ChangeLog revision de2362d3
1commit 068a59e010ce6bfcd54f5a18cc08c55c54b8618d 2Author: Alex Deucher <alexander.deucher@amd.com> 3Date: Wed Oct 1 23:38:52 2014 -0400 4 5 radeon: bump version for release 6 7 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 8 9commit 5fee565e420d3efbf72fdf4f63c3d5d93d8ceddb 10Author: Alex Deucher <alexander.deucher@amd.com> 11Date: Wed Oct 1 23:36:31 2014 -0400 12 13 radeon: require libdrm_radeon 2.4.58 for latest SI/CI pci ids 14 15 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 16 17commit 56c7fb8ada4cd9cf096c6b06a8d368d286d74e68 18Author: Thomas DeRensis <tderensis@gmail.com> 19Date: Wed Sep 10 21:20:32 2014 -0400 20 21 radeon: fix build warnings regarding const qualifier 22 23 Signed-off-by: Thomas DeRensis <tderensis@gmail.com> 24 25commit b9a9b244336ed50d7df4f36135f875a3bb6ca948 26Author: Thomas DeRensis <tderensis@gmail.com> 27Date: Thu Sep 4 21:45:19 2014 -0400 28 29 radeon: fix a leak in radeon_vbo_get_bo() 30 31 The dma_bo struct was never freed in the error path. Reported by cppcheck. 32 33 Signed-off-by: Thomas DeRensis <tderensis@gmail.com> 34 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 35 36commit ba5fcc4d6b6eae3f2c55c6229a53d24d160789c1 37Author: Alex Deucher <alexander.deucher@amd.com> 38Date: Thu Aug 21 11:32:45 2014 -0400 39 40 radeon/kms: fix mullins pci id 41 42 strange errant change from: 43 39fef269f521c92a2a31c80447e9401bacb3797a 44 45 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 46 47commit efef34930650d6d80f7b527f4cee76d9e5954ace 48Author: Alex Deucher <alexander.deucher@amd.com> 49Date: Thu Aug 21 11:30:17 2014 -0400 50 51 radeon/kms: add new SI pci ids 52 53 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 54 55commit 39fef269f521c92a2a31c80447e9401bacb3797a 56Author: Alex Deucher <alexander.deucher@amd.com> 57Date: Thu Aug 21 11:27:49 2014 -0400 58 59 radeon/kms: add new CIK pci ids 60 61 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 62 63commit 7dab6b2fe9cf690b0dae1c127d0374322bd8278e 64Author: Michel Dänzer <michel.daenzer@amd.com> 65Date: Wed Aug 20 17:51:35 2014 +0900 66 67 Revert "glamor: Set environment variable RADEON_THREAD=0" 68 69 This reverts commit 4b5060f357a3cb248c9359c92c1e9c42ef6434c8. 70 71 Further testing shows that disabling the thread is only a moderate win in 72 some cases, but a much bigger loss in some other cases. 73 74 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 75 76commit 1ff383360886c5cee1bb75abcc20675aca0336a6 77Author: Alex Deucher <alexander.deucher@amd.com> 78Date: Tue Aug 12 12:27:12 2014 -0400 79 80 radeon: fix warnings when building against older xservers 81 82 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 83 84commit 94202cbfbca05a503acdc1cca2f8409d141173af 85Author: Alex Deucher <alexdeucher@gmail.com> 86Date: Fri Aug 1 21:55:40 2014 +0200 87 88 radeon: enable hawaii accel conditionally (v3) 89 90 Only if the kernel has the new CP firmware. 91 92 v2: check value of ACCEL_WORKING2 93 v3 (Andreas Boll): 94 - check for value 2 or 3 of ACCEL_WORKING2 95 - update man page 96 97 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> (v2) 98 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 99 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 100 101commit 03930edd49f6b8a8d79910c7be5408b47db9649b 102Author: Andreas Boll <andreas.boll.dev@gmail.com> 103Date: Mon Aug 4 16:23:13 2014 +0200 104 105 radeon: remove definitions already present in radeon_drm.h 106 107 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 108 109commit 91849fba0742ef61ba327e71fc3ce8f754af0a6f 110Author: Andreas Boll <andreas.boll.dev@gmail.com> 111Date: Mon Aug 4 16:23:12 2014 +0200 112 113 radeon: drop radeon_drm.h 114 115 Now we use libdrm's radeon_drm.h. 116 117 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 118 119commit b4658901bf1d619f8ff9b5f94344894f935ee6c8 120Author: Andreas Boll <andreas.boll.dev@gmail.com> 121Date: Mon Aug 4 16:23:11 2014 +0200 122 123 radeon: move RADEON_TILING_{MASK, LINEAR} from radeon_drm.h to radeon.h 124 125 This allows us to drop radeon_drm.h from xf86-video-ati and use instead 126 radeon_drm.h from libdrm. 127 128 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 129 130commit 656bae7361c1e018553ef6d6d8c9efad616a4513 131Author: Andreas Boll <andreas.boll.dev@gmail.com> 132Date: Mon Aug 4 16:23:10 2014 +0200 133 134 radeon: drop redundant radeon_drm.h includes 135 136 Already included via radeon.h. 137 138 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 139 140commit fbf575cb010e558a87fad0de45660738c8180896 141Author: Michel Dänzer <michel.daenzer@amd.com> 142Date: Wed Jul 30 17:03:24 2014 +0900 143 144 Add Emacs .dir-local.el file 145 146 Based on the one from the Gallium radeon winsys, but enabling tabs for 147 indentation. 148 149commit 4b5060f357a3cb248c9359c92c1e9c42ef6434c8 150Author: Michel Dänzer <michel.daenzer@amd.com> 151Date: Thu Jun 19 18:27:59 2014 +0900 152 153 glamor: Set environment variable RADEON_THREAD=0 154 155 Reviewed-by: Marek Olšák <marek.olsak@amd.com> 156 157commit 9b54caf6509a9c02dd17c9c43d6be8f7ddc98054 158Author: Hans de Goede <hdegoede@redhat.com> 159Date: Mon Jul 28 15:55:12 2014 +0200 160 161 configure: Include xorg-server.h before 162 glamor.h 163 164 glamor.h cannot be included without first including xorg-server.h, this also 165 applies to including it from configure snippets. 166 167 Without this the configure glamor checks fail on systems with the latest 168 glibc, throwing this error: 169 170 In file included from /usr/include/xorg/misc.h:115:0, 171 from /usr/include/xorg/screenint.h:50, 172 from /usr/include/xorg/scrnintstr.h:50, 173 from /usr/include/xorg/glamor.h:32, 174 from conftest.c:61: 175 /usr/include/xorg/os.h:579:2: error: expected identifier or '(' before 176 '__exten 177 strndup(const char *str, size_t n); 178 ^ 179 180 This is caused by HAVE_STRNDUP not being set (it is set from xorg-server.h), 181 causing os.h to redefine it. 182 183 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 184 185commit c4ae0e2cbcc0e2ebf9f13ee92d59b5120254a1dc 186Author: Michel Dänzer <michel.daenzer@amd.com> 187Date: Mon Jun 30 10:20:12 2014 +0900 188 189 Handle CRTC DPMS from output DPMS hooks 190 191 This fixes at least two issues: 192 193 The CRTC DPMS hook isn't called after a modeset, so the vertical blank 194 interrupt emulation code considered the CRTC disabled after a modeset. As 195 a side effect, page flipping was no longer used after a modeset. 196 197 This change also makes sure the vertical blank interrupt emulation code 198 runs before the hardware CRTC is disabled and after it's enabled from the 199 output DPMS hook. The wrong order could cause gnome-shell to hang after 200 a suspend/resume and/or DPMS off/on cycle. 201 202 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 203 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 204 205commit c0c3cac8613e31e310242695d0955b452f116e25 206Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 207Date: Wed Jun 25 16:21:57 2014 +0200 208 209 bump version post release 210 211commit 906a0ec9224146098bb4581486129d2934d36495 212Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 213Date: Wed Jun 25 16:10:21 2014 +0200 214 215 bump version for release 216 217commit cc615d06db0332fc6e673b55632bcc7bf957b44b 218Author: Michel Dänzer <michel.daenzer@amd.com> 219Date: Wed Jun 4 16:58:53 2014 +0900 220 221 Rename Option "NoAccel" to "Accel" 222 223 Removes the need for a double negation when forcing acceleration on. 224 225 Note that this change is backwards compatible, as the option parser 226 automagically handles the 'No' prefix. 227 228 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 229 230commit 851b2cf8714618843725f6d067915375485ade9d 231Author: Adam Jackson <ajax@redhat.com> 232Date: Wed May 21 09:34:32 2014 -0400 233 234 kms: Use own thunk function instead of shadowUpdatePackedWeak 235 236 I plan to delete the Weak functions from a future server. 237 238 Signed-off-by: Adam Jackson <ajax@redhat.com> 239 240commit b2dba2906f0b2284f17f53fd5251ba0f03d52a8b 241Author: Michel Dänzer <michel.daenzer@amd.com> 242Date: Thu May 15 16:07:53 2014 +0900 243 244 Don't disable acceleration on >= SI on attempts to force EXA 245 246 Also make this case clear in the log file: 247 248 (WW) RADEON(0): EXA not supported, using glamor 249 250 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 251 252commit be1469cc23aba46daf3293b3d09c5f2e792e7f42 253Author: Alex Deucher <alexander.deucher@amd.com> 254Date: Fri May 2 17:48:07 2014 -0400 255 256 radeon: enable tiling for mullins 257 258 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 259 260commit 2ae8e4b8d12f5c9bab6655eb8cd3c5c1d5cfb10e 261Author: Samuel Li <samuel.li@amd.com> 262Date: Tue Nov 12 15:30:42 2013 -0500 263 264 radeon: add Mullins pci ids. 265 266 Signed-off-by: Samuel Li <samuel.li@amd.com> 267 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 268 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 269 270commit 5c86a3461597647224c33d5190b4531aeeb2655f 271Author: Samuel Li <samuel.li@amd.com> 272Date: Thu Apr 17 15:17:28 2014 -0400 273 274 radeon: add support for Mullins. 275 276 Signed-off-by: Samuel Li <samuel.li@amd.com> 277 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 278 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 279 280commit fdb90ffc50acbb7d5ba0598470f9feeac6ce55fc 281Author: Alex Deucher <alexander.deucher@amd.com> 282Date: Fri May 2 17:45:45 2014 -0400 283 284 radeon: require libdrm_radeon 2.4.54 for mullins support 285 286 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 287 288commit 06e3c8c53ef4bd159f5864eabf726438d008b49a 289Author: Dave Airlie <airlied@redhat.com> 290Date: Wed Apr 23 13:39:42 2014 +1000 291 292 radeon: fix use-after-free in modesetting cleanup 293 294 noticed while looking at something else. 295 296 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 297 Signed-off-by: Dave Airlie <airlied@redhat.com> 298 299commit dbac18c361f9e514ecb40d0617f9d68b65a542e0 300Author: Michel Dänzer <michel.daenzer@amd.com> 301Date: Mon Apr 28 17:51:56 2014 +0900 302 303 Revert "Adapt to load_cursor_argb signature change in xserver 1.15.99.902" 304 305 This reverts commit 48d3dbc8a0d3bfde88f46e402e530438f9317715. 306 307 xserver Git has been updated to be backwards compatible with the 308 previous API. 309 310commit c84230d686c078aac1dc98d82153f8b02521b2e1 311Author: Michel Dänzer <michel.daenzer@amd.com> 312Date: Fri Apr 25 09:17:51 2014 +0900 313 314 dri2: Handle PRIME for source buffer as well in radeon_dri2_copy_region2 315 316 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=77810 317 318 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 319 320commit 48d3dbc8a0d3bfde88f46e402e530438f9317715 321Author: Michel Dänzer <michel.daenzer@amd.com> 322Date: Tue Apr 15 17:45:35 2014 +0900 323 324 Adapt to load_cursor_argb signature change in xserver 1.15.99.902 325 326 Apart from the compiler warning below, not doing this may result in 327 accidentally using software cursors. 328 329 ../../src/drmmode_display.c:808:5: warning: initialization from incompatible pointer type [enabled by default] 330 .load_cursor_argb = drmmode_load_cursor_argb, 331 ^ 332 ../../src/drmmode_display.c:808:5: warning: (near initialization for 'drmmode_crtc_funcs.load_cursor_argb') [enabled by default] 333 334 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 335 336commit 409786a8f780d78a83bf0bddea5d37117ff6fa39 337Author: Michel Dänzer <michel.daenzer@amd.com> 338Date: Thu Apr 10 11:43:04 2014 +0900 339 340 glamor: Fix test for creating shared pixmaps 341 342 The pixmap usage hint is not a bitmask in general. The test for 343 CREATE_PIXMAP_USAGE_SHARED was incorrectly triggering for a glamor internal 344 usage hint being added in the xserver tree. 345 346 Tested-by: Ed Tomlinson <edtoml@gmail.com> 347 348commit aecf1c4e5f4718adcfb85836830d065d3f4f97a5 349Author: Michel Dänzer <michel.daenzer@amd.com> 350Date: Thu Apr 10 15:10:56 2014 +0900 351 352 dri2: Fix conflicting CreatePixmap usage flag definitions 353 354 RADEON_CREATE_PIXMAP_TILING_MICRO_SQUARE was the same as 355 RADEON_CREATE_PIXMAP_DRI2. 356 357 Disambiguate the definitions and rearrange them to try and prevent this 358 from happening again. 359 360 Tested-by: Ed Tomlinson <edtoml@gmail.com> 361 362commit b50da3b96c212086cb58501dbe988d64f1f35b6d 363Author: Hans de Goede <hdegoede@redhat.com> 364Date: Fri Apr 11 09:44:37 2014 +0200 365 366 Fix building on older servers without xf86platformBus.h 367 368 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 369 370commit ed0cfbb4fe77146b0b38f777bc28f3a4ea6da07f 371Author: Hans de Goede <hdegoede@redhat.com> 372Date: Fri Mar 7 13:27:30 2014 +0100 373 374 Add support for server managed fds 375 376 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 377 378commit 3d7861fe112f25874319d4cdc12b745fbcd359cf 379Author: Hans de Goede <hdegoede@redhat.com> 380Date: Mon Mar 17 10:38:13 2014 +0100 381 382 Add radeon_get_drm_master_fd helper function 383 384 This is a preparation patch for adding server-managed-fd support without it 385 turning into a goto fest. 386 387 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 388 389commit a63342ad15408071437c80b411d14196f3288aed 390Author: Hans de Goede <hdegoede@redhat.com> 391Date: Mon Mar 17 10:36:55 2014 +0100 392 393 radeon_open_drm_master get rid of unnecessary goto 394 395 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 396 397commit bdc412044f6ced056cd57320d1b2ee0d967c2191 398Author: Michel Dänzer <michel.daenzer@amd.com> 399Date: Thu Mar 13 16:40:19 2014 +0900 400 401 Build against glamor in the xserver tree if available 402 403commit 921a153f9964ca452e1241f76c7f7d653f42ceaf 404Author: Alex Deucher <alexander.deucher@amd.com> 405Date: Tue Mar 11 12:24:13 2014 -0400 406 407 update man page to reflect tiling changes for CI parts 408 409 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 410 411commit 5eee8a4d5c86bb1cc34d8caf2f2b64b53c241fa5 412Author: Alex Deucher <alexander.deucher@amd.com> 413Date: Mon Mar 10 16:20:09 2014 -0400 414 415 radeon: enable tiling by default on CIK 416 417 Now that mesa 10.1 is released, we can enable this by 418 default for CIK parts. Tiling improves memory bandwidth 419 utilization. 420 421 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 422 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 423 424commit 691ec3d99c30111a4789830dfccb6eb5d3c40187 425Author: Alex Deucher <alexander.deucher@amd.com> 426Date: Mon Mar 10 16:17:34 2014 -0400 427 428 radeon: require libdrm 2.4.51 429 430 Required for proper tiling support on CIK parts. 431 432 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 433 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 434 435commit 515bcf14d514f9dcaaf30fd0bf1ef6dd6ba9a0cd 436Author: Michel Dänzer <michel.daenzer@amd.com> 437Date: Tue Mar 4 12:34:26 2014 +0900 438 439 Allow enabling glamor on R500 (and R300) class 3D engines as well. 440 441 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=75709 442 443 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 444 445commit ea6d0affe52d82becadab6fb1c87f9261b0605a2 446Author: Michel Dänzer <michel.daenzer@amd.com> 447Date: Mon Feb 24 13:12:21 2014 +0900 448 449 Only log debugging output about initializing colormaps when we're doing so 450 451 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 452 453commit 8da17f30c70f4494ce22ad781a1cee17041812f3 454Author: Jérôme Glisse <jglisse@redhat.com> 455Date: Mon Feb 24 14:50:25 2014 -0500 456 457 evergreen: fix shader constant upload on ppc 458 459 The number of dword we have to swap is (16*4) ie 16 vectors of 460 4 floats each not 16 floats. Never hit this issue before because 461 we never had more than 4 constant vector. 462 463 Signed-off-by: Jérôme Glisse <jglisse@redhat.com> 464 465commit cadb6b493942a84bfeb298751dce0dee39257a06 466Author: Alex Deucher <alexander.deucher@amd.com> 467Date: Fri Feb 21 08:33:21 2014 -0500 468 469 radeon: don't install colormap handling if there are no crtcs 470 471 Fixes a crash on cards with 0 crtcs. 472 473 Discussion: 474 http://lists.freedesktop.org/archives/dri-devel/2014-February/054186.html 475 476 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 477 478commit 7cd972a85705341dd8306eefc558ed9e5def05d7 479Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 480Date: Thu Feb 20 11:14:03 2014 +0100 481 482 return immediately in preinit when called with PROBE_DETECT 483 484 This fixes a crash with Xorg -configure. 485 486 Bug: 487 https://bugs.launchpad.net/ubuntu/+source/xorg/+bug/1278046 488 489 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 490 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 491 492commit 8de6f7b2f476d3baa9c7e2bb3544e4bafaad46b7 493Author: Alex Deucher <alexander.deucher@amd.com> 494Date: Fri Jan 24 11:04:30 2014 -0500 495 496 bump version post release 497 498 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 499 500commit 0333f5bda27dc0ec2edc180c7a4dc9a432f13f97 501Author: Alex Deucher <alexander.deucher@amd.com> 502Date: Fri Jan 24 10:19:49 2014 -0500 503 504 radeon: bump version for release 505 506 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 507 508commit fc4167f2a85d9cba65078d8fc6f08c7a619ad66e 509Author: Alex Deucher <alexander.deucher@amd.com> 510Date: Fri Jan 24 10:17:08 2014 -0500 511 512 Require glamor 0.6.0 513 514 This is required for Xv support and a number of important 515 performance improvements. 516 517 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 518 519commit f2a0a5cf6c5a21e2a02280e110a4eb8e6609dace 520Author: Michel Dänzer <michel.daenzer@amd.com> 521Date: Wed Jan 22 11:04:42 2014 +0900 522 523 Don't require the glamoregl module to be pre-loaded with xserver >= 1.15 524 525 The issues with loading it on demand have been fixed in xserver 1.15. 526 527 Inspired by Jérôme Glisse on IRC. 528 529commit 3213df16d61302148be0088c8f93c6a5a88558f1 530Author: Michel Dänzer <michel.daenzer@amd.com> 531Date: Wed Jan 8 11:30:59 2014 +0900 532 533 dri2: Make last_vblank_seq local unsigned to match dpms_last_seq 534 535 Without this, I was occasionally running into gnome-shell hangs due to 536 wildly off vblank sequence values. Doesn't seem to happen anymore with 537 this change. 538 539 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 540 541commit bcc454ea2fb239e13942270faec7801270615b9c 542Author: Alex Deucher <alexander.deucher@amd.com> 543Date: Mon Jan 6 09:52:50 2014 -0500 544 545 radeon/exa: Always use a scratch surface for UTS to vram 546 547 If we don't, we may hit a buffer that crosses the 548 visible vram boundary resulting in a sigbus when the 549 CPU accesses the buffer beyond the PCI aperture. 550 This will introduce an extra copy in certain cases. 551 552 This is based on Michel's patch from bug 44099 updated 553 for all asic families. 554 555 Bug: 556 https://bugs.freedesktop.org/show_bug.cgi?id=44099 557 558 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 559 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 560 561commit 04ef035c9315b4a6fbf1b14720be87cee4099a9f 562Author: Alex Deucher <alexander.deucher@amd.com> 563Date: Tue Dec 24 15:14:35 2013 -0500 564 565 drm/radeon: fix SUMO2 pci id 566 567 0x9649 is sumo2, not sumo. 568 569 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 570 571commit bfbff3b246db509c820df17b8fcf5899882ffcfa 572Author: Robert Millan <rmh@freebsd.org> 573Date: Fri Dec 20 11:03:14 2013 +0000 574 575 radeon: Restore kernel module load on FreeBSD. 576 577 Since the introduction of a call to drmCheckModesettingSupported() 578 in radeon_kernel_mode_enabled(), with abort condition if such call 579 fails, the drmOpen() call in radeon_open_drm_master() no longer 580 takes the responsibility of loading the radeon kernel module. 581 582 However at least on FreeBSD (and GNU/kFreeBSD), X is still relied 583 on to load the modules it needs. This commit restores the old 584 behaviour of loading kernel modules on these systems. 585 586 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=72852 587 Signed-off-by: Robert Millan <rmh@freebsd.org> 588 589commit 796c9a0cb587f528326bede11fa3f3eb7d3edaf1 590Author: Robert Millan <rmh@freebsd.org> 591Date: Fri Dec 20 10:55:09 2013 +0000 592 593 radeon: Set first parameter of drmOpen() to NULL 594 595 Since the introduction of a call to drmCheckModesettingSupported() 596 in radeon_kernel_mode_enabled(), with abort condition if such call 597 fails, the first argument to drmOpen() call in radeon_open_drm_master() 598 has become a no-op red herring. 599 600 Such argument (a kernel module name) is supposed to result in load 601 of specified kernel module. However, this will never happen. The 602 problem is that if the code containing drmOpen() call is reached, it 603 means that drmCheckModesettingSupported() check has previously 604 succeeded, which implies the module is already loaded. 605 606 So, drmOpen() will never load a kernel module. But it gives the 607 impression that it will. 608 609 In order to avoid this confusion, this commit replaces it with NULL, 610 like xf86-video-intel driver does. 611 612 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=72849 613 Signed-off-by: Robert Millan <rmh@freebsd.org> 614 615commit d571d6af70ef27efd1ed6420eb892bdde963ed7a 616Author: Alex Deucher <alexander.deucher@amd.com> 617Date: Tue Sep 24 11:39:10 2013 -0400 618 619 radeon/kms: add Hawaii pci ids 620 621 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 622 623commit e38a92e00d015a6b80a1f3a16d58c61f084b066f 624Author: Alex Deucher <alexander.deucher@amd.com> 625Date: Tue Sep 24 11:35:00 2013 -0400 626 627 radeon: add support for Hawaii 628 629 Disabled by default until the acceleration code stablizes. 630 631 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 632 633commit e4cd0f4392ea11c93088ad429f36eaaf9bcbf505 634Author: Alex Deucher <alexander.deucher@amd.com> 635Date: Wed Oct 30 15:21:26 2013 -0400 636 637 radeon: enable tiling on SI by default (v2) 638 639 Now that mesa 9.2 is out with support for tiling 640 on SI asics, we can enable it here. Tiling improves 641 memory bandwidth utilization. 642 643 V2: update man page 644 645 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 646 647commit 3b38701a72fa1cad1e4610a2f4330b3da4cc6391 648Author: Vadim Girlin <vadimgirlin@gmail.com> 649Date: Fri Nov 1 10:36:39 2013 -0400 650 651 radeon: disable 2D tiling on buffers < 128 pixels 652 653 Seems to run into alignment problems with certain 654 card configurations. 655 656 bug: 657 https://bugs.freedesktop.org/show_bug.cgi?id=70675 658 659 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 660 661commit 0c921edf0162fed616cea9d02e168b719243bcd2 662Author: Jerome Glisse <jglisse@redhat.com> 663Date: Wed Oct 30 13:01:02 2013 -0400 664 665 radeon/glamor: with new pixmap for dri2 drawable no need to create new texture 666 667 When creating a new pixmap/bo for dri2 drawable there is no need to create a 668 new texture, instead the texture associated with the new pixmap should be use 669 otherwise there is a missmatch between the bo backing the texture for the 670 drawable and the bo used by dri2 client. 671 672 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 673 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 674 675commit f1dc677e79cd7a88d7379a934ebc7d87a3b18805 676Author: Christopher James Halse Rogers <raof@ubuntu.com> 677Date: Mon Sep 23 12:25:29 2013 -0700 678 679 EXA/evergreen: Paranoia around linear tiling. (v2) 680 681 The last two bytes of tiling_mode contain the actual tiling mode; the rest are 682 extra tiling configuration bits. These configuration bits are not necessarily 683 zero for a linear buffer, so mask them out before checking for linearity 684 685 v2: Also fix up evergreen_textured_videofuncs.c 686 687 agd5f: remove trailing whitespace 688 689 Signed-off-by: Christopher James Halse Rogers <raof@ubuntu.com> 690 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 691 692commit 67fb82a3f0759b171fea21b475a70fa825693570 693Author: Alex Deucher <alexander.deucher@amd.com> 694Date: Tue Oct 1 09:35:30 2013 -0400 695 696 radeon: fix the non-glamor build harder... 697 698 I need to stop pushing patches first thing in the morning. 699 700 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 701 702commit afc0374fdec3e24ece25805724459463e9a19f5e 703Author: Alex Deucher <alexander.deucher@amd.com> 704Date: Tue Oct 1 09:32:02 2013 -0400 705 706 drm/radeon: fix non-glamor build 707 708 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 709 710commit 2d791370dfc5570eb74d7a1fb3baf4d4c8ecf243 711Author: Dave Airlie <airlied@redhat.com> 712Date: Mon Sep 23 07:57:15 2013 +0100 713 714 radeon: use glamor Xv support if present. 715 716 This creates adaptors using glamor if possible. 717 718 Signed-off-by: Dave Airlie <airlied@redhat.com> 719 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 720 721commit cbb99f659ee7b18ded0008a606e41ded38c1a194 722Author: Alex Deucher <alexander.deucher@amd.com> 723Date: Mon Sep 30 08:55:16 2013 -0400 724 725 Revert "radeon: add glamor Xv support (v2)" 726 727 This causes problems if glamor doesn't have Xv support 728 enabled. I just noticed that Dave has a better version, 729 so use that instead. 730 731 This reverts commit 4fc1fa920584ace2c84d75af82d06962d0c84ec8. 732 733commit 4fc1fa920584ace2c84d75af82d06962d0c84ec8 734Author: Alex Deucher <alexander.deucher@amd.com> 735Date: Thu Sep 26 11:02:55 2013 -0400 736 737 radeon: add glamor Xv support (v2) 738 739 v2: guard new glamor Xv bits with USE_GLAMOR 740 741 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 742 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 743 744commit 41dfe327ac8740ac2cd84def96b5947224e422e7 745Author: Alex Deucher <alexander.deucher@amd.com> 746Date: Thu Sep 26 11:00:49 2013 -0400 747 748 radeon: fix limit handling for cards with >4G of ram 749 750 We can overflow the 32-bit limit. 751 752 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 753 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 754 755commit c45e728107269c6f51599dad4f6a02ccfef703f1 756Author: Michel Dänzer <michel.daenzer@amd.com> 757Date: Wed Sep 18 10:57:52 2013 +0200 758 759 DRI2: Install client callback only once 760 761 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=60182 762 Acked-by: Alex Deucher <alexander.deucher@amd.com> 763 764commit b955ddd0d41801e4ca0c30a70a5d0b27c3f366c8 765Author: Alex Deucher <alexander.deucher@amd.com> 766Date: Mon Sep 16 10:58:22 2013 -0400 767 768 glamor: require 0.5.1 or newer 769 770 0.5.1 fixes a number of issues. 771 772 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 773 774commit 282587cd0709850e7bacb1d8307065d95dc2c97d 775Author: Alex Deucher <alexander.deucher@amd.com> 776Date: Mon Sep 16 09:41:13 2013 -0400 777 778 radeon: enable glamor by default (v4) 779 780 It's required for newer asics, so enable it by 781 default. 782 783 v2: update the autoconf help. Require --disable-glamor 784 to build without glamor support. 785 v3: default to yes rather than auto 786 v4: fix help text 787 788 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 789 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 790 791commit fdb7563a5cbc736b09c2864b67a93b475c98b2bd 792Author: Alex Deucher <alexander.deucher@amd.com> 793Date: Thu Jan 24 21:17:11 2013 -0500 794 795 radeon/kms: add berlin pci ids 796 797 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 798 799commit 8927d33f76ee12bc618fecfc59fc7ff1fcedcd5e 800Author: Mark Kettenis <kettenis@openbsd.org> 801Date: Thu Aug 22 10:32:46 2013 -0400 802 803 Fix shadowfb on big-endian machines 804 805 For shadowfb, the framebuffer needs to have the RADEON_TILING_SURFACE 806 flag set, otherwise the appropriate byte swapping won't happen. 807 808 See https://bugs.freedesktop.org/show_bug.cgi?id=66663 809 810 Signed-off-by: Mark Kettenis <kettenis@openbsd.org> 811 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 812 813commit fa83d3d1636c315bc43dd622d407eb058e3ef976 814Author: Alex Deucher <alexander.deucher@amd.com> 815Date: Tue Aug 20 09:29:23 2013 -0400 816 817 radeon: disallow glamor on pre-R600 asics 818 819 I'm not sure they can handle the shaders properly, especially 820 only older parts like r300. 821 822 This will avoid display corruption problems reported by people 823 using glamor on older asics by falling back to EXA if they try 824 and enable glamor. 825 826 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 827 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 828 829commit d0323622ee9b97a5f246baffbb2c65930a78ed14 830Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 831Date: Wed Aug 7 11:27:07 2013 +0200 832 833 bump version post release 834 835 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 836 837commit 9c97cca5c24409ca8447c99f051a12fd2d494e79 838Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 839Date: Wed Aug 7 10:48:17 2013 +0200 840 841 radeon: bump version for release 842 843 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 844 845commit 16270cfb202ab67dd152644ef019b2f1ee4d0341 846Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 847Date: Wed Aug 7 10:29:33 2013 +0200 848 849 add bicubic_table.py to EXTRA_DIST 850 851 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 852 853commit 2cb9197ca7a337c911f38b5de562a2364b922b86 854Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 855Date: Wed Aug 7 10:28:52 2013 +0200 856 857 kill unused radeon_driver.c 858 859 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 860 861commit c5cbfcf575b0b4aea6f797558ae974c1453c8e07 862Author: Alex Deucher <alexander.deucher@amd.com> 863Date: Tue Jul 30 10:08:25 2013 -0400 864 865 drmmode: add support for multi-screen reverse optimus 866 867 Initial reverse optimus didn't consider multiple screens, so 868 this overhauls the code to use the new X server interface, 869 and allows for multiple outputs on the dGPU to be used with 870 the iGPU doing the rendering. Ported from Dave's nouveau 871 patch. 872 873 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 874 875commit 429d5b797769895eb4f5fef816ce4e2f3a342031 876Author: Dave Airlie <airlied@redhat.com> 877Date: Tue Jan 8 15:56:37 2013 +1000 878 879 radeon: add support for reverse prime (v2) 880 881 This adds support for reverse prime configurations 882 883 v2: fix compilation with older xservers 884 885 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 886 887commit 4de9356a2900ae0fb380a2350791ef045629cd05 888Author: Alex Deucher <alexander.deucher@amd.com> 889Date: Mon Aug 5 17:57:16 2013 -0400 890 891 radeon: fix naming clashes with multiple GPUs (v3) 892 893 The compat naming code for UMS causes problems 894 with multiple GPU as you may end up with the same 895 output name on multiple GPUs. Adjust the naming on 896 secondary GPUs to avoid conflicts. 897 898 v2: integrate Dave's fixes for nouveau 899 v3: keep compat with existing naming on primary GPU 900 901 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 902 903commit 2ae6bb18fefddb309920fa69c9b56c3a7f3db7b4 904Author: Grigori Goronzy <greg@chown.ath.cx> 905Date: Wed Jul 31 12:01:20 2013 +0200 906 907 EXA/evergreen/ni: replace magic number 908 909 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 910 911commit 6a278369c05a298a4367306d986467a9ceacae8c 912Author: Raul Fernandes <rgfernandes@gmail.com> 913Date: Tue Jul 30 09:26:05 2013 -0400 914 915 EXA/6xx/7xx: optimize non-overlapping Copy 916 917 In case dst and src rectangles of a Copy operation in the same surface 918 don't overlap, it is safe to skip the scratch surface. This is a 919 common case. 920 921 Based on evergreen/ni patch from Grigori Goronzy. 922 923 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 924 925commit 4375a6e75e5d41139be7031a0dee58c057ecbd07 926Author: Grigori Goronzy <greg@chown.ath.cx> 927Date: Mon Jul 22 02:30:28 2013 +0200 928 929 EXA/evergreen/ni: accelerate PictOpOver with component alpha 930 931 Subpixel text rendering is typically done with a solid src and a 932 pixmap mask. Traditionally, this cannot be accelerated in a single 933 pass and requires two passes [1]. However, we can cheat a little 934 with a constant blend color. 935 936 We can use: 937 938 const.A = src.A / src.A 939 const.R = src.R / src.A 940 const.G = src.G / src.A 941 const.B = src.B / src.A 942 943 dst.A = const.A * (src.A * mask.A) + (1 - (src.A * mask.A)) * dst.A 944 dst.R = const.R * (src.A * mask.R) + (1 - (src.A * mask.R)) * dst.R 945 dst.G = const.G * (src.A * mask.G) + (1 - (src.A * mask.G)) * dst.G 946 dst.B = const.B * (src.A * mask.B) + (1 - (src.A * mask.B)) * dst.B 947 948 This only needs a single source value. src.A is cancelled down in 949 the right places. 950 951 [1] http://anholt.livejournal.com/32058.html 952 953commit 94d0d14914a025525a0766669b556eaa6681def7 954Author: Grigori Goronzy <greg@chown.ath.cx> 955Date: Thu Jul 18 16:06:23 2013 +0200 956 957 EXA/evergreen/ni: fast solid pixmap support 958 959 Solid pixmaps are currently implemented with scratch pixmaps, which 960 is slow. This replaces the hack with a proper implementation. The 961 Composite shader can now either sample a src/mask or use a constant 962 value. 963 964commit 5bb04351c43a91a1d60348b7293544da05d75e72 965Author: Grigori Goronzy <greg@chown.ath.cx> 966Date: Fri Jul 27 17:31:53 2012 +0200 967 968 EXA/evergreen/ni: optimize non-overlapping Copy 969 970 In case dst and src rectangles of a Copy operation in the same surface 971 don't overlap, it is safe to skip the scratch surface. This is a 972 common case. 973 974commit c08e09b7bec441c4bf93b4cae4de1260754bf940 975Author: Grigori Goronzy <greg@chown.ath.cx> 976Date: Sat May 18 13:46:03 2013 +0200 977 978 Fix RADEON_FALLBACK logging 979 980commit c16c59f8f9b6aa7a4a6a6465582ad98f02a3606a 981Author: Mark Kettenis <kettenis@openbsd.org> 982Date: Sun Jul 7 13:44:13 2013 +0200 983 984 Always go through DFS/UTS when byte swapping is needed. 985 986 Before commit ef9bfb262db7004bef3704e5d914687e50d3fca4 and 987 e5bd99faa3b6629a55168386d5dfa936ee4e97ae, byte swapping for the front buffer 988 used to be done in hardware. Now that this no longer happens we need to let 989 DFS/UTS ihandle the byte swapping. 990 991 See https://bugs.freedesktop.org/show_bug.cgi?id=66663 . 992 993 Signed-off-by: Mark Kettenis <kettenis@openbsd.org> 994 995commit 1239dbbd8c8d9b55756c1de52cad353171a06522 996Author: Alex Deucher <alexander.deucher@amd.com> 997Date: Sun Jul 7 20:22:10 2013 -0400 998 999 radeon: bump libdrm requirement to 2.4.46 for CIK support 1000 1001 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1002 1003commit 869e0b187b2d07e637c4482fe50d9f1b0c09f4e4 1004Author: Alex Deucher <alexander.deucher@amd.com> 1005Date: Thu Jan 24 21:12:28 2013 -0500 1006 1007 radeon/kms: add kabini pci ids 1008 1009 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1010 1011commit 0edcefc4c97e4572431c38fe170032470e22e0b7 1012Author: Alex Deucher <alexander.deucher@amd.com> 1013Date: Fri Jun 7 14:56:26 2013 -0400 1014 1015 radeon/kms: add bonaire pci ids 1016 1017 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1018 1019commit ebc32b27af25b23604e725eb50d844a8d26116bb 1020Author: Alex Deucher <alexander.deucher@amd.com> 1021Date: Fri Aug 31 18:17:24 2012 -0400 1022 1023 radeon: update cursor handling for CIK 1024 1025 CIK asics have 128x128 hw cursors 1026 1027 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1028 1029commit 4adaea996454b1ad5185f9c0f37667dbfc266495 1030Author: Alex Deucher <alexander.deucher@amd.com> 1031Date: Fri Jun 7 14:47:36 2013 -0400 1032 1033 radeon: add family ids for CIK 1034 1035 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1036 1037commit 3626ab147b6749a355e7d21710447f13b9796c77 1038Author: Leo Liu <leo.liu@amd.com> 1039Date: Fri Jun 14 10:27:20 2013 -0400 1040 1041 radeon: fix a memory leak in get_modes() callback. 1042 1043 leak happens when looping xrandr prop. 1044 1045 Signed-off-by: Leo Liu <leo.liu@amd.com> 1046 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1047 1048commit c0f7d03cb29166c6c00dd6fb698ba3f70c59dbdf 1049Author: Alex Deucher <alexdeucher@gmail.com> 1050Date: Mon Jun 17 11:38:27 2013 -0400 1051 1052 man: fix up previous commit 1053 1054 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1055 1056commit 03b5edc9dd3859eed2504155970613aaf88df852 1057Author: Eric S. Raymond <esr@thyrsus.com> 1058Date: Fri Jun 7 13:23:14 2013 -0400 1059 1060 Fix some incomplete list entries. 1061 1062 This necessarily involved adding some content, which I was able to 1063 Google for. Please review, correct, and make more specific. 1064 1065 Bodiless .TP entries screw up doclifter's page parsing. Please 1066 don't do that anymore. 1067 1068commit 7ce43639a8e53eefb94e5255a39c1232b6bdbf66 1069Author: Michel Dänzer <michel.daenzer@amd.com> 1070Date: Thu Jun 6 09:58:27 2013 +0200 1071 1072 autogen.sh: Restore passing --enable-maintainer-mode to configure 1073 1074 Looks like this was accidentally dropped when the GNOME Build API was 1075 implemented. 1076 1077commit 365e571d9de0b5979425c34210789afca4ea5f68 1078Author: Samuel Li <samuel.li@amd.com> 1079Date: Wed Mar 20 11:59:58 2013 -0400 1080 1081 radeon: use direct mapping for fast fb access. 1082 1083 Signed-off-by: Samuel Li <samuel.li@amd.com> 1084 1085commit bd2557ea5ef84b975060e929d5ece53ec464336f 1086Author: Ilija Hadzic <ilijahadzic@gmail.com> 1087Date: Wed May 8 22:39:48 2013 -0400 1088 1089 DRI2: add interpolated blanks to frame number in event handlers 1090 1091 The 'frame' argument passed to event handlers is the plain 1092 CRTC vblank counter that stops progressing when the 1093 associated display is in DPMS-off mode. If we have a 1094 DPMS-off period the frame counter and MSC will 1095 diverge, which can cause some higher-level functions 1096 to return incorrect values. 1097 1098 This patch fixes the problem by adding interpolated 1099 vblanks to the frame counter before using it in handler 1100 functions. 1101 1102 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1103 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1104 1105commit 385a92b4fbe8f53b359ef6c463704414d00476fa 1106Author: Ilija Hadzic <ilijahadzic@gmail.com> 1107Date: Wed May 8 22:39:47 2013 -0400 1108 1109 DRI2: support scheduling emulated events with zero delay 1110 1111 Now that we fully emulating a running CRTC through DPMS-off state 1112 it is possible to come up with a zero delay when scheduling 1113 a swap or MSC-wait (e.g., if a call into respective wait function 1114 was entered very late). This patch wraps the TimerSet function 1115 into our own radeon_dri2_schedule_event such that the latter 1116 calls the event right away if zero delay is specified. 1117 1118 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1119 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1120 1121commit e87b52e6ad41ad7a87e43b818d80e7d522d9c68d 1122Author: Ilija Hadzic <ilijahadzic@gmail.com> 1123Date: Wed May 8 22:39:46 2013 -0400 1124 1125 DRI2: make wait_msc and get_msc work on disabled CRTCs 1126 1127 Now that the running CRTC is emulated through DPMS-off 1128 states, wait_msc and get_msc must also work on disabled 1129 CRTCs. When CRTC is disabled, we must extrapolate the 1130 MSC count from present time and last MSC when CRTC was 1131 running and also use timers facility in wait_msc. 1132 1133 v2: CRTC-private now stores frame rate instead of nominal 1134 vblank period. 1135 1136 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1137 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1138 1139commit f1584152bbeda98cc1907359f1c7954a84c8837d 1140Author: Ilija Hadzic <ilijahadzic@gmail.com> 1141Date: Wed May 8 22:39:45 2013 -0400 1142 1143 DRI2: move rename and rework radeon_dri2_deferred_swap 1144 1145 radeon_dri2_deferred_swap will be used to generate 1146 real events (not just fallbacks) so now it needs to 1147 generate real timestamp and frame counter. Also 1148 this function will be used both by schedule_swap 1149 and wait_msc, so give it a more generic name: 1150 radeon_dri2_deferred_event 1151 1152 v2: - Extrapolate the frame number from the time 1153 of actual execution of the function instead 1154 of using the MSC calculated (extrapolated) 1155 at event scheduling time. 1156 - CRTC-private now stores frame rate instead of 1157 nominal vblank period. 1158 1159 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1160 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1161 1162commit 84bce7943b735aee19e26be4d47fdf140564f3f5 1163Author: Ilija Hadzic <ilijahadzic@gmail.com> 1164Date: Wed May 8 22:39:44 2013 -0400 1165 1166 DRI2: track CRTC in event record 1167 1168 Tracking the CRTC associated with an event will save us 1169 some lookups later in event handlers. 1170 1171 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1172 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1173 1174commit e40d5390b3efdea3e02267413350410d8e6a2970 1175Author: Ilija Hadzic <ilijahadzic@gmail.com> 1176Date: Wed May 8 22:39:43 2013 -0400 1177 1178 DRI2: hook up vblank extrapolation to schedule_swap 1179 1180 This patch hooks up swap-scheduling function with 1181 vblank-extrapolation function. Rather than waiting for 1182 fixed time, we calculate exactly how much we should wait 1183 and what we should update target_msc to using 1184 radeon_dri2_extrapolate_msc_delay helper function 1185 and schedule the swap completion using DIX's timer facility. 1186 1187 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1188 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1189 1190commit ffaa5abf207415159cdb28e90da49b95f497ef61 1191Author: Ilija Hadzic <ilijahadzic@gmail.com> 1192Date: Wed May 8 22:39:42 2013 -0400 1193 1194 DRI2: add vblank extrapolation function 1195 1196 Implement a helper function that will be called when emulating 1197 the running CRTC. The function should be called only when CRTC 1198 is in DPMS-off state. It will look at the vblank count and the 1199 time that was recorded last time the CRTC was running and 1200 calculate how long one must wait (from present time) until 1201 the target_msc is reached if the CRTC were running. 1202 1203 v2: - CRTC-private now stores frame rate instead of nominal 1204 vblank period. 1205 - DIX's timer facility can sometimes wake up the scheduled 1206 functions more than a millisecond earlier. To avoid 1207 generating an old MSC, we have to add more margin when 1208 converting the delay in microseconds to milliseconds. 1209 1210 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1211 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1212 1213commit 80ae2291d082b57c70d27a80182f00d760fb3d3a 1214Author: Ilija Hadzic <ilijahadzic@gmail.com> 1215Date: Wed May 8 22:39:41 2013 -0400 1216 1217 DRI2: adjust MSC by the number of interpolated vblanks 1218 1219 Normally, MSC value equals the vblank count of the CRTC 1220 on which the drawable resides. However, we are now interpolating 1221 vblanks through DPMS-off state, so if the CRTC is turned 1222 off at some point, the vlbank count does not change, but 1223 MSC does change by the amount that equals the cumulative 1224 number of interpolated vblanks. 1225 1226 So each time we interact with the CRTC's vblank facility, 1227 we have to adjust the vblank counter: in the request, we 1228 have to subtract the number of interpolated vblanks from 1229 MSC and in the reply we have to add the number of interpolated 1230 vblanks to the MSC. 1231 1232 This patch only makes the MSC correct when CRTC is on. 1233 The subsequent patch will construct the MSC when CRTC is not 1234 running. 1235 1236 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1237 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1238 1239commit 5fd2eb5d12cea32927a9e6c6ce4afd18aa7d046a 1240Author: Ilija Hadzic <ilijahadzic@gmail.com> 1241Date: Wed May 8 22:39:40 2013 -0400 1242 1243 drmmode: calculate interpolated vblanks while in dpms-off state 1244 1245 This adds provisions for interpolating vblanks while the CRTC 1246 is in DPMS-off state. When entering DPMS-off state, we 1247 record the last vblank time, sequence number and frame rate 1248 in CRTC-private structure. 1249 1250 When going back to DPMS-on state we read the current 1251 time and calculate how long we have been off. Then we derive 1252 how many vblanks that would have been had the CRTC remained 1253 running. These are the interpolated vblanks. 1254 1255 Finally, we accumulate the number of interpolated vblanks 1256 in CRTC-private structure to get the number of interpolated 1257 vblanks over the system lifetime. 1258 1259 v2: Track frame rate instead of vblank period. The former 1260 eliminates some roundoff errors. 1261 1262 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1263 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1264 1265commit e41ad30d09be4962cfb1942b0b9f63875dbb2d2a 1266Author: Ilija Hadzic <ilijahadzic@gmail.com> 1267Date: Wed May 8 22:39:39 2013 -0400 1268 1269 drmmode: remove dead code from drmmode_crtc_dpms 1270 1271 We are about to modify this function, so let's clean it 1272 up first. 1273 1274 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1275 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1276 1277commit 373671d2eed30e30b3bdee7e40426cf58c127234 1278Author: Ilija Hadzic <ilijahadzic@gmail.com> 1279Date: Wed May 8 22:39:38 2013 -0400 1280 1281 drmmode: add drmmode_get_current_ust function 1282 1283 The new helper function retrieves current time in the format 1284 that is compatible with vblank timestamps. 1285 1286 v2: - fix an incorrect statement in a comment 1287 - add a #define so that don't depend on libdrm patches that 1288 have not yet been accepted upstream 1289 1290 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1291 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1292 1293commit 006fbbd1d38a089b50ab3197d32815689ed249fa 1294Author: Ilija Hadzic <ilijahadzic@gmail.com> 1295Date: Wed May 8 22:39:37 2013 -0400 1296 1297 DRI2: make populate_vbl_request_type external 1298 1299 We'll need to use this function in other .c files so 1300 remove static annotation from it. Also, add radeon_ 1301 prefix because this function is now part of global namespace, 1302 so it would be nice to know where it is coming from. 1303 1304 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1305 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1306 1307commit 34660d8757ff16acd1686ff2872f4600d92a68bf 1308Author: Ilija Hadzic <ilijahadzic@gmail.com> 1309Date: Wed May 8 22:39:36 2013 -0400 1310 1311 DRI2: change signature of populate_vbl_request_type 1312 1313 This function no longer uses info argument, so nuke it. 1314 1315 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1316 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1317 1318commit ab6e85d1f5dd4657591b314b85fd2453f99c3712 1319Author: Michel Dänzer <michel.daenzer@amd.com> 1320Date: Fri Mar 22 10:51:05 2013 +0100 1321 1322 glamor: Keep old private in radeon_pixmap_set_bo 1323 1324commit 0a84f2760eccf07781ae6c3f55d2d246c155c9ce 1325Author: Alex Deucher <alexander.deucher@amd.com> 1326Date: Mon May 13 17:06:00 2013 -0400 1327 1328 radeon: add Hainan pci ids 1329 1330 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1331 1332commit 6cee1eaebfcef2b2d2c51ec30c31260fd98b564e 1333Author: Alex Deucher <alexander.deucher@amd.com> 1334Date: Mon May 13 17:01:30 2013 -0400 1335 1336 radeon: add hainan chip family 1337 1338 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1339 1340commit 8a9e9b21cbb44c23d37bf18c983e2c3f0fe33e87 1341Author: Michel Dänzer <michel.daenzer@amd.com> 1342Date: Thu Apr 25 17:10:12 2013 +0200 1343 1344 glamor: Fix build against xserver < 1.13 1345 1346 Those didn't define CREATE_PIXMAP_USAGE_SHARED. 1347 1348commit 6216b5aa3793abe71b864170bc9d4d745f027328 1349Author: Alex Deucher <alexander.deucher@amd.com> 1350Date: Thu Apr 25 14:19:37 2013 -0400 1351 1352 radeon: add new richland pci ids 1353 1354 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1355 1356commit fd5903f15b36555bf1ccc1c3865c9fffd499128a 1357Author: Alex Deucher <alexander.deucher@amd.com> 1358Date: Thu Apr 25 14:17:51 2013 -0400 1359 1360 radeon: add new SI pci ids 1361 1362 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1363 1364commit b2826580e4d246d45871b413e4deecb070162577 1365Author: Jerome Glisse <jglisse@redhat.com> 1366Date: Mon Apr 8 13:45:18 2013 -0400 1367 1368 radeonsi: add 2d tiling support 1369 1370 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 1371 1372commit 6e74aacc5e5da3b51744153dad1645caa6ea4ce3 1373Author: Michel Dänzer <michel.daenzer@amd.com> 1374Date: Wed Mar 27 11:57:29 2013 +0100 1375 1376 glamor: Only enable for depth >= 24 1377 1378 It can't handle lower depths yet. 1379 1380 Reported-by: Igor Vagulin <igor.vagulin@gmail.com> 1381 1382commit 33d8408eec806355c2e55726679ec50ef3b769f1 1383Author: Michel Dänzer <michel.daenzer@amd.com> 1384Date: Tue Mar 12 19:03:06 2013 +0100 1385 1386 glamor: Enable by default on SI 1387 1388 We should now handle its major initialization failure cases gracefully. 1389 1390 Reviewed-by: Jerome Glisse <jglisse@redhat.com> 1391 1392commit ad1d8874fab30257ba8ec8bee8ffa42a8fa9b5c0 1393Author: Michel Dänzer <michel.daenzer@amd.com> 1394Date: Fri Mar 15 18:28:27 2013 +0100 1395 1396 glamor: Initial PRIME pixmap sharing hooks. 1397 1398 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57200 1399 1400 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1401 Tested-by: Christoph Haag <haagch.christoph@googlemail.com> 1402 Tested-by: linedot <linedot@xcpp.org> 1403 1404commit 35b384a00a0c76ea556f55787fccc95ecd51bc0c 1405Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1406Date: Thu Mar 21 15:34:21 2013 +0100 1407 1408 drmmode: handle failure of drmModeGetConnector 1409 1410 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1411 1412commit 751c7259cf23a29fdd2653881e2db63ce3cf1ca5 1413Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1414Date: Thu Mar 21 10:40:42 2013 +0100 1415 1416 Move fd cleanup to correct place. 1417 1418 Oops, just after previous commit I realized that it 1419 needs to be done from RADEONFreeRec, because when preinit fails 1420 info pointer can be null in FreeScreen. 1421 1422 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1423 1424commit 1643b8cd2dc53ed36916e11003590c7037b4ddd6 1425Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1426Date: Thu Mar 21 10:33:19 2013 +0100 1427 1428 radeon: add refcounts to fix up zaphod open/close. 1429 1430 Oops, turns out my previous commits were buggy. 1431 Adding proper refcounts will handle this correctly. 1432 1433 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1434 1435commit a4d1bc43d4d10e2ca1e512c1010e962c587d6c16 1436Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1437Date: Wed Mar 20 17:13:44 2013 +0100 1438 1439 Cleanup drmmode in CloseScreen. 1440 1441 Prevents a use-after-free. 1442 1443 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1444 1445commit 3cb3ab8704813d7c14b9e547a30e37598d70c85a 1446Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1447Date: Wed Mar 20 17:12:29 2013 +0100 1448 1449 Close drm_fd in CloseScreen. 1450 1451 This prevents a small fd leak. 1452 1453 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1454 1455commit 462d8d3b86038438ab854072e546cc7a9efacdd7 1456Author: Michel Dänzer <michel.daenzer@amd.com> 1457Date: Tue Mar 12 18:59:19 2013 +0100 1458 1459 glamor: Bail if the glamoregl module wasn't loaded early 1460 1461 Trying to initialize glamor in that case crashes. 1462 1463 Reviewed-by: Jerome Glisse <jglisse@redhat.com> 1464 1465commit 3e5350bedf50a1a140e0eba08c39ed2451fb9115 1466Author: Alex Deucher <alexander.deucher@amd.com> 1467Date: Fri Mar 8 14:07:42 2013 -0500 1468 1469 radeon: add Richland pci ids 1470 1471 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1472 1473commit bdbf766d8170f6ea6c16b9780ef5a4007730588b 1474Author: Alex Deucher <alexander.deucher@amd.com> 1475Date: Fri Mar 8 14:00:09 2013 -0500 1476 1477 radeon: mark trinity APUs with IGP flag 1478 1479 Was missing on trinity. 1480 1481 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1482 1483commit 5ac335eaae135c0e66cc4643908f41174e2f1123 1484Author: Michel Dänzer <michel.daenzer@amd.com> 1485Date: Fri Mar 8 17:37:41 2013 +0100 1486 1487 Only parse Option "AccelMethod" if glamor was enabled at build time 1488 1489 Will hopefully make it more obvious when it wasn't. 1490 1491commit f1059e3f0934ccf1ae183099dbd4de700a7c81ac 1492Author: Colin Walters <walters@verbum.org> 1493Date: Wed Jan 4 17:37:06 2012 -0500 1494 1495 autogen.sh: Implement GNOME Build API 1496 1497 http://people.gnome.org/~walters/docs/build-api.txt 1498 1499 Signed-off-by: Adam Jackson <ajax@redhat.com> 1500 1501commit 343b01c9bf35b125cd0c3df8db7c01a5fb227bda 1502Author: Alex Deucher <alexander.deucher@amd.com> 1503Date: Thu Jan 24 19:57:23 2013 -0500 1504 1505 radeon/kms: add oland pci ids 1506 1507 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1508 1509commit 9980f1e0e0a5df5fb0c791687e5c98c13089d7d1 1510Author: Alex Deucher <alexander.deucher@amd.com> 1511Date: Thu Jan 24 19:52:04 2013 -0500 1512 1513 radeon: add oland chip family 1514 1515 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1516 1517commit d89e545826969910226cb96e73b64f18e92d2064 1518Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1519Date: Wed Jan 30 17:21:05 2013 +0100 1520 1521 bump version post release 1522 1523 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1524 1525commit 21a4f3eaca98c1c54bffcdf3926c631cec9c409d 1526Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1527Date: Wed Jan 30 16:55:27 2013 +0100 1528 1529 radeon: bump version for release 1530 1531 And fixup distcheck.. 1532 1533 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1534 1535commit 102ee4a24e4c9ba0c0b65d3ca8599dc7e0216c1e 1536Author: Alex Deucher <alexdeucher@gmail.com> 1537Date: Fri Jan 11 09:52:32 2013 -0500 1538 1539 radeon/r200: program RE_WIDTH_HEIGHT properly 1540 1541 Values are inclusive. Mesa already sets these 1542 correctly. Also bump EXA/Xv limits from 2047 to 1543 2048. 1544 1545 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1546 1547commit 47689ef8b878a56ea094e21a6090b228e67d605e 1548Author: Michel Dänzer <michel.daenzer@amd.com> 1549Date: Fri Jan 11 15:41:48 2013 +0100 1550 1551 Make radeon_glamor_create_textured_pixmap() return TRUE without USE_GLAMOR. 1552 1553 Otherwise rotation won't work when glamor support isn't built in. 1554 1555 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1556 1557commit 1a8c05fd5559be842d8a038df73de34d4b261c15 1558Author: Michel Dänzer <michel.daenzer@amd.com> 1559Date: Fri Jan 11 13:22:24 2013 +0100 1560 1561 Make CRTC shadow pixmaps usable by glamor. 1562 1563 Fixes rotation with glamor. 1564 1565 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1566 1567commit e5e22d788b103c04d3460a221a3b9dc3be69e0a4 1568Author: Michel Dänzer <michel.daenzer@amd.com> 1569Date: Thu Jan 10 17:00:16 2013 +0100 1570 1571 DRI2/r[345]xx: 16 bit depth buffers use square micro tiling. 1572 1573 Fixes piglit sanity.tests (with 3D drivers which allocate depth buffers from 1574 the X server). 1575 1576 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1577 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1578 1579commit f5d2682584c74587890a9795ece7a8917214a83f 1580Author: Michel Dänzer <michel.daenzer@amd.com> 1581Date: Thu Jan 10 17:00:15 2013 +0100 1582 1583 DRI2: Properly use buffer depth passed in by client. 1584 1585 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1586 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1587 1588commit 4e35b2f530e2ca8c7b7220cacd05c661de43d20d 1589Author: Michel Dänzer <michel.daenzer@amd.com> 1590Date: Thu Jan 10 12:10:52 2013 +0100 1591 1592 Drop support for X servers older than 1.7. 1593 1594 The main purpose is to drop DRI2 compatibility code paths which are getting 1595 awkward to deal with and at this point are probably only tested lightly if at 1596 all. xserver 1.7 was released more than 3 years ago, and 1.6.2 was already 1597 required before. 1598 1599 Mostly mechanical, there's probably potential for more cleanups. 1600 1601 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 1602 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1603 1604commit 857d729eb3672443f89c0516f8226eacb37dad2b 1605Author: Dave Airlie <airlied@redhat.com> 1606Date: Mon Jan 7 14:13:33 2013 +1000 1607 1608 radeon: fix damage reporting for slave pixmaps 1609 1610 This is definitely more correct, not sure if it'll fix any bugs. 1611 1612 Signed-off-by: Dave Airlie <airlied@redhat.com> 1613 1614commit 6981a5c087165b126c15ba0025cffdba218ab652 1615Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 1616Date: Wed Dec 19 10:35:43 2012 -0500 1617 1618 DRI2: limit the swap rate when CRTC is in DPMS-off state 1619 1620 If drawable is displayed on a CRTC and relevant CRTC is in 1621 DPMS off state, defer the swap by a fixed (hard-coded) time. 1622 1623 This patch fixes a bug that caused an application to render 1624 at uncontrolled rate when CRTC goes into DPMS "off" state, 1625 thus thrashing the GPU and CPU and likely offsetting the 1626 power savings achieved by shutting off the display. 1627 1628 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1629 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 1630 1631commit 3657672207322be651cdb94a811337b7c5668c84 1632Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 1633Date: Wed Dec 19 10:35:42 2012 -0500 1634 1635 DRI2: change signature of radeon_dri2_drawable_crtc 1636 1637 Return pointer to the CRTC instead of CRTC ID and expose 1638 consider_disabled option in arguments of this function. 1639 1640 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1641 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 1642 1643commit 61d0aec40e2521488c2fe43e7a6823e5c87d94d7 1644Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 1645Date: Wed Dec 19 10:35:41 2012 -0500 1646 1647 video: add option to include disabled CRTCs in best CRTC search 1648 1649 This patch adds an option called consider_disabled to 1650 radeon_pick_best_crtc function. If this option is set 1651 and searching for best-fit CRTC yields nothing, the search 1652 will be widened to include CRTCs in DPMS "off" state. 1653 The new option is not used yet; it will be in the patches 1654 to follow. 1655 1656 v2: Do not look at the CRTC mode, we only care about its box. 1657 1658 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1659 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 1660 1661commit a60d2152e928a7011fc7c44a885a34c3cdd4f0fe 1662Author: Michel Dänzer <michel.daenzer@amd.com> 1663Date: Thu Dec 27 12:13:57 2012 +0100 1664 1665 glamor: Enable full 2D acceleration for SI as well. 1666 1667 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1668 1669commit 04ebe14fe775d18b534e89b50b90ee293ae1910e 1670Author: Michel Dänzer <michel.daenzer@amd.com> 1671Date: Thu Dec 27 11:54:27 2012 +0100 1672 1673 glamor: Release the drawable after passing to glamor_push_pixels 1674 1675 Ported from xf86-video-intel commit ca5c028c2b4d9bf02002acd484054fe427ea8d09 . 1676 1677 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1678 1679commit cee7d61b144162974238b282d76a70aa2f14cc0d 1680Author: Alex Deucher <alexdeucher@gmail.com> 1681Date: Tue Dec 18 15:49:47 2012 -0500 1682 1683 exa/Xv: fix coordinate limits on AVIVO IGPs 1684 1685 RS6xx asics are r4xx derived, but seem to have r3xx 1686 limitations as far as clipping is concerned. Spotted 1687 by Michel on IRC. 1688 1689 Fixes: 1690 https://bugs.freedesktop.org/show_bug.cgi?id=58469 1691 1692 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1693 1694commit 3c7d024c2671dac541b8f2daed55040f4fd5d62d 1695Author: Michel Dänzer <michel.daenzer@amd.com> 1696Date: Wed Dec 12 10:43:55 2012 +0100 1697 1698 Remove dead option definitions. 1699 1700 These were only used with UMS. 1701 1702 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1703 1704commit bdbd84a6c99943f4e012c92ad9cb920cbd676621 1705Author: Michel Dänzer <michel.daenzer@amd.com> 1706Date: Mon Dec 10 23:35:41 2012 +0100 1707 1708 Remove local variables total_size_bytes. 1709 1710 Not actually used for anything. Refactoring gone wrong? 1711 1712 Also remove a comment that doesn't seem to make sense anymore. 1713 1714 Stumbled over this due to a warning by clang: 1715 1716 ../../src/radeon_kms.c:1562:6: warning: variable 'total_size_bytes' is uninitialized when used here [-Wuninitialized] 1717 total_size_bytes += (64 * 4 * 64); 1718 ^~~~~~~~~~~~~~~~ 1719 ../../src/radeon_kms.c:1557:25: note: initialize the variable 'total_size_bytes' to silence this warning 1720 int total_size_bytes; 1721 ^ 1722 = 0 1723 1724 Reviewed-by: Alex Deucher <aleander.deucher@amd.com> 1725 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1726 1727commit cf0a81547bcde32bdd2b080c0a546a49dd2eb628 1728Author: Michel Dänzer <michel.daenzer@amd.com> 1729Date: Tue Dec 11 11:48:33 2012 +0100 1730 1731 Remove unused local variable 'path'. 1732 1733 ../../src/radeon_probe.c: In function 'radeon_platform_probe': 1734 ../../src/radeon_probe.c:270:11: error: unused variable 'path' [-Werror=unused-variable] 1735 1736 Reviewed-by: Alex Deucher <aleander.deucher@amd.com> 1737 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1738 1739commit d268038a5bdd1655f7533c8dd2edf74fbc82bbee 1740Author: Michel Dänzer <michel.daenzer@amd.com> 1741Date: Mon Jul 2 09:07:14 2012 +0200 1742 1743 Remove dead code flagged by gcc -Wunused-but-set-variable. 1744 1745 Reviewed-by: Alex Deucher <aleander.deucher@amd.com> 1746 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1747 1748commit 793e1b0ea579cce7489aee57a95f20384d6ecce8 1749Author: Alex Deucher <alexdeucher@gmail.com> 1750Date: Thu Dec 6 09:21:19 2012 -0500 1751 1752 radeon: fix ring count in R300PrepareTexturedVideo 1753 1754 Fixes: 1755 https://bugs.freedesktop.org/show_bug.cgi?id=31364 1756 1757 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1758 1759commit f402805b22e4f4f735924d59c69474b24e623a25 1760Author: Jerome Glisse <jglisse@redhat.com> 1761Date: Fri Nov 30 10:07:51 2012 -0500 1762 1763 radeon: avoid copying over itself ddx fb 1764 1765 In some rare case it seems that the old fb could already be the 1766 ddx fb. As copying from the same bo will trigger a ttm deadlck 1767 detection which will result in infinite kernel/userspace loop. 1768 Avoid doing any copy as anyway it's useless. 1769 1770 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 1771 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 1772 1773commit 60cd6ceaf44b506433aebf6b3a639a17604dfddd 1774Author: Alex Deucher <alexdeucher@gmail.com> 1775Date: Wed Nov 21 18:42:56 2012 -0500 1776 1777 radeon: add new SI pci id 1778 1779 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1780 1781commit 53af6aa99dfb67b5223c28d5129a79694165048d 1782Author: Jerome Glisse <jglisse@redhat.com> 1783Date: Wed Nov 14 12:15:47 2012 -0500 1784 1785 radeon: avoid segfault when pixmap exceed GPU capabilities 1786 1787 We might get a request for a pixmap/drawable that is too big 1788 for GPU capabilities in the dri2 get buffer path. In face of 1789 such things just return NULL to dri2 get buffer request. The 1790 GL driver should then use something like transparent black to 1791 avoid something too ugly on the screen. 1792 1793 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 1794 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 1795 1796commit 71eeb0833489ece745f1e5dc41379ec63161ee81 1797Author: Alex Deucher <alexdeucher@gmail.com> 1798Date: Tue Nov 6 09:43:55 2012 -0500 1799 1800 bump verion post release 1801 1802 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1803 1804commit cf377c991f4c1cd4eee1a8530ba9fdf6f9f63301 1805Author: Alex Deucher <alexdeucher@gmail.com> 1806Date: Tue Nov 6 09:12:39 2012 -0500 1807 1808 radeon: bump version for release 1809 1810 7.0.0, RIP UMS 1811 1812 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1813 1814commit 20bfc652ce40008ea561db2984bccf137409c7fd 1815Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 1816Date: Mon Sep 24 17:19:09 2012 -0400 1817 1818 radeon/radeon_platform_probe: fix Zaphod mode breakage 1819 1820 Using radeon_platform_probe function breaks the Zaphod mode because 1821 it attempts to call xf86AddEntityToScreen multiple times, but nobody 1822 calls xf86SetEntityShared prior to that. Consequently, calls for all 1823 but first device instance fail. 1824 1825 Prior to introduction of platform bus, the logic was that the Probe 1826 function would make the entity sharable, which would cause Xserver 1827 to later make it shared prior to adding it to screen. With the 1828 platform bus loading, add to screen happens in the probe 1829 function so we have to make it shared there. 1830 1831 v2: do not make the entity shared if it was not previously marked 1832 sharable. Should fix: 1833 1834 https://bugs.freedesktop.org/show_bug.cgi?id=56680 1835 https://bugs.freedesktop.org/show_bug.cgi?id=56663 1836 1837 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1838 1839commit b2807b53083af5db0b85645666b0127be16623c7 1840Author: Alex Deucher <alexdeucher@gmail.com> 1841Date: Fri Nov 2 17:57:18 2012 -0400 1842 1843 man: note that 2D tiling requires Mesa 9.0 or newer 1844 1845 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1846 1847commit 1038e29162d869e2ce37f9b8f5f7c8ea37c3e53d 1848Author: Alex Deucher <alexdeucher@gmail.com> 1849Date: Fri Nov 2 10:04:46 2012 -0400 1850 1851 Revert "radeon/radeon_platform_probe: fix Zaphod mode breakage" 1852 1853 This reverts commit 39eac3104c2f08b4d78aab3f88fd104301eb4711. 1854 1855 This breaks non-zaphod configurations with xserver 1.13. 1856 See: 1857 https://bugs.freedesktop.org/show_bug.cgi?id=56680 1858 https://bugs.freedesktop.org/show_bug.cgi?id=56663 1859 1860commit d748894c84e891662cde6d7a5748f74a5aa5175e 1861Author: Alex Deucher <alexdeucher@gmail.com> 1862Date: Fri Nov 2 08:50:16 2012 -0400 1863 1864 man: update man pages for trinity (ARUBA) 1865 1866 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1867 1868commit 6740e8e90b934611c3bb0f9e8732d66e25b8c727 1869Author: Andreas Boll <andreas.boll.dev@gmail.com> 1870Date: Fri Nov 2 13:30:44 2012 +0100 1871 1872 radeon/man: ColorTiling2D is by default on for r6xx-cayman 1873 1874 Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> 1875 1876commit 39eac3104c2f08b4d78aab3f88fd104301eb4711 1877Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 1878Date: Mon Sep 24 17:41:32 2012 -0400 1879 1880 radeon/radeon_platform_probe: fix Zaphod mode breakage 1881 1882 Using radeon_platform_probe function breaks the Zaphod mode because 1883 it attempts to call xf86AddEntityToScreen multiple times, but nobody 1884 calls xf86SetEntityShared prior to that. Consequently, calls for all 1885 but first device instance fail. 1886 1887 Prior to introduction of platform bus, the logic was that the Probe 1888 function would make the entity sharable, which would cause Xserver 1889 to later make it shared prior to adding it to screen. With the 1890 platform bus loading, add to screen happens in the probe 1891 function so we have to make it shared there. 1892 1893 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 1894 1895commit 2e96aa641bc8a5c263a5b643961383cdf530d044 1896Author: Marek Olšák <maraeo@gmail.com> 1897Date: Sun Oct 28 12:50:49 2012 -0400 1898 1899 radeon: turn on 2D tiling by default on r6xx-cayman 1900 1901 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1902 1903commit e830c039479539735d0836082abef24f6841c803 1904Author: Marek Olšák <maraeo@gmail.com> 1905Date: Sun Oct 28 12:49:15 2012 -0400 1906 1907 radeon: log if 2D tiling is enabled 1908 1909 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 1910 1911commit ce1b745dcb60dc516ad999756240b78e72a7aa54 1912Author: Alex Deucher <alexander.deucher@amd.com> 1913Date: Tue Oct 16 13:05:46 2012 -0400 1914 1915 radeon: add new pci ids for SI 1916 1917 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 1918 1919commit e3519be285bdab50ef2193f98826a351a4414fae 1920Author: Dave Airlie <airlied@redhat.com> 1921Date: Tue Oct 16 16:09:23 2012 +1000 1922 1923 radeon/dri2: fix the window->pixmap math properly. 1924 1925 Okay kwin did things different, and I really hope this time this is the 1926 correct transformation operation. It works here with gears under kwin 1927 and under mutter and under xcompmgr. 1928 1929 Signed-off-by: Dave Airlie <airlied@redhat.com> 1930 1931commit bd9e2c064d93afbe4ab84febb2a43ae6a95ab17b 1932Author: Dave Airlie <airlied@redhat.com> 1933Date: Thu Oct 11 14:36:11 2012 +1000 1934 1935 radeon/dri2: fix damage reporting on pageflip 1936 1937 when you use MPX or slave usb devices, the damage posting was incorrect, 1938 and resulted in misrendered boxes on the screen. 1939 1940 Fix it by reporting damage correctly. 1941 1942 Signed-off-by: Dave Airlie <airlied@redhat.com> 1943 1944commit e8cb0b721e6ea251f85c799ca0563bfa59a2d37c 1945Author: Adam Jackson <ajax@redhat.com> 1946Date: Wed Sep 26 09:27:54 2012 -0400 1947 1948 Remove mibstore.h 1949 1950 Signed-off-by: Adam Jackson <ajax@redhat.com> 1951 1952commit 8637f772347c958fa3beea6c0dc9c22255db70e5 1953Author: Michel Dänzer <michel.daenzer@amd.com> 1954Date: Thu Sep 13 19:50:35 2012 +0200 1955 1956 glamor: Force acceleration for DRI2 copies. 1957 1958 Improves performance of non-pageflipped 3D apps on SI. 1959 1960 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1961 1962commit 40b87f0e9db2642ca8a70e994fb4d3742e199640 1963Author: Michel Dänzer <michel.daenzer@amd.com> 1964Date: Wed Sep 12 18:44:10 2012 +0200 1965 1966 glamor: Fix DRI2 crash introduced by PRIME changes. 1967 1968 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 1969 1970commit 646ffd49d2cf0d3130301e8ad2f49f620bd9ef1b 1971Author: Dave Airlie <airlied@redhat.com> 1972Date: Mon Sep 10 15:31:06 2012 +1000 1973 1974 radeon: fix build against old servers 1975 1976 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=54711 1977 1978 Reported-by: wolput@onsneteindhoven.nl 1979 Signed-off-by: Dave Airlie <airlied@redhat.com> 1980 1981commit 7c7f2775604da61537c24c249884aaa220e06e13 1982Author: Dave Airlie <airlied@redhat.com> 1983Date: Mon Sep 10 11:27:45 2012 +1000 1984 1985 radeon: don't fail to load if we have no modes and gpu driver. 1986 1987 This ensures radeon loads on output less GPUs as a GPU driver. 1988 1989 There are server bugs that also need to be fixed. 1990 1991 Signed-off-by: Dave Airlie <airlied@redhat.com> 1992 1993commit 7fe16dd037ac688726869816888bd77fe4356f8e 1994Author: Dave Airlie <airlied@gmail.com> 1995Date: Wed Sep 5 10:41:51 2012 +1000 1996 1997 radeon: fix typo in glamor paths 1998 1999 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=54519 2000 2001 Reported-By: scaty@dcinformatique.com 2002 Signed-off-by: Dave Airlie <airlied@redhat.com> 2003 2004commit cdc985e9190ce54dc6274ebc0e87ab954b15bb0d 2005Author: Dave Airlie <airlied@gmail.com> 2006Date: Tue Sep 4 18:18:50 2012 +1000 2007 2008 radeon: fix build against older servers 2009 2010 Another silly one. 2011 2012 Signed-off-by: Dave Airlie <airlied@redhat.com> 2013 2014commit bd0e9c51407cb99f1fe4f0acfedc37ee0c354040 2015Author: Dave Airlie <airlied@gmail.com> 2016Date: Tue Sep 4 18:17:06 2012 +1000 2017 2018 radeon/dri2: fix build against older servers. 2019 2020 Older servers won't have DRI2UpdatePrime, so fix build. 2021 2022 bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=54433 2023 Signed-off-by: Dave Airlie <airlied@redhat.com> 2024 2025commit 75c1f62f44d7b3709c1fa103d59733866967b841 2026Author: Dave Airlie <airlied@redhat.com> 2027Date: Mon Sep 3 13:09:02 2012 +1000 2028 2029 radeon: add capability and provider support 2030 2031 Signed-off-by: Dave Airlie <airlied@redhat.com> 2032 2033commit ca8b61e916998458203dfa4b9d0c997fa7d90ba9 2034Author: Dave Airlie <airlied@redhat.com> 2035Date: Mon Sep 3 13:08:37 2012 +1000 2036 2037 radeon: allocated shared pixmaps as untiled gtt. 2038 2039 Signed-off-by: Dave Airlie <airlied@redhat.com> 2040 2041commit 8c1bf9d8fe3948b72795984e625ef46b2f0bf654 2042Author: Dave Airlie <airlied@redhat.com> 2043Date: Mon Sep 3 13:08:05 2012 +1000 2044 2045 radeon: add pixmap sharing hooks. 2046 2047 This hooks into EXA and the dirty tracking to add sharing and output 2048 offload support. 2049 2050 Signed-off-by: Dave Airlie <airlied@redhat.com> 2051 2052commit 3add8df8122697acfe126d4857f3946ce44b8305 2053Author: Dave Airlie <airlied@redhat.com> 2054Date: Mon Sep 3 13:07:07 2012 +1000 2055 2056 radeon: add dri2 offload support. 2057 2058 This adds the new dri2 interfaces required for radeon to be offloaded to. 2059 2060 Signed-off-by: Dave Airlie <airlied@redhat.com> 2061 2062commit f7502a11c8ef9c453ceb40d26109977116df88c2 2063Author: Dave Airlie <airlied@redhat.com> 2064Date: Mon Sep 3 12:56:56 2012 +1000 2065 2066 radeon: add shared support to pixmaps. 2067 2068 this just adds the interface and shared support to the pixmap. 2069 2070 Signed-off-by: Dave Airlie <airlied@redhat.com> 2071 2072commit f71139a2afe8fffb628331402bf829a6d67c9fff 2073Author: Dave Airlie <airlied@redhat.com> 2074Date: Mon Sep 3 12:37:22 2012 +1000 2075 2076 radeon: add platform bus loading support. 2077 2078 This allows the radeon driver to be loaded via the platform bus mechanism. 2079 2080 Signed-off-by: Dave Airlie <airlied@redhat.com> 2081 2082commit 9911e72ea79e000ee0492d19a434aa98b83b5066 2083Author: Dave Airlie <airlied@redhat.com> 2084Date: Fri Aug 17 17:02:51 2012 +1000 2085 2086 radeon: damage full screen after pageflip 2087 2088 in order for offload to work we need to cause a full 2089 screen damage to be registered after pageflip. 2090 2091 Signed-off-by: Dave Airlie <airlied@redhat.com> 2092 2093commit 0873b8cef21abf1e7286731758f0db1e7f2e7c97 2094Author: Dave Airlie <airlied@redhat.com> 2095Date: Fri Aug 3 16:45:50 2012 +1000 2096 2097 ati: fix stub driver loader to load all drivers properly 2098 2099 Signed-off-by: Dave Airlie <airlied@redhat.com> 2100 2101commit cdf522f5992adf7b76f9153c777e6dddcf7758a6 2102Author: Adam Jackson <ajax@redhat.com> 2103Date: Wed Aug 15 13:38:28 2012 -0400 2104 2105 Implement ->driverFunc 2106 2107 Copied from fbdev, makes it so we can run without iopl. 2108 2109 Signed-off-by: Adam Jackson <ajax@redhat.com> 2110 2111commit 763f387ccac2399f1d58ec112c8bed45196bea5e 2112Author: Adam Jackson <ajax@redhat.com> 2113Date: Wed Aug 15 12:49:10 2012 -0400 2114 2115 Don't bother to include vbe.h 2116 2117 The only VBE code here went away when we dropped UMS. 2118 2119 Signed-off-by: Adam Jackson <ajax@redhat.com> 2120 2121commit aabcf5ea30195274c79ae94c0fe6edc965f5c123 2122Author: Tomas Chvatal <tchvatal@suse.cz> 2123Date: Mon Aug 13 22:58:18 2012 +0200 2124 2125 Add switch for udev to disable automagic detection. 2126 2127 Signed-off-by: Tomas Chvatal <tchvatal@suse.cz> 2128 2129commit aef90993621331f01d17b95a751d4924d77eed0c 2130Author: Alex Deucher <alexdeucher@gmail.com> 2131Date: Mon Aug 6 11:00:45 2012 -0400 2132 2133 radeon: add some new SI pci ids 2134 2135 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2136 2137commit 6ef1ad6a46348d3aecd8d1f5e94431ca2298853c 2138Author: Michel Dänzer <michel.daenzer@amd.com> 2139Date: Fri Jul 13 11:15:25 2012 +0200 2140 2141 Deal more gracefully with DRI2 being unavailable at build or run time. 2142 2143 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2144 2145commit ef8a404391036d8aa814dbda2407c789b8a64b92 2146Author: Michel Dänzer <michel.daenzer@amd.com> 2147Date: Thu Jul 5 20:14:48 2012 +0200 2148 2149 Initial SI support. 2150 2151 Defaults to shadowfb. 3D acceleration is available with glamor. 2D 2152 acceleration is disabled until the radeonsi driver can handle glamor's 2153 shaders. 2154 2155 v2: add chip flags (Alex Deucher) 2156 2157 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2158 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2159 2160commit e9edd2f5002c642b59f028b3ec076d604ae8ce9d 2161Author: Michel Dänzer <michel.daenzer@amd.com> 2162Date: Wed Jun 20 08:40:07 2012 +0200 2163 2164 Initial glamor support. 2165 2166 Enable at build time with --enable-glamor and runtime with 2167 2168 Option "AccelMethod" "glamor" 2169 2170 The most notable lack of functionality is XVideo. Use something like VDPAU for 2171 now. 2172 2173 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2174 2175commit 9eac8021f3d33a63156f9f5d43a220e88bb3f8db 2176Author: Michel Dänzer <michel.daenzer@amd.com> 2177Date: Wed Jun 27 14:48:47 2012 +0200 2178 2179 EXA: Factor out pixmap BO allocation into a helper function. 2180 2181 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2182 2183commit 060c7836e7f7777bacca4e23f57c5985beab33bc 2184Author: Michel Dänzer <michel.daenzer@amd.com> 2185Date: Wed Jul 4 11:49:18 2012 +0200 2186 2187 Some cosmetic fixups for the radeon manpage. 2188 2189 ATI -> ATI/AMD 2190 PCIE -> PCIe 2191 2192 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2193 2194commit c3a9f64fe40c78331b9340c3832d344f38b11564 2195Author: Michel Dänzer <michel.daenzer@amd.com> 2196Date: Wed Jul 4 11:46:30 2012 +0200 2197 2198 Drop UMS specific parts from the radeon manpage. 2199 2200 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2201 2202commit 5ebebfdb4fadada157d1369f35a8610c0fa72e02 2203Author: Dave Airlie <airlied@redhat.com> 2204Date: Fri Jun 29 15:03:43 2012 +0100 2205 2206 radeon: fix conflict with fix in master. 2207 2208 Signed-off-by: Dave Airlie <airlied@redhat.com> 2209 2210commit ff12ddf613ad42e4566d6e10fc6cf33a4e445736 2211Author: Dave Airlie <airlied@redhat.com> 2212Date: Fri Jun 29 15:00:44 2012 +0100 2213 2214 configure: bump to 6.99.99 2215 2216 this is in prep for 7.0 release. 2217 2218 Signed-off-by: Dave Airlie <airlied@redhat.com> 2219 2220commit 50689ec8dbd4a68527b2ac16cecac298b8d441d0 2221Merge: 6a60fcd ae682a6 2222Author: Dave Airlie <airlied@redhat.com> 2223Date: Fri Jun 29 14:59:47 2012 +0100 2224 2225 Merge remote-tracking branch 'origin/kms-only' 2226 2227 This merges the removal of the UMS code and subsequent simplifications. 2228 2229 Hell yes. 2230 2231 * origin/kms-only: (24 commits) 2232 radeon: fix radeonchipsets properly 2233 radeon: fix RADEONChipsets 2234 radeon: migrate remainder of radeon_driver.c to rest of driver 2235 radeon: avoid including large static struct twice. 2236 radeon: drop some more unused macros/inlines 2237 radeon: drop another unused struct member. 2238 radeon: more unused stuff 2239 radeon: drop cardType shouldn't matter to userspace. 2240 radeon: drop radeonGetPixmapOffset 2241 radeon: move more functions into texture video 2242 radeon: drop legacy memory, merge into radeon video 2243 radeon: drop unused txoffset vars. 2244 radeon: drop vb_mc_addr, not needed anymore 2245 radeon: drop offset member of r600 accel object 2246 radeon: move macros into radeon.h 2247 radeon: drop more wrapper macros. 2248 radeon: further macro cleanups. 2249 radeon: make exa copy funcs static. 2250 radeon: drop FUNC_NAME macro and ONCE_ONLY stuff 2251 radeon: drop radeon_commonfuncs. 2252 ... 2253 2254 Conflicts: 2255 configure.ac 2256 2257commit 6a60fcdf060a0f553d8d4f7939c2a05aa04bedfa 2258Author: Dave Airlie <airlied@redhat.com> 2259Date: Fri Jun 29 14:57:58 2012 +0100 2260 2261 configure: bump version after release. 2262 2263 Signed-off-by: Dave Airlie <airlied@redhat.com> 2264 2265commit 008eb209d506ed7195fd544dfa91be9783946fd5 2266Author: Dave Airlie <airlied@redhat.com> 2267Date: Fri Jun 29 14:46:55 2012 +0100 2268 2269 configure: bump version number to 6.14.6 pre-release 2270 2271 Signed-off-by: Dave Airlie <airlied@redhat.com> 2272 2273commit 72fea2635a56c4f572c07fc50fc58f703a21eb4f 2274Author: Dave Airlie <airlied@redhat.com> 2275Date: Fri Jun 29 14:46:17 2012 +0100 2276 2277 radeon: require libdrm 2.4.36 for KMS support. 2278 2279 This is due to some commits to the surface manager that fix bugs. 2280 2281 Signed-off-by: Dave Airlie <airlied@redhat.com> 2282 2283commit 179b035835bfa99283d44ba2d5ee17e686196535 2284Author: Michel Dänzer <michel.daenzer@amd.com> 2285Date: Wed Jun 27 19:26:51 2012 +0200 2286 2287 Fix up displayWidth vs. virtualX confusion in radeon_setup_kernel_mem(). 2288 2289 It was using the pitch (displayWidth) for the virtual width (virtualX). This 2290 prevented using page flipping in some cases, as displayWidth was already 2291 overaligned for virtualX, so the DRI2 front and back buffers ended up having 2292 a different pitch. 2293 2294 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2295 Reviewed-by: Jerome Glisse <jglisse@redhat.com> 2296 2297commit 5dc5012320536f7b1e86a81ed3b823e3626828e1 2298Author: Dave Airlie <airlied@redhat.com> 2299Date: Mon Jun 25 09:23:04 2012 +0100 2300 2301 radeon: dixPrivateKeyRegistered is only in server 1.9 2302 2303 just fallback to the old behaviour on older servers. 2304 2305 should fix: https://bugs.freedesktop.org/show_bug.cgi?id=51388 2306 Review-over-irc: Michel Dänzer <michel.daenzer@amd.com> 2307 Signed-off-by: Dave Airlie <airlied@redhat.com> 2308 2309commit ae682a6a1781ab023ab36e3154d6b3e88da63d64 2310Author: Dave Airlie <airlied@redhat.com> 2311Date: Mon Jun 18 13:11:31 2012 +0100 2312 2313 radeon: fix radeonchipsets properly 2314 2315commit 6514bcdc0fc199b4f3a8b4273bd29e0fe255d232 2316Author: Dave Airlie <airlied@redhat.com> 2317Date: Mon Jun 18 13:02:03 2012 +0100 2318 2319 radeon: fix RADEONChipsets 2320 2321 reported on irc by okias. 2322 2323 Signed-off-by: Dave Airlie <airlied@redhat.com> 2324 2325commit dd4ab5aba9047d522dbcfe8a341c368312e1a471 2326Author: Dave Airlie <airlied@redhat.com> 2327Date: Sat Jun 16 08:47:32 2012 +0100 2328 2329 radeon: migrate remainder of radeon_driver.c to rest of driver 2330 2331 This helps make a few more things static and the driver generally 2332 smaller. 2333 2334 Signed-off-by: Dave Airlie <airlied@redhat.com> 2335 2336commit a19237fad2492d70777167c631f4e6e1fae0e908 2337Author: Dave Airlie <airlied@redhat.com> 2338Date: Sat Jun 16 08:19:35 2012 +0100 2339 2340 radeon: avoid including large static struct twice. 2341 2342 Just extern this in the one file that needs access. 2343 2344 Signed-off-by: Dave Airlie <airlied@redhat.com> 2345 2346commit c8a5c5a0dd9adc46b03b898c9c1719dbbde05ba1 2347Author: Dave Airlie <airlied@redhat.com> 2348Date: Fri Jun 15 19:52:14 2012 +0100 2349 2350 radeon: drop some more unused macros/inlines 2351 2352 These are no longer used anywhere. 2353 2354 Signed-off-by: Dave Airlie <airlied@redhat.com> 2355 2356commit 3097f273443b5b8fce097e2b1e09a17b02d06a40 2357Author: Dave Airlie <airlied@redhat.com> 2358Date: Fri Jun 15 19:45:15 2012 +0100 2359 2360 radeon: drop another unused struct member. 2361 2362 Signed-off-by: Dave Airlie <airlied@redhat.com> 2363 2364commit cb97e75acab84b67b7b7358860788638efc9b344 2365Author: Dave Airlie <airlied@redhat.com> 2366Date: Fri Jun 15 19:38:45 2012 +0100 2367 2368 radeon: more unused stuff 2369 2370 Remove all CurrentLayout stuff. 2371 2372 Signed-off-by: Dave Airlie <airlied@redhat.com> 2373 2374commit 81593e7deb688fa3108a0589c1418459ec0df4de 2375Author: Dave Airlie <airlied@redhat.com> 2376Date: Fri Jun 15 18:52:52 2012 +0100 2377 2378 radeon: drop cardType shouldn't matter to userspace. 2379 2380 Signed-off-by: Dave Airlie <airlied@redhat.com> 2381 2382commit 997b0dddfeb180043ece5eff0ac0d6a1a751f9b5 2383Author: Dave Airlie <airlied@redhat.com> 2384Date: Fri Jun 15 18:43:01 2012 +0100 2385 2386 radeon: drop radeonGetPixmapOffset 2387 2388 no longer needed. 2389 2390 Signed-off-by: Dave Airlie <airlied@redhat.com> 2391 2392commit 639061edb3ac35a24e4d6633c9b6384434dcac64 2393Author: Dave Airlie <airlied@redhat.com> 2394Date: Fri Jun 15 18:38:48 2012 +0100 2395 2396 radeon: move more functions into texture video 2397 2398 make more thing static. 2399 2400 Signed-off-by: Dave Airlie <airlied@redhat.com> 2401 2402commit 85711a0a6a553232b603dbbb2b2d234c35391578 2403Author: Dave Airlie <airlied@redhat.com> 2404Date: Fri Jun 15 17:26:50 2012 +0100 2405 2406 radeon: drop legacy memory, merge into radeon video 2407 2408 This merges these two functions into textured video code. 2409 2410 Signed-off-by: Dave Airlie <airlied@redhat.com> 2411 2412commit 445ca1d66ce7b37762731fe67407cc1b978bb129 2413Author: Dave Airlie <airlied@redhat.com> 2414Date: Fri Jun 15 17:13:48 2012 +0100 2415 2416 radeon: drop unused txoffset vars. 2417 2418 These vars aren't being used anymore. 2419 2420 Signed-off-by: Dave Airlie <airlied@redhat.com> 2421 2422commit 832244de9e6d7aa1b8fbeb29c9a6a86923744d60 2423Author: Dave Airlie <airlied@redhat.com> 2424Date: Fri Jun 15 17:09:00 2012 +0100 2425 2426 radeon: drop vb_mc_addr, not needed anymore 2427 2428 This field is totally unused now, so drop it. 2429 2430 Signed-off-by: Dave Airlie <airlied@redhat.com> 2431 2432commit 6559b6e28bf111b64c3aeb5ec1260acfeb3e0b7f 2433Author: Dave Airlie <airlied@redhat.com> 2434Date: Fri Jun 15 17:05:54 2012 +0100 2435 2436 radeon: drop offset member of r600 accel object 2437 2438 This is always 0 now. 2439 2440 Signed-off-by: Dave Airlie <airlied@redhat.com> 2441 2442commit 857179e538c9752fab239a21312d0fcc80022151 2443Author: Dave Airlie <airlied@redhat.com> 2444Date: Fri Jun 15 16:55:23 2012 +0100 2445 2446 radeon: move macros into radeon.h 2447 2448 Drop separate header file, there isn't many separate macros now. 2449 2450 Signed-off-by: Dave Airlie <airlied@redhat.com> 2451 2452commit 148e889149f5ecc3dc2473f2a9e01572eeb037d3 2453Author: Dave Airlie <airlied@redhat.com> 2454Date: Fri Jun 15 16:52:24 2012 +0100 2455 2456 radeon: drop more wrapper macros. 2457 2458 Signed-off-by: Dave Airlie <airlied@redhat.com> 2459 2460commit ad8629e3323efe484dd1421d94dd934e1bfafe56 2461Author: Dave Airlie <airlied@redhat.com> 2462Date: Fri Jun 15 16:49:13 2012 +0100 2463 2464 radeon: further macro cleanups. 2465 2466 drop some more of the wrapper macros. 2467 2468 Signed-off-by: Dave Airlie <airlied@redhat.com> 2469 2470commit 96644f96f1a2df1d7d1ca8cbdc9a105affdb4f32 2471Author: Dave Airlie <airlied@redhat.com> 2472Date: Fri Jun 15 16:34:53 2012 +0100 2473 2474 radeon: make exa copy funcs static. 2475 2476 These are no longer used outside this area. 2477 2478 Signed-off-by: Dave Airlie <airlied@redhat.com> 2479 2480commit 83904b820b81d226940ec6d663246f243f92e7ba 2481Author: Dave Airlie <airlied@redhat.com> 2482Date: Fri Jun 15 16:30:41 2012 +0100 2483 2484 radeon: drop FUNC_NAME macro and ONCE_ONLY stuff 2485 2486 Drop all the CP vs MMIO macros that name functions. 2487 2488 Signed-off-by: Dave Airlie <airlied@redhat.com> 2489 2490commit b422d8085aab4e01a1ac4ffb247c73fe51aafa0d 2491Author: Dave Airlie <airlied@redhat.com> 2492Date: Fri Jun 15 16:14:54 2012 +0100 2493 2494 radeon: drop radeon_commonfuncs. 2495 2496 modify all the macros and drop it into radeon_accel.c 2497 2498 Signed-off-by: Dave Airlie <airlied@redhat.com> 2499 2500commit 29dd3f84eecd2892a438f004403521de8986a91d 2501Author: Dave Airlie <airlied@redhat.com> 2502Date: Fri Jun 15 16:11:03 2012 +0100 2503 2504 radeon: misc cleanups. 2505 2506 Signed-off-by: Dave Airlie <airlied@redhat.com> 2507 2508commit 678c1f3425c0fa7e98795c8e351da77d04652e02 2509Author: Dave Airlie <airlied@redhat.com> 2510Date: Fri Jun 15 16:06:27 2012 +0100 2511 2512 make pci chipsets static 2513 2514 Signed-off-by: Dave Airlie <airlied@redhat.com> 2515 2516commit 95cbbe999418c72dc407069d31e759d70f49ab16 2517Author: Dave Airlie <airlied@redhat.com> 2518Date: Fri Jun 15 16:01:53 2012 +0100 2519 2520 drop RING_LOCALS/ACCEL_PREAMBLE 2521 2522 these were unused now. 2523 2524 Signed-off-by: Dave Airlie <airlied@redhat.com> 2525 2526commit 18d5ae3bd9075ac1a2ee21b071ac133e2e634b62 2527Author: Dave Airlie <airlied@redhat.com> 2528Date: Fri Jun 15 10:05:03 2012 +0100 2529 2530 radeon: drop all UMS/DRI1/XAA/overlay support. 2531 2532 This overhauls the radeon driver and removes all the old UMS-only code, 2533 it drops all the UMS, DRI1, XAA, overlay Xv, video capture, tv tuners 2534 2535 There are probably a lot more cleanups that will fall out of this afterwards. 2536 2537 So far this is compile/build tested. 2538 2539 Signed-off-by: Dave Airlie <airlied@redhat.com> 2540 2541commit 248e912c487636d7352cfad43c03fc9f19fc2215 2542Author: Michel Dänzer <michel.daenzer@amd.com> 2543Date: Fri Jun 8 12:18:48 2012 +0200 2544 2545 UMS: Fix CRTC DPMS state check. 2546 2547 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2548 2549commit 58883711136fa6671d4f3250d4ee973e2953ebb4 2550Author: Alex Deucher <alexdeucher@gmail.com> 2551Date: Thu Jun 7 11:47:21 2012 -0400 2552 2553 configure: bump version post release 2554 2555 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2556 2557commit 3a9103f855c67bbcd7b97ee2e96836353e0d789d 2558Author: Alex Deucher <alexdeucher@gmail.com> 2559Date: Thu Jun 7 11:27:08 2012 -0400 2560 2561 configure: bump for release 2562 2563 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2564 2565commit f1693ee09e9b251eb75e3e5fe0ee7699fec6fb98 2566Author: Alex Deucher <alexdeucher@gmail.com> 2567Date: Thu Jun 7 11:23:31 2012 -0400 2568 2569 configure: bump libdrm_radeon requirement 2570 2571 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2572 2573commit 77a056c1725c9c44a328ed324687bdf35144e9e2 2574Author: Michel Dänzer <michel.daenzer@amd.com> 2575Date: Wed Jun 6 12:29:37 2012 +0200 2576 2577 Don't page-flip or wait on a CRTC while we're VT-switched away. 2578 2579 We don't know what the CRTC state is. 2580 2581 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2582 2583commit 1d9f1eb7e65c57c52759ccbfa51e8bdd5a10acac 2584Author: Michel Dänzer <michel.daenzer@amd.com> 2585Date: Wed Jun 6 12:16:08 2012 +0200 2586 2587 Don't wait on a CRTC which has been disabled via DPMS (bug #49761). 2588 2589 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=49761 . 2590 2591 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2592 2593commit 4603285aa8efaf2614f15a38254465cec2075f11 2594Author: Alex Deucher <alexdeucher@gmail.com> 2595Date: Tue Jun 5 09:48:59 2012 -0400 2596 2597 radeon: add new PCI ids 2598 2599 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2600 2601commit 9307609420b4b209767d2057b4803bcb16cc1455 2602Author: Dave Airlie <airlied@redhat.com> 2603Date: Sat Jun 2 17:16:34 2012 +0100 2604 2605 radeon: improve smooth startup fallbacks. 2606 2607 If we can't handover the framebuffer, memset it to black. 2608 2609 mostly ported from nouveau. 2610 2611 Signed-off-by: Dave Airlie <airlied@redhat.com> 2612 2613commit 89a4c79a43a2b6963e41d7812e8fe5beedb556af 2614Author: Dave Airlie <airlied@redhat.com> 2615Date: Sat Jun 2 17:14:14 2012 +0100 2616 2617 radeon: fix smooth startup with tiling enabled. 2618 2619 We need to use the surface we worked out when we allocated the front bo, 2620 not work out a new surface from scratch. 2621 2622 This fixes smooth handover from plymouth to gdm on F17. 2623 2624 Signed-off-by: Dave Airlie <airlied@redhat.com> 2625 2626commit c1b9b2c9d71a87a0b98ccdb542da8a30c8c4d99a 2627Author: Dave Airlie <airlied@redhat.com> 2628Date: Fri Jun 1 12:22:06 2012 +0100 2629 2630 radeon: fix compat api for newest planned API. 2631 2632 Signed-off-by: Dave Airlie <airlied@redhat.com> 2633 2634commit 68c51d148b22ee8cbe98b5d5441e3494804ee808 2635Author: Dave Airlie <airlied@redhat.com> 2636Date: Fri Jun 1 12:21:34 2012 +0100 2637 2638 drmmode: drop flags arg to adjust frame. 2639 2640 Signed-off-by: Dave Airlie <airlied@redhat.com> 2641 2642commit b1e9c3081d2b975682a62389aa29063b79f9655c 2643Author: Dave Airlie <airlied@redhat.com> 2644Date: Thu May 24 17:54:41 2012 +0100 2645 2646 radeon: move include down further 2647 2648 fixes tinderbox build. 2649 2650 Signed-off-by: Dave Airlie <airlied@redhat.com> 2651 2652commit 5ed959585f35a4eff2cd803ae222b46d164f210d 2653Author: Dave Airlie <airlied@redhat.com> 2654Date: Thu May 24 08:17:11 2012 +0100 2655 2656 radeon: update compat-api.h for block handler + enable/disable fb. 2657 2658 This updates the compat stuff for the latest block handler code, 2659 and the enable/disable interface. 2660 2661 Signed-off-by: Dave Airlie <airlied@redhat.com> 2662 2663commit 2de17efdcb502f91fb145130c12be3cb164f03c5 2664Author: Dave Airlie <airlied@redhat.com> 2665Date: Thu May 24 08:16:39 2012 +0100 2666 2667 radeon/xvmc: fix build with new API. 2668 2669 This was missing the compat include. 2670 2671 Signed-off-by: Dave Airlie <airlied@redhat.com> 2672 2673commit a1f47768309aa8b14b13f05ba01bb490024c4cbc 2674Author: Dave Airlie <airlied@redhat.com> 2675Date: Thu May 24 07:55:14 2012 +0100 2676 2677 radeon: fix a few more RHDAtomBiosFunc uses 2678 2679 I missed these in my initial search/replace for some reason. 2680 2681 Signed-off-by: Dave Airlie <airlied@redhat.com> 2682 2683commit c26e40cf60eb6a2bffb5ec7cba1f1bbffcbf1d53 2684Author: Dave Airlie <airlied@redhat.com> 2685Date: Wed May 23 12:12:54 2012 +0100 2686 2687 radeon/atombios: rewrite atombios parser code to not use xf86Screens. 2688 2689 Just pass the ScrnInfoPtr around instead. 2690 2691 Signed-off-by: Dave Airlie <airlied@redhat.com> 2692 2693commit 14122214edabd2a574b66577289a7c8732f4e953 2694Author: Dave Airlie <airlied@redhat.com> 2695Date: Wed May 23 11:57:35 2012 +0100 2696 2697 radeon/generic_bus: stop passing scrnIndex 2698 2699 Just pass a pointer to the screen, removes usage of xf86Screens lookup 2700 2701 Signed-off-by: Dave Airlie <airlied@redhat.com> 2702 2703commit 2101e485480e3ffd6a17ed12f5287b8b1e4e0223 2704Author: Dave Airlie <airlied@redhat.com> 2705Date: Tue May 22 16:17:55 2012 +0100 2706 2707 compat for new server API 2708 2709commit 9d2549057ffced996bfaeb8df4901ef57da636c7 2710Author: Dave Airlie <airlied@redhat.com> 2711Date: Wed May 16 15:45:53 2012 +0100 2712 2713 ati: convert to new screen conversion APIs 2714 2715 The compat header takes care of the old server vs new server. 2716 2717 this commit was autogenerated from util/modular/x-driver-screen-scrn-conv.sh 2718 2719 Signed-off-by: Dave Airlie <airlied@redhat.com> 2720 2721commit 80a45bfc26e20d1d481f13cb15c0720d165a8306 2722Author: Dave Airlie <airlied@redhat.com> 2723Date: Wed May 16 15:42:59 2012 +0100 2724 2725 radeon: add compat-api.h 2726 2727commit 84c3900b5179a8d2cc5f925e94ed5b2bbdee1ad9 2728Author: Dave Airlie <airlied@redhat.com> 2729Date: Wed May 23 08:49:45 2012 +0100 2730 2731 radeon/xaa: drop scrnIndex parameter to some functions 2732 2733 This isn't needed, and makes api changes later easier. 2734 2735 Signed-off-by: Dave Airlie <airlied@redhat.com> 2736 2737commit 3fb694b308ebadd1b849836059b6b56bb19385f7 2738Author: Thierry Vignaud <thierry.vignaud@gmail.com> 2739Date: Wed May 16 14:43:53 2012 +0200 2740 2741 UMS/EXA: Add reminder for potential solid picture performance issue. 2742 2743commit 4b9bad959438725e4434e2aa4f142542d8dcc260 2744Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 2745Date: Fri May 11 05:25:32 2012 +0200 2746 2747 radeon: use GB_GR and BG_RG formats for packed yuv video for r600+ 2748 2749 Those formats were invented for exactly that purpose so use them. 2750 This saves some code and also some hw resources (only need one 2751 sampler instead of two for packed yuv). 2752 Only tested on EG. 2753 2754commit 37786e9027b8c8d1f9ec9928915784dd28853766 2755Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 2756Date: Fri May 11 05:25:33 2012 +0200 2757 2758 radeon: avoid rounding errors in texture coords for textured xv on EG+ 2759 2760 make sure the division is done with floats, otherwise the coordinate 2761 can be wrong up to 1 texel. 2762 Particularly visible with clipping and small source scaled up (since one 2763 texel can be a shift of several pixels) but could be seen even unscaled. 2764 Should provide more accurate coords without clipping too depending on the 2765 scale factor probably. 2766 This is a straight port of 688c8a54a00b01e73a11970ad2abe858f8c7c5c4 2767 when I apparently forgot the eg code... 2768 2769commit faea3aafa8c9a7c1bc866ffcd847972c5569c8bc 2770Author: Anisse Astier <anisse@astier.eu> 2771Date: Thu May 10 17:22:59 2012 +0200 2772 2773 Fail more gracefully when drm surface manager can't be initialized 2774 2775 Should make bugs like https://bugs.freedesktop.org/show_bug.cgi?id=48138 2776 easier to diagnose. 2777 2778 [ Michel Dänzer: Appended newline to error message. ] 2779 2780 Signed-off-by: Anisse Astier <anisse@astier.eu> 2781 Singed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2782 2783commit b0b7d8d26fd107df342b5c87b0a38e5bb08101a9 2784Author: Michel Dänzer <michel.daenzer@amd.com> 2785Date: Wed May 9 11:08:49 2012 +0200 2786 2787 EXA/UMS: Synchronize to the GPU before writing solid colour to scratch pixmap. 2788 2789 UMS doesn't do this automagically. It's a big hammer that will probably suck 2790 for performance, but I don't have any better ideas right now. 2791 2792 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2793 2794commit 6bda7ceda645e838723883d133d614def1511d16 2795Author: Michel Dänzer <michel.daenzer@amd.com> 2796Date: Thu May 3 15:07:30 2012 +0200 2797 2798 EXA: Attempt to fix solid picture acceleration with UMS. 2799 2800 Only compile tested, but should fix 2801 https://bugs.freedesktop.org/show_bug.cgi?id=49182 . 2802 2803 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2804 2805commit b33cf9cf9dce5add8aaccab10ba5aaa64247c8ce 2806Author: Alex Deucher <alexdeucher@gmail.com> 2807Date: Wed May 2 09:55:16 2012 -0400 2808 2809 radeon: add MacModel entry for SAM440ep embedded board 2810 2811 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2812 2813commit 0bda305f7ab2a4720b3fea3f318ab2a73be151e5 2814Author: Alex Deucher <alexander.deucher@amd.com> 2815Date: Fri Apr 13 13:28:10 2012 -0400 2816 2817 EXA: Support acceleration of solid pictures on R2xx. 2818 2819 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2820 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2821 2822commit a7754b076e355fef3aea082b6e2d3aefbb8e7a9b 2823Author: Alex Deucher <alexander.deucher@amd.com> 2824Date: Fri Apr 13 13:24:46 2012 -0400 2825 2826 EXA: Support acceleration of solid pictures on R1xx. 2827 2828 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2829 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2830 2831commit cac2d7ed54918579418fc762558497d3ec547fad 2832Author: Alex Deucher <alexander.deucher@amd.com> 2833Date: Sat Apr 14 08:53:39 2012 -0400 2834 2835 EXA: Support acceleration of solid pictures on Evergreen/NI. 2836 2837 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 2838 Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> 2839 2840commit d88b9700137ee407c483f263bb55c77cd6f92fef 2841Author: Michel Dänzer <michel.daenzer@amd.com> 2842Date: Fri Apr 13 09:35:38 2012 +0200 2843 2844 EXA: Support acceleration of solid pictures on R3xx-R7xx. 2845 2846 Allocate 1x1 scratch pixmaps to hold the solid picture colours. 2847 2848 This works around https://bugs.freedesktop.org/show_bug.cgi?id=47266 and might 2849 improve performance in other cases as well. 2850 2851 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2852 2853commit 66b586b9b9cdaf70f0fcd547b5a04f044d848d44 2854Author: Michel Dänzer <michel.daenzer@amd.com> 2855Date: Mon Apr 16 15:33:36 2012 +0200 2856 2857 RADEONCopySwap: Fix RADEON_HOST_DATA_SWAP_16BIT case. 2858 2859 It was the same code as for RADEON_HOST_DATA_SWAP_32BIT. This caused bus errors 2860 on FreeBSD/PPC, but I'm not sure how it could not cause problems anywhere... 2861 2862 Reported-by: Andreas Tobler <andreast@fgznet.ch> 2863 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2864 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 2865 2866commit 1e656cd5973bbcf85b63406638ede0bccd65e28f 2867Author: Michel Dänzer <michel.daenzer@amd.com> 2868Date: Mon Apr 16 15:44:01 2012 +0200 2869 2870 Make radeon_setup_kernel_mem failures more graceful and verbose. 2871 2872 So that bugs like https://bugs.freedesktop.org/show_bug.cgi?id=48138 can be 2873 diagnosed more easily. 2874 2875 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2876 2877commit d282719a9c2fb0ee32830aa75b8dfbb9392954ed 2878Author: Jerome Glisse <jglisse@redhat.com> 2879Date: Wed Apr 4 17:08:30 2012 -0400 2880 2881 r6xx-r9xx: force 1D tiling for buffer with height < 64 2882 2883 Due to some old kernel issue, height is 8 aligned insided the ddx 2884 For buffer with height btw 57 & 63 this lead ddx to believe it can 2885 allocate a 2D tiled surface while mesa will not align height and 2886 will assume 1D tiled leading to disagreement and rendering issue. 2887 This patch force buffer with height < 64 to be 1D tiled. 2888 2889 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 2890 2891commit de2419e5509398328903ca61f4ea80852cba6bc4 2892Author: Alex Deucher <alexdeucher@gmail.com> 2893Date: Thu Mar 29 00:28:39 2012 -0400 2894 2895 configure: bump version post release 2896 2897 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2898 2899commit 9425c50e93903fb64d9e569cfdc1e2c35d16ce25 2900Author: Alex Deucher <alexdeucher@gmail.com> 2901Date: Thu Mar 29 00:19:12 2012 -0400 2902 2903 configure: bump version for release 2904 2905 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2906 2907commit dc18d771713ecd893c7d5833da6e0661093161dc 2908Author: Alex Deucher <alexdeucher@gmail.com> 2909Date: Wed Mar 28 17:32:53 2012 -0400 2910 2911 configure: bump libdrm requirement for TN support 2912 2913 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2914 2915commit 03535904a3e1542b3924d0a062c4b022ca196888 2916Author: Alex Deucher <alexdeucher@gmail.com> 2917Date: Tue Mar 27 09:48:28 2012 -0400 2918 2919 radeon: man page updates 2920 2921 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2922 2923commit 6ed191c33caa33c12c2c6dafcba3a5ab1bf4a02f 2924Author: Alex Deucher <alexdeucher@gmail.com> 2925Date: Tue Mar 20 19:57:53 2012 -0400 2926 2927 radeon/kms: add TN pci ids 2928 2929 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2930 2931commit 55d65fcf33eb383e3fbc7a1d469ab68a70a7ab37 2932Author: Alex Deucher <alexdeucher@gmail.com> 2933Date: Tue Mar 20 19:54:57 2012 -0400 2934 2935 radeon/kms: add support for TN (trinity) APUs 2936 2937 - KMS only 2938 - Includes full EXA/Xv support 2939 2940 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2941 2942commit b5cf9bd693cf4090956add4c33c4fae9c3069a03 2943Author: Marek Olšák <maraeo@gmail.com> 2944Date: Wed Mar 7 11:01:40 2012 -0500 2945 2946 r6xx: initialize SX_MISC 2947 2948 If Mesa set it to 1, the DDX would not render anything = the monitor would 2949 basically freeze. 2950 2951 agd5f: update emit count as well. 2952 2953 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2954 2955commit 355dc4295912c153f5333421594fa90aa119a056 2956Author: Michel Dänzer <michel.daenzer@amd.com> 2957Date: Tue Mar 6 15:52:40 2012 +0100 2958 2959 DRI2: Unreference buffers immediately when event wait info is invalidated. 2960 2961 Deferring this could result in trying to unreference buffers from a previous 2962 server generation, i.e. accessing freed memory. 2963 2964 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2965 Tested-by: Christian König <Christian.koenig@amd.com> 2966 2967commit fe51469b2e02e4d565050bab077985270fb58a9b 2968Author: Michel Dänzer <michel.daenzer@amd.com> 2969Date: Tue Mar 6 15:52:40 2012 +0100 2970 2971 Re-register DRM FD wakeup handler for each server generation. 2972 2973 Fixes hang when trying to use DRI2 swap scheduling after a server reset. 2974 2975 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 2976 Tested-by: Christian König <Christian.koenig@amd.com> 2977 2978commit 878454ae8d8e96dd27a19d0b30940d014c4cd7e2 2979Author: Hans Verkuil <hverkuil@xs4all.nl> 2980Date: Fri Feb 24 09:35:39 2012 -0500 2981 2982 Fix ConnectorTable crash in radeon_output.c 2983 2984 The sam440ep PPC board requires a ConnectorTable xorg.conf option, but putting 2985 in that option causes the radeon driver to crash. I finally traced it to a 2986 copy-and-paste bug in radeon_output.c as a result of a major rework in commit 2987 82f12e5a40c1fbcb91910a0f8b725c34fff02aae. 2988 2989 The actual crash occurred in RADEONPrintPortMap(). 2990 2991 Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> 2992 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 2993 2994commit 688c8a54a00b01e73a11970ad2abe858f8c7c5c4 2995Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 2996Date: Sat Feb 18 21:12:34 2012 +0100 2997 2998 radeon: avoid rounding errors in texture coords for textured xv 2999 3000 make sure the division is done with floats, otherwise the coordinate 3001 can be wrong up to 1 texel. 3002 Particularly visible with clipping and small source scaled up (since one 3003 texel can be a shift of several pixels) but could be seen even unscaled. 3004 Should provide more accurate coords without clipping too depending on the 3005 scale factor probably. 3006 Changed for r100-r600, though only tested on r300. 3007 3008commit 2778b56252124ef6f636a493d2e1457b43911c37 3009Author: Jerome Glisse <jglisse@redhat.com> 3010Date: Mon Feb 13 20:42:57 2012 -0500 3011 3012 radeon: r6xx-eg use linear general when using scratch bo 3013 3014 In path where we need to use scratch bo as temporary area, 3015 consider it as linear buffer. Not linear aligned. Fix some 3016 case such as in bugs: 3017 3018 https://bugs.freedesktop.org/show_bug.cgi?id=45827 3019 3020 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 3021 3022commit c66ae235700f5efe64eb168327551b8f1d153c9c 3023Author: Michel Dänzer <michel.daenzer@amd.com> 3024Date: Mon Feb 13 10:43:58 2012 +0100 3025 3026 Handle new xorg_list API. 3027 3028 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=45937 3029 3030 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3031 3032commit 7ff277e22c629308915307bbee96eb25ff77f8b9 3033Author: Alex Deucher <alexdeucher@gmail.com> 3034Date: Fri Feb 10 13:04:59 2012 -0500 3035 3036 radeon: fix crash in drmmode_create_bo_pixmap() 3037 3038 Only init surface on r6xx+. Return NULL rather than 3039 FALSE. 3040 3041 Fixes: 3042 https://bugs.freedesktop.org/show_bug.cgi?id=45829 3043 3044 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3045 3046commit 60b949f34df5db05e0e102cc3daa33469aa50cfc 3047Author: Alex Deucher <alexdeucher@gmail.com> 3048Date: Fri Feb 10 13:11:26 2012 -0500 3049 3050 radeon/kms: reusing fd message is not an error 3051 3052 It's standard behavior. 3053 3054 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3055 3056commit e20284409937d784847339b5d466a95012d85940 3057Author: Michel Dänzer <michel.daenzer@amd.com> 3058Date: Fri Feb 3 12:21:59 2012 +0100 3059 3060 EXA/r6xx+: Only set write domain or read domains, not both. (Bug #43893) 3061 3062 Avoids an accounting bug in libdrm_radeon 2.4.31 or older. 3063 3064 See https://bugs.freedesktop.org/show_bug.cgi?id=43893 3065 3066 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3067 3068commit 5a7f64c5170ca424c9dca739662018e30df13413 3069Author: Michel Dänzer <michel.daenzer@amd.com> 3070Date: Wed Feb 8 11:35:35 2012 +0100 3071 3072 Fix UMS build failure. 3073 3074 And some UMS specific warnings. 3075 3076 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3077 3078commit 87f776b966f4200c97a989536d4b71822ae4c0b3 3079Author: Michel Dänzer <michel.daenzer@amd.com> 3080Date: Wed Feb 8 11:10:27 2012 +0100 3081 3082 Remove unused local variable 'height'. 3083 3084 Pointed out by gcc -Wunused-variable. 3085 3086 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3087 3088commit f63262e2f112a348c45f0dcecd891c8b6d9c5ee8 3089Author: Michel Dänzer <michel.daenzer@amd.com> 3090Date: Wed Feb 8 10:28:45 2012 +0100 3091 3092 evergreen: Initialize source surface member for textured video. 3093 3094 Fixes crash reported by Ole Salscheider on IRC. 3095 3096 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3097 3098commit c8f104d38870f14049402bbc14f662c151caeeef 3099Author: Jerome Glisse <jglisse@redhat.com> 3100Date: Tue Feb 7 15:04:37 2012 -0500 3101 3102 radeon: fix tiling for weird resolution 3103 3104 Should also fix xv for some case. 3105 3106 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 3107 3108commit 615033f2b5e3817e335e9d022fc9fdcf8ac8b11a 3109Author: Jerome Glisse <jglisse@redhat.com> 3110Date: Tue Dec 13 11:08:19 2011 -0500 3111 3112 r600-evergreen: use common surface allocator for tiling v11 3113 3114 Use libdrm common surface code so mesa,ddx have same idea 3115 about tiling surface and what their pitch should be and 3116 the alignment constraint. 3117 3118 v2 fix remaining issue add new option to conditionaly enable 3119 v3 fix fbcon copy and r600 exa copy path 3120 v4 fix non tiled path 2D tiling on GPU >= R600, set it to false 3121 as default 3122 v5 adapt to pixel/element size split of libdrm/radeon 3123 v6 update to properly handle falling back to 1d tiled 3124 v6 final fix to tile split value on evergreen and newer 3125 v7 fix default array mode on r6xx, fix height alignment issue 3126 on evergreen 3127 v8 fix tile split value 3128 v9 add stencil tile split support, simplify dri2 for stencil 3129 with evergreen 3130 v10 Try to fix xv path regarding tiling. Adapt to libdrm API 3131 change. Try to fix case where there is no surface which 3132 means non tiled bo. 3133 v11 check for proper libdrm 3134 3135 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 3136 3137commit 36c190671081967bac6fff48aaf66d67b639a48c 3138Author: Michel Dänzer <michel.daenzer@amd.com> 3139Date: Wed Feb 1 13:21:02 2012 +0100 3140 3141 Fix vline range calculations. 3142 3143 The range passed in is in pixmap coordinates, so the CRTC offset needs to be 3144 added to the clamping limits and subtracted from the clamped range for 3145 pre-AVIVO display engines. 3146 3147 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3148 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3149 3150commit bb0e1531ac6949d38025d7dcb19234fee33b2acf 3151Author: Michel Dänzer <michel.daenzer@amd.com> 3152Date: Wed Feb 1 13:07:11 2012 +0100 3153 3154 Check for empty vline ranges after clamping. 3155 3156 The clamping could turn a previously non-empty range into an empty one. 3157 3158 Also, start == stop means the range is empty. 3159 3160 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3161 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3162 3163commit 13b3aed4ef9afbcbaea1dcf0ed1acb162b240a3f 3164Author: Alex Deucher <alexdeucher@gmail.com> 3165Date: Tue Jan 10 09:35:09 2012 -0500 3166 3167 EXA/r6xx+: fix rop setting for overlapping copies 3168 3169 Need to use GXCopy for the src to temp copy, then 3170 the original rop for the temp to dest copy. 3171 3172 Noticed by: Frank Huang 3173 3174 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3175 3176commit eb6d769a087b2ed5952f477fc3f0b0625810a287 3177Author: Egbert Eich <eich@suse.de> 3178Date: Tue Nov 15 18:50:56 2011 +0100 3179 3180 DPMS: Split non-modeset CRTC DPMS function. 3181 3182 RADEONRestore() calls crtc->funcs->dpms() after most of the mode setting 3183 subsystems have been restored. This function enables the CRTCs but does 3184 more: it calls DRM pre- and post-modeset ioctls and sets up the palettes 3185 (LUTs). 3186 None of these two things are needed. Accessing the palette registers after 3187 restoring the PLLs can even lead to lockups. 3188 Thus the CRTC DPMS function is split into two parts: one that just enables 3189 /disables the CRTC and one which wraps this function and does the rest. 3190 Now the inner function can be called directly from RADEONRestore() as 3191 there is no need to go thru the RandR hooks in this function while the 3192 RandR hook uses the wrappering function so the full functionality is 3193 preserved from an RandR point of view. 3194 3195 Signed-off-by: Egbert Eich <eich@freedesktop.org> 3196 Reviewed-by: Alex Deucher <alexdeucher@gmail.com> 3197 3198commit ac51e331895b216d288bc7bd108a38b362214668 3199Author: Egbert Eich <eich@suse.de> 3200Date: Mon Nov 14 19:10:01 2011 +0100 3201 3202 UMS: Fix lockups in palette save/restore on pre-AVIVO chips. 3203 3204 The reintroduction of palette save/restore in 5efdf514 causes some 3205 pre-AVIVO chips to lock up. An investigation revealed that accessing 3206 palette registers when the associated PLL is not running is causing 3207 this. With UMS the PLL setup that is saved has been done by the BIOS 3208 typically. 3209 A similar issue was observed when VGA palette save/restore had 3210 been reinitroduced with 80eee856: 3211 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=480312 3212 and has been worked around for Linux without further investigation 3213 by 87e66ce7. 3214 To fix the issue we now 3215 a. introduce 'on-demand' palette saving (ie the palette is 3216 saved before it is first altered). This guarantees that 3217 the palette register are only associated when the associated 3218 CRTC is active and thus the PLLs are powered up and running. 3219 b. move palette restore before PLL restore. 3220 c. eliminate generic VGA palette save/restore which seems to be 3221 unneeded when the palette is restored natively. 3222 It is believed that this caused the behavior described in 3223 https://bugs.freedesktop.org/show_bug.cgi?id=18407#c27 3224 3225 Signed-off-by: Egbert Eich <eich@freedesktop.org> 3226 Reviewed-by: Alex Deucher <alexdeucher@gmail.com> 3227 3228commit 0a8d04eeac95f4db9d03ee31070bd825a7feb0b2 3229Author: Matthieu Herrb <matthieu.herrb@laas.fr> 3230Date: Sun Jan 1 18:27:54 2012 +0100 3231 3232 Update for new vgaHW API. 3233 3234 Signed-off-by: Matthieu Herrb <matthieu.herrb@laas.fr> 3235 Reviewed-by: Mark Kettenis <kettenis@openbsd.org> 3236 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3237 3238commit ae45d7e6d8e6844cd4586c9ee97c21b257fa788f 3239Author: Michel Dänzer <michel.daenzer@amd.com> 3240Date: Wed Dec 28 11:48:36 2011 +0100 3241 3242 DRI2: Can't use page flipping for pixmaps. (bug #42913) 3243 3244 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=42913 . 3245 3246 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3247 3248commit aacbd629b02cbee3f9e6a0ee452b4e3f21376bd3 3249Author: Alex Deucher <alexdeucher@gmail.com> 3250Date: Mon Dec 12 09:32:30 2011 -0500 3251 3252 radeon: add some new pci ids 3253 3254 fixes: 3255 https://bugs.freedesktop.org/show_bug.cgi?id=43739 3256 3257 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3258 3259commit bc54e415e2fd344939c5c788ea0686133a7e2c69 3260Author: Dave Airlie <airlied@redhat.com> 3261Date: Tue Dec 6 15:47:45 2011 +0000 3262 3263 radeon: add original radeon to always tiled. 3264 3265 and actually enable it for M7, previous commit only did one function. 3266 3267 Signed-off-by: Dave Airlie <airlied@redhat.com> 3268 3269commit ba46c7b0cf72d157748981eb3224d5eefb6200aa 3270Author: Dave Airlie <airlied@redhat.com> 3271Date: Tue Dec 6 13:42:49 2011 +0000 3272 3273 radeon: refine always tiled depth check 3274 3275 So it appears the M7 family always tiles its depth buffer also. 3276 3277 Signed-off-by: Dave Airlie <airlied@redhat.com> 3278 3279commit 98b2d5fe1722a43c4bbe7711ed7180a3fb65305f 3280Author: Dave Airlie <airlied@redhat.com> 3281Date: Mon Dec 5 18:44:28 2011 +0000 3282 3283 radeon: r200 depth buffers are always tiled 3284 3285 When we do the allocations we need to make sure the always tiled 3286 nature is taken into account. 3287 3288 Signed-off-by: Dave Airlie <airlied@redhat.com> 3289 3290commit 7dcefc69d9fbceae27cd03083c815e01a19b527e 3291Author: Alex Deucher <alexdeucher@gmail.com> 3292Date: Mon Dec 5 09:21:48 2011 -0500 3293 3294 Xv: Evergreen+ asics support 16k surfaces 3295 3296 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3297 3298commit 72e386d42516e7cd3c2cbf2fffc9174cd3ec8451 3299Author: Alex Deucher <alexander.deucher@amd.com> 3300Date: Wed Nov 30 19:38:35 2011 -0500 3301 3302 radeon: add some new pci ids 3303 3304 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3305 3306commit 3853c3020d05175ae180b9a188dec7c425bdd0b8 3307Author: Dave Airlie <airlied@redhat.com> 3308Date: Mon Nov 28 18:38:30 2011 +0000 3309 3310 fixup xinerama since 9151f3b1c2ebcc34e63195888ba696f2183ba5e2 3311 3312 since the driver would call RRFirstOutput without checking if randr has 3313 been enabled, and it would crash in privates code. 3314 3315 reported by vereteran on #radeon 3316 3317 Signed-off-by: Dave Airlie <airlied@redhat.com> 3318 Acked-on-irc-by: Michel Dänzer <michel.daenzer@amd.com> 3319 3320commit d669c34f140c000f88c4b4e464e44e6c8694f581 3321Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 3322Date: Mon Nov 21 11:35:40 2011 +1100 3323 3324 ddx/evergreen: Fix endian of ALU constants 3325 3326 The constants are written directly into a buffer object shared with the 3327 card and we "forget" to swap them. This patch fixes it by doing the swap 3328 in evergreen_set_alu_consts() in-place (ie, it modifies the buffer), 3329 which should be fine with the way we use it in the ddx. 3330 3331 This makes everything work fine on my caicos card on a G5 including some 3332 quik tests with Xv, gnome3 shell, etc... 3333 3334 Thanks a lot to Jerome Glisse for holding my hand through debugging that 3335 (and finding the actual bug). 3336 3337 Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> 3338 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3339 3340commit 422bdd4fe6cb728e1dd08a56f6ee2d0f009cbfcb 3341Author: Alex Deucher <alexdeucher@gmail.com> 3342Date: Mon Nov 14 09:39:16 2011 -0500 3343 3344 radeon: add missing FireMV pci id 3345 3346 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3347 3348commit 534fb6e413a909a9d1afd57d1c711844b7c5ddf4 3349Author: Dave Airlie <airlied@redhat.com> 3350Date: Fri Nov 11 10:26:51 2011 +0000 3351 3352 ati: enable bg none when fbcon succeeds and we are built against ABI after 10. 3353 3354 One less patch to keep carrying in Fedora. 3355 3356 Signed-off-by: Dave Airlie <airlied@redhat.com> 3357 3358commit 89452c08048c98fb5cc3dc551b3824be40d52cf2 3359Author: Michel Dänzer <michel.daenzer@amd.com> 3360Date: Tue Nov 8 11:23:11 2011 +0100 3361 3362 UMS: Guard references to PCITAG / pciTag with XSERVER_LIBPCIACCESS (bug #42690) 3363 3364 Should fix https://bugs.freedesktop.org/show_bug.cgi?id=42690 . 3365 3366 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3367 3368commit 5ec34ed95948f7164184551615c1fc4c3eef3b98 3369Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 3370Date: Thu Nov 3 20:16:47 2011 -0400 3371 3372 DRI/DRI2: remove hard-coded limitation to 6 crtcs 3373 3374 DRM's hard limit to the number of CRTCs is 32. ATI DDX unnecessarily 3375 clips this limit to 6 by hard coding initial assumption for 3376 output->possible_crtcs mask to 0x7f (before it gets trimmed down to 3377 what's really possible for a given output) and by allocating only 6 3378 entries for for cursor_bo[] array in RADEONInfoRec. 3379 3380 Fix this and thus allow the ATI DDX to deal with as many CRTCs 3381 as the DRM allows (32), so it is ready if anything with >6 CRTCs 3382 comes out. 3383 3384 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3385 3386commit 4853ab2cdc3b97948c7cd69eaf4fff54f59774fc 3387Author: Michel Dänzer <michel.daenzer@amd.com> 3388Date: Fri Nov 4 12:15:53 2011 +0100 3389 3390 Turn compile time check into runtime check. 3391 3392 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3393 3394commit bcdb54fe16ebf2e239b84eebf20e8adfe5094bff 3395Author: Alex Deucher <alexdeucher@gmail.com> 3396Date: Thu Oct 20 11:11:35 2011 -0400 3397 3398 check for xserver 1.9.4.901 to enable tiling by default 3399 3400 Previous xservers had a bug in the EXA code which caused 3401 display corruption in some cases. 3402 3403 See: 3404 https://bugs.freedesktop.org/show_bug.cgi?id=33929 3405 3406 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3407 3408commit edde94cba5321e6e51e7fd4d79dde5abc4944495 3409Author: Michel Dänzer <michel.daenzer@amd.com> 3410Date: Wed Nov 2 13:40:05 2011 +0100 3411 3412 Bump version post release. 3413 3414commit 93459f842c2d8dc178a1954b8e05150fcb96ac9a 3415Author: Michel Dänzer <michel.daenzer@amd.com> 3416Date: Wed Nov 2 12:51:15 2011 +0100 3417 3418 Bump version for 6.14.3 release. 3419 3420commit fe3ce559b7f07d39ded39abb38576846d6eb185b 3421Author: Jeremy Huddleston <jeremyhu@apple.com> 3422Date: Tue Nov 1 20:27:03 2011 -0700 3423 3424 Build fix for -Werror=int-to-pointer-cast 3425 3426 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 3427 3428commit d3d6a5da07aaec5961e51c9a8f90c1490ee101b6 3429Author: Tormod Volden <debian.tormod@gmail.com> 3430Date: Sat Oct 29 23:08:46 2011 +0200 3431 3432 radeon: do not include xf86PciInfo.h 3433 3434 We already use atipciids.h instead most places. 3435 3436 Signed-off-by: Tormod Volden <debian.tormod@gmail.com> 3437 3438commit 70da7001e81363ed6ef2c4727c512daf53ae29fe 3439Author: Jeremy Huddleston <jeremyhu@apple.com> 3440Date: Sat Oct 29 20:15:09 2011 -0700 3441 3442 Use malloc/calloc/realloc/free directly 3443 3444 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 3445 3446commit a3bb07efb1757c33d70e2e1928219d12a4dd6498 3447Author: Michel Dänzer <michel.daenzer@amd.com> 3448Date: Tue Oct 25 17:43:58 2011 +0200 3449 3450 EXA >= R6xx / KMS: Avoid running out of CS space at inconvenient times. 3451 3452 Otherwise we may end up with things not properly set up at the beginning of the 3453 next CS. 3454 3455 Fixes http://bugs.debian.org/645007 . 3456 3457 In contrast to the Composite code for < R6xx, this isn't necessary with UMS, 3458 as the draw packet only uses constant space in the indirect buffer, and nothing 3459 else can mess with the 3D state between indirect buffers. 3460 3461 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3462 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3463 3464commit 23788c4a5e3b6affb9b183e1393edd0e5ca4550e 3465Author: Michel Dänzer <michel.daenzer@amd.com> 3466Date: Thu Oct 27 19:12:22 2011 +0200 3467 3468 EXA < R6xx: Make sure 2D state is re-emitted after running out of CS space. 3469 3470 Otherwise it's basically luck what the 2D state ends up being at the beginning 3471 of the next CS. 3472 3473 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3474 3475commit 6e0e1a821accc6ca95f4134e49b66a6b168c1934 3476Author: Michel Dänzer <michel.daenzer@amd.com> 3477Date: Tue Oct 25 17:39:57 2011 +0200 3478 3479 Make radeon_dri2_create_buffer(s) more robust. (Bug #30047) 3480 3481 In particular, handle and propagate failure to allocate GPU accessible memory, 3482 instead of crashing. Fixes https://bugs.freedesktop.org/show_bug.cgi?id=30047 . 3483 3484 Also take care not to leak resources in error paths. 3485 3486 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3487 3488commit 856583dbca9319c77fed40daa9956e81a0068f9e 3489Author: Jerome Glisse <jglisse@redhat.com> 3490Date: Thu Oct 20 14:17:14 2011 -0400 3491 3492 radeon/kms: fallback to vesa if GPU is not supported by UMS 3493 3494 For GPU not supported by UMS, test in probe so that we properly 3495 fallback to vesa. 3496 3497 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 3498 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3499 3500commit f6f1b4f7789cfef763c063e671b76b60be8bdd8e 3501Author: Maarten Lankhorst <m.b.lankhorst@gmail.com> 3502Date: Tue Oct 18 14:30:39 2011 +0200 3503 3504 dri2: Add vdpau driver name entry 3505 3506 libvdpau has a driver loading mechanism that looks for a dri2 driver 3507 first before falling back to nvidia, so lets use that. 3508 3509 Allows use of libvdpau_rx00 without having to set things up separately, 3510 similar to the patch to xf86-video-nouveau. 3511 3512 Signed-off-by: Maarten Lankhorst <m.b.lankhorst@gmail.com> 3513 Reviewed-by: Christian König <deathsimple@vodafone.de> 3514 Reviewed-by: Michel Dänzer <michel@daenzer.net> 3515 Tested-by: Michel Dänzer <michel@daenzer.net> 3516 3517commit a6b2bd2d184f10d4c56c4ee17186aedb238a36ec 3518Author: Alex Deucher <alexdeucher@gmail.com> 3519Date: Wed Sep 21 17:00:16 2011 -0400 3520 3521 UMS: fix DDIA enable on some rs690 systems 3522 3523 DVOOutputControl checks the value of of bios scratch reg 3 3524 on some tables and assumes the encoder is already enabled 3525 if the DFP2_ACTIVE bit is set. Clear that bit so the table 3526 sets the DDIA enable bit properly. 3527 3528 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3529 3530commit d78860ba53d9bfcf6c28e1cfd2d970709b5e20fa 3531Author: Michel Dänzer <michel.daenzer@amd.com> 3532Date: Fri Aug 12 11:21:33 2011 +0200 3533 3534 Only call radeon_dri2_close_screen() if DRI2 was enabled. 3535 3536 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3537 3538commit 11330ca5dc61a70fe4507e63230f9133ca22d891 3539Author: Michel Dänzer <michel.daenzer@amd.com> 3540Date: Fri Aug 12 11:21:32 2011 +0200 3541 3542 Remove dead variable remain_size_bytes. 3543 3544 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3545 3546commit fcf0cca9c0ab0f692b222f619aee8f1cdad3b519 3547Author: Michel Dänzer <michel.daenzer@amd.com> 3548Date: Tue Sep 20 12:34:05 2011 +0200 3549 3550 KMS Color Tiling requires xserver which supports EXA_MIXED_PIXMAPS. 3551 3552 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3553 3554commit c96e6fb8a5f5be2319fdb4c431c1ba5279997fe8 3555Author: Alex Deucher <alexdeucher@gmail.com> 3556Date: Sat Sep 17 08:26:12 2011 -0400 3557 3558 man: note that the list of marketing names is non-exhaustive 3559 3560 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3561 3562commit 1a51fce6f6ab169c882a86b936909c0820f27a68 3563Author: Alex Deucher <alexdeucher@gmail.com> 3564Date: Thu Sep 15 17:52:25 2011 -0400 3565 3566 update man page with new marking names 3567 3568 Note that the driver support all currently 3569 shipping asics and the names in the man page 3570 are just a sampling. 3571 3572 Fixes: 3573 https://bugs.freedesktop.org/show_bug.cgi?id=40808 3574 3575 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3576 3577commit 64f237a4cf3ef5bcb3163e88e1447ff275a1eefa 3578Author: Michel Dänzer <michel.daenzer@amd.com> 3579Date: Thu Aug 18 19:11:08 2011 +0200 3580 3581 Convert register ranges for >= r6xx from enums to defines. 3582 3583 Avoids lots of "comparison between 'enum <anonymous>' and 'enum <anonymous>'" 3584 warnings with newer versions of gcc. See 3585 https://bugs.freedesktop.org/show_bug.cgi?id=38238 . 3586 3587 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3588 3589commit 95991fcce45f0dd904f76b9f98f9c7ed5708e4fa 3590Author: Michel Dänzer <michel.daenzer@amd.com> 3591Date: Wed Aug 17 11:10:34 2011 +0200 3592 3593 Bail if we're trying to start up in UMS mode on KMS. 3594 3595 Ideally, the display manager will start the X server again, and everything 3596 will be fine and dandy. But in the worst case, at least we won't hit the 3597 hardware behind the KMS driver's back. 3598 3599 (This change intentionally makes (ab)use of the fact that Bool is defined as 3600 int). 3601 3602 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3603 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3604 3605commit f95a41b7851565c282d22f8d679db1377428f165 3606Author: Michel Dänzer <michel.daenzer@amd.com> 3607Date: Thu Aug 11 11:22:57 2011 +0200 3608 3609 video: Don't round up bottom/right edge for clipping source width/height. 3610 3611 It's not necessary: If the top/left edge was rounded down, this will be 3612 compensated by the subtraction. 3613 3614 Worse, if the original source width/height is odd, rounding up may result in 3615 reading past the end of the source data. 3616 3617 Fixes http://bugs.debian.org/637258 . 3618 3619 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3620 Reviewed-by: Alex Deucher <alexander.deucher@amd.com> 3621 3622commit 93fc0843a1e31dc9237433bc2bf17df79e956d26 3623Author: Michel Dänzer <michel.daenzer@amd.com> 3624Date: Wed Aug 10 17:44:37 2011 +0200 3625 3626 Change my e-mail address to something that still works, and always will, I hope. 3627 3628commit 9151f3b1c2ebcc34e63195888ba696f2183ba5e2 3629Author: Michel Dänzer <michel.daenzer@amd.com> 3630Date: Tue Aug 9 19:13:26 2011 +0200 3631 3632 Prefer the CRTC of the primary output for synchronization. 3633 3634 See https://bugs.freedesktop.org/show_bug.cgi?id=39696 . 3635 3636 Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> 3637 3638commit 3b9fdc807dd7e52af0576299cefba596040f6f2f 3639Author: Alex Deucher <alexdeucher@gmail.com> 3640Date: Wed Aug 3 16:20:13 2011 -0400 3641 3642 r5xx+: Fix vline setup with crtc offsets 3643 3644 On r5xx+, vline is relative to to the viewport, not 3645 the scanlines. Based on initial patch and investigation 3646 from Herbert Pötzl (Bertl) on IRC. 3647 3648 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3649 3650commit 104b2d7c071f29266b1bc4184a74e9714d14febc 3651Author: Alex Deucher <alexander.deucher@amd.com> 3652Date: Mon Aug 1 10:05:30 2011 -0400 3653 3654 kms: fix possible leak in pageflip code 3655 3656 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3657 3658commit d29bab632e9ecccba518d4107d52620bf75eb1cf 3659Author: Ville Syrjala <syrjala@sci.fi> 3660Date: Wed May 4 23:51:27 2011 +0300 3661 3662 kms: Move flip_count and co. to a per swap structure 3663 3664 If multiple drawables are doing page flipping, the global drmmode 3665 structure can't be used to keep per swap information. For example 3666 flip_count can increase prematurely due to another swap request, 3667 and then the previous swap request never gets completed, leading to a 3668 stuck client. Move the relevant pieces of data to a strucuture that 3669 gets allocated once per swap request and shared by all involved CRTCs. 3670 3671 Signed-off-by: Ville Syrjala <syrjala@sci.fi> 3672 3673commit 9493563c1ef4b51af0ee8a44cb4e7c5bb280347e 3674Author: Ville Syrjala <syrjala@sci.fi> 3675Date: Wed May 4 23:51:26 2011 +0300 3676 3677 dri2: Update front buffer pixmap and name before exchanging buffers 3678 3679 Buffer exchange assumes that the front buffer pixmap and name 3680 information is accurate. That may not be the case eg. if the window 3681 has been (un)redirected since the buffer was created. 3682 3683 Signed-off-by: Ville Syrjala <syrjala@sci.fi> 3684 3685commit 8c9266ed2da22a510243f9a952c14d4423f48a2b 3686Author: Alex Deucher <alexander.deucher@amd.com> 3687Date: Fri Jul 15 10:44:57 2011 -0400 3688 3689 radeon: add some new NI pci ids 3690 3691 Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 3692 3693commit f59c3b294b0f715fc96e2bbe25893f2b31aa488b 3694Author: Christian König <deathsimple@vodafone.de> 3695Date: Thu Jul 14 11:49:06 2011 +0200 3696 3697 Register XvMC video decoding acceleration 3698 3699commit e8d0d437957b15252dfad775796a3949ed50dbcf 3700Author: Dave Airlie <airlied@redhat.com> 3701Date: Tue Jul 12 11:43:25 2011 -0400 3702 3703 evergreen: Emit SQ_LDS_RESOURCE_MGMT 3704 3705 Avoids rendering problems when compute changes this reg. 3706 3707 Fixes: 3708 https://bugs.freedesktop.org/show_bug.cgi?id=39119 3709 3710 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3711 3712commit 9bb31158466e6168116d841d12c8b4303f11c4a6 3713Author: Alex Deucher <alexdeucher@gmail.com> 3714Date: Wed Jun 22 12:24:28 2011 -0400 3715 3716 evergreen: fix num_banks for 2D tiling config 3717 3718 The field is encoded. 3719 3720 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3721 3722commit 122b471f734aa07427b01d4bec35ff1ac28290b5 3723Author: Alex Deucher <alexdeucher@gmail.com> 3724Date: Fri Jun 17 04:02:16 2011 -0400 3725 3726 dri2: fix copy pasto in a6154c00c64932332e8f6e334661ffd579cfd894 3727 3728 Reported-by: Nils Wallménius <nils.wallmenius@gmail.com> 3729 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3730 3731commit 122bedcbcf45cb583cf51b2fd04ed2805e0ca60b 3732Author: Alex Deucher <alexdeucher@gmail.com> 3733Date: Thu Jun 16 12:57:11 2011 -0400 3734 3735 dri2: missing bit from a6154c00c64932332e8f6e334661ffd579cfd894 3736 3737 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3738 3739commit a6154c00c64932332e8f6e334661ffd579cfd894 3740Author: Alex Deucher <alexdeucher@gmail.com> 3741Date: Thu Jun 16 12:48:43 2011 -0400 3742 3743 dri2/eg+: fix size and alignment of depth/stencil buffers 3744 3745 Base alignment may be 256B or 512B depending on the group 3746 size. Also need to check against front size for virtualX. 3747 3748 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3749 3750commit 41eb1fbb3d9da64feb4a96df7c575e44136b1538 3751Author: Alex Deucher <alexdeucher@gmail.com> 3752Date: Mon Jun 13 12:50:39 2011 -0400 3753 3754 kms/man: update ColorTiling info 3755 3756 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3757 3758commit 481baa5c74271cd7ce38bae3965d2bc4b8809058 3759Author: Alex Deucher <alexdeucher@gmail.com> 3760Date: Mon Jun 13 12:44:07 2011 -0400 3761 3762 kms: enable ColorTiling by default on r6xx-cayman asics 3763 3764 Currently only 1D tiling as 2D tiling still has some corner 3765 cases to fix up. 3766 3767 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3768 3769commit cbcc57b0fa6f581be777bef648f2bf3efe7443ee 3770Author: Alex Deucher <alexdeucher@gmail.com> 3771Date: Mon Apr 4 12:52:00 2011 -0400 3772 3773 radeon: add llano pci ids 3774 3775 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3776 3777commit 36afd1e1055eeadb2396dadcc68b214655bd90a9 3778Author: Alex Deucher <alexdeucher@gmail.com> 3779Date: Tue May 31 16:03:36 2011 -0400 3780 3781 radeon: add support for llano APUs 3782 3783 - KMS only 3784 - Includes full EXA/Xv support 3785 3786 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3787 3788commit 557f46dc2f18734ecf1f18dee7e951e0bf062e63 3789Author: Dave Airlie <airlied@redhat.com> 3790Date: Fri May 27 07:22:08 2011 +1000 3791 3792 bump version after release 3793 3794commit 2fca40ea65d9f2a6f8451c324bb4b82786f34f76 3795Author: Dave Airlie <airlied@redhat.com> 3796Date: Thu May 26 12:52:21 2011 +1000 3797 3798 radeon: bump version for release of 6.14.2 3799 3800commit fab868c5f7d8cafdb0176d2751f216819a5ba66a 3801Author: Alex Deucher <alexdeucher@gmail.com> 3802Date: Wed May 25 01:09:12 2011 -0400 3803 3804 EXA: make evergreen_fix_scissor_coordinates static 3805 3806 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3807 3808commit fe5c42f5155361006b687da824181418f688809f 3809Author: Alex Deucher <alexdeucher@gmail.com> 3810Date: Tue May 24 22:32:01 2011 -0400 3811 3812 cayman: endian fixes for shaders 3813 3814 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3815 3816commit 470ecd02347c32e79316046d01a7d5dad0e2fe99 3817Author: Alex Deucher <alexdeucher@gmail.com> 3818Date: Tue May 24 18:31:47 2011 -0400 3819 3820 EXA/Xv: add workarounds for eg/cayman scissors bugs 3821 3822 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3823 3824commit b913e7ba2b60d47a6660699210cc3cf6f5dc52c2 3825Author: Alex Deucher <alexdeucher@gmail.com> 3826Date: Tue May 24 18:02:52 2011 -0400 3827 3828 Revert "cayman: add scissors workaround." 3829 3830 Needs a proper workaround for a hw bug. 3831 3832 This reverts commit b77d374b0d11f48c33cfffdb4157c4ec4b05ea15. 3833 3834commit 82cb33c3f0e1ba802d7a94f3159b3c5c86cd4043 3835Author: Dave Airlie <airlied@redhat.com> 3836Date: Wed May 18 14:49:17 2011 +1000 3837 3838 cayman: enable all accel 3839 3840commit b77d374b0d11f48c33cfffdb4157c4ec4b05ea15 3841Author: Dave Airlie <airlied@redhat.com> 3842Date: Tue May 24 15:53:58 2011 +1000 3843 3844 cayman: add scissors workaround. 3845 3846 wasted a lot of time getting to this. 3847 3848commit ffeab7a7058298e15294a3b2c740c731e36dda1d 3849Author: Alex Deucher <alexdeucher@gmail.com> 3850Date: Mon Apr 18 18:16:51 2011 -0400 3851 3852 cayman: fix dword counts default state 3853 3854 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3855 3856commit 3cbfae361bf5e779d3364f0f31cfd25bd0f59e65 3857Author: Alex Deucher <alexdeucher@gmail.com> 3858Date: Wed Mar 2 20:48:19 2011 -0500 3859 3860 cayman: add spi state to default state 3861 3862 changed in e3145801b80fd4be4cf770128876e86e89bda66f 3863 3864 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3865 3866commit b8ade97c9d0fa5aacb0e3166868bb72e9bc679a6 3867Author: Alex Deucher <alexdeucher@gmail.com> 3868Date: Wed Mar 2 20:44:19 2011 -0500 3869 3870 cayman: first pass at exa/Xv shaders 3871 3872 Main differences with evergreen: 3873 - 4-way rather than 5-way 3874 - END_OF_PROGRAM bit removed from CF istructions, use 3875 CF_INST_END instead. 3876 - MEGA_FETCH* fields removed from VTX commands 3877 - no more VC, all fetches go through the TC 3878 3879 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3880 3881commit 01b646ed800732985c1638b147716641a99082f9 3882Author: Alex Deucher <alexdeucher@gmail.com> 3883Date: Wed Mar 2 20:39:38 2011 -0500 3884 3885 cayman: add a default state function 3886 3887 The rest of the state functions can be shared 3888 with evergreen. I've noted where there are 3889 differences. 3890 3891 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3892 3893commit 42eecc6f4fb1570769490bdaeac06817c6c36a7e 3894Author: Alex Deucher <alexdeucher@gmail.com> 3895Date: Wed Mar 2 20:20:56 2011 -0500 3896 3897 cayman: add 3D register headers 3898 3899 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3900 3901commit e1d28e011f4a5139cbc778973c63158ed2746716 3902Author: Alex Deucher <alexdeucher@gmail.com> 3903Date: Wed Mar 2 20:13:50 2011 -0500 3904 3905 kms/cayman: stub out exa support 3906 3907 Just fallbacks for now. 3908 3909 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3910 3911commit 21e44a20b8b1b64079ee77f45aaa5010206ed7b6 3912Author: Michel Dänzer <daenzer@vmware.com> 3913Date: Thu May 12 09:23:38 2011 +0200 3914 3915 UMS: Fix comparison of unsigned variable against < 0. 3916 3917 Pointed out by clang: 3918 3919 ../../src/radeon_crtc.c:242:18: error: comparison of unsigned expression < 0 is always false [-Werror,-Wtautological-compare] 3920 error = error < 0 ? 0xffffffff : error; 3921 ~~~~~ ^ ~ 3922 3923 If a UMS regression is bisected to this commit, the assignment should probably 3924 just be removed, as it's a no-op in the current form. 3925 3926commit 3b893d81982c9381393c92625e308541e0071b05 3927Author: Michel Dänzer <daenzer@vmware.com> 3928Date: Fri May 13 09:10:02 2011 +0200 3929 3930 KMS: Fix output properties logic error. 3931 3932 Pointed out by clang: 3933 3934 ../../src/drmmode_display.c:1023:30: error: use of logical && with constant operand; switch to bitwise & or remove constant [-Werror,-Wconstant-logical-operand] 3935 if (props && (props->flags && DRM_MODE_PROP_ENUM)) { 3936 ^ ~~~~~~~~~~~~~~~~~~ 3937 3938 Reviewed-by: Alex Deucher <alexdeucher@gmail.com> 3939 3940commit f83d58cf5b33686139067f8f898b8e566ba5c253 3941Author: Nicolas Kaiser <nikai@nikai.net> 3942Date: Fri May 13 00:56:31 2011 +0200 3943 3944 man: fix typos 3945 3946 Signed-off-by: Nicolas Kaiser <nikai@nikai.net> 3947 3948commit 90abffbd30f44b9cf76a6e28103ddcb5419b4522 3949Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 3950Date: Fri May 6 09:45:23 2011 -0400 3951 3952 DRI2: fix high-crtc/vblank oversight/bug 3953 3954 improvements to high-crtc handling done in 3955 f0b7d7b449cc77bb2b281d81108507f8bc2e6018 introduced a bug that caused 3956 the populate_vbl_request_type to never use the high-crtc field even 3957 when it should. The reason is that the offending patch put the code 3958 under #ifdef DRM_VBLANK_HIGH_CRTC_MASK which is not visible outside the 3959 enum type, so #else was always taken in compilation type. This patch 3960 fixes it by basing #ifdef on (pre-processor visible) 3961 DRM_VBLANK_HIGH_CRTC_SHIFT constant 3962 3963 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 3964 3965commit 62a4cd180fe884dca24586d453395472516e6496 3966Author: Alex Deucher <alexdeucher@gmail.com> 3967Date: Wed May 4 01:13:55 2011 -0400 3968 3969 fusion: fix tiling enable logic 3970 3971 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3972 3973commit 76638ca687b02d3b1494b9868f817fd4fd892c64 3974Author: Alex Deucher <alexdeucher@gmail.com> 3975Date: Wed May 4 01:06:22 2011 -0400 3976 3977 fusion: enable tiling if DFS works 3978 3979 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3980 3981commit a6d2dba6573a3512d550d7e442bf42ea03012bbc 3982Author: Dave Airlie <airlied@redhat.com> 3983Date: Wed May 4 10:44:43 2011 +1000 3984 3985 radeon: add add hw DFS support for fusion 3986 3987 Fusion had a bug setting up the VM on earlier kernels so we need to work 3988 around that and only enable accel on a new enough kernel. 3989 3990 Signed-off-by: Dave Airlie <airlied@redhat.com> 3991 3992commit 859e052af49e68a826b77a9135c7f067dc331a06 3993Author: Alex Deucher <alexdeucher@gmail.com> 3994Date: Tue May 3 15:15:04 2011 -0400 3995 3996 radeon: add some new pci ids 3997 3998 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 3999 4000commit 8f8bbf628c6eed037f57bc8c155f0ecdacbebad1 4001Author: Alex Deucher <alexdeucher@gmail.com> 4002Date: Sun May 1 13:19:15 2011 -0400 4003 4004 man: add cayman to man page 4005 4006 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4007 4008commit 903e90c31cf0319be9297529aa7b8daa1756cf63 4009Author: Alex Deucher <alexdeucher@gmail.com> 4010Date: Wed Apr 20 03:10:08 2011 -0400 4011 4012 EXA/Xv: used cached bo tiling flags for accel setup on 6xx+ 4013 4014 This avoids calling into the kernel for each bo in the accel 4015 code. This is a follow on to: 4016 cc7d1fa39da40a532fcdbe6c7924ca47a879e66a 4017 4018 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4019 4020commit 982c22f16c8eeee9be81779fbfe17d8d3f9b6897 4021Author: Adam Jackson <ajax@redhat.com> 4022Date: Thu Apr 14 16:04:50 2011 -0400 4023 4024 R520: Fix textures larger than 2k 4025 4026 Ported from the equivalent fix in Mesa. 4027 4028commit cc7d1fa39da40a532fcdbe6c7924ca47a879e66a 4029Author: Michel Dänzer <daenzer@vmware.com> 4030Date: Mon Apr 4 17:37:12 2011 +0200 4031 4032 EXA: Cache BO tiling flags. 4033 4034 Calling into the kernel every time is quite expensive, and nobody else should 4035 ever change the tiling flags. 4036 4037 There's still more to do along the same lines for >= R6xx. 4038 4039commit f0b7d7b449cc77bb2b281d81108507f8bc2e6018 4040Author: Michel Dänzer <daenzer@vmware.com> 4041Date: Tue Apr 5 13:36:01 2011 +0200 4042 4043 DRI2: Some cleanups for the scheduling mess. 4044 4045 * Fix build against libdrm that doesn't define *_VBLANK_HIGH_CRTC*. 4046 * If we have more than two CRTCs but can't use DRM_VBLANK_HIGH_CRTC_MASK, don't 4047 enable scheduling in the first place rather than relying on 4048 DRM_VBLANK_SECONDARY magically doing something sensible for higher CRTCs. 4049 * Only set up client state tracking when scheduling is enabled. 4050 * Only declare pRADEONEnt when it's needed, and break long lines. 4051 4052commit 7acf9bc833de539fa2259a051c66a99445a54bc4 4053Author: Alex Deucher <alexdeucher@gmail.com> 4054Date: Mon Apr 4 11:08:37 2011 -0400 4055 4056 radeon: add some additional ontario pci ids 4057 4058 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4059 4060commit 0af6386f85a0f1ba14864a3334164733a10a6cb8 4061Author: Ilija Hadzic <ihadzic@research.bell-labs.com> 4062Date: Thu Mar 24 13:33:27 2011 -0400 4063 4064 xf86-video-ati: (revised #2) add support for vblank on crtc > 1 4065 4066 Hi Alex, 4067 4068 Enclosed is a revised version of two patches sent on Mar 18 and Mar 22, 4069 respectively. Details summarized in these two threads: 4070 http://lists.freedesktop.org/archives/dri-devel/2011-March/009463.html 4071 http://lists.freedesktop.org/archives/dri-devel/2011-March/009582.html 4072 4073 This patch reconciles the DDX with the change in libdrm sent to this list 4074 earlier today. Specifically, it refers to a symbol that has been renamed 4075 from DRM_CAP_HIGH_CRTC to DRM_CAP_VBLANK_HIGH_CRTC. It *supersedes* the 4076 previous patch (i.e. apply it to the master branch as it exists at the 4077 time of this writing, not as an incremental patch to the one sent previously). 4078 4079 Regards, 4080 4081 Ilija 4082 4083 Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> 4084 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4085 4086commit fe2e0ad3ffa58f40311319c950b842e2928a5740 4087Author: matthew green <mrg@eterna.com.au> 4088Date: Mon Mar 21 12:17:58 2011 -0400 4089 4090 bug fix for r6xx/r7xx UMS 4091 4092 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4093 4094commit 83978ad5fe37581e2b2f3fbd9c073d91b2ae1d50 4095Author: Alex Deucher <alexdeucher@gmail.com> 4096Date: Thu Mar 17 19:07:43 2011 -0400 4097 4098 bump version post release 4099 4100 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4101 4102commit 38d9368e59b2990bf32a028ece2132451b402350 4103Author: Alex Deucher <alexdeucher@gmail.com> 4104Date: Thu Mar 17 18:59:55 2011 -0400 4105 4106 bump for release 4107 4108 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4109 4110commit cdfc007eccc9b130fc08012deef304488eb6e469 4111Author: Dave Airlie <airlied@redhat.com> 4112Date: Wed Mar 16 10:40:42 2011 +1000 4113 4114 radeon/exa: correct function name 4115 4116 this corrects the function name so it matches the contents. 4117 4118 Signed-off-by: Dave Airlie <airlied@redhat.com> 4119 4120commit b93d7658f669f6dc1cfacebcfe955a1e113a537c 4121Author: Dave Airlie <airlied@redhat.com> 4122Date: Wed Mar 16 10:37:22 2011 +1000 4123 4124 radeon/exa: fix scaling check 4125 4126 This caused a regression in firefox, as these numbers are xFixed 4127 values hence 1 is actually 65536. 4128 4129 Should fix: https://bugzilla.redhat.com/show_bug.cgi?id=685084 4130 4131 Signed-off-by: Dave Airlie <airlied@redhat.com> 4132 4133commit 8567b8cd9e136ef69906e02286b3752db2404741 4134Author: Alex Deucher <alexdeucher@gmail.com> 4135Date: Tue Mar 15 13:13:06 2011 -0400 4136 4137 APU: no tiling yet 4138 4139 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4140 4141commit 4d350497012fa31a417ada662006e2d64db2a4b5 4142Author: Dave Airlie <airlied@redhat.com> 4143Date: Tue Mar 15 10:32:19 2011 +1000 4144 4145 radeon: exa shaders don't handle scaling either. 4146 4147 rendercheck tsrccoords test fails. 4148 4149 Signed-off-by: Dave Airlie <airlied@redhat.com> 4150 4151commit 6319a33cb7bd8dba1eaf47bee1e2608cee5d7600 4152Author: Alex Deucher <alexdeucher@gmail.com> 4153Date: Wed Mar 2 19:38:04 2011 -0500 4154 4155 kms: add cayman pci ids 4156 4157 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4158 4159commit fd65def45a57e3242596274d270791df8abc8e29 4160Author: Alex Deucher <alexdeucher@gmail.com> 4161Date: Thu Jan 13 17:02:31 2011 -0500 4162 4163 kms: no accel yet for cayman 4164 4165 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4166 4167commit 585e2a3b803b6d3e070f2427e9cb49bab0949239 4168Author: Alex Deucher <alexdeucher@gmail.com> 4169Date: Thu Jan 13 16:58:01 2011 -0500 4170 4171 kms: add cayman chip family 4172 4173 KMS only. No UMS support for cayman. 4174 4175 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4176 4177commit 6835e22256ba7a74c02504383f63baaba7e87156 4178Author: Alex Deucher <alexdeucher@gmail.com> 4179Date: Wed Mar 2 12:20:45 2011 -0500 4180 4181 r6xx+: truncate point sampled coordinates 4182 4183 By default the hardware rounds texcoords. However, 4184 for point sampled textures, the expected behavior is 4185 to truncate. When we have point sampled textures, 4186 set the truncate bit in the sampler. 4187 4188 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4189 4190commit 8a8862e4165dc5be779bce5e327a31da4826db3b 4191Author: Sedat Dilek <sedat.dilek@googlemail.com> 4192Date: Sat Feb 26 21:04:22 2011 +0100 4193 4194 kms: Fix warning XNFprintf is deprecated (v2) 4195 4196 The first take of my patch missed the breakage against pre-1.10 4197 xservers and now I know XNFasprintf() is not a 1:1 replacement 4198 for XNFprintf(). 4199 The idea for the fix seemed to be so obvious, but as usual "really" 4200 test the patch you are sending to any ML. 4201 I am awfully sorry for the overhasty bad hack. 4202 Thanks to Alan and Alex for the help to fix it correctly, now. 4203 4204 Tested-By: Sedat Dilek <sedat.dilek@gmail.com> 4205 Signed-off-by: Sedat Dilek <sedat.dilek@gmail.com> 4206 4207commit 88304fe9c0ce69012d44a77a5368c35236511dac 4208Author: Alex Deucher <alexdeucher@gmail.com> 4209Date: Mon Feb 28 11:29:19 2011 -0500 4210 4211 kms: don't rotate if acceleration is not enabled 4212 4213 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4214 4215commit 99141761e947e2153ed1a3bc485dc5716a22acb1 4216Author: Alex Deucher <alexdeucher@gmail.com> 4217Date: Mon Feb 28 11:20:44 2011 -0500 4218 4219 kms: don't enable tiling if accel is off 4220 4221 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4222 4223commit 7928c120fb629a782846bcd5734d9fa14cbb099d 4224Author: Dave Airlie <airlied@redhat.com> 4225Date: Mon Feb 28 14:11:57 2011 +1000 4226 4227 xv: fix height alignments for U/V planes 4228 4229 The kernel CS checks showed we were incorrectly aligning the 4230 U/V allocations for when we used them as texture sources. 4231 4232 This should fix 4233 https://bugs.freedesktop.org/show_bug.cgi?id=34567 4234 4235 Signed-off-by: Dave Airlie <airlied@redhat.com> 4236 4237commit e55520c0e582f2a500b3b01c2cb78750e15256ca 4238Author: Michel Dänzer <daenzer@vmware.com> 4239Date: Sat Feb 26 15:27:59 2011 +0100 4240 4241 Revert "kms: Fix warning XNFprintf is deprecated" 4242 4243 This reverts commit cb4dc7ab66016d0c980800f8dbf71bbe993889d3. 4244 4245 Looks like this caused https://bugs.freedesktop.org/attachment.cgi?id=43852 , 4246 but even if it didn't, the following needs to be addressed first: 4247 4248 cc1: warnings being treated as errors 4249 ../../src/radeon_kms.c: In function ‘radeon_open_drm_master’: 4250 ../../src/radeon_kms.c:413: error: passing argument 1 of ‘XNFasprintf’ from incompatible pointer type 4251 /usr/local/include/xorg/Xprintf.h:57: note: expected ‘char **’ but argument is of type ‘char *’ 4252 ../../src/radeon_kms.c:413: error: passing argument 2 of ‘XNFasprintf’ makes pointer from integer without a cast 4253 /usr/local/include/xorg/Xprintf.h:57: note: expected ‘const char * restrict’ but argument is of type ‘uint16_t’ 4254 ../../src/radeon_kms.c:413: error: assignment makes pointer from integer without a cast 4255 make[2]: *** [radeon_kms.lo] Error 1 4256 4257commit cb4dc7ab66016d0c980800f8dbf71bbe993889d3 4258Author: Sedat Dilek <sedat.dilek@gmail.com> 4259Date: Fri Feb 25 17:21:52 2011 -0500 4260 4261 kms: Fix warning XNFprintf is deprecated 4262 4263 Noticed when building ati/radeon ddx against xserver 1.10-rc3. 4264 4265 [ build.log ] 4266 ../../src/radeon_kms.c:412:5: warning: 'XNFprintf' is deprecated (declared at /usr/include/xorg/os.h:273) 4267 4268 agd5f: don't break the build against pre-1.10 xservers. 4269 4270 Signed-off-by: Sedat Dilek <sedat.dilek@gmail.com> 4271 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4272 4273commit ecfdb209afe2aafc378baab8c511f5df7b000270 4274Author: Sedat Dilek <sedat.dilek@googlemail.com> 4275Date: Fri Feb 25 21:48:14 2011 +0100 4276 4277 UMS: Fix build against xserver 1.10-rc3 4278 4279 This issue was introduced due to last minute backout of RandR-1.4 4280 in xserver 1.10-rc3. 4281 4282 Switch to "#ifdef RANDR_14_INTERFACE" as suggested by Keith Packard. 4283 See also <http://lists.x.org/archives/xorg-devel/2011-February/019643.html>. 4284 4285 Note: 4286 The ddx needs a rebuild as the X video driver ABI changed to version 10.0. 4287 4288 Reported-by: Alex Deucher <alexdeucher@gmail.com> 4289 CC: Keith Packard <keithp@keithp.com> 4290 Signed-off-by: Sedat Dilek <sedat.dilek@gmail.com> 4291 4292commit 04293a3a84acd96ccdf8e5e865cf973325ca43a1 4293Author: Alex Deucher <alexdeucher@gmail.com> 4294Date: Thu Feb 24 02:33:41 2011 -0500 4295 4296 dri2: disable pageflipping for transformed displays 4297 4298 Based on Ben's nouveau patch. 4299 4300 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4301 4302commit 91070cfd75d5607c4a72ace780f830f0ddb40e84 4303Author: Alex Deucher <alexdeucher@gmail.com> 4304Date: Wed Feb 23 00:55:28 2011 -0500 4305 4306 Xv: use aligned height to calculate Xv buffer size 4307 4308 Make sure it's large enough for the tiling mode so the 4309 CS checker doesn't complain. Should fix: 4310 https://bugs.freedesktop.org/show_bug.cgi?id=34567 4311 4312 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4313 4314commit acd54a48a6680956ed903fc890204c0c910b6712 4315Author: Mark Kettenis <kettenis@openbsd.org> 4316Date: Sun Feb 20 22:51:04 2011 +0100 4317 4318 Properly restore CLOCK_CNTL_INDEX register 4319 4320 RADEONRestore() restores the CLOCK_CNTL_INDEX register before calling 4321 RADEONRestorePLLRegisters(), which modifies the CLOCK_CNTL_INDEX register 4322 to access the PLL registers. As a result we may end up with the wrong clock 4323 being selected when exiting X. This happens on platforms where the driver 4324 doesn't save and restore the VGA state. 4325 4326 Signed-off-by: Mark Kettenis <kettenis@openbsd.org> 4327 4328commit d7c2ba4286ea32401c0026cb92ec167f7b5753d8 4329Author: Alex Deucher <alexdeucher@gmail.com> 4330Date: Wed Feb 16 18:30:34 2011 -0500 4331 4332 man: add btc chips to the list 4333 4334 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4335 4336commit d0f8bffbbc26818eccdd305ef142387dc4975bba 4337Author: Alex Deucher <alexdeucher@gmail.com> 4338Date: Wed Feb 16 15:09:20 2011 -0500 4339 4340 kms: add tiling support for evergreen/NI 4341 4342 - requires 2.6.37 or newer 4343 - 1D only at the moment 4344 - disabled by default 4345 4346 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4347 4348commit 0471d8412acd82e281a35fc4c6bb2d53b1ff5802 4349Author: Alex Deucher <alexdeucher@gmail.com> 4350Date: Wed Feb 16 15:07:35 2011 -0500 4351 4352 kms: EXA/Xv tiling fixes 4353 4354 - properly set tiling flags for temp surfaces 4355 - fix CB non_disp_tiling bits on evergreen 4356 4357 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4358 4359commit 4d7e1498f7d9eb50e2eddabca193fc27bde24f0e 4360Author: Alex Deucher <alexdeucher@gmail.com> 4361Date: Tue Feb 15 11:24:23 2011 -0500 4362 4363 fix Xv on pre-r6xx asics 4364 4365 broken by: 4366 a9c1358099:"Xv: fix textured video alignment" 4367 4368 Reported by: 4369 Artur Skawina <art.08.09@gmail.com> 4370 4371 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4372 4373commit a9a59717d11af37a2dda5555f6a83c5b65449527 4374Author: Alex Deucher <alexdeucher@gmail.com> 4375Date: Sat Feb 12 05:34:50 2011 -0500 4376 4377 kms: use worst case base/pitch align if we don't have drm tiling info 4378 4379 To avoid CS rejection. 4380 4381 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4382 4383commit 5f9b7fdb4ce27f8c6a64134bb4d5ebeb1c958cca 4384Author: Alex Deucher <alexdeucher@gmail.com> 4385Date: Sat Feb 12 05:19:12 2011 -0500 4386 4387 kms: always get the tiling info even if we aren't tiling 4388 4389 We need the proper group size to calculate alignment so the 4390 ddx and drm agree. 4391 4392 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4393 4394commit 355f0f7a67ce239b7fd2f2e81bcaaa88b2a09f03 4395Author: Alex Deucher <alexdeucher@gmail.com> 4396Date: Sat Feb 12 04:52:52 2011 -0500 4397 4398 kms: fix rotate pitch align 4399 4400 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4401 4402commit f1dc419c989addc4737aed06ec8b8acdb4d40063 4403Author: Alex Deucher <alexdeucher@gmail.com> 4404Date: Fri Feb 11 17:21:10 2011 -0500 4405 4406 kms: evergreen/ni big endian accel support 4407 4408 Based on 6xx/7xx patches from Cédric Cano. 4409 4410 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4411 4412commit e8dc728a549323f1babe337b9d42ad504af1ca39 4413Author: Cédric Cano <ccano@interfaceconcept.com> 4414Date: Fri Feb 11 17:06:47 2011 -0500 4415 4416 ums: atombios endian fixes 4417 4418 agd5f: fix a few more cases. 4419 4420 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4421 4422commit 48ffad957f1dbca909515ffa00629f4caa68706b 4423Author: Cédric Cano <ccano@interfaceconcept.com> 4424Date: Fri Feb 11 17:00:31 2011 -0500 4425 4426 kms: 6xx/7xx big endian accel support 4427 4428 agd5f: minor cleanups 4429 4430 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4431 4432commit 151b22bd7c3b1002a7261538611fb2b468815c86 4433Author: Cédric Cano <ccano@interfaceconcept.com> 4434Date: Fri Feb 11 16:33:16 2011 -0500 4435 4436 kms/6xx+: endian swap cursor uploads 4437 4438 agd5f: remove kms surface for cursor since 4439 we now do the swap in the cursor load to avoid 4440 breaking pre-6xx cards with BE. 4441 4442 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4443 4444commit 309fac041f998e20338d320aabaeda5cc1bb6348 4445Author: Alex Deucher <alexdeucher@gmail.com> 4446Date: Fri Feb 11 14:32:45 2011 -0500 4447 4448 kms/exa/xv: fix alignment checking in accel state setup 4449 4450 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4451 4452commit 25fe3ec4c6dfcdf6ba72a032bc9cb2c84d84029a 4453Author: Alex Deucher <alexdeucher@gmail.com> 4454Date: Fri Feb 11 14:03:56 2011 -0500 4455 4456 kms/exa: UTS/DFS base alignment fixes 4457 4458 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4459 4460commit 3d10278ce511f5dabb68ed86ee43eaaf43585983 4461Author: Alex Deucher <alexdeucher@gmail.com> 4462Date: Fri Feb 11 13:51:11 2011 -0500 4463 4464 Xv: fix textured video alignment harder 4465 4466 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4467 4468commit a9c135809945fb372ce390f06918b9fd8e455b3f 4469Author: Alex Deucher <alexdeucher@gmail.com> 4470Date: Fri Feb 11 13:21:43 2011 -0500 4471 4472 Xv: fix textured video alignment 4473 4474 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4475 4476commit e5bd99faa3b6629a55168386d5dfa936ee4e97ae 4477Author: Alex Deucher <alexdeucher@gmail.com> 4478Date: Fri Feb 11 12:14:55 2011 -0500 4479 4480 kms: remove RADEON_TILING_SURFACE flag on front bo 4481 4482 Should have been removed in: 4483 ef9bfb262db7004bef3704e5d914687e50d3fca4 4484 4485 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4486 4487commit 92ffce1892d24a98d789aad57a4230cadb0c812a 4488Author: Alex Deucher <alexdeucher@gmail.com> 4489Date: Fri Feb 11 03:12:43 2011 -0500 4490 4491 EXA/r6xx+: properly account for height alignment in copy temp buffer 4492 4493 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4494 4495commit 39104c6e8461cf49c1bb03a18858ad75a9d98b46 4496Author: Alex Deucher <alexdeucher@gmail.com> 4497Date: Thu Feb 10 14:31:43 2011 -0500 4498 4499 remove EVERGREENSetAccelState() 4500 4501 It was a duplicate of the R600 variant. 4502 4503 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4504 4505commit 7ac3a2e0bcdadff7c7172a9f833f526b526da16b 4506Author: Alex Deucher <alexdeucher@gmail.com> 4507Date: Thu Feb 10 14:24:50 2011 -0500 4508 4509 6xx+: switch to linear aligned rather than linear general 4510 4511 linear aligned is supposedly more performant, but more 4512 importantly, linear general only works on the CB without 4513 slices. The texture blocks technically don't support 4514 linear general although, I think linear general gets 4515 upgraded to linear aligned in the hw which is why it 4516 currently works. 4517 4518 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4519 4520commit e3145801b80fd4be4cf770128876e86e89bda66f 4521Author: Alex Deucher <alexdeucher@gmail.com> 4522Date: Thu Feb 10 14:14:55 2011 -0500 4523 4524 evergreen/NI: consolidate spi setup 4525 4526 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4527 4528commit be67ded05621aff9c85525372fd119071d3278ec 4529Author: Alex Deucher <alexdeucher@gmail.com> 4530Date: Thu Feb 10 14:06:38 2011 -0500 4531 4532 6xx/7xx: consolidate spi setup 4533 4534 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4535 4536commit 2c5ae1724307e0dba5d0306fe27c1e15a7390a2f 4537Author: Alex Deucher <alexdeucher@gmail.com> 4538Date: Thu Feb 10 13:47:53 2011 -0500 4539 4540 evergreen/ni: consolidate CB state handling 4541 4542 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4543 4544commit c52d817b51e13447802fe338be2d247ffba2a669 4545Author: Alex Deucher <alexdeucher@gmail.com> 4546Date: Thu Feb 10 13:35:02 2011 -0500 4547 4548 6xx/7xx: consolidate remaining CB state 4549 4550 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4551 4552commit 5f9bc127d302404432b631e3e774192950f57b42 4553Author: Alex Deucher <alexdeucher@gmail.com> 4554Date: Thu Feb 10 03:28:01 2011 -0500 4555 4556 kms/r6xx+: clean up pitch/height alignment in EXA UTS/DFS 4557 4558commit a4899db96029acde6cd400fc0541693a487898e7 4559Author: Alex Deucher <alexdeucher@gmail.com> 4560Date: Sat Feb 5 14:11:42 2011 -0500 4561 4562 UMS: fix spelling in error message 4563 4564 Noticed by: Warren Block <wblock@wonkity.com> 4565 4566 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4567 4568commit 84450ea9221f32296c1f786a6aa126a08bb9756f 4569Author: Alex Deucher <alexdeucher@gmail.com> 4570Date: Sat Feb 5 14:05:48 2011 -0500 4571 4572 UMS/DCE3.2: fix segfault 4573 4574 atombios_pick_dig_encoder() needs to be called 4575 with a valid crtc assigned. 4576 4577 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4578 4579commit a27b5dbd9fba689f1814c0888d1b7459b6c6b1a1 4580Author: Alex Deucher <alexdeucher@gmail.com> 4581Date: Thu Feb 3 19:06:22 2011 -0500 4582 4583 Bump version post release 4584 4585 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4586 4587commit 0a1a0513a61f392580bde39cca4880f2c19abc8d 4588Author: Alex Deucher <alexdeucher@gmail.com> 4589Date: Thu Feb 3 18:09:17 2011 -0500 4590 4591 bump version for release 4592 4593 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4594 4595commit ef9bfb262db7004bef3704e5d914687e50d3fca4 4596Author: Alex Deucher <alexdeucher@gmail.com> 4597Date: Tue Feb 1 19:24:44 2011 -0500 4598 4599 kms/pre-6xx: fix pageflipping with tiling 4600 4601 Use UTS/DFS to tile/untile as appropriate for sw access on 4602 pre-6xx. 4603 4604 Fixes: 4605 https://bugs.freedesktop.org/show_bug.cgi?id=33738 4606 4607 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4608 4609commit bb16dd77321e5a64a3cb2d2ca9982117799ac1a8 4610Author: Alex Deucher <alexdeucher@gmail.com> 4611Date: Tue Feb 1 19:20:00 2011 -0500 4612 4613 kms/r6xx+: fix tiling and pageflipping harder 4614 4615 Thanks for Michel for final fix. Fixes: 4616 https://bugs.freedesktop.org/show_bug.cgi?id=33738 4617 for r6xx+ asics. A similar approach for pre-r6xx 4618 asics is pending. 4619 4620 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4621 4622commit 9c124f781049ef11a8b19894a29b7f62975b3011 4623Author: Alex Deucher <alexdeucher@gmail.com> 4624Date: Tue Feb 1 18:59:40 2011 -0500 4625 4626 radeon: remove device 0x4243 4627 4628 0x4243 is a PCI bridge, not a gpu. 4629 4630 See: 4631 https://bugs.freedesktop.org/show_bug.cgi?id=33815 4632 4633 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4634 4635commit 66eb81b62e5ae8e1d7bd44ed8a179e5ec1ca69af 4636Author: Michel Dänzer <daenzer@vmware.com> 4637Date: Mon Jan 31 12:26:20 2011 +0100 4638 4639 UMS: Slightly improve xserver version check. 4640 4641 This will ensure the xserver definitions necessary to compile the new code are 4642 there. xf86CrtcSetMode will be unresolved at runtime with a few xserver 4643 commits, but I'm afraid this is the best we can do. 4644 4645 See https://bugs.freedesktop.org/show_bug.cgi?id=33719 . 4646 4647commit 3dc28c86eb57cac819e4ed650acfe1f7df6ef565 4648Author: Alex Deucher <alexdeucher@gmail.com> 4649Date: Thu Jan 27 13:09:26 2011 -0500 4650 4651 ums: fix build against xserver >= 1.10 4652 4653 xf86CrtcSetMode was removed in: 4654 http://cgit.freedesktop.org/xorg/xserver/commit/?id=8b35118c03590a7ad3786d3284bafb3f40fcb8cc 4655 4656commit f58e1354b78bf6b70120bddfe1566da3b0723f72 4657Author: Jerome Glisse <jglisse@redhat.com> 4658Date: Wed Jan 26 16:13:30 2011 -0500 4659 4660 dp: fix displayport support by syncing with KMS code 4661 4662 Warning the dp clock value are divided by 10 in ddx (10 times 4663 bigger than kernel value) this is somethings very picky. 4664 4665 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 4666 4667commit 0f90133ac61c5e1d80b6fa9256e079349940029f 4668Author: Alex Deucher <alexdeucher@gmail.com> 4669Date: Mon Jan 24 12:40:22 2011 -0500 4670 4671 vbo: warn if we add a bo with a pointer to the wait list 4672 4673commit 42529603ecf86fcfd0a8f3495d9db9ce1ee9b4c3 4674Author: Alex Deucher <alexdeucher@gmail.com> 4675Date: Mon Jan 24 12:30:56 2011 -0500 4676 4677 vbo: always flush the cbuf bo in flush_indirect() 4678 4679 Always flush the cbuf in case we end up with a cbuf mapped in 4680 Prepare*(), but never end up issuing a draw call since the cbuf 4681 may be in use by multiple ops. 4682 4683 The CS check for the cbuf bo is no longer necessay because 4684 the cbuf bo is NULL at this point due to the radeon_vbo_put() 4685 just prior. 4686 4687 Fixes: 4688 https://bugs.freedesktop.org/show_bug.cgi?id=32188 4689 4690commit 9b0e03e98ab739efb4031f81cc4a1a50b3d87a42 4691Author: Alex Deucher <alexdeucher@gmail.com> 4692Date: Mon Jan 24 00:13:25 2011 -0500 4693 4694 vbo: avoid extra cs revalidation in vbo alloc 4695 4696 Make sure we have a valid bo before revalidating. 4697 4698commit fadee0409a8e13b78bbccb83dd70f590fee23d57 4699Author: Alex Deucher <alexdeucher@gmail.com> 4700Date: Fri Jan 21 17:30:28 2011 -0500 4701 4702 vbo: balance radeon_vbo_get() and radeon_vbo_put() 4703 4704 bo_ref() and bo_map() in radeon_vbo_get() 4705 bo_unmap() and bo_unref() in radeon_vbo_put() 4706 4707 rather than doing the bo_map() separately in 4708 radeon_vbo_space(). 4709 4710commit af7d81625a8cf873e6efc881489b3eda9861bd03 4711Author: Alex Deucher <alexdeucher@gmail.com> 4712Date: Fri Jan 21 17:01:25 2011 -0500 4713 4714 evergreen: use vb_offset rather than vb_start_op for cbuf offset 4715 4716commit 4817fac5f728b777939e2e2bee16b842c9e1367b 4717Author: Alex Deucher <alexdeucher@gmail.com> 4718Date: Fri Jan 21 15:59:32 2011 -0500 4719 4720 vbo: remove useless radeon_vb_discard() 4721 4722 We already reset vb_start_op to -1 in the 4723 UMS/KMS ib discard functions. 4724 4725commit 0a03f03a65aad925ba2d9c76b1d3356184607bf9 4726Author: Alex Deucher <alexdeucher@gmail.com> 4727Date: Thu Jan 20 17:07:32 2011 -0500 4728 4729 evergreen/ni: fix Xv artifacts 4730 4731 Port of the 6xx/7xx fix to evergreen. Bad texture size 4732 for texture cache flush. 4733 4734commit f41cf83750ba9a2e0797fdb21ea9104b85ce53aa 4735Author: Marton Balint <cus@fazekas.hu> 4736Date: Thu Jan 20 17:04:56 2011 -0500 4737 4738 r6xx/7xx: fix Xv artifacts 4739 4740 bad textures size for cache flushes. 4741 4742 Fixes: 4743 https://bugs.freedesktop.org/show_bug.cgi?id=22007 4744 4745commit 57fbddfc21d8c6794f378489b764cc2a0ad4a48c 4746Author: Michel Dänzer <daenzer@vmware.com> 4747Date: Tue Jan 18 16:07:52 2011 +0100 4748 4749 Fix crash freeing KMS video memory. 4750 4751 Where's that brown paper bag? :} 4752 4753commit 6548bb9836253c586023ffe5ad1497ddabaa50fc 4754Author: Michel Dänzer <daenzer@vmware.com> 4755Date: Tue Jan 18 10:23:41 2011 +0100 4756 4757 Fix KMS textured video leaks (bug #33193). 4758 4759 v2: Fix radeon_legacy_free_memory() argument type error pointed out by 4760 Marton Balint, refactor video memory freeing logic into helper function. 4761 4762 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=33193 . 4763 4764commit edc3496b55577ee8509ddd9188e6f2bcdf7169a1 4765Author: Dave Airlie <airlied@redhat.com> 4766Date: Mon Jan 17 07:55:29 2011 +1000 4767 4768 radeon: add back flush in block handler. 4769 4770 this is also needed for certain things like stipple rendering. 4771 4772 Signed-off-by: Dave Airlie <airlied@redhat.com> 4773 4774commit 63d7dece3f4be8b14012b3a2bedd850831c0437b 4775Author: Alex Deucher <alexdeucher@gmail.com> 4776Date: Thu Jan 13 11:24:26 2011 -0500 4777 4778 rs880 fix typo in HD 4250 string 4779 4780 Noticed by Nigel Taylor 4781 4782 Fixes: 4783 https://bugs.freedesktop.org/show_bug.cgi?id=33057 4784 4785commit c5b3db18d888552328e9718ea022794fc5bde352 4786Author: Alex Deucher <alexdeucher@gmail.com> 4787Date: Tue Jan 11 16:21:18 2011 -0500 4788 4789 kms: fix pitch aligment for scanout 4790 4791 Display has slightly stricter pitch alignment requirements 4792 than other blocks. Factor that in when aligning pitch. 4793 4794 Fixes: 4795 https://bugs.freedesktop.org/show_bug.cgi?id=32997 4796 4797commit bbd7adce889359b5eb3239b73e904b3ede283e12 4798Author: Alex Deucher <alexdeucher@gmail.com> 4799Date: Tue Jan 11 15:41:03 2011 -0500 4800 4801 radeon: fix yet another pitch align 4802 4803commit af2e6d7d2f1b3d8f8f6b0acfb2b7b0cfaff7bcdb 4804Author: Alex Deucher <alexdeucher@gmail.com> 4805Date: Tue Jan 11 14:42:17 2011 -0500 4806 4807 radeon: fix pitch align in pageflip code 4808 4809commit e5d0a400d08da2358fac9c2ad12042f125525736 4810Merge: 0e432df e27e9b4 4811Author: Alex Deucher <alexdeucher@gmail.com> 4812Date: Mon Jan 10 18:16:49 2011 -0500 4813 4814 Merge branch 'kms-pflip' of git+ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 4815 4816commit 0e432dff9e06a183acaeb20db29cbd03ff0f4b82 4817Author: Alex Deucher <alexdeucher@gmail.com> 4818Date: Thu Jan 6 20:56:45 2011 -0500 4819 4820 NI: add pci ids 4821 4822commit 34d61785b9bc13287bd7ab8bdd8a1b99a3df4440 4823Author: Alex Deucher <alexdeucher@gmail.com> 4824Date: Tue Nov 23 20:54:57 2010 -0500 4825 4826 Add EXA/Xv acceleration support for NI chips 4827 4828commit 97322c00faf08461b941edf97efe86d8b082a0ce 4829Author: Alex Deucher <alexdeucher@gmail.com> 4830Date: Wed Nov 24 12:18:04 2010 -0500 4831 4832 Bail on NI cards with UMS 4833 4834 NI cards require KMS. 4835 4836commit c678b79f99238473df6ff1bedbae840950cdf88c 4837Author: Alex Deucher <alexdeucher@gmail.com> 4838Date: Tue Nov 23 20:45:25 2010 -0500 4839 4840 add NI family ids 4841 4842commit e27e9b4e50ad42885ad2e25be897cdf29aa59712 4843Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 4844Date: Sun Dec 19 03:56:17 2010 +0100 4845 4846 ddx/ati: Bugfix for pageflip consistency check. 4847 4848 target_msc wasn't passed from vblank event handler 4849 to pageflip routine, due to a missing initalization. 4850 Now fixed. 4851 4852 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 4853 4854commit 30c5f217f91ed1ee15b5fff0002521d98682efb7 4855Author: Alex Deucher <alexdeucher@gmail.com> 4856Date: Fri Dec 17 13:40:40 2010 -0500 4857 4858 exa/dfs/ontario: always use the CPU for DFS for now 4859 4860 There are issues with rendering to gart memory that haven't 4861 been sorted out completely yet. 4862 4863commit e97079c1b621433b4aa3ecfdd2aa0884aac5df76 4864Author: Alex Deucher <alexdeucher@gmail.com> 4865Date: Thu Dec 16 18:49:55 2010 -0500 4866 4867 DCE4.1: fix DIGx routing 4868 4869 Works more like DCE4.0. With only two DIGx blocks 4870 the routing is hardcoded based on link. 4871 This fixes blank screen issues when changing crtc 4872 routing due to incorrect crtc to dig mapping. 4873 4874 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4875 4876commit 3b1fb12dec7df4e9434b2c51568ac8d997638922 4877Author: Alex Deucher <alexdeucher@gmail.com> 4878Date: Sun Dec 12 19:16:47 2010 -0500 4879 4880 man: minor formatting fixes 4881 4882commit b1af52e522158788a9613d6f1a2309a3987ad59f 4883Author: Alex Deucher <alexdeucher@gmail.com> 4884Date: Tue Dec 7 19:17:09 2010 -0500 4885 4886 man: add rs880 info 4887 4888commit f9bbb26dd97254b66de11bb2abd821aa293ecba5 4889Author: Matt Turner <mattst88@gmail.com> 4890Date: Thu Dec 2 17:00:11 2010 -0500 4891 4892 Add missing call to vbeFree (bug 4417) 4893 4894 Also kill unneeded vbe.h include from radeon_bios.c. 4895 4896 Signed-off-by: Matt Turner <mattst88@gmail.com> 4897 4898commit 9fc16ff3f8806971cb495e8838af1e5f9149d34f 4899Author: Alex Deucher <alexdeucher@gmail.com> 4900Date: Fri Dec 3 02:04:28 2010 -0500 4901 4902 radeon: remove duplicated define 4903 4904 xfixedtofloat macro was defined several times 4905 4906commit f074762fff4253a457cb48519bb33a2e2c90f8b9 4907Author: Alex Deucher <alexdeucher@gmail.com> 4908Date: Fri Dec 3 01:25:41 2010 -0500 4909 4910 radeon/exa: fallback for non-affine transforms 4911 4912 We need to implement a texture lookup with perspective 4913 divide for non-affine transforms. For now just fallback. 4914 Should fix: 4915 https://bugs.freedesktop.org/show_bug.cgi?id=31799 4916 although it appears either EXA or the xserver gets this 4917 wrong too. 4918 4919commit 774b114baa932078f8bf91ea302850a8c54c2c07 4920Author: Nicolas Joly <njoly@pasteur.fr> 4921Date: Thu Dec 2 16:02:44 2010 -0500 4922 4923 ati: remove leftover word in ati.man 4924 4925 fixes: 4926 https://bugs.freedesktop.org/show_bug.cgi?id=28297 4927 4928commit 69639ef377a9d6701cdef902f8a1c5e0b58cf833 4929Author: Alex Deucher <alexdeucher@gmail.com> 4930Date: Thu Dec 2 02:12:24 2010 -0500 4931 4932 radeon/kms: pageflipping man page updates 4933 4934commit f48af8a6cfa1ac665f07b8f9712e94b77bc4f5e9 4935Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 4936Date: Tue Nov 23 04:41:53 2010 +0100 4937 4938 ddx/ati: Fix reporting of pageflip completion events on multi-head. 4939 4940 When a drawable is page-flipped on multiple crtc's (fullscreen 4941 drawable on mirror-mode or multi-head x-screen), only one pageflip 4942 event is finally delivered, after the last participating crtc signals 4943 flip completion, this to avoid visual corruption. 4944 4945 Old code returned vblank count and timestamps of flip completion 4946 of this last crtc, instead of the values of the "master crtc", the 4947 one that was used for initially scheduling/triggering the pagflip 4948 via vblank events. (master = radeon_dri2_drawable_crtc(drawable)) 4949 4950 This patch makes sure that the pageflip completion values of the 4951 "master" crtc are returned, otherwise client applications will 4952 get confused by the random (msc, ust) values returned by whichever 4953 crtc was the last to complete its flip. Without this, the returned 4954 values change randomly and jump forward and backward in time and 4955 count. 4956 4957 The patch also implements a consistency check on returned vblank 4958 count values of pageflip completion. Impossible values are detected, 4959 a x-warning is logged and returned (msc,ust) values are marked invalid, 4960 so clients could perform error handling. Such a warning would indicate 4961 bugs in the pageflip completion routine of future kms drivers or the 4962 ddx to aid driver debugging. 4963 4964 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 4965 4966commit 122536ee0aeb1eef1a9d80d5e464dcb423dc2837 4967Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 4968Date: Mon Nov 22 04:11:07 2010 +0100 4969 4970 ddx/ati: Add option "SwapbuffersWait" to control vsync of DRI2 swaps. 4971 4972 A new optional kms driver option "SwapbuffersWait" is defined 4973 for xorg.conf, which defaults to "on". If "on", DRI2 bufferswaps 4974 will be synchronized to vsync, otherwise not. 4975 4976 This currently only affects copy-swaps, not pageflipped swaps. 4977 It also requires a swap_interval setting of zero by the OpenGL 4978 client. 4979 4980 Ideally, we'd provide a way for dri2 to pass the current swap 4981 interval to the ddx so we could change this dynamically. 4982 4983 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 4984 4985commit 0de680730294bd623f6b3e189faa7b88a09d3a2a 4986Author: Alex Deucher <alexdeucher@gmail.com> 4987Date: Wed Nov 17 17:39:39 2010 -0500 4988 4989 radeon/kms: add pageflip support 4990 4991 requires radeon drm 2.8.0 or higher 4992 4993 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 4994 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 4995 4996commit fccdca8db34010f566bd068c74cdef0f4a8cb7f5 4997Author: Alex Deucher <alexdeucher@gmail.com> 4998Date: Wed Nov 17 17:37:25 2010 -0500 4999 5000 radeon/kms: allow tiled front buffer on 6xx/7xx 5001 5002 Use UTS/DFS to tile/untile as appropriate for sw access. 5003 Also enables pageflipping with tiling enabled. 5004 5005commit 035f7f3ab529ca19b853066792af8a23d08a0f53 5006Author: Dave Airlie <airlied@redhat.com> 5007Date: Wed Dec 1 10:23:14 2010 +1000 5008 5009 radeon: add some missing math.h and assert.h includes 5010 5011commit 0f924e83d7be6fca7e024acdf973089efdf3b627 5012Author: Dave Airlie <airlied@redhat.com> 5013Date: Wed Dec 1 10:23:33 2010 +1000 5014 5015 radeon: make sure EXA is off when EXA is disabled with --disable-exa 5016 5017commit e142e55c56d8440d7185b594c696ec9eeb699cda 5018Author: Alex Deucher <alexdeucher@gmail.com> 5019Date: Tue Nov 30 15:51:21 2010 -0500 5020 5021 evergreen: set default group_bytes to 256 5022 5023 512 seems to cause issues on certain cards. 5024 5025 Fixes: 5026 https://bugs.freedesktop.org/show_bug.cgi?id=31724 5027 5028commit 90f831361844f1b80b3f6bb718ff5ac584d73d48 5029Author: Alex Deucher <alexdeucher@gmail.com> 5030Date: Mon Nov 29 18:09:05 2010 -0500 5031 5032 evergreen: use dot4 for transforms 5033 5034commit 3cae361d0448b6e231c80f53d64bdbbdd74dc4cf 5035Author: Alex Deucher <alexdeucher@gmail.com> 5036Date: Mon Nov 29 17:44:47 2010 -0500 5037 5038 6xx/7xx: clean up gpr/const handling in shaders 5039 5040commit d9bcac516f2a810acb300b29169e56a2df0b47ac 5041Author: Alex Deucher <alexdeucher@gmail.com> 5042Date: Mon Nov 29 17:23:30 2010 -0500 5043 5044 r6xx/r7xx use dot4 for transforms 5045 5046commit 5d3f33729be0639cef17372345b2dab6127e39d9 5047Author: Alex Deucher <alexdeucher@gmail.com> 5048Date: Wed Nov 24 10:59:09 2010 -0500 5049 5050 radeon/kms: fix output enumeration 5051 5052 If a system has a mix of DVI-I, and DVI-D connectors, or a mix 5053 of HDMI-A and HDMI-B connectors, the enumeration was per 5054 connector type, but the naming was the same for multiple 5055 connectors (DVI-[0-n] for DVI-I, -D, -A, and HDMI-[0-n] 5056 for HDMI-A, and -B), you'd end up with multiple connectors 5057 with the same name. Fix this by tracking the number of 5058 connectors across similar connector types. 5059 5060commit dc470b5382ffa27bd4a673b21fb6b978d3716ca0 5061Author: Alex Deucher <alexdeucher@gmail.com> 5062Date: Tue Nov 23 01:46:15 2010 -0500 5063 5064 radeon/man: divide driver options into UMS and KMS 5065 5066commit 4cbb7100df4b983d77c20a77431cf2f5101bde36 5067Author: Alex Deucher <alexdeucher@gmail.com> 5068Date: Tue Nov 23 01:22:04 2010 -0500 5069 5070 update man page to include ontario 5071 5072commit 718dfd443897d70d97fc19afddfbcfe1e39a9c59 5073Author: Alex Deucher <alexdeucher@gmail.com> 5074Date: Mon Nov 22 13:11:50 2010 -0500 5075 5076 Ontario: add workaround for small pixmap issues with DFS 5077 5078 On past asics these kind a problems were usually due to a bad 5079 backend map config. I suspect in this case, and the previous 5080 cases it may be related to using linear buffers rather than 5081 tiled. 5082 5083 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5084 5085commit 61c97b0812c820564c7bc9d64a998c109bcd46b3 5086Author: Alex Deucher <alexdeucher@gmail.com> 5087Date: Mon Nov 22 13:10:16 2010 -0500 5088 5089 Add EXA/Xv acceleration support for Ontario Fusion APUs 5090 5091 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5092 5093commit 0659f35dfbeda7dd3ff58da1deb6484fabadc332 5094Author: Alex Deucher <alexdeucher@gmail.com> 5095Date: Mon Nov 22 13:07:00 2010 -0500 5096 5097 Add Ontario fusion APU pci ids 5098 5099 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5100 5101commit fa0cec331e6c94cebf82270dcd6c179b15851c5e 5102Author: Alex Deucher <alexdeucher@gmail.com> 5103Date: Mon Nov 22 12:56:45 2010 -0500 5104 5105 ontario: add UMS modesetting support 5106 5107 Accel not enabled yet. 5108 5109 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5110 5111commit 7ad1f01573e930aa6238eaeb0e11708ae548dfce 5112Author: Jerome Glisse <jglisse@redhat.com> 5113Date: Fri Nov 19 10:27:40 2010 -0500 5114 5115 evergreen: fix typo for fetch resource offset 5116 5117 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 5118 5119commit 3455a3b58532ea3ad901a317126968ab6cbb21b7 5120Author: Alex Deucher <alexdeucher@gmail.com> 5121Date: Wed Nov 17 17:32:41 2010 -0500 5122 5123 radeon/kms: fix buffer base alignment for tiling 5124 5125 On r6xx+, 2D tiling can require larger than 4k base alignment. 5126 5127commit beb7fecd0191e38fb238134ba612985062cf9770 5128Author: Dave Airlie <airlied@redhat.com> 5129Date: Mon Nov 15 11:59:58 2010 +1000 5130 5131 dce4: improve the VT switching code. 5132 5133 add in lots more blocks of regs to save/restore 5134 5135commit 51e51f864a691e51b5193d102398ced667e0beb1 5136Author: Dave Airlie <airlied@redhat.com> 5137Date: Thu Nov 11 18:48:40 2010 +1000 5138 5139 evergreen: remove warnings since last commit 5140 5141commit abbf73ee990512ac16ca77e8bb23288495e1f9f4 5142Author: Dave Airlie <airlied@redhat.com> 5143Date: Thu Nov 11 09:38:15 2010 +1000 5144 5145 evergreen: add UMS VT switch support. 5146 5147 This isn't perfect, but it brings back text VTs here on the 5148 DAC and DVI outputs. 5149 5150commit e1dfaf93d06bc5eafdbc2e1823d19204ce8f242c 5151Author: Dave Airlie <airlied@redhat.com> 5152Date: Thu Nov 11 09:15:27 2010 +1000 5153 5154 avivo: use arrays to store the crtc/pll info for save/restore 5155 5156 this is just prep work for evergreen VT save/restore 5157 5158commit 0c2834e67df6e143a7d15f373faaddca1dda6b18 5159Author: Michel Dänzer <daenzer@vmware.com> 5160Date: Fri Nov 5 09:30:40 2010 +0100 5161 5162 Fix incorrect CS size in one textured video path. (Bug #31364) 5163 5164 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=31364 . 5165 5166commit 2a7b080727f61559cf8444c93a5088f4155a8fdf 5167Author: Alex Deucher <alexdeucher@gmail.com> 5168Date: Fri Oct 29 16:56:45 2010 -0400 5169 5170 radeon/kms: more alignment fixes 5171 5172commit bf60af579382a0d48f7a65a4cec88759cc8b683c 5173Author: Dave Airlie <airlied@redhat.com> 5174Date: Thu Oct 28 13:06:54 2010 +1000 5175 5176 dri2: reference count the client privates key/callback 5177 5178 This lets multi-screen work better, but still having issues after server 5179 recycle, but it doesn't crash at least. 5180 5181commit fb22d0c06a7dc42216230e198ff443d8035e9d21 5182Author: Dave Airlie <airlied@redhat.com> 5183Date: Thu Oct 28 11:10:03 2010 +1000 5184 5185 radeon: only register block handler once per fd. 5186 5187 should fix https://bugs.freedesktop.org/show_bug.cgi?id=29726 5188 5189 the problem is of course the second head instance tries to access the 5190 fd and fails, however I think this might break syncing on the second 5191 head but not sure, but its better than just hanging up the X server 5192 5193commit 040140b5d7cbeaf5d83687bc87e4a236ff9f0cbd 5194Author: Alex Deucher <alexdeucher@gmail.com> 5195Date: Wed Oct 27 20:16:17 2010 -0400 5196 5197 drm/radeon: unify fb bo alignment harder 5198 5199 More duplicated paths discoved... 5200 5201commit f07f9b7b61c05f2de1d61bb0e2f71bd017c8d36a 5202Author: Alex Deucher <alexdeucher@gmail.com> 5203Date: Wed Oct 27 12:37:42 2010 -0400 5204 5205 kms/radeon: unify fb bo alignment handling 5206 5207 Previously there were 3 different paths with what should 5208 have had duplicated code: 5209 - EXACreatePixmap2 5210 - Initial front buffer creation 5211 - Randr resize 5212 5213 This patch attempts to unify the alignment across all 3. 5214 5215 This may fix tiling issues in some cases and should make 5216 buffer pitches match for pageflipping. 5217 5218commit d31046ba6c8eee9b7decc3875697d37c38bc38f3 5219Author: Michel Dänzer <daenzer@vmware.com> 5220Date: Wed Oct 27 12:12:04 2010 +0200 5221 5222 Re-use result of GetScreenPixmap call. 5223 5224 Fixes compile warning due to local variable ppix being unused when building 5225 against current xserver Git. 5226 5227commit 4dd7f835c17b00707f1d8d4e36a24380ba52761e 5228Author: Michel Dänzer <daenzer@vmware.com> 5229Date: Wed Oct 27 12:09:03 2010 +0200 5230 5231 Fix another stray xfree() call. 5232 5233commit f815e9c3c61b1ed73e7dd9383587efcdd8dfc07f 5234Author: Michel Dänzer <daenzer@vmware.com> 5235Date: Wed Oct 27 11:29:12 2010 +0200 5236 5237 Don't try to delete DRI2 event list entries that were never added. (Bug #31086) 5238 5239 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=31086 . 5240 5241commit df92245e1fd0ead2581b675cd96ac8bcc8cb65d7 5242Author: Alex Deucher <alexdeucher@gmail.com> 5243Date: Sun Oct 24 14:27:27 2010 -0400 5244 5245 r100: add some missing bits in SE_COORD_FMT 5246 5247 Texture coordinates work fine with or without these, 5248 but this should be more correct I think although 5249 I don't think it matters since we aren't sending w 5250 anyway. 5251 5252commit fc7aa6e3bbb86eaa2f34b827629f7e82cde65ac7 5253Author: Alex Deucher <alexdeucher@gmail.com> 5254Date: Sun Oct 24 11:56:48 2010 -0400 5255 5256 UMS/radeon: fix typo in restore palette 5257 5258commit 5efdf514ab0e5439114e8dd7a49105812155eb69 5259Author: Alex Deucher <alexdeucher@gmail.com> 5260Date: Sun Oct 24 11:45:19 2010 -0400 5261 5262 UMS/radeon: fix screen dimming on VT switch 5263 5264 Save and restore the palettes on VT switch. The restore 5265 has to be done after the vga restore to work properly as 5266 determined by Jonathan Kollasch. 5267 5268 Fixes: 5269 https://bugs.freedesktop.org/show_bug.cgi?id=18407 5270 5271commit 1e3ede62c39307d0210cb2044badd619a4f44fec 5272Author: Alex Deucher <alexdeucher@gmail.com> 5273Date: Sun Oct 24 11:24:07 2010 -0400 5274 5275 UMS/atom: default to DVI mode if we don't have monitor hdmi info 5276 5277 Fixes: 5278 https://bugs.freedesktop.org/show_bug.cgi?id=30330 5279 5280commit 51f26a7f01b68e373867c03881868ca8830664d6 5281Author: Mark Schreiber <mark7@alumni.cmu.edu> 5282Date: Sat Apr 10 10:55:08 2010 -0700 5283 5284 Correct spelling and grammar 5285 5286commit 0f575984594e6977b4a7cb794dad8c6e56c850f1 5287Author: Alex Deucher <alexdeucher@gmail.com> 5288Date: Tue Oct 19 19:21:39 2010 -0400 5289 5290 Xv: limit overlay to 2047x2047 5291 5292 Fixes: 5293 https://bugs.freedesktop.org/show_bug.cgi?id=15391 5294 5295commit 0f184630cd4760199430320215e4280438fc4ce5 5296Author: Alex Deucher <alexdeucher@gmail.com> 5297Date: Tue Oct 19 17:13:06 2010 -0400 5298 5299 radeon: fix compilation on xserver >= 1.10 5300 5301 Fixes: 5302 https://bugs.freedesktop.org/show_bug.cgi?id=30451 5303 5304 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5305 5306commit 179b0e15924ae2757efaece6488ae835ba6436e3 5307Author: Dave Airlie <airlied@redhat.com> 5308Date: Tue Oct 19 13:37:22 2010 +1000 5309 5310 r600: for 2d tiles we need to align to channels no banks. 5311 5312commit 809cacbc23ab37aa7952665f9b2e37c03a34f173 5313Author: Adam Jackson <ajax@redhat.com> 5314Date: Mon Oct 18 12:32:11 2010 -0400 5315 5316 kms: Let the server know if we got more than 1 EDID block 5317 5318 Otherwise things like xf86MonitorIsHDMI() won't work right. 5319 5320 Signed-off-by: Adam Jackson <ajax@redhat.com> 5321 5322commit 042037e485396981cf4a420c247104ba5d016ca0 5323Author: Alex Deucher <alexdeucher@gmail.com> 5324Date: Thu Oct 14 17:00:51 2010 -0400 5325 5326 evergreen: requires same cursor fix as avivo chips 5327 5328commit 0f37458c253aec46d98db6661c4def711df8fc17 5329Author: Adam Jackson <ajax@redhat.com> 5330Date: Thu Oct 14 14:37:34 2010 -0400 5331 5332 Don't allow the config file to override BIOS location 5333 5334 Signed-off-by: Adam Jackson <ajax@redhat.com> 5335 5336commit 2918df3a312fc600e83f84c298b9653d2eb3a8d5 5337Author: Alex Deucher <alexdeucher@gmail.com> 5338Date: Wed Oct 13 17:16:04 2010 -0400 5339 5340 evergreen: increase XV_CRTC Xv attribute range 5341 5342 Cover all 6 possible crtcs. 5343 5344commit 0d1f9fd03d0196dda5c1ce34b2e68b007c1874f2 5345Author: Alex Deucher <alexdeucher@gmail.com> 5346Date: Sun Oct 10 12:06:22 2010 -0400 5347 5348 DCE3+: UMS modesetting fixes 5349 5350 - add mising LVTMA case statement for DCE3.0 dig encoder 5351 - some DCE4 systems have EN/DISABLE_OUTPUT actions 5352 5353commit acf4024aa0ef192355b2bd8281507b1c2e6b7de4 5354Author: Alex Deucher <alexdeucher@gmail.com> 5355Date: Fri Oct 8 16:30:36 2010 -0400 5356 5357 ATOM DDC fixes for UMS 5358 5359 - Header size was already subtraced from table size 5360 - Only hw capable ddc pads are shared with aux 5361 5362commit 74fd2b91477106a26a2d9fb4b11c885910996041 5363Author: Alex Deucher <alexdeucher@gmail.com> 5364Date: Thu Oct 7 17:33:00 2010 -0400 5365 5366 DCE3.2 UMS: fix duallink 5367 5368 Should fix: 5369 https://bugs.freedesktop.org/show_bug.cgi?id=30686 5370 5371commit 8c544272f218fd8bd93a898b96afedc2498b27bb 5372Author: Alex Deucher <alexdeucher@gmail.com> 5373Date: Thu Oct 7 17:25:47 2010 -0400 5374 5375 fix the non-kms build 5376 5377 Should fix: 5378 https://bugs.freedesktop.org/show_bug.cgi?id=30685 5379 5380commit 81360adffb2a66b9a95a38671f9227a9718c9841 5381Author: Oldřich Jedlička <oldium.pro@seznam.cz> 5382Date: Mon Sep 6 21:14:52 2010 +0200 5383 5384 radeon: proper DRI2 pending events handling when client gone. (v6) 5385 5386 Properly handle asynchronous DRI2 events for disconnected clients. 5387 Track client's pending requests and mark them as invalid when the 5388 client disconnects. 5389 5390 This is based on the version from Alban Browaeys in bug #29065. 5391 5392 v1 (Alban Browaeys): Based upon a detailed explanation from Oldřich 5393 Jedlička and comments from Christopher James Halse Rogers. 5394 on http://lists.x.org/archives/xorg-driver-ati/2010-August/016780.html . 5395 5396 v2: Updated version to apply on master. Removed unnecessary 5397 client_index field from _DRI2FrameEvent. Added freeing/removing from 5398 list to failed paths of radeon_dri2_schedule_wait_msc and 5399 radeon_dri2_schedule_swap. 5400 5401 v3: Adopt to older xorg-server that doesn't have dixRegisterPrivateKey. 5402 5403 v4: Conditional include of list.h, unreachable return removed. 5404 5405 v5: Distribute list.h as xorg_list.h, remove xorg-server version check. 5406 Use the version from xorg-server when available (checked in 5407 configure.ac). 5408 5409 v6: Removed xorg_list.h, made DRI2 scheduling features dependent on 5410 list.h presence. 5411 5412commit adee138f007e26307f1aab7f8fe066150c12e55d 5413Author: Jesse Adkins <jesserayadkins@gmail.com> 5414Date: Tue Sep 28 13:29:50 2010 -0700 5415 5416 Purge cvs tags. 5417 5418 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com> 5419 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 5420 5421commit f1bf9cbb471d60ee4b0422bc4e896e3b86f4d75f 5422Author: Alex Deucher <alexdeucher@gmail.com> 5423Date: Wed Oct 6 11:44:42 2010 -0400 5424 5425 radeon: man page updates 5426 5427 - tv-out is not supported on carding using 5428 the rage theatre chip for tv-out 5429 - add evergreen cards to the list 5430 5431commit 0921ecc1c751df0dd56e0b1d0b78ab53d7164904 5432Author: Alex Deucher <alexdeucher@gmail.com> 5433Date: Mon Oct 4 12:34:47 2010 -0400 5434 5435 evergreen: use EXACreatePixmap2 if available 5436 5437commit 5bdb6434975584eef90eb9e5955b9c2a14b7f327 5438Author: Alex Deucher <alexdeucher@gmail.com> 5439Date: Mon Oct 4 12:31:51 2010 -0400 5440 5441 evergreen: port Karl's UTS/DFS changes to evergreen 5442 5443commit be8f45cbd313b68ad663f303c64edb4525b8f981 5444Merge: e843faf ff5f466 5445Author: Alex Deucher <alexdeucher@gmail.com> 5446Date: Mon Oct 4 12:36:56 2010 -0400 5447 5448 Merge branch 'evergreen_accel' of git+ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 5449 5450commit ff5f466e6ca0ee541a255facff6c9198976dfd52 5451Author: Alex Deucher <alexdeucher@gmail.com> 5452Date: Fri Oct 1 15:46:43 2010 -0400 5453 5454 evergreen/dri2: fix depth allocation for depth+stencil 5455 5456 evergreen uses separate allocations for depth and stencil, 5457 so to handle that, create a depth buffer large enough to 5458 handle both. This is required for using the stencil 5459 buffer in mesa. 5460 5461commit e843faf355c864beab81e74f0e39f8ad53d4c2bf 5462Author: Alex Deucher <alexdeucher@gmail.com> 5463Date: Thu Sep 30 19:30:35 2010 -0400 5464 5465 radeon: fix fbLocation for >32 bit MC addresses 5466 5467 If the fbLocation was at an address >32 bits, we'd fail. 5468 Change fbLocation to uint64_t and properly cast when needed. 5469 5470commit 886febc882053e09294225e85b102f965041b62b 5471Author: Alex Deucher <alexdeucher@gmail.com> 5472Date: Thu Sep 30 19:20:17 2010 -0400 5473 5474 r6xx: fix bad mask when setting up HDP_NONSURFACE_BASE 5475 5476 This fails for MC addresses >32 bits 5477 5478commit 16589b862feb796e15dc6e471d25604a4f5e7c1c 5479Author: Alex Deucher <alexdeucher@gmail.com> 5480Date: Mon Sep 27 18:28:43 2010 -0400 5481 5482 bump version post release 5483 5484commit cc5005af61f45a3552f7358dc5aa711e42f5af54 5485Author: Alex Deucher <alexdeucher@gmail.com> 5486Date: Mon Sep 27 18:20:53 2010 -0400 5487 5488 bump version for release 5489 5490commit 7f8820fcec8c90bf2f823170bd08a23e8b4ff7af 5491Author: Michael Cree <mcree@orcon.net.nz> 5492Date: Mon Sep 27 13:39:12 2010 -0400 5493 5494 Fix some unaligned 32bit accesses in the AtomBios code. 5495 5496 On the Alpha architecture unaligned 32bit accesses incur a software 5497 trap to the kernel and pollute the kernel logs. Fixed by use of the 5498 ldl_u() interface. 5499 5500 Signed-off-by: Michael Cree <mcree@orcon.net.nz> 5501 5502commit c4f834cdfbe96aa47ac5fb039f9dd7aa9730c8a3 5503Author: Nicolas Reinecke <nr@das-labor.org> 5504Date: Mon Sep 27 13:33:55 2010 -0400 5505 5506 radeon: Convert remaining x(c)alloc/xfree to m/calloc/free. 5507 5508 Fixes deprecation warnings missed out by 5509 f7a91ece264af9f3fd2fc18e99aefcda93ce9f5c 5510 5511commit f8fb9312d791af1f77020e8c2d35bb30841ed9aa 5512Author: Karl Tomlinson <karlt+@karlt.net> 5513Date: Sun Aug 22 22:46:33 2010 +1200 5514 5515 RADEONPrepareAccess_CS: fallback to DFS when pixmap is in VRAM 5516 5517 This avoids costly CPU VRAM reads and lets EXA manage a system memory cache 5518 of the portions of pixmaps needed for unaccelerated operations. 5519 5520 https://bugs.freedesktop.org/show_bug.cgi?id=27139 5521 5522commit 35c4ff936601ee083f51510a5192fb97d622a483 5523Author: Karl Tomlinson <karlt+@karlt.net> 5524Date: Sun Aug 22 22:28:06 2010 +1200 5525 5526 radeon: complete UTS and DFS even when a scratch BO is not necessary 5527 5528 Turns on the big-endian paths even for little-endian systems, and adds 5529 similar paths to the r6xx/r7xx functions. 5530 5531 This makes UTS and DFS reliable, which will let PrepareAccess (with 5532 mixed pixmaps) choose to fail based on whether the pixmap is in VRAM 5533 (to avoid CPU reads). 5534 5535commit d46381a3a6bf10903803f5acaa7aa0ce06373b96 5536Author: Karl Tomlinson <karlt+@karlt.net> 5537Date: Sun Aug 22 21:02:45 2010 +1200 5538 5539 radeon: complete big endian UTS and DFS even when scratch allocation fails. 5540 5541 On big endian systems, PrepareAccess will fail when byte-swapping is 5542 required so UploadToScreen and DownloadFromScreen cannot rely on 5543 fallback to PrepareAccess. 5544 5545 When scratch BO space allocation fails, this patch merely adds simple 5546 fallback to direct CPU access without any GPU blit. This sometimes 5547 requires a CS flush even in UploadToScreen. 5548 (No allocation retry after a flush is added here.) 5549 5550commit 4ced4e1eff67946e306c0c67c9ed59dd5f3c4ba9 5551Author: Karl Tomlinson <karlt+@karlt.net> 5552Date: Sun Aug 22 20:04:42 2010 +1200 5553 5554 RADEONDownloadFromScreenCS: flush CS writes before mapping BO for read 5555 5556 If unflushed CS operations write to the pixmap BO, then these need to be 5557 flushed before mapping the BO for read. This currently only affects big 5558 endian systems and only when the operation writes to the GTT domain. 5559 5560commit a4eef8faffbb1ea2f742273ee855f4e6f992e5c8 5561Author: Karl Tomlinson <karlt+@karlt.net> 5562Date: Sat Aug 21 22:29:34 2010 +1200 5563 5564 FinishAccess_CS: set bo_mapped to FALSE on unmap 5565 5566 This is actually only necessary when PrepareAccess may behave differently on 5567 different calls with the same pixmap, which currently doesn't happen. 5568 5569 However resetting bo_mapped is necessary to let PrepareAccess (with mixed 5570 pixmaps) choose to fail based on whether the pixmap is in VRAM (to avoid CPU 5571 reads). 5572 5573commit bfebe039af0c0282d04eb6234b6e6d1e02097146 5574Author: Karl Tomlinson <karlt+@karlt.net> 5575Date: Sat Aug 21 21:44:39 2010 +1200 5576 5577 DownloadFromScreenCS: download via a scratch BO if pixmap domain is unknown 5578 5579 radeon_bo_is_busy() may return without setting the domain out-parameter. 5580 If this happens, then download via a scratch GTT BO to avoid CPU VRAM read. 5581 5582commit e26a59e9db8067882327f872e3d2d760ce4c66f3 5583Author: Mathieu Bérard <mathieu@mberard.eu> 5584Date: Thu Sep 9 19:32:38 2010 -0400 5585 5586 Fix NULL possible deref in evergreen_cp_wait_vline_sync 5587 5588commit b90cb61ccb0f4f80e0627141f223354a9371d47d 5589Author: Alex Deucher <alexdeucher@gmail.com> 5590Date: Tue Sep 7 11:51:16 2010 -0400 5591 5592 radeon: set interlaced and doublescan enabled for randr outputs 5593 5594 interlaced used to work without setting these parameters. Changes 5595 in the xserver seem to require them now. 5596 5597 Should fix: 5598 https://bugs.freedesktop.org/show_bug.cgi?id=29591 5599 5600commit 7cc0d3311f23ad569af004e0e4a0e2efbac107e5 5601Author: Alex Deucher <alexdeucher@gmail.com> 5602Date: Fri Sep 3 00:32:27 2010 -0400 5603 5604 evergreen: fix Xv 5605 5606 VS const buffer offset was wrong. 5607 5608 fixes: 5609 https://bugs.freedesktop.org/show_bug.cgi?id=29788 5610 5611commit 7c6ce4e62693f446d7d3c8a86502ccc03c0e55b0 5612Author: Alex Deucher <alexdeucher@gmail.com> 5613Date: Thu Sep 2 20:34:56 2010 -0400 5614 5615 evergreen: add additional default state 5616 5617 Note, you also need a drm patch to fix the GPU hangs: 5618 drm/radeon/kms/evergreen: fix gpu hangs in userspace accel code 5619 5620 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5621 5622commit 2b98ec1f7e931019a4ab699a56d5dfaa395946fb 5623Author: Alex Deucher <alexdeucher@gmail.com> 5624Date: Wed Sep 1 13:24:19 2010 -0400 5625 5626 r6xx/r7xx: fix typos 5627 5628 some stray - signs 5629 5630commit 966ac1be81da76c8aa4ea46b63f3ca5358a2c021 5631Author: Alex Deucher <alexdeucher@gmail.com> 5632Date: Fri Aug 27 18:22:21 2010 -0400 5633 5634 evergreen: work around bad data in some i2c tables 5635 5636 The 7th entry in a lot of evergreen i2c gpio tables is partially 5637 zeroed. Fix the entry. 5638 5639 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5640 5641commit 91f707d308d4bbf16c3d62d046cf280fef5a8f4c 5642Author: Heikki Lindholm <holin@iki.fi> 5643Date: Fri Aug 27 02:26:24 2010 -0400 5644 5645 xv: fix non-kms/non-dri Xv column ordering on big endian systems 5646 5647 Column order is wrong on big endian systems, primarly because of a 5648 bits / bytes mix up with the bpp variable. Fix tested with r100 and 5649 r300, screen depth 16 and 32 with YV12 and YUY2 (overlay, textured video), 5650 RGBA and RGBT (overlay). 5651 5652 Should fix: https://bugs.freedesktop.org/show_bug.cgi?id=29041 5653 5654 Signed-off-by: Heikki Lindholm <holin@iki.fi> 5655 5656commit 6a2c8587a4e05a8be2a2e975a6660942cfe115d6 5657Author: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com> 5658Date: Fri Aug 27 13:14:33 2010 +1000 5659 5660 dri2: Reference count DRI2 buffers 5661 5662 When a client calls ScheduleSwap we set up a kernel callback when the 5663 relevent vblank event occurs. However, it's possible for the client 5664 to go away between calling ScheduleSwap and the vblank event, 5665 resulting in the buffers being destroyed before they're passed to 5666 radeon_dri2_frame_event_handler. 5667 5668 Add reference-counting to the buffers and take a reference in 5669 radeon_dri2_schedule_swap to ensure the buffers won't be destroyed 5670 before the vblank event is dealt with. 5671 5672 This parallels the approach taken by the Intel DDX in commit 5673 0d2392d44aae95d6b571d98f7ec323cf672a687f. 5674 5675 Fixes: http://bugs.freedesktop.org/show_bug.cgi?id=29065 5676 5677 v2: Don't write completion events to the client if it has quit. 5678 v3: Don't try to unref the NULL buffers from a DRI2_WAITMSC event. 5679 Take a ref in schedule_swap earlier, so the offscreen fallback 5680 doesn't incorrectly destroy the buffers. 5681 5682 Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com> 5683 Signed-off-by: Dave Airlie <airlied@redhat.com> 5684 5685commit e9928fe036e9382fd7bc353f3f05531445f08977 5686Author: Dave Airlie <airlied@redhat.com> 5687Date: Wed Aug 25 10:42:39 2010 +1000 5688 5689 remove explicit buffer submit from copy region 5690 5691 port of 0be3e95c844247746742805830860ace9f546d99 from intel driver. 5692 5693 Remove explicit batchbuffer submit in DRI2 copyregion 5694 5695 Now that we submit from the flush callback chain, we know we'll always 5696 submit before the client receives the reply or event that blocks it from 5697 rendering the next frame. 5698 5699 Signed-off-by: Dave Airlie <airlied@redhat.com> 5700 5701commit 9f13049ddf06f6f2138851a548cfb82f12a52f42 5702Author: Dave Airlie <airlied@redhat.com> 5703Date: Wed Aug 25 08:56:37 2010 +1000 5704 5705 radeon: add correct flushing for direct rendered 5706 5707 this is a port of 69d65f9184006eac790efcff78a0e425160e95aa from the Intel 5708 driver. 5709 5710 Submit batch buffers from flush callback chain 5711 5712 There are a few cases where the server will flush client output buffers 5713 but our block handler only catches the most common (before going into select 5714 If the server flushes client buffers before we submit our batch buffer, 5715 the client may receive a damage event for rendering that hasn't happened yet 5716 5717 Instead, we can hook into the flush callback chain, which the server will 5718 invoke just before flushing output. This lets us submit batch buffers 5719 before sending out events, preserving ordering. 5720 5721 Fixes 28438: [bisected] incorrect character in gnome-terminal under compiz 5722 https://bugs.freedesktop.org/show_bug.cgi?id=28438 5723 5724 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net> 5725 5726 Signed-off-by: Dave Airlie <airlied@redhat.com> 5727 5728commit 141cbc283fddeb67c2a6c47a0f0f5c2aa2bfb430 5729Author: Alex Deucher <alexdeucher@gmail.com> 5730Date: Tue Aug 24 18:01:05 2010 -0400 5731 5732 evergreen Xv: remove stray line that caused a segfault 5733 5734 fixes: 5735 https://bugs.freedesktop.org/show_bug.cgi?id=29788 5736 5737commit f170dddd3538a587f2363ef5fa10a4b484e762da 5738Author: Alex Deucher <alexdeucher@gmail.com> 5739Date: Mon Aug 23 18:58:12 2010 -0400 5740 5741 evergreen: use vbo pool for constant buffers 5742 5743commit eede93b057bbdddcde5da3220a3f8c6f73784a7e 5744Author: Alex Deucher <alexdeucher@gmail.com> 5745Date: Mon Aug 23 13:16:27 2010 -0400 5746 5747 radeon: move vbo data to a separate struct 5748 5749 this way we can share the vbo code with const buffers 5750 5751commit d8abf27dbd14f5eb746c5e8b8b1436ad292d8ec6 5752Author: Alex Deucher <alexdeucher@gmail.com> 5753Date: Mon Aug 23 01:31:27 2010 -0400 5754 5755 evergreen: fix dword counts in default state 5756 5757commit 5a7c9d94733a0db1d3565447acc9f0e751db5950 5758Author: Alex Deucher <alexdeucher@gmail.com> 5759Date: Mon Aug 23 00:43:48 2010 -0400 5760 5761 radeon: fix legacy lvds dpms sequence 5762 5763 Take from my kms commit. 5764 5765 Should fix: 5766 https://bugs.freedesktop.org/show_bug.cgi?id=19459 5767 5768 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5769 5770commit 6930d2c981221757b1e11ef194809f085753a611 5771Author: Alex Deucher <alexdeucher@gmail.com> 5772Date: Fri Aug 20 16:55:21 2010 -0400 5773 5774 Add initial EXA and Xv support for evergreen 5775 5776 Based on the r6xx/r7xx code updated for evergreen. 5777 Still causes GPU hangs in some cases. We haven't 5778 tracked down why yet. Might be related to constant 5779 buffer persistence. 5780 5781 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5782 5783commit bdd41fecdb19c83c6c7b793016b61d38065dcd13 5784Author: Alex Deucher <alexdeucher@gmail.com> 5785Date: Fri Aug 20 01:17:13 2010 -0400 5786 5787 evergreen: set encoder type to DVI for HDMI 5788 5789 Fixes: 5790 http://bugs.freedesktop.org/show_bug.cgi?id=27452 5791 5792 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5793 5794commit ad8ea1f6e5fcb4f163622cf5eab953ea812b5829 5795Author: Alex Deucher <alexdeucher@gmail.com> 5796Date: Thu Aug 19 17:04:35 2010 -0400 5797 5798 DCE3+: switch pads to ddc mode when doing i2c 5799 5800 The pins for ddc and aux are shared so you need to switch the 5801 mode when doing ddc. The ProcessAuxChannel table already sets 5802 the pin mode to DP. This should fix unreliable ddc issues 5803 on DP ports using non-DP monitors. 5804 5805 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5806 5807commit fd686668289258ffaf6b81057545e50612aac6a8 5808Author: Dave Airlie <airlied@redhat.com> 5809Date: Thu Aug 12 12:59:18 2010 +1000 5810 5811 radeon: fixup non-kms build 5812 5813commit 5a9865d90c23c4ce0f46d380ea9119eac87a99eb 5814Author: Dave Airlie <airlied@redhat.com> 5815Date: Thu Aug 12 11:18:06 2010 +1000 5816 5817 evergreen: fix multi-head cursor support 5818 5819commit 8351bb9f085fde5dc47b115375efcc61adc23859 5820Author: Dave Airlie <airlied@redhat.com> 5821Date: Tue Aug 10 15:05:45 2010 +1000 5822 5823 radeon: take 8/10 encoding into a/c on displayport link 5824 5825commit b91e0efcb24eea32e6978c146c5409eeeeac0a62 5826Author: Dave Airlie <airlied@redhat.com> 5827Date: Tue Aug 10 15:05:25 2010 +1000 5828 5829 dce32: remove rmx workaround 5830 5831commit 94bc1b7156cd0866566dc44a823c7e051bb45175 5832Author: Dave Airlie <airlied@redhat.com> 5833Date: Tue Aug 10 13:24:52 2010 +1000 5834 5835 atombios: fixup set crtc source like KMS 5836 5837 This removes a bunch of strict aliasing warnings and fixes the 5838 codepaths up like the latest KMS code, including a workaround for a bug 5839 on evergreen. 5840 5841commit 9bc716eb62d4e0eed2902b92437a42634eef6ba1 5842Author: Dave Airlie <airlied@redhat.com> 5843Date: Tue Aug 10 13:23:21 2010 +1000 5844 5845 atombios: move adjust pixel clock around to follow KMS code flow 5846 5847 This reworks the pixel clock adjusting code to follow the KMS style, 5848 also fixes warnings in this code. 5849 5850commit 31de43bf9d9eb93cc2b2150474ea7404beabe49d 5851Author: Dave Airlie <airlied@redhat.com> 5852Date: Tue Aug 10 13:21:39 2010 +1000 5853 5854 displayport: retry on timeout 5855 5856 this is ported from KMS 5857 5858commit 9ef67335583d36080d227e8bce1966afe08e0486 5859Author: Dave Airlie <airlied@redhat.com> 5860Date: Tue Aug 10 13:21:01 2010 +1000 5861 5862 evergreen: don't call YUV table on evergreen 5863 5864commit 1cce55e8ba43e7958cb67147aeaeb068826ab99f 5865Author: Dave Airlie <airlied@redhat.com> 5866Date: Tue Aug 10 13:20:13 2010 +1000 5867 5868 evergreen: add support to parse firmware info for ext dp clk 5869 5870commit bbffd67d3296344e8735b007cdee83146d38369c 5871Author: Dave Airlie <airlied@redhat.com> 5872Date: Tue Aug 10 13:14:54 2010 +1000 5873 5874 atombios: realign digital transmitter/encoder setup with kms 5875 5876commit 6244153467665f5007e2fc7786b4bcc4b0b96030 5877Author: Dave Airlie <airlied@redhat.com> 5878Date: Tue Aug 10 12:57:22 2010 +1000 5879 5880 update atombios.h to latest from kernel 5881 5882commit bb7c77ca75e857f90791b0dd1c04c8e2f19d0e3c 5883Author: Alex Deucher <alexdeucher@gmail.com> 5884Date: Mon Aug 9 22:18:53 2010 -0400 5885 5886 atom: upstream parser update 5887 5888 fixes digital output problems on evergreen asics 5889 5890commit cc0a167ff2db9b097891883ded147af1d67e4407 5891Author: Alex Deucher <alexdeucher@gmail.com> 5892Date: Tue May 25 18:17:15 2010 -0400 5893 5894 r6xx/r7xx: default to micro (1D) tiling for now 5895 5896 SW de-tiling used in the r600c 3D driver has issues with 5897 certain configurations. 5898 5899 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5900 5901commit b5bfdbd70d9671250957ccd41dfc8818850d257e 5902Author: Alex Deucher <alexdeucher@gmail.com> 5903Date: Thu Aug 5 17:26:28 2010 -0400 5904 5905 r6xx/r7xx: add support for tiling with kms (v3) 5906 5907 Requires radeon drm 2.6.0 and updated mesa. 5908 5909 v2: - fix lockup due to not emiting DB_DEPTH_INFO 5910 https://bugs.freedesktop.org/show_bug.cgi?id=28342 5911 - fix drm minor version to deal with evergreen accel 5912 v3: rebase on current ddx git 5913 5914 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 5915 5916commit a3c59c6f6be7067421e348142da0ca13428dcd57 5917Author: Alex Deucher <alexdeucher@gmail.com> 5918Date: Wed Aug 4 12:07:36 2010 -0400 5919 5920 radeon: add new pci ids 5921 5922commit a456587b77ae357750179a50f8db2a17c0f2738e 5923Author: Alex Deucher <alexdeucher@gmail.com> 5924Date: Mon Aug 2 14:24:41 2010 -0400 5925 5926 r6xx/r7xx: move syrface sync emit to the functions that emit surface info 5927 5928 reduces code duplication. 5929 5930commit 8eba977cab1878ba247da8160771d41194d8014f 5931Author: Alex Deucher <alexdeucher@gmail.com> 5932Date: Thu Jun 10 13:34:59 2010 -0400 5933 5934 EXA: move more common helpers to exa_shared 5935 5936commit 71c1a2704af23b61439cee5ce784f7fe267a8a26 5937Author: Alex Deucher <alexdeucher@gmail.com> 5938Date: Thu Jun 10 11:52:43 2010 -0400 5939 5940 remove unused reg headers from radeon_exa_shared.c 5941 5942commit d73aef78919005369af1b60df138439b4b6105c3 5943Author: Alex Deucher <alexdeucher@gmail.com> 5944Date: Mon Aug 2 13:27:12 2010 -0400 5945 5946 r6xx/r7xx: set VGT_MAX_VTX_INDX to a larger value 5947 5948commit dacaf5d827b58c39f9e5a7ac0530f9ea6e257347 5949Author: Dave Airlie <airlied@redhat.com> 5950Date: Mon Aug 2 08:33:51 2010 +1000 5951 5952 fix make distcheck 5953 5954commit c79ce215a01b45fc63b483da167ae37ec7aefad6 5955Author: Dave Airlie <airlied@redhat.com> 5956Date: Sun Aug 1 16:51:48 2010 +1000 5957 5958 radeon/r600: restructure exa + vbo to provide more sharing 5959 5960 This is a precursor for r300/500 vbo support. 5961 5962 Signed-off-by: Dave Airlie <airlied@redhat.com> 5963 5964commit 82254b59268140c4102ae3cd713743ae2be15c00 5965Author: Alex Deucher <alexdeucher@gmail.com> 5966Date: Fri Jul 30 17:15:05 2010 -0400 5967 5968 r6xx/r7xx: unify composite mask and non-mask pixel shader 5969 5970commit 1c17f3a192f644e8e38b5cfb1470f49434bfba27 5971Author: Alex Deucher <alexdeucher@gmail.com> 5972Date: Fri Jul 30 16:34:54 2010 -0400 5973 5974 r6xx/r7xx: clean up composite vertex shader 5975 5976 keep CF, ALU, Fetch instructions in separate groups 5977 5978commit f9d6c0de231357f96e2e0de71e6c9221bcb36bd4 5979Author: Gaetan Nadon <memsize@videotron.ca> 5980Date: Fri Jul 23 13:28:42 2010 -0400 5981 5982 The local copy of the modes code is no longer required. 5983 5984 The server 1.2 as shipped in the tarball on the web does not contain the 5985 modes code. It was added just after and found in git branch server-1.2-branch. 5986 5987 The modes code was initially included in version ati 6.8.0 and fails to compile 5988 with server 1.2 as it requires randr 1.2. The modes code is included in server 5989 versions 1.3 and later, so there is no need to provide an unknown version of 5990 the modes code in the ati driver tarball. It will never be used. 5991 5992 This patch makes the ati driver requiring server 1.3 or later. 5993 Version 6.8.0 configures and builds ok on server 1.3 5994 Master branch post 6.13.1 configures and builds ok on server 1.3 5995 5996 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 5997 5998commit 0028419acb0762eeb950de5fe702c93e70301612 5999Author: Gaetan Nadon <memsize@videotron.ca> 6000Date: Sat Jul 24 10:51:18 2010 -0400 6001 6002 config: add AM_PROG_CC_C_O for per-target compilation flags 6003 6004 Per-target compilation flags (theatre200_drv_la_CFLAGS) are required 6005 when multiple targets which require different compiler flags, 6006 are build in the same makefile. 6007 6008 Automake issues a command with -c and -o flags which not all compilers 6009 support. The object fles are prefixed with theatre200_drv_la. 6010 The macro AM_PROG_CC_C_O must then be used to provide this feature 6011 on compilers that do not have it. If not, a warning is issued at make time. 6012 6013 This macros checks for compiler support and if missing, uses a "compile" 6014 script it generates in the package root directory. 6015 6016 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6017 6018commit aca0a8669b538d58f018f95c9b22e6b3ec1ffe50 6019Author: Gaetan Nadon <memsize@videotron.ca> 6020Date: Wed Jul 21 16:49:04 2010 -0400 6021 6022 config: add comments for main statements 6023 6024commit 66e614f7115efeec237b3b916d9637e8b3e8985c 6025Author: Gaetan Nadon <memsize@videotron.ca> 6026Date: Wed Jul 21 16:07:00 2010 -0400 6027 6028 config: replace deprecated use of AC_OUTPUT with AC_CONFIG_FILES 6029 6030 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6031 6032commit 16e5510c90ef1ba2bbaab78d18943f080b86d809 6033Author: Gaetan Nadon <memsize@videotron.ca> 6034Date: Wed Jul 21 14:37:41 2010 -0400 6035 6036 config: replace deprecated AC_HELP_STRING with AS_HELP_STRING 6037 6038 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6039 6040commit b36d8c09e91382f4cfa71635374ec88f5b676d1c 6041Author: Gaetan Nadon <memsize@videotron.ca> 6042Date: Wed Jul 21 14:05:22 2010 -0400 6043 6044 config: replace deprecated AM_CONFIG_HEADER with AC_CONFIG_HEADERS 6045 6046 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6047 6048commit cd9351b04c2d6982b28c647a63d550eb3e1937eb 6049Author: Alex Deucher <alexdeucher@gmail.com> 6050Date: Wed Jul 21 13:48:24 2010 -0400 6051 6052 r6xx/r7xx: group op variable state 6053 6054 Group the op variable state into one emit block, re-order 6055 to reduce dwords emitted. 6056 6057commit 5f838c664e8010f4e51afecd4100d73a96fe1209 6058Author: Gaetan Nadon <memsize@videotron.ca> 6059Date: Wed Jul 21 09:27:42 2010 -0400 6060 6061 config: complete AC_INIT m4 quoting 6062 6063 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6064 6065commit 48ec2e65c268b426ab9a3e214d174447cf5b5936 6066Author: Gaetan Nadon <memsize@videotron.ca> 6067Date: Wed Jul 21 08:33:38 2010 -0400 6068 6069 config: remove unrequired AC_SUBST for LIBDRM and LIBUDEV 6070 6071 These macros are called by the PKG_CHECK_MODULES macro. 6072 6073 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6074 6075commit 1aabb7eb7d8f06c7481151145db3b9a722ce4ef0 6076Author: Gaetan Nadon <memsize@videotron.ca> 6077Date: Tue Jul 20 21:54:11 2010 -0400 6078 6079 config: remove unrequired AC_SUBST([DRI_CFLAGS]) 6080 6081 This macro is called by PKG_CHECK_MODULES 6082 6083 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6084 6085commit 8f92b349821a3ee5ed8df55273d905b9605385aa 6086Author: Gaetan Nadon <memsize@videotron.ca> 6087Date: Tue Jul 20 21:44:57 2010 -0400 6088 6089 config: remove unrequired AC_SUBST([XORG_CFLAGS]) 6090 6091 This macro is called by PKG_CHECK_MODULES 6092 6093 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6094 6095commit 5b483b832f9c4a5b92ffb7f72a470669201f1fba 6096Author: Gaetan Nadon <memsize@videotron.ca> 6097Date: Tue Jul 20 20:24:42 2010 -0400 6098 6099 config: remove unrequired AC_HEADER_STDC 6100 6101 Autoconf says: 6102 "This macro is obsolescent, as current systems have conforming 6103 header files. New programs need not use this macro". 6104 6105 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6106 6107commit 6574e3a16eb3631ee7e00ee60a8c9ba95c8b84ef 6108Author: Gaetan Nadon <memsize@videotron.ca> 6109Date: Tue Jul 20 19:41:30 2010 -0400 6110 6111 config: remove AC_PROG_CC as it overrides AC_PROG_C_C99 6112 6113 XORG_STRICT_OPTION from XORG_DEFAULT_OPTIONS calls 6114 AC_PROG_C_C99. This sets gcc with -std=gnu99. 6115 If AC_PROG_CC macro is called afterwards, it resets CC to gcc. 6116 6117 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6118 6119commit 593eff2924c2ad161d8b490fbf6d7e433fbe2a80 6120Author: Gaetan Nadon <memsize@videotron.ca> 6121Date: Tue Jul 20 18:45:18 2010 -0400 6122 6123 config: update AC_PREREQ statement to 2.60 6124 6125 Unrelated to the previous patches, the new value simply reflects 6126 the reality that the minimum level for autoconf to configure 6127 all x.org modules is 2.60 dated June 2006. 6128 6129 ftp://ftp.gnu.org/gnu/autoconf/autoconf-2.60.tar.gz 6130 6131 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6132 6133commit c2ab6ffc25aa6759cbbb4c1fbbd4a136b38983bf 6134Author: Gaetan Nadon <memsize@videotron.ca> 6135Date: Tue Jul 20 16:15:29 2010 -0400 6136 6137 config: upgrade to util-macros 1.8 for additional man page support 6138 6139 Use MAN_SUBST now supplied in XORG_MANPAGE_SECTIONS 6140 The value of MAN_SUBST is the same for all X.Org packages. 6141 6142commit cdeb1949c820242f05a8897d3ddd0718f204dacf 6143Author: Jerome Glisse <jglisse@redhat.com> 6144Date: Thu Jul 15 16:21:41 2010 -0400 6145 6146 kms: don't call cursor helper if using software cursor 6147 6148 Fix : 6149 https://bugzilla.redhat.com/show_bug.cgi?id=601713 6150 https://bugzilla.redhat.com/show_bug.cgi?id=598358 6151 6152 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 6153 6154commit 06691376b1ee963c711420edaf5a03eab6f5658f 6155Author: Dave Airlie <airlied@redhat.com> 6156Date: Wed Jul 7 13:15:03 2010 +1000 6157 6158 fix build on non-kms 6159 6160commit 052cf0169ae70d5448af6dc4db840b2fc195569b 6161Author: Dave Airlie <airlied@redhat.com> 6162Date: Wed Jul 7 11:10:46 2010 +1000 6163 6164 configure.ac: bump version post release 6165 6166commit ad999e633ff41d27eed9d2c6535e163a7181b0bd 6167Author: Dave Airlie <airlied@redhat.com> 6168Date: Wed Jul 7 10:49:22 2010 +1000 6169 6170 set version for release 6171 6172commit 37b348059b1c15d7b381cd3df3db52bd9ee6613e 6173Author: Alex Deucher <alexdeucher@gmail.com> 6174Date: Wed Jun 30 12:56:48 2010 -0400 6175 6176 remove rv100 quirk 6177 6178 Some RV100 cards with 2 VGA ports show up with DVI+VGA, however 6179 some boards with DVI+VGA have the same subsystem ids. Better 6180 to have a VGA port show up as DVI than having a non-useable 6181 DVI port. 6182 6183 reported by DHR in irc. 6184 6185 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6186 6187commit c8ea942bd0e9d1c055d50b94440aa4dae425f20b 6188Author: Cooper Yuan <cooperyuan@gmail.com> 6189Date: Tue Jun 29 20:34:57 2010 +0800 6190 6191 Remove HDP_SOFT_RESET function, there is no need to reset HDP block. 6192 6193 This commit can fix an issue reported on DELL server, system gets hang 6194 during soft resetting while another application tries to access PCI 6195 configuration space. 6196 6197commit 139b38bf67ec10d876cc56df833541d497ae4fa4 6198Author: Alex Deucher <alexdeucher@gmail.com> 6199Date: Thu Jun 24 14:55:09 2010 -0400 6200 6201 r6xx shader: use ADDR() for CF_DWORD0 6202 6203 no change in functionality 6204 6205commit 801e83227a59a29eea425ea612083bbf2b536c30 6206Author: Wolfram <bugzilla1@malloc.de> 6207Date: Mon Jun 21 18:59:19 2010 -0400 6208 6209 r6xx/r7xx: fix ums cmd buffer leak 6210 6211 Fixes: 6212 https://bugs.freedesktop.org/show_bug.cgi?id=27957 6213 6214commit b13d719080b75fc6db4d15d2d323b8fce8b7ad06 6215Author: Alex Deucher <alexdeucher@gmail.com> 6216Date: Mon Jun 21 18:49:44 2010 -0400 6217 6218 r6xx/r7xx: fix miscount in state emit 6219 6220commit c3c5c8e2cc91b51a24effdffb85281216eed731d 6221Author: Alex Deucher <alexdeucher@gmail.com> 6222Date: Mon Jun 21 14:30:17 2010 -0400 6223 6224 r6xx/r7xx accel: add back some additional default state 6225 6226 This adds back everything removed in c29157bbf5b0dd26857675282ab094082fbaed0d 6227 except CB_FOG_*, CB_CLEAR_* and the VPORT transforms. Those shouldn't 6228 be needed as we aren't using fog or viewport transforms. We probably don't 6229 need all the state that was added back either but I can't reproduce any 6230 problems here, so it's hard to say which parts are problematic. 6231 6232 Should fix: 6233 https://bugs.freedesktop.org/show_bug.cgi?id=28629 6234 and several corruption reports on #radeon. 6235 6236 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6237 6238commit 800cb2088fec698d0626063a9ab198ff534938c0 6239Author: Michel Dänzer <daenzer@vmware.com> 6240Date: Mon Jun 21 08:15:14 2010 +0200 6241 6242 DRI2: Fix up confusion between windows and pixmaps. 6243 6244 Fixes crashes with xserver master, where looking up a pixmap private on a 6245 window no longer happens to work. 6246 6247commit f7a91ece264af9f3fd2fc18e99aefcda93ce9f5c 6248Author: Michel Dänzer <daenzer@vmware.com> 6249Date: Mon Jun 14 08:52:16 2010 +0200 6250 6251 Convert x(c)alloc/xfree to m/calloc/free. 6252 6253 Fixes deprecation warnings with xserver master and should also work with older 6254 xservers. 6255 6256commit ea37d24b1b6d4cbcf73e680846de25b72af216e3 6257Author: Dave Airlie <airlied@redhat.com> 6258Date: Mon Jun 21 13:55:15 2010 +1000 6259 6260 radeon: fix support for 1.9 server master. 6261 6262 This moves pixmap and mode set into CSR where its allowed. Should work fine on 6263 old servers also. 6264 6265 Signed-off-by: Dave Airlie <airlied@redhat.com> 6266 6267commit fdd8ecafd054f65842351aee6ee6fba7af6613b2 6268Author: Alex Deucher <alexdeucher@gmail.com> 6269Date: Wed Jun 16 19:02:10 2010 -0400 6270 6271 r6xx/r7xx: macro safety fixes 6272 6273commit 4651d77211b508cb6b76931807780e317f232220 6274Author: Alex Deucher <alexdeucher@gmail.com> 6275Date: Wed Jun 16 12:28:36 2010 -0400 6276 6277 radeon: fix depth 16 with ums 6278 6279 Fixes: 6280 https://bugs.freedesktop.org/show_bug.cgi?id=28494 6281 https://bugzilla.redhat.com/show_bug.cgi?id=554967 6282 6283 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6284 6285commit 1e1d6a515428b6884fea586d180346fc74ef75a1 6286Author: Alex Deucher <alexdeucher@gmail.com> 6287Date: Wed Jun 16 12:20:03 2010 -0400 6288 6289 r3xx-r5xx Xv: disable bicubic filtering by default 6290 6291 - makes Xv more consistent with r1xx/r2xx/r6xx/r7xx 6292 - Xv attributes like brightness, contrast, hue, etc. only work 6293 when bicubic is disabled. 6294 - avoids performance issues on some systems when sampling from textures 6295 in GART with kms. 6296 - It can be re-enabled with using xvattr 6297 6298 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6299 6300commit c29157bbf5b0dd26857675282ab094082fbaed0d 6301Author: Alex Deucher <alexdeucher@gmail.com> 6302Date: Tue Jun 15 19:39:42 2010 -0400 6303 6304 r6xx/r7xx: remove unnecessary state emit 6305 6306 No need to emit state that doesn't impact that we use for EXA/Xv. 6307 6308 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6309 6310commit 35280545fcfb911f189d7657bd1040a28450fe7b 6311Author: Alex Deucher <alexdeucher@gmail.com> 6312Date: Tue Jun 15 19:21:32 2010 -0400 6313 6314 r6xx/r7xx: reorg default and clipping state emit 6315 6316 Group ordered registers to save command buffer space. 6317 Reduces the default and clipping state from 256 to 160 dwords. 6318 6319 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6320 6321commit a2528a734c1d4e8639c49e5d222e3630a93ffbfd 6322Author: Alex Deucher <alexdeucher@gmail.com> 6323Date: Tue Jun 15 17:00:34 2010 -0400 6324 6325 r6xx/r7xx accel: remove some duplicate emits and minor clean up 6326 6327 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6328 6329commit 5f093357f18eb9bea641394ab86a92a1766d8f2e 6330Author: Gaetan Nadon <memsize@videotron.ca> 6331Date: Sat Jun 12 15:11:46 2010 -0400 6332 6333 COPYING: replace stub file with actual copyright notices 6334 6335 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 6336 6337commit 0c2118586d3edeecc2473b5d685472df4b5e70fa 6338Author: Dave Airlie <airlied@redhat.com> 6339Date: Fri Jun 11 15:53:18 2010 +1000 6340 6341 radeon: fixup last fix, use CURSOR_WIDTH not RADEON_CURSOR_WIDTH. 6342 6343commit 9a117f768cb8261327fd3c324da9c98875785cc1 6344Author: Dave Airlie <airlied@redhat.com> 6345Date: Fri Jun 11 14:28:24 2010 +1000 6346 6347 radeon: set proper stride for cursor in tiling flags. 6348 6349 definitely incorrect, but hope it doesn't break anything. 6350 6351 Signed-off-by: Dave Airlie <airlied@redhat.com> 6352 6353commit b6346ede94f0d0b11ee04770cf52508cb0a5e6c6 6354Author: Dave Airlie <airlied@redhat.com> 6355Date: Thu Jun 10 20:13:58 2010 -0400 6356 6357 drmmode: fix big endian issue with properties. 6358 6359 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. 6360 6361 Signed-off-by: Dave Airlie <airlied@redhat.com> 6362 6363commit 1971dc6d758dea13b9fb6d1c8f516c165628b5e5 6364Author: Alex Deucher <alexdeucher@gmail.com> 6365Date: Tue Jun 8 11:18:28 2010 -0400 6366 6367 radeon: fix rn50 cloning with kms 6368 6369 Since they only have one crtc sometimes the xserver doesn't assign 6370 a crtc to one of the outputs even though both outputs have common modes 6371 which results in only one monitor being enabled. Assign a crtc in 6372 preinit so that both outputs light up. 6373 6374 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6375 6376commit 426114b4a99d37b394efe3336968bb0ab9b6e9ae 6377Author: Dave Airlie <airlied@redhat.com> 6378Date: Tue Jun 8 11:34:35 2010 +1000 6379 6380 xv: fix Xv on M6/RV100 under KMS. 6381 6382 pRADEONEnt->HasCRTC2 wasn't setup under KMS. 6383 6384 Signed-off-by: Dave Airlie <airlied@redhat.com> 6385 6386commit f64bf0de8e2de7c1bf9cc0c614603dd23c9060ad 6387Author: Alex Deucher <alexdeucher@gmail.com> 6388Date: Thu Jun 3 14:48:21 2010 -0400 6389 6390 evergreen: explicitly disable accel on evergreen 6391 6392 Previously we relied on the drm not having accel enabled 6393 to make sure evergreen used shadowfb, now we when we enable 6394 accel in the drm, we need to make sure the ddx doesn't try 6395 and use it until it's implemented. 6396 6397 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6398 6399commit 428125c095b18c2a2864c1aef24ac0f384b6be54 6400Author: Marek Olšák <maraeo@gmail.com> 6401Date: Sun May 16 20:11:17 2010 +0000 6402 6403 r3xx-r5xx: enable color tiling by default on KMS 6404 6405 Signed-off-by: Marek Olšák <maraeo@gmail.com> 6406 6407commit 6712ce4f4715c8ce7c2fcddb52ca9b103bdd634c 6408Author: Marek Olšák <maraeo@gmail.com> 6409Date: Sun May 16 20:09:31 2010 +0000 6410 6411 rs740: follow macro-switch too 6412 6413 Signed-off-by: Marek Olšák <maraeo@gmail.com> 6414 6415commit 842fa162e9d3bbad2bc44d3732bbc8e5a54402ea 6416Author: Alex Deucher <alexdeucher@gmail.com> 6417Date: Tue May 18 11:21:01 2010 -0400 6418 6419 dri2: use radeon_pick_best_crtc() 6420 6421 It's already exposed and used by exa and Xv and has 6422 the same functionality. radeon_covering_crtc() can 6423 be dropped eventually. 6424 6425commit 64e6858aaf5d0e39ecc9f9804689012f3a38660a 6426Author: Alex Deucher <alexdeucher@gmail.com> 6427Date: Mon May 17 12:50:44 2010 -0400 6428 6429 r2xx exa: fix typo in 27c3326863deae36bc35e2c3b73fffd400208ff1 6430 6431 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6432 6433commit 27c3326863deae36bc35e2c3b73fffd400208ff1 6434Author: Alex Deucher <alexdeucher@gmail.com> 6435Date: Mon May 17 09:53:32 2010 -0400 6436 6437 r1xx/r2xx EXA: handle RepeatNone properly with transforms 6438 6439 Fallback with unsupported ops like r3xx, r6xx. 6440 6441 Should fix fdo bug 28142. 6442 6443 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 6444 6445commit 30591320ec46e491ba20904cc64f3405b51c6505 6446Author: Jerome Glisse <jglisse@redhat.com> 6447Date: Fri May 7 17:35:09 2010 +0200 6448 6449 kms: add support for the MSC swap & sync API 6450 6451 This patch is mostly a port over of Intel ddx code for 6452 MSC support. It needs a radeon KMS module with version 6453 2.4 which has the query for hw crtc id. 6454 6455 Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> 6456 Signed-off-by: Jerome Glisse <jglisse@redhat.com> 6457 6458commit 766024dcc61c83490540910ce752f9bfe6dddba4 6459Author: Marek Olšák <maraeo@gmail.com> 6460Date: Mon May 3 22:56:27 2010 +0200 6461 6462 r3xx-r5xx: fix texturing with small macrotiled pixmaps 6463 6464 Pixmaps smaller than a macrotile cannot be used as textures because hardware 6465 automatically switches to macro-linear and therefore sampled pixels are 6466 messed up. This behavior is known as MACRO_SWITCH. 6467 6468 The only sane workaround seems to be not using macrotiling for small pixmaps. 6469 6470 The function RADEONMacroSwitch has been ported from r300g and implements 6471 MACRO_SWITCH the same way it's implemented in hardware. It's been well tested 6472 in r300g. 6473 6474 This commit also fixes blit-based framebuffer reads, which are used for tiled 6475 surfaces in r300g, when ColorTiling is enabled. 6476 6477 Signed-off-by: Marek Olšák <maraeo@gmail.com> 6478 6479commit 6aedd57f6c2f08d0151a8bd1c5893d40d3db709a 6480Author: Alex Deucher <alexdeucher@gmail.com> 6481Date: Tue May 4 14:30:52 2010 -0400 6482 6483 radeon: fix pll_out_min default for pre-avivo 6484 6485 Fixes fdo bug 27960 6486 6487commit b2ecd1bb507ed4e0d9ba6ebd498c51da2e8136fa 6488Author: Michel Dänzer <daenzer@vmware.com> 6489Date: Fri Apr 23 18:28:22 2010 +0200 6490 6491 FB/MMIO_cnt should be ints, not pointers. 6492 6493 Pointed out by compiler warnings. 6494 6495 https://bugs.freedesktop.org/show_bug.cgi?id=27817 6496 6497commit 1ad41054989f7c3edec373ccf09dceda6d7c94b4 6498Author: Alex Deucher <alexdeucher@gmail.com> 6499Date: Thu Apr 22 16:48:21 2010 -0400 6500 6501 r3xx: fix gb pipe setup for SE cards 6502 6503 Noticed by Tormod Volden 6504 6505commit 59e287d7c484b4addd4a06d013670577639c7ae2 6506Author: Alex Deucher <alexdeucher@gmail.com> 6507Date: Thu Apr 22 02:05:40 2010 -0400 6508 6509 r3xx-r5xx: fix vertex units 6510 6511 Noticed by Tormod Volden. 6512 6513 RV3xx is 2, RV560,RV570 is 8 6514 6515commit 2a6b409496f26da0436972b5feae6ea035dde08d 6516Author: Owain Ainsworth <zerooa@googlemail.com> 6517Date: Mon Apr 19 14:37:33 2010 +0100 6518 6519 Reference count shared driver mappings. 6520 6521 With MMIO it wasn't *such* a bit deal if we leaked the smallish mapping. 6522 with FB it could be a larger deal. So instead of worrying about this, 6523 reference count the mappings in the entity structure and unmap them when 6524 no one cares anymore. 6525 6526 Prompted by a discussion with airlied 6527 6528 Signed-off-by: Dave Airlie <airlied@redhat.com> 6529 6530commit 761f0de5556e46f166280476185977f720efe586 6531Author: Owain G. Ainsworth <zerooa@googlemail.com> 6532Date: Mon Apr 19 04:05:36 2010 +0100 6533 6534 Make consistent use of fbOffset and share fb mappings. 6535 6536 What we were doing previously was mapping the framebuffer for zaphod for 6537 only this driver instances chunk, however, fbOffset was (rightly) set to 6538 the offset into the whole framebuffer we were using. 6539 6540 Since in some cases we did operations on the FB virtual address + 6541 fbOffset (for example zeroing the framebuffer on entervt) we were 6542 actually pissing all over ourselves in those cases. 6543 6544 Fix this by implementing shared fb mappings like we do for MMIO already, 6545 and whenever we wish to refer to our area of FB space we always use 6546 fbOffset. Fixes zaphod for some users on r600 chipsets, my 4870 is still 6547 behaving strangely on screen 0, but I suspect that is another bug. 6548 6549 Once calculation (in PreInitAccel) is now wrong because of this, however 6550 dri on zaphod does now happen so this is irrelavent, add a comment to 6551 that effect. 6552 6553 Signed-off-by: Dave Airlie <airlied@redhat.com> 6554 6555commit 2059d628c5fb03222a62502cc1b5724bf296a7b3 6556Author: Owain G. Ainsworth <zerooa@googlemail.com> 6557Date: Mon Apr 19 04:05:35 2010 +0100 6558 6559 Kill per-context SAREA support. 6560 6561 It has never been used, and since the world is changing it almost 6562 certainly never will be. Good riddance. 6563 6564 Signed-off-by: Dave Airlie <airlied@redhat.com> 6565 6566commit c4b46f56279cc18c8114235ca41cfb41ffede67f 6567Author: Owain G. Ainsworth <zerooa@googlemail.com> 6568Date: Mon Apr 19 04:05:34 2010 +0100 6569 6570 Remove remnants of the ShowCache option. 6571 6572 The rest of it died a long time ago. 6573 6574 Signed-off-by: Dave Airlie <airlied@redhat.com> 6575 6576commit 03b34c5fb12c4df180d4df113d2b654f6afd7230 6577Author: Dave Airlie <airlied@redhat.com> 6578Date: Mon Apr 19 17:58:07 2010 +1000 6579 6580 radeon: add FireMV 2400 pci id. 6581 6582 From RH#581927 - thanks to Carsten Clasohm 6583 6584 Signed-off-by: Dave Airlie <airlied@redhat.com> 6585 6586commit 6c07816025f70e522986614c29c300ea13e5d932 6587Author: Alex Deucher <alexdeucher@gmail.com> 6588Date: Fri Apr 16 18:50:17 2010 -0400 6589 6590 r5xx texvid: deal with large numbers of verts 6591 6592 should fix fdo bug 25884 6593 6594commit 5ee320a6b1b4c65fe592c8ac4d1981799242d59e 6595Author: Alex Deucher <alexdeucher@gmail.com> 6596Date: Fri Apr 16 18:42:56 2010 -0400 6597 6598 r3xx texvid: deal with large numbers of verts 6599 6600 should fix fdo bug 25884 6601 6602commit 47af3f4f266232517486238917d82fc5ca9c82e6 6603Author: Alex Deucher <alexdeucher@gmail.com> 6604Date: Fri Apr 16 20:50:18 2010 -0400 6605 6606 r2xx texvid: deal with large numbers of verts 6607 6608 should fix fdo bug 25884 6609 6610commit 57577d5cd0641b7cad02242478699bcfece59227 6611Author: Alex Deucher <alexdeucher@gmail.com> 6612Date: Fri Apr 16 20:40:56 2010 -0400 6613 6614 r1xx texvid: deal with large numbers of verts 6615 6616 should fix fdo bug 25884 6617 6618commit 31a888e78fa403f2bddacee098a29f36eaa969cb 6619Author: Alex Deucher <alexdeucher@gmail.com> 6620Date: Fri Apr 16 15:52:59 2010 -0400 6621 6622 r5xx textured video: split into prepare and draw functions 6623 6624commit f7ebe4a127965bb9e46d20c60331bcc4dab76fc5 6625Author: Alex Deucher <alexdeucher@gmail.com> 6626Date: Fri Apr 16 15:46:24 2010 -0400 6627 6628 r3xx textured video: split into prepare and draw functions 6629 6630commit 10799b4ce0c18c5a7d92d688549e8e777344f111 6631Author: Alex Deucher <alexdeucher@gmail.com> 6632Date: Fri Apr 16 15:38:49 2010 -0400 6633 6634 r2xx textured video: split into prepare and draw functions 6635 6636commit bd1cf42201a1f918080bf34786a6de327cb31b5f 6637Author: Alex Deucher <alexdeucher@gmail.com> 6638Date: Fri Apr 16 15:31:40 2010 -0400 6639 6640 r1xx textured video: split into prepare and draw functions 6641 6642commit c1136f94b80346065893f8a43c0fbf60c8c9b057 6643Author: Alex Deucher <alexdeucher@gmail.com> 6644Date: Fri Apr 16 15:23:14 2010 -0400 6645 6646 Xv: track vtx_count, is_planar in port priv 6647 6648commit 2f680b631357661511456f07b7b4ba1aa2cbf30c 6649Author: Alex Deucher <alexdeucher@gmail.com> 6650Date: Fri Apr 16 13:43:41 2010 -0400 6651 6652 radeon/kms: fix possible crtc mask for evergreen 6653 6654commit 4656f5dff1ed72fa2c7a1484305f2aef7b65ff2b 6655Author: Alex Deucher <alexdeucher@gmail.com> 6656Date: Fri Apr 16 13:36:25 2010 -0400 6657 6658 radeon/kms: fix crash when using more than two heads 6659 6660 fixes fdo bug 27692 6661 6662commit a69e749d0562887af6bd236c38802472e54640c4 6663Author: Dave Airlie <airlied@redhat.com> 6664Date: Fri Apr 16 13:58:54 2010 +1000 6665 6666 kms: add uevent support. 6667 6668 When we get a hotplug event from the kernel we should notify the client side to reconfigure displays. 6669 6670 based on work by ajax in intel driver. 6671 6672 Signed-off-by: Dave Airlie <airlied@redhat.com> 6673 6674commit 22a46dddd375b2b9399e12fdf168fa5292ff17a4 6675Author: Alex Deucher <alexdeucher@gmail.com> 6676Date: Tue Apr 13 11:08:59 2010 -0400 6677 6678 radeon: add support for pll algo selection 6679 6680 tv-out on atom systems is very particular about it's 6681 dividers. force it to use the old algo. 6682 6683 Should fix fdo bug 27593. 6684 6685commit c1b817c45477c21234abaaebe78feb7ea4fd92b1 6686Author: Alex Deucher <alexdeucher@gmail.com> 6687Date: Mon Apr 12 16:32:04 2010 -0400 6688 6689 atom: disable TV encoder when VGA is in use 6690 6691 Switching between TV and VGA caused problems on some 6692 systems since the TV encoder was left enabled. 6693 6694 Should fix fdo bug 25520. 6695 6696commit 878814aeab7031ff6169fcaf2be869b7595c8699 6697Author: Alex Deucher <alexdeucher@gmail.com> 6698Date: Mon Apr 12 11:16:14 2010 -0400 6699 6700 Xv overlay: set scalerwidth to 1920 on r3xx/r4xx chips 6701 6702 reported on phoronix forums. 6703 6704commit 276fe739f1caa95778213054375a65288b80f320 6705Author: Alex Deucher <alexdeucher@gmail.com> 6706Date: Fri Apr 9 20:03:09 2010 -0400 6707 6708 radeon: add some new pci ids 6709 6710commit eb5665688ef9b52f03f61546351d0848cab54740 6711Author: Alex Deucher <alexdeucher@gmail.com> 6712Date: Thu Apr 8 00:31:52 2010 -0400 6713 6714 radeon: don't setup Xv on rn50 6715 6716 It has no overlay or 3D engine. 6717 6718 fixes fdo bug 27528 6719 6720commit 3c6c6afcbfc6f2c582e13a7c37fe2014b0875b22 6721Author: Alex Deucher <alexdeucher@gmail.com> 6722Date: Mon Apr 5 12:12:56 2010 -0400 6723 6724 change version post release 6725 6726commit fb7911912e60b2cdbc2152b96847775b767ca3ef 6727Author: Alex Deucher <alexdeucher@gmail.com> 6728Date: Mon Apr 5 11:59:24 2010 -0400 6729 6730 bump version for release 6731 6732commit 476a1c6e8b42807b897b8c6f8550ec42b2d5c10e 6733Author: Alex Deucher <alexdeucher@gmail.com> 6734Date: Thu Apr 1 12:42:56 2010 -0400 6735 6736 radeon: fix possible gpio i2c table overrun 6737 6738 GPIO_I2C_INFO does not always have ATOM_MAX_SUPPORTED_DEVICE 6739 entries. Should fix Novell bug 589022. 6740 6741commit 6baa96c44ca93b88acf5233335cee233e59d5af4 6742Author: Dave Airlie <airlied@redhat.com> 6743Date: Thu Apr 1 11:24:32 2010 +1000 6744 6745 r600/exa: further cleanup, use the object struct in the accel state. 6746 6747 This cleans up the accel state objects as well. 6748 6749 Signed-off-by: Dave Airlie <airlied@redhat.com> 6750 6751commit 87f49a24019967552c5478d6af89ba438b4261c8 6752Author: Dave Airlie <airlied@redhat.com> 6753Date: Thu Apr 1 10:52:58 2010 +1000 6754 6755 r600: cleanup wasteful variables. 6756 6757 Just assign directly to the structs. 6758 6759 Signed-off-by: Dave Airlie <airlied@redhat.com> 6760 6761commit 478319e55d52fce241d040912fea952071e3b742 6762Author: Dave Airlie <airlied@redhat.com> 6763Date: Thu Apr 1 10:45:52 2010 +1000 6764 6765 r600: reduce function call overhead. 6766 6767 Create a small accel object that can be used to reduce the amount 6768 of parameters passed to SetAccelState. This can be cleaned up a lot more. 6769 6770 Signed-off-by: Dave Airlie <airlied@redhat.com> 6771 6772commit bc93395b3eb5e3511c1b62af90693269f4fa6e13 6773Author: Alex Deucher <alexdeucher@gmail.com> 6774Date: Wed Mar 31 22:30:24 2010 -0400 6775 6776 r6xx+ EXA: fix swapped domains in kms UTS 6777 6778 Noticed by Dave. 6779 6780 Should fix fdo bug 27284 6781 6782commit 3a33b1a92c328b690bdc1aafa0011e4326109815 6783Author: Owain G. Ainsworth <zerooa@googlemail.com> 6784Date: Tue Mar 30 17:14:02 2010 +0100 6785 6786 Add support for MacModel autodetection on OpenBSD. 6787 6788 This allows radeon-equipped macs to work without an xorg.conf on openbsd. 6789 Patch originally written by Mark Kettenis (kettenis@openbsd.org) 6790 6791commit c7eeda8c3f3514ba95ebf2893fbe124bf526b3df 6792Author: Alex Deucher <alexdeucher@gmail.com> 6793Date: Fri Mar 26 01:38:36 2010 -0400 6794 6795 radeon: add support eDP connectors with and LCD device tag 6796 6797 should fix fdo bug 27322 6798 6799commit ff323e11485fa5030ac278c836bf80a535c22249 6800Author: Alex Deucher <alexdeucher@gmail.com> 6801Date: Fri Mar 26 01:24:16 2010 -0400 6802 6803 radeon: fix name string for eDP 6804 6805commit 53ac06161eb2b8cffb1b88e24a9a21cfd12e5883 6806Author: Alex Deucher <alexdeucher@gmail.com> 6807Date: Tue Mar 23 13:34:38 2010 -0400 6808 6809 r6xx+ EXA/Xv: add a R600SetAccelState function 6810 6811 This moves CS bo checking and alignment checks into 6812 a central location. It also cleans up the code. 6813 6814commit 7a044472dfea7cf05ba4c2b87be30e94e2ae0b62 6815Author: Alex Deucher <alexdeucher@gmail.com> 6816Date: Tue Mar 23 01:27:22 2010 -0400 6817 6818 r6xx+ EXA: always use a temp surface for overlapping copies 6819 6820 The scanline based fallback code was complex and added a lot of 6821 overhead. It also didn't work with kms. 6822 6823commit d33bddaedae81126ab7f0023af1c6443bb0b4c4f 6824Author: Alex Deucher <alexdeucher@gmail.com> 6825Date: Mon Mar 22 18:15:23 2010 -0400 6826 6827 r6xx+ EXA: always use the accel_state state in DoPrepareCopy 6828 6829commit dda3f5a99e7a2dc5d57860f4d07df3498e1e21df 6830Author: Alex Deucher <alexdeucher@gmail.com> 6831Date: Mon Mar 22 18:05:36 2010 -0400 6832 6833 r6xx EXA/Xv: track src/dst domains 6834 6835 Much of the code is shared, so track the src/dst 6836 domains so we make sure the uses consistent domains 6837 for each bo. 6838 6839commit 5c256808cb5fea955eea96ffe9196473715156aa 6840Author: Alex Deucher <alexdeucher@gmail.com> 6841Date: Wed Mar 17 23:47:50 2010 -0400 6842 6843 XAA: disable render accel 6844 6845 It's been reported broken for a while. Should fix 6846 fdo bug 27151, others. 6847 6848commit a0683be5cc082bdbdd3bc4e9b52f39f423650946 6849Author: Dave Airlie <airlied@redhat.com> 6850Date: Thu Mar 18 12:36:25 2010 +1000 6851 6852 radeon: avoid using DRI1 init path on DRI2 driver. 6853 6854 I was playing with multi-seat and found this code, fixed 6855 it up to be sane and more DRI2 like. 6856 6857 Signed-off-by: Dave Airlie <airlied@redhat.com> 6858 6859commit 080a5414593e9b59ed923f26aa6057747b0c868f 6860Author: Alex Deucher <alexdeucher@gmail.com> 6861Date: Tue Mar 16 12:33:39 2010 -0400 6862 6863 kms: fix ums naming compat for DisplayPort 6864 6865commit 819b4015349b5d8c5ffa5f979097599774fce5bb 6866Author: Alex Deucher <alexdeucher@gmail.com> 6867Date: Mon Mar 15 13:47:29 2010 -0400 6868 6869 dump version for rc release 6870 6871commit d3482a947e3731be4ed0b00b4e3079470700dc4c 6872Author: Michael Cree <mcree@orcon.net.nz> 6873Date: Fri Mar 12 22:23:31 2010 +1300 6874 6875 Fix some word accesses in AtomBios to work on all architectures. 6876 6877 The UINT16LE_TO_CPU(), etc., macros are used in the AtomBios code to 6878 fix up endian issues but they do not address bad alignment or assist 6879 architectures that cannot perform hardware byte or word accesses. 6880 This patch inserts use of the ldw_u(), etc., interface of the Xserver 6881 into certain AtomBios accesses to address alignment issues. 6882 6883 This resolves Debian bug 572311, namely that the driver when compiled 6884 for generic Alpha architecture (i.e. doesn't use the byte-word extension) 6885 resulted in no display output on certain Radeon cards. 6886 6887 Signed-off-by: Michael Cree <mcree@orcon.net.nz> 6888 6889commit 488c9fd8300505cc6c0c2f8f0f00849f27cc5d63 6890Author: Alex Deucher <alexdeucher@gmail.com> 6891Date: Mon Mar 15 12:25:57 2010 -0400 6892 6893 r6xx/r7xx: fix domain handling in accel code 6894 6895 Noticed by Pauli and Michel on IRC. 6896 6897 Improves GetImage performace by a factor of ~10. 6898 6899commit 2ace2591d92fb6d3ce7a6453edb04b36a6c49a32 6900Author: Alex Deucher <alexdeucher@gmail.com> 6901Date: Mon Mar 15 12:03:58 2010 -0400 6902 6903 radeon: remove some leftover debugging output 6904 6905commit 67e81c8f17ddde6eba633d2a5aef528e1d598d89 6906Author: Andrzej Hajda <andrzej.hajda@wp.pl> 6907Date: Wed Mar 10 18:19:35 2010 -0500 6908 6909 radeon: add support for pal on legacy IGP chips 6910 6911 Based on my initial non-working patch. 6912 6913 Fixes some element of fdo bug 12007 6914 6915commit 3a44f1cb0d2bb748692b1024003de8ee88ca77a5 6916Author: Alex Deucher <alexdeucher@gmail.com> 6917Date: Tue Mar 9 09:44:01 2010 -0500 6918 6919 atom: i2c gpio fixes 6920 6921 Basically a port of my kms patch. This allows us 6922 to remove some quirks. 6923 6924commit d85cb40f516c67305e818302bec7ee817df4144c 6925Author: Matt Turner <mattst88@gmail.com> 6926Date: Sun Mar 7 14:24:35 2010 -0500 6927 6928 Don't check for Xinerama. 6929 6930 It doesn't seem to be used anywhere, so don't require it. 6931 6932 CC: Jerome Glisse <jglisse@redhat.com> 6933 CC: Alex Deucher <alexdeucher@gmail.com> 6934 CC: Dave Airlie <airlied@redhat.com> 6935 Signed-off-by: Matt Turner <mattst88@gmail.com> 6936 6937commit e7b41f8cb082ed462d29bf3fc440072424cbd852 6938Author: Alex Deucher <alexdeucher@gmail.com> 6939Date: Fri Mar 5 19:16:11 2010 -0500 6940 6941 radeon: disable frac fb div with new pll code 6942 6943 fixes fdo bug 26897 6944 6945commit 14aff767490c253cbcdd411f812e50b91673119e 6946Author: Alex Deucher <alexdeucher@gmail.com> 6947Date: Wed Mar 3 13:31:19 2010 -0500 6948 6949 radeon: add new RS880 pci id 6950 6951commit e6dc886634b38e4a36af7b5f0b23299d5acd7244 6952Author: Dave Airlie <airlied@redhat.com> 6953Date: Tue Mar 2 10:25:15 2010 +1000 6954 6955 radeon: bump configure.ac 6956 6957commit 4975658f05c387b39b3e96a292a7683f17645c2c 6958Author: Dave Airlie <airlied@redhat.com> 6959Date: Sat Feb 27 16:47:19 2010 +1000 6960 6961 pciids: hopefully fix HP 6962 6963commit e76b90b399c3cc0f0998c0209300c46f97505498 6964Author: Alex Deucher <alexdeucher@gmail.com> 6965Date: Fri Feb 26 15:01:28 2010 -0500 6966 6967 rv740: disable dfs workaround for drm 1.32+ 6968 6969 rv740 pipe setup was fixed in 2.6.33 6970 6971commit 2de0af5f5b806f9dbfdb1e9b6a5cf96d9433961d 6972Author: Matt Turner <mattst88@gmail.com> 6973Date: Wed Feb 24 22:46:28 2010 -0500 6974 6975 Use RADEON_ALIGN instead of open coding it. 6976 6977 Also fix some RADEON_ALIGN(x, 63), which would return incorrect results 6978 for odd x. Though this shouldn't happen, it's still not right. You 6979 wouldn't ever write (x + 62) & ~62 which is clearly wrong (and what it 6980 expands to). 6981 6982 CC: Jerome Glisse <jglisse@redhat.com> 6983 CC: Alex Deucher <alexdeucher@gmail.com> 6984 CC: Dave Airlie <airlied@redhat.com> 6985 Signed-off-by: Matt Turner <mattst88@gmail.com> 6986 6987commit c7e81d2f3a372e0d5f751dd0c5091aec2b56d936 6988Author: Matt Turner <mattst88@gmail.com> 6989Date: Wed Feb 24 22:46:27 2010 -0500 6990 6991 Use/define RADEON_GPU_PAGE_SIZE instead of sprinkling 4096 everywhere. 6992 6993 Also, define RADEON_BUFFER_ALIGN in terms of it, and replace some 6994 RADEON_ALIGN(x, RADEON_BUFFER_ALIGN) with RADEON_ALIGN(x, 6995 RADEON_GPU_PAGE_SIZE) since this is really what was intended. 6996 6997 CC: Jerome Glisse <jglisse@redhat.com> 6998 CC: Alex Deucher <alexdeucher@gmail.com> 6999 CC: Dave Airlie <airlied@redhat.com> 7000 Signed-off-by: Matt Turner <mattst88@gmail.com> 7001 7002commit c0a5c9403dff254e1669df606a4193794270aeff 7003Author: Dave Airlie <airlied@redhat.com> 7004Date: Fri Feb 26 19:17:28 2010 +1000 7005 7006 pci: add HP vendor id 7007 7008commit 3970cc5563900f08ba5236b23f765c926616c345 7009Author: Cooper Yuan <cooperyuan@gmail.com> 7010Date: Fri Feb 26 16:14:12 2010 +0800 7011 7012 add more triple-head server ID 7013 7014commit b499eee7506f7478649e0000e9f8fcd00ac9d88d 7015Author: James Le Cuirot <chewi@aura-online.co.uk> 7016Date: Thu Feb 25 21:13:18 2010 -0500 7017 7018 clean up kms zaphod handling 7019 7020 Taken from James patch on bug 24523 7021 7022 agd5f: adapt to the current code 7023 7024commit 433c8617341f5768255826435a2b09afba684f02 7025Author: Alex Deucher <alexdeucher@gmail.com> 7026Date: Thu Feb 25 21:03:13 2010 -0500 7027 7028 update ZaphodHeads option 7029 7030 Now that the screen section mapping is fixed, 7031 Make the option per-instance and allow multiple 7032 outputs to be specified; e.g., DVI-0 and HDMI-0 7033 associated with instance 0 and LVDS and VGA-0 7034 associated with instance 1. 7035 7036commit f3e20b01c77f02bdf3483ceee27a1f00d240c33d 7037Author: James Le Cuirot <chewi@aura-online.co.uk> 7038Date: Thu Feb 25 19:10:19 2010 -0500 7039 7040 zaphod: Fix entity instance assignment. 7041 7042 Each screen needs to have a pointer back to its 7043 device's entity instance, Currently Screen1 is 7044 pointing to Screen0's instance so when it comes 7045 to match up the screen sections with the detected 7046 screens, Screen0 gets chosen in both cases. 7047 7048 See bug 24523 7049 7050commit 1b7e9a2e50f77819b3aff4e37ba39eaec69ff932 7051Author: Alex Deucher <alexdeucher@gmail.com> 7052Date: Tue Feb 23 11:39:55 2010 -0500 7053 7054 radeon: fixes for zaphodheads option 7055 7056 Needed for systems with more than two outputs. 7057 Both KMS and non-KMS zaphod work on systems with 7058 more than two outputs now. 7059 7060commit bd83e5f1c75b8c00b43846c7225568917fc9a8ce 7061Author: Dave Airlie <airlied@redhat.com> 7062Date: Tue Feb 23 19:27:24 2010 +1000 7063 7064 zaphod: hopefully fix kms + zaphod 7065 7066commit 0c3468d812e3790ce03d9e76779ae81e7b7b82d5 7067Author: Alex Deucher <alexdeucher@gmail.com> 7068Date: Mon Feb 22 17:34:51 2010 -0500 7069 7070 AVIVO: set frac fb divider pll flag 7071 7072 matches previous behavior 7073 7074commit 212e152536c4325e6799018891d9aee132681f48 7075Author: Alex Deucher <alexdeucher@gmail.com> 7076Date: Mon Feb 22 17:27:24 2010 -0500 7077 7078 radeon: update new pll algo 7079 7080 - add support for pre-avivo chips 7081 - add support for fixed post/ref dividers 7082 - add support for non-fractional fb dividers 7083 7084 By default avivo chips use the new algo and 7085 pre-avivo chips use the old algo. Use the 7086 "NewPLL" option to toggle between them (set to 7087 TRUE for the new algo, FALSE for the old). 7088 7089commit e68d3a3890fc81c51f2006b5548da1e8756ad2fd 7090Author: Alex Deucher <alexdeucher@gmail.com> 7091Date: Thu Feb 18 14:02:00 2010 -0500 7092 7093 RS600: add connector quirk 7094 7095 System lists DVI port as HDMI. 7096 7097 fixes fdo bug 26605 7098 7099commit a3b730eceb522c7ac1ef3dd6f6c7d773118d03f7 7100Author: Jerome Glisse <jglisse@redhat.com> 7101Date: Thu Feb 18 14:13:48 2010 +0100 7102 7103 r6xx/kms: when reseting BO force default state emission 7104 7105 In KMS world each cs need to fully initialize the 3D engine 7106 when we were erasing cs in IBDiscard for r6xx we weren't 7107 forcing reemission of default state. This patch force this 7108 emission. Fix : 7109 7110 http://bugs.freedesktop.org/show_bug.cgi?id=26603 7111 http://bugzilla.kernel.org/show_bug.cgi?id=15284 7112 7113commit 579cdcf9b4e38c791a497b747a055fc0a07d8dd6 7114Author: Alex Deucher <alexdeucher@gmail.com> 7115Date: Wed Feb 17 12:22:48 2010 -0500 7116 7117 radeon: add ZaphodHeads option 7118 7119 Allows users that want to use zaphod mode to select 7120 which xrandr outputs are assigned to which head. E.g., 7121 7122 Option "ZaphodHeads" "LVDS,VGA-0" 7123 will assign LVDS to the first zaphod driver instance 7124 and VGA-0 to the second instance. 7125 7126commit 47136fa347d1756523239746b4c74cd5278a1118 7127Author: Michel Dänzer <daenzer@vmware.com> 7128Date: Mon Feb 15 16:44:39 2010 +0100 7129 7130 EXA: Fix order of coordinates passed to radeon_pick_best_crtc(). 7131 7132 The wrong order prevented tearing avoidance from working with EXA/DRI2. 7133 7134commit 6e95905f759b844373be62ec513c78153ed00222 7135Author: Michel Dänzer <daenzer@vmware.com> 7136Date: Mon Feb 15 17:02:16 2010 +0100 7137 7138 Allocate BOs for 'video offscreen surfaces' and 'TV-in' in VRAM. 7139 7140 I think these can only really work in VRAM, though right now there's probably 7141 nothing to prevent them from getting evicted... 7142 7143commit 78e7047c5235b09858b66dd3688d39aaa27d7589 7144Author: Pauli Nieminen <suokkos@gmail.com> 7145Date: Mon Feb 15 13:40:37 2010 +0200 7146 7147 Allocate Xv buffers to GTT. 7148 7149 KMS doesn't have acceleration for upload to vram. memcpy/memmove to VRAM 7150 directly is very slow (40M/s in benchmark) which causes visible problems 7151 to video. 7152 7153 Allocating video buffer in GTT will give good performance (350-450M/s) 7154 for memmove operation. This is nice performance boost for Xv under KMS. 7155 7156 Signed-off-by: Pauli Nieminen <suokkos@gmail.com> 7157 7158commit 3ec25e59854b6b03ad763bc374d3475a50f562d8 7159Author: Pauli Nieminen <suokkos@gmail.com> 7160Date: Mon Feb 15 13:45:28 2010 +0200 7161 7162 Use memcpy instead of memove for RADEONSwapCopy. 7163 7164 memcpy has about 25% better performance than memmove when 7165 destination is GTT (wc caching). Changing memmove to memcpy 7166 in SwapCopy will reduce CPU time spent moving data between 7167 host and GPU. 7168 7169 memcpy will specially help Xv when playing high resolution 7170 videos. 7171 7172 The swap path doesn't support overlapping memory copies either 7173 so extra safety is not worth the performance hit in commonly 7174 used code path. 7175 7176 Signed-off-by: Pauli Nieminen <suokkos@gmail.com> 7177 7178 [ Michel Dänzer: Fixed up whitespace ] 7179 7180commit 221ef11b31756deb7134801730e76c040e841f5c 7181Author: Alex Deucher <alexdeucher@gmail.com> 7182Date: Fri Feb 12 14:15:10 2010 -0500 7183 7184 r6xx/r7xx: implement EXA vline support 7185 7186commit 8ad4025affe1fb2f417e3a3031d74f83be5df253 7187Author: Alex Deucher <alexdeucher@gmail.com> 7188Date: Fri Feb 12 11:34:47 2010 -0500 7189 7190 radeon: consolidate crtc selector for vline wait 7191 7192 Use the Xv version as it takes into account the area 7193 covered by the op rather than just picking the largest 7194 crtc area. 7195 7196commit 4f9d1714a7382594b834d446bbe502663f6a2778 7197Author: Alex Deucher <alexdeucher@gmail.com> 7198Date: Fri Feb 5 12:27:51 2010 -0500 7199 7200 r500: fix relocs for Xv planar video 7201 7202 fixes fdo bug 26445 7203 7204commit b7ca1ab11ac0e4e4be5a1e7789ae8633b460a2d4 7205Author: Dave Airlie <airlied@redhat.com> 7206Date: Fri Feb 5 13:37:20 2010 +1000 7207 7208 rs400/rs480: mc idle bit is bit 2 like original radeon 7209 7210 backport from a KMS fix, the rs400/480 mc idle is bit 2 not bit 4. 7211 7212 Signed-off-by: Dave Airlie <airlied@redhat.com> 7213 7214commit 97c387e269672b146a95b45fbef5c8c18e473e36 7215Author: Gaetan Nadon <memsize@videotron.ca> 7216Date: Fri Jan 29 16:08:17 2010 -0500 7217 7218 config: remove dead LINUXDOC macro usage 7219 7220 This module does not generate LINUXDOC documentation. 7221 7222 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7223 7224commit 05fa2dee0c63d56fd523ee1ebab479bcb9f5e89d 7225Author: Alex Deucher <alexdeucher@gmail.com> 7226Date: Thu Feb 4 13:13:58 2010 -0500 7227 7228 evergreen: minor clean up in transmitter setup 7229 7230 both link and encoder must be set for linkb 7231 7232commit 76eea5e02776f7beeb8f4491a374bd36792eda92 7233Author: Alex Deucher <alexdeucher@gmail.com> 7234Date: Thu Feb 4 10:52:14 2010 -0500 7235 7236 r600: reduce number of cache flushes 7237 7238 We don't need to flush so often. Next step 7239 would be to move the flushing to the drm and only 7240 flush after each command buffer rather than each 7241 draw. 7242 7243commit 8d63d70f7ebaf9d250f0449d3720ef47516c05df 7244Author: Alex Deucher <alexdeucher@gmail.com> 7245Date: Wed Feb 3 15:22:22 2010 -0500 7246 7247 evergreen: use external clock source for DP PHY 7248 7249 DP CRTC clock always comes from DCPLL. This frees 7250 up PPll1/2 for non-DP-mode PHYs and CRTCs 7251 7252commit ecbc26431914216a8b207e81451282ea07c8b92f 7253Author: Alex Deucher <alexdeucher@gmail.com> 7254Date: Wed Feb 3 03:52:07 2010 -0500 7255 7256 evergreen: blank/unblank DP in dpms calls 7257 7258commit 710a2fd07235349f084ec40626cc28fbae523d27 7259Author: Alex Deucher <alexdeucher@gmail.com> 7260Date: Wed Feb 3 03:00:24 2010 -0500 7261 7262 evergreen: pll fixes 7263 7264 SetPixelClock needs to be called for DP outputs with a 7265 valid PLL. 7266 7267commit 9ab2377ee23c5dc360ba135a77aa6b181a1862e6 7268Author: Alex Deucher <alexdeucher@gmail.com> 7269Date: Tue Feb 2 15:15:36 2010 -0500 7270 7271 evergreen: DP requires coherent mode 7272 7273commit 634da80056525a48ec17ffb81e2fb59b168bea9e 7274Author: Alex Deucher <alexdeucher@gmail.com> 7275Date: Tue Feb 2 15:09:57 2010 -0500 7276 7277 evergreen: fix units on frac_fb_div 7278 7279commit a887818f491f6c7315c56c4e0d0b702c4c6aa4ac 7280Author: Alex Deucher <alexdeucher@gmail.com> 7281Date: Mon Feb 1 11:01:47 2010 -0500 7282 7283 evergreen: add pci ids 7284 7285commit cb2772b69480268c059766c4f6b209ce590ede0e 7286Author: Alex Deucher <alexdeucher@gmail.com> 7287Date: Fri Jan 29 12:59:46 2010 -0500 7288 7289 evergreen: add atombios crtc/pll functions 7290 7291commit bd8e04cb7b39f38b6958273582a9b324a9f0759a 7292Author: Alex Deucher <alexdeucher@gmail.com> 7293Date: Mon Feb 1 10:07:43 2010 -0500 7294 7295 evergreen: add atom support for digital outputs 7296 7297 analog is already supported by the existing code. 7298 7299commit 8ad40d3c32ad5b91725bd37fcade6bed504df421 7300Author: Alex Deucher <alexdeucher@gmail.com> 7301Date: Tue Jan 26 15:39:44 2010 -0500 7302 7303 evergreen: add crtc set base/format support 7304 7305commit 2c9cf0a07ff9e5a4989861bc2fdfd71d841013a8 7306Author: Alex Deucher <alexdeucher@gmail.com> 7307Date: Thu Dec 10 15:29:14 2009 -0500 7308 7309 evergreen: add lut support 7310 7311commit 65246545c3dd9bfef73e8a0f200bd5909b89a167 7312Author: Alex Deucher <alexdeucher@gmail.com> 7313Date: Tue Jan 26 12:10:02 2010 -0500 7314 7315 evergreen: add hw cursor support 7316 7317commit c05cad56b69d239fa2e69905d15f4f08b9db4c55 7318Author: Alex Deucher <alexdeucher@gmail.com> 7319Date: Thu Jan 28 01:28:52 2010 -0500 7320 7321 evergreen: add base asic support 7322 7323commit ed63e1b1abe8810b5da6b4140892337eef08a9ea 7324Author: Alex Deucher <alexdeucher@gmail.com> 7325Date: Fri Jan 29 12:42:33 2010 -0500 7326 7327 evergreen: add register and utility defines 7328 7329commit 019260ec4a9f9fbf2ac63a2ca3314aa308471f7e 7330Author: Alex Deucher <alexdeucher@gmail.com> 7331Date: Thu Dec 10 14:57:39 2009 -0500 7332 7333 evergreen: add chip enums 7334 7335commit e86ae564d8b94d2cc8d3c7cceb452a197ab6387a 7336Author: Alex Deucher <alexdeucher@gmail.com> 7337Date: Mon Feb 1 12:03:23 2010 -0500 7338 7339 atom: update to the latest upstream atombios.h changes 7340 7341commit 77b13a02c70842a58e0590d0243f0ae016c5a640 7342Author: Alex Deucher <alexdeucher@gmail.com> 7343Date: Mon Feb 1 12:16:20 2010 -0500 7344 7345 r500: fix CS section size missmatch in Xv code 7346 7347commit 5c11264517d1b9808415d47e9dedc345610f0f86 7348Author: Dave Airlie <airlied@linux.ie> 7349Date: Fri Jan 29 15:10:33 2010 +1000 7350 7351 radeon: further digital encoder cleanups. 7352 7353 this just cleans up a few more bits of the digital encoder setup. 7354 7355 Signed-off-by: Dave Airlie <airlied@redhat.com> 7356 7357commit 872a11fa9efb0b76cef83089ea1bb80792b94930 7358Author: Alex Deucher <alexdeucher@gmail.com> 7359Date: Thu Jan 28 01:19:23 2010 -0500 7360 7361 radeon: clean up dig encoder handling 7362 7363 - switch the var name to dig_encoder 7364 - quiet coherent messages 7365 - clean up dig encoder selection 7366 7367 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7368 7369commit c1a73ff51b371ecc6f57a81a33f644344b4e3f70 7370Author: Dave Airlie <airlied@linux.ie> 7371Date: Thu Jan 28 12:58:29 2010 +1000 7372 7373 radeon: add support for picking the digitial encoder 7374 7375 Testing on a W500 laptop found the LVDS + DisplayPort would conflict 7376 over the choice of digital encoder, and only one would work. 7377 7378 This patch fixes the DCE3 case where LVTMA requires the second digital 7379 encoder and lets DP/DVI use the first in most cases. 7380 7381 Signed-off-by: Dave Airlie <airlied@redhat.com> 7382 7383commit 95b56275d80dcee48a7927df124602c34cd72235 7384Author: Alex Deucher <alexdeucher@gmail.com> 7385Date: Wed Jan 27 15:48:25 2010 -0500 7386 7387 avivo: add some hotplug detect regs 7388 7389commit 5a4327f7784361933484895c9af751ccfa242d48 7390Author: Alex Deucher <alexdeucher@gmail.com> 7391Date: Tue Jan 26 16:06:00 2010 -0500 7392 7393 radeon: minor pll updates 7394 7395 add new fixed post divider option 7396 7397commit cbd642af7c76469d701471daea0c8d167567ccf8 7398Author: Alex Deucher <alexdeucher@gmail.com> 7399Date: Tue Jan 26 12:03:45 2010 -0500 7400 7401 radeon: use mmio bar size rather than hardcoded number for register ops 7402 7403 newer asics have larger mmio bars 7404 7405commit e5933fd74923a07aa401bdb4ecd8d1d641148440 7406Author: Alex Deucher <alexdeucher@gmail.com> 7407Date: Mon Jan 25 13:18:47 2010 -0500 7408 7409 KMS: fix no accel option 7410 7411 NO_ACCEL wasn't handled correctly with KMS. 7412 7413 fixes fdo bug 26198 7414 7415commit ba99a82ced1a6101365876723d22bcfec4607537 7416Author: Dave Airlie <airlied@redhat.com> 7417Date: Mon Jan 25 13:50:13 2010 +1000 7418 7419 radeon/kms: only enable EXA pixmaps under kms if we support render accel 7420 7421 I've seen RN50s with 64MB of RAM that are slow as molasses, this 7422 should fix them. 7423 7424 RH bug 556400 7425 7426 Signed-off-by: Dave Airlie <airlied@redhat.com> 7427 7428commit de40d36264364434541377c7f65f3836dc514fc0 7429Author: Kusanagi Kouichi <slash@ac.auone-net.jp> 7430Date: Fri Jan 22 17:18:51 2010 +0900 7431 7432 Remove unnecessary checks. 7433 7434 Signed-off-by: Kusanagi Kouichi <slash@ac.auone-net.jp> 7435 7436commit 30a19b75cc82b8e04c45e6684b84f9a4ccc0505b 7437Author: Alan Coopersmith <alan.coopersmith@sun.com> 7438Date: Tue Jan 19 14:54:04 2010 -0800 7439 7440 Fix configure with --disable-dri 7441 7442 AM_CONDITIONAL calls can't be made inside if statements, since 7443 automake won't know what to do when the if isn't true. 7444 7445 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 7446 7447commit bd011784c0609f1fd03dcb784744c8f5cdef33da 7448Author: Alex Deucher <alexdeucher@gmail.com> 7449Date: Tue Jan 19 11:59:38 2010 -0500 7450 7451 ATOM: Upstream parser updates 7452 7453commit 95d63e408cc88b6934bec84a0b1ef94dfe8bee7b 7454Author: Jerome Glisse <jglisse@redhat.com> 7455Date: Mon Jan 18 10:46:50 2010 +0100 7456 7457 r6xx/r7xx: emit relocation for FRAG & TILE buffer 7458 7459 FRAG & TILE buffer are unused but still they need 7460 to be associated with a valid relocation so that 7461 userspace can't try to abuse them to overwritte 7462 GART and then try to write anywhere in system 7463 memory. 7464 7465commit bbaf71fb46e3e0acdadcb91e9b1c73396bd9aa8e 7466Author: Alan Coopersmith <alan.coopersmith@sun.com> 7467Date: Fri Jan 15 15:09:27 2010 -0800 7468 7469 Update Sun license notices to current X.Org standard form 7470 7471 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 7472 7473commit 3d158716a60d5a113a541ea2f680b81a1be41ad2 7474Author: Dave Airlie <airlied@redhat.com> 7475Date: Thu Jan 14 11:21:33 2010 +1000 7476 7477 displayport: fix DDC on DVI->DP convertors. 7478 7479 Seems we have to not do auxch DDC if we aren't talking to a DP 7480 sink. 7481 7482commit 6a363f68415d37c302151581f2a86855dba39b67 7483Author: Alex Deucher <alexdeucher@gmail.com> 7484Date: Mon Jan 11 15:13:45 2010 -0500 7485 7486 rv100: fix typo in fix for bug 25992 7487 7488 Noticed by Maarten Maathuis. 7489 7490commit 7b01e1ee29f681bf1735ecded6445d12beeb52d8 7491Author: Alex Deucher <alexdeucher@gmail.com> 7492Date: Mon Jan 11 12:40:06 2010 -0500 7493 7494 rv100: reject modes >135 Mhz with DVI 7495 7496 Due to heat issues. fixes bug 25992 7497 7498 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7499 7500commit 74da9a66a420a40bb5d562f07e97ce6ac2eb010f 7501Author: Alex Deucher <alexdeucher@gmail.com> 7502Date: Fri Jan 8 16:51:43 2010 -0500 7503 7504 radeon/kms: add new strings for eDP and TV 7505 7506commit 61977f61f516b54d1fd9df34d0f0892fbf5227d7 7507Author: Alex Deucher <alexdeucher@gmail.com> 7508Date: Thu Jan 7 02:15:31 2010 -0500 7509 7510 radeon: fallback to i2c for edid if aux fails 7511 7512 In case of a DP->DVI adapter for example. 7513 7514commit 958d073869404f60e56dc0cc70b3e7de85904694 7515Author: Alex Deucher <alexdeucher@gmail.com> 7516Date: Thu Jan 7 02:12:00 2010 -0500 7517 7518 radeon: add initial support for eDP (embedded DisplayPort) 7519 7520 Should fix fdo bug 25931 7521 7522commit 944ae73f7cd5f9a1622b38e06a58565812cb102f 7523Author: Alex Deucher <alexdeucher@gmail.com> 7524Date: Thu Jan 7 01:52:35 2010 -0500 7525 7526 radeon: updated ObjectID.h 7527 7528commit 48aa5064aff4b9adf768e480df2312d4375e9c40 7529Author: Alex Deucher <alexdeucher@gmail.com> 7530Date: Mon Jan 4 11:05:44 2010 -0500 7531 7532 radeon: fix LVDS power sequence on Mac cards 7533 7534 Noticed by John R. Dunning. Fix taken from radeonfb. 7535 7536 I'm not sure if this sequence would be useful on any PC 7537 laptops or not so make it mac specific for now. 7538 7539commit 36bd69affc996c92c40b7360a7fbaa1a3a46abfd 7540Author: Gaetan Nadon <memsize@videotron.ca> 7541Date: Wed Dec 16 15:52:18 2009 -0500 7542 7543 configure.ac: remove unused sdkdir=$(pkg-config...) statement 7544 7545 The sdkdir variable isn't use, so remove the statement. 7546 7547 Acked-by: Dan Nicholson <dbn.lists@gmail.com> 7548 7549 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7550 7551commit 9d0f3af7278dc939fd4e6f3ea69d9f488a9fbed7 7552Author: Alex Deucher <alexdeucher@gmail.com> 7553Date: Tue Dec 22 08:45:27 2009 -0500 7554 7555 radeon: add cvt timing if we only have panel w/h 7556 7557 fixes mac laptops without an edid 7558 7559 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7560 7561commit 6e1f5553c6d7e3b5d089af2e3d587efe95936855 7562Author: Alex Deucher <alexdeucher@gmail.com> 7563Date: Mon Dec 21 17:32:40 2009 -0500 7564 7565 ATOM: add new power table defs 7566 7567commit 4b05c47ac657f9a93d76221269761ed64c81f716 7568Author: Dave Airlie <airlied@redhat.com> 7569Date: Thu Dec 17 14:30:23 2009 +1000 7570 7571 radeon: use new libdrm api 7572 7573commit 1d620800bd79290967fa487decf798e318c45f25 7574Author: Ingmar Vanhassel <ingmar@exherbo.org> 7575Date: Sun Dec 20 21:04:56 2009 +0100 7576 7577 Replace shave with automake silent-rules 7578 7579 Which are enabled by default in Xorg's util-macros 1.3 if automake-1.11 7580 is used. 7581 7582 Signed-off-by: Ingmar Vanhassel <ingmar@exherbo.org> 7583 7584commit 52f6a2ff682ad1f8e26ef1cd7d380493dcd57752 7585Author: Gaetan Nadon <memsize@videotron.ca> 7586Date: Wed Dec 16 15:29:53 2009 -0500 7587 7588 src/Makefile.am: missing 3 header files in tarball 7589 7590 radeon_dummy_bufmgr.h 7591 radeon_vbo.h 7592 simple_list.h 7593 Replace leading spaces with tab for radeon_atomwrapper.h 7594 7595 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 7596 7597commit 3a30210d50b27f8772fc5045133940246764fce9 7598Author: Matthias Hopf <mhopf@suse.de> 7599Date: Tue Dec 15 10:53:48 2009 -0500 7600 7601 fix 200M freezes on VT switch if CRTC is disabled 7602 7603 It appears that RS4xx chips need to have the crtc 7604 enabled when the timing is programmed. 7605 7606 agd5f: minor fixes/cleanup of the original patch 7607 7608commit 299d395bd3f294239dee58ab7d607d7d2c657f61 7609Author: Corbin Simpson <MostAwesomeDude@gmail.com> 7610Date: Sun Dec 13 14:51:17 2009 -0800 7611 7612 dri: be more useful if disabling DRI due to KMS/UMS mismatch. 7613 7614 It's only fair. 7615 7616 Compile-tested only. 7617 7618 Signed-off-by: Dave Airlie <airlied@redhat.com> 7619 7620commit 0e5c9d87b5d7e0751df71cc8958ca5ccaed25104 7621Author: Alex Deucher <alexdeucher@gmail.com> 7622Date: Thu Dec 10 14:27:43 2009 -0500 7623 7624 ATOM: split set base/format into separate functions 7625 7626 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7627 7628commit 46630da5fd6f45bb8ea150b870162997480d69c7 7629Author: Alex Deucher <alexdeucher@gmail.com> 7630Date: Wed Dec 9 12:55:25 2009 -0500 7631 7632 AVIVO: add new PLL code 7633 7634 This should hopefully help the problems with flickering 7635 and blinking monitors reported on some systems. If there 7636 are problems, the old PLL algorithm can be selected with: 7637 Option "NewPLL" "FALSE" 7638 in the device section of your X config. 7639 7640 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7641 7642commit f082b1693d6f7f763ccf5a8436a89890ca2c6129 7643Author: Matthijs Kooijman <matthijs@stdin.nl> 7644Date: Thu Dec 3 12:10:14 2009 -0500 7645 7646 radeon: fix crtc2 dpms 7647 7648 noticed by Matthijs Kooijman on fdo bug 22140 7649 7650commit 3a96fbf3b84522cf1ba1b176e82b662222331c14 7651Author: Alex Deucher <alexdeucher@gmail.com> 7652Date: Wed Dec 2 18:02:33 2009 -0500 7653 7654 radeon: only read RADEON_PPLL_REF_DIV for ref div on pre-avivo 7655 7656 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7657 7658commit f03450796d2e9247a1228c4e2abb1dfad7aecddf 7659Author: Kusanagi Kouichi <slash@ac.auone-net.jp> 7660Date: Wed Dec 2 01:36:37 2009 -0500 7661 7662 radeon: Lift hardcoded limit from RADEONQueryImageAttributes 7663 7664 The dimension of an XvImage is limited to 2048 x 2048 even if an adaptor 7665 supports larger image. 7666 XvCreateImage and XvShmCreateImage lower the width or height of an image. 7667 XvPutImage and XvShmPutImage return BadValue. 7668 The cause is that 2048 is hardcoded in RADEONQueryImageAttributes. 7669 7670commit efbc2c80ab02879edf3b7b3d65b16c45ddce5017 7671Merge: 88a50a3 a612813 7672Author: Dave Airlie <airlied@redhat.com> 7673Date: Wed Dec 2 14:30:36 2009 +1000 7674 7675 Merge remote branch 'origin/displayport' 7676 7677commit 88a50a30df11a06263209340a42251851f8e2334 7678Author: Dave Airlie <airlied@redhat.com> 7679Date: Mon Nov 30 15:32:12 2009 +1000 7680 7681 r600: fix multi-operation in single batch support. 7682 7683 This ports the mesa DMA buffer handling with the 3 lists, 7684 7685 Signed-off-by: Dave Airlie <airlied@redhat.com> 7686 7687commit b2597deea3a3953ff50d54ff37e3c043eac409f4 7688Author: Kusanagi Kouichi <slash@ac.auone-net.jp> 7689Date: Mon Nov 30 11:10:50 2009 -0500 7690 7691 R600: fix some warnings 7692 7693commit 4d90dc3cb248e9d61c0c490bba80f6727fafd318 7694Author: Alex Deucher <alexdeucher@gmail.com> 7695Date: Mon Nov 30 11:05:38 2009 -0500 7696 7697 atom: pull misc mode info for lvds panel mode from bios tables 7698 7699 sync polarity, etc. This will likely fix LVDS problems 7700 on some laptops. 7701 7702 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7703 7704commit a612813c5f329f00a9271e7bb69abd2630a49403 7705Author: Alex Deucher <alexdeucher@gmail.com> 7706Date: Fri Nov 27 17:20:30 2009 -0500 7707 7708 radeon: clean up DP code 7709 7710 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7711 7712commit a8dbf7c23481501987971a9c0b6cb0760f86127f 7713Author: Dave Airlie <airlied@redhat.com> 7714Date: Fri Nov 27 10:16:00 2009 +1000 7715 7716 r600/accel: cleanup and merge a lot of the accel functions 7717 7718 All of the drawing ops were the exact same modulo the vtx size, 7719 this along with the vertex buffer wrapping code could all be consolidated 7720 into a smaller set of functions. 7721 7722 This also adds 2 VBO which we switch between, and merges a #define to 7723 enable the multiple operations in one CS under KMS mode. 7724 7725 Multi-operation still isn't working though. 7726 7727 Signed-off-by: Dave Airlie <airlied@redhat.com> 7728 7729commit 971e463b2cfdc1ba1c78a24fb439d33dd9155dfc 7730Author: Alex Deucher <alexdeucher@gmail.com> 7731Date: Thu Nov 26 12:45:15 2009 -0500 7732 7733 dce3: CV1/TV1OutputControl tables didn't go away until dce3.2 7734 7735 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7736 7737commit 95385f0906f371a1ed6e60eb3e597e699c7a3222 7738Author: Jerome Glisse <jglisse@redhat.com> 7739Date: Thu Nov 26 12:21:21 2009 +0100 7740 7741 kms: Fix resizing when acceleration is disabled 7742 7743 When acceleration is disabled we need to reallocate 7744 a new shadow framebuffer and we should also avoid 7745 calling any EXA function as EXA is disabled in such 7746 case. 7747 7748commit 8195385d3cedf70d89ca576371af6bca5045bbd2 7749Author: Dave Airlie <airlied@redhat.com> 7750Date: Thu Nov 26 15:43:54 2009 +1000 7751 7752 dce3: no CV1 or TV1 Output Control tables anymore. 7753 7754 According to the DCE3 docs we should only use DAC1/2 not 7755 CV1/TV OutputControl, also my rv730 bios doesn't have any 7756 CV1/TV tables. 7757 7758 Signed-off-by: Dave Airlie <airlied@redhat.com> 7759 7760commit 381b14d5422131ea8ff942ab06a328f6beca0d3f 7761Author: Dave Airlie <airlied@redhat.com> 7762Date: Thu Nov 26 15:28:15 2009 +1000 7763 7764 atombios: CV mode retrival was broken. 7765 7766 a) crev and frev are reversed 7767 b) my rv730 bios only has one mode in it, so 7768 bounds check the table size. 7769 7770commit a67a13c315a9235be052662e2a3270686cccc0f0 7771Author: Dave Airlie <airlied@redhat.com> 7772Date: Thu Nov 26 13:41:36 2009 +1000 7773 7774 radeon: fix typo in unused as of yet CV code 7775 7776commit cd9bfb5892eebef6dff14ac414dff074db5ddee4 7777Author: Dave Airlie <airlied@redhat.com> 7778Date: Thu Nov 26 06:43:13 2009 +1000 7779 7780 fixes displayport for me here 7781 7782commit 256de862a3c017dcb4ed0d8689294a552eda0d18 7783Author: Dave Airlie <airlied@redhat.com> 7784Date: Thu Nov 26 06:35:39 2009 +1000 7785 7786 displayport: cleanup call dp from dpms on instead. 7787 7788commit 05551295c5e0946745163f17e5c1d3d41b94bcbf 7789Author: Kristian Høgsberg <krh@bitplanet.net> 7790Date: Tue Nov 24 21:09:03 2009 -0500 7791 7792 dri2: Use drmGetDeviceNameFromFd() instead of open coded loop. 7793 7794commit 0061c4db1d3aecdca13efb5133ab8784dc37df95 7795Author: Dave Airlie <airlied@redhat.com> 7796Date: Wed Nov 25 15:39:10 2009 +1000 7797 7798 Revert "r600: enable multiple operations in one CS" 7799 7800 turn this off for now, on my rv635 desktop, I started getting blanks 7801 in places in firefox and the odd bit of font corruption, need to 7802 track that down. 7803 7804 This reverts commit 985a065518b1d33599de33f7fe082d3302db58a6. 7805 7806commit 8b28534bcf877557a5681fa3b4f107c74615d3de 7807Author: Dave Airlie <airlied@redhat.com> 7808Date: Wed Nov 25 13:24:44 2009 +1000 7809 7810 radeon/exa: change option to determine exa pixmap usage. 7811 7812 This moves to a boolean instead of using VRAM sizing. 7813 7814 as per Michel's suggestions on list. 7815 7816 Signed-off-by: Dave Airlie <airlied@redhat.com> 7817 7818commit f0acb16fa19844453adc6db3399977fba7c8a0db 7819Author: Dave Airlie <airlied@redhat.com> 7820Date: Wed Nov 25 12:04:47 2009 +1000 7821 7822 r600/xv: drop inited 3d false in xv code 7823 7824commit 985a065518b1d33599de33f7fe082d3302db58a6 7825Author: Dave Airlie <airlied@redhat.com> 7826Date: Wed Nov 25 11:54:08 2009 +1000 7827 7828 r600: enable multiple operations in one CS 7829 7830 This switches on multiple ops in a single CS under KMS/DRI2. 7831 7832 It gets for on a Pentium D 3 + rv740 from 330,000 to 500,000 7833 with x11perf -aa10text. 7834 7835 It also knocks a couple of seconds of gtkperf -a 7836 7837 Signed-off-by: Dave Airlie <airlied@redhat.com> 7838 7839commit af816ac752820255f245793b53a7cca5a4a49cd4 7840Author: Dave Airlie <airlied@redhat.com> 7841Date: Wed Nov 25 11:53:07 2009 +1000 7842 7843 r600: fixup problems with EXA operation reset for multiple ops 7844 7845 To put multiple ops into one CS, you can't just discard the whole 7846 IB. This add supports for reset the CS cdw to the correct place 7847 after an op discards. 7848 7849 Still doesn't enable the final accel bits. 7850 7851commit 3d8dcbc29323a3c644100bec13aa93f024653bd3 7852Author: Dave Airlie <airlied@redhat.com> 7853Date: Wed Nov 25 11:43:57 2009 +1000 7854 7855 kms: flush cs on close screen. 7856 7857 this is needed for server recycle. 7858 7859 Signed-off-by: Dave Airlie <airlied@redhat.com> 7860 7861commit 19f1a357944f9d8f4567a2691a68067ec033ccb7 7862Author: Dave Airlie <airlied@redhat.com> 7863Date: Wed Nov 25 11:41:52 2009 +1000 7864 7865 radeon: fix check for no work in operation 7866 7867commit 3a460a14b9603159f10d89da27b559c36a184e27 7868Author: Dave Airlie <airlied@redhat.com> 7869Date: Wed Nov 25 10:33:17 2009 +1000 7870 7871 r600: refactor code to help future acceleration speedups. 7872 7873 This changes the vertex buffer index to be an offset, and 7874 records the start of the vb for each operation and uses 7875 that to set the operations up. 7876 7877 This still flushes after each operation to make sure we have 7878 no regressions in non-kms/kms cases. 7879 7880 Signed-off-by: Dave Airlie <airlied@redhat.com> 7881 7882commit 797a3f0c71c94477eec565ea2c95553c6f66d9fd 7883Author: Dave Airlie <airlied@redhat.com> 7884Date: Wed Nov 25 10:07:59 2009 +1000 7885 7886 kms: allow prepare copy to fail without dying 7887 7888commit ba76acc76e0d0a51a1c488e4d5494a71e155a05b 7889Author: Dave Airlie <airlied@redhat.com> 7890Date: Tue Nov 24 15:38:47 2009 +1000 7891 7892 radeon/kms: drop special r600 indirect flush 7893 7894 This just merges it with the main kms cs flush 7895 7896commit 3c527d752d3cced219c6113acb932511b16e35fb 7897Author: Dave Airlie <airlied@redhat.com> 7898Date: Tue Nov 24 15:11:12 2009 +1000 7899 7900 r600: enable mixed pixmaps 7901 7902commit b9eb2380df5145c85b2ef694a04628c055c47429 7903Author: Alex Deucher <alexdeucher@gmail.com> 7904Date: Mon Nov 23 12:12:02 2009 -0500 7905 7906 DCE3.2: fix uniphy2 dvi issues 7907 7908 In some cases the atom transmitter table sets the 7909 golden value of this reg differently which some monitors 7910 don't like. I haven't had time to dig further, so this 7911 works around it for now. 7912 7913 Fixes fdo bug 24313 7914 7915 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7916 7917commit d0dd5122d8d75b7c4a72df07dc7f562b2e75dd5e 7918Author: Gaetan Nadon <memsize@videotron.ca> 7919Date: Mon Nov 23 09:25:05 2009 -0500 7920 7921 Makefile.am: add ChangeLog and INSTALL on MAINTAINERCLEANFILES 7922 7923 Now that the INSTALL file is generated. 7924 Allows running make maintainer-clean. 7925 7926commit 3d7888aa0493f46d826f4d80dea36f1f642f3c6b 7927Author: Dave Airlie <airlied@redhat.com> 7928Date: Mon Nov 23 15:26:15 2009 +1000 7929 7930 radeon: fix warning in printf 7931 7932commit 3a1a8b7b73424a769baf917bba2223ea6515b596 7933Author: Dave Airlie <airlied@redhat.com> 7934Date: Mon Nov 23 15:07:10 2009 +1000 7935 7936 radeon: surround mixed pixmaps usage with ifdef check for old servers 7937 7938commit e20af9c94982ec6487dae42c141d33cf6a7a2eb4 7939Author: Dave Airlie <airlied@redhat.com> 7940Date: Fri Nov 20 12:15:02 2009 +1000 7941 7942 radeon: avoid using hw pixmaps when we have little VRAM. 7943 7944 This patch returns NULL for pixmap creation when we are using 7945 mixed pixmaps and the pixmap has a size. 7946 7947 The size check is necessary for the front buffer. 7948 7949 We add a flag to force pixmap creation for certain pixmaps 7950 that need to be hw, like the DRI2 and Xv ones. 7951 7952 Idea from Michel and workarounds from Ben Skeggs. 7953 7954 v2: add Option "EXALowVRAM" to allow configuring this, value in MBs. 7955 7956 Signed-off-by: Dave Airlie <airlied@redhat.com> 7957 7958commit 543338842ba7fe7da4902947d386ea1cabd1cdf0 7959Author: Alex Deucher <alexdeucher@gmail.com> 7960Date: Fri Nov 20 16:38:43 2009 -0500 7961 7962 dp fix timing 7963 7964 aux channel atom command table delay is 10 usec units 7965 this gets the link training working, but alas, no image. 7966 7967commit dd3eab848cf352bb96c3d01fe6028d8a4a8e451e 7968Author: Alex Deucher <alexdeucher@gmail.com> 7969Date: Fri Nov 20 01:06:49 2009 -0500 7970 7971 r300: render target limit is 2560 7972 7973 limit was increased to 4021 in r4xx. 7974 7975 fixes fdo bug 25191. 7976 7977 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 7978 7979commit df35d40d4ec260cdab57a9b36e4e96b2a6c2a827 7980Author: Alex Deucher <alexdeucher@gmail.com> 7981Date: Thu Nov 19 20:01:42 2009 -0500 7982 7983 DP: hack that makes DP work. 7984 7985 Something's wrong with the status check. 7986 7987commit f42412ec4dbd21f562a7fe4239c0a1f17ef1b8f2 7988Author: Alex Deucher <alexdeucher@gmail.com> 7989Date: Thu Nov 19 19:40:54 2009 -0500 7990 7991 wip 7992 7993commit eb9bc133fc426e67b397e661bfd22bf62009d9d3 7994Author: Dave Airlie <airlied@redhat.com> 7995Date: Fri Nov 20 09:23:31 2009 +1000 7996 7997 kms: recalculate the flush limits after screen resize. 7998 7999 When we resize the front buffer we need to reduce the flush limits 8000 appropriately. 8001 8002 Signed-off-by: Dave Airlie <airlied@redhat.com> 8003 8004commit f7f58ef4c042e492618665a6c5555e8e67387ab3 8005Author: Dave Airlie <airlied@redhat.com> 8006Date: Fri Nov 20 09:22:39 2009 +1000 8007 8008 radeon: r100/r200 have a 2047 scissor limit 8009 8010 We were overflowing this in my case with a 2704 width desktop, 8011 so videos were stopping around 700 bytes across the screen. 8012 8013 can I haz shatter already? 8014 8015 Signed-off-by: Dave Airlie <airlied@redhat.com> 8016 8017commit 54c09778d79de1516ee511d6d7d5b54dd4bf9130 8018Author: Alex Deucher <alexdeucher@gmail.com> 8019Date: Thu Nov 19 13:32:30 2009 -0500 8020 8021 DP wip 8022 8023 - use coherent 8024 - fix up lane/clock numbers 8025 8026commit 02d017a0552db2c8a48506c5b26c48a1a2ab75b6 8027Author: Gaetan Nadon <memsize@videotron.ca> 8028Date: Wed Oct 28 14:41:41 2009 -0400 8029 8030 INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206 8031 8032 Automake 'foreign' option is specified in configure.ac. 8033 Remove from Makefile.am 8034 8035commit 396ee9b7e7e3b1dd64532270c4e32cdb33857596 8036Author: Gaetan Nadon <memsize@videotron.ca> 8037Date: Wed Oct 28 14:09:09 2009 -0400 8038 8039 INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206 8040 8041 Add missing INSTALL file. Use standard GNU file on building tarball 8042 README may have been updated 8043 Remove AUTHORS file as it is empty and no content available yet. 8044 Remove NEWS file as it is empty and no content available yet. 8045 8046commit 1dcda0c4d5d506e18a7989ff05f2ad8569a73f23 8047Author: Gaetan Nadon <memsize@videotron.ca> 8048Date: Mon Oct 26 12:54:21 2009 -0400 8049 8050 Several driver modules do not have a ChangeLog target in Makefile.am #23814 8051 8052 The git generated ChangeLog replaces the hand written one. 8053 Update configure.ac to xorg-macros level 1.3. 8054 Use XORG_DEFAULT_OPTIONS which replaces four XORG_* macros 8055 Update Makefile.am to add ChangeLog target if missing 8056 Remove ChangeLog from EXTRA_DIST or *CLEAN variables 8057 This is a pre-req for the INSTALL_CMD 8058 8059commit 5b517bcc353e7111d2365a213505f27cfae30924 8060Author: Gaetan Nadon <memsize@videotron.ca> 8061Date: Thu Oct 22 12:34:16 2009 -0400 8062 8063 .gitignore: use common defaults with custom section # 24239 8064 8065 Using common defaults will reduce errors and maintenance. 8066 Only the very small or inexistent custom section need periodic maintenance 8067 when the structure of the component changes. Do not edit defaults. 8068 8069commit 605c869076985b69350c893cf2bbcc90de8dc6bf 8070Author: Alex Deucher <alexdeucher@gmail.com> 8071Date: Wed Nov 18 01:51:12 2009 -0500 8072 8073 more dp cleanup 8074 8075 - clean up handling of linkb, etc. 8076 - add support for ATOM_TRANSMITTER_ACTION_SETUP_VSEMPH 8077 to existing transmitter control function 8078 - fix lots of warnings 8079 8080commit bce1911f66f446b793125ac8cd5463bf6292629e 8081Author: Alex Deucher <alexdeucher@gmail.com> 8082Date: Wed Nov 18 00:04:33 2009 -0500 8083 8084 radeon: pull previous displayport work into master 8085 8086commit 437113124bbd6fb166825169eabec4dfde900dd9 8087Author: Alex Deucher <alexdeucher@gmail.com> 8088Date: Tue Nov 17 14:16:22 2009 -0500 8089 8090 radeon: deal with connectors sourced to the same encoder 8091 8092 Some systems have multiple connectors connected to the 8093 same encoder; e.g., DVI and HDMI connected to the same 8094 encoder with the same ddc line. Since we expose 8095 connectors as xrandr outputs, randr treats them separately 8096 which results in it trying to source the same encoder to 8097 different crtcs. If we have an HDMI and DVI-D port on the 8098 same encoder, pick the one to be considered connected 8099 based on the edid (HDMI if edid indicates HDMI, DVI 8100 otherwise). Also, don't turn off (dpms) encoders that are 8101 shared. 8102 8103 Fixes fdo bug 21767. 8104 8105 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8106 8107commit 9b8605d4e2bf7703c7848f3e2022a8c7328a3acd 8108Author: Alex Deucher <alexdeucher@gmail.com> 8109Date: Tue Nov 17 11:45:09 2009 -0500 8110 8111 radeon: fix potential memory leak in ddc code 8112 8113 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8114 8115commit ce8299962003de572122561a6f7f61eaccf633b2 8116Author: Dave Airlie <airlied@redhat.com> 8117Date: Tue Nov 17 15:23:30 2009 +1000 8118 8119 kms: check for 0 mask for cloned outputs. 8120 8121 Signed-off-by: Dave Airlie <airlied@redhat.com> 8122 8123commit 06d68f7fa3870f9601e837e1834d33e8395008b1 8124Author: Dave Airlie <airlied@redhat.com> 8125Date: Tue Nov 17 15:10:04 2009 +1000 8126 8127 kms: add cloning support. 8128 8129 We have to translate from the kernel encoder cloning to the randr 8130 connector cloning, we do this by constructing an encoder mask per output 8131 and an encoder cloning mask per output then comparing them. 8132 8133 Signed-off-by: Dave Airlie <airlied@redhat.com> 8134 8135commit 7587ce16ab0380337fe801f457c1d1d9b4141cc5 8136Author: Alex Deucher <alexdeucher@gmail.com> 8137Date: Mon Nov 16 19:16:03 2009 -0500 8138 8139 r600: remove un-needed format conversions 8140 8141 we do the normalization and xforms in the vertex 8142 shader. 8143 8144 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8145 8146commit ba605b5c9ff48765f2b66960aa4cc81b6235d012 8147Author: Alex Deucher <alexdeucher@gmail.com> 8148Date: Mon Nov 16 19:00:52 2009 -0500 8149 8150 r600: fix num format in vtx fetch 8151 8152 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8153 8154commit 0c4710c67a2fee2061fc3da43c9f908585693cfa 8155Author: Alex Deucher <alexdeucher@gmail.com> 8156Date: Thu Nov 12 11:11:17 2009 -0500 8157 8158 radeon: man page updates 8159 8160commit 2af2744c80f3307e6d84edba9479be5bd36e1d03 8161Author: Alex Deucher <alexdeucher@gmail.com> 8162Date: Wed Nov 11 13:12:24 2009 -0500 8163 8164 radeon: fix support for external tmds on some legacy chips 8165 8166 Should fix bug 11801. 8167 8168 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8169 8170commit f17657fdc83d8f4c0386d2c7dade98de5b94acbe 8171Author: Alex Deucher <alexdeucher@gmail.com> 8172Date: Mon Nov 9 16:59:11 2009 -0500 8173 8174 r600/r700: typo, fix mask of DB_ALPHA_TO_MASK 8175 8176 noticed by glisse. 8177 8178commit 0ee7763fb33bf7fbc4ae1e8372cbc74578bfb720 8179Author: Alex Deucher <alexdeucher@gmail.com> 8180Date: Mon Nov 9 15:09:34 2009 -0500 8181 8182 IGP: some IGP chips report as AGP 8183 8184 Set bus type appropriately. fixes bug 25002 8185 8186 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8187 8188commit da1fcddaade58e371c7bee555ce32944c689f810 8189Author: Alex Deucher <alexdeucher@gmail.com> 8190Date: Thu Nov 5 14:13:49 2009 -0500 8191 8192 radeon: properly fix unused label warnings 8193 8194commit 6e496f46fdc90d0a6cef41e9d45a6458aaed1eca 8195Author: Alex Deucher <alexdeucher@gmail.com> 8196Date: Thu Nov 5 14:11:55 2009 -0500 8197 8198 Revert "Fix warning about unused defined labels in radeon_exa_funcs.c." 8199 8200 This reverts commit b6ac42cce2156880ac18b32e15acbff40b2f53e9. 8201 8202 This patch is very broken. 8203 8204commit b6ac42cce2156880ac18b32e15acbff40b2f53e9 8205Author: Tomáš Chvátal <scarabeus@gentoo.org> 8206Date: Sat Oct 31 23:07:07 2009 +0100 8207 8208 Fix warning about unused defined labels in radeon_exa_funcs.c. 8209 8210 Signed-off-by: Tomáš Chvátal <scarabeus@gentoo.org> 8211 8212commit 87d7235790866f9c19ef08972d5237f09d940cd9 8213Author: Alex Deucher <alexdeucher@gmail.com> 8214Date: Thu Nov 5 10:23:03 2009 -0500 8215 8216 EXA: fallback if no pMaskPicture->pDrawable 8217 8218 A solid or gradient mask could be used for blending 8219 the source picture onto the destination picture. 8220 8221 Fixes fdo bug 24838 8222 8223 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8224 8225commit 873897c54429e230a23011a9dd2f1069f4eb7752 8226Author: Alex Deucher <alexdeucher@gmail.com> 8227Date: Thu Nov 5 00:51:11 2009 -0500 8228 8229 DCE3+: call transmitter init on mode set 8230 8231 Generally this is done at post, but might not always 8232 be done with softboot or for connectors on docking 8233 stations. 8234 8235 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8236 8237commit 3028374488cc0f34942ac372c8d05cf15898a613 8238Author: Alex Deucher <alexdeucher@gmail.com> 8239Date: Wed Nov 4 18:48:03 2009 -0500 8240 8241 ATOM/DCE3+: fix up usPixelClock calculation for Transmitter tables 8242 8243 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8244 8245commit 52973e8f2c4104de60d00c0e5c29a31f77bf3de7 8246Author: Alex Deucher <alexdeucher@gmail.com> 8247Date: Sun Nov 1 13:58:03 2009 -0500 8248 8249 radeon/atom: fix tv-out on r5xx 8250 8251 fixes bug 24796 8252 8253 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8254 8255commit 19b4f5f9b647cd470616b8ec1d34f1e4afa357ad 8256Author: Alex Deucher <alexdeucher@gmail.com> 8257Date: Tue Oct 27 11:55:20 2009 -0400 8258 8259 ATOM: fix up DVO for DCE 3.x 8260 8261 AdjustDisplayPll takes care of this for us. 8262 8263commit 5a0019126a57138ee506d9a66738c9e8b75cbb96 8264Author: Alex Deucher <alexdeucher@gmail.com> 8265Date: Tue Oct 27 11:39:06 2009 -0400 8266 8267 radeon: fix DVO on AVIVO chips 8268 8269 DVO generally requires 2x ppll on AVIVO cards due to the 8270 way the tmds chip is wired up. Because of this, disable 8271 cloning of DVO outputs. 8272 8273 fixes fdo bug 21857. 8274 8275 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8276 8277commit 2a6eeec72d6bb04a32225883f431c1d2e8cff123 8278Author: Alex Deucher <alexdeucher@gmail.com> 8279Date: Tue Oct 27 11:18:44 2009 -0400 8280 8281 atom: loosen pll min output limits 8282 8283 Limiting the pll output range is a good thing generally as 8284 it limits the number of possible pll combinations for a given 8285 frequency presumably to the ones that work best on each card. 8286 That's why the limits are in the bios tables. However, certain 8287 duallink DVI monitors seem to like pll combinations that would 8288 be limited by this at least on pre-DCE 3.0 r6xx hardware. This 8289 might need to be adjusted per family or per clock range in the 8290 future. 8291 8292 Fixes fdo bug 24727. 8293 8294 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8295 8296commit f0d9d80fee4176eaba9435a9539f29d0eefe2a87 8297Author: Luca Tettamanti <kronos.it@gmail.com> 8298Date: Sat Oct 24 16:45:23 2009 -0400 8299 8300 Disable color tiling on r600/r700 8301 8302 Not supported yet. 8303 8304commit ac499c3c0ea92f83ebd8127bb8cb5de625de6c94 8305Author: Alex Deucher <alexdeucher@gmail.com> 8306Date: Fri Oct 23 10:48:55 2009 -0400 8307 8308 radeon: clamp the internal FB map to the aperture size v2 8309 8310 We don't use the invisible memory yet and on cards with 8311 large amounts of vram this can cause the top of GART 8312 calculation to overflow. 8313 8314 Fixes bug fdo bug 24301: 8315 http://bugs.freedesktop.org/show_bug.cgi?id=24301 8316 8317 v2: only clamp cards with more than 512 MB. This seems 8318 to cause problems on some older cards due to the way the 8319 drm and ddx set up the internal memory map. 8320 8321commit e57b54daf318ff59315a1c9ed21934f288c5c782 8322Author: Alex Deucher <alexdeucher@gmail.com> 8323Date: Thu Oct 22 16:07:58 2009 -0400 8324 8325 R7xx: program additional CUR/GRPH regs for 40 bit addresses 8326 8327 The *_HIGH regs are reversed. The secondary ones are in the 8328 primary block and vice versa. 8329 8330commit 4cf06dfba617529291ce4b4c306c4fc1bba110ee 8331Author: Alex Deucher <alexdeucher@gmail.com> 8332Date: Thu Oct 22 02:08:45 2009 -0400 8333 8334 ATOM: use indirect addressing for regs >= 0x10000 8335 8336commit 66b194a78c470cb3978f310828dd96c3f3e96944 8337Author: Alex Deucher <alexdeucher@gmail.com> 8338Date: Wed Oct 21 15:58:18 2009 -0400 8339 8340 ATOM: fix up get clock info 8341 8342 Newer revisions of the atom firmware table have 8343 changed. This helps select better pll dividers 8344 in some cases. Noticed by Mathias Froehlich. 8345 8346commit bd89b7501f294ac645390ef144df569953c81dc4 8347Author: Alex Deucher <alexdeucher@gmail.com> 8348Date: Sun Oct 18 16:21:36 2009 -0400 8349 8350 radeon: fixup CustomEDID option 8351 8352 - add support for analog option to force edid to analog 8353 mode 8354 - fix compilation on older xservers 8355 - updated man page 8356 8357commit 255ac8d54e3dc12e579e05edb41b40946a00eab8 8358Author: Alex Deucher <alexdeucher@gmail.com> 8359Date: Tue Oct 13 12:27:52 2009 -0400 8360 8361 radeon: add XV_CRTC attribute for textured video 8362 8363 This attribute allows the user to override which 8364 crtc is synced with when XV_VSYNC is enabled. This 8365 is useful for clone modes where the user can might want 8366 to override the default. 8367 8368commit 801d900add4731f3c424ebb780ad2dbd7a1c6a25 8369Author: Alex Deucher <alexdeucher@gmail.com> 8370Date: Mon Oct 12 10:00:57 2009 -0400 8371 8372 ATOM: minor cleanups to pll setup 8373 8374commit f4407962cd7b272e0860319f11f6a6583ef226c2 8375Author: Csillag Kristof <csillag.kristof@united-consult.hu> 8376Date: Fri Oct 9 18:00:09 2009 -0400 8377 8378 Fix some issues with CustomEDID support 8379 8380 - Fixes a memory allocation problem introduced with the previous 8381 version of the CustomEDID patch (now in GIT), and 8382 - Enhances the syntax of the CustomEDID option so that I can force 8383 digital output, even if the EDID was acquired over an analog link. 8384 8385commit de55995e82c3875f70b6394fff440d695d062113 8386Author: Alex Deucher <alexdeucher@gmail.com> 8387Date: Fri Oct 9 11:07:30 2009 -0400 8388 8389 r600 EXA: fix up mask reg mixup 8390 8391commit a168caf5349b757873ad89d977772feaa7fb6e58 8392Author: Alex Deucher <alexdeucher@gmail.com> 8393Date: Fri Oct 9 10:32:26 2009 -0400 8394 8395 r600 EXA: fix some compile warnings 8396 8397commit fb8dda2928bd0def2b34c8c8e414129e650d6eb5 8398Author: Alex Deucher <alexdeucher@gmail.com> 8399Date: Fri Oct 9 10:30:21 2009 -0400 8400 8401 r600 EXA: properly check planemask 8402 8403 We only support per-component masks. 8404 8405commit 1b25bad2d8ebe481ffb4b0fd3a85ab636bd4fec2 8406Author: Dave Airlie <airlied@linux.ie> 8407Date: Fri Oct 9 20:09:52 2009 +1000 8408 8409 Revert "radeon: clamp the internal FB map to the aperture size" 8410 8411 This reverts commit 5f846360c46f5a989f5d0fde6d251cdbd61d4968. 8412 8413 Numerous reports of system hangs since this, I'm guessing 8414 some sort of conflict with the drm memory setup. This code 8415 has always been fragile between kernel/userspace drivers. 8416 8417commit a956d478043b44c49aed39c7aa3d576ef32cadc5 8418Author: Dave Airlie <airlied@linux.ie> 8419Date: Fri Oct 9 20:08:31 2009 +1000 8420 8421 Revert "kms: setup colormap and gamma correctly." 8422 8423 This reverts commit 60d9685abddccec17c1a9a5ec48cbe9c92543e0f. 8424 8425 which breaks colormap on 1.6 X server 8426 8427 Conflicts: 8428 8429 src/drmmode_display.c 8430 8431commit a493feb94c0d84aaf5bfd84c6940d7f03974dd86 8432Author: Dave Airlie <airlied@redhat.com> 8433Date: Fri Oct 9 15:05:25 2009 +1000 8434 8435 kms: don't use scratch pixmaps at all 8436 8437 use our own wrapper to allocate a pixmap for wrapping a bo in. 8438 8439commit 0caffbd6fd6b356b9b7d715c015285eaa91e3e36 8440Author: Dave Airlie <airlied@redhat.com> 8441Date: Fri Oct 9 15:04:56 2009 +1000 8442 8443 drmmode: reload cursors on modeset 8444 8445commit 02e12ae6be7bc1976f63848fa2854d320d5ab36e 8446Author: Alex Deucher <alexdeucher@gmail.com> 8447Date: Wed Oct 7 19:56:58 2009 -0400 8448 8449 ATOM: rework crtc modeset 8450 8451 - clean up tv timing handling 8452 - unify SetCRTCTiming and SetCRTCDTDTiming interfaces 8453 8454commit d499eeaf22f77a1294e99aa38a50aa6810bb684a 8455Author: Alex Deucher <alexdeucher@gmail.com> 8456Date: Wed Oct 7 17:43:11 2009 -0400 8457 8458 ATOM: reorder crtc dpms based on bios recommendations 8459 8460commit 4ccd2a21f9c50515b4246f35454f76ef49006c76 8461Author: Alex Deucher <alexdeucher@gmail.com> 8462Date: Wed Oct 7 16:30:03 2009 -0400 8463 8464 ATOM: add support for AdjustDisplayPll table 8465 8466 Depending on the output and clock, this table will 8467 adjust the pixelclock accordingly. 8468 8469commit 5f846360c46f5a989f5d0fde6d251cdbd61d4968 8470Author: Alex Deucher <alexdeucher@gmail.com> 8471Date: Wed Oct 7 17:01:55 2009 -0400 8472 8473 radeon: clamp the internal FB map to the aperture size 8474 8475 We don't use the invisible memory yet and on cards with 8476 large amounts of vram this can cause the top of GART 8477 calculation to overflow. 8478 8479 Fixes bug fdo bug 24301: 8480 http://bugs.freedesktop.org/show_bug.cgi?id=24301 8481 8482commit 186bf719c394116341abff471fb4b4803650da9a 8483Author: Dave Airlie <airlied@redhat.com> 8484Date: Wed Oct 7 17:23:36 2009 +1000 8485 8486 radeon: fix rotation since mixed pixmaps. 8487 8488 Passing null in here stops the mixed pixmap code doing it wrong, 8489 but I'm not 100% sure how it is meant to work, we have the 8490 same issue with the real front bo by the looks of it. 8491 8492commit e08411af1aa8c7d4233ba593b84360397cdbb307 8493Author: Roel Kluin <roel.kluin@gmail.com> 8494Date: Tue Oct 6 18:31:17 2009 -0400 8495 8496 radeon: Fix duplicated bit settings 8497 8498 [agd5f: adapted from kms patch] 8499 8500 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8501 8502commit e59ae08270711512e64b70f79b6476cc2c52d230 8503Author: Michel Dänzer <daenzer@vmware.com> 8504Date: Tue Oct 6 00:13:09 2009 +0200 8505 8506 Don't call radeon_cs_flush_indirect() without KMS. 8507 8508 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=24316 . 8509 8510commit 457646d734558672402b6bcd15cadb66741edbed 8511Author: Jerome Glisse <jglisse@redhat.com> 8512Date: Mon Oct 5 15:52:42 2009 +0200 8513 8514 kms: assume accel is working is info ioctl return -EINVAL 8515 8516 -EINVAL should only happen if the info ioctl doesn't support 8517 the get accel query. This patch assume that on such case accel 8518 is working. 8519 8520commit 9d596562496863d65850306d2126d8df98464de4 8521Author: Dave Airlie <airlied@linux.ie> 8522Date: Mon Oct 5 18:35:20 2009 +1000 8523 8524 kms: only touch gamma stuff on 1.7 8525 8526 for correct colormap operation you probably want 1.7 8527 8528commit 876f43d4682df4d47d3e158f75e0ca6040f41c70 8529Author: Alex Deucher <alexdeucher@gmail.com> 8530Date: Mon Oct 5 02:23:03 2009 -0400 8531 8532 R1/2xx EXA: Minimise number of draw primitives used for Composite operations. 8533 8534commit 86cafb8affea448bdf58300044e755201b822d2a 8535Author: Dave Airlie <airlied@redhat.com> 8536Date: Mon Oct 5 14:17:25 2009 +1000 8537 8538 kms: don't use scratch pixmaps when copying fbcon. 8539 8540 scratch pixmaps seem to interact badly with mixed pixmaps, it appears 8541 some state may be getting left around in the privates somewhere, since 8542 scratch pixmap headers don't get destroyed. 8543 8544commit f8471512ea9f1d38140dfe98a0f832e9f935f51b 8545Author: Csillag Kristof <csillag.kristof@united-consult.hu> 8546Date: Sun Oct 4 18:18:56 2009 -0400 8547 8548 radeon: add support for Custom EDID 8549 8550 Allows you to specify an edid per output from a file 8551 to override what is detected by DDC. Useful for 8552 problematic monitors or KVM switches that block 8553 DDC. Specifying an EDID that is not compatible with 8554 your monitor could damage your monitor so use with 8555 caution. 8556 8557 agd5f: cache the custom edid at startup so we don't 8558 have to read it from file every time the output is 8559 queried. 8560 8561commit 60d9685abddccec17c1a9a5ec48cbe9c92543e0f 8562Author: Dave Airlie <airlied@redhat.com> 8563Date: Mon Oct 5 12:41:30 2009 +1000 8564 8565 kms: setup colormap and gamma correctly. 8566 8567 We need to program the gamma at modeset. no need to use 8568 old cmap interface at all. 8569 8570commit b0b10e548fd69b3ff67a0f9b71456b89c61cb4d6 8571Author: Alex Deucher <alexdeucher@gmail.com> 8572Date: Sun Oct 4 17:07:32 2009 -0400 8573 8574 R6/7xx EXA: Add support for BGRA picture formats. 8575 8576 Also add mysteriously missing case statements for ABGR destination formats. 8577 8578commit 8404c37b2ff95aba31db97676f106ec2f30fda7f 8579Author: Dave Airlie <airlied@linux.ie> 8580Date: Sun Oct 4 18:18:51 2009 +1000 8581 8582 radeon: fix build without kms 8583 8584commit 9460ea864b12ec1fbd11c5d9a20bb5a4279d9d3d 8585Author: Michel Dänzer <daenzer@vmware.com> 8586Date: Sat Oct 3 16:33:33 2009 +0200 8587 8588 Fix KMS on big endian machines. 8589 8590 Requires at least xserver 1.7 to work properly. 8591 8592 Also make sure the front buffer is and stays tiled if colour tiling is enabled. 8593 8594commit aee6b6f2c120baf477b4845ddc1a94637c31db2f 8595Author: Michel Dänzer <daenzer@vmware.com> 8596Date: Sat Oct 3 16:33:33 2009 +0200 8597 8598 EXA: Bail earlier from PrepareCopy hook on trivially unsupportable conditions. 8599 8600commit 30031b169def5caadb394225919a6079a0957142 8601Author: Michel Dänzer <daenzer@vmware.com> 8602Date: Sat Oct 3 16:33:33 2009 +0200 8603 8604 R3/5xx EXA: Add support for BGRA picture formats. 8605 8606 Also add mysteriously missing case statements for ABGR destination formats. 8607 8608commit 4b4ce36081ca151c24e028c54b59986f41731a73 8609Author: Michel Dänzer <daenzer@vmware.com> 8610Date: Sat Oct 3 16:33:32 2009 +0200 8611 8612 R3/5xx EXA: Minimise number of draw primitives used for Composite operations. 8613 8614 This should reduce the kernel CS checker overhead, if nothing else. 8615 8616 I'll leave porting this to other chipset families to others who can test it. 8617 8618commit eade1e5be159c9f2965d611925596d33cab11d6d 8619Author: Michel Dänzer <daenzer@vmware.com> 8620Date: Sat Oct 3 16:33:32 2009 +0200 8621 8622 EXA: Don't always flush when switching between 2D/3D engines with KMS. 8623 8624 This seems to work fine here now and help 2D performance quite a bit, let's see 8625 if it breaks anybody else's setup... 8626 8627commit 926e414fc835ace141c066830d11d8ce32dbb06c 8628Author: Michel Dänzer <daenzer@vmware.com> 8629Date: Sat Oct 3 16:33:32 2009 +0200 8630 8631 KMS: Double-buffer textured video source image upload. 8632 8633 In order to avoid stalling on previous frame. 8634 8635 OTOH without KMS we can't do this but have to wait for the previous frame to 8636 finish rendering. 8637 8638commit 6a3b75fae4147244212298ff55c7ab36c3d80d30 8639Author: Michel Dänzer <daenzer@vmware.com> 8640Date: Sat Oct 3 16:33:31 2009 +0200 8641 8642 KMS: Don't hide HW cursor when it's updated. 8643 8644 Inspired by the intel driver; might avoid/reduce HW cursor flicker in some 8645 cases. 8646 8647commit c4ba15b4871c70cced7f1761e24be692cdb28d2b 8648Author: Michel Dänzer <daenzer@vmware.com> 8649Date: Sat Oct 3 16:33:31 2009 +0200 8650 8651 Fix some compiler warnings. 8652 8653commit cc45856a18dd3e6f7e44d9eb507b31419da70977 8654Author: Kenneth Graunke <kenny@whitecape.org> 8655Date: Fri Oct 2 14:31:36 2009 -0400 8656 8657 radeon: Remove gatos message from driver startup 8658 8659 Remove a message on driver startup which directs people to the old GATOS 8660 website; said site contains highly out of date software which is no longer 8661 necessary. 8662 8663commit d302481493ff61eb7f8648b7afe7407b6b653cf7 8664Author: Jerome Glisse <jglisse@redhat.com> 8665Date: Wed Sep 16 15:18:40 2009 +0200 8666 8667 radeon/kms: fallback to shadowfb if kernel report acceleration is off 8668 8669 This will fallback to shadowfb is GPU accel failed somewhere during 8670 KMS kernel init. 8671 8672commit 7968e1fb89f6b59d1654df48249bf4b81990c008 8673Author: Alex Deucher <alexdeucher@gmail.com> 8674Date: Fri Sep 25 19:05:38 2009 -0400 8675 8676 radeon: fix segfault in MMIO path in RADEONInit3DEngine 8677 8678 fixes fdo bug 24158. 8679 8680 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8681 8682commit 0380a9cf98d34e88dd81370ade5525680ec89c02 8683Author: Alex Deucher <alexdeucher@gmail.com> 8684Date: Fri Sep 25 16:46:43 2009 -0400 8685 8686 kms/r600: add support for vline relocs 8687 8688 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8689 8690commit 3efecebb10de7f7bacf9f8c57ae20fd508097294 8691Author: Alex Deucher <alexdeucher@gmail.com> 8692Date: Fri Sep 25 14:44:41 2009 -0400 8693 8694 radeon: fix vline handling for kms 8695 8696 drm crtc ids do not correspond to actual hw crtcs, 8697 as such the vline stuff was never enabled for Xv. 8698 8699 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8700 8701commit 9733dcde0a21b7503aa20254724f2910b541b990 8702Author: Dave Airlie <airlied@linux.ie> 8703Date: Sat Sep 26 06:15:58 2009 +1000 8704 8705 r600: fix build 8706 8707commit ff18595eace42ddfc931f4f948cb5adf59ebcf52 8708Author: Dave Airlie <airlied@linux.ie> 8709Date: Sat Sep 26 06:04:51 2009 +1000 8710 8711 r600/xv: fix dst bo write domain 8712 8713commit 718a611f3cc7f5eea95cd73a9dcc7913c922fcbc 8714Author: Dave Airlie <airlied@linux.ie> 8715Date: Sat Sep 26 06:03:51 2009 +1000 8716 8717 radeon: fix zaphod 8718 8719commit 86c367be517e992aabc1bcfea2a2dde1a1ae520c 8720Author: Alex Deucher <alexdeucher@gmail.com> 8721Date: Fri Sep 25 13:03:44 2009 -0400 8722 8723 r600: fix bo accounting for Xv 8724 8725 fixes Xv after changes in: 8726 8f80e37eed3ec028718b4e71bbb9b598847fd94e 8727 8728 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8729 8730commit 8f80e37eed3ec028718b4e71bbb9b598847fd94e 8731Author: Dave Airlie <airlied@linux.ie> 8732Date: Fri Sep 25 11:37:38 2009 +1000 8733 8734 r600: space check wasn't correct. 8735 8736 the r6xx space check code was incorrect for the DFS case, since 8737 the dst bo was in GTT but the hardcoded cp_start function 8738 assumed the bos were where it thought. 8739 8740 Ripped out assumptions and replaced with code more like other radeons. 8741 8742commit 90669f6cb5ada9067d6fccc7f54c69b367862f42 8743Author: Alex Deucher <alexdeucher@gmail.com> 8744Date: Thu Sep 24 10:40:40 2009 -0400 8745 8746 radeon: remove old pre-randr rotation cruft 8747 8748 leftover from the randr 1.2 conversion. 8749 8750 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> 8751 8752commit 97a4e747bfac14f34646c55ddf639e8fe22f2f55 8753Author: Michael Olbrich <m.olbrich@pengutronix.de> 8754Date: Sun Sep 20 14:19:41 2009 +0200 8755 8756 use AC_CHECK_HEADER instead of AC_CHECK_FILE 8757 8758 AC_CHECK_FILE is not possible when cross-compiling. Use AC_CHECK_HEADER / 8759 AC_PREPROC_IFELSE instead. 8760 8761 [ Michel Dänzer: Shuffled things around slightly to make it work on my setup ] 8762 8763 Signed-off-by: Michael Olbrich <m.olbrich@pengutronix.de> 8764 Signed-off-by: Michel Dänzer <daenzer@vmware.com> 8765 8766commit 579070ed9f0218a975e840331b6fe9fa8fd05789 8767Author: Alex Deucher <alexdeucher@gmail.com> 8768Date: Sat Sep 19 19:15:14 2009 -0400 8769 8770 kms: disable dga 8771 8772 Can be dangerous with kms: 8773 http://lists.freedesktop.org/archives/xorg-devel/2009-September/002113.html 8774 8775commit da7487f6ac7b32ed7866af53da2925aa52ed7aae 8776Author: Alex Deucher <alexdeucher@gmail.com> 8777Date: Sat Sep 19 13:08:02 2009 -0400 8778 8779 kms: init dga using xf86DiDGAInit 8780 8781 fixes cursor problems in games like UT2004 and Penumbra Overture 8782 8783 see fdo bug 24034 8784 8785commit 52279251fae9df99c569c16e2522bbd346d8ec38 8786Author: Dave Airlie <airlied@redhat.com> 8787Date: Fri Sep 18 14:29:32 2009 +1000 8788 8789 avivo: disable VGA rendering core when starting X. 8790 8791 this thing can be rendering to VRAM when we don't expect it. 8792 turn it off. 8793 8794 Signed-off-by: Dave Airlie <airlied@redhat.com> 8795 8796commit 2c46bafcb77c4125a27c18ad6ca2f6de5f143a2d 8797Author: Pauli Nieminen <suokkos@gmail.com> 8798Date: Thu Sep 17 15:13:51 2009 +0300 8799 8800 Fix configuration to disable KMS if not finding libdrm_radeon. 8801 8802 Most of KMS building was disabled if LIBDRM_RADEON was not detected. But at 8803 least configure was reporting KMS enabled even tough it was realy disabled. 8804 8805commit ee9f6d802988a4d7f7985687036db854f4a14fe5 8806Author: Pauli Nieminen <suokkos@gmail.com> 8807Date: Mon Sep 14 16:36:48 2009 +0300 8808 8809 radeon: Add missing libdrm_radeon cflags to compiler paramaters. 8810 8811 This fixes compilation if libdrm_radeon is installed to non-standard location like /opt. 8812 8813commit ac853ca0a950d6147cd6a0b4a1b089f5d2f080ea 8814Author: Alex Deucher <alexdeucher@gmail.com> 8815Date: Fri Sep 11 14:25:31 2009 -0400 8816 8817 atom: fix typo in asus quirks 8818 8819 Should be DVI-I, not DVI-D 8820 8821commit a88f60e7b789cd23a79e6e18a7e3fa008d15aa77 8822Author: Alex Deucher <alexdeucher@gmail.com> 8823Date: Thu Sep 10 13:42:31 2009 -0400 8824 8825 r600: add CS UTS and DFS hooks 8826 8827commit 0bb0ff0e55aaf8e64017d204e376a14f0a7384a3 8828Author: Dave Airlie <airlied@linux.ie> 8829Date: Thu Sep 10 11:48:00 2009 +1000 8830 8831 r6xx: fix EXA crash with gtkperf -a 8832 8833 unmapping was getting unbalanced. 8834 8835commit 10a58d54857484b6e89763ecf463f54dc290c702 8836Author: Dave Airlie <airlied@redhat.com> 8837Date: Wed Sep 9 14:24:49 2009 +1000 8838 8839 kms: add property support. 8840 8841 this adds support to the userspace DDX to get properties from the kernel 8842 and expose them over randr. Its most liberated from the Intel driver. 8843 8844commit fad9ee6cdae5ae45e6406e3071a84276f1b59a85 8845Author: Alex Deucher <alexdeucher@gmail.com> 8846Date: Tue Sep 8 17:50:50 2009 -0400 8847 8848 radeon: replace dga code with DiDGA 8849 8850commit 651fe5a47435c6a5cc1cdfb33c5c6601d692adcb 8851Author: Adam Jackson <ajax@redhat.com> 8852Date: Tue Sep 8 13:21:07 2009 +1000 8853 8854 radeon: fix dri2 struct zeroing 8855 8856commit 6990f2ac6478bf92929a4400ef84fb2142699204 8857Merge: 917f2d7 853f4c3 8858Author: Dave Airlie <airlied@redhat.com> 8859Date: Tue Sep 8 11:26:32 2009 +1000 8860 8861 Merge branch 'r6xx-cs' 8862 8863commit 853f4c3d1ea8f975ab2855f18d3ae336a4095091 8864Author: Dave Airlie <airlied@redhat.com> 8865Date: Tue Sep 8 11:25:39 2009 +1000 8866 8867 r600: more alignment fixups + vb map/unmap 8868 8869 I'm not so sure the vb map/unmap is a good idea, I think 8870 it pretty much locksteps the cpu/gpu, so we should really 8871 work out if we really need to flush this often, since 8872 mesa doesn't have to and we are just doing 3D ops. 8873 8874commit 917f2d7cd1815279710939edc27133ca6609460f 8875Author: Alex Deucher <alexdeucher@gmail.com> 8876Date: Sun Sep 6 10:53:29 2009 -0400 8877 8878 RV280: add agp quirk 8879 8880 deb bug 545040 8881 8882commit 78fcbf577ad6eba6399cc39f74b7ce5f9c8e265e 8883Author: Michel Dänzer <daenzer@vmware.com> 8884Date: Thu Sep 3 14:55:05 2009 +0200 8885 8886 EXA: Check for solid/gradient pictures the same way for all generations. 8887 8888 In particular, also catch them for >= R300. 8889 8890commit 794ae743c305331741e7cf45dd71c755fe325ed5 8891Author: Michel Dänzer <daenzer@vmware.com> 8892Date: Wed Sep 2 09:19:17 2009 +0200 8893 8894 EXA: Fix Composite restart dst/src switcheroo. 8895 8896 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=22636 . 8897 8898 Now, where's that brown paper bag? :} 8899 8900commit c4ab50c5eafca3d04129a72453496eb8afb26b34 8901Author: Alex Deucher <alexdeucher@gmail.com> 8902Date: Tue Sep 1 13:43:00 2009 -0400 8903 8904 radeon: add some notes about DVO setup 8905 8906commit 22fbd6476f4b3513378e848affc4dd269e9d4498 8907Author: Dave Airlie <airlied@itt42.(none)> 8908Date: Tue Sep 1 15:22:57 2009 +1000 8909 8910 r100/r200: seriously WRAP is illegal for rectangular textures 8911 8912 unless we scale all the bos up to POT 8913 8914commit 8f4196e88855f10762254fca9e0a0988e7b5562f 8915Author: Alex Deucher <alexdeucher@gmail.com> 8916Date: Mon Aug 31 19:41:59 2009 -0400 8917 8918 r6xx/r7xx: various CS fixes from Dave 8919 8920commit a32384f9b9ed73c14484a380dfa14652ed478ad6 8921Author: Alex Deucher <alexdeucher@gmail.com> 8922Date: Mon Aug 31 13:26:05 2009 -0400 8923 8924 r1xx: fix textured video 8925 8926 Copy/paste bug from when kms support was added. 8927 8928commit 3d6bf0f8cf128ee67d448491be44d345547b75c5 8929Author: Pauli Nieminen <suokkos@gmail.com> 8930Date: Fri Aug 28 17:05:44 2009 +0300 8931 8932 radeon: Fix DRI2BufferPtr to be DRI2Buffer2Ptr for xserver 1.6. 8933 8934 Bump requirement for xserver in KMS mode to 1.6.2 for DRI2Buffer2Ptr. 8935 8936commit 6521dbf3884185ca5692d35d879d251890cc4f08 8937Author: Roland Scheidegger <sroland@tungstengraphics.com> 8938Date: Sat Aug 29 12:11:03 2009 +0200 8939 8940 r100/r200: fix section size mismatch for textured video 8941 8942 this fixes the driver complaining though reportedly it still does 8943 not work (on r100) 8944 8945commit 39dfac41eebbcdacb1ae0c17cea325ea2966c4a5 8946Author: Michel Dänzer <daenzer@vmware.com> 8947Date: Fri Aug 28 12:20:04 2009 +0200 8948 8949 EXA: RENDER repeat fix and more cleanup. 8950 8951 Apparently changing pPict->repeatType causes badness, see 8952 http://bugs.freedesktop.org/show_bug.cgi?id=23560 . 8953 8954 Now we just use a normalized repeatType value derived from pPict->repeat and 8955 pPict->repeatType everywhere. 8956 8957commit 7d36f13afa5da742aa6c989a2652e92c96e5f80e 8958Author: Michel Dänzer <daenzer@vmware.com> 8959Date: Thu Aug 27 09:52:11 2009 +0200 8960 8961 KMS: Remove explicit radeon_bo_wait calls before radeon_bo_map. 8962 8963 libdrm_radeon always takes care of this implicitly now. 8964 8965commit 066c90e8f5b93c80195dc43486ad3efa083a1f3a 8966Author: Michel Dänzer <daenzer@vmware.com> 8967Date: Thu Aug 27 09:52:11 2009 +0200 8968 8969 KMS: Don't use a blit for UploadToScreen if the pixmap BO will be idle anyway. 8970 8971 Direct CPU writes should be at least as fast in that case. 8972 8973commit 174b61bb786a841cebd354e23b4b0caa85b541aa 8974Author: Michel Dänzer <daenzer@vmware.com> 8975Date: Thu Aug 27 09:52:11 2009 +0200 8976 8977 KMS: DownloadFromScreen improvements. 8978 8979 * Drop superfluous RADEONDownloadFromScreenGTT function, EXA does the same 8980 thing when we return FALSE. 8981 * Take unflushed operations into account for determining which GEM domain the 8982 pixmap BO will end up in. 8983 * Only use a blit if it ends up in VRAM. 8984 8985commit 7623e169e1f7d5afbd4108de03f28098bca519db 8986Author: Michel Dänzer <daenzer@vmware.com> 8987Date: Thu Aug 27 09:52:11 2009 +0200 8988 8989 EXA: Allocate pixmap BOs in VRAM. 8990 8991 This is the intention with EXA in xserver Git, and IME it improves performance 8992 significantly even with older versions. 8993 8994commit b90b823790657707b5bf7249095833bef959b0e8 8995Author: Michel Dänzer <daenzer@vmware.com> 8996Date: Thu Aug 27 09:52:11 2009 +0200 8997 8998 DRI2: Set tiling for depth/stencil buffers with all DRI2 interface versions. 8999 9000commit 27d9664f8862d1948dd32021f8e9bcb94f23e4ef 9001Author: Michel Dänzer <daenzer@vmware.com> 9002Date: Thu Aug 27 09:52:10 2009 +0200 9003 9004 EXA: RENDER repeat cleanups. 9005 9006 We can't rely on the server setting pPict->repeatType to something sensible 9007 when pPict->repeat is FALSE, but we can do it ourselves. 9008 9009commit 9b5517fcf7f2397a1a814d396fc9dc24769a6d9f 9010Author: Michel Dänzer <daenzer@vmware.com> 9011Date: Thu Aug 27 09:52:10 2009 +0200 9012 9013 Throttle DRI2 buffer swaps / frontbuffer flushes. 9014 9015commit e87f0f50f31a59ca1f60d4582d4a57ed00854fb7 9016Author: Alex Deucher <alexdeucher@gmail.com> 9017Date: Wed Aug 26 02:13:38 2009 -0400 9018 9019 r6xx/r7xx: set EXA_HANDLES_PIXMAPS 9020 9021commit 3212c26b90c0f6f1a7248b4da3ed985a9c2e9381 9022Author: Alex Deucher <alexdeucher@gmail.com> 9023Date: Wed Aug 26 01:42:10 2009 -0400 9024 9025 r6xx/r7xx: more WIP 9026 9027commit 5a08e68cc254fb255e631b456e331c32456ef0e7 9028Author: Alex Deucher <alexdeucher@gmail.com> 9029Date: Tue Aug 25 19:24:41 2009 -0400 9030 9031 r6xx/r7xx: fix some define problems in Xv code 9032 9033commit 9aa214e125b7927d62b9fe124a851d0373c24d7e 9034Author: Alex Deucher <alexdeucher@gmail.com> 9035Date: Tue Aug 25 18:45:49 2009 -0400 9036 9037 r6xx/r7xx: fix reloc for vtx buffer 9038 9039commit bba51187055932ecd466f5f817428d6c773747b9 9040Author: Alex Deucher <alexdeucher@gmail.com> 9041Date: Tue Aug 25 18:37:15 2009 -0400 9042 9043 R6xx/r7xx: unmap vb bo when done 9044 9045commit 599adfc1f5e6d708be7ad30f4871de3046775727 9046Author: Alex Deucher <alexdeucher@gmail.com> 9047Date: Tue Aug 25 18:13:14 2009 -0400 9048 9049 r6xx/r7xx: fix flipped domains 9050 9051commit 2e83cca8d7efaf1a6836cfb9ea5893fd9d70175f 9052Author: Alex Deucher <alexdeucher@gmail.com> 9053Date: Tue Aug 25 18:05:43 2009 -0400 9054 9055 r6xx/r7xx: more cs exa wip 9056 9057commit 65852de027989c105246fa4e4eed432f29525a22 9058Author: Alex Deucher <alexdeucher@gmail.com> 9059Date: Tue Aug 25 17:29:42 2009 -0400 9060 9061 r6xx/r7xx EXA: WIP 9062 9063commit 69ec7a35e2a0a3d802ec093a6aab2d7ed2cc88be 9064Author: Alex Deucher <alexdeucher@gmail.com> 9065Date: Tue Aug 25 16:14:02 2009 -0400 9066 9067 r6xx/r7xx: first pass at kms accel support 9068 9069 Adapted from various patches from Dave and Jerome. 9070 9071commit 04692f80678ffea15ac33820381ce45865e697b8 9072Author: Dave Airlie <airlied@linux.ie> 9073Date: Wed Aug 26 05:31:56 2009 +1000 9074 9075 radeon: afaics these aren't radeon regs but r128 regs. 9076 9077 so remove them from radeon use. 9078 9079commit 9cf965bbc977f0523437c0ecf1d7363b17de2468 9080Author: Alex Deucher <alexdeucher@gmail.com> 9081Date: Tue Aug 25 12:47:34 2009 -0400 9082 9083 R6xx/r7xx: add begin/end batch macros 9084 9085commit b6368cc572c79bce9a9366242c727c13cab3f006 9086Author: Alex Deucher <alexdeucher@gmail.com> 9087Date: Tue Aug 25 12:14:33 2009 -0400 9088 9089 r6xx/r7xx: move more common state to default state setup 9090 9091commit fc74e1194c980d978667e02c60a29a761a694bde 9092Author: Alex Deucher <alexdeucher@gmail.com> 9093Date: Tue Aug 25 11:39:56 2009 -0400 9094 9095 RV740: disable small DFS transfers 9096 9097 Seems problematic on all rv740 chips, so until we 9098 can find a solution, disable them. 9099 9100commit 1c909ecd88e75c6e16af7aa068e32a7a60b0af37 9101Author: Jerome Glisse <jglisse@redhat.com> 9102Date: Tue Aug 25 11:21:41 2009 +0200 9103 9104 radeon: Fix X recycling with KMS 9105 9106 On X recycling CloseScreen drop master so grab it again 9107 in InitScreen and avoid to reallocate buffer & various 9108 others cs/bo utilities on X recycling. 9109 9110commit 39aaf9addc923852598ba6fb914459671b78e492 9111Author: Dave Airlie <airlied@linux.ie> 9112Date: Mon Aug 24 18:59:22 2009 +1000 9113 9114 r600: use exa fallback code for solid/copy codepaths. 9115 9116 Solid seem to try and handle 1bpp which really no. 9117 9118commit e4b2d57ddca8739ec0b452565713398d036be0d5 9119Author: Dave Airlie <airlied@redhat.com> 9120Date: Mon Aug 24 14:11:24 2009 +1000 9121 9122 r600: fix bad unit in texture setup 9123 9124commit bac224912c750dc1c85ff2d9b8526dad6c23b572 9125Author: Dave Airlie <airlied@redhat.com> 9126Date: Sat Aug 22 21:17:59 2009 +1000 9127 9128 radeon: don't spec any initial placement for pixmaps. 9129 9130 allow the first use to decide placement. 9131 9132commit 77f98717d825162da106c6898cdbcbdf5c984ae6 9133Author: Dave Airlie <airlied@redhat.com> 9134Date: Sat Aug 22 21:16:25 2009 +1000 9135 9136 exa/cs: add DFS from GTT optimisation 9137 9138 This uses the new libdrm busy interface, once I had this in place 9139 I added a error if this happened and it does on my desktop here, 9140 so may as well add the optimisation that used to be in my old KMS tree. 9141 9142 Signed-off-by: Dave Airlie <airlied@redhat.com> 9143 9144commit b1b77a4d6fb7404af9568644e1a8e050fdfa956e 9145Author: Alex Deucher <alexdeucher@gmail.com> 9146Date: Wed Aug 19 13:29:04 2009 -0400 9147 9148 rs600: add support for DisplayPriority HIGH 9149 9150 Still haven't implemented the actual watermark 9151 calculation, but this forces the display requests 9152 to urgent if the user specifies DisplayPriority HIGH. 9153 9154commit 72e0d1b2cb11a67b8e4be4c74913ee44dc051c5b 9155Author: Dave Airlie <airlied@redhat.com> 9156Date: Wed Aug 19 19:01:30 2009 +1000 9157 9158 radeon/kms: add initial colortiling support (disabled by default). 9159 9160 This requires an X server from git with createpixmap2 support fixed up in it. 9161 9162 On 1.6 and previous it won't do any tiling, if you are running git server 9163 please upgrade to latest git. 9164 9165 Option "AllowColorTiling" "true" to enable and do some testing 9166 9167commit 479a6daefe46f985c415b0d000b1b1b820f3924e 9168Author: Alex Deucher <alexdeucher@gmail.com> 9169Date: Tue Aug 18 02:53:29 2009 -0400 9170 9171 AVIVO: add support for DisplayPriority HIGH 9172 9173 If the DisplayPriority option is set to HIGH, 9174 force the diplay mem requests to urgent. 9175 9176commit 8fd6f69c84103b277de70181988e843da23c85b0 9177Author: Dave Airlie <airlied@redhat.com> 9178Date: Mon Aug 17 15:50:21 2009 +1000 9179 9180 radeon: move detected monitor type debugging. 9181 9182 We should print this after DAC detection to be useful for TV 9183 9184commit d25185f33d06a8162469e18688e95ee19a5d94d8 9185Author: Dave Airlie <airlied@redhat.com> 9186Date: Mon Aug 17 15:31:16 2009 +1000 9187 9188 atom: recall crtc source script after tv dpms on 9189 9190 For some reason we lost the DACB source when we dpms tv back on. 9191 9192commit a971f1cfbbda1e3d16fb41f4a07f3b43fd59bc0d 9193Author: Dave Airlie <airlied@redhat.com> 9194Date: Mon Aug 17 15:25:17 2009 +1000 9195 9196 atombios: store dac detect result into save register. 9197 9198 We end up writing back an empty save register over the dac 9199 detection results so the atombios table can no longer 9200 figure out the S-video vs composite so sets the wrong one up. 9201 9202 this gets tv-out on my r580 with a composite connector working. 9203 9204commit efef7c01ad38e078de2fa3f9e528e4ef7d05d00a 9205Author: Dave Airlie <airlied@redhat.com> 9206Date: Mon Aug 17 13:44:20 2009 +1000 9207 9208 atom/tvout: set the correct scaler up for tv out 9209 9210 If we need to use the second scaler actually set the second scaler 9211 up. 9212 9213commit 032a8a06315d12e66240e7ef74b5b165f6d0d1a9 9214Author: Dave Airlie <airlied@redhat.com> 9215Date: Mon Aug 17 12:18:23 2009 +1000 9216 9217 kms: make tv out match the connector name 9218 9219commit 36e51c4d63836863dd7a17cbf6e5a9b7c1bbd31a 9220Author: Dave Airlie <airlied@redhat.com> 9221Date: Mon Aug 17 12:17:31 2009 +1000 9222 9223 r100/r200: dont emit wrong clamp modes. 9224 9225 for rect textures you don't want to use wrap clamping which 9226 is the default. 9227 9228commit 595f7c3a60ca982ca2836a0174e1f36f9d1a4fec 9229Author: Dave Airlie <airlied@itt42.(none)> 9230Date: Mon Aug 17 09:38:09 2009 +1000 9231 9232 radeon/tv: fix typo in crtc register modifications 9233 9234commit 92baeafde8f0c1eed2b6769e96f8093e3d6d7478 9235Author: Alex Deucher <alexdeucher@gmail.com> 9236Date: Tue Aug 11 15:29:17 2009 -0400 9237 9238 RV280: add agp quirk 9239 9240 from Hartmut Niemann 9241 9242commit 04288fdb8b8c62fc2eb32fd467fdf05151ad121a 9243Author: Jerome Glisse <jglisse@redhat.com> 9244Date: Tue Aug 11 19:32:52 2009 +0200 9245 9246 radeon/r600: abstract vertex buffer for r600 & r700 hardware 9247 9248 Abstract vertex buffer, preliminary work before others change 9249 to bring cs support along IB support. 9250 9251commit 72daee2a18bfb809260d1bcbe80cc14afea3b0fa 9252Author: Adam Jackson <ajax@redhat.com> 9253Date: Mon Aug 10 13:59:49 2009 -0400 9254 9255 Fix misleading filenames in DRI setup failure messages 9256 9257commit cd99d9f0d715f1f74de4fe22f2fd30046f2c7568 9258Author: Michel Dänzer <daenzer@vmware.com> 9259Date: Fri Aug 7 11:56:50 2009 +0200 9260 9261 Properly let radeon_cs_flush_indirect() re-emit 2D state if necessary. 9262 9263 Also remove a superfluous assignment. 9264 9265commit 577ff3ce922e457cc32f80d4365cb1da81552e72 9266Author: Michel Dänzer <daenzer@vmware.com> 9267Date: Fri Aug 7 11:56:49 2009 +0200 9268 9269 Add support for EXA_MIXED_PIXMAPS in xserver master. 9270 9271commit e755fa56d03aa338d0c6345ed41e32aa3115ad4b 9272Author: Michel Dänzer <daenzer@vmware.com> 9273Date: Fri Aug 7 11:56:49 2009 +0200 9274 9275 Minor fixes for KMS EXA DownloadFromScreen hook. 9276 9277 Check we can handle the bpp, and remove superfluous flush. 9278 9279commit 4cebafae81c1738a1c330d52a2c5248869f06411 9280Author: Alex Deucher <alexdeucher@gmail.com> 9281Date: Fri Aug 7 11:56:40 2009 +0200 9282 9283 Add KMS EXA UploadToScreen hook. 9284 9285 Fixups by Michel Dänzer: Doesn't seem to be a win in terms of raw numbers, but 9286 should allow more pipelining and avoiding BO memory waste for small glyph 9287 pixmaps. 9288 9289commit 9243791322e36b9231e6a3f04024ad66325385e3 9290Author: Dave Airlie <airlied@redhat.com> 9291Date: Fri Aug 7 14:07:32 2009 +1000 9292 9293 r200: fixup scissors for DDX. 9294 9295 a) turn of R200_RE_CNTL - SCISSOR_ENABLE - this save us emitting R200_RE_TOP_LEFT, note scissor is still enabled. 9296 b) disable aux scissors. 9297 9298commit bd03977e320591ca55b1a2fbb32414c53cb3f72e 9299Author: Alex Deucher <alexdeucher@gmail.com> 9300Date: Tue Aug 4 15:21:44 2009 -0400 9301 9302 radeon: fix compile with !kms 9303 9304commit 4fc1e67e606daf9227e74518c54315b60cd38301 9305Author: Alex Deucher <alexdeucher@gmail.com> 9306Date: Tue Aug 4 15:17:39 2009 -0400 9307 9308 radeon: re-add asus connector quirks 9309 9310 fixes bug 19943 9311 9312commit 22074cf0e58fddba743924532625e6fca49b6bdc 9313Author: Jerome Glisse <jglisse@redhat.com> 9314Date: Tue Aug 4 21:09:25 2009 +0200 9315 9316 radeon/kms: add simple DownloadFromScreen implementation 9317 9318 What we want to do is add userspace object support to radeon 9319 kernel modesetting. Also this DFS is dumb and might endup doing 9320 blit from GTT to GTT. 9321 9322commit fce31b61a88522733863a9b4e9f1c935c439cb4e 9323Author: Alex Deucher <alexdeucher@gmail.com> 9324Date: Tue Aug 4 11:16:05 2009 -0400 9325 9326 RS880: enable accel 9327 9328commit 6b1b4b9214d58175727f7992fc42bc7244c42bf2 9329Author: Alex Deucher <alexdeucher@gmail.com> 9330Date: Mon Aug 3 15:55:04 2009 -0400 9331 9332 radeon: add some new r7xx pci ids 9333 9334commit f564460e94c9d0f1cf3ff4b8535481b2b8b4e9c1 9335Author: Alex Deucher <alexdeucher@gmail.com> 9336Date: Sat Aug 1 16:53:47 2009 -0400 9337 9338 radeon: use XAA in some cases 9339 9340 Use XAA in low memory situations or when the DRI is disabled. 9341 Using shadowfb might also be a viable option, maybe even a better option... 9342 fixes bug 21683 9343 9344commit c71b2f050e8996787eae463eddbfdb5864ffa65a 9345Author: Bryce Harrington <bryce@canonical.com> 9346Date: Sat Aug 1 16:50:43 2009 -0400 9347 9348 radeon: AGPMode quirk needed for SiS 9349 9350 fixes bug 23065 9351 9352commit e3659ed06fc5bb8817f1dbd7c2d6bc94c67b30f7 9353Author: Bryce Harrington <bryce@canonical.com> 9354Date: Sat Aug 1 16:48:48 2009 -0400 9355 9356 radeon: AGPMode quirk needed for IBM Thinkpad T40 with Mobility M7 LW 9357 9358 fixes bug 23064 9359 9360commit 2391531ed6b7c11ddd5ab91b2369821cc5f8b8a7 9361Author: Bryce Harrington <bryce@canonical.com> 9362Date: Sat Aug 1 16:41:41 2009 -0400 9363 9364 radeon: AGPMode quirk needed for HP Omnibook 6200 9365 9366 fixed bug 23063 9367 9368commit 474eda02257152ced52364f38cbad24c20aebbc0 9369Author: Alex Deucher <alexdeucher@gmail.com> 9370Date: Sat Aug 1 16:06:52 2009 -0400 9371 9372 radeon: reload bicubic Xv texture on VT switch 9373 9374 fixed bicubic filtering after VT switch or suspend/resume 9375 9376commit 447a2ce1b88aa2d6d5713e93174c4002617720f7 9377Author: Dave Airlie <airlied@redhat.com> 9378Date: Thu Jul 30 09:54:01 2009 +1000 9379 9380 radeon: fix r600 shadow fb mode with cursor allocation in wrong place 9381 9382 From RHEL QE testing, we could end up with the cursor at 0 since 9383 we think EXA is in use when really it isn't. the info->useEXA = FALSe 9384 might be unnecessary but better to be explicit 9385 9386 Signed-off-by: Dave Airlie <airlied@redhat.com> 9387 9388commit 2b9a5be3bb22fad1d52ec9eea7733c5e956250b7 9389Author: Alex Deucher <alexdeucher@gmail.com> 9390Date: Wed Jul 29 03:34:11 2009 -0400 9391 9392 radeon: fix typo in object header to connector type conversion 9393 9394 Should fix bug 19943 9395 9396commit 63c873cbd4d1d21d9f688028c0900c79fadc42c1 9397Author: Dave Airlie <airlied@redhat.com> 9398Date: Tue Jul 28 15:22:40 2009 +1000 9399 9400 ati: change to using ABI version check 9401 9402commit 21a621c297ac71c65c239ea960c38706e718b91c 9403Author: Dave Airlie <airlied@redhat.com> 9404Date: Tue Jul 28 13:32:28 2009 +1000 9405 9406 ati: update for resources/RAC API removal 9407 9408commit 5ad2519c2a7c0df389b2cd7cf7151c7e4b7252a8 9409Author: Michel Dänzer <daenzer@vmware.com> 9410Date: Mon Jul 27 14:22:02 2009 +0200 9411 9412 KMS: Explicitly specify VRAM and GTT domains when allocating BOs for pixmaps. 9413 9414 Without this, apparently they are initially allocated in system RAM (or 9415 possibly GTT), so any GPU rendering to them first triggers a copy of the whole 9416 uninitialized data to VRAM... 9417 9418 We may want to explore more sophisticated schemes in the future, but for now 9419 this seems to improve KMS 2D performance quite a bit. 9420 9421commit 57f2c83a22f27567506c555af431f89e6031204c 9422Author: Alex Deucher <alexdeucher@gmail.com> 9423Date: Sat Jul 25 14:19:38 2009 -0400 9424 9425 radeon: Set PCI/PCIE bus type properly per asic 9426 9427 - r1xx-rv350 chips have the old pci gart 9428 - rv380+ chips have newer pcie gart 9429 9430 Select the right kind regardless of whether the user selects 9431 PCI or PCIE. 9432 9433commit 2afc46fa74ce7730f766a3370d323c6b59694186 9434Author: Jerome Glisse <jglisse@redhat.com> 9435Date: Thu Jul 23 20:13:15 2009 +0200 9436 9437 radeon: fix KMS shadowfb for r6xx and newer hw. 9438 9439 We need to map front buffer and also to avoid EXA initialization 9440 when falling back to shadowfb on newer r6xx/r7xx when KMS is enabled. 9441 9442commit 328e4c816348b6f867df30b39856dbe78c8e0dcc 9443Author: Alex Deucher <alexdeucher@gmail.com> 9444Date: Thu Jul 23 13:28:36 2009 -0400 9445 9446 RS780: fill in MC access functions 9447 9448 Also, r6xx/r7xx don't have indirect MC space, so 9449 don't try to access it. 9450 9451commit f96d5b255425fbd02be2cad26edb590d474a5640 9452Author: Alex Deucher <alexdeucher@gmail.com> 9453Date: Thu Jul 23 13:25:04 2009 -0400 9454 9455 R6xx/R7xx: clip rendering to destination surface dimensions 9456 9457commit e372f845b0defaf2d2c9ef3cbbf7498e09d9372e 9458Author: Michel Dänzer <daenzer@vmware.com> 9459Date: Sun Jul 19 16:05:29 2009 +0200 9460 9461 Guard reference to CRTC active field. 9462 9463 Fixes build against older xserver. 9464 9465commit e38305aebdc95f80f5b4b3e5ba541ea67dc05f01 9466Author: Michel Dänzer <daenzer@vmware.com> 9467Date: Sat Jul 18 22:23:45 2009 +0200 9468 9469 Also drop DRM master in KMS CloseScreen. 9470 9471 The LeaveVT hook isn't always called when the server dies, e.g. when quitting 9472 from the GDM greeter. This may cause existing servers to fall over if the dying 9473 server process still exists when they try to re-acquire master and set a mode. 9474 9475 Also use drmSet/DropMaster() rather than ioctl() directly. 9476 9477commit a43c660a00147bfae5ca601f4720b2680b75211f 9478Author: Michel Dänzer <daenzer@vmware.com> 9479Date: Sat Jul 18 22:21:59 2009 +0200 9480 9481 Set CRTC active flag to TRUE after a successful KMS mode set. 9482 9483 Otherwise some things like changing the colour map won't work properly. 9484 9485commit f32069f42b3e4643f15148d0b96164def00dcc74 9486Author: Michel Dänzer <daenzer@vmware.com> 9487Date: Sat Jul 18 22:20:08 2009 +0200 9488 9489 Set the STALL bit in the CRTC_GUI_TRIG_VLINE registers. 9490 9491 As recommended by the register reference when using the WAIT_CRTC_VLINE bit in 9492 the WAIT_UNTIL register, as we are. 9493 9494commit 27bd9fc9b2c0eeff488c1f26f1355a7fa3655520 9495Author: Michel Dänzer <daenzer@vmware.com> 9496Date: Sat Jul 18 22:14:27 2009 +0200 9497 9498 Enable vsync for DRI2 region copies. 9499 9500commit 4b1b28865ffb17f58a70d955212fd51cb1fb55ab 9501Author: Alex Deucher <alexdeucher@gmail.com> 9502Date: Thu Jul 16 17:09:33 2009 -0400 9503 9504 R6xx/r7xx: reduce the amount of default state setup 9505 9506 Most of the default state setup was extraneous or duplicated 9507 in the accel ops. There's no need to reset all the consts 9508 everytime you reset the default state for example. This leads 9509 to a 10x reduction in default state. 9510 9511commit 7e6557b81026d8a4d6c837839e68c28b151f8a97 9512Author: Peter Hutterer <peter.hutterer@who-t.net> 9513Date: Thu Jul 16 11:33:57 2009 +1000 9514 9515 Update to xextproto 7.1 support. 9516 9517 DPMS header was split into dpms.h (client) and dpmsconst.h (server). Drivers 9518 need to include dpmsconst.h if xextproto 7.1 is available. 9519 9520 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 9521 9522commit 95d431e8b260dd43cf6d93b90eecab1dc0ec7c26 9523Author: Dave Airlie <airlied@redhat.com> 9524Date: Wed Jul 15 12:07:01 2009 +1000 9525 9526 radeon: emit colorpitch relocs. 9527 9528 This causes relocs to be emitted for the colorpitch. 9529 9530 Shouldn't have no effect on current kernels, but will keep DDX 9531 going on kms kernels when tiling lands 9532 9533 Also contains a missing reloc space for r200 9534 9535commit 8312763977b22e108b7d003cbf3256e0165e4439 9536Author: Alex Deucher <alexdeucher@gmail.com> 9537Date: Tue Jul 14 18:13:27 2009 -0400 9538 9539 Add an r420 AGP quirk 9540 9541 fixes bug 22726 9542 9543commit 5473eeeb45468fd200fec49a8506281628a5e4b5 9544Author: Alex Deucher <alexdeucher@gmail.com> 9545Date: Tue Jul 14 14:05:13 2009 -0400 9546 9547 R3/4/5xx: only upload the bicubic texture once 9548 9549 Upload the bicubic texture once during textured video init 9550 rather than once per frame. Suggested by Michel Daenzer on 9551 IRC. 9552 9553commit 0485f27bc3d75cb6ab320e8164dbe6ea2713c78e 9554Author: Alex Deucher <alexdeucher@gmail.com> 9555Date: Tue Jul 14 13:11:12 2009 -0400 9556 9557 R3/4/5xx: fix bicubic Xv filtering with KMS changes 9558 9559 fixes bug 22730 9560 9561commit ca4a9efef987108266ae59afbb5142d4d2000c8e 9562Author: Alex Deucher <alexdeucher@gmail.com> 9563Date: Mon Jul 13 11:28:41 2009 -0400 9564 9565 Add configure option to disable kms support 9566 9567 Useful to building with mixed kms/non-kms setups 9568 9569commit 035e8d1d5593c12828bb079de4e663cf1b1f1674 9570Author: Edward O'Callaghan <eocallaghan@auroraux.org> 9571Date: Sun Jul 12 13:45:29 2009 +0200 9572 9573 Fix a warning 9574 9575 Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com> 9576 9577commit 76af48c43f829e7aebacc9f2a623823fa26ee22b 9578Author: Gaetan Nadon <memsize@videotron.ca> 9579Date: Wed Jul 8 09:58:56 2009 -0400 9580 9581 xf86-video-ati: use XORG_CHANGELOG macro to create ChangeLog. #22611 9582 9583 Build break: Makefile.am: command not found: git-log 9584 Adding the macro in configure.ac and use it in Makefile.am 9585 Refer to: https://bugs.freedesktop.org/show_bug.cgi?id=22611 9586 Tested: running autogen.sh, make and 'make dist' 9587 9588 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 9589 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 9590 9591commit 43db263d301082e84e9bc304816bcbb206fe280e 9592Author: Alex Deucher <alexdeucher@gmail.com> 9593Date: Thu Jul 9 12:48:06 2009 -0400 9594 9595 RV280: another AGP quirk 9596 9597 fdo bug 12544 9598 9599commit 69b5e5496f10a9f566d2e563862c96cb41952eb6 9600Author: Alex Deucher <alexdeucher@gmail.com> 9601Date: Thu Jul 9 12:42:45 2009 -0400 9602 9603 RV280: Add an AGP quirk 9604 9605 lp bug 370205 via fdo bug 12544 9606 9607commit a6d6c900e60a3685ee9e93c34eb7f6d237c45fd8 9608Author: Alex Deucher <alexdeucher@gmail.com> 9609Date: Thu Jul 9 11:30:57 2009 -0400 9610 9611 RS740: fixup display latency setting as per rs690 9612 9613 Both chips program the same in this regard 9614 9615commit ea407570d39ace3162d372eda56bf791bfd80c24 9616Author: Dave Airlie <airlied@redhat.com> 9617Date: Thu Jul 9 10:46:23 2009 +1000 9618 9619 radeon: add shave support. 9620 9621 clean up output a lot 9622 9623commit 8c03c1fdb5ea35570064946557050c87ca30582a 9624Author: Alex Deucher <alexdeucher@gmail.com> 9625Date: Wed Jul 8 18:34:57 2009 -0400 9626 9627 R6xx/R7xx: fix hangs on x2 cards with PM options 9628 9629 Changing the PCIE lanes on x2 cards results in a hang, 9630 so for now, disable it. 9631 9632 Fixes fdo bug 22669 9633 9634commit 9645838c57f6b40837fdce23ce7f9faefb3d9966 9635Author: Michel Dänzer <daenzer@vmware.com> 9636Date: Wed Jul 8 19:18:02 2009 +0200 9637 9638 Further non-KMS fixes / cleanups. 9639 9640 Only compile tested. 9641 9642commit 2d86ec4ddb1fb83be95ed9cdcbabf4d568a6d7b6 9643Author: Alex Deucher <alexdeucher@gmail.com> 9644Date: Wed Jul 8 12:33:34 2009 -0400 9645 9646 Fix the build for the !XF86DRM_MODE case 9647 9648 compile tested only. 9649 9650commit a68afc2a80b8581cc3d67e9b5a95a10b0d4ead0e 9651Author: Michel Dänzer <daenzer@vmware.com> 9652Date: Wed Jul 8 09:15:28 2009 +0200 9653 9654 Fix RADEON_TRACE_FALL build. 9655 9656commit 9ca0aeaad01b7c682396b30af48668f287a59d6e 9657Author: Michel Dänzer <daenzer@vmware.com> 9658Date: Wed Jul 8 09:13:36 2009 +0200 9659 9660 Pass on CS relocation offset values. 9661 9662 Not actually used ATM, but just in case... 9663 9664commit f718dd794477fbdd5eeb4183f43ff25517c63034 9665Author: Michel Dänzer <daenzer@vmware.com> 9666Date: Wed Jul 8 09:11:18 2009 +0200 9667 9668 Don't set EXA UploadTo/DownloadFromScreen hooks when they'd always return FALSE. 9669 9670commit 42e76d0b24b76fbcc6b08a4ea5003dbec368645c 9671Author: Michel Dänzer <daenzer@vmware.com> 9672Date: Wed Jul 8 08:51:10 2009 +0200 9673 9674 Don't always flush CS when switching from 2D to 3D. 9675 9676 This works fine here and doubles x11perf -aa10text scores, but may cause 9677 problems on other setups according to Dave. Let's see... 9678 9679commit 94fe049c35187ddc3ec0472e9309c3a22879f415 9680Author: Michel Dänzer <daenzer@vmware.com> 9681Date: Wed Jul 8 08:36:22 2009 +0200 9682 9683 KMS CS fixes, take two. 9684 9685 Avoid flushing CS in the middle of (setting up state for) an operation, 9686 properly finish/restart the operation. 9687 9688commit 0519f15af2fe5e7e5a6c94e7203a96fe363c643e 9689Author: Michel Dänzer <daenzer@vmware.com> 9690Date: Tue Jul 7 10:04:09 2009 +0200 9691 9692 Fix DRI2 with current xserver Git. 9693 9694commit 7ad11fec2579807371360cb680e593f723ebaa1a 9695Author: Dave Airlie <airlied@redhat.com> 9696Date: Mon Jul 6 18:43:08 2009 +1000 9697 9698 radeon: add make dist headers 9699 9700commit e68fbf743ec88e0fa3a9f1ffe83daf251e250b21 9701Author: Dave Airlie <airlied@redhat.com> 9702Date: Mon Jul 6 18:32:32 2009 +1000 9703 9704 git: fix git-log to git log 9705 9706commit 5eeb1fd19c6797cc76ce308570dffe34cb82b24d 9707Author: Dave Airlie <airlied@linux.ie> 9708Date: Mon Jul 6 17:45:50 2009 +1000 9709 9710 radeon: oops fix compile without libdrm 9711 9712commit 98824f37fc74dc1fec2677515d0dc3cfc263f00a 9713Author: Dave Airlie <airlied@redhat.com> 9714Date: Mon Jul 6 15:29:53 2009 +1000 9715 9716 radeon: move state changes into flush function 9717 9718commit 503b51343267248b4c6994b88238dd5384528f03 9719Author: Dave Airlie <airlied@redhat.com> 9720Date: Mon Jul 6 15:20:13 2009 +1000 9721 9722 radeon: taken from kms-support + CS fixes commit. 9723 9724 re-emit 2D state and reset 3D state emit on flush 9725 9726commit d78d50591851d14543f0935d051a59ef29751bbc 9727Author: Dave Airlie <airlied@redhat.com> 9728Date: Mon Jul 6 14:45:07 2009 +1000 9729 9730 radeon: port to new space checking in libdrm interface 9731 9732 This uses a new libdrm interface which shares code with mesa. 9733 9734 It also fixes the bo to flush when full instead of never flushing. 9735 9736 It survives gtkperf -a here which the driver didn't before now 9737 9738commit e1200cb89218930d01330ba0114e013438655cce 9739Author: Dave Airlie <airlied@redhat.com> 9740Date: Mon Jul 6 15:10:29 2009 +1000 9741 9742 Revert "KMS CS fixes." 9743 9744 This reverts commit 11cf8a28eb46f48d101888552266921bee8b63fc. 9745 9746 I'll bring this back in a few minutes, I've got a big change introducing 9747 a new cs space accounting scheme which this messes up and I think 9748 probably conflicts with what I've done. 9749 9750 I promise i'll go over this and pick the bits that are still useful :) 9751 9752commit 11cf8a28eb46f48d101888552266921bee8b63fc 9753Author: Michel Dänzer <daenzer@vmware.com> 9754Date: Sun Jul 5 15:14:59 2009 +0200 9755 9756 KMS CS fixes. 9757 9758 * Flush CS before running out of space. 9759 * Don't flush CS every time when switching between 2D and 3D engine, but make 9760 sure to re-initialize 2D/3D state after flushing. 9761 9762 Doesn't seem quite perfect yet (some rendering seems to get dropped on the floor 9763 sometimes), but this should fix a lot of CS failures people have been seeing. 9764 9765commit 8d2f712eaf1e569fd92bbe2db5aceb43b7b367d1 9766Author: Dave Airlie <airlied@linux.ie> 9767Date: Fri Jul 3 10:40:59 2009 +1000 9768 9769 radeon: don't set up overlay under KMS until we figure out how. 9770 9771commit 2a622168606d98c404708b879e6417c8f7d7305c 9772Author: Michel Dänzer <daenzer@vmware.com> 9773Date: Thu Jul 2 15:03:53 2009 +0200 9774 9775 Test BO pointer before passing it to radeon_bo_unref(). 9776 9777commit 36a4ea5a9c7062f4c8aa7480bb7488ce2062390f 9778Author: Michel Dänzer <daenzer@vmware.com> 9779Date: Thu Jul 2 14:52:44 2009 +0200 9780 9781 Warning fixes. 9782 9783commit 65465d56c1bf943113518a5b2f57b3170213bab9 9784Author: Dave Airlie <airlied@linux.ie> 9785Date: Thu Jul 2 18:38:38 2009 +1000 9786 9787 radeon: fix up configure check better 9788 9789commit ad17108b72c9adf773a122b4c054c7b92e6a8b9d 9790Author: Dave Airlie <airlied@linux.ie> 9791Date: Thu Jul 2 17:25:57 2009 +1000 9792 9793 radeon: fix it properly. 9794 9795 I blame the flu 9796 9797commit e1b9cb9ea665eaa84e64a386299efa0819d55df4 9798Author: Dave Airlie <airlied@redhat.com> 9799Date: Thu Jul 2 17:18:47 2009 +1000 9800 9801 radeon: fix compile without kms 9802 9803commit de427fac6912a625f7e4d8b982a1f7a1ba80dfbe 9804Author: Dave Airlie <airlied@redhat.com> 9805Date: Thu Jul 2 15:41:43 2009 +1000 9806 9807 radeon/dri2: fix dri2 - needs newest libdrm_radeon 9808 9809commit 2b671adba54ebff1756dd990123f4bade6e35d61 9810Author: Dave Airlie <airlied@redhat.com> 9811Date: Thu Jul 2 15:25:19 2009 +1000 9812 9813 radeon/dri2: call proper gem bo function 9814 9815commit 851b2b8cab883e4422745cca1827aa8ac9f193a3 9816Author: Dave Airlie <airlied@redhat.com> 9817Date: Thu Jul 2 15:17:13 2009 +1000 9818 9819 radeon/kms: add textured video BO size checks 9820 9821commit 9d092060236e8587697fdd57a7a01f3c0dddd680 9822Author: Dave Airlie <airlied@redhat.com> 9823Date: Thu Jul 2 15:17:03 2009 +1000 9824 9825 kms: drop DGA 9826 9827commit 207174ea964c561076bdd5297a2fb6f136b4b645 9828Author: Dave Airlie <airlied@redhat.com> 9829Date: Thu Jul 2 15:16:47 2009 +1000 9830 9831 radeon: fix CS imbalance 9832 9833commit 4d24a9d2c0be0c025266cc8cdf7e8d7bb065cdb1 9834Author: Dave Airlie <airlied@redhat.com> 9835Date: Thu Jul 2 14:55:45 2009 +1000 9836 9837 radeon: enable KMS build 9838 9839commit 9bcd3ae7966eb7a17b922bcd33e0d6f016245614 9840Author: Dave Airlie <airlied@redhat.com> 9841Date: Thu Jul 2 14:55:18 2009 +1000 9842 9843 radeon/kms: fixup 3d state emission for KMS 9844 9845 This fixes font rendering under KMS (should have looked at dmesg) 9846 9847commit ac1607bb23f0c7584648407a3ac3fb27142e186e 9848Author: Dave Airlie <airlied@redhat.com> 9849Date: Thu Jul 2 14:55:05 2009 +1000 9850 9851 kms: no need to reference fifo slots 9852 9853commit 1782ce28953184776c90eb1255208a3e0ad245f0 9854Author: Dave Airlie <airlied@redhat.com> 9855Date: Tue Jun 30 16:55:26 2009 +1000 9856 9857 radeon: add KMS support (still disabled) 9858 9859 This adds DRI2 + KMS + driver pixmaps support to the driver. 9860 9861 I've decided to just do a completely separate KMS driver file 9862 instead of hacking the crap out of radeon_driver.c. So now 9863 I do the KMS check in radeon_probe.c time and set the DDX 9864 pointed up to a completely different set at this stage. 9865 9866 This avoids a lot of if (kms) type crap in the code at 9867 the expense of making sure we make changes to both files 9868 if necessary. 9869 9870 This code is still disabled in configure.ac as I broke EXA composite 9871 rendering somehow in KMS mode 9872 9873commit ac4bd24a66c1bdda0293f770a3f891e2b88cc8ee 9874Author: Alex Deucher <alexdeucher@gmail.com> 9875Date: Tue Jun 30 16:00:06 2009 -0400 9876 9877 EXA: fix the build on servers without EXA_HANDLES_PIXMAPS 9878 9879commit 49a0ca19ee4109cdd39a43fbf4bedf3f0557d6ec 9880Author: Michel Dänzer <daenzer@vmware.com> 9881Date: Tue Jun 30 17:22:50 2009 +0200 9882 9883 Warning fixes. 9884 9885commit e932836691aeaec37794fdaed2dabb22710fd171 9886Author: Dave Airlie <airlied@redhat.com> 9887Date: Tue Jun 30 16:24:37 2009 +1000 9888 9889 radeon: initial preparation for kms patch. 9890 9891 This patch contains most of the changes to the EXA and texture video 9892 accel code. 9893 9894 It adds a few bits of pixmap support but doesn't actually do anything 9895 useful KMS yet. 9896 9897 Testing this should not have any regressions over what we have already, 9898 biggest worries are r6xx, I've fixed a textured video one, but no idea 9899 what other might lurk 9900 9901 It won't build against libdrm radeon yet either 9902 9903commit bb04b450ed00ca4b1aa44c33085567d47b33b547 9904Author: Fredrik Höglund <fredrik@kde.org> 9905Date: Mon Jun 29 22:38:26 2009 +0200 9906 9907 Check if the composite op is supported in R200CheckComposite. 9908 9909commit f39cafc5c4a93a16ac93756473ebb1f33c7881db 9910Author: Alex Deucher <alexdeucher@gmail.com> 9911Date: Sun Jun 28 20:42:49 2009 -0400 9912 9913 radeon: call hdmi-b connectors dvi in randr 9914 9915 Technically HDMI-B is a glorfied DL DVI, but this is confusing 9916 to users, so call it DVI when reporting randr outputs. This 9917 allows us to remove two quirks where the bios was actually 9918 correct. 9919 9920commit 407e676b04117c7ca3dcc0cc072f3552b03c64e5 9921Author: Roland Scheidegger <sroland@tungstengraphics.com> 9922Date: Sun Jun 28 16:54:32 2009 -0400 9923 9924 R6xx/R7xx: fix pixel centers 9925 9926 Make sure we are using GL mode (centers at 0.5) rather 9927 than D3D mode (centers at 0.0). This also fixes fdo 9928 bug 21963. 9929 9930commit 68001981f22173ff949720055dba89291f284474 9931Author: Alex Deucher <alexdeucher@gmail.com> 9932Date: Sun Jun 28 16:20:10 2009 -0400 9933 9934 Add some missing M96 pci ids 9935 9936 fixes fdo bug 22404 9937 9938commit 06d5d8d7fd92b58b2467c782e57a80a3520a2c90 9939Author: Adam Jackson <ajax@redhat.com> 9940Date: Thu Jun 25 13:51:36 2009 -0400 9941 9942 Add a (harmlessly) missing break statement. 9943 9944commit 7d1d968fdc8de1e0ab0be916a1f3a4a398af47aa 9945Author: Adam Jackson <ajax@redhat.com> 9946Date: Thu Jun 25 13:49:18 2009 -0400 9947 9948 Cosmetic cleanup to digital output detection 9949 9950 No functional change, but this logic would need extending anyway for 9951 DisplayID, so unify it now. 9952 9953commit 96581746eea2f4e29a04f9266ec9649f25480424 9954Author: Adam Jackson <ajax@redhat.com> 9955Date: Wed Jun 24 17:05:47 2009 -0400 9956 9957 atom: Fix phantom VGA connector when HDMI-B present. 9958 9959 HDMI-B is really a glorified DL-DVI connector, the analog pins work fine. 9960 9961commit 496adc4dc7cf53fc6acfecb1158f1491d5f36168 9962Author: Michel Dänzer <daenzer@vmware.com> 9963Date: Fri Jun 19 14:37:01 2009 +0200 9964 9965 Revert "Build fixes for -Werror that aren't ready for prime time yet." 9966 9967 This reverts commit ab20ea1530ec8a3a9fd4089775c4e157e50fe1bf. 9968 9969 Whoops, pushed wrong branch... 9970 9971commit ab20ea1530ec8a3a9fd4089775c4e157e50fe1bf 9972Author: Michel Dänzer <daenzer@vmware.com> 9973Date: Sun Jun 14 16:50:07 2009 +0200 9974 9975 Build fixes for -Werror that aren't ready for prime time yet. 9976 9977commit 77e3537d312175a25f0e21cc07c3a96f78c3b35a 9978Author: Markus Gapp <markus.gapp@gmx.net> 9979Date: Sat Jun 13 11:51:29 2009 -0400 9980 9981 Add quirk for asus hd3450 board 9982 9983 Fixes bug 22266 9984 9985commit e8e0bb1ba29761c94d473e42ca6ada161cfff557 9986Author: Alex Deucher <alexdeucher@gmail.com> 9987Date: Fri Jun 12 11:07:43 2009 -0400 9988 9989 Fix count in 3d state setup after 43374c7420e378918bec062f4cbd581f16adb6f0 9990 9991commit 808c90a24c48da7fa97e15e2f12be5bb8fd8cc96 9992Author: Alex Deucher <alexdeucher@gmail.com> 9993Date: Thu Jun 11 16:09:40 2009 -0400 9994 9995 Only enable frac fb divs on rs600/rs690/rs740 for now 9996 9997 fixes bug 21413 9998 9999commit d095bb3157faaee10b0130ea9b1032e0b533aaee 10000Author: Alex Deucher <alexdeucher@gmail.com> 10001Date: Thu Jun 11 12:34:15 2009 -0400 10002 10003 Re-enable frac feedback dividers on AVIVO hardware 10004 10005 Preferring a freq <= target frequency seems to fix the 10006 previous issues with frac fb divs. Appears to work fine 10007 on all hardware I've tested on. See fdo bug 22229 for 10008 details. Should fix bugs: 22229, 21553, 21413, and 10009 possibly other ones related to unstable images at certain 10010 resolutions. 10011 10012commit 55fbdbae83d1563b472f49d0436c9298e390be66 10013Author: Alex Deucher <alexdeucher@gmail.com> 10014Date: Thu Jun 11 12:31:53 2009 -0400 10015 10016 Add PLL flag to prefer frequencies <= the target freq 10017 10018 This appears to be needed when using fractional feedback 10019 dividers. Based on a patch from Tom Hirst. See fdo 10020 bug 22229 for more details. 10021 10022commit 43374c7420e378918bec062f4cbd581f16adb6f0 10023Author: Alex Deucher <alexdeucher@gmail.com> 10024Date: Tue Jun 9 14:39:54 2009 -0400 10025 10026 r1xx/r2xx: clip 3D rendering to destination buffer 10027 10028commit 74b0a38c75c1d6a6344015d32d84b43d9e80df1d 10029Author: Alex Deucher <alexdeucher@gmail.com> 10030Date: Tue Jun 9 14:26:21 2009 -0400 10031 10032 Consolidate indentical RADEONDoneSolid/RADEONDoneCopy functions 10033 10034commit f6dff8836729cc9d29264080c9cfa0e1df6182cf 10035Author: Alex Deucher <alexdeucher@gmail.com> 10036Date: Tue Jun 9 13:31:15 2009 -0400 10037 10038 r3xx/r4xx/r5xx: clip 3D rendering to destination buffer 10039 10040commit 48dbf09e1309c815ac8b733145a0310083a29933 10041Author: Alex Deucher <alexdeucher@gmail.com> 10042Date: Tue Jun 9 13:12:40 2009 -0400 10043 10044 r3xx/r4xx/r5xx: refactor exa prepare composite 10045 10046 - consolidate code duplicated between r3xx/r4xx and r5xx 10047 - make pixel shader setup easier to read and less 10048 redundant 10049 10050commit ec1a3ccdf61c2226617b571cc4a209a542e9d7e3 10051Author: Alex Deucher <alexdeucher@gmail.com> 10052Date: Mon Jun 8 15:31:57 2009 -0400 10053 10054 R6xx/r7xx: don't clamp tex coords in composite VS 10055 10056 Fixes broken repeat modes. see fdo bugs 21818, 22139 10057 10058commit 32c218c5633fb7ffeb19e717bf645c1b25dd4ef5 10059Author: Alan Coopersmith <alan.coopersmith@sun.com> 10060Date: Thu Jun 4 07:23:03 2009 -0700 10061 10062 Add link to patch submission instructions to README 10063 10064 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 10065 10066commit 39f736d6cb65fc7c3fd769478023fc87c1af5979 10067Author: Alex Deucher <alexdeucher@gmail.com> 10068Date: Thu Jun 4 09:57:22 2009 -0400 10069 10070 atom: connector quirk 10071 10072 Some r6xx cards have a VGA and an HDMI port with 10073 a shared ddc line listed as an HDMI port with both 10074 analog and digital encoders. 10075 10076 Fixes bug 19943 10077 10078commit 7599dc40855c33a5fbd8e9bbc2b4cd62752fb7df 10079Author: Alex Deucher <alexdeucher@gmail.com> 10080Date: Fri May 29 16:29:50 2009 -0400 10081 10082 Add RV740 (HD4770) support 10083 10084 Verified by ernstp on IRC 10085 10086commit ee0424078a3ea26f58c2998c9e58e065fb63f1a0 10087Author: Alex Deucher <alexdeucher@gmail.com> 10088Date: Wed May 27 18:37:58 2009 -0400 10089 10090 radeon: remove dead code 10091 10092commit b34df233115c0d82d7bcf82e041afbc55981ce82 10093Author: Alex Deucher <alexdeucher@gmail.com> 10094Date: Tue May 26 17:47:01 2009 -0400 10095 10096 R3xx/R4xx: increase EXA render target limits 10097 10098 now that we are using 1/12 subpixel mode 10099 10100commit 1457bc089d1395d18cf582a4d52444005f013753 10101Author: Matthieu Herrb <matthieu.herrb@laas.fr> 10102Date: Tue May 26 17:21:07 2009 -0400 10103 10104 radeon: fix segfault if Option "NoAccel" is set 10105 10106commit 4079f97958cf1d5fd76a83aa61b93d6289bb9b1e 10107Author: Michel Dänzer <daenzer@vmware.com> 10108Date: Sat May 23 13:51:00 2009 +0200 10109 10110 Bump version to mark Git snapshot. 10111 10112 The fourth version component isn't visible in the log file. 10113 10114commit 88a9e98341d96e5e7f48b69aed597d1bada6313a 10115Author: Alex Deucher <alexdeucher@gmail.com> 10116Date: Tue May 19 17:11:34 2009 -0400 10117 10118 Pre-atom: Improve engine clock setup code 10119 10120 Also remove SetMemoryClk() for pre-atom cards for now 10121 as it requires quite a bit more asic specific work. 10122 To set the mclk we'll need to use the mem reset/dll tables 10123 in the bios. 10124 10125commit 1429133ffdbfe046249fcaaa4cbee432a1012e55 10126Author: Michel Dänzer <daenzer@vmware.com> 10127Date: Mon May 18 17:59:00 2009 +0200 10128 10129 EXA: Set new EXA_SUPPORTS_OFFSCREEN_OVERLAPS flag when we have a 2D engine. 10130 10131commit e9cca1e7dc521d3756bb2d97f1d703b9ad65cb47 10132Author: Alex Deucher <alexdeucher@gmail.com> 10133Date: Mon May 18 10:48:25 2009 -0400 10134 10135 r4xx: Add some missing pci ids 10136 10137 Noticed by Romain in bug 21798 10138 10139commit 76fffd51027820b3a6ea10de2be67d5fd4f3e843 10140Author: Alex Deucher <alexdeucher@gmail.com> 10141Date: Fri May 15 16:35:13 2009 -0400 10142 10143 R100: switch to normalized texture coordinates 10144 10145 for consistency with the other asics 10146 10147commit 1a1eef31e36f98b7a30e27e567c1b66e0e4cf7dc 10148Author: Alex Deucher <alexdeucher@gmail.com> 10149Date: Thu May 14 15:48:54 2009 -0400 10150 10151 radeon: try harder when checking if a card is posted 10152 10153 fixes debian bug 524280, possibly others. 10154 10155commit 22e39392297fa11003df90c175db3c449d8f9853 10156Author: Michel Dänzer <daenzer@vmware.com> 10157Date: Thu May 14 10:24:07 2009 +0200 10158 10159 Pre-R600 EXA: Fix coordinate limits off-by-one error. 10160 10161 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=21598 . 10162 10163commit fa09b058c7a17689989e600ffd465856a058579d 10164Author: Alex Deucher <alexdeucher@gmail.com> 10165Date: Wed May 13 16:10:40 2009 -0400 10166 10167 R6xx/R7xx Xv: normalize texture coordinates in the vertex shader 10168 10169commit 026b6f820d6caea17d2a082193e850713d5770a8 10170Author: Alex Deucher <alexdeucher@gmail.com> 10171Date: Wed May 13 15:48:32 2009 -0400 10172 10173 R6xx/R7xx: do EXA transforms in the vertex shader 10174 10175commit cd89241396d1931b04cfbdd8d553be16dbf9c360 10176Author: Alex Deucher <alexdeucher@gmail.com> 10177Date: Tue May 12 17:30:02 2009 -0400 10178 10179 R3xx-R5xx: do EXA transforms in the vertex shader 10180 10181commit 15ee78d37f8b64c3e6d234e7ab37a31e82327b6b 10182Author: Alex Deucher <alexdeucher@gmail.com> 10183Date: Mon May 11 02:22:13 2009 -0400 10184 10185 RV770: add missing pci id 10186 10187 Fixes bug 21209 10188 10189commit 5ff51ea83f9fdd836e830c77e3561eaef4b83183 10190Author: Alex Deucher <alexdeucher@gmail.com> 10191Date: Fri May 8 13:19:26 2009 -0400 10192 10193 R7xx: add support for reading disabled bios images 10194 10195commit 76fdae9018330aa5af0d57339baff3ca4834e1b5 10196Author: Alex Deucher <alexdeucher@gmail.com> 10197Date: Thu May 7 02:59:32 2009 -0400 10198 10199 R3xx/R4xx tex vid: no need to check if IS_R300_3D 10200 10201 function is r3xx/r4xx specific already 10202 10203commit 23a8ada06a5b1ccc975e89f129c00382ecbf29df 10204Author: Alex Deucher <alexdeucher@gmail.com> 10205Date: Thu May 7 02:34:22 2009 -0400 10206 10207 R3xx/R4xx tex vid: increase the guardband limit for rendering with a tri 10208 10209 limit goes from 2880 to 4021 now that we've switched to 1/12 subpixel mode. 10210 10211 This allows us to render with a clipped tri at higher resolutions preventing 10212 diagonal tearing. 10213 10214commit 0c3541f7013c53fb8096ada63cc7b3577524aa4c 10215Author: Alex Deucher <alexdeucher@gmail.com> 10216Date: Wed May 6 12:24:35 2009 -0400 10217 10218 AVIVO: set default cursor mode in cursor init 10219 10220 Should fix bug 21557 10221 10222commit da58e351b6398fa18b1d9c1a57a6e49b285f610f 10223Author: Alex Deucher <alexdeucher@gmail.com> 10224Date: Mon May 4 16:03:53 2009 -0400 10225 10226 AVIVO: fix cursoe corruption bands for real 10227 10228 Don't have to leave both cursors enabled, just have to use 10229 the same mode for both cursors whether or not they are enabled. 10230 10231commit 2c8e130f73c680d4a7381b2ef37982b82c6ee478 10232Author: Alex Deucher <alexdeucher@gmail.com> 10233Date: Mon May 4 13:40:17 2009 -0400 10234 10235 radeon: switch to EXA by default 10236 10237commit 8155a62df6817358a12cc2d613c1470ca55996ef 10238Author: Alex Deucher <alexdeucher@gmail.com> 10239Date: Mon May 4 12:57:36 2009 -0400 10240 10241 AVIVO: move cusor offscreen when disabling 10242 10243 Adapted from Yang's patch. Setting size to zero is 10244 actually a 1x1 cursor. 10245 10246commit a1c64ea5224009779ccad66b0f84d861eae966ac 10247Author: Alex Deucher <alexdeucher@gmail.com> 10248Date: Sun May 3 19:42:44 2009 -0400 10249 10250 r3xx-r5xx: switch to 1/12 subpixel precision 10251 10252 - based on similar patch in Jerome's cs ddx tree 10253 - also fix clipping offsets 10254 - should eventually allow for 4k render targets 10255 - mesa driver uses 1/12 mode, this avoids changing the 10256 subpixel mode when switching between ddx and mesa 10257 10258commit 3880bc3c95764acccd62cfa438932458df95daf6 10259Author: Alex Deucher <alexdeucher@gmail.com> 10260Date: Sun May 3 19:01:28 2009 -0400 10261 10262 Remove some old debugging info 10263 10264commit 3362381aa3a97c1da564f8a479ed4e7b0d6368b8 10265Author: Alex Deucher <alexdeucher@gmail.com> 10266Date: Sun May 3 18:52:00 2009 -0400 10267 10268 r2xx-r4xx: fix typo in last i2c fix and clarify hw i2c pin sel 10269 10270commit e01b8dc827ce1913927e1a4fad05a99a2dd9e805 10271Author: Alex Deucher <alexdeucher@gmail.com> 10272Date: Sun May 3 13:09:39 2009 -0400 10273 10274 RS690: gpio/connector fixes 10275 10276 Fixes bug 21521 10277 10278commit 1f70c9f05df9017d87b37f887e1eccd6d0568a02 10279Author: Alex Deucher <alexdeucher@gmail.com> 10280Date: Thu Apr 30 19:18:13 2009 -0400 10281 10282 R3xx-R5xx: don't set TX_OFFSET_RS in RS_INST_COUNT 10283 10284 Isn't necessary and seems to cause problems for RS690 users 10285 10286commit de17d36c1f710305870e3c636cae3f742f1cb6d6 10287Author: Alex Deucher <alexdeucher@gmail.com> 10288Date: Thu Apr 30 19:13:09 2009 -0400 10289 10290 AVIVO: first pass at display watermark setup 10291 10292 This helps with flickering and blanking when 10293 there is contention for MC bandwidth. 10294 10295commit 7d9f643ae3d07e51e644a5979ca90bc2c102bc89 10296Author: Alex Deucher <alexdeucher@gmail.com> 10297Date: Wed Apr 29 12:38:15 2009 -0400 10298 10299 radeon: fix several dpms issues 10300 10301 - atom dpms was unblanking for standby/suspend 10302 - return if r600+ in radeon_crtc_modeset_ioctl() 10303 - remove seprate standby/suspend handling in legacy 10304 crtc dpms. we turn the outputs off, so turn the 10305 crtcs off too. 10306 - disable the crtcs in legacy crtc dpms 10307 - move radeon_crtc_modeset_ioctl() calls to radeon_crtc_dpms() 10308 so they will get called for both atom and legacy paths 10309 10310 Should fix bug 21321 10311 10312commit 35183ef528411eb0122cf48550d93f921d291d12 10313Author: Alex Deucher <alexdeucher@gmail.com> 10314Date: Wed Apr 29 01:55:10 2009 -0400 10315 10316 AVIVO: better fix for cursor flickering/corruption issues 10317 10318 Should prevent ghost cursor from being left on screen. 10319 10320commit 53e75cbd0ba8f39799d7b776cdc59d60aa2a6916 10321Author: Alex Deucher <alexdeucher@gmail.com> 10322Date: Wed Apr 29 01:05:31 2009 -0400 10323 10324 R6xx/r7xx: fix CURSOR_SWAPPING_* macros 10325 10326 r6xx/r7xx have different swapper regs 10327 10328commit 4e948e2aaa59267a9aa797396add2c6b7b4c1e7c 10329Author: Alex Deucher <alexdeucher@gmail.com> 10330Date: Tue Apr 28 19:28:25 2009 -0400 10331 10332 AVIVO: fix cursor corruption/flickering issues 10333 10334 When both crtcs are enabled, both cursors need to be enabled 10335 or you get corruption bands. 10336 10337 Thanks to Yang Zhao for figuring this out. 10338 10339commit 4d0beaf02b28d5a5dac4ff6053593a2f979f92ec 10340Author: Alex Deucher <alexdeucher@gmail.com> 10341Date: Tue Apr 28 12:33:24 2009 -0400 10342 10343 RV410: SE variants only have 1 quad pipe 10344 10345 Should fix EXA corruption with the 3D engine. 10346 Mesa and drm patches forthcoming. 10347 10348 Reported by Kano on IRC. 10349 10350commit abd565ae71d9844d5afcd658628ca7778ce26ab1 10351Author: Alex Deucher <alexdeucher@gmail.com> 10352Date: Mon Apr 27 14:21:38 2009 -0400 10353 10354 AVIVO: disable fractional feedback divs 10355 10356 Doesn't seem to work well in all cases. Needs more investigation. 10357 fixes bug 21413 10358 10359commit 93626b1277f57bd559cf71b263221209b02e0046 10360Author: Alex Deucher <alexdeucher@gmail.com> 10361Date: Mon Apr 27 03:35:24 2009 -0400 10362 10363 AVIVO: make sure cursor width isn't negative 10364 10365commit 32af0f097bccc4499d67e969d46ba35e7487c5e2 10366Author: Dave Airlie <airlied@linux.ie> 10367Date: Sat Apr 25 19:39:34 2009 +1000 10368 10369 Revert "r300/r500: set R clamp mode." 10370 10371 This reverts commit dab3d61619d91ae6a3da3e9b29b8258313653813. 10372 10373commit b2c8f57bccbd46164907049a26ad2dd209d7b108 10374Author: Dave Airlie <airlied@linux.ie> 10375Date: Sat Apr 25 19:39:19 2009 +1000 10376 10377 Revert "r3xx/r5xx: fix typo in last commit" 10378 10379 This reverts commit 1a7b982457ead5f21c6e62a409ca0dbf86005607. 10380 10381commit 253731dce6ab25122f2eda559233c098536b7143 10382Author: Dave Airlie <airlied@linux.ie> 10383Date: Sat Apr 25 19:39:05 2009 +1000 10384 10385 Revert "radeon: attempt to fix R clamping for repeat again" 10386 10387 This reverts commit 5998f262a52cb85b334fcc1fe24c7b2ae474ce93. 10388 10389commit 5998f262a52cb85b334fcc1fe24c7b2ae474ce93 10390Author: Dave Airlie <airlied@linux.ie> 10391Date: Fri Apr 24 17:30:57 2009 +1000 10392 10393 radeon: attempt to fix R clamping for repeat again 10394 10395commit 7ed27b926f28a1f088eace5034a7d9985f13752d 10396Author: Alex Deucher <alexdeucher@gmail.com> 10397Date: Thu Apr 23 14:06:06 2009 -0400 10398 10399 r2xx/r3xx/r4xx: further i2c fixups 10400 10401 - hw i2c engine has pin selection on r2xx/r2xx/r3xx chips 10402 - also switch hw i2c pin sel for external tmds 10403 10404commit 0da80f34b5aa5999d2030ffbc5187328fa4e4ae5 10405Author: Alex Deucher <alexdeucher@gmail.com> 10406Date: Wed Apr 22 18:42:36 2009 -0400 10407 10408 radeon: minor cleanup and warning fix 10409 10410commit 1a7b982457ead5f21c6e62a409ca0dbf86005607 10411Author: Alex Deucher <alexdeucher@gmail.com> 10412Date: Thu Apr 23 10:09:22 2009 -0400 10413 10414 r3xx/r5xx: fix typo in last commit 10415 10416 Noticed by osiris on IRC 10417 10418commit dab3d61619d91ae6a3da3e9b29b8258313653813 10419Author: Dave Airlie <airlied@redhat.com> 10420Date: Thu Apr 23 15:46:44 2009 +1000 10421 10422 r300/r500: set R clamp mode. 10423 10424 On rs690 without this set, we appear to end up fetching from well 10425 outside the texture sometimes into memory that isn't mapped. 10426 10427 This "shouldn't" have any effect. 10428 10429commit bbfd4625478e37d10fd58b23f46153f349370a52 10430Author: Alex Deucher <alexdeucher@gmail.com> 10431Date: Wed Apr 22 14:47:16 2009 -0400 10432 10433 radeon PM: update quirks 10434 10435commit e3c564114eea4d032f920fcebd4040f1934c21b0 10436Author: Alex Deucher <alexdeucher@gmail.com> 10437Date: Wed Apr 22 14:31:44 2009 -0400 10438 10439 AVIVO: fix for cursor corruption issue 10440 10441commit 24e4b73b4fbbb2c790e6120ede3caaa4e7e58359 10442Author: Alex Deucher <alexdeucher@gmail.com> 10443Date: Wed Apr 22 11:42:18 2009 -0400 10444 10445 radeon pll: add support for fractional feedback divs 10446 10447 Allows us to hit dot clocks much closer, especially on 10448 chips with non-27 Mhz reference clocks like most IGP chips. 10449 10450 This should fix most flickering and blanking problems with 10451 non-exact dot clocks. 10452 10453commit efa0825a86a8dc0f03ebb42c576ed26189e9d4bb 10454Author: Alex Deucher <alexdeucher@gmail.com> 10455Date: Tue Apr 21 13:51:11 2009 -0400 10456 10457 Tex vid: fix top clipping 10458 10459commit aa59b5b55e6ac64940fb3f0fa61f7b8fe36ba745 10460Author: Alex Deucher <alexdeucher@gmail.com> 10461Date: Tue Apr 21 12:06:24 2009 -0400 10462 10463 RS690: add quirk for acer board 10464 10465 thanks to Gino Badouri for testing 10466 10467commit 0e5164d968316bbb59b4484a7df087854738068f 10468Author: Alex Deucher <alexdeucher@gmail.com> 10469Date: Tue Apr 21 04:47:43 2009 -0400 10470 10471 Tex vid: rework part 2 10472 10473 - use pPriv->w/h directly for tex coords 10474 - take src x/y offset into account when calculating tex coords 10475 - when copying data, only optimize for y clipping. In order 10476 to deal with the x clipping optimization, the copy routines 10477 or tex coords would need to be fixed. This should fix clipping 10478 problems with the current code. 10479 10480commit 06a1b4169b506f69dd807380d9ce7210b3a3bc23 10481Author: Alex Deucher <alexdeucher@gmail.com> 10482Date: Tue Apr 21 01:21:51 2009 -0400 10483 10484 R4xx: avoid a PM related hang 10485 10486commit 228b9ac87668241b09980fca17e9d866891285e6 10487Author: Alex Deucher <alexdeucher@gmail.com> 10488Date: Tue Apr 21 01:02:34 2009 -0400 10489 10490 radeon PM: add some quirks 10491 10492commit b6febf7368d3df3f1d3fee6a87aa71c54e77cdcc 10493Author: Alex Deucher <alexdeucher@gmail.com> 10494Date: Mon Apr 20 18:04:18 2009 -0400 10495 10496 Tex vid: cleanup and unify common code 10497 10498commit d096d1189c8dc5e19f68cc605ccffe1a8fef048b 10499Author: Alex Deucher <alexdeucher@gmail.com> 10500Date: Mon Apr 20 13:05:18 2009 -0400 10501 10502 Endian fixes for object table 10503 10504commit 101ce28ec4af696549b7d6defebb5b4ad412176c 10505Author: Alex Deucher <alexdeucher@gmail.com> 10506Date: Sun Apr 19 12:13:18 2009 -0400 10507 10508 Remove old messages 10509 10510 R5xx/r6xx support isn't really experimental anymore 10511 10512commit db177c70ce88af19c8e05eb916a15f3e124876eb 10513Author: Alex Deucher <alexdeucher@gmail.com> 10514Date: Fri Apr 17 01:05:15 2009 -0400 10515 10516 Update Xv info in man page 10517 10518commit 00266177bed2dc8693df497ca3ec19f2dc4adc05 10519Author: Alex Deucher <alexdeucher@gmail.com> 10520Date: Thu Apr 16 20:33:28 2009 -0400 10521 10522 R3xx/R5xx: only apply Xv attributes if bicubic is disabled 10523 10524 Provides consistent output 10525 10526commit 85323a7f84381fef7fad20c7f7ec601637af9aa7 10527Author: Alex Deucher <alexdeucher@gmail.com> 10528Date: Thu Apr 16 11:10:51 2009 -0400 10529 10530 R3xx/R4xx: set tex caching for Y texture when doing planar rendering 10531 10532 Doesn't affect performance, but docs indicate its the right 10533 thing to do. 10534 10535commit 5ea5df22c038fc8f00984acc760e9d8c962bf902 10536Author: Alex Deucher <alexdeucher@gmail.com> 10537Date: Wed Apr 15 20:26:34 2009 -0400 10538 10539 Tex vid: remove remnants of XV_HWPLANAR 10540 10541 no longer needed as bicubic is the only thing that uses 10542 the old csc code. 10543 10544commit 9091b3f5f13dbea83ffd89679dac600e9f280bb2 10545Author: Alex Deucher <alexdeucher@gmail.com> 10546Date: Wed Apr 15 20:07:42 2009 -0400 10547 10548 R3xx/R4xx: fix up planar shader 10549 10550 We were overwriting the coord fetch address with the first 10551 tex fetch. Seemed to work however, luck I guess. Reorder 10552 the fetches to write to temp0 last. 10553 10554commit 14c13faeb9f9b7717a25fcc1ca97d46cc6ee0031 10555Author: Alex Deucher <alexdeucher@gmail.com> 10556Date: Wed Apr 15 19:53:12 2009 -0400 10557 10558 R5xx: add shader-based csc 10559 10560 - native planar support 10561 - Xv attributes 10562 10563commit 832efc7b90f5eb2da99512fcb902ab4838d2dcd1 10564Author: Alex Deucher <alexdeucher@gmail.com> 10565Date: Tue Apr 14 17:56:49 2009 -0400 10566 10567 R3xx/R4xx: Implement shader-based csc for packed formats 10568 10569commit 32625118c27041265d25811c00d25ab7e82fb340 10570Author: Alex Deucher <alexdeucher@gmail.com> 10571Date: Tue Apr 14 16:38:40 2009 -0400 10572 10573 tex vid: fix attribute setup typo for XV_COLORSPACE 10574 10575commit adf0912006b4f1597784dbfcc563d5c6d1c5667d 10576Author: Alex Deucher <alexdeucher@gmail.com> 10577Date: Tue Apr 14 16:16:33 2009 -0400 10578 10579 R6xx/R7xx: implement Xv attributes 10580 10581 - brightness, contrast, hue, etc. 10582 - TODO: implement gamma 10583 10584commit 8810fe92b5aed08888584c6914482586b59f71ab 10585Author: Alex Deucher <alexdeucher@gmail.com> 10586Date: Tue Apr 14 11:26:35 2009 -0400 10587 10588 R200: clean code, always use shader based csc 10589 10590 - consolidate common r2xx csc shader code 10591 - always use shader based csc for both packed and planar 10592 formats 10593 - always use native planar csc on r1xx 10594 10595commit 17685fefba68d188c7c0fe7a079180ec0722c046 10596Author: Alex Deucher <alexdeucher@gmail.com> 10597Date: Mon Apr 13 20:06:11 2009 -0400 10598 10599 Tex vid: general cleanup 10600 10601 - convert macros to more meaningful VTX_OUT_4 and VTX_OUT_6 10602 names to reflect that they actually do 10603 - fix indenting 10604 10605commit 093ab4c9a33b0b396b78c061c3321dc044bdccdc 10606Author: Alex Deucher <alexdeucher@gmail.com> 10607Date: Mon Apr 13 19:48:35 2009 -0400 10608 10609 R1xx: add support for native planar textured Xv 10610 10611commit ec0cb51df81c6c9a1de640d227fa9c9c33161083 10612Author: Alex Deucher <alexdeucher@gmail.com> 10613Date: Mon Apr 13 17:21:20 2009 -0400 10614 10615 R2xx tex vid: append verts for clip boxes 10616 10617 rather than sending a new draw packet for each rect 10618 10619commit fde075a30a8ee2c333aa1bbe8fbd177258b085ba 10620Author: Alex Deucher <alexdeucher@gmail.com> 10621Date: Mon Apr 13 17:13:51 2009 -0400 10622 10623 R1xx tex vid: append verts for clip boxes 10624 10625 rather than sending a new draw packet for each rect 10626 10627commit 12839fc17a2cca4ac14b9757bdaa63ba4679f96f 10628Author: Alex Deucher <alexdeucher@gmail.com> 10629Date: Mon Apr 13 17:04:31 2009 -0400 10630 10631 Tex vid: split by family 10632 10633commit a30737b337edb31528174b483c9094941a5d41bb 10634Author: Roland Scheidegger <sroland@tungstengraphics.com> 10635Date: Mon Apr 13 15:36:07 2009 -0400 10636 10637 r200/r300: implement brightness/contrast/hue/saturation/gamma controls for textured video 10638 10639 This implements 10640 contrast/brightness/hue/saturation controls for r200/r300 plus gamma (same 10641 gamma value for all channels used though separate values would be trivial) 10642 control for r300. 10643 Some issues left: 10644 - only r200/r300 10645 - still can't be combined with bicubic 10646 - controls will silently cease to work if the format used is packed and not 10647 planar (except for rv250) 10648 - gamma range is from 100 to 10000 corresponding to 0.1 and 10.0 like used in 10649 overlay. However, usable range is far smaller. Over 2.0 picture gets dark 10650 pretty quickly, and below 0.6 or so black seems to turn into purple (I've 10651 verified that even with gamma 1.0 black actually often seems to be RGB 1/0/1 so 10652 this explains this since that gets amplified by low gamma values - not sure if 10653 this is a rounding problem somewhere, bogus reference values or is somehow 10654 expected). 10655 - gamma adds a bit too many instructions for my taste (7) though the 10656 alternative (3 texture lookups + some swizzling instructions) doesn't seem any 10657 better. 10658 10659commit 9dd33cc8346073300312cd5f68de72d9bea1dd6b 10660Author: Dave Airlie <airlied@redhat.com> 10661Date: Fri Apr 17 10:30:21 2009 +1000 10662 10663 radeon: tweak ddc fix for all r300s. 10664 10665 Alex seems to think this might affect some other chips 10666 10667commit a846ec14dd65bb12f314f83ffb45c4c4be564a5e 10668Author: Dave Airlie <airlied@redhat.com> 10669Date: Fri Apr 17 10:24:15 2009 +1000 10670 10671 radeon: fix DDC on rv410 VGA ports for non-atom use. 10672 10673 The rv410 hw i2c block appears to hold down whatever i2c lines 10674 it is connected to when in reset, so when doing DDC change hw i2c to 10675 point at different i2c lines. 10676 10677commit efa1476fc0765b53298624198137dc5cc71d37a8 10678Author: Alex Deucher <alexdeucher@gmail.com> 10679Date: Thu Apr 16 16:43:02 2009 -0400 10680 10681 PM: fix reversed logic in pm block handler 10682 10683commit 2b95de17781959457a809c8fecc6bbb08336c83f 10684Author: Alex Deucher <alexdeucher@gmail.com> 10685Date: Thu Apr 16 12:52:41 2009 -0400 10686 10687 RS400/RS480: attempt to fix ClockGating on RS4xx chips 10688 10689commit c0215d2f2a27755fb04d20aaa4320a5e7da7ee04 10690Author: Alex Deucher <alexdeucher@gmail.com> 10691Date: Thu Apr 16 11:58:19 2009 -0400 10692 10693 RS690/RS740: fix connector enumeration in some cases 10694 10695 Should fix Masta-G's issue reported on IRC. 10696 10697commit 573339114660f06e5f257648ba413bab0d6083ce 10698Author: Alex Deucher <alexdeucher@gmail.com> 10699Date: Thu Apr 16 11:19:01 2009 -0400 10700 10701 R3xx/R4xx EXA: set explicit tex cache regions 10702 10703 When using a mask, set explicit cache partitions for 10704 each texture. Gives 1% performance improvement in 10705 x11perf with anti-aliased text. 10706 10707commit eea9800890b56bac9c07b7bd9c9e33fae2938af3 10708Author: Alex Deucher <alexdeucher@gmail.com> 10709Date: Thu Apr 16 10:41:28 2009 -0400 10710 10711 Revert "DCE3/3.2: cleanup dpms after previous changes" 10712 10713 This reverts commit f8c7d6a6162196a743f6885ecaf63ba50de1722a. 10714 10715 This is apparently still needed for some setups, however, I can't 10716 reproduce this locally anymore. 10717 10718commit 099cb27d760af61b967239d960bc77f71e828655 10719Author: Alex Deucher <alexdeucher@gmail.com> 10720Date: Wed Apr 15 16:23:01 2009 -0400 10721 10722 Change ForceLowPowerMode to 4 pcie lanes 10723 10724 This fixes corruption for some users 10725 10726commit 5abcea88d1f3307a7d08ae7757c4b535df1503bf 10727Author: Alex Deucher <alexdeucher@gmail.com> 10728Date: Wed Apr 15 15:52:04 2009 -0400 10729 10730 PM: don't touch clock gating regs unless ClockGating option is set 10731 10732 Fixes an M26 hang reported by tormod on IRC 10733 10734commit bae88f81e2a01c0d547132b5b91e717ff9d0497d 10735Author: Alex Deucher <alexdeucher@gmail.com> 10736Date: Wed Apr 15 14:38:35 2009 -0400 10737 10738 man page: fix missing break 10739 10740commit 4b3a3785f7b06205eeec6ca6b8913ecc2b94bd1f 10741Author: Alex Deucher <alexdeucher@gmail.com> 10742Date: Sat Apr 4 11:39:35 2009 -0400 10743 10744 Unify ForceLowPowerMode and DynamicPM Options 10745 10746 Force lower power mode and switch to a lower mode 10747 when idle. 10748 10749commit 63cb57c5eb0d6bc083f54eb100cb972b0e7bca69 10750Author: Alex Deucher <alexdeucher@gmail.com> 10751Date: Wed Apr 1 16:22:02 2009 -0400 10752 10753 Add DynamicPM Option 10754 10755 Dyanmically switch between power states. Switch to a low 10756 power state when the system is idle (DPMS off). 10757 10758commit a34a8b37afbea6ed4bf8ca42364195e174250c48 10759Author: Alex Deucher <alexdeucher@gmail.com> 10760Date: Wed Apr 1 15:23:17 2009 -0400 10761 10762 Set default low power PCIE lanes to 2 10763 10764 1 lane seems to cause occasional corruption when 10765 blitting to/from gart memory. 10766 10767commit 1f0dc778dc25f4f85fedd73c55c847cab2c79fc5 10768Author: Alex Deucher <alexdeucher@gmail.com> 10769Date: Wed Apr 1 15:12:27 2009 -0400 10770 10771 Add support for setting the number PCIE lanes 10772 10773commit 7e10b6222e8f44a3ecc6aaea55a7a7680d133bb3 10774Author: Alex Deucher <alexdeucher@gmail.com> 10775Date: Wed Apr 1 15:05:38 2009 -0400 10776 10777 Add ForceLowPowerMode Option 10778 10779 Force the chip to a low power mode at the expense 10780 of performance. 10781 10782commit 533d01a5933cd491bbc09cd463ea62475abf4bf2 10783Author: Alex Deucher <alexdeucher@gmail.com> 10784Date: Tue Mar 31 17:24:29 2009 -0400 10785 10786 radeon: Add functions to set sclk/mclk using atombios 10787 10788commit adb099409768e695b9928fa6aa5760f93dadd9af 10789Author: Alex Deucher <alexdeucher@gmail.com> 10790Date: Tue Mar 31 17:13:11 2009 -0400 10791 10792 radeon: Add functions to set sclk/mclk on r1xx-r4xx 10793 10794commit 68e2a959ccc3d1a5d0731f1b55fdf1b2412635b2 10795Author: Alex Deucher <alexdeucher@gmail.com> 10796Date: Tue Mar 31 17:02:45 2009 -0400 10797 10798 Add unified ClockGating Option 10799 10800 - replaces DynamicClocks Option as the name was misleading 10801 - unified interface for atom and com based bioses 10802 - fix up clock gating code for newer r3xx asics 10803 10804commit 336cee0e6d19068582b189b2c747d92cb2139d2f 10805Author: Dave Airlie <airlied@redhat.com> 10806Date: Mon Apr 13 18:11:48 2009 +1000 10807 10808 radeon: legacy bandwidth calcs for later r3xx/r4xx cards. 10809 10810 This is an attempt to rationalise the code using the register info 10811 files available to me here. 10812 10813 Further info is required: 10814 r350 check for stop_req > 15 then subtract 10: what other chips need it? 10815 get rs480 working instead of using magic. 10816 10817commit 367ba7af76e7f63f3186a6e8d9bbe21aa31fc6d1 10818Author: Alex Deucher <alexdeucher@gmail.com> 10819Date: Fri Apr 10 09:37:15 2009 -0400 10820 10821 radeon: fix tmds_pll randr property 10822 10823commit 15f1a2f31a7bc33f276a65283dbad0c106bee73e 10824Author: Alex Deucher <alexdeucher@gmail.com> 10825Date: Thu Apr 9 20:17:03 2009 -0400 10826 10827 radeon: remove old crtc size heuristics 10828 10829 IIRC, the old randr code used to use this to use for front buffer sizing, 10830 but it has since been changed. 10831 10832commit ce9077b2f5a5e749c0078dfb79d26ef8711a823a 10833Author: Alex Deucher <alexdeucher@gmail.com> 10834Date: Thu Apr 9 19:59:19 2009 -0400 10835 10836 R6xx/R7xx: cleanup macro 10837 10838commit 6532aeddacdeda3aa534264741c8648e1c449e76 10839Author: Michel Dänzer <daenzer@vmware.com> 10840Date: Thu Apr 9 09:41:56 2009 +0200 10841 10842 Clear the framebuffer before initializing modes. 10843 10844 Fixes garbage being visible shortly on server startup or when VT switching back 10845 to X. 10846 10847commit 0c19053975925fed637aaa41d440e0cab8f6b596 10848Author: Alex Deucher <alexdeucher@gmail.com> 10849Date: Wed Apr 8 13:34:23 2009 -0400 10850 10851 radeon: chip family typo 10852 10853commit 2888dd9ae9689b1cd72115dc0ceea1f5957299b0 10854Author: Alex Deucher <alexdeucher@gmail.com> 10855Date: Wed Apr 8 13:10:01 2009 -0400 10856 10857 R6xx/R7xx: set proper 3D client driver name for r600 10858 10859commit 61a083116a2c89e145aa19da951243536e07542e 10860Author: Alex Deucher <alexdeucher@gmail.com> 10861Date: Wed Apr 8 10:44:30 2009 -0400 10862 10863 Bump version post release 10864 10865commit 1b02b93895c31a0c9d641e47b46dce43b40edd97 10866Author: Alex Deucher <alexdeucher@gmail.com> 10867Date: Wed Apr 8 10:34:44 2009 -0400 10868 10869 Bump for release 10870 10871commit da74b94a5da8f0bcf4ae99b20e594d2f0f4fc6f6 10872Author: Alex Deucher <alexdeucher@gmail.com> 10873Date: Tue Apr 7 18:59:06 2009 -0400 10874 10875 ATOM: code cleanup 10876 10877 - remove cases that aren't possible (e.g., no TMDSA on DCE3 cards) 10878 - condense duplicate cases 10879 10880commit a24aa7250d09ae68ceaaba987314388b5e52096c 10881Author: Alex Deucher <alexdeucher@gmail.com> 10882Date: Tue Apr 7 18:52:08 2009 -0400 10883 10884 DCE3/3.2: further fixes 10885 10886 disable the dig transmitter and encoder before setup and enable 10887 10888commit 284e1943f6726ed08e796f99e49fbd90ac7fa8a4 10889Author: Alex Deucher <alexdeucher@gmail.com> 10890Date: Wed Apr 1 15:07:01 2009 -0400 10891 10892 Cast info->sclk to int 10893 10894commit b580e388e0e31843308c783c1ca859672dc6ca0e 10895Author: Alex Deucher <alexdeucher@gmail.com> 10896Date: Wed Apr 1 14:59:43 2009 -0400 10897 10898 Fix missing ')' in PCIE macro 10899 10900commit 8ec27be3c8043a6ddc816b0b1480dd19f8d73832 10901Author: Alex Deucher <alexdeucher@gmail.com> 10902Date: Tue Apr 7 11:33:44 2009 -0400 10903 10904 radeon: fix Xv vsync for multi-head 10905 10906commit f8c7d6a6162196a743f6885ecaf63ba50de1722a 10907Author: Alex Deucher <alexdeucher@gmail.com> 10908Date: Sat Apr 4 16:05:20 2009 -0400 10909 10910 DCE3/3.2: cleanup dpms after previous changes 10911 10912commit a707d355c3c6ff92252c5a060a1fc32d97547552 10913Author: Alex Deucher <alexdeucher@gmail.com> 10914Date: Sat Apr 4 15:20:01 2009 -0400 10915 10916 DCE3/3.2: fix up transmitter/encoder setup 10917 10918 Should fix bug 21050 10919 10920commit 215e12f9c0e8ac62c23af1add776ef88f9a0dc54 10921Author: Eduard Fuchs <edfuchs@uni-kassel.de> 10922Date: Thu Apr 2 12:36:59 2009 -0400 10923 10924 Enable byte swapping for r6xx/r7xx Hardware 10925 10926 Patch from Eduard Fuchs with some cleanup from me. 10927 10928 Tested at 32 bpp on MPC8641HPCN board (PowerPC) with 10929 HD2400 PCIe card 10930 10931commit 0ded9fab6bc3fecc1976ae8369b788659e4d7793 10932Author: Alex Deucher <alexdeucher@gmail.com> 10933Date: Thu Apr 2 12:09:00 2009 -0400 10934 10935 Add RV790 (HD 4890) support 10936 10937commit 0c6f20a9404c6a4210c5180bdca61c13fc186754 10938Author: Alex Deucher <alexdeucher@gmail.com> 10939Date: Tue Mar 31 15:51:11 2009 -0400 10940 10941 Add PCIE register accessors 10942 10943commit 229d4e20e5adcdf9207944c19ca721477936ff02 10944Author: Alex Deucher <alexdeucher@gmail.com> 10945Date: Tue Mar 31 15:39:42 2009 -0400 10946 10947 Unify mc_idle() 10948 10949commit f44b486848094b75479abef61c1aa075aede835c 10950Author: Alex Deucher <alexdeucher@gmail.com> 10951Date: Tue Mar 31 15:33:11 2009 -0400 10952 10953 COMBIS: Clarify case 0 for LCD DDC table 10954 10955commit e6f3c48899a92174a55f6fe7327412bc88039d87 10956Author: Alex Deucher <alexdeucher@gmail.com> 10957Date: Tue Mar 31 15:30:40 2009 -0400 10958 10959 COMBIOS: fix default sclk/mclk from bios 10960 10961commit 75bf5272936093bc024559e5744d97d8e79943b1 10962Author: Alex Deucher <alexdeucher@gmail.com> 10963Date: Tue Mar 31 15:17:37 2009 -0400 10964 10965 mac mini: 3rd time's the charm 10966 10967 fixes 10fbbac4bed59e12ad794ac873dd415d9ee4e340 10968 fixes bugs 20796 and 20979 10969 10970commit 10fbbac4bed59e12ad794ac873dd415d9ee4e340 10971Author: Alex Deucher <alexdeucher@gmail.com> 10972Date: Tue Mar 31 15:14:27 2009 -0400 10973 10974 Mac mini: fix connector setup for real 10975 10976 fixes f061308e7abcf93d1612681b935387fee23e0a24 10977 fixes bugs 20796 and 20979 10978 10979commit 8f2a03bfc86422669a46c2853050979f099a17a8 10980Author: Alex Deucher <alexdeucher@gmail.com> 10981Date: Mon Mar 30 11:30:34 2009 -0400 10982 10983 r1xx-r3xx TV: fix white noise 10984 10985 should fix bug 13872 10986 10987commit f061308e7abcf93d1612681b935387fee23e0a24 10988Author: Alex Deucher <alexdeucher@gmail.com> 10989Date: Fri Mar 27 01:14:12 2009 -0400 10990 10991 mac mini: fix connector setup regression 10992 10993 fixes bug 20796 10994 10995commit 47134ca77988f8c3b42d55ac4fcbb5dc182878d4 10996Author: Alex Deucher <alexdeucher@gmail.com> 10997Date: Thu Mar 26 17:18:47 2009 -0400 10998 10999 radeon: re-enable r/g/b intensity attributes for overlay 11000 11001 Noticed by sroland on IRC. 11002 11003commit e2b5e48e2652daf03fcae852d8368f29ea4abcd9 11004Author: Alex Deucher <alexdeucher@gmail.com> 11005Date: Thu Mar 26 13:42:25 2009 -0400 11006 11007 R6xx/R7xx EXA: rework composite pixel shader 11008 11009 - move to vram storage 11010 - move swizzle logic to tex setup 11011 11012commit 24dbd3f93c640f819a9f5029f5a49e41ab875bd4 11013Author: Alex Deucher <alexdeucher@gmail.com> 11014Date: Thu Mar 26 12:01:07 2009 -0400 11015 11016 R6xx/R7xx: clean up logic in EXA Comp PS setup 11017 11018commit 076e4e3ac1c81086aecb4e70dd30fd5d26ec7bc8 11019Author: Alex Deucher <alexdeucher@gmail.com> 11020Date: Thu Mar 26 11:36:26 2009 -0400 11021 11022 R6xx/R7xx: clean up bool const code 11023 11024 3 regs: 1 bit per bool, 32 bools per ps/vs/gs 11025 11026commit d2c3964fe04be42fe538f36439ed5ffca96e436a 11027Author: Roland Scheidegger <sroland@tungstengraphics.com> 11028Date: Wed Mar 18 01:55:12 2009 +0100 11029 11030 fix textured video allocation bug 11031 11032 size needs to be calculated after dstPitch adjustments, got already fixed 11033 for earlier than R600 chips by planar textured yuv patches, clean this up. 11034 11035commit 18e56eb179fde28477487c63e6f9ebf7579e2cd5 11036Author: Roland Scheidegger <sroland@tungstengraphics.com> 11037Date: Thu Mar 5 02:07:46 2009 +0100 11038 11039 don't convert planar yuv to packed for r200 11040 11041 uses 3 textures for planar yuv and does yuv->rgb conversion in the shader. 11042 Similar to r300 code, but might have precision issues - hardware alu should 11043 have enough precision but hardware consts are only 8bit and we'd want 11044 at least 11. 11045 This also enables textured video on rv250 (and also supports packed yuv 11046 on that chip by using basically the same shader with packed data). 11047 11048commit 58530bf4912800f9e09ebaea42a13cff8a80c19e 11049Author: Roland Scheidegger <sroland@tungstengraphics.com> 11050Date: Sat Feb 21 04:46:31 2009 +0100 11051 11052 don't convert planar yuv to packed for r300 11053 11054 uses 3 textures for planar yuv and does yuv->rgb conversion in the shader. 11055 small performance advantage, but manual texture cache setting is necessary 11056 otherwise it may be measurably slower (but probably not relevant) in some 11057 cases. 11058 Unlike some other drivers, using MADs instead of DP3s, since this requires 11059 less instructions due to no MOVs are required, the end result is the same 11060 though the constants need to be different. 11061 Use of this is user settable for now (XV_HWPLANAR attrib). 11062 11063commit 97e19d96ba65a3df2fa3bbf73cfcc01b6dc3e796 11064Author: Roland Scheidegger <sroland@tungstengraphics.com> 11065Date: Tue Dec 30 22:23:39 2008 +0100 11066 11067 clip fixes 11068 11069 This fixes some oddities observed when the video is only partly visible. 11070 Instead of recalculating the geometry of the video, always use the same. 11071 Also fixes a assignment present twice, and another issue (bring in line with 11072 what the overlay code does). 11073 11074commit 7aaf5f08e4147b75c4975e6d680d3f5a93424164 11075Author: Alex Deucher <alexdeucher@gmail.com> 11076Date: Mon Mar 23 12:55:02 2009 -0400 11077 11078 R4xx ATOM: fix crtc routing for CRT1 11079 11080 should fix bug 20814 11081 11082commit 2fe6b393d9dbb193d4f0a3351ad76618cda07e2c 11083Author: Alex Deucher <alexdeucher@gmail.com> 11084Date: Sun Mar 22 01:31:30 2009 -0400 11085 11086 radeon: add new chip ids 11087 11088commit 1ddee7cd6fd267b2fc86f21af27c5425eb0835a4 11089Author: Alex Deucher <alexdeucher@gmail.com> 11090Date: Fri Mar 20 13:44:00 2009 -0400 11091 11092 DCE3.2: use RMX for for non-native modes on DVI 11093 11094 DVI seems to have issues with low dotclocks, so 11095 use the scaler instead. 11096 11097 Fixes bug 20754 11098 11099commit 44081e8fd2bf974272d5447b45dd7b206cf939c1 11100Author: Alex Deucher <alexdeucher@gmail.com> 11101Date: Thu Mar 19 20:48:16 2009 -0400 11102 11103 R6xx/R7xx EXA: fix maxPitchBytes 11104 11105 should now allow accel up the hw max of 8192x8192 11106 11107commit e8e2367b66d9a4f5977d95f52b37cdbad75e409a 11108Author: Alex Deucher <alexdeucher@gmail.com> 11109Date: Thu Mar 19 20:09:32 2009 -0400 11110 11111 Add new pci ids 11112 11113commit 68e83f10adb23fc359c1da881f5d666cd08b4067 11114Author: Alex Deucher <alexdeucher@gmail.com> 11115Date: Wed Mar 18 20:58:19 2009 -0400 11116 11117 bump version post release 11118 11119commit a196aae19603c8651dc502f5cb0df4d55befe163 11120Author: Alex Deucher <alexdeucher@gmail.com> 11121Date: Wed Mar 18 20:52:12 2009 -0400 11122 11123 bump for release 11124 11125commit 537da7018b0fd2ad263fc5cdc55b3b85cf8e9205 11126Author: Alex Deucher <alexdeucher@gmail.com> 11127Date: Wed Mar 18 19:22:55 2009 -0400 11128 11129 ATOM: fix up tv-out support 11130 11131 - fix purple hue when using ATOMTvOut option 11132 - fix TV load detection on newer r6xx/r7xx chips 11133 - disable CV for now 11134 11135commit c0e2513ab128ddd5be0ed626d9e31777a98983ef 11136Author: Alex Deucher <alexdeucher@gmail.com> 11137Date: Wed Mar 18 14:08:30 2009 -0400 11138 11139 R6xx/R7xx AGP: disable gart data transfers 11140 11141 UTS/DFS/Xv 11142 11143 Doesn't seem to be reliable on AGP. 11144 11145 fixes bugs: 20436, 20236, several reports on ML and IRC 11146 11147commit e3e1f0c7b892b094b292081021a0e97932b8e079 11148Author: Fabio <fabio.ped@libero.it> 11149Date: Wed Mar 18 12:13:53 2009 -0400 11150 11151 man page update 11152 11153commit 6ae54d3ea023968887bf0cb0ec1be2115621d554 11154Author: Alex Deucher <alexdeucher@gmail.com> 11155Date: Wed Mar 18 11:50:22 2009 -0400 11156 11157 AVIVO: Fix dac load detection 11158 11159commit 04e36b794c216e4fa77c780fd989df1ef1c7f0d9 11160Author: Fabio <fabio.ped@libero.it> 11161Date: Tue Mar 17 11:29:26 2009 -0400 11162 11163 man page updates 11164 11165 from bug 20648 11166 11167commit a6855c370194b6df307ea33724fe17a85d67607e 11168Author: Alex Deucher <alexdeucher@gmail.com> 11169Date: Mon Mar 16 13:52:42 2009 -0400 11170 11171 radeon: add another AGP quirk 11172 11173 bug 20647 11174 11175commit a55ced5ee20c07e743c7c0978803fd10589c1531 11176Author: Alex Deucher <alexdeucher@gmail.com> 11177Date: Sun Mar 15 12:08:04 2009 -0400 11178 11179 Fix fallout from r6xx/r7xx EXA merge in IB handling 11180 11181 Noticed Christiaan van Dijk. 11182 Should fix bug 20510 11183 11184commit a065eeb9b301ab452cc8fbd3cac1fd9a3841c4e1 11185Author: Alex Deucher <alexdeucher@gmail.com> 11186Date: Sat Mar 14 18:40:28 2009 -0400 11187 11188 atom: fix up bios scartch register usage 11189 11190 the bios scratch registers keep 3 types of state for each output: 11191 1. dpms 11192 2. active 11193 3. connected 11194 11195 some of the command tables use the active flags. We used to set the active 11196 flags based on dpms state, but instead set them based on whether they are 11197 connected or not. 11198 11199commit 5c4ef924d082fd3b279f62fda7a97f563b009d38 11200Author: Alex Deucher <alexdeucher@gmail.com> 11201Date: Sat Mar 14 18:16:49 2009 -0400 11202 11203 radeon: fix typo in bios scratch reg setup 11204 11205commit 96fce5a56269509a935e8ee621a5e17cc9005ddc 11206Author: Alex Deucher <alexdeucher@gmail.com> 11207Date: Fri Mar 13 19:29:58 2009 -0400 11208 11209 Bump post-release 11210 11211commit 8da1b82a70de7b3ef38905bd77c58a20fd44ddd6 11212Author: Alex Deucher <alexdeucher@gmail.com> 11213Date: Fri Mar 13 19:23:19 2009 -0400 11214 11215 bump for release 11216 11217commit 28d9d699dcdff394aa561505f7f54f08db800828 11218Author: Alex Deucher <alexdeucher@gmail.com> 11219Date: Fri Mar 13 18:18:31 2009 -0400 11220 11221 radeon: man page updates 11222 11223commit 06963796d00b5cb6704412a7d4417fcefae4aa4d 11224Author: Adam Jackson <ajax@redhat.com> 11225Date: Fri Mar 13 18:14:43 2009 -0400 11226 11227 uniphy: start adding DisplayPort setup 11228 11229commit 1db81afe4f8db55e02a65cd22ad84c6bbb27e8ef 11230Author: Alex Deucher <alexdeucher@gmail.com> 11231Date: Fri Mar 13 17:51:56 2009 -0400 11232 11233 radeon: add support for 30 bit LUTs 11234 11235 All radeons have them. Thanks to Yang Zhao for figuring 11236 this out. 11237 11238commit 5bb326c64ad3bf5c93943280dcbcb25c8910d1aa 11239Author: Alex Deucher <alexdeucher@gmail.com> 11240Date: Fri Mar 13 14:10:58 2009 -0400 11241 11242 R6xx/R7xx: EXA VSync Option not supported yet 11243 11244commit 8c2d40894448329e721360811c124998a17ba2ba 11245Author: Alex Deucher <alexdeucher@gmail.com> 11246Date: Fri Mar 13 14:03:43 2009 -0400 11247 11248 R6xx/R7xx: fix up vline stuff along the lines of previous chips 11249 11250 Avoids an additional function call. 11251 11252commit 176c1b43fc3ee48e87c134446754360b3683c24b 11253Author: Alex Deucher <alexdeucher@gmail.com> 11254Date: Fri Mar 13 13:28:42 2009 -0400 11255 11256 R6xx/R7xx: return in RADEONWaitForIdleMMIO() if accel is off 11257 11258 Don't wait for engine idle if we haven't initialized the engine. 11259 Behavior should be consistent with versions of the driver prior 11260 to accel support being added. 11261 11262 Should fix bug 20645 11263 11264commit d538a4508248adf2debaf0cc870a2d1c48abd986 11265Author: Alex Deucher <alexdeucher@gmail.com> 11266Date: Fri Mar 13 13:22:03 2009 -0400 11267 11268 R6xx/r7xx: clarify accel messages 11269 11270 We only support EXA and and only with DRI. 11271 11272commit 1b7090a4c0d6a38701fbfb928fac8e15d40fb81f 11273Author: Adam Jackson <ajax@redhat.com> 11274Date: Fri Mar 13 17:41:59 2009 -0400 11275 11276 Add some more cases to the DVI single-link checks. 11277 11278commit 104e13da74de37477e3a31a5de96c5ebb8fe888b 11279Author: Cooper Yuan <cooperyuan@gmail.com> 11280Date: Fri Mar 13 11:47:43 2009 -0400 11281 11282 radeon: save bios scratch registers in Preinit() 11283 11284 The display detect routines can change these which causes 11285 havok with some bioses. 11286 11287commit f70d7a0fd9ac3659d93511ab362eb80a7ff49777 11288Author: Dave Airlie <airlied@redhat.com> 11289Date: Fri Mar 13 12:56:29 2009 +1000 11290 11291 r600: reload shaders into VRAM on resume 11292 11293 As VRAM gets zeroed out over s/r, we need to reload the 11294 shaders. 11295 11296commit 1fe0dcfe98dfe30499ba53c1f208c4ef10bef001 11297Author: Yang Zhao <yang@yangman.ca> 11298Date: Thu Mar 12 18:50:32 2009 -0400 11299 11300 R6xx/R7xx shader: Fix OFFSET_[XYZ] macro for TEX_DWORD2 to accept floats 11301 11302 Values for OFFSET_[XYZ] are 5-bits two's-complement fixed-point with 11303 one-bit after decimal point. Values in [-8.0, 7.5] are valid. Inputs 11304 that do not exactly land on 0.5 increments are rounded towards 0 to the 11305 nearest increment. 11306 11307commit f356ca7795531e0c080775efb598141eb7db44ba 11308Author: Alex Deucher <alexdeucher@gmail.com> 11309Date: Thu Mar 12 03:42:40 2009 -0400 11310 11311 RS600: fix page table size for rs600 as well 11312 11313commit ef69336584c45e35aabb797a06c4030e75d66507 11314Author: Dave Airlie <airlied@redhat.com> 11315Date: Thu Mar 12 14:54:18 2009 +1000 11316 11317 r600: fix sizing of PCI GART table for r600 11318 11319commit 945ccbbd4fa2b65ccdfb23716c178c95b036734d 11320Author: Alex Deucher <alexdeucher@gmail.com> 11321Date: Tue Mar 10 09:53:14 2009 -0400 11322 11323 radeon: clean more thoroughly in RADEONFreeRec() 11324 11325commit 03ec2099fdc859b23e5fa56ac8696ec045f1b110 11326Author: 오유연(Yu-yeon Oh) <yuyeon_oh@tmax.co.kr> 11327Date: Tue Mar 10 09:51:08 2009 -0400 11328 11329 radeon_driver.c small memory bug 11330 11331commit 60dac878cc8626b0821c4c7a6a6dfa12a5597cfa 11332Author: Alex Deucher <alexdeucher@gmail.com> 11333Date: Mon Mar 9 20:48:39 2009 -0400 11334 11335 R4xx: add R4xxATOM option 11336 11337 If set, enables the use of atombios for modesetting 11338 on r4xx cards. 11339 11340commit 758fcc33c232c32e7b66261fca51de161d1a4800 11341Author: Dave Airlie <airlied@redhat.com> 11342Date: Tue Mar 10 10:34:13 2009 +1000 11343 11344 r600: enable DRI by default 11345 11346 fallback to shadowfb if DRI fails pre init 11347 11348commit 7735e411954fd2cb6105c30ebd5a1b209974ce22 11349Author: Alex Deucher <alexdeucher@gmail.com> 11350Date: Mon Mar 9 17:49:25 2009 -0400 11351 11352 R6xx/r7xx: clarify EXA message 11353 11354commit 1d355798f3d421cabf2a3a0ba32a580483476ab0 11355Author: Alex Deucher <alexdeucher@gmail.com> 11356Date: Mon Mar 9 15:13:01 2009 -0400 11357 11358 AVIVO: fix panning 11359 11360 low 2 bits of X need to be 0. 11361 11362commit d1add18f3d238c755f9875b5e5c211e4af742482 11363Author: root <root@benzylpiperazine.boston.devel.redhat.com> 11364Date: Fri Mar 6 17:53:59 2009 -0500 11365 11366 output: Filter out dual-link modes from DP->DVI connections 11367 11368 There's not enough pins on a DP->DVI connector for this to possibly work. 11369 11370commit 04481bf450650d1ea5aadabc9213b899b0fe8119 11371Author: root <root@benzylpiperazine.boston.devel.redhat.com> 11372Date: Fri Mar 6 17:32:15 2009 -0500 11373 11374 atom: Enable DisplayPort source to DVI/HDMI sink 11375 11376 DP source to DP sink still doesn't work yet, of course. 11377 11378commit 0ba73e040b94590867f8b1071a26da2526a3c375 11379Author: Adam Jackson <ajax@redhat.com> 11380Date: Fri Mar 6 15:38:00 2009 -0500 11381 11382 output: Allow for multiple DisplayPort outputs. 11383 11384commit 8f7c17a6b77ca368f067c2acb803cc8f9ef19351 11385Author: Tormod Volden <debian.tormod@gmail.com> 11386Date: Fri Mar 6 09:42:39 2009 -0500 11387 11388 RV350: Add AGPMode quirk for Thinkpad 11389 11390 fixes bug 15096 (and Ubuntu bug #195051) 11391 11392commit 5dc4b69fbdd0e8ca48cc00321f193260a8f10bd0 11393Author: Alex Deucher <alexdeucher@gmail.com> 11394Date: Thu Mar 5 12:26:04 2009 -0500 11395 11396 radeon: add a few more default common modes for lvds 11397 11398commit 3b0fc22ad9e5c0f120a74c4f3d8e48c270f0ff29 11399Author: Alex Deucher <alexdeucher@gmail.com> 11400Date: Wed Mar 4 19:04:34 2009 -0500 11401 11402 R6xx/R7xx: wire up DMAForXv option like older asics 11403 11404 Select between SW and HW-assisted uploads 11405 11406commit 52f06ace04ad8141effc45fb6a0107a05bb46a73 11407Author: Alex Deucher <alexdeucher@gmail.com> 11408Date: Wed Mar 4 18:51:44 2009 -0500 11409 11410 R6xx/R7xx: disable XV_BICUBIC attribute 11411 11412 It's not implemented yet. 11413 11414commit b1fd883b59b85fed8782e035890098908902f4ce 11415Author: Alex Deucher <alexdeucher@gmail.com> 11416Date: Wed Mar 4 11:17:16 2009 -0500 11417 11418 RN50: fix up cloning on servers 11419 11420 Since the rn50 only has one crtc, but often has multiple outputs, 11421 occasionally one of the outputs doesn't get a crtc assigned eventhough 11422 the outputs may have common modes. If we see an output with modes, but 11423 no crtc, assign it so all the attached outputs light up. 11424 11425 fixes bugs 19457 and similar ones 11426 11427commit 592bcac52f113a95923a8f1cb8427e7552d5670b 11428Author: Alex Deucher <alexdeucher@gmail.com> 11429Date: Wed Mar 4 10:53:20 2009 -0500 11430 11431 ATOM: don't use fixed ref div for LVDS 11432 11433 The bios doesn't provide one. If there are problems 11434 with certain panels, we may need to adjust the PLL flags 11435 in certain cases. 11436 11437 Should fix bug 20458 11438 11439commit e88b0eac8cab8b54b769fd7b4ad3b3b65a90de09 11440Author: Alex Deucher <alexdeucher@gmail.com> 11441Date: Wed Mar 4 02:52:36 2009 -0500 11442 11443 radeon: just add some common modes for LVDS 11444 11445 avoids needing to hack around with the edid 11446 11447commit 2bb319fd80d20a3d4e5cc9416891bc7e4c3e431e 11448Author: Alex Deucher <alexdeucher@gmail.com> 11449Date: Wed Mar 4 02:50:45 2009 -0500 11450 11451 Revert "radeon: adjust LVDS so that default modes get added" 11452 11453 This reverts commit 1a2b16561d19ec9c027c562902f5fc086c856994. 11454 11455commit 1a2b16561d19ec9c027c562902f5fc086c856994 11456Author: Alex Deucher <alexdeucher@gmail.com> 11457Date: Tue Mar 3 20:21:17 2009 -0500 11458 11459 radeon: adjust LVDS so that default modes get added 11460 11461 we can scale, so add the default modes. for panels 11462 with an EDID, set the continous freq bit. for panels without 11463 and EDID, add a FAKE edid with the continous freq bit set. 11464 11465 Based on similar code in the Intel driver. 11466 11467commit 71117970df36cbe689ef15e9a6cca24439b4cd62 11468Author: Alex Deucher <alexdeucher@gmail.com> 11469Date: Tue Mar 3 19:40:30 2009 -0500 11470 11471 AVIVO: add aspect scaling mode 11472 11473 No luck yet for aspect on pre-avivo chips 11474 11475commit d586a2c6f821c821a4a7708a3382acb63187534f 11476Author: Alex Deucher <alexdeucher@gmail.com> 11477Date: Tue Mar 3 12:17:30 2009 -0500 11478 11479 Rotation: don't rotate if acceleration is not active 11480 11481commit fdc64948119e2218921b4dc0a492522442668875 11482Author: Alex Deucher <alexdeucher@gmail.com> 11483Date: Mon Mar 2 16:33:46 2009 -0500 11484 11485 R6xx/r7xx: remove some unneeded code I missed in the last commit 11486 11487commit 4ad1c4decfee653dbbc1ea2ca4270487be622382 11488Author: Alex Deucher <alexdeucher@gmail.com> 11489Date: Mon Mar 2 15:38:35 2009 -0500 11490 11491 R6xx/R7xx EXA: cleanup composite texture setup 11492 11493commit b7164ac4ad55e5d0fc474df8ae762b469b91ba30 11494Author: Alex Deucher <alexdeucher@gmail.com> 11495Date: Mon Mar 2 04:08:09 2009 -0500 11496 11497 R6xx/R7xx EXA: combine composite mask/non-mask VS 11498 11499 Also fix set_bool_const() 11500 the CF bool consts are not contiguous by shader type 11501 There are 96 boolean constants (32 each for PS, VS, GS) and 11502 they are ordered as follows: 11503 ps, vs, gs ... ps, vs, gs 11504 11505commit fa98f424de739be2c6005b740a74bbf1ee968a8b 11506Author: Alex Deucher <alexdeucher@gmail.com> 11507Date: Mon Mar 2 02:28:57 2009 -0500 11508 11509 R6xx/R7xx Xv: combine packed and planar shaders 11510 11511 use a bool const to select the tex fetch routine 11512 11513commit ccde35c3eda3fff0de29eb8c6fdc392629724a34 11514Author: Christian Koenig <deathsimple@vodafone.de> 11515Date: Sun Mar 1 23:38:37 2009 -0500 11516 11517 R6xx/R7xx: move shaders to r600_shader.c and fixup Xv PS 11518 11519 patches from Christian Koenig with some adjustments from me 11520 11521commit a8e631c1b1c9b46602aeca66f8e7e68154d0bfc8 11522Author: Alex Deucher <alexdeucher@gmail.com> 11523Date: Sun Mar 1 22:30:31 2009 -0500 11524 11525 R6xx/R7xx: code cleanups 11526 11527commit 528061d51bb4248d6fabec9579dead32a730467a 11528Author: Alex Deucher <alexdeucher@gmail.com> 11529Date: Sun Mar 1 18:28:13 2009 -0500 11530 11531 R6xx/R7xx: write vertexes directly to the IB 11532 11533 Reduces the vertex buffer setup overhead 11534 11535commit 96a0bafa16dbb67a3d10fa6fe45be13930495ddc 11536Author: Alex Deucher <alexdeucher@gmail.com> 11537Date: Sun Mar 1 17:58:36 2009 -0500 11538 11539 R6xx/R7xx: switch emit functions to macros 11540 11541 This improves performance due to decreased function call 11542 overhead. 11543 11544commit cbd579c41a16bddcbb8793eac2a41f2e958bef4d 11545Author: Alex Deucher <alexdeucher@gmail.com> 11546Date: Sun Mar 1 13:16:20 2009 -0500 11547 11548 Print a message when we have a shared DDC line 11549 11550commit a7f465f73363fce409870f62173d518b1bc02ae6 11551Author: Alex Deucher <alexdeucher@gmail.com> 11552Date: Sun Mar 1 12:39:31 2009 -0500 11553 11554 R300: Add AGP quirk 11555 11556 bug 7770 11557 11558commit 017bc7f34de2e40919f0e711205b78ee4e3643f6 11559Author: Tormod Volden <debian.tormod@gmail.com> 11560Date: Sat Feb 28 10:33:35 2009 -0500 11561 11562 M9+: Add AGP quirk for Sony Vaio 11563 11564commit 49b57767d0d2c041517b0764c2ed2d2ba5a7092c 11565Author: Bryce Harrington <bryce@bryceharrington.org> 11566Date: Fri Feb 27 20:31:03 2009 -0500 11567 11568 Quirk for RV280 on 82865G/PE/P DRAM Controller/Host-Hub 11569 11570 Solves issue where display suffers from line corruption. 11571 (See https://edge.launchpad.net/ubuntu/+bug/300304) 11572 11573commit 150abd4123664eab65ea43585dbdb9b3151c46af 11574Author: Tormod Volden <lists.tormod@gmail.com> 11575Date: Fri Feb 27 11:23:32 2009 -0500 11576 11577 Janitor: cosmetic clean-up of AGPMode quirk table 11578 11579commit c7d344f0ad255761d5ae30da59cf0a8b532dfc68 11580Author: Michel Dänzer <daenzer@vmware.com> 11581Date: Fri Feb 27 17:05:01 2009 +0100 11582 11583 EXA: Make sure Prepare/FinishAccess hooks can handle EXA_PREPARE_AUX* indices. 11584 11585 And signal this to the EXA core using the new EXA_SUPPORTS_PREPARE_AUX flag. 11586 11587commit f4d57105fc869e9318f447a1514f1b284a0a4ec0 11588Author: Michel Dänzer <daenzer@vmware.com> 11589Date: Fri Feb 27 09:23:24 2009 +0100 11590 11591 Revert "EXA: Adapt to EXA changes in xserver Git." 11592 11593 This reverts commit de358736dc696559ba99c71cf5b2a97508201630. 11594 11595commit 3285a4af17b19f2d1c24eebfb6aa51cddcf3ca50 11596Author: Alex Deucher <alexdeucher@gmail.com> 11597Date: Thu Feb 26 17:31:13 2009 -0500 11598 11599 DCE3.2: fix up Save()/Restore() 11600 11601commit 4fc978f3d83ef6370c9715e7ca5350640ccbfe10 11602Author: Alex Deucher <alexdeucher@gmail.com> 11603Date: Thu Feb 26 12:10:08 2009 -0500 11604 11605 R6xx/R7xx: use shadowfb if DRI is disabled 11606 11607commit 000756e052a291230e5c95e48b69a5aa9c4fab0e 11608Merge: 22d7746 8373f43 11609Author: Alex Deucher <alexdeucher@gmail.com> 11610Date: Thu Feb 26 11:44:13 2009 -0500 11611 11612 Merge branch 'r6xx-r7xx-support' of git+ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 11613 11614commit 22d7746bfc1bcbb3269afac88971254088ea9ece 11615Author: Alex Deucher <alexdeucher@gmail.com> 11616Date: Thu Feb 26 11:00:46 2009 -0500 11617 11618 RS600: enable the DRI by default 11619 11620commit 8373f4399b03961f2c928a9275d47e9f41bd92bb 11621Author: Alex Deucher <alexdeucher@gmail.com> 11622Date: Wed Feb 25 09:16:38 2009 -0500 11623 11624 R6xx/R7xx EXA: same surface and same coords equals nop 11625 11626 should fix bug 20305 11627 11628commit c74727015453ff3c3d6d06b812ebca9eb19a9767 11629Author: Alex Deucher <alexdeucher@gmail.com> 11630Date: Tue Feb 24 11:08:46 2009 -0500 11631 11632 R6xx/R7xx EXA: init copy_area to NULL 11633 11634commit 95ce13572dc2d9f5dd6cf55c23411e275c0aadf1 11635Author: Alex Deucher <alexdeucher@gmail.com> 11636Date: Tue Feb 24 10:51:13 2009 -0500 11637 11638 R6xx/R7xx EXA: Optimize temp surface for overlapping copies 11639 11640 - allocate temp surface in PrepareCopy() 11641 - fall back to old OverlapCopy() path if we are not able 11642 to allocate a temp surface 11643 11644commit 1a7db3fc2a0277d724d60d028064d8ef75019c28 11645Author: Mark van Doesburg <mark.vandoesburg@hetnet.nl> 11646Date: Tue Feb 24 10:44:19 2009 -0500 11647 11648 R6xx/R7xx EXA: use a temp surface for overlapping copy 11649 11650commit 957c2ba3459af152ad16a3f6dd4631044f9df6f9 11651Author: Alex Deucher <alexdeucher@gmail.com> 11652Date: Tue Feb 24 09:43:58 2009 -0500 11653 11654 RS600: fix MC addr mask 11655 11656 Noticed by osiris on IRC 11657 11658commit de358736dc696559ba99c71cf5b2a97508201630 11659Author: Michel Dänzer <daenzer@vmware.com> 11660Date: Tue Feb 24 08:30:23 2009 +0100 11661 11662 EXA: Adapt to EXA changes in xserver Git. 11663 11664 The PrepareAccess hook can now get six different indices. 11665 11666 Also remove superfluous instances of #include "exa.h". 11667 11668 Note that you may need to run autoheader manually to get config.h.in updated. 11669 11670commit 752b0218af36a45b333a12f66dfdce8c1e662aa8 11671Author: Michel Dänzer <daenzer@vmware.com> 11672Date: Tue Feb 24 08:13:06 2009 +0100 11673 11674 EXA: Pass pScrn and info into RadeonCompositeTile. 11675 11676 Saves some cycles looking them up. 11677 11678commit e2e508bc08d993fc108297e9cd2fa0ce183282bc 11679Author: Michel Dänzer <daenzer@vmware.com> 11680Date: Tue Feb 24 08:12:42 2009 +0100 11681 11682 Only call RADEONWaitForVLine if it might actually do anything useful. 11683 11684 Otherwise it and RADEONBiggerCrtcArea show up pretty high in x11perf -aa*text 11685 profiles. 11686 11687commit 5ab8699c97c050424cf813b9a60550aba32446a2 11688Author: Michel Dänzer <daenzer@vmware.com> 11689Date: Tue Feb 24 08:12:23 2009 +0100 11690 11691 Fix compile warning when building without EXA. 11692 11693commit 132e61741185f4a1114e66a34bdbf21e08e4d5d8 11694Author: Alex Deucher <alexdeucher@gmail.com> 11695Date: Mon Feb 23 12:13:41 2009 -0500 11696 11697 radeon: re-enable load detection output attribute for TV/CV 11698 11699commit 918d511f6f26d2fdeb63d6e32cd348707b4a2140 11700Author: Alex Deucher <alexdeucher@gmail.com> 11701Date: Mon Feb 23 10:22:19 2009 -0500 11702 11703 RS600: fix up MC setup 11704 11705 Please test if you have an RS600 11706 11707commit 27f8ca2cce65be2bcb3375231886d5444d251808 11708Author: Alex Deucher <alexdeucher@gmail.com> 11709Date: Thu Feb 19 21:18:07 2009 -0500 11710 11711 R6xx/R7xx: add wait for idle MMIO path 11712 11713commit 1f783117f79f95f1744681ccc7d80dc0e7b53057 11714Author: Alex Deucher <alexdeucher@gmail.com> 11715Date: Thu Feb 19 21:07:06 2009 -0500 11716 11717 RBBM_GUICNTL is pre-r6xx only 11718 11719commit f3933ed200dbf61b05a114f7697f2bb68d92abe7 11720Author: Alex Deucher <alexdeucher@gmail.com> 11721Date: Thu Feb 19 20:12:33 2009 -0500 11722 11723 Don't write new HDP location until we've written the new FB location 11724 11725commit 7f6801abb4bbd79830999d99ccdbd672a9984df6 11726Author: Alex Deucher <alexdeucher@gmail.com> 11727Date: Thu Feb 19 19:02:52 2009 -0500 11728 11729 radeon: one more 32 -> 64 just to be safe 11730 11731 missed this in cadae2a8d9979b2d61eae7d1f7a9ab8f5eb28e83 11732 11733commit ca22390c8a5256bbc8eeff775d5eebf0e892d890 11734Author: Alex Deucher <alexdeucher@gmail.com> 11735Date: Thu Feb 19 18:58:40 2009 -0500 11736 11737 R6xx/R7xx: fix up a few more paths 11738 11739commit cadae2a8d9979b2d61eae7d1f7a9ab8f5eb28e83 11740Author: Alex Deucher <alexdeucher@gmail.com> 11741Date: Thu Feb 19 18:50:49 2009 -0500 11742 11743 Fix MC setup on systems with more than 512 MB of VRAM 11744 11745 many thanks to app4des on IRC for helping me sort this out. 11746 11747commit 42492c70c481c88911337eefb97ba9d030adf7b6 11748Author: Alex Deucher <alexdeucher@gmail.com> 11749Date: Thu Feb 19 10:20:28 2009 -0500 11750 11751 R6xx/R7xx: wait for MC idle when changing the MC 11752 11753 should fix bug 19984 11754 11755commit e6475282486f4895bc68f6b093ecbb1aa6d25f72 11756Author: Alex Deucher <alexdeucher@gmail.com> 11757Date: Wed Feb 18 20:21:13 2009 -0500 11758 11759 R6xx/R7xx Xv: fix some missing bits from last commit 11760 11761commit adff8906c9899dde7711382577a63f4a726437ca 11762Author: Alex Deucher <alexdeucher@gmail.com> 11763Date: Wed Feb 18 20:01:57 2009 -0500 11764 11765 R6xx/R7xx EXA/Xv: properly deal with running out of vertex buffer space 11766 11767 As noted by mhopf, if VGT_MAX/MIN_INDX, etc. regs change, you need to re-emit 11768 CB blocks to avoid a hang. So, just set the VGT_MAX_INDX to a reasonably large value 11769 in the default state and don't touch them when drawing. When we run out of VB space, 11770 just draw the current buffer, grab a new one, and continue. 11771 11772commit cf85d4a1d43a3209c7ca9307aede2c2c243f7130 11773Author: Alex Deucher <alexdeucher@gmail.com> 11774Date: Wed Feb 18 19:02:56 2009 -0500 11775 11776 R6xx/R7xx: reset 3D state after VT switch 11777 11778commit 6545f0a2967414fa518a78440060b8b69c0146ee 11779Author: Alex Deucher <alexdeucher@gmail.com> 11780Date: Wed Feb 18 18:15:18 2009 -0500 11781 11782 R6xx/R7xx: fixup accel paths 11783 11784commit 61311b1de019c5a2344928a47347a966c229b08a 11785Author: Alex Deucher <alexdeucher@gmail.com> 11786Date: Wed Feb 18 17:27:05 2009 -0500 11787 11788 Bump version post release 11789 11790commit 0d3d1f1f7b450dabd3ad7d2df26ad1c72709b29b 11791Author: Alex Deucher <alexdeucher@gmail.com> 11792Date: Wed Feb 18 17:24:57 2009 -0500 11793 11794 bump for release 11795 11796commit 76a32f8391317513538dafee49cbb61c0d756356 11797Author: Alex Deucher <alexdeucher@gmail.com> 11798Date: Wed Feb 18 16:31:44 2009 -0500 11799 11800 R6xx: Connector quirk for asus board 11801 11802 bug 19943 11803 11804commit 97b8482dba4e99088b59d3ab1396be98a70e823e 11805Author: Alex Deucher <alexdeucher@gmail.com> 11806Date: Wed Feb 18 13:27:15 2009 -0500 11807 11808 Set default RMX type to FULL on LVDS 11809 11810commit 7d22b1799b34010d34c3600d6cb02ffe839a0780 11811Author: Alex Deucher <alexdeucher@gmail.com> 11812Date: Wed Feb 18 12:55:40 2009 -0500 11813 11814 DCE3.2+: allow output cloning 11815 11816commit 1a237a40958c006c56b80850bd77b2ac6c17e030 11817Author: Alex Deucher <alexdeucher@gmail.com> 11818Date: Wed Feb 18 11:46:20 2009 -0500 11819 11820 ATOM: reset crtc initialized flag on CloseScreen() 11821 11822 additional fix needed for bug 16781 11823 11824commit 9a108f0a0b7203458673ce6221e747a166d39617 11825Author: Alex Deucher <alexdeucher@gmail.com> 11826Date: Tue Feb 17 19:52:27 2009 -0500 11827 11828 ATOM: don't unblank uninitialized crtcs 11829 11830 If the crtc timing isn't setup, you might get stuck in a loop 11831 in the BlankCRTC command table 11832 11833 fixes bug 16781 11834 11835commit 3dff20e276615e8b77177689a4a5f8d91b3e8eac 11836Author: Alex Deucher <alexdeucher@gmail.com> 11837Date: Tue Feb 17 13:14:38 2009 -0500 11838 11839 R6xx/R7xx EXA: switch to surface sync packet 11840 11841commit 31bd6d28dd35021e0e706d4f094989deb856c26a 11842Author: Alex Deucher <alexdeucher@gmail.com> 11843Date: Tue Feb 17 11:24:02 2009 -0500 11844 11845 Fix crtc routing on pre-DCE3.2 systems 11846 11847 This should fix peterz and bug 20074 11848 11849commit 3c9e00c7f2ed494976713e8f77b729e0d76409db 11850Author: Alex Deucher <alexdeucher@gmail.com> 11851Date: Mon Feb 16 16:18:42 2009 -0500 11852 11853 RV350: add another AGP quirk 11854 11855commit 036a17bec5dabbb91640b907811da060dac154d6 11856Author: Alex Deucher <alexdeucher@gmail.com> 11857Date: Mon Feb 16 13:10:45 2009 -0500 11858 11859 RV280: add another agp quirk 11860 11861commit 00ac9d2f8d58725d0d3e6b1bf7f728d9ae7e6109 11862Author: David Miller <davem@davemloft.net> 11863Date: Sun Feb 15 18:26:14 2009 +1000 11864 11865 GART: Save/restore GART table consistently. 11866 11867 Always save the GART table with the RADEON_SURFACE_CNTL register 11868 zero'd out to make sure we always use the same endinanness. 11869 11870 fixed up as per Michel's suggestion for endianness. 11871 11872 Signed-off-by: David S. Miller <davem@davemloft.net> 11873 Signed-off-by: Dave Airlie <airlied@linux.ie> 11874 11875commit 2de90c8349409eb892edc57fd24b61b679eb7719 11876Author: David Miller <davem@davemloft.net> 11877Date: Sun Feb 15 18:24:01 2009 +1000 11878 11879 DRI: Fix page size used in RADEONDRIGetPciAperTableSize(). 11880 11881 The ATI GART has a fixed size of 4096 bytes. So using 11882 the system page size here is wrong. 11883 11884 Signed-off-by: David S. Miller <davem@davemloft.net> 11885 Signed-off-by: Dave Airlie <airlied@linux.ie> 11886 11887commit 7cde00f49649e25fd5816927c7a5e28b608fabcd 11888Author: Alex Deucher <alexdeucher@gmail.com> 11889Date: Thu Feb 12 17:39:44 2009 -0500 11890 11891 R6xx/R7xx: switch to drm for wait for idle 11892 11893 THIS REQUIRES AN UPDATED DRM 11894 11895commit e22cd4011b9be437ba89bff568e7fb82b4907d99 11896Author: Yang Zhao <yang@yangman.ca> 11897Date: Thu Feb 12 14:46:53 2009 -0500 11898 11899 R6xx/R7xx EXA: Further optimizations to overlapping copy 11900 11901 Diagonal overlapping copies can be reduced to either horizontal- or 11902 vertical-only offset, and the one with fewer copies is picked. 11903 11904commit da08b760bcf3d04d775c4440fafec10657bb1863 11905Author: Alex Deucher <alexdeucher@gmail.com> 11906Date: Thu Feb 12 13:53:11 2009 -0500 11907 11908 R6xx/R7xx EXA: handle running out of vertex space in the copy path 11909 11910commit e3be312b0b73982c24f1f5d9cf76d7caafae0853 11911Author: Alex Deucher <alexdeucher@gmail.com> 11912Date: Thu Feb 12 13:48:36 2009 -0500 11913 11914 R6xx/R7xx EXA: properly handle non repeat cases in the texture setup 11915 11916commit 3ff1eb7e50fef22adb6280dd7de77c3ecafdb451 11917Author: Michel Dänzer <daenzer@vmware.com> 11918Date: Thu Feb 12 19:24:41 2009 +0100 11919 11920 EXA: If making a pixmap offscreen fails, return ~0ULL as texture offset. 11921 11922 This allows AIGLX to fall back to the non-zero-copy code rather than messing up 11923 the 3D driver. 11924 11925commit c19aa4fc8da7ac4745624098b146bcc42c0436dd 11926Author: Michel Dänzer <daenzer@vmware.com> 11927Date: Thu Feb 12 19:22:12 2009 +0100 11928 11929 EXA: The source tiling code can't handle RepeatReflect yet. 11930 11931commit adc35636aa1e936ec74cba8931ccffe2b5c11656 11932Author: Alex Deucher <alexdeucher@gmail.com> 11933Date: Thu Feb 12 13:10:53 2009 -0500 11934 11935 R1xx/R2xx EXA: fix non repeat texture setup 11936 11937commit c6e1bf75c399807716b99486ab3301aeba60fa79 11938Author: Alex Deucher <alexdeucher@gmail.com> 11939Date: Thu Feb 12 12:49:24 2009 -0500 11940 11941 R3xx-R5xx EXA: fix texture setup for non-repeat case 11942 11943 fixes bug 19923 11944 11945commit 0804dde9012cf33f614a1c2ee6d5d0def040a9af 11946Author: Alex Deucher <alexdeucher@gmail.com> 11947Date: Thu Feb 12 12:02:57 2009 -0500 11948 11949 AVIVO: fix dualhead/rotation for real 11950 11951commit 3e5ac32f747bef70903e9cdf41652a827eece707 11952Author: Alex Deucher <alexdeucher@gmail.com> 11953Date: Wed Feb 11 17:40:20 2009 -0500 11954 11955 Revert "R6xx/R7xx: handle running out of vertex buffer space" 11956 11957 This reverts commit 4fd7228de7ad88edf825dbc3039df877795a9479. 11958 11959 Causes lockups. 11960 11961commit bd141aa73a77f68301715fb3b5664e2082327a80 11962Author: Alex Deucher <alexdeucher@gmail.com> 11963Date: Wed Feb 11 15:31:47 2009 -0500 11964 11965 R6xx/R7xx Xv: switch packed over to Yang's new shader code 11966 11967commit cb4a1ceaa32d6847b146a31291772e1183972ee7 11968Author: Yang Zhao <yang@yangman.ca> 11969Date: Wed Feb 11 15:18:00 2009 -0500 11970 11971 R6xx/R7xx Xv: Planar - Properly scale Y'CbCr values before converting to RGB 11972 11973 According to MPEG-2 spec, Y' and Cb/Cr values are scaled to [16, 235] 11974 and [16, 240], respectively, when packed into bytes. Properly take care 11975 of the reverse scaling before translating to RGB. 11976 11977 Conversion matrix has been simplified to remove 3rd column, as the fitting 11978 to [-0.5, 0.5] can be done with scaling. 11979 11980 Redundant MOV instructions were also removed, and now only 3 GPRs are required. 11981 11982commit 4fd7228de7ad88edf825dbc3039df877795a9479 11983Author: Alex Deucher <alexdeucher@gmail.com> 11984Date: Wed Feb 11 14:46:02 2009 -0500 11985 11986 R6xx/R7xx: handle running out of vertex buffer space 11987 11988 draw current VB and start a new one. 11989 11990commit ec60ef094762901ede4df5cde55c3f162e8c667b 11991Author: Alex Deucher <alexdeucher@gmail.com> 11992Date: Wed Feb 11 13:05:27 2009 -0500 11993 11994 R6xx/R7xx: be more verbose about what function ran out of VB space 11995 11996commit e7dedbc355970407eddb38370b58a8c96d204c1d 11997Author: Alex Deucher <alexdeucher@gmail.com> 11998Date: Wed Feb 11 12:45:14 2009 -0500 11999 12000 R6xx/R7xx: Move engine idle to sync functions 12001 12002commit 4a759a907ecd571460c7ff64cd6288380359b04b 12003Author: Alex Deucher <alexdeucher@gmail.com> 12004Date: Wed Feb 11 12:29:50 2009 -0500 12005 12006 adjust alignment 12007 12008commit d98359b20e1eb8eae893581ab199c299c8352dc4 12009Author: Alex Deucher <alexdeucher@gmail.com> 12010Date: Wed Feb 11 12:03:56 2009 -0500 12011 12012 Revert "Radeon EXA: wait for the engine to be idle before sw access" 12013 12014 This reverts commit 0c4694c58f2ec152b19023626bc7ced891d9da65. 12015 12016 This patch just adds latency and thus isn't a real solution 12017 12018commit d6c50b221cbab6e726948e2310c1def9fc38da64 12019Author: Alex Deucher <alexdeucher@gmail.com> 12020Date: Wed Feb 11 11:05:02 2009 -0500 12021 12022 R6xx/R7xx Xv: add support for packed uploads 12023 12024commit 8e437e996cc3f2c424c342701f4aa6bcf72ad08e 12025Author: Alex Deucher <alexdeucher@gmail.com> 12026Date: Wed Feb 11 10:53:50 2009 -0500 12027 12028 R6xx/R7xx Xv: Add native support for packed formats 12029 12030commit 0c4694c58f2ec152b19023626bc7ced891d9da65 12031Author: Alex Deucher <alexdeucher@gmail.com> 12032Date: Tue Feb 10 17:08:37 2009 -0500 12033 12034 Radeon EXA: wait for the engine to be idle before sw access 12035 12036 Really we need proper fencing with timestamps but this should 12037 help in the short term. 12038 12039 This should help with minor corruption issues and possibly 12040 lockups. 12041 12042commit 13ab61d90bb58112bef4ce7267ef127935315f80 12043Author: Fabio <fabio.ped@libero.it> 12044Date: Tue Feb 10 10:50:30 2009 -0500 12045 12046 man page updates 12047 12048 bug 20030 12049 12050commit 615f2f202cee5f0dd2c6b29329f843bb1c9c70d5 12051Author: Alex Deucher <alexdeucher@gmail.com> 12052Date: Tue Feb 10 10:17:47 2009 -0500 12053 12054 Bump version post RC release 12055 12056commit 1b4afc1c9f8458f9ab3434418f4ccf959f532ac1 12057Author: Alex Deucher <alexdeucher@gmail.com> 12058Date: Mon Feb 9 19:50:37 2009 -0500 12059 12060 R6xx/R7xx: Add checks to make sure we don't overrun VB space 12061 12062commit ff6efe513e0c3148812d6a9578d2265522417053 12063Author: Alex Deucher <alexdeucher@gmail.com> 12064Date: Mon Feb 9 17:43:24 2009 -0500 12065 12066 ATOM: warning fixes 12067 12068commit 87ad9006f3dafb45e131c9a3d6df713321f39998 12069Author: Alex Deucher <alexdeucher@gmail.com> 12070Date: Mon Feb 9 17:40:41 2009 -0500 12071 12072 RV350: add AGP quirk 12073 12074 bug 19981 12075 12076commit 610b8b65fd7aa97f9c03b2ff6ae607aa959eab0b 12077Author: Alex Deucher <alexdeucher@gmail.com> 12078Date: Mon Feb 9 16:33:20 2009 -0500 12079 12080 Bump for rc release 12081 12082commit e85b7a1c5948b5123eddf9dfbb9e002d74fbe9fb 12083Author: Alex Deucher <alexdeucher@gmail.com> 12084Date: Mon Feb 9 14:13:16 2009 -0500 12085 12086 R6xx/R7xx Xv: fix cache flush buffer size for planar 12087 12088commit 2a893bac1faffd28dce6d9a9693108196c71e6d3 12089Author: Alex Deucher <alexdeucher@gmail.com> 12090Date: Mon Feb 9 13:18:17 2009 -0500 12091 12092 R6xx/R7xx Xv: add accelerated uploads for planar formats 12093 12094commit 231aee18a73805be2f6c962e94a8345dd89fd0df 12095Author: Alex Deucher <alexdeucher@gmail.com> 12096Date: Mon Feb 9 13:02:27 2009 -0500 12097 12098 R6xx/R7xx Xv: implement native shader for planar formats 12099 12100commit 6c76bfe8105e3cf4e7e6ea1bfe1235be2079110f 12101Author: Alex Deucher <alexdeucher@gmail.com> 12102Date: Mon Feb 9 09:47:22 2009 -0500 12103 12104 R6xx/R7xx UTS: move actual upload to separate function 12105 12106 So it can be shared with Xv 12107 12108commit 04859362ce64ed4ae48305a83c92059899c1db4b 12109Author: Alex Deucher <alexdeucher@gmail.com> 12110Date: Sun Feb 8 20:13:06 2009 -0500 12111 12112 Add some missing r6xx/r7xx pci ids 12113 12114commit 132e4c575dc4675f4995e45f08c53c26bffd999a 12115Author: Yang Zhao <yang@yangman.ca> 12116Date: Sat Feb 7 13:54:51 2009 -0500 12117 12118 R6xx/R7xx EXA: Optimize overlapping copy 12119 12120 Overlapping copy is now done in chunks proportional to the 12121 non-overlapping area. 12122 12123 Diagonal overlaps are also handled properly. 12124 12125commit 8e9ef8ff581892cbe1b7ea56d48b9a1abd70179d 12126Author: Pierre Ossman <pierre@ossman.eu> 12127Date: Sat Feb 7 18:57:47 2009 +0100 12128 12129 Xv vsync support on r6xx/r7xx cards. 12130 12131commit 2222f0fd700f100b2e91fac2babe7d1b53f56c3e 12132Author: Pierre Ossman <pierre@ossman.eu> 12133Date: Sat Feb 7 18:56:42 2009 +0100 12134 12135 Fix bad range adjustment in VLINE code. 12136 12137commit 8ce6c024e31f6a3f5ae6c882738b1e64ae2944a1 12138Author: Alex Deucher <alexdeucher@gmail.com> 12139Date: Sat Feb 7 10:32:01 2009 -0500 12140 12141 R6xx/R7xx Xv: fix typos in cache flushing commands 12142 12143commit bba5d806cc46328ad9e4c521bfeeb2a59f96862b 12144Author: Alex Deucher <alexdeucher@gmail.com> 12145Date: Fri Feb 6 19:44:17 2009 -0500 12146 12147 AVIVO: better fix for rotation 12148 12149 This should handle initial rotation as well. 12150 12151commit 28cf3492f431f325cb08d7c6aade43d8886df3f6 12152Author: Alex Deucher <alexdeucher@gmail.com> 12153Date: Fri Feb 6 19:20:18 2009 -0500 12154 12155 AVIVO: fix rotation 12156 12157 When rotation is active, virtualX and virtualY change. 12158 12159commit 1b45936ae614244aa49b1a5d3c7fc39773c4f9b6 12160Author: Alex Deucher <alexdeucher@gmail.com> 12161Date: Fri Feb 6 18:29:55 2009 -0500 12162 12163 R6xx/R7xx EXA: Fix typo in DFS 12164 12165 noticed by pzad in IRC 12166 12167commit c06d89e16d5b2553142e8641e66080e1770c1563 12168Author: Alex Deucher <alexdeucher@gmail.com> 12169Date: Fri Feb 6 18:07:22 2009 -0500 12170 12171 R6xx/R7xx EXA: fallback on overlapping blits for now 12172 12173 Leave this disabled until we get a proper solution. 12174 12175commit 1d5fc3febf3470b94c423a1eda5e0683856909df 12176Author: Alex Deucher <alexdeucher@gmail.com> 12177Date: Fri Feb 6 14:43:30 2009 -0500 12178 12179 Revert "r6xx/r7xx EXA: Optimize overlapping copy" 12180 12181 This reverts commit 0dfadc1843e0d14b9cc1ee19a72f4fd60a2c495b. 12182 12183 This doesn't always work properly. 12184 12185commit def317e22d072405cd95ddb19d17eacd784ffd9e 12186Author: Alex Deucher <alexdeucher@gmail.com> 12187Date: Fri Feb 6 11:04:59 2009 -0500 12188 12189 R6xx/R7xx EXA: add accelerated UTS/DFS hooks 12190 12191 I'm not sure how much of a win these are. I need to 12192 do some benchmarking. 12193 12194commit 0dfadc1843e0d14b9cc1ee19a72f4fd60a2c495b 12195Author: Yang Zhao <yang@yangman.ca> 12196Date: Fri Feb 6 10:29:39 2009 -0500 12197 12198 r6xx/r7xx EXA: Optimize overlapping copy 12199 12200 When source and destination blocks are only offset horizontally, it 12201 appears to be unnecessary to perform careful, segment-by-segment copy. 12202 The code path that does this is taken out completely. 12203 12204 For the case where offset is only vertical, copying is now done by 12205 height of the non-overlapping area each time, instead of always 12206 line-by-line. 12207 12208commit 3d17bd199423e92d201c20f047d5e699942af976 12209Author: Alex Deucher <alexdeucher@gmail.com> 12210Date: Thu Feb 5 19:25:03 2009 -0500 12211 12212 Revert "R6xx/R7xx EXA: improve overlapping copy performance" 12213 12214 This reverts commit b24827c9d211e8a35da53b665385a7733d19910d. 12215 12216 This seems to cause corruption in some cases. 12217 12218commit bf74055f543e7f0664741620fb1fe827ebc12711 12219Author: Alex Deucher <alexdeucher@gmail.com> 12220Date: Thu Feb 5 17:31:43 2009 -0500 12221 12222 r6xx/r7xx EXA: fix corruption when doing sw access 12223 12224 need to wait until the engine is idle. Ideally we wait 12225 on a timestamp shadowed in memory, but polling the 12226 GRBM_STATUS reg will do for now. 12227 12228commit b24827c9d211e8a35da53b665385a7733d19910d 12229Author: Alex Deucher <alexdeucher@gmail.com> 12230Date: Thu Feb 5 16:21:37 2009 -0500 12231 12232 R6xx/R7xx EXA: improve overlapping copy performance 12233 12234 send vertices for each line of the copy, but only draw once 12235 12236commit 729fe756f809a41494dc161252d82313956e4a4b 12237Author: Alex Deucher <alexdeucher@gmail.com> 12238Date: Thu Feb 5 16:08:35 2009 -0500 12239 12240 r6xx/r7xx EXA: cleanup overlapping copy 12241 12242commit 16e01a5796a8e8b86ad25d3aa45b9e7044dc72d1 12243Author: Alex Deucher <alexdeucher@gmail.com> 12244Date: Tue Feb 3 19:46:10 2009 -0500 12245 12246 Fix encoder accounting 12247 12248 Should fix bug 19924 12249 12250commit d7bf7b9c17add31dc9a115b56a563c7f9bf2cdb6 12251Author: Alex Deucher <alexdeucher@gmail.com> 12252Date: Tue Feb 3 18:28:38 2009 -0500 12253 12254 EXA: fix and re-enable Solid() on R7xx 12255 12256 missing last bit in alu clause in solid PS 12257 12258commit 9abb09d5f43320bd27b336612d39d264fe93549f 12259Author: Alex Deucher <alexdeucher@gmail.com> 12260Date: Tue Feb 3 11:17:13 2009 -0500 12261 12262 Fix bad rv710 pci id 12263 12264commit e5b916770946a9eebcb4bd1e6f698220db8c718a 12265Author: Alex Deucher <alexdeucher@gmail.com> 12266Date: Tue Feb 3 10:48:41 2009 -0500 12267 12268 Allow rotation on r6xx/r7xx 12269 12270commit d1f071c7f1dad6babfbcfcc2cb2b722a4987f372 12271Author: Alex Deucher <alexdeucher@gmail.com> 12272Date: Tue Feb 3 10:44:10 2009 -0500 12273 12274 Initial R6xx/R7xx EXA and textured video support 12275 12276commit c88c3ef6f3db266c1aacba5297b8dfc8b66bf00e 12277Author: Wolke Liu <wolke.liu@amd.com> 12278Date: Mon Feb 2 17:01:34 2009 -0500 12279 12280 AVIVO: Save/restore vga pll registers 12281 12282 This fixes some VT switch issues on some chips 12283 12284commit 6fac3cefd1f46161c1e276ba40e72da2823aa9f6 12285Author: Alex Deucher <alexdeucher@gmail.com> 12286Date: Sat Jan 31 14:21:02 2009 -0500 12287 12288 Return NULL for encoder if no active device is assigned 12289 12290 fixes bug 19855 12291 12292commit db178c852d00e9b85513d750b5f8f7495f9e3360 12293Author: Alan Coopersmith <alan.coopersmith@sun.com> 12294Date: Fri Jan 30 20:37:41 2009 -0800 12295 12296 Add README with pointers to mailing list, bugzilla & git repos 12297 12298commit fa8e5a4fc236f8f15f462cb0d6164b194a65a118 12299Author: Michel Dänzer <daenzer@vmware.com> 12300Date: Fri Jan 30 16:16:26 2009 +0100 12301 12302 EXA: Accelerate Composite of RepeatPad/Reflect pictures when possible. 12303 12304commit 3a6e958c59b4b44fe620c8abe760ed1abf0007b1 12305Author: airlied <airlied@roo.(none)> 12306Date: Fri Jan 30 07:56:14 2009 +1000 12307 12308 rs780: include RS780 in the InitMemory to leave alone 12309 12310 This stops the MC_FB_LOCATION getting reassigned to 0, which 12311 is really bad thing to happen. 12312 12313 I've had reported memory corruption on these cards so hopefully this 12314 fixes it. 12315 12316commit bd8021d46e9066e4cd116c03a7b7adcfe2557aff 12317Merge: 4f88dd8 20d5dd3 12318Author: Alex Deucher <alexdeucher@gmail.com> 12319Date: Wed Jan 28 14:33:14 2009 -0500 12320 12321 Merge branch 'atom-rework' and fix conflicts 12322 12323commit 20d5dd387da555e895e2b73fb53e2b026dd91003 12324Author: Alex Deucher <alexdeucher@gmail.com> 12325Date: Wed Jan 28 14:12:26 2009 -0500 12326 12327 Move encoder specific data to encoder dev_priv 12328 12329commit 4f88dd83f290fb38e41256a7f1804008c0f28139 12330Merge: 1c5678f 2ca970c 12331Author: Alex Deucher <alexdeucher@gmail.com> 12332Date: Wed Jan 28 13:13:13 2009 -0500 12333 12334 Merge branch 'atom-tvout' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 12335 12336 Conflicts: 12337 12338 src/radeon_atombios.c 12339 12340commit 1a62360e8260eaf5f98fa6be3f89f64c8cae7d32 12341Author: Alex Deucher <alexdeucher@gmail.com> 12342Date: Mon Jan 26 10:11:28 2009 -0500 12343 12344 Encoders not assigned yet, use supported devices 12345 12346 need to move this to encoder setup 12347 12348commit 44d9ceacb1fa6e1859d5d8bd1c55405aea24ad85 12349Author: Alex Deucher <alexdeucher@gmail.com> 12350Date: Mon Jan 26 03:47:32 2009 -0500 12351 12352 Fix legacy output setup 12353 12354 don't have assigned encoders at this point. Need to 12355 eventually move this stuff to encoder dev_priv. 12356 12357commit 0c24cbcea56c8f36374aca0f8ba849adfbf0ecfe 12358Author: Alex Deucher <alexdeucher@gmail.com> 12359Date: Mon Jan 26 03:24:19 2009 -0500 12360 12361 ATOM: print useful output info for DPMS events 12362 12363commit 2f92b885457c5cee2098ebbf8c6fe5a5eb98ec48 12364Author: Alex Deucher <alexdeucher@gmail.com> 12365Date: Mon Jan 26 03:12:44 2009 -0500 12366 12367 Warning fixes 12368 12369commit 4cbf6189f3a7945a1141b0451e4f7881ceaaadcd 12370Author: Alex Deucher <alexdeucher@gmail.com> 12371Date: Mon Jan 26 02:59:51 2009 -0500 12372 12373 Move legacy output setup functions to legacy_output.c 12374 12375commit 93ed767255b60fbdf5b416b9bd06c366036a5141 12376Author: Alex Deucher <alexdeucher@gmail.com> 12377Date: Mon Jan 26 02:40:41 2009 -0500 12378 12379 Fix off by one when printing encoder name 12380 12381 - also cleanup some debugging output 12382 12383commit b515eee4b1cafae78ed39cd91c18b4f5e671b20b 12384Author: Alex Deucher <alexdeucher@gmail.com> 12385Date: Mon Jan 26 01:50:18 2009 -0500 12386 12387 Additional output cleanup 12388 12389commit 1c5678fc452dbc93135e9ece50a4cff6583f6078 12390Author: Michel Dänzer <daenzer@vmware.com> 12391Date: Sat Jan 24 11:10:00 2009 +0100 12392 12393 Drop memcpy fallbacks from EXA UploadToScreen and DownloadFromScreen hooks. 12394 12395 While in theory it's possible for the PrepareAccess hook to fail on big endian 12396 platforms, in practice it isn't at this point because there's no other users of 12397 the surface registers. 12398 12399commit 5cbd6082b9f8d7009ee9ef5e036673f99e1bfb05 12400Author: Michel Dänzer <daenzer@vmware.com> 12401Date: Sat Jan 24 11:07:12 2009 +0100 12402 12403 Don't transform EXA Composite mask coordinates when there's no mask. 12404 12405commit 7e54c4b8acafead8de5d0dd6fdc1b619f832dd6f 12406Author: Thomas Jaeger <ThJaeger@gmail.com> 12407Date: Fri Jan 23 20:03:27 2009 -0500 12408 12409 Fall back to software for unsupported repeat modes 12410 12411commit 289b4a377cfff0c2a8ec938951cfbae1e270570e 12412Author: Alex Deucher <alexdeucher@gmail.com> 12413Date: Fri Jan 23 20:00:43 2009 -0500 12414 12415 Remove OutputType and other cruft 12416 12417commit 5f98508effbdc1fd62d0df15798722c147d88da1 12418Author: Maciej Cencora <m.cencora@gmail.com> 12419Date: Fri Jan 23 19:33:25 2009 -0500 12420 12421 Make sure gb_num_pipes is initialized when DRI is disabled 12422 12423 Found with valgrind 12424 12425commit 0faab02cc732414af3cfb2014d3ddc1139aff063 12426Author: Alex Deucher <alexdeucher@gmail.com> 12427Date: Fri Jan 23 17:35:01 2009 -0500 12428 12429 Remove some unused cruft 12430 12431commit 025082160c53c01a44f4d681cc63bf26ac4561d5 12432Author: Alex Deucher <alexdeucher@gmail.com> 12433Date: Fri Jan 23 17:31:18 2009 -0500 12434 12435 track encoder state 12436 12437commit 82f12e5a40c1fbcb91910a0f8b725c34fff02aae 12438Author: Alex Deucher <alexdeucher@gmail.com> 12439Date: Fri Jan 23 16:21:33 2009 -0500 12440 12441 Remove TMDSType, DACType, LVDSType from output rec 12442 12443commit 7b42b57d748f77ff6ef6f8ade7cbc983a98d7204 12444Author: Alex Deucher <alexdeucher@gmail.com> 12445Date: Wed Jan 21 16:28:00 2009 -0500 12446 12447 few more logic pasto's bits I missed 12448 12449commit fef1c12fbe4ceb4676258850d1fa5cac994f7c21 12450Author: Alex Deucher <alexdeucher@gmail.com> 12451Date: Wed Jan 21 15:34:01 2009 -0500 12452 12453 Fix compilation with RADEON_TRACE_FALL set 12454 12455 noticed by curaga on phoronix forum 12456 12457commit 1305376b009d17d65b4935896dc35cb7530f7c52 12458Author: Alex Deucher <alexdeucher@gmail.com> 12459Date: Wed Jan 21 15:20:47 2009 -0500 12460 12461 Move active_device setup to detect() 12462 12463 Also add encoder debugging printouts 12464 12465commit 9e0cc0ca5d7bc68756ae92ee3b5b2392366be71a 12466Author: Alex Deucher <alexdeucher@gmail.com> 12467Date: Wed Jan 21 14:47:08 2009 -0500 12468 12469 Fix logic cut and paste error 12470 12471commit 12010ae9296635ffa5f42ce0cd14482aaa290cf7 12472Author: Alex Deucher <alexdeucher@gmail.com> 12473Date: Wed Jan 21 13:52:11 2009 -0500 12474 12475 More legacy rework 12476 12477commit 3f8bbede8bf639050f773780888b89cbdcb425b8 12478Author: Alex Deucher <alexdeucher@gmail.com> 12479Date: Wed Jan 21 13:45:11 2009 -0500 12480 12481 fix legacy crtc routing and add some debugging info 12482 12483commit 80e0162ea2c1f793964731268251aba2d3243127 12484Author: Alex Deucher <alexdeucher@gmail.com> 12485Date: Wed Jan 21 13:28:14 2009 -0500 12486 12487 ATOM: fix encoder init 12488 12489commit aea27d9caabb0e54e868a21e6e547733a4ca2709 12490Author: Alex Deucher <alexdeucher@gmail.com> 12491Date: Wed Jan 21 12:33:49 2009 -0500 12492 12493 Switch legacy output code to use new encoder objects 12494 12495commit 9b44a4b57b00b150d041a5d82cb2df42159c062b 12496Author: Alex Deucher <alexdeucher@gmail.com> 12497Date: Wed Jan 21 11:32:24 2009 -0500 12498 12499 ATOM: more output cleanup 12500 12501commit 9e283fa9de34407fa6f0d5a8006a611b104f4bf2 12502Author: Alex Deucher <alexdeucher@gmail.com> 12503Date: Wed Jan 21 03:16:20 2009 -0500 12504 12505 Fixup encoder setup on pre-ATOM chips 12506 12507commit 5850abbdcd22b70bdbccd3642f8ba1a27e38e7cd 12508Author: Alex Deucher <alexdeucher@gmail.com> 12509Date: Wed Jan 21 02:28:47 2009 -0500 12510 12511 clean up encoder setup 12512 12513 - make sure not to duplicate encoders 12514 - use single encoder array 12515 12516commit 152d1d4c179c262be4d0a5618400aa13b4820342 12517Author: Alex Deucher <alexdeucher@gmail.com> 12518Date: Tue Jan 20 00:58:18 2009 -0500 12519 12520 First pass at converting legacy code to encoder objects 12521 12522commit 7cc7ba8bf9ce2f0eebfd9b48aa648c416c333ff4 12523Author: Alex Deucher <alexdeucher@gmail.com> 12524Date: Mon Jan 19 18:42:12 2009 -0500 12525 12526 ATOM: round 1 of output rework 12527 12528commit 25f3878858e6efb486888e5271b60d4ba08c73c6 12529Author: Alex Deucher <alexdeucher@gmail.com> 12530Date: Mon Jan 19 15:54:42 2009 -0500 12531 12532 start to re-org outputs 12533 12534commit cbeb310d35e5ea25ba7a45a2c555468391cdf29e 12535Author: Alex Deucher <alexdeucher@gmail.com> 12536Date: Fri Jan 16 19:53:49 2009 -0500 12537 12538 ATOM: switch to define for external tmds 12539 12540commit d0d157abced43f822885fa26f14a4924766594cd 12541Author: Alex Deucher <alexdeucher@gmail.com> 12542Date: Fri Jan 16 19:51:07 2009 -0500 12543 12544 ATOM: combine DAC setup functions 12545 12546 plus a few other clean-ups 12547 12548commit f6f969338f8187529c7ae9b2adc980b71a3f5837 12549Author: Alex Deucher <alexdeucher@gmail.com> 12550Date: Fri Jan 16 18:10:34 2009 -0500 12551 12552 DCE30: LVTMA requires DIG2 encoder 12553 12554 Fixes bug 19544 12555 12556commit 549ae63aecc2624a69b41ce4db75883d03d59c23 12557Author: Alex Deucher <alexdeucher@gmail.com> 12558Date: Thu Jan 15 19:26:40 2009 -0500 12559 12560 RV280 Add another AGP quirk 12561 12562commit 0e69068a8ed82e3fb7349d8224933451955e5f16 12563Author: Tormod Volden <debian.tormod@gmail.com> 12564Date: Tue Jan 13 20:31:17 2009 -0500 12565 12566 [PATCH] Add yet another AGP quirk for RV280 12567 12568commit b437da34cb4f7c637674fb432ad81b9bf32244a3 12569Author: Christiaan van Dijk <dvbmail@xs4all.nl> 12570Date: Mon Jan 12 11:54:15 2009 -0500 12571 12572 R3xx/R4xx: Maximize the use of clipped triangles for Xv rendering 12573 12574 While trying to get tear free video on my RS690 I found some problems in the code for rendering the 12575 video as a clipped triangle. The size limitations are quite strict and scaling the video output window 12576 creates all kinds of weird effects. This patch maximizes the use of this function and resolves the 12577 scaling issues. There could be a problem with zero target height/width, not sure if this is handled 12578 elsewhere. 12579 12580commit 3d106b990fa8daba96e590c57b164640a2faae7f 12581Author: Alan Coopersmith <alan.coopersmith@sun.com> 12582Date: Fri Jan 9 16:26:04 2009 -0800 12583 12584 Remove xorgconfig & xorgcfg from See Also list in man page 12585 12586commit 14d29a223028c183ba11e717a3737285d0ac8de1 12587Author: Alex Deucher <alexdeucher@gmail.com> 12588Date: Wed Jan 7 17:04:42 2009 -0500 12589 12590 RV280: add another AGP quirk 12591 12592commit 1795706a6bb13272a58e8fa1bd5ad2603d553dde 12593Author: Nicos Gollan <gtdev@spearhead.de> 12594Date: Wed Jan 7 14:25:24 2009 -0500 12595 12596 [PATCH] Fixed enumerations in radeon-output.c 12597 12598commit f69ad5aa17bb907105c1657803628d52b5ff3f11 12599Author: Alex Deucher <alexdeucher@gmail.com> 12600Date: Wed Jan 7 11:33:53 2009 -0500 12601 12602 Bump version post release 12603 12604commit a0dd5d7ee3f038a9bfe051db8dbfac4934a81581 12605Author: Alex Deucher <alexdeucher@gmail.com> 12606Date: Tue Jan 6 19:08:28 2009 -0500 12607 12608 Bump version for release 12609 12610commit c3fb8bb279959512a4ced644a64cc660c5cd97bd 12611Author: Alex Deucher <alexdeucher@gmail.com> 12612Date: Tue Jan 6 17:41:22 2009 -0500 12613 12614 ATOM: rework encoder/transmitter setup 12615 12616commit 9bb6b8ceaf772e1ce63610812a12260ee3860c27 12617Author: Alex Deucher <alexdeucher@gmail.com> 12618Date: Tue Jan 6 16:25:26 2009 -0500 12619 12620 ATOM: refactor output dpms 12621 12622commit 467995d960bdefb1911eabfc2a047cf816013101 12623Author: Alex Deucher <alexdeucher@gmail.com> 12624Date: Tue Jan 6 15:18:41 2009 -0500 12625 12626 ATOM: Adjust PLL setup for recent atom changes 12627 12628commit ee6310174c44809c761dbdb423e56e61392a0e10 12629Author: Alex Deucher <alexdeucher@gmail.com> 12630Date: Tue Jan 6 14:59:22 2009 -0500 12631 12632 ATOM: handle cases where TMDS uses linkb 12633 12634 fixes bug 18564 12635 12636commit 690999b682c0f954e27d9f1b3a3d8db4ef865c21 12637Author: Alex Deucher <alexdeucher@gmail.com> 12638Date: Tue Jan 6 14:55:49 2009 -0500 12639 12640 ATOM: rework object table parsing 12641 12642commit e23baa2fa8b25b68cd473aec7a3d8cf3ee7b0ea8 12643Author: Alex Deucher <alexdeucher@gmail.com> 12644Date: Tue Jan 6 14:38:06 2009 -0500 12645 12646 R100: Fix typo in e55598cc1b35d398c9eb2c3946df71456775f621 12647 12648 Noticed by Andreas Stenglein 12649 12650commit d5cb24075928690377fd2dfaa32abfebff9990df 12651Author: Alex Deucher <alexdeucher@gmail.com> 12652Date: Sun Jan 4 21:04:52 2009 -0500 12653 12654 ATOM: add format rev 2 for digitial command tables 12655 12656 should fix bug 18911 12657 12658commit 1813456af3def38c6f53103054976f3092491746 12659Author: Alex Deucher <alexdeucher@gmail.com> 12660Date: Sun Jan 4 21:01:05 2009 -0500 12661 12662 ATOM: fail on more tables with unhandled revisions 12663 12664commit 90ce00a3dbf37833a71db234372695460c8f0342 12665Author: Alex Deucher <alexdeucher@gmail.com> 12666Date: Sun Jan 4 19:13:53 2009 -0500 12667 12668 AGP: Print both device and subsytem pci ids for agp 12669 12670 helpful for adding quirks with libpciaccess 12671 12672commit 006d736d890a6b2f1d7eb8ee20ccf467a939516f 12673Author: Alex Deucher <alexdeucher@gmail.com> 12674Date: Sun Jan 4 19:08:37 2009 -0500 12675 12676 Another AGP quirk 12677 12678 debian bug 510208 12679 12680commit c39ca66a7e216b7492fc5e3680ac535831cfb9a1 12681Author: Alex Deucher <alexdeucher@gmail.com> 12682Date: Tue Dec 30 19:05:58 2008 -0500 12683 12684 R2xx: add another AGP quirk 12685 12686 from http://www.phoronix.com/forums/showthread.php?t=14626 12687 12688commit 877be1b268f275b8ad83d5e245a0f0d3b1a05091 12689Author: Fabio <fabio.ped@libero.it> 12690Date: Mon Dec 29 18:11:10 2008 -0500 12691 12692 man page updates 12693 12694 bug 19329 12695 12696commit 837f7e8a48ea35d21de24b620830a2c6db854861 12697Author: Dave Airlie <airlied@linux.ie> 12698Date: Tue Dec 30 06:50:54 2008 +1000 12699 12700 atombios: endian fix in r600 object table handling 12701 12702 This fixes the r600 object table on r600. 12703 12704commit 45f8b911659dba7f43b9b70bde4b9c3be51b0b5b 12705Author: Alex Deucher <alexdeucher@gmail.com> 12706Date: Mon Dec 29 10:25:33 2008 -0500 12707 12708 RS600: remove gpio quirk for ddc 12709 12710 Doesn't seem necessary on RS600. I'll try and find out more. 12711 12712commit e913990030c536427bfecdb7b3d1e18f1beb90d2 12713Author: Alex Deucher <alexdeucher@gmail.com> 12714Date: Mon Dec 29 10:22:20 2008 -0500 12715 12716 Revert "radeon: maybe special handling only needed for DVI port not HDMI one." 12717 12718 This reverts commit c877450a75d3fd4a2f200af197c691669966f6f4. 12719 12720 This breaks RS690 setups. I think for now we just apply this quirk to 12721 RS690/RS740. I'll follow up with the bios folks about RS600. 12722 12723commit c877450a75d3fd4a2f200af197c691669966f6f4 12724Author: Dave Airlie <airlied@linux.ie> 12725Date: Mon Dec 29 09:05:45 2008 +1000 12726 12727 radeon: maybe special handling only needed for DVI port not HDMI one. 12728 12729 fixes ciaranm's rs600 board on radeon DDC 12730 12731commit 2a9bc6be3246ec230b181391221e3a46beb94a23 12732Author: Dave Airlie <airlied@linux.ie> 12733Date: Mon Dec 29 09:05:16 2008 +1000 12734 12735 radeon: a-bit motherboard has no DVI output 12736 12737 thanks to ciaranm for finding this. 12738 12739commit eccf3fed2be54f5c02fefef32e39c4084acbbe22 12740Author: Alex Deucher <alexdeucher@gmail.com> 12741Date: Fri Dec 26 19:51:58 2008 -0500 12742 12743 Make sure to include radeon_drm.h in the tarball 12744 12745commit e55598cc1b35d398c9eb2c3946df71456775f621 12746Author: Alex Deucher <alexdeucher@gmail.com> 12747Date: Fri Dec 26 19:31:55 2008 -0500 12748 12749 R2xx EXA: limit composite to 2047 like r1xx 12750 12751 See bug 19269 12752 12753commit 631123d144d088d4f77a0599c34deaa31d551d71 12754Author: Dave Airlie <airlied@redhat.com> 12755Date: Tue Dec 23 10:08:46 2008 +1000 12756 12757 radeon: update to 6.9.0.91 12758 12759 do a prerelease for end of year. 12760 12761commit 047f7603174f6047090f4fafbe6488d697eb78bc 12762Author: Dave Airlie <airlied@linux.ie> 12763Date: Tue Dec 23 08:16:25 2008 +1000 12764 12765 radeon: pass distcheck 12766 12767commit c0c33dab44e6966b1702d4e8cfba3537fc6e2d5c 12768Author: Patrick Haller <patrick.haller@haller-consult.com> 12769Date: Mon Dec 22 03:06:23 2008 -0500 12770 12771 Fix off by one in EXA composite limit checking 12772 12773 Patch from Patrick, with some updates from me: 12774 - fix r200 limits 12775 - note about r100 limits 12776 12777commit d01a609a2b07da4ca0f182e79459432584ec61f6 12778Author: Alex Deucher <alexdeucher@gmail.com> 12779Date: Mon Dec 22 02:18:43 2008 -0500 12780 12781 R3xx-R5xx: better fix for xv primitive tearing issues 12782 12783 R5xx: always use single clipped triangle 12784 R3xx/R4xx: use single clipped triangle up to guardband limit, 12785 then use quad. 12786 12787commit 4d009d14c9c72f6e1bc8100ec8d9b8037ee4e514 12788Merge: d52882d 4e96278 12789Author: Dave Airlie <airlied@redhat.com> 12790Date: Mon Dec 22 16:00:01 2008 +1000 12791 12792 Merge branch 'atom-tvout' 12793 12794commit 4e96278b581e296c1203d97a6d7aa3bff3977222 12795Author: Dave Airlie <airlied@redhat.com> 12796Date: Mon Dec 22 15:58:35 2008 +1000 12797 12798 atombios/tv: add an option to enable atom tv-out for users. 12799 12800 This code is still experimental but we will allow users to enable it for experimental reasons 12801 12802commit d52882d8188830dd52fa112dadcf5ea7f3e5fd5e 12803Author: Dave Airlie <airlied@redhat.com> 12804Date: Mon Dec 22 09:27:47 2008 +1000 12805 12806 radeon: setup 3D engine even when no DRI. 12807 12808 This should fix some missing font issues in EXA without DRI cases. 12809 12810commit 2346fd7cbd90dcdce2b361d374a53da064fa6a12 12811Author: Dave Airlie <airlied@redhat.com> 12812Date: Thu Dec 18 16:35:13 2008 +1000 12813 12814 radeon: add all new pci ids for rv730/rv710 families 12815 12816commit 2ca970c7aa1f3f212f03bf3287d6b60eeafd68ae 12817Author: Alex Deucher <alexdeucher@gmail.com> 12818Date: Wed Dec 17 13:07:19 2008 -0500 12819 12820 Allow arbitrary tv-out modes 12821 12822commit 5da147eec70abb7c9125a8940089fd3ff8bdea95 12823Author: Alex Deucher <alexdeucher@gmail.com> 12824Date: Wed Dec 17 05:37:58 2008 -0500 12825 12826 missed one in last commit 12827 12828commit 6bfbb930c549e67f68969f7f0483380417f755b5 12829Author: Alex Deucher <alexdeucher@gmail.com> 12830Date: Wed Dec 17 05:36:09 2008 -0500 12831 12832 Make sure we hit the right bios reg 12833 12834commit 6ed0e68060fd2775e2d4027ff1533729e87848ba 12835Author: Alex Deucher <alexdeucher@gmail.com> 12836Date: Wed Dec 17 04:54:50 2008 -0500 12837 12838 properly handle EnableYUV 12839 12840 A bit of a hack. we really need to restructure the 12841 bios scratch reg handling. 12842 12843commit f21e673aced1bb5138bbb80e6d9724530eb95ebf 12844Author: Alex Deucher <alexdeucher@gmail.com> 12845Date: Wed Dec 17 03:17:45 2008 -0500 12846 12847 Fix colors on tv-out 12848 12849 Call EnableYUV to set the transform and enable/disable it. 12850 should also fix green tint with tv connected at boot issues. 12851 12852commit b2b43905a5385a8bb0b59b8e50952863d8dacb59 12853Author: Alex Deucher <alexdeucher@gmail.com> 12854Date: Wed Dec 17 00:59:52 2008 -0500 12855 12856 Pre-avivo: fix FP setup 12857 12858 - make sure to clear various shadow timing bits 12859 - crtc1 select bit was set wrong. 12860 - should fix bugs like 19100 12861 12862commit 3754cc23abac99880eda87d06aa42e16081b6c8a 12863Author: Alex Deucher <alexdeucher@gmail.com> 12864Date: Wed Dec 17 00:56:19 2008 -0500 12865 12866 R2xx: switch EXA composite to rects rather quads 12867 12868 For conistency with Xv path and to reduce potential diagonal tearing 12869 12870commit be12cf8fcc7a5a3c3ee2c630474e2f9d9368f4d4 12871Author: Dave Airlie <airlied@redhat.com> 12872Date: Wed Dec 17 15:28:25 2008 +1000 12873 12874 radeon: r500 tv-out force scaler values to nice set that looks correct 12875 12876commit 01f7a50d76f706a456de7a4eb8323939d905c82b 12877Author: Dave Airlie <airlied@redhat.com> 12878Date: Wed Dec 17 13:43:21 2008 +1000 12879 12880 r500: re-enable TV out 12881 12882commit 45b9ff5e250d69de0df41f68b33524f7e2ce13b7 12883Author: Dave Airlie <airlied@redhat.com> 12884Date: Wed Dec 17 13:42:10 2008 +1000 12885 12886 radeon: r500 PAL timings are slightly incorrect 12887 12888 this stablises the PAL image at least 12889 12890commit 435d996949185201a97493e131f9920a32863acd 12891Author: Dave Airlie <airlied@redhat.com> 12892Date: Wed Dec 17 11:50:48 2008 +1000 12893 12894 radeon: tv timings have crev/frev wrong way around for some reason. 12895 12896commit b9549e89cbdf8fff4aa7bd9f3adfb7d5732508cc 12897Author: Dave Airlie <airlied@redhat.com> 12898Date: Wed Dec 17 11:40:53 2008 +1000 12899 12900 atombios: re-enable TV dpms 12901 12902commit db85ff2e0c52a030546c0cc8ed0b851418f3e045 12903Author: Dave Airlie <airlied@redhat.com> 12904Date: Wed Dec 17 11:37:17 2008 +1000 12905 12906 atom/r600: re-enable TV outputs 12907 12908commit ff7b52c9a48f2e417acf6bec3b72546ad9b33471 12909Author: Dave Airlie <airlied@redhat.com> 12910Date: Wed Dec 17 11:36:38 2008 +1000 12911 12912 radeon: use latest atombios constants for TV encoders in set_crtc_source 12913 12914commit 1e0c7d3f0bf6be67b976fdea8fa2df0b4b16435f 12915Author: Dave Airlie <airlied@redhat.com> 12916Date: Wed Dec 17 11:35:06 2008 +1000 12917 12918 atom: spc3 only requires ucEncoderMode for DIG outputs 12919 12920commit 105cfd0b8696347ef71ad9daa941abb308bd6c2c 12921Author: Dave Airlie <airlied@redhat.com> 12922Date: Wed Dec 17 11:31:15 2008 +1000 12923 12924 atom/tv: enable scaler can set the TV output scaler up. 12925 12926 Add the tv standard to the scaler enable for TV use. 12927 12928commit 2a768abbd2f18abe203571213260bccc225f70fd 12929Author: Dave Airlie <airlied@redhat.com> 12930Date: Wed Dec 17 11:29:07 2008 +1000 12931 12932 atom/tv: fixup the tv out default if bios doesn't specify 12933 12934commit b7fdd5da08bfec705691363b8890699c794dce53 12935Author: Dave Airlie <airlied@redhat.com> 12936Date: Wed Dec 17 11:24:56 2008 +1000 12937 12938 atombios: fixup parsing of TV tables on r500 and r600 cards. 12939 12940 The TV tables changed between for later cards. 12941 12942commit e3b08cb59716b525b3063b184eab70e32b9d66f0 12943Author: Dave Airlie <airlied@redhat.com> 12944Date: Tue Dec 16 15:43:16 2008 +1000 12945 12946 atombios: correct fix for previous issue. 12947 12948 I committed an old patch. 12949 12950commit 1431bc6900b0e0d7f486b5b3ba281873472b3fdc 12951Author: Dave Airlie <airlied@redhat.com> 12952Date: Tue Dec 16 15:31:34 2008 +1000 12953 12954 atombios: fix issue with Clearing of MM_INDEX 0. 12955 12956 While playing with tv-out I discovered this issue with clearing 12957 the MM_INDEX register. 12958 12959commit f72367aff2d09161d30066a00673ba291b292a36 12960Author: Dave Airlie <airlied@redhat.com> 12961Date: Tue Dec 16 12:32:24 2008 +1000 12962 12963 radeon: add initial rv730 pciids. 12964 12965 Lots more to come, this just adds a few to start. 12966 12967commit 4d2429f3ec4d7d3e1512b268c7ff5cbff8d186b4 12968Author: Dave Airlie <airlied@redhat.com> 12969Date: Tue Dec 16 12:19:00 2008 +1000 12970 12971 DCE32: add support for DCE3.2 digital outputs. 12972 12973 This adds support for the digital outputs on the RV730. 12974 12975commit 4d3792fd1b9f733107582436540f483d8993199f 12976Author: Dave Airlie <airlied@redhat.com> 12977Date: Fri Dec 12 10:34:57 2008 +1000 12978 12979 radeon: fixes from Alex for some output engines 12980 (cherry picked from commit 1e8ac6ea7b35ccbc3649c96ead60f69158f66ebc) 12981 12982commit f892154dbd0d8b6dd8abdf2196ab0c7c18327412 12983Author: Alex Deucher <alexdeucher@gmail.com> 12984Date: Mon Dec 15 19:41:32 2008 -0500 12985 12986 R3xx-R5xx: switch back to quad rendering for Xv 12987 12988 The large clipped triangle trick hits the guardband limits on 12989 large displays. This gets textured video working again 12990 until we sort out how to deal with the situation better. 12991 12992 See bug 19046 12993 12994commit 62344e460f5d0cac81f94c535cda77d3d76528a9 12995Author: Alex Deucher <alexdeucher@gmail.com> 12996Date: Mon Dec 15 11:30:42 2008 -0500 12997 12998 Remove remnants of unused DDCMode option 12999 13000commit f1efee6a11dc9436276fb12cfb90aa79ada1192b 13001Author: Alex Deucher <alexdeucher@gmail.com> 13002Date: Mon Dec 15 11:27:17 2008 -0500 13003 13004 Remove mention of MergedFB from server log 13005 13006 noticed by Fabio, bug 19098 13007 13008commit 94bffac6968057cdedc792f8540fe44cb6ecfd7c 13009Author: Fabio <fabio.ped@libero.it> 13010Date: Mon Dec 15 11:22:56 2008 -0500 13011 13012 man page updates 13013 13014 from bug 19098 with some minor fixes from me 13015 13016commit c7899905ec824071f9cabbd1e271c4d76010b733 13017Author: Fabio <fabio.ped@libero.it> 13018Date: Mon Dec 15 10:58:23 2008 -0500 13019 13020 Remove duplicate ATI in some device names 13021 13022 from bug 19098 13023 13024commit 89a8fd5337c77abf1f3166e69ed061d85b5c20f0 13025Author: Dave Airlie <airlied@redhat.com> 13026Date: Mon Dec 15 14:41:10 2008 +1000 13027 13028 r600: don't idle engine 13029 (cherry picked from commit b77729448e487d37707ce028d8edc59352d1cc2d) 13030 13031commit 4ce255a9db8027f2471ba50e57724d3e4d5b16ba 13032Author: Fabio <fabio.ped@libero.it> 13033Date: Fri Dec 12 09:53:33 2008 -0500 13034 13035 Fix formatting in man page 13036 13037 fixes bug 19039 13038 13039commit 0e99017cb9a6d46b891705df73678c2705439532 13040Author: Joseph Adams <joeyadams3.14159@gmail.com> 13041Date: Thu Dec 11 00:57:03 2008 -0500 13042 13043 Add more eMac modes, fix crash in last patch 13044 13045commit fb46c30d316ab3ec54c54f3aec91d5164070a423 13046Author: Alex Deucher <alexdeucher@gmail.com> 13047Date: Wed Dec 10 17:16:56 2008 -0500 13048 13049 Add support for Apple eMacs 13050 13051 Thanks to Joseph Adams for helping me sort this out. 13052 Verified on eMac G4/1.0 with radeon 7500 13053 Later eMacs with radeon 9200 or 9600 chips may have 13054 different ddc setups. Need to verify. 13055 13056commit c0bcea9150ef215fa614733cef9a5b71a55a33bd 13057Author: Alex Deucher <alexdeucher@gmail.com> 13058Date: Wed Dec 10 11:40:47 2008 -0500 13059 13060 Add another AGP quirk 13061 13062 fixes bug 12544 13063 13064commit 607b02d6abf78147902a5944bd006897ad95e982 13065Author: Alex Deucher <alexdeucher@gmail.com> 13066Date: Wed Dec 10 03:54:35 2008 -0500 13067 13068 Add randr 1.3 panning support 13069 13070commit 24007acbf36bcfd09cb5e74ce4a8d918d393cf34 13071Author: Benjamin Close <Benjamin.Close@clearchain.com> 13072Date: Wed Dec 10 02:17:56 2008 -0500 13073 13074 [PATCH] xf86misc extension has been removed, update driver to reflect it 13075 13076commit 69e445f7f2ebd299070e44457b434bbc28b7e05c 13077Author: Dave Airlie <airlied@linux.ie> 13078Date: Tue Dec 9 18:59:11 2008 +1000 13079 13080 radeon: memset the pll setting function 13081 13082 Thanks to Peter Zijlstra for making me debug this and providing 13083 feedback. 13084 13085commit c72b732a2f5597cd72bd4ebe207799ae07f2681e 13086Author: Adrian Friedli <adi@koalatux.ch> 13087Date: Mon Dec 8 10:42:46 2008 -0500 13088 13089 Add another AGP quirk 13090 13091commit 979ad04d901676cb41aea5cf47a8a2e9c19064dd 13092Author: Alex Deucher <alexdeucher@gmail.com> 13093Date: Fri Dec 5 17:53:57 2008 -0500 13094 13095 Add missing .TP in man page 13096 13097commit b89d434011787fb71593395f462d142375783e35 13098Author: Pierre Ossman <pierre@ossman.eu> 13099Date: Fri Dec 5 17:52:02 2008 -0500 13100 13101 Fix scissor setup for Xv 13102 13103 We need to adjust the scissoring for each clip box, so move the setup 13104 to inside the while loop. 13105 13106commit 83c1115333cbe763f86422fa5aabf4ab21951e68 13107Author: Alex Deucher <alexdeucher@gmail.com> 13108Date: Fri Dec 5 11:52:18 2008 -0500 13109 13110 R200: Remove scissor setup from init3d() 13111 13112 We aren't using them, so no need to set them up. 13113 13114commit 7593d5399cdb2f42ecbcb5193580385b286a9db9 13115Author: Alex Deucher <alexdeucher@gmail.com> 13116Date: Fri Dec 5 11:18:15 2008 -0500 13117 13118 set coherent mode consistently (enabled for now) 13119 13120 It's set to enabled in the output setup, but then was set to 13121 disabled when teh output properties were setup. Set it up 13122 consistently. 13123 13124commit ad2579f8898251105a6b36b745afd1ce1dab103e 13125Author: Dave Airlie <airlied@linux.ie> 13126Date: Fri Dec 5 20:03:53 2008 +1000 13127 13128 radeon: damn you bios, HDMI cannot have DACs. 13129 13130commit bae30856f2cb3884387ef347327af6ff00b75854 13131Author: Alex Deucher <alexdeucher@gmail.com> 13132Date: Fri Dec 5 08:15:59 2008 +0100 13133 13134 Make VSync for EXA and Xv configurable 13135 13136commit 33638d9e388b330e2f4eb4debd05ba09924cf176 13137Author: Pierre Ossman <pierre@ossman.eu> 13138Date: Thu Dec 4 23:29:31 2008 +0100 13139 13140 Optimise RADEONWaitForVLine 13141 13142 Only avoid the vlines we are rendering to, instead of the entire 13143 screen. This way we don't stall the card for longer than we 13144 absolutely have to. 13145 13146 EXA calls fixed by Alex Deucher. 13147 13148commit d1690f5cc096e2f735c8b407c370a1c1cd7a8afc 13149Author: Pierre Ossman <pierre@ossman.eu> 13150Date: Wed Dec 3 21:49:37 2008 +0100 13151 13152 Improve tearing avoidance for Xvideo in two steps 13153 13154 - Fix up VLINE handling to trigger whenever scanout is outside the 13155 visible area. 13156 13157 - Render the video as a scissored triangle as R300+ cannot render a 13158 quad in a single pass. 13159 13160commit 4d98acbca2e630056bf56cdcd0e23007fded2ced 13161Author: Alex Deucher <alexdeucher@gmail.com> 13162Date: Wed Jul 16 03:14:06 2008 -0400 13163 13164 First pass at tear-free accel 13165 13166 if the dest pixmap is the front buffer, stall the pipe 13167 until the vline is outside the active area. 13168 For EXA, pick crtc based on the larger mode area; 13169 ideally we'd have one pixmap per crtc. 13170 For Xv, use dst window area to determine crtc. 13171 13172commit 82d67b70d0fb747cf7b1eb0b64244ae56cb84572 13173Author: Pierre Ossman <pierre@ossman.eu> 13174Date: Thu Dec 4 15:38:13 2008 -0500 13175 13176 Switch r200 Xv to use rect lists rather than quads to avoid tearing 13177 13178 Patch from Pierre with fixes from me. 13179 13180commit 9eccd0ad0827a1b8ec9abb2cebf5768d204a5ce9 13181Author: Alex Deucher <alexdeucher@gmail.com> 13182Date: Thu Dec 4 12:28:12 2008 -0500 13183 13184 Fix up US setup after r3xx bicubic merge 13185 13186commit e13fba853ba19e6b0f081b9b3d9fa76c38a0f82b 13187Merge: d296337 79bbdd9 13188Author: Alex Deucher <alexdeucher@gmail.com> 13189Date: Thu Dec 4 12:25:29 2008 -0500 13190 13191 Merge branch 'bicubic' of git://git.infradead.org/users/drzeus/xf86-video-ati 13192 13193commit d29633708b5451f5541e88371d831ae03019a7f5 13194Author: Dave Airlie <airlied@redhat.com> 13195Date: Thu Dec 4 09:08:48 2008 +1000 13196 13197 radeon: apply atom quirks to object table 13198 13199commit 941ac0d3ab839dc5aaaef2f3b7f5c7085b2894c5 13200Author: Alex Deucher <alexdeucher@gmail.com> 13201Date: Wed Dec 3 17:31:25 2008 -0500 13202 13203 Make sure 3D state is emited when switching to 3D IB 13204 13205 - fixes bug 18864 13206 13207commit 79bbdd984c925e37f5b3db2605339f1a21377fcf 13208Author: Pierre Ossman <pierre@ossman.eu> 13209Date: Wed Dec 3 20:33:36 2008 +0100 13210 13211 Change the XV_BICUBIC attribute to a tristate, where the third state 13212 automatically turns the filter on and off as appropriate. 13213 13214commit bbdf821c179242266fd86edf9e688a6be9cf985f 13215Author: Pierre Ossman <pierre@ossman.eu> 13216Date: Wed Dec 3 19:42:50 2008 +0100 13217 13218 Make the R300 bicubic shader program a bit easier to understand by 13219 having comments that describe the operation, not how it has to be 13220 implemented in the hardware. 13221 13222commit 8f4b22932c91f1d51fcbf1d1a99d852fcdaafdcc 13223Author: Pierre Ossman <pierre@ossman.eu> 13224Date: Wed Dec 3 19:31:31 2008 +0100 13225 13226 Now that we have a bicubic code path that messes up the pixel shader, 13227 we need to restore it in the non-bicubic code. 13228 13229commit 0851daa6e7db77e440188d774ef8b3236146bcb9 13230Author: Pierre Ossman <pierre@ossman.eu> 13231Date: Wed Dec 3 19:17:36 2008 +0100 13232 13233 Fix macros so that they are safe to use with expressions as parameters. 13234 13235commit e04bcb0066781f61e97787856261e3380036d304 13236Author: Pierre Ossman <pierre@ossman.eu> 13237Date: Wed Dec 3 19:16:03 2008 +0100 13238 13239 Fix node setup on R300 bicubic pixel shader. It was backwards and used 13240 the wrong base for sizes. 13241 13242commit 927aee181f3637bda1b761caf93fbc0652e1207b 13243Author: Alex Deucher <alexdeucher@gmail.com> 13244Date: Wed Dec 3 01:39:38 2008 -0500 13245 13246 Only flush IB when direct rendering is enabled 13247 13248commit ca74ebf8813b7fab8d08465ef9600f001c948e94 13249Author: Alex Deucher <alexdeucher@gmail.com> 13250Date: Tue Dec 2 22:26:21 2008 -0500 13251 13252 move 3d state init after RADEON_SWITCH_TO_3D() 13253 13254commit fc079c5267baf431bbecee7744e484783d393152 13255Author: Dave Airlie <airlied@redhat.com> 13256Date: Tue Dec 2 22:11:57 2008 -0500 13257 13258 Don't mix 2D and 3D in the same IB 13259 13260 fix from radeon-gem-cs 13261 13262commit 678693aabd0b4c1812ab44bf5abb4c8394b254bb 13263Author: Dave Airlie <airlied@redhat.com> 13264Date: Tue Dec 2 22:02:21 2008 -0500 13265 13266 Stability fixes from radeon-gem-cs 13267 13268commit f231e4d9b2c683dc16f42cf09484fae926a892c1 13269Author: Dave Airlie <airlied@redhat.com> 13270Date: Wed Dec 3 11:39:04 2008 +1000 13271 13272 radeon: fix HDMI having DACs in some BIOSes 13273 13274commit fe73d9a7dfe8ec5c8f1a8dc08e14b4e138aa9276 13275Author: Bryce Harrington <bryce@canonical.com> 13276Date: Tue Dec 2 19:47:45 2008 -0500 13277 13278 Add another AGP quirk 13279 13280commit 6e0c48958c6a50a639368e369ce72d813256aee4 13281Author: Alex Deucher <alexdeucher@gmail.com> 13282Date: Mon Dec 1 18:50:40 2008 -0500 13283 13284 Add quirk for r200 card with the primary dac wired to both ports 13285 13286 refactor load detection setup as a byproduct 13287 should fix bug 18719 13288 13289commit fa496d7b0397d9be57db90d0860928e9ced73cca 13290Author: Pierre Ossman <pierre@ossman.eu> 13291Date: Mon Dec 1 08:29:29 2008 +0100 13292 13293 Fix comments for R500 fragment shader to reflect the code. 13294 13295commit 0d95fca114194664d9ab612c90d3222912cf41ff 13296Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13297Date: Sat Aug 30 12:38:43 2008 -0700 13298 13299 Fix indentation on IS_R300_3D Xv code. 13300 13301commit 36a62b8c47def3f23a0b735875f59d23ed3c095a 13302Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13303Date: Sat Aug 30 11:15:20 2008 -0700 13304 13305 r3xx: Various Xv fixes. 13306 13307commit 01ed5462fad56f5b7bf867041d00443bdf81c653 13308Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13309Date: Thu Aug 7 18:08:54 2008 -0700 13310 13311 Switch to Mesa-style 24-bit float packing. 13312 Seems like Mesa's got it right, so we should follow suit. 13313 13314commit 85bbe889045aa434f71a3eb0ba8153b7d2cc5023 13315Author: Maciej Cencora <m.cencora@gmail.com> 13316Date: Thu Aug 7 07:06:46 2008 -0700 13317 13318 Enable bicubic Xv on r3xx+. Still some buggies in the actual display, but looks pretty good. 13319 13320commit bf6263cff50bb85c19ff6dfebb739333fa8aca18 13321Author: Maciej Cencora <m.cencora@gmail.com> 13322Date: Thu Aug 7 07:00:36 2008 -0700 13323 13324 Add r3xx-specific bicubic filtering code. FP, RS, mostly, everything else is shared with r5xx. 13325 13326commit da021c36bbdf3bca31ee50ebe01cdb9495c09b36 13327Author: Dave Airlie <airlied@linux.ie> 13328Date: Sun Nov 30 18:44:24 2008 +1000 13329 13330 radeon_drm.h: remove kernel defines 13331 13332 remove these we should update from an installed kernel copy not from the 13333 original kernel copy. 13334 13335commit 783cdb7374941bb1d2b63eea375fbf2f1b808cc3 13336Author: Dave Airlie <airlied@redhat.com> 13337Date: Thu Nov 27 15:29:06 2008 +1000 13338 13339 r100/r200: EXA misrenders dst == a8 + dst alpha use. 13340 13341 Fedora BZ#469556 13342 13343 I've played with various dst and texture formats and rewritten large 13344 parts of this to no avail, so I'm punting for now. 13345 13346commit 2d7cfc421713286edd5121b56999246f9ae70286 13347Author: Eygene Ryabinkin <rea-fbsd@codelabs.ru> 13348Date: Tue Nov 25 13:25:33 2008 -0500 13349 13350 Catch unsubstituted macro for PCI region base access 13351 13352 also fix up some warnings. 13353 13354commit af7690b9c300d37c3a0c51e27e0ad2ca009224a2 13355Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 13356Date: Tue Nov 25 02:57:30 2008 -0500 13357 13358 [PATCH] radeon: Fix PCI usage of 32-bit driver on 64-bit platform 13359 13360 The radeon driver is storing PCI addresses in unsigned long's which 13361 won't work well on 32-bit platforms with 64-bit physical address space 13362 such as PowerPC 4xx. This fixes it by using unsigned long long instead. 13363 13364commit 065938617c0feab17f4274a5350de02a692ba065 13365Author: Ramon van der Stelt <rvanderstelt@xs4all.nl> 13366Date: Tue Nov 25 02:23:46 2008 -0500 13367 13368 Interlaced mode fixes 13369 13370 see bug 12626 13371 13372commit 36a7dc6ea1e1929e986ab1159497c71521cb2f10 13373Author: Bryce Harrington <bryce@canonical.com> 13374Date: Tue Nov 25 01:04:37 2008 -0500 13375 13376 Additional AGP quirks 13377 13378 bug 18693 13379 13380commit 2acb8e4ed220fccb21b22a53c5142ab3a35d32ef 13381Author: Alex Deucher <alexdeucher@gmail.com> 13382Date: Tue Nov 25 15:07:09 2008 +1000 13383 13384 radeon: fixup shared DDC lines for some rv610 cards. 13385 13386 Some cards share DDC between a HDMI and VGA output, and expect the 13387 driver to use load detect or EDID to figure it out. 13388 13389 airlied- shipped in RHEL5 13390 13391 Signed-off-by: Dave Airlie <airlied@redhat.com> 13392 13393commit b01e35168de40d192fd7a9ce6884b9c7419afbd4 13394Author: Alex Deucher <alexdeucher@gmail.com> 13395Date: Mon Nov 24 21:06:42 2008 -0500 13396 13397 Fix up posted logic 13398 13399 noticed by benh on IRC 13400 13401commit d7a03e93372b17d642b42f5490464616ae7c99c1 13402Author: Alex Deucher <alexdeucher@gmail.com> 13403Date: Mon Nov 24 15:52:05 2008 -0500 13404 13405 Get hardcoded edid from the bios for servers that support it 13406 13407 should fix bug 11300 13408 13409commit 3858e31fa9d77d52a5cabb02f53f06385eab40f7 13410Author: Alex Deucher <alexdeucher@gmail.com> 13411Date: Mon Nov 24 14:39:41 2008 -0500 13412 13413 Don't support rotation is accel is not available 13414 13415 Fixes bug 18068 13416 13417commit 3d27876d91ef69d71f17b6cf8c07c26bf12857b4 13418Author: Dave Airlie <airlied@redhat.com> 13419Date: Mon Nov 10 15:47:13 2008 +1000 13420 13421 radeon: always align dstPitch to 64 bytes for textured video 13422 13423 bz# 18454 13424 13425commit 6c8b352264540ccc4622a7c0b704ce452dcd707d 13426Author: Dave Airlie <airlied@redhat.com> 13427Date: Wed Oct 29 15:53:57 2008 +1000 13428 13429 pciid: add HD3300 - 790GX chipset. 13430 13431 RH BZ (#466706) 13432 13433commit 902eaf768142c6c7dcc487e10775027b84cd1f9a 13434Author: Alex Deucher <alexdeucher@gmail.com> 13435Date: Thu Nov 6 15:46:43 2008 -0500 13436 13437 Check for LVDS on all IGP chips 13438 13439 - fixes bug 18395 13440 13441commit 44abaa0ea8e5e3bbe833fa73e6d011618bb6d7e7 13442Author: Adam Jackson <ajax@redhat.com> 13443Date: Mon Nov 3 15:09:28 2008 -0500 13444 13445 Fix HDMI output setup on DIG1/DIG2. 13446 13447 The only reliable way to detect an HDMI sink is to check for the 13448 appropriate EDID extension block, so this will only work right on 13449 servers with xf86DoEEDID(). RANDR 1.2 will call this internally for 13450 us if it exists, but pre-1.6 servers are out of luck. 13451 13452commit 937b7ac2a259cf504a19dcf62a58b1db1afb8eb9 13453Author: Bryce Harrington <bryce@canonical.com> 13454Date: Wed Oct 29 01:31:50 2008 -0400 13455 13456 Add AGP quirk table 13457 13458 - set known good default agp modes for problematic chip/bridge combos 13459 - should fix bug 17360, others 13460 13461commit 67bc2ae7f227a2afedc8e699cf0458fa7960f301 13462Author: Julien Cristau <jcristau@debian.org> 13463Date: Sun Oct 26 16:05:04 2008 +0100 13464 13465 Bug#7148: Fix manpage formatting 13466 13467commit 98fdd78a1229584b922f816f331a45bfc178c961 13468Author: Dave Airlie <airlied@linux.ie> 13469Date: Tue Oct 21 18:36:22 2008 +1000 13470 13471 atombios: oops typo 13472 13473commit 763dff6c117a10b6a57ba1854c5b798359789e44 13474Author: Dave Airlie <airlied@linux.ie> 13475Date: Tue Oct 21 18:35:44 2008 +1000 13476 13477 atombios: set missing pixel clock 13478 13479commit 435cf7da68186f2601c4b888296117d4f652c625 13480Author: Alex Deucher <alexdeucher@gmail.com> 13481Date: Wed Oct 15 10:52:14 2008 -0400 13482 13483 Add LCD dithering quirk for macbook pro 13484 13485 fixes bug 17897. 13486 13487commit c0e6cb6d0eeef8f2ea60d840e1cd668fa92cd7f9 13488Author: Alex Deucher <alexdeucher@gmail.com> 13489Date: Fri Oct 10 11:09:27 2008 -0400 13490 13491 Fix grey level dithering setup 13492 13493commit 9eb2b6cd28224f0cbbabf3df451696ae1a845b9c 13494Author: Dave Airlie <airlied@redhat.com> 13495Date: Fri Oct 10 16:18:05 2008 +1000 13496 13497 radeon: hopefully fix plls for 30" monitors. 13498 13499 Only do the HIGH trick on the DCE3 cards. Tested on rv610/630/770. 13500 13501commit 0975e007ec7933acf9cd8ad2f281313ef8d32ac9 13502Author: Dave Airlie <airlied@redhat.com> 13503Date: Thu Oct 9 17:16:28 2008 +1100 13504 13505 radeon: rename radeon_memory to radeon_legacy_memory. 13506 13507 This is a setup patch for the kms/memory management code. 13508 13509commit d744dc013cf5ca3141c0fdf55e36d713fe9f53ba 13510Author: Adam Jackson <ajax@redhat.com> 13511Date: Wed Oct 8 15:50:25 2008 -0400 13512 13513 Fix connector table mapping for lvtma/kaleidoscope/uniphy. 13514 13515 Fixes DPMS on LVDS. 13516 13517commit dc795ba66a8c0eb2c3fdb86bd42dfc17e0aa3b6e 13518Author: Alex Deucher <alexdeucher@gmail.com> 13519Date: Tue Oct 7 12:32:47 2008 -0400 13520 13521 Fixups from last DFP/LCD refactor 13522 13523 - fix DFP1 -> LCD1 13524 - use _INDEX rather than _SUPPORT defs 13525 13526commit 4a9e8d9fe24278b56aef51677fd35289d01584a7 13527Author: Alex Deucher <alexdeucher@gmail.com> 13528Date: Mon Oct 6 18:00:08 2008 -0400 13529 13530 set grey levels correctly for temporal dithering 13531 13532commit 5708624d1151877dd66a462d4a75fed6774604a7 13533Author: Alex Deucher <alexdeucher@gmail.com> 13534Date: Mon Oct 6 15:11:56 2008 -0400 13535 13536 Turn coherent mode off by default on DVI 13537 13538commit a5c5ce96279d01eb519bfb92b94c06a58acb7f07 13539Author: Alex Deucher <alexdeucher@gmail.com> 13540Date: Mon Oct 6 15:07:31 2008 -0400 13541 13542 Refactor atom LCD/DFP output setup 13543 13544 Simplify the code, properly handle dithering, coherent mode, encoding, etc. 13545 Should fix bug 17897. 13546 13547commit f9826a5694b7adb6920eb5bdf45d840d8fb14d53 13548Author: Alex Deucher <alexdeucher@gmail.com> 13549Date: Fri Oct 3 15:24:12 2008 -0400 13550 13551 Add support for DDC via atom commands for RV410 13552 13553 The atom calls use the hw i2c engine for DDC. For some 13554 reason, sw i2c doesn't seem to work on the VGA GPIO on 13555 RV410 chips, so we use atom in that case. 13556 13557 This fixes the longstanding VGA DDC problems on RV410/M26 13558 chips. 13559 13560commit 23c17c0121d43c2fd11bb5cfbaa4b02abac2b16d 13561Author: Alex Deucher <alexdeucher@gmail.com> 13562Date: Fri Oct 3 14:30:33 2008 -0400 13563 13564 man page updates 13565 13566 Update render accel and page flipping information. 13567 fixes bug 17887. 13568 13569commit 4ac6e499089aec3c79af5dbede5fe631b105ce3d 13570Author: Alex Deucher <alexdeucher@gmail.com> 13571Date: Wed Oct 1 18:53:15 2008 -0400 13572 13573 proper fix for initial rotation 13574 13575 The initial modeset was setting the MC regs, so just set them 13576 rather than the full modeset. 13577 13578commit 346228b20c69a965c9e7d67157f42c1d36a5b6c3 13579Author: Alex Deucher <alexdeucher@gmail.com> 13580Date: Wed Oct 1 14:38:27 2008 -0400 13581 13582 Workaround to make initial rotation work 13583 13584 Acceleration needs to be initialized before setting the mode 13585 for initial rotation to work. Changing the order in RADEONScreenInit() 13586 doesn't work because RADEONDRIKernelInit() hangs in the ioctl to 13587 initialize the CP if the mode is not set prior (even if the heads 13588 are turned off after setting the mode). The workaround is to set the 13589 modes twice, once before accel and once after. 13590 13591 This needs to be looked into further. 13592 13593commit c359c2a31caf9f75b9fc6b6bcbc3e9dc1fe404ba 13594Author: Michel Dänzer <michel@tungstengraphics.com> 13595Date: Tue Sep 30 11:04:54 2008 +0200 13596 13597 Pass base offset into RADEONDisplayVideo() explicitly. 13598 13599 This prevents the base offset from flipping back and forth if double buffering 13600 is enabled and the two buffers start on different sides of a 4 MB boundary. 13601 13602 Should fix http://bugs.freedesktop.org/show_bug.cgi?id=17254 ; please don't 13603 reopen that bug report without attaching a log file captured after reproducing 13604 the problem with this fix. 13605 13606commit ba21d5ae7fbbf5828c44c41655aff8b182453e2e 13607Author: Michel Dänzer <michel@tungstengraphics.com> 13608Date: Tue Sep 30 11:03:04 2008 +0200 13609 13610 Update GEN_INT_CNTL register value after calling DRM modeset ioctl. 13611 13612 Fixes sync-to-vblank hangs after mode switch with DRM vblank-rework. 13613 13614commit 3110646daf05fa09faaa70066cc8c3db1bb07cec 13615Author: Michel Dänzer <michel@tungstengraphics.com> 13616Date: Sun Sep 14 11:46:36 2008 +0200 13617 13618 Fix -Werror build. 13619 13620commit d82f2938f69402c641a1c8362fdc513419b27659 13621Author: Alex Deucher <alexdeucher@gmail.com> 13622Date: Fri Sep 26 13:51:24 2008 -0400 13623 13624 atombios updates from upstream 13625 13626commit d0d58b39e49c5ce00bc8bd12f721ad8c7f554b91 13627Author: Alex Deucher <alexdeucher@gmail.com> 13628Date: Wed Sep 24 11:54:34 2008 -0400 13629 13630 radeon: Further pll adjustments 13631 13632commit d100a6af8f828eb94f8ba6e8a96c24389b5cf46f 13633Author: Alex Deucher <alexdeucher@gmail.com> 13634Date: Fri Sep 19 14:04:59 2008 -0400 13635 13636 cleanup macbook quirk 13637 13638commit 1218e38c75351a20f7414ba6fcb3545689339442 13639Author: Alex Deucher <alexdeucher@gmail.com> 13640Date: Fri Sep 19 13:52:07 2008 -0400 13641 13642 more macbook fixes 13643 13644commit 83e1f132965deadedda7a622184eb400cd4c120c 13645Author: Alex Deucher <alexdeucher@gmail.com> 13646Date: Fri Sep 19 09:36:59 2008 -0400 13647 13648 Fix copy/paste error in the last commit 13649 13650 DVI-D -> DVI-I 13651 13652commit 4af825f7b13cba8c3b204e72a07cab5009331240 13653Author: Alex Deucher <alexdeucher@gmail.com> 13654Date: Fri Sep 19 00:29:20 2008 -0400 13655 13656 radeon: Add quirk for macbook 13657 13658 - should fix bug 17625 13659 13660commit 38aa450fce6a91b35c191fd07112490a62cc29c6 13661Author: Alex Deucher <alexdeucher@gmail.com> 13662Date: Tue Aug 26 21:50:23 2008 -0400 13663 13664 fix some fallout from the common allocator 13665 13666 should fix bug 17317 13667 13668commit a3cc1d7a421456186024c5c069e403d374a0f0b9 13669Author: Dave Airlie <airlied@linux.ie> 13670Date: Tue Aug 26 08:03:20 2008 +1000 13671 13672 radeon: fix powerpc build 13673 13674commit 6cebfe257f7ddad855ee743e4eb899bd6fac7f46 13675Author: Alex Deucher <alexdeucher@gmail.com> 13676Date: Fri Jul 11 19:32:06 2008 -0400 13677 13678 Switch EXA path back to static cursor allocation 13679 13680 pre-AVIVO cards have address limits for the cursor offset 13681 13682commit 4dff54a3c8d7c9f2d6ec50354ff0b92f1b7fcbdf 13683Author: Alex Deucher <alexdeucher@gmail.com> 13684Date: Fri Jul 11 17:28:03 2008 -0400 13685 13686 Switch cursors over to generic allocator 13687 13688commit 5b1978a4796bcc31ac2f01d303dc8f8f44323025 13689Author: Alex Deucher <alexdeucher@gmail.com> 13690Date: Mon Aug 25 10:05:28 2008 -0400 13691 13692 Bicubic fixes from the last cherry-pick 13693 13694commit 7daba77ded1c718e93ae8c372a39a6e85228d513 13695Author: Alex Deucher <alexdeucher@gmail.com> 13696Date: Fri Jul 11 02:02:38 2008 -0400 13697 13698 Convert randr, Xv to a common allocator 13699 13700commit 610fe1a937da78f4ac813ac919c158dde8f42442 13701Author: Alex Deucher <alexdeucher@gmail.com> 13702Date: Thu Jul 10 22:47:35 2008 -0400 13703 13704 Switch to using radeon_drm.h from the drm 13705 13706 modelled after Matthias' similar rhd change 13707 13708commit a6561f2ec673b38907f7181235386f32e60c32ba 13709Author: Alex Deucher <alexdeucher@gmail.com> 13710Date: Thu Jul 10 22:31:13 2008 -0400 13711 13712 Move DRI to a separate struct 13713 13714 based on Matthias' similar work in rhd 13715 13716commit 75ef8dc214715d3c5c50996b293933842903ba65 13717Author: Egbert Eich <eich@freedesktop.org> 13718Date: Thu Jul 10 21:49:55 2008 -0400 13719 13720 Cleanups from rhd port 13721 13722 - remove unused vars 13723 - remove static exa render vars 13724 13725commit 71ad140fa11f3a504c38d6bddf40e3a3c0a20e60 13726Author: Alex Deucher <alexdeucher@gmail.com> 13727Date: Thu Jul 10 21:24:16 2008 -0400 13728 13729 Move accel state to a separate struct 13730 13731commit 5b2e095c31b88d8495a4f86e6cb46b49fa4acd65 13732Author: Alex Deucher <alexdeucher@gmail.com> 13733Date: Thu Jul 10 20:07:44 2008 -0400 13734 13735 Move CP into a separate struct 13736 13737commit 2145309230e36aee7758bd244deb1e72ada0c065 13738Author: Alex Deucher <alexdeucher@gmail.com> 13739Date: Mon Aug 25 08:47:59 2008 -0400 13740 13741 match textured video macro names with other accel code 13742 13743 OUT_VIDEO_REG() -> OUT_ACCEL_REG() 13744 etc. 13745 13746commit 01daef0f095fbbaee701d5fe97f3dd7838b5f915 13747Author: Alex Deucher <alexdeucher@gmail.com> 13748Date: Mon Aug 25 08:26:16 2008 -0400 13749 13750 Additional cleanups and re-arragement following bicubic merge 13751 13752commit 0a51d08c24af040fe48690662b1a912acad51700 13753Author: Alex Deucher <alexdeucher@gmail.com> 13754Date: Mon Aug 25 06:43:14 2008 -0400 13755 13756 Whitespace cleanup from bucubic merge 13757 13758commit c0170fef510b53a6ca2c6ea7a99119235229c929 13759Author: Maciej Cencora <m.cencora@gmail.com> 13760Date: Thu Aug 7 06:53:39 2008 -0700 13761 13762 Add needed FP registers, etc. for r3xx bicubic Xv. 13763 13764commit d9c38326cf70f57ab777ffdf9520b8cdea9d9cb6 13765Author: Dennis Kasprzyk <onestone@opencompositing.org> 13766Date: Tue Aug 5 18:45:38 2008 +0200 13767 13768 Fix typos. 13769 13770commit d38ceba62aa5cee76baa342ce7719a983a79f09e 13771Author: Dennis Kasprzyk <onestone@opencompositing.org> 13772Date: Tue Aug 5 16:21:57 2008 +0200 13773 13774 Fix bicubic fp calculation. 13775 13776commit 6f9c7fde75edc0399559f975db9a6c4dc22714ae 13777Author: Dennis Kasprzyk <onestone@opencompositing.org> 13778Date: Tue Aug 5 15:38:42 2008 +0200 13779 13780 Fixed bicubic fragment program comments. 13781 13782commit 48b09ca40ccb28b5584069316fd38786a78c1dd3 13783Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13784Date: Sat Aug 2 02:36:21 2008 -0700 13785 13786 Switch from 32-bit floats to 16-bit half-floats. 13787 Massive bandwidth savings, or so I'm told. Yay? 13788 13789commit ebbb7fb634fcadf28ff99c1df2c3db89fd56932d 13790Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13791Date: Sat Aug 2 01:45:43 2008 -0700 13792 13793 Change floats to uint32_t hex. 13794 Useful for moving to 16-bit half-floats. 13795 13796commit 3b46162d8a90be3524cb513d42b9ad3d0bea45f5 13797Author: Dennis Kasprzyk <onestone@opencompositing.org> 13798Date: Fri Aug 1 15:24:01 2008 +0200 13799 13800 Move some ALU instructions after the TEX instructions, so we can do something usefull while we are waiting for the texture values. 13801 13802commit 2ecdec4bafc97212dde4d6908ee4ccf618adc0e1 13803Author: Dennis Kasprzyk <onestone@opencompositing.org> 13804Date: Fri Aug 1 04:03:12 2008 +0200 13805 13806 Another uneeded SEM_AQUIRE. 13807 13808commit e93b5d1b80d6203f63543b7b678e2f1d9221b5b3 13809Author: Dennis Kasprzyk <onestone@opencompositing.org> 13810Date: Fri Aug 1 03:00:26 2008 +0200 13811 13812 Smarter usage of the texture semaphore. 13813 13814commit 286f7aa18f19bd21b263701adab38b736dbeda0f 13815Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13816Date: Thu Jul 31 17:01:44 2008 -0700 13817 13818 Enable bicubic filtering for all r5xx HW. 13819 13820commit 242aa4f630b4c60aefa3c12dc459a4d4d0b334a0 13821Author: Dennis Kasprzyk <onestone@opencompositing.org> 13822Date: Fri Aug 1 01:11:28 2008 +0200 13823 13824 Remove one constant. 13825 13826commit 413eacb0538977b0b3c92df074d40510f4539abc 13827Author: Dennis Kasprzyk <onestone@opencompositing.org> 13828Date: Fri Aug 1 00:52:09 2008 +0200 13829 13830 Heavy optimizations. 13831 13832commit a0c4a949cb49e5ac1e857aef08a8742b9f7b49da 13833Author: Dennis Kasprzyk <onestone@opencompositing.org> 13834Date: Thu Jul 31 21:19:18 2008 +0200 13835 13836 Remove uneeded negations. 13837 13838commit abb2b2e757d92591ab9277824b9d9746ba98f875 13839Author: Dennis Kasprzyk <onestone@opencompositing.org> 13840Date: Thu Jul 31 20:31:03 2008 +0200 13841 13842 Implement LPR in one instruction. 13843 13844commit c370b74bec13194573348583c38adf710b880e79 13845Author: Dennis Kasprzyk <onestone@opencompositing.org> 13846Date: Thu Jul 31 19:50:49 2008 +0200 13847 13848 Set helper texture filter correctly. 13849 13850commit 5e85d5a5d0c50b29086ec0c219c8b52d25dbc2e9 13851Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13852Date: Wed Jul 30 19:45:15 2008 -0700 13853 13854 Update bicubic tables. 13855 Now including the 2048x1 texture as well. 13856 13857commit 17e5e9573e59c3d82d51c261b9c5005f6aec7d43 13858Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13859Date: Wed Jul 30 11:57:25 2008 -0700 13860 13861 Force R580-only for bicubic. 13862 Initial reading of docs suggest RV560 and RV570 can't handle it, 13863 but they're welcome to test. 13864 13865commit 83b52473d0e7102265918b07be053fcda17a14b4 13866Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13867Date: Wed Jul 30 01:03:57 2008 -0700 13868 13869 Bump bicubic cutoff to R580. 13870 RV530 chipsets can't handle fullscreen bicubic... 13871 13872commit 69a4998d1286bcdd7bfe874cd5628bc1cc232bae 13873Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13874Date: Wed Jul 30 01:03:46 2008 -0700 13875 13876 Oops, made a mistake with vertices. 13877 13878commit a4ec30a677906ec2ff9824c7ddca586655f6d1a8 13879Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13880Date: Wed Jul 30 00:54:32 2008 -0700 13881 13882 Merge upstream changes to vertices, and also add Xv attributes for textured video, including bicubic filtering. 13883 13884commit f3b81c7582aed307fa44e134ee161cd8a3158657 13885Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13886Date: Mon Jul 28 19:50:10 2008 -0700 13887 13888 Fix constants. 13889 13890commit 7dbb7023ba023ec1a38be63af9c9f49e40222b7b 13891Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13892Date: Mon Jul 28 17:30:41 2008 -0700 13893 13894 Finally got the fragment program fully working for bicubic filtering on r5xx. 13895 13896commit 0e4dd73b9ebc6f608eeff945b4d463a00c02e07c 13897Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13898Date: Mon Jul 28 12:16:50 2008 -0700 13899 13900 More bicubic FP buggies. 13901 13902commit a87647e0c27e0950f4d0d8203a1242a994ad3419 13903Author: Dennis Kasprzyk <onestone@compiz-fusion.org> 13904Date: Sun Jul 27 10:43:01 2008 -0700 13905 13906 Fix texture size, texture filter, vertex offsets, etc. 13907 13908commit 232aa3e943fef4c4037b255c3b64a0aaff90ab5c 13909Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13910Date: Sat Jul 26 10:56:56 2008 -0700 13911 13912 Make vertices emit properly. 13913 *bangs head against wall* 13914 13915commit 8c84f67b93d926095633830aa8d95930a48b1c7b 13916Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13917Date: Sun Jul 20 14:25:44 2008 -0700 13918 13919 Try to get tex coords from the VAP to RS properly when bicubic filtering is enabled. 13920 I'm soo bad at this... :c 13921 13922commit a4a7d5f5967c51c394229de5eccaec44cfec8f50 13923Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13924Date: Sat Jul 19 13:02:02 2008 -0700 13925 13926 Upload pixel shader to card for r5xx. This was ridiculous. Also it doesn't work yet... 13927 13928commit 20c1db2d7c110ab5c1117a57b169baa1ab070518 13929Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13930Date: Fri Jul 18 13:48:08 2008 -0700 13931 13932 Upload bicubic filter to card. This was a LOT easier than I had feared, to be honest. 13933 13934commit a760e628134c6d7d42ec3c98118b6e5f6fcd3e7f 13935Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13936Date: Fri Jul 18 12:21:20 2008 -0700 13937 13938 Allocate memory for the bicubic filter texture. 13939 13940commit b6c9e2bb5365de82315c6814f915e57b0c4fa444 13941Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13942Date: Fri Jul 18 11:24:14 2008 -0700 13943 13944 Fixed typos in the bicubic texture tables. Whoops, looks like I'm still asleep. 13945 13946commit ca51f4f37e1dbf53bf7ffc0e8f612e9609e11209 13947Author: Corbin Simpson <MostAwesomeDude@gmail.com> 13948Date: Fri Jul 18 11:06:34 2008 -0700 13949 13950 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". 13951 13952commit 1cf7a5494fa94e8d9f30f9b2905dfbe6d4faa445 13953Author: Bryce Harrington <bryce@bryceharrington.org> 13954Date: Wed Aug 20 09:46:59 2008 -0400 13955 13956 radeon: Fix pasto in connector table setup for vga powerbooks 13957 13958 fixes bug 17214 13959 13960commit a55e85f742d1334bf88e4681e553f025d2de38df 13961Author: Michel Dänzer <michel@tungstengraphics.com> 13962Date: Thu Aug 14 15:21:51 2008 -0400 13963 13964 Make sure video offerlay offsets don't exceed the hardware limit of 128 MB. 13965 13966 Always set the overlay base address such that the buffer offsets are as small 13967 as possible. This could still break in theory if the buffers were more than 13968 128 MB apart, but in reality this can't happen ATM because we always allocate 13969 a single memory area for all buffers. 13970 13971 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=16845 . 13972 13973commit 92ee21df344a989778e37369c7beb3904a00ead6 13974Author: Alex Deucher <alexdeucher@gmail.com> 13975Date: Thu Aug 14 14:49:45 2008 -0400 13976 13977 PLL adjustments 13978 13979 Seems higher dotclocks prefer a higher FB div. 13980 Someone with a lot of should try and find out where 13981 the div sweet spots are for various dotclock ranges. 13982 fixes bug 17125 13983 13984commit 0d5e0347af4322713075193154b8a348de4a0b52 13985Author: Alex Deucher <alexdeucher@gmail.com> 13986Date: Wed Aug 13 14:17:34 2008 -0400 13987 13988 Remove reset of 3D scissor registers when using the CP in the ddx 13989 13990 They should only affect 3D and init3d() should take care of that case 13991 noticed by libv on IRC. 13992 13993commit 7e456e9c427d48151b166bcac08c3e8a8b180f01 13994Author: Dave Airlie <airlied@linux.ie> 13995Date: Tue Aug 12 08:18:30 2008 +1000 13996 13997 radeon: add 0x9441 for hd 4870 x2 13998 13999 thanks to Michael Larabel for testing 14000 14001commit 4dbdeea7c9316575fba26b41fd347452e42cdcf2 14002Author: Alex Deucher <alexdeucher@gmail.com> 14003Date: Fri Aug 8 17:39:48 2008 -0400 14004 14005 Further cleanup and unification of i2c code 14006 14007 - unify the ddc and i2c code 14008 - add gpio mask support for legacy chips 14009 - remove the magic gpio dance for ancient monitors 14010 (if you have an ancient monitor that ddc stops 14011 working on let me know). This should speed up DDC 14012 on legacy chips. 14013 14014 -- radeon sw gpio i2c -- 14015 4 sets of gpio registers for clk and data and corresponding bit masks 14016 mask - locks the gpio for use by sw 14017 get - reads the value off the gpio pad 14018 put - sets the gpio direction to output 14019 a - "other stuff" On legacy chips you clear them if you want 14020 to use a gpio for i2c. In some cases they are used for the 14021 output value when the gpio in the output state. 14022 14023commit 268c848130ec1770bb645a74197b6aca7fc95abc 14024Author: Alex Deucher <alexdeucher@gmail.com> 14025Date: Fri Aug 8 15:50:07 2008 -0400 14026 14027 Fix VT switching on M6 chips 14028 14029 Some M6 chips have a faulty MEM_SIZE register that in 14030 some cases reports 0 on 8 MB cards. On EnterVT we check 14031 the MEM_SIZE reg as a check to see if the card is posted or 14032 not. Since this reg returns 0, the driver attempts to post 14033 the card which can lead to a hang. Switch this to check if 14034 either crtc is active as is done in the bios init code. 14035 fixes bug 13994 14036 14037commit 33f88f7fc90d9d93fdcbba9ad59dd70a6596bc3f 14038Author: Alex Deucher <alexdeucher@gmail.com> 14039Date: Thu Aug 7 17:57:07 2008 -0400 14040 14041 RS4xx DDC fixes take 3 14042 14043commit df0d1ef53100f0a19c5b5fdc349f5186c8d9bd87 14044Author: Alex Deucher <alexdeucher@gmail.com> 14045Date: Wed Aug 6 19:13:56 2008 -0400 14046 14047 RS4xx: Fix up ddc gpio 14048 14049 - I mixed up the regs previously 14050 14051commit 18429390440a829fb24ed3afd99ccf8278138496 14052Author: Alex Deucher <alexdeucher@gmail.com> 14053Date: Wed Aug 6 00:35:15 2008 -0400 14054 14055 Remove un-needed dac check for single crtc cards 14056 14057 Connector tables should be setup properly already and this 14058 check breaks systems with 2 DACs and 1 crtc like the es1000. 14059 14060commit 001c535687d8588873037ee5363d0a709f44b418 14061Author: Alex Deucher <alexdeucher@gmail.com> 14062Date: Tue Aug 5 17:14:48 2008 -0400 14063 14064 IGP: fix typo in IGP quirk handling 14065 14066commit eb65ddf70d182b6457e1ef5ebb820456039e8f6d 14067Author: Alex Deucher <alexdeucher@gmail.com> 14068Date: Tue Aug 5 15:01:33 2008 -0400 14069 14070 IGP: attempt to get external TMDS working 14071 14072 RS4xx chips have a tmds init table in the mobile info table 14073 14074commit 8b8990917809b9a35c6e9c1b9e3b12ff81c6dbb3 14075Author: Alex Deucher <alexdeucher@gmail.com> 14076Date: Tue Aug 5 12:16:06 2008 -0400 14077 14078 IGP: Updated quirk handling 14079 14080 - Hopefully finally fix DDC on RS4xx chips 14081 - RS3xx/RS4xx do not have internal TMDS 14082 - general quirk handling cleanup 14083 - we don't currently support mm gpio for external TMDS 14084 so return false. 14085 14086commit cb0deba5412a575d36f2f99377120b123506c946 14087Author: Calvin Fong <hoiwai930@gmail.com> 14088Date: Mon Aug 4 02:04:15 2008 -0400 14089 14090 IGP: add support for NTSC tv-out on legacy IGP chips 14091 14092 Patch from Calvin with some cleanups from me. 14093 No support for PAL yet. 14094 14095commit 942b18aca91819fa65d853cd15ffd1cd720cbd68 14096Author: Dave Airlie <airlied@redhat.com> 14097Date: Mon Aug 4 14:38:21 2008 +1000 14098 14099 radeon: make r600 use i2c table lookup for ddc. 14100 14101 This may fix DDC on rv770 cards. 14102 14103commit 1f3eee3682f3598a303c9c3accfbe01b245cacf9 14104Author: Alex Deucher <alexdeucher@gmail.com> 14105Date: Tue Jul 29 20:29:32 2008 -0400 14106 14107 Change prim types for exa and textured video to help avoid tearing 14108 14109 - r1xx - switch from tri fan to rect list 14110 - r2xx/r3xx/r4xx/r5xx - switch from tri fan to quad list 14111 14112commit 5b5441f8cc119db0d1e03dd35bd06015a26270dd 14113Author: Brice Goglin <bgoglin@debian.org> 14114Date: Tue Jul 29 23:31:14 2008 +0200 14115 14116 Increase default GART size to 32MB on >=R300 14117 14118commit 0a505297f09aefb1b4432176a263bfdf6f256f77 14119Author: Alex Deucher <alexdeucher@gmail.com> 14120Date: Mon Jul 28 17:12:41 2008 -0400 14121 14122 Fix error in driver connector table for powerbook w/ vga 14123 14124commit d5799ac53c3e1b2ca1da97780b171a44060c3aad 14125Author: Alex Deucher <alexdeucher@gmail.com> 14126Date: Mon Jul 28 11:09:10 2008 -0400 14127 14128 Add quirk for oem x300 card 14129 14130 - debian bug 492457 14131 14132commit 64dbadddcf6d069c0472f37afeab89d3e31e937d 14133Author: Owain Gordon Ainsworth <oga@openbsd.org> 14134Date: Mon Jul 28 10:40:58 2008 +1000 14135 14136 radeon: don't call ioctl unless DRI is initialised 14137 14138commit 1c5858484da4fb1c9bc3ac3b4d7a97863ab99730 14139Author: Alex Deucher <alexdeucher@gmail.com> 14140Date: Mon Jul 21 23:47:45 2008 -0400 14141 14142 First pass at InitDispBandwidth() for AVIVO chips 14143 14144 - support for LB allocation 14145 - MC priority bumps for display1/2 on RV515 variants and RS690 14146 If you are having display underflow problems (flickering on sides of 14147 screen in high res modes, etc.) on RV515 or RS690 boards, try setting: 14148 Option "DisplayPriority" "HIGH" in your config. 14149 - still no support for full display watermark programming yet 14150 14151 Something similar might be useful in rhd as well. 14152 14153commit b0378bb145c8a915c943bef7d17f2cdecfccc891 14154Author: Alex Deucher <alexdeucher@gmail.com> 14155Date: Mon Jul 21 13:47:09 2008 -0400 14156 14157 Interlaced mode fixups for AVIVO chips 14158 14159commit c18fad622a3c4f9572051120d83af68b625b5686 14160Author: Alex Deucher <alexdeucher@gmail.com> 14161Date: Mon Jul 21 10:36:48 2008 -0400 14162 14163 Clear display priority bits before resetting them 14164 14165commit dc231ff8e063313d2bcf5acccad67a9f8a7e3314 14166Author: Alex Deucher <alexdeucher@gmail.com> 14167Date: Mon Jul 21 10:30:41 2008 -0400 14168 14169 Clean up legacy display watermark setup 14170 14171 - makes crtc1 and crtc2 watermark setup independant. 14172 - fixes the case where only crtc2 is active 14173 14174commit f9034214f070fe3054cd9cefd7034252de234f38 14175Author: Michel Dänzer <michel@tungstengraphics.com> 14176Date: Mon Jul 21 09:09:02 2008 +0200 14177 14178 Call DRM modeset ioctl after the IRQ has been initialized. 14179 14180 This lets the DRM know it can safely disable the vblank interrupts. 14181 14182commit ecb6347a3d7071890600c98c2addef3a8ca260ee 14183Author: Alex Deucher <alexdeucher@gmail.com> 14184Date: Sat Jul 19 11:34:16 2008 -0400 14185 14186 Add oem quirk for external tmds setup on Dell Inspiron 8600 14187 14188 Noticed by fnord42 on IRC. 14189 14190commit df53d12a06fad41f00cff849458cb358ab5e2098 14191Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 14192Date: Thu Jul 17 14:37:05 2008 +1000 14193 14194 Fix console switch on R500 14195 14196 This patch fixes the console switch for me on R5xx. 14197 14198 There are two aspects to it: 14199 14200 - Fix the ordering of avivo_restore() to better match what's 14201 happening in the driver & ATOM, properly locking/unlocking and 14202 only enabling the CRTCs after everything has been properly 14203 programmed. 14204 14205 - Don't ASIC_INIT if the card has any CRTC enabled. This is the 14206 best I came up with for avoiding spurrious ASIC_INIT on cards that 14207 -are- POSTed but don't have the BIOS coming from c0000 on x86. The 14208 problem with spurrious ASIC_INIT is that we do it before we do 14209 RADEONSave(), so that screws up the console switch. 14210 14211 Note that I think we also should save/restore the palette, I don't think 14212 we do. right now, it's a minor issue for me because I fixed offb to be 14213 able to set it on AVIVO's but it might still have to be done in the long 14214 run. 14215 14216 Tested with a VGA analog setup on DACA and a DVI setup on TMDS 0. I 14217 haven't tested any other combo but that should get us going. 14218 14219 Cheers, 14220 Ben. 14221 14222 Signed-off-by: Dave Airlie <airlied@redhat.com> 14223 14224commit e8c07270e1fc2ee455320370692aaf90abac2773 14225Author: Alex Deucher <alexdeucher@gmail.com> 14226Date: Tue Jul 15 17:50:32 2008 -0400 14227 14228 ATOM: fix for r4xx cards with no VRAM_Usage tables 14229 14230 fixes bug 15954 14231 14232commit da41e71efd41907d9347a902720bce60b32550e5 14233Author: Alex Deucher <alexdeucher@gmail.com> 14234Date: Mon Jul 14 13:21:38 2008 -0400 14235 14236 ATOM: Add support for UpdateCRTC_DoubleBufferRegisters cmd table 14237 14238 locks/unlocks the crtc/grph/mode regs before updating the crtc 14239 14240commit a5e0cf13dc7ace6cf0e44e18b73b9a9266e740ab 14241Author: Julien Cristau <jcristau@debian.org> 14242Date: Sun Jul 6 12:17:28 2008 +0200 14243 14244 Link with -lpciaccess and -ldrm if needed 14245 14246 This makes sure the driver ends up with a DT_NEEDED reference to 14247 the libraries it's using. 14248 14249commit f8da849fd89322a54cc4d722767da755358eab70 14250Author: Michel Dänzer <michel@tungstengraphics.com> 14251Date: Sun Jul 13 11:31:03 2008 +0200 14252 14253 Only declare local variable when it's used. 14254 14255commit a6db4dc65aec560f3b1f37ff73ec6b0bd8207295 14256Author: Dave Airlie <airlied@linux.ie> 14257Date: Sat Jul 12 18:47:07 2008 +1000 14258 14259 atombios: fix typo in mode conversion for panel 14260 14261commit 19a0d5879851eff23a3e7d5cdea55bd784051115 14262Author: Dave Airlie <airlied@linux.ie> 14263Date: Sat Jul 12 17:04:25 2008 +1000 14264 14265 atombios: use macro to get offset into table 14266 14267commit a9746114369d516072d841ec356ec3ba3d0be71a 14268Author: Dave Airlie <airlied@linux.ie> 14269Date: Sat Jul 12 10:46:36 2008 +1000 14270 14271 pciid: add radeon HD3850. 14272 14273 pci id from legume on #radeon 14274 14275commit ab14f725676e4d4e45278c64b03fe2d328a3e439 14276Author: Alex Deucher <alexdeucher@gmail.com> 14277Date: Fri Jul 11 19:05:00 2008 -0400 14278 14279 R300: NUM_FPU adjustments for VAP_CNTL 14280 14281commit 810c28cc2660b73e03e4d27be97988fc5321a06f 14282Author: Alex Deucher <alexdeucher@gmail.com> 14283Date: Fri Jul 11 15:31:57 2008 -0400 14284 14285 Make sure cursor offsets are properly aligned when using EXA 14286 14287commit 7e67d0163579a44f104e8f354a929ac9b33e4c21 14288Author: Alex Deucher <alexdeucher@gmail.com> 14289Date: Fri Jul 11 14:30:17 2008 -0400 14290 14291 Fix cursor with multi-head and rotation 14292 14293commit 9086d008fb8c3cde0633b90ce19ffbf4eded388d 14294Author: Alex Deucher <alexdeucher@gmail.com> 14295Date: Tue Jul 8 09:57:04 2008 -0400 14296 14297 Add quirk for Falcon NW laptop 14298 14299commit 61f82ace0210251beb0bcc492218a75a193e1deb 14300Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 14301Date: Tue Jul 8 21:58:43 2008 +1000 14302 14303 atombios: add support for other endians. 14304 14305 This is a cleaned up (in as much as atombios can be..) of benh's patch. 14306 14307 airlied - removed benh's debugging for now, it might need to be put back later.. 14308 14309commit b4d1a47b5c6eafda5e274398eebe2701b030f22e 14310Author: Dave Airlie <airlied@linux.ie> 14311Date: Tue Jul 8 11:09:01 2008 +1000 14312 14313 rv770 initial support 14314 14315 adds pci ids and one register from AMD code 14316 14317commit 7ae4cec8cc8c90aee5dc4fa7abcce22321d4f4eb 14318Author: Roland Scheidegger <sroland@tungstengraphics.com> 14319Date: Mon Jul 7 14:39:47 2008 -0400 14320 14321 clamp tex coords (r100/r200) for textured video 14322 14323 fixes bug 14904 14324 14325commit 879f9a37856642b337e3c0233e38fd5443338eef 14326Author: Alex Deucher <alexdeucher@gmail.com> 14327Date: Fri Jul 4 13:54:49 2008 -0400 14328 14329 Revert "rotate: fix rotation in conf file." 14330 14331 This reverts commit 22c6ca0d5d226c725ba448e5db63fdba9e32332c. 14332 14333 There seems to be an ordering issue in that causes a nasty hang 14334 if the DRI is enabled. This needs more investigation. 14335 14336commit 22c6ca0d5d226c725ba448e5db63fdba9e32332c 14337Author: Dave Airlie <airlied@redhat.com> 14338Date: Thu Jul 3 20:14:59 2008 +1000 14339 14340 rotate: fix rotation in conf file. 14341 14342 need to set modes after init accel architecture 14343 14344 reported-by: Christoph Bartoschek <bartoschek@or.uni-bonn.de> 14345 14346commit ce1a3edd52b647744cfa4433301befb437d211bb 14347Author: Dave Airlie <airlied@redhat.com> 14348Date: Thu Jul 3 20:05:54 2008 +1000 14349 14350 radeon: drop all use of CPMode. 14351 14352 We never test the other codepath and I don't think I've ever recommended it 14353 for anyone. 14354 14355commit c037b4ce8769ad840a257e22b1e4ad58b8ed96fa 14356Author: Wolke Liu <wolke.lui@amd.com> 14357Date: Tue Jul 1 10:45:39 2008 -0400 14358 14359 Add pci id for FireMV 2400 14360 14361commit 72992668ca96717bf69bcea95a5deeaf7c57e826 14362Author: Alex Deucher <alex@botchco.com> 14363Date: Mon Jun 30 14:09:09 2008 -0400 14364 14365 RS4xx: enable the DRI by default on all variants 14366 14367 Some RS4xx chips had the DRI disabled by default. 14368 With the recent drm and ddx changes these chips have been 14369 reported to work now with the DRI. 14370 14371commit 0378c8ed88c829f09b5c6d51314325eaf2685fdb 14372Author: Michel Dänzer <michel@tungstengraphics.com> 14373Date: Sat Jun 28 14:06:45 2008 +0200 14374 14375 Restore versioning of interface for Mesa DRI driver. 14376 14377 This doesn't have anything to do with the package version. 14378 14379commit cfaa23d925e3c062cf87ea844566ac11ea02d69e 14380Author: Alex Deucher <alex@botchco.com> 14381Date: Fri Jun 27 20:36:54 2008 -0400 14382 14383 RADEON: IGP VGA fixes take 2 14384 14385 It seems only RS300/350/400/480 are missing the primary dac 14386 See bug 15708 14387 14388commit 31c27ffcb3c4c5334cf726ecd4e293a678b2a1ea 14389Author: Alex Deucher <alex@botchco.com> 14390Date: Fri Jun 27 20:29:04 2008 -0400 14391 14392 Revert "IGP: attempt to fix VGA on IGP chips" 14393 14394 This reverts commit e78e8a21b4040cd7f1983c241c860d9209398396. 14395 14396commit af79a77e6d8206b18bba97cf4e0770921dc439a7 14397Author: Brice Goglin <Brice.Goglin@ens-lyon.org> 14398Date: Fri Jun 27 17:09:13 2008 +0200 14399 14400 use PACKAGE_VERSION_* for radeon version 14401 14402 now radeon shows verion 6.9.0 instead of 4.3.0. 14403 14404 RADEON_VERSION_MAJOR_TILED is dropped from ddxDriverMajorVersion 14405 since RADEON_VERSION_MAJOR is bigger now. 14406 14407commit c83fbdfa076c107012b7dfbbfbbb2feede00542b 14408Author: Alex Deucher <alex@botchco.com> 14409Date: Thu Jun 26 19:48:45 2008 -0400 14410 14411 Bump for 6.9.0 release 14412 14413commit 32e1d1daf4f68ea37624afaa2bc1ea7742e1163b 14414Author: Brad Smith <brad@comstyle.com> 14415Date: Thu Jun 26 10:22:03 2008 -0400 14416 14417 RADEON: sys/endian.h needs sys/types.h on BSDs 14418 14419 See bug 16512 14420 14421commit dca522355a9039eca6efaba3b36397b246800f94 14422Author: Brad Smith <brad@comstyle.com> 14423Date: Thu Jun 26 10:13:59 2008 -0400 14424 14425 RADEON: fix copy/paste error in accel code 14426 14427commit f4292e110105910d0c19bf2db28e2682b27af2c7 14428Author: Dave Airlie <airlied@redhat.com> 14429Date: Fri Jun 20 15:31:43 2008 +1000 14430 14431 Revert "Revert "ATOM: disable TV-out for now"" 14432 14433 Oops I really didn't mean to do this, I was testing something and it slipped 14434 past. 14435 14436 This reverts commit dd18caa4b2efc430eaae0c4362b65641f9bef440. 14437 14438commit bb1cfcbfbc1ace1f257fc3bf6401501f1a6da827 14439Author: Alex Deucher <alex@botch2.com> 14440Date: Wed Jun 25 15:50:16 2008 -0400 14441 14442 RADEON: adjustments to Jerome's last commit 14443 14444 - Flush caches and wait for idle after drawing 14445 - Make sure 3D is idle too (after composite or textured video) 14446 14447commit 07be302f6dba61dd15918963fae66502230c74b7 14448Merge: 8c9b8de 5245974 14449Author: Jerome Glisse <glisse@freedesktop.org> 14450Date: Wed Jun 25 10:30:45 2008 +0200 14451 14452 Merge branch 'master' of git://anongit.freedesktop.org/xorg/driver/xf86-video-ati 14453 14454commit 8c9b8de0373797cba6f0a27e6b6f461e1070fef9 14455Author: Jerome Glisse <glisse@freedesktop.org> 14456Date: Wed Jun 25 10:28:26 2008 +0200 14457 14458 radeon: flush & wait for 2d & dma idle after 2d blit 14459 14460 This should help to avoid 2d & 3d engine to step on each 14461 other dma transaction. 14462 14463commit 52459745ec05de88adbc087e9566fe6d97ef424b 14464Author: Dave Airlie <airlied@linux.ie> 14465Date: Wed Jun 25 17:09:11 2008 +1000 14466 14467 ati: hopefully fix byteswap mess for those other OSes. 14468 14469commit 67a6ac0001bc9d062aa426384a11a41fa7a1c09a 14470Author: Alex Deucher <alex@botch2.com> 14471Date: Tue Jun 24 21:06:37 2008 -0400 14472 14473 bump for rc release 14474 14475commit bd68507d2d66e03d8bcde5f6e7ea9b2dbfe8b8a0 14476Author: Alex Deucher <alex@botch2.com> 14477Date: Tue Jun 24 20:59:58 2008 -0400 14478 14479 RADEON: warning fix 14480 14481commit 8e534d69c9b19fc085f7c5ca9e18f5ea04f6fc12 14482Author: Alex Deucher <alex@botch2.com> 14483Date: Tue Jun 24 20:08:35 2008 -0400 14484 14485 RADEON: cleanups 14486 14487 - fix some warnings 14488 - RS400 and RS480 are separate families now (update default tmds and dac2 14489 tables) 14490 14491commit e78e8a21b4040cd7f1983c241c860d9209398396 14492Author: Alex Deucher <alex@botch2.com> 14493Date: Tue Jun 24 19:53:28 2008 -0400 14494 14495 IGP: attempt to fix VGA on IGP chips 14496 14497 VGA has never worked on some IGP chips. While the chip only has 14498 one DAC, it appears to use a mix of Primary DAC and TVDAC controls. 14499 14500 See bug 15708 14501 14502commit faa4b4b8dbe9b8a452cfa60d53874350bb04e0cb 14503Author: Jiří Paleček <jpalecek@web.de> 14504Date: Tue Jun 24 19:20:50 2008 -0400 14505 14506 RADEON: remove extraneous line from new pll code 14507 14508commit 72feaa37ea07620f5f2ead438dbc72a1c8883cd3 14509Author: Jiří Paleček <jpalecek@web.de> 14510Date: Mon Jun 23 15:53:58 2008 -0400 14511 14512 RADEON: PLL tweaks 14513 14514 Patch from Jiří Paleček (see debian bug 465864) with some tweaks 14515 by me. 14516 14517 - abort rather than programming bad dividers if no pll dividers can be found 14518 - improve the pll selection algorithm 14519 - in general, prefer lower ref dividers 14520 14521 I've tested this patch on a wide variety of chips (r1xx-r6xx) and clocks. 14522 14523commit 9c2f909ea437a63a408d2398ecabe0b378dbb982 14524Author: Alex Deucher <alex@botch2.com> 14525Date: Mon Jun 23 10:38:15 2008 -0400 14526 14527 RADEON: adjust randr crtc/output prepare/commit ordering 14528 14529 This fixes some occasional mode change problems with multiple heads active. 14530 It seems radeons generally like to turn on the whole output/crtc setup 14531 in one shot. 14532 14533commit aea9bf75cf0774afd3e65fcf3fd3851f5fb21ca3 14534Author: Alex Deucher <alex@botch2.com> 14535Date: Sat Jun 21 10:57:05 2008 -0400 14536 14537 0x1002 0x5657 is actually RV410 14538 14539 See bug 14289 14540 14541 com_bios_scratch.diff 14542 14543commit 38ce8a984f96056b7345bcc06505ba27e0e6d5b4 14544Author: Dave Airlie <airlied@redhat.com> 14545Date: Fri Jun 20 14:16:22 2008 +1000 14546 14547 legacy: use entity MMIO for dpms as this can crossover between zaphod infos 14548 14549commit 1a7d9bc53512b0a5240176c949e6abf1ae2fb1fd 14550Author: Dave Airlie <airlied@redhat.com> 14551Date: Fri Jun 20 14:14:21 2008 +1000 14552 14553 atombios: use MMIO from the entity not the info 14554 14555commit ef624b88903b1a87ef5b6388e18291f75776b93d 14556Author: Alex Deucher <alex@botch2.com> 14557Date: Thu Jun 19 18:20:52 2008 -0400 14558 14559 RADEON: fix read past the end of an array 14560 14561commit cfe814a481d8cf2005d738a0ca9782f1ed4177f5 14562Author: Dave Airlie <airlied@linux.ie> 14563Date: Fri Jun 20 07:51:27 2008 +1000 14564 14565 r600: don't add fb offset here to make shadowfb work. 14566 14567 discovered on irc with wpwrak. 14568 14569commit 5b323a2cbbc412b6dd4e782438b64ee996558d05 14570Merge: dd18caa eed2415 14571Author: Dave Airlie <airlied@linux.ie> 14572Date: Fri Jun 20 07:49:16 2008 +1000 14573 14574 Merge branch 'master' of ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 14575 14576commit eed241553748125e902c44dcc8cf8e0605be071b 14577Author: Alex Deucher <alex@botch2.com> 14578Date: Tue Jun 17 17:42:31 2008 -0400 14579 14580 radeon: X_ERROR -> X_WARNING for num gb pipes 14581 14582 This just means your drm is old, not fatal or anything 14583 14584commit dd18caa4b2efc430eaae0c4362b65641f9bef440 14585Author: Dave Airlie <airlied@linux.ie> 14586Date: Sat Jun 14 18:37:56 2008 +1000 14587 14588 Revert "ATOM: disable TV-out for now" 14589 14590 This reverts commit effa245914823371e052cd9aa1143a02350891e7. 14591 14592commit b416e97a1f16ef700ba1aaca317dee82b7a3cf64 14593Author: Alex Deucher <alex@botch2.com> 14594Date: Fri Jun 13 02:21:31 2008 -0400 14595 14596 RADEON: man page updates 14597 14598commit 88ce87d79069662de28ecb3c10c140919927d584 14599Author: Alex Deucher <alex@botchco.com> 14600Date: Thu Jun 12 15:10:47 2008 -0400 14601 14602 Bump for 6.9.0rc1 release 14603 14604commit ada41723fdc8cbeeda12bb4ae2d16e8d1ba215de 14605Author: Alex Deucher <alex@botch2.com> 14606Date: Wed Jun 11 14:29:36 2008 -0400 14607 14608 R3xx: use DSTCACHE_CTLSTAT rather than RB2D_DSTCACHE_CTLSTAT 14609 14610 According to the hw guys, you should use DSTCACHE_CTLSTAT to 14611 flush the 2D dst cache rather than RB2D_DSTCACHE_CTLSTAT. 14612 14613commit 3d70c398afa0831387af81559453b77c218a54f9 14614Author: Dave Airlie <airlied@redhat.com> 14615Date: Thu Jun 12 06:18:26 2008 +1000 14616 14617 atom: fix another endian define 14618 14619commit f1794e4c34d801371461afed37e7bb7d1fc79305 14620Author: Dave Airlie <airlied@redhat.com> 14621Date: Thu Jun 12 06:10:30 2008 +1000 14622 14623 atom: fixup includes in cd operations to get correct endian bits in atombios.h 14624 14625commit f29976d12cc2020cd80c90fa3fd4b25c031559ba 14626Author: Dave Airlie <airlied@ppcg5.localdomain> 14627Date: Wed Jun 11 19:33:02 2008 +1000 14628 14629 radeon: add lots of endian bits into the atombios parser 14630 14631commit 09eb220971b5d2bfd7d1ff6f552060967a133152 14632Author: Dave Airlie <airlied@ppcg5.localdomain> 14633Date: Wed Jun 11 19:21:22 2008 +1000 14634 14635 atombios: fixup atombios bitfields for endianness 14636 14637 reverse the bits for big endian 14638 14639commit b80a3e85bf9dbfc8d96a745876d75d38bea97c1a 14640Author: Alex Deucher <alex@botchco.com> 14641Date: Mon Jun 9 20:29:47 2008 -0400 14642 14643 R5xx: fix RADEONSetAgpBase() for R5xx chips 14644 14645 Is there any reason to still do this in the ddx? 14646 Maybe real old drms? 14647 14648commit ab7936708dfbee4d3676262f8010e001b73a4a38 14649Author: Alex Deucher <alex@botchco.com> 14650Date: Mon Jun 9 17:05:04 2008 -0400 14651 14652 RS4xx: RC410 chips are actually RS400 based 14653 14654commit a54ca3c699c103e2e3df87101dca929a3eee4377 14655Author: Dave Airlie <airlied@linux.ie> 14656Date: Thu Jun 5 18:31:53 2008 +1000 14657 14658 radeon: fix typo in memory calcs 14659 14660 noticed by ghepeu on irc. 14661 14662commit 6e4e6d2a8f29f92efc219dca24ea31d1f37d5a0f 14663Author: Alex Deucher <alex@botchco.com> 14664Date: Tue Jun 3 20:32:20 2008 -0400 14665 14666 RADEON: minor cleanups 14667 14668commit 9d3afbf5fa4110928a9f965df2733c79db92ea99 14669Author: Michel Dänzer <michel@tungstengraphics.com> 14670Date: Tue Jun 3 11:40:49 2008 +0200 14671 14672 Call DRM_IOCTL_MODESET_CTL ioctl to avoid problems with DRM post vblank-rework. 14673 14674commit effa245914823371e052cd9aa1143a02350891e7 14675Author: Alex Deucher <alex@botchco.com> 14676Date: Sun Jun 1 20:47:50 2008 -0400 14677 14678 ATOM: disable TV-out for now 14679 14680 It only sort of works in NTSC mode and isn't handled correctly 14681 in all cases yet. Until we sort out the remaining details leave it 14682 disabled. 14683 See bugs: 16186, 16178, 16185 14684 14685commit 8504c6b0e40477ee544ad7f5366d569bdc53d6f0 14686Author: Alex Deucher <alex@botch2.com> 14687Date: Fri May 30 11:19:03 2008 -0400 14688 14689 RADEON: update RADEONGetVRAMType() for newer chips 14690 14691commit 7cb695a329a9b543ad61af08c2d7d6eaf56de35e 14692Author: Alex Deucher <alex@botch2.com> 14693Date: Fri May 30 10:10:56 2008 -0400 14694 14695 RADEON: add Int10 option 14696 14697 On some radeons you can't read the bios without initializing int10. 14698 On ATOM-based secondary cards, intitializing int10 tends to hang the card. 14699 14700commit 714b2c63ec1c4a1410a6d521c03e9d1f90937c01 14701Author: Alex Deucher <alex@botch2.com> 14702Date: Thu May 29 02:31:00 2008 -0400 14703 14704 RADEON: don't enable legacy bios table based init yet 14705 14706 Needs more work and we need to figure out how best to decide 14707 when to use it. 14708 14709commit e20b08525a64888287ec4a369d8f7dbde95c655d 14710Author: Alex Deucher <alex@botch2.com> 14711Date: Wed May 28 14:43:47 2008 -0400 14712 14713 ATOM: fail of we are not able to get a bios image 14714 14715commit d4fa17b426b0cafac3184b2ea9b675e2ff154ee9 14716Author: Alex Deucher <alex@botch2.com> 14717Date: Wed May 28 09:50:02 2008 -0400 14718 14719 R1xx: Gamma fixes for overlay 14720 14721commit c7e6a50f60a04bbbf1c1ddbf4738a5f75c94f950 14722Author: Alex Deucher <alex@botchco.com> 14723Date: Wed May 28 09:05:19 2008 -0400 14724 14725 RADEON: only attempt to read vbios from legacy ISA space if it's primary 14726 14727commit faea008806802ec0e045754ec1eca492ebae320e 14728Author: Alex Deucher <alex@botch2.com> 14729Date: Tue May 27 18:36:01 2008 -0400 14730 14731 R3/4/5xx: use get_param to get the num_gb_pipes from the drm 14732 14733commit 965a5dbcd9dc4bf1cdd7f2bbdec15e9733b2e090 14734Author: Alex Deucher <alex@botch2.com> 14735Date: Tue May 27 16:48:41 2008 -0400 14736 14737 RADEON: improve support for secondary cards 14738 14739 this should fix bugs 16115, 16035 14740 14741commit 5f951a5573f0c7572230c9aa4d3f75d67f91ed71 14742Author: Alex Deucher <alex@botch2.com> 14743Date: Wed May 21 00:35:14 2008 -0400 14744 14745 ATOM: remove duplicate code 14746 14747commit edce33e87fb79a540d8c285f205d4c3f2c2bc9f4 14748Author: Dave Airlie <airlied@redhat.com> 14749Date: Wed May 28 06:43:40 2008 +1000 14750 14751 radeon: fix surface access on avivo chips. 14752 14753 This should fix VT switch on vesafb 14754 14755commit 307bd65a25ee425d8359dd30572b002ce2338f91 14756Author: Michael Babej <mbabej@redhat.com> 14757Date: Tue May 27 19:13:49 2008 +1000 14758 14759 r600: fix vt switch issue 14760 14761 RH BZ 441492 14762 14763commit bf48c9eb8ec592515be9d1732d60283af715674d 14764Author: Dave Airlie <airlied@linux.ie> 14765Date: Sat May 24 11:51:20 2008 +1000 14766 14767 atombios: for LVDS set use ref div as per legacy 14768 14769commit 04500c8419b1aeaeac7968492b82e1d7cad1d05b 14770Author: Alan Coopersmith <alan.coopersmith@sun.com> 14771Date: Tue May 20 20:34:31 2008 -0700 14772 14773 Strip ^M's from src/AtomBios/includes/ObjectID.h 14774 14775commit 521a0488d1c414209c3534dc242000faa332a441 14776Author: Alan Coopersmith <alan.coopersmith@sun.com> 14777Date: Tue May 20 20:33:37 2008 -0700 14778 14779 On non-gcc compilers, use C99's __func__ instead of gcc's __FUNCTION__ 14780 14781commit b7c80d0c86646105d2bce5d4d59ba6c45aa7cafc 14782Author: Tobias Diedrich <ranma+freedesktop@tdiedrich.de> 14783Date: Tue May 20 19:16:54 2008 -0400 14784 14785 R200/R300: fix gamma setup for overlay 14786 14787 More pending. See bug 16001 14788 14789commit 69423482e2e94637142a9ba675589a1449a346a8 14790Author: Alex Deucher <alex@botch2.com> 14791Date: Tue May 20 18:57:13 2008 -0400 14792 14793 RADEON: cleanup connection detection and remove duplicate code 14794 14795commit 55e4469f59c82bb5762673de5f3f27d18b0bd9a3 14796Author: Alex Deucher <alex@botch2.com> 14797Date: Tue May 20 17:46:58 2008 -0400 14798 14799 RADEON: enable cloning on multi-crtc cards 14800 14801commit 12f185634071980041aaac6265d89708b181b215 14802Author: Alex Deucher <alex@botch2.com> 14803Date: Tue May 20 16:05:41 2008 -0400 14804 14805 ATOM: Ignore invalid connector entries 14806 14807commit 19e97f74e39fc2b35727708ac429de33d0b70162 14808Author: Egbert Eich <eich@freedesktop.org> 14809Date: Tue May 20 11:16:15 2008 +0200 14810 14811 Change RMX code to follow the programming algorithm suggested by ATI. 14812 14813 ATI provides the following algorithm to calculate the RMX scaling ratios 14814 in its programming specs: 14815 when RMX_AUTO_RATIO_HORZ_INC set to 1,Horizontal auto ratio result trucated, 14816 and then incremented by 1. 14817 Horz_Ratio = ( ((Active display width in characters (including overscan) + 1) 14818 / (Panel width in characters)) x 4096 + 1 ) 14819 else 14820 Horz_Ratio = ( ((Active display width in characters (including overscan)) 14821 / (Panel width in characters)) x 4096 + 1 ) 14822 14823 when RMX_AUTO_RATIO_VERT_INC set to 1, Vertical auto ratio result trucated, 14824 and then incremented by 1. 14825 Vert_Ration = ( ((Active display width in characters (including overscan) + 1) 14826 / (Panel width in characters)) x 4096 + 1) 14827 else 14828 Vert_Ration = ( ((Active display width in characters (including overscan)) 14829 / (Panel width in characters)) x 4096 + 1) 14830 14831 This patch implements this behavor. Additionally it avoids the use of floats. 14832 14833commit a4f3d0088ba763ed8eab1d331959b5ecde8262e8 14834Author: Alex Deucher <alex@botch2.com> 14835Date: Tue May 20 11:42:53 2008 -0400 14836 14837 R3/4/5xx: fixup logic from last commit 14838 14839 Spotted by otaylor and MrCooper 14840 14841commit 130e55738047f2a073bcc47e3e1400f7b694a81b 14842Author: Alex Deucher <alex@botch2.com> 14843Date: Tue May 20 11:32:42 2008 -0400 14844 14845 R3/4/5xx: fix EXA rotation 14846 14847 xrandr uses PictOpSrc for rotation which we were falling back on since 14848 render semanties require alpha=0 for REPEAT_NONE when there is no alpha 14849 channel and there is a transform. If the dst has no alpha channel we 14850 should be ok I think. 14851 14852 Otayler and I discussed this on IRC. the more general fix would be 14853 to clip the dst to the src and mask for bounded ops like in/add/over 14854 in the pixel exact transform case. 14855 14856commit cc9f510770700228e5d597c872e926a4e99bd950 14857Author: Corbin Simpson <MostAwesomeDude@gmail.com> 14858Date: Wed May 14 16:49:01 2008 -0400 14859 14860 RADEON: missing stdint.h includes 14861 14862commit 71fa57f871dba03260dba2180ce1dab44048ac1a 14863Author: Alex Deucher <alex@cube.(none)> 14864Date: Wed May 14 13:36:38 2008 -0400 14865 14866 Add RS600 support 14867 14868commit 582c1a1b2c7b1032e9f9f54ca36100c57f580c5c 14869Author: Alex Deucher <alex@cube.(none)> 14870Date: Tue May 13 20:50:25 2008 -0400 14871 14872 RS4xx: Split out RS400 and RS480 as separate families 14873 14874 RS400 (intel based IGP) and RS480 (AMD based IGP) have different 14875 MC setups and need to be handled differently 14876 14877commit 708e7c98f636738fbcc47a597bc94b309a4dc1c4 14878Author: Michel Dänzer <michel@tungstengraphics.com> 14879Date: Tue May 13 11:00:38 2008 +0200 14880 14881 Reinstate CARD* types that are part of external interfaces. 14882 14883 The compiler pointed this out... 14884 14885commit 908b7b940e3ef296836bc94680ebb8ab67650e31 14886Author: Matt Turner <mattst88@gmail.com> 14887Date: Mon May 12 12:06:33 2008 -0400 14888 14889 Replace CARD{8,16,32} with uint{8,16,32}_t 14890 14891 As has been done with xf86-video-intel, replace all CARD* datatypes with 14892 uint*_t datatypes available from stdint.h. 14893 14894commit 87e66ce76430890ab4939ffcd42f72b9288eb598 14895Author: Avi Rozen <avi.rozen@gmail.com> 14896Date: Mon May 12 08:33:38 2008 -0400 14897 14898 RADEON: fix lockup on start 14899 14900 see debian bug 480312 14901 14902commit 94bf8f01bd43cb103fffecfe04d04a214f892baf 14903Author: Dave Airlie <airlied@redhat.com> 14904Date: Mon May 12 20:02:51 2008 +1000 14905 14906 radeon: fix set_cursor_colours remove fb offset 14907 14908commit 1d0f1d31e2ed1d91ee87cb3e02ce48c8c07aa418 14909Author: Dave Airlie <airlied@redhat.com> 14910Date: Mon May 12 15:44:35 2008 +1000 14911 14912 radeon: rs485 vs rs485m.. mobile? non-mobile? bi? 14913 14914 So it appears we have in the wild two chipsets with the same PCI ID (0x1002:0x5974) 14915 that are mobile and non-mobile. 14916 14917 the RH bug references is a desktop system. 14918 The Dell Vostro 1100 also has this chipset with LVDS bits. 14919 14920commit 25e0c3945a51ae8c14b3a847ec75a256e1397f24 14921Author: Alex Deucher <alexdeucher@gmail.com> 14922Date: Mon May 12 10:28:33 2008 +1000 14923 14924 radeon: add initial support for cloning outputs. (single-CRTC cards only) 14925 14926 airlied - This code was originally written by Alex, and I've modified it to 14927 only run on single-CRTC cards for now until we can test it some more. 14928 14929commit 94405eb1c9e4e0ababc6aef03b753d6ed9eb7838 14930Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 14931Date: Fri May 9 05:28:44 2008 -0400 14932 14933 RADEON: fixup ifdef from last commit 14934 14935commit 18e20bc22a55ecfee9798c01079d7b24d19f0051 14936Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 14937Date: Thu May 8 08:45:51 2008 -0400 14938 14939 RADEON: disable MMIO paths for EXA composite/texvid on IGP/R5xx 14940 14941 The MMIO paths eventually lead to a hang on r5xx/IGP. I haven't 14942 been able to find out why yet. 14943 14944commit c3532268875fd24e6519bea2fb1b814d612bbdb4 14945Author: Dave Airlie <airlied@linux.ie> 14946Date: Wed May 7 02:37:18 2008 +1000 14947 14948 radeon: fix zaphod EXA with texture video 14949 14950commit ffc437f3606ab8ceba1ff152e4bb08988a58b54c 14951Author: Dave Airlie <airlied@linux.ie> 14952Date: Wed May 7 02:30:28 2008 +1000 14953 14954 avivo: fix zaphod cursor in theory 14955 14956commit e36ef14e3a1087e1fe41baa26ade2937f396001f 14957Author: Dave Airlie <airlied@linux.ie> 14958Date: Wed May 7 01:39:28 2008 +1000 14959 14960 radeon: fix textured-xv on zaphod 14961 14962commit 8fc19bee27c0f151d2ab3354f6ac0992b358436d 14963Author: Dave Airlie <airlied@linux.ie> 14964Date: Wed May 7 01:38:24 2008 +1000 14965 14966 radeon: zaphod: fix render accel for EXA 14967 14968commit fc41b9042a5220a8419cc7b69ca3850cae6b903c 14969Author: Dave Airlie <airlied@linux.ie> 14970Date: Wed May 7 01:32:23 2008 +1000 14971 14972 radeon: fix EXA pixmap offset on zaphod 14973 14974commit 4568cb820d567c6909a4be956d7e79b91232535e 14975Author: Dave Airlie <airlied@linux.ie> 14976Date: Wed May 7 01:19:39 2008 +1000 14977 14978 radeon: zaphod fixes for pciaccess not allowing multiple MMIO maps 14979 14980commit ca81fa086b21633a7fd926833fb6d1d4fa080646 14981Author: Dave Airlie <airlied@linux.ie> 14982Date: Wed May 7 01:12:01 2008 +1000 14983 14984 radeon: zaphod fix for cursor on second head 14985 14986 We don't need to add fbOffset here as the mmap we have of the framebuffer 14987 starts half way. 14988 14989commit 24b60c8965f6a0b3f0c2bb1e7236b4d6642c5918 14990Author: Julien Cristau <jcristau@debian.org> 14991Date: Fri May 2 15:30:45 2008 -0400 14992 14993 Add a test for __GLIBC__ to CD_Common_Types.h. 14994 14995 Atombios redefines the standard types but the definitions conflict 14996 with the ones from glibc (causes build failures on GNU/Hurd 14997 and GNU/kFreeBSD). 14998 14999commit f051359ac09c6b9416e39b9ca7d9dc0880aa1557 15000Author: thegraze <thegraze@googlemail.com> 15001Date: Fri May 2 14:02:16 2008 -0400 15002 15003 ATOM: add support for DragonFlyBSD 15004 15005commit 3d469cbc3225d890a895dac7cbc1ab7e08054b48 15006Author: Alex Deucher <alex@cube.(none)> 15007Date: Wed Apr 30 18:33:04 2008 -0400 15008 15009 RADEON: lock the cursors when updating 15010 15011 this should fix occasional corruption seen when updating 15012 the cursor. 15013 15014commit 445b71021843665ba32f37b2ce5c9d2857c07cc7 15015Author: Alex Deucher <alex@cube.(none)> 15016Date: Tue Apr 29 21:01:41 2008 -0400 15017 15018 RADEON: assorted fixes 15019 15020 - free rotate pixmaps on VT switch 15021 - save crtc/output status so we only turn on 15022 crtcs/outputs if they are off 15023 - show/hide cursors when changing modes 15024 15025commit 070cce5255a5c311f9d8b85ec54bd56655014933 15026Author: Stephan Wolf <stephan@letzte-bankreihe.de> 15027Date: Mon Apr 28 11:26:37 2008 -0400 15028 15029 R3xx+: further fix for IGP chips 15030 15031 see bug 15538 15032 15033commit 211e0041c7fc2df494b77428553943a2b526ee4e 15034Author: Alex Deucher <alex@cube.(none)> 15035Date: Sun Apr 27 21:08:00 2008 -0400 15036 15037 IGP: fix EXA composite corruption 15038 15039commit 656b06bdde129ca4fc370f5a2cf7311c9179b0ff 15040Author: Alex Deucher <alex@botch2.com> 15041Date: Sun Apr 27 20:20:49 2008 -0400 15042 15043 RADEON: remove duplicate register define 15044 15045 Also add more bit defs to wait_until register 15046 15047commit 8a9820a3aa49bc667f90ac291a27e4d7b4ae38b3 15048Author: Alex Deucher <alex@botch2.com> 15049Date: Sun Apr 27 19:02:22 2008 -0400 15050 15051 RADEON: decrease crtc/output verbosity 15052 15053commit c5d62fa0e8f52c3264ff9db3ff10cdf5a806bfc0 15054Author: Owen Taylor <otaylor@huygens.home.fishsoup.net> 15055Date: Thu Apr 17 13:14:53 2008 +0200 15056 15057 Emulate repeats by drawing in tiles 15058 15059 When we can't turn on hardware repeats, because the texture 15060 is non-power-of-two, or has padding at the ends of lines, 15061 try to draw the image in multiple tiles rather than falling 15062 back to software. (We can only do this when there is no 15063 transform.) 15064 15065commit eeb7b74bb6c813b0e3afa4b704f6ffb0d0aab92b 15066Author: Owen Taylor <otaylor@huygens.home.fishsoup.net> 15067Date: Thu Apr 17 13:14:25 2008 +0200 15068 15069 Turn on wrapping when repeating on R100 + R200 15070 15071 Actually enable repeats for R100 and R200. This corresponds 15072 to a R300 change made in the patch in: 15073 http://bugs.freedesktop.org/show_bug.cgi?id=15333 15074 15075commit e511f39dfef503006cf249b9f6934091eaade9b5 15076Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 15077Date: Thu Apr 17 05:04:34 2008 -0400 15078 15079 R300+: move more common code into init3d() 15080 15081 - pre-load r3xx tex instructions 15082 - setup RS instructions in init3d() 15083 15084commit 99435b7c18d931ea620044d0fdb4cc93dfcc6331 15085Author: Owen Taylor <otaylor@redhat.com> 15086Date: Thu Apr 17 02:46:11 2008 -0400 15087 15088 Radeon: Omit mask coordinates 15089 15090 Adapted from Owen's patch on bug 15546 15091 This fixes the slowness with aatext on r300 15092 and may speed up other chips marginally. 15093 15094commit 37614e1db9a595fbe8a21d7a045895e11d272db9 15095Author: Alex Deucher <alex@botch2.com> 15096Date: Tue Apr 15 09:48:16 2008 -0400 15097 15098 fix up some things from the last commit 15099 15100commit 1286fe5ce1c77453d57817b9b26b1bdb32ca7bc8 15101Author: Alex Deucher <alex@botch2.(none)> 15102Date: Mon Apr 14 20:02:14 2008 -0400 15103 15104 R300+: properly setup vap_cntl 15105 15106 this fixes tcl/pvs on RV515 among other things 15107 15108commit f72a4b805db26f10f69330b88459cbeae661189b 15109Author: Alex Deucher <alex@botch2.com> 15110Date: Mon Apr 14 14:10:40 2008 -0400 15111 15112 EXA: Don't wait for 3D idle after each Composite() 15113 15114 wait in CompositeDone() instead 15115 15116commit 4cd4acf1092aeb696b086a382a033aee471d2de9 15117Author: Alex Deucher <alex@botch2.com> 15118Date: Mon Apr 14 11:50:59 2008 -0400 15119 15120 R300: move more common code to init3d() 15121 15122commit 3c523c9a07402e17dff588fad842224c57e98223 15123Author: Alex Deucher <alex@botch2.com> 15124Date: Mon Apr 14 11:21:42 2008 -0400 15125 15126 R3xx+: 3D engine documentation and minor cleanups 15127 15128 - document the R300 exa/textured video code 15129 - minor cleanups of textured video code to clarify meaning 15130 15131commit ce025bbb2496d4de94b8d4ac450c64441b64ee04 15132Author: Alex Deucher <alex@botch2.com> 15133Date: Sat Apr 12 21:22:03 2008 -0400 15134 15135 R300+: consolidate some tcl/non-tcl paths 15136 15137 - Move more code to init3d() 15138 - MMIO textured video seems more reliable now on newer chips 15139 15140commit 11b54a319c7c9dd52e3fb13372697059dafe1cd3 15141Author: Alex Deucher <alex@botch2.com> 15142Date: Sat Apr 12 16:50:22 2008 -0400 15143 15144 R3xx+: fix XAA + textured video on non-TCL path 15145 15146commit dd15a2f5906725116b8cd9954243099055e88e37 15147Author: Alex Deucher <alex@botch2.com> 15148Date: Sat Apr 12 16:49:03 2008 -0400 15149 15150 R3xx+: more fixes to 2D/3D engine init 15151 15152commit f3e68d4b7afd2e23675bf6361c496814c9cb4b94 15153Author: Alex Deucher <alex@botch2.com> 15154Date: Fri Apr 11 10:59:07 2008 -0400 15155 15156 Fix exa glyph corruption on newer chips 15157 15158commit b59686d6427cbf8b35e36b020cbbc6a0c5149b22 15159Author: Alex Deucher <alex@botch2.com> 15160Date: Fri Apr 11 10:15:25 2008 -0400 15161 15162 R300+: pre-load vertex programs in init3D() 15163 15164commit acc5833a35ad6c29a57f659607afb27eebdc2ea5 15165Author: Alex Deucher <alex@botch2.com> 15166Date: Thu Apr 10 17:52:52 2008 -0400 15167 15168 R3xx+: consolidate more tcl code 15169 15170commit 6f8f75bd19ef1919c0291141675be2d0e29b3251 15171Author: Alex Deucher <alex@botch2.com> 15172Date: Thu Apr 10 17:08:50 2008 -0400 15173 15174 R3xx+: consolidate some common 3D code 15175 15176commit 4b9234e1c4f7c7f419cb4245d64f3f9756c98bb6 15177Author: Alex Deucher <alex@botch2.com> 15178Date: Thu Apr 10 16:58:22 2008 -0400 15179 15180 R3xx+: tcl wip 15181 15182commit 865c463e3afb4759758f569132be8bf1386da5cc 15183Author: Alex Deucher <alex@botch2.com> 15184Date: Thu Apr 10 16:51:04 2008 -0400 15185 15186 R300+: textured video tcl cleanup 15187 15188commit 79c8d4ca36a1c3e5fe759d4ccc379c36af8f1676 15189Author: Alex Deucher <alex@botch2.com> 15190Date: Thu Apr 10 16:28:18 2008 -0400 15191 15192 RADEON: cleanup 15193 15194commit c4821a287d29a65f3bcb7d60dc72ec13c0384008 15195Author: Alex Deucher <alex@botch2.com> 15196Date: Thu Apr 10 16:20:17 2008 -0400 15197 15198 Revert "R3xx/R5xx: move more VAP, etc. state setup into common init3d() function" 15199 15200 This reverts commit 305a3310963a5dd07b3495015b06aa8c7c4e6b02. 15201 15202 Conflicts: 15203 15204 src/radeon_commonfuncs.c 15205 src/radeon_exa_render.c 15206 src/radeon_textured_videofuncs.c 15207 15208commit 0032c80bf30bab189204e3e6929e18a19d753138 15209Author: Alex Deucher <alex@botch2.com> 15210Date: Thu Apr 10 14:35:00 2008 -0400 15211 15212 RADEON: store tcl status in driver rec 15213 15214commit 9e2ffe66d106abe34a670d2edc9905ed62c485e8 15215Author: Alex Deucher <alex@botch2.com> 15216Date: Thu Apr 10 14:24:04 2008 -0400 15217 15218 R3xx+: use the right register for engine flush 15219 15220commit e1a9f26c2d2cbca9ad159e723ec95b95be1ef349 15221Author: Alex Deucher <alex@botch2.com> 15222Date: Thu Apr 10 14:12:15 2008 -0400 15223 15224 R3xx+: minor textured video fixes 15225 15226 - set shader output swizzling correctly 15227 - flush the right cache register on r3xx+ 15228 15229commit d79040906cd25bd494feb5901f465bbd050aa923 15230Author: Alex Deucher <alex@botch2.com> 15231Date: Thu Apr 10 13:59:58 2008 -0400 15232 15233 R3xx+: EXA/textured video fixes 15234 15235 - get pipe config based on GB_PIPE_SELECT where applicable 15236 (adapted from a similar patch from Dave) 15237 - only flush the dst cache after submitting vertices, freeing 15238 the cache lines stalls the pipe 15239 - no need to wait for 3D idle after submitting vertices 15240 - fix PURGE_CACHE() and PURGE_ZCACHE() for r3xx+ 15241 - fix depth 16 with EXA composite 15242 15243commit 0a96173cc38e506728d4c3f2dd383ba56e856578 15244Author: Michel Dänzer <michel@tungstengraphics.com> 15245Date: Mon Apr 7 18:15:34 2008 +0200 15246 15247 Increase default CP timeout. 15248 15249 Helps avoid spurious timeouts causing problems, see 15250 http://bugs.freedesktop.org/show_bug.cgi?id=15203 . 15251 15252commit 255fbf465f5e7db2609a5a151bfa810249db52a0 15253Author: Owen W. Taylor <otaylor@fishsoup.net> 15254Date: Thu Apr 3 02:25:41 2008 -0400 15255 15256 Fix rendering of transformed sources for REPEAT_NONE with EXA on >= R300. 15257 15258 Use the border color when possible, otherwise fall back to software. 15259 15260commit bc0407e53237d7968808110bc0243076377acf6e 15261Author: Alex Deucher <alex@cube.(none)> 15262Date: Fri Apr 4 18:40:16 2008 -0400 15263 15264 ATOMBIOS: Add support for DynamicClocks option 15265 15266 This patch adds support for dynamic clock gating and static 15267 power management using the atom command tables. In some cases 15268 the bios may already set this up during post, so YMMV. 15269 15270 I was only able to test on desktop cards, so I haven't tested 15271 to see how much (if any) power this saves or how it affects the 15272 thermal footprint. 15273 15274commit 5f5e21bb50555c56bd371576074c28c929307ff1 15275Author: Alex Deucher <alex@cube.(none)> 15276Date: Fri Apr 4 14:29:45 2008 -0400 15277 15278 RADEON: warning fixes 15279 15280commit c8e9a973aaded24aad567a0e36d0c78a05d6b2fd 15281Author: Alex Deucher <alex@cube.(none)> 15282Date: Fri Apr 4 14:26:19 2008 -0400 15283 15284 RADEON: add some quirks 15285 15286commit 091963a635b79884afe77c026eabb48972fbe175 15287Author: Alex Deucher <alex@botch2.com> 15288Date: Thu Apr 3 22:35:16 2008 -0400 15289 15290 Minor cleanup 15291 15292commit 950e9860643c20acde0eca4e4ff26baacc1f2b69 15293Author: Alex Deucher <alex@botch2.com> 15294Date: Thu Apr 3 22:11:48 2008 -0400 15295 15296 Revert "RADEON: memmap rework 1" 15297 15298 This reverts commit dd8ee1b444f4b973a1e0fadca5f943f2162b5e94. 15299 15300 Conflicts: 15301 15302 src/radeon.h 15303 src/radeon_driver.c 15304 15305 This rework seems to have caused more trouble than it was worth. 15306 15307commit 88a1fe4a94c5d11aff22734b21c89890e4428cd5 15308Author: Alex Deucher <alex@botch2.com> 15309Date: Thu Apr 3 22:04:43 2008 -0400 15310 15311 Revert "RADEON: remove driver rec copies of mc info, use save rec directly" 15312 15313 This reverts commit be0858a84fbdf74c0b844f462933a221d48c707d. 15314 15315 Conflicts: 15316 15317 src/radeon_driver.c 15318 15319commit c40a7aa3989576a8144213e2f31b892d21df8686 15320Author: Owen W. Taylor <otaylor@fishsoup.net> 15321Date: Thu Apr 3 14:43:55 2008 -0400 15322 15323 R3xx/R5xx: Fix pitch and clamp mode for repeating textures 15324 15325 - We can always use TXPITCH on a R300 even when repeating, 15326 (previous check for pitch matching width was also wrong) 15327 - Fix clamp mode for repeating textures to be WRAP 15328 15329commit a8593482c1f2e0f2dbac06c2e5325ba8c83ed9ff 15330Author: Dave Airlie <airlied@redhat.com> 15331Date: Wed Apr 2 09:58:05 2008 +1000 15332 15333 atombios: fix the dual-head hopefully. 15334 15335 tested on r600 with DVI and VGA 15336 15337commit 61d883d116fab3e9b513432d65e705afc5bb39f1 15338Author: Dave Airlie <airlied@redhat.com> 15339Date: Wed Apr 2 09:57:38 2008 +1000 15340 15341 Revert "Revert "atombios: fixup the width/height to use the mode values not the scrn ones"" 15342 15343 This reverts commit fc9af578997b6f22ee8b17e83f37d98689291b0e. 15344 15345 I see your revert and raise you one... 15346 15347commit fc9af578997b6f22ee8b17e83f37d98689291b0e 15348Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 15349Date: Tue Apr 1 09:25:45 2008 -0400 15350 15351 Revert "atombios: fixup the width/height to use the mode values not the scrn ones" 15352 15353 This reverts commit c2b1c8b706a6c7c1fd0af80091958473133d54e7. 15354 15355 These registers hold surface size. Using the mode values 15356 breaks dualhead. 15357 15358commit 959509dd54de053f526b534e379a46934127231f 15359Author: Dave Airlie <airlied@redhat.com> 15360Date: Mon Mar 31 14:29:44 2008 +1000 15361 15362 radeon: use correct DDC interfaces so quirks get applied 15363 15364 Radeon seemed to mess up applying certain quirks, hopefully this will fix it. 15365 15366commit 18f5f1cd2f52afed89fc11ade0920f3dfea87306 15367Author: Dave Airlie <airlied@redhat.com> 15368Date: Mon Mar 31 14:11:49 2008 +1000 15369 15370 radeon: split quirks into separate function and new quirk for IBM RN50 15371 15372 Add a connector table quirk for the IBM RN50. 15373 15374commit c2b1c8b706a6c7c1fd0af80091958473133d54e7 15375Author: Dave Airlie <airlied@redhat.com> 15376Date: Sun Mar 30 11:44:14 2008 +1000 15377 15378 atombios: fixup the width/height to use the mode values not the scrn ones 15379 15380 this fixes it properly, legacy appears to be okay. 15381 15382commit c5edea3d8c9254d3a21e390b8309e39e4c9635db 15383Author: Dave Airlie <airlied@redhat.com> 15384Date: Sun Mar 30 11:11:22 2008 +1000 15385 15386 r500/r600: fix rotation to fill screen 15387 15388 I'm not 100% sure this is the correct fix (maybe we shouldn't be using scrn 15389 virtualX/Y)... this will fix it for now until I get more time. 15390 15391commit 9c62c820ba45ebc14d5f36f5d7885863800b6adb 15392Author: Michel Dänzer <michel@tungstengraphics.com> 15393Date: Fri Mar 28 12:37:29 2008 +0100 15394 15395 Include config.h, so FGL_LINUX can actually be defined when it's tested... 15396 15397commit a00d9260a85b94a522c442aee24bc5ea4dc31c5c 15398Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 15399Date: Thu Mar 27 20:03:13 2008 -0400 15400 15401 RADEON: fix lid issues on AVIVO chips for real this time :) 15402 15403commit f0e89c09074b2c7e641f73692bb39b0bf68eb49c 15404Author: Alex Deucher <alex@botch2.com> 15405Date: Thu Mar 27 19:15:18 2008 -0400 15406 15407 Revert "RADEON: attempt to fix lid issues" 15408 15409 This reverts commit 9b4473c1d830b88866dd22e8174a07195bd6fcf4. 15410 This doesn't help. 15411 15412commit 1442d396b938049b83f009a78ddabe2bf85641b6 15413Author: Dave Airlie <airlied@redhat.com> 15414Date: Thu Mar 27 14:02:51 2008 +1000 15415 15416 radeon: size bios to max of bar vs 64k. 15417 15418 reported by dwmw2: rhbz 438299 15419 15420commit de2f609ff0004ef8b74727bfebc2c74fb91205ea 15421Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 15422Date: Wed Mar 26 18:35:21 2008 -0400 15423 15424 AVIVO: no need to call PreinitXv() on AVIVO chips as they have no overlay 15425 15426commit 75884c257bc2bcfa5b498a77d4c403f09face036 15427Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 15428Date: Wed Mar 26 18:16:47 2008 -0400 15429 15430 XAA: update message about render so as to not confuse users 15431 15432commit 9b4473c1d830b88866dd22e8174a07195bd6fcf4 15433Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 15434Date: Wed Mar 26 18:01:29 2008 -0400 15435 15436 RADEON: attempt to fix lid issues 15437 15438 On some laptops the bios attempts to re-program the chip 15439 when a lid event comes in. This should hopefully prevent 15440 the bios from doing that. 15441 15442commit 8b144830fe9b4a0cee4745023de5e7d387070f60 15443Author: Alex Deucher <alex@samba.(none)> 15444Date: Tue Mar 25 01:15:05 2008 -0400 15445 15446 RV250: disable textured video due to HW bug 15447 15448 The YUV->RGB conversion in the texture engine is broken 15449 on RV250 so the colors come out wrong. 15450 15451commit 1789f11ab91633d3928f8b71988d51ff44bda9d1 15452Author: Alex Deucher <alex@samba.(none)> 15453Date: Mon Mar 24 19:03:30 2008 -0400 15454 15455 R3xx/R5xx: flush PVS state before enabling pvs-bypass 15456 15457commit 305a3310963a5dd07b3495015b06aa8c7c4e6b02 15458Author: Alex Deucher <alex@samba.(none)> 15459Date: Mon Mar 24 14:25:03 2008 -0400 15460 15461 R3xx/R5xx: move more VAP, etc. state setup into common init3d() function 15462 15463 Also some minor code cleanups 15464 15465commit 399b1d405e602c62d6deebea6d7e1f38886cd8e2 15466Author: Alex Deucher <alex@samba.(none)> 15467Date: Mon Mar 24 13:04:57 2008 -0400 15468 15469 R3xx/R5xx: use non VAP/TCP for textured video 15470 15471 Just extra state to emit. 15472 15473commit cd77ec18f32a7b36acb655c927bbfd7044019f97 15474Author: Dave Airlie <airlied@redhat.com> 15475Date: Mon Mar 24 18:42:21 2008 +1000 15476 15477 r300: don't bother with VAP/TCL for render. 15478 15479 We just send more data to the card to process per transaction, without getting 15480 any actual gains, as we already pre-compute the vertices without needing 15481 any clipping or transforms from the card. 15482 15483 Perhaps some stuff could be done on-card, but so far the code is a lot 15484 faster if we avoid sending this extra info. 15485 15486 pre: 150000 glyphs/sec 15487 post: 185000 glyphs/sec 15488 15489commit 301c6739b88676a0c78fc72194e993f894b8dc28 15490Author: Alex Deucher <alex@botch2.com> 15491Date: Sun Mar 23 11:14:02 2008 -0400 15492 15493 RS4xx: Revert back to previous fifo settings for now 15494 15495 Setup of these registers needs more investigation. 15496 15497commit 9bea60b3eb378de5e1d44cc02a2763f4feae7882 15498Author: Alex Deucher <alex@botch2.com> 15499Date: Sat Mar 22 11:46:15 2008 -0400 15500 15501 RS4xx: more work on disp/disp2 fifo setup 15502 15503commit 90f11c3986c28daa7b600b9662da145af325d264 15504Author: Alex Deucher <alex@botch2.com> 15505Date: Sat Mar 22 11:29:51 2008 -0400 15506 15507 RS4xx: missed this on the last commit. 15508 15509commit 6d5066a451017a2683addc9e2496987626795dda 15510Author: Alex Deucher <alex@samba.(none)> 15511Date: Fri Mar 21 16:21:54 2008 -0400 15512 15513 RS4xx: attempt to set up disp/disp2 fifos correctly 15514 15515 If you have an XPRESS chip, please test!!! 15516 15517commit fb1cffac05ae20c8365b25a2042b0ae961880faf 15518Author: Alex Deucher <alex@samba.(none)> 15519Date: Fri Mar 21 15:24:36 2008 -0400 15520 15521 RS4xx: attempt to fix TMDS/DVO support 15522 15523 XPRESS chips added a second set of FP control registers. 15524 I don't have the hw to test however. 15525 15526commit 5e3b21284482df9974c9a58f248f0100def2bb0c 15527Author: Alex Deucher <alex@samba.(none)> 15528Date: Wed Mar 19 19:15:05 2008 -0400 15529 15530 Disable the setting of HARDWARE_CURSOR_BIT_ORDER_MSBFIRST 15531 15532 See bug 11796 15533 15534commit 17cd42ed31814ba329a6a68edd0d75390a7da40e 15535Author: Matt Turner <mattst88@gmail.com> 15536Date: Wed Mar 19 18:17:10 2008 -0400 15537 15538 Enable BSR in Log2 functions 15539 15540 This patch edits RADEONLog2 and ATILog2 to use the x86 BSR instruction instead 15541 of looping through bits. It should provide a somewhat of a speed increase in 15542 this function on x86 and AMD64 architectures. 15543 15544 Note: the BSR instruction was added with the 80386 CPU and is therefore not 15545 compatible with earlier CPUs, though I highly doubt it's even possible to use a 15546 286 in conjunction with a Radeon. 15547 15548 The inline assembly also works with Intel's compiler (icc). 15549 15550commit c83827b4d2b6f03c54429e757a756eb99ff8be6b 15551Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 15552Date: Wed Mar 19 17:58:34 2008 -0400 15553 15554 [PATCH] Compile warning fixes. 15555 15556 Minor changes to avoid declarations mixed with code. 15557 Ansified functions with empty prototype to specify they don't 15558 receive arguments. 15559 Added some prototypes to radeon.h, and major reorder on radeon.h 15560 adding prototypes in alphabetical order and specifying to file that 15561 defines it. 15562 15563commit bed9754ad21d6c0a7f61067b04ba31c430a7cecb 15564Merge: 55e446b f71ac0e 15565Author: Alex Deucher <alex@samba.(none)> 15566Date: Wed Mar 19 16:06:41 2008 -0400 15567 15568 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into r3xx-render 15569 15570commit 55e446b5bc091e6c7b3c2e9ae20b45130555c246 15571Author: Alex Deucher <alex@samba.(none)> 15572Date: Wed Mar 19 13:15:32 2008 -0400 15573 15574 R3xx/R5xx: Make sure to clamp the output of the FS 15575 15576commit b6aa4279cbe68cc8e4523795e9714fb798b62d98 15577Author: Alex Deucher <alex@samba.(none)> 15578Date: Wed Mar 19 12:45:01 2008 -0400 15579 15580 R5xx: bump textured video limits to 4096 15581 15582commit 4a445a3e8c4c5ecd9d4ef8daa26906c3ceaa94a1 15583Author: Alex Deucher <alex@samba.(none)> 15584Date: Wed Mar 19 12:31:51 2008 -0400 15585 15586 RADEON: add new macros to distinguish between R3xx and R5xx 3D 15587 15588commit 85d0c9e8d22ccc72bec87b3fd44da5d7609293e0 15589Author: Alex Deucher <alex@samba.(none)> 15590Date: Wed Mar 19 12:07:33 2008 -0400 15591 15592 RADEON: fixed textured video with XAA and tiling 15593 15594commit f5951db7b3522e0fe6af7f46a170c9c9a60a9bff 15595Author: Alex Deucher <alex@samba.(none)> 15596Date: Wed Mar 19 12:01:50 2008 -0400 15597 15598 RV515: fix textured video and EXA Composite 15599 15600 There seems to be an issue with the PVS setup on RV515, but 15601 bypassing it seems to work fine. 15602 15603commit 13573879fe56368ad06234712b677c23fabc56c6 15604Author: Dave Airlie <airlied@redhat.com> 15605Date: Wed Mar 19 15:06:47 2008 +1000 15606 15607 r500: make it work from startup. 15608 15609 I'm not sure why this worked or what is going wrong here, really the 15610 VAP internal architecture escapes me :) 15611 15612commit d331dd64d644a18ec99a2136cd0943b5edca1f03 15613Author: Alex Deucher <alex@samba.(none)> 15614Date: Tue Mar 18 19:44:26 2008 -0400 15615 15616 R3xx/R5xx: remove extra return after last commit 15617 15618commit bc34df7a9c35cdd38c49d5c22471f3f487a33d6e 15619Author: Alex Deucher <alex@samba.(none)> 15620Date: Tue Mar 18 19:39:47 2008 -0400 15621 15622 R3xx/R5xx: switch an ErrorF() to RADEONFALLBACK() 15623 15624commit 6f03f8fe0ecf4181dcf125049cf63bece0451fb2 15625Author: Alex Deucher <alex@samba.(none)> 15626Date: Tue Mar 18 19:36:05 2008 -0400 15627 15628 R3xx: we only use 2 temps, not 3 15629 15630commit 8bb71ab4a3eb4fb6ef7f709e87c8df387cb70ee3 15631Author: Tilman Sauerbeck <tilman@code-monkey.de> 15632Date: Tue Mar 18 14:36:08 2008 -0400 15633 15634 R3xx/R5xx: fix up a8-src-something_with_colors 15635 15636commit c362591d9b496df30668543158e4de44de742dc3 15637Author: Alex Deucher <alex@samba.(none)> 15638Date: Tue Mar 18 11:15:17 2008 -0400 15639 15640 R3xx/R5xx: remove some cruft 15641 15642commit 89fe6d2c7d7471e6088558130f6e49f46c31dd47 15643Author: Dave Airlie <airlied@linux.ie> 15644Date: Tue Mar 18 09:43:43 2008 -0400 15645 15646 R5xx: fix typ in r5xx render accel 15647 15648 This gets render working on r5xx 15649 15650commit 79b40ebcd8dedfc83e484c1024beeeaccc6124f3 15651Author: Alex Deucher <alex@samba.(none)> 15652Date: Tue Mar 18 02:46:49 2008 -0400 15653 15654 R5xx: first pass at render support (untested) 15655 15656commit 71292c8f193230255d1d980c2e996bb01d04fab6 15657Author: Alex Deucher <alex@samba.(none)> 15658Date: Tue Mar 18 00:45:37 2008 -0400 15659 15660 R5xx: bump tex/dst limits to 4096 15661 15662commit 30b52f8aa6a471455284f59b5b27252743892b13 15663Author: Alex Deucher <alex@samba.(none)> 15664Date: Mon Mar 17 23:20:10 2008 -0400 15665 15666 R3xx/R5xx: whitespace cleanup and cruft removal 15667 15668commit 9c9f1b538ed710c3066775fba0a8e936b63087b1 15669Author: Alex Deucher <alex@samba.(none)> 15670Date: Mon Mar 17 23:01:37 2008 -0400 15671 15672 R3xx: get masks working and cleanup 15673 15674 RS offset was wrong for mask texture 15675 15676commit ef94febd74f8ee63081b61e42f093a5a2b8fbf1e 15677Author: Alex Deucher <alex@samba.(none)> 15678Date: Mon Mar 17 22:27:19 2008 -0400 15679 15680 R3xx: minor adjustments 15681 15682commit f71ac0e40b9d950bcb3bba42a75d41f45b6ed1bf 15683Author: Alban Browaeys <prahal@yahoo.com> 15684Date: Mon Mar 17 20:48:48 2008 -0400 15685 15686 RADEON: Revert to old behavior when resetting the memmap on VT switch 15687 15688 Not sure why this needs to be done twice. Should fix bug 14980 15689 Probably needs more investigation. 15690 15691commit bedbbf196dc97ee5142e7dfae16fb6f317fca5a7 15692Author: Alex Deucher <alex@samba.(none)> 15693Date: Mon Mar 17 20:16:25 2008 -0400 15694 15695 R3xx: some progress 15696 15697commit af0e626c132de2dd9958fec657fcc85d4c0fe5e1 15698Author: Alex Deucher <alex@samba.(none)> 15699Date: Mon Mar 17 18:07:12 2008 -0400 15700 15701 R3xx: fix errant w 15702 15703commit 29ea5bfc0eb3194e2454fc3ee863df54f0300880 15704Author: Alex Deucher <alex@cube.(none)> 15705Date: Mon Mar 17 16:41:57 2008 -0400 15706 15707 RADEON: fix typo in RADEONAdjustMemMapRegisters() 15708 15709commit ab317e85c5ab1a249a510c34aeb3a908be1a66fc 15710Author: Alex Deucher <alex@cube.(none)> 15711Date: Mon Mar 17 15:28:09 2008 -0400 15712 15713 RADEON: make sure var is initialized properly in RADEONAdjustMemMapRegisters() 15714 15715commit 208d307227e15f37a6af5194398ed23266ff743a 15716Author: Dave Airlie <airlied@linux.ie> 15717Date: Sun Mar 16 19:39:23 2008 +1000 15718 15719 radeon: the 0x5974 appears to be a mobility chip... 15720 15721 After debugging with partymola on #radeon, adding this allowed his 15722 Dell Vostro 1000 to work properly 15723 15724commit 9bc7c2ec4048e1677547c1d60c51ccb954f7589a 15725Author: Alex Deucher <alex@samba.(none)> 15726Date: Fri Mar 14 20:12:22 2008 -0400 15727 15728 R3xx: odds and ends... 15729 15730 still not working. 15731 - swizzle US output for BGR formats 15732 - no need to write to temps in ALU ops, 15733 write to output only 15734 - flush the PVS before updating 15735 15736commit 96bea7906c4706fcd57a9cd8f1ce3feab6ac676d 15737Author: Alex Deucher <alex@samba.(none)> 15738Date: Fri Mar 14 15:59:36 2008 -0400 15739 15740 R3xx: theoretical support for component alpha 15741 15742 masks are still broken so... 15743 15744commit cffe3dcc8991cd7c457a9c1a9f41055aa9ea3436 15745Author: Alex Deucher <alex@samba.(none)> 15746Date: Fri Mar 14 14:37:43 2008 -0400 15747 15748 R3xx: VS WIP 15749 15750commit b73f52a50dfd6ff8d92f04d6b510c39582c6ac67 15751Author: Alex Deucher <alex@samba.(none)> 15752Date: Fri Mar 14 14:20:49 2008 -0400 15753 15754 R3xx/R5xx: enable VS for mask texture 15755 15756commit 569a14ca9be1e18fe9921edc816ac3dc32d6cca7 15757Author: Alex Deucher <alex@samba.(none)> 15758Date: Fri Mar 14 13:32:12 2008 -0400 15759 15760 R3xx/R5xx: Fix magic numbers in vertex shaders 15761 15762commit 4878997529601d62e257aa1c9112bd460561de73 15763Author: Alex Deucher <alex@samba.(none)> 15764Date: Thu Mar 13 21:23:40 2008 -0400 15765 15766 R3xx: make sure to set the FS code size correctly 15767 15768commit 22f46b88ef05afb6a6b6d70007ac4980a446430e 15769Author: Alex Deucher <alex@samba.(none)> 15770Date: Thu Mar 13 20:25:33 2008 -0400 15771 15772 R3xx: attempt to setup the rasterizer properly for mask texture 15773 15774 Not working yet 15775 15776commit 081fc9e892fa3d2e07b7db65b2e2719646255463 15777Author: Alex Deucher <alex@samba.(none)> 15778Date: Thu Mar 13 18:38:26 2008 -0400 15779 15780 R3xx: more mask work 15781 15782commit 2bf0236c03538ace3ce6d0e68f0829fc47d1385b 15783Author: Alex Deucher <alex@samba.(none)> 15784Date: Thu Mar 13 18:32:00 2008 -0400 15785 15786 R3xx: enable composite for non-mask cases 15787 15788commit 74286ba41302107d2fc626fee2181f7c4bc18164 15789Author: Alex Deucher <alex@samba.(none)> 15790Date: Thu Mar 13 18:25:32 2008 -0400 15791 15792 R3xx: add basic mask support 15793 15794commit a2bbe10d866567911b68f222b4758624bfe9bf84 15795Author: Alex Deucher <alex@samba.(none)> 15796Date: Thu Mar 13 18:16:53 2008 -0400 15797 15798 R300: setup source selects and output swizzling 15799 15800commit b9974ecce7d1932595226004858b08a7a6b188dc 15801Author: Alex Deucher <alex@samba.(none)> 15802Date: Thu Mar 13 17:35:38 2008 -0400 15803 15804 R3xx: set the texture id and add some register info 15805 15806commit 0ef700b7da5e554a0d0d166f3fde85ff45c9eb1f 15807Author: Alex Deucher <alex@samba.(none)> 15808Date: Thu Mar 13 17:02:25 2008 -0400 15809 15810 R3xx/R5xx: enable blending 15811 15812commit b35c09a597c93a1d9f06ef0091c96822b0653f98 15813Author: Dave Airlie <airlied@redhat.com> 15814Date: Thu Mar 13 18:42:29 2008 +1000 15815 15816 xv: fixup XAA on r500 textured video 15817 15818 the XAA area should never end up tiled. This may break with nooffscreen pixmaps 15819 15820commit d4446461c3630caff166456c351ace34f57cc119 15821Author: Matt Turner <mattst88@gmail.com> 15822Date: Tue Mar 11 21:20:53 2008 -0400 15823 15824 Properly fix uninitialized variables warnings 15825 15826 According to commit 9fd13e6773371c82b9799a5bda7c96ffa5cafe8c to 15827 xf86-video-intel by Kristian Høgsberg, there is a better way to fix the 15828 possibly initialized variables warnings. This patch will use Kristian's fix. 15829 15830commit 20adfd7390d9b1f100e0c4a14f175377b8335c82 15831Author: Alex Deucher <alex@cube.(none)> 15832Date: Tue Mar 11 20:09:35 2008 -0400 15833 15834 RADEON: enable output attributes that require a modeset immediately 15835 15836 This should fix bug 14915 15837 15838commit 53ba7f5771b0b53fb0d3bc29d64bdd3813756d10 15839Author: Alex Deucher <alex@cube.(none)> 15840Date: Tue Mar 11 19:12:40 2008 -0400 15841 15842 RADEON: fix vblank interrupts after VT switch or suspend/resume 15843 15844commit e946c097f0438afbea6f3dd37ee39d67d415708c 15845Author: Matt Turner <mattst88@gmail.com> 15846Date: Tue Mar 11 19:07:58 2008 -0400 15847 15848 [PATCH] Fix a few warnings 15849 15850commit 8e160508520c0a24ca90aad182f296461ca0d9b6 15851Author: Alex Deucher <alex@cube.(none)> 15852Date: Tue Mar 11 18:11:13 2008 -0400 15853 15854 DCE3: add support for PCIEPHY (untested) 15855 15856commit fbded88a2925f9f049936dad0736721e7b84a6ee 15857Author: Alex Deucher <alex@cube.(none)> 15858Date: Tue Mar 11 14:10:31 2008 -0400 15859 15860 ATOM: remove some cruft 15861 15862commit 3263f6e4a410281d620c288a92bb4521f7b6fc06 15863Author: Alex Deucher <alex@cube.(none)> 15864Date: Tue Mar 11 14:05:48 2008 -0400 15865 15866 DCE3: enable DPMS on DIG ports 15867 15868commit eb90e235b58c94f3d4d75394725ab2fe246a42ff 15869Author: Alex Deucher <alex@cube.(none)> 15870Date: Tue Mar 11 13:53:54 2008 -0400 15871 15872 DCE3: adjust PLL for DCE3 chips 15873 15874 this fixes stability issues on digital outputs and certain modes. 15875 15876commit 552615ccc5360baafb8bb41698c1ca27816fd4b2 15877Author: Alex Deucher <alex@cube.(none)> 15878Date: Tue Mar 11 13:38:29 2008 -0400 15879 15880 ATOMBIOS: enable load detection by default on both DACs 15881 15882 Load detection is reliable with atom, so enable it by default 15883 on both DACA and DACB, rather than just DACA. 15884 15885commit 78b10487cf222c96f8944ba25e2ea970506b3535 15886Author: Alex Deucher <alex@cube.(none)> 15887Date: Tue Mar 11 13:16:00 2008 -0400 15888 15889 DCE3: add output attribute to enable/disable coherent mode 15890 15891 Enabled by default. The TMDS transmitter can be programmed 15892 slightly differently depending on the chips in the panel. If you 15893 have problems with tmds on a particular panel, try disabling it. 15894 15895commit d20be31c46fbec623af4c3628a7c603ceacf500f 15896Author: Alex Deucher <alex@botch2.(none)> 15897Date: Mon Mar 10 21:05:43 2008 -0400 15898 15899 RV550: MC setup is like RV515 not RV530 15900 15901commit 38606b08b68842fbcc81c233009c1117269f3be9 15902Author: Matthieu Herrb <matthieu@bluenote.herrb.net> 15903Date: Sat Mar 8 23:22:59 2008 +0100 15904 15905 Makefile.am: nuke RCS Id 15906 15907commit 9d710ee1a44cf2f3a948fbdbe17ef09521cbe744 15908Author: Alex Deucher <alex@cube.(none)> 15909Date: Fri Mar 7 15:09:14 2008 -0500 15910 15911 AVIVO: clean up some unused variables 15912 15913commit c28c30c9f3d7bfebfd56a5c982c96f0090982054 15914Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 15915Date: Fri Mar 7 14:10:49 2008 -0500 15916 15917 RADEON: Fix crash in last commit 15918 15919commit c3a3635865e380c784a226c8ead069d4716d6b75 15920Author: Dave Airlie <airlied@redhat.com> 15921Date: Thu Mar 6 20:17:45 2008 -0500 15922 15923 RADEON: fix tiling/interlaced interaction with randr 1.2 15924 15925commit df1b94dc4eb1f35b636dbf2ec0ab1c2da9937c0d 15926Author: Alex Deucher <alex@botch2.(none)> 15927Date: Thu Mar 6 19:22:08 2008 -0500 15928 15929 DCE3: Ignore outputs with DIN connectors for now 15930 15931commit cb2dc19387c7b6494c47c76d683cf38a48700768 15932Author: Alex Deucher <alex@cube.(none)> 15933Date: Thu Mar 6 18:33:12 2008 -0500 15934 15935 AVIVO: fix typo from a previous commit 15936 15937 Leave tv dpms hook disabled or you may get bad interactions 15938 with the shared DAC 15939 15940commit 77355de48057e5e7e0d5b3f3cf5a7a92220a53b1 15941Author: Alex Deucher <alex@cube.(none)> 15942Date: Thu Mar 6 17:46:00 2008 -0500 15943 15944 AVIVO: don't add outputs for invalid connectors 15945 15946commit 600dbe080997a01ceaf6be86723189d518bc1281 15947Merge: 594743a 5b7875d 15948Author: Alex Deucher <alex@cube.(none)> 15949Date: Thu Mar 6 17:31:37 2008 -0500 15950 15951 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 15952 15953commit 594743a99811a8b0f391412892414fcd158eeb56 15954Author: Alex Deucher <alex@cube.(none)> 15955Date: Thu Mar 6 17:30:21 2008 -0500 15956 15957 AVIVO: fix up memsize detection for IGP chips 15958 15959commit 5b7875d0cbfbdbcd1515c4e942d30de298b49dff 15960Author: Doug Chapman <doug.chapman@hp.com> 15961Date: Thu Mar 6 14:31:06 2008 -0500 15962 15963 Bug #14826: Fix a bogus check around xf86SetOperatingState. 15964 15965commit 651fe23f4c650ed91843dec48db24e18e8b91219 15966Merge: 3de2dc8 766f464 15967Author: Adam Jackson <ajax@redhat.com> 15968Date: Thu Mar 6 14:28:27 2008 -0500 15969 15970 Merge branch 'master' of git+ssh://git.freedesktop.org/git/xorg/driver/xf86-video-ati 15971 15972commit 41171c25cd235bafad26bcbabced16ead4b8c54b 15973Author: Alex Deucher <alex@cube.(none)> 15974Date: Thu Mar 6 14:05:18 2008 -0500 15975 15976 DCE3.0: add support for crtc memreq table 15977 15978commit 766f464dfdfccadef23e4232f2bce5db22195513 15979Author: Alex Deucher <alex@cube.(none)> 15980Date: Thu Mar 6 13:35:43 2008 -0500 15981 15982 RADEON: take 2 on proper pragma pack support for bsds 15983 15984 See bug 14594. Based on suggestion by Henry Zhao 15985 15986commit a842ce9ca6494e724a7828cead9b61c9ef02b6aa 15987Author: Alex Deucher <alex@cube.(none)> 15988Date: Thu Mar 6 12:32:18 2008 -0500 15989 15990 DCE3.0: Minor fixups 15991 15992commit 8a1ba374033591c725a78923aa30829e4de2a5ae 15993Author: Alex Deucher <alex@cube.(none)> 15994Date: Thu Mar 6 09:53:51 2008 -0500 15995 15996 RADEON: option to override TVDAC adj values from bios with driver defaults 15997 15998 If you have a washed out image on the tv dac, try this option. 15999 Option "DefaultTVDACAdj" "TRUE" 16000 16001commit 0ed48f8f651a28e189f9fee8c6b593da0178d21c 16002Author: Alex Deucher <alex@cube.(none)> 16003Date: Wed Mar 5 18:41:01 2008 -0500 16004 16005 AVIVO: Initial support for DCE 3.0 using atombios 16006 16007 DACs are working well, DIG support (DVI, HDMI, LVDS, etc.) 16008 still has some issues. 16009 16010commit 2901e99f1942842856cd39c1dcc8b22f3cf7d9e3 16011Author: Alex Deucher <alex@botch2.(none)> 16012Date: Wed Mar 5 10:40:06 2008 -0500 16013 16014 RADEON: fix fetching of dac2 adj values from newer bios tables 16015 16016commit 74eb981287d76836327830bd51272f605a07e0cc 16017Author: Alex Deucher <alex@botch2.(none)> 16018Date: Mon Mar 3 12:02:44 2008 -0500 16019 16020 ATOMBIOS: fix atombios parser support on *bsd 16021 16022 bsd requires a different pragma pack than Linux. 16023 See bug 14594. 16024 16025commit f7769ea86e265f347eb58c517ccb5ef8b35eec27 16026Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 16027Date: Sun Mar 2 14:49:21 2008 -0500 16028 16029 [PATCH] Ensure symbols used by other modules are visible. 16030 16031 The xf86-video-ati drivers are one of the cases where LoaderSymbol is 16032 widely used in some obscure ways. This patch fixes the problem, and 16033 allows compiling with -fvisibility=hidden. 16034 16035commit a4398ac3ad77216f2c8aa628425bef5f2912a0a9 16036Author: Alex Deucher <alex@cube.(none)> 16037Date: Sat Mar 1 18:52:26 2008 -0500 16038 16039 RS6xx: change isIGP checks to CHIP_FAMILY_RS690 16040 16041 these paths are only relevant on RS6xx chips 16042 16043commit 67d4d04836c05293b844bc505f303cfb04c0f8a4 16044Author: Alex Deucher <alex@cube.(none)> 16045Date: Sat Mar 1 18:33:18 2008 -0500 16046 16047 RADEON: use xf86SetDesiredModes() in screeninit and enterVT 16048 16049 this should restore the proper output state on VT switches 16050 16051commit be0858a84fbdf74c0b844f462933a221d48c707d 16052Author: Maciej Cencora <m.cencora@gmail.com> 16053Date: Sat Mar 1 18:11:58 2008 -0500 16054 16055 RADEON: remove driver rec copies of mc info, use save rec directly 16056 16057 info->mc_* were used and the immediately copied into info->ModeReg 16058 ones. Just use the ModeReg copies directly. 16059 16060commit dd8ee1b444f4b973a1e0fadca5f943f2162b5e94 16061Author: Alex Deucher <alex@cube.(none)> 16062Date: Sat Mar 1 16:23:51 2008 -0500 16063 16064 RADEON: memmap rework 1 16065 16066 Don't restore memmap regs on every mode switch. 16067 Just do memmap save/restore/setup on server start and VT switch. 16068 16069commit 1f6a23000001f3d1c21b5c04f94714a8caa7aa8b 16070Author: Alex Deucher <alex@cube.(none)> 16071Date: Sat Mar 1 15:53:42 2008 -0500 16072 16073 RADEON: only restore legacy dac regs on legacy radeons 16074 16075commit dee6cef8e62d0651c00319e03eea92940fd24aa4 16076Author: Alex Deucher <alex@botch2.(none)> 16077Date: Sat Mar 1 14:39:32 2008 -0500 16078 16079 RS4xx: enable exa render accel and textured video 16080 16081 RS6xx paths seem to work fine on RS4xx 16082 16083commit 129f737efe4e8d1a368e7db4b063bdcd9339cb09 16084Author: Alex Deucher <alex@cube.(none)> 16085Date: Sat Mar 1 14:32:30 2008 -0500 16086 16087 AVIVO: save/restore regs by block 16088 16089 Save/Restore the entire block for each output. 16090 This should fix VT switch problems. 16091 16092commit b069aadaa63a95d7a71b5cfbab83577b49501094 16093Author: Alex Deucher <alex@botch2.(none)> 16094Date: Fri Feb 29 22:36:02 2008 -0500 16095 16096 AVIVO: LVDS panels need dithering enabled 16097 16098 Fixes bug 14760 16099 16100commit fe87bdee815372b4b4d7d4c705e34681625b90f2 16101Author: Alex Deucher <alex@botch2.(none)> 16102Date: Fri Feb 29 13:10:13 2008 -0500 16103 16104 AVIVO: disable pageflipping on avivo chips until we have proper drm support 16105 16106commit fb3678c7f511d539a51cd090cb8b5041d7d2ba26 16107Author: Alex Deucher <alex@botch2.(none)> 16108Date: Fri Feb 29 13:01:21 2008 -0500 16109 16110 R5xx: fix register count when sending fragment program for textured video 16111 16112commit a66d37d1a896ec934989592457c2beff8e6f1639 16113Author: Alex Deucher <alex@botch2.(none)> 16114Date: Fri Feb 29 04:07:05 2008 -0500 16115 16116 fix off-by-one in last commit 16117 16118commit e56062960be0c8d3947861dd5e0691fce6516b99 16119Author: Alex Deucher <alex@cube.(none)> 16120Date: Thu Feb 28 19:16:39 2008 -0500 16121 16122 AVIVO: save/restore scaler regs 16123 16124commit ae1c39a9b3e666404d0931679c9078c2e125a8bc 16125Author: Alex Deucher <alex@cube.(none)> 16126Date: Thu Feb 28 18:53:55 2008 -0500 16127 16128 RS6xx: rework output parsing 16129 16130 Turns out it's not as complex as I originially thought. 16131 IGP chips just have non-standard GPIO entires for DDC. 16132 16133commit d8d6c9fe4ae7e1ab67dd041a251e901d97c29ed6 16134Author: Alex Deucher <alex@botch2.(none)> 16135Date: Thu Feb 28 17:01:14 2008 -0500 16136 16137 RS6xx: fix typos in previous commit 16138 16139 Noted by Maciej Cencora on IRC 16140 16141commit 46547ae8bdbc5c10f1fd028b95ec4c5c31a5b318 16142Author: Alex Deucher <alex@cube.(none)> 16143Date: Thu Feb 28 14:29:30 2008 -0500 16144 16145 AVIVO: disable dithering on DFPs 16146 16147 This should fix the color banding some people have noticed. 16148 Also save/restore DDIA regs on RS6xx 16149 16150commit 72a53d6f20ac29b3baddb7d8af04f19b76d2e04f 16151Author: Michel Dänzer <michel@tungstengraphics.com> 16152Date: Thu Feb 28 17:38:04 2008 +0100 16153 16154 Handle EXA coordinate limits more cleverly. 16155 16156 Generally set the 2D engine limits, and only enforce the 3D engine limits in the 16157 CheckComposite hook. This should still prevent useless migration of pixmaps the 16158 3D engine can't handle but allows for basic acceleration of bigger ones. 16159 16160 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=14708 . 16161 16162commit 5249f450a2487475a95531603cc8668db2c21c33 16163Author: Michel Dänzer <michel@tungstengraphics.com> 16164Date: Thu Feb 28 12:23:58 2008 +0100 16165 16166 Fix 16 bit packed YUV XVideo playback on big endian systems with DRI disabled. 16167 16168 http://bugs.freedesktop.org/show_bug.cgi?id=14668 16169 16170commit e40d75fd8b2aece9dae8076fac822a4a83025fb2 16171Author: Alex Deucher <alex@samba.(none)> 16172Date: Wed Feb 27 22:53:10 2008 -0500 16173 16174 R500: fragment program clean up and magic number conversion 16175 16176commit 140dadba36b2191f0e18e41dd987785abd5f55d2 16177Author: Alex Deucher <alex@samba.(none)> 16178Date: Wed Feb 27 22:21:12 2008 -0500 16179 16180 R300: fix up magic numbers in fragment program 16181 16182commit e521476bb5e2dfabc93747e43eb911a8a101357e 16183Author: Alex Deucher <alex@botch2.(none)> 16184Date: Wed Feb 27 21:26:55 2008 -0500 16185 16186 R300/R400: bump up the clip limits for textured video 16187 16188 This allows up to 2560x2560 (hw limit) 16189 16190commit 10db46f11d7e1c055c9ad6034c65ad163dad17dc 16191Author: Alex Deucher <alex@samba.(none)> 16192Date: Wed Feb 27 15:28:50 2008 -0500 16193 16194 AVIVO: make sure we select the right LUT for each crtc 16195 16196commit ea944f38dcfd871b27345698afea1cb986ecb049 16197Author: Alex Deucher <alex@samba.(none)> 16198Date: Wed Feb 27 14:37:52 2008 -0500 16199 16200 R300+: update RADEONCP_REFRESH() to reflect new location of scissor regs 16201 16202commit b865faf95666e2172c3eec143f77fe9c524e4983 16203Author: Alex Deucher <alex@samba.(none)> 16204Date: Wed Feb 27 14:05:44 2008 -0500 16205 16206 R100/R200: move r100/r200 specific 3D setup into appropriate blocks 16207 16208 R3xx+ doesn't have these regs. 16209 16210commit ee485ee0fc550414cd39cdb8fd2769fe0bc996a8 16211Author: George Sapountzis <gsap7@yahoo.gr> 16212Date: Wed Feb 27 18:41:31 2008 +0200 16213 16214 radeon: pass distcheck 16215 16216commit ff110c781299fb729e93572ba3d630ca59e8127d 16217Author: George Sapountzis <gsap7@yahoo.gr> 16218Date: Wed Feb 27 18:41:17 2008 +0200 16219 16220 add warning for r128 16221 16222commit b4a344f36332a12a51c9e896d93e9cb0672d31ef 16223Author: George Sapountzis <gsap7@yahoo.gr> 16224Date: Wed Feb 27 18:40:50 2008 +0200 16225 16226 add warning for mach64 16227 16228commit 81ce4097c12ab64a52d0992f981f588f5a297b7b 16229Author: George Sapountzis <gsap7@yahoo.gr> 16230Date: Wed Feb 27 18:40:23 2008 +0200 16231 16232 drop r128 16233 16234commit 873ff2033f1d0b84c91c48e829aa945f41b490b7 16235Author: George Sapountzis <gsap7@yahoo.gr> 16236Date: Wed Feb 27 18:40:09 2008 +0200 16237 16238 drop r128 (build system) 16239 16240commit 2b8b2676e7b2fd2952f5307a934ef0a11006f838 16241Author: George Sapountzis <gsap7@yahoo.gr> 16242Date: Wed Feb 27 18:39:49 2008 +0200 16243 16244 drop mach64 16245 16246commit 05bbbec11a04e7f6321b92c9dbec4ad4c0d130ff 16247Author: George Sapountzis <gsap7@yahoo.gr> 16248Date: Wed Feb 27 18:39:32 2008 +0200 16249 16250 drop mach64 (build system) 16251 16252commit 68888189cf8d460ef6f8f2f1431a6ffe9fcd8134 16253Author: Alex Deucher <alex@samba.(none)> 16254Date: Wed Feb 27 03:09:28 2008 -0500 16255 16256 RADEON: fix textured video uploads 16257 16258commit 7cfad216a3f1486be194c972b9b6bcf070358586 16259Author: Alex Deucher <alex@samba.(none)> 16260Date: Wed Feb 27 02:13:10 2008 -0500 16261 16262 RADEON: fix clipping for textured video 16263 16264commit a2dca1d68d751def34ef3c6f836574173737bf76 16265Author: Alex Deucher <alex@samba.(none)> 16266Date: Tue Feb 26 16:38:29 2008 -0500 16267 16268 RADEON: Convert textured video to use pipelined uploads 16269 16270commit b4fa1ce9d2da04d94521a82d3c2e95f0fe985ccc 16271Author: Alex Deucher <alex@samba.(none)> 16272Date: Tue Feb 26 14:45:25 2008 -0500 16273 16274 R300/R500: fix up clipping 16275 16276 Based on patch from Peter Zubaj. 16277 16278commit 44e527a117ab0a363135ff066c7f7e0c12e3dc89 16279Author: Alex Deucher <alex@samba.(none)> 16280Date: Tue Feb 26 14:33:24 2008 -0500 16281 16282 R300: fix cordinate clamping in render code 16283 16284 Based on Peter's fix for textured video 16285 16286commit 00ec17ad53d7ad43f19c9b723794ac1b8ef86826 16287Author: Peter Zubaj <pzubaj at marticonet.sk> 16288Date: Tue Feb 26 14:26:14 2008 -0500 16289 16290 R300: Fix clamping and pitch for textured video 16291 16292commit 4207793f9b6c891cb17ba715223d2002668885e3 16293Author: Alex Deucher <alex@samba.(none)> 16294Date: Tue Feb 26 13:16:22 2008 -0500 16295 16296 R300/R500: set the number graphics pipes properly 16297 16298 This should fix the checkerboard issues on r300/r350 cards. 16299 16300commit 823d8bf49e4c0a8c839354179942c5a1944c178d 16301Author: Alex Deucher <alex@samba.(none)> 16302Date: Tue Feb 26 12:25:14 2008 -0500 16303 16304 R300/R500: clean up magic numbers in textured video 16305 16306commit 8ea75b268f11794f4a7e7bac52cb256490ed3fd2 16307Author: Dave Airlie <airlied@linux.ie> 16308Date: Tue Feb 26 16:29:19 2008 +1000 16309 16310 regs: fix spelling properly 16311 16312commit f2816064a6c2c4c35ccba74b9aa80547e25c012e 16313Author: Dave Airlie <airlied@linux.ie> 16314Date: Tue Feb 26 16:28:24 2008 +1000 16315 16316 regs: fix spelling mistake 16317 16318 pointed out by plaes on irc 16319 16320commit 9d2ca30b90607085578dde1f314db663bd5f82ec 16321Author: Alex Deucher <alex@samba.(none)> 16322Date: Mon Feb 25 17:34:00 2008 -0500 16323 16324 R300/R500: clean up magic numbers in render code 16325 16326commit 153ad6fcf704cbf9f811d9986cd4baf04e82c9d2 16327Author: Dave Airlie <airlied@linux.ie> 16328Date: Mon Feb 25 07:10:48 2008 +1000 16329 16330 fixup check for EXA composite pointed out by Alan Swanson 16331 16332commit 85e470e64f629de72e361c77770e2e29998d1bf4 16333Merge: 27ddb39 1b84c76 16334Author: Alex Deucher <alex@samba.(none)> 16335Date: Sun Feb 24 05:37:22 2008 -0500 16336 16337 Merge master and fix conflicts 16338 16339 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 16340 16341 Conflicts: 16342 16343 src/radeon_commonfuncs.c 16344 16345commit 27ddb39b12a0b54e099fd5274c4c91f08e2d2822 16346Author: Alex Deucher <alex@samba.(none)> 16347Date: Sun Feb 24 05:30:11 2008 -0500 16348 16349 R300: clean up magic numbers in RADEONInit3DEngine 16350 16351commit d4c20f33ad6a1f88615cd7e09ad3638896873f9e 16352Author: Alex Deucher <alex@samba.(none)> 16353Date: Sun Feb 24 04:46:10 2008 -0500 16354 16355 R300: replace magic numbers in cache flush 16356 16357commit e52f1c8d2647b81d891ec0728dd582941a76c83f 16358Author: Alex Deucher <alex@samba.(none)> 16359Date: Sun Feb 24 04:43:18 2008 -0500 16360 16361 R300: fill in some more 3D bitfields 16362 16363commit 1b84c76f27c8d24cb42beae26abf000721901c1c 16364Author: Dave Airlie <airlied@redhat.com> 16365Date: Sun Feb 24 19:20:36 2008 +1100 16366 16367 rs690: initial textured video support 16368 16369commit 68158124366db883a901e960fe5f05f8df5efa42 16370Author: Dave Airlie <airlied@redhat.com> 16371Date: Tue Feb 19 19:51:18 2008 +1100 16372 16373 rs690: initial rs690 rotate 16374 16375 suffers same problem as r500 with clipping 16376 16377commit e614bb6965588bf09dcb87f5e08e67120ec9847f 16378Author: Dave Airlie <airlied@redhat.com> 16379Date: Sun Feb 24 01:46:05 2008 -0500 16380 16381 r500: convert fragprog to use register values 16382 16383commit 6ce9ee47c75620b2e5d211c5d59d17271a6a7b19 16384Author: Dave Airlie <airlied@redhat.com> 16385Date: Sat Feb 23 22:49:34 2008 -0500 16386 16387 r500: add textured video Xv adapter support 16388 16389commit 9aaf8b33b22b6ba112869558ae54e021b9487ad2 16390Author: Dave Airlie <airlied@redhat.com> 16391Date: Sat Feb 23 22:16:25 2008 -0500 16392 16393 r500: initial rotate support - not fully working yet. 16394 16395 Just an example of how to setup and run the r500 3D engine for rotation. 16396 this rotates for me but I get some strange clipping on the bottom of my screen 16397 16398commit 05dc3e4fc19d056ce99a7b110665adab2ca1ea21 16399Author: Adam Jackson <ajax@redhat.com> 16400Date: Sat Feb 23 20:29:51 2008 -0500 16401 16402 Clarify R500 US setup. 16403 16404commit 080606ad528972623e1ed1124d8efe7705a73446 16405Author: Adam Jackson <ajax@redhat.com> 16406Date: Sat Feb 23 20:21:17 2008 -0500 16407 16408 Add the R500 US index/data pair. 16409 16410commit edfb3b6bbf0ee17ace8e6ba704a6f54e249fec63 16411Author: Alex Deucher <alex@samba.(none)> 16412Date: Sat Feb 23 19:59:33 2008 -0500 16413 16414 RADEON: no textured video yet on XPRESS chips 16415 16416 Still need to sort out the VAP and PVS stuff 16417 16418commit 4146bfe5d00e40a86d17826fac50d04b2469621d 16419Author: Alex Deucher <alex@samba.(none)> 16420Date: Sat Feb 23 19:21:52 2008 -0500 16421 16422 R500: fix typo in new r5xx regs 16423 16424commit d9be9f34b0d3313e7b22b2a8bb0a8924ad3116bf 16425Author: Alex Deucher <alex@samba.(none)> 16426Date: Sat Feb 23 19:06:30 2008 -0500 16427 16428 RADEON: add textured video support for r1xx-r4xx radeons 16429 16430 Based on the kdrive ati video code by Eric Anholt. 16431 R3xx/R4xx still have some clipping issues in certain situations 16432 16433commit 9dc4acad79196e9d5d94dd710773bfa83456d47f 16434Author: Alex Deucher <alex@cube.(none)> 16435Date: Sat Feb 23 18:29:00 2008 -0500 16436 16437 RS6xx: gpio entry for DDIA varies depending on the number of DFP ports 16438 16439commit ed0a93edf28155308e7ab9d8705581bb38455ea0 16440Author: Adam Jackson <ajax@redhat.com> 16441Date: Sat Feb 23 15:02:17 2008 -0500 16442 16443 Fix R500_US_CONFIG. 16444 16445commit bc2bd6f841b51aeed3b6b4a47dbe758c200bc5a6 16446Author: Adam Jackson <ajax@redhat.com> 16447Date: Sat Feb 23 14:34:18 2008 -0500 16448 16449 Add R500 unified shader register block. 16450 16451commit a7b5c3bb74fc4de5e38a75ac31656445ce823464 16452Author: Alex Deucher <alex@cube.(none)> 16453Date: Fri Feb 22 19:35:11 2008 -0500 16454 16455 RS6xx: fix DDC on DDIA output (usually HDMI port) 16456 16457commit 3327a681e21101cc6f6e162f4e29f9937b69ccc3 16458Author: Alex Deucher <alex@cube.(none)> 16459Date: Fri Feb 22 17:05:56 2008 -0500 16460 16461 ATOM: properly set up DDIA output on RS6xx boards 16462 16463commit 1d0e9ab8b9451101b1b91943546f6c5833c21b3f 16464Author: Michel Dänzer <michel@tungstengraphics.com> 16465Date: Wed Feb 20 10:21:49 2008 +0100 16466 16467 radeon: Fix typo flagged by gcc -Wall. 16468 16469commit b5bd442b60dbc72fe4c1e928ab864aeb0fd7a3cb 16470Author: Alex Deucher <alex@botch2.(none)> 16471Date: Tue Feb 19 20:47:40 2008 -0500 16472 16473 R100: fix render accel for transforms 16474 16475 Not sure why we had a separate broken path for r100 vertex 16476 submission. 16477 16478commit a0a73208a21546ac120fb9a463261836c9ea7b55 16479Author: Alex Deucher <alex@botch2.(none)> 16480Date: Tue Feb 19 20:11:19 2008 -0500 16481 16482 RADEON: restore clock gating and CP clock errata on VT switch 16483 16484 This may help people with hangs on resume 16485 16486commit b77e2aff7453a9f370beba37ca3c25b92b3f97ff 16487Author: Alex Deucher <alex@botch2.(none)> 16488Date: Tue Feb 19 19:55:41 2008 -0500 16489 16490 RADEON: fix DDC types 5 and 6 16491 16492commit af82172a82f2bdf96e571def659a1c70f92dfdbf 16493Author: Alex Deucher <alex@botch2.(none)> 16494Date: Tue Feb 19 19:39:35 2008 -0500 16495 16496 RADEON: update man page with supported chips 16497 16498commit 1302d1b019c8091986aaf9c86b25c1f36a037788 16499Author: Alex Deucher <alex@botch2.(none)> 16500Date: Tue Feb 19 00:34:31 2008 -0500 16501 16502 Bump for 6.8 release 16503 16504commit 830d29e8b280c11e798fb5bf5278bf24ecbac7bf 16505Author: Alex Deucher <alex@cube.(none)> 16506Date: Mon Feb 18 21:17:12 2008 -0500 16507 16508 ATOM: fix RS6xx connector table parsing and fix some warnings 16509 16510commit 03aa4cc6d6e8c715a1c1d677cc1845223505b358 16511Author: Roland Scheidegger <sroland@tungstengraphics.com> 16512Date: Mon Feb 18 20:19:58 2008 -0500 16513 16514 bring back to life planar-to-packed conversion for rs4xx 16515 16516 Here's a patch to bring back the code for converting planar yuv to packed yuv, 16517 if a RS400 family chip is used (though I've no idea if they really all fail 16518 with planar yuv). 16519 16520 fixes bug 12744 16521 16522commit 66b0b3f43bb714bc007169d5ba6dabe34cfc008b 16523Author: George Sapountzis <gsap7@yahoo.gr> 16524Date: Mon Feb 18 10:33:59 2008 +0200 16525 16526 r128/radeon: probably correct fix for non pci access code 16527 16528commit d055b9e800ae50d08cca9db75fc666ce1da9ab52 16529Author: Dave Airlie <airlied@linux.ie> 16530Date: Sat Feb 16 09:13:43 2008 +1000 16531 16532 r128/radeon: hopeful fix for non pci access code 16533 16534commit c773bc6a314327da29e21e4ebac6fa7f3e98a0a7 16535Author: Dave Airlie <airlied@linux.ie> 16536Date: Sat Feb 16 09:05:07 2008 +1000 16537 16538 r128/radeon: fix build without pciaccess 16539 16540commit 690a52da5248f47a587a878d05fce9784957970b 16541Author: Dave Airlie <airlied@linux.ie> 16542Date: Sat Feb 16 08:33:36 2008 +1000 16543 16544 mach64: fix non pciaccess build 16545 16546commit fc85188fd95bf78b7f965cdde3e22b644c74ff81 16547Author: George Sapountzis <gsap7@yahoo.gr> 16548Date: Fri Feb 15 18:36:10 2008 +0200 16549 16550 ati: convert to pci probe 16551 16552 add pciids for each subdriver, make no use of the match_data functionality. 16553 16554 thanks to Alex Deucher for reviewing and testing. 16555 16556commit 665bd7e2f61cac3e029bbad5024034e5136deec1 16557Author: George Sapountzis <gsap7@yahoo.gr> 16558Date: Fri Feb 15 18:34:56 2008 +0200 16559 16560 ati wrapper: translate Device lines 16561 16562commit a596c1618f72179a45289a50a1f9e89462ce9667 16563Author: George Sapountzis <gsap7@yahoo.gr> 16564Date: Fri Feb 15 18:34:31 2008 +0200 16565 16566 r128: do not compile in PciChipsets twice 16567 16568commit 99cd8ff9a7e15fc2b4e55f8bc020f584173a8c2d 16569Author: George Sapountzis <gsap7@yahoo.gr> 16570Date: Fri Feb 15 18:33:51 2008 +0200 16571 16572 ati: drop duplicate pci-id defines 16573 16574commit 146b01b51069dc227d0b125babb3f6957c9b9de2 16575Author: George Sapountzis <gsap7@yahoo.gr> 16576Date: Fri Feb 15 18:33:29 2008 +0200 16577 16578 mach64: clean probe a little 16579 16580 do not report I/O ports now, they are reported later. 16581 16582commit f47d461331a032f9bdcf6f63336e848778cec6cc 16583Author: George Sapountzis <gsap7@yahoo.gr> 16584Date: Fri Feb 15 18:32:56 2008 +0200 16585 16586 mach64: minor cosmetic, I2C type 16587 16588commit 2c66f2e812195167df9ca113044d46deece776ac 16589Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 16590Date: Fri Feb 15 10:59:30 2008 +1000 16591 16592 make distcheck pass 16593 16594commit ed87f367ddab7366f84917b43b31d76df4ce1110 16595Author: Alex Deucher <alex@botch2.(none)> 16596Date: Wed Feb 13 12:53:46 2008 -0500 16597 16598 RADEON: disable LVDS if panel size is 0x0 16599 16600 if we can't get the panel size from the bios or edid 16601 or a user specified option, assume it's not connected. 16602 16603commit 422d7f441fdbb318d22d44db661ac9bd61387bd2 16604Author: Alex Deucher <alex@cube.(none)> 16605Date: Tue Feb 12 15:35:46 2008 -0500 16606 16607 R6xx: when both crtcs are in use make sure they are both enabled 16608 16609 sometimes setting a mode on one crtc can cuase a blank screen on the other. 16610 make sure they are both enabled if they should be. 16611 16612commit 860f5af75274cb236f536e1da09da6bd9a579530 16613Author: Alex Deucher <alex@cube.(none)> 16614Date: Tue Feb 12 14:46:49 2008 -0500 16615 16616 R6xx: fix up use of bios scratch regs to reflect the new offsets 16617 16618commit 32f2119b43a0faf6069d8cc0816f0d9f7914c07f 16619Author: Alex Deucher <alex@cube.(none)> 16620Date: Tue Feb 12 14:20:02 2008 -0500 16621 16622 R6xx: bios scratch regs moved. 16623 16624commit 8d64be6ebd7f50d4bcb587afeee8252c1367dc77 16625Author: Alex Deucher <alex@samba.(none)> 16626Date: Tue Feb 12 12:53:09 2008 -0500 16627 16628 RADEON: make sure we always set up the FP crtc shadow registers for crtc0 16629 16630 The behavior changed when I added rmx center mode support. In cases where 16631 crtc0 drives a DAC this can lead to a blank screen. 16632 16633commit 810d192ee046077a894e0fb5f2dfd6a7c0130766 16634Author: Alex Deucher <alex@cube.(none)> 16635Date: Mon Feb 11 19:23:01 2008 -0500 16636 16637 R6xx: add missing objects 16638 16639commit 85043439426e534e561259ce98bebdd8508b36a9 16640Author: Alex Deucher <alex@botch2.(none)> 16641Date: Mon Feb 11 16:36:58 2008 -0500 16642 16643 R6xx: make sure we set up the HDP base properly 16644 16645commit 9ab5d2ec7c583c74f364d7cfbb54bcd2cd8ae2f5 16646Author: Alex Deucher <alex@botch2.(none)> 16647Date: Mon Feb 11 15:26:51 2008 -0500 16648 16649 RADEON: always restore crtc1 before crtc0 when using both crtcs 16650 16651 In some rare cases restoring crtc0 first will result in a blank screen 16652 on crtc1. If you are having issues with a blank screen on crtc1 16653 that used to work on 6.6.3 or before, this should help. 16654 16655commit e33edca75bd9df0aa19a33e74c38a6d02610befd 16656Author: Alex Deucher <alex@botch2.(none)> 16657Date: Mon Feb 11 00:33:12 2008 -0500 16658 16659 RADEON: Implement proper (hopefully) BIOS/DRIVER interaction for ATOM bios 16660 16661 Tested on atom-based Desktop cards. It'd nice to get some testing 16662 on atom-based laptops. 16663 16664commit 6524e33435a786f7de0064cdd1b04c1120d21593 16665Author: Alex Deucher <alex@botch2.(none)> 16666Date: Sun Feb 10 18:52:52 2008 -0500 16667 16668 RADEON: Implement proper (hopefully) BIOS/DRIVER interaction for COM bios 16669 16670 Tested on my M10-based laptop. 16671 16672commit 8606c1bd175893c77e8c758246a1aed166c8be51 16673Author: George Sapountzis <gsap7@yahoo.gr> 16674Date: Fri Feb 8 19:07:03 2008 +0200 16675 16676 mach64: factor out BIOS panel info parsing 16677 16678commit 9f33218c80f5a6d6d9464aa3db8ae25a4759f351 16679Author: George Sapountzis <gsap7@yahoo.gr> 16680Date: Fri Feb 8 19:06:41 2008 +0200 16681 16682 mach64: minor cosmetic, LCD panel id 16683 16684commit 8cd5a465a03834b4b2f3a19f8d36fc9d2e18c6d4 16685Author: George Sapountzis <gsap7@yahoo.gr> 16686Date: Fri Feb 8 19:06:27 2008 +0200 16687 16688 mach64: minor cosmetic, DAC type 16689 16690commit 7f4db96123fdcba9874226352802d702c2a498bd 16691Author: Dave Airlie <airlied@redhat.com> 16692Date: Fri Feb 8 10:47:10 2008 +1000 16693 16694 add rv670 pciids 16695 16696commit e8899b9978291c62a65f468c92f340f65ad5479d 16697Author: Alex Deucher <alex@botch2.(none)> 16698Date: Thu Feb 7 19:27:38 2008 -0500 16699 16700 R6xx: fix ddc after my i2c rework 16701 16702 Seems r6xx does something different for its i2c table, 16703 revert to the old behavior for now. 16704 16705commit 435de6c4e46ff2bebd4cee58888a66b936cd3fdf 16706Author: Alex Deucher <alex@samba.(none)> 16707Date: Thu Feb 7 19:14:13 2008 -0500 16708 16709 RADEON: sync up with latest upstream versions 16710 16711 atombios.h 16712 ObjectID.h 16713 16714commit 692789a293970f70b88ccb6adcf0676d8b495ae2 16715Author: George Sapountzis <gsap7@yahoo.gr> 16716Date: Thu Feb 7 18:03:37 2008 +0200 16717 16718 mach64: factor out BIOS multimedia parsing 16719 16720commit 933328ffd6d1d872a18d3de8624c4df845a64588 16721Author: George Sapountzis <gsap7@yahoo.gr> 16722Date: Thu Feb 7 18:03:03 2008 +0200 16723 16724 mach64: complement hint for sparc 16725 16726commit 956c8c81f3ff434930a0cb17b027b2f8e4eeabb2 16727Author: George Sapountzis <gsap7@yahoo.gr> 16728Date: Thu Feb 7 18:02:31 2008 +0200 16729 16730 mach64: consolidate refclk #2 16731 16732commit dce4cc26a8e2bf53805ec63763243f3ff6b4a6d3 16733Author: George Sapountzis <gsap7@yahoo.gr> 16734Date: Thu Feb 7 18:02:17 2008 +0200 16735 16736 mach64: consolidate refclk #1 16737 16738commit f7ed807f0d82a7446ebc4acdd4e94df44a675f19 16739Author: George Sapountzis <gsap7@yahoo.gr> 16740Date: Thu Feb 7 18:01:59 2008 +0200 16741 16742 mach64: cosmetic 16743 16744commit cda1cd198f33c26ef1b51532a2126468369743b8 16745Author: George Sapountzis <gsap7@yahoo.gr> 16746Date: Thu Feb 7 18:01:33 2008 +0200 16747 16748 mach64: factor out BIOS clock parsing 16749 16750commit 73ff279469be9c7cbf9f533b85fcb553694ff413 16751Author: George Sapountzis <gsap7@yahoo.gr> 16752Date: Thu Feb 7 18:00:55 2008 +0200 16753 16754 mach64: BIOSBase is no longer used 16755 16756commit caea326cc6f1932bb299f451be013651a5749ea7 16757Author: Dave Airlie <airlied@linux.ie> 16758Date: Wed Feb 6 06:36:13 2008 +1000 16759 16760 r300: move fragprog setup code to prepare composite for now 16761 16762commit 470cd6a401c6a3e8fea981a8fe97c28be3cfb81d 16763Author: Dave Airlie <airlied@linux.ie> 16764Date: Wed Feb 6 06:04:13 2008 +1000 16765 16766 r300: remove r300 specific vertex emission 16767 16768 Set the vertex and fragment engine to expect the mask coords. 16769 16770commit f65e8dfac23adfa199026765fe3a1ea08cf4da67 16771Author: Alex Deucher <alex@cube.(none)> 16772Date: Sun Feb 3 00:09:59 2008 -0500 16773 16774 RADEON: rework i2c handling 16775 16776 Split out clk, data, and lock regs and masks. some cards use different 16777 regs and masks for each. For cards with ATOMBIOS, use the i2c bios 16778 table to grab the i2c data. 16779 16780commit a38a903debc0a50dbc73f59dc2741bbea76d2bd9 16781Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 16782Date: Fri Feb 1 02:59:07 2008 -0500 16783 16784 RADEON: don't restore LVDS_PLL_CNTL for now 16785 16786 seems to cause problems with resume for some users. 16787 this needs further investigation. 16788 see bug 12596 16789 16790commit 5d7bea2b62c86868baf1960665a40249a43bafc5 16791Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 16792Date: Fri Feb 1 02:53:22 2008 -0500 16793 16794 RADEON: remove redundant RADEONDisableDisplays() 16795 16796 use RADEONBlank() instead 16797 16798commit 73b437ce232c94c0067a0d2f70538b6e1e8c07a7 16799Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 16800Date: Fri Feb 1 02:47:06 2008 -0500 16801 16802 RADEON: remove unused "aspect" scaler option 16803 16804commit f1fb9e4daa29bc379f653f847254db1496b625fd 16805Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 16806Date: Fri Feb 1 02:39:18 2008 -0500 16807 16808 RADEON: Implement "center" mode for RMX on legacy radeons 16809 16810commit bcd590103e04bfdb4f12413beacebf344f07e88e 16811Author: Alex Deucher <alex@botch2.(none)> 16812Date: Tue Jan 29 12:12:54 2008 -0500 16813 16814 RADEON: update man page to reflect AGP 1x default revert 16815 16816commit 09348a83d06ba9d3129499c4daedd44a68771530 16817Author: Alex Deucher <alex@botch2.(none)> 16818Date: Tue Jan 29 12:09:24 2008 -0500 16819 16820 Seems the default is more reliable... we can't win. 16821 16822 Revert "radeon: Default to 1x again with non-v3 AGP cards." 16823 16824 This reverts commit b653e5a628bfa4dfb168e96f93f41eb910f409fb. 16825 16826commit 0c26806245381b925b73ff9f3b747e3dcf0ebd6f 16827Author: Alex Deucher <alex@botch2.(none)> 16828Date: Tue Jan 29 10:26:48 2008 -0500 16829 16830 RADEON: Add new RV380 pci id 16831 16832 bug 14289 16833 16834commit ce77ed78a877023da72dbe51609aef9a07e250b1 16835Author: Maciej Cencora <m.cencora@gmail.com> 16836Date: Mon Jan 28 19:02:56 2008 -0500 16837 16838 RS690: Implement MC idle check 16839 16840commit b7de4ff52cfbdd85ee65000613632e21b92af24e 16841Author: George Sapountzis <gsap7@yahoo.gr> 16842Date: Sat Jan 26 19:28:05 2008 +0200 16843 16844 mach64: add hint for sparc and minor cosmetic. 16845 16846commit 09d713aa0ed6367b4457420b3c2832fe8eca9b00 16847Author: Alex Deucher <alex@botch2.(none)> 16848Date: Wed Jan 23 18:05:41 2008 -0500 16849 16850 Revert "RADEON: adjust PAL mode hstart" 16851 16852 This reverts commit 719a9a376e34d99032af75e3f7b002670ccb816b. 16853 16854 This breaks TV out on some cards. 16855 16856commit 719a9a376e34d99032af75e3f7b002670ccb816b 16857Author: Alex Deucher <alex@botch2.(none)> 16858Date: Sun Jan 20 18:40:53 2008 -0500 16859 16860 RADEON: adjust PAL mode hstart 16861 16862 Most people seem to get a more aligned picture with this setting 16863 16864commit b2db8657fb888cff6d64c6dcb182caac389776ce 16865Author: Alex Deucher <alex@botch2.(none)> 16866Date: Sun Jan 20 18:33:22 2008 -0500 16867 16868 RADEON: re-work i2c for DDC 16869 16870 Unify the radeon/avivo paths and grab the data/clk masks from bios 16871 if available 16872 16873commit a43003b24022a833e604f41b2873c0350b34181c 16874Author: Alex Deucher <alex@botch2.(none)> 16875Date: Sat Jan 19 18:49:53 2008 -0500 16876 16877 RADEON: get dac2 adj values from the bios tables 16878 16879commit d4596c52ac9994be26e9ec2d7d57b3892c34abdb 16880Author: Alex Deucher <alex@botch2.(none)> 16881Date: Sat Jan 19 17:17:26 2008 -0500 16882 16883 RADEON: grab pll_in_min/pll_in_max from bios tables if available 16884 16885commit 9a0947c812d0d38d1bca6a91140ac210831a6cb4 16886Author: Alex Deucher <alex@botch2.(none)> 16887Date: Sat Jan 19 13:57:50 2008 -0500 16888 16889 ATOM: Use LVDS edid from bios if available 16890 16891commit 7238258c12def8ef273e5362f716d165f720c5a5 16892Author: Kusanagi Kouichi <slash@ma.neweb.ne.jp> 16893Date: Sat Jan 19 15:04:21 2008 +0100 16894 16895 radeon: Partial fix for XVideo RGB image distortions. 16896 16897commit 32be3cf9d6c34e60ff8c3d6cfe9f73f1869c50e4 16898Author: Brice Goglin <Brice.Goglin@ens-lyon.org> 16899Date: Fri Jan 18 14:42:14 2008 -0500 16900 16901 RADEON: print the name of the output when printing the EDID 16902 16903commit 4ba9430ee97dbce8f77db8de6ce9b753a75e453d 16904Author: George Sapountzis <gsap7@yahoo.gr> 16905Date: Fri Jan 18 18:18:18 2008 +0200 16906 16907 mach64: workaround for corruption at upper-left 16908 16909 commit possible workaround, it's reported multiple times ... 16910 16911commit 12c00111b68c9cf4872a424258c6f8b7247aac47 16912Author: George Sapountzis <gsap7@yahoo.gr> 16913Date: Fri Jan 18 18:17:40 2008 +0200 16914 16915 use stand-alone drivers for each chip family. 16916 16917 Do not load the ati wrapper when the user specifies the sub-driver name in 16918 the Driver line of xorg.conf. Also, for -configure cause the wrapper to fail 16919 and let each sub-driver speak for themselves. 16920 16921commit 24c7d134cd450f9e2cca85e4a2fc3253d250be04 16922Author: Alex Deucher <alex@botch2.(none)> 16923Date: Fri Jan 18 09:50:38 2008 -0500 16924 16925 RADEON: make sure EXA Composite is actually disabled on XPRESS chips. 16926 16927commit eaf425436008092abe81208321a2b3b6698a5d79 16928Author: Dave Airlie <airlied@redhat.com> 16929Date: Fri Jan 18 20:11:57 2008 +1000 16930 16931 fixup register 6594 save/restore 16932 16933commit 3de2dc88cf26ff5932f11cecdf975777b8aa2a4a 16934Author: Adam Jackson <ajax@redhat.com> 16935Date: Wed Jan 16 14:55:05 2008 -0500 16936 16937 Bump CRTC size limits on AVIVO chips so 30" displays work without tweaking. 16938 16939 Note that the CRTC size limits we're using right now are _not_ the 16940 hardware limits, they're just heuristics until we can resize the front 16941 buffer properly. 16942 16943commit 2a89a31481d71a56a9930073cf99d3ae7b4290e1 16944Author: Alex Deucher <alex@samba.(none)> 16945Date: Thu Jan 17 15:08:17 2008 -0500 16946 16947 RADEON: use radeon_output->Flags for tracking RMX rather than mode->Flags 16948 16949commit 495e3119250ffb48489debbaabe560d23753cc43 16950Author: Alex Deucher <alex@samba.(none)> 16951Date: Thu Jan 17 14:56:19 2008 -0500 16952 16953 AVIVO: Add support for RMX 16954 16955 Both centered and expansion modes are supported. Select 16956 using output attributes. 16957 16958commit 6bd510a211f25d52e74791e4a429cd2218ced541 16959Author: Alex Deucher <alex@samba.(none)> 16960Date: Wed Jan 16 18:09:49 2008 -0500 16961 16962 RADEON: add a message about render accel on newer cards 16963 16964commit 85bf3439fe2579aec48f5cd8d65a9d51b1ae8535 16965Author: Alex Deucher <alex@samba.(none)> 16966Date: Wed Jan 16 17:52:06 2008 -0500 16967 16968 R300: only init3d on r3xx 16969 16970commit dbb2ca471dfbff245b30c5055871dee0dc0e3d15 16971Author: Alex Deucher <alex@samba.(none)> 16972Date: Wed Jan 16 17:10:02 2008 -0500 16973 16974 R300: only enable render accel on non-IGP r3xx/r4xx chips for now 16975 16976commit 3c72b100bcfacee600644669b586e86cfd32754e 16977Author: Alex Deucher <alex@samba.(none)> 16978Date: Wed Jan 16 16:55:42 2008 -0500 16979 16980 R300: First pass at render accel 16981 16982 This first pass is pretty limited. All it currently supports 16983 is transforms for rotation. No blending yet. 16984 16985 Based on inital implementation from Wolke Liu with 16986 additional lock-up fixes by Dave Airlie. 16987 16988commit 2ba3562d2af911fdd90881049599e239d27260bc 16989Author: George Sapountzis <gsap7@yahoo.gr> 16990Date: Sat Jan 12 17:11:59 2008 +0200 16991 16992 ati wrapper: xf86PciInfo.h is enough 16993 16994commit c2caeb11a97dad5379d70881c5c0fd834a8c3d54 16995Author: George Sapountzis <gsap7@yahoo.gr> 16996Date: Sat Jan 12 16:18:34 2008 +0200 16997 16998 ati wrapper: add DriverRec's and use them 16999 17000commit 19e1b180fec6f83a474e125465bc60111c0f43e0 17001Author: George Sapountzis <gsap7@yahoo.gr> 17002Date: Sat Jan 12 16:18:07 2008 +0200 17003 17004 mach64: load for both "ati" and "mach64" as driver names 17005 17006 similar to r128/radeon 17007 17008commit 92f54400d5450b29b3a76d5ecc927cf0d73e156e 17009Author: George Sapountzis <gsap7@yahoo.gr> 17010Date: Sat Jan 12 16:17:47 2008 +0200 17011 17012 mach64: add version (need not match with ati) 17013 17014 similar to r128/radeon 17015 17016commit 311ec7b6c54a50a4b8a5a445f7283da2b0b2e0f5 17017Author: George Sapountzis <gsap7@yahoo.gr> 17018Date: Sat Jan 12 16:17:21 2008 +0200 17019 17020 atimisc: rename to mach64 17021 17022commit 5244e235262290eab8a3546f449295c12ed8f6c7 17023Author: Alex Deucher <alex@samba.(none)> 17024Date: Mon Jan 14 16:11:09 2008 -0500 17025 17026 ATOMBIOS: disable the scaler by default on avivo cards 17027 17028 Fixes bug 14001 17029 17030commit 729da30c80d6545b75c8faea44754634f477cc09 17031Merge: 000741e 625a885 17032Author: Alex Deucher <alex@samba.(none)> 17033Date: Mon Jan 14 10:05:01 2008 -0500 17034 17035 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 17036 17037commit 625a885a964488da2a35065eb597a17ee57b22a9 17038Author: George Sapountzis <gsap7@yahoo.gr> 17039Date: Wed Jan 9 17:39:56 2008 +0200 17040 17041 ati wrapper: use pci ids 17042 17043commit 48865200ca9f1daf87e52620b9f8175f88dd886f 17044Author: George Sapountzis <gsap7@yahoo.gr> 17045Date: Wed Jan 9 19:01:04 2008 +0200 17046 17047 radeon: remove stray include 17048 17049commit a5a6b873353611cb7e46e5e375f039957c7051a7 17050Author: George Sapountzis <gsap7@yahoo.gr> 17051Date: Wed Jan 9 13:03:32 2008 +0200 17052 17053 radeon: remove stray _X_EXPORT 17054 17055commit 80c9974b6cdc0147d777df6990b3f3aacd87969d 17056Author: George Sapountzis <gsap7@yahoo.gr> 17057Date: Tue Jan 8 18:15:47 2008 +0200 17058 17059 r128: pci-rework conversion 17060 17061 compile-tested only 17062 17063commit cd4b39403d74f314e1c2cfa4cf0e4fe542891dc3 17064Author: George Sapountzis <gsap7@yahoo.gr> 17065Date: Sat Mar 10 23:12:15 2007 +0200 17066 17067 Drop symbol lists from r128, radeon, theatre. 17068 17069 compile-tested only 17070 17071commit 63b4b8213fabc5a57f897b60b6eaa9f78d86e6ff 17072Author: George Sapountzis <gsap7@yahoo.gr> 17073Date: Wed Mar 21 20:24:22 2007 +0200 17074 17075 [mach64] Drop symbol lists. 17076 17077 egrep LOADER\|SymLists\|Symbols src/*.[hc] 17078 17079 not needed after server commit bca9364f3f4a2376edbcf57a34f704ce28be21ba, i.e. 17080 xserver 1.2 17081 17082 compile-tested only 17083 17084commit 000741e250e54122b0adc91694eb4bfa320a70fb 17085Author: Alex Deucher <alex@samba.(none)> 17086Date: Thu Jan 10 14:49:48 2008 -0500 17087 17088 RADEON: clean up output handling 17089 17090commit 10e7636c02478b8ffe183bb0c46229ca0d6584e1 17091Author: Kristian Høgsberg <krh@bitplanet.net> 17092Date: Wed Jan 9 12:47:39 2008 -0500 17093 17094 RADEON: fix crtc routing on r4xx cards when using atom to init DVO chip 17095 17096commit 3af671f5963810dbfd63abc9889b1d46b68f404c 17097Author: Alex Deucher <alex@botch2.(none)> 17098Date: Wed Jan 9 11:30:25 2008 -0500 17099 17100 RADEON: restore FP2 regs before external encoders 17101 17102 This may fix krh's dvi problem 17103 17104commit 2a54c6bb09ade2ec8f998dfc1624017029d47fa3 17105Author: Alex Deucher <alex@botch2.(none)> 17106Date: Tue Jan 8 18:43:54 2008 -0500 17107 17108 RADEON: Make default output actually work... 17109 17110commit fa3e2055225c27e25465fc46786da1b7574fd3cc 17111Author: Alex Deucher <alex@botch2.(none)> 17112Date: Mon Jan 7 01:13:09 2008 -0500 17113 17114 RADEON: add default outputs if no connected devices 17115 17116 If no connected devices found at server startup, default 17117 to something so the server comes up. LVDS on mobility chips, 17118 DAC or TMDS on others. 17119 17120commit d972cc9237eb90b49b11f8d2bdc5b61f628911dc 17121Author: Alex Deucher <alex@botch2.(none)> 17122Date: Sat Jan 5 17:19:06 2008 -0500 17123 17124 RADEON: Fix TVStandard option 17125 17126commit 45656b9d5a426053da2a094de39c2690c0c6f799 17127Author: Alex Deucher <alex@botch2.(none)> 17128Date: Sat Jan 5 12:00:55 2008 -0500 17129 17130 R128: Like powerpc, don't use VGA by default on sparc 17131 17132commit b8e8db4675d07e45782de0d7c67ee0fd85eaedb3 17133Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17134Date: Fri Jan 4 20:16:19 2008 -0500 17135 17136 RADEON: fix tvdac load detection at server start up 17137 17138commit 3ba7f393d0669df36848715799de8affc10a5534 17139Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17140Date: Fri Jan 4 20:00:48 2008 -0500 17141 17142 RADEON: further fixup for pScrn->pScreen issue 17143 17144 the previous fix seems to cause the driver to hang on 17145 some cards. 17146 17147commit a0de9c0844f9e066e0f02e8cd8045bdd278e6494 17148Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17149Date: Fri Jan 4 19:48:30 2008 -0500 17150 17151 RADEON: improve ntsc image centering 17152 17153commit 03d2f25801c8a8ec15030f06008df112d07c1a2d 17154Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17155Date: Fri Jan 4 18:19:54 2008 -0500 17156 17157 RADEON: fix compile 17158 17159commit 308848783ed9ae27aed7c7de6ee813d375ef495e 17160Author: Dave Airlie <airlied@linux.ie> 17161Date: Sat Jan 5 09:11:55 2008 +1000 17162 17163 i2c: a line mux of 0 is valid 17164 17165commit 94d5a432f72801f821d1c4ce952baba17552659d 17166Author: Dave Airlie <airlied@linux.ie> 17167Date: Fri Jan 4 17:16:54 2008 +1000 17168 17169 atombios: i2c lines are all done with sw so we can accept any of them. 17170 17171 hch reported this working on his laptop with 0x7e60 as EDID for LVDS 17172 17173commit 2e4473b63d65801ae8ac5a8391de232b2201d958 17174Author: Alex Deucher <alex@samba.(none)> 17175Date: Thu Jan 3 15:47:50 2008 -0500 17176 17177 RADEON: fix crash when setting rotation in the config file 17178 17179 xf86CrtcRotate() accesses pScrn->pScreen which is not set 17180 during ScreenInit(). This should also be fixed in the server. 17181 See bug 12129 17182 17183commit c652208861bffca94f06b7f67688ce220e050bfb 17184Author: Michel Dänzer <michel@tungstengraphics.com> 17185Date: Thu Jan 3 17:54:58 2008 +0100 17186 17187 radeon: Adapt manpage to reality wrt AGP options. 17188 17189commit ab451e4b7a5423d61b57cf0646599267d8504af4 17190Author: Michel Dänzer <michel@tungstengraphics.com> 17191Date: Thu Jan 3 17:52:39 2008 +0100 17192 17193 radeon: Miscellaneous warning fixes. 17194 17195commit 394c52273328e90518568b694ee79dc1a8dab651 17196Author: Dave Airlie <airlied@linux.ie> 17197Date: Thu Jan 3 18:56:16 2008 +1000 17198 17199 r500: tvout avoid doing dpms here it makes my tv mode go all crappy 17200 17201 need to investigate further 17202 17203commit 1c647279f021d01e110980727b7c7dd7efae1642 17204Author: Dave Airlie <airlied@linux.ie> 17205Date: Thu Jan 3 11:55:28 2008 +1000 17206 17207 r600: change devices list depending on connector 17208 17209commit f36db6e10d32a68b32d20ae4ad02cfc0bfd1c9c3 17210Author: Dave Airlie <airlied@linux.ie> 17211Date: Thu Jan 3 11:27:47 2008 +1000 17212 17213 r600: fixup crash on unknown output type 17214 17215 not sure this is 100% the correct approach 17216 17217commit 1accfdd590828e95e0d68a576c8ee05a06a86e43 17218Author: Alex Deucher <alex@samba.(none)> 17219Date: Wed Jan 2 19:48:28 2008 -0500 17220 17221 RADEON: various avivo tv-out related clean-ups 17222 17223commit ce34090c758ac91171cb6adb9b8a36e4adbf99cf 17224Merge: 2180f04 30cab1d 17225Author: Alex Deucher <alex@samba.(none)> 17226Date: Wed Jan 2 16:41:36 2008 -0500 17227 17228 Merge branch 'master' of git+ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 17229 17230commit 2180f04b6fb94a638f6274cb4455d5688b324dbc 17231Author: Alex Deucher <alex@samba.(none)> 17232Date: Wed Jan 2 16:41:06 2008 -0500 17233 17234 RADEON: preliminary support for ATOM RMX 17235 17236 not functional yet. 17237 17238commit dab4dc285154d40303aadaa849b85f8e251e578e 17239Author: Alex Deucher <alex@samba.(none)> 17240Date: Wed Jan 2 16:27:19 2008 -0500 17241 17242 RADEON: add support for ATOM component video table 17243 17244 Component output is still not working. 17245 17246commit 30cab1dbebb7bdb925f2fe9f91b84183312bbbfd 17247Author: Alex Deucher <alex@botch2.(none)> 17248Date: Wed Jan 2 15:41:01 2008 -0500 17249 17250 RADEON: Make sure all old IGP chips have HasSingleDac set 17251 17252 fix the csv file and re-gen the headers. 17253 17254commit e8e585651215b011e3ad07c59d0eab9107ccd8c6 17255Author: Dave Airlie <airlied@redhat.com> 17256Date: Wed Jan 2 10:14:46 2008 +1000 17257 17258 PLL/r600: tweak pll to pick first one found instead of keeping going 17259 17260commit 14aa4060ad27ecb3d40b2b17ee4cf7cc55a121dd 17261Author: Dave Airlie <airlied@redhat.com> 17262Date: Wed Jan 2 09:49:44 2008 +1000 17263 17264 r600: fix tv-out output naming 17265 17266commit f65374f5e15bfd391a1838a686cd87d3bab8043d 17267Author: Maciej Cencora <m.cencora@gmail.com> 17268Date: Mon Dec 31 09:44:34 2007 +1000 17269 17270 atombios: initial rs690 patches 17271 17272commit a674f683e6699c30664d9cd468a64de991c3fd7e 17273Author: Dave Airlie <airlied@linux.ie> 17274Date: Sun Dec 30 17:43:17 2007 +1000 17275 17276 atombios: enable TV-out detection code - tv out works for me with this 17277 17278commit aa7c28cbd943bb525698515d444cb5097880e364 17279Author: Dave Airlie <airlied@linux.ie> 17280Date: Sun Dec 30 17:40:37 2007 +1000 17281 17282 atombios: enable support for using tv timings 17283 17284 enable support for the atombios tv timings selection by programming the crtc 17285 with the tv timings if a tv is detected on the output 17286 17287commit 0bc3fd595a73e12a424571697d164a09a6a4c072 17288Author: Dave Airlie <airlied@linux.ie> 17289Date: Sun Dec 30 16:39:58 2007 +1000 17290 17291 atombios: add support for reading tv standard from atombios 17292 17293 fix typo in atombios header file 17294 17295commit bfa22d676a6f333503104041f62222f4de9bb7d8 17296Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17297Date: Wed Dec 26 12:29:47 2007 -0500 17298 17299 RADEON: fix typo that broke tv load detection 17300 17301commit 0c99554c6fab1192f5e8595469c21b5f33e1eb4f 17302Author: David Miller <davem@davemloft.net> 17303Date: Wed Dec 26 02:19:12 2007 -0500 17304 17305 [RADEON]: Add missing break in SCLK calculation. 17306 17307commit 6e0d5cc1c62fbfc1962fa0d6333f0c0a8c6327bd 17308Author: David Miller <davem@davemloft.net> 17309Date: Wed Dec 26 02:17:34 2007 -0500 17310 17311 [RADEON]: Like powerpc, don't use VGA by default on sparc. 17312 17313commit 2b1fae668ddabbc72e5fc31365302ea722174df1 17314Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17315Date: Wed Dec 26 02:04:06 2007 -0500 17316 17317 RADEON: fix PAL tv-out 17318 17319 Many thanks to Andrew Randrianasulu for providing me with 17320 pll reg dumps 17321 17322commit d736eb5732da573162c70712dc4e8b0114986702 17323Author: Alex Deucher <alex@botch2.(none)> 17324Date: Mon Dec 24 12:34:15 2007 -0500 17325 17326 RADEON: default "IgnoreLidStatus" to TRUE 17327 17328 Seems there are lots of busted ACPI lid status and people 17329 starting X with the lid closed. 17330 17331commit 653da558148cc601bc1f80253e92ef98c75ef37a 17332Author: Alex Deucher <alex@botch2.(none)> 17333Date: Mon Dec 24 01:11:56 2007 -0500 17334 17335 RADEON: restore crtc regs before VGA regs. 17336 17337 It seems some radeons don't restore text console properly if 17338 the crtc regs are restored after the VGA regs. 17339 Thanks to Sverre Froyen for helping me track this down 17340 17341commit ad3325f672a241449ca239c8ee3a24b6d7703d76 17342Author: Alex Deucher <alex@botch2.(none)> 17343Date: Sun Dec 23 17:18:42 2007 -0500 17344 17345 RADEON: Add "IgnoreLidStatus" option 17346 17347 Generally, users that start X with the laptop lid closed 17348 want to use one or more external monitors rather than the 17349 internal panel and do not want the internal panel to be on 17350 by default. Others, it seems, want to always have the 17351 internal panel on, regardless of the lid. I can't win. 17352 Enable this option to force the latter. 17353 17354commit 20eedf348a527e1e0a5450bc22d7564895034a66 17355Author: Alex Deucher <alex@botch2.(none)> 17356Date: Sun Dec 23 11:27:55 2007 -0500 17357 17358 RADEON: fix pll input setup on mac cards 17359 17360 the function was exiting before the complete setup was finished. 17361 17362commit 4f2e833e8ebaba3ad85ec5314fff8fa05395b679 17363Author: Arkadiusz Miskiewicz <arekm@maven.pl> 17364Date: Fri Dec 21 18:56:34 2007 -0500 17365 17366 configure.ac fixes 17367 17368commit 2b6e8e2b8f74e94560de89693ecbc7260536591e 17369Author: Arkadiusz Miskiewicz <arekm@maven.pl> 17370Date: Fri Dec 21 18:10:17 2007 -0500 17371 17372 RADEON: various cleanups 17373 17374commit 4c6f60e3b19ac55ab1255c79df03b1df5950864e 17375Author: Alex Deucher <alex@botch2.(none)> 17376Date: Fri Dec 21 17:33:04 2007 -0500 17377 17378 RADEON: clean up prototypes 17379 17380commit 8c761afdcb9baf1649b93449692fb9ab67bc2c80 17381Author: Alex Deucher <alex@botch2.(none)> 17382Date: Fri Dec 21 16:24:49 2007 -0500 17383 17384 RADEON: more re-org 17385 17386 move save/restore routines into legacy_crtc/output 17387 17388commit 910773d3a6c717f9d4762ea7b9ee6c3ae061781e 17389Author: Alex Deucher <alex@botch2.(none)> 17390Date: Fri Dec 21 15:53:15 2007 -0500 17391 17392 RADEON: more re-org. move XAA Mem init to radeon_accel.c 17393 17394commit 0631a23bd103f9b74e525da2c41304eab60c6f17 17395Author: Alex Deucher <alex@botch2.(none)> 17396Date: Fri Dec 21 15:40:18 2007 -0500 17397 17398 RADEON: fix rn50 reversion from last merge 17399 17400commit 5b917797a13c6caa80028d1842a284598e874288 17401Author: Alex Deucher <alex@botch2.(none)> 17402Date: Fri Dec 21 15:30:20 2007 -0500 17403 17404 RADEON: remove no longer used radeon_display.c 17405 17406commit bf14aa5f88fc3b4e69d71db5b23248b8bb2018d0 17407Author: Alex Deucher <alex@botch2.(none)> 17408Date: Fri Dec 21 15:29:47 2007 -0500 17409 17410 RADEON: re-org legacy crtc/output code into separate files 17411 17412commit be7f8fd338f5af8b632f16a83db41e15d00af469 17413Author: Dave Airlie <airlied@linux.ie> 17414Date: Fri Dec 21 10:03:49 2007 +1000 17415 17416 fix mode bandwidth configure check 17417 17418commit 3f9b597dedc45379b0bc0b631f3f924c403bca48 17419Author: Dave Airlie <airlied@linux.ie> 17420Date: Fri Dec 21 09:55:42 2007 +1000 17421 17422 fixup clip test include handling 17423 17424commit 9a5b501332c0a1f10af20845af48c9ddd2ce26a0 17425Author: Dave Airlie <airlied@linux.ie> 17426Date: Fri Dec 21 09:45:55 2007 +1000 17427 17428 set ddc line correctly post-merge 17429 17430commit 3c31b96afa20913ad947e68fe0c3a662e5eafbdd 17431Merge: eb99c3c f5e8c18 17432Author: Dave Airlie <airlied@linux.ie> 17433Date: Fri Dec 21 09:36:22 2007 +1000 17434 17435 Merge remote branch 'origin/atombios-support' 17436 17437 Conflicts: 17438 17439 src/radeon_display.c 17440 src/radeon_driver.c 17441 17442commit eb99c3c5c9a2249cb84920f0f225e525fc3a4144 17443Author: Alex Deucher <alex@botch2.(none)> 17444Date: Thu Dec 20 18:14:38 2007 -0500 17445 17446 Bump for RC release 17447 17448commit 8d49ff1da917b7f8240267953ef6ce4ff04daecb 17449Author: Alex Deucher <alex@botch2.(none)> 17450Date: Thu Dec 20 01:09:57 2007 -0500 17451 17452 RADEON: check for xf86_crtc_clip_video_helper() in configure.ac 17453 17454 use xf86_crtc_clip_video_helper() from the server if available. 17455 17456commit fb7a4e24f2da3561ef81371ca4013a4f13806e91 17457Author: Adam Jackson <ajax@redhat.com> 17458Date: Wed Dec 19 19:15:19 2007 -0500 17459 17460 Fix RN50 mode filtering. 17461 17462 The old code would attempt to limit the maximum pixel size of the screen 17463 by limiting the maximum PLL frequency. This ends up confusing the PLL 17464 computation code since sometimes your maximum freq can be lower than your 17465 minimum freq. More to the point it's just wrong, maximum PLL frequency 17466 isn't the same thing as maximum pixel clock, and even that isn't the same 17467 thing as maximum scanout pixels per second. 17468 17469 The correct thing to do is filter by the mode's effective memory bandwidth. 17470 17471commit c1b0b69cc50516c3b7e881b0eb46cb3cd2e9dce6 17472Merge: 0e66348 ce4fa1c 17473Author: Alex Deucher <alex@botch2.(none)> 17474Date: Wed Dec 19 11:54:50 2007 -0500 17475 17476 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati 17477 17478commit 0e6634870d1ab38ee8c83f6bda1ba60364997853 17479Author: Alex Deucher <alex@botch2.(none)> 17480Date: Wed Dec 19 11:54:27 2007 -0500 17481 17482 RADEON: skip empty connectors when creating outputs 17483 17484commit 6afbf718c151dc3c5c59bd3136b58a93a114d798 17485Author: Alex Deucher <alex@botch2.(none)> 17486Date: Wed Dec 19 11:48:38 2007 -0500 17487 17488 RADEON: add support for legacy radeons with DVI and no connector table 17489 17490commit f5e8c185001e62e744310667c2d1bd3fe6542a62 17491Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 17492Date: Wed Dec 19 10:38:58 2007 +1000 17493 17494 more endian related fixage 17495 17496commit 98b247066d00db66abe91f518cd93b5c4da4cfb4 17497Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 17498Date: Wed Dec 19 10:25:41 2007 +1000 17499 17500 fix big endian build since zaphod fixups 17501 17502commit ce4fa1cedec0cf56b9979dfaa12a8d3a7c643df4 17503Author: Arkadiusz Miskiewicz <arekm@maven.pl> 17504Date: Tue Dec 18 15:34:14 2007 -0500 17505 17506 RADEON: fix fd leak in lid detect code 17507 17508commit 20568f66f9a9a60a33bd9a69ccc14a891c656836 17509Author: Arkadiusz Miskiewicz <arekm@maven.pl> 17510Date: Tue Dec 18 15:32:10 2007 -0500 17511 17512 RADEON: more cleanups and warning fixes 17513 17514commit 1496194200adbcb044ec3977367a0908262e389c 17515Author: Arkadiusz Miskiewicz <arekm@maven.pl> 17516Date: Tue Dec 18 15:29:53 2007 -0500 17517 17518 RADEON: driver cleanups, warning fixes 17519 17520commit bd7206fa120495037e3fea0c920d0031d7715bf6 17521Author: Alex Deucher <alex@botch2.(none)> 17522Date: Tue Dec 18 03:03:11 2007 -0500 17523 17524 RADEON: fix another merge error 17525 17526 this broken legacy radeons 17527 17528commit 65a3ac7530e11bb7d818a988fd0cf1dde7688fa4 17529Author: Alex Deucher <alex@samba.(none)> 17530Date: Tue Dec 18 00:15:38 2007 -0500 17531 17532 RADEON: more PLL tweaks 17533 17534commit d93a0e10b8bc6e3797a3cf6c1e28ca413a7c38e4 17535Author: Alex Deucher <alex@samba.(none)> 17536Date: Mon Dec 17 20:32:45 2007 -0500 17537 17538 RADEON: post div tweaks for legacy radeon 17539 17540commit 03b8b49f6f502c45552b018fd8c44d366b2d576f 17541Author: Alex Deucher <alex@samba.(none)> 17542Date: Mon Dec 17 20:20:04 2007 -0500 17543 17544 RADEON: fix typo from merge 17545 17546commit 19b9d3708852b7efe2b05249c8359dadb924dd94 17547Merge: cf685f3 29706ca 17548Author: Alex Deucher <alex@samba.(none)> 17549Date: Mon Dec 17 20:07:32 2007 -0500 17550 17551 Merge branch 'atombios-support' of git+ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 17552 17553commit cf685f37ec874f0aacd09e7c4eb0402c6daec1b0 17554Merge: 2a134af 44d07c4 17555Author: Alex Deucher <alex@samba.(none)> 17556Date: Mon Dec 17 20:07:07 2007 -0500 17557 17558 Merge branch 'master' of git+ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 17559 17560 merge master and fix conflicts 17561 17562commit 29706ca585ebd9b7b3521521a17016322e9ecccb 17563Author: Dave Airlie <airlied@linux.ie> 17564Date: Tue Dec 18 10:55:38 2007 +1000 17565 17566 fixup shadow setup on !r600 17567 17568commit 44d07c4ccce9acb5bd21a17acb082e91f7225764 17569Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17570Date: Mon Dec 17 18:56:12 2007 -0500 17571 17572 RADEON: typo from last commit 17573 17574commit 4da3782239921eb377216d4de4a9cc5bb55e0e8a 17575Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17576Date: Mon Dec 17 18:51:31 2007 -0500 17577 17578 RADEON: add output enable masks 17579 17580 add output enable masks for outputs that drive 17581 more than one connector. Make sure we don't turn off 17582 an output that's driving another connector. 17583 17584commit 5c5d2d19b2b032a06dd333b4ecc029aac342fb93 17585Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17586Date: Mon Dec 17 18:15:55 2007 -0500 17587 17588 RADEON: whitespace clean-ups 17589 17590commit 9f1d8220315c8894a17f2cc328025dc682b0c6e0 17591Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17592Date: Mon Dec 17 18:04:05 2007 -0500 17593 17594 RADEON: more PLL fixes 17595 17596 - reduce the calculation accuracy 17597 - certain LVDS panels seem to only like certain ref_divs 17598 - add pll flags to handle special cases 17599 - adjust the pll limits on legacy cards 17600 17601commit 2a134af01bc85de758ad83a85b23d1bba4a1a0f5 17602Author: Dave Airlie <airlied@redhat.com> 17603Date: Mon Dec 17 15:00:36 2007 +1000 17604 17605 r600: add shadow support to r600 driver to at least make 2d useable 17606 17607commit 614414611a9f246cbc74f579a79987fff97cf571 17608Author: Dave Airlie <airlied@redhat.com> 17609Date: Mon Dec 17 11:10:14 2007 +1000 17610 17611 radeon: cleanup pitch calculation and make r600 work again 17612 17613commit 79a375dbc7f323e2f551490a35f44ec36bed877c 17614Author: George Wu <geo@flood.OCF.Berkeley.EDU> 17615Date: Mon Dec 17 10:55:36 2007 +1000 17616 17617 r600: might as well fix VT for R600 17618 17619commit bc213ee723a45f2c396b4ed211a50f7642349973 17620Author: Alex Deucher <alex@samba.(none)> 17621Date: Sun Dec 16 14:54:00 2007 -0500 17622 17623 RADEON: fix sclock printout 17624 17625commit 4747c1f3cd4167b6a51d4864a297719ea48b9346 17626Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17627Date: Sun Dec 16 14:07:29 2007 -0500 17628 17629 RADEON: Make sure LVDS_EN bit is set when enabling LVDS 17630 17631commit a9817b2cf436a536dbc43ad77abc3bdcc53d346d 17632Author: Alex Deucher <alex@samba.(none)> 17633Date: Sat Dec 15 20:51:53 2007 -0500 17634 17635 RADEON: clean up units in PLL calculation 17636 17637commit b653e5a628bfa4dfb168e96f93f41eb910f409fb 17638Author: Michel Dänzer <michel@tungstengraphics.com> 17639Date: Sat Dec 15 00:50:10 2007 +0100 17640 17641 radeon: Default to 1x again with non-v3 AGP cards. 17642 17643 Seems more reliable in general than what was set up by firmware - fingers 17644 crossed... 17645 17646commit 6229825fa5d6715569098afbb21a40f7a2e7e6be 17647Author: Michel Dänzer <michel@tungstengraphics.com> 17648Date: Sat Dec 15 00:48:26 2007 +0100 17649 17650 radeon: Warning fixes. 17651 17652 The lid detection code probably wouldn't work on other non-x86 platforms 17653 though... 17654 17655commit 818ccf0fd4b5879171c5f20526d5a58638f8fde5 17656Author: Fredrik Höglund <fredrik@kde.org> 17657Date: Fri Dec 14 23:56:12 2007 +0100 17658 17659 RADEON: Fix the vertex coordinates for transformed pictures 17660 17661 This partially fixes transformed pictures on R100/R200 based 17662 cards. The texture still doesn't appear to be clamped correctly, 17663 but since that doesn't matter for rotations at perpendicular 17664 angles, I'm committing this now so randr rotation and reflection 17665 will work properly. 17666 17667commit 3cfbcf4cafbdfdb33411d16e51fb1f77cd0f93dd 17668Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17669Date: Fri Dec 14 17:11:00 2007 -0500 17670 17671 RADEON: Fix PLL set up on certain notebooks 17672 17673 Some LVDS panels require specific PLL dividers as 17674 specified in the bios tables. Make sure to use them 17675 if the output is LVDS. 17676 17677commit a84d446fd301d456bcea8f7abdc52e5a30776412 17678Author: Alex Deucher <alex@botch2.(none)> 17679Date: Fri Dec 14 02:17:14 2007 -0500 17680 17681 RADEON: select fb_div0 for LVDS on RV410 (x700) mobility 17682 17683 Fixes bug 8038 17684 I wonder if desktop RV410 need a similar fix. 17685 If your x700 laptop panel has problems after this let me know. 17686 17687commit b3eed3d87f76779b5a62a3115f99a31484dc38e0 17688Author: Alex Deucher <alex@samba.(none)> 17689Date: Fri Dec 14 00:20:10 2007 -0500 17690 17691 RADEON: fix typo in previous cursor fix 17692 17693commit 814c6c48aebba2e45ce257289b922cd7e92caf2a 17694Author: Alex Deucher <alex@samba.(none)> 17695Date: Thu Dec 13 18:45:09 2007 -0500 17696 17697 RADEON: rework PLL calculation 17698 17699 - Take into account the limits from the bios tables 17700 - Unify the PLL calculation between legacy and avivo chips 17701 17702commit f5ac34983411e4c4f41ab1817dce582830f398fd 17703Merge: f2b2e08 6ccf5b3 17704Author: Alex Deucher <alex@samba.(none)> 17705Date: Wed Dec 12 22:37:44 2007 -0500 17706 17707 Merge branch 'master' of git+ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 17708 17709 merge and fix conflicts 17710 17711commit f2b2e0804183b52d9d3f56ad85b3552ece76c544 17712Author: Alex Deucher <alex@samba.(none)> 17713Date: Wed Dec 12 22:18:37 2007 -0500 17714 17715 RADEON: fix rotation on avivo chips 17716 17717 There are still some issues, but this is better than before. 17718 17719commit 6ccf5b33d27218ae1c45ab93c122438ed536d8ba 17720Author: Alex Deucher <alex@botch2.(none)> 17721Date: Wed Dec 12 20:12:06 2007 -0500 17722 17723 RADEON: only enable vblanks if we want them 17724 17725 should fix bug 13610 17726 17727commit 1668f2056f56370f1b5681c13f1e14904e301216 17728Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17729Date: Wed Dec 12 19:39:08 2007 -0500 17730 17731 RADEON: use /proc/acpi to determine lid status 17732 17733 Linux only 17734 17735commit 33a39947f7f79533cd90007a17d57b20126642c6 17736Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17737Date: Wed Dec 12 18:50:18 2007 -0500 17738 17739 RADEON: fix cursors when using rotation 17740 17741 allocate separate cursor buffers for each crtc 17742 17743commit 9e5efdecd12092031a4aebce58747cb4a6f48f28 17744Author: Arkadiusz Miskiewicz <arekm@maven.pl> 17745Date: Tue Dec 11 23:53:03 2007 -0500 17746 17747 sparse fixes and cleanups from arekm 17748 17749commit 372bf41818fdafc6a9d2914aee3a8e359f668f02 17750Author: Alex Deucher <alex@botch2.(none)> 17751Date: Tue Dec 11 14:04:58 2007 -0500 17752 17753 RADEON: handle HMDI properly (untested) and fix some merge leftovers 17754 17755commit 3c22ad977c25d5ca2811821fcac6bb8ecd79994a 17756Merge: c9a0cee f3d2ec3 17757Author: Alex Deucher <alex@botch2.(none)> 17758Date: Tue Dec 11 13:11:15 2007 -0500 17759 17760 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 17761 17762 fix conflicts 17763 17764commit f3d2ec3a5ae61215c792018320158750e7aa937c 17765Author: Alex Deucher <alex@botch2.(none)> 17766Date: Tue Dec 11 11:57:27 2007 -0500 17767 17768 RADEON: rewrite PLL computation 17769 17770 Algorithm adapted from BeOS radeon driver with some tweaks by me. 17771 Actually calulate and use the reference divider rather than using the bios default. 17772 Also, always calculate the PLL, rather than falling back to bios dividers. 17773 This should fix bugs 12913, 13590, 13533, possibly others. 17774 17775commit c9a0cee97ca69e8fe1e1937c7670fa903214cded 17776Author: Dave Airlie <airlied@linux.ie> 17777Date: Tue Dec 11 06:03:46 2007 +1000 17778 17779 more zaphod fixes - some other work maybe needed 17780 17781commit 9b125312ab6edc585e4f5931a6a6de81e13b6acc 17782Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17783Date: Mon Dec 10 13:53:15 2007 -0500 17784 17785 RADEON: only update crtc values when RMX is active 17786 17787commit 3a161e1b5d80361e318ced8da5c19e797749d693 17788Author: Alex Deucher <alex@botch2.(none)> 17789Date: Mon Dec 10 00:57:26 2007 -0500 17790 17791 RADEON: bios PLL cleanup 17792 17793commit 731830297f2fc4a416882aacfb0d9b5f8ed32520 17794Author: Dave Airlie <airlied@linux.ie> 17795Date: Mon Dec 10 15:50:38 2007 +1000 17796 17797 fixup some warnings 17798 17799commit 2818e2b02ca90c9dfa50905b5311b2ae83ac3b0c 17800Author: Dave Airlie <airlied@linux.ie> 17801Date: Mon Dec 10 15:43:52 2007 +1000 17802 17803 add more to configure.ac for using out-of-tree mode src 17804 17805commit cc3c36100986f9d8060bc2d433373d4806f8e730 17806Author: Dave Airlie <airlied@linux.ie> 17807Date: Mon Dec 10 15:25:56 2007 +1000 17808 17809 add support for building against legacy servers similiar to Intel codepaths 17810 17811commit 9c278cb7fa7f18d13bde053fd75221cfba9da377 17812Merge: 6451ea2 cc167b9 17813Author: Dave Airlie <airlied@linux.ie> 17814Date: Mon Dec 10 15:18:03 2007 +1000 17815 17816 Merge branch 'zaphod-lolz' of git://git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 17817 17818 Conflicts: 17819 17820 src/radeon.h 17821 src/radeon_crtc.c 17822 src/radeon_driver.c 17823 src/radeon_output.c 17824 17825commit 6451ea2dcc4fac762442f699935864f4a8d445f7 17826Merge: 0d89556 5896ca4 17827Author: Dave Airlie <airlied@linux.ie> 17828Date: Mon Dec 10 15:08:42 2007 +1000 17829 17830 Merge branch 'master' into atombios-support 17831 17832commit 5896ca4097d439f59f90f397939132c061c3c364 17833Author: LisaWu <liswu@ati.com> 17834Date: Fri Dec 7 09:45:05 2007 +0100 17835 17836 radeon: Use %u instead of %d for unsigned value. 17837 17838commit df44f8380268c27d3978c4e91d736f093322b8b8 17839Author: Michel Dänzer <michel@tungstengraphics.com> 17840Date: Fri Dec 7 09:41:47 2007 +0100 17841 17842 radeon: Use gettimeofday instead of xf86getsecs. 17843 17844commit cc167b9bb7f1c3b8579e51e7bc2fca2f8eba6bd1 17845Author: Dave Airlie <airlied@redhat.com> 17846Date: Fri Dec 7 15:41:36 2007 +1000 17847 17848 disable tiling for zaphod heads 17849 17850commit 2ce8d192533a8c795714c5a9fb308ec74db40287 17851Author: Dave Airlie <airlied@redhat.com> 17852Date: Fri Dec 7 15:35:21 2007 +1000 17853 17854 don't add fboffset to info->FB it already is mapped at the offset 17855 17856commit 0dcd926d3092100854b3e362d6659d4950508aeb 17857Author: Dave Airlie <airlied@redhat.com> 17858Date: Fri Dec 7 14:45:04 2007 +1000 17859 17860 radeon: bring back zaphod all is forgiven. 17861 17862 You've whined, you've cried, you've nagged, and you're guilt trippin has 17863 made me do it... It actually wasn't as hard as I thought it would be. 17864 17865 Still not perfect, couple of things to fix yet 17866 17867commit bb5ede557bf32a42eef158ff0fbcfe1c6ede098a 17868Author: Dave Airlie <airlied@redhat.com> 17869Date: Fri Dec 7 14:30:32 2007 +1000 17870 17871 radeon: move savedreg/modereg into entity instead of info 17872 17873commit 64ab1cdf343a9a69e7e9e64f0bba77c54a94e9d0 17874Author: James Cloos <cloos@jhcloos.com> 17875Date: Thu Dec 6 15:51:12 2007 -0500 17876 17877 Add missing PHONY line for automatic ChangeLog generation 17878 17879commit 0d89556bfa41a3acbd6afe85b062e3a21f2ca057 17880Author: Dave Airlie <airlied@ppcg5.localdomain> 17881Date: Thu Dec 6 19:23:06 2007 +1100 17882 17883 powerpc: build fixes from last merge 17884 17885commit dbe3d2608ecc9896db9c23b3a347b50748c51e13 17886Merge: 48e31cd 21ed435 17887Author: Dave Airlie <airlied@redhat.com> 17888Date: Thu Dec 6 14:22:03 2007 +1000 17889 17890 Merge branch 'master' into atombios-support 17891 17892 Conflicts: 17893 17894 src/radeon_output.c 17895 17896commit 21ed435398e4a398dd8a0a5d7c1d4cc45e916332 17897Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17898Date: Tue Dec 4 17:08:58 2007 -0500 17899 17900 RADEON: add MacModel imac-g5-isight for iMac G5 iSight 17901 17902 Thanks to Étienne Bersac for helping to figure this out. 17903 17904commit 54bfd522405d9fdfb69d3a59e111ac3d63483dbb 17905Author: Étienne Bersac <bersace03@laposte.net> 17906Date: Tue Dec 4 14:22:42 2007 -0500 17907 17908 RADEON: fix typo 17909 17910commit 5022d006cfc06ca0395981526b2c2c94c6878567 17911Author: Michel Dänzer <michel@tungstengraphics.com> 17912Date: Sun Dec 2 17:27:33 2007 +0100 17913 17914 radeon: Further XVideo fixes. 17915 17916 * Make sure pitch constraints are always met for DMA upload blits. 17917 * RGB24 is not affected by endianness. 17918 17919commit 6ed55b70b23dfdc7b41103ea59c1df2bda5e41e6 17920Author: Kusanagi Kouichi <slash@ma.neweb.ne.jp> 17921Date: Sun Dec 2 17:18:46 2007 +0100 17922 17923 radeon: Fix crash with XVideo 24bit RGB images. 17924 17925 See https://bugs.freedesktop.org/show_bug.cgi?id=13274 . 17926 17927commit a697b590899bb7704ec4d7ae9a9c3cbbfcaef382 17928Author: Michel Dänzer <michel@tungstengraphics.com> 17929Date: Sun Dec 2 17:11:20 2007 +0100 17930 17931 Fix build against xserver master. 17932 17933 (DE)ALLOCATE_LOCAL are gone. 17934 17935commit 48e31cdaa0caa21573879af5b9267773fe89176a 17936Author: George Wu <geo@ocf.berkeley.edu> 17937Date: Sun Dec 2 15:25:09 2007 +1000 17938 17939 RADEON/R600: small code cleanup 17940 17941commit 00b4480aa2c5d7f751e34fc964f431b90b14c8d2 17942Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 17943Date: Sat Dec 1 14:18:40 2007 -0500 17944 17945 RADEON: add options for force TV out as detected and to set TV standard 17946 17947 Also fix a typo in internal tv-out parsing 17948 17949commit 1e029fef5fe264f2ced445b80bf6070abcb84b82 17950Author: Alex Deucher <alex@samba.(none)> 17951Date: Sat Dec 1 00:58:51 2007 -0500 17952 17953 RADEON: move GPIO lookup to a separate function 17954 17955commit dcbef1ba9dfcf35c28e058832a55adf00afb472e 17956Author: Alex Deucher <alex@samba.(none)> 17957Date: Sat Dec 1 00:35:25 2007 -0500 17958 17959 RADEON: fix typo in previous commit 17960 17961 check gpio table revision before connector table revision 17962 17963commit fdce0598a2228c48c84deae1d7bebb2d7b3e979b 17964Author: Alex Deucher <alex@samba.(none)> 17965Date: Sat Dec 1 00:15:34 2007 -0500 17966 17967 RADEON: convert atombios connector table parsing to use ATOM structs 17968 17969 convert and add hpd info 17970 17971commit e3d7de9cc956aec5f940ad6db09e826b3a69523a 17972Author: Alex Deucher <alex@botch2.(none)> 17973Date: Fri Nov 30 20:14:42 2007 -0500 17974 17975 RADEON: remove unused cruft from last atom import 17976 17977commit d5d83411e8a884154d671aad440524507cce313e 17978Author: Alex Deucher <alex@botch2.(none)> 17979Date: Fri Nov 30 20:11:42 2007 -0500 17980 17981 RADEON: save/restore avivo crtc cursor control 17982 17983 this should prevent the cursor from showing up on in text 17984 mode or vesafb etc. after running the driver. 17985 17986commit af0196f7bf0d1d5d211391149c18935d64ed2b06 17987Merge: d9858a2 0175b79 17988Author: Alex Deucher <alex@botch2.(none)> 17989Date: Fri Nov 30 16:40:28 2007 -0500 17990 17991 Merge branch 'master' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 17992 17993 merge master and fix conflicts 17994 17995commit d9858a2b3744b99003cfb9f31b743a2d31b322e9 17996Author: Dave Airlie <airlied@linux.ie> 17997Date: Sat Dec 1 06:49:59 2007 +1000 17998 17999 radeon: add in pll spread spectrum workaround 18000 18001commit 0175b79987ef4d7b0ce8238c3bdde989e504516a 18002Author: Alex Deucher <alex@botch2.(none)> 18003Date: Fri Nov 30 15:37:42 2007 -0500 18004 18005 RADEON: rework MacModel option 18006 18007 this brings in some previous research from Michel Dänzer, 18008 Sjoerd Simons, and myself. Hopefully, the driver will pick 18009 the correct MacModel in more cases. This also changes the 18010 default connector table for desktop Macs to dual DVI rather 18011 than DVI+VGA as that seems to be the case more often than not. 18012 External TMDS chips are handled separately now as well. 18013 Eventually we should add an option to allow the user to specify 18014 what external TMDS chip they need, but we don't have enough info 18015 yet, so we'll rely on OF to init the external chip in most cases 18016 for now. 18017 18018commit e1945f1f25a34310bd58ce128c8ff27ecc985618 18019Merge: b368b0f df7777b 18020Author: Alex Deucher <alex@botch2.(none)> 18021Date: Fri Nov 30 14:30:55 2007 -0500 18022 18023 Merge branch 'atombios-support' of ssh://agd5f@git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 18024 18025commit b368b0f22cd1d7ef9b4c65d82929c76f3b82d573 18026Author: Alex Deucher <alex@botch2.(none)> 18027Date: Fri Nov 30 14:29:27 2007 -0500 18028 18029 RADEON: disable atom pll set for r4xx cards 18030 18031 the clocks do not get set correctly in all cases. this needs 18032 further investigation. 18033 18034commit 5af15739571c09260750bcfd3620e16fd7fec862 18035Author: Alex Deucher <alex@botch2.(none)> 18036Date: Fri Nov 30 14:24:30 2007 -0500 18037 18038 RADEON: small cleanup of pll code 18039 18040commit df7777bff40c1feabcc12d2148ad6ac5213efbb3 18041Author: George Wu <geo@ocf.berkeley.edu> 18042Date: Fri Nov 30 17:49:33 2007 +1000 18043 18044 Add LVTMA PWRSEQ registers to fix VT switching for LVDS 18045 18046commit 9840a0fd4fc8c980533fcd4a02c55cd0d5634b6d 18047Author: Alex Deucher <alex@botch2.(none)> 18048Date: Thu Nov 29 13:27:37 2007 -0500 18049 18050 RADEON: add MacModel "mini-internal" for minis with internal TMDS 18051 18052 Some macs (minis and powerbooks) use internal tmds, others use external tmds 18053 and not just for dual-link TMDS, it shows up with single-link as well. 18054 Unforunately, there doesn't seem to be any good way to figure it out. 18055 18056commit 017c939cf0a2b12fbdc1681cc70c28b23ae3b397 18057Author: Alex Deucher <alex@samba.(none)> 18058Date: Thu Nov 29 02:52:14 2007 -0500 18059 18060 RADEON: implement CLUT adjust support 18061 18062commit 9963b0fe01feb6dd0cb555b874a48f6fa3b255cb 18063Author: Alex Deucher <alex@samba.(none)> 18064Date: Thu Nov 29 00:46:23 2007 -0500 18065 18066 RADEON: fix cursor offset on avivo chips 18067 18068commit 6c56e3d7655b17e93e8823aefe34b05291104695 18069Author: Alex Deucher <alex@botch2.(none)> 18070Date: Tue Nov 27 15:27:36 2007 -0500 18071 18072 RADEON: switch r4xx to atombios load detection 18073 18074 works great 18075 18076commit 7561242e5b79bc2798ca3aace2b79e1a36949488 18077Author: Alex Deucher <alex@botch2.(none)> 18078Date: Tue Nov 27 14:50:36 2007 -0500 18079 18080 RADEON: re-org load detection for legacy chips 18081 18082commit febdcc2dccd42acbcd68ae630b7811cae5c58e8a 18083Author: Dave Airlie <airlied@linux.ie> 18084Date: Wed Nov 28 05:10:57 2007 +1000 18085 18086 legacy: fix fb/agp read/writes 18087 18088commit bb8545146959b748994be055d5b3de66ec66c8b2 18089Author: Alex Deucher <alex@samba.(none)> 18090Date: Mon Nov 26 17:34:51 2007 -0500 18091 18092 RADEON: first pass at TV/Component video 18093 18094 Untested and not likely to work just yet. 18095 18096commit e2bde646b864dca9056d9ecfe23a0d905647ea9a 18097Author: Alex Deucher <alex@samba.(none)> 18098Date: Mon Nov 26 14:35:57 2007 -0500 18099 18100 RADEON: move crtc output source selection into atombios_output.c 18101 18102 The function fits better as an output function and should now 18103 work with clones as well. 18104 18105commit 16c9be4107678a2a58d3418b7f1cc94d695ca8d6 18106Author: Alex Deucher <alex@samba.(none)> 18107Date: Mon Nov 26 14:20:54 2007 -0500 18108 18109 RADEON: add default connector table for avivo chips 18110 18111commit 6f080d00e6f4f84d5e0d6b4eff302bf42c230e81 18112Author: Arkadiusz Miskiewicz <arekm@maven.pl> 18113Date: Mon Nov 26 12:43:30 2007 -0500 18114 18115 RADEON: fix backlight control on some laptops 18116 18117 It seems the bios scratch regs are involved in backlight control 18118 on some laptops. This patch fixes the problematic laptops and doesn't 18119 seem to break the previous bios lid and output control fixes. 18120 18121commit dcf22aed87366f4625fb5042cb84fecccd9ceece 18122Author: Alex Deucher <alex@botch2.(none)> 18123Date: Mon Nov 26 11:10:03 2007 -0500 18124 18125 RADEON: only return status unknown for XPRESS chips 18126 18127 this seems to cause more issues than it attempted to fix 18128 so limit it to XPRESS chips for now. 18129 18130commit 206e280f02324641b4fe5a1986e26adf0e021fd4 18131Author: Alex Deucher <alex@botch2.(none)> 18132Date: Mon Nov 26 09:39:27 2007 -0500 18133 18134 RADEON: fix typo in man page 18135 18136commit 4e792db655dc92d2864db36b4d8f6714908de8e8 18137Author: Dave Airlie <airlied@redhat.com> 18138Date: Fri Nov 23 15:44:44 2007 +1000 18139 18140 r500: set default minimum pll freq 18141 18142commit a13b4c69c105c096dd05e6de2c5c154c9a8bcc71 18143Author: Dave Airlie <airlied@redhat.com> 18144Date: Fri Nov 23 15:25:06 2007 +1000 18145 18146 r5xx: cleanup pll code.. 18147 18148 Clean the PLL code to use the radeon pll structs. 18149 18150commit 5d792b5673bbf4784eb0ec059221e9b57232a122 18151Author: Dave Airlie <airlied@redhat.com> 18152Date: Fri Nov 23 15:03:13 2007 +1000 18153 18154 radeon: fix up memory mapping issues for vt switch 18155 18156commit 558a2ef266c1ca517c7fb464b0ccfef83238c913 18157Author: Dave Airlie <airlied@redhat.com> 18158Date: Fri Nov 23 14:39:32 2007 +1000 18159 18160 fix silly spaces 18161 18162commit 1bda4424a7031de437acfca9c41d4b3668e36051 18163Author: Dave Airlie <airlied@redhat.com> 18164Date: Fri Nov 23 14:39:19 2007 +1000 18165 18166 r600: add hi agp address for mc 18167 18168commit 133234c31a294f24a3968a576aad2bb8b89d0f6a 18169Author: Dave Airlie <airlied@redhat.com> 18170Date: Fri Nov 23 14:15:18 2007 +1000 18171 18172 atombios: use values from object header 18173 18174commit dbf6eae7e7a4bd1bc60fefdc7ab6276ed3f097c4 18175Author: Dave Airlie <airlied@redhat.com> 18176Date: Fri Nov 23 11:55:05 2007 +1000 18177 18178 atombios: add initial object header parsing for r600 cards 18179 18180commit 6b103915c11fc79d2efc43c44fc2a00c3bc64ede 18181Author: Dave Airlie <airlied@redhat.com> 18182Date: Fri Nov 23 09:24:20 2007 +1000 18183 18184 r500: make vt switch work again for me 18185 18186commit d24208276aad7669feeed527dced60c76d39eae6 18187Author: Dave Airlie <airlied@redhat.com> 18188Date: Fri Nov 23 08:50:25 2007 +1000 18189 18190 avivo: fix typo in register saving 18191 18192commit 197a62704742a4a19736c2637ac92d1dc5ab34ed 18193Author: Adam Jackson <ajackson@redhat.com> 18194Date: Thu Nov 22 20:26:23 2007 +1000 18195 18196 radeon: fix openoffice/render bug on r100 chips 18197 18198commit 64010fc4eae8359c01e430f64252931242c91435 18199Author: Dave Airlie <airlied@linux.ie> 18200Date: Thu Nov 22 20:25:31 2007 +1000 18201 18202 Revert "Disable RENDER acceleration by default on some RV200 chips." 18203 18204 This reverts commit 145da701bf4fb9c0ad9f95620b20914ae0126852. 18205 18206 pull in fix from ajax next commit 18207 18208commit 145da701bf4fb9c0ad9f95620b20914ae0126852 18209Author: Stefan Dirsch <sndirsch@suse.de> 18210Date: Thu Nov 22 08:38:09 2007 +0100 18211 18212 Disable RENDER acceleration by default on some RV200 chips. 18213 18214 Novell Bug #341699: Render acceleration is known to be broken 18215 on at least "Radeon 7500 QW" and "Radeon Mobility M7 LW". 18216 18217commit e810c3ae9908cd57e95b1b091cded87cbfc12fdc 18218Author: Roland Scheidegger <sroland@tungstengraphics.com> 18219Date: Thu Nov 22 02:37:55 2007 +0100 18220 18221 really do not set up surface regs for depth buf on r100-class igps (bug #13080) 18222 18223 fix the if condition testing for these chips... 18224 18225commit 0cc7e94849c1525750fabd04cf58f2dee88372e0 18226Author: Alex Deucher <alex@botch2.(none)> 18227Date: Wed Nov 21 17:06:17 2007 -0500 18228 18229 RADEON: reorder crtc/pll setup 18230 18231commit d56bde98efceaa8344e62f8e98db90c4bb642331 18232Author: Alex Deucher <alex@botch2.(none)> 18233Date: Wed Nov 21 17:03:39 2007 -0500 18234 18235 RADEON: fix crtc to output routing 18236 18237 Thanks to AMD for the information 18238 18239commit a12390c832abe423def60e39cd5a9118e5910339 18240Merge: d531792 e74dca1 18241Author: Alex Deucher <alex@botch2.(none)> 18242Date: Wed Nov 21 02:24:48 2007 -0500 18243 18244 Merge branch 'atombios-support' of git://git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 18245 18246commit d5317922f29a57b6c4127826a2fc126c5fd7c117 18247Author: Alex Deucher <alex@botch2.(none)> 18248Date: Wed Nov 21 02:23:37 2007 -0500 18249 18250 RADEON: attempt to fix crtc to output routing 18251 18252 The output routing seems to be based on the output ids from the bios 18253 connector tables and the connected status in the bios scratch regs. 18254 I don't fully understand this yet, but this seems to work 18255 for the most part, however changing modes can sometimes 18256 lead to a blanked head. This can be remedied by forcing 18257 a dpms off cycle. 18258 18259commit 81ce299bffd75540925b4c8234adf11226147165 18260Author: Alex Deucher <alex@botch2.(none)> 18261Date: Wed Nov 21 01:35:44 2007 -0500 18262 18263 RADEON: provide clearer debugging info 18264 18265commit e74dca19416b13f97db9d1fc06299b988057d6a4 18266Author: Dave Airlie <airlied@redhat.com> 18267Date: Wed Nov 21 16:24:25 2007 +1000 18268 18269 re-enable mobility chips 18270 18271commit 9c5b813dd6b3492cbc9833bc59792a5cec457e51 18272Author: Alex Deucher <alex@botch2.(none)> 18273Date: Wed Nov 21 01:22:42 2007 -0500 18274 18275 RADEON: major re-org and clean up of atom output control 18276 18277 - use radeon_output->devices to determine output 18278 - clean up and simplify dpms and mode set 18279 18280commit 7634cb6b96f938bc6615eb2c49ae75aaefd04cce 18281Author: Alex Deucher <alex@botch2.(none)> 18282Date: Wed Nov 21 00:10:14 2007 -0500 18283 18284 RADEON: make naming consistent and remove some cruft 18285 18286commit 908748343fc9a6cdc38af0fc028c63a82766da3f 18287Author: Alex Deucher <alex@botch2.(none)> 18288Date: Wed Nov 21 00:05:42 2007 -0500 18289 18290 RADEON: store devices ids from bios for each driver output 18291 18292commit 3975da2ea8cb628f7f66c3f26c5dfa181cd1c532 18293Merge: e283aa3 295ce27 18294Author: Alex Deucher <alex@botch2.(none)> 18295Date: Tue Nov 20 23:52:29 2007 -0500 18296 18297 Merge branch 'atombios-support' of git://git.freedesktop.org/git/xorg/driver/xf86-video-ati into atombios-support 18298 18299commit e283aa332adf0134243a4fa3d14263719cd8a3fd 18300Author: Alex Deucher <alex@botch2.(none)> 18301Date: Tue Nov 20 23:35:46 2007 -0500 18302 18303 RADEON: add LVDS atom support 18304 18305commit e4bc3e1e7bb45571367d41b5328ff2590810b0f9 18306Author: Alex Deucher <alex@botch2.(none)> 18307Date: Tue Nov 20 18:09:29 2007 -0500 18308 18309 RADEON: enable/disable the right TMDS controller 18310 18311commit 3e47683ffaa44a89cda9bcddf530643befb27efa 18312Author: Alex Deucher <alex@botch2.(none)> 18313Date: Tue Nov 20 18:01:15 2007 -0500 18314 18315 RADEON: fixup for bios tables with wrong connector types 18316 18317commit 7412952eb1d1e9857cdab8417f7305f676900827 18318Author: Alex Deucher <alex@botch2.(none)> 18319Date: Tue Nov 20 18:00:12 2007 -0500 18320 18321 RADEON: switch to using ATOM defines for bios device table 18322 18323commit 295ce277bb0a44b9539b3dba575e7aff279dc2d0 18324Author: Dave Airlie <airlied@redhat.com> 18325Date: Wed Nov 21 08:56:40 2007 +1000 18326 18327 add missing files for make dist 18328 18329 noted by ndim on #radeonhd 18330 18331commit 20083b0695987b25e442ecbdec24f3cb6f1ac2ae 18332Author: Dave Airlie <airlied@redhat.com> 18333Date: Wed Nov 21 08:53:44 2007 +1000 18334 18335 LVDS on r500/r600 needs some work disable mobile chips for now 18336 18337commit e4b8a4479ddea9b083b3a763dc0b9302e7b9a82a 18338Author: Dave Airlie <airlied@redhat.com> 18339Date: Wed Nov 21 08:06:12 2007 +1000 18340 18341 r600: add memory controller regs from AMD 18342 18343commit aa88da974b97ea1e9bbb47b3494543575c09d912 18344Author: Dave Airlie <airlied@redhat.com> 18345Date: Wed Nov 21 08:01:35 2007 +1000 18346 18347 rs690 is !> r600 18348 18349commit c8872603454e6a4ffed9fc7d9adc2c364a429608 18350Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 18351Date: Tue Nov 20 22:33:39 2007 +1000 18352 18353 radeon: restructure pci ids to avoid effort later 18354 18355 This uses a single file with all the pciids and parameters for radeon 18356 family and setup in it. I don't run the perl script at build time to avoid 18357 a perl dependency on build but adding pci ids should be done via the csv file 18358 with openoffice or gnumeric if possible. 18359 18360commit a5b34c2f1f7d5346c4489cb30e07291d1217026c 18361Author: Dave Airlie <airlied@redhat.com> 18362Date: Tue Nov 20 16:37:29 2007 +1000 18363 18364 r600: use standard memory controller setup paths 18365 18366commit 4a523da5221d53f2efa49da2326500e9b0b9f14d 18367Author: Dave Airlie <airlied@redhat.com> 18368Date: Tue Nov 20 15:31:11 2007 +1000 18369 18370 r600: get r600 to work non-accelerated. 18371 18372 DDC still not working yet 18373 18374commit 80023441ba46882bc810ff3790c7148059f155f5 18375Author: Dave Airlie <airlied@redhat.com> 18376Date: Tue Nov 20 14:10:23 2007 +1000 18377 18378 r600: block r600 startup due to lack of memory controller info 18379 18380commit f6fbbacc17bf9b1073d3e993b225987fd9173182 18381Author: Dave Airlie <airlied@redhat.com> 18382Date: Tue Nov 20 13:41:55 2007 +1000 18383 18384 atombios: add warnings for r500 and r600 18385 18386commit 45a8b083c123b820c008f04ab857a64a8facec14 18387Author: Dave Airlie <airlied@redhat.com> 18388Date: Tue Nov 20 13:37:00 2007 +1000 18389 18390 atombios: add all r5xx and r6xx pci ids 18391 18392commit 5d023e2c3c2ab44ea57ffadc9607025d602c376c 18393Merge: 0d1e0c7 c887260 18394Author: Dave Airlie <airlied@redhat.com> 18395Date: Tue Nov 20 13:02:43 2007 +1000 18396 18397 Merge branch 'master' into atombios-support 18398 18399 Conflicts: 18400 18401 src/radeon_chipset.h 18402 src/radeon_driver.c 18403 src/radeon_probe.c 18404 18405commit 0d1e0c7805b3d8e56ccb49465e6b144afb7bdc51 18406Author: Dave Airlie <airlied@redhat.com> 18407Date: Tue Nov 20 09:08:04 2007 +1000 18408 18409 r5xx: add 71c5 for macbook pro 18410 18411commit d5909b30595c103bb5f42cd1704330f944bba49c 18412Author: Dave Airlie <airlied@redhat.com> 18413Date: Tue Nov 20 08:15:58 2007 +1000 18414 18415 r5xx: cleanups after last merge 18416 18417commit fe2f7a09050fb7a345a1f52239f8f3c4f1053891 18418Merge: 744c8cb 49055d8 18419Author: Dave Airlie <airlied@redhat.com> 18420Date: Tue Nov 20 08:04:32 2007 +1000 18421 18422 Merge branch 'master' into agd-atom-merge 18423 18424 Conflicts: 18425 18426 src/radeon_cursor.c 18427 src/radeon_output.c 18428 18429commit 744c8cb6c293fcaa687566f52901644e699baace 18430Merge: e258fbe e530af7 18431Author: Dave Airlie <airlied@redhat.com> 18432Date: Tue Nov 20 07:56:33 2007 +1000 18433 18434 Merge branch 'agd-atom' of ssh://people.freedesktop.org/~agd5f/xf86-video-ati-atom into agd-atom 18435 18436commit e258fbe411d255a1044b61d7ff738aee3fb5b7f4 18437Author: Dave Airlie <airlied@redhat.com> 18438Date: Mon Nov 19 16:35:05 2007 +1000 18439 18440 makes 2-headed cursor work 18441 18442commit e530af79adf51b3e95a0eca676c915a34dcbf4a7 18443Merge: 69e197f 52aba8d 18444Author: Alex Deucher <alex@botch2.(none)> 18445Date: Mon Nov 19 00:59:30 2007 -0500 18446 18447 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 18448 18449commit 69e197f2c8002aacf2587754c8d3bd63c88f85b1 18450Merge: 5e8940f 862dcab 18451Author: Alex Deucher <alex@botch2.(none)> 18452Date: Mon Nov 19 00:57:34 2007 -0500 18453 18454 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 18455 18456commit fca47ad083449f4cf9063dd970cdcebea6a7f110 18457Author: Dave Airlie <airlied@redhat.com> 18458Date: Mon Nov 19 15:53:40 2007 +1000 18459 18460 add z3ro's pciids 18461 18462commit 5e8940fa6e33d09091aa4bcf04b0f9e79596e1b8 18463Author: Alex Deucher <alex@botch2.(none)> 18464Date: Mon Nov 19 00:52:38 2007 -0500 18465 18466 fix logic in connector table check for TVs and switch counter to symbolic names 18467 18468commit c19123fd9483758eb6b286c3dffcb6d79d5b1ee5 18469Author: Dave Airlie <airlied@redhat.com> 18470Date: Mon Nov 19 15:46:58 2007 +1000 18471 18472 add firegl card on ajaxs machine 18473 18474commit f02f340e466a415b4e01648ca1e323f4ce125885 18475Author: Alex Deucher <alex@botch2.(none)> 18476Date: Mon Nov 19 00:39:19 2007 -0500 18477 18478 Don't assign a gpio for TV 18479 18480commit 52aba8d73189ba959f19c0437499d5e7a8829827 18481Merge: 862dcab 5e8940f 18482Author: Dave Airlie <airlied@redhat.com> 18483Date: Mon Nov 19 15:30:46 2007 +1000 18484 18485 Merge branch 'agd-atom' of ssh://people.freedesktop.org/~agd5f/xf86-video-ati-atom into agd-atom 18486 18487commit 862dcabfe0c10751d815e5cdd7436c10c2c2db10 18488Author: Dave Airlie <airlied@redhat.com> 18489Date: Mon Nov 19 15:30:08 2007 +1000 18490 18491 r520: nail i2c enable/disable issue 18492 18493commit f3dd7f413b670eeb6b8639f6677d72050ad5fe04 18494Author: Alex Deucher <alex@botch2.(none)> 18495Date: Mon Nov 19 00:19:39 2007 -0500 18496 18497 Don't detect TV out for now 18498 18499commit 94de0e22d7229ca71e18e1e849d8545d9ca7bafe 18500Author: Dave Airlie <airlied@redhat.com> 18501Date: Mon Nov 19 14:47:33 2007 +1000 18502 18503 i2c: fix bus enable stuff 18504 18505commit 3f1fc7eef13ea02fa5119e9b51d499841b801f2d 18506Author: Alex Deucher <alex@botch2.(none)> 18507Date: Mon Nov 19 00:02:14 2007 -0500 18508 18509 CRTs/DFPs may share a DVI port, but TV and CV don't 18510 18511commit 8f84c5ad4c4af14612ea68fe6f24d0d527f00acc 18512Author: Alex Deucher <alex@botch2.(none)> 18513Date: Sun Nov 18 23:43:06 2007 -0500 18514 18515 fix typo in loop 18516 18517commit 384cd8f52c89d089c6559e2eedbae45641fcd14e 18518Merge: f3f0e4e 234b607 18519Author: Dave Airlie <airlied@redhat.com> 18520Date: Mon Nov 19 14:02:55 2007 +1000 18521 18522 Merge branch 'agd-atom' of ../xf86-video-ati into agd-atom 18523 18524commit f3f0e4ec92c935c89ddb2f4241fe4335a521b439 18525Author: Alex Deucher <alex@botch2.(none)> 18526Date: Sun Nov 18 23:14:01 2007 -0500 18527 18528 RADEON: unify connectortype handling 18529 18530commit 234b6073054ac7630e82781683e666b94b2f12de 18531Author: Dave Airlie <airlied@redhat.com> 18532Date: Mon Nov 19 14:02:09 2007 +1000 18533 18534 restore avivo memory map registers at correct places 18535 18536commit 459a30ba511fe2fa8051380a9741fcfd9bb401ef 18537Author: Dave Airlie <airlied@redhat.com> 18538Date: Mon Nov 19 13:44:38 2007 +1000 18539 18540 fix type for r520 agp code 18541 18542commit 760af92412ef0d5cc44e52e7cec11fd80c4aaaeb 18543Author: Alex Deucher <alex@botch2.(none)> 18544Date: Sun Nov 18 22:34:59 2007 -0500 18545 18546 RADEON: unify DDC line handling 18547 18548commit e73bf6290da20dd61798ace775999ce1cb550934 18549Author: Dave Airlie <airlied@redhat.com> 18550Date: Mon Nov 19 13:32:16 2007 +1000 18551 18552 add x1900xt support 18553 18554commit 2e37937bacd624d616b91c41006c113791ebe98d 18555Author: Alex Deucher <alex@botch2.(none)> 18556Date: Sun Nov 18 20:18:50 2007 -0500 18557 18558 RADEON: step one in output rework 18559 18560 re-organize the output type 18561 18562commit 679e7a2e0d1b213524b8109193483bc9840fb116 18563Author: Alex Deucher <alex@botch2.(none)> 18564Date: Sun Nov 18 17:56:51 2007 -0500 18565 18566 Few fixes from the last commit. 18567 18568 Update parser works fine on r4xx. 18569 18570commit 1cd7cc3e6758ab1012f3ced6e958a1517f45557f 18571Author: Alex Deucher <alex@botch2.(none)> 18572Date: Sun Nov 18 17:44:36 2007 -0500 18573 18574 WIP: new atom code comples. 18575 18576 I commented out the object parsing for the time being as 18577 using it will require some thought as to new output 18578 related data structures. 18579 18580commit b155fa872ee4ca5d801e942aee6e619cef104f35 18581Author: Alex Deucher <alex@botch2.(none)> 18582Date: Sat Nov 17 00:34:56 2007 -0500 18583 18584 WIP: more new ATOM integration work 18585 18586commit 67db114d97abed7a607467e5d67c7b4ffa2c347e 18587Merge: 7d06a87 ea15346 18588Author: Alex Deucher <alex@botch2.(none)> 18589Date: Fri Nov 16 14:29:53 2007 -0500 18590 18591 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 18592 18593commit ea1534659de87d3d75eb20d808d039cff22cb537 18594Author: Dave Airlie <airlied@linux.ie> 18595Date: Fri Nov 16 18:46:02 2007 +1000 18596 18597 avivo: fixup some i2c stuff 18598 18599commit 7d06a8791839ce6b22e2449646832b79cebf1b21 18600Author: Alex Deucher <alex@botch2.(none)> 18601Date: Fri Nov 16 02:43:00 2007 -0500 18602 18603 WIP: sync up with the latest ATOM bios code in rhd 18604 18605 doesn't compile ATM 18606 18607commit 3614d80ceb9a7b3615b0baab3cf2dd34ed4ab464 18608Author: Dave Airlie <airlied@linux.ie> 18609Date: Fri Nov 16 17:22:39 2007 +1000 18610 18611 add missing hpd register 18612 18613commit 9a2715fda97ac0ebcb45650a416e0652aab575b8 18614Author: Dave Airlie <airlied@linux.ie> 18615Date: Fri Nov 16 17:22:25 2007 +1000 18616 18617 make i2c unlock/lock registers for gpios 18618 18619commit 20dc549fff9a4137c93ebed449d05e0c437b6bc1 18620Author: Dave Airlie <airlied@linux.ie> 18621Date: Fri Nov 16 17:01:33 2007 +1000 18622 18623 avivo i2c: consolidate the avivo i2c code 18624 18625commit 3e62730f79a13883a65a568bc821bc56055a4ab7 18626Author: Dave Airlie <airlied@linux.ie> 18627Date: Fri Nov 16 15:19:00 2007 +1000 18628 18629 atombios: fixup warnings in atombios files 18630 18631commit cca7af3c4910983f7f090792986fcbfa0dc97cfb 18632Author: Dave Airlie <airlied@linux.ie> 18633Date: Fri Nov 16 15:04:01 2007 +1000 18634 18635 remove avivo_reg.h 18636 18637commit d39eb2077c6b2fc094ccd952772528eb9428c587 18638Author: Dave Airlie <airlied@linux.ie> 18639Date: Fri Nov 16 15:00:50 2007 +1000 18640 18641 radeon: rename a large section of avivo regs to documented names 18642 18643commit b7774c28dde72a205a40be78003df72eabfb9b1f 18644Author: Dave Airlie <airlied@linux.ie> 18645Date: Fri Nov 16 14:48:36 2007 +1000 18646 18647 Add copyright headers 18648 18649commit 3cfcd2164b400bd0d1cb4ede8eeb01abba9d75c8 18650Merge: efac14e 718bfd3 18651Author: Alex Deucher <alex@botch2.(none)> 18652Date: Thu Nov 15 23:25:39 2007 -0500 18653 18654 Merge branch 'agd-atom' of /home/alex/git/airlied/xf86-video-ati2 into agd-atom 18655 18656commit efac14e669a0c6184f8848191eb49ffb21934ee1 18657Author: Dave Airlie <airlied@linux.ie> 18658Date: Thu Nov 15 23:17:25 2007 -0500 18659 18660 r5xx: fix typo for crtc offset 18661 18662commit e6db621c37ff615be286462f000d67a662c5c331 18663Author: Alex Deucher <alex@botch2.(none)> 18664Date: Thu Nov 15 23:15:56 2007 -0500 18665 18666 fix INMC() and OUTMC() on !AVIVO chips 18667 18668 WR_EN is bit 8 so don't use OUTREG8. 18669 18670commit 52ba3fdd1ce05983fabedff234cfaf4c60fba38d 18671Author: Alex Deucher <alex@botch2.(none)> 18672Date: Thu Nov 15 23:12:30 2007 -0500 18673 18674 atombios_dac_detect() takes care of primary vs tv dac itself 18675 18676commit 49055d8aff91ff12186feaf5343c8fd2f96bcba0 18677Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18678Date: Thu Nov 15 22:56:09 2007 -0500 18679 18680 RADEON: set proper defaults for tv dac BGADJ/DACADJ 18681 18682 we should get these values from the bios tables, but for now use 18683 some reasonable defaults. This should fix the washed out color 18684 problems on bugs 1082 and 12844. 18685 18686commit 821acf38b716ab87c3d07263d6e4a139fe54803f 18687Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18688Date: Thu Nov 15 22:28:42 2007 -0500 18689 18690 RADEON: Make sure we set the MT properly for connected status unknown 18691 18692commit a94123f33ec6584fbdfc4b9ecd543d1357de8814 18693Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18694Date: Thu Nov 15 22:19:54 2007 -0500 18695 18696 Revert "Portability fix from netbsd" 18697 18698 This reverts commit c9264aa53bf1470ad9104d1e7c4a8ce13c49c270. 18699 This breaks damage support. See bug 13244 18700 18701commit 718bfd3b61879172eee819fdab7080d5d4c0a756 18702Author: Dave Airlie <airlied@linux.ie> 18703Date: Fri Nov 16 10:37:04 2007 +1000 18704 18705 r5xx: fix typo for crtc offset 18706 18707commit e9d721c31372db045550f9562534b28f16121bb9 18708Author: Roland Scheidegger <sroland@tungstengraphics.com> 18709Date: Tue Nov 13 23:42:42 2007 +0100 18710 18711 ignore sometime bogus agp_mode bit from chip (bug #13190) 18712 18713 bit is wrong on at least X700 cards with rialto pcie-agp bridge chip, 18714 should be safe to use just the bit from the bridge hopefully to make 18715 agp setup work on these cards and not adversely affect others. 18716 18717commit b865886d00907899297ae864358eb26d9980975d 18718Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18719Date: Sun Nov 11 20:38:41 2007 -0500 18720 18721 bump for RC release 18722 18723commit 7aeb35e5ad1aed6e78a3d8565fbfbfe66232ab45 18724Author: Alex Deucher <alex@botch2.(none)> 18725Date: Sun Nov 11 19:29:30 2007 -0500 18726 18727 fix from last commit 18728 18729commit 2ea95900547165e86ad3f8a41ce3331a05bad60e 18730Author: Alex Deucher <alex@botch2.(none)> 18731Date: Sun Nov 11 18:43:03 2007 -0500 18732 18733 Add full parsing support for atom bios connector table 18734 18735commit 7ce730828c293f0810dfdc554df48dfd76e35c49 18736Author: Alex Deucher <alex@botch2.(none)> 18737Date: Sun Nov 11 12:37:01 2007 -0500 18738 18739 rework crtc output source setup 18740 18741commit d61b6c78aa7810a2f9b9e2d9d95aab4295de80ce 18742Author: Alex Deucher <alex@botch2.(none)> 18743Date: Sun Nov 11 11:58:17 2007 -0500 18744 18745 make sure i2c bus exists before using it 18746 18747commit 342e3e207efda42ba679731c30dfb9d5e9d5643f 18748Author: Alex Deucher <alex@botch2.(none)> 18749Date: Fri Nov 9 17:11:43 2007 -0500 18750 18751 combine outputs based on id 18752 18753commit 8078c299d5941460243944d55051547c1a4d3791 18754Author: Alex Deucher <alex@botch2.(none)> 18755Date: Fri Nov 9 16:35:08 2007 -0500 18756 18757 use atom to program plls on r4xx 18758 18759commit 5febe2c96642f61d006abe6e8081e69d5b95adc0 18760Author: Alex Deucher <alex@botch2.(none)> 18761Date: Fri Nov 9 16:24:56 2007 -0500 18762 18763 turn off vga control when using ext modes 18764 18765commit 83f170c32c08c74a9e3466ffa0e0a0606c74427b 18766Author: Alex Deucher <alex@botch2.(none)> 18767Date: Thu Nov 8 19:33:13 2007 -0500 18768 18769 add pci ids for X1550PRO 18770 18771commit 5cdcaba0f6e9de2d15cfcc109ab97d9fd423e3bf 18772Author: Alex Deucher <alex@botch2.(none)> 18773Date: Thu Nov 8 19:28:03 2007 -0500 18774 18775 make sure to assign gpio 18776 18777commit 2dcb852778301b9284a2b4906dcf64f95ed638b7 18778Author: Alex Deucher <alex@botch2.(none)> 18779Date: Thu Nov 8 18:39:23 2007 -0500 18780 18781 pull in another of Dave's fixes 18782 18783commit 96273016a0bbdfa4d3a4e6275a3b09eeeadaa534 18784Author: Alex Deucher <alex@botch2.(none)> 18785Date: Wed Nov 7 01:22:30 2007 -0500 18786 18787 fix and move crtc source set up atombios_output.c 18788 18789 it's really more output related. 18790 18791commit 5c495c81cc3bcd4a38d06954243ed3bdc85bdc07 18792Author: Alex Deucher <alex@botch2.(none)> 18793Date: Wed Nov 7 01:04:11 2007 -0500 18794 18795 add support for initing external tmds via ATOM 18796 18797commit 5c13d9355280e6de44ebbf8de7ea89a6b91c7388 18798Author: Alex Deucher <alex@botch2.(none)> 18799Date: Wed Nov 7 00:17:28 2007 -0500 18800 18801 add avivo output stuff 18802 18803commit 68e7f5c67e2e9d2162b469ce31f452f3f89756b5 18804Author: Alex Deucher <alex@botch2.(none)> 18805Date: Tue Nov 6 23:43:29 2007 -0500 18806 18807 more avivo updates 18808 18809commit 303562dfb57e13c027b2aa9289d54e547c829ff1 18810Author: Alex Deucher <alex@botch2.(none)> 18811Date: Tue Nov 6 23:06:46 2007 -0500 18812 18813 add additional connector types 18814 18815commit 0d3e0735f710cb7b9505e4330997aa332f73c102 18816Author: Alex Deucher <alex@botch2.(none)> 18817Date: Tue Nov 6 22:59:25 2007 -0500 18818 18819 First round of avivo support 18820 18821commit 0abfe3150ce3eed4db93ccc2975bd4622dfa54a7 18822Author: Alex Deucher <alex@botch2.(none)> 18823Date: Tue Nov 6 18:47:00 2007 -0500 18824 18825 Add atombios files 18826 18827commit 20f01950e42babc308b4470df6a3c6628c932003 18828Author: Alex Deucher <alex@botch2.(none)> 18829Date: Tue Nov 6 18:04:43 2007 -0500 18830 18831 for r4xx ATOM cards, just use ATOM for PLL 18832 18833 while crtc timing and pll seem to work fine, output setup 18834 and routing don't seem to work too reliably with atom. 18835 AMD claims ATOM was still pretty new with r4xx so 18836 it's probably better to stick with direct programming for 18837 some things. 18838 18839commit 37af2f6b52c8f217b205486742a6b46a864da8ed 18840Author: Michel Dänzer <michel@tungstengraphics.com> 18841Date: Mon Nov 5 10:58:25 2007 +0100 18842 18843 radeon: Use indirect register access for (un)hiding the cursor. 18844 18845 This avoids hangs that occur when touching these registers while the card is 18846 busy with some cards. 18847 18848commit 5659e2f1b08e44d0c4f21a403e91b4801ef78f62 18849Author: Michel Dänzer <michel@tungstengraphics.com> 18850Date: Mon Nov 5 10:32:04 2007 +0100 18851 18852 radeon: Fix build warning on powerpc. 18853 18854commit d41d73f5f9d8baee7137e94da2c7852ea0412180 18855Author: Michel Dänzer <michel@tungstengraphics.com> 18856Date: Mon Nov 5 10:30:23 2007 +0100 18857 18858 Revert "radeon: Reinstate sync in radeon_crtc_show/hide_cursor." 18859 18860 This reverts commit 0241cac643fa1c08a45ea44f5c670b290e760ad8. 18861 18862 Taking the DRI lock in paths called from the SIGIO handler isn't safe. For 18863 example, the SIGIO handler may be called while the lock is held by an AIGLX 18864 context, resulting in deadlock. See e.g. 18865 https://bugs.freedesktop.org/show_bug.cgi?id=13005 . 18866 18867commit 78a3eabff382e8ebe33df2039076fb083bcc361b 18868Author: Alex Deucher <alex@botch2.(none)> 18869Date: Sun Nov 4 14:11:26 2007 -0500 18870 18871 WIP: get ATOM crtc stuff working on r4xx 18872 18873commit 5e9ebd8e496b72b051053d637c63b2956b7861d3 18874Author: Alex Deucher <alex@botch2.(none)> 18875Date: Sat Nov 3 20:46:17 2007 -0400 18876 18877 Initial support for r4xx 18878 18879 - add r3xx/r4xx MC access macros and functions 18880 18881commit c106075ccb81ca2ee4894743e676fd37653c8dce 18882Author: Alex Deucher <alex@botch2.(none)> 18883Date: Sat Nov 3 18:29:20 2007 -0400 18884 18885 More of Dave's ATOM init code. 18886 18887commit e18f5d61806b445ad77d93e258fbce9422b52bb6 18888Author: Alex Deucher <alex@botch2.(none)> 18889Date: Sat Nov 3 18:20:55 2007 -0400 18890 18891 Initial integration of Atom code and some of Dave's code. 18892 18893 ATOM builds, but it's not hooked up yet. 18894 18895commit aaafc6a8f33d4736c37c43d3fd5e43085a2590fb 18896Author: Brice Goglin <bgoglin@debian.org> 18897Date: Sun Oct 28 21:10:32 2007 +0100 18898 18899 Fix typo in 'disposing cursor' 18900 18901 Bug #12984. 18902 18903commit a5c55c418319c1315ad73c305ef48b44fe1d9580 18904Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18905Date: Sat Oct 27 19:46:11 2007 -0400 18906 18907 RADEON: disable FP DETECT for DVI. 18908 18909 These bits don't seem to be reliable or routed right 18910 in all cases. 18911 18912commit 49cf7cb3db36ce7734f7e314a040a240191d6477 18913Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18914Date: Thu Oct 25 21:23:09 2007 -0400 18915 18916 RADEON: Fix IgnoreEDID option 18917 18918commit 70bbc0cfe699588f8ae4123efca0e1baffc91000 18919Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18920Date: Thu Oct 25 20:56:11 2007 -0400 18921 18922 RADEON: clean up mode handling and improve reliability of DDC 18923 18924 - Reduce the number of times we do DDC and don't 18925 skip the magical GPIO dance when doing DDC. 18926 - If no DDC modes are found, add the screen modes 18927 18928commit 1a4f590c2ac071f0c348f38e9f26fa6093722453 18929Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18930Date: Thu Oct 25 19:25:36 2007 -0400 18931 18932 RADEON: document DefaultTMDSPLL option 18933 18934commit 2d26e0d0bcf25e210dc7771e76638dcfafcc2356 18935Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18936Date: Thu Oct 25 19:18:18 2007 -0400 18937 18938 RADEON: Add driver option to force TV DAC load detection on 18939 18940commit a9306b7986467c0794c91625ad2ece597b718462 18941Author: Alex Deucher <alex@botch2.(none)> 18942Date: Thu Oct 18 20:06:58 2007 -0400 18943 18944 RADEON: fix possible crash in radeon_crtc_mode_set() 18945 18946 when radeon_crtc_mode_set() is called during ScreenInit(), 18947 we don't have pScrn->pScreen yet. 18948 18949commit 115b4b65ef679a46d90b7fc8ac22ace37038b3fc 18950Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18951Date: Wed Oct 17 20:25:47 2007 -0400 18952 18953 RADEON: Remove LVDSBiosNativeMode Option 18954 18955 Shouldn't be needed any longer as I've finally sorted out 18956 the LVDS issues due to the crtc setup. 18957 18958commit f2eb5c1cc69a4f7b0754ec6b2efde4fa1650ab6d 18959Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 18960Date: Tue Oct 16 18:54:28 2007 -0400 18961 18962 RADEON: finish fixing LVDS/RMX 18963 18964 For panels, both the mode and crtc values need to be updated based 18965 on the native mode timing. This shoudl fix any remaining problems 18966 with RMX. 18967 18968commit 93561688087493fccb7ac387d728feec61212180 18969Author: Michel Dänzer <michel@tungstengraphics.com> 18970Date: Tue Oct 16 10:57:06 2007 +0200 18971 18972 radeon: Handle 64 bit PCI resource bases better. 18973 18974commit 08945fc8a14ba57e32350e85daa3354619e9e72e 18975Author: Alex Deucher <alex@botch2.(none)> 18976Date: Mon Oct 15 21:12:46 2007 -0400 18977 18978 RADEON: allow the user to set any tv standard 18979 18980 allow the user to set any tv standard rather than 18981 limiting it to what the bios says is supported. 18982 18983commit 571548b2b95ce289f26a77c11deb6266f0b18516 18984Author: Alex Deucher <alex@botch2.(none)> 18985Date: Mon Oct 15 20:33:56 2007 -0400 18986 18987 RADEON: use native timing for RMX 18988 18989 RMX should work again with native timings after the previous 18990 patch. 18991 18992commit 2f87bff293a343b40c1be096933a5ae126632468 18993Author: Alex Deucher <alex@botch2.(none)> 18994Date: Mon Oct 15 20:06:28 2007 -0400 18995 18996 RADEON: Fix subtle change in crtc reg init 18997 18998 At some point we changed how hsync_wid and vsync_wid were clipped. 18999 Previously we used the field size as a mask when building the 19000 register. This got changed to setting the value to the field 19001 size if it was greater. this probably explains number stange 19002 mode bugs. 19003 19004commit fd5bb7bb5e968127b87102320eccc1222f205e5a 19005Author: Michel Dänzer <michel@tungstengraphics.com> 19006Date: Sat Oct 13 18:56:43 2007 +0200 19007 19008 radeon: Improve detection of default value for Option "MacModel" on Linux. 19009 19010 * Detect all PowerBooks with dual link DVI (according to developer.apple.com) 19011 and both Mac Mini G4 models. Other PowerBooks and iBooks should be covered 19012 by the 'detected as' line in /proc/cpuinfo. 19013 * Give specific instructions for overriding and reporting incorrect detection. 19014 * Only perform detection when Option "MacModel" isn't present with a valid 19015 value. 19016 * Close /proc/cpuinfo file handle when done with it. 19017 * Coding style cleanups. 19018 19019commit fbf121add5584049627f07345fc502b2aefc88ce 19020Author: Eugen Dedu <Eugen.Dedu@pu-pm.univ-fcomte.fr> 19021Date: Sat Oct 13 16:51:58 2007 +0200 19022 19023 radeon: Derive default value of Option "MacModel" from /proc/cpuinfo on Linux. 19024 19025commit 5db3afaa1fdb69d382ac769ef40191a4b964d28e 19026Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19027Date: Thu Oct 11 20:25:20 2007 -0400 19028 19029 RADEON: return status unknown for flaky chip/connector combinations 19030 19031 This should at least get something on the screen. 19032 XPRESS chips, I'm looking in your general direction... 19033 19034commit 1148d332f52f3780897aae3fcd5d6a67687c42cd 19035Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19036Date: Thu Oct 11 19:09:45 2007 -0400 19037 19038 RADEON: fix potential crash in mode handling 19039 19040commit 905b19df33d6f912336b65895e97775ffec07728 19041Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19042Date: Thu Oct 11 18:55:08 2007 -0400 19043 19044 RADEON: small cleanup from last commit 19045 19046commit 7afd04c1e4ffa6e4e5ba08ae90ba002237dc282b 19047Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19048Date: Thu Oct 11 00:16:45 2007 -0400 19049 19050 RADEON: tell the bios not to muck with the hardware while the driver is active 19051 19052 by toggling the appropriate bios scratch regs you can tell 19053 the bios not the touch the hw while the driver is active. 19054 This should prevent the bios from scrambling the hardware 19055 when users open the lid or toggle bios hotkeys. 19056 19057 fixes bug 12567 19058 19059commit 1b231d28fdda5cdc44bb9d2075d4edfd8f17e21f 19060Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19061Date: Wed Oct 10 21:54:46 2007 -0400 19062 19063 RADEON: attempt to detect lid status on legacy bios 19064 19065 I'm not sure how reliable this is. The lid closed 19066 bit only seems to be set if the lid is closed at boot 19067 so LVDS will default to on if you restart X with the 19068 lid closed after boot. 19069 19070commit 47ef583b993607bc3a372dd125c6e5228c4ec7bf 19071Author: Alex Deucher <alex@botch2.(none)> 19072Date: Wed Oct 10 19:35:31 2007 -0400 19073 19074 RADEON: more XPRESS connector table hacks 19075 19076 Seems XPRESS desktop chips with DVI-D list a proprietary 19077 connector type rather than DVI-D. try and do the right thing. 19078 Also, XPRESS chips seem to include connector type 8 and ddc 19079 type 8 which are unknown at this time. 19080 19081commit c9264aa53bf1470ad9104d1e7c4a8ce13c49c270 19082Author: Joerg Sonnenberger <joerg@britannica.bec.de> 19083Date: Tue Oct 9 23:18:29 2007 -0400 19084 19085 Portability fix from netbsd 19086 19087commit 80eee856938756e1222526b6c39cee8b5252b409 19088Author: Matthieu Herb <> 19089Date: Tue Oct 9 16:17:50 2007 -0400 19090 19091 RADEON: fix console restore on netbsd 19092 19093 Include the mode restore bugfix from monolithic Xorg, that is derived 19094 from the version in xsrc which in turn was provided by Matthieu Herb 19095 over 3 years ago on the XFree86 lists. Suggested by various 19096 developers, hold-back due to the working state in xorg-server 1.1.1. 19097 Tracing down the exact change showed that the changed default color 19098 depth made this issue a lot more prominent again. Discussed with Eric 19099 Anholt. 19100 19101commit 547fef4c6382f8a4951c086ee531b804cba31075 19102Author: Lars Weiler <pylon@gentoo.org> 19103Date: Tue Oct 9 16:10:34 2007 -0400 19104 19105 include stdint.h to fix the build on some platforms 19106 19107 fixes bug 12733 19108 19109commit ddec5783103045084a66a34d71326cc6e4a3f598 19110Author: George Sapountzis <gsap7@yahoo.gr> 19111Date: Tue Oct 9 13:27:27 2007 +0300 19112 19113 theatre compiles with pci-rework. 19114 19115commit 051435610a66735fd455bbb526fa294fcfe8c0b6 19116Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19117Date: Sun Oct 7 19:39:47 2007 -0400 19118 19119 RADEON: still more LVDS fixes 19120 19121 Seems some laptops need the native mode from the bios for 19122 LVDS while others seem to prefer a CVT mode. Add an option 19123 to pick the preferred mode. The default it to use the bios 19124 table timing. 19125 19126commit 0b03a73b7dcb4aa192c42f2a4c842d324c358122 19127Author: Dave Airlie <airlied@linux.ie> 19128Date: Sat Oct 6 20:11:19 2007 +1000 19129 19130 radeon: add support for DDC type 6 19131 19132 thanks to arekm on xorg-devel for testing/feedback 19133 19134commit 83d7506a603889b6c18736cff9b2cd3296c95c84 19135Author: Alex Deucher <alex@botch2.(none)> 19136Date: Fri Oct 5 20:25:04 2007 -0400 19137 19138 Bump for RC release 19139 19140commit fe66cc4e02a52e1d014ae5de192c65e8abd7f58a 19141Author: Tormod Volden <bugzi06.fdo.tormod@xoxy.net> 19142Date: Fri Oct 5 20:10:55 2007 -0400 19143 19144 RADEON: rename to PCI_CHIP_RS485_5975 everywhere 19145 19146 RS482_5975 is actually RS485. see bug 12048 19147 19148commit 314e8fdbe5369e4c1786d176d03d0d2a046f25d5 19149Author: Tormod Volden <bugzi06.fdo.tormod@xoxy.net> 19150Date: Fri Oct 5 20:05:54 2007 -0400 19151 19152 RADEON: Disable DRI by default for RS482 cards 19153 19154 fixes bug 12048 19155 19156commit abbafe6db7c0d24efc61a05950d4214c00d3f043 19157Author: Michel Dänzer <michel@tungstengraphics.com> 19158Date: Thu Oct 4 11:33:09 2007 +0200 19159 19160 radeon: Fix compile warnings. 19161 19162commit 1110957ba703475847b8e72023c6edcbd1068aef 19163Author: Michel Dänzer <michel@tungstengraphics.com> 19164Date: Thu Oct 4 11:32:48 2007 +0200 19165 19166 radeon: Fix for pci-rework. 19167 19168 Surprisingly easy, thanks to George's pci-rework changes. 19169 19170commit 7addf41885ec5658f531624a9c24ea5bd7d22d19 19171Merge: 3a958ba 78c6bd3 19172Author: Michel Dänzer <michel@tungstengraphics.com> 19173Date: Thu Oct 4 11:06:08 2007 +0200 19174 19175 Merge remote branch 'origin/pci-rework' 19176 19177commit 3a958ba136c3fae5a4ddd56373ac7cd47046f10e 19178Author: Alex Deucher <alex@botch2.(none)> 19179Date: Wed Oct 3 22:59:14 2007 -0400 19180 19181 RADEON: remove some cruft 19182 19183commit b6bda79f72df5e5bf9c6b71fa3298e765da506bd 19184Author: Alex Deucher <alex@botch2.(none)> 19185Date: Wed Oct 3 22:56:27 2007 -0400 19186 19187 RADEON: more fixes for single crtc chips 19188 19189commit 5f5c4e6ad61c45c24f1443b91b4bc5375efdebc0 19190Author: Alex Deucher <alex@botch2.(none)> 19191Date: Wed Oct 3 22:46:21 2007 -0400 19192 19193 RADEON: remove RADEONSaveMode() 19194 19195 It's only called by RADEONSave(), so just move its 19196 contents into RADEONSave(). this mirrors RADEONRestore(). 19197 19198commit bfede412b3a3cd11769a580b167c528734146096 19199Author: Alex Deucher <alex@botch2.(none)> 19200Date: Wed Oct 3 22:19:16 2007 -0400 19201 19202 RADEON: minor fixes for external TMDS 19203 19204commit 597dffce9bdc200003d0be880235258386a0bdd7 19205Author: Alex Deucher <alex@botch2.(none)> 19206Date: Wed Oct 3 21:43:08 2007 -0400 19207 19208 RADEON: final fix for RMX/LVDS 19209 19210 It seems that on quite a few laptops the LVDS H/V timing 19211 from the bios tables are way off or just doesn't work period. 19212 Either we are using it wrong or we need some additional checks 19213 when we parse it. Only the dot clock seems to really matter, 19214 so use cvt modes and update the dotclock. This seems to work 19215 correctly in all cases. 19216 19217commit cc0c2d8e61600652b1f9cb3dc49db2ef62b1e40d 19218Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19219Date: Wed Oct 3 00:59:37 2007 -0400 19220 19221 RADEON: Finally sort out LVDS modes 19222 19223 - The panel timing from the bios tables is for the crtc, 19224 not the native panel mode, so add cvt modes for the panel 19225 and fix up the crtc values in mode_fixup(). 19226 - rename and reconfigure RADEONValidateFPModes() to what it 19227 really does: add screen modes. I suppose for backwards 19228 compatibility we ought to add the screen modes to every 19229 reasonable output, but everyone should really use xrandr 19230 or an output monitor section to add custom modes. Hopefully 19231 this will go away at some point. 19232 19233commit 0ca184c3c35032df39ea7ce5d2d4aba1a97b6426 19234Author: Alex Deucher <alex@botch2.(none)> 19235Date: Tue Oct 2 19:38:18 2007 -0400 19236 19237 RADEON: fix RADEONInitDispBandwidth() on single crtc cards 19238 19239 Should fix bug 12490 19240 19241commit a5fb2a5d54cdd562a97881cdea1a4af35b483aab 19242Author: Dave Airlie <airlied@linux.ie> 19243Date: Mon Oct 1 08:48:27 2007 +1000 19244 19245 radeon: fix powerpc build 19246 19247commit 22519fde1e002f28d6036d448fcd18452d00f1bb 19248Author: Alex Deucher <alex@botch2.(none)> 19249Date: Sun Sep 30 13:11:20 2007 -0400 19250 19251 RADEON: add support for ext tmds table and ext tmds chip init 19252 19253 This probably won't work on all chips as the various gpio lines 19254 seem to need special magic to to actually talk to the i2c slave 19255 chips. 19256 19257commit dcc376e2d2a13329dd03f1bc4b471329757a6f5f 19258Author: Alex Deucher <alex@botch2.(none)> 19259Date: Wed Sep 26 23:27:38 2007 -0400 19260 19261 RADEON: remove cruft 19262 19263commit 99ceaefa18c6e07b55106cca0ea8996fa73667be 19264Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19265Date: Sun Sep 30 12:02:54 2007 -0400 19266 19267 RADEON: more work on LVDS mode validation and fixups 19268 19269 Hopefully this will fix up the weird mode issues with LVDS and 19270 native and RMX modes, but I can't seem to get quite the right 19271 combo to fix everyone. 19272 19273commit d808781d48adf01e80b5bb476bae2d2f599030f1 19274Author: Dave Airlie <airlied@redhat.com> 19275Date: Fri Sep 28 14:50:42 2007 +1000 19276 19277 rc410: disable DRI by default due to it not working 19278 19279commit 4d9f944e40c6b9ac7e80428dff49c6e83fda1d8f 19280Author: Michel Dänzer <michel@tungstengraphics.com> 19281Date: Thu Sep 27 17:50:15 2007 +0200 19282 19283 radeon: Fix build on little endian platforms. 19284 19285 This was accidentally broken when moving away from using pixmap data pointers 19286 directly. 19287 19288commit 5902e913010a2fb0c17efe01f9aade73f9646f16 19289Author: Michel Dänzer <michel@tungstengraphics.com> 19290Date: Thu Sep 27 12:53:34 2007 +0200 19291 19292 radeon: EXA pitch fixes. 19293 19294 The combined pitch/offset registers only support pitches up to 16320 bytes. 19295 19296 With EXA >= 2.3, set the maxPitchBytes field accordingly. With older versions, 19297 limit maxX such that the pitch of 32bpp pixmaps doesn't exceed the limit. 19298 19299 Also check the limit in RADEONGetOffsetPitch just in case. 19300 19301 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=12551 . 19302 19303commit ca728973bce504e03bf6228d18287906c35eaa3f 19304Author: Michel Dänzer <michel@tungstengraphics.com> 19305Date: Thu Sep 27 11:14:21 2007 +0200 19306 19307 radeon: Don't use pixmap data pointer directly with EXA. 19308 19309 Fixes segfault with current xserver master without AccelDFS. 19310 19311commit 21593d04d222b05dbba9abd31eaa3bfb91d999b6 19312Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19313Date: Mon Sep 24 23:01:20 2007 -0400 19314 19315 RADEON: more clean up of mode code 19316 19317 also print out LVDS mode info from bios for both 19318 legacy and atom bioses 19319 19320commit 22cccb99d762567757d3fd38795f71e943f1faf9 19321Author: Alex Deucher <alex@botch2.(none)> 19322Date: Mon Sep 24 00:10:01 2007 -0400 19323 19324 bump for RC release 19325 19326commit fad9e7b00de051f632000eb2304ac23a524f7c8e 19327Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19328Date: Sun Sep 23 23:43:26 2007 -0400 19329 19330 RADEON: Rework LVDS mode validation 19331 19332 The old validation code wasn't really well suited to randr. 19333 This fixes several issues: 19334 - missing display size for panels with edid 19335 - broken duplicate modes 19336 19337commit bf9674b9df351108e069d037ec10869a4154a881 19338Author: Alex Deucher <alex@botch2.(none)> 19339Date: Sun Sep 23 15:01:34 2007 -0400 19340 19341 RADEON: properly set default RMX types 19342 19343commit b27135bce8b41d69290613b440a338b0a7fe0200 19344Author: Alex Deucher <alex@botch2.(none)> 19345Date: Sun Sep 23 14:44:38 2007 -0400 19346 19347 RADEON: fix up LVDS handling for r3xx and newer 19348 19349commit d5cf268be2c9b3e320d7101f5213f7d5609b6308 19350Author: Alex Deucher <alex@botch2.(none)> 19351Date: Sat Sep 22 18:31:51 2007 -0400 19352 19353 RADEON: Fix crash when Xv window is outside of either crtc 19354 19355commit 8aca0ca6be7122e0ece20985a1862553a1b225c0 19356Author: Alex Deucher <alex@botch2.(none)> 19357Date: Sat Sep 22 17:56:50 2007 -0400 19358 19359 Revert "RADEON: fix crash when Xv window is outside of either crtc" 19360 19361 This reverts commit 9109e62e3be7f96b41b534ab517fdf1baf458806. 19362 19363 This breaks ABI. better fix to come. 19364 19365commit d6e1d9be1cd5786a24ed4bb8e45fa7e04df56622 19366Author: Alex Deucher <alex@botch2.(none)> 19367Date: Sat Sep 22 16:14:06 2007 -0400 19368 19369 RADEON: Don't make the entity as shareable 19370 19371 This should prevent people from mistakenly trying to run 19372 zaphod mode 19373 19374commit 81114af1cdddc0b10d076f2e38c7a00c1223cc48 19375Author: Alex Deucher <alex@botch2.(none)> 19376Date: Sat Sep 22 15:51:23 2007 -0400 19377 19378 RADEON: preliminary support for mac mini 19379 19380 Option "MacModel" "mini" 19381 may not be 100% correct yet 19382 19383commit 6c482e453bc8156886294d0c1b8f3f1b3dcf4b36 19384Author: Alex Deucher <alex@botch2.(none)> 19385Date: Sat Sep 22 15:11:20 2007 -0400 19386 19387 RADEON: Fix RMX on LVDS 19388 19389 LVDS + RMX doesn't seem to like having the crtc values tweaked. 19390 19391commit 9109e62e3be7f96b41b534ab517fdf1baf458806 19392Author: Michel Dänzer <daenzer@debian.org> 19393Date: Sat Sep 22 14:19:10 2007 -0400 19394 19395 RADEON: fix crash when Xv window is outside of either crtc 19396 19397 Should fix bug 12175 19398 19399commit d2ce4a5003ce1291ea2327b2c00a0b24408fe26c 19400Author: Alex Deucher <alex@botch2.(none)> 19401Date: Sat Sep 22 10:41:55 2007 -0400 19402 19403 RADEON: RMX updates 19404 19405 - add option to turn off RMX 19406 - turn off RMX by default on DVI 19407 - add infrastructure to support more RMX modes 19408 19409commit f95b9ab729376083bf0d12987ee260ec2aba721e 19410Author: Alex Deucher <alex@botch2.(none)> 19411Date: Fri Sep 21 00:12:37 2007 -0400 19412 19413 RADEON: remove more old cruft 19414 19415commit 5a6f74103f0ec0d451d0e2573442efe5922848af 19416Author: Maciej Cencora <m.cencora@gmail.com> 19417Date: Thu Sep 20 23:56:08 2007 -0400 19418 19419 RADEON: fix video in on RV380 (tested on X600 VIVO) 19420 19421commit c72a365386e19f9257db041d44b09ad499cc9f6a 19422Author: Alex Deucher <alex@botch2.(none)> 19423Date: Thu Sep 20 23:49:57 2007 -0400 19424 19425 RADEON: fix up dvo support (still no external chip init) 19426 19427commit 5e4d98470b6412a686883c554e7eb7badbe78c4d 19428Author: Alex Deucher <alex@botch2.(none)> 19429Date: Thu Sep 20 23:22:48 2007 -0400 19430 19431 RADEON: round 3 on the PLLs. should fix the LVDS issues 19432 19433commit c5e2a2f09af807006c7ea493a8e90ff77abe207c 19434Author: Alex Deucher <alex@botch2.(none)> 19435Date: Wed Sep 19 19:58:28 2007 -0400 19436 19437 bump for RC release 19438 19439commit 46ff78b9f010ca24178d4363761be00eb3ecb632 19440Author: Alex Deucher <alex@botch2.(none)> 19441Date: Wed Sep 19 19:49:58 2007 -0400 19442 19443 RADEON: add default connector setup for single crtc chips 19444 19445commit 509ca0cb0e1cde905b47db2bbac6f2a58523b279 19446Author: Alex Deucher <alex@botch2.(none)> 19447Date: Wed Sep 19 19:41:17 2007 -0400 19448 19449 RADEON: fix external TMDS parsing in legacy bios connector table 19450 19451commit 5a59547ef7e986c9613023d941c3354cd476faf1 19452Author: Alex Deucher <alex@botch2.(none)> 19453Date: Wed Sep 19 19:28:09 2007 -0400 19454 19455 Revert "RADEON: initial pass at external TMDS support" 19456 19457 This reverts commit 4000a710c93dd2d82891e4082bc7fa922ba9c5f4. 19458 19459 This needs to be reworked and needs more soak time. so revert 19460 for now. 19461 19462commit dc333884c540d536bffe51a5ebfaf1822af6a91c 19463Author: Alex Deucher <alex@botch2.(none)> 19464Date: Tue Sep 18 22:23:12 2007 -0400 19465 19466 RADEON: fix plls again 19467 19468 previous fix seemed to break other chips. Lets see how 19469 this goes. 19470 19471commit 271e541088f455d1cfedff87e88c9a4fdbdbe424 19472Author: Alex Deucher <alex@botch2.(none)> 19473Date: Tue Sep 18 22:12:31 2007 -0400 19474 19475 RADEON: Remove more mergedfb cruft 19476 19477commit 8ae69c496eba701e744cca0605f73242673f7b3f 19478Author: Alex Deucher <alex@botch2.(none)> 19479Date: Tue Sep 18 20:12:21 2007 -0400 19480 19481 RADEON: adjust pll restore 19482 19483 some chips seem to be pickier than others. fixes bug 12467 19484 19485commit 8a84dc8ee260716026fa661a6dced2000621efa2 19486Author: Michel Dänzer <michel@tungstengraphics.com> 19487Date: Tue Sep 18 11:53:39 2007 +0200 19488 19489 radeon: Remove declaration of unused variable. 19490 19491commit 2fbbd9c5f0ec27e72780e8b054bc59502f1e5651 19492Author: Alex Deucher <alex@botch2.(none)> 19493Date: Tue Sep 18 00:45:56 2007 -0400 19494 19495 RADEON: note that MergedFB support has been replaced with randr 1.2 19496 19497commit 81b708dbd35c29bf2d67e8c655af97b60e61498b 19498Author: Alex Deucher <alex@botch2.(none)> 19499Date: Mon Sep 17 23:54:03 2007 -0400 19500 19501 RADEON: turn off load detection on tv dac by default 19502 19503 it's just too unreliable and even when it is most people 19504 don't want it. enable it via output attribute or force 19505 the outputs on with xrandr 19506 19507commit c3ab53159d42157985a25e9125eaa5362273b73d 19508Author: Alex Deucher <alex@botch2.(none)> 19509Date: Mon Sep 17 23:28:51 2007 -0400 19510 19511 RADEON: XPRESS connector tables are crap 19512 19513 besides specifying the pimary DAC (which they don't have), 19514 most seem to specify DDC_CRT2 for DDC, which does not seem 19515 to ever work. Some have reported success with DDC_MONID so 19516 lets give that a try. 19517 19518commit 1fafa970fa78ddea29a8804b6cd0aea52e4fa2e1 19519Author: Alex Deucher <alex@botch2.(none)> 19520Date: Mon Sep 17 23:10:13 2007 -0400 19521 19522 RADEON: ignore edid digital bit on analog ports 19523 19524 Some monitors have this bit set even on analog ports 19525 19526commit 4000a710c93dd2d82891e4082bc7fa922ba9c5f4 19527Author: Alex Deucher <alex@botch2.(none)> 19528Date: Mon Sep 17 23:04:17 2007 -0400 19529 19530 RADEON: initial pass at external TMDS support 19531 19532 - Based on Dave Airlie's initial work and the dvo support 19533 in the intel driver. 19534 - Only sil164 is supported at the moment. 19535 - Once we get some testing and such, we ought to move the dvo 19536 drivers out of the drivers so they can be shared among all drivers 19537 - Doesn't seem to work on my card (r4xx ATOM) card ATM 19538 - Legacy bios table programming sequence not implemented yet 19539 19540commit 38515d402555eaa61c686d42973e59f659b07466 19541Author: Alex Deucher <alex@botch2.(none)> 19542Date: Sat Sep 15 11:39:47 2007 -0400 19543 19544 RADEON: don't restore bios scratch regs 19545 19546 Leave them as programmed by the bios 19547 fixes bug 12424 19548 19549commit 78c6bd305f2ea70629d1dda3c0c48c067451f5fb 19550Author: George Sapountzis <gsap7@yahoo.gr> 19551Date: Mon Sep 10 19:00:16 2007 +0300 19552 19553 [mach64] Revert to multiple mappings per BAR. 19554 19555 libpciaccess added subrange mappings and mtrr handling. 19556 19557 compile-tested only. 19558 19559commit 0241cac643fa1c08a45ea44f5c670b290e760ad8 19560Author: Michel Dänzer <michel@tungstengraphics.com> 19561Date: Sat Sep 8 00:22:40 2007 +0200 19562 19563 radeon: Reinstate sync in radeon_crtc_show/hide_cursor. 19564 19565 Make sure the DRI lock is held though, as these can be called asynchronously. 19566 19567 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=12245 . 19568 19569commit 49933e2f7d590811f2bc8c0d51a09f3b7f14845e 19570Author: Alex Deucher <alex@botch2.(none)> 19571Date: Sat Sep 1 12:23:07 2007 -0400 19572 19573 RADEON: automatically disable tiling if requested virtual desktop exceeds surface limits 19574 19575 Also, tweak default desktop sizes and add some informational messages 19576 19577commit d49a60bc939c9f3e9d57c23e44263f5bb52f09fb 19578Author: Sjoerd Simons <sjoerd@luon.net> 19579Date: Thu Aug 30 11:40:13 2007 +0200 19580 19581 radeon: Fix Option "MacModel". 19582 19583commit 2b860726fff058623e0e8e1b0aca092c246875ce 19584Author: Michel Dänzer <michel@tungstengraphics.com> 19585Date: Thu Aug 30 11:38:28 2007 +0200 19586 19587 radeon: Don't build currently unused RADEONSavePalette. 19588 19589commit 12187a6aa93049c002a4171344d03c713f7f3c5d 19590Author: Alex Deucher <alex@botch2.(none)> 19591Date: Wed Aug 29 23:11:30 2007 -0400 19592 19593 RADEON: Add quirk and connector tables for apple laptops 19594 19595 As far as I can tell there are three apple laptop variants: 19596 ibook - LVDS, TVDAC drives TV or VGA via dongle 19597 powerbook-duallink - LVDS, TV, External TMDS/Primary DAC 19598 powerbook - LVDS, TV, Internal TMDS/Primary DAC 19599 use Option "MacModel" "<string>" 19600 to enable the appropriate quirks where string is one of the above 19601 19602 We can't yet init the external TMDS directly, but if OF inits it, 19603 it should work. This should also fix bug 9955. 19604 19605 Please test! 19606 19607commit 61c1fdaa8553581944f78a11e6f9aa76163a468a 19608Author: Alex Deucher <alex@botch2.(none)> 19609Date: Tue Aug 28 23:47:19 2007 -0400 19610 19611 RADEON: add option to force tmds pll to default table 19612 19613 Also rework the tmds pll output attribute handling a bit 19614 19615commit 0d9087bc60bb95c770b899cfed29699c02bdac49 19616Author: Alex Deucher <alex@botch2.(none)> 19617Date: Tue Aug 28 23:08:20 2007 -0400 19618 19619 RADEON: remove some cruft forgotten in a previous commit 19620 19621commit 6ff0645ecfe65727e8ef5d5e6215b4e03078e1a5 19622Author: George Sapountzis <gsap7@yahoo.gr> 19623Date: Tue Aug 28 15:37:11 2007 +0300 19624 19625 [mach64] Convert to pci-rework, keeping source-code compatibility. 19626 19627 It still uses the old probe method though, this is due to the ati wrapper. 19628 19629commit 7b38d9a1209f87255e5bb0aefe46a363ce4fb6ef 19630Merge: 2e3d43a 673f799 19631Author: George Sapountzis <gsap7@yahoo.gr> 19632Date: Tue Aug 28 16:01:12 2007 +0300 19633 19634 Merge branch 'master' into pci-rework 19635 19636 Conflicts: 19637 19638 src/ati.c 19639 19640commit 673f799729824f4439dd5f681f75dd5aab50947f 19641Author: Alex Deucher <alex@botch2.(none)> 19642Date: Tue Aug 28 00:42:30 2007 -0400 19643 19644 RADEON: Update tv attributes immediately 19645 19646commit ad6f7ad1b2ccae0bc0a416b9b0ca22709c9d5199 19647Author: Alex Deucher <alex@botch2.(none)> 19648Date: Tue Aug 28 00:08:41 2007 -0400 19649 19650 RADEON: remove the "default" tv_standard option 19651 19652commit 17e0f9e6cbfdb115034d327bd34d46339fd632b7 19653Author: Alex Deucher <alex@botch2.(none)> 19654Date: Mon Aug 27 23:59:03 2007 -0400 19655 19656 RADEON: enable load detection for tvdac if output count for tvdac < 2 19657 19658commit 42839fb5a8584196e7b18375bff6c426ed0347d9 19659Author: Alex Deucher <alex@botch2.(none)> 19660Date: Mon Aug 27 23:44:13 2007 -0400 19661 19662 RADEON: make load detection an output attribute for analog outputs 19663 19664 Since TV/VGA/DVI-I can share the TV DAC, we often get false detection 19665 of all inputs that share that DAC. Make load detection an output 19666 attribute. Enabled by default on primary dac and on cards where 19667 tv dac is (usually) dedicated to tv (non-IGP mobilities). 19668 19669commit a5a1055d64ab4fa16bfb03a412ae6c4fe69ff65d 19670Author: Alex Deucher <alex@botch2.(none)> 19671Date: Mon Aug 27 22:42:22 2007 -0400 19672 19673 RADEON: make tmds pll an output attribute 19674 19675 sometimes the bios tmds plls are busted for certain 19676 monitors. sometimes the dirver tables are. Let the user 19677 pick at run time. 19678 19679commit a12e4aa01bf1c5723c3c791ff9bdc26eef21d5ea 19680Author: Alex Deucher <alex@botch2.(none)> 19681Date: Sun Aug 26 18:51:29 2007 -0400 19682 19683 Bump for new release 19684 19685commit d43596e5f5d7c60f96b57bc3e743a9b40eb7109d 19686Author: Alex Deucher <alex@botch2.(none)> 19687Date: Sun Aug 26 18:07:50 2007 -0400 19688 19689 RADEON: Fix rotation. works now 19690 19691commit 47eb3327c258bb0cfd9a1d5677624b9988a39057 19692Author: Alex Deucher <alex@botch2.(none)> 19693Date: Sun Aug 26 15:43:22 2007 -0400 19694 19695 RADEON: minor tweak to tv out 19696 19697commit f36720377737210c985b196d9a988efdd767f1c7 19698Author: Alex Deucher <alex@samba.(none)> 19699Date: Sun Aug 26 14:13:06 2007 -0400 19700 19701 RADEON: fix depth 16 palette for real this time 19702 19703commit f2b13f1457bf860b075310d3962254be0ed7bea3 19704Author: Alex Deucher <alex@botch2.(none)> 19705Date: Sun Aug 26 13:27:19 2007 -0400 19706 19707 RADEON: Only update pixclks_cntl when updating tv routing 19708 19709 No need to re-set all of the pll2 stuff 19710 19711commit 5d044b9f74c7aa7e12f2822896fed881e2ca9d19 19712Author: Alex Deucher <alex@botch2.(none)> 19713Date: Sat Aug 25 21:03:08 2007 -0400 19714 19715 RADEON: fix crtc clipping for Xv 19716 19717commit 3fd2d22a02812d5f86cdc1c9503f48362b0c362b 19718Author: Alex Deucher <alex@botch2.(none)> 19719Date: Sat Aug 25 17:37:35 2007 -0400 19720 19721 RADEON: remove fbdev option 19722 19723 FBDev support is currently broken, and it not really compatible 19724 with randr 19725 19726commit 3469e1aa08792890fa6a5c72da52a1992a0b382c 19727Author: Alex Deucher <alex@botch2.(none)> 19728Date: Fri Aug 24 20:42:13 2007 -0400 19729 19730 RADEON: add extra green data in depth 16 19731 19732 Apparently some radeons need this? 19733 19734commit 71f650d1bc432514516f7ac64a5e8a54c5227881 19735Author: Michel Dänzer <michel@tungstengraphics.com> 19736Date: Fri Aug 24 09:21:39 2007 +0200 19737 19738 Require xorg-server >= 1.3 for RandR 1.2. 19739 19740commit d7230939f523610c57f92bdfc72966bdbc6f1070 19741Author: Michel Dänzer <michel@tungstengraphics.com> 19742Date: Fri Aug 24 09:21:14 2007 +0200 19743 19744 64 bit warning fixes. 19745 19746 For printf vs. CARD32, use %u or %x and and a cast to unsigned. 19747 19748commit 91c45fedfd155a153dcd2c3f3e30986bfbd44e6f 19749Author: Dave Airlie <airlied@nx6125b.(none)> 19750Date: Fri Aug 24 15:05:01 2007 +1000 19751 19752 radeon: don't disable dac if either tv or vga is using it 19753 19754 On my rs480 I had to vt switch to get hotplug VGA working due to the tv-out 19755 code turning off the dac when the vga code was actually using it. 19756 19757commit 056ca6bb5adf974290693b55de6cd6880d2132d1 19758Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19759Date: Thu Aug 23 20:19:47 2007 -0400 19760 19761 RADEON: NONE to None to match other port info 19762 19763commit 13fd53286bdda2c55683bdb5f63e7d345f6c63ef 19764Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19765Date: Thu Aug 23 20:17:51 2007 -0400 19766 19767 RADEON: set (hopefully) reasonable default max desktop sizes 19768 19769 Based on the amount of vram. We really need ttm... 19770 19771commit 53bad86ca48a9b6529c1f0989ee568d9d48841c6 19772Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19773Date: Thu Aug 23 19:24:59 2007 -0400 19774 19775 RADEON: clean up the logic in crtc_mode_set() 19776 19777commit 4712dedea225e9e07177aebda2ffc6290d1f53c7 19778Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19779Date: Thu Aug 23 18:51:34 2007 -0400 19780 19781 RADEON: Always assume LVDS is connected 19782 19783 Not all bioses seem to set the right scratch bits. If we have 19784 LVDS (via bios table or otherwise) assume it's connected. 19785 19786commit b0f170c5f736ecba1a5899d602c4173fe9b9b1fa 19787Author: Michel Dänzer <michel@tungstengraphics.com> 19788Date: Thu Aug 23 12:55:40 2007 +0200 19789 19790 radeon: Remove unnecessary #include <time.h>. 19791 19792commit ac54c0e4360099697755d14b1030def73d8235b0 19793Author: Michel Dänzer <michel@tungstengraphics.com> 19794Date: Wed Aug 22 14:33:59 2007 +0200 19795 19796 radeon: Warning fixes. 19797 19798commit de26e406f52b3b13f03eee2b8023924ec6406f0a 19799Author: Alon Ziv <alonz@nolaviz.org> 19800Date: Mon Jul 30 22:47:59 2007 +0300 19801 19802 radeon: Sane handling of timeouts in WaitForVerticalSync(2). 19803 19804 RADEONWaitForVerticalSync() and RADEONWaitForVerticalSync2() need to wait 19805 for a timeout specified in milliseconds; looping around usleep() causes 19806 the timeout to be unnecessarily long, as the OS may sleep longer than 19807 requested (on Linux the minimum actual sleep value may be several ms). 19808 19809 The new logic uses gettimeofday() in the loop to see when the (absolute) 19810 timeout has arrived. 19811 19812 Signed-off-by: Alon Ziv <alonz@nolaviz.org> 19813 19814commit c66e5de26ae93caa368213f3cce139aacec955d2 19815Author: Sascha Sommer <saschasommer@freenet.de> 19816Date: Thu Aug 23 12:11:51 2007 +0200 19817 19818 radeon: Round down RMX stretch ratios. 19819 19820 Fixes issues with RMX scaling, see 19821 https://bugs.freedesktop.org/show_bug.cgi?id=8983 . 19822 19823commit 633c1fff10a3be4c9f48c1995e330d60bf6abbb2 19824Author: Michel Dänzer <michel@tungstengraphics.com> 19825Date: Thu Aug 23 12:11:41 2007 +0200 19826 19827 radeon: Sync pages when enabling page flipping with EXA as well. 19828 19829 Exclude the DRI window(s) though to avoid scribbling over 3D rendering. 19830 19831commit 4f8010ce22043c0f8d60c0f49d270ce98c9d2466 19832Author: Michel Dänzer <michel@tungstengraphics.com> 19833Date: Thu Aug 23 12:11:12 2007 +0200 19834 19835 radeon: Don't synchronize DRI windows between pages when possible. 19836 19837commit 7b527054a7c81d1d1dbc79d41b9e53064dab68cb 19838Author: Michel Dänzer <michel@tungstengraphics.com> 19839Date: Thu Aug 23 12:10:33 2007 +0200 19840 19841 radeon: Restore memmap registers even if only AGP location changed. 19842 19843commit d7ba9f001c0ab645984526afd0e64d1c6a6d654a 19844Author: Michel Dänzer <michel@tungstengraphics.com> 19845Date: Thu Aug 23 11:39:54 2007 +0200 19846 19847 radeon: Change a test to info->IsIGP from several IGP families. 19848 19849commit 8c7c22e22b6076abc80e4e1aaa8d1f4cf2f3ed14 19850Author: Michel Dänzer <michel@tungstengraphics.com> 19851Date: Thu Aug 23 11:38:17 2007 +0200 19852 19853 radeon: Wait for pending overlay flip to finish before emitting new one. 19854 19855commit 92fa7cc00688d7bfc1fb72e645ac30c6d92669c6 19856Author: Michel Dänzer <michel@tungstengraphics.com> 19857Date: Thu Aug 23 11:38:16 2007 +0200 19858 19859 radeon: Don't call RADEONDRIRefreshArea when the damaged region is empty. 19860 19861commit 5cb20c2dc5eca9d7d7d78e9924ea1b90076e7253 19862Author: Dave Airlie <airlied@linux.ie> 19863Date: Thu Aug 23 18:46:41 2007 +1000 19864 19865 radeon: bug 11899 + debian 435040 - test for usefbdev before calling int10 19866 (cherry picked from commit 3334c247540b113f9c4a5ce1879d8648a45a959a) 19867 19868commit 9d6261c024c697a211da4b6300420153eb72a264 19869Author: Roland Bär <roland@verifysoft.de> 19870Date: Thu Aug 23 18:33:34 2007 +1000 19871 19872 radeon: bug 11860 fix some mem leaks 19873 19874commit 7bc1f862bc5f992f213143fbafef52459ba7db4a 19875Author: Roland Bär <roland@verifysoft.de> 19876Date: Thu Aug 23 18:37:35 2007 +1000 19877 19878 radeon: bug 11861 - dead code removal in radeon_video.c 19879 (cherry picked from commit 8e3a6f83016cd8c4cfd43ceee4cbf0a8dc018b2a) 19880 19881commit 9d38c8aa1a7d6fb1af41ee8abdb4a95f94843538 19882Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 19883Date: Thu Aug 23 20:10:24 2007 +1000 19884 19885 radeon: cleanup some warnings 19886 19887commit e4c8969b48a6c8dcc4e7f9852479d24a0204fc0d 19888Author: Dave Airlie <airlied@redhat.com> 19889Date: Thu Aug 23 19:56:21 2007 +1000 19890 19891 updated release numbering for randr 1.2 19892 19893commit 81a8093f2ffdce59d4e8a44f65f3d5c771fd6425 19894Merge: c08e6ec 53a67e3 19895Author: Dave Airlie <airlied@redhat.com> 19896Date: Thu Aug 23 19:51:47 2007 +1000 19897 19898 Merge remote branch 'origin/randr-1.2' 19899 19900commit c08e6ec9e7ac06caee53689b0ec50ef6a7a0ba37 19901Merge: 5793e87 a0d0fcd 19902Author: Dave Airlie <airlied@clockmaker.usersys.redhat.com> 19903Date: Thu Aug 23 19:42:19 2007 +1000 19904 19905 Merge branch 'master' into randr-merge 19906 19907 Conflicts: 19908 19909 src/radeon.h 19910 src/radeon_bios.c 19911 src/radeon_display.c 19912 src/radeon_dri.c 19913 src/radeon_driver.c 19914 src/radeon_modes.c 19915 src/radeon_probe.h 19916 src/radeon_video.c 19917 19918commit 53a67e31904bec9a3aa1bd24de8034dcafea1d2a 19919Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19920Date: Tue Aug 21 21:20:41 2007 -0400 19921 19922 RADEON: Fix color problem on pre-R3xx chips tv-out 19923 19924commit e9719e8e02eef46717ae9b4d8c7998466dac30cb 19925Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19926Date: Tue Aug 21 21:17:20 2007 -0400 19927 19928 RADEON: more tv out fixes and clean up 19929 19930commit 36c22a49580d86a6518b67f31a78bd53d39491af 19931Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 19932Date: Tue Aug 21 20:28:39 2007 -0400 19933 19934 RADEON: fix tv-out on R3xx 19935 19936 R3xx apparently needs the tv clock forced on. 19937 19938commit 9470bd67731059f26859ed5f0bea3ade09e2c80c 19939Author: Alex Deucher <alex@botch2.com> 19940Date: Mon Aug 20 20:54:06 2007 -0400 19941 19942 RADEON: Add DefaultConnectorTable option 19943 19944 This option skips the parsing the BIOS connector table 19945 and falls back to chip specific defaults. 19946 Also remove man page section for the now gone bioshotkeys 19947 option. 19948 19949commit a0d0fcd3bab765b4db25e04884fd8a342abb9c66 19950Author: iLisa Wu <liswu@ati.com> 19951Date: Mon Aug 20 15:44:07 2007 +0200 19952 19953 Fix crash with no valid mode in xorg.conf's modelist and empty Virtual 19954 19955 If the resolution defined in xorg.conf failed to find a matching mode in the 19956 supported modelist, and no virtual desktop dimensions are defined the xorg.conf 19957 either, virtual X and Y dimension will be set to 0 which will cause Xserver 19958 crash. 19959 19960 (Novell bugzilla #296856, closed) 19961 19962commit b275febdb0918e8cebdffbb433b0eeb3ff4d3746 19963Author: Alex Deucher <alex@samba.(none)> 19964Date: Sun Aug 19 20:55:32 2007 -0400 19965 19966 RADEON: turn off TVCLK when blanking tv encoder 19967 19968commit a90d675832ddb02c81ace010ccbf02619b70edac 19969Author: Alex Deucher <alex@botch2.com> 19970Date: Thu Aug 16 21:55:14 2007 -0400 19971 19972 RADEON: fix Xv clipping and overlay sourcing 19973 19974 - Basically just copied from the intel driver. I'm planning to push 19975 this to the server soon, but add it now to get things working 19976 and to provide compat for older servers. 19977 19978 - Overlay crtc source control attribute now called XV_CRTC 19979 The old attribute XV_SWITCHCRT has been removed. If anyone cares, 19980 we can add it back as an alias to XV_CRTC 19981 XV_CRTC: -1 auto, 0 crtc0, 1 crtc1 19982 19983commit 5793e8753d11432bf95c7c6dd80c811e16aba058 19984Author: Alex Deucher <alex@botch2.com> 19985Date: Wed Aug 15 19:26:36 2007 -0400 19986 19987 RADEON: Remove RADEONRestoreMode() 19988 19989 Since we no longer use it to write modes all it did 19990 was restore some of the regs for a console restore. 19991 Just move the relevant bits into RADEONRestore() and 19992 remove it. 19993 19994commit 6f011aaabaf18d66ffc255ad76aaf938b2396302 19995Author: Lisa Wu <Lisa.Wu@amd.com> 19996Date: Wed Aug 15 19:17:51 2007 -0400 19997 19998 RADEON: fix console restore on r3xx and r4xx 19999 20000 When restoring the console we need to: 20001 1) we need to restore DAC registers after all other registers are 20002 restored and CRTCs are enabled. 20003 2) we need to enable CRTC2 registers before CRTC1 registers 20004 20005commit c8dad98abb042c6abbbee18f9ae4db72084bc513 20006Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20007Date: Wed Aug 15 00:55:24 2007 -0400 20008 20009 RADEON: make sure crtc routing is correct in r300_detect_tv() 20010 20011 Also remove some unused variable 20012 20013commit f8cd74435f0072dbf5f6e83d67d2d5e1f4e82c91 20014Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20015Date: Tue Aug 14 23:50:57 2007 -0400 20016 20017 RADEON: minor cleanup 20018 20019commit 6f398cd07ea734dd66a8eac71b629e59123d75b8 20020Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20021Date: Tue Aug 14 23:42:32 2007 -0400 20022 20023 RADEON: Implement improved tv load detection for r300 20024 20025 The previous implementation resulted in false positives 20026 on occasion. This method works much more reliably. 20027 Based on beos code by Thomas Kurschel 20028 20029commit 366a1d4c240ac93622caff97b652696db99bf2e6 20030Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20031Date: Tue Aug 14 23:25:37 2007 -0400 20032 20033 RADEON: Turn off tv encoder in disableoutputs() 20034 20035commit 5b4a04c23e3f1ec2490418b111f417c16463c709 20036Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20037Date: Tue Aug 14 23:11:52 2007 -0400 20038 20039 RADEON: Make sure RMX is always programmed for crtc1 20040 20041 If the RMX registers are left as programmed by the bios 20042 this can lead to a blank screen when crtc1 is feeding a 20043 DAC. Fix found and reported by Lisa Wu <liswu@ati.com> 20044 Fixes bug 11985 20045 20046commit db2a828b2f21b92cd654b309d137204334975b89 20047Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20048Date: Tue Aug 14 22:47:29 2007 -0400 20049 20050 RADEON: Add load detection 20051 20052 Based on the beos driver by Thomas Kurschel and the 20053 existing load detection code in this driver. 20054 20055commit 6fc3ddbbb6fbbee1f6076c776e5b46c0c772b6d4 20056Author: Alex Deucher <alex@botch2.com> 20057Date: Sun Aug 12 12:00:20 2007 -0400 20058 20059 RADEON: order the VGA and DVI ports correctly 20060 20061commit b7738d2ef82e1759adf78e5db1291f8739b4166f 20062Author: Alex Deucher <alex@botch2.com> 20063Date: Sat Aug 11 17:50:42 2007 -0400 20064 20065 RADEON: fix connector setup when there's no bios tables 20066 20067commit d0895f67e327bb268fd59fcfd8fc22678d804f57 20068Author: Luc Verhaegen <libv@skynet.be> 20069Date: Thu Aug 9 12:11:40 2007 +0200 20070 20071 AGPFastWrite risk reduction. 20072 20073 Actively warn the user in the log about the effects of AGPFastWrite and 20074 sanitise AGPFastWrite handling while we're here. 20075 20076commit c01000bc684f6a23a38a52f0808182cefaa686bb 20077Author: Alex Deucher <alex@botch2.com> 20078Date: Wed Aug 8 00:07:16 2007 -0400 20079 20080 RADEON: fix typo in comment 20081 20082commit 5c3f49e651c36f3bd14fa29359e24825d8f7f77f 20083Author: Lisa Wu <liswu@ati.com> 20084Date: Wed Aug 8 00:05:47 2007 -0400 20085 20086 RADEON: make sure RADEON_CRTC2_CRT2_ON bit state is properly accounted for 20087 20088 - when the TV DAC is used for crtc1 make sure to keep this bit set when 20089 initializing crtc2. 20090 - fixes bug 11894 20091 20092commit 08fe7ad00fba523775e95b5e6295fe23a5119d60 20093Author: Alex Deucher <alex@botch2.com> 20094Date: Tue Aug 7 23:16:05 2007 -0400 20095 20096 RADEON: fix ConnectorTable option after connector table re-work 20097 20098 We may want to add the possiblity of more connectors to this option 20099 20100commit f56b90a60393ed187c0e39e149b2a3a32331ea52 20101Author: Alex Deucher <alex@botch2.com> 20102Date: Tue Aug 7 23:08:02 2007 -0400 20103 20104 RADEON: Make sure the default TV standard is supported 20105 20106commit a88a0e77eec514325b4d07bf7be9fb04e6f8e244 20107Author: Alex Deucher <alex@botch2.com> 20108Date: Tue Aug 7 00:43:20 2007 -0400 20109 20110 RADEON: minor fix to legacy bios connector table 20111 20112commit 3752808d6c08a9727370ef8d79088e787791e131 20113Author: Alex Deucher <alex@botch2.com> 20114Date: Sun Aug 5 16:13:39 2007 -0400 20115 20116 RADEON: Always set MonType in radeon_detect() 20117 20118 - Always set the the MonType in radeon detect even if no 20119 monitor is detected so users can still force disconnected 20120 outputs on. For DVI-I users will have to set the 20121 dvi_monitor_type attribute to force digital or analog, 20122 everything else should just work. 20123 20124commit b4ec3e436afb5bc99ec755cbd96eee9a1ec492a5 20125Author: Alex Deucher <alex@botch2.com> 20126Date: Sun Aug 5 15:37:04 2007 -0400 20127 20128 RADEON: add tv out properties 20129 20130commit 1fc2a1120e7c05938e2bd72d3c7837ecff8bc9da 20131Author: Alex Deucher <alex@botch2.com> 20132Date: Sun Aug 5 14:41:57 2007 -0400 20133 20134 RADEON: switch DVI-I monitor type attribute to string 20135 20136commit 7d8eb3751d74bd8a1fb9fa2d2fcb9c4c895f6ba4 20137Author: Alex Deucher <alex@botch2.com> 20138Date: Sun Aug 5 14:31:18 2007 -0400 20139 20140 RADEON: make backlight attribute lower case 20141 20142commit e71bb88bc9052af8866fb6945dbc06dbb0e6d1c0 20143Author: Alex Deucher <alex@botch2.com> 20144Date: Sun Aug 5 14:26:15 2007 -0400 20145 20146 RADEON: switch RMX attribute to string type 20147 20148 Still not actually hooked up. 20149 20150commit 284323135ec04635dfa9cabd5790a35b953abca2 20151Author: Alex Deucher <alex@botch2.com> 20152Date: Sun Aug 5 04:35:24 2007 -0400 20153 20154 RADEON: typo 20155 20156commit d1abdad167aa24ac970c69422435df443c82ebd6 20157Author: Alex Deucher <alex@botch2.com> 20158Date: Sun Aug 5 03:45:02 2007 -0400 20159 20160 RADEON: fixes 20161 20162 - fix output ordering 20163 - set tv-out to return un-connected for now in radeon_detect() 20164 20165commit 0cca25d8d6a0cb0d29b68e6cd9c699d1390aede1 20166Author: Alex Deucher <alex@botch2.com> 20167Date: Sun Aug 5 03:19:24 2007 -0400 20168 20169 RADEON: Fix DVI-I support in ATOM bios connector table parsing 20170 20171commit ba5496ae7973786802962bf649dd91c219531749 20172Author: Alex Deucher <alex@botch2.com> 20173Date: Sun Aug 5 02:27:32 2007 -0400 20174 20175 RADEON: refactor output init to handle multiple DVI or VGA 20176 20177 - refactor output init to handle multiple DVI or VGA with the new 20178 bios table parsing 20179 20180commit 2ec22783ddf4c522df9e5fd1b2003854486d7a2b 20181Author: Alex Deucher <alex@botch2.com> 20182Date: Sun Aug 5 01:39:35 2007 -0400 20183 20184 RADEON: attempt to do the right thing for standards other than PAL or NTSC 20185 20186commit cc8e1d95f1b90a259beea4e8cc4d7e29af660919 20187Author: Alex Deucher <alex@botch2.com> 20188Date: Sun Aug 5 01:14:36 2007 -0400 20189 20190 RADEON: Major rework of BIOS table parsing 20191 20192 - greatly simplify ATOM and legacy connector table parsing 20193 - use bios tables to detect LVDS and TV outputs 20194 - add support for TV table parsing (legacy only) 20195 20196commit 1de52d91ff3a04b9b587b858e1e5be40d3a7fd0a 20197Author: Dave Airlie <airlied@linux.ie> 20198Date: Sat Aug 4 17:58:58 2007 +1000 20199 20200 update configure.ac for 6.6.193 release 20201 20202commit 371001c0433db1d17e468f3ea99ea57f922145a3 20203Author: Dave Airlie <airlied@linux.ie> 20204Date: Sat Aug 4 17:51:19 2007 +1000 20205 20206 radeon: remove unused variables 20207 20208commit 90946c53d7a4b23d03270ad6da0450759a11de3d 20209Author: Dave Airlie <airlied@linux.ie> 20210Date: Sat Aug 4 17:51:06 2007 +1000 20211 20212 ati: add -Wall for gcc 20213 20214 As per luc's "suggestion" in Novell bug. 20215 20216 This may generate warnings on 64-bit until we get rid of CARD32 20217 20218commit e30a145934df8f6a7f71290d6c75e4239f9d52f7 20219Author: Dave Airlie <airlied@linux.ie> 20220Date: Sat Aug 4 17:44:46 2007 +1000 20221 20222 radeon: cleanup some pieces of the dpms/blank register programming 20223 20224commit a156db5e8b037ed12a448f70045453baf9d0c504 20225Author: Luc Verhaegen <libv@skynet.be> 20226Date: Sat Aug 4 17:37:18 2007 +1000 20227 20228 Clean up PortInfo to CRTC mapping 20229 20230 Also sanitise blanking and DPMS functions 20231 20232 Fixes from Novell Bug 264720, and fd.o 10772 20233 20234commit b61a49f2a5401560f85e11bcdd005287433cad12 20235Author: Alex Deucher <alex@botch2.com> 20236Date: Fri Aug 3 19:27:59 2007 -0400 20237 20238 RADEON: Change indexing of TV constants table in preparation for standard re-work 20239 20240commit 288fa627274cb399059262d4f8bd844fc220a042 20241Author: Alex Deucher <alex@botch2.com> 20242Date: Thu Aug 2 02:37:16 2007 -0400 20243 20244 RADEON: avoid a divide by 0 and only save tv out regs if the chip has them 20245 20246commit b66a1bc7994b33d349c1519761e431959311c85f 20247Author: Alex Deucher <alex@botch2.com> 20248Date: Thu Aug 2 02:11:20 2007 -0400 20249 20250 RADEON: fix tv-out enable/disable 20251 20252commit d86592c8d5ce45d81d8a726c263e870e94fbcf11 20253Author: Alex Deucher <alex@botch2.com> 20254Date: Thu Aug 2 00:50:51 2007 -0400 20255 20256 RADEON: add missing break 20257 20258commit 971feb34843225030fff05b3f9d3801534fbf2d4 20259Author: Alex Deucher <alex@botch2.com> 20260Date: Thu Aug 2 00:50:04 2007 -0400 20261 20262 RADEON: move tv dac enable to enabledisplay() 20263 20264commit 98d7e00437bea78e03180eb30ff30de3455d9d1a 20265Author: Alex Deucher <alex@botch2.com> 20266Date: Thu Aug 2 00:20:50 2007 -0400 20267 20268 RADEON: limit tv modes to the only one we can program at the moment 20269 20270commit 5c549c1d42f7bbc556942af13aff2661fae856f2 20271Author: Alex Deucher <alex@botch2.com> 20272Date: Wed Aug 1 23:45:07 2007 -0400 20273 20274 RADEON: remove unused elements 20275 20276commit b03978028fd975eb6946503d3a56a49c5a67f339 20277Author: Alex Deucher <alex@botch2.com> 20278Date: Wed Aug 1 22:39:16 2007 -0400 20279 20280 RADEON: convert hard coded tv out values to calculations 20281 20282commit cf54222f1fa37366b2c2b39c82f8afc02f32e63c 20283Author: Alex Deucher <alex@botch2.com> 20284Date: Tue Jul 31 02:01:49 2007 -0400 20285 20286 RADEON: more fixes... 20287 20288commit 22d460d3ad991223aa1fbd7e5edeb45e36c65dc0 20289Author: Alex Deucher <alex@botch2.com> 20290Date: Tue Jul 31 01:28:05 2007 -0400 20291 20292 RADEON: fix name of tv output 20293 20294commit 4822a2b837334f408f962646ab5ea4f8b0335ac9 20295Author: Alex Deucher <alex@botch2.com> 20296Date: Tue Jul 31 01:18:40 2007 -0400 20297 20298 RADEON: tv-out fixes. works now. tested on rv350. 20299 20300 VT siwtch is busted, and xrandr doesn't play nice yet. 20301 uncomment code in radeon_output.c to test. 20302 20303commit fe494c9db2995bb8ce7a028ecf9626e0cb0cf506 20304Author: Alex Deucher <alex@botch2.com> 20305Date: Sun Jul 29 15:26:34 2007 -0400 20306 20307 RADEON: add info about tv out code and authorship 20308 20309commit 6b9b7a7bdc290d07de9b226691ec8025af8db896 20310Author: Alex Deucher <alex@botch2.com> 20311Date: Sun Jul 29 15:23:14 2007 -0400 20312 20313 RADEON: Initial pass at integrated tv out support 20314 20315 Based on the GATOS tv-out support by Federico Ulivi <fulivi@lycos.com> 20316 and information from ati with substantial rework by myself. 20317 20318 Code is not actually hooked up yet. 20319 20320commit 8d043db1817d94edeb72ab208dfea60026715d48 20321Author: Alex Deucher <alex@botch2.com> 20322Date: Wed Jul 25 20:37:58 2007 -0400 20323 20324 RADEON: Compute PLL VCO gain 20325 20326commit 62f06d89da3f7160d5e4df8d7ce6fe1a94e9d07c 20327Author: Alex Deucher <alex@botch2.com> 20328Date: Wed Jul 25 20:22:25 2007 -0400 20329 20330 RADEON: write out saved vclk and pixclk values 20331 20332commit 165a07cbbfcd94e3d1fac434b8fada8d29428a09 20333Author: Brice Goglin <Brice.Goglin@ens-lyon.org> 20334Date: Wed Jul 25 19:01:02 2007 +0200 20335 20336 Minor fixes in the manpages 20337 20338 - ati.man 20339 . add references to radeon and r128 manpages in SEE ALSO 20340 (Xavier Bestel in Debian bug #386001). 20341 . remove the manpage suffix from atimisc reference in the text 20342 since there is no such manpage for now. 20343 - radeon.man 20344 . typo in "specifying" (reported by A. Costa in Debian bug #432059). 20345 . replace a non-ascii character with the corresponding groff escape 20346 sequence so that it works whatever the locale (Julien Cristau). 20347 20348commit 9cc3ab8320162f371bba15dc131f23c5de2013fc 20349Author: Alex Deucher <alex@botch2.com> 20350Date: Wed Jul 25 02:24:14 2007 -0400 20351 20352 RADEON: hack around console restore hang on r4xx 20353 20354 r4xx seems to hang when unblanking the crtc(s) when 20355 restoring the console. This needs to be investigated 20356 further. 20357 20358commit 193d58b0ea9de3e326017e50b88e2f3726854506 20359Author: Alex Deucher <alex@botch2.com> 20360Date: Wed Jul 25 01:27:58 2007 -0400 20361 20362 RADEON: fix randr on r4xx 20363 20364 Seems the way we parse the ATOM connector table results 20365 in reversed connectors. 20366 20367commit 262e32bd9ea4123b116362b33b5798753b2568fe 20368Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20369Date: Thu Jul 19 23:48:57 2007 -0400 20370 20371 RADEON: fix palette when depth != 24 20372 20373commit 18b00b47a483e7854727e99126808ddf361e7a4a 20374Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20375Date: Thu Jul 19 19:01:42 2007 -0400 20376 20377 R128: don't clip modes to panel on laptops when only using the CRT port 20378 20379 fixes bug 5832 20380 20381commit 882fe7631586b0a7919f808588a2ea4fb555f7e8 20382Author: Brice Goglin <brice.goglin@ens-lyon.org> 20383Date: Thu Jul 19 18:44:27 2007 -0400 20384 20385 MACH64: avoid crash in Xv code 20386 20387 fixes bug 11054 20388 20389commit 7b3e22e96cfc98621bd20fa76317e6d8f7242165 20390Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20391Date: Thu Jul 5 20:37:10 2007 -0400 20392 20393 RADEON: set default values for crtc_offset_cntls 20394 20395commit 2a35ccfb8ffee9f0ef6f8d4f0eeb80a471543caa 20396Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20397Date: Thu Jul 5 20:04:04 2007 -0400 20398 20399 RADEON: Fix tiling on r1xx and r2xx 20400 20401 spotted by Andrew Randrianasulu <randrik@mail.ru> 20402 fixes bug 11357 20403 20404commit ac9cbaf32176cf144bc694ac879e7c9e2920f762 20405Author: Michel Dänzer <michel@tungstengraphics.com> 20406Date: Mon Jul 2 10:06:02 2007 +0200 20407 20408 radeon: Fully zero-initialize info->CRT2pScrn->monitor. 20409 20410 This prevents modes from getting rejected due to the uninitialized maxPixClock 20411 value. 20412 20413 Fixes http://bugs.freedesktop.org/show_bug.cgi?id=9816 . 20414 20415commit bdcae622100c81a4d9a53938542b64908bacd195 20416Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20417Date: Fri Jun 29 01:15:57 2007 -0400 20418 20419 RADEON: save crtc on/off state for console restore 20420 20421 - prevents a possible hang if console is only using 1 crtc 20422 20423commit 21be0d3a1f4700572c6425800596785d6850626f 20424Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20425Date: Fri Jun 29 00:38:50 2007 -0400 20426 20427 RADEON: simplify console restore 20428 20429 - still need to track crtcs for blank/unblank when restoring text console 20430 20431commit d9bf28b55d98fea2d285c9c46362aaf2175d0a46 20432Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20433Date: Thu Jun 28 23:52:28 2007 -0400 20434 20435 RADEON: factor out surface_cntl init into one function 20436 20437commit 9f193985627be8e6ea1418a424e825ddbc4957b2 20438Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20439Date: Thu Jun 28 23:43:13 2007 -0400 20440 20441 RADEON: move crtc base setups to new functions 20442 20443commit 0f361e9e80a29d287fa42436c32c657e3c102539 20444Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20445Date: Thu Jun 28 23:08:07 2007 -0400 20446 20447 RADEON: fix corruption after 3D apps run 20448 20449commit 1d4630067d293d26284d5fe230debef3913ec6cf 20450Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20451Date: Thu Jun 28 22:24:04 2007 -0400 20452 20453 RADEON: set info->CurrentLayout.displayWidth in RADEONSetPitch() 20454 20455commit 2754d1ba01fc3367019487e0c0f59d74c950aaa7 20456Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20457Date: Thu Jun 28 22:15:22 2007 -0400 20458 20459 RADEON: Switch cursor back to memcpy() 20460 20461commit 7901bcafa92dccd319ddb5de4627d806a39f15f9 20462Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20463Date: Mon Jun 25 18:39:14 2007 -0400 20464 20465 RADEON: make sure we unblank in RADEONRestore() 20466 20467 reported by Andrew Randrianasulu (bugs 11357, 11146) 20468 20469commit 6a3599d6155d073a3116c4b13bdf04b44bb9e087 20470Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20471Date: Mon Jun 25 18:20:54 2007 -0400 20472 20473 RADEON: make sure we set the stride properly 20474 20475commit 771c37fcac8104b894120cc3fc0154b31b0318d1 20476Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20477Date: Mon Jun 25 17:53:26 2007 -0400 20478 20479 RADEON: attempt to fix cursor on big endian (only tested on x86) 20480 20481commit 5b5b90c2cea7e36895354f5872acd3fc769653f9 20482Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20483Date: Mon Jun 25 16:27:54 2007 -0400 20484 20485 RADEON: fix possible segfault on mobility chips with MM tables 20486 20487 spotted by Stefan Buehler 20488 20489commit 01bd5eb713500d3e3d4351865d460c8d1e476454 20490Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20491Date: Mon Jun 25 16:16:18 2007 -0400 20492 20493 RADEON: fix possible segfault on mobility chips with MM tables 20494 20495 reported by Stefan Buehler 20496 20497commit 52cc1dc1491559a9055f3ba6dd54064bb382ad86 20498Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20499Date: Mon Jun 25 15:49:19 2007 -0400 20500 20501 RADEON: tiling and FB size fixes 20502 20503 - fix tiling on older radeons 20504 - allow the user to specify FB size using Virtual line in config 20505 Current default is 1600x1200 20506 - reset ecp_div for Xv in randr crtc mode set 20507 20508commit 3bb46c03fddd3bd79bdab887366aeeced0290a3a 20509Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20510Date: Fri Jun 22 01:12:38 2007 -0400 20511 20512 RADEON: bump crtc max to 1200 so 1600x1200 will work out of the box 20513 20514 - a better fix will be forthcoming... 20515 20516commit b9fac5695bf8f123b391e3b162c4865258a685dd 20517Author: Andrew Randrianasulu <randrik@mail.ru> 20518Date: Fri Jun 22 01:01:14 2007 -0400 20519 20520 RADEON: Fix VT switch hangs 20521 20522 - re-order DRI resume functions in EnterVT() 20523 - fixes bug 11287 20524 20525commit 92e65d5e0d6817ff4c9a08020a0a9b3a8c3c98b0 20526Author: Andrew Randrianasulu <randrik@mail.ru> 20527Date: Fri Jun 22 00:55:00 2007 -0400 20528 20529 RADEON: Fix VT switch hangs 20530 20531 - reorder RADEONDRISetVBlankInterrupt() and RADEONDRIResume() 20532 - see bug 11287 20533 20534commit ec1fd2c388ec2c4033644776ee8588405a3d25f1 20535Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20536Date: Tue Jun 19 00:39:59 2007 -0400 20537 20538 RADEON: Split FP reg restore to only touch regs relevant to the output 20539 20540 - only restore output specific regs 20541 - restore fp2_gen_cntl on R200 DAC restore as R200 DAC2 is connected via 20542 DVO 20543 20544commit 0f5a92667281ca3f80218bc67d4031125a5e4793 20545Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20546Date: Mon Jun 18 23:35:21 2007 -0400 20547 20548 RADEON: only touch LVDS on mobility chips 20549 20550commit b72ff160f908bf3aa9f64705377e92d80360a4f7 20551Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20552Date: Mon Jun 18 21:51:45 2007 -0400 20553 20554 RADEON: only touch LVDS on mobility chips 20555 20556commit 80313621ffa8c409ae63fc1b28c15fc4abdc3a7b 20557Author: Henry Zhao <henryz@localhost.localdomain> 20558Date: Fri Jun 15 17:00:05 2007 -0700 20559 20560 Update CRT2pScrn->monitor->Last to reflect the 20561 last mode of CRT2pScrn. See bug 11278. 20562 20563commit 5868af53204e8fc210735e68e891e15085285629 20564Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20565Date: Tue Jun 12 23:53:48 2007 -0400 20566 20567 RADEON: yet another LVDS... 20568 20569commit f19a6f7ee5bf4ec632e7813359f167599c08e823 20570Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20571Date: Tue Jun 12 23:44:41 2007 -0400 20572 20573 RADEON: additional LVDS off fix (missed in previous commit) 20574 20575commit 4f2a1ba9c2fc20042d67132f986a86a9783a245e 20576Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20577Date: Tue Jun 12 23:41:24 2007 -0400 20578 20579 RADEON: additional LVDS off fix (missed in previous commit) 20580 20581commit 9fc416ce08cc5d74faa7b184e80ce43ed2f4fbee 20582Author: Michel Dänzer <michel@tungstengraphics.com> 20583Date: Tue Jun 12 23:35:09 2007 -0400 20584 20585 radeon: Fix panel size detection from registers with stretched mode programmed. 20586 20587commit c4045cbc5e6975520efef422ce9df2a7d99a7ff4 20588Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20589Date: Tue Jun 12 23:32:18 2007 -0400 20590 20591 RADEON: make sure to clear RADEON_LVDS_ON when turning LVDS off 20592 20593 - fixes bug 3483 20594 20595commit 090888d3baf61a70640d28e99afdce20952cdea0 20596Author: Tormod Volden <bugzi06.fdo.tormod@xoxy.net> 20597Date: Tue Jun 12 23:30:18 2007 -0400 20598 20599 RADEON: limit PanelPwrDly to 2000 ms on ATOM bios as per legacy bios 20600 20601 - fixes bug 11238 20602 20603commit 44748a826b52924a6e050215d8d62755e0ac7fb2 20604Author: Tormod Volden <bugzi06.fdo.tormod@xoxy.net> 20605Date: Tue Jun 12 23:23:49 2007 -0400 20606 20607 RADEON: limit PanelPwrDly to 2000 ms on ATOM bios as per legacy bios 20608 20609 - fixes bug 11238 20610 20611commit c292fc64499ff4cc135c07deda99cf4169f8fef4 20612Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20613Date: Tue Jun 12 23:20:18 2007 -0400 20614 20615 RADEON: make sure RADEON_LVDS_ON is cleared when turning LVDS off 20616 20617 - fixes bug 3483 20618 20619commit c6a3286d6f6ddda89115d98d45665dadd78bf41d 20620Author: Michel Dänzer <michel@tungstengraphics.com> 20621Date: Mon Jun 11 09:39:38 2007 +0200 20622 20623 radeon: Fix panel size detection from registers with stretched mode programmed. 20624 20625commit 7a6b3c0f77c69019268f585c51c3dcbcc99014d4 20626Author: Michel Dänzer <michel@tungstengraphics.com> 20627Date: Mon Jun 11 09:39:38 2007 +0200 20628 20629 radeon: Don't loop indefinitely if no mode matches detected panel size. 20630 20631commit f31fd9ce598841c505a0b5ed32bf124f49fea332 20632Author: Michel Dänzer <michel@tungstengraphics.com> 20633Date: Mon Jun 11 09:39:38 2007 +0200 20634 20635 radeon: Fix some more ErrorFs when setting mode. 20636 20637commit 9dd9f09a9021653e5590d27bd7aa0e7af85416ad 20638Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20639Date: Thu Jun 7 15:32:31 2007 -0400 20640 20641 RADEON: fix some issues with bios table init code 20642 20643 - fix some issues with bios table init code 20644 - re-org preinit to better handle bios table init of cards 20645 20646commit 9ad311eeb0ac2b70a862dbb9de278154ba9142ec 20647Author: Tilman Sauerbeck <tilman@code-monkey.de> 20648Date: Thu Jun 7 11:03:02 2007 +0200 20649 20650 Fixed 'make dist'. 20651 20652commit 2527f2b69aa7dffa3ba4359c45955e3185bdf0e6 20653Author: Alex Deucher <alex@botch2.com> 20654Date: Wed Jun 6 00:08:45 2007 -0400 20655 20656 RADEON: implement support for posting cards based on x86 bios tables (untested) 20657 20658 This is based on the netbsd radeonfb driver by Garrett D'Amore. 20659 The code is not hooked up yet, but should allow you to post cards 20660 without needing to execute bios code. This should be useful on 20661 non-x86 platforms, for posting secndary cards, and for suspend/resume. 20662 Works on legacy bioses only (no ATOM support). 20663 20664commit e67d1420bf65055ecb6fdfe6b1b1f53aae83854a 20665Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20666Date: Tue Jun 5 22:31:50 2007 -0400 20667 20668 RADEON: fixup RADEONDisplayVideo() to better handle the crtc being used 20669 20670commit 7886405308e3288d5c86b6f2c7dbfa8ff865139c 20671Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20672Date: Mon Jun 4 21:16:56 2007 -0400 20673 20674 RADEON: fix indenting 20675 20676commit f54ad565a69deb52547fd04e123f56fc1294cd0a 20677Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20678Date: Mon Jun 4 21:13:16 2007 -0400 20679 20680 RADEON: no need to go through all the crtcs, we've got what we need 20681 20682commit 687879bff716ad01f9f158860deb8ba770faab99 20683Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20684Date: Mon Jun 4 21:08:40 2007 -0400 20685 20686 RADEON: add support for RMX on DVI, make sure RMX is only enabled on crtc1 20687 20688commit e6161e472ff266f69547704a61040228a8704b06 20689Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20690Date: Mon Jun 4 20:32:55 2007 -0400 20691 20692 RADEON: remove some cruft 20693 20694commit be0ce38232ea4f5679c5829ab663939144c6e617 20695Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20696Date: Mon Jun 4 20:22:15 2007 -0400 20697 20698 RADEON: number the outputs if there are more than one of the same 20699 20700commit 8ec617f6493dd0aea5d11f92e3d58c3feef8c8fd 20701Author: Dave Airlie <airlied@linux.ie> 20702Date: Sun Jun 3 17:32:43 2007 +1000 20703 20704 radeon: disable irqs at server start until 3D app starts 20705 20706commit 2d40fa55e8d7a1cfb204d66ca4a4d95a3b13d5b5 20707Author: Dave Airlie <airlied@linux.ie> 20708Date: Sun Jun 3 17:32:43 2007 +1000 20709 20710 radeon: disable irqs at server start until 3D app starts 20711 20712commit dcb64a4d3947e5a9fbda4b72e29a5b6102370f07 20713Author: Dave Airlie <airlied@linux.ie> 20714Date: Sun Jun 3 17:10:49 2007 +1000 20715 20716 radeon: disable vbl interrupts when no 3d is running on a new enough drm 20717 20718commit 73d8e3ec8536b4777490b7ba457566f02233811f 20719Author: Dave Airlie <airlied@linux.ie> 20720Date: Sun Jun 3 17:07:59 2007 +1000 20721 20722 randr-1.2 increase dri minor version for mesa to know we can do vbl on both 20723 20724commit 63f0d4ed0a98830ecbe18c6e4174689111a59b68 20725Author: Dave Airlie <airlied@linux.ie> 20726Date: Sun Jun 3 16:49:14 2007 +1000 20727 20728 randr-1.2: add support for vblank on both heads 20729 20730 Also disable vbl when 3D isn't running 20731 20732commit d7775c1b38b5bdc439a27ec2c3c3a03a5b24cf57 20733Author: Dave Airlie <airlied@linux.ie> 20734Date: Sun Jun 3 12:46:31 2007 +1000 20735 20736 randr-1.2: make native mode preferred 20737 20738 If we have a native mode from the BIOS make it preferred mode 20739 20740commit 5a9516fe4f17854acd4fbf3a8eadf5139081dbd4 20741Author: Paul TBBle Hampson <Paul.Hampson@pobox.com> 20742Date: Sat Jun 2 14:28:18 2007 -0400 20743 20744 RADEONProbePLLParameters sets pll->reference_div, and 20745 RADEONGetPanelInfoFromReg uses it. 20746 20747 This ensures that the former is called before the latter. 20748 (this should fix randr on ppc) 20749 20750commit 802804461ebdca9a951e7e562ec68fd08d8eae01 20751Author: Matthieu Herrb <matthieu@deville.herrb.com> 20752Date: Thu May 31 22:23:57 2007 -0600 20753 20754 Fix build whithout XF86DRI 20755 20756commit 0e0946e0aa2527794e07473199e851bbfbc47cb1 20757Author: Michel Dänzer <michel@tungstengraphics.com> 20758Date: Wed May 30 18:56:53 2007 +0200 20759 20760 radeon: Lots of warning fixes. 20761 20762 Move code where it's used, remove unused variables, etc. 20763 20764commit 31c1be420d5277dd15505bd73e6144827a0580cd 20765Author: Dave Airlie <airlied@nx6125b.(none)> 20766Date: Wed May 30 17:49:01 2007 +1000 20767 20768 remove these syncs, at least on rs480 it doesn't break 20769 20770 probably requires testing on other r300 based cards, with the syncs in 20771 we hang when moving the cursor into the second CRTC. 20772 20773commit 7fc02657c4d740941fbda5a8823cf45de3eca3f8 20774Merge: 800bf53 4c61c0e 20775Author: Dave Airlie <airlied@nx6125b.(none)> 20776Date: Wed May 30 17:27:22 2007 +1000 20777 20778 Merge branch 'origin' into randr-1.2-test 20779 20780 Conflicts: 20781 20782 src/radeon_cursor.c 20783 src/radeon_display.c 20784 src/radeon_driver.c 20785 20786commit 800bf53279e2c2bf854682bbfd6fa16d03afed00 20787Author: Matthieu Herrb <matthieu@deville.herrb.com> 20788Date: Tue May 29 23:31:13 2007 -0600 20789 20790 Fix build without XF86DRI 20791 20792commit 4c61c0ee91a2ffeefce30972a584486f1df1d1ae 20793Author: Matthieu Herrb <matthieu@deville.herrb.com> 20794Date: Tue May 29 21:35:35 2007 -0600 20795 20796 Fix build without XF86DRI 20797 20798commit 5337e7bd0069a3f2c4ab22b21a19471427ad3d81 20799Author: Dave Airlie <airlied@nx6125b.(none)> 20800Date: Wed May 30 08:10:44 2007 +1000 20801 20802 radeon: add bios quirk for nx6125 monid 20803 20804commit bff809dc8ed07ac39e9b576a87916486a5e37156 20805Author: Dave Airlie <airlied@nx6125b.(none)> 20806Date: Wed May 30 08:02:26 2007 +1000 20807 20808 rs480: more unknown regs 20809 20810 Hardcode the values from a working fglrx run, this works for me now 20811 20812 I've no idea what it might do for anyone else 20813 20814commit 104105fee5c3945d3f210e6a4cb73ab492c61543 20815Author: Dave Airlie <airlied@nx6125b.(none)> 20816Date: Tue May 29 19:09:33 2007 +1000 20817 20818 rs480: make second crtc work with magic number in magic register. 20819 20820 I've no idea why or what this does. 20821 20822commit 5aa603bcabbb077dec169c48438c2e2ebe1195d7 20823Author: Dave Airlie <airlied@nx6125b.(none)> 20824Date: Tue May 29 07:23:24 2007 +1000 20825 20826 rs480: only has single dac 20827 20828commit c52322354fe64725733842b3356798c50e7735d5 20829Merge: dd6a966 975da59 20830Author: Dave Airlie <airlied@nx6125b.(none)> 20831Date: Tue May 29 07:21:48 2007 +1000 20832 20833 Merge branch 'origin' 20834 20835commit c2637a01f9aca4032262c66ade305f5fe2c54294 20836Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20837Date: Sat May 26 18:26:35 2007 -0400 20838 20839 RADEON: remove some dead code from the last commit 20840 20841commit bbb769c4107bfcae682e46d026e54cbfb67d62cd 20842Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20843Date: Sat May 26 18:25:06 2007 -0400 20844 20845 RADEON: more rotation work (still not there) 20846 20847 - once again borrowed heavily from intel 20848 20849commit a69f90be9384244744fabfa76469ede9cd26ac98 20850Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20851Date: Sat May 26 17:54:48 2007 -0400 20852 20853 RADEON: first pass at rotation (not working yet) 20854 20855 - based heavily on intel 20856 20857commit 3b619d88a65a5801c85d120f2d39704194433f10 20858Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20859Date: Sat May 26 17:06:18 2007 -0400 20860 20861 RADEON: clean up and add comment regarding clones 20862 20863commit 089ae4afd23322bf602d340c531c36a4c8b461ed 20864Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20865Date: Sat May 26 15:49:11 2007 -0400 20866 20867 RADEON: Switch SetOutputType() to use names rather than numbers 20868 20869commit c4cef0c9481257a744ac99dfc7beb988ce51a8ee 20870Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20871Date: Sat May 26 13:12:37 2007 -0400 20872 20873 RADEON: make sure we have the lock when SYNCing 20874 20875commit 899c62e987d5c7524817ba85675e39a8a28e0232 20876Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20877Date: Sat May 26 12:54:01 2007 -0400 20878 20879 RADEON: Make sure LVDS and FP2 routing info gets written 20880 20881commit 18857184ffa6847815d349c020b003f8401e36ee 20882Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20883Date: Sat May 26 12:27:00 2007 -0400 20884 20885 RADEON: fix handling of DRI lock 20886 20887commit 3a61453efb4f04492cef823b6dd1273b55c6a785 20888Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20889Date: Sat May 26 01:38:09 2007 -0400 20890 20891 RADEON: implement backlight control for LVDS 20892 20893 This code is currently disabled as I'm not sure which 20894 laptops actually use this method for backlight control. 20895 My laptop seems to use another method as adjusting the 20896 backlight level doesn't seem to touch LVDS_GEN_CNTL. 20897 Maybe just macs? 20898 20899commit ceec3f62257bafe4771e75d3c4f1d2a517d7acf8 20900Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20901Date: Fri May 25 23:29:51 2007 -0400 20902 20903 RADEON: implement auto/analog/digital output property for DVI-I (untested) 20904 20905commit e3e9c608651e7cbb9851489274815f47d2a8dbbb 20906Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20907Date: Fri May 25 19:26:24 2007 -0400 20908 20909 RADEON: more clean of last commits 20910 20911commit a7d2c9c0f6155c4e65a2e73f3832c0d8ca7af5fc 20912Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20913Date: Fri May 25 19:09:11 2007 -0400 20914 20915 RADEON: fixup last commit and add support for ext TMDS detect 20916 20917commit e50ca35ce2ea79dadb38ce14e459eed836452ff7 20918Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20919Date: Fri May 25 18:56:32 2007 -0400 20920 20921 RADEON: clean up non-DDC probe, add detect for DVI 20922 20923commit abb9b57e7b4162ab6c21fd1e809d24f13f7e1ea1 20924Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20925Date: Wed May 23 21:37:58 2007 -0400 20926 20927 RADEON: clean up some logic 20928 20929commit 679236defd28bea8874004377436347950ff7bec 20930Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20931Date: Wed May 23 21:18:40 2007 -0400 20932 20933 RADEON: add output properties (not functional yet) 20934 20935commit 975da595f032c145ad74079ff8edeaead779dc7b 20936Author: Michel Dänzer <michel@tungstengraphics.com> 20937Date: Tue May 22 10:56:47 2007 +0200 20938 20939 radeon: Provide new DRI texOffsetStart hook when available with EXA. 20940 20941commit e38cad5633c974467f6417051eda8d487ac3686e 20942Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20943Date: Tue May 22 00:09:46 2007 -0400 20944 20945 RADEON: Only add valid connectors 20946 20947commit 8275151baac22c34149cef0b7d922771d24abc3e 20948Author: Michel Dänzer <michel@tungstengraphics.com> 20949Date: Mon May 21 10:25:48 2007 +0200 20950 20951 radeon: HW cursor cleanup. 20952 20953 Don't needlessly turn the HW cursor on/off in RADEONLoadCursor*(). 20954 20955 Besides cleaning up the code, this semms to avoid some HW cursor related 3D 20956 lockups, see https://bugs.freedesktop.org/show_bug.cgi?id=10815 . My best 20957 guess is that this is because the engine is now always idled before touching 20958 the CRTC registers. 20959 20960commit 137e3fc1899078af0f72303ab0a4e6cf35804a7b 20961Author: Michel Dänzer <michel@tungstengraphics.com> 20962Date: Mon May 21 10:25:48 2007 +0200 20963 20964 radeon: Suppress debugging output by default. 20965 20966 It can be enabled at runtime by increasing the log verbosity level. 20967 20968 Also change the prefix from (**) to (II) to make grepping the log file for 20969 defaults overridden by xorg.conf more useful again. 20970 20971 Turn some MC related debugging output into normal informational output as it's 20972 useful for recognizing corner cases that can cause stability issues. 20973 20974commit aed193a47a939451d9a6d05b02653b1d73e1d523 20975Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20976Date: Sun May 20 21:13:59 2007 -0400 20977 20978 RADEON: fix crtc1 (un)blanking code after last commit 20979 20980 - crtc1 MC control is on CRTC_GEN_CNTL 20981 - fix indenting 20982 20983commit 870c8043068a0f44b53d1148371b2cc1e3970a7b 20984Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20985Date: Sun May 20 21:03:10 2007 -0400 20986 20987 RADEON: several updates 20988 20989 - move output init from InitCrtc() functions to the output mode_set() 20990 - take the crtc off the MC when blanking 20991 - move EnableDisplay() to output commit() 20992 - clean up some dead code 20993 - don't enable dacs in initcrtc() functions, this is taken care of in EnableDisplay() 20994 20995commit 09bfc8ed000f95ede5b73f2bad69edc1a4d9bac6 20996Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 20997Date: Sun May 20 18:06:22 2007 -0400 20998 20999 update to 6.6.192 for rc release 21000 21001commit 764cb73e8dec4040cdd418d249fc504399fca3ee 21002Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21003Date: Sun May 20 17:26:26 2007 -0400 21004 21005 Fix regular/"xinerama"/zaphod dualhead mode 21006 21007 - logic in RADEONUnblank() was wrong 21008 - Calling RADEONSetupConnectors() on second instance screwed up the port info 21009 - still seem to be HW cursor issues with zaphod mode 21010 21011commit 9a147fef8e0e2ede2a0008c4ecfbd9b00c8dc5f6 21012Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21013Date: Fri May 18 23:45:11 2007 -0400 21014 21015 RADEON: if connector is VGA set TMDS to none 21016 21017commit bbd6faff4c6acb48970d774375c8a61861405f96 21018Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21019Date: Fri May 18 23:30:19 2007 -0400 21020 21021 RADEON: Change default LVDS i2c line for powerpc 21022 21023commit 7940ec364f3cbf02ba64b92c74cbaad4555baf38 21024Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21025Date: Fri May 18 00:23:08 2007 -0400 21026 21027 RADEON: cleanup 21028 21029commit f71bfde7352ef858c1041037d7dc77c237e315a4 21030Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21031Date: Fri May 18 00:12:03 2007 -0400 21032 21033 RADEON: more cleanup 21034 21035commit f711e266d6927dec648f6ff26f15c6f48643f78c 21036Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21037Date: Thu May 17 23:59:07 2007 -0400 21038 21039 RADEON: turn off all outputs in screeninit(). We'll turn on the ones we want later 21040 21041commit bfc1c372d7475b7fa6bffb8a79dec1dc2f11ec59 21042Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21043Date: Thu May 17 23:45:29 2007 -0400 21044 21045 RADEON: More cleanup 21046 21047commit c60b3bc9e3e7463bdb42e2478be8cc3f22c63c68 21048Author: Paul TBBle Hampson <Paul.Hampson@pobox.com> 21049Date: Thu May 17 23:26:54 2007 -0400 21050 21051 Fix for infinite loop in RADEONGetLVDSInfo 21052 21053commit 11289f8206cab1a94bd64a3938cf9af50f19497e 21054Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21055Date: Tue May 15 00:48:31 2007 -0400 21056 21057 RADEON: remove some debugging code 21058 21059commit fba1a11e287ebf04cf311645d31299896bad7283 21060Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21061Date: Tue May 15 00:44:29 2007 -0400 21062 21063 RADEON: Move DAC regs to their own Save() function 21064 21065commit 9ed00e959fcdab2739a7e64ca1e303a6faf28f15 21066Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21067Date: Tue May 15 00:36:05 2007 -0400 21068 21069 RADEON: Lots of small fixes 21070 21071 - remove extra crtc2_base set 21072 - remove some debugging code 21073 - clean up RestoreMode() 21074 - make sure a DDC reg gets assigned for LCD_DDC 21075 - make sure we adjust the right frame in AdjustFrame() 21076 - rename RADEONMapControllers() to RADEONPrintPortMap() to 21077 reflect what it actually does now 21078 - make i2c bus name match the DDC port 21079 - remove or comment out un-needed code 21080 21081commit 5106c8fe5f74865cac8b00937739a4efde9fc254 21082Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21083Date: Sun May 13 21:48:26 2007 -0400 21084 21085 RADEON: Further cleanup 21086 21087commit 224a73e41e9be344d5644203e7ebd5a3a8272604 21088Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21089Date: Sun May 13 21:41:59 2007 -0400 21090 21091 RADEON: remove remnants of "old" multi-head support 21092 21093 The old screen based multi-head code is broken at this point 21094 and I have no intention of fixing it, so to clean things up 21095 I've removed it. 21096 21097commit 9c2f20a83a5fed14225f4c3ebcd8ca41e9d4bd48 21098Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21099Date: Sun May 13 21:13:23 2007 -0400 21100 21101 RADEON: additional cleanups 21102 21103commit e187321ab8dd58d2b2fe92c062d070ba4820a2bf 21104Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21105Date: Sun May 13 19:49:10 2007 -0400 21106 21107 RADEON: add new ConnectorTable option and re-add PanelSize option 21108 21109commit e60a7bcaf5611ad4706b1d1442ce4cae49145f42 21110Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21111Date: Sun May 13 18:14:29 2007 -0400 21112 21113 RADEON: remove mergedfb, etc. cruft from man page 21114 21115commit 51caa5dccc4a2e14d1b41bb7868ad20468dd4cfe 21116Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21117Date: Sun May 13 18:10:02 2007 -0400 21118 21119 RADEON: remove old mergedfb and dualhead options that are no longer used 21120 21121commit c5da9d4040cb08598d171d20f84d3f6c20a033e0 21122Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21123Date: Sun May 13 18:05:38 2007 -0400 21124 21125 RADEON: remove more dead code 21126 21127commit e776fa9ecc53d63f916a5447a76fb2bb3b824167 21128Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21129Date: Sun May 13 15:43:39 2007 -0400 21130 21131 RADEON: re-org randr code 21132 21133 - move crtc stuff to radeon_crtc.c 21134 - move output stuff to radeon_output.c 21135 21136commit 58ce388452b7bc790c438d75c9cf4a0f69f0d7b2 21137Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21138Date: Sun May 13 15:05:01 2007 -0400 21139 21140 RADEON: Remove dead code 21141 21142commit 66e8e6c8348d007930730e90295588efe8108844 21143Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21144Date: Sun May 13 14:25:03 2007 -0400 21145 21146 RADEON: several fixes 21147 21148 - give better DDC names 21149 - disable DAC if connector is DVI-D 21150 - fix indentation in ProbeDDCModes() 21151 21152commit 1c16c2ce9c5b02b03d23da965127d82eea4c4039 21153Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21154Date: Sun May 13 13:54:04 2007 -0400 21155 21156 RADEON: fix VT switch 21157 21158commit 7c66e903368f77ecc4d3bd1c9f08d2adbd85e83c 21159Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21160Date: Sun May 13 13:37:59 2007 -0400 21161 21162 RADEON: move crtc offset handling into init/save/restore functions 21163 21164commit aec078eb0740651fba8ec602e8239bd679efc8ad 21165Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21166Date: Sun May 13 11:57:57 2007 -0400 21167 21168 RADEON: fix up DDCConnected() 21169 21170commit 0550c37ecc434b8075fb3c367d100ff27625bb64 21171Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21172Date: Sun May 13 11:44:50 2007 -0400 21173 21174 RADEON: don't need to pass pScrn to EnableDisplay() 21175 21176commit 117220527de9fd3158f600645bcfcaf46847f45f 21177Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21178Date: Sun May 13 11:37:35 2007 -0400 21179 21180 RADEON: remove hardcoded output limit and PortInfo stuff 21181 21182commit 7e5c29961ac2a9e9dbe5d6d2d73d11cd018d62b5 21183Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21184Date: Fri May 11 18:00:40 2007 +0200 21185 21186 RADEON: Fix RMX after the last commit 21187 21188commit ab5603edd8fc3ef0560bdfb6a6d9c6af2a49d1e5 21189Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21190Date: Fri May 11 17:34:35 2007 +0200 21191 21192 RADEON: Move LVDS, TMDS, DAC properties to the output rec 21193 21194commit 94eb0681de0641e490f06486468617a727fefe86 21195Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21196Date: Fri May 11 14:39:48 2007 +0200 21197 21198 RADEON: switch output dpms to use RADEONEnableDisplay() 21199 21200commit 673ede5578d5d9caf2adf0445fe1e684b034eea5 21201Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21202Date: Fri May 11 14:28:17 2007 +0200 21203 21204 RADEON: re-arrange output mode setting 21205 21206commit 0070a7d787adaae99f7bc2659be4b0f49f439db5 21207Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21208Date: Fri May 11 13:06:15 2007 +0200 21209 21210 RADEON: more re-org 21211 21212 - move crtc mode setting around 21213 - add dri lock/unlock to crtc lock/unlock calls 21214 21215commit bba456232ac9a6218aa7fbd504d6093fa72860cf 21216Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21217Date: Thu May 10 15:33:51 2007 +0200 21218 21219 RADEON: fix VT switch 21220 21221commit 1779a12a947401e5c6bcf784b47e9b3c80d37204 21222Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 21223Date: Thu May 10 06:22:35 2007 -0700 21224 21225 Add cscope files to .gitignore 21226 21227commit 0cb23277666db3b30438c6f88840d861e04df414 21228Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21229Date: Thu May 10 15:20:56 2007 +0200 21230 21231 RADEON: randr driver re-org checkpoint 21232 21233 - split the mode setting per-crtc 21234 - reduce start up flicker 21235 21236commit 33c370b1d8350945f80ac12097d3e91243a400f2 21237Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 21238Date: Wed May 9 16:16:39 2007 -0700 21239 21240 RADEON: 21241 - fix an ugly modesetting bug: if we happened to set the mode on 21242 CRTC1 before CRTC2, CRTC2's RestoreMode function would clobber 21243 CRTC1's CRTC_OFFSET register since we never updated 21244 ModeReg.crtc_offset... so make AdjustFrame use ModeReg and pull 21245 the call to it up before RestoreMode, seems to work ok here. 21246 21247commit 6263248a0044777a352e4ee7380b4b8f9afd091b 21248Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 21249Date: Wed May 9 14:52:00 2007 -0700 21250 21251 RADEON: 21252 - use fixup_mode hook to set RADEON_USE_RMX flag so panel scaling 21253 works 21254 - use valid_mode hook to prune invalid default modes from list 21255 - use adjusted_mode in crtc_mode_set (using adjusted_mode from 21256 fixup hook) 21257 21258commit d2497009e395800fbde5777465f3087a54b94418 21259Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21260Date: Wed May 9 22:23:45 2007 +0200 21261 21262 randr checkpoint 21263 21264 - server still hangs if you start with external monitor connected 21265 - RMX not working 21266 - more than 2 outputs now possible (untested) 21267 21268commit 4488f0737d5268168eab41440b7a3b5732efb15e 21269Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21270Date: Wed May 9 16:41:13 2007 +0200 21271 21272 more randr re-work 21273 21274 - remove RADEONQueryConnectedDisplays(); randr takes care of this now 21275 - print edid after randr detection 21276 21277commit 4d992386e2ab9d8c50f0484445564325dfb42930 21278Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21279Date: Wed May 9 16:12:21 2007 +0200 21280 21281 enable all outputs on each crtc on mode restore 21282 21283commit 2618cf2aa8ed76411b943eb90c95869814c2f151 21284Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21285Date: Wed May 9 15:48:40 2007 +0200 21286 21287 More re-org to allow more than 2 outputs 21288 21289 - Move radeon output crtc map into crtc_set_mode 21290 - in modeinit, set up all outputs attached to crtc 21291 21292commit 61b9e79cbeee6f735a4c82ec8a802aee85d8b890 21293Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21294Date: Wed May 9 15:11:23 2007 +0200 21295 21296 start to clean up MapControllers 21297 21298commit 6a724dd798c1a7b461672993c02be83a7bccded6 21299Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21300Date: Wed May 9 14:48:17 2007 +0200 21301 21302 minor cleanup 21303 21304commit 03860fed24b4f76cc7f1f4210ec8f8040fa04777 21305Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21306Date: Tue May 8 19:27:48 2007 +0200 21307 21308 remove some old mergedfb cruft 21309 21310commit 76670f665ebec7cdf40a04bf9379cb3ad4417507 21311Merge: 83f81ed a3ee422 21312Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21313Date: Tue May 8 18:41:25 2007 +0200 21314 21315 Merge branch 'master' into randr-1.2 and fix conflicts 21316 21317commit 83f81ed5e3c33c94c80500316c37a7cbfc51f41f 21318Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21319Date: Mon May 7 22:54:16 2007 +0200 21320 21321 RADEON: Fix cursor handling for randr 21322 21323commit 5fc21cd34436919300018d8d4850fc67db284eb3 21324Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 21325Date: Mon May 7 13:31:20 2007 -0700 21326 21327 - fix randr 1.2 on pre-RV350 chips by enabling DAC_MACRO_CNTL writes 21328 - set dac_cntl on non-primary crtcs 21329 - set XCRT_CNT_EN in CRTC_EXT_CNTL just because 21330 - fix warnings in calls to xf86PrintModeline (wants screen index not 21331 pScrnInfo) 21332 21333commit 99b3df154317f0209618e532282a3e7ad091c00f 21334Author: Alex Deucher <alex@t41p.hsd1.va.comcast.net> 21335Date: Mon May 7 19:02:35 2007 +0200 21336 21337 RADEON - update randr cursor handling, LVDS setup fix 21338 21339 - quick fix for the cursor handling to update to the latest 21340 server bits 21341 - make sure connector type is CONNECTOR_PROPRIETARY for LVDS. 21342 21343commit a3ee42207aab77d93655a82fdcb32be38268b85f 21344Author: Dave Airlie <airlied@linux.ie> 21345Date: Wed Apr 25 09:52:22 2007 +1000 21346 21347 radeon: another attempt at fixing the mergedfb refresh rate 21348 21349 This attempts to keep it inside the 32-bit limit when multiplying things out 21350 later in the randr tree. 21351 21352 Let me know if I screwed this up.. 21353 21354commit 3828237200fc16d4d32664fb8358950c213d4897 21355Author: Dave Airlie <airlied@nx6125b.(none)> 21356Date: Sun Apr 22 11:36:00 2007 +1000 21357 21358 radeon: add support for DDC on some laptop chipsets 21359 21360 I noticed fglrx has DDC for the panel in the rs480 laptop, however radeon 21361 didn't pick it up, so I valgrinded fglrx and spotted 0x1a0/0x1a4 accesses 21362 I actually noticed this before from the BIOS but never figured it out. 21363 21364 So now I get DDC from the LCD on this laptop. 21365 21366commit dd6a966e862b774a8e8b9e1a085309219673efad 21367Author: Dave Airlie <airlied@nx6125b.(none)> 21368Date: Sun Apr 22 11:36:00 2007 +1000 21369 21370 radeon: add support for DDC on some laptop chipsets 21371 21372 I noticed fglrx has DDC for the panel in the rs480 laptop, however radeon 21373 didn't pick it up, so I valgrinded fglrx and spotted 0x1a0/0x1a4 accesses 21374 I actually noticed this before from the BIOS but never figured it out. 21375 21376 So now I get DDC from the LCD on this laptop. 21377 21378commit c81ed9bd7b37c9d02141d10f6c7bad3d0c57032f 21379Author: Dave Airlie <airlied@linux.ie> 21380Date: Sat Apr 21 18:58:40 2007 +1000 21381 21382 radeon: fix build on older server 21383 21384commit 406eec71116a58d42288a7f1c809a92d5bda7350 21385Author: Dave Airlie <airlied@linux.ie> 21386Date: Sat Apr 21 18:56:28 2007 +1000 21387 21388 radeon: fix build since patches for IBM don't actually build 21389 21390commit ad119960095b4b64f4c6793f65950c9967ce4989 21391Author: Matthias Hopf <mhopf@suse.de> 21392Date: Thu Apr 19 11:54:46 2007 +0200 21393 21394 Disable RMX for IBM Lewis server. 21395 21396 Due to the hardware layout RMX ddc_mode has to be set. 21397 If ddc_mode is set, RADEONValdiateFPModes() shouldn't be called. 21398 Bugzilla #10620 (3). 21399 21400commit 16ef77df4ebaf5ea13baa82972aaf98e71ac32ee 21401Author: Matthias Hopf <mhopf@suse.de> 21402Date: Wed Apr 18 17:36:15 2007 +0200 21403 21404 Set sync polarity restriction flags even for non-"digital separate" monitors. 21405 21406 According to Lisa Wu, this is correct regarding the VESA EEDID standard. 21407 Bugzilla #10620 (2), original patch by Lisa Wu @ATI 21408 21409commit 0abce69f0d826a7ca1a41d963cd4730b6e01c145 21410Author: Matthias Hopf <mhopf@suse.de> 21411Date: Wed Apr 18 17:32:52 2007 +0200 21412 21413 Fix inconsistent use of Mode lists. 21414 21415 Some scans used to only check every second entry, some stopped at the entry 21416 before the last entry. 21417 Bugzilla #10620 (1), original patch by Lisa Wu @ATI 21418 21419commit aea801cf9a5ce519a53d6fffd9a3a2e526ec79ea 21420Author: Matthias Hopf <mhopf@suse.de> 21421Date: Fri Apr 13 16:16:05 2007 +0200 21422 21423 Fix crash if MergedFB and secondary head not found 21424 21425 If the secondary head isn't found (Monitor unplugged etc.) but MergedFB 21426 is configured, the driver segfaults because it tries to access the mode 21427 list private structures, which are not filled in. 21428 21429commit 07ddffb32e6293c77b32c94b87ec468caef3d6f5 21430Author: Matthias Hopf <mhopf@suse.de> 21431Date: Wed Apr 11 14:36:51 2007 +0200 21432 21433 Fixed typo in mode list generation. 21434 21435commit 4effa67ea75736a31b9e78a7b35acf74b067c43e 21436Author: Dave Airlie <airlied@linux.ie> 21437Date: Mon Apr 9 22:08:31 2007 +1000 21438 21439 radeon: add support for enabling direct rendering on RS480 21440 21441 Thanks to Matthew Garrett and Ubuntu for the hw loan to get this working. 21442 21443 Still no 3D driver support but at least you should get CP acceleration for 21444 2D now. 21445 21446commit 6b25a4c48796e022a093f3072574ffe9709ecaf4 21447Author: Michel Dänzer <michel@tungstengraphics.com> 21448Date: Wed Apr 4 10:47:29 2007 +0200 21449 21450 radeon: Link nearest modes by default for clone mode. 21451 21452 This makes sure all modes of both CRTCs will be available by default with 21453 MergedFB. 21454 21455commit 3c892f163ec1fa9be6e733aab091c9b718f41efc 21456Author: Michel Dänzer <michel@tungstengraphics.com> 21457Date: Wed Apr 4 09:52:37 2007 +0200 21458 21459 radeon: Always sort modes when adding to list. 21460 21461 This makes sure mode lists will always be sorted from larger to smaller. 21462 21463commit 3a8190ccc79969925257e7b980b78d79053d208d 21464Author: Michel Dänzer <michel@tungstengraphics.com> 21465Date: Wed Apr 4 09:47:07 2007 +0200 21466 21467 radeon: Don't shrink virtual size based on secondary modes. 21468 21469commit 9b1e97284ce185d358ca756a235d2cee346fa53f 21470Author: Henry Zhao <henryz@localhost.localdomain> 21471Date: Sat Mar 31 23:01:52 2007 -0800 21472 21473 10205: Radeon driver's own mode validation code does not work properly 21474 21475commit 9c2dcd19be8fc2cc29e637d1e9748e66196e3900 21476Author: Henry Zhao <henryz@localhost.localdomain> 21477Date: Sat Mar 31 20:10:03 2007 -0800 21478 21479 9337: EDID modes do not participate in validation for CRT monitor 21480 21481commit 1acd6d6fa42acec07fb11aeb189f492ddb021cb4 21482Author: Michel Dänzer <michel@tungstengraphics.com> 21483Date: Fri Mar 30 11:06:10 2007 +0200 21484 21485 radeon: Guard some MergedFB specific code with info->MergedFB tests. 21486 21487 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=10442 . 21488 21489commit 1a71106c0e4fe5f650239dc694163fdf52d33663 21490Author: Michel Dänzer <michel@tungstengraphics.com> 21491Date: Tue Mar 27 10:13:21 2007 +0200 21492 21493 radeon: Fix typo. 21494 21495commit f87e57d4d773a019d1cc8a10425c57480430f6a4 21496Author: Dave Airlie <airlied@linux.ie> 21497Date: Tue Mar 27 18:08:54 2007 +1000 21498 21499 radeon: fix up crtc debug dereference problem 21500 21501commit 66b4a571a4e7960da6807d3f30955aa08e89ccc6 21502Author: Dave Airlie <airlied@linux.ie> 21503Date: Tue Mar 27 17:00:37 2007 +1000 21504 21505 update number to 6.6.191 for rc release 21506 21507commit fca30a6b581cb6c1466ab1bc316df8fed5d82b60 21508Author: Alex Deucher <alex@botch2.com> 21509Date: Mon Mar 26 23:26:51 2007 -0400 21510 21511 fix NULL dereference 21512 21513 when IsSecondary is true, crtc1 is NULL 21514 Noticed by Sverre Froyen. 21515 21516commit 97d8d1ed10d069343f2b3172ba64ca421821a602 21517Author: George Sapountzis <gsap7@yahoo.gr> 21518Date: Sat Mar 24 20:02:12 2007 +0200 21519 21520 Move atichip.c from ati to atimisc. 21521 21522commit aa8f5b02ebc9be60df48722588261627d6a457e8 21523Author: George Sapountzis <gsap7@yahoo.gr> 21524Date: Sat Mar 24 19:53:02 2007 +0200 21525 21526 [mach64] Use Mach64Chipsets[] instead of ATIChipNames[]. 21527 21528 atimisc is PCI-only now, we can get the chip name with xf86TokenToString(). 21529 21530commit 39e896a1e688ea2d2d21f88c1c5d34c5810aac1c 21531Author: George Sapountzis <gsap7@yahoo.gr> 21532Date: Sat Mar 24 19:47:18 2007 +0200 21533 21534 [mach64] Set pATI->ChipRevision correctly, instead of overriding. 21535 21536 There is no need to override pATI->ChipRevision for GX/CX, as it is only 21537 reported with a printf. 21538 21539commit f046a910ca117279fbabc6281b2e23439ec9ea4e 21540Author: George Sapountzis <gsap7@yahoo.gr> 21541Date: Fri Mar 23 22:19:17 2007 +0200 21542 21543 Drop probing by driver name from "Device" section. 21544 21545 atimisc is PCI-only now, so we only need to scan the PCI space. 21546 21547commit 9cd175d9cd4ed710fccb303664c77519ecaf1e21 21548Author: George Sapountzis <gsap7@yahoo.gr> 21549Date: Fri Mar 23 22:12:48 2007 +0200 21550 21551 Fold FillIn() back to Probe(). 21552 21553commit d7a8cd0e476034796fc38e25a28cd28d05ea4a13 21554Author: George Sapountzis <gsap7@yahoo.gr> 21555Date: Fri Mar 23 22:10:03 2007 +0200 21556 21557 Drop now unneeded _X_EXPORT's. 21558 21559commit 58626d8b78f26f0d9c480033d3c3a12e541342b1 21560Author: George Sapountzis <gsap7@yahoo.gr> 21561Date: Fri Mar 23 21:30:19 2007 +0200 21562 21563 Move {atimach64,r128,radeon}_probe.c from ati to subdrivers. 21564 21565 Subdrivers are now loaded from the wrapper Probe function rather than at screen 21566 creation time. 21567 21568 The wrapper Identify callback only prints chip families now, chip lists are 21569 printed when a subdriver is loaded. This also avoids duplication of subdriver 21570 Identify callbacks. 21571 21572 Unknown radeons should still get a list of known radeon and then probe fails... 21573 21574 Probe for atimisc last to avoid needless loading in most cases (r128, radeon). 21575 21576commit 2e3d43af1e5077cc61dd8668551a6291368d9ed2 21577Author: George Sapountzis <gsap7@yahoo.gr> 21578Date: Mon Mar 19 19:05:04 2007 +0200 21579 21580 [mach64] Compute MMIO address once for each case (PIO, MMIO). 21581 21582 For an MMIO-only driver, the extra MMIO address probing is not needed since we 21583 already compute the MMIO address during probe and do not make any BIOS calls. 21584 21585 For a PIO driver, this extra probing could even result in a wrong MMIO address 21586 since the checking is done by reading through PIO, not MMIO. The MMIO address 21587 is later corrected but having the extra probe there is useless. 21588 21589commit e7defc0e930f1b2d83623a769b2bfdb05c66a5fe 21590Author: George Sapountzis <gsap7@yahoo.gr> 21591Date: Mon Mar 19 19:04:10 2007 +0200 21592 21593 [mach64] Move setting of PIO and MMIO addresses up, after BIOS init. 21594 21595 Partially revert "Minor refactoring of pATI->Block0Base computation." 21596 21597 This parially reverts commit 2c8b33b761e4411451aea1eea3a89d629aeea40d. 21598 21599commit 3dfdba195f1ff794ab00585ab80e7ea6ac92fcf4 21600Author: George Sapountzis <gsap7@yahoo.gr> 21601Date: Sun Mar 18 16:40:07 2007 +0200 21602 21603 [mach64] Make mapping code friendlier to pci-rework source-code compatibility. 21604 21605commit 321db6d87c9110654ab386e5ec270116eca1e04f 21606Author: George Sapountzis <gsap7@yahoo.gr> 21607Date: Sun Mar 18 16:15:20 2007 +0200 21608 21609 [mach64] Add macros for pci-rework source-code compatibilty. 21610 21611commit dfd07b6e99020d1db43d7ce0cae4423d8c6b1f05 21612Author: George Sapountzis <gsap7@yahoo.gr> 21613Date: Sun Mar 18 16:38:26 2007 +0200 21614 21615 [mach64] Drop checking that BAR base + size fits in 32 bit. 21616 21617 I can't see how these tests could fail on 32-bit. OTOH they are potentially 21618 problematic with 64-bit PCI addresses. 21619 21620commit e674338a98c50800637b8ebc01adf3aec2a3eb38 21621Author: George Sapountzis <gsap7@yahoo.gr> 21622Date: Sun Mar 18 15:50:41 2007 +0200 21623 21624 [mach64] Single mapping per BAR. 21625 21626 - We reset write-combining of the FB when MMIO registers fall in the linear 21627 aperture. 21628 21629 - We set MMIOInLinear in all three aperture mappings since the mapping code 21630 depends on it. 21631 21632 - We map the linear aperture in its entirety for MMIOInLinear also. 21633 This simplifies mapping of HW cursor and the LE aperture on BE archs, as 21634 they always fall in the linear aperture. 21635 21636 - I assumed that BAR[2] is page-aligned (we know it's 4K-aligned for sure). 21637 21638commit e7969de8ec572d8801806f9525a9a081a58a03f1 21639Author: George Sapountzis <gsap7@yahoo.gr> 21640Date: Sun Mar 18 15:35:25 2007 +0200 21641 21642 [mach64] Drop poking the VGA MMIO aperture. 21643 21644 We will not be able to handle this when we do mappings for PCI BARs only and 21645 it should be handled with a CPIO driver anyway. 21646 21647commit 166c760a86165330175023e07c4b2bd6891633c5 21648Author: Michel Dänzer <michel@tungstengraphics.com> 21649Date: Tue Mar 20 09:16:02 2007 +0100 21650 21651 Fix advertised minimum minor version of the DRI module. 21652 21653 We don't automatically require bumped minor versions. 21654 21655commit 1bdd376dbd57de8925244f0808f974d6d8cff39d 21656Author: Michel Dänzer <michel@tungstengraphics.com> 21657Date: Tue Mar 20 09:14:39 2007 +0100 21658 21659 radeon: Only sync to hardware when really necessary with EXA. 21660 21661 In particular, don't sync again after accelerated DownloadFromScreen, which 21662 syncs implicitly. This avoids calling into the kernel when it's not necessary, 21663 which can be relevant in some situations. 21664 21665commit 3cfa3a5c8daf03aaad6fc30d275709f6eb717d29 21666Author: Michel Dänzer <michel@tungstengraphics.com> 21667Date: Tue Mar 20 09:13:24 2007 +0100 21668 21669 radeon: Unify code to release the CP. 21670 21671commit e174d8df8c801fad95e5f79cff69187c200bee6e 21672Author: Michel Dänzer <michel@tungstengraphics.com> 21673Date: Tue Mar 20 09:12:51 2007 +0100 21674 21675 radeon: Minor BlockHandler cleanups. 21676 21677 Don't flush indirect buffer in BlockHandler; it's done in LeaveServer. 21678 21679 Also set the EXA engine mode to unknown only at the end. 21680 21681commit 113fb4b61e709a9b54fc2ef73efce06011e771c1 21682Author: Michel Dänzer <michel@tungstengraphics.com> 21683Date: Tue Mar 20 09:12:09 2007 +0100 21684 21685 radeon: Disable CP line acceleration on RV280s. 21686 21687 There have been several reports of stability issues with things like the 21688 xscreensaver hack hypercube. 21689 21690commit 4651d00b183cb498879d605c4b93cd3a0c63cb33 21691Author: Michel Dänzer <michel@tungstengraphics.com> 21692Date: Tue Mar 20 09:10:19 2007 +0100 21693 21694 radeon: Make sure 3D clients will re-upload textures to video RAM after LeaveVT. 21695 21696 Walk the SAREA texList and bump the age of every active object, so their owners 21697 will consider them kicked out when they grab the HW lock next time. 21698 21699commit 2d2fb54ba370c1df9ef5102e83c17a7ff5c55403 21700Author: Michel Dänzer <michel@tungstengraphics.com> 21701Date: Thu Mar 15 17:26:56 2007 +0100 21702 21703 atimisc: Fix strict aliasing violation flagged by -Wall. 21704 21705commit cfb82631fb683a0dc9bbc8f2d214646a8f1bca0b 21706Author: Michel Dänzer <michel@tungstengraphics.com> 21707Date: Thu Mar 15 17:12:43 2007 +0100 21708 21709 radeon: Fix build warnings. 21710 21711commit 50a524445ed1ea3c4fab56de772921ba86a21d35 21712Author: George Sapountzis <gsap7@yahoo.gr> 21713Date: Thu Mar 8 17:43:48 2007 +0200 21714 21715 Move atioption.c from ati to atimisc. 21716 21717commit 52684241ba1ffb393cb655ecc09cbd0cef5b1717 21718Author: George Sapountzis <gsap7@yahoo.gr> 21719Date: Thu Mar 8 07:07:02 2007 +0200 21720 21721 Move atiprobe.c from ati to atimisc. 21722 21723commit 2a1cd107a593630001799d6cd9e72c64222553b2 21724Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 21725Date: Tue Mar 6 14:35:50 2007 -0800 21726 21727 Add prepare/commit hooks to output and crtc func table 21728 21729commit 703c6fc0142ffc600285c13fe6dafecf988c0a1d 21730Author: Jesse Barnes <jbarnes@jbarnes-mobile.amr.corp.intel.com> 21731Date: Tue Mar 6 10:12:42 2007 -0800 21732 21733 Update for new CRTC resize hooks. 21734 21735commit 1e270b411cf691c69194932b864ee2c9db292263 21736Author: George Sapountzis <gsap7@yahoo.gr> 21737Date: Sun Mar 4 18:21:49 2007 +0200 21738 21739 [mach64] Bug 5586: overlay scaler limited to 720 pixels on ATI Rage Pro. 21740 21741commit 54f9b8c1d01ecb97e2d4d247918f0efc86ed6037 21742Author: George Sapountzis <gsap7@yahoo.gr> 21743Date: Sun Mar 4 18:16:40 2007 +0200 21744 21745 [mach64] Xv: use single surface and encoding structs. 21746 21747commit e1e55b533d3ae528c8da37dcb77ed906d05697fb 21748Author: Dave Airlie <airlied@linux.ie> 21749Date: Sun Mar 4 19:11:34 2007 +1100 21750 21751 radeon: add option to set pci aperture size from config file 21752 21753 This requires a drm > 1.26 to work 21754 21755commit 58ee31d015cf8bec0edca62a46faec0b3505be8c 21756Author: Dave Airlie <airlied@linux.ie> 21757Date: Sun Feb 25 23:29:09 2007 +1100 21758 21759 fix typo 21760 21761commit 3cfe94d5438961b869766dacbcd13fde8b770ca2 21762Merge: 31c018c 3b43788 21763Author: Dave Airlie <airlied@linux.ie> 21764Date: Sun Feb 25 23:27:19 2007 +1100 21765 21766 Merge branch 'radeon-randrv12-v4' 21767 21768 Conflicts: 21769 21770 src/radeon_display.c 21771 src/radeon_driver.c 21772 21773commit 3b43788c45f51ad2d3e8e64383c412f4ddd97207 21774Author: Dave Airlie <airlied@linux.ie> 21775Date: Sun Feb 25 23:17:31 2007 +1100 21776 21777 cleanup radeon code against master server 21778 21779commit 24c6fa7cfac5602ba9e6e2f331bcac52fab258e5 21780Author: Dave Airlie <airlied@linux.ie> 21781Date: Sun Feb 25 21:43:54 2007 +1100 21782 21783 make radeon randr build against master server 21784 21785commit 4374895d393bf170134ccffc188521fe515d5a77 21786Author: George Sapountzis <gsap7@yahoo.gr> 21787Date: Sat Feb 24 21:25:13 2007 +0200 21788 21789 [mach64] Consolidate adjustments of mode timings, part 3. 21790 21791 Cosmetic. 21792 21793commit 8b9a4ec8af90b2f9f0d0a3a7f6627061df7e01b3 21794Author: George Sapountzis <gsap7@yahoo.gr> 21795Date: Sat Feb 24 21:22:49 2007 +0200 21796 21797 [mach64] Consolidate adjustments of mode timings, part 2. 21798 21799 Factor out to seperate function. 21800 21801commit bb226c9e7218d2a65c056fe74cb7eece0550670f 21802Author: George Sapountzis <gsap7@yahoo.gr> 21803Date: Sat Feb 24 21:20:53 2007 +0200 21804 21805 [mach64] Consolidate adjustments of mode timings, part 1. 21806 21807 Move LCD block from atimode.c to atimach64.c . 21808 21809commit 31c018ca4a18ce426b29006f103f56eee7f985fa 21810Author: Roland Scheidegger <sroland@tungstengraphics.com> 21811Date: Wed Feb 21 13:47:37 2007 +0100 21812 21813 add missing part of last commit... 21814 21815commit df07fa14da73e92d1a6ee0173468ab5c075d1cbb 21816Author: Roland Scheidegger <sroland@tungstengraphics.com> 21817Date: Wed Feb 21 13:05:42 2007 +0100 21818 21819 fix alignment issues with planar yuv and a bug with packed uyvy 21820 21821 respect that all source planar yuv planes are already dword aligned. 21822 Some attempts to fix up odd widths and odd heights (which are a bit strange 21823 for 4:2:0 formats). They still don't quite work 100% correctly (at the borders) 21824 but neither do they with packed yuv formats. 21825 While here, fix totally broken packed UYVY format by inserting missing break... 21826 Both bugs reported by Felipe Contreras. 21827 21828commit dfcb431adfbbaaee0d262d32735585555a0cbde4 21829Author: Dave Airlie <airlied@linux.ie> 21830Date: Tue Feb 20 21:20:28 2007 +1100 21831 21832 radeon: move some i2c registers out into common register file 21833 21834 These regs are more useful than just multimedia 21835 21836commit 7a25512415bb40b772491bb4a773dfe02ac8b71a 21837Author: Alex Deucher <alex@botch2.com> 21838Date: Fri Feb 16 00:44:11 2007 -0500 21839 21840 Add "DRI" option 21841 21842 This option allows you to disable the DRI per card. It also 21843 removes the "RN50Force3D" option as it is now covered by this 21844 option. RN50 users should set this to TRUE if they want to force 21845 the DRI on. 21846 21847commit fa30ec6d5cd9bf4eb1a960592ca7311175219e4b 21848Author: George Sapountzis <gsap7@yahoo.gr> 21849Date: Tue Feb 13 15:35:32 2007 +0200 21850 21851 [mach64] PreInit: maxPitch, minor cosmetic. 21852 21853commit 1777dcc1956d910073e13322767d9a3de41a949b 21854Author: George Sapountzis <gsap7@yahoo.gr> 21855Date: Tue Feb 13 15:03:40 2007 +0200 21856 21857 [mach64] PreInit: pitchInc is local. 21858 21859commit 9cdf4fcc3a93a4e20244286392dc31a0b8e6a10a 21860Author: George Sapountzis <gsap7@yahoo.gr> 21861Date: Tue Feb 13 11:53:42 2007 +0200 21862 21863 [mach64] PreInit: Use goto bail. 21864 21865 Match ATILock - ATIUnlock. 21866 21867commit 482b85144fee42b6543dcc97a75899e363e05e9e 21868Author: George Sapountzis <gsap7@yahoo.gr> 21869Date: Mon Feb 12 14:27:37 2007 +0200 21870 21871 [mach64] Enable RENDER acceleration on the Pro variants. 21872 21873commit 63248f0b4308a4487cda3aa22daa36e3e0d38d14 21874Author: Dave Airlie <airlied@linux.ie> 21875Date: Mon Feb 12 19:37:36 2007 +1100 21876 21877 fix LVDS by moving bios reading around 21878 21879commit 6a25f620d40cbb063de94aa6b5267296cd5a0670 21880Author: David Airlie <airlied@asimov.stargames.com.au> 21881Date: Mon Feb 12 15:59:43 2007 +1100 21882 21883 remove assignments to pOutput before it is created 21884 21885commit 9234d8045c5fefbd1a781209409e55a13e3e5370 21886Author: David Airlie <airlied@asimov.stargames.com.au> 21887Date: Mon Feb 5 11:53:18 2007 +1100 21888 21889 remove all the locking glxgears runs fine over screen changes without it 21890 21891commit eda9fb15e8bc042661a91a3d8c921006dfb3ddd9 21892Author: George Sapountzis <gsap7@yahoo.gr> 21893Date: Tue Nov 21 04:02:41 2006 +0200 21894 21895 Consolidate atixv.c w/ atimach64xv.c, part 2. 21896 21897 Move {Initialize,Close}XVideo from atixv.c to atimach64xv.c . 21898 21899commit 67b240ddd4a6549a23885afc723b6efa7c52db4c 21900Author: George Sapountzis <gsap7@yahoo.gr> 21901Date: Tue Oct 31 16:44:10 2006 +0200 21902 21903 Consolidate atixv.c w/ atimach64xv.c, part 1. 21904 21905 Merge ATIXVInitializeAdaptor() in ATIMach64XVInitialiseAdaptor(). 21906 21907commit 4dac7896549392f7378cad9620618cec28f41a16 21908Author: George Sapountzis <gsap7@yahoo.gr> 21909Date: Sun Oct 29 03:55:39 2006 +0200 21910 21911 Consolidate aticursor.c w/ atimach64cursor.c 21912 21913commit 23b12b9ff55224e8c3c45eb58ea5fbf2f747d82e 21914Author: George Sapountzis <gsap7@yahoo.gr> 21915Date: Tue Oct 31 16:10:07 2006 +0200 21916 21917 Consolidate atiaccel.c w/ atimach64accel.c, part 2. 21918 21919 Merge ATIInitializeAcceleration() in ATIMach64AccelInit(). 21920 21921commit bc3b222b57d7edf7bdce00a53e8abd9047c314a4 21922Author: George Sapountzis <gsap7@yahoo.gr> 21923Date: Sat Oct 28 19:29:12 2006 +0300 21924 21925 Consolidate atiaccel.c w/ atimach64accel.c, part 1. 21926 21927 Move ATIResizeOffscreenLinear() from atiaccel.c to atimach64xv.c and make it 21928 static. It is only used by XV memory management functions. 21929 21930commit d876889add41c8052d4f3feaa21ddf384f5a7054 21931Author: George Sapountzis <gsap7@yahoo.gr> 21932Date: Tue Nov 21 04:47:06 2006 +0200 21933 21934 Drop non-PCI support from atividmem.c 21935 21936 - we require a PCI device (pVideo != NULL) 21937 - map the VGA framebuffer with xf86MapDomainMemory() 21938 21939commit 2c8b33b761e4411451aea1eea3a89d629aeea40d 21940Author: George Sapountzis <gsap7@yahoo.gr> 21941Date: Tue Nov 21 04:51:02 2006 +0200 21942 21943 Minor refactoring of pATI->Block0Base computation. 21944 21945commit a659b7235332e19659c56cc9d7af362bd8885a29 21946Author: George Sapountzis <gsap7@yahoo.gr> 21947Date: Sun Feb 4 11:28:54 2007 +0200 21948 21949 Cleanup checks for pATI->PCIInfo. 21950 21951 Fixes a bug where shadowfb would not be enabled by default in the non-accel 21952 case because pATI->PCIInfo is now set after ATIProcessOptions() is called. 21953 21954commit 9d77aabdff919360f0c9333105436c31f1f5749a 21955Merge: ff8ea19 57822be 21956Author: George Sapountzis <gsap7@yahoo.gr> 21957Date: Mon Feb 5 19:16:51 2007 +0200 21958 21959 Merge branch 'mach64-pci-1' 21960 21961commit d7317604c843e21c1df048a9253ed55a1957cccd 21962Author: David Airlie <airlied@linux.ie> 21963Date: Fri Feb 2 14:54:00 2007 +1100 21964 21965 get rid of mergetype and displaytype from radeon.h 21966 21967commit 15c68ac19ff9971ae02cf52036ba36ddea1a6759 21968Author: David Airlie <airlied@linux.ie> 21969Date: Fri Feb 2 14:30:16 2007 +1100 21970 21971 rename some variables to output from connector 21972 21973commit 72ef23ef7dd159d827e8e122482a58928d532268 21974Author: David Airlie <airlied@linux.ie> 21975Date: Fri Feb 2 14:26:29 2007 +1100 21976 21977 fixup mode probing a bit neater 21978 21979commit 96acf6b2b242454345cc4b9cfc7ca07e0b597b43 21980Author: David Airlie <airlied@linux.ie> 21981Date: Fri Feb 2 14:26:19 2007 +1100 21982 21983 bring over setproperty from intel code 21984 21985commit 76bc53f9b153880730ab61dcd2b6e4e7717e4058 21986Author: David Airlie <airlied@linux.ie> 21987Date: Fri Feb 2 14:15:03 2007 +1100 21988 21989 move i2c buses into outputs 21990 21991commit 47fb9ce657f018177a35b449a4d716dc03be9327 21992Author: David Airlie <airlied@linux.ie> 21993Date: Fri Feb 2 11:45:40 2007 +1100 21994 21995 fixup callers to crtcsetmode fixes my cursors 21996 21997commit 720730b1b0de632488d3b9818210ec5e9c7f07ae 21998Author: David Airlie <airlied@linux.ie> 21999Date: Thu Feb 1 16:43:38 2007 +1100 22000 22001 add locks for crtcs and some missing lines from intel update 22002 22003commit 6748732658850ea506f623a3622aa7135513ffd0 22004Author: David Airlie <airlied@linux.ie> 22005Date: Thu Feb 1 16:26:31 2007 +1100 22006 22007 update to latest intel codebase modulo using their mode set 22008 22009 and it breaks my cursor 22010 22011commit a77f08298dc7e097025e3f7f92e3665c0ef30095 22012Author: David Airlie <airlied@linux.ie> 22013Date: Mon Jan 29 16:12:24 2007 +1100 22014 22015 radeon: always reset cursor regs 22016 22017commit ff8ea19fcdce099732f9359e53cd62b9a04bfa6d 22018Author: Alex Deucher <alex@botch2.hsd1.va.comcast.net> 22019Date: Tue Jan 30 21:11:09 2007 -0500 22020 22021 add option to force the DRI on for RN50 chips 22022 22023 Allow user to force the DRI on for RN50 chips. 22024 3D is not guaranteed to work on these chips, 22025 however in some cases it does. 22026 fixes bug 9802. 22027 22028commit 57822be75740f339445f2375d44632560f4bbe57 22029Author: George Sapountzis <gsap7@yahoo.gr> 22030Date: Tue Nov 21 03:08:26 2006 +0200 22031 22032 Drop pATI->Chipset, no longer used. 22033 22034commit 8b7c88e9ed549777413136c0f2ccc61747b010d8 22035Author: George Sapountzis <gsap7@yahoo.gr> 22036Date: Thu Dec 28 14:46:10 2006 +0200 22037 22038 Add option "probe_sparse" to force probing sparse I/O bases. 22039 22040commit 3654b786f1ad1f4e6bfb2fff5ff057254958f7c0 22041Author: George Sapountzis <gsap7@yahoo.gr> 22042Date: Wed Dec 27 17:59:14 2006 +0200 22043 22044 Sane probe, based on work by Luc Verhaegen <libv@skynet.be>. 22045 22046commit 0117c9bde0d0bbb487bc259cf55e8066ebd63f6a 22047Author: George Sapountzis <gsap7@yahoo.gr> 22048Date: Wed Dec 27 17:49:45 2006 +0200 22049 22050 Drop checking of sparse I/O bases conflicts 22051 22052 This will be handled with an option somehow. 22053 22054commit 87cf2f1874ca8a7cc492d2e2b6918ba69a8aa713 22055Author: George Sapountzis <gsap7@yahoo.gr> 22056Date: Mon Aug 14 17:55:14 2006 +0300 22057 22058 Minor refactoring of module (un)loading for TV_OUT. 22059 22060commit 0b31333c9147af21c50493872779da12a3668e5d 22061Author: George Sapountzis <gsap7@yahoo.gr> 22062Date: Mon Nov 20 22:08:49 2006 +0200 22063 22064 Drop support for non-programmable clocks. 22065 22066 From README.ati: 22067 22068 Clocks for supported programmable clock generators: 22069 22070 The driver currently supports all programmable clock generators known 22071 to exist on Mach64 adapters. 22072 22073 Clocks for unsupported programmable clock generators: 22074 22075 This case is unlikely to occur, but is documented for the sake of 22076 completeness. 22077 22078 Thus: 22079 22080 - check for (pATI->ProgrammableClock > ATI_CLOCK_FIXED) && 22081 (pATI->ProgrammableClock < ATI_CLOCK_MAX) 22082 22083 - drop "probe_clocks" option 22084 22085 - pATIHW->ClockUnmap is no longer used 22086 22087 - pATIHW->ClockMap is only used with NewHW.crtc which is always ATI_CRTC_MACH64 22088 and has the identity map, so drop it 22089 22090 - (pATI->ProgrammableClock != ATI_CLOCK_INTERNAL) => (pATI->depth <= 8) 22091 22092commit c9a9ea5ebd1c27881c1fac19842cfa3af55c126e 22093Author: George Sapountzis <gsap7@yahoo.gr> 22094Date: Mon Nov 20 15:09:41 2006 +0200 22095 22096 Banked memory is no longer needed. 22097 22098 - (pATI->BankInfo.BankSize = 0) in all cases, cull pATI->BankInfo 22099 22100 - only keep the minimal pATIHW.SetBank interface for save/restore 22101 22102 - clean ATISwap() a little, (NewHW.crtc != ATI_CRTC_VGA) 22103 22104 - (UseSmallApertures == TRUE) <=> pATI->VGAAdapter 22105 22106commit 383e963e275d351ea3631c352f5795340162d69f 22107Author: George Sapountzis <gsap7@yahoo.gr> 22108Date: Sun Jan 28 00:44:31 2007 +0200 22109 22110 Always use the linear aperture. 22111 22112 - drop (pATI->OptionLinear == FALSE) 22113 22114 - AcceleratorVideoRAM is always set, i.e. VGAVideoRAM is not used 22115 22116 - pATI->LinearBase is always set 22117 22118 - xf86LinearVidMem() is now checked in atipreinit() for both CPIO and MMIO 22119 22120commit 9b126f45db27496c1ad16db65a61fe641a018983 22121Author: George Sapountzis <gsap7@yahoo.gr> 22122Date: Mon Nov 20 20:38:59 2006 +0200 22123 22124 We always use the accelerator CRTC within the xserver. 22125 22126 - cull (pATI->NewHW.crtc != ATI_CRTC_MACH64). 22127 22128commit 0abcb2b2c84a50e64bb6aa9e94760659ecf33add 22129Author: George Sapountzis <gsap7@yahoo.gr> 22130Date: Sun Jan 28 00:19:15 2007 +0200 22131 22132 Clean pATI->Adapter and pATI->VGAAdapter. 22133 22134 - cull (pATI->Adapter != ATI_ADAPTER_MACH64) 22135 22136 - treat pATI->VGAAdapter as Bool 22137 22138commit 817b4b5338dba4d8eb862f1f79ef0edb75aad7f4 22139Author: George Sapountzis <gsap7@yahoo.gr> 22140Date: Wed Dec 27 21:16:18 2006 +0200 22141 22142 Keep PCI mach64 only, drop: 22143 22144 - Chip < ATI_CHIP_88800GXC 22145 - Chipset != ATI_CHIPSET_ATI 22146 - Adapter != ATI_ADAPTER_MACH64 22147 22148 - depth < 8 22149 22150commit d350860e29f043e98bfb1da74b26280f1755ab6f 22151Author: George Sapountzis <gsap7@yahoo.gr> 22152Date: Wed Dec 27 20:56:45 2006 +0200 22153 22154 Keep PCI mach64 only (atipreinit.c, atimode.c), drop: 22155 22156 - Chip < ATI_CHIP_88800GXC 22157 - Chipset != ATI_CHIPSET_ATI 22158 - Adapter != ATI_ADAPTER_MACH64 22159 22160 - depth < 8 22161 22162 atimode.c only: 22163 22164 - NewHW.crtc != ATI_CRTC_MACH64 22165 22166 This allows to drop VGACalculate(), VGAWonderCalculate() cruft early. 22167 22168commit 5fa77f2d122e7267911a15235338d8d3f9eece2e 22169Author: George Sapountzis <gsap7@yahoo.gr> 22170Date: Wed Dec 27 14:50:48 2006 +0200 22171 22172 Drop probing for VGA, VGAWonder, 8514, Mach32. 22173 22174 - ChipHasSUBSYS_CNTL 22175 - Coprocessor 22176 - SharedAccelerator 22177 - SharedVGA <=> (VGAAdapter != ATI_ADAPTER_NONE) 22178 22179commit 1b2a9e1735da63bbb8cb41939668f789cee22579 22180Author: George Sapountzis <gsap7@yahoo.gr> 22181Date: Mon Nov 20 02:35:56 2006 +0200 22182 22183 Drop AVOID_NON_PCI. 22184 22185 This was not set anyway, because configure.ac would compute ATIMISC_NON_PCI and 22186 then test ATI_AVOID_NON_PCI to set AVOID_NON_PCI... 22187 22188commit 2fb84dd2fc1a663c7b9d77e706189439ffe08381 22189Author: Alex Deucher <alex@botch2.hsd1.va.comcast.net> 22190Date: Thu Jan 25 02:01:42 2007 -0500 22191 22192 I think we can say r300 support is not experimental 22193 22194commit e155659f4b006edff31f1fda88b32bf608dbeece 22195Author: Alex Deucher <alex@botch2.hsd1.va.comcast.net> 22196Date: Thu Jan 25 01:59:27 2007 -0500 22197 22198 update man page. fixes bug 6457 22199 22200 Most cards support 2D and 3D now 22201 22202commit fa12fe1cc90dd745f3eea35a07d4f3efd652e7b4 22203Author: Dave Airlie <airlied@linux.ie> 22204Date: Tue Jan 23 19:37:51 2007 +1100 22205 22206 remove unused restore palette function 22207 22208commit 4d3649b84a3325043c38cece4499de0095cebd71 22209Author: Dave Airlie <airlied@linux.ie> 22210Date: Tue Jan 23 19:37:21 2007 +1100 22211 22212 remove isused and isactive 22213 22214commit 9149e763865598c307cbefc753ff6ebdeeaf32ae 22215Author: David Airlie <airlied@asimov.stargames.com.au> 22216Date: Tue Jan 23 15:20:11 2007 +1100 22217 22218 radeon: fixup problem with cursor not going off 22219 22220 Also fixup secondary dac detect to return unknown for now 22221 22222commit 2a13a3b641d9acf4f50472e1c8ba07633c3b78d6 22223Author: David Airlie <airlied@asimov.stargames.com.au> 22224Date: Tue Jan 23 10:36:06 2007 +1100 22225 22226 fixup init disp bandwidth 22227 22228 This probably break old-style dualhead badly 22229 22230commit cbd84bed13582e82f8b2e84aa152602474c09cd4 22231Author: Dave Airlie <airlied@linux.ie> 22232Date: Mon Jan 22 22:19:46 2007 +1100 22233 22234 removed even more mergedfb remains 22235 22236commit 6296882135b183425a219efb75374d3b172de370 22237Author: Dave Airlie <airlied@linux.ie> 22238Date: Mon Jan 22 22:14:48 2007 +1100 22239 22240 remove more mergedfb relics 22241 22242commit 26e1bab2c7a4ad52710746265b3da495b54dab36 22243Author: Dave Airlie <airlied@linux.ie> 22244Date: Mon Jan 22 22:11:55 2007 +1100 22245 22246 radeon: fixup some of the video code after randr 22247 22248 This may not be complete, I need to test on dual-head system later 22249 22250commit 9008e1caa45e4a18e6f3289c7b17036730fe578a 22251Author: Dave Airlie <airlied@linux.ie> 22252Date: Mon Jan 22 21:16:47 2007 +1100 22253 22254 remove unused merged code 22255 22256commit e29a32d711553fcb4ca9928122ac285fe0b0c1a7 22257Author: Dave Airlie <airlied@linux.ie> 22258Date: Mon Jan 22 21:09:53 2007 +1100 22259 22260 add back overlay CRTC chooser this may not work yet 22261 22262commit aa9d04ba94a3cd60b248231da517e2817591fc69 22263Author: Dave Airlie <airlied@linux.ie> 22264Date: Sun Jan 21 20:29:56 2007 +1100 22265 22266 remove alot more mergedfb support 22267 22268commit e8be0056e8ff666f63a294691661f5dab289203e 22269Author: Dave Airlie <airlied@linux.ie> 22270Date: Thu Jan 18 16:41:25 2007 +1100 22271 22272 remove mergedfb 22273 22274commit a648050a3cc60f92b1ca0b3d707aadf93d076d91 22275Author: Dave Airlie <airlied@linux.ie> 22276Date: Wed Jan 17 17:28:07 2007 +1100 22277 22278 move some cursor code around for show/hide 22279 22280commit d2ecfb507282726122bb8b0d17fd3637d0ae7d46 22281Author: Dave Airlie <airlied@linux.ie> 22282Date: Wed Jan 17 17:16:27 2007 +1100 22283 22284 make dri work again 22285 22286commit 369f7c85ceff983defb7657b80ec9cd3e5440b07 22287Author: Dave Airlie <airlied@linux.ie> 22288Date: Wed Jan 17 11:14:09 2007 +1100 22289 22290 comment out unblanking on VT switch 22291 22292commit bdb66a2042f02c4b57bd3c0181a00b39fcbdb232 22293Author: Dave Airlie <airlied@linux.ie> 22294Date: Sun Jan 14 21:07:05 2007 +1100 22295 22296 radeon: get randr-1.2 mode switching mostly working on my laptop 22297 22298 The main change is to fix the dac macro + cntl register writes for rv350 and 22299 above, this still has a problem with resetting the same mode after connect 22300 disconnect cycle, need to talk to keithp 22301 22302commit aa0e7337815d4daca4df0671768621330b759011 22303Author: Dave Airlie <airlied@linux.ie> 22304Date: Sun Jan 14 19:17:32 2007 +1100 22305 22306 use more randr-ish names for variables 22307 22308commit 9d37f23aea43b74a7ec640e6b03617c8392e2572 22309Author: Dave Airlie <airlied@linux.ie> 22310Date: Sun Jan 14 19:10:11 2007 +1100 22311 22312 radeon: move blank/unblank to use randr 22313 22314commit 6bd4fe42789c38e7e804826715214ce6badcca6a 22315Author: Dave Airlie <airlied@linux.ie> 22316Date: Sun Jan 14 18:38:59 2007 +1100 22317 22318 disable dpms on/off as it was writing regs before we wanted to 22319 22320commit 3949288ed26a91e180b178f4796f2f7e1bdc8ed6 22321Author: Dave Airlie <airlied@linux.ie> 22322Date: Sun Jan 14 16:26:39 2007 +1100 22323 22324 radeon: destroy and fixup LVDS crtc 22325 22326commit f5ffb9fb3bdd7cec020240c5831ae92ee744972c 22327Author: Timo Jyrinki <timo.jyrinki@hut.fi> 22328Date: Fri Jan 12 23:08:50 2007 +1100 22329 22330 ati: fix detection of some newer radeons using ati wrapper 22331 22332 This is mainly due to the cards having a different resource 1. 22333 Fixes 6796 22334 22335commit 1e690c57ad1dfaeb760ef6e799faa6398e9a0551 22336Author: Dave Airlie <airlied@linux.ie> 22337Date: Fri Jan 12 23:03:10 2007 +1100 22338 22339 always create a second CRTC I think this should solve any problems 22340 22341commit 25b36a4c56422ad8e25a4a1c55055c2f062213ca 22342Author: Dave Airlie <airlied@linux.ie> 22343Date: Fri Jan 5 16:13:16 2007 +1100 22344 22345 minor randr1.2 fixups 22346 22347commit abe8791e4f9fa3e88273897b351387cd33822734 22348Author: Dave Airlie <airlied@linux.ie> 22349Date: Fri Jan 5 15:09:43 2007 +1100 22350 22351 hook up DPMS through xf86 22352 22353commit 8a9c68c2234b91ed38555f892afdad30b9e4b455 22354Author: Dave Airlie <airlied@linux.ie> 22355Date: Fri Jan 5 15:09:15 2007 +1100 22356 22357 set binding for new heads 22358 22359commit d6a2b8aeb5a97ee907fd45b574bc6e4ab4b3aede 22360Author: Dave Airlie <airlied@linux.ie> 22361Date: Fri Jan 5 09:33:54 2007 +1100 22362 22363 fix LUT 22364 22365commit ef1d36e56dec1fec37cee2dfd9cb5bf8ce2c485c 22366Author: Roland Scheidegger <sroland@tungstengraphics.com> 22367Date: Wed Jan 3 15:56:23 2007 +0100 22368 22369 fix uninitialized mode regs (bug 9495) 22370 22371 Since the reorganization of the mode setting code, the mode registers relying 22372 on state already set (by bios) were not read, thus clearing out all bits the 22373 driver does not touch. At the very least, this could lead to completely 22374 nonfunctional to misbehaving dvi output (see bug 9495). Fix this by using the 22375 SavedReg values, which also makes it more obvious that those are bits which 22376 were not set by the driver previously, but come from register readback. 22377 (cherry picked from 9506f7015a1e442f4ca0bd3bfae555ec7e8a5f37 commit) 22378 22379commit 55aa832157bdebcba2d58896777942d108c352b0 22380Author: Dave Airlie <airlied@linux.ie> 22381Date: Thu Jan 4 17:57:31 2007 +1100 22382 22383 fix cursor handling 22384 22385commit 4f8a7cafdc77e98dc44f9eced876560b1ee01117 22386Author: Dave Airlie <airlied@linux.ie> 22387Date: Thu Jan 4 16:23:39 2007 +1100 22388 22389 hook up crtc modesetting 22390 22391commit 52f749c8a613ee316044abe82156ee270412ced8 22392Author: Dave Airlie <airlied@linux.ie> 22393Date: Thu Jan 4 16:23:26 2007 +1100 22394 22395 fix typo 22396 22397commit d7ff61c6822cbede7f5b59b411048d33dbae9ee4 22398Author: Dave Airlie <airlied@linux.ie> 22399Date: Thu Jan 4 13:38:26 2007 +1100 22400 22401 hook up detection of output 22402 22403commit 50439d7e9f4f6f9933bacd59f8bb3e655a35dbc2 22404Author: Dave Airlie <airlied@linux.ie> 22405Date: Thu Jan 4 13:21:57 2007 +1100 22406 22407 fixup some function calls 22408 22409commit e067bfee92f2b0877108355619b2fb9188a9d15a 22410Author: Dave Airlie <airlied@linux.ie> 22411Date: Thu Jan 4 13:21:45 2007 +1100 22412 22413 add more files for back compat 22414 22415commit f1f34627ffbe2136ac3e023c01c0430412919ded 22416Author: Dave Airlie <airlied@linux.ie> 22417Date: Thu Jan 4 12:50:03 2007 +1100 22418 22419 make restore mode non-static 22420 22421commit 249ec67c296b34d0c7c1d78602628d2f7ce6a045 22422Author: Dave Airlie <airlied@linux.ie> 22423Date: Thu Jan 4 12:43:03 2007 +1100 22424 22425 more minor fixes to get radeon up to speed 22426 22427commit a43c1d55f5f855d9e6ae939dd4eec1c607b6d514 22428Author: Dave Airlie <airlied@linux.ie> 22429Date: Thu Jan 4 12:29:51 2007 +1100 22430 22431 bring radeon randr code inline with intel randr code 22432 22433commit c54a31565e3ea50f1af7b81b1bf0c0e437099f47 22434Author: Dave Airlie <airlied@linux.ie> 22435Date: Thu Jan 4 10:39:01 2007 +1100 22436 22437 radeon: fixup fastwrites to be always off 22438 22439 My 8500 in i845 doesn't work with fastwrites even setup by the firmware. 22440 22441commit dda2bbfdc2e0da6e8b9cb8d1bf78b91f1bb8131f 22442Author: Dave Airlie <airlied@linux.ie> 22443Date: Thu Jan 4 10:29:54 2007 +1100 22444 22445 radeon: don't touch Controller[1] on single CRTC cards. 22446 22447 This might fix bug 9371 22448 22449commit 9506f7015a1e442f4ca0bd3bfae555ec7e8a5f37 22450Author: Roland Scheidegger <sroland@tungstengraphics.com> 22451Date: Wed Jan 3 15:56:23 2007 +0100 22452 22453 fix uninitialized mode regs (bug 9495) 22454 22455 Since the reorganization of the mode setting code, the mode registers relying 22456 on state already set (by bios) were not read, thus clearing out all bits the 22457 driver does not touch. At the very least, this could lead to completely 22458 nonfunctional to misbehaving dvi output (see bug 9495). Fix this by using the 22459 SavedReg values, which also makes it more obvious that those are bits which 22460 were not set by the driver previously, but come from register readback. 22461 22462commit d5c6306385cbc7b13a132ae72ed04e04e24d9886 22463Author: George Sapountzis <gsap7@yahoo.gr> 22464Date: Tue Dec 19 09:45:24 2006 +0200 22465 22466 Use portable XORG_RELEASE_VERSION macro. 22467 22468commit 97df0a32b14e34f22dce9cd3b375c1ac5fb1c5eb 22469Author: Michel Dänzer <michel@tungstengraphics.com> 22470Date: Tue Dec 19 11:28:21 2006 +0100 22471 22472 radeon: Advertise 2D engine limits to EXA when we're not using the 3D engine. 22473 22474 Also round up to the maximum width and height, as that's what EXA compares. 22475 22476commit 295823d0879a5b574bb79843a6acd43adb9259e5 22477Author: Michel Dänzer <michel@tungstengraphics.com> 22478Date: Tue Dec 19 11:19:20 2006 +0100 22479 22480 radeon: Default to leaving AGP transfer mode and fast writes unchanged. 22481 22482 Based on the assumption that firmware should have set up the card and host 22483 bridge appropriately for these settings, this may actually be safer, at least 22484 for the transfer rate; leaving fast writes enabled is hopefully safe as well, 22485 it certainly is on my sytem. 22486 22487 See https://bugs.freedesktop.org/show_bug.cgi?id=9284 . 22488 22489commit 81f3b4070b70483d6da4f7587e52a64ac69c8ca0 22490Author: Eamon Walsh <ewalsh@tycho.nsa.gov> 22491Date: Mon Dec 18 14:06:15 2006 -0500 22492 22493 Revert "Convert callers of LookupWindow() to dixLookupWindow()." 22494 22495 This reverts commit 48ff33a1770f3684cd50184db8f1944a456d9974. 22496 22497commit 87592ffb717da1f0a1767a38918d16d60953599c 22498Author: Dave Airlie <airlied@linux.ie> 22499Date: Sun Dec 17 20:10:58 2006 +1100 22500 22501 add setdpi call 22502 22503commit 41cd50487bc85708a3f791dfa70bf85223d91a65 22504Author: Dave Airlie <airlied@linux.ie> 22505Date: Sun Dec 17 20:08:07 2006 +1100 22506 22507 add displayWidth set 22508 22509commit 555e0f988c571aa47bc62ec6d9612a71f1a1f59b 22510Author: Dave Airlie <airlied@linux.ie> 22511Date: Sun Dec 17 19:59:19 2006 +1100 22512 22513 add randr pre init 22514 22515commit 3e2c72d7ac80ed874bcf2887a5253dac1c9bb02c 22516Author: Dave Airlie <airlied@linux.ie> 22517Date: Sun Dec 17 19:49:08 2006 +1100 22518 22519 fixup even more remnants of old code 22520 22521commit 4356b031886e00b3ed5ac1dbceeadebd7fc29fc2 22522Author: Dave Airlie <airlied@linux.ie> 22523Date: Sun Dec 17 19:47:45 2006 +1100 22524 22525 fix remnant of old code 22526 22527commit 17d39502a8dbea395051007aebffdf23644e9ee1 22528Author: Dave Airlie <airlied@linux.ie> 22529Date: Sun Dec 17 19:44:39 2006 +1100 22530 22531 use getddcmodes properly 22532 22533commit bf0b364cbd8682f297e2d110ece1f72e7151340d 22534Author: Dave Airlie <airlied@linux.ie> 22535Date: Sun Dec 17 17:31:08 2006 +1100 22536 22537 fix up output namings 22538 22539commit 4962a5430844114de864418d803f3182b90792ff 22540Author: Dave Airlie <airlied@linux.ie> 22541Date: Sun Dec 17 16:58:07 2006 +1100 22542 22543 remove some code that moved into server 22544 22545commit b47013f41249516ae82e22bd75caf573da2cc13c 22546Author: Dave Airlie <airlied@linux.ie> 22547Date: Sat Dec 16 12:02:37 2006 +1100 22548 22549 hook up mode detect to new randr code 22550 22551commit 48ff33a1770f3684cd50184db8f1944a456d9974 22552Author: Eamon Walsh <ewalsh@tycho.nsa.gov> 22553Date: Fri Dec 15 17:43:39 2006 -0500 22554 22555 Convert callers of LookupWindow() to dixLookupWindow(). 22556 22557commit 62a5e7565b66d7b7d65609c034f34b55bd266617 22558Author: Dave Airlie <airlied@linux.ie> 22559Date: Fri Dec 15 19:48:42 2006 +1100 22560 22561 bring over latest code from Intel git repo. 22562 22563 this code doesn't work yet just realigns with intel work 22564 22565commit 29124400c7f193317d41d8cfd748371a239cfea1 22566Author: Dave Airlie <airlied@linux.ie> 22567Date: Sun Dec 10 18:00:17 2006 +1100 22568 22569 hook up randr crtc setting 22570 22571commit 120c854f185c1e7711cf0dee19303fdb8894d49d 22572Author: Roland Scheidegger <sroland@tungstengraphics.com> 22573Date: Fri Dec 8 02:51:52 2006 +0100 22574 22575 radeon: avoid unnecessary OUTPLL/INPLL calls when displaying video 22576 22577 It is not necessary to always emit a OUTPLL/INPLL pair when we display 22578 a video frame. On some chips there are erratas for which the workarounds 22579 cause a 10ms delay by those calls. This is related to #5876 though those 22580 affected may suffer from other slowness issues too. 22581 22582commit ccd37b3e88cdcfe03b7c707b6082339eb8f11e6b 22583Author: Michel Dänzer <michel@tungstengraphics.com> 22584Date: Thu Dec 7 18:41:48 2006 +0100 22585 22586 radeon: Move Xv option handling code into RADEONPreInitXv(). 22587 22588 This unclutters RADEONPreInit() somewhat, but more importantly moves comparison 22589 against info->ChipFamily after that's initialized. 22590 22591commit 02be8c04a268eaa71bdfe891331758715fef4e71 22592Author: Michel Dänzer <michel@tungstengraphics.com> 22593Date: Thu Dec 7 18:31:58 2006 +0100 22594 22595 radeon: Make cache flushing for R300 series less invasive. 22596 22597 Instead of calling the DRM CP idle ioctl, just emit the cache flush commands 22598 into the CP stream. 22599 22600commit be46f7b6f7a729695a0affe2328077666283af12 22601Author: Michel Dänzer <michel@tungstengraphics.com> 22602Date: Thu Dec 7 16:27:10 2006 +0100 22603 22604 Fix radeon compile warnings. 22605 22606commit 59c5b4f69c97fb233144677ab18095f88aed006b 22607Author: Daniel Stone <daniel@fooishbar.org> 22608Date: Wed Dec 6 18:48:55 2006 +0200 22609 22610 Makefile.am: re-add ChangeLog to CLEANFILES 22611 22612commit 7611f14dc697b75cd7bb6493bb65b3a7ead379b8 22613Author: Daniel Stone <daniel@fooishbar.org> 22614Date: Wed Dec 6 18:47:33 2006 +0200 22615 22616 Makefile.am: fix changelog hook 22617 Make changelog hook safe for all situations. 22618 22619commit f2a9d2d61addc4e3d5e128d8078a04c9a136be6a 22620Author: Roland Scheidegger <sroland@tungstengraphics.com> 22621Date: Wed Dec 6 00:16:41 2006 +0100 22622 22623 radeon: fix up packed yuv broken by planar yuv patch 22624 22625 fix the forgotten leftuv value for packed yuv which is 22626 needed to get correct uv starting pixel (fixes broken clipping / 22627 non-null src start pixel as tvtime uses) 22628 22629commit d2aa57a4bdd4a0deb2c659bb07f3f57d1116f0d1 22630Author: Roland Scheidegger <sroland@tungstengraphics.com> 22631Date: Tue Dec 5 01:30:13 2006 +0100 22632 22633 radeon: use overlay scaler native planar yuv capability. 22634 22635 Radeons can do planar yuv just fine, there is no need to convert all data 22636 to packed yuv manually. This saves some cpu cycles as well as some 22637 (graphic card) ram. 22638 22639commit 5d5fa1b86e5179b061f0db47fe0227d1b84c37f8 22640Author: Dave Airlie <airlied@linux.ie> 22641Date: Mon Dec 4 18:53:33 2006 +1100 22642 22643 update radeon driver to Intel driver xf86Crtc interfaces 22644 22645commit 51d1cf19e71dd5de47f2c6467f4a1685eefd9e1e 22646Author: Dave Airlie <airlied@linux.ie> 22647Date: Sun Dec 3 16:30:01 2006 +1100 22648 22649 Add radeon randr 1.2 initial attempt at support 22650 22651 This doesn't do a huge amount yet 22652 22653commit eb17c9aed2144701ad7bd1042b2905446e4d708a 22654Author: airlied <airlied@optimus.localdomain> 22655Date: Sat Nov 18 15:48:02 2006 +1100 22656 22657 move radeon over to not using X's mode management like the intel driver. 22658 22659 this builds and at least seems to pick a mode.. might be near to hooking 22660 up randr1.2 soon 22661 22662commit 941b5120916ebff69ec0ee4ca7a46105d5306e5d 22663Author: airlied <airlied@optimus.localdomain> 22664Date: Sat Nov 18 10:41:12 2006 +1100 22665 22666 make radeon modes use generic code 22667 22668commit 608427283ea59cdfe3d85e1cf32719bc3809dc16 22669Author: airlied <airlied@optimus.localdomain> 22670Date: Sat Nov 18 10:35:31 2006 +1100 22671 22672 add more functions from i830 driver 22673 22674 This adds more generic functions from the i830 driver to the radeon 22675 22676commit d5b8cafc185b7d9fb909cb18a08615f81d89eaf8 22677Author: airlied <airlied@optimus.localdomain> 22678Date: Sat Nov 18 10:22:38 2006 +1100 22679 22680 add generic mode handling code from Intel driver 22681 22682commit 9cfa82e1670ad85746926995972a535ddf03ee07 22683Author: Dave Airlie <airlied@linux.ie> 22684Date: Sun Dec 3 15:56:38 2006 +1100 22685 22686 move monitor detection into single function 22687 22688commit 032a00d155ecdcdcab3ca0d047f552fe6b338a97 22689Author: airlied <airlied@optimus.localdomain> 22690Date: Sat Nov 18 10:23:12 2006 +1100 22691 22692 make DDC function more generic 22693 22694commit 2632ae24b3b50c69567e92611ce7c2cd2ee98baa 22695Author: Dave Airlie <airlied@linux.ie> 22696Date: Sun Dec 3 15:22:38 2006 +1100 22697 22698 re-program bios 4 scratch so we know if the LCD panel is there post startup 22699 22700commit ddec94e8f2f5746bad0e998d15303e771803d22b 22701Author: Dave Airlie <airlied@linux.ie> 22702Date: Sun Dec 3 11:07:23 2006 +1100 22703 22704 move crtc to connector mapping to a connector to crtc mapping. 22705 22706 Still not allowing any clone modes, but heading in the correct direction 22707 I hope... there is a chance this will regress something from superpatch.. 22708 22709commit 218cb7f8db35d0ce8f9fc4ece72bd5ec3f52106f 22710Author: Dave Airlie <airlied@linux.ie> 22711Date: Sun Dec 3 09:53:35 2006 +1100 22712 22713 radeon: dynamically allocate connectors 22714 22715commit f22fd02bc7f7ef67e33c9bd93eae2d7488669b5e 22716Author: Dave Airlie <airlied@linux.ie> 22717Date: Sat Dec 2 19:48:00 2006 +1100 22718 22719 make CRTCs malloc'ed rather than part of Ent 22720 22721 This is leading towards randr-1.2 believe me :-) 22722 22723commit 13efd1f75fbd29a4d1f69d7d5a5bb2b8d90ad820 22724Author: Dave Airlie <airlied@linux.ie> 22725Date: Sat Dec 2 19:16:41 2006 +1100 22726 22727 move HasCRTC2 to pRADEONEnt from info 22728 22729commit 3933b86390eb2f4198d490acd4914fb470c40f47 22730Author: Michel Dänzer <michel@tungstengraphics.com> 22731Date: Fri Dec 1 18:00:17 2006 +0100 22732 22733 Make log file output clear about origin of used overlay scaler buffer width. 22734 22735commit 563dfb52f5b24d2f765fae0394e599c11dcf0b5b 22736Author: Michel Dänzer <michel@tungstengraphics.com> 22737Date: Fri Dec 1 17:56:19 2006 +0100 22738 22739 Increase default overlay scaler buffer width for RV350. 22740 22741 Works with 1920x1080 video on my M10. 22742 22743commit 984ea3d9b7c27d8183ca94ad9d7905f03613fe10 22744Author: Michel Dänzer <michel@tungstengraphics.com> 22745Date: Fri Dec 1 17:42:23 2006 +0100 22746 22747 Fix build warning. 22748 22749commit 9f5ea3981449f29ff204eb154166e8fc813205fa 22750Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 22751Date: Wed Nov 29 16:46:57 2006 +0100 22752 22753 radeon: bug #1462, predownscale to make HD video work. 22754 22755 Use the overlay scaler's predownscale capability to make videos with large 22756 horizontal resolution work if it exceeds the scaler buffer width. Make the 22757 scaler buffer width user-configurable since we don't know it for all chips, 22758 and using predownscaling may otherwise reduce quality even if it wouldn't 22759 be needed. This should fix bug #1462. 22760 22761commit 35a5eaf5adf103ef57f8d4590dae25b0c50bd780 22762Author: airlied <airlied@optimus.localdomain> 22763Date: Tue Nov 28 17:31:05 2006 +1100 22764 22765 fixup EXA after previous commit 22766 22767commit ba047ae863aac869595a31ea94f102084636fd29 22768Author: airlied <airlied@optimus.localdomain> 22769Date: Tue Nov 28 17:26:10 2006 +1100 22770 22771 radeon: initial move framebuffer away from zero prep. 22772 22773 In theory the driver should be able to handle the front buffer not at VRAM 0 22774 In practice it didn't.. this is cleanup for at least XAA parts of the driver 22775 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...) 22776 22777commit 0d42cab2d889070ab6de05990e2b19724595f7af 22778Author: Alex Deucher <alex@t41p.(none)> 22779Date: Wed Nov 22 13:44:49 2006 -0500 22780 22781 clean up previous VIP commit. 22782 22783commit 8e0f57571228729cc40ed59266484ecc43819f44 22784Author: Alex Deucher <alex@t41p.(none)> 22785Date: Wed Nov 22 13:38:01 2006 -0500 22786 22787 Don't mess with the VIP bus on mobility chips. Seems to cause fan 22788 problems on asus laptops (see bug 7463). We don't support video-in 22789 on any laptops at the moment anyway. 22790 22791commit e4850610596a08082a528c38186268b72012e78c 22792Author: Alain Péteut <alain.peteut@gmx.ch> 22793Date: Wed Nov 22 12:54:55 2006 -0500 22794 22795 enable i2c for PCI_CHIP_RV250_If. allows AIW fuctionality to work. 22796 patch and testing from Alain Péteut 22797 22798commit ff38c2c9cf0c6acbdaf53b681ec721487cf411f8 22799Author: Michel Dänzer <michel@tungstengraphics.com> 22800Date: Sun Nov 19 17:11:46 2006 +0100 22801 22802 radeon: Bug #8989: Fix page flipping artifacts with XAA. 22803 22804 We still have to force-sync the pages on enabling page flipping with XAA as the 22805 second page may have been clobbered by the offscreen pixmap cache since they 22806 were last synced. 22807 22808commit 2dd0892817090709b2fb7ffba920595ada00fcad 22809Author: Michel Dänzer <michel@tungstengraphics.com> 22810Date: Sun Nov 19 17:09:11 2006 +0100 22811 22812 Fix compile warnings. 22813 22814commit 8c1b9269aaac746989bdb9be02915722bcf6fec0 22815Author: airlied <airlied@optimus.localdomain> 22816Date: Fri Nov 17 18:47:50 2006 +1100 22817 22818 forgot SetPitch function is called 22819 22820commit 3a0338bb487739a1740ddfb4d36d53cfa7f798c6 22821Author: airlied <airlied@optimus.localdomain> 22822Date: Fri Nov 17 18:37:00 2006 +1100 22823 22824 split mode handling code into separate modes source file 22825 22826commit 13e0a95564220d42ad8060b43b0aaaba3002157a 22827Author: David Airlie <airlied@asimov.stargames.com.au> 22828Date: Mon Nov 13 09:35:18 2006 +1100 22829 22830 fix up issue with bindings and mergedfb 22831 22832commit 63444770ff30afeb134befed5bb3a5c28b7bc4e9 22833Author: George Sapountzis <gsap7@yahoo.gr> 22834Date: Thu Nov 9 22:01:28 2006 +0200 22835 22836 [mach64] Prefer double-buffered visuals for GLX. 22837 22838 So that 3D apps apropriately fail in DRI... 22839 22840commit 0411f68d71c75cda38ad46ad5c0f5cc3cda64464 22841Author: George Sapountzis <gsap7@yahoo.gr> 22842Date: Thu Nov 9 22:00:56 2006 +0200 22843 22844 [mach64] Use asynchronous DMA mode by default. 22845 22846commit 8dd668db05e8f1282fa0e29a91e38c7cbdc97aa1 22847Author: Dave Airlie <airlied@linux.ie> 22848Date: Sun Nov 12 16:33:21 2006 +1100 22849 22850 radeon: abstract init display bandwidth from mergedfb 22851 22852commit 45a9f434b77ad892efcdba267a935135b972dd42 22853Author: Dave Airlie <airlied@linux.ie> 22854Date: Sun Nov 12 16:12:56 2006 +1100 22855 22856 add radeon Init2 function, to abstract some mergedfb mode setting 22857 22858commit d0828804aec0852afaca44ee41c4792edc40dbb4 22859Author: Dave Airlie <airlied@linux.ie> 22860Date: Sun Nov 12 15:59:02 2006 +1100 22861 22862 add initial controller binding, remove uses of mergedfb test 22863 22864commit 78061ca460c9d6fd338ccf273bbfa152f471622e 22865Author: Dave Airlie <airlied@linux.ie> 22866Date: Mon Oct 23 16:49:59 2006 +1000 22867 22868 radeon: separate static controller setup from non-static 22869 22870commit f917f65e8d2b13d8b9bfc4320f4a935f5a5c4bff 22871Author: George Sapountzis <gsap7@yahoo.gr> 22872Date: Thu Nov 2 21:16:53 2006 +0200 22873 22874 Officially require libdrm 2.2.0 22875 22876 Note that this is only for the headers at compile-time. The driver still 22877 uses libdrm 1.2.0 from Xserver at run-time. 22878 22879commit ca8c148942371c9445f32985625ce38eabc016e1 22880Author: George Sapountzis <gsap7@yahoo.gr> 22881Date: Thu Oct 26 15:40:46 2006 +0300 22882 22883 Handle .99 in version patch level. 22884 22885commit 23cec754158c94b43225b2070fc0a0b003e7bd25 22886Author: George Sapountzis <gsap7@yahoo.gr> 22887Date: Thu Oct 26 15:39:49 2006 +0300 22888 22889 [mach64] EXA: reserve DRI buffers only when DRI is active. 22890 22891commit 9cd78af1d06f753e7bd7e1acaff25398bde45f6e 22892Author: Dave Airlie <airlied@linux.ie> 22893Date: Mon Oct 23 16:52:07 2006 +1000 22894 22895 radeon: set preferred modes 22896 22897commit 914a5f0117ccb76292bdd81ea4a2a79d33319550 22898Author: Michel Dänzer <michel@tungstengraphics.com> 22899Date: Mon Oct 23 10:13:24 2006 +0200 22900 22901 radeon: Pitch doesn't matter when height == 1. 22902 22903commit 2bcb51d66edaa944379cf8c8ca1ba91fffdc20a8 22904Author: Michel Daenzer <michel@tungstengraphics.com> 22905Date: Sun Oct 22 17:26:28 2006 +0200 22906 22907 Bug #6756: Attempt to fix repeat picture acceleration. 22908 22909 Always use normalized texture coordinates on R200 and fall back if a POT 22910 texture pitch doesn't match the HW's implicit pitch. 22911 22912commit dabffb8335027b60ca1fc554423e196dfb9acd6d 22913Author: Henry Zhao <henry.zhao@sun.com> 22914Date: Tue Oct 17 21:17:47 2006 -0400 22915 22916 improved results for mergedfb xrandr refresh rates - bug 6966 22917 22918commit 831fc4a009841f68fe161b6e0db5b212d7d8079b 22919Author: Henry Zhao <henry.zhao@sun.com> 22920Date: Sun Oct 15 17:12:06 2006 -0400 22921 22922 fix bug 6966 (Henry Zhao, Shem Multinymous) 22923 22924commit 5425be996692425e57884e5a4c159fc79510e89f 22925Author: Alex Deucher <alex@t41p.(none)> 22926Date: Sun Oct 15 16:49:24 2006 -0400 22927 22928 clean up magic number 22929 22930commit 457178f418983041ea59df703a17513a47d17044 22931Author: Alex Deucher <alex@t41p.(none)> 22932Date: Sun Oct 15 16:40:08 2006 -0400 22933 22934 Disable all ouputs on initial mode set. The ones we want will get 22935 enabled by RADEONEnableDisplay(). 22936 22937commit 3fb3f410931a13542fdd4c51c4c9ecada7b37d76 22938Author: Alex Deucher <alex@t41p.(none)> 22939Date: Sun Oct 15 13:29:48 2006 -0400 22940 22941 Rework the output init code. Divide into separate functions 22942 for each output. 22943 22944commit 45e13f3c8786e75276cce85de4af653713d9b4d8 22945Author: Alex Deucher <alex@t41p.(none)> 22946Date: Sun Oct 15 13:18:01 2006 -0400 22947 22948 Remove paneloff option. This is already covered by monitorlayout and 22949 shouldn't be needed with xrandr++. 22950 Also remove commented out Bioshotkey code. 22951 22952commit 4abb5b7afea8d37e95461335e52b16168e555dab 22953Author: Alex Deucher <alex@t41p.(none)> 22954Date: Sun Oct 15 13:10:22 2006 -0400 22955 22956 Remove Bioshotkeys option. Leave the bios as configured at boot. 22957 Most distros enable this option by defalt anyway, and with xrandr++ 22958 it won't be an issue anyway. 22959 22960commit 2933140b5f1f8a6e270460d8918a14943bd0f174 22961Author: Michel Dänzer <michel@tungstengraphics.com> 22962Date: Sun Oct 15 17:05:02 2006 +0200 22963 22964 radeon: Use xf86XVFillKeyHelperDrawable() when available. 22965 22966commit 4edbfb9eb952f9b3f5f597749192c17baf15cdce 22967Author: Michel Dänzer <michel@tungstengraphics.com> 22968Date: Sun Oct 15 17:03:13 2006 +0200 22969 22970 radeon: Track screen damage all the time for page flipping. 22971 22972 This way we don't ever have to forcedly synchronize the pages so we hopefully 22973 never clobber one page's contents when we shouldn't. 22974 22975commit c9c2ac2b42398f5bfad909f3289e2545bb7b3888 22976Author: Michel Dänzer <michel@tungstengraphics.com> 22977Date: Sat Oct 14 15:49:22 2006 +0200 22978 22979 radeon: Don't try to synchronize pages when the CP isn't running. 22980 22981 Also move the DamageEmpty() call into RADEONDRIRefreshArea(). 22982 22983commit c19915a401ccf69046bed03c8d5067f2e72a094e 22984Author: Dave Airlie <airlied@linux.ie> 22985Date: Fri Oct 6 09:55:40 2006 +1000 22986 22987 always turn on DAC for scanning the CRT connect. 22988 22989 Ubuntu 63994 - myself and BenH discussed this as a good idea on irc 22990 22991commit b8547aa89f93a7fefbefb1c0d2743999930ac785 22992Author: Dave Airlie <airlied@linux.ie> 22993Date: Wed Oct 4 01:44:24 2006 +1000 22994 22995 update version number to 6.6.99 in preparation for a 6.7 branch at some stage 22996 22997commit 23a6f97e08fd49e1cae03cd97cae67a5f06b7634 22998Author: Alex Deucher <alex@samba.(none)> 22999Date: Tue Oct 3 13:48:09 2006 -0400 23000 23001 FP timing regs required for both internal and external TMDS 23002 23003commit 829be884702043fe4c9e349488d2585bf0561ae9 23004Author: Alex Deucher <alex@samba.(none)> 23005Date: Tue Oct 3 13:41:21 2006 -0400 23006 23007 remove bogus registers 23008 23009commit cd900580dbc8928ebc28514162b4c1dbfe9fd5ae 23010Author: George Sapountzis <gsap7@yahoo.gr> 23011Date: Tue Oct 3 00:32:22 2006 +0300 23012 23013 [mach64] Fix build for !defined(USE_EXA). 23014 23015commit 22860a3c0685a25f982983654303ad02aedc02ec 23016Author: George Sapountzis <gsap7@yahoo.gr> 23017Date: Sun Jul 16 19:23:56 2006 +0300 23018 23019 Bug 6242: [mach64] Use private DMA buffers. 23020 23021 Map the DMA buffers read-only. This eliminates a security problem where a 23022 client can alter the contents of the DMA buffer after submission to the DRM. 23023 23024commit cb53fe7c1e45937746e43437ae6adb0355306ae9 23025Author: Alex Deucher <alex@t41p.(none)> 23026Date: Sun Oct 1 18:46:05 2006 -0400 23027 23028 fix LVDS plus internal TMDS 23029 23030commit 145c716de6d42778e447581578f3e4db13c9a653 23031Author: David Woodhouse <dwmw2@infradead.org> 23032Date: Sun Oct 1 18:22:14 2006 +0200 23033 23034 Fix radeon vs fbdev pitch calculation. 23035 23036 Fetch the correct displayWidth from fbdevhw after setting the mode. 23037 23038commit d43ad88fa3913437f6987ab5ab46a38f0cb555a8 23039Author: Michel Dänzer <michel@tungstengraphics.com> 23040Date: Sat Sep 30 19:25:20 2006 +0200 23041 23042 radeon: Add support for page flipping with EXA. 23043 23044 Also use the damage layer directly instead of via shadowfb and blit dirty 23045 rectangles to the second page in LeaveServer in order to try and improve the 23046 tradeoff between performance and correctness. 23047 23048commit 146ea328b7962f56c6ab47fbe2adc7561b36189c 23049Author: Michel Dänzer <michel@tungstengraphics.com> 23050Date: Sat Sep 30 19:22:26 2006 +0200 23051 23052 radeon: Fix function name typo which could probably lead to an unresolved symbol. 23053 23054 Also remove declaration of an unused variable. 23055 23056commit 6671c1b01bf29d8f1cacf9306ef658b967d8a3cf 23057Author: Michel Dänzer <michel@tungstengraphics.com> 23058Date: Sat Sep 30 18:29:45 2006 +0200 23059 23060 radeon: Always call RADEONGetAccessibleVRAM(). 23061 23062 This fixes issues with Option "UseFBDev" and the new memmap code. 23063 23064 OTOH, there is potential for this to break IGPs, but that's what Bugzilla and 23065 git-bisect are for... 23066 23067commit c979c7f87fc449cc9631820101edff82c3013e66 23068Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 23069Date: Sat Sep 30 15:14:42 2006 +0200 23070 23071 radeon: Check ROM more thoroughly before treating it as an x86 BIOS. 23072 23073 This prevents the ROMs on some Mac cards getting treated as x86 BIOSen, 23074 resulting in garbage values. 23075 23076commit 37c114acbabe35881c39642e293519e2635907aa 23077Author: Michel Dänzer <michel@tungstengraphics.com> 23078Date: Sat Sep 30 15:11:42 2006 +0200 23079 23080 Remove some dead code. 23081 23082 Fix two compile warnings. 23083 23084commit 6d7b0763173f56050411d7bdf9a583b9bfc050c4 23085Author: Adam Jackson <ajax@benzedrine.nwnk.net> 23086Date: Fri Sep 29 18:50:44 2006 -0400 23087 23088 Undo unintended configure.ac change. 23089 23090 Dammit, git, grow a sensible UI already. 23091 23092commit f96ddd32d3a9e64dc0adec5c50386d6db5b182ee 23093Author: Ronald Wahl <rwahl@gmx.de> 23094Date: Fri Sep 29 17:51:42 2006 -0400 23095 23096 Hook up DFP power to DPMS events. 23097 23098commit f6b42149c5a3fee663575fe6bbe78b62695bdbc8 23099Author: Alex Deucher <alex@samba.(none)> 23100Date: Fri Sep 29 00:44:36 2006 -0400 23101 23102 tvdac handling was reversed between r200 and other chips 23103 23104commit c0660495080719c052d6393ede707755929102cd 23105Author: Dave Airlie <airlied@linux.ie> 23106Date: Wed Sep 27 18:12:52 2006 +1000 23107 23108 radeon: check DDC info for both connectors not just the pScrn one. 23109 23110 Otherwise we disable DDC before we've checked the CRT if LVDS is in use 23111 23112commit 5b98545618141e86164f3ce66469f9e16b1cf6d4 23113Author: Dave Airlie <airlied@linux.ie> 23114Date: Wed Sep 27 18:10:30 2006 +1000 23115 23116 radeon: don't move connectors around if LVDS connector is connector 1 23117 23118 On my laptop this allow me to plug or not plug a monitor without affecting 23119 my lvds 23120 23121commit 74c725bd86348b5bf9bba0c50e8a6a7885d7ff7b 23122Author: Dave Airlie <airlied@linux.ie> 23123Date: Mon Sep 25 11:35:32 2006 +1000 23124 23125 radeon: cleanup some bits from superpatch 23126 23127commit 9827afaa40720a58e2fc029c0bf92ad2fe223d11 23128Merge: 10b4b46 ad8259c 23129Author: Dave Airlie <airlied@linux.ie> 23130Date: Mon Sep 25 11:25:17 2006 +1000 23131 23132 Merge branch 'radeon-sp' 23133 23134commit ad8259c814629e741ed5567923f40879cc0c7051 23135Author: Dave Airlie <airlied@linux.ie> 23136Date: Mon Sep 25 11:25:03 2006 +1000 23137 23138 radeon : add back tv-dac register write 23139 23140commit 10b4b46c16ff3748856b732f9a39de40ba197112 23141Author: Dave Airlie <airlied@linux.ie> 23142Date: Mon Sep 25 09:08:47 2006 +1000 23143 23144 remove all CVS annotations from the ATI driver files 23145 23146commit 976dec04786e9b050945c290840b4cdc978c0b23 23147Author: Dave Airlie <airlied@linux.ie> 23148Date: Mon Sep 25 09:00:21 2006 +1000 23149 23150 radeon: put back in reverseddc code 23151 23152 BenH claims it is needed, agd5f can fix it up later if it isn't. 23153 23154commit 980fb2f6bd7641c8c57769b0c67e3561903e89a9 23155Author: Dave Airlie <airlied@linux.ie> 23156Date: Sat Sep 23 08:31:56 2006 +1000 23157 23158 radeon: add some missing bits of superpatch for crtc registers 23159 23160commit 1ba4f36821e60289cad937abbb0edb273c88436c 23161Author: Dave Airlie <airlied@linux.ie> 23162Date: Sat Sep 23 08:25:42 2006 +1000 23163 23164 radeon: finish radeon_driver.c import for register changes 23165 23166commit 547543bbefe605a453bfa5ae6d063ae02c5f040e 23167Author: Alex Deucher <alexdeucher@gmail.com> 23168Date: Sat Sep 23 08:21:59 2006 +1000 23169 23170 radeon: re-organise FP and CRTC register setting routines 23171 23172commit fd978140bcb7670f28c684c06c2b6c611c26bef4 23173Author: Dave Airlie <airlied@linux.ie> 23174Date: Sat Sep 23 08:09:57 2006 +1000 23175 23176 radeon: initial CRTC register setting alignment 23177 23178commit 406f4911e7ce821002c111d6bcdec35f5b56c943 23179Author: Dave Airlie <airlied@linux.ie> 23180Date: Sat Sep 23 07:59:24 2006 +1000 23181 23182 radeon: setup the radeon init function to use secondary 23183 23184commit 452f0f8079d65679905ed5178a256534ef0db0e4 23185Author: Dave Airlie <airlied@linux.ie> 23186Date: Sat Sep 23 07:41:57 2006 +1000 23187 23188 radeon: bring over final set of changes for radeon_display.c from susperpatch 23189 23190 This brings over Alex's final set of changes... 23191 23192 i'm unsure about the disappearance of reverseddc - alex to review 23193 23194commit a8b9aa69259a0706d0688d2656ece5a292799242 23195Author: Dave Airlie <airlied@linux.ie> 23196Date: Sat Sep 23 07:38:25 2006 +1000 23197 23198 radeon: align tabs and make display detection like superpatch 23199 23200commit 8d4ae98c528a2309dc3c545893dbba2e7f76916c 23201Author: Dave Airlie <airlied@linux.ie> 23202Date: Sat Sep 23 07:26:44 2006 +1000 23203 23204 radeon: align some output with superpatch 23205 23206commit 65b54847c5e79b2ec3cb1a08b265229443c9deb9 23207Author: Dave Airlie <airlied@linux.ie> 23208Date: Sat Sep 23 07:22:31 2006 +1000 23209 23210 radeon: change function name to Displays rather than Monitors 23211 23212commit 7e16d9fff96dcdff8c25dbbe2a51be0023b7985a 23213Author: Dave Airlie <airlied@linux.ie> 23214Date: Sat Sep 23 07:21:15 2006 +1000 23215 23216 radeon: make GetConnectedMonitors a static void 23217 23218commit 6f63b3dc8b3e2299f5f02abc1c565176d2d46cee 23219Author: Dave Airlie <airlied@linux.ie> 23220Date: Sat Sep 23 07:19:43 2006 +1000 23221 23222 radeon: some minor alignments and cleanups for info-> 23223 23224commit c9f3ced3b1c194cacf1ab9436ee6c2d4efd3a8e9 23225Author: Alex Deucher <alexdeucher@gmail.com> 23226Date: Sat Sep 23 07:16:10 2006 +1000 23227 23228 radeon: add reverse display option to switch displays on non-mobility cards 23229 23230commit e752443c540d9a94522050c61ef7c33d912fece5 23231Author: Dave Airlie <airlied@linux.ie> 23232Date: Fri Sep 22 07:10:10 2006 +1000 23233 23234 radeon: flesh out mapcontrollers 23235 23236commit 7a0e4eb47655c4701fc3a4b33b6c8e9f378ccc9e 23237Author: Dave Airlie <airlied@linux.ie> 23238Date: Fri Sep 22 07:05:56 2006 +1000 23239 23240 radeon: add some info to the map controllers 23241 23242commit 75bf69c868be6164d0041a32f52244f1ef83e050 23243Author: Dave Airlie <airlied@linux.ie> 23244Date: Fri Sep 22 07:03:52 2006 +1000 23245 23246 radeon: add simple map controllers call 23247 23248commit 8ba5abd1329347c7fbf4b4d09c95da0d23b4884a 23249Author: Dave Airlie <airlied@linux.ie> 23250Date: Fri Sep 22 06:58:41 2006 +1000 23251 23252 radeon: move comment 23253 23254commit e8d77b7c36fa791a2350a6f4b15728b15d8f3de0 23255Author: Dave Airlie <airlied@linux.ie> 23256Date: Fri Sep 22 06:55:51 2006 +1000 23257 23258 radeon: refactor blank/unblank 23259 23260commit 8534723f5710e2916b4a22efe22a6b7365cc8db6 23261Author: Alex Deucher <alexdeucher@gmail.com> 23262Date: Fri Sep 22 06:36:03 2006 +1000 23263 23264 radeon: make DVI_I default connector 23265 23266commit d7fc79fd9f094ac462de0883af7e6030cb6195a6 23267Author: Dave Airlie <airlied@linux.ie> 23268Date: Fri Sep 22 06:35:34 2006 +1000 23269 23270 radeon: add enable display function 23271 23272commit 4a54886d510f26b29d27e5c9a73647554291b1a6 23273Merge: 70264d6 9fa176c 23274Author: Dave Airlie <airlied@linux.ie> 23275Date: Fri Sep 22 06:00:04 2006 +1000 23276 23277 Merge branch 'master' into radeon-spm1 23278 23279 Conflicts: 23280 23281 src/radeon_driver.c 23282 23283commit 9fa176c7989030e7340cc9b85d0b6a7b34303091 23284Author: Alex Deucher <alex@samba.(none)> 23285Date: Tue Sep 19 13:58:17 2006 -0400 23286 23287 Provide all resolutions (instead only the largest one) in MergedFB clone 23288 mode without config file (Henry Zhao) 23289 23290commit e742aeb28c7d9d6e75932c408bcc7c44af52e303 23291Author: Alex Deucher <alex@samba.(none)> 23292Date: Tue Sep 19 13:45:33 2006 -0400 23293 23294 fix large negative refresh rates with mergedfb (Henry Zhao) 23295 fixes bug 6966 23296 23297commit 9c0c805704f1df1b66adc2a6b9313597ae5b86cc 23298Author: Dave Airlie <airlied@linux.ie> 23299Date: Mon Sep 18 19:41:21 2006 +1000 23300 23301 radeon: fix rn50 memmap 23302 23303 This should fixup the rn50/m6/m7 memory map problem that has been reported 23304 since BenH's memory mapping changes 23305 23306commit 70264d645724718a48ca229fedc04e33681b7d31 23307Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23308Date: Mon Sep 18 09:56:25 2006 +1000 23309 23310 radeon; put back old TV DAC code for now 23311 23312commit 53eb843ad8ec5a292dff0844330ecfa9b7b2a0cb 23313Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23314Date: Mon Sep 18 07:55:21 2006 +1000 23315 23316 radeon: move code around in InitCrtcRegisters 23317 23318commit bcae4352079109a9dd16117bf9ec19dd215180bf 23319Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23320Date: Mon Sep 18 07:33:48 2006 +1000 23321 23322 radeon: cleanup DPMS set on/off code to use new controller structure 23323 23324commit 87622f56b3dd7e2d2f00ad71e5b569eac31fb2b1 23325Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23326Date: Mon Sep 18 07:15:54 2006 +1000 23327 23328 radeon: cleanup DacPowerSet 23329 23330commit a417fa081926b2f84e2cbdc355cc7462d1ebdfad 23331Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23332Date: Mon Sep 18 07:15:41 2006 +1000 23333 23334 radeon: fixup InitFPRegisters function call to not pass info 23335 23336commit 8bff57a938e92dfc8729e40f7aa9ad9a96262131 23337Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23338Date: Mon Sep 18 07:02:35 2006 +1000 23339 23340 radeon: copy of overupdated bios hotkeys code from Alex 23341 23342commit 76352c684d5e840c9f7ac0781e3ecb32951b32db 23343Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23344Date: Mon Sep 18 06:54:28 2006 +1000 23345 23346 radeon: minor cleanup of masking for crtc registers 23347 23348commit a7b97697ed13ac25844890439ffba31524e03a16 23349Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23350Date: Mon Sep 18 06:53:23 2006 +1000 23351 23352 radeon: initialise controller structures 23353 23354commit 362ee3778764633cb4c9388aa329c5b715885038 23355Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23356Date: Mon Sep 18 06:41:14 2006 +1000 23357 23358 radeon: move InitFPRegisters up to align with superpatch 23359 23360commit 11fb9a14741746c752b0001c1b1d39a54e69657a 23361Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23362Date: Mon Sep 18 06:35:50 2006 +1000 23363 23364 radeon: add tv dac adj lookup 23365 23366commit a15ac59c4d1a15380910be2a7438111b09161c81 23367Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23368Date: Mon Sep 18 06:35:28 2006 +1000 23369 23370 radeon: re-add ATI patch for rn50 modes 23371 23372commit d80ea2d67b63ea2f2ce0277952af186b69b19e7d 23373Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23374Date: Mon Sep 18 06:30:30 2006 +1000 23375 23376 radeon: re-add DAC power on/off code around detect 23377 23378commit 0a93936f9f048b8837f1c765fea7ccb61f072697 23379Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23380Date: Mon Sep 18 06:25:15 2006 +1000 23381 23382 radeon: restore secondary head behaviour 23383 23384commit 6da5a0bfdedb2062ec835f45514c22917abdcf4c 23385Author: Alex Deucher <alexdeucher@gmail.com> 23386Date: Mon Sep 18 06:16:58 2006 +1000 23387 23388 radeon: move to using Alex's controller structure 23389 23390commit cd1bc1d1febf0a79d835c7dc8b353123278dec51 23391Author: Alex Deucher <alexdeucher@gmail.com> 23392Date: Mon Sep 18 02:44:09 2006 +1000 23393 23394 radeon: add save/restore of some new registers 23395 23396 move some registers about 23397 23398commit 9b5c4cf8a98b0c20bb821f5003495481fbd0c10c 23399Author: Alex Deucher <alexdeucher@gmail.com> 23400Date: Mon Sep 18 02:28:32 2006 +1000 23401 23402 radeon: add pre init controller function 23403 23404commit 311e776c4429485d29aaf11d350cab8e6fe7564b 23405Author: Alex Deucher <alexdeucher@gmail.com> 23406Date: Mon Sep 18 02:21:49 2006 +1000 23407 23408 radeon: add missing tv-out registers to register list 23409 23410commit 65cb77c79c2c5cb536494200d4fba6815d7e771c 23411Author: Alex Deucher <alexdeucher@gmail.com> 23412Date: Mon Sep 18 02:21:29 2006 +1000 23413 23414 radeon: restructure files along lines of Alex's superpatch 23415 23416commit 25fa71ca0a24190cc8be2e80769443e0e09303b3 23417Author: Henry Zhao <henryz@ati.com> 23418Date: Mon Sep 18 03:15:54 2006 +1000 23419 23420 Power up DAC around check for CRT connectedness 23421 23422commit 39fcac431eba60adf2d4e357cfc811dbae36a4e1 23423Author: Henry Zhao <henryz@ati.com> 23424Date: Mon Sep 18 03:09:43 2006 +1000 23425 23426 Bump up clock rate for ES1000 to allow higher resolutions 23427 23428commit 7d9e2b450f0fa42553156f57572a6dab70a0ea13 23429Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 23430Date: Tue Sep 12 07:04:23 2006 +1000 23431 23432 radeon: this stops mergedfb working on my ATI M7 development board 23433 23434 I've no idea what this is for, it came from an ATI codedrop and it may 23435 fix some laptop or production system, but it looks to me more like a dirty 23436 hack than a true fix, as I have an M7 which I need to specify CRT for the 23437 second head and it breaks mergedfb for me. 23438 23439commit ad1778eb1eb59de14ec309b99131931cbccf64e9 23440Author: Michel Dänzer <michel@tungstengraphics.com> 23441Date: Sun Sep 10 23:32:44 2006 +0200 23442 23443 Revert "* [ Bug 6966 ] large negative refresh rates reported" 23444 23445 This reverts caaed927a07ffbac68b08246185ef93c1e7bb98c commit. 23446 23447 See https://bugs.freedesktop.org/show_bug.cgi?id=8137 . 23448 23449commit 6e92981b024b59c2237f949321f6b2591dada519 23450Author: Tilman Sauerbeck <tilman@code-monkey.de> 23451Date: Sun Sep 3 21:11:10 2006 +0200 23452 23453 Removed the warning re. the r300 DRI driver status. 23454 23455commit caaed927a07ffbac68b08246185ef93c1e7bb98c 23456Author: Henry Zhao <henryz@localhost.localdomain> 23457Date: Thu Aug 31 13:27:45 2006 -0700 23458 23459 * [ Bug 6966 ] large negative refresh rates reported 23460 * [ Bug 6309 ] radeon/r128 fails to read hsync/vsync rates when rang 23461 * descriptor is missing 23462 * [ Bug 5473 ] Blank screen with Radeon Mobility X700 (Acer Ferrari 4005) 23463 * Reboot should not be needed for external CRT to function when connected 23464 later on Ferrari 4000 23465 * Modes need to be sorted, and sorted correctly 23466 * Provide all resolutions (instead only the largest one) in MergedFB clone 23467 mode without config file 23468 * Bump up clock rate for ES1000 to allow higher resolution 23469 23470commit 6d2028b84ff2b82c849b74776258f8e52c952e36 23471Author: Dave Airlie <airlied@linux.ie> 23472Date: Fri Aug 25 10:05:44 2006 +1000 23473 23474 and of course it takes two commits to do a simple commit 23475 23476commit e90c6feddbcd719fda3bbc3d48fd55045e9d5adc 23477Author: Dave Airlie <airlied@linux.ie> 23478Date: Fri Aug 25 09:59:58 2006 +1000 23479 23480 add changelog hook from nv driver 23481 23482commit 1e06212f792d1ed1f7545e16824b86fbd34fa284 23483Author: Adam Jackson <ajax@benzedrine.nwnk.net> 23484Date: Thu Aug 24 19:52:30 2006 -0400 23485 23486 Enable CPIO in the atimisc driver on ia64 platforms. 23487 23488 This fixes many mach64s, but potentially breaks mach32. Mach64 is widespread 23489 on ia64 machines but PCI mach32's are quite rare on any platform. 23490 23491commit dfa47b09b5b8d7eeaa3f8ae629ebb3c7efb51511 23492Author: Adam Jackson <ajax@benzedrine.nwnk.net> 23493Date: Thu Aug 24 19:50:46 2006 -0400 23494 23495 Disable DRI on xpress 200{,m} since we know it's broken. 23496 23497commit 1dcd9b1f75ba2307beb42556aa7f2f945171d5fb 23498Author: Adam Jackson <ajax@benzedrine.nwnk.net> 23499Date: Thu Aug 24 19:42:26 2006 -0400 23500 23501 Disable DRI on RN50 in a more pleasant way. 23502 23503 Rather than publish a non-existant driver name for RN50, just don't bother 23504 initializing it in the first place. 23505 23506commit 64b18e246586b1043a12f381a9d4c834464ed9b5 23507Author: Dave Airlie <airlied@linux.ie> 23508Date: Fri Aug 25 09:40:04 2006 +1000 23509 23510 update version numbers for release 23511 23512commit be229d97e1f0282f2099c7ace7cb257a2738a58f 23513Author: Dave Airlie <airlied@linux.ie> 23514Date: Fri Aug 25 09:39:46 2006 +1000 23515 23516 make colortiling default to on all cards 23517 23518commit 750ed51bc3d5119e2d2a91a05ae5a971dcd3b88a 23519Author: George Sapountzis <gsap7@yahoo.gr> 23520Date: Mon Aug 7 17:09:02 2006 +0300 23521 23522 [mach64] RENDER support: drop caching DP_WRITE_MASK. 23523 23524 Caching DP_WRITE_MASK causes artifacts, e.g. with evas_xrender_x11_test. 23525 23526commit b2beea2fa8949874d3d57fb9b43fe85cc08a8bff 23527Author: George Sapountzis <gsap7@yahoo.gr> 23528Date: Sat Aug 5 18:36:24 2006 +0300 23529 23530 [mach64] RENDER support: save/restore, cache texture registers. 23531 23532commit 43aaed99950640c3695b3c2b91faabf00c6338a7 23533Author: George Sapountzis <gsap7@yahoo.gr> 23534Date: Sat Aug 12 22:12:07 2006 +0300 23535 23536 [mach64] RENDER support (bug #6877). 23537 23538 RENDER accleration is disabled by default. It is strongly recommended that the 23539 patch from bug #6772 and the 'exa-damagetrack branch' are merged in the xserver 23540 before enabling RENDER acceleration. To enable RENDER acceleration, add the 23541 following in xorg.conf: 23542 23543 Section "Device" 23544 [...] 23545 23546 Option "AccelMethod" "exa" 23547 Option "RenderAccel" "true" 23548 EndSection 23549 23550commit d113f40fd97408661193895ff3aa82262f100143 23551Author: Dave Airlie <airlied@linux.ie> 23552Date: Sat Aug 12 17:16:33 2006 +1000 23553 23554 radeon: make some local data static 23555 23556 This is inspired by a patch in XFree86 from David Dawes, it isn't the 23557 same as we have different gamma curves. 23558 23559commit a1da2b5e0a38feae01107676a76bdc5c185ad34b 23560Author: George Sapountzis <gsap7@yahoo.gr> 23561Date: Sat Aug 5 03:18:34 2006 +0300 23562 23563 [mach64] Halve max w/h for hardware-accelerated operations. 23564 23565 EXA hits these limits for some operations on offscreen pixmaps and GTPRO seems 23566 to support the reduced limits. 23567 23568commit a66c8d2f9855caa10142b39f155dd8232b1f2389 23569Author: George Sapountzis <gsap7@yahoo.gr> 23570Date: Sat Aug 5 03:16:56 2006 +0300 23571 23572 [mach64] EXA: add memcpy-based UTS/DFS. 23573 23574 EXA hits more optimized paths when it does not have to fallback because of 23575 missing UTS/DFS. 23576 23577commit 0fb8503d60cf646c91a806f9ed322f9ceff18baa 23578Author: George Sapountzis <gsap7@yahoo.gr> 23579Date: Sat Aug 5 03:16:15 2006 +0300 23580 23581 [mach64] XAA: minor refactoring of memory manager setup. 23582 23583 - DRI: factor out to separate function and add comments on layout 23584 - no DRI: factor out to separate function, maxScanlines is always equal to 23585 ATIMach64MaxY 23586 23587commit e203d86643d5d70bf18248712d05b72b79aee705 23588Author: George Sapountzis <gsap7@yahoo.gr> 23589Date: Sat Aug 5 03:15:42 2006 +0300 23590 23591 [mach64] EXA support. 23592 23593commit dc1e289a611a17090e6dc7ae8a8d3f26d20df4eb 23594Author: Samuel Thibault <samuel.thibault@ens-lyon.org> 23595Date: Sat Jul 29 01:05:35 2006 +0300 23596 23597 Bug #6623: Fix I2C bus Mach64 initialisation failure. 23598 23599commit b377f403e0cd5253402ef3945e5944d1a2763a94 23600Author: Marc Aurele La France <tsi@xfree86.org> 23601Date: Sat Jul 29 01:03:34 2006 +0300 23602 23603 Fix warnings for Option:1 bit fields. 23604 23605commit 91b8b0ca41ad0b9659f0982a05148cab8558e9d5 23606Author: Marc Aurele La France <tsi@xfree86.org> 23607Date: Sat Jul 29 01:02:48 2006 +0300 23608 23609 Bug #807: copy throttling. 23610 23611 Fix corruption associated with the engine randomly not waiting for a copy 23612 operation to commit its results. 23613 23614commit 7800407a2db4b7307192bf3b43fffb9c4748b9c0 23615Author: Marc Aurele La France <tsi@xfree86.org> 23616Date: Sat Jul 29 01:00:56 2006 +0300 23617 23618 Bug #807: invalidate read-back cache. 23619 23620 Fix corruption when scaling 1-pixel wide pixmaps, in which case the first CPU 23621 read of the framebuffer will return stale data. 23622 23623 This bug always manifests itself in mach64 EXA (to be committed soon), with 23624 this patch mach64 EXA passes rendercheck. 23625 23626commit 49b11c540d37152eee4907ab30353ede01e4fb56 23627Author: George Sapountzis <gsap7@yahoo.gr> 23628Date: Sat Jul 29 00:52:28 2006 +0300 23629 23630 Bug #807: split out TestRegisterCaching block for DP regs. 23631 23632commit 290c5aaa580428bc4748a47e4b9bcc22b90fc8d4 23633Author: George Sapountzis <gsap7@yahoo.gr> 23634Date: Sat Jul 29 00:51:53 2006 +0300 23635 23636 Bug #807: split out TestRegisterCaching block for DP regs. 23637 23638commit fba8c839b47e2c8d6a6d65950c3431ff5b870aa0 23639Author: Michel Dänzer <michel@tungstengraphics.com> 23640Date: Sun Jul 23 18:51:47 2006 +0200 23641 23642 Bug #6819: Fix broken rendering with *8b8g8r8 source formats. 23643 23644 Use the corresponding texture formats with R200 family. 23645 23646 Unfortunately, it looks like the R100 family can't accelerate these source 23647 formats, so we have to fall back for them. 23648 23649commit 60ebe97e2d5d27c2bf4213e201cef74503809cb6 23650Author: Tilman Sauerbeck <tilman@code-monkey.de> 23651Date: Mon Jul 3 19:37:28 2006 +0200 23652 23653 Move variable declarations before statements to be C89 compatible 23654 23655commit 6900ff3c5e574881b1470e2363b200e99ab8bf3e 23656Author: Alex Williamson <alex.williamson@hp.com> 23657Date: Mon Jul 3 13:09:47 2006 +0200 23658 23659 Bug #6872: Disable digital output while reading ROM with Radeon 7000. 23660 23661 This is a workaround for ATI Radeon 7000 erratum F1. 23662 23663commit 920ac1f56fad6f48a30e65836ce054815f270738 23664Author: Michel Dänzer <michel@tungstengraphics.com> 23665Date: Mon Jul 3 01:21:18 2006 +0200 23666 23667 Do even less cache flushing when the previous engine mode is known. 23668 23669 OTOH, flush everything when the previous engine mode is unknown, and mark the 23670 engine mode as unknown in a couple more cases. 23671 23672commit dd0dd4bab76907b4b1b6d814f24e73e57d71f587 23673Author: Eric Anholt <anholt@freebsd.org> 23674Date: Mon Jul 3 01:10:10 2006 +0200 23675 23676 Bug #6762: Remove gratuitous cache flushing. 23677 23678commit 22f165f290f57e638d198f24a37f62fc16b72bae 23679Author: Michel Dänzer <michel@tungstengraphics.com> 23680Date: Mon Jul 3 00:39:45 2006 +0200 23681 23682 Build and warning fixes for !defined(USE_XAA). 23683 23684commit ea5105230c7c3519730b8d3c250e9ba9a4e4374a 23685Author: Michel Dänzer <michel@tungstengraphics.com> 23686Date: Mon Jul 3 00:03:22 2006 +0200 23687 23688 Reserve HW cursor memory statically instead of from EXA offscreen. 23689 23690 As a result, quite a bit of code can be removed that attempted to deal with the 23691 EXA offscreen memory disappearing, with varying success. Because the HW cursor 23692 memory is now always immediately after the front buffer and before the back 23693 buffer, this also fixes bug #6808. 23694 23695 As the HW cursor memory is now reserved statically with EXA and XAA, change 23696 RADEONUseHWCursor() to check info->cursor instead of info->cursor_offset, for 23697 which 0 might become valid with HW cursor enabled in the future. 23698 23699 Change info->cursor_offset from unsigned long to CARD32 as the corresponding 23700 register is 32 bits wide, and to avoid issues with printf. 23701 23702commit 66586f905857ea5bf37cb5d34949cd450a9ed067 23703Author: Michel Dänzer <michel@tungstengraphics.com> 23704Date: Sun Jul 2 23:38:20 2006 +0200 23705 23706 Clean up some redundant code. 23707 23708commit 1159c85861dc507ebf323aca82e32fcc67726420 23709Author: Michel Dänzer <michel@tungstengraphics.com> 23710Date: Sun Jul 2 18:24:28 2006 +0200 23711 23712 Bug #6548: Fix playback of interlaced video with more than 1023 lines. 23713 23714 This will still break with 2048 lines... 23715 23716commit 11bedb7c845fc6074383890d5525315aebe05f2c 23717Author: Michel Dänzer <michel@tungstengraphics.com> 23718Date: Sun Jul 2 17:03:23 2006 +0200 23719 23720 Bug #6939: Disable RENDER acceleration by default on RN50. 23721 23722 Based on patch from Weixing Zhang, but still allows enabling it with Option 23723 "RenderAccel". 23724 23725commit 4a0c22e32901a510c7aad070be25c31519c284e3 23726Author: Weixing Zhang <anzhang@ati.com> 23727Date: Sun Jul 2 16:55:25 2006 +0200 23728 23729 Bug #6939: Disable 3D acceleration on RN50 by setting invalid ClientDriverName. 23730 23731commit 9c4a1e3300d30b23806988aa9328fc6ef92e2663 23732Author: Frank Gevaerts <frank@gevaerts.be> 23733Date: Thu Jun 29 18:21:22 2006 +0200 23734 23735 Fix build for !defined(USE_EXA). 23736 23737commit ee0544b670567491b7770f469a64a6cd85519ba1 23738Author: Michel Dänzer <michel@tungstengraphics.com> 23739Date: Thu Jun 29 18:19:58 2006 +0200 23740 23741 Remove unused variables. 23742 23743commit 15c556f5a8c9069d3dd00ce139d46c39b7298951 23744Author: Michel Dänzer <michel@tungstengraphics.com> 23745Date: Thu Jun 29 18:18:21 2006 +0200 23746 23747 Don't declare unused variables on __powerpc__ and __alpha__. 23748 23749commit 055ac2836ec0afd51cf554d043a59068c36a07d3 23750Author: Michel Dänzer <michel@tungstengraphics.com> 23751Date: Thu Jun 29 18:14:45 2006 +0200 23752 23753 Fix strict aliasing violation flagged by gcc -Wall. 23754 23755commit 4b1904017caa976c138594a86e75feaf470e72b5 23756Author: Dave Airlie <airlied@linux.ie> 23757Date: Mon Jun 26 20:34:34 2006 +1000 23758 23759 radeon: force CP and VIP clocks on some r300 and rv100 chips. 23760 23761 fglrx appears to do this on r300 and Jerome Glisse has spent most of his 23762 life tracking down what caused the 9800 to lockup. This is my attempt to 23763 fix this. Please report if it works for you. 23764 23765 Signed-off-by: Dave Airlie <airlied@linux.ie> 23766 23767commit c5d8c8db3e75976687bfb364f0a7b235c7ae8793 23768Author: Michel Dänzer <michel@tungstengraphics.com> 23769Date: Sat Jun 24 16:01:19 2006 +0200 23770 23771 Bug #7274: Fix corruption with 'small' accelerated DownloadFromScreen transfers. 23772 23773commit f25ba3dc0e7850f0b1b78f54c0e7211bf2e81b3a 23774Author: Tilman Sauerbeck <tilman@code-monkey.de> 23775Date: Fri Jun 16 11:38:18 2006 +0200 23776 23777 Remove obsolete FIXME. 23778 23779commit 764013ec61b9043f1487513658e374569f7154a9 23780Author: Tilman Sauerbeck <tilman@code-monkey.de> 23781Date: Fri Jun 16 10:56:02 2006 +0200 23782 23783 Use AC_CHECK_HEADER() to check for exa.h 23784 23785commit 3c701d49b3eb7adb82267cec53163e475913aca3 23786Author: Tilman Sauerbeck <tilman@code-monkey.de> 23787Date: Fri Jun 16 09:59:29 2006 +0200 23788 23789 Use AC_MSG_RESULT() instead of echo for check results. 23790 23791commit dbef62615f38aebf4cdc579e3b26a7e7c719abc4 23792Author: Tilman Sauerbeck <tilman@code-monkey.de> 23793Date: Fri Jun 16 08:50:49 2006 +0200 23794 23795 conditionally declare pPCI and PciReg, too 23796 23797commit a01c01787cfbe3706ae7747cde1657abd79ecdaa 23798Author: Michel Dänzer <michel@tungstengraphics.com> 23799Date: Wed Jun 21 09:38:47 2006 +0200 23800 23801 Bug #7283: Build fixes for !defined(XF86DRI). 23802 23803 Thanks to Matthieu Herrb for pointing out some of these. 23804 23805commit 79d1c573133e64ddf4446b6bb0eaab4a5a6c73ab 23806Author: Michel Dänzer <michel@tungstengraphics.com> 23807Date: Sun Jun 18 17:39:09 2006 +0200 23808 23809 Implement accelerated EXA DownloadFromScreen hook. 23810 23811 x11perf -getimage numbers only increase by about 20-30% on my PowerBook with an 23812 M9, but by about 100 times(!) with a PCIe X550. I suspect the former could 23813 perform better with PCI as opposed to AGP transfers, which would also remove 23814 the need to disable this by default with AGP. 23815 23816commit 2bcdc7cfd013bd457d35f927390e5cc3544f30f1 23817Author: Michel Dänzer <michel@tungstengraphics.com> 23818Date: Sun Jun 18 16:01:00 2006 +0200 23819 23820 Remove #include <fbdevhw.h>, not used here. 23821 23822commit 25a5ccb63a9e3074b4c8eaf48844877d84ff76bb 23823Author: Michel Dänzer <michel@tungstengraphics.com> 23824Date: Sun Jun 18 15:26:55 2006 +0200 23825 23826 Prefer visuals with stencil when it's hardware acclerated. 23827 23828 Also make the formatting of the if statements consistent. 23829 23830commit 8fcc436f19f9c1371295af2d0c45608fb43c584b 23831Author: Michel Dänzer <michel@tungstengraphics.com> 23832Date: Sun Jun 18 15:16:08 2006 +0200 23833 23834 Take part of the blame for the radeon driver. 23835 23836commit c8366d7b457565a06acdf6b40c5448c7f13f1641 23837Author: Michel Dänzer <michel@tungstengraphics.com> 23838Date: Sun Jun 18 14:24:14 2006 +0200 23839 23840 Add Option "DepthBits". 23841 23842 This allows getting higher depth buffer precision and stencil hardware 23843 acceleration in depth 16 or saving some video RAM at the cost of these features 23844 in depth 24. 23845 23846commit 23a21d00db6f3b88b54c44648f968018a03677d3 23847Author: Tilman Sauerbeck <tilman@code-monkey.de> 23848Date: Fri Jun 16 16:37:07 2006 +1000 23849 23850 [PATCH] move the declaration of pMach64 into the correct ifndef statement 23851 Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de> 23852 Signed-off-by: Dave Airlie <airlied@linux.ie> 23853 23854commit c6b33db9d2c751e34c2152b0bb7e0e84a812589f 23855Author: Dave Airlie <airlied@linux.ie> 23856Date: Fri Jun 16 16:22:03 2006 +1000 23857 23858 ati: prepare for release 23859 23860 Update some versions for release 23861 23862commit 9c534b5330534fc6cf392936a9b200a5128c787c 23863Author: Michel Dänzer <michel@tungstengraphics.com> 23864Date: Thu Jun 15 16:09:55 2006 +0200 23865 23866 Some more ignores. 23867 23868commit 3d68698647fb17a6f5ae9a225f552cb0130fa946 23869Author: Michel Dänzer <michel@tungstengraphics.com> 23870Date: Thu Jun 15 16:07:15 2006 +0200 23871 23872 Fix semantic glitches pointed out by gcc -Wall. 23873 23874 The radeon driver builds warning-free with gcc -Wall here now. 23875 23876commit ab2b48e0df328fd08a2412911126b3184a601aea 23877Author: Michel Dänzer <michel@tungstengraphics.com> 23878Date: Thu Jun 15 16:06:13 2006 +0200 23879 23880 Remove another unused variable. 23881 23882commit 56bd5a370472663af5607cd327f757367b238db0 23883Author: Michel Dänzer <michel@tungstengraphics.com> 23884Date: Thu Jun 15 15:35:05 2006 +0200 23885 23886 Fix some more cosmetic warnings. 23887 23888commit ec93e08b24f7c0e583970d08bfe18ad142e6b047 23889Merge: 051c49d 6ca9862 23890Author: Michel Dänzer <michel@tungstengraphics.com> 23891Date: Thu Jun 15 15:03:27 2006 +0200 23892 23893 Merge branch 'master' of ssh+git://git.freedesktop.org/git/xorg/driver/xf86-video-ati 23894 23895commit 6ca986242c80bcfe7ca2d435a772bbdf034892c4 23896Author: Dave Airlie <airlied@linux.ie> 23897Date: Thu Jun 15 21:24:46 2006 +1000 23898 23899 remove unused variable 23900 23901commit 9a5f5bc0d3f5e62b79355993ca6246382394daee 23902Author: Dave Airlie <airlied@linux.ie> 23903Date: Thu Jun 15 21:24:16 2006 +1000 23904 23905 remove all printf specifier warnings on Linux 23906 23907commit 051c49d0f002c72d1d8f15f6df264d8b093d4e95 23908Merge: a840afc dfac819 23909Author: Michel Dänzer <michel@tungstengraphics.com> 23910Date: Thu Jun 15 13:10:44 2006 +0200 23911 23912 Merge branch 'master' of ssh+git://git.freedesktop.org/git/xorg/driver/xf86-video-ati 23913 23914commit dfac8191dbda29ef7f6ce33d7356fee89c17d59b 23915Author: Dave Airlie <airlied@linux.ie> 23916Date: Thu Jun 15 21:05:16 2006 +1000 23917 23918 fix last ati*.[ch] format specifiers 23919 23920commit 33525540f3a9d6760c6edcfe765d2a004a6b894f 23921Author: Dave Airlie <airlied@linux.ie> 23922Date: Thu Jun 15 21:03:01 2006 +1000 23923 23924 fix more printf formats 23925 23926commit 6bcf6f29af24b6725e097dd72af8e8b9f6d0676c 23927Author: Dave Airlie <airlied@linux.ie> 23928Date: Thu Jun 15 20:59:33 2006 +1000 23929 23930 some more unused variable cleanups 23931 23932commit a840afc75a0453b4694e3f987ebdbddc6844a169 23933Merge: c90c49d d511ef4 23934Author: Michel Dänzer <michel@tungstengraphics.com> 23935Date: Thu Jun 15 12:46:58 2006 +0200 23936 23937 Merge branch 'master' of ssh+git://git.freedesktop.org/git/xorg/driver/xf86-video-ati 23938 23939commit d511ef486a15b582e5d1708a17a00701f5cf1ccd 23940Author: Dave Airlie <airlied@linux.ie> 23941Date: Thu Jun 15 20:39:22 2006 +1000 23942 23943 ati/r128: add includes to get C function decls 23944 23945 Also fix a void pointer cast 23946 23947commit f46c3bebe821ee0031770d98d04bd406d1abb9ed 23948Author: Dave Airlie <airlied@linux.ie> 23949Date: Thu Jun 15 20:34:56 2006 +1000 23950 23951 fix some more unused variables and bad parameter passing 23952 23953commit 6c3f81683abab8ff6b8f2fcaf01a2a1a63af5ef6 23954Author: Dave Airlie <airlied@linux.ie> 23955Date: Thu Jun 15 20:25:00 2006 +1000 23956 23957 ati: fix -Wall in atidri.c 23958 23959 atidri.c has some parameter passing bugs, fixed those and cleaned other 23960 warnings. 23961 23962commit 58c6aac0669101481ef6b8217e5cede4573ece42 23963Author: Dave Airlie <airlied@linux.ie> 23964Date: Thu Jun 15 19:58:11 2006 +1000 23965 23966 theatre: fix all -Wall warnings in theatre* 23967 23968 This fixes all the warnings, it doesn't clean up the horrible mess that is 23969 theatre*.[ch]. 23970 23971commit c90c49da3844f93d2d29070b15bb7def72b98b0f 23972Author: Michel Dänzer <michel@tungstengraphics.com> 23973Date: Thu Jun 15 10:29:46 2006 +0200 23974 23975 Collapse .gitignore files into single toplevel file, sorted with LC_ALL=C. 23976 23977 Suggested by Keith Packard. 23978 23979commit b7a3e4c9b33f72ae447f99c0c765f24fce988142 23980Author: Michel Dänzer <michel@tungstengraphics.com> 23981Date: Thu Jun 15 09:45:12 2006 +0200 23982 23983 Fix some implicit function declarations. 23984 23985commit 556ddc9c41020271e92389a3900d097ec443c625 23986Author: Michel Dänzer <michel@tungstengraphics.com> 23987Date: Thu Jun 15 09:30:43 2006 +0200 23988 23989 .cvsignore -> .gitignore 23990 23991commit 4708327969a21fc171dcd80a1b506028cc8e4f90 23992Author: Dave Airlie <airlied@linux.ie> 23993Date: Thu Jun 15 07:37:03 2006 +1000 23994 23995 ati: remove Changelog from ATI driver. 23996 23997commit ecbec8379c24bd6c16ced674d5e6affcd3985e46 23998Author: Michel Daenzer <michel@daenzer.net> 23999Date: Thu Jun 8 17:29:12 2006 +0000 24000 24001 Bug #5678: Fix setting of transfer mode for AGPv3 devices (Tilman 24002 Sauerbeck). 24003 24004commit 874ff105bc63c4fc242fbabaec04fde2cb81fdd4 24005Author: Daniel Stone <daniel@fooishbar.org> 24006Date: Sat Jun 3 10:50:33 2006 +0000 24007 24008 Bug #1889: Stop the CCE before an engine reset. (Conn) 24009 24010commit f9db5595fe4d62c54b3a501b2af64b4e2b7035cd 24011Author: Michel Daenzer <michel@daenzer.net> 24012Date: Sun May 21 14:33:34 2006 +0000 24013 24014 Derive ATI_VERSION_{MAJOR,MINOR,PATCH} from package version in order to 24015 hopefully confuse bug triagers a little less. 24016 24017commit 25ba85cab8b3d831da646fef19f01769410fc958 24018Author: Alex Deucher <agd5f@yahoo.com> 24019Date: Wed May 10 02:46:37 2006 +0000 24020 24021 - remove obsolete option "OverlayOnCRT2" (bug 6488) 24022 24023commit ede75fe2b1850a4d9f03c724b7c1d1ed65e60f2f 24024Author: Michel Daenzer <michel@daenzer.net> 24025Date: Tue May 9 14:40:23 2006 +0000 24026 24027 Drop support for radeon DRM <= 1.7. The 2.4 kernel has version 1.7, and 24028 various problems up to system hangs on X server startup have been 24029 reported with that and DRI enabled. 24030 24031commit bc93a46c4fe499f23b9c0ff52e2e36b4c587f7d6 24032Author: Michel Daenzer <michel@daenzer.net> 24033Date: Mon May 8 09:29:34 2006 +0000 24034 24035 Bugzilla #4640 <https://bugs.freedesktop.org/show_bug.cgi?id=4640> Patch 24036 #5273 <https://bugs.freedesktop.org/attachment.cgi?id=5273>: 24037 Add cases for FOURCC_RGBA32 and FOURCC_RGB24. This prevents clients from 24038 receiving incorrect data sizes with these formats. (Jan Schmidt) 24039 24040commit 2b0cdd9448a24ea067b0d78f319b99c1041df2e0 24041Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24042Date: Wed May 3 04:25:27 2006 +0000 24043 24044 Enable decoding of version 4 TMDS PLL config tables, they actually work, 24045 the problem in the comment only concerned pre-production cards that Hui 24046 tested apparently. Also set back the default value to what it was 24047 initially, fixes some more cards that don't have a proper table (Mac 24048 Mini for example) 24049 24050commit 36c04b88416905b9ae305583e8e34fefdbfafc50 24051Author: Michel Daenzer <michel@daenzer.net> 24052Date: Mon May 1 17:49:41 2006 +0000 24053 24054 Bugzilla #6755 <https://bugs.freedesktop.org/show_bug.cgi?id=6755> Patch 24055 #5536 <https://bugs.freedesktop.org/attachment.cgi?id=5536>: 24056 Change HostDataBlit interface to take dst_offset_pitch and coordinates 24057 instead of just a destination pointer, as the latter is not sufficient 24058 with tiling. Also, use HW clipping to avoid overwriting destination 24059 data outside of the specified width. 24060 Adapt to new HostDataBlit interface. This fixes corruption with 24061 UploadToScreen to the front buffer (from exaPutImage). 24062 24063commit b90d3710e67caad0686efab491c2b76ae877920c 24064Author: Michel Daenzer <michel@daenzer.net> 24065Date: Sat Apr 29 21:30:23 2006 +0000 24066 24067 Bugzilla #1992 <https://bugs.freedesktop.org/show_bug.cgi?id=1992> Patch 24068 #4988 <https://bugs.freedesktop.org/attachment.cgi?id=4988>: Fix Option 24069 "DisplayPriority" "HIGH" for R300 family. 24070 24071commit 30f3ce5c5ba205a23748def11de019ee727897f3 24072Author: Michel Daenzer <michel@daenzer.net> 24073Date: Sat Apr 29 20:34:57 2006 +0000 24074 24075 Properly de-initialize EXA driver data, so EXA can be initialized again in 24076 the next generation. 24077 24078commit d705429c230c9cc1645ae540a51dba62cae9f23e 24079Author: Michel Daenzer <michel@daenzer.net> 24080Date: Sat Apr 29 20:18:48 2006 +0000 24081 24082 Fix tests for (un)forcing HW cursor. The broken tests caused problems when 24083 switching to console with 3D clients running. 24084 24085commit 6142cb079d4a5844e4863c800baf4ae19c636af5 24086Author: Michel Daenzer <michel@daenzer.net> 24087Date: Sat Apr 29 20:12:38 2006 +0000 24088 24089 Prefer double-buffered visuals. As a consequence, the root window's visual 24090 is double-buffered by default, which is desirable in a number of 24091 situations, in particular when running a GLX compositing manager with 24092 AIGLX. 24093 24094commit eb841974d041b19461b7fb726687df8938b2e39b 24095Author: Eric Anholt <anholt@freebsd.org> 24096Date: Fri Apr 28 02:55:08 2006 +0000 24097 24098 Bug #6761: Fix font rendering in non-DRI mode by replacing RB2D_DSTCACHE_* 24099 with RB3D_DSTCACHE_*. The RB2D versions are read-only mirrors of the 24100 RB3D. This is masked when DRI is enabled because the DRM uses the right 24101 registers in its flushing. 24102 24103commit 85e2ee6de6fd347aad66785ec7119f92013cd994 24104Author: Dave Airlie <airlied@linux.ie> 24105Date: Wed Apr 26 08:38:47 2006 +0000 24106 24107 Make the ATI driver use the PCI BAR size to determine the MMIO register 24108 aperture size. Otherwise it tries to overmap things if they are up near 24109 the end of memory. 24110 24111commit 13b19edf5accb7427436d3f0b1c9a9efb4fe98fa 24112Author: Eric Anholt <anholt@freebsd.org> 24113Date: Wed Apr 26 07:26:15 2006 +0000 24114 24115 Add support for some component-alpha operations (InReverse, OutReverse) 24116 necessary for testing upcoming general component alpha acceleration 24117 assistance in EXA. Tested with rendercheck on an r100. 24118 24119commit 44a45bcb82d49349f4fdfae3e4b8f24f98d99b46 24120Author: Eric Anholt <anholt@freebsd.org> 24121Date: Tue Apr 25 23:31:56 2006 +0000 24122 24123 Use a CPPFLAGS with XORG_CFLAGS so that the exa test can actually find its 24124 headers and not fail. Also correct some nearby spelling. 24125 24126commit cc1825b503ce7e785cbb4611fd177477a1081b6e 24127Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24128Date: Mon Apr 24 07:44:52 2006 +0000 24129 24130 Fix enabling of external VGA output on some cards with dual entity setup 24131 instead of MergedFB. This was broken due to bogus manipulations of a 24132 register between the 2 entities. This fix is a big "violent" in that it 24133 always enables the analog output regardless of what is plugged. It 24134 should work though (at the expense of some additional power 24135 consumption). A proper fix should be easy to do once Alex gets his 24136 output mapping rework in. 24137 24138commit a66145bc9d487f7e19237cf292f12750cb959f61 24139Author: Dave Airlie <airlied@linux.ie> 24140Date: Fri Apr 21 00:38:44 2006 +0000 24141 24142 Bugzilla 6672: fix interrupts on radeon since memmap changes 24143 24144commit f1f057df9aaba42986d9e20afe6f609a90318333 24145Author: Dave Airlie <airlied@linux.ie> 24146Date: Fri Apr 21 00:34:43 2006 +0000 24147 24148 Add support for backing up the PCIE GART table on VT switch. This makes 24149 suspend/resume work a lot better on PCIE cards. 24150 24151commit 31396df3b0101571bb85a4a66b67f5a8b0c992a9 24152Author: Adam Jackson <ajax@nwnk.net> 24153Date: Sat Apr 8 01:19:17 2006 +0000 24154 24155 Bump to 6.6.0 and release. This module's versioning is completely screwed, 24156 so until we split out mach64/r128/radeon please just increment the 24157 package number whenever you make significant changes. 24158 24159commit d706f494ed22afd369991e8c75a2d4f5aed15305 24160Author: Aaron Plattner <aplattner@nvidia.com> 24161Date: Fri Apr 7 23:06:02 2006 +0000 24162 24163 Add a DrawablePtr argument to the XV functions to pave the way for 24164 redirected video. 24165 24166commit f94685aec26b65f60c00a4b1bcd8f26bd80eb6e4 24167Author: Eric Anholt <anholt@freebsd.org> 24168Date: Tue Apr 4 07:42:36 2006 +0000 24169 24170 Bug #5632: Fix up ati.4 to explain that it is a wrapper driver for the 24171 other three. While here, touch up radeon.4, and add an authors list 24172 gleaned from checking copyright headers. 24173 24174commit 58c9b1d84f514755945e7a0d25fd5530af34ce95 24175Author: Eric Anholt <anholt@freebsd.org> 24176Date: Sun Apr 2 21:45:12 2006 +0000 24177 24178 Add missing .cvsignore lines. 24179 24180commit f1ce6170ef21e1f9769337c3138cd8d65c33e40a 24181Author: Alex Deucher <agd5f@yahoo.com> 24182Date: Sat Apr 1 23:02:40 2006 +0000 24183 24184 - Fix dpi when switching from clone to dualhead with MergedFB. 24185 - Add ConstantDPI option to force a particlar dpi across mode changes Both 24186 based on Thomas Winischhofer's sis code. 24187 24188commit 607f18cfbfc2a7b5509f8b567bf510fce31b361e 24189Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24190Date: Mon Mar 27 06:12:57 2006 +0000 24191 24192 Fix a regression of my latest fix that broke color tiling on some setups. 24193 RADEONRestoreMemMapRegisters() must not clear CRTC_OFFSET_CNTL and 24194 friends unconditionally but only when there is a map change or it will 24195 clear them after the DRI updates the AGP location, thus turning tiling 24196 back off on the CRTC. Also remove a duplicate call to 24197 RADEONInitCommonRegisters(). 24198 24199commit bf1c4471137f10b7de9d79519ca2240058997e18 24200Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 24201Date: Thu Mar 23 18:04:17 2006 +0000 24202 24203 commit the actual changes instead of just the Changelog... 24204 24205commit aa5fff9bae4818bbf99e8d9caa8f2ce9905294ca 24206Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 24207Date: Thu Mar 23 17:59:25 2006 +0000 24208 24209 Use programmable tap coefficients for (hopefully) better video image 24210 quality when downscaling. Remove unnecessary code programming the tap 24211 coefficients twice. Disable some code which is supposed to improve 24212 filtering quality but causes very visible image quality problems at 24213 least with rv250 when upscaling. Fix potentially broken FOURCC_RGB16T 24214 format (untested if it was broken, untested if it works now...). Change 24215 some magic numbers to (new) macro defines. 24216 24217commit 17c52ff15ebe2cf220d5988ee13b599c41799886 24218Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24219Date: Thu Mar 23 01:37:15 2006 +0000 24220 24221 Call RADEONRestoreCommonRegisters() before RADEONRestoreCrtc2Registers() 24222 instead of after. I can't find any good reason why we did it backward 24223 until now and it definitely causes problems as the workaround for VT 24224 switch in RADEONRestoreCommonRegisters() will break output to the 24225 secondary head in various situations. Also does an unrelated minor 24226 update to a comment. 24227 24228commit 673e4d7501861066fdd0874919a4524b820b944b 24229Author: Kristian Høgsberg <krh@redhat.com> 24230Date: Thu Mar 23 00:06:56 2006 +0000 24231 24232 Add atipciids.h. 24233 24234commit 5dcb4152dca6ac1fe9342f44eb48c0198b15772e 24235Author: Kristian Høgsberg <krh@redhat.com> 24236Date: Wed Mar 22 22:30:14 2006 +0000 24237 24238 Convert use of xf86fopen() and other xf86 wrapped libc symbols to use libc 24239 symbols directly. The xf86* versions aren't supposed to be used 24240 directly. 24241 Drop libc wrapper; don't include xf86_ansic.h and add includes now missing. 24242 24243commit 21acdf371e3cf913c5affbd3e86641cfb115311c 24244Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24245Date: Sat Mar 18 00:08:24 2006 +0000 24246 24247 Clear the offsets when updating the memory map instead of when enabling the 24248 CRTCs and try to make that code more reliable. Doesn't disable the 24249 CRTCs beforehand. 24250 24251commit 323ecb92e40d71c5ef994b41b6d8dedba6dd6203 24252Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 24253Date: Fri Mar 17 03:00:53 2006 +0000 24254 24255 Fix various small cosmetic issues. Change a driver message, get the order 24256 right for requesting drm versions, replace the use of some numbers with 24257 the respective macro defines in radeon_video.c, and add some more macro 24258 defines. None of that really matters. 24259 24260commit dc552edf191e14249421f8b27146874f16b80c1a 24261Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24262Date: Thu Mar 16 21:53:58 2006 +0000 24263 24264 Add missing return statement (the driver would not start randomly), add 24265 RV410 to the list of new generation cards for the memory map setup and 24266 fix a bogus message. 24267 24268commit 901409f47cc990721c1648b1c08e23450e6a4452 24269Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24270Date: Thu Mar 16 04:32:22 2006 +0000 24271 24272 Fix incorrect CRTC2_OFFSET values when using old-style dual head (not 24273 MergedFB). Fixes Xinerama for me. 24274 24275commit 4d6656062129da0489eb4bc898871379ba891d8f 24276Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24277Date: Thu Mar 16 03:09:34 2006 +0000 24278 24279 Add various workarounds that seem to fix some remaning lockup scenarios I'm 24280 experiencing with the driver when setting the memory map. Some of the 24281 magic delays are a bit dodgy but they seem to work, I suppose I can't 24282 do better now without help from ATI. Also removed some really too noisy 24283 debug messages. 24284 24285commit 36799e3ea7ddde5fce528e8bc17ea942933a9111 24286Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24287Date: Wed Mar 15 04:03:37 2006 +0000 24288 24289 Fix various issues with DRI & server recycle by putting DRICloseSreen in 24290 the normal CloseScreen callback chain at a spot that matches the call 24291 to DRIFinishScreenInit in order to unwrap at the right time. 24292 24293commit 50923b6a9325592edc8b02b17646f96b9b8c4a01 24294Author: Luc Verhaegen <libv@skynet.be> 24295Date: Wed Mar 15 00:46:33 2006 +0000 24296 24297 Add some autoconf to keep this driver from building EXA when a slightly 24298 older server with older exa is installed. 24299 The way in which this problem was treated by the actually related 24300 developers explains a lot about this driver. 24301 24302commit 74f4ed4275866475b3c182fae962df2fc0946b85 24303Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24304Date: Mon Mar 13 02:57:17 2006 +0000 24305 24306 Fix typo in the function setting the color space transformation factors for 24307 old radeon 24308 24309commit 65625c6c2053792cd14b54c904bf46f360ef1565 24310Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24311Date: Mon Mar 13 00:09:44 2006 +0000 24312 24313 Earliest fix introduced a bug due to another chicken & egg problem in card 24314 detection. Now split PreInitConfig into PreInitChipType and PreInitVRAM 24315 so that PreInitDRI can be done just in between. Restores proper AGPx4 24316 functionality (when enabled) 24317 24318commit 3a290115af41b0ecb94eb201d7e7abb42ebbfec6 24319Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24320Date: Sun Mar 12 06:53:27 2006 +0000 24321 24322 Stop using xf86PciInfo.h, instead use a local copy of the PCI IDs we need 24323 in atipciids.h so we can update the ATI driver independently of the 24324 server when new chips are added 24325 24326commit d6b8f0a5eeb332563ec3ec7082f03be1797d26d8 24327Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24328Date: Sat Mar 11 04:19:47 2006 +0000 24329 24330 Add #ifdef XF86DRI guards around references to info->allowPageFlip and 24331 info->depthMoves, since they are only defined if building with DRI. 24332 24333commit 69fe90bc501df9bd9e74b894bfca79a4025725bd 24334Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 24335Date: Fri Mar 10 13:00:49 2006 +0000 24336 24337 Use "FBTexPercent" option for XAA too (only for increasing memory reserved 24338 for textures). 24339 24340commit 8033d7423267fc1b3725df6a077dd910b586daaf 24341Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24342Date: Thu Mar 9 23:26:27 2006 +0000 24343 24344 More memory map fixes !!! This time, we add a way to get the DRM kernel 24345 module version early, in order to properly limit the framebuffer size 24346 when a module that doesn't support the new map is loaded. That will 24347 improve backward compatibility. I also made the minimum required kernel 24348 module version 1.3 which hopefully shouldn't be a problem for anybody, 24349 and cleaned up the ordering in which we test for DRM version to enable 24350 features 24351 24352commit c637939d3c4ce5cdddbc5200d5ffbca4c3927050 24353Author: Michel Daenzer <michel@daenzer.net> 24354Date: Thu Mar 9 15:41:16 2006 +0000 24355 24356 RN50: Skip modes that exceed memory bandwidth. 24357 Bugzilla #5766 <https://bugs.freedesktop.org/show_bug.cgi?id=5766> Patch 24358 #4636 <https://bugs.freedesktop.org/attachment.cgi?id=4636> 24359 - Acknowledge that RN50 only has one CRTC, and use this to distinguish it 24360 from RV100. 24361 - Fix detection of RN50 memory type and bus width. 24362 - Model RN50 memory bandwidth limits by capping the pixel clock range based 24363 on memory clock, bpp and memory bus width. (ATI Technologies Inc.) 24364 24365commit 890b95a3ddad7634f3aea635cb236b221e902acc 24366Author: Eric Anholt <anholt@freebsd.org> 24367Date: Thu Mar 9 06:06:25 2006 +0000 24368 24369 Update drivers that support EXA for the new EXA ABI. This consists of 24370 moving all the accel and card members into the driver ptr, filling in 24371 the exa_major/ exa_minor fields, and always using LoadSubModule so we 24372 can check the module version up front. Only tested on ATI. 24373 24374commit e19ccfa18fee046c54d7a9401e4e0098b2300598 24375Author: Michel Daenzer <michel@daenzer.net> 24376Date: Fri Mar 3 16:41:41 2006 +0000 24377 24378 Add Option "FBTexPercent" to override the amount of video RAM reserved for 24379 OpenGL textures with EXA. 24380 24381commit ca00035eccb6e3495b27f8e64f2d016d8d2a9ce2 24382Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24383Date: Thu Mar 2 02:05:17 2006 +0000 24384 24385 Fix the fix ... I didn't interpret PciInfo->size properly and forgot that 24386 the function works in Kb not bytes... Ooops. 24387 24388commit b2145aea36bb035bff048366c607b967d70fff49 24389Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24390Date: Wed Mar 1 21:35:14 2006 +0000 24391 24392 Extend the alignement workaround to post-rv280 chips as well (thanks Hui) 24393 and also limit the CPU accessible memory to the size of the PCI BAR 24394 size (yeah, it was passed to the driver, I just got blind for a while 24395 it seems) 24396 24397commit 163581a82fe9395672c31b41cd67aa44158f5201 24398Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24399Date: Tue Feb 28 23:34:03 2006 +0000 24400 24401 Workaround for rv280 bug which needs the framebuffer memory mapping to be 24402 aligned to its own size. Also add back limit of 128Mb of mapped fb 24403 space for now until I fully sort out what's up with dual function 24404 cards. I would really need to access the PCI BAR sizes here but that 24405 information seem to be missing from the PCI info passed to the driver. 24406 24407commit 5ff8fb2bea1fe7d14db53c19ff3a3a56668ec417 24408Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 24409Date: Tue Feb 28 00:52:53 2006 +0000 24410 24411 Add the newly added (and tons of previously forgotten ones) pci ids to the 24412 ati wrapper too (probably not required for it to work as everything 24413 should default to radeon anyway there). 24414 24415commit 1574b41a0f894f9b6e30d3ee94e596695a4c64f0 24416Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24417Date: Sun Feb 26 10:01:32 2006 +0000 24418 24419 Fix page flipping with XAA. The new code ended up calling ShadowFBInit() 24420 before XAAInit(), which breaks ShadowFB. 24421 24422commit 1f43a584a684af8078631425886cce236500e0e6 24423Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 24424Date: Sat Feb 25 01:57:05 2006 +0000 24425 24426 Add pci ids known to exist (see #4284 for instance). There are still 24427 entries which probably don't really exist (cancelled cards and such), 24428 leave them as-is. Fix the name of some entries, mostly based on the 24429 .inf file of the newest catalyst driver. Use own family id for rv410 24430 and rs400, though there is no different code (yet?). 24431 24432commit 5c141bb15d1163e04c012a0cdf0699d534f0be37 24433Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24434Date: Thu Feb 16 23:27:44 2006 +0000 24435 24436 Memory map and misc fixes. This completely reworks the way the driver 24437 detects available & accessible video memory and initializes the card 24438 internal memory map. Along the way, I re-ordered some of the stuffs in 24439 pre-init in order to remove some dodgy codepath, fixed various init, 24440 exit and vt switch issues, and overall made the driver more robust. The 24441 full memory map fix requires an updated DRM. The X driver no longer 24442 tries to position the AGP aperture, this becomes entirely DRM 24443 responsibility. 24444 24445commit 8c6e5a6eb50d371fd4ec700c26aa628cb5f45731 24446Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24447Date: Wed Feb 15 23:36:55 2006 +0000 24448 24449 The engine setup would trigger a bogus line write before the proper 24450 addresses are setup, thus causing the card to try to bus master over 24451 the system bus to some stale location (usually 0 but could depend on 24452 whatever was used before X). Remove that useless init bit, and always 24453 setup some good enough engine source & destination pointers in case we 24454 have some other spurrious engine activity going on. 24455 24456commit d4ec0223029d8fe03a836a5a6e9a7cc8897b4c90 24457Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 24458Date: Mon Feb 13 05:30:13 2006 +0000 24459 24460 Fix EXA acceleration of textures with byteswap or tiling From John Clemens 24461 <john@deater.net> 24462 24463commit 79c60ebbc34aaf0ac0b0723898d5b63015547178 24464Author: Eric Anholt <anholt@freebsd.org> 24465Date: Tue Feb 7 04:39:59 2006 +0000 24466 24467 Fix a copy'n'paste-o in the EXA render implementation that resulted in 24468 [ax]8b8g8r8 pict formats not being supported on r200, but no other 24469 issues. 24470 24471commit 8d74f3d49e73626f5d658ba6586a7c1487dae141 24472Author: Daniel Stone <daniel@fooishbar.org> 24473Date: Thu Jan 19 14:20:37 2006 +0000 24474 24475 Bug #5656: Set all Radeon XPRESS 200 R[CS]4xx-based chips to have IsIGP = 24476 1, fixing modesetting issues. 24477 24478commit e490f4b86e6bba70f7c884fca3cf41d159323689 24479Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24480Date: Fri Jan 13 01:37:43 2006 +0000 24481 24482 XF86_VERSION_CURRENT -> XORG_VERSION_CURRENT 24483 24484commit ac296d9137b02753bc79a9930d168855adea9c95 24485Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24486Date: Fri Jan 13 01:02:01 2006 +0000 24487 24488 When failing to parse MetaModes, reset CRT2pScrn to NULL after free'ing it 24489 so we don't crash when trying to use it later (such as in 24490 RADEONFreeRec). 24491 24492commit dd98ce78407c9eddd7adaad1d73d6910e6949b87 24493Author: Adam Jackson <ajax@nwnk.net> 24494Date: Sun Jan 8 19:02:16 2006 +0000 24495 24496 Bug #5523: Make radeon compile without USE_XAA. (Tilman Sauerbeck) 24497 24498commit a99429b2bc6407ff2a094e6ee30ce2032f734966 24499Author: Kevin E Martin <kem@kem.org> 24500Date: Wed Dec 21 02:29:59 2005 +0000 24501 24502 Update package version for X11R7 release. 24503 24504commit 96e225b950c74e64f3f788be8a8489c5075ea55a 24505Author: Adam Jackson <ajax@nwnk.net> 24506Date: Mon Dec 19 16:25:52 2005 +0000 24507 24508 Stub COPYING files 24509 24510commit 6bea785ea0bdf01673a6ddc45765ae7372fdd371 24511Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24512Date: Mon Dec 19 09:07:50 2005 +0000 24513 24514 Typo fixes, mailing list & url updates, and other changes to prepare for 24515 X11R6.9 & 7.0 releases. 24516 24517commit 07d5cf8fc37eb72dbb344081ff07d59dbf568e9c 24518Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24519Date: Thu Dec 15 23:15:37 2005 +0000 24520 24521 Bugzilla #5238 <https://bugs.freedesktop.org/show_bug.cgi?id=5238> Patch 24522 #3978 <https://bugs.freedesktop.org/attachment.cgi?id=3978> r128 man 24523 page dualhead note update (Alex Deucher) 24524 24525commit 6db5cbfbf6d405d9871892d1255dd7442e4539e9 24526Author: Kevin E Martin <kem@kem.org> 24527Date: Thu Dec 15 00:21:32 2005 +0000 24528 24529 Update package version number for final X11R7 release candidate. Bump 24530 driver version number. 24531 24532commit 726fd3a81dfbd1b0c8e4ede104a48523daecab92 24533Author: Adam Jackson <ajax@nwnk.net> 24534Date: Wed Dec 14 19:41:28 2005 +0000 24535 24536 Bug #1760: Fix ati wrapper multihead to be dlloader friendly. 24537 24538commit 328a2d74dabe32ce35e879799412666feedb0eef 24539Author: Adam Jackson <ajax@nwnk.net> 24540Date: Fri Dec 9 18:25:02 2005 +0000 24541 24542 Bug #3869: Don\'t walk off the end of an array in connector detection. 24543 (Eric Anholt) 24544 24545commit 03961a05438e916b6725f4e547f23a84ead13c5e 24546Author: Adam Jackson <ajax@nwnk.net> 24547Date: Fri Dec 9 18:15:31 2005 +0000 24548 24549 Bug #1106: Fix memory allocation for Render. (Michel Daenzer) 24550 24551commit 091e5c584124f9b849c0f46802b2f0d8b0cc2388 24552Author: Kevin E Martin <kem@kem.org> 24553Date: Thu Dec 8 17:55:18 2005 +0000 24554 24555 Add configure options to allow hard-coded paths to be changed. 24556 24557commit e38e2df15c9a4b08dee6386a225e98827bb35aad 24558Author: Kevin E Martin <kem@kem.org> 24559Date: Thu Dec 8 17:54:40 2005 +0000 24560 24561 Allow hard-coded paths to be configurable. 24562 24563commit b8203604bf9bd3813668313a1c8c17040b9d93ce 24564Author: Kevin E Martin <kem@kem.org> 24565Date: Tue Dec 6 22:48:33 2005 +0000 24566 24567 Change *man_SOURCES ==> *man_PRE to fix autotools warnings. 24568 24569commit a86920e812bfd40ec9facfc234220e83066dd263 24570Author: Kevin E Martin <kem@kem.org> 24571Date: Sat Dec 3 05:49:32 2005 +0000 24572 24573 Update package version number for X11R7 RC3 release. 24574 24575commit 010adc8988846e6f76f111a9fd32eca021d12fbc 24576Author: Kevin E Martin <kem@kem.org> 24577Date: Sat Dec 3 05:45:17 2005 +0000 24578 24579 Bump version numbers for X11R6.9/X11R7 RC3. 24580 24581commit a5efa6724e7fd8318e262060b084630632dfcb44 24582Author: Kevin E Martin <kem@kem.org> 24583Date: Fri Dec 2 02:16:07 2005 +0000 24584 24585 Remove extraneous AC_MSG_RESULT. 24586 24587commit feeef3cba1f4736a96daf3002ec902ced2874198 24588Author: Adam Jackson <ajax@nwnk.net> 24589Date: Wed Nov 30 18:51:22 2005 +0000 24590 24591 Bump libdrm dep to 2.0. 24592 24593commit 7ca2b3fa30fedd07951f7c6e73ed4d2a6f6606ea 24594Author: Adam Jackson <ajax@nwnk.net> 24595Date: Tue Nov 29 23:29:58 2005 +0000 24596 24597 Only build dlloader modules by default. 24598 24599commit e167e5151e32d54919e96cbfdcf932e16b1cab2a 24600Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24601Date: Mon Nov 28 22:04:08 2005 +0000 24602 24603 Change *mandir targets to use new *_MAN_DIR variables set by xorg-macros.m4 24604 update to fix bug #5167 (Linux prefers *.1x man pages in man1 subdir) 24605 24606commit 0015b1a007bed61aaf389f8f927e4ae0568d7da7 24607Author: Eric Anholt <anholt@freebsd.org> 24608Date: Mon Nov 21 10:49:09 2005 +0000 24609 24610 Add .cvsignores for drivers. 24611 24612commit 00b8f6ccfbb5163c47e9def0e4ecc6ac7be7e8af 24613Author: Eric Anholt <anholt@freebsd.org> 24614Date: Fri Nov 11 11:05:42 2005 +0000 24615 24616 Bug #1028: Don't (incorrectly) mark the 24/8 z/stencil case as slow, which 24617 resulted in whining at runtime about the visual not being supported. 24618 24619commit 33326847ccc9cf4cd6a23a257e5392e348e6dd22 24620Author: Kevin E Martin <kem@kem.org> 24621Date: Wed Nov 9 21:31:20 2005 +0000 24622 24623 Bump version numbers for X11R6.9/X11R7 RC2. 24624 24625commit 68e11246bd6f6a641807230834dce4e7fd6911af 24626Author: Kevin E Martin <kem@kem.org> 24627Date: Wed Nov 9 21:15:14 2005 +0000 24628 24629 Update package version number for X11R7 RC2 release. 24630 24631commit 1582d3b34643babcbb794d2df8d7c9a755d569c9 24632Author: Matthias Hopf <mhopf@suse.de> 24633Date: Wed Nov 9 17:42:57 2005 +0000 24634 24635 Bug #4998: Add ';' and ',' as valid separators for MetaModes as well. 24636 24637commit 2d22d78d157e07e6ec6a3e2d03fc17f102f6e00e 24638Author: Kevin E Martin <kem@kem.org> 24639Date: Wed Nov 9 07:30:03 2005 +0000 24640 24641 Fix DoLoadableServer NO build. 24642 24643commit 999f67763a5ea9fef7b7b576262f0da079efcfee 24644Author: Michel Daenzer <michel@daenzer.net> 24645Date: Tue Nov 8 16:30:48 2005 +0000 24646 24647 bugzilla #988 (https://bugs.freedesktop.org/show_bug.cgi?id=988) Fix typo 24648 which may or may not have had a negative impact on stability with R300 24649 class cards. 24650 24651commit 5aa401efa0b8d81da418eafeb857142a33f85da9 24652Author: Michel Daenzer <michel@daenzer.net> 24653Date: Tue Nov 8 14:05:00 2005 +0000 24654 24655 bugzilla #4494 (https://bugs.freedesktop.org/show_bug.cgi?id=4494) Call 24656 RADEONDGAInit() even if EXA is enabled; the problematic DGA 24657 acceleration hooks are already disabled in that case. 24658 This fixes DGA issues such as DGA mouse (typically used by games in 24659 fullscreen mode) not working. 24660 Fix suggested by Eric Anholt. 24661 24662commit 345180131a7b2752773dc2a3ee1221ad1e05c815 24663Author: Michel Daenzer <michel@daenzer.net> 24664Date: Tue Nov 8 13:35:23 2005 +0000 24665 24666 bugzilla #4916 (https://bugs.freedesktop.org/show_bug.cgi?id=4916) Fix 24667 segfault with MergedFB and EXA enabled and HW cursor temporarily 24668 disabled. 24669 24670commit 23911421ffa7bdd2e191a72c41bf1f26c92c1a67 24671Author: Michel Daenzer <michel@daenzer.net> 24672Date: Tue Nov 8 09:44:41 2005 +0000 24673 24674 bugzilla #4951 (https://bugs.freedesktop.org/show_bug.cgi?id=4951) 24675 attachment #3689 (http://bugs.freedesktop.org/attachment.cgi?id=3689) 24676 Fix HW cursor getting silently disabled with EXA. 24677 24678commit e1259c0117a190cf4c6e7be07d5aa83e03afb3a0 24679Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 24680Date: Mon Nov 7 19:28:40 2005 +0000 24681 24682 Replaced the variadic macros(gcc) by macros according to C99 standard 24683 24684commit 67bdaae04ab37f38a6d0c8dc4b6328c3ba3b3393 24685Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24686Date: Sun Nov 6 07:41:29 2005 +0000 24687 24688 Sun bug #6325220: Xorg dumps core in dual monitor configuration with radeon 24689 driver if CRT2Position not specified in xorg.conf 24690 <http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6325220> 24691 (Stuart Krietman - Sun Microsystems) 24692 24693commit 733ebf6e4a7c7b2f1a861077d43da2a853e4c0d7 24694Author: Kevin E Martin <kem@kem.org> 24695Date: Tue Nov 1 15:08:52 2005 +0000 24696 24697 Update pkgcheck depedencies to work with separate build roots. 24698 24699commit dd0eb76337bfdaec8f456380a6ced9e0487e190f 24700Author: Adam Jackson <ajax@nwnk.net> 24701Date: Fri Oct 21 18:40:18 2005 +0000 24702 24703 Bug #4829: Remove some useless DRI symbol checks. 24704 24705commit e6c1859e3a7c84b882e0c30d0783255ed1258d9e 24706Author: Donnie Berkholz <spyderous@gentoo.org> 24707Date: Thu Oct 20 23:01:58 2005 +0000 24708 24709 Bug #4831 <https://bugs.freedesktop.org/show_bug.cgi?id=4831>. Check for 24710 x86_64 in addition to amd64 in host_cpu. 24711 24712commit 33460fb393c9591babcc43b30a73eef3971765c2 24713Author: Kevin E Martin <kem@kem.org> 24714Date: Wed Oct 19 02:48:02 2005 +0000 24715 24716 Update package version number for RC1 release. 24717 24718commit 77d5c3549db4f6a7ff9f969f885d7a4e58ab2b16 24719Author: Kevin E Martin <kem@kem.org> 24720Date: Tue Oct 18 15:46:37 2005 +0000 24721 24722 Do the following to make the drivers pass distcheck: 24723 - Only define EXTRA_DIST when there are actually man pages to be processed 24724 - Add missing man pages to keyboard and ati drivers EXTRA_DIST 24725 24726commit 35e717078c840414b0d6a0e37b3ffed0a7b102aa 24727Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24728Date: Tue Oct 18 00:01:52 2005 +0000 24729 24730 Use @DRIVER_MAN_SUFFIX@ instead of $(DRIVER_MAN_SUFFIX) in macro 24731 substitutions to work better with BSD make 24732 24733commit ff6e5b0412fe347575992c83bb7cb47b6f96f390 24734Author: Adam Jackson <ajax@nwnk.net> 24735Date: Mon Oct 17 22:57:32 2005 +0000 24736 24737 More 1.7 braindamage: define EXTRA_DIST in terms of @DRIVER_NAME@ instead 24738 of indirectly 24739 24740commit a2e7c55850d383e3af948d445b1645d8a47811cf 24741Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24742Date: Mon Oct 17 00:09:01 2005 +0000 24743 24744 Use sed & cpp to substitute variables in driver man pages 24745 24746commit 524fb1440760a14c53fb3f238aad6c3cdb7e6a08 24747Author: Dave Airlie <airlied@linux.ie> 24748Date: Sun Oct 9 10:36:50 2005 +0000 24749 24750 Secure the Radeon PCIE GART tables in framebuffer memory by not allowing 24751 the clients to get access to a secure area of framebuffer. 24752 24753commit e526211b86ec5343002978debfcf78f74a10c69e 24754Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 24755Date: Tue Oct 4 20:01:03 2005 +0000 24756 24757 Check in generated README files 24758 24759commit 431d0e144ae4c294e1f6110fc3db497ce9bd9cfd 24760Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 24761Date: Mon Oct 3 21:22:07 2005 +0000 24762 24763 - For all drivers that have a <driver>.sgml file, add code in their build 24764 system to build the README file at make dist time 24765 - in util/macros/xorg-macros.m4, add a new XORG_CHECK_LINUXDOC macro that 24766 will check if the required tools and files exist, and if so set a 24767 conditional. 24768 - util/modular/symlink.sh 24769 - Link all the <driver>.sgml to xf86-video-<driver>/README.sgml 24770 - Add all the README.<driver> to the list of excluded files 24771 - xc/programs/Xserver/hw/xfree86/doc/sgml/SiS.sgml: Various changes to make 24772 it spew less warnings when the text file is built. 24773 24774commit 9147c20e5ff662ac511a19e7623b406dffee2242 24775Author: Adam Jackson <ajax@nwnk.net> 24776Date: Mon Oct 3 06:14:58 2005 +0000 24777 24778 Bug #4001: Don't crash when PanelSize is unset. (Alan Coopersmith) 24779 24780commit 7e2090e3ebf0bd699d66d336353c097d658a86ae 24781Author: Matthieu Herrb <matthieu.herrb@laas.fr> 24782Date: Sat Oct 1 20:42:06 2005 +0000 24783 24784 missed one missing return value. 24785 24786commit a843e4281134de1545d538cb8970179b1d6977f6 24787Author: Matthieu Herrb <matthieu.herrb@laas.fr> 24788Date: Sat Oct 1 17:09:48 2005 +0000 24789 24790 add missing return value on success in RADEONSetupMemXAA(). 24791 24792commit 98f7d043f894c73fecaf14440c1a24b6c2fc4b06 24793Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 24794Date: Wed Sep 28 15:10:38 2005 +0000 24795 24796 Make ati driver distcheck 24797 radeon_commonfuncs.c \ 24798 radeon_exa_render.c \ 24799 radeon_exa_funcs.c \ 24800 24801commit e7b1af5d1677e53820dd1c0926cc8b5a4284901e 24802Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 24803Date: Tue Sep 27 23:55:15 2005 +0000 24804 24805 Move the RADEONEnablePageFlip/RADEONChangeSurfaces calls back to the end of 24806 the function (the code above will cause pixmap cache to be freed and 24807 copied around, and these functions must be called after that otherwise 24808 pixmap damage will occur). May be related to #4440, #4456 (unclear if 24809 these setups involve 3d apps) 24810 24811commit bc3c26fe365347c427e04385c1df95a5eec14f28 24812Author: Eric Anholt <anholt@freebsd.org> 24813Date: Tue Sep 27 23:34:11 2005 +0000 24814 24815 Fix Radeon MMIO Render acceleration with EXA by writing out floating- point 24816 coordinates rather than integers. 24817 24818commit 58dfb95c446b73fbc34784f0894cf89152dab043 24819Author: Adam Jackson <ajax@nwnk.net> 24820Date: Sat Sep 24 18:45:45 2005 +0000 24821 24822 Disable the {Open,Close}FullScreen DRI protocol. Remove empty FullScreen 24823 stubs from drivers, comment the non-empty ones. 24824 24825commit 65b5f0017e29903565aff0b7527d955ed016c9a6 24826Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24827Date: Tue Sep 20 16:53:34 2005 +0000 24828 24829 Only include radeon_dri.h & radeon_sarea.h if XF86DRI is defined. 24830 Include "compiler.h" to define __inline__ on non-gcc compilers 24831 24832commit 0b7441b19a1a5df91aca6b1a02944afb636bc523 24833Author: Eric Anholt <anholt@freebsd.org> 24834Date: Sun Sep 18 02:32:23 2005 +0000 24835 24836 Break EXA ABI while we still can. Add coordinates to the UploadToScreen 24837 hook so we can upload a subset of a pixmap, and convert the current 24838 drivers to respect that. Use this support to directly UploadToScreen in 24839 exaGlyphs, providing a 47.4% +/-2.4% decrease in wall time for ls -lR 24840 programs/Xserver in an antialiased gnome-terminal on an M6 (n=3, caches 24841 hot). I would have bumped major version, only I can't tell what the 24842 EXA_VERSION_* is supposed to be doing as opposed to the module version. 24843 24844commit 64e7db2698ee48af92594c16d72e33c16372c9d7 24845Author: Eric Anholt <anholt@freebsd.org> 24846Date: Sat Sep 17 20:02:01 2005 +0000 24847 24848 - Don't try to upload 0 byte-per-pixel (PICT_a1) data using 24849 RADEONHostDataBlit. 24850 - Disable the shortcut for switching from 3d to 3d in radeon_exa.c. It 24851 appears that we do need the cache flush here, thought it's not clear 24852 why. Disable the 2d to 2d shortcut while here, since I'm unsure of what 24853 we're doing. Exposed by the following bit: 24854 - Bug #4485: Add a new routine, exaGlyphs, to handle font drawing. Glyphs 24855 were being accumulated in from non-migratable scratch pixmaps, causing 24856 the destination pixmap to move towards screen but the migration 24857 necessary for source never to happen, leading to abysmal performance. 24858 Instead, copy the scratch glyph data into a real pixmap first, then 24859 composite from that into the destination, allowing for migration. time 24860 ls -lR from programs/Xserver showed 26.9% (+/- 6.3%) decrease in wall 24861 time (n=3). 24862 - Create exaDrawableUse* wrapping exaPixmapUse*, but which are aware of 24863 windows needing backing store. Makes migration code prettier, and 24864 ensures that composited windows will be migrated as normal when we turn 24865 off cw for EXA. (issue brought up by keithp) 24866 24867commit 3c71198a088e71a105e274f176c8ceb503615d46 24868Author: Dave Airlie <airlied@linux.ie> 24869Date: Sat Sep 17 07:47:51 2005 +0000 24870 24871 Switch radeon driver to using a card type rather than having a IsPCI flag, 24872 allow for PCI/AGP/PCIE cards. Set PCI gart allocation to only happen on 24873 PCIE cards. Add Radeon PCIE card detection using capability bits. 24874 24875commit 10cf3523f236a017a1ca7e5ee57e02176b924059 24876Author: Eric Anholt <anholt@freebsd.org> 24877Date: Wed Sep 14 22:06:41 2005 +0000 24878 24879 Bug #2573: Remove the disabling of automatic cache flushing on non-r300 24880 cards. It has been implicated in hanging at least r300s and some r100s, 24881 and seemed to be the cause of some bad rendering for me. 24882 24883commit d68894a463a3bb08c97e91ff6fc4b65b29183ff6 24884Author: Eric Anholt <anholt@freebsd.org> 24885Date: Tue Sep 13 22:04:12 2005 +0000 24886 24887 Remove the hardcoded USE_EXA/USE_XAA defines now that they've been 24888 hardcoded in configure.ac (config.h). 24889 24890commit baa9b5b6e19cf9794bf810e3c582bf1d7f808663 24891Author: Dave Airlie <airlied@linux.ie> 24892Date: Mon Sep 12 09:27:07 2005 +0000 24893 24894 Fix Radeon EXA for when RENDER isn't defined 24895 24896commit 8c89610389197dc94d3bceaf44d9d2ea91848f1b 24897Author: Dave Airlie <airlied@linux.ie> 24898Date: Mon Sep 12 09:17:42 2005 +0000 24899 24900 Move GART offset/size outside XAA 24901 24902commit eba44d9747adde8385f2a242dee5a8e73ce5a205 24903Author: Daniel Stone <daniel@fooishbar.org> 24904Date: Mon Sep 12 09:04:36 2005 +0000 24905 24906 Define USE_XAA and USE_EXA for the time being. 24907 24908commit a255e7b0984890daad98b74ace0c572965fe61f5 24909Author: Eric Anholt <anholt@freebsd.org> 24910Date: Sun Sep 11 21:00:44 2005 +0000 24911 24912 Attempt to fix up the modular pieces for EXA support. (untested) 24913 24914commit 3e0d9c945a6a71cc476d27341ef18618529a91a8 24915Author: Eric Anholt <anholt@freebsd.org> 24916Date: Sun Sep 11 20:58:53 2005 +0000 24917 24918 Add support for EXA to the radeon driver. Building EXA and XAA support is 24919 controlled at compile time, plus the runtime option of Option 24920 "AccelMethod" "EXA" or "XAA". The XAA support appears to remain as 24921 before, while the EXA pieces need just a little more polishing. Notable 24922 features: 24923 - Render acceleration working on Radeon 100 and 200-series with DRI on. 24924 - DRI works with EXA Notable issues: 24925 - DGA disabled in the EXA case. 24926 - Backbuffer moves disabled in the EXA case. 24927 - No textured XVideo. 24928 - MMIO render acceleration is close but still has some issues. 24929 - Memory pressure while using Composite is really troublesome with DRI on. 24930 This patch is based on an initial patch by Zack Rusin, with significant 24931 work by Benjamin Herrenschmidt and myself. 24932 24933commit fd62082b68ac3aadd8ffc441352d75d88334904e 24934Author: Dave Airlie <airlied@linux.ie> 24935Date: Sun Sep 11 08:51:38 2005 +0000 24936 24937 Add support for allocating PCI GART table in framebuffer memory. This is 24938 needed to support PCIE Radeons using a new DRM. 24939 24940commit cb63f8d9c0563fb0eff28e2be6d4adf5666540d2 24941Author: Dave Airlie <airlied@linux.ie> 24942Date: Tue Sep 6 10:03:19 2005 +0000 24943 24944 Handle VIP timeouts more gracefully -- impose a hard limit of 10ms on 24945 waiting for VIP_BUSY so we don't lock up hard, spinning. 24946 24947commit dbd83cf18e1bf5e7e31e1ac150227bd27e8b1739 24948Author: Adam Jackson <ajax@nwnk.net> 24949Date: Sun Aug 28 20:51:14 2005 +0000 24950 24951 Bug #4148: Bump required DRM version for r300 to 1.17. 24952 24953commit ce0589ff3b9fcf8c6bab64da4dba40ba540a6a03 24954Author: Adam Jackson <ajax@nwnk.net> 24955Date: Sun Aug 28 18:10:34 2005 +0000 24956 24957 Bug #1109: Fix VGA init on Rage Mobility 7500 (Marc Le France) 24958 24959commit 250c8732a54924eeb459880f2905f8e59d038e60 24960Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 24961Date: Sun Aug 28 18:00:23 2005 +0000 24962 24963 Modified the licens type from GPL to a X/MIT one 24964 24965commit 94c0196652c3cfa0558f61b55f5df9bf50f9ac49 24966Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 24967Date: Fri Aug 26 20:04:01 2005 +0000 24968 24969 #include "extensions/dpms.h" -> <X11/extensions/dpms.h> 24970 24971commit 107e42cabe1cc41fe86b0218c3229b5f34b2df84 24972Author: Daniel Stone <daniel@fooishbar.org> 24973Date: Fri Aug 26 07:00:24 2005 +0000 24974 24975 Fix 'ugly' warning in r128. (Branden Robinson, Debian #069) 24976 24977commit d3f5911c85ccfaaf50c05ce80d56dac13fd15701 24978Author: Daniel Stone <daniel@fooishbar.org> 24979Date: Fri Aug 26 05:30:52 2005 +0000 24980 24981 Only use vgahw symbols if we're building vgahw for this architecture. TODO: 24982 Fix this for the modular tree (maybe an unconditional yes?). Define 24983 that we're using a DPMS server and include DPMS headers. (Debian #024b) 24984 24985commit 879213455740d2d653d38c712275f07610e4dc9c 24986Author: Alex Deucher <agd5f@yahoo.com> 24987Date: Fri Aug 26 02:11:24 2005 +0000 24988 24989 - fix pitch increment (Aapo Tahkola, Roland Scheidegger) 24990 24991commit 6519cc71e4387ef979de90be517145611511ff1a 24992Author: Daniel Stone <daniel@fooishbar.org> 24993Date: Thu Aug 25 03:34:58 2005 +0000 24994 24995 Fix bug in Radeon connector table parsing -- parse all entries, not just 24996 the first. (Ben Herrenschmidt) 24997 24998commit 8734a5d0bafb60f5e38b2cee10587f01fc016379 24999Author: Daniel Stone <daniel@fooishbar.org> 25000Date: Thu Aug 25 03:07:50 2005 +0000 25001 25002 Blacklist Radeon M300 from multimedia I2C code. Handle I2C timeouts more 25003 gracefully -- impose a hard limit of 10ms on waiting for I2C_GO so we 25004 don't lock up hard, spinning. 25005 25006commit 578564b4f7fce1f02d056c2b12ffb7803d268829 25007Author: Adam Jackson <ajax@nwnk.net> 25008Date: Mon Aug 22 03:05:52 2005 +0000 25009 25010 Explicit license statement. 25011 25012commit c9005c91bf6aaf4728a6606284f5d24e0b4a8282 25013Author: Daniel Stone <daniel@fooishbar.org> 25014Date: Thu Aug 18 09:03:46 2005 +0000 25015 25016 Update autogen.sh to one that does objdir != srcdir. 25017 25018commit 508a7421a1386b5b783a867d860b858a55f216d3 25019Author: Alan Hourihane <alanh@fairlite.demon.co.uk> 25020Date: Mon Aug 15 07:30:05 2005 +0000 25021 25022 Egbert's 64bit fixes for mixed 32/64bit clients 25023 25024commit 204290412fbe8b4ac1a0f4f97bc7aec2847a1df7 25025Author: Daniel Stone <daniel@fooishbar.org> 25026Date: Sun Aug 14 16:14:51 2005 +0000 25027 25028 Make DGA, CPIO, non-PCI, and TV-Out support detected per-platform, using 25029 the same semantics as in drivers/ati/Imakefile. 25030 25031commit adb093bdf3cea8108a40bcd72f09924b416de0c1 25032Author: Alex Deucher <agd5f@yahoo.com> 25033Date: Mon Aug 8 23:42:36 2005 +0000 25034 25035 - radeon XPRESS chips are actually RV3xx based 25036 25037commit 9c11018beb186b4b303524685fb62629b9ca863e 25038Author: Vladimir Dergachev <volodya@mindspring.com> 25039Date: Sun Aug 7 14:46:55 2005 +0000 25040 25041 Modified: 25042 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 25043 Update warning about DRI support on R300 and later cards. 25044 25045commit a28b2c76c7df2fe65a93eed34bc85c2cf5afb314 25046Author: Vladimir Dergachev <volodya@mindspring.com> 25047Date: Sun Aug 7 14:40:58 2005 +0000 25048 25049 Modified: 25050 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 25051 Properly configure fbLocation and related registers for R300 cards. (Dave 25052 Airlie) 25053 25054commit 692fb0f8b51abeccaf70b51fc1defc62132b15e9 25055Author: Alex Deucher <agd5f@yahoo.com> 25056Date: Sat Aug 6 17:23:16 2005 +0000 25057 25058 - missing line in mergedfb option parsing. Spotted by Christoph Bauer 25059 25060commit 14e77a39320a7a074b5c2eeeea9750b469abc978 25061Author: Ian Romanick <idr@us.ibm.com> 25062Date: Fri Aug 5 23:04:32 2005 +0000 25063 25064 I have resolved this by making all the "Direct rendering disabled" messages 25065 X_WARNING and all the "Direct rendering enabled" messages X_INFO. 25066 Ideally, we need to factor some of this stuff out to common code. There 25067 is a *LOT* of duplication among the *_dri.c files under 25068 programs/Xserver/hw/xfree86/drivers. 25069 I'm trying to clean up as much low hanging fruit as I can and got tired of 25070 waiting for ajax. :P 25071 25072commit 8e88d36f78c13025847310b411402f1d46a2e0cd 25073Author: Adam Jackson <ajax@nwnk.net> 25074Date: Sun Jul 31 17:19:27 2005 +0000 25075 25076 Post-merge fixups: 25077 - Disable building the gamma driver 25078 - Fix numerous Imakefiles to get header paths and subdirs right 25079 - Symlink client-side GLX lib source from Mesa 25080 - Add new shader subdirs to the Mesa core build 25081 - Tweak the DDX drivers to not ask for "xf86dri.h"; they don't need it, 25082 since they don't generate protocol, and it was breaking the build. 25083 25084commit fb9f712fd1044feb8ce6a99aabaeb5966123e268 25085Author: Kevin E Martin <kem@kem.org> 25086Date: Fri Jul 29 21:22:42 2005 +0000 25087 25088 Various changes preparing packages for RC0: 25089 - Verify and update package version numbers as needed 25090 - Implement versioning scheme 25091 - Change bug address to point to bugzilla bug entry form 25092 - Disable loadable i18n in libX11 by default (use --enable-loadable-i18n to 25093 reenable it) 25094 - Fix makedepend to use pkgconfig and pass distcheck 25095 - Update build script to build macros first 25096 - Update modular Xorg version 25097 25098commit ee45e1fb5da793fa14770b28fd662e69d52b2c6a 25099Author: Michel Daenzer <michel@daenzer.net> 25100Date: Fri Jul 29 19:45:14 2005 +0000 25101 25102 bugzilla #3911 (https://bugs.freedesktop.org/show_bug.cgi?id=3911) 25103 attachment #3191 (http://bugs.freedesktop.org/attachment.cgi?id=3191) 25104 Disable bus mastering while updating MC_FB_LOCATION and friends to 25105 prevent the X server from hanging on startup every now and then under 25106 some circumstances. (ATI Technologies Inc.) 25107 25108commit 44a66ec5a01c3b4dfa046a66b80a3b7e9dfff537 25109Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 25110Date: Fri Jul 15 15:02:47 2005 +0000 25111 25112 Install theathre drivers in @moduledir@/multimedia 25113 25114commit dbcf104689ca368523715eb5945bcd800985799b 25115Author: Kevin E Martin <kem@kem.org> 25116Date: Fri Jul 15 03:39:58 2005 +0000 25117 25118 Fix filename typo 25119 25120commit 2256fabf1cd69c950639f236f61e59e68cc8e6fa 25121Author: Kevin E Martin <kem@kem.org> 25122Date: Thu Jul 14 22:30:47 2005 +0000 25123 25124 Fix check for headers. Use {XORG,DRI}_CFLAGS subst instead of replacing 25125 CFLAGS. 25126 25127commit 520317863ffb365eca17c96b4d545adb554bbbd6 25128Author: Kevin E Martin <kem@kem.org> 25129Date: Thu Jul 14 19:35:48 2005 +0000 25130 25131 Make compiling with DRI support configurable/autodetecable. Autodetect DRI 25132 sdk headers. Expand help string for xorg-module-dir option. 25133 25134commit ab5264838ae34bb29791bc08d5a4f853a995faf1 25135Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 25136Date: Thu Jul 14 17:10:01 2005 +0000 25137 25138 fuck you cvs 25139 25140commit a608c4422665ce1e933483b7c2a25ab2daae5930 25141Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 25142Date: Thu Jul 14 16:03:36 2005 +0000 25143 25144 Try adding build system for ATI driver again. Who knows, maybe CVS works 25145 today. 25146 25147commit d715561d28d82591adf5c7504068c1818eb2245c 25148Author: Adam Jackson <ajax@nwnk.net> 25149Date: Mon Jul 11 02:29:45 2005 +0000 25150 25151 Prep for modular builds by adding guarded #include "config.h" everywhere. 25152 25153commit 5b37b695019b7a4522a96a63f1f6b035160e6a25 25154Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 25155Date: Mon Jul 4 23:44:59 2005 +0000 25156 25157 Bug #2604: Disable workaround for horrible flickering when scrolling 25158 vertically in a virtual screen when color tiling and page flip are both 25159 enabled, as it potentially does more harm than good. The flickering 25160 still needs fixing probably though. 25161 25162commit 08b886850f115f404187acd619aa2c2ca18ae3c7 25163Author: Daniel Stone <daniel@fooishbar.org> 25164Date: Fri Jul 1 22:43:11 2005 +0000 25165 25166 Change all misc.h and os.h references to <X11/foo.h>. 25167 25168commit 1c7c3372776d6d116facabbf055f27af3e96f59f 25169Author: Alex Deucher <agd5f@yahoo.com> 25170Date: Fri Jul 1 04:16:42 2005 +0000 25171 25172 - fix depth tiling for r3/4xx chips (Aapo Tahkola) 25173 - adjust limits for tiled surfaces on r3/4xx 25174 25175commit e8c2574557950bfa150806432d566cd38c47158f 25176Author: Alex Deucher <agd5f@yahoo.com> 25177Date: Fri Jul 1 02:56:04 2005 +0000 25178 25179 - Fix Support for Philips FM1236/F tuner on ATI AIW 9600 XT (Jeff Smith) 25180 Bug 3401 25181 25182commit d258e1c1566d005693260230712c9d78f1447dba 25183Author: Alex Deucher <agd5f@yahoo.com> 25184Date: Thu Jun 30 04:46:11 2005 +0000 25185 25186 - change back part of a comment I seem to have accidently changed. 25187 25188commit 6d9c02d58eae1966e8fa8d7d305849c6e57434ad 25189Author: Alex Deucher <agd5f@yahoo.com> 25190Date: Thu Jun 30 02:02:10 2005 +0000 25191 25192 - small mergedfb fix from Thomas Winischhofer 25193 25194commit 85365b791542cab221c37c9b3b4593ea0c6c9e2d 25195Author: Alex Deucher <agd5f@yahoo.com> 25196Date: Thu Jun 30 01:40:17 2005 +0000 25197 25198 - add support for colortiling on r3/4xx hardware (disabled by default). In 25199 order for pageflipping to work on r3/4xx we will need to update the the 25200 sarea and drm so that the crtc1/2 xytiling regs get updated. Initially 25201 worked out by Aapo Tahkola. 25202 25203commit 3ec99c9cd821aeacdd57a9bbc20c4e4ec4711790 25204Author: Alex Deucher <agd5f@yahoo.com> 25205Date: Thu Jun 30 00:56:01 2005 +0000 25206 25207 - Increase the priority of display 0/1 in the memory controller for r3/4xx 25208 hardware when displaypriority is set to HIGH. Fixes display problems in 25209 high res modes. Originally reported by Aapo Tahkola. 25210 25211commit a7523e6c0be7a600963e9de10cf0060d16117bf6 25212Author: Alex Deucher <agd5f@yahoo.com> 25213Date: Thu Jun 30 00:33:36 2005 +0000 25214 25215 - fix some OUTREGs that should be OUTPLLs in dynamicclocks code 25216 25217commit c7e1d31c6aec3a99061c77c4d6f57e2e874e37f1 25218Author: Adam Jackson <ajax@nwnk.net> 25219Date: Sun Jun 26 04:38:12 2005 +0000 25220 25221 Bug #3628: Fix video driver submodule visibility. 25222 25223commit 1beca962ecee5b5c51eb2fb8451fcc2a2ed4b97c 25224Author: Adam Jackson <ajax@nwnk.net> 25225Date: Sat Jun 25 21:16:53 2005 +0000 25226 25227 Bug #3626: _X_EXPORT tags for video and input drivers. 25228 25229commit 9a4895c4c2fd9614d83adfe3967f4f46dc4dfff6 25230Author: Eric Anholt <anholt@freebsd.org> 25231Date: Wed Jun 22 22:50:49 2005 +0000 25232 25233 Add missing detection of the PCI ID for 0x5b62, Radeon X600 (Vladimir 25234 Kushnir). 25235 25236commit 43e3cc7c21160f4536ed860f97d02545c267dfbc 25237Author: Alex Deucher <agd5f@yahoo.com> 25238Date: Wed Jun 15 23:40:50 2005 +0000 25239 25240 Sync up radeon mergedfb with Thomas' latest sis mergedfb changes: 25241 - non-rectangular mergedfb desktops 25242 - screen offsets 25243 25244commit b67f9f5c875efaa41bded7ed2f38b2a5fbd17597 25245Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 25246Date: Sat Jun 4 20:26:28 2005 +0000 25247 25248 Sync with 6.8.2 branch: Bug #591 25249 (https://bugs.freedesktop.org/show_bug.cgi?id=591) attachment #308 25250 (https://bugs.freedesktop.org/attachment.cgi?id=308): Fix video timing 25251 problems with Sony PCG-C1VN Vaio Picturebook notebook && co. Patch by 25252 Marc Aurele La France 25253 25254commit daae70d9083e9c15358f160f0df3176e344f1cb9 25255Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 25256Date: Sat Jun 4 20:04:36 2005 +0000 25257 25258 Sync with 6.8.2 branch: Bugzilla #1306 25259 (https://bugs.freedesktop.org/show_bug.cgi?id=1306) attachment #1291 25260 (https://bugs.freedesktop.org/attachment.cgi?id=1291): Fix SEGV in 25261 "radeon" driver. Patch by Kevin E. Martin <kem@freedesktop.org> 25262 25263commit c27c193f192b117232dbe39575d3d5ab7c6a1b1e 25264Author: Felix Kuehling <fxkuehl@gmx.de> 25265Date: Fri May 27 22:33:23 2005 +0000 25266 25267 Fix for bugzilla https://bugs.freedesktop.org/show_bug.cgi?id=3414 (patch 25268 https://bugs.freedesktop.org/attachment.cgi?id=2782): Don't use RMX if 25269 we are Dell Server. 25270 25271commit 3b674815e0a6d19b0719e71261c0909b92173aed 25272Author: Adam Jackson <ajax@nwnk.net> 25273Date: Tue May 24 23:26:50 2005 +0000 25274 25275 Typo fix from bug #3163 commit 25276 25277commit 0845d1db801ae74688464bd4c41a7e40e4b50186 25278Author: Egbert Eich <eich@suse.de> 25279Date: Mon May 23 09:50:20 2005 +0000 25280 25281 Explicitely disable DPMS power states during close screen. Fix macro that 25282 takes a value and a mask of bits to modify to be behave evquvalently if 25283 the bits that should remain unchanged are set or unset in the value 25284 (Bugzilla #3369). 25285 25286commit 74b86b036def72e5180a46871b8452ba9eea9bb1 25287Author: Adam Jackson <ajax@nwnk.net> 25288Date: Wed May 18 17:47:35 2005 +0000 25289 25290 Bug #3163: Create new DRIINFO_*_VERSION macros to indicate the version 25291 number of the DRIInfoRec, disambiguating it from the XF86DRI protocol 25292 version number. Modify DRIQueryVersion to return the libdri version 25293 number, which all DDXes were requesting implicitly. Fix the DDXes to 25294 check for the DRIINFO version they were compiled against. 25295 25296commit 0f4a8fc6aa5157a8b48beb49f30511665cfa4b63 25297Author: Matthieu Herrb <matthieu.herrb@laas.fr> 25298Date: Wed Apr 20 21:07:44 2005 +0000 25299 25300 programs/Xserver/hw/xfree86/drivers/ati/theatre.h Build fix for 25301 DoLoadableServer = NO systems. 25302 25303commit 3355d12f832909b05e8b022fad784c9f7af7341f 25304Author: Daniel Stone <daniel@fooishbar.org> 25305Date: Wed Apr 20 12:25:22 2005 +0000 25306 25307 Fix includes right throughout the Xserver tree: 25308 change "foo.h" to <X11/foo.h> for core headers, e.g. X.h, Xpoll.h; 25309 change "foo.h", "extensions/foo.h" and "X11/foo.h" to 25310 <X11/extensions/foo.h> for extension headers, e.g. Xv.h; 25311 change "foo.[ch]" to <X11/Xtrans/foo.[ch]> for Xtrans files. 25312 25313commit 31cb4758eae4703d4de31a0b291847cefa855fc3 25314Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 25315Date: Sun Apr 17 23:28:09 2005 +0000 25316 25317 Added fifo_read and fifo_write in tge generic bus definition 25318 25319commit 2b363e02c7c122f07e6fc406fb8ee52d97b55cad 25320Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 25321Date: Sun Apr 17 23:26:36 2005 +0000 25322 25323 Added new options for the ATI driver: RageTheatreMicrocPath and 25324 RageTheatreMicrocType 25325 25326commit 6286efeba564c63ec198c091e4bba22027586901 25327Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 25328Date: Sun Apr 17 23:23:45 2005 +0000 25329 25330 Addeded support for FI1216ME (PAL) version of the tuner, detection of 25331 UDA1380 sound processor 25332 25333commit 9a63c216e90622cd96d4555155df06dd66c3235b 25334Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 25335Date: Sun Apr 17 23:18:26 2005 +0000 25336 25337 Use the theatre_detect to determine the chip ID and then load the 25338 coresponding theatre module 25339 25340commit 247057762813f12f171b1fa05f31c5ace4c245a7 25341Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 25342Date: Sun Apr 17 23:15:31 2005 +0000 25343 25344 Added functions that access the FIFO on RT200 25345 25346commit bcccf9259784f21d84d52c526aa68df0787118b4 25347Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 25348Date: Sun Apr 17 23:13:45 2005 +0000 25349 25350 Added RT200 registers 25351 25352commit da9d27f8c8cccf2f40a244b25ffe0d3162c41a63 25353Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 25354Date: Sun Apr 17 23:09:28 2005 +0000 25355 25356 This is the theatre chip detection module 25357 25358commit e4623bbce161c41ccac65210fd2539211e8af715 25359Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 25360Date: Sun Apr 17 23:07:54 2005 +0000 25361 25362 Now the detection of the theatre chip is in other module called 25363 theatre_detect 25364 25365commit 405b4dcb067346fce7fd9f6d8a6507a58c740ac7 25366Author: Bogdan Diaconescu <b_diaconescu@yahoo.com> 25367Date: Sun Apr 17 23:06:17 2005 +0000 25368 25369 Added the RageTheatre200 video demodulator support 25370 25371commit 310eb24a45b145efbd47196a1c3ec750cf004154 25372Author: Daniel Stone <daniel@fooishbar.org> 25373Date: Mon Apr 4 23:07:08 2005 +0000 25374 25375 Release SDA/SDL I2C lines from an asserted-low state after a DDC probe, 25376 which caused Apple Studio Display monitors to shut off after a second 25377 or so (Ben Herrenschmidt). 25378 25379commit a416667f980a8679352be897af2fb3215da2d000 25380Author: Alex Deucher <agd5f@yahoo.com> 25381Date: Thu Mar 31 23:18:10 2005 +0000 25382 25383 - Add new Radeon pci ids (ATI devrel), bug 2827 25384 25385commit 2e7365ffb767099faf3f191dc65341dbb3ef6247 25386Author: Vladimir Dergachev <volodya@mindspring.com> 25387Date: Thu Mar 31 18:26:05 2005 +0000 25388 25389 Modified: 25390 programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 25391 Patch from Antti Ajanki - provide VBI capture support for PAL and SECAM 25392 (You need GATOS km module to actually get at the data, the change 25393 merely configures the hardware to handle it correctly). 25394 25395commit 9a3c1c8e4cba2885b707db0f855e11dc6e640b49 25396Author: Michel Daenzer <michel@daenzer.net> 25397Date: Tue Mar 29 06:14:34 2005 +0000 25398 25399 bugzilla #2844 (https://bugs.freedesktop.org/show_bug.cgi?id=2844) 25400 attachment #2230 (http://bugs.freedesktop.org/attachment.cgi?id=2230) 25401 Wait for idle before uploading HW cursor data to the framebuffer to 25402 avoid lockups when moving the mouse on some machines. 25403 25404commit 97863f112850be5206ea555297354c1c104fcdbe 25405Author: Kevin E Martin <kem@kem.org> 25406Date: Tue Mar 29 03:49:04 2005 +0000 25407 25408 bugzilla #2840 (https://bugs.freedesktop.org/show_bug.cgi?id=2840) 25409 attachment #2227 (https://bugs.freedesktop.org/attachment.cgi?id=2227) 25410 Fix build problems introduced by previous Radeon PLL errata 25411 workarounds. 25412 25413commit 04cbb8d174cc5e3c7ecdd4e171170213ecb3c7ed 25414Author: Dave Airlie <airlied@linux.ie> 25415Date: Sat Mar 26 00:53:01 2005 +0000 25416 25417 bugzilla #2057 (https://bugs.freedesktop.org/show_bug.cgi?id=2057) 25418 attachment #1516 (https://bugs.freedesktop.org/attachment.cgi?id=1516) 25419 Add TVOUT Support for Mach64 (Leif Delgass, fixed up for Xorg by me). 25420 25421commit 5be4bf9000bdf58584a10a6b8e285d0f173304fa 25422Author: Alex Deucher <agd5f@yahoo.com> 25423Date: Thu Mar 24 06:45:52 2005 +0000 25424 25425 - radeon PLL access errata workarounds (Benjamin Herrenschmidt, acked by 25426 Hui Yu) 25427 25428commit 34af3da80304a13b6407dcc0f4db54d6c8bb7a41 25429Author: Michel Daenzer <michel@daenzer.net> 25430Date: Fri Mar 18 22:36:47 2005 +0000 25431 25432 Add support for production version of ATI RN50/ES1000. (ATI Technologies 25433 Inc.) 25434 25435commit b56a5f96048b920a7751318dde211e881f53b2bb 25436Author: Vladimir Dergachev <volodya@mindspring.com> 25437Date: Thu Mar 17 03:36:25 2005 +0000 25438 25439 Modified: 25440 programs/Xserver/hw/xfree86/drivers/ati/radeon_mergedfb.c Back out 25441 RadeonWaitForIdleMMIO that I put in earlier. 25442 From discussion on dri-devel we do not need to be as careful with INREG() 25443 access on Radeons as we are on mach64 chips. 25444 The fix is backed out due to: 25445 it breaks DRI on SMP systems (due to absense of the lock in mouse handling 25446 functions) 25447 I am no longer able to reproduce the lockup I saw before 25448 25449commit 01a9c1157cdfbc5f74b035383e060b8419979563 25450Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 25451Date: Tue Mar 8 23:41:37 2005 +0000 25452 25453 Fix accidentally removed depth pitch adjustment if displayWidth is not a 25454 multiple of 32 pixels (bug #788) 25455 25456commit 9b994b40710103b263364153429ebdb9ea3f58ed 25457Author: Vladimir Dergachev <volodya@mindspring.com> 25458Date: Fri Mar 4 22:21:40 2005 +0000 25459 25460 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c Let through all chips 25461 R300 and above - presumably the filtering should take place at the DRM 25462 driver level. Correct cut'n'paste position misprint. Big thanks for 25463 Michel Danzer for pointing out both improvements. 25464 25465commit 7d37a9f1f9fffe98067c3b5bc4749cc799fb0c77 25466Author: Vladimir Dergachev <volodya@mindspring.com> 25467Date: Fri Mar 4 05:48:31 2005 +0000 25468 25469 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c Turns out Radeon R420 25470 chips work fine with R300 microcode. Thanks go to Rune Petersen for 25471 careful investigation. 25472 25473commit f1edbb7852d85f660188498f578a0a697c2094fe 25474Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 25475Date: Tue Feb 22 00:39:15 2005 +0000 25476 25477 Fix building radeon driver with BuildXF86DRI set to NO, the breakage was 25478 caused by the tiling support (bug #2481) 25479 25480commit b853dbb50fb7c46fb5a70f349a3f264f0bb94e8c 25481Author: Vladimir Dergachev <volodya@mindspring.com> 25482Date: Sat Feb 19 16:00:12 2005 +0000 25483 25484 Modified: 25485 programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 25486 Fix an INREG without WaitForIdleMMIO. This one should be unlikely to 25487 trigger, but we should still do this in the name of correctness. 25488 25489commit 84290849b4aee336e9db0779d93fc2d7ed8881b1 25490Author: Vladimir Dergachev <volodya@mindspring.com> 25491Date: Sat Feb 19 00:57:23 2005 +0000 25492 25493 Modified: 25494 programs/Xserver/hw/xfree86/drivers/ati/radeon_mergedfb.c 25495 Add a call to RADEONWaitForIdleMMIO() within RADEONChooseCursorCRTC() as 25496 this function uses OUTREGP() which, internally, calls INREG(). 25497 Thus (due to SilkenMouse) INREG() could be called while CP engine was busy. 25498 I have observed this lockup on RV350 + R300 3d driver and have no idea 25499 why no issues were reported for R200 cards. Perhaps, few of them have 25500 enough memory to run merged fb and 3d simultaneously ? 25501 This fix is the right thing to do, however, we might also need to put fifo 25502 management calls into other cursor handling functions (in particular 25503 those that write cursor image directly to video memory). 25504 25505commit 60070063e2130af0b9d5d34eed7adbae0c0fa67b 25506Author: Michel Daenzer <michel@daenzer.net> 25507Date: Fri Feb 18 19:55:35 2005 +0000 25508 25509 Bug #2576: Add support for ATI RN50/ES1000. (ATI Technologies Inc.) 25510 25511commit 32032cdc78097840b7e2685fdec11bc0816aa0ac 25512Author: Adam Jackson <ajax@nwnk.net> 25513Date: Fri Feb 18 16:42:11 2005 +0000 25514 25515 Bug #2567: Fix reversed memset arguments. (David Krause) 25516 25517commit d5fadd6d86ae2a2f8c1bb55e36ab3d6d49649b46 25518Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 25519Date: Wed Feb 9 02:22:39 2005 +0000 25520 25521 do not enable color tiling on r300 chips and newer, since it doesn't work 25522 (yet) (Bugzilla #2504) 25523 25524commit 5ff35300e66a81ecc17c4e9396eb6d158f5c8e1b 25525Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 25526Date: Wed Feb 2 21:35:30 2005 +0000 25527 25528 fix error with color tiling and hw scrolling, it looks like rv100 doesn't 25529 need the special treatment, after all (reported by Stephane Marchesin) 25530 25531commit 7a9031683914bb1d17d223634e5dbf2e18955651 25532Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 25533Date: Mon Jan 31 20:10:28 2005 +0000 25534 25535 change option EnableColorTiling to ColorTiling 25536 25537commit a2b123f7672ea678290798786ecb408a86d8c683 25538Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 25539Date: Wed Jan 26 18:23:41 2005 +0000 25540 25541 (Stephane Marchesin, me) Add support for color (framebuffer) tiling. 25542 Convert all XAA accel code to use explicit dst and src offset pitch 25543 control to make it work. Use the new drm surface ioctls to set up the 25544 tiled surfaces (if dri is enabled). 25545 25546commit e0df1e41724cd009fe7c833d2f9555f5d87e3202 25547Author: Roland Scheidegger <rscheidegger_lists@hispeed.ch> 25548Date: Wed Jan 26 17:39:56 2005 +0000 25549 25550 fix issues with mergedfb + pageflip (fix up sarea frame values) 25551 25552commit aeb01807263c47f49abaf00031d8505e8ebda82c 25553Author: Egbert Eich <eich@suse.de> 25554Date: Mon Jan 24 15:40:21 2005 +0000 25555 25556 Don't exclude 'RADEONInfoPtr info = ...' declaration even when not building 25557 with DRI support (Bugzilla #2370) 25558 25559commit c6c30cfa354eb456638f10771a7cb02271696af2 25560Author: Vladimir Dergachev <volodya@mindspring.com> 25561Date: Thu Jan 13 17:35:49 2005 +0000 25562 25563 Modified: 25564 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 25565 Some of the newer Radeons (R420 for example) have MM_TABLE's with size 0x33 25566 (instead of 0x0c that we know). Furthermore, they lockup during access 25567 of I2C bus ! 25568 Add a check to validate MM_TABLE. Note: the check is not more stringent as 25569 other card manufacturers (like Asus or Tekram) might make cards with 25570 slightly different MM_TABLE. 25571 25572commit 414c035a965b9db15656058c0e9cfeb3acf0158f 25573Author: Alex Deucher <agd5f@yahoo.com> 25574Date: Fri Dec 24 18:26:19 2004 +0000 25575 25576 - Fix MCLK / SCLK calculation (Benjamin Herrenschmidt) bug 2144 25577 - formatting cleanup 25578 25579commit 7c7b278883676412f5709cf812587b1408628133 25580Author: Alex Deucher <agd5f@yahoo.com> 25581Date: Sun Dec 19 19:49:33 2004 +0000 25582 25583 - Add support for Dualhead on M3/M4 (bug 1760) 25584 25585commit 08a45118b53129f3b9c30004404b906c7d7690be 25586Author: Alex Deucher <agd5f@yahoo.com> 25587Date: Sun Dec 19 19:29:10 2004 +0000 25588 25589 - Make dualhead ram allocation cleaner and related messages clearer 25590 25591commit 97cfd7b1489a9510cd233d6dc7154c210f32b9f5 25592Author: Alex Deucher <agd5f@yahoo.com> 25593Date: Sun Dec 19 06:56:37 2004 +0000 25594 25595 - Add DRI resume support (Micha Feigin) bug 2063 25596 25597commit 5b8577ae3200d967b4316156f16028cbcfd21983 25598Author: Alex Deucher <agd5f@yahoo.com> 25599Date: Sun Dec 19 06:11:06 2004 +0000 25600 25601 - fix warnings (Ronny V. Vindenes) 25602 25603commit 447124a663a24a7fffb0cd241d18dd896a43098b 25604Author: Alex Deucher <agd5f@yahoo.com> 25605Date: Sat Dec 18 21:37:48 2004 +0000 25606 25607 - fix warning and potential error in radeon_driver.c 25608 25609commit ac6166158a58b377e7383950c169956f78b62093 25610Author: Alex Deucher <agd5f@yahoo.com> 25611Date: Fri Dec 17 19:19:32 2004 +0000 25612 25613 - Add VGAAccess option to r128 driver. Based on Benh's radeon patch. r128 25614 should now work on ppcs without usefbdev. To use this on ppc make sure 25615 you set "usefbdev" to false. (bug 2089) 25616 - Apply Benjamin Herrenschmidt's radeon patches (bug 2064) 25617 25618commit ac860ed6bbed84f6fc39a9acf53ee87646cfc8dc 25619Author: Vladimir Dergachev <volodya@mindspring.com> 25620Date: Fri Dec 17 16:50:36 2004 +0000 25621 25622 Modified: 25623 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 25624 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c 25625 Move DMA robustness fix into radeon_dri.c::RADEONEnterServer() as per 25626 suggestion by Michel D�nzer. 25627 I could not trigger a lockup, even with r300_demo (possibly it has code 25628 that flushes cache inside ?), so this must be good enough.. 25629 25630commit 3ba5b525c4d6390184fdf79b28c954d5caadc6a4 25631Author: Vladimir Dergachev <volodya@mindspring.com> 25632Date: Thu Dec 16 16:26:56 2004 +0000 25633 25634 Modified: 25635 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 25636 Add DMA robustness fix from Nicolai Haehnle. 25637 TODO: move this fix into DRM driver or other parts of Xserver when R300 25638 support is more stable. 25639 This fix affects only R300 and later variants and results in small slowdown 25640 when using CP acceleration in comparison to the previous version of the 25641 driver. 25642 We are still a lot faster in CP mode than using MMIO register access. 25643 25644commit c7b5f4a9eb49c0edd690c972be1015c98408ffaf 25645Author: Adam Jackson <ajax@nwnk.net> 25646Date: Wed Dec 15 19:35:33 2004 +0000 25647 25648 Bug #2008: Fix gcc4 compile problem. 25649 25650commit 8c1969f22fba2e1c9ed34d5ca77bee3a9f577640 25651Author: Vladimir Dergachev <volodya@mindspring.com> 25652Date: Sun Dec 12 20:42:29 2004 +0000 25653 25654 Modified: 25655 programs/Xserver/hw/xfree86/drivers/ati/radeon_render.c 25656 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 25657 Make DMA blitting code more robust. 25658 25659commit 8a9077bc70fd800983f138a224b95301992a2296 25660Author: Vladimir Dergachev <volodya@mindspring.com> 25661Date: Sun Dec 12 19:28:37 2004 +0000 25662 25663 Modified: 25664 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 25665 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c 25666 Enable DRM support permanently, allowing its use for any CHIP_FAMILY > 25667 R300. Fix bug in naming Mesa3d driver (it was defaulting to r200) 25668 25669commit 845df6dd96785b59c44c670009c4fcdb20e147ab 25670Author: Alex Deucher <agd5f@yahoo.com> 25671Date: Sun Dec 12 17:29:45 2004 +0000 25672 25673 - missing diff from agp 8x support 25674 25675commit 36b92808e4ff1311f91543388c18cbe8e9eb4aab 25676Author: Vladimir Dergachev <volodya@mindspring.com> 25677Date: Sun Dec 12 16:05:35 2004 +0000 25678 25679 Modified: 25680 programs/Xserver/hw/xfree86/drivers/ati/radeon_accel.c 25681 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 25682 programs/Xserver/hw/xfree86/drivers/ati/radeon_dri.c 25683 programs/Xserver/hw/xfree86/drivers/ati/radeon_common.h 25684 programs/Xserver/hw/xfree86/drivers/ati/radeon_version.h Add R300 (and 25685 later) DRM support, protected by X_R300_DRM option. 25686 Note: at the moment this is for 2d support only, if enabled running 25687 glxgears locks up my machine, but DMAForXv=yes works fine. 25688 25689commit 68ea67ce98b8f3ad4c17f987e0e2b92f79a96fbf 25690Author: Alex Deucher <agd5f@yahoo.com> 25691Date: Sun Dec 12 02:00:48 2004 +0000 25692 25693 - add AGP 8x support (Hui Yu, Michel Daenzer) bug 1819 25694 25695commit 93675d288d5ebb35d4724ebbfc455a703e25a311 25696Author: Alex Deucher <agd5f@yahoo.com> 25697Date: Sun Dec 12 00:14:32 2004 +0000 25698 25699 - only mess with Dynamicclocks on mobility chips (Benh, me) bug 1912 25700 - fix display detection (Hui Yu) bug 1559 25701 - Add new option "BIOSHotkeys" to allow the bios to control output 25702 switching (off by default) bug 2000 25703 - update the savage man page to document the "CrtOnly" option 25704 25705commit ed217003127028483fd2956851d438977169d154 25706Author: Daniel Stone <daniel@fooishbar.org> 25707Date: Mon Dec 6 13:27:56 2004 +0000 25708 25709 Apply Debian patch to re-POST via VBE if driver thinks it has no video 25710 memory (a symptom of incompletely coming up from D3 power state), which 25711 fixes some Dell laptops (closes: #1890). 25712 25713commit 2b8ab42b05b32710fa4f71a25e2f84192e637fbd 25714Author: Michel Daenzer <michel@daenzer.net> 25715Date: Mon Dec 6 06:27:53 2004 +0000 25716 25717 When direct rendering is enabled, use hostdata blits to transfer data from 25718 system memory to video RAM, which should reduce CPU usage especially 25719 with larger videos. Can be disabled via Option "DMAForXv" if there 25720 should be any stability issues, but it's been stable for me during a 25721 week of testing. Based on a patch by Nikolaus Meine 25722 <meine@tnt.uni-hannover.de>. Probably fix endianness issues in some 25723 newer XVideo code, untested. 25724 Also use hostdata blits to transfer RENDER image data to video RAM to avoid 25725 idling the accelerator engine. Increases RENDER performance 25726 significantly for me. 25727 These changes were only tested on an M9 in a Titanium PowerBook but should 25728 work with all Radeons where direct rendering is supported. 25729 25730commit 960d15ae3a44efe7c02a3d6d0acbabe63e07be74 25731Author: Daniel Stone <daniel@fooishbar.org> 25732Date: Sat Dec 4 22:34:54 2004 +0000 25733 25734 Disable interrupt handling after the DRM handler has been removed, 25735 preventing lockups (closes: #1886). 25736 Allow pitches with a granularity of 32, rather than 64, because some 25737 clamshell iBooks have 800-byte pitches. Change Xv allocation 25738 granularity to 8 bytes, rather than 16 bytes, also (closes: #1888). 25739 25740commit 4706dfb7804a46e3cc5aafc3644f5a1134ab3d19 25741Author: Daniel Stone <daniel@fooishbar.org> 25742Date: Sat Dec 4 22:24:06 2004 +0000 25743 25744 Retry CCE idle commands until the (higher) timeout is reached, and properly 25745 stop the engine if we totally fail -- gets rid of serious performance 25746 issues and lockups with r128 DRI. 25747 25748commit bce508a288d6b95708dbbfc6b235e2d95b6b96ad 25749Author: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk> 25750Date: Sat Dec 4 00:42:59 2004 +0000 25751 25752 Encoding of numerous files changed to UTF-8 25753 25754commit e9c9fd279ed43cc0d779266efcea69a67541c794 25755Author: Egbert Eich <eich@suse.de> 25756Date: Tue Nov 30 08:38:44 2004 +0000 25757 25758 Make Xorg the default server to install. 25759 Avoid PIO access on IA64. Some IA64 machine check if legacy ports outside 25760 the VGA range are accessed. The ATi driver however does this to probe 25761 for ISA Mach8/32/64. Since no IA64 has ISA slots this restriction 25762 should not be relevant to the user. 25763 Avoid recursive calls of xf86scanpci(). This function normally detects that 25764 it has been called before by checking if the PCI structure is filled 25765 out. So far if this was not the case (because PCI probing has failed 25766 for some reason) the function is traversed again. With the chipset 25767 specific PCI bus probing this can lead to an endless recursive loop as 25768 the post-probing code calls xf86scanpci() from within this function. 25769 The OS specific PCI code for Linux worked only if bus 0 was populated as it 25770 checked for the presence of /proc/bus/pci/00. Fixed to check for 25771 /proc/bus/pci/<bus_to_look_for> instead. 25772 25773commit fffbcb761c178e025e3d55d428a803b1284481e0 25774Author: Matthieu Herrb <matthieu.herrb@laas.fr> 25775Date: Sat Nov 27 15:14:15 2004 +0000 25776 25777 xc/programs/Xserver/hw/xfree86/driver/ati/radeon.man 25778 xc/programs/Xserver/hw/xfree86/driver/voodoo/voodo.man 25779 xc/programs/Xserver/hw/xfree86/driver/wsfb/wsfb.man XF86Config -> 25780 __xconfigfile__ 25781 25782commit ad6347d187f9afa053c755aa3a2da9d821916cb2 25783Author: Daniel Stone <daniel@fooishbar.org> 25784Date: Tue Nov 23 21:27:43 2004 +0000 25785 25786 Use fp_horz_stretch instead of fp_vert_stretch for horizontal stretch 25787 factor calculation in panel size derivation in the Radeon driver 25788 (closes: #1881). 25789 25790commit ac054e427c6ce8cd5b89459dd3f16ca20e6b2d6a 25791Author: Vladimir Dergachev <volodya@mindspring.com> 25792Date: Sun Oct 24 18:17:36 2004 +0000 25793 25794 Add registers to facilitate work with stereo 3d. 25795 25796commit 445ed4b9b61347c866ee67dc0cb110904d667b84 25797Author: Alex Deucher <agd5f@yahoo.com> 25798Date: Fri Oct 22 03:51:58 2004 +0000 25799 25800 - make the overlay work on crtc2 again 25801 - make PutVideo() MergedFB aware 25802 25803commit 54cffc3fe273cf3c9ed2a8aef833d4bd337a8a09 25804Author: Egbert Eich <eich@suse.de> 25805Date: Mon Oct 18 15:19:44 2004 +0000 25806 25807 Fixing one of the infamous 'control reaches end of non void function 25808 problems'. 25809 25810commit 7a97d363327e0b2c8646d961e00d05f99766b54f 25811Author: Egbert Eich <eich@suse.de> 25812Date: Mon Oct 18 14:21:45 2004 +0000 25813 25814 Made handling of DevelDrivers for x86-64 more conformant to other 25815 platforms. 25816 Compress all font encodings (Stefan Dirsch). 25817 Fixed warnings. 25818 Turn on forwarding XNSpotLocation event to XIM server in OffTheSpot and 25819 Root mode (bugzilla #1580, James Su). 25820 Added another compose key combination for the Euro symbol (Stefan Dirsch). 25821 Added support for UTF-8 in ja_JP, ko_KR and zh_TW locales (Mike Fabian). 25822 Changed default encoding for ru from KOI8-R to ISO8859-5 (Mike Fabian). 25823 This is the encoding that is also used by glibc. We may break other 25824 libcs - lets see who complains. 25825 Added explanation for DESTDIR to install to a different directory than /. 25826 Added some early bailouts to atiprobe if PCI structure pointer is NULL to 25827 prevent sig11. 25828 XV support doesn't depend on 2D acceleration any more. This patch removes 25829 this limitation from the NSC driver. This is a patch that I have 25830 committed to XFree86 a while ago but never ported over to X.Org. 25831 Matthieu Herrb contributed some build fixes. 25832 Fixing SetDPMSTimers() so that DPMS*Time == 0 disables the appropriate 25833 timer. This takes advantage of the fact that TimerSet() with a timeout 25834 argument 0 behaves like TimerCanel(). 25835 Use /dev/xconsole (named pipe) or devpts for system logger (Werner Fink). 25836 Create missing links for backward compatibility to XFree86 (Stefan Dirsch). 25837 Changed comment to mention xorg. 25838 Changed cursor for the 'kill' action from XC_draped_box to XC_Pirate. If 25839 you don't like it we can change it back (original author unkown). 25840 Added 'pic' to the man page rendering command pipeline (Werner Fink). 25841 Added missing return value (Stefan Dirsch, Roland Mainz) 25842 25843commit c7ab340fecb94a89c25d6bc7f686b7c97cd3067f 25844Author: Alex Deucher <agd5f@yahoo.com> 25845Date: Sun Oct 17 22:45:57 2004 +0000 25846 25847 - Fix DFP detection in r128 driver (bug 1386, Klaus Kusche) 25848 - Fix EmulateWheelInertia defaults (bug 1115, Andrew Pimlott) 25849 25850commit ace6e1b7933cb415b8efc41dffa335938f9c89b5 25851Author: Vladimir Dergachev <volodya@mindspring.com> 25852Date: Sun Oct 17 17:57:33 2004 +0000 25853 25854 Modified: 25855 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c On my Inspiron 25856 8600 both internal and external ports are marked DAC_PRIMARY in BIOS. 25857 So be extra careful - only swap when the first port is not DAC_PRIMARY 25858 25859commit 22405e81b812c239b5d8653b402b5f846e6e10b1 25860Author: Alex Deucher <agd5f@yahoo.com> 25861Date: Sun Oct 17 16:52:02 2004 +0000 25862 25863 - Add Thomas' latest sis MergedFB enhancements to the radeon driver 25864 25865commit 186efa31b2cb7285717678bb1c4df4cd9e182395 25866Author: Vladimir Dergachev <volodya@mindspring.com> 25867Date: Wed Oct 13 23:02:42 2004 +0000 25868 25869 Modified: 25870 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 25871 xc/programs/Xserver/hw/xfree86/drivers/i2c/fi1236.h Squash annoying warning 25872 about fi1236_dump_status 25873 25874commit 43b24861e36023613a99627c57ac44721ed1a548 25875Author: Egbert Eich <eich@suse.de> 25876Date: Tue Oct 12 19:13:43 2004 +0000 25877 25878 Set fbdev mode as the default mode on PPC (Olaf Hering). 25879 Added support for IBM space saver keyboard (Stefan Dirsch). Added support 25880 for Cherry CyMotion Master XPress (Marcus Schaefer). 25881 Change order of SetDisplayDevice(), HWRestore(), UnbindGART() and 25882 RestoreBIOSMemSize() to be exactly opposite to the Save procedure in 25883 EnterVT() (Matthias Hopf, Alan Hourihane). 25884 Fix text mode restauration by removing the assumption that the register 25885 which determines which head is programmed is set. to the active head by 25886 the BIOS (Mark Vojkovich). 25887 When I wrote the resource code 5 years ago I made some assumptions which 25888 turned out to be false: I've assumed that the bus number of the PCI 25889 hostbridge would be the PCI bus the bridge links to. This is not 25890 correct. Fixing this assumption is not easy. However I hope that the 25891 attached patch will make the system work regardless as it 'ignores' 25892 host bridges from which the target bus is not known. This should not 25893 matter at all as we really don't care about host bridges (unless we 25894 have bridge specific code which retrieves information about the 25895 bridge). 25896 Fixed server crash on reset when a structure allocated in PreInit() was 25897 freed on CloseScreen(). 25898 Fixed ring buffer lock ups that happened because the structure that 25899 contained ringbuffer data was not zeroed after allocation. 25900 Fixed numerous warnings due to signed unsigned comparisons. 25901 programs/Xserver/hw/xfree86/drivers/nv/nv_driver.c: 25902 (NVBacklightEnable): Changed the order in which the sequencer registers 25903 and the backlight control registers are written. The sequencer control 25904 register need to be written first otherwise DPMS screen blanking 25905 produces vertical bars on a mobile device. lib/Xau/Imakefile: Build 25906 libXau static library PIC so it can be linked into toolkits that would 25907 like to wrap its functionality. 25908 25909commit 515b7c4b748f70c17aee149a33ee1b41632daa1a 25910Author: Alex Deucher <agd5f@yahoo.com> 25911Date: Mon Oct 11 23:20:26 2004 +0000 25912 25913 - fix resume on radeons (bug 1220) 25914 - clean up some old cruft in the savage i2c code 25915 25916commit 07e1aaee862c4835601c704e14fcb12e28f4e78e 25917Author: Egbert Eich <eich@suse.de> 25918Date: Mon Oct 11 09:58:04 2004 +0000 25919 25920 Improving DPMS handling on VT swich and server termination/abort: previous 25921 version called the driver directly and too late. 25922 Unblank secondary screen explicitely. Don't rely on the value read during 25923 register save as the BIOS have blanked the secondary head. 25924 Checking if server isn't switched away before calling sync. Sanity check 25925 for possible bugs in aother areas of the code. 25926 Fixing default amount of of allocated video memory from AGP for i810: Use 25927 16MB if less than 192MB are installed else use 24MB (Matthias Hopf). 25928 25929commit a8c77480e7f4ddf8c45bea3ecec7b6564c0d6452 25930Author: Vladimir Dergachev <volodya@mindspring.com> 25931Date: Fri Oct 8 00:56:05 2004 +0000 25932 25933 Modified: 25934 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c Remove redundant 25935 and incorrect line. 25936 25937commit ec831d4fba23dc5f4d0a1450aa63efe5ff5937a2 25938Author: Vladimir Dergachev <volodya@mindspring.com> 25939Date: Tue Oct 5 01:16:01 2004 +0000 25940 25941 Modified: 25942 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre.h Patch from Matthieu 25943 Herrb - fix compilation with DoLoadableServer=NO 25944 25945commit f1dadde8f642b332a2dcd2e2a5779f25fec7a3de 25946Author: Adam Jackson <ajax@nwnk.net> 25947Date: Mon Oct 4 21:52:47 2004 +0000 25948 25949 Correct radeon man page, IGP is supported for 3D but R400 is not. 25950 25951commit e3dba9841521466629828501c4fdccbba3d6af89 25952Author: Vladimir Dergachev <volodya@mindspring.com> 25953Date: Mon Oct 4 15:24:34 2004 +0000 25954 25955 Modified: 25956 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c Do not probe for 25957 Rage Theatre if no multimedia table is present and no parameters have 25958 been specified. 25959 25960commit 4aadb784928ebbd60dce172c9a9a80e8d84e943f 25961Author: Vladimir Dergachev <volodya@mindspring.com> 25962Date: Mon Oct 4 14:53:11 2004 +0000 25963 25964 Modified: 25965 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c This is really 25966 embarassing - I found a part of code that had <cr> at the end of the 25967 lines. I am certain I never saw it before - talk about code rot ! 25968 25969commit db5a27081d9d3ab4cc1481a7d110d11103b3de19 25970Author: Vladimir Dergachev <volodya@mindspring.com> 25971Date: Sun Oct 3 23:54:40 2004 +0000 25972 25973 Modified: 25974 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_mm_i2c.c Remove debugging 25975 statement that cause loading of DLL modules to fail. 25976 25977commit 37a8d8a53f12f829783011c1e077fefb5729e618 25978Author: Vladimir Dergachev <volodya@mindspring.com> 25979Date: Sun Oct 3 14:41:45 2004 +0000 25980 25981 Modified: 25982 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c Fix compilation 25983 with gcc 3.4.x (patch by Ronny V. Vindenes) 25984 25985commit 6dbcd503fe21c6dec27c2b97adc0f39ba0a7ab2a 25986Author: Vladimir Dergachev <volodya@mindspring.com> 25987Date: Sun Oct 3 03:07:17 2004 +0000 25988 25989 Modified: 25990 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.[c,h] 25991 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre.c 25992 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_reg.h Port the rest of 25993 GATOS Radeon-specific code. Remove "checkpoint" noise from logs. Test 25994 with AIW Radeon 7500 25995 25996commit 7cb911291d459f08f54444d09081d3378fe73b8b 25997Author: Vladimir Dergachev <volodya@mindspring.com> 25998Date: Sun Oct 3 00:01:14 2004 +0000 25999 26000 Modified: 26001 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon.h 26002 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 26003 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.[c,h] 26004 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre.c Detect and initialize 26005 Rage Theatre (RT100) 26006 26007commit 6c1996174b9f7afc8f4e33c35e4864b31696d4d6 26008Author: Vladimir Dergachev <volodya@mindspring.com> 26009Date: Sat Oct 2 20:55:42 2004 +0000 26010 26011 Modified: 26012 xc/programs/Xserver/hw/xfree86/drivers/ati/Imakefile 26013 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_reg.h 26014 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.[c,h] 26015 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_vip.c Add VIP bus access 26016 code, hook it into initialization system. 26017 26018commit 136649a1c2065624a9cf17386054c6595aa6d85e 26019Author: Vladimir Dergachev <volodya@mindspring.com> 26020Date: Sat Oct 2 20:35:04 2004 +0000 26021 26022 Modified: 26023 xc/programs/Xserver/hw/xfree86/drivers/ati/theatre* 26024 xc/programs/Xserver/hw/xfree86/drivers/ati/generic_bus.h Add Rage Theatre 26025 (RT100) module code and hook it into the build system 26026 26027commit 7a17b1c294fa8c7af0b1cdea0618cb6060479356 26028Author: Vladimir Dergachev <volodya@mindspring.com> 26029Date: Sat Oct 2 01:59:13 2004 +0000 26030 26031 Modified: 26032 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon.h 26033 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.c 26034 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_video.h 26035 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_mm_i2c.c Add code to 26036 access and initialize multimedia i2c bus. Hook it up. 26037 26038commit e0123e24b6895dde7b385528da24c4e6e9d5a4cb 26039Author: Vladimir Dergachev <volodya@mindspring.com> 26040Date: Sat Oct 2 01:36:43 2004 +0000 26041 26042 Add more multimedia registers. 26043 26044commit 576f53be4f0fa54bde7167872b7508b98e5c9454 26045Author: Vladimir Dergachev <volodya@mindspring.com> 26046Date: Sat Oct 2 00:43:08 2004 +0000 26047 26048 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 26049 xc/programs/Xserver/hw/xfree86/drivers/ati/radeon.h Add TV-in specific 26050 options. Add TV-in specific fields to radeon per-screen structure, add 26051 code to initialize them. Preserve MM_TABLE (instead of whole VBIOS as 26052 was done in GATOS drivers) 26053 26054commit 9a9ba82172428b09312fdd6e11da04b21c8529fc 26055Author: Adam Jackson <ajax@nwnk.net> 26056Date: Wed Sep 22 04:38:03 2004 +0000 26057 26058 Bug #1252, #1253, #1255, #1256: Various typo fixes from Dave Jones. 26059 26060commit 700b0dc5148d0170c9223765d208ccef447242df 26061Author: Adam Jackson <ajax@nwnk.net> 26062Date: Tue Sep 21 18:10:34 2004 +0000 26063 26064 Convert the ATI drivers to ANSI function prototypes. 26065 26066commit 17e5f3c1169066be1004306d12b5398cf01cbd35 26067Author: Eric Anholt <anholt@freebsd.org> 26068Date: Fri Sep 17 05:16:45 2004 +0000 26069 26070 Bugzilla #1170: Fix an overcounting of DMA in BEGIN_RING that wasn't 26071 harmful. Bugzilla #1169: Add counting for 26072 BEGIN_RING/OUT_RING/ADVANCE_RING to avoid these sorts of problems in 26073 the future. 26074 26075commit 29bbee6b8a8dd1b883e07547691ff8e1289459ce 26076Author: Adam Jackson <ajax@nwnk.net> 26077Date: Fri Sep 17 03:04:50 2004 +0000 26078 26079 Bug #1192: Remove cfb support from drivers where its use is an option. 26080 Delete xf24_32bpp, as s3virge was the last user. Fix up some comments 26081 to refer to fb rather than cfb. 26082 26083commit cec06d4126f5f31d14ace73798c71ec75e544c45 26084Author: Alex Deucher <agd5f@yahoo.com> 26085Date: Thu Sep 16 22:15:05 2004 +0000 26086 26087 Update comment 26088 26089commit dbeb8d28ac3469f4a7d7d8dae979ae876d4cdcba 26090Author: Kevin E Martin <kem@kem.org> 26091Date: Fri Sep 3 23:26:19 2004 +0000 26092 26093 Updated release notes for bug #999. 26094 Update Radeon man page to note that RenderAccel is now enabled by default 26095 on certain chips. 26096 26097commit c829f667c8a79bbdd40e1f924d57638e7bd425dc 26098Author: Michel Daenzer <michel@daenzer.net> 26099Date: Mon Aug 30 04:31:50 2004 +0000 26100 26101 Hotfix for Eric Anholt's fix for bug #1156 : Clear the 26102 RADEON_NONSURF_AP0_SWP_16BPP bit for 4 bpp data, otherwise it breaks at 26103 depth 16. 26104 26105commit 677246740b5b538e73a7fbde661ec6a97319430b 26106Author: Adam Jackson <ajax@nwnk.net> 26107Date: Fri Aug 27 20:31:26 2004 +0000 26108 26109 Back out my last change, dupe. 26110 26111commit a8eaa15264ff0db1596d653e91260c9a39a8b5ee 26112Author: Adam Jackson <ajax@nwnk.net> 26113Date: Fri Aug 27 18:32:40 2004 +0000 26114 26115 Bug #1061 (documentation-only update): Document the RenderAccel option in 26116 the radeon manpage. 26117 26118commit 360dab062f895c19e1784146e6684f75a20c5b24 26119Author: Eric Anholt <anholt@freebsd.org> 26120Date: Fri Aug 27 03:36:28 2004 +0000 26121 26122 Bug #1156: Fix the build on PPC by splitting the byte-order-setting code 26123 into a separate routine. While here, fix the byte swapping in the 32bpp 26124 case. (Michel Daenzer) 26125 26126commit 9b19bbb4d37b79c97365aa492cd93931d492b3a8 26127Author: Eric Anholt <anholt@freebsd.org> 26128Date: Fri Aug 27 03:23:47 2004 +0000 26129 26130 Bug #1157: Fix Radeon line acceleration in the DMA case (Michel Daenzer). 26131 26132commit 838488b6b46359334e5e781589dbee3832909628 26133Author: Kevin E Martin <kem@kem.org> 26134Date: Fri Aug 27 02:38:56 2004 +0000 26135 26136 Fix Radeon compilation without Xinerama (Bug #1179, Simon Schubert). 26137 26138commit 5f54d4f45e2adc6b15e98152db88743c243aa128 26139Author: Kevin E Martin <kem@kem.org> 26140Date: Wed Aug 25 00:30:41 2004 +0000 26141 26142 Fix hangs during startup with Radeon driver (Bug #1007, Sjoerd Simons). 26143 Fix i810 driver startup on Sony VAIO laptop (Bug #1084, Dave Airlie). 26144 Fix several library build problem (Bug #1147). 26145 Fix build problems (Bugs #1148 and #1149). 26146 26147commit dde880e876f455f1fd66efbcc055e177886397c9 26148Author: Eric Anholt <anholt@freebsd.org> 26149Date: Tue Aug 17 20:22:55 2004 +0000 26150 26151 Reenable Radeon Render acceleration now that interactions with 3D clients 26152 should be resolved. 26153 26154commit 279efef503fdbd442a9bd00f0073d9f6006d9e61 26155Author: Eric Anholt <anholt@freebsd.org> 26156Date: Sun Aug 15 00:44:57 2004 +0000 26157 26158 Fix unresolved symbol warnings in r128 driver (Adam Jackson) 26159 26160commit 955f2b89f94aaa20b5eafd045b01e3eeef3d29ad 26161Author: Kevin E Martin <kem@kem.org> 26162Date: Thu Aug 12 05:00:22 2004 +0000 26163 26164 Fix context switching on Radeons and build problem when XF86DRI is not 26165 defined (Bug #922, Eric Anholt). 26166 26167commit 8f3ea53f26493a5f02410297a587d1d95c067094 26168Author: Adam Jackson <ajax@nwnk.net> 26169Date: Thu Aug 12 02:23:46 2004 +0000 26170 26171 oops, these didn't added somehow. 26172 26173commit 6ec74414bd33e9d8dfd5119058d15f3a6206e9d9 26174Author: Adam Jackson <ajax@nwnk.net> 26175Date: Thu Aug 12 01:03:06 2004 +0000 26176 26177 Factor out R{128,ADEON}Chipsets into their own headers so they can be 26178 sanely included in both the ati core and the r128/radeon submodules; 26179 r128 and radeon can now be directly loaded under dlloader. Teach imstt 26180 about the dlloader- frendly way to talk to fbdevhw. (Partial for bugs 26181 #393 and #400.) 26182 26183commit a5b1e4dd7adacd17b63582e477df3a585efb249a 26184Author: Adam Jackson <ajax@nwnk.net> 26185Date: Wed Aug 11 01:20:28 2004 +0000 26186 26187 Bug #978: Silence more unresolved symbol warnings from the ati driver. 26188 Patch from Stefan Dirsch. 26189 26190commit 1356884132491010a2e41eb2b5be4a227b9b8138 26191Author: Kevin E Martin <kem@kem.org> 26192Date: Wed Aug 11 01:14:14 2004 +0000 26193 26194 Fix Radeon build problem with #define DoLoadableServer NO 26195 26196commit f5c93cb015e334975c2797a23b68798372a284c9 26197Author: Adam Jackson <ajax@nwnk.net> 26198Date: Tue Aug 10 04:01:42 2004 +0000 26199 26200 Bug #978: Silence unresolved symbol warnings from the ati driver. Reported 26201 by Michel Daenzer. 26202 26203commit a2da1caf5848a78b10afd99811fd1a5677af6786 26204Author: Kevin E Martin <kem@kem.org> 26205Date: Mon Aug 9 22:37:22 2004 +0000 26206 26207 Fix install problem on platforms not using xorg.cf/xfree86.cf (Bug #339, 26208 Harold L. Hunt II, Alexander Gottwald). 26209 Fix crash when using X core font in zh_CN.UTF-8 locale (Bug #368, Yu Shao, 26210 David Dawes). 26211 Fix glXMakeCurrent(Dpy, None, NULL) crash (Bug #719, Adam Jackson). 26212 HP-PA build fix (Bug #828, Guy Martin, Paul Anderson). 26213 Fix SDK build for GATOS and Wacom driver (Bug #829, Bryan Stine). 26214 Fix attempt to read video ROM before enabling it (Bug #843, Ivan 26215 Kokshaysky, Mike A. Harris). 26216 Fix detection of primary adapter (Bug #843, Ivan Kokshaysky, Mike A. 26217 Harris). 26218 Clarify xset man page description of how to use the keyboard repeat rate 26219 settings (Bug #846, Mike A. Harris). 26220 Fix problem where print-screen key would get remapped to sys-req in certain 26221 keymaps, which broke GNOME printscreen functionality (Bug #847, Owen 26222 Taylor). 26223 Fix several render problems: 26224 - MMIO mode support 26225 - Hang on IGP chips 26226 - VT switching hang 26227 - 3D render corruption (Bug #922, Hui Yu). 26228 26229commit 3171664c6a02424736864204b731249097fa7aae 26230Author: Kevin E Martin <kem@kem.org> 26231Date: Fri Aug 6 20:07:35 2004 +0000 26232 26233 Disable Render acceleration in Radeon driver temporarily until it no longer 26234 conflicts with the 3D code. 26235 26236commit 27ef7adf3c1f6c3ca79b42b468d08b8542348cdd 26237Author: Michel Daenzer <michel@daenzer.net> 26238Date: Wed Aug 4 13:17:31 2004 +0000 26239 26240 Use info->ModeReg.surface_cntl to restore byte swapping for the framebuffer 26241 aperture on big endian machines, remove superfluous local variables and 26242 register reads. 26243 Adapt framebuffer aperture byte swapping to texture format before copying 26244 data to offscreen area and restore it afterwards on big endian 26245 machines, fixes Render acceleration there. 26246 reviewed by: Hui Yu <hyu@ati.com>, Kevin E. Martin <kem@freedesktop.org> 26247 26248commit 2efdc9f847fa9ff3096667b16df84817fc65605d 26249Author: Eric Anholt <anholt@freebsd.org> 26250Date: Wed Aug 4 10:05:37 2004 +0000 26251 26252 - Add two new XAA hooks, SetupForCPUToScreenTexture2 and 26253 SetupForCPUToScreenAlphaTexture2. These add a dstFormat argument after 26254 the previous format argument, which the driver needs to use to properly 26255 set up the destination format. Two new arrays are added for the list of 26256 destination formats supported that correspond to the previous format 26257 arrays for sources. 26258 - Make Render acceleration only occur when the new hook for that 26259 acceleration type is supplied and the dst format list is set, along 26260 with the src format list being set. Without knowing the destination 26261 format, the Render acceleration couldn't properly support all the 26262 destinations it might encounter. 26263 - Bump XAA module minor version. 26264 - Update the Radeon Render acceleration to use the new hooks when the XAA 26265 module is sufficiently new. Fix a bug in the src/dst alpha booleans for 26266 ops, and use them to set blend_cntl to support destinations without 26267 alpha. Add missing PICT_a1r5g5b5 texture format, and add list 26268 terminator. (!) 26269 26270commit 1e16c66551eed750330bb74a92208c45a7112cd0 26271Author: Matthieu Herrb <matthieu.herrb@laas.fr> 26272Date: Tue Aug 3 09:42:20 2004 +0000 26273 26274 programs/Xserver/hw/xfree86/drivers/ati/radeon_merged_fb.c Add missing 26275 argument to xf86DrvMsg(). 26276 26277commit fd440a1e2110e05d116207800c123c281fbf5e19 26278Author: Matthieu Herrb <matthieu.herrb@laas.fr> 26279Date: Tue Aug 3 09:38:03 2004 +0000 26280 26281 programs/Xserver/hw/xfree86/drivers/ati/r128_driver.c 26282 programs/Xserver/hw/xfree86/drivers/ati/r128_probe.h 26283 programs/Xserver/hw/xfree86/drivers/ati/radeon_driver.c 26284 programs/Xserver/hw/xfree86/drivers/ati/radeon_probe.h Fix Constification 26285 problems. 26286 26287commit dd0f86c4cb046c79a65144f767e97fd99e7f61c9 26288Author: Kevin E Martin <kem@kem.org> 26289Date: Sun Aug 1 23:22:01 2004 +0000 26290 26291 Fix Radeon crash on server exit or generation 26292 26293commit c64f90b673f5c487dfe059f1636b04a1c2ab7dc2 26294Author: Adam Jackson <ajax@nwnk.net> 26295Date: Sat Jul 31 01:21:18 2004 +0000 26296 26297 Change several LoaderSymbol calls introduced by the bug #400 patch to 26298 *Weak() resolver functions. 26299 26300commit b091b4b074cd61c9067ba95ef016f7a29a38c38d 26301Author: Hui YU <hyu@ati.com> 26302Date: Fri Jul 30 22:20:21 2004 +0000 26303 26304 Support for New radeon chips: R420/M18, R423, RV370/M22, RV380/M24, RS300. 26305 Add special handlings for DELL triple-head server (RV100). Misc. bug 26306 fixes for flat panel, host aperture, etc (Bug #946) 26307 26308commit d04f61bfaf6952e832d52e1cd6014e86435aebd9 26309Author: Adam Jackson <ajax@nwnk.net> 26310Date: Fri Jul 30 20:30:51 2004 +0000 26311 26312 Bug #400 (partial): Driver fixes for the dlloader. When using dlloader, all 26313 framebuffer formats except cfb and the overlay modes should work, and 26314 r128 and radeon need to be loaded from the ati driver (both issues to 26315 be fixed soon). Tested on i740, s3virge, mach64, tdfx, vesa, and vga 26316 drivers. elfloader users shouldn't be affected. 26317 26318commit 6dcb78bb818c3c1c7fe406bee5379f9bef023f4b 26319Author: Alex Deucher <agd5f@yahoo.com> 26320Date: Wed Jul 28 02:35:30 2004 +0000 26321 26322 - add gamma correction support to the radeon video overlay 26323 26324commit 26029dc9378bbd95835d397402bfc9e6b4f1a478 26325Author: Alex Deucher <agd5f@yahoo.com> 26326Date: Mon Jul 26 23:14:37 2004 +0000 26327 26328 - Add Radeon DynamicClocks option 26329 - Add small fixes and clean ups from ati's last code drop (typo_fixes, 26330 remove_fudge, laptop, xvfix) 26331 - fix possible segfault in mga_dri.c (Ryan Underwood) 26332 - Add Xv support to pre-nm2160 neomagic chipsets 26333 26334commit 9f251b16e702da486f824307e72a6f80c6987b5d 26335Author: Adam Jackson <ajax@nwnk.net> 26336Date: Mon Jul 26 19:14:58 2004 +0000 26337 26338 Bug #848: Clean up gatos notice messages. 26339 26340commit 6f1e01ecdb11913fd768b9918f516b8379b7858f 26341Author: Eric Anholt <anholt@freebsd.org> 26342Date: Fri Jul 23 05:36:14 2004 +0000 26343 26344 - Add a new option, BuildDevelDRIDrivers, which controls whether 26345 DevelDRIDrivers is non-empty. DevelDRIDrivers contains those drivers 26346 aren't built by default either due to being unusable on that 26347 architecture (but buildable for testing) or due to security concerns, 26348 as in the mach64 and savage cases. 26349 - Merge the Mach64 DDX's DRI support, under the XF86DRI_DEVEL define, which 26350 is set only when BuildDevelDRIDrivers is set. 26351 - Fix up of the sets of DRI drivers built for various architectures. 26352 - Fix build of unichrome DRI driver, which is now enabled on x86. 26353 26354commit b68322f467b7e524092af15e33ab872abb99b6a2 26355Author: Eric Anholt <anholt@freebsd.org> 26356Date: Thu Jul 22 07:03:04 2004 +0000 26357 26358 Merge and resolve conflicts from DRI/DRM/Mesa 20040721 imports. 26359 26360commit 57de5e285d6271fc4b2bc8a951a98be6308d649d 26361Author: Eric Anholt <anholt@freebsd.org> 26362Date: Fri Jul 16 04:38:54 2004 +0000 26363 26364 Bugzilla #803,804: Fix build in the !XF86DRI case after DRI and Radeon 26365 Render acceleration merges. 26366 26367commit 9b0ec803bd588d9af778bfd307ee6d958e4a1e73 26368Author: Adam Jackson <ajax@nwnk.net> 26369Date: Sat Jul 10 14:22:35 2004 +0000 26370 26371 Bugzilla #849: Fix DRI pagesize assumptions in radeon and r128 drivers. 26372 from mharris@, reviewed and applied to DRI by airlied@. 26373 26374commit ccd95bfb291a0c70272fab6822ef379b83dc298d 26375Author: Egbert Eich <eich@suse.de> 26376Date: Tue Jul 6 14:37:47 2004 +0000 26377 26378 Separated Intel drivers from default DriDrivers to avoid building them on 26379 IA64 (Egbert Eich). 26380 Fixed wrong function prototype (Egbert Eich). 26381 Don't test for generic VGA on IA64 (Egbert Eich). 26382 Fixed a segfault when accessing a structure before verifying the pointer 26383 exists (Egbert Eich). 26384 Added a showcache option for debugging (Egbert Eich). 26385 Increase default video RAM size to 16MB when DRI is enabled and more than 26386 128MB are available (Egbert Eich). Fixed lockups during mode switch. 26387 Problem was introduced when attempting to copy the behavior during 26388 LeaveVT()/EnterVT() but but forgetting to call I810DRILeave() before 26389 I810DRIEnter(). The entire DRILeave()/Enter() scenario has been 26390 commented out as it didn't seem to be necessary (Egbert Eich). 26391 Fix TweakMemorySize() (tested with i855/i865) (Egbert Eich). 26392 increased MAX_DEVICES to 128 (Egbert Eich). 26393 Use OS provided PCI config space access as default method (Egbert Eich). 26394 Added support for Linux 2.6 proc file format. 26395 Fixed unaligned accesses to pieces of the VBE info block. VESA did not 26396 align elements to size (Egbert Eich). 26397 26398commit 6bdf3d36ea3086432af5157540fde4e0aa0c44dc 26399Author: Eric Anholt <anholt@freebsd.org> 26400Date: Sun Jun 20 07:34:55 2004 +0000 26401 26402 Add file missed in last commit: Bug #748: 26403 - Add Render acceleration for Radeon 100 and 200-series cards, enabled by 26404 default. Notable performance gains include 27fps in cairogears instead 26405 of 6fps (compared to 50 with glitz), and my disks are now the 26406 bottleneck in an ls -lR in gnome-terminal. Only supported in DRI mode 26407 because the MMIO submission hangs the card so far, but the code is left 26408 in because it may be supportable soon. 26409 - Add xorg.conf options to disable Render acceleration and to force the 26410 subpixel order in the server (Hui Yu). 26411 Many thanks to Hui Yu at ATI for the code this was based off of, Michel 26412 Daenzer for bugfixes and suggestions, and LinuxFund for sponsoring the 26413 work. 26414 26415commit 8c1b37420ea890f76edb97aed61ee427a871cd3f 26416Author: Eric Anholt <anholt@freebsd.org> 26417Date: Sun Jun 20 07:26:08 2004 +0000 26418 26419 Bug #748: 26420 - Add Render acceleration for Radeon 100 and 200-series cards, enabled by 26421 default. Notable performance gains include 27fps in cairogears instead 26422 of 6fps (compared to 50 with glitz), and my disks are now the 26423 bottleneck in an ls -lR in gnome-terminal. Only supported in DRI mode 26424 because the MMIO submission hangs the card so far, but the code is left 26425 in because it may be supportable soon. 26426 - Add xorg.conf options to disable Render acceleration and to force the 26427 subpixel order in the server (Hui Yu). 26428 Many thanks to Hui Yu at ATI for the code this was based off of, Michel 26429 Daenzer for bugfixes and suggestions, and LinuxFund for sponsoring the 26430 work. 26431 26432commit 6ecf374d500afe6da494dfdd6566396ec65b6d6a 26433Author: Eric Anholt <anholt@freebsd.org> 26434Date: Wed Jun 16 09:43:59 2004 +0000 26435 26436 Merge DRI-trunk-20040613 changes in programs/Xserver/hw/xfree86/drivers, 26437 with the following notes: 26438 - Savage and Mach64 (= ati/ati*.[ch] changes) DRI not merged due to 26439 insecurity. 26440 - VIA driver converted to new drmContext and drmHandle names. 26441 - Radeon driver merge conflicted in many places, and MergedFB at least 26442 could probably use some checking at this point. 26443 26444commit bea8085e04136b0ef513c17bb65c54069ec531e1 26445Author: Eric Anholt <anholt@freebsd.org> 26446Date: Wed Jun 16 09:25:58 2004 +0000 26447 26448 DRI trunk-20040613 import 26449 26450commit 7b588cda922992a8ee2d04853ba1533cf43592a3 26451Author: Eric Anholt <anholt@freebsd.org> 26452Date: Wed Jun 16 09:25:58 2004 +0000 26453 26454 Initial revision 26455 26456commit 6cbb5bbff43d65a762d432659793333868837f0b 26457Author: Eric Anholt <anholt@freebsd.org> 26458Date: Wed Jun 16 09:22:56 2004 +0000 26459 26460 DRI XFree86-4_3_99_12-merge import 26461 26462commit 3586c3c7955365a2af0b8a286ea0d0488ddfd137 26463Author: Egbert Eich <eich@suse.de> 26464Date: Fri Apr 23 19:26:46 2004 +0000 26465 26466 Merging XORG-CURRENT into trunk 26467 26468commit 538768e0f472026325be9dd37dc4bfe7ea7d3b75 26469Author: Egbert Eich <eich@suse.de> 26470Date: Sun Mar 14 08:33:20 2004 +0000 26471 26472 Importing vendor version xf86-4_4_99_1 on Sun Mar 14 00:26:39 PST 2004 26473 26474commit 5f179e796d1cd310790d06eebb474ad67114ec84 26475Author: Egbert Eich <eich@suse.de> 26476Date: Wed Mar 3 12:12:18 2004 +0000 26477 26478 Importing vendor version xf86-4_4_0 on Wed Mar 3 04:09:24 PST 2004 26479 26480commit 493c44580191beff63575c4032f772fbbf1b06eb 26481Author: Egbert Eich <eich@suse.de> 26482Date: Thu Feb 26 13:35:52 2004 +0000 26483 26484 readding XFree86's cvs IDs 26485 26486commit 942a47fd3c5e8df85642672c01072bf97285bc30 26487Author: Egbert Eich <eich@suse.de> 26488Date: Thu Feb 26 09:23:18 2004 +0000 26489 26490 Importing vendor version xf86-4_3_99_903 on Wed Feb 26 01:21:00 PST 2004 26491 26492commit dda79b5ee3eb49f924a3213592d93ba981c99394 26493Author: Kaleb Keithley <kaleb@freedesktop.org> 26494Date: Mon Feb 23 20:35:02 2004 +0000 26495 26496 Import most of XFree86 4.4RC3. This import excludes files which have the 26497 new license. If we want to, later we can import 4.4RC3 again and pick 26498 up the files that have the new license, but for now the vendor branch 26499 is "pure." 26500 26501commit 08104f18ac497561c969694ff709da6c361ccea7 26502Author: Egbert Eich <eich@suse.de> 26503Date: Thu Jan 29 08:08:34 2004 +0000 26504 26505 Importing vendor version xf86-012804-2330 on Thu Jan 29 00:06:33 PST 2004 26506 26507commit 162d76b9c48ed17d6e645057db58c9f13436f25f 26508Author: Kaleb Keithley <kaleb@freedesktop.org> 26509Date: Fri Dec 19 20:55:06 2003 +0000 26510 26511 XFree86 4.3.99.902 (RC 2) 26512 26513commit 195772ec74f7a9b7d5bf294b3bc2c3b960e08a31 26514Author: Kaleb Keithley <kaleb@freedesktop.org> 26515Date: Thu Dec 4 22:03:16 2003 +0000 26516 26517 XFree86 4.3.99.901 (RC 1) 26518 26519commit 73105e438e90f950249cb265ba99001d4db3dc60 26520Author: Kaleb Keithley <kaleb@freedesktop.org> 26521Date: Tue Nov 25 19:28:36 2003 +0000 26522 26523 XFree86 4.3.99.16 Bring the tree up to date for the Cygwin folks 26524 26525commit 0a53864f9581196604d04fd28b4e94a9b0b73d18 26526Author: Kaleb Keithley <kaleb@freedesktop.org> 26527Date: Tue Nov 25 19:28:36 2003 +0000 26528 26529 Initial revision 26530 26531commit c56aaef79028261e0a3be0c86ffd3f16b6bdd9ba 26532Author: Kaleb Keithley <kaleb@freedesktop.org> 26533Date: Fri Nov 14 16:48:55 2003 +0000 26534 26535 XFree86 4.3.0.1 26536 26537commit d09f463a5d1ce73e0b65d5276fbcca393fa2da46 26538Author: Kaleb Keithley <kaleb@freedesktop.org> 26539Date: Fri Nov 14 16:48:55 2003 +0000 26540 26541 Initial revision 26542 26543commit d9af6dc32652502d84ea8da5d57a5ab45429c4ad 26544Author: Kaleb Keithley <kaleb@freedesktop.org> 26545Date: Fri Nov 14 15:54:49 2003 +0000 26546 26547 R6.6 is the Xorg base-line 26548