:55:06 2003 +0000 28347 28348 XFree86 4.3.99.902 (RC 2) 28349 28350commit 195772ec74f7a9b7d5bf294b3bc2c3b960e08a31 28351Author: Kaleb Keithley <kaleb@freedesktop.org> 28352Date: Thu Dec 4 22:03:16 2003 +0000 28353 28354 XFree86 4.3.99.901 (RC 1) 28355 28356commit 73105e438e90f950249cb265ba99001d4db3dc60 28357Author: Kaleb Keithley <kaleb@freedesktop.org> 28358Date: Tue Nov 25 19:28:36 2003 +0000 28359 28360 XFree86 4.3.99.16 Bring the tree up to date for the Cygwin folks 28361 28362commit 0a53864f9581196604d04fd28b4e94a9b0b73d18 28363Author: Kaleb Keithley <kaleb@freedesktop.org> 28364Date: Tue Nov 25 19:28:36 2003 +0000 28365 28366 Initial revision 28367 28368commit c56aaef79028261e0a3be0c86ffd3f16b6bdd9ba 28369Author: Kaleb Keithley <kaleb@freedesktop.org> 28370Date: Fri Nov 14 16:48:55 2003 +0000 28371 28372 XFree86 4.3.0.1 28373 28374commit d09f463a5d1ce73e0b65d5276fbcca393fa2da46 28375Author: Kaleb Keithley <kaleb@freedesktop.org> 28376Date: Fri Nov 14 16:48:55 2003 +0000 28377 28378 Initial revision 28379 28380commit d9af6dc32652502d84ea8da5d57a5ab45429c4ad 28381Author: Kaleb Keithley <kaleb@freedesktop.org> 28382Date: Fri Nov 14 15:54:49 2003 +0000 28383 28384 R6.6 is the Xorg base-line 28385