ChangeLog revision dd52494d
1commit ec2b45d1bf4c31c5420b48c213bf9ee1656adf9d 2Author: Ilia Mirkin <imirkin@alum.mit.edu> 3Date: Mon Jan 28 23:25:58 2019 -0500 4 5 Bump version to 1.0.16 6 7 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 8 9commit 2376d1ebf2d9a96bc2ebf21d53a9f9841ce5c15b 10Author: Ilia Mirkin <imirkin@alum.mit.edu> 11Date: Sat Jan 26 23:00:19 2019 -0500 12 13 update known chipsets list 14 15 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 16 Reviewed-by: Rhys Kidd <rhyskidd@gmail.com> 17 18commit ca13913aaf7e3995c3b49d3d80aad7e6cb33291a 19Author: Ilia Mirkin <imirkin@alum.mit.edu> 20Date: Wed Jan 23 00:29:51 2019 -0500 21 22 Revert "wfb: Fix missing init function decls behind FB_ACCESS_WRAPPER" 23 24 This reverts commit d7ba24fb6e4fce43adc3c73369c5d1035ae080a9. 25 26 When built and run against Xorg 1.20.3, part way through init, I got an 27 error about wfbPictureInit not existing despite not having wfb_enabled 28 set. 29 30 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 31 32commit fb2d4d27837ebf3ae2fdcf091495f52d2ad8c822 33Author: Ilia Mirkin <imirkin@alum.mit.edu> 34Date: Wed Jan 23 00:28:05 2019 -0500 35 36 present: rotated crtc's work fine 37 38 The "pick best" logic takes rotation into account. However flipping a 39 rotated CRTC can't work, so we disable that. 40 41 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 42 43commit 92db605f0f5ac14957e38e59782ef3ed4808de05 44Author: Rhys Kidd <rhyskidd@gmail.com> 45Date: Mon Jan 21 22:59:54 2019 -0500 46 47 Adapt to PixmapDirtyUpdateRec::src being a DrawablePtr 48 49 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 50 51commit 8a90309401963aa43f89f91e204ec9cac5ade704 52Author: Rhys Kidd <rhyskidd@gmail.com> 53Date: Mon Jan 21 00:52:23 2019 -0500 54 55 config: Suggest upstream to find macros in case they are missing 56 57 Added upstream suggestion to error message in case xorg macros or 58 xserver macros are missing. No actual package manager references in 59 suggestion since these are 'user' and/or 'distribution' preferences. 60 61 (Cherry picked from xf86-video-intel 90538ecfff9c324d61419f451d41d26ab026f232) 62 63 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 64 65commit fcddd2c8e6e4a738bf28028e6d4ad5be34c8b2ae 66Author: Rhys Kidd <rhyskidd@gmail.com> 67Date: Mon Jan 21 00:52:22 2019 -0500 68 69 config: Complete bug report URL for Xorg DDX nouveau driver 70 71 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 72 73commit a566c883b3663b685e467788b8b5de6d99ef9ae2 74Author: Rhys Kidd <rhyskidd@gmail.com> 75Date: Mon Jan 21 00:52:21 2019 -0500 76 77 config: Set automake options consistently in one location 78 79 Avoid use of two manners of setting automake configuration options, 80 and consolidate them into AM_INIT_AUTOMAKE macro. 81 82 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 83 84commit 56e67e09925b0b5544884b279b9cc5552964fd25 85Author: Rhys Kidd <rhyskidd@gmail.com> 86Date: Mon Jan 21 00:52:20 2019 -0500 87 88 config: remove unused sdkdir=$(pkg-config...) statement 89 90 The sdkdir variable is not used, so remove the statement. 91 92 (Copied from xf86-video-ati 36bd69affc996c92c40b7360a7fbaa1a3a46abfd) 93 94 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 95 96commit db51fcf52067e4412c96bb4c84d97ceac6c857a6 97Author: Rhys Kidd <rhyskidd@gmail.com> 98Date: Mon Jan 21 00:52:19 2019 -0500 99 100 config: Remove unrequired AC_SUBST for LIBDRM_NOUVEAU and LIBUDEV 101 102 These macros are called by the PKG_CHECK_MODULES macro, for all 103 versions of pkg-config >= 0.24.0 (released May 23, 2010). 104 105 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 106 107commit f5250b1d46b6442550140b149341e06f7f27ae8d 108Author: Rhys Kidd <rhyskidd@gmail.com> 109Date: Mon Jan 21 00:52:18 2019 -0500 110 111 config: Add comments for main statements 112 113 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 114 115commit ed653f76b17ba4b252fe405e7bd39b36142dae05 116Author: Rhys Kidd <rhyskidd@gmail.com> 117Date: Mon Jan 21 00:52:17 2019 -0500 118 119 config: Use LT_INIT option instead of AC_DISABLE_STATIC 120 121 Since libtool 2.2 (released March 2008) the preferred initiation of 122 libtool is with the LT_INIT() macro, instead of AC_PROG_LIBTOOL. 123 124 As we are using LT_INIT anyway we might as well pass the disable-static 125 option directly to it instead of using the AC_DISABLE_STATIC macro. 126 127 See: 128 - https://www.gnu.org/software/libtool/manual/html_node/LT_005fINIT.html 129 - https://www.x.org/wiki/NewModuleGuidelines/ 130 131 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 132 133commit fe3fc188f17f000afae4c307ad5e54f1596fc6ff 134Author: Rhys Kidd <rhyskidd@gmail.com> 135Date: Mon Jan 21 00:52:16 2019 -0500 136 137 config: Utilize util-macros 1.8 for additional man page support 138 139 Use MAN_SUBST now supplied in XORG_MANPAGE_SECTIONS 140 The value of MAN_SUBST is the same for all X.Org packages. 141 142 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 143 144commit caaac38112fa9066089099655866e0e9f5e3a647 145Author: Rhys Kidd <rhyskidd@gmail.com> 146Date: Mon Jan 21 00:52:15 2019 -0500 147 148 config: remove unrequired AC_HEADER_STDC 149 150 Autoconf says: 151 "This macro is obsolescent, as current systems have conforming 152 header files. New programs need not use this macro". 153 154 Was previously required to ensure the system has C header files conforming 155 to ANSI C89 (ISO C90). Specifically, this macro checks for stdlib.h, 156 stdarg.h, string.h, and float.h. 157 158 This autoconf option was used to provide conditional fallback support 159 via defined STDC_HEADERS. 160 161 xf86-video-nouveau does not do so, so this macro is obsolete and unused, 162 so let's drop it. 163 164 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 165 166commit 4cf6d02ecbd5d2201d66324ba33f9fcbed67092f 167Author: Rhys Kidd <rhyskidd@gmail.com> 168Date: Mon Jan 21 00:52:14 2019 -0500 169 170 config: Use XORG_DEFAULT_OPTIONS which replaces six XORG_* macros 171 172 This macro expands into several macros and provide several build functions [0]. 173 Refer to the module generated aclocal.m4 file as this is subject to change. 174 175 At present, this includes: 176 - XORG_CWARNFLAGS: platform adjusted compiler warning flags 177 - XORG_STRICT_OPTION: additional warning flags 178 - XORG_RELEASE_VERSION: defines variables for major, minor and patch level 179 - XORG_CHANGELOG: contains a makefile rule which creates the ChangeLog file from Git 180 - XORG_INSTALL: contains a makefile rule which provides the INSTALL file in the module root directory 181 - XORG_MANPAGE_SECTIONS: platform adjusted man page section number 182 183 The above macros invoke the following Autoconf macros: 184 - AC_PROG_INSTALL 185 - AC_PROG_CC_C99 186 - AC_PROG_SED 187 - AC_CANONICAL_HOST 188 189 [0] https://www.x.org/wiki/NewModuleGuidelines/ 190 191 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 192 193commit 945f95775a21cbcc721b579f7dc220f1474d1d25 194Author: Rhys Kidd <rhyskidd@gmail.com> 195Date: Mon Jan 21 00:52:13 2019 -0500 196 197 config: libtoolize: put macros in AC_CONFIG_MACRO_DIR, 'm4' 198 199 As suggested by one tool called by autoreconf. Necessitates change to: 200 - configure.ac: Call AC_CONFIG_MACRO_DIR() 201 - Makefile.am: Set ACLOCAL_AMFLAGS = -I m4 202 - Create placeholder m4 folders 203 - Update gitignore files 204 205 >From the autoconf documentation: 206 207 "Starting from version 2.58, autoconf provide the macro AC_CONFIG_MACRO_DIR 208 to declare where additional macro files are to be put and found. The 209 argument passed to this macro is commonly m4." 210 https://autotools.io/autoconf/macros.html 211 212 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 213 214commit 9b60c3bc74c69d607d44c864f60f5fda0b11cd48 215Author: Rhys Kidd <rhyskidd@gmail.com> 216Date: Mon Jan 21 00:52:12 2019 -0500 217 218 gitignore: Update to mirror modern X.org drivers 219 220 Avoids git showing autoconf/automake generated files (e.g. tarballs) that 221 are local build artifacts and should not be committed. 222 223 Cherry picked from xf86-video-intel as at: 224 https://gitlab.freedesktop.org/xorg/driver/xf86-video-intel/blob/7c9dbc980b760e0053d83ca2d7cb147613285680/.gitignore 225 226 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 227 228commit 1761769d43ab3ab31b5e5d92e6014fdea21830bd 229Author: Rhys Kidd <rhyskidd@gmail.com> 230Date: Mon Jan 21 00:52:11 2019 -0500 231 232 config: remove AC_PROG_CC as it overrides AC_PROG_C_C99 233 234 XORG_STRICT_OPTION from XORG_DEFAULT_OPTIONS calls 235 AC_PROG_C_C99. This sets gcc with -std=gnu99. 236 If AC_PROG_CC macro is called afterwards, it resets CC to gcc. 237 238 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 239 240commit e27c0bb6c70b1a68ddf2aeeb0e4346e0a41b8d9c 241Author: Rhys Kidd <rhyskidd@gmail.com> 242Date: Mon Jan 21 00:52:10 2019 -0500 243 244 config: Rely upon list.h as at least xorg-server 1.8 required 245 246 Fixes: edd1608 ("xorg_list: define the xorg_list* symbols") 247 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 248 249commit 6700548071763daf30ab74dbcfb6543ea0db101d 250Author: Rhys Kidd <rhyskidd@gmail.com> 251Date: Mon Jan 21 00:52:09 2019 -0500 252 253 config: Only check for presence of libdrm package once 254 255 After a version-conditioned check for libdrm was added, no longer necessary 256 to check a second time, especially without a version number check. 257 258 Fixes: 779b8d4 ("Link against libdrm") 259 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 260 261commit 9b2b008eb76c6f4c31d5f71ca01a06254a03f7c9 262Author: Rhys Kidd <rhyskidd@gmail.com> 263Date: Mon Jan 21 00:52:08 2019 -0500 264 265 config, nv: Use xorg-macros PACKAGE_VERSION_{MAJOR, MINOR, PATCHLEVEL} 266 267 Use xorg-macros provided PACKAGE_VERSION_{MAJOR,MINOR,PATCHLEVEL} rather 268 than rolling our own in xf86-video-nouveau. 269 270 Macro has been available since xorg-macros 1.1.5 (we rely on >= 1.8) 271 provided that XORG_DEFAULT_OPTIONS is called. 272 273 Fixes: b82aacd ("configure: require xorg-macros 1.8") 274 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 275 276commit b098bdae321182d2f18d791e7f1c665dec21f571 277Author: Rhys Kidd <rhyskidd@gmail.com> 278Date: Mon Jan 21 00:52:07 2019 -0500 279 280 config: Fix typo in comment 281 282 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 283 284commit 764b42babd38b545ccb3b43cd81c989bddf71c77 285Author: Rhys Kidd <rhyskidd@gmail.com> 286Date: Sun Jan 20 22:21:34 2019 -0500 287 288 xv/nv30, nv40: Mark local NV30GetSurfaceFormat and NV40GetSurfaceFormat as static 289 290 Avoids warnings with gcc 8.2: 291 292 nv30_xv_tex.c:138:1: warning: no previous prototype for ‘NV30GetSurfaceFormat’ [-Wmissing-prototypes] 293 NV30GetSurfaceFormat(PixmapPtr ppix, int *fmt_ret) 294 ^~~~~~~~~~~~~~~~~~~~ 295 296 nv40_xv_tex.c:138:1: warning: no previous prototype for ‘NV40GetSurfaceFormat’ [-Wmissing-prototypes] 297 NV40GetSurfaceFormat(PixmapPtr ppix, int *fmt_ret) 298 ^~~~~~~~~~~~~~~~~~~~ 299 300 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 301 302commit 0fa6ba423593cb5c8e43b8f6569f90f9fcc2b36c 303Author: Rhys Kidd <rhyskidd@gmail.com> 304Date: Sun Jan 20 22:21:33 2019 -0500 305 306 xv: Mark local NVSetupTexturedVideo function as static 307 308 Avoids warning with gcc 8.2: 309 310 nouveau_xv.c:2136:1: warning: no previous prototype for ‘NVSetupTexturedVideo’ [-Wmissing-prototypes] 311 NVSetupTexturedVideo (ScreenPtr pScreen, XF86VideoAdaptorPtr *textureAdaptor) 312 ^~~~~~~~~~~~~~~~~~~~ 313 314 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 315 316commit 2905ea76571faed91e7d17b5dc6e44758e71ff0b 317Author: Rhys Kidd <rhyskidd@gmail.com> 318Date: Sun Jan 20 22:21:32 2019 -0500 319 320 dri2: Mark local create/destroy buffer and copy region functions as static 321 322 Avoids warnings with gcc 8.2: 323 324 nouveau_dri2.c:38:1: warning: no previous prototype for ‘nouveau_dri2_create_buffer2’ [-Wmissing-prototypes] 325 nouveau_dri2_create_buffer2(ScreenPtr pScreen, DrawablePtr pDraw, unsigned int attachment, 326 ^~~~~~~~~~~~~~~~~~~~~~~~~~~ 327 nouveau_dri2.c:109:1: warning: no previous prototype for ‘nouveau_dri2_create_buffer’ [-Wmissing-prototypes] 328 nouveau_dri2_create_buffer(DrawablePtr pDraw, unsigned int attachment, 329 ^~~~~~~~~~~~~~~~~~~~~~~~~~ 330 nouveau_dri2.c:117:1: warning: no previous prototype for ‘nouveau_dri2_destroy_buffer2’ [-Wmissing-prototypes] 331 nouveau_dri2_destroy_buffer2(ScreenPtr pScreen, DrawablePtr pDraw, DRI2BufferPtr buf) 332 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~ 333 nouveau_dri2.c:131:1: warning: no previous prototype for ‘nouveau_dri2_destroy_buffer’ [-Wmissing-prototypes] 334 nouveau_dri2_destroy_buffer(DrawablePtr pDraw, DRI2BufferPtr buf) 335 ^~~~~~~~~~~~~~~~~~~~~~~~~~~ 336 nouveau_dri2.c:137:1: warning: no previous prototype for ‘nouveau_dri2_copy_region2’ [-Wmissing-prototypes] 337 nouveau_dri2_copy_region2(ScreenPtr pScreen, DrawablePtr pDraw, RegionPtr pRegion, 338 ^~~~~~~~~~~~~~~~~~~~~~~~~ 339 nouveau_dri2.c:215:1: warning: no previous prototype for ‘nouveau_dri2_copy_region’ [-Wmissing-prototypes] 340 nouveau_dri2_copy_region(DrawablePtr pDraw, RegionPtr pRegion, 341 ^~~~~~~~~~~~~~~~~~~~~~~~ 342 343 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 344 345commit eb93278fadaab508ba8c2db65d2d3af2bd421efa 346Author: Rhys Kidd <rhyskidd@gmail.com> 347Date: Sun Jan 20 22:21:31 2019 -0500 348 349 wfb: Remove declaration for undefined function nouveau_wfb_init() 350 351 Fixes: 794a277 ("nv50: use libwfb for pixmap access when driver pixmaps enabled") 352 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 353 354commit 5233f3c84a436e01dabac531c2615e0e9af5f4ac 355Author: Rhys Kidd <rhyskidd@gmail.com> 356Date: Sun Jan 20 21:31:19 2019 -0500 357 358 xv: Avoid shadowed declaration of 'int ret' in NVPutImage 359 360 Warning reported by gcc 8.2: 361 362 nouveau_xv.c: In function ‘NVPutImage’: 363 nouveau_xv.c:1369:7: warning: declaration of ‘ret’ shadows a previous local [-Wshadow] 364 int ret = BadImplementation; 365 ^~~ 366 nouveau_xv.c:1073:6: note: shadowed declaration is here 367 int ret, i; 368 ^~~ 369 370 Fixes: a50bba6 ("xv: misc cleanups") 371 Cc: Ben Skeggs <bskeggs@redhat.com> 372 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 373 374commit 22dd49f60be54c5649a9e80c87c9bfc99e2974b8 375Author: Rhys Kidd <rhyskidd@gmail.com> 376Date: Sun Jan 20 21:31:18 2019 -0500 377 378 xv: Avoid shadowed declaration of 'int i' in NVPutImage 379 380 int i is accessed outside immediate scope so leave declaration at 381 the highest common scope level: 382 383 1073: int ret, i; 384 ... 385 // Highest common scope 386 ... 387 1193: if (newTTSize <= destination_buffer->size) { 388 ... 389 // Used in this scope 390 ... 391 1248: } else { 392 ... 393 // Used in this scope 394 ... 395 1316: } 396 // Not used in this scope 397 398 Warning reported by gcc 8.2: 399 400 nouveau_xv.c: In function ‘NVPutImage’: 401 nouveau_xv.c:1195:7: warning: declaration of ‘i’ shadows a previous local [-Wshadow] 402 int i = 0; 403 ^ 404 nouveau_xv.c:1073:11: note: shadowed declaration is here 405 int ret, i; 406 ^ 407 408 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 409 410commit d7ba24fb6e4fce43adc3c73369c5d1035ae080a9 411Author: Rhys Kidd <rhyskidd@gmail.com> 412Date: Sun Jan 20 19:20:08 2019 -0500 413 414 wfb: Fix missing init function decls behind FB_ACCESS_WRAPPER 415 416 Fixes warning with xserver 1.20 and gcc 8.2: 417 418 nv_driver.c:1443:9: warning: implicit declaration of function ‘wfbScreenInit’; did you mean ‘fbScreenInit’? [-Wimplicit-function-declaration] 419 ret = wfbScreenInit(pScreen, FBStart, pScrn->virtualX, 420 ^~~~~~~~~~~~~ 421 fbScreenInit 422 423 (See xserver 706e6d9cd074da606016ed4ecff51e9c2a822087) 424 425 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 426 427commit dbba3246c989a88f8ba5b0c906af58779bef0cdd 428Author: Rhys Kidd <rhyskidd@gmail.com> 429Date: Sun Jan 20 14:50:35 2019 -0500 430 431 xv: Silence build warning regarding const qualifier 432 433 Silences warning with gcc 8.2: 434 435 nouveau_xv.c: In function ‘NVInitVideo’: 436 nouveau_xv.c:2247:68: warning: passing argument 2 of ‘vlCreateAdaptorXvMC’ discards ‘const’ qualifier from pointer target type [-Wdiscarded-qualifiers] 437 adaptorsXvMC[0] = vlCreateAdaptorXvMC(pScreen, textureAdaptor[0]->name); 438 ~~~~~~~~~~~~~~~~~^~~~~~ 439 In file included from nouveau_xv.c:42: 440 vl_hwmc.h:6:63: note: expected ‘char *’ but argument is of type ‘const char *’ 441 XF86MCAdaptorPtr vlCreateAdaptorXvMC(ScreenPtr pScreen, char *xv_adaptor_name); 442 ~~~~~~^~~~~~~~~~~~~~~ 443 444 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 445 Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu> 446 447commit cc284803fd75e0e340a5481342e0d0f7a5d7fa15 448Author: Rhys Kidd <rhyskidd@gmail.com> 449Date: Sat Jan 19 15:45:36 2019 -0500 450 451 Check for xf86CursorResetCursor() 452 453 If it's available, xorg-server calls it on each mode configuration change. It 454 does what xf86_reload_cursors does (and more), so we don't need to call 455 the latter anymore. 456 457 Avoids gcc 8.2 warning: 458 459 drmmode_display.c: In function ‘drmmode_set_mode_major’: 460 drmmode_display.c:525:2: warning: ‘xf86_reload_cursors’ is deprecated [-Wdeprecated-declarations] 461 xf86_reload_cursors(crtc->scrn->pScreen); 462 ^~~~~~~~~~~~~~~~~~~ 463 In file included from nv_type.h:10, 464 from nv_include.h:69, 465 from drmmode_display.c:36: 466 /usr/include/xorg/xf86Crtc.h:1068:37: note: declared here 467 static _X_INLINE _X_DEPRECATED void xf86_reload_cursors(ScreenPtr screen) {} 468 ^~~~~~~~~~~~~~~~~~~ 469 470 (Ported from radeon commit d670c5c9851b4eff21c845d26c7d7e4eb5ee0fa9) 471 472 Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> 473 474commit a4283f7fee5b0376f62ba305a2a47ee3dc5a822f 475Author: Ilia Mirkin <imirkin@alum.mit.edu> 476Date: Fri Nov 10 23:28:08 2017 -0500 477 478 drmmode: update logic for dynamic connectors, paths, and tiles 479 480 This largely copies the code from modesetting with minor adjustments. 481 482 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 483 484commit 8dd1118369460e03ab4963987901dc9e13a2d53d 485Author: Mario Kleiner <mario.kleiner.de@gmail.com> 486Date: Tue Jun 12 17:58:06 2018 +0200 487 488 Fix colormap handling at screen depth 30. (v2) 489 490 The various clut handling functions like a setup 491 consistent with the x-screen color depth. Otherwise 492 we observe improper sampling in the gamma tables 493 at depth 30. 494 495 Tested at depths 16, 24 and 30 and tested at depths 496 24 and 30 that xgamma and gamma table animations work, 497 and with measurement equipment to make sure identity 498 gamma ramps actually are identity mappings at the output. 499 500 v2: Also deal with X-Server 1.19 and earlier, which as of 501 v1.19.6 lack a fix to color palette handling and can 502 not deal with depths/bpc > 24/8 bpc. On < 1.20 we skip 503 xf86HandleColormaps() setup at > 8 bpc. This disables 504 color palette handling on such servers at > 8 bpc, but 505 still keeps RandR gamma table handling intact. 506 507 Tested on 1.19.6 and 1.20.0 to do the right thing. 508 509 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 510 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 511 512commit c9ce4b8e86525b6e0ecc814b2c5ebb242fb1f107 513Author: Ilia Mirkin <imirkin@alum.mit.edu> 514Date: Sun Dec 9 13:40:55 2018 -0500 515 516 drmmode: provide better error when failing to set gamma 517 518 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 519 520commit ac8f7b949e50d96f281b09f7fe96c92ea4d178ec 521Author: Ilia Mirkin <imirkin@alum.mit.edu> 522Date: Fri Feb 9 23:00:27 2018 -0500 523 524 dri3: don't check permissions on render node 525 526 The permission check fails if udev sets the render node to 0666 but 527 leaves the card at 0660, as is done in at least udev-236. 528 529 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 530 531commit 6e2b4bed7b226dce262c7e14f72a29b003791be8 532Author: Ilia Mirkin <imirkin@alum.mit.edu> 533Date: Sun Feb 4 13:50:29 2018 -0500 534 535 nv50/xv: add support for depth 30 xv output 536 537 Ostensibly it should probably be looking at the pixmap format. However 538 it's always the window pixmap, so we can assume it's what we expect. 539 540 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 541 542commit ea83b9bd8c88dad67e6494031f1f65eed823b6f1 543Author: Ilia Mirkin <imirkin@alum.mit.edu> 544Date: Sun Dec 31 02:22:27 2017 -0500 545 546 dri3: remove bogus condition for creating pixmap 547 548 Not clear what the depth % 8 was trying to protect against, but it was 549 breaking 30bpp visuals with DRI3. Add it in to ensure that bitsPerPixel 550 % 8 is 0, since there is plenty of bpp/8 math in the driver. 551 552 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 553 554commit 048baf53872f2f93628f2c8637ec832fb4006057 555Author: Ilia Mirkin <imirkin@alum.mit.edu> 556Date: Sun Dec 10 23:40:51 2017 -0500 557 558 man: remove reference to glamor under DRI option 559 560 GLAMOR backend was removed a while ago, but the help text stayed on. 561 562 Reported-by: Jason Zaman <jason@perfinion.com> 563 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 564 565commit 69aecdd305defdd014bc92b82acc6000988511cf 566Author: Adam Jackson <ajax@redhat.com> 567Date: Tue Jun 13 09:23:42 2017 -0400 568 569 modesetting: Validate the atom for enum properties 570 571 The client could have said anything here, and if what they said doesn't 572 actually name an atom NameForAtom() will return NULL, and strcmp() will 573 be unhappy about that. 574 575 [copied from xserver d4995a3936ae283b9080fdaa0905daa669ebacfc] 576 577 Signed-off-by: Adam Jackson <ajax@redhat.com> 578 579commit a7c190e0c230ed2d5e047ed6aba12d71a4866950 580Author: Lyude <lyude@redhat.com> 581Date: Fri Apr 21 14:41:17 2017 -0400 582 583 Bump version to 1.0.15 584 585 Signed-off-by: Lyude <lyude@redhat.com> 586 587commit 3047e99a641d2b8720371aae292750206f5e91fc 588Author: Ben Skeggs <bskeggs@redhat.com> 589Date: Thu Apr 20 15:57:33 2017 +1000 590 591 fix null pointer deref when building against >=libdrm 2.4.78 592 593 A new pointer got added to drmEventContext. 594 595 As a result of us both: 596 - Setting drmEventContext.version to "latest" AND 597 - Not zeroing the struct 598 599 We end up thwarting libdrm's compatibility checks, resulting in 600 libdrm choosing to call through an invalid pointer. 601 602 Fix this by zeroing the struct. 603 604 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 605 606commit e9418e434311336e905b70553a5ed740838d90ad 607Author: Mariusz Bialonczyk <manio@skyboo.net> 608Date: Wed Mar 29 22:07:50 2017 +0200 609 610 Do not register hotplug without RandR 611 612 When using Xinerama, RandR is automatically disabled, and calling RR 613 routines will trigger an assert() because the RR keys/resources are 614 not set, leading to an Xserver abort. 615 616 Hotplug makes little sense without RandR, so no need to install a 617 udev monitor if RandR is not available. 618 619 Ported from xf86-video-intel commit 620 1a489142c8e6a4828348cc9afbd0f430d3b1e2d8, original work by: 621 Chris Wilson <chris@chris-wilson.co.uk> 622 623 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98383 624 Signed-off-by: Mariusz Bialonczyk <manio@skyboo.net> 625 Acked-by: Ilia Mirkin <imirkin@alum.mit.edu> 626 627commit e472b47d15634a864c8c981ed588d882aceaf26b 628Author: Ilia Mirkin <imirkin@alum.mit.edu> 629Date: Tue Mar 21 21:25:33 2017 -0400 630 631 Add Pascal family support, identical to Maxwell 632 633 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 634 Tested-by: Lyude <lyude@redhat.com> 635 636commit b71de83b7fae0abeb311251e6144294d319062cf 637Author: Lyude <lyude@redhat.com> 638Date: Tue Mar 7 18:44:43 2017 -0500 639 640 Bump version to 1.0.14 641 642 Signed-off-by: Lyude <lyude@redhat.com> 643 644commit 924083938c8f209d8f6ff472caf8692a644f7e78 645Author: Lyude <lyude@redhat.com> 646Date: Fri Mar 3 18:27:42 2017 -0500 647 648 Consider CRTCs disabled when DPMS is off 649 650 It turns out there's a difference in X between a CRTC being "disabled" 651 and simply having it's DPMS turned off. This is problematic though, 652 because if DPMS is turned off you can't really use the CRTC as a normal 653 CRTC anyway since page flipping and vblanks will be non-functional. As a 654 result, we've been considering DPMS-on CRTCs as enabled and attempt to 655 perform pageflips, vblank waits, etc. on them which inevitably fails. and 656 usually breaks the display the first time any of the CRTCs have their 657 DPMS turned on. 658 659 This was a problem that didn't really show itself until kernel 4.10 when 660 atomic modesetting was added which caused nouveau to stop trying to 661 fulfill pageflips and vblank waits on disabled CRTCs. I'm not sure how 662 pageflipping disabled CRTCs ever worked in the first place, but since 663 not doing so is the proper behavior anyway I haven't investigated any 664 further. 665 666 So, copy the ms_crtc_on() function from the modesetting driver and add 667 it here as drmmode_crtc_on(), then use that in all of the places where 668 we should be checking for both DPMS off and disabled CRTCs. 669 670 This fixes issues with the X ceasing to function (usually) after the 671 first time a CRTC has it's DPMS turned on. Reproduction recipe: 672 673 - Load up gnome-shell on a machine 674 - Wait for the display to timeout from inactivity and turn itself off 675 - Shake the cursor or press something on the keyboard. Chances are the 676 monitor will come back on, but the display remains black until the 677 next time the X server is restarted. 678 679 Signed-off-by: Lyude <lyude@redhat.com> 680 Reviewed-by: Adam Jackson <ajax@redhat.com> 681 682commit 1516d35b06c9cda399bef01d992805d6e63dcbae 683Author: Ilia Mirkin <imirkin@alum.mit.edu> 684Date: Sun Oct 16 15:10:10 2016 -0400 685 686 recognize and accelerate GM20x 687 688 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 689 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 690 691commit b00b73c3aa0da1d4cee5c9f580ca65a7bd344e0f 692Author: Ilia Mirkin <imirkin@alum.mit.edu> 693Date: Mon Oct 17 23:23:41 2016 -0400 694 695 copy: add maxwell/pascal copy engine classes 696 697 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 698 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 699 700commit 5a3ada3b0ff3d89441faf35d331bb8b6cf5e51c0 701Author: Ilia Mirkin <imirkin@alum.mit.edu> 702Date: Sun Oct 16 15:03:35 2016 -0400 703 704 nvc0: refactor TIC uploads to allow different specifics per generation 705 706 This flips GM10x to using the updated format, which is what I tested 707 with. However GM20x and GP10x also use this TIC format. 708 709 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 710 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 711 712commit 514441f32c5b948ce87668234ab9f01e8ec28dac 713Author: Ilia Mirkin <imirkin@alum.mit.edu> 714Date: Sun Oct 16 12:42:18 2016 -0400 715 716 nvc0: rename BEGIN_IMC0 to IMMED_NVC0 717 718 For consistency with mesa. It wasn't used anywhere previously. 719 720 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 721 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 722 723commit 4459c04a6bc208924fac8082920f6786538bc5cd 724Author: Ilia Mirkin <imirkin@alum.mit.edu> 725Date: Sun Oct 16 12:38:38 2016 -0400 726 727 nvc0: make use of the new hwdefs for TEX_CB_INDEX 728 729 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 730 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 731 732commit ea6fd12ea059c6af832bd2180b26d75e00d4ea71 733Author: Ilia Mirkin <imirkin@alum.mit.edu> 734Date: Sun Oct 16 12:38:02 2016 -0400 735 736 hwdefs: update nvc0_3d, add gm107_texture for new TIC format 737 738 These are copied directly from the mesa repository. 739 740 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 741 Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 742 743commit a24ded627e42d453546c553d609edc073f59cd57 744Author: Ilia Mirkin <imirkin@alum.mit.edu> 745Date: Thu Sep 25 20:15:51 2014 -0400 746 747 exa: add GM10x acceleration support 748 749 rendercheck -f a8r8g8b8 passes as much as on a GK208, and xv appears to 750 work. Very lightly tested. 751 752 Instead of sticking coordinates into pushbufs, the vertex shader is 753 modified to read them from a constbuf, indexed by vertex id. This 754 approach could be used for all nvc0 generations, but I didn't want to 755 rock the boat. 756 757 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 758 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 759 760commit e6479845ec0db20dc733c621b7967b751840a552 761Author: Ilia Mirkin <imirkin@alum.mit.edu> 762Date: Tue Sep 20 00:31:06 2016 -0400 763 764 Bump version to 1.0.13 765 766 - Fixes for reverse prime offload 767 - ABI 23 support for Xorg 1.19 768 769 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu 770 771commit 6473b68762b0dca2dfccfdfc74100398b7459296 772Author: Keith Packard <keithp@keithp.com> 773Date: Tue Jul 19 08:29:40 2016 -0700 774 775 Use NotifyFd for drm and udev fds 776 777 NotifyFd is available after API 22, and must be used after API 23. 778 779 Signed-off-by: Keith Packard <keithp@keithp.com> 780 781commit a964931e8dc4e06c27749edb4993a656dc261d75 782Author: Adam Jackson <ajax@redhat.com> 783Date: Tue Jul 19 10:03:56 2016 -0400 784 785 Adapt Block/WakeupHandler signature for ABI 23 786 787 Signed-off-by: Adam Jackson <ajax@redhat.com> 788 789commit 12f77348e7a9579b167b41228dec9e6f97b74de8 790Author: Hans de Goede <hdegoede@redhat.com> 791Date: Fri Jun 3 14:46:10 2016 +0200 792 793 Properly cleanup fb for reverse-prime-offload 794 795 drmmode_set_scanout_pixmap(pix) adds drmmod->fb_id through a call 796 to drmmode_xf86crtc_resize(), but on a subsequent 797 drmmode_set_scanout_pixmap(NULL) it would not remove the fb. 798 799 This keeps the crtc marked as busy, which causes the dgpu to not 800 being able to runtime suspend, after an output attached to the dgpu 801 has been used once. Which causes burning through an additional 10W 802 of power and the laptop to run quite hot. 803 804 This commit adds the missing remove fb call, allowing the dgpu to runtime 805 suspend after an external monitor has been plugged into the laptop. 806 807 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 808 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 809 810commit 1da8a937be19e41c51a3d516bd98cee988bca44b 811Author: Ben Skeggs <bskeggs@redhat.com> 812Date: Thu Jun 2 11:23:53 2016 +1000 813 814 exa/nv50-: fix some potential incomplete pushes 815 816 I don't think these should actually be able to happen, given that we 817 already reject the unknown formats in the Check() functions. But, 818 just in case... 819 820 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 821 822commit bb57f9af9d179af0962456dca369953521ea8c2c 823Author: Ben Skeggs <bskeggs@redhat.com> 824Date: Thu Jun 2 10:17:58 2016 +1000 825 826 fix use of out-of-scope data 827 828 Reported by Coverity. 829 830 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 831 832commit b824d36c28124955eda4aced5e637aa75eea4d6c 833Author: Ilia Mirkin <imirkin@alum.mit.edu> 834Date: Wed Jan 13 11:55:06 2016 -0500 835 836 HAS_DIRTYTRACKING_ROTATION also supports multiple CRTCs 837 838 A port of commit ff9a6b6f from xf86-video-ati. 839 840 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93670 841 Reported-by: Thomas D. <thfrdue@gmx.de> 842 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 843 844commit b18bc036bf9997ea65f5cbd824fd057931e604fe 845Author: Ben Skeggs <bskeggs@redhat.com> 846Date: Tue Dec 8 15:52:25 2015 +1000 847 848 Bump version to 1.0.12 849 850 Various random fixes have been pending for a long while now... 851 852 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 853 854commit 6e6d8ac1c7b4ee047a7b40b95dea1e65a7c3211a 855Author: Mario Kleiner <mario.kleiner.de@gmail.com> 856Date: Sun Jun 28 02:33:49 2015 +0200 857 858 Take shift in crtc positions for ZaphodHeads configs into account. 859 860 In multi-x-screen ZaphodHeads configurations, there isn't a 861 one-to-one mapping of kernel provided drmmode crtc index 862 to the index of the corresponding xf86Crtc inside the 863 xf86CrtcConfig crtc array anymore, ie. for kernel provided 864 drmmode->mode_res->crtcs[i], the i'th crtc won't correspond 865 to the xf86Crtc in the i'th slot of the x-screens xf86CrtcConfig 866 anymore, once ZaphodHeads has only selected a subset of all crtcs 867 of a graphics card for a given x-screen, instead of all crtcs. 868 869 This breaks the mapping of bit positions in the bit masks returned 870 in kencoder->possible_crtcs and kencoder->possible_clones. A 1 bit 871 in position i of those masks allows use of the kernels i'th crtc for 872 the given kencoder. The X-Servers dix code checks those bit masks 873 for valid xf86Output -> xf86Crtc assignments, assuming that the i'th 874 slot xf86CrtcConfigPtr config->crtc[i] corresponds to bit i in the 875 xf86Output->possibe_crtcs bitmask, and bails if the bitmask doesn't 876 allow the specified assignment of crtc to output. If ZaphodHeads 877 breaks the assumption of bit i <-> crtc slot i this ends in failure. 878 879 Take this shift of crtc index positions wrt. encoder bitmask bit 880 positions into account by bit-shifting positions accordingly when 881 assigning encoder->possible_crtcs to output->possible_crtcs, so 882 the proper indices match up again for validation by the dix. 883 884 This problem wasn't apparent last year when testing the ZaphodHeads 885 support on some Kepler cards, as apparently the encoder->possible_crtcs 886 bitmasks returned for those cards by the kernel just had all 4 887 lsb bits set for all tested encoders/output, so each of the cards 4 888 crtcs could go with each output and things worked by chance. 889 890 The current code breaks, e.g., on 2010 MacBookPro with nv50, where 891 one crtc is hardwired to the internal lvds panel, and one crtc 892 is hardwired to the external DP connector, resulting in a failure 893 where dual-display on single-x-screen works fine, but assigning 894 each output to a separate x-screen via ZaphodHeads fails due to 895 the mismatched encoder->possible_crtcs bitmasks. 896 897 This patch fixes the problem. 898 899 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 900 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 901 902commit 1ff13a922535924681b91452235b017e43a4c6f6 903Author: Ilia Mirkin <imirkin@alum.mit.edu> 904Date: Sun Sep 13 15:29:15 2015 -0400 905 906 fix build after glamor removal 907 908 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 909 910commit 3bb943b33bbe8e593afc00d13fe867dd7a708c3c 911Author: Ilia Mirkin <imirkin@alum.mit.edu> 912Date: Thu Aug 21 20:36:41 2014 -0400 913 914 xv: use correct max width/height settings 915 916 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=70931 917 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 918 919commit 3e2e0faa2ee1cce9c1bb5c7ad80d0592460f3edc 920Author: Ilia Mirkin <imirkin@alum.mit.edu> 921Date: Sat Jul 11 17:16:27 2015 -0400 922 923 remove maxwell GM10x support for now 924 925 There is no EXA acceleration, and now no GLAMOR either. 926 User is better off with modesetting until EXA makes an appearance. 927 928 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 929 930commit 1f8b2b348526d94e9bde4a285f331a4934c11078 931Author: Ilia Mirkin <imirkin@alum.mit.edu> 932Date: Sat Jul 11 17:08:52 2015 -0400 933 934 glamor: remove 935 936 If you want glamor, just use modesetting instead. It's hooked up 937 incorrectly in nouveau, and there doesn't seem to be any point in fixing 938 it when the modesetting driver does it all properly. 939 940 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 941 942commit 6296145654b78518f3299bb5887f224f0d3810fd 943Author: Mario Kleiner <mario.kleiner.de@gmail.com> 944Date: Wed Jul 29 14:39:02 2015 +0200 945 946 Add Option "DRI" to allow selection of maximum DRI level. (v2) 947 948 Allow user to select the maximum level of DRI implementation 949 to use, DRI2 or DRI3. 950 951 exa accel supports both DRI2 and, if the kernel supports 952 rendernodes, also DRI3. However, DRI3 still seems to have 953 some bugs on current implementations, and additionally it 954 doesn't work well at all for X-Servers older than 1.16.3 955 due to X-Server bugs. Therefore we default to DRI2 on exa, 956 but allow the user to enable DRI3 with this new option. 957 958 nouveau's glamor accel backend currently doesn't work under 959 DRI2 at all, so we continue to use DRI3 whenever it is 960 supported and ignore this new option for now. 961 962 Also add a bit more output about status of Present and 963 DRI3 to aid debugging. 964 965 Note: This was originally meant to be a boolean parameter, 966 to just select between DRI3 on and off, but changed 967 here to a DRI level to make it consistent with the 968 same option in the released Intel-ddx. 969 970 v2: Use fixed up Bool return type of nouveau_present_init(). 971 972 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 973 Cc: Ilia Mirkin <imirkin@alum.mit.edu> 974 Cc: Emil Velikov <emil.l.velikov@gmail.com> 975 Cc: Martin Peres <martin.peres@free.fr> 976 Cc: Ben Skeggs <bskeggs@redhat.com> 977 Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu> 978 979commit 0deb1fcc6d1099db698ff8c512a4884708a458ec 980Author: Mario Kleiner <mario.kleiner.de@gmail.com> 981Date: Wed Jul 29 14:39:01 2015 +0200 982 983 present: Fixup return type of nouveau_present_init() 984 985 Make it a Bool consistently, as declared in header. 986 987 Reported-by: Ilia Mirkin <imirkin@alum.mit.edu> 988 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 989 Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu> 990 991commit 5212dbd7e5eba35fdd6578b30f5b0c8c8543081b 992Author: Ilia Mirkin <imirkin@alum.mit.edu> 993Date: Tue Jul 14 19:23:01 2015 -0400 994 995 Avoid build fail without COMPOSITE 996 997 Reviewed-by: Dave Airlie <airlied@redhat.com> 998 999commit fdc528306e4018b314f2efcb2be4e37672bddd75 1000Author: Vadim Rutkovsky <vrutkovs@redhat.com> 1001Date: Thu Jul 9 23:23:37 2015 +0200 1002 1003 Adapt to new dirty tracking api 1004 1005 Mirrors commit b6d871bf2 (radeon: adopt for new X server dirty tracking 1006 APIs.) in xf86-video-ati. 1007 1008 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91284 1009 1010commit dfd827cefb940f46b18dc77c129f0a603c7c95fa 1011Author: Samuel Pitoiset <samuel.pitoiset@gmail.com> 1012Date: Wed May 20 17:08:58 2015 +0200 1013 1014 fix a wrong use of a logical operator in drmmode_output_dpms() 1015 1016 This is probably a typo error which has been introduced in 2009... 1017 This fixes the following warning detected by Clang : 1018 1019 drmmode_display.c:907:30: warning: use of logical '&&' with constant operand [-Wconstant-logical-operand] 1020 if (props && (props->flags && DRM_MODE_PROP_ENUM)) { 1021 1022 Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 1023 Reviewed-by: Dave Airlie <airlied@redhat.com> 1024 1025commit 27234dbe00caf6dfd4efb24fc7bbc026fa0134fc 1026Author: Ilia Mirkin <imirkin@alum.mit.edu> 1027Date: Wed May 27 20:26:42 2015 -0400 1028 1029 nv50/exa: add comments showing the code that makes up the programs 1030 1031 As decoded by envydis. Perhaps these should move into separate files 1032 similar to nvc0, but this seems like a simple intermediate solution. 1033 1034 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1035 1036commit 08ef36dc6766080f173ae2b7e660b3182cab9f2d 1037Author: Ilia Mirkin <imirkin@alum.mit.edu> 1038Date: Wed May 20 18:36:17 2015 -0400 1039 1040 older hardware prefers copy to clear, so use that 1041 1042 Doing a solid copy of 0, or a clear, has the same effect. This is a 1043 partial revert of commit 5439a9905eab8c. Pre-nv50 hardware succeeds with 1044 copies more often. 1045 1046 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1047 1048commit 278ad73475bd137eac8a49ec7a22406bfc2867e7 1049Author: Ilia Mirkin <imirkin@alum.mit.edu> 1050Date: Tue May 19 16:26:16 2015 -0400 1051 1052 nv04-nv40: don't attempt to do 32-bit shifts 1053 1054 A 32-bit shift is a no-op, which will also make the new planemask get 1055 or'd with ~0, thus negating the usefulness of the subsequent 1056 planemask != ~0 check. Only do this if it's a less-than-32-bit per pixel 1057 format, in which case it will have the desired effect of setting the 1058 high bits. 1059 1060 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1061 Reviewed-by: Martin Peres <martin.peres@free.fr> 1062 1063commit 7583471e1fe181dc7c26b7d0a0d16f70eb10dd21 1064Author: Ilia Mirkin <imirkin@alum.mit.edu> 1065Date: Tue May 19 16:11:17 2015 -0400 1066 1067 Check before trying a solid fill 1068 1069 Pre-nv50 has all sorts of funny requirements for non-copy alu 1070 operations, and will bail out of solid fills left and right. Account for 1071 that case and fall back to the memset. 1072 1073 Reported-by: Andrew Randrianasulu <randrianasulu@gmail.com> 1074 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1075 Reviewed-by: Martin Peres <martin.peres@free.fr> 1076 1077commit 212fc5354ed60ba282d5eabf468c0e2dd0d0afe6 1078Author: Ilia Mirkin <imirkin@alum.mit.edu> 1079Date: Mon Mar 23 20:05:23 2015 -0400 1080 1081 nv50: explicitly disable ZETA buffer 1082 1083 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1084 1085commit 734b8404c0a1fa9ce8ee62a0d4cc5baae7c8fcdf 1086Author: Ilia Mirkin <imirkin@alum.mit.edu> 1087Date: Sat Mar 21 11:46:35 2015 -0400 1088 1089 use defined method names where available 1090 1091 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1092 1093commit 5439a9905eab8c8d0f86e9963df6c893633bdbfd 1094Author: Maarten Lankhorst <dev@mblankhorst.nl> 1095Date: Mon Mar 23 19:41:38 2015 +0100 1096 1097 Actually do a clear instead of doing a bogus copy. 1098 1099 Oops! 1100 1101commit f0fdc156039eaa79fb1ad82fcc6437f2432306c6 1102Author: Maarten Lankhorst <dev@mblankhorst.nl> 1103Date: Sat Mar 21 22:30:10 2015 +0100 1104 1105 Try using EXA ops harder in drmmode_fbcon_copy fallback 1106 1107 The memcpy in the fallback path fails with SIGBUS on GK20A because 1108 this memory is not coherent, try to use a solid fill instead. 1109 1110commit 7319b54291d59e1cd3cbe4f39ea8cc6dc8bf45bc 1111Author: Maarten Lankhorst <dev@mblankhorst.nl> 1112Date: Fri Mar 20 23:24:50 2015 +0100 1113 1114 Simplify nouveau_allocate_surface's pitch calculation. 1115 1116 Judging from previous commits all linear buffers should have a 256 byte alignment. 1117 1118commit 91e51a7571e6e8cb9338fc168f817f56438b1901 1119Author: Maarten Lankhorst <dev@mblankhorst.nl> 1120Date: Fri Mar 20 23:17:13 2015 +0100 1121 1122 Use drmGetRenderDeviceNameFromFd to obtain the dri3 node. 1123 1124commit d29d8baad4d845ec9d3d1d1bc8d1c1e1f6f6f086 1125Author: Maarten Lankhorst <dev@mblankhorst.nl> 1126Date: Sat Mar 14 18:00:00 2015 +0100 1127 1128 Use the correct class and shader ops for GK20A. 1129 1130 Signed-off-by: Maarten Lankhorst <dev@mblankhorst.nl> 1131 1132commit 6829d4c672fc05f4d3d993ac1bfde21d4b6c1200 1133Author: Maarten Lankhorst <dev@mblankhorst.nl> 1134Date: Sat Mar 14 11:58:12 2015 +0100 1135 1136 Make the ddx load with platform devices 1137 1138 Signed-off-by: Maarten Lankhorst <dev@mblankhorst.nl> 1139 1140commit 3fb97d78202213c1221a1d3ad8a5cebda78fdf44 1141Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 1142Date: Thu Oct 23 09:29:14 2014 +0200 1143 1144 fix null pointer deref 1145 1146commit 90ba0bb6a371a5beebdf6d363bcd73659b7c9f1f 1147Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 1148Date: Thu Oct 23 08:25:01 2014 +0200 1149 1150 check memory allocation in nouveau_glamor_set_shared_pixmap_backing 1151 1152 As reported by Roy Spliet. 1153 1154commit 4dfa13c77a1f6cee1affea05b5f3a0b26827d366 1155Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 1156Date: Thu Oct 23 08:21:55 2014 +0200 1157 1158 rework initialization to prevent some crashes 1159 1160commit 485902e61b18aff864226db62ba10399aab40128 1161Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 1162Date: Wed Oct 22 15:52:22 2014 +0200 1163 1164 fix !DRI3 case 1165 1166commit 79d438ec4d23280a242d36c2d81f2f43f6e90ada 1167Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 1168Date: Wed Oct 22 15:42:55 2014 +0200 1169 1170 fix crash in nouveau_glamor_set_shared_pixmap_backing 1171 1172commit 217f6577afe44b10f7f8e066eaf55419fb0f5ee1 1173Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 1174Date: Wed Oct 22 13:16:33 2014 +0200 1175 1176 move down call to nouveau_present_init until after fb init. 1177 1178commit 241e7289f25a342a457952b9b0e539c2f0b81d99 1179Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 1180Date: Wed Oct 22 11:04:16 2014 +0200 1181 1182 enable dri3 support without glamor 1183 1184commit 762b22f5a2b9deed1c6b18d7098f8fddf2f4b759 1185Author: Ilia Mirkin <imirkin@alum.mit.edu> 1186Date: Tue Sep 23 21:44:12 2014 -0400 1187 1188 kepler: remove unnecessary texbars, add live-only tex for nvf0 1189 1190 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1191 1192commit ce9d6e165cd810f06d158e3cfbb4b9f14b060ded 1193Author: Ilia Mirkin <imirkin@alum.mit.edu> 1194Date: Tue Sep 23 21:37:46 2014 -0400 1195 1196 allow path to envyas binary to be specified 1197 1198 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1199 1200commit 02c1aee91ae22b58e777716ffd38397f9df0a087 1201Author: Ilia Mirkin <imirkin@alum.mit.edu> 1202Date: Thu Sep 4 15:27:53 2014 -0400 1203 1204 copy: don't bother trying to create copy on pre-nva3 1205 1206 Only the NVA3/5/8/F tesla chips have a copy engine. Don't bother trying 1207 to create one on earlier tesla chips (and thus printing an error on 1208 failure). 1209 1210 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1211 1212commit 3cd4c8494c296c7583dfa7f0823a272c9e932e03 1213Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1214Date: Thu Aug 28 03:57:48 2014 +0200 1215 1216 Bump version to 1.0.11 for release. 1217 1218 Highlights: 1219 1220 - Support for server managed fd's. 1221 - Glamor support. 1222 - Maxwell support. 1223 - DRI3 and initial Present support. 1224 - vsync'ed kms pageflip performance fixes when running on Linux 3.13+ 1225 - Multi-display vsync, vblank, swap scheduling, timestamping fixes. 1226 - Multi x-screen support fixes. 1227 - ZaphodHead support on for multiple outputs per x-screen. 1228 - EXA nv-10 fixes. 1229 - Enable sync of swaps to vblank by default (Option GLXVblank "on"). 1230 - Disable pseudo-triplebuffering by default (Option SwapLimit "1"). 1231 1232 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1233 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1234 1235commit 6d49f3f7aaaec872b6f9841f744a0bf09cc98492 1236Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1237Date: Thu Aug 28 03:49:55 2014 +0200 1238 1239 dri2: Set default SwapLimit to 1 for double-buffering. 1240 1241 This used to be 2 for pseudo-triplebuffering on XOrg 1.12+. 1242 1243 However, it caused problems like unthrottled swapping 1244 at up to twice the video refresh rate for redirected 1245 windows under desktop composition, which can't be easily 1246 fixed under dri2. 1247 1248 Default to double-buffering. Users can override this, but 1249 dri3 provides better solutions for this. 1250 1251 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1252 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1253 1254commit 10729a442be3cf018c5806c95ec4218977917ef3 1255Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1256Date: Mon Aug 18 09:53:27 2014 +0200 1257 1258 xv/dri2/dri3: Switch to optimized crtc selection for drawables. 1259 1260 Make crtc selection consistent with ati and intel ddx. 1261 1262 Pick the crtc to use for vblank events, swap scheduling and 1263 kms-pageflip completion events as the one with maximum pixel area 1264 intersection between its viewport and the drawable. If multiple 1265 crtc's viewports display the same amount of area of a drawable, 1266 select the crtc whose output is the designated "primary output" 1267 for the x-screen. The latter allows, e.g., in a clone display 1268 setup, to select the important output on which tearing should be 1269 minimized. The former makes sure to minimize or avoid tearing on 1270 the display which shows the largest part of the drawable. 1271 1272 This method is transplanted from the ati ddx. 1273 1274 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1275 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1276 1277commit be96a5a62df5e1f286690311a6bc4df4f2a99e57 1278Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1279Date: Sun Aug 17 02:30:36 2014 +0200 1280 1281 Add file "compile" to .gitignore 1282 1283 Some auto-generated file by new make systems to deal with some 1284 old compilers. 1285 1286 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1287 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1288 1289commit 16c885ce9c7bebcfa1252a5d500956dd5f551043 1290Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1291Date: Tue Aug 12 11:08:24 2014 +0200 1292 1293 Fix event handling on multi-x-screen configs. 1294 1295 Only register the wakeup handler and event socket 1296 once per card fd and server generation, as the fd 1297 and device file is shared between all x-screens for 1298 a given card during a given server generation. 1299 1300 Without this fix, vblank and kms-pageflip completion 1301 event processing don't work properly, as the server 1302 doesn't kick the wakeup handler for gpu events, and 1303 therefore the desktop will freeze, unless the user 1304 manually kicks the wakeup handler by moving the mouse 1305 or hitting the keyboard. 1306 1307 Add proper reference counting and checks to make it so. 1308 This fix is derived from a similar and proven fix in 1309 the ati ddx for the same problem. 1310 1311 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1312 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1313 1314commit 7eab6806927ea8938b9bd143d1702e87fde2f2cc 1315Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1316Date: Wed Aug 13 06:48:10 2014 +0200 1317 1318 Make new ZaphodHeads work across server regenerations. 1319 1320 Reset the mask of already assigned crtc's at start 1321 of each new server generation. 1322 1323 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1324 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1325 1326commit 6ad053788e804f3a8bd36ee3653ec95e7df62623 1327Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1328Date: Wed Aug 6 07:37:32 2014 +0200 1329 1330 dri2 and xv: Fix NVxxSyncToVBlank() for multi-x-screen's. 1331 1332 Use helper function drmmode_head() to choose proper hw-crtc 1333 to sync to. Fixes xv blits, and dri2 Copy-Swaps. 1334 1335 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1336 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1337 1338commit 17de663a71bcad6f69414eb6bc6bf9b76a6bf5f4 1339Author: Ben Skeggs <bskeggs@redhat.com> 1340Date: Mon Aug 11 09:48:42 2014 +1000 1341 1342 fix glamor header check 1343 1344 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1345 1346commit 64ca88570b71289a019a2d130b1a30dc474ef52d 1347Author: Ben Skeggs <bskeggs@redhat.com> 1348Date: Mon Aug 11 08:59:30 2014 +1000 1349 1350 present: use drmmode_head() where it's appropriate 1351 1352 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1353 1354commit 6e71bdf5a9da15c60694fa8543243e37a32c5faf 1355Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1356Date: Mon Jun 23 11:25:41 2014 +0200 1357 1358 Allow/Fix use of multiple ZaphodHead outputs per x-screen. 1359 1360 Defining multiple ZaphodHead outputs per x-screen in a 1361 multiple x-screen's per gpu configuration caused all 1362 outputs except one per x-screen to go dark, because 1363 there was a fixed mapping x-screen number -> crtc number, 1364 limiting the number of crtc's per x-screen to one. 1365 1366 On a ZaphodHead's setup, be more clever and assign 1367 as many crtc's to a given x-screen as there are 1368 ZaphodHeads defined for that screen, assuming 1369 there are enough unused crtc's available. 1370 1371 Tested on a triple display setup with different combos 1372 of one, two or three ZaphodHeads per one, two or three 1373 x-screens. 1374 1375 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1376 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1377 1378commit fcf91a275e96f14e49a93dc3f8dd6798cf7d98f3 1379Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1380Date: Mon Jun 23 11:13:57 2014 +0200 1381 1382 dri2: Fix drmWaitVblank on multi-x-screen configs. 1383 1384 Old mapping of xf86Crtc -> display head index was 1385 too simple and only worked on a single x-screen config. 1386 1387 Introduce a new helper function drmmode_hw_head_index() 1388 which returns proper mapping also for multi x-screens per 1389 gpu and ZaphodHeads setups and other non-standard multi- 1390 display configurations. 1391 1392 Tested on triple-head setup with various ZaphodHead and 1393 server layout combos. 1394 1395 This helper will be also needed for dri3/present code. 1396 1397 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1398 1399commit 29572e44c24613be9dec0c98ff2731e46ce6203b 1400Author: Ilia Mirkin <imirkin@alum.mit.edu> 1401Date: Sun Aug 10 02:25:15 2014 -0400 1402 1403 exa/nv10: src ops should use the SF macro 1404 1405 Doesn't actually end up mattering. 1406 1407 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1408 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1409 1410commit 35e17905e1c54558059f6b72b8594a3377f1a57c 1411Author: Ilia Mirkin <imirkin@alum.mit.edu> 1412Date: Sun Aug 10 02:25:14 2014 -0400 1413 1414 exa/nv10: fix blend function setup to check the correct mask 1415 1416 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1417 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1418 1419commit dfac18c67a17782a0263daf8cb4ee487374abd2c 1420Author: Ilia Mirkin <imirkin@alum.mit.edu> 1421Date: Sun Aug 10 02:25:13 2014 -0400 1422 1423 exa/nv10: use color 0 when the pict format has no RGB component 1424 1425 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1426 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1427 1428commit ff7010e06b6da0fe27143b47b8a4ffd2f12894b7 1429Author: Ilia Mirkin <imirkin@alum.mit.edu> 1430Date: Sun Aug 10 02:25:12 2014 -0400 1431 1432 exa/nv10: use same clip settings as mesa driver 1433 1434 The higher 0x800 was getting overwritten by the 0x7ff anyways, so it 1435 wasn't doing any good. The mesa driver just uses 0x800 for the low 1436 portion and doesn't set the 8 bit in the higher portion, so do the same 1437 thing here. 1438 1439 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1440 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1441 1442commit edd1608c397ebcf487b84948031f7807178e5e4d 1443Author: Emil Velikov <emil.l.velikov@gmail.com> 1444Date: Tue Jul 1 22:31:47 2014 +0100 1445 1446 xorg_list: define the xorg_list* symbols 1447 1448 ...when building against pre 1.12 x. 1449 The build will fail as earlier versions of X were missing the xorg_ 1450 prefix of the symbols - struct and util functions. 1451 1452 Reported-by: Pali Rohár <pali.rohar@gmail.com> 1453 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1454 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1455 1456commit f6c1c755d8230494c8115e2c7899a85663f5c176 1457Author: Ben Skeggs <bskeggs@redhat.com> 1458Date: Tue Jul 1 12:56:49 2014 +1000 1459 1460 fix a harmless typo 1461 1462 fdo#80738 1463 1464 Reported-by: Paulo Sérgio Travaglia <pstglia@gmail.com> 1465 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1466 1467commit 096fb1ebc10b14ba1b747832efe378369bcf44e6 1468Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1469Date: Mon May 26 05:39:50 2014 +0200 1470 1471 dri2: Add support for handling more than 2 crtc's 1472 1473 Need to use the DRM_VBLANK_HIGH_CRTC bits to allow 1474 selecting crtc's with id > 1 on latest gpu's with 1475 up to four display engines. 1476 1477 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1478 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1479 1480commit 4771ff5c668047a2dc56e4c6131175eed837e757 1481Author: Marcin Slusarz <marcin.slusarz@gmail.com> 1482Date: Sun Jun 22 01:32:42 2014 +0200 1483 1484 present: build only when glamor is enabled 1485 1486 nouveau_present_flip_exec references glamor_fd_from_pixmap, 1487 which do not exist when glamor is disabled 1488 1489 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> 1490 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1491 1492commit 41c1c4ee867753422d6486f13bf20b45905a5dd5 1493Author: Marcin Slusarz <marcin.slusarz@gmail.com> 1494Date: Sun Jun 22 01:31:46 2014 +0200 1495 1496 glamor: fix build without glamor.h 1497 1498 xorg-server can be built without glamor, which leads to: 1499 1500 CC nouveau_xv.lo 1501 In file included from nouveau_xv.c:41:0: 1502 nouveau_glamor.h:12:20: fatal error: glamor.h: No such file or directory 1503 compilation terminated. 1504 1505 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> 1506 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1507 1508commit f0fa8313714c2a5b16e784b257b5ff79da3a443b 1509Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1510Date: Thu Jun 19 17:11:34 2014 +0200 1511 1512 dri2: Enable sync of bufferswaps to Vblank by default. 1513 1514 Make this consistent with other drivers default behaviour. 1515 1516 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1517 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1518 1519commit 02d7e647ce3a4a5d48c4ee94c94bebcac2d4827d 1520Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1521Date: Sat Jun 21 01:26:16 2014 +0200 1522 1523 dri2: Fix kms pageflip completion timestamping. 1524 1525 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1526 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1527 1528commit 4d92001ebe6bec8c5eaab8156421e62a7e4dedc8 1529Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1530Date: Fri Mar 14 18:27:53 2014 +0100 1531 1532 Add kernel async_flip cap detection. Part II of double-sync fix. 1533 1534 Query if kernel supports the async_flip cap, thereby needs the 1535 new sync behavior. Linux 3.13+ nouveau-kms have this cap and 1536 need this fix. 1537 1538 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1539 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1540 1541commit 54dfbb2fed393b65d846e6ed9672acb47e0f19de 1542Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1543Date: Fri Mar 14 18:23:04 2014 +0100 1544 1545 dri2: Fix double-sync of pageflips on Linux 3.13+ - Part I 1546 1547 Linux 3.13 and later sync kms pageflips to vblank in the kms 1548 driver, so we must not emit a sync to vblank pushbuf in the ddx 1549 on such kernels, or maximum framerate will be cut into half! 1550 1551 A sync-to-vblank-pushbuf is emitted for copyswaps as in the past, 1552 also for older kernels which don't support async_pageflip's and 1553 don't sync by themselves. 1554 1555 This adds the implementation, but not the detection logic for 1556 async_pageflip support in the kernel. 1557 1558 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1559 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1560 1561commit 303402e00d55b5296311738184cd61f4aadab74d 1562Author: Mario Kleiner <mario.kleiner.de@gmail.com> 1563Date: Sat Jun 21 00:09:39 2014 +0200 1564 1565 Fix nouveau_copy_init() 1566 1567 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 1568 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1569 1570commit 65a70c4edd1f0df808a3a4dea6e8c9846cc7c920 1571Author: Ben Skeggs <bskeggs@redhat.com> 1572Date: Fri Jun 20 16:57:18 2014 +1000 1573 1574 another build fix... 1575 1576commit 882ebb4000803f6025e1bb3213a31b19f6d1a2c6 1577Author: Ben Skeggs <bskeggs@redhat.com> 1578Date: Mon Nov 18 14:42:02 2013 +1000 1579 1580 present: initial support 1581 1582 Until glamor grows its own implementation. 1583 1584 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1585 1586commit 0d94f32fce2759c4b0f4d22b99f3ced09c2aa5a0 1587Author: Ben Skeggs <bskeggs@redhat.com> 1588Date: Fri Jun 20 09:47:39 2014 +1000 1589 1590 kms: implement a more generic event mechanism 1591 1592 Present wants to use this too. 1593 1594 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1595 1596commit 97feb3b2104287eb385ffba8329f304e2808d808 1597Author: Ben Skeggs <bskeggs@redhat.com> 1598Date: Fri Jun 20 09:29:08 2014 +1000 1599 1600 dri2: move page flip handling out of drmmode 1601 1602 Nothing to see here... This is to (hopefully) avoid busting DRI2 while 1603 implementing Present support. 1604 1605 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1606 1607commit 60b876eab3708f1c277027addb5514974452644c 1608Author: Ben Skeggs <bskeggs@redhat.com> 1609Date: Fri Jun 20 09:20:37 2014 +1000 1610 1611 dri2: move away from directly touching drmmode on page flips 1612 1613 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1614 1615commit 36dda1341008dd87a6b5459f0616776097d11d9f 1616Author: Ben Skeggs <bskeggs@redhat.com> 1617Date: Fri Jun 20 09:04:30 2014 +1000 1618 1619 dri2: shuffle some functions around 1620 1621 Nothing to see here... 1622 1623 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1624 1625commit 8bdec8ebd68ba5eaa696d8a666e18774bff3b2ad 1626Author: Ben Skeggs <bskeggs@redhat.com> 1627Date: Fri Jun 20 12:37:33 2014 +1000 1628 1629 glamor: fix randr resize 1630 1631 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1632 1633commit 3c45b590a31157abca140096b141aa83fecb671f 1634Author: Ben Skeggs <bskeggs@redhat.com> 1635Date: Fri Jun 20 08:28:30 2014 +1000 1636 1637 sync: fix build against DRI3-less server 1638 1639 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1640 1641commit 88ae58d786b0c41148e692dbfe8e2646e7533d9e 1642Author: Ilia Mirkin <imirkin@alum.mit.edu> 1643Date: Thu Jun 19 02:37:45 2014 -0400 1644 1645 update man page with new chips, AccelMethod option 1646 1647 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1648 1649commit 5cb36ba57493bf7edf560bf370c87eff79f1d1a6 1650Author: Ben Skeggs <bskeggs@redhat.com> 1651Date: Thu Jun 19 14:54:20 2014 +1000 1652 1653 default to glamor on maxwell 1654 1655 We have no RENDER/Xv acceleration in EXA for Maxwell. 1656 1657 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1658 1659commit 81148bb1dbc7007c021c59411d56cb31cfc74ef2 1660Author: Ben Skeggs <bskeggs@redhat.com> 1661Date: Thu Jun 19 14:54:20 2014 +1000 1662 1663 glamor: provide dri3 support when enabled 1664 1665 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1666 1667commit 86024ceef015ffe31a204cc5bc6c326a19363ff1 1668Author: Ben Skeggs <bskeggs@redhat.com> 1669Date: Thu Jun 19 14:54:20 2014 +1000 1670 1671 glamor: initial support (no dri) 1672 1673 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1674 1675commit fd0ce8839f307693d86c7602dd926ce79e6b777d 1676Author: Ben Skeggs <bskeggs@redhat.com> 1677Date: Thu Jun 19 14:54:20 2014 +1000 1678 1679 add support for maxwell, minus Xv/renderaccel 1680 1681 More extensive acceleration will be via glamor, at least to begin with. It 1682 seems to make sense to jump on the bandwagon now, and deal with any issues 1683 that arise before it it becomes the standard for XWayland. 1684 1685 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1686 1687commit 8ad0465e15bfc502abc259e886d6dcb448e0141a 1688Author: Ben Skeggs <bskeggs@redhat.com> 1689Date: Thu Jun 19 14:54:20 2014 +1000 1690 1691 50/c0/e0 -> tesla/fermi/kepler 1692 1693 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1694 1695commit 4b138ab18f58c6d459e21dc319615f536c8e69c8 1696Author: Ben Skeggs <bskeggs@redhat.com> 1697Date: Thu Jun 19 14:54:20 2014 +1000 1698 1699 merge nv_dma.c into nv_accel_common.c 1700 1701 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1702 1703commit 5c44e6b86195c22e092db5dae06e70208e8b66a0 1704Author: Ben Skeggs <bskeggs@redhat.com> 1705Date: Thu Jun 19 14:54:20 2014 +1000 1706 1707 pull copy engine handling into one place, a place that isn't also exa 1708 1709 As a side-effect, we now allow Kepler to use async copies too. 1710 1711 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1712 1713commit c375bacbdaeabd0288855c13b4717d0694be9e56 1714Author: Ben Skeggs <bskeggs@redhat.com> 1715Date: Thu Jun 19 14:54:20 2014 +1000 1716 1717 exa: move module load into nouveau_exa_init 1718 1719 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1720 1721commit b24cae0bf5db6ece21439d4c6ff3668aed4c78d6 1722Author: Ben Skeggs <bskeggs@redhat.com> 1723Date: Thu Jun 19 14:54:20 2014 +1000 1724 1725 dri2: move "is supported" checks out of nv_driver.c 1726 1727 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1728 1729commit 07c26a0670dd2558acf8e1159f2f68989a7f753d 1730Author: Ben Skeggs <bskeggs@redhat.com> 1731Date: Thu Jun 19 14:54:20 2014 +1000 1732 1733 replace NoAccel with AccelMethod in preparation for glamor support 1734 1735 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1736 1737commit e7cf48d9529160f7add2fef72d2e42be9370594a 1738Author: Ben Skeggs <bskeggs@redhat.com> 1739Date: Thu Jun 19 14:54:20 2014 +1000 1740 1741 silence a warning 1742 1743 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1744 1745commit 4a18dd74c0cc6e75937eb78347e8ea6c81d1373a 1746Author: Dave Airlie <airlied@redhat.com> 1747Date: Mon May 26 17:30:24 2014 +1000 1748 1749 autogen: add support for gnome autobuilder 1750 1751 Signed-off-by: Dave Airlie <airlied@redhat.com> 1752 1753commit 8604a741baa5360164a882bd9f3c8474840744db 1754Author: Hans de Goede <hdegoede@redhat.com> 1755Date: Fri Apr 11 14:20:43 2014 +0200 1756 1757 Fix building on older servers without xf86platformBus.h 1758 1759 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 1760 1761commit 0c13843e300d1a9db2bfed18dc1b5df38c5adfdf 1762Author: Hans de Goede <hdegoede@redhat.com> 1763Date: Mon Mar 17 17:37:31 2014 +0100 1764 1765 Add support for server managed fds 1766 1767 v2 (Martin Peres): 1768 - Do not define "NVEntPtr pNVEnt" if not needed to avoid some warnings 1769 1770 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 1771 Signed-off-by: Martin Peres <martin.peres@free.fr> 1772 1773commit 47f08aa6d713da4e6ae9c35ce691e790ffaed09d 1774Author: Hans de Goede <hdegoede@redhat.com> 1775Date: Mon Mar 17 17:37:30 2014 +0100 1776 1777 Add a NVOpenNouveauDevice helper function 1778 1779 This is a preparation patch for adding support for server managed fds 1780 1781 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 1782 Signed-off-by: Martin Peres <martin.peres@free.fr> 1783 1784commit 480f0998ffed6d9a5c6656dba75182f00fd88a1b 1785Author: Ben Skeggs <bskeggs@redhat.com> 1786Date: Thu Nov 7 14:56:48 2013 +1000 1787 1788 bump to 1.0.10 for release 1789 1790 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1791 1792commit 59395c738ebdc21087109075ada67ea8d7783141 1793Author: Ben Skeggs <bskeggs@redhat.com> 1794Date: Thu Nov 7 14:56:06 2013 +1000 1795 1796 recognise GK208 as a Kepler board 1797 1798 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1799 1800commit 03642323a954d0adb11662a98e92dfa20bee9728 1801Author: Dave Airlie <airlied@gmail.com> 1802Date: Wed Jul 31 11:19:50 2013 +1000 1803 1804 nouveau: fix build against older servers. 1805 1806 Older servers had no GPU support. 1807 1808 Signed-off-by: Dave Airlie <airlied@redhat.com> 1809 1810commit 27323ce68d58fcb82a648ab14abbc99acdc36f44 1811Author: Dave Airlie <airlied@gmail.com> 1812Date: Wed Jul 31 11:01:15 2013 +1000 1813 1814 nouveau: add gpu identifier to connector names for secondary gpus 1815 1816 this avoids clashes with identifiers from other GPUs. 1817 1818 Signed-off-by: Dave Airlie <airlied@redhat.com> 1819 1820commit 1df177f35a05db505577cdc929e63fde906a704b 1821Author: Dave Airlie <airlied@gmail.com> 1822Date: Wed Jul 31 10:51:03 2013 +1000 1823 1824 nouveau: fix build on ppc by wrapping immintrin include. 1825 1826 Signed-off-by: Dave Airlie <airlied@redhat.com> 1827 1828commit 300c5a32d79cf507b5fede96710335bcd316eccb 1829Author: Dave Airlie <airlied@redhat.com> 1830Date: Tue Jul 30 15:28:06 2013 +1000 1831 1832 bump to 1.0.9 for release 1833 1834 Signed-off-by: Dave Airlie <airlied@redhat.com> 1835 1836commit be44e7804862b4c276ed4d4717b1212920f428e6 1837Author: Dave Airlie <airlied@gmail.com> 1838Date: Tue Jul 30 15:26:46 2013 +1000 1839 1840 nouveau: fix crash when xinerama is enabled. 1841 1842 Signed-off-by: Dave Airlie <airlied@redhat.com> 1843 1844commit 588252556d1c84bf6793bc22edd48e6650a0f1b9 1845Author: Dave Airlie <airlied@gmail.com> 1846Date: Tue Jul 30 15:14:22 2013 +1000 1847 1848 fix make distcheck. 1849 1850 add all the nvf0 stuff. 1851 1852 Signed-off-by: Dave Airlie <airlied@redhat.com> 1853 1854commit 29aecda1b5bdb95700894083874492a6d5dfc08a 1855Author: Dave Airlie <airlied@gmail.com> 1856Date: Tue Jul 30 14:10:51 2013 +1000 1857 1858 drmmode: add support for multi-screen reverse optimus 1859 1860 Initial reverse optimus didn't consider multiple screens, so 1861 this overhauls the code to use the new X server interface, 1862 and allows for multiple outputs on the nvidia to be used with 1863 the intel doing the rendering. 1864 1865 Signed-off-by: Dave Airlie <airlied@redhat.com> 1866 1867commit d7ae886260b480da0ea4cad22eb6f7b7d12f07b0 1868Author: Ilia Mirkin <imirkin@alum.mit.edu> 1869Date: Mon Jul 29 21:03:54 2013 -0400 1870 1871 nouveau: use uppercase for chipset name 1872 1873 Right now the log will output e.g. NVaa instead of NVAA, which is how we 1874 usually talk about it. 1875 1876 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1877 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1878 1879commit 4492cf90c2a7fe27badf6f3f0f296755a65e1e83 1880Author: Ilia Mirkin <imirkin@alum.mit.edu> 1881Date: Mon Jul 29 02:40:39 2013 -0400 1882 1883 xv: speed up YV12 -> NV12 conversion using SSE2 if available 1884 1885 memcpy() goes from taking 45% to 66% of total function time, which 1886 translates to a 30% decrease in NVPutImage runtime. 1887 1888 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1889 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1890 1891commit 08c667b06d28756d0d32f82bd736b0bdf2b4448e 1892Author: Ilia Mirkin <imirkin@alum.mit.edu> 1893Date: Mon Jul 29 02:40:38 2013 -0400 1894 1895 xv: fix last pixel for big-endian machines in YV12 -> NV12 conversion 1896 1897 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1898 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1899 1900commit 8c1c5d4f212958559132506d51dcfeeb17f60371 1901Author: Ben Skeggs <bskeggs@redhat.com> 1902Date: Wed May 15 21:34:13 2013 +1000 1903 1904 nvf0: implement EXA and XVideo acceleration 1905 1906 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1907 1908commit 779b8d44b5ff84b87b52600a2f010ac19cd396ad 1909Author: Sven Joachim <svenjoac@gmx.de> 1910Date: Thu Jun 13 20:41:53 2013 +0200 1911 1912 Link against libdrm 1913 1914 The DDX uses symbols from libdrm but currently does not link against 1915 it, but rather relies on indirect linkage from libdrm_nouveau. This 1916 could lead to insufficient dependencies in distribution packages, seen 1917 in http://bugs.debian.org/633786. 1918 1919 Signed-off-by: Sven Joachim <svenjoac@gmx.de> 1920 1921commit e96ae2b129df2a50d1d3839084df4fb1868d914d 1922Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1923Date: Wed Jun 12 11:05:23 2013 +0200 1924 1925 fixup the dri2 check 1926 1927 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1928 1929commit 7f20a6af25ae28452ec9c971bfa2227857e6ee93 1930Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1931Date: Wed Jun 12 10:46:39 2013 +0200 1932 1933 bump to 1.0.8 1934 1935commit 1fda669c0e5a25eac7f9f1e9a743120cae50ff51 1936Author: Emil Velikov <emil.l.velikov@gmail.com> 1937Date: Sat Feb 16 20:48:47 2013 +0000 1938 1939 configure: printout the configuration info 1940 1941 Let the bikeshedding begin 1942 1943 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1944 1945commit 45dbcaac301efce04d0bb492337f4febf8129ed9 1946Author: Emil Velikov <emil.l.velikov@gmail.com> 1947Date: Sat Feb 16 20:48:45 2013 +0000 1948 1949 nouveau: mandate dri2 build 1950 1951 Building nouveau without dri2 is just silly 1952 1953 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1954 1955commit ea8d225fe100ecabb72e3cc0a92372f16afb10b3 1956Author: Emil Velikov <emil.l.velikov@gmail.com> 1957Date: Sat Feb 16 20:48:44 2013 +0000 1958 1959 dri1: purge the final references 1960 1961 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1962 1963commit b5d1a9bd269c9d2758fee37492cdcccc976fcc3b 1964Author: Emil Velikov <emil.l.velikov@gmail.com> 1965Date: Sat Feb 16 20:48:43 2013 +0000 1966 1967 nouveau: Do not load dri {sub,}module 1968 1969 Nouveau does not use dri1, thus loading XFree86-DRI is not needed 1970 On the other hand, we do use dri2 1971 1972 As a side effect, purge the 'set-but-unused' variable pLibDRMVersion 1973 1974 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1975 1976commit 7c3f6cd8d6b14aea561fb877f6006d339d156f35 1977Author: Emil Velikov <emil.l.velikov@gmail.com> 1978Date: Sat Feb 16 20:48:42 2013 +0000 1979 1980 nouveau: stop using dri1 function DRICreatePCIBusID 1981 1982 Step 1 to completely rip out dri1 out of nouveau 1983 1984 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1985 1986commit 3b1a45325d5fd46f0e7fad17168742b4c9a2465a 1987Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1988Date: Wed Jun 12 10:30:32 2013 +0200 1989 1990 update makefiles for make distcheck 1991 1992commit b82aacd3e72b52240181569c8b0abb55c1f716d5 1993Author: Emil Velikov <emil.l.velikov@gmail.com> 1994Date: Sat Feb 16 12:48:46 2013 -0800 1995 1996 configure: require xorg-macros 1.8 1997 1998 Explicitly check for xorg-macros 1.8, as a subtle hint 1999 to update your build deps if building your own ddx 2000 2001 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 2002 2003commit 4f667b40f41e311fbe302237efe173bcc24fb8bb 2004Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2005Date: Mon Jun 10 14:29:59 2013 +0200 2006 2007 fix a xfree from previous patch 2008 2009 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2010 2011commit 2e34d11541e1f987c7ca0556592cb16f4d232299 2012Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2013Date: Mon Jun 10 13:06:35 2013 +0200 2014 2015 nouveau: enable support for normal optimus configuration 2016 2017 nvd7 has no display engine, only enable sink output capability 2018 if there is a crtc. 2019 2020 drmmode doesn't work without crtc's, so prevent crashes by skipping 2021 the drmmode code in that case. 2022 2023 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2024 2025commit bf72ae1f6574c540f0afc2d7845d41df43507a8f 2026Author: Ben Skeggs <bskeggs@redhat.com> 2027Date: Wed May 15 22:05:01 2013 +1000 2028 2029 nve0: envyas isn't overly clever.. avoid short-form instructions 2030 2031 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2032 2033commit 41cc2f24c43d2656ac722f594aaebad93e37dfec 2034Author: Ben Skeggs <bskeggs@redhat.com> 2035Date: Wed May 15 21:13:11 2013 +1000 2036 2037 nvc0-nve0: decompile shaders into their source, add basic build system 2038 2039 Far, far easier to maintain this way... 2040 2041 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2042 2043commit 1edb5d7ae450a32fe6607d536653cf733444e7e2 2044Author: Ben Skeggs <bskeggs@redhat.com> 2045Date: Wed May 15 20:47:44 2013 +1000 2046 2047 nvc0-nve0: split shader code from their headers 2048 2049 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2050 2051commit c18439f759bec267a3b67d375dcbfeb8a6fd7b6a 2052Author: Dave Airlie <airlied@redhat.com> 2053Date: Mon May 13 13:55:51 2013 +1000 2054 2055 nouveau: clean up nv entity index declarations. 2056 2057 Signed-off-by: Dave Airlie <airlied@redhat.com> 2058 2059commit d3b52efe959f255784f5ead16d7276ca0fb4cdb1 2060Author: Dave Airlie <airlied@redhat.com> 2061Date: Mon May 13 13:35:12 2013 +1000 2062 2063 nouveau: attempt to fix zaphod since dri1 code removal 2064 2065 j_v on #nouveau bisected b1a630b48210d6a3c44994fce1b73273000ace5c has 2066 breaking zaphod, on review it was trying to open the drm fd a second time 2067 which was unnecessary. 2068 2069 Avoid the problem by storing the nv fd in an entity and have share it between 2070 the two scrn info recs. 2071 2072 Signed-off-by: Dave Airlie <airlied@redhat.com> 2073 2074commit 50426ac83f9f077791e4df8c04fc0ecefc06e8ab 2075Author: Marek Vasut <marex@denx.de> 2076Date: Fri May 3 15:48:55 2013 +0200 2077 2078 nouveau: Allow 3D accelerators with output ports 2079 2080 This patch let's cards with PCI class 0x30200 (3D controller) be 2081 operated by the nouveau driver as well. The nv Quadro NVS 450 is 2082 one such card, where the first GPU has PCI class 0x30000 and the 2083 other GPU has PCI class 0x30200. By ignoring the 1 << 9 bit in 2084 the PCI class mask, displays attached to the other GPU can also 2085 be used now. 2086 2087 Signed-off-by: Marek Vasut <marex@denx.de> 2088 Cc: Dave Airlie <airlied@redhat.com> 2089 Cc: Ben Skeggs <skeggsb@gmail.com> 2090 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2091 2092commit 2fa3397e348161a3394e2b456f065921272a056a 2093Author: Ilia Mirkin <imirkin@alum.mit.edu> 2094Date: Fri May 3 09:59:39 2013 -0400 2095 2096 nouveau_xv: Avoid reading off the end of the source image on NV50+ 2097 2098 The 'w' argument to NVCopyNV12ColorPlanes is used to index into the 2099 source image. line_len is rounded up to 8 on NV50+, so if the source 2100 image (+ left offset) is not rounded to 8, NVCopyNV12ColorPlanes could 2101 read past the end of the array and crash X. This change can cause the 2102 last few horizontal pixels of dst to not be initialized, but they 2103 should be truncated by the renderer anyways. 2104 2105 See https://bugs.freedesktop.org/show_bug.cgi?id=63263 2106 2107 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 2108 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2109 2110commit 81f9ab38f0a5de7217227dd17a4d0e50dffd0d28 2111Author: Ben Skeggs <bskeggs@redhat.com> 2112Date: Thu May 2 11:38:07 2013 +1000 2113 2114 nvf0: recognise as supported chipset 2115 2116 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2117 2118commit 653c95d60af0f7c19a1d2f3b76dd17bb169d9c9b 2119Author: Ben Skeggs <bskeggs@redhat.com> 2120Date: Mon Apr 29 10:07:59 2013 +1000 2121 2122 nve0: explicitly ask kernel for copy2 object 2123 2124 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2125 2126commit 6771424d79e541d2fa7253a582db3dc9108fd97d 2127Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2128Date: Wed Mar 27 09:50:03 2013 +0100 2129 2130 bump to 1.0.7 2131 2132 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2133 2134commit 84998320162a74a0861b7be1fcc230e50f08424a 2135Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2136Date: Mon Mar 25 13:55:37 2013 +0100 2137 2138 Clean up some errors on closing. 2139 2140 If forced close happens, all ioctl's will fail. Some of the handlers 2141 also need to be unregistered before the module is unloaded entirely. 2142 2143commit a80785f79268ed3701fc69cbd9194ad12822c82c 2144Author: Dave Airlie <airlied@redhat.com> 2145Date: Mon Feb 11 09:56:25 2013 +1000 2146 2147 nouveau: fix build against old servers (part 2) 2148 2149 Should fix next bit of 2150 https://bugs.freedesktop.org/show_bug.cgi?id=60369 2151 2152 Signed-off-by: Dave Airlie <airlied@redhat.com> 2153 2154commit 05dc9b57896689a3dbd9e6aeb16d57c49515b02e 2155Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2156Date: Sun Feb 3 21:12:46 2013 +0100 2157 2158 nv30/exa: fix shaders on big-endian machines 2159 2160 Direct port of commit d1bc38b6673c54af61196056c489383fba8dced8 2161 "nv40/exa: fix shaders on big-endian machines" to nv30. 2162 2163 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=60050 2164 2165 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2166 2167commit 61a277488f7195914cbdda10355db0d758fc801c 2168Author: Dave Airlie <airlied@redhat.com> 2169Date: Thu Feb 7 10:47:20 2013 +1000 2170 2171 nouveau: fix build on older X servers 2172 2173 should fix https://bugs.freedesktop.org/show_bug.cgi?id=60369 2174 2175 Reported-by: fabio.ped@libero.it 2176 Signed-off-by: Dave Airlie <airlied@redhat.com> 2177 2178commit e8f222fd10e21f39488d3d8c697e4948cd787cd0 2179Author: Dave Airlie <airlied@redhat.com> 2180Date: Wed Feb 6 16:21:50 2013 +1000 2181 2182 nouveau: make prime regression fix more robust. 2183 2184 This makes this fix more robust and less crashy. 2185 2186 Signed-off-by: Dave Airlie <airlied@redhat.com> 2187 2188commit d762631c9306b6580b34db1e7eb57bbcac901390 2189Author: Dave Airlie <airlied@redhat.com> 2190Date: Wed Feb 6 15:52:33 2013 +1000 2191 2192 nouveau: align shared buffers to 256 2193 2194 We were getting 0xa5 error code on 2D transfers from tiled VRAM to untiled 2195 GART when using USB offload devices, this fixes this by upping the alignment 2196 on the 2D engine for shared buffers. 2197 2198 Signed-off-by: Dave Airlie <airlied@redhat.com> 2199 2200commit c6980b2add5752a7c285a846b2c9353c9d18afd4 2201Author: Dave Airlie <airlied@redhat.com> 2202Date: Wed Feb 6 15:24:09 2013 +1000 2203 2204 nouveau: fix regression in PRIME since 58284cf32778d54cde139423450bc33360784503 2205 2206 Signed-off-by: Dave Airlie <airlied@redhat.com> 2207 2208commit 59fb3f0252e9b769c5877cfe83d8dbc51cdff232 2209Author: Dave Airlie <airlied@redhat.com> 2210Date: Wed Feb 6 14:58:08 2013 +1000 2211 2212 nouveau: create shared pixmaps in GART 2213 2214 this creates the shared pixmaps in GART, and makes sure they are untiled. 2215 2216 Signed-off-by: Dave Airlie <airlied@redhat.com> 2217 2218commit 1fdd7db94b55c65ea62cc9eaefff620b20e9e4ea 2219Author: Dave Airlie <airlied@redhat.com> 2220Date: Mon Jan 7 15:28:53 2013 +1000 2221 2222 nouveau: add reverse prime support 2223 2224 This allows the nvidia card to scanout Intel cards rendering. 2225 2226 Signed-off-by: Dave Airlie <airlied@redhat.com> 2227 2228commit 58284cf32778d54cde139423450bc33360784503 2229Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 2230Date: Tue Jan 29 16:38:21 2013 +1000 2231 2232 dri2: fix missing frame throttle on copy_region buffer swaps 2233 2234 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2235 2236commit 912d418fdfd2e99eef1e5c631c76dda1d82cf451 2237Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2238Date: Mon Jan 21 00:15:38 2013 +0100 2239 2240 nouveau: put cursor BO in GART 2241 2242 Keeping it in VRAM wastes CPU time, because cursor_set ioctl reads 2243 handed BO back to RAM, just to write it to actual cursor BO. 2244 2245 Here (nv92/core i7), this patch decreases overall cpu usage of 2246 drmmode_load_cursor_argb from 4.6ms to ~90us and lets nv50_crtc_cursor_set 2247 disappear from perf traces. 2248 2249 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2250 2251commit 8f934fad5d4934936b3265f272ca81e73c60d7dc 2252Author: Dave Airlie <airlied@redhat.com> 2253Date: Mon Jan 7 14:13:10 2013 +1000 2254 2255 bump to 1.0.6 2256 2257 Signed-off-by: Dave Airlie <airlied@redhat.com> 2258 2259commit 00834273e22f7234a3755269100ea3f10b521e35 2260Author: Dave Airlie <airlied@redhat.com> 2261Date: Mon Jan 7 14:04:40 2013 +1000 2262 2263 nouveau: fix dirty redisplay damage handling 2264 2265 This processes the damage in the correct order. 2266 2267 Signed-off-by: Dave Airlie <airlied@redhat.com> 2268 2269commit 389b8a8dd91a48dee524eac96e60813c507f1a82 2270Author: Ben Skeggs <bskeggs@redhat.com> 2271Date: Mon Jan 7 14:05:06 2013 +1000 2272 2273 bump to 1.0.5 2274 2275 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2276 2277commit efe9c57af6ff4c12936e49ac3e188c54b8b2d5bd 2278Author: Ben Skeggs <bskeggs@redhat.com> 2279Date: Mon Jan 7 11:35:56 2013 +1000 2280 2281 nvc0-/exa: fix typo which caused breakage in rendercheck tmcoords test 2282 2283 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2284 2285commit 5cd86445272818ec2f598f4685a5d7f24e403d07 2286Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2287Date: Sun Nov 25 22:59:49 2012 +0100 2288 2289 add some missing newlines in error messages 2290 2291 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2292 2293commit 3b33a0b426b8915f302247dcd40c64fb1a6dcdd2 2294Author: Emil Velikov <emil.l.velikov@gmail.com> 2295Date: Sun Nov 4 18:06:18 2012 +0000 2296 2297 kms: Let the server know if we got more than 1 EDID block 2298 2299 Verbatim copy from xf86-video-radeon commit 809cacb 2300 2301 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=32341 2302 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 2303 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2304 2305commit 018ab2d4634f3c8923038337ce9600973e2f0002 2306Author: Ben Skeggs <bskeggs@redhat.com> 2307Date: Fri Nov 9 12:16:00 2012 +1000 2308 2309 bump version to 1.0.4 2310 2311 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2312 2313commit 0d32d964dbba3864af0946aa8894cef858eb1319 2314Author: Dave Airlie <airlied@redhat.com> 2315Date: Wed Nov 7 14:49:03 2012 +1000 2316 2317 nouveau: drop unused nv private members. 2318 2319 these are no longer needed. 2320 2321 Signed-off-by: Dave Airlie <airlied@redhat.com> 2322 2323commit b1a630b48210d6a3c44994fce1b73273000ace5c 2324Author: Dave Airlie <airlied@redhat.com> 2325Date: Wed Nov 7 14:45:14 2012 +1000 2326 2327 nouveau: drop DRI1 device open interface. 2328 2329 this is a DRI1 left over, the driver is meant to open the device itself now. 2330 2331 this old interface did crazy stuff like drmAddMap. 2332 2333 Reviewed-by: Ben Skeggs <bskeggs@redhat.com> 2334 Signed-off-by: Dave Airlie <airlied@redhat.com> 2335 2336commit 27a1a0616304e9b9f0ae842899b7d614f1026578 2337Author: Emil Velikov <emil.l.velikov@gmail.com> 2338Date: Fri Nov 2 03:57:42 2012 +0000 2339 2340 nouveau: NVPlatformProbe, handle shareable entities 2341 2342 During NVPciProbe, xf86ConfigPciEntity() implicitly calls checks and sets 2343 the entity as shared 2344 Here we are only allocating the screen, thus need to explicitly call the 2345 functions 2346 2347 ZaphodHead and Xinerama should finally work 2348 2349 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=56347 2350 2351 Reported-By: Vladimir Chebotarev <exc@rbcmail.ru> 2352 Reported-By: Damian Nowak <nowaker@geozone.pl> 2353 Tested-By: Damian Nowak <nowaker@geozone.pl> 2354 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 2355 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2356 2357commit e34cfbd5bd23f7f15372af52d8a39a5715ce7310 2358Author: Emil Velikov <emil.l.velikov@gmail.com> 2359Date: Fri Nov 2 03:57:41 2012 +0000 2360 2361 nouveau: Factor out common code to NVHasKMS() 2362 2363 As the name suggests checks if it has kernel mode setting, 2364 prints out the interface version and checkes if the chipset 2365 is supported 2366 2367 Function is used in NVPciProbe and NVPlatformProbe 2368 2369 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 2370 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2371 2372commit 78f24dc3798d635fca2e20856a09012c28bbbb8b 2373Author: Emil Velikov <emil.l.velikov@gmail.com> 2374Date: Fri Nov 2 03:57:40 2012 +0000 2375 2376 nouveau: introduce NVInitScrn() 2377 2378 Common/helper function for NV{Pci,Platform}Probe() 2379 2380 This introduces some the following functions to the 2381 NVPlatformProbe codepath 2382 * xf86SetEntitySharable() 2383 * xf86SetEntityInstanceForScreen() 2384 2385 Necessary for dualhead, xinerama and/or zaphodhead 2386 2387 v2: Rename nouveau_init_scrn() to NVInitScrn() 2388 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 2389 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2390 2391commit 47457fbd5006070ea9a06238a5a69ca8167dac77 2392Author: Emil Velikov <emil.l.velikov@gmail.com> 2393Date: Fri Nov 2 03:57:39 2012 +0000 2394 2395 nouveau: Remove unneeded xf86GetEntityInfo() 2396 2397 Retrieving the whole entity information is not needed as we only 2398 use the index - stored already in entity_num 2399 2400 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 2401 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2402 2403commit 20995bb5920021668b8b607f886201c643ee0e9a 2404Author: Dave Airlie <airlied@redhat.com> 2405Date: Fri Nov 2 09:04:43 2012 +1000 2406 2407 nouveau: fix regression in virtualbox drawing hopefully. 2408 2409 Signed-off-by: Dave Airlie <airlied@redhat.com> 2410 2411commit 8c3e1623b0be15f8cc590d893bfd19be87bd079a 2412Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2413Date: Wed Oct 24 11:19:04 2012 +0200 2414 2415 bump version to 1.0.3 2416 2417commit 6644a3a724a11150eba0c0082c8d4fe167b1755e 2418Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2419Date: Tue Oct 23 20:17:29 2012 +0200 2420 2421 exa: remove exaMarkSync from download/upload screen 2422 2423 They're noops on nouveau since sync markers are unused and 2424 download/upload are always synchronous anyway from exa's POV. 2425 2426 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2427 2428commit 2a757b7dffa85c30879653a1e52e33b8f4fa8806 2429Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2430Date: Tue Oct 23 17:04:37 2012 +0200 2431 2432 exa: fix comparison in download_from_screen 2433 2434 Presumably a copy/paste error from upload_to_screen. 2435 2436 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2437 2438commit 72230cd5a7846f7a26f929262a745dc0bdc34a3d 2439Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2440Date: Tue Oct 23 10:37:27 2012 +0200 2441 2442 exa: fix up download_from/upload_to screen 2443 2444 Add a missing return TRUE to download_from_screen. 2445 Add a missing exaMarkSync to memcpy in upload_to_screen. 2446 Add a really loud warning to both if memcpy to/from tiled is done. 2447 2448 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2449 2450commit 1d7890ecb8b3188177ab85608ea232812672d923 2451Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2452Date: Tue Oct 23 09:53:39 2012 +0200 2453 2454 remove unused members from nv_type.h 2455 2456commit 0b2a548316b0ce88ba19d5e05fdca205fc139962 2457Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2458Date: Mon Oct 22 13:12:06 2012 +0200 2459 2460 nvc0/exa: make solid fill actually work 2461 2462 Some investigation showed that the code for accelerating solid fill 2463 didn't actually work because NVC0EXACheckTexture would fail. 2464 2465 Making the check the same as in NV50EXACheckTexture fixes this, and 2466 seems to get acceleration for solid fills working right. 2467 2468 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2469 2470commit b4231dd715a8a7f86b04519b5f4e8a8d93c2f561 2471Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2472Date: Tue Oct 9 09:06:59 2012 +0200 2473 2474 dri2: Fix potential race and crash for swap at next vblank. 2475 2476 This fixes a potential race + crash that wasn't properly 2477 handled by commit 248de8cdbd6d0bc062633b49896fa4791148cd3b 2478 and happened at least on one users machine. 2479 2480 That commit wrongly assumed no special action would be needed 2481 for swaps at next vblank while triple-buffering is enabled on 2482 XOrg server 1.12 or later. 2483 2484 Closer inspection of the x-server main dispatch loop shows 2485 it is possible that the client manages to get the server 2486 to dispatch a new DRI2GetBuffersWithFormat() call before 2487 the server calls the vblank event handler and executes 2488 the nouveau_dri2_finish_swap() routine. Such a race would 2489 cause a crash, as described in above commit. 2490 2491 This commit handles the "swap at next vblank" case by 2492 calling nouveau_dri2_finish_swap() immediately without 2493 the roundtrip (queue vblank_event -> kernel -> deliver event 2494 -> x-server processes event -> nouveau vblank event handler), 2495 before control gets returned to the client. 2496 2497 This avoids the race while retaining triple-buffering. As 2498 a bonus, time-critical swaps at next vblank get processed 2499 without roundtrip delay, increasing the chance of not 2500 skipping a frame due to vblank miss while sync to vblank is 2501 on. 2502 2503 Thanks to Anssi for reporting this problem on the nouveau 2504 mailing list at 12th July 2012 and for testing this patch. 2505 2506 Reported-by: Anssi Hannula <anssi.hannula@iki.fi> 2507 Tested-by: Anssi Hannula <anssi.hannula@iki.fi> 2508 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2509 2510commit 01c9a9cff7d6232ca6eafa59dd60833a0d9cdc5b 2511Author: Ben Skeggs <bskeggs@redhat.com> 2512Date: Thu Oct 18 18:38:21 2012 +1000 2513 2514 shadowfb: fix segfault due to reading outside of shadow buffer 2515 2516 Probably caused by the new libdrm port, the new libdrm sets the bo 2517 size field to the *allocated* size and not the *requested* size, 2518 making the max_height calculation here invalid. 2519 2520 Switched to using virtualX/virtualY as the bounds, which should 2521 hopefully do the right thing.. 2522 2523 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2524 2525commit 297fd0d0755bda698be1d0b30cc60a41d7673c0b 2526Author: Dave Airlie <airlied@redhat.com> 2527Date: Tue Oct 16 16:15:16 2012 +1000 2528 2529 nouveau/dri2: fix pixmap/window offset calcs. 2530 2531 This should fix prime rendering under kwin, and not break it under the 2532 others. 2533 2534 Signed-off-by: Dave Airlie <airlied@redhat.com> 2535 2536commit 5c9379b14cdabc81fd5d4c916dcd93dc77c4f683 2537Author: Sebastian Keller <sebastian-keller@gmx.de> 2538Date: Tue Sep 25 11:35:35 2012 +0200 2539 2540 exa: use CLAMP_TO_EDGE for RepeatPad 2541 2542 This fixes border rendering in some gtk3 themes. 2543 https://bugs.freedesktop.org/show_bug.cgi?id=55310 2544 2545 v2 (Ben Skeggs): implement same fix for fermi/kepler too 2546 2547 Signed-off-by: Sebastian Keller <sebastian-keller@gmx.de> 2548 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2549 2550commit 302b5da84b4890c067f5f7e8b0f4e689909bbb51 2551Author: Adam Jackson <ajax@redhat.com> 2552Date: Tue Sep 25 10:44:32 2012 -0400 2553 2554 Remove mibstore.h 2555 2556 Signed-off-by: Adam Jackson <ajax@redhat.com> 2557 2558commit 20f3830ddf868c0d1a46c18fa381585c076bfb8d 2559Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 2560Date: Wed Sep 12 13:42:43 2012 +0200 2561 2562 bump version to 1.0.2 2563 2564commit 239f39977ebb64c9d545c6f3c905de3c0ad840d0 2565Author: Dave Airlie <airlied@redhat.com> 2566Date: Thu Jul 19 13:46:44 2012 +1000 2567 2568 nouveau: enable provider/capabilities 2569 2570 this adds the provider object for nouveau and enables the capabilities 2571 on it. 2572 2573 Signed-off-by: Dave Airlie <airlied@redhat.com> 2574 2575commit 28abba276e36fd0176f416996d44297ddf13f5cd 2576Author: Dave Airlie <airlied@redhat.com> 2577Date: Thu Jul 19 13:46:12 2012 +1000 2578 2579 nouveau/dri2: add support for DRI2 offload 2580 2581 This adds the nouveau interfaces for DRI2 offload, this allows nouveau 2582 to act as an offload slave. 2583 2584 Signed-off-by: Dave Airlie <airlied@redhat.com> 2585 2586commit b0be0ac86bae871c9f1724c40298d50d7f2b08e4 2587Author: Dave Airlie <airlied@redhat.com> 2588Date: Thu Jul 19 13:45:32 2012 +1000 2589 2590 nouveau: add pixmap tracking/updating support. 2591 2592 This should allow nouveau to act as an output master for driving output 2593 slaves. 2594 2595 Signed-off-by: Dave Airlie <airlied@redhat.com> 2596 2597commit baf485b1d304d741073f058d524ca698af3c15cd 2598Author: Dave Airlie <airlied@redhat.com> 2599Date: Thu Jul 19 13:42:13 2012 +1000 2600 2601 nouveau/exa: add support for shared pixmaps. 2602 2603 This adds the pixmap export/import hooks, along with code for nv50/nvc0 2604 to validate shared buffers in GTT space. 2605 2606 Signed-off-by: Dave Airlie <airlied@redhat.com> 2607 2608commit 2f48b8f462a03cb92db9e9a7ae1957eb27473965 2609Author: Dave Airlie <airlied@redhat.com> 2610Date: Thu Jul 19 13:40:23 2012 +1000 2611 2612 nouveau: add platform bus support 2613 2614 This adds the ability to load nouveau via the new platform apis. 2615 2616 Signed-off-by: Dave Airlie <airlied@redhat.com> 2617 2618commit 3aa9a296f508089a247f92b829036ca6a55b7b42 2619Author: Adam Jackson <ajax@redhat.com> 2620Date: Wed Aug 15 13:26:51 2012 -0400 2621 2622 Remove vbe and int10 includes 2623 2624 This driver has never used int10 or vbe, presumably these were just 2625 copypasta from nv. 2626 2627 Signed-off-by: Adam Jackson <ajax@redhat.com> 2628 2629commit 9b9be11e99930136a591059db1bf4ceaca8806e7 2630Author: Adam Jackson <ajax@redhat.com> 2631Date: Thu Jul 19 18:22:44 2012 -0400 2632 2633 Implement ->driverFunc 2634 2635 Copied from fbdev, makes it so we can run without iopl. 2636 2637 Signed-off-by: Adam Jackson <ajax@redhat.com> 2638 2639commit 6629066b87142cdd1d9e881b61ef55c97aedf949 2640Author: Ben Skeggs <bskeggs@redhat.com> 2641Date: Fri Aug 3 17:05:33 2012 +1000 2642 2643 nvc0/nve0: support sync-to-vblank if kernel is new enough 2644 2645 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2646 2647commit d1bc38b6673c54af61196056c489383fba8dced8 2648Author: Ben Skeggs <bskeggs@redhat.com> 2649Date: Fri Jul 6 16:23:50 2012 +1000 2650 2651 nv40/exa: fix shaders on big-endian machines 2652 2653 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2654 2655commit 0d2b414aff634c996e0468bb0df1894d6c28c09e 2656Author: Ben Skeggs <bskeggs@redhat.com> 2657Date: Tue Jun 19 10:12:37 2012 +1000 2658 2659 bump version to 1.0.1 2660 2661commit f84c28c9e6a0079e527655994689cb9c2ba51999 2662Author: Ben Skeggs <bskeggs@redhat.com> 2663Date: Tue Jun 19 09:49:02 2012 +1000 2664 2665 exa: fix remaining signed datatype issues 2666 2667 Wasn't expecting this situation, apparently it happens.. 2668 2669 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2670 2671commit 794141f22179a09ba6b2a094ba72316c298fee8b 2672Author: Maarten Maathuis <madman2003@gmail.com> 2673Date: Tue Jun 19 00:28:59 2012 +0200 2674 2675 exa: fix various issues in transfer bo handling 2676 2677 Thanks to "Christoph Bumiller <e0425955@student.tuwien.ac.at>" for hinting 2678 at the cause of the memory leak. 2679 2680 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 2681 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2682 2683commit 661d0cf42219a063f8faa0518f3490ddd38209ff 2684Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 2685Date: Mon Jun 18 17:13:55 2012 +0200 2686 2687 nve0: fix and improve NV12 shader 2688 2689 Only the 1st set of coordinates is valid. 2690 NVC0 was fixed in 9a3e579f637267b32efc46cfce5d1a36a41323b1. 2691 2692commit a907efe44177298cb1c42a236aab087c14b71657 2693Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 2694Date: Mon Jun 18 13:50:27 2012 +0200 2695 2696 nvc0/accel: oops fix method size in previous patch 2697 2698commit 8805e20bb9c1cfb7477f3eb7def64c5711fc91ea 2699Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 2700Date: Mon Jun 18 13:41:43 2012 +0200 2701 2702 nvc0/accel: use correct (signed) vertex data type in PUSH_VTX1/2s 2703 2704 Could have kept s16 but decided against mask and shift. 2705 2706commit 6dbf4ea12600275775123f9f564469454415da55 2707Author: Ben Skeggs <bskeggs@redhat.com> 2708Date: Mon Jun 18 13:28:07 2012 +1000 2709 2710 bump version to 1.0.0 2711 2712 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2713 2714commit 5e74cee90eeb4e8e638b371a5f1a2c6737be70ff 2715Author: Ben Skeggs <bskeggs@redhat.com> 2716Date: Mon Jun 18 15:36:47 2012 +1000 2717 2718 distcheck fixes 2719 2720commit 383aef5a8882a65efdf34e8d2681cd5ccec51b22 2721Author: Ben Skeggs <bskeggs@redhat.com> 2722Date: Mon Jun 18 13:31:37 2012 +1000 2723 2724 exa: drop massive gart allocation in favour of smaller ones, as needed 2725 2726 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2727 2728commit 52836c15b1c0e4f28d92ca9e6c76872481917fd4 2729Author: Ben Skeggs <bskeggs@redhat.com> 2730Date: Mon Jun 18 13:20:18 2012 +1000 2731 2732 nv50/exa: somehow fix firefox tabs not being drawn properly since vp xfrm 2733 2734 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2735 2736commit 36d3f8c28c55c05ccf4efa985831d2d249bad2fe 2737Author: Ben Skeggs <bskeggs@redhat.com> 2738Date: Wed Jun 13 22:41:58 2012 +1000 2739 2740 dri2: kill debug spam 2741 2742 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2743 2744commit 4dbc132f22721e3da30eb2e7fc97dea5b8458df6 2745Author: Viktor Novotný <noviktor@seznam.cz> 2746Date: Sat May 26 22:15:20 2012 +0200 2747 2748 nv04/exa: Reset destination surface offset in the same call of NV04EXACopy 2749 2750 Fixes FDO bug #48954. 2751 2752 Signed-off-by: Viktor Novotný <noviktor@seznam.cz> 2753 Reviewed-by: Marcin Slusarz <marcin.slusarz@gmail.com> 2754 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2755 2756commit aab58ee9a8025422416458d637c203af5ba008fd 2757Author: Dave Airlie <airlied@redhat.com> 2758Date: Tue Jun 5 12:34:06 2012 +0100 2759 2760 nouveau: i is used inside the function 2761 2762 fixes build, reported by tallica on irc. 2763 2764 Signed-off-by: Dave Airlie <airlied@redhat.com> 2765 2766commit 619e99731f772ff8d9d93cd1d6d83de5d450574b 2767Author: Dave Airlie <airlied@redhat.com> 2768Date: Tue Jun 5 11:02:54 2012 +0100 2769 2770 nouveau: port to compat server API. 2771 2772 This ports to the new server API. 2773 2774 Signed-off-by: Dave Airlie <airlied@redhat.com> 2775 2776commit d2e16c62b5054f34a439e504bfa261fb71f5d7de 2777Author: Dave Airlie <airlied@redhat.com> 2778Date: Tue Jun 5 10:57:30 2012 +0100 2779 2780 nouveau: drop flags arg to adjust frame 2781 2782 Signed-off-by: Dave Airlie <airlied@redhat.com> 2783 2784commit fadf83d7b373282ccbf0fa0c01928a35ff717a5e 2785Author: Ben Skeggs <bskeggs@redhat.com> 2786Date: Thu May 31 15:46:57 2012 +1000 2787 2788 nvc0/exa: add support for async UTS/DFS copies 2789 2790 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2791 2792commit 1600f7f202721c4b29c29e94f7fb69dc313b99dc 2793Author: Ben Skeggs <bskeggs@redhat.com> 2794Date: Thu May 31 15:40:45 2012 +1000 2795 2796 nva3/exa: add support for async UTS/DFS copies 2797 2798 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2799 2800commit ace77b6b1304826f4004bde23809b55d476b0615 2801Author: Ben Skeggs <bskeggs@redhat.com> 2802Date: Tue May 29 21:21:57 2012 +1000 2803 2804 disable fermi accel on 0.0.16 interface 2805 2806 Kepler accel support broke some assumption made by the older kernel 2807 interface, and Fermi shares the same code. It can't work (without 2808 some annoying hacks anyway) with the 0.0.16 kernel anymore. 2809 2810commit 7041e30ab8beb627bbf569367961a658e79c2bdc 2811Author: Dave Airlie <airlied@redhat.com> 2812Date: Wed May 23 14:18:24 2012 +0100 2813 2814 vl_hwmc: add missing compat include. 2815 2816 Reported-by: tallica on irc. 2817 Signed-off-by: Dave Airlie <airlied@redhat.com> 2818 2819commit 2abf8467cfb7a7648ce73ba5bcbbc62219d65d6d 2820Author: Dave Airlie <airlied@redhat.com> 2821Date: Wed May 23 11:29:05 2012 +0100 2822 2823 nouveau: add compat-api.h to makefile. 2824 2825 Signed-off-by: Dave Airlie <airlied@redhat.com> 2826 2827commit 1d861ad716861c57b2b81531d21840d7c8de024b 2828Author: Dave Airlie <airlied@redhat.com> 2829Date: Wed May 23 11:15:06 2012 +0100 2830 2831 nouveau: convert two more xf86Screens access to macros 2832 2833 for some reason the script missed these two, just fix them manually. 2834 2835 Signed-off-by: Dave Airlie <airlied@redhat.com> 2836 2837commit 5625fb84efc699e65da0062ae101915a49f2969b 2838Author: Dave Airlie <airlied@redhat.com> 2839Date: Wed May 23 11:13:30 2012 +0100 2840 2841 nouveau: convert scrn/screen to using new interfaces 2842 2843 This commit was generated with the util/modular/x-driver-screen-scrn-conv.sh 2844 2845 Signed-off-by: Dave Airlie <airlied@redhat.com> 2846 2847commit 9f037f7be516ebf8fa4a67a5c9ff258594ab353e 2848Author: Dave Airlie <airlied@redhat.com> 2849Date: Wed May 23 11:12:59 2012 +0100 2850 2851 nouveau: add initial compat-api header 2852 2853 This just provides the scrn conversion and picture stuff so far. 2854 2855 Signed-off-by: Dave Airlie <airlied@redhat.com> 2856 2857commit 581564466c47ab0435ecca3cf4d82c03e1b30259 2858Author: Francisco Jerez <currojerez@riseup.net> 2859Date: Tue May 8 14:02:44 2012 +0200 2860 2861 dri2: Don't try to page-flip pixmaps (fdo bug 49351). 2862 2863 Reported-by: Marcin Slusarz <marcin.slusarz@gmail.com> 2864 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 2865 2866commit b1efc807392d372e6a232aad9d16ae6e1b1c1916 2867Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2868Date: Sun Apr 15 13:46:42 2012 +0200 2869 2870 nv50: add missing pushbuf space check in NV50SyncToVBlank 2871 2872 Regression from "WIP: port to new libdrm". 2873 2874commit 174f17072005971fee6e4eb76a273c3cc1aa10d3 2875Author: Ben Skeggs <bskeggs@redhat.com> 2876Date: Tue Apr 24 21:55:03 2012 +1000 2877 2878 nv40/exa: return false from gradient picture setup 2879 2880 Not implemented yet. Shouldn't hit this anyway, fallback in check_composite, 2881 but, best be safe :) 2882 2883 Reported-by: Maarten Maathuis <madman2003@gmail.com> 2884 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2885 2886commit 60edf2a87b928f413385443335493cb27da30a48 2887Author: Ben Skeggs <bskeggs@redhat.com> 2888Date: Tue Apr 24 10:54:51 2012 +1000 2889 2890 nv40/exa: support for solid pictures 2891 2892 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2893 2894commit 067b8416ad1ab3e71ebbb8db8bfb82ff6041939b 2895Author: Ben Skeggs <bskeggs@redhat.com> 2896Date: Mon Apr 23 17:49:57 2012 +1000 2897 2898 nv30/exa: implement support for solid pictures using register combiners 2899 2900 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2901 2902commit b09a61780ea6710b679a6a099e35987e7a5f559f 2903Author: Ben Skeggs <bskeggs@redhat.com> 2904Date: Mon Apr 23 14:49:22 2012 +1000 2905 2906 nv10/exa: implement support for solid pictures 2907 2908 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2909 2910commit 9583753a205c585ec844dfb7168513a2c2a157ba 2911Author: Ben Skeggs <bskeggs@redhat.com> 2912Date: Fri Apr 20 20:52:01 2012 +1000 2913 2914 nv10/exa: use fixed-function texture matrix for transforms 2915 2916 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2917 2918commit c21c0ae57717bfb0fceab82505009a5382d1428d 2919Author: Ben Skeggs <bskeggs@redhat.com> 2920Date: Fri Apr 20 20:30:44 2012 +1000 2921 2922 nv30/exa: use fixed-function texture matrix for src/mask transforms 2923 2924 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2925 2926commit 24a5ca723539a940bc2dabcc50aa631f0d9d4219 2927Author: Ben Skeggs <bskeggs@redhat.com> 2928Date: Fri Apr 20 15:24:45 2012 +1000 2929 2930 nv40/exa: do transforms in vertex program 2931 2932 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2933 2934commit 9a3e579f637267b32efc46cfce5d1a36a41323b1 2935Author: Ben Skeggs <bskeggs@redhat.com> 2936Date: Thu Apr 19 16:58:52 2012 +1000 2937 2938 nvc0/exa: port recent nv50 changes (vp transform, solid pictures, etc) 2939 2940 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2941 2942commit 8ea09db1a1e8dff7c341dc3da17edefda7e56e6d 2943Author: Ben Skeggs <bskeggs@redhat.com> 2944Date: Thu Apr 19 13:46:03 2012 +1000 2945 2946 nv50/exa: support solid pictures 2947 2948 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2949 2950commit b48bcc094beecf521899dd63c8fdbccfd534e5cd 2951Author: Ben Skeggs <bskeggs@redhat.com> 2952Date: Thu Apr 19 13:21:27 2012 +1000 2953 2954 nv50/exa: perform texcoord transformations in vertex program 2955 2956 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2957 2958commit 5ac2ca8c56ec8b055878c8ac4cbc8ca74379abda 2959Author: Ben Skeggs <bskeggs@redhat.com> 2960Date: Thu Apr 19 09:52:24 2012 +1000 2961 2962 nv50: implement and use PUSH_DATAu 2963 2964 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2965 2966commit 6c41b3d7a8799daa9d0b34dcfbfc5c891c24fffa 2967Author: Ben Skeggs <bskeggs@redhat.com> 2968Date: Tue Apr 17 11:07:05 2012 +1000 2969 2970 nv30/nv40: upload shaders directly from rankine/curie init 2971 2972 Also removes A8 surface shader hack, and implements proper shaders for 2973 the specific purpose. 2974 2975 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2976 2977commit ec019b3a50475b7ff1b5e0a5dbe14e5ef677bd9e 2978Author: Ben Skeggs <bskeggs@redhat.com> 2979Date: Tue Apr 17 09:30:49 2012 +1000 2980 2981 Keep a single buffer for random accel data, rather than 3 different ones 2982 2983 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2984 2985commit fb3a36b1e5af0f81bb266da894d3442eed8e4e55 2986Author: Ben Skeggs <bskeggs@redhat.com> 2987Date: Wed Mar 28 16:38:30 2012 +1000 2988 2989 nve0: initial exa/xv acceleration for kepler chipsets 2990 2991 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2992 2993commit 95be16d8cd88db1ace0ac4568b4c67b51fd30fd9 2994Author: Ben Skeggs <bskeggs@redhat.com> 2995Date: Thu Mar 29 19:55:27 2012 +1000 2996 2997 nvc0: create scratch buffer before initialising m2mf 2998 2999 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3000 3001commit 4896c87484e1cf1e5421de377b6b83890948664d 3002Author: Ben Skeggs <bskeggs@redhat.com> 3003Date: Thu Mar 29 19:47:21 2012 +1000 3004 3005 nvc0: switch to kepler's subchannel assignments (see note) 3006 3007 This will bust for kernels with nouveau ABI version < 1.0.0... It should 3008 be possible to add compat for 0.0.16 however... 3009 3010 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3011 3012commit 4225b486cb8ab98d27ba23874096ff0879d45575 3013Author: Ben Skeggs <bskeggs@redhat.com> 3014Date: Thu Mar 29 15:56:02 2012 +1000 3015 3016 nvc0: tidy and cull a heap of fermi init (stuff in default grctx) 3017 3018 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3019 3020commit 007dd62a8c5244bc5e71fe8f32520fb35b67a237 3021Author: Ben Skeggs <bskeggs@redhat.com> 3022Date: Thu Mar 29 15:28:03 2012 +1000 3023 3024 nvc0: punt shaders out to a header file 3025 3026 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3027 3028commit 1c7651532975cf738b3aac4ec3c3b46187e24eea 3029Author: Ben Skeggs <bskeggs@redhat.com> 3030Date: Thu Mar 29 14:25:15 2012 +1000 3031 3032 nvc0: do tic/tsc binding only once, at startup 3033 3034 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3035 3036commit 180d8778764913ed061e68520603c0925477faba 3037Author: Ben Skeggs <bskeggs@redhat.com> 3038Date: Thu Mar 29 14:21:25 2012 +1000 3039 3040 nvc0: replace direct inline m2mf with inline helper to do the same 3041 3042 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3043 3044commit e70d801ae9287eab5e82f4d467dc8cd4be1b31a8 3045Author: Ben Skeggs <bskeggs@redhat.com> 3046Date: Fri Dec 9 18:15:32 2011 +1000 3047 3048 WIP: port to new libdrm 3049 3050 calim: nv50,nvc0/xv: fix src_w/h width/height confusion 3051 3052commit ab7291d368ddc66ad21c5ad5caa0faeee42ccaf1 3053Author: Ben Skeggs <bskeggs@redhat.com> 3054Date: Wed Mar 21 14:56:55 2012 +1000 3055 3056 xv: don't try to init textured video without 3d engine object 3057 3058 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3059 3060commit fe7e0425a4a43b2a7d4743a0a90a86724c3b6775 3061Author: Ben Skeggs <bskeggs@redhat.com> 3062Date: Mon Mar 12 11:59:29 2012 +1000 3063 3064 recognise kepler 3065 3066 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3067 3068commit f5d1cd2cb6808838ae1a188cef888eaa9582c76d 3069Author: Ben Skeggs <bskeggs@redhat.com> 3070Date: Tue Mar 6 08:16:51 2012 +1000 3071 3072 remove use of xf86PciInfo.h, deprecated now apparently 3073 3074 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3075 3076commit 248de8cdbd6d0bc062633b49896fa4791148cd3b 3077Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3078Date: Thu Feb 16 00:45:24 2012 +0100 3079 3080 dri2: Fix corner case crash for swaplimit > 1 3081 3082 If a swaplimit > 1 is set on a server which 3083 supports the swaplimit api (XOrg 1.12.0+), 3084 the following can happen: 3085 3086 1. Client calls glXSwapBuffersMscOML() with a 3087 swap target > 1 vblank in the future, or a 3088 client calls glXSwapbuffers() while the swap 3089 interval is set to > 1 (unusual but possible). 3090 3091 2. nouveau_dri2_finish_swap() is therefore called 3092 only at the target vblank, instead of immediately. 3093 3094 3. Because of the deferred execution of 3095 nouveu_dri2_finish_swap(), the OpenGL client 3096 can call x-servers DRI2GetBuffersWithFormat() 3097 before nouveau_dri2_finish_swap() executes and 3098 it deletes pixmaps that would be needed by 3099 nouveau_dri2_finish_swap() --> Segfault --> Crash. 3100 3101 Prevent this: When a swap is scheduled into the 3102 future, we temporarily reduce the swaplimit to 1 3103 until nouveau_dri2_finish_swap() is done, then 3104 restore it to its original value. This throttles 3105 the client inside the server in DRI2ThrottleClient() 3106 before it can call the evil DRI2GetbuffersWithFormat(). 3107 3108 The client will still be released one video refresh 3109 interval before swap completion, so there is still 3110 some potential win. 3111 3112 This doesn't affect the common case of swapping at 3113 the next vblank, where this throttling is not needed 3114 or done. 3115 3116 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3117 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3118 3119commit 247465ec5e21ce7151ef1e6a3e7644bc74d48a5b 3120Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3121Date: Thu Feb 16 00:45:23 2012 +0100 3122 3123 dri2: Reimplement hack for triple-buffering on old X-Servers. 3124 3125 X-Servers before 1.12.0 don't have the DRI2SwapLimit() 3126 API. On these, default to a swaplimit of 1 - double-buffering. 3127 3128 This patch implements support for swap limit of 2, 3129 triple-buffering, on old x-servers via Francisco Jerez 3130 previous hack: 3131 3132 Return DRI2SwapComplete() before the swap has completed, 3133 so clients don't get blocked on the pending swap. This 3134 allows for a "triple-buffering look-alike" behaviour, but 3135 breaks the swap scheduling and timestamping defined 3136 in the OML_sync_control spec, so applications which 3137 rely on conformant behaviour will break with a swap 3138 limit of 2 on pre 1.12.0 x-servers. 3139 3140 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3141 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3142 3143commit a1902ffaab1f6807fdaac0fce49f784e7208d1a5 3144Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3145Date: Thu Feb 16 00:45:22 2012 +0100 3146 3147 dri2: Allow vblank controlled swaps for redirected windows. Part II 3148 3149 This part implements proper throttling for clients. For 3150 vblank synchronized blits, it defers DRI2SwapComplete() 3151 until 1 vblank after the framebuffer blit is submitted to 3152 the gpu. 3153 3154 Rationale: 3155 3156 For unredirected windows, this is the earliest time the 3157 "blit swap" can complete, as blits are submitted one vblank 3158 before the target vblank and synchronized with vblank in the 3159 gpu. This makes swap completion timestamps at least reasonable. 3160 3161 For redirected windows, the compositor will probably pick 3162 up the "blit swapped" frontbuffer pixmap of the window quickly, 3163 but defer its own recomposition to the next vblank, at least 3164 if sync to vblank for the compositor is on. 3165 3166 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3167 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3168 3169commit c8572d329014afc0c79b5765468dcdfb65be0868 3170Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3171Date: Thu Feb 16 00:45:21 2012 +0100 3172 3173 dri2: Allow vblank controlled swaps for redirected windows. Part I 3174 3175 Make sure that swaps for redirected windows under a 3176 compositor are still scheduled via vblank events, to 3177 avoid effects like 2900 fps swaps under a compositor. 3178 3179 See discussion with Francisco Jerez at: 3180 3181 http://lists.freedesktop.org/archives/nouveau/2011-September/009278.html 3182 http://lists.freedesktop.org/archives/nouveau/2011-September/009292.html 3183 3184 This is part I of the agreed upon band-aid, in a separate patch. 3185 3186 It allows to use vblank related functions on redirected 3187 windows and thereby fixes functions from sgi_sync_control 3188 and oml_sync_control extension, e.g., glXWaitForMscOML(), 3189 glXGetSyncValuesOML(), glXWaitVideoSyncSGI, ... 3190 3191 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3192 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3193 3194commit 919643888045b102f3e10dfdeb6210ab14b6ecab 3195Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3196Date: Thu Feb 16 00:45:20 2012 +0100 3197 3198 dri2: Fixes to swap scheduling. 3199 3200 Fix some small off-by-one errors and a mismatch 3201 between 32 bit kernel interfaces for vblank count 3202 and 64 bit dri2 interfaces for target_msc et al. 3203 3204 Return corrected target_msc to swap scheduling in 3205 x-server. 3206 3207 A revised version of the patch discussed here: 3208 http://lists.freedesktop.org/archives/nouveau/2011-September/009143.html 3209 3210 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3211 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3212 3213commit 5ede4538366a352aa9726560f8d42e85df4aec82 3214Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3215Date: Thu Feb 16 00:45:19 2012 +0100 3216 3217 dri2: Update front buffer pixmap and name before exchanging buffers 3218 3219 Buffer exchange assumes that the front buffer pixmap and name 3220 information is accurate. That may not be the case eg. if the window 3221 has been (un)redirected since the buffer was created. 3222 3223 This is a translation to nouveau of a fix that was originally developed 3224 by Ville Syrjala <syrjala@sci.fi> for the ati/radeon ddx to fix the 3225 same bug there. 3226 3227 See thread at: 3228 3229 http://lists.x.org/archives/xorg-devel/2011-May/021908.html 3230 3231 Fixes FDO bug #35452. 3232 3233 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3234 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3235 3236commit 6fdf60b7288d49f889c80705aabc1db3bc327ba3 3237Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3238Date: Thu Feb 16 00:45:18 2012 +0100 3239 3240 dri2: Add support for DRI2SwapLimit() API. 3241 3242 Uses the new DRI2SwapLimit() API of X-Server 1.12+ 3243 to allow to change the maximum number of pending 3244 swaps on a drawable before the OpenGL client is 3245 throttled by the server. 3246 3247 The new optional xorg.conf parameter "SwapLimit" 3248 allows to select a new swap limit >= 1. The default 3249 swap limit is 2 for triple-buffering on XOrg 1.12+, 3250 1 for double-buffering on older servers, as we can't 3251 change the swap limit there. 3252 3253 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3254 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3255 3256commit 820916e722db2997a53128f2f370d14a02d4401c 3257Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3258Date: Thu Feb 16 00:45:17 2012 +0100 3259 3260 dri2: Implement handling of pageflip completion events. 3261 3262 Requests pageflip completion events from the kernel. 3263 Implements pageflip completion handler to finalize 3264 and timestamp swaps. 3265 3266 Completion handler includes a consistency check, and 3267 disambiguation if multiple crtc's are involved in a 3268 pageflip (e.g., clone mode, extendend desktop). Only 3269 the timestamp of the crtc whose vblank event initially 3270 triggered the swap is used, but handler waits for flip 3271 completion on all involved crtc's before completing the 3272 swap and releasing the old framebuffer. 3273 3274 This code is almost identical to the code used in the 3275 ati/radeon ddx and intel ddx. 3276 3277 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3278 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3279 3280commit 4da68cedbefef4f72c00bb088e179071490882c8 3281Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3282Date: Thu Mar 1 19:11:14 2012 +0100 3283 3284 dri2: Fix can_exchange() to allow page-flipping on more mesa versions. 3285 3286 can_exchange() returns false and thereby prevents page 3287 flipping on some drawables where page flipping would 3288 work fine. This due to non-matching drawable depths values 3289 between front buffer pixmap and back buffer pixmap, because 3290 front buffer pixmaps inherit the depth of the screen, typically 3291 24 bits, whereas the depth value of back buffer pixmaps for a 3292 given RGB8 or RGBA8 visual depends on the mesa version in use, 3293 either 24 bits or 32 bits. 3294 3295 Use bitsPerPixel instead of depth to decide if drawable 3296 is flippable. This will still catch really incompatible 3297 formats like 32 bpp vs. 16 bpp buffers. 3298 3299 Tested for screen DefaultDepth 24 and also 30 bits 3300 (for RGB10 framebuffers) on NV-50. 3301 3302 The problem was fixed in the same way in the ati & intel ddx. 3303 3304 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 3305 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3306 3307commit b0d7f4ddd94154491d04dcb9698d63d61e9240dc 3308Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3309Date: Mon Jan 9 15:17:27 2012 +0100 3310 3311 dri2: CreateBuffer format may be 0, use depth in that case 3312 3313commit 3d9f6b32e0ad72dc8c7db07d039d87e83d8fdb1d 3314Author: Ben Skeggs <bskeggs@redhat.com> 3315Date: Fri Dec 9 17:45:31 2011 +1000 3316 3317 nv50/xv: bump max dims to 8k 3318 3319 Untested, but I don't see why this wouldn't work... 3320 3321commit 57c62228e2e67370feeeb421c369fbf940d30ca4 3322Author: Ben Skeggs <bskeggs@redhat.com> 3323Date: Fri Dec 9 17:44:07 2011 +1000 3324 3325 xv: stop using custom m2mf code 3326 3327commit ad402333157c437ccd2c639f5b3f3bcb81e78d6c 3328Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3329Date: Sat Dec 10 13:32:24 2011 +0100 3330 3331 dri2: fix allocation of Z16 depth attachments 3332 3333commit 3d2a752ca8aafee5e1e94dabfd7deec439890e95 3334Author: Ben Skeggs <bskeggs@redhat.com> 3335Date: Thu Dec 1 11:31:49 2011 +1000 3336 3337 move away from libdrm's BEGIN_RING (and variants) macros 3338 3339 subchan auto-assignment is now gone. 3340 3341commit b6cfdc01aeb5c2ab401ee81fff66bd0560abdfcc 3342Author: Ben Skeggs <bskeggs@redhat.com> 3343Date: Thu Dec 1 11:25:10 2011 +1000 3344 3345 Append _CLASS to object class id definitions 3346 3347 I want to be able to use ring macros of the same name 3348 3349commit 1dadb5826363d0d8ed1a72209be18277550f8345 3350Author: Ben Skeggs <bskeggs@redhat.com> 3351Date: Thu Dec 1 09:06:24 2011 +1000 3352 3353 exa: create general m2mf rect copy for each chipset, use for UTS/DFS 3354 3355 No point having duplicated UTS/DFS logic in each implementation. This also 3356 splits nv50 out into its own code, in preparation for it being made to not 3357 use relocs at all. 3358 3359 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3360 3361commit b5534a15a5f5e1c3bd6c92e3ed8391ff628c09bb 3362Author: Marcin Slusarz <marcin.slusarz@gmail.com> 3363Date: Wed Nov 30 18:56:54 2011 +0100 3364 3365 fix typo in conversion to rnn headers 3366 3367commit 5975ce34a12d1943b2852d0f5fbe09358754c2ad 3368Author: Ben Skeggs <bskeggs@redhat.com> 3369Date: Wed Nov 30 17:40:02 2011 +1000 3370 3371 convert entire ddx to rnn headers 3372 3373commit 9fa0c6c74d0b9f7d66ebe3c99184eaa13abeb1a0 3374Author: Marcin Slusarz <marcin.slusarz@gmail.com> 3375Date: Sun Nov 6 20:32:10 2011 +0100 3376 3377 Default to 16bpp for low memory cards. 3378 3379commit 5aa30ae6554c957609929ef29d1b27c89e9a7f06 3380Author: Marcin Slusarz <marcin.slusarz@gmail.com> 3381Date: Sun Nov 6 20:32:09 2011 +0100 3382 3383 nv50/exa: wait for enough space for at least one composite operation 3384 3385commit 3d54426c4833a7e058e50624452f89139921e64b 3386Author: Marcin Slusarz <marcin.slusarz@gmail.com> 3387Date: Sun Nov 6 20:32:08 2011 +0100 3388 3389 add missing return in nv50_accel_init failure path 3390 3391commit d575a28a5020ff083836d3d1cf50018a99f49c9f 3392Author: Maarten Lankhorst <m.b.lankhorst@gmail.com> 3393Date: Sat Oct 8 10:33:20 2011 +0200 3394 3395 dri2: Add vdpau driver name entry 3396 3397 libvdpau has a driver loading mechanism that looks for a dri2 driver 3398 first before falling back to nvidia, so lets use that. 3399 3400commit 656cfe3bdb8b487827b710a0cea35ae1bcd5044e 3401Author: Marcin Slusarz <marcin.slusarz@gmail.com> 3402Date: Sat Sep 10 23:14:53 2011 +0200 3403 3404 xv: support AI44/IA44 3405 3406 It's needed for osd in mplayer's xvmc output. 3407 3408commit 169512fbe91f0671a90dfee5e280357f0a4ef701 3409Author: Maxim Levitsky <maximlevitsky@gmail.com> 3410Date: Tue Aug 23 13:10:59 2011 +0200 3411 3412 dri2: Disable the "exchange" swapbuffers path for the moment (bug 35930). 3413 3414 Allow page flipping only for scanout buffer for now as simple swapping 3415 between off-screen pixmaps confuses compiz because there is no 3416 syncronization method to tell it about that swap 3417 3418 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3419 3420commit b806e3f97a73701f057c45b1f45233e69e19f113 3421Author: Bryan Cain <bryancain3@gmail.com> 3422Date: Wed Jul 20 13:48:36 2011 -0500 3423 3424 g3dvl: update for compatibility with latest pipe-video branch in Mesa 3425 3426 Signed-off-by: Bryan Cain <bryancain3@gmail.com> 3427 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3428 3429commit de9d1ba7efeba64f319efa00df183d3cb78f24af 3430Author: Emil Velikov <emil.l.velikov@gmail.com> 3431Date: Tue Jul 19 00:35:38 2011 +0100 3432 3433 drmmode_display: Resolve missing brackets 3434 3435 Correct some missing/misplaced brackets in drmmode_pre_init() 3436 The issue was exposed when trying a 4 monitor desktop using two 3437 cards/gpus 3438 3439 Resolves https://bugs.freedesktop.org/show_bug.cgi?id=39099 3440 3441 Reported-By: Damian Nowak <nowaker@geozone.pl> 3442 Tested-By: Damian Nowak <nowaker@geozone.pl> 3443 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 3444 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3445 3446commit ef4957492956df54fafa6c75d576a07ce678ea0a 3447Author: Ben Skeggs <bskeggs@redhat.com> 3448Date: Mon Jul 11 12:48:34 2011 +1000 3449 3450 kms: fix multiple rotations in noaccel mode 3451 3452commit ab89aa022f0c78154e525b5f5787feee8024171f 3453Author: Ben Skeggs <bskeggs@redhat.com> 3454Date: Fri Jun 17 10:10:09 2011 +1000 3455 3456 nvc0: accept 0xdx chipsets 3457 3458 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3459 3460commit 7dd8987b8b69f56d04a199207122d2e51b9caa6e 3461Author: Ben Skeggs <bskeggs@redhat.com> 3462Date: Fri Jun 17 10:07:36 2011 +1000 3463 3464 nvc0: assume 0x9097's presence on all chipsets 3465 3466 This appears to be the case thus far, and is what the 3D driver does. 3467 3468 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3469 3470commit 98ee2105dbd8685e59c1967a587fccb6135743ac 3471Author: Ben Skeggs <bskeggs@redhat.com> 3472Date: Mon May 30 10:10:58 2011 +1000 3473 3474 kms: clear new fb to black on fb resize 3475 3476commit 8378443bd3b26b57ef2ae424a700e01ead813d33 3477Author: Ben Skeggs <bskeggs@redhat.com> 3478Date: Thu Mar 24 02:13:12 2011 +1000 3479 3480 nv50-nvc0/exa: fix bug causing surface state to not be reemitted after flush 3481 3482 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3483 3484commit 92db2bc192a074d4b7fc3e9c16e9aa62a8deaeb6 3485Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3486Date: Thu Mar 3 22:17:23 2011 +0100 3487 3488 nvc0/accel: allow nvc1,nvc3,nvc4 in 3D engine init and use 9097 3489 3490commit ace98a492353e6de712f4f717e6d3f562e3591f0 3491Author: Ben Skeggs <bskeggs@redhat.com> 3492Date: Tue Mar 1 13:54:38 2011 +1000 3493 3494 dri2: return an error rather than crashing if we can't allocate a buffer 3495 3496 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3497 3498commit bc5dec2ca7ca7edc340a99bd73946e228117dfd8 3499Author: Ben Skeggs <bskeggs@redhat.com> 3500Date: Thu Feb 24 15:15:00 2011 +1000 3501 3502 dri2: disable page flipping if any crtc is rotated 3503 3504 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3505 3506commit c123618470b5ff32c6377a4cae0f148d4a518a5d 3507Author: Francesco Marella <francesco.marella@gmail.com> 3508Date: Thu Feb 17 03:48:16 2011 +0100 3509 3510 Fix gcc 4.6.0 set but unused variable warnings 3511 3512 Signed-off-by: Francesco Marella <fmarl@paranoici.org> 3513 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3514 3515commit 3531b2aa13e9c78288d96fe5e1492f0b6e899b00 3516Author: Francesco Marella <francesco.marella@gmail.com> 3517Date: Thu Feb 17 03:47:46 2011 +0100 3518 3519 Page flipping man page update 3520 3521 Signed-off-by: Francesco Marella <fmarl@paranoici.org> 3522 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3523 3524commit 85a706df6f7986158c964e3afd0aa8e352ff5cea 3525Author: Francesco Marella <francesco.marella@gmail.com> 3526Date: Thu Feb 17 03:47:08 2011 +0100 3527 3528 Update configure.ac with autoupdate 3529 3530 Signed-off-by: Francesco Marella <fmarl@paranoici.org> 3531 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3532 3533commit 42c16ff33dcca6cd9bae62fe0d71a40452e03581 3534Author: Ben Skeggs <bskeggs@redhat.com> 3535Date: Wed Feb 16 13:51:54 2011 +1000 3536 3537 nv50: attempt to allocate compressed Z buffers 3538 3539 This should automagically fallback to the non-compressed version on kernels 3540 that don't support compression. 3541 3542 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3543 3544commit 46acb7e095e9c0d55b90070b77c64679fcfb9d4e 3545Author: Marcin Slusarz <marcin.slusarz@gmail.com> 3546Date: Sun Feb 13 20:46:42 2011 +0100 3547 3548 fix drmSetMaster error reporting 3549 3550 On error, drmSetMaster returns -1 and sets errno. 3551 3552 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3553 3554commit 54764980411b4b8124c36edb8730612b04f292a9 3555Author: Maarten Maathuis <madman2003@gmail.com> 3556Date: Thu Feb 10 20:12:07 2011 +0100 3557 3558 Also flush from the BlockHandler. 3559 3560 - The fact that commit 02c50db307a4e4eeb87f7db5d401bc53cce3b21f 3561 fixed missing text for some people proves this is needed. 3562 That commit should have only improved latency a bit. 3563 - The FlushCallback is still needed for DRI2 texture_from_pixmap. 3564 - See bugs.fd.o #33977 for discussion. 3565 3566 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 3567 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3568 3569commit 02c50db307a4e4eeb87f7db5d401bc53cce3b21f 3570Author: Maarten Maathuis <madman2003@gmail.com> 3571Date: Mon Feb 7 19:37:47 2011 +0100 3572 3573 exa: Flush (S)IFC to the frontbuffer immediately. 3574 3575 This avoids high latency while typing with core fonts for example. 3576 3577 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 3578 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3579 3580commit e4fb9fd2f790ed66e4be61ca00d55522525619f4 3581Author: Ben Skeggs <bskeggs@redhat.com> 3582Date: Wed Feb 9 13:46:32 2011 +1000 3583 3584 dri2: allow page flipping to be disabled in xorg.conf 3585 3586 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3587 3588commit 38e8809bb415bae5c182fc79c8fc62992c5e4ed0 3589Author: Ben Skeggs <bskeggs@redhat.com> 3590Date: Mon Jan 17 09:14:31 2011 +1000 3591 3592 nvc0: merge initial acceleration support 3593 3594 Squashed commit of the following: 3595 3596 commit e8ce1ef351ee51bec51211dca6fda88fdbfbefee 3597 Author: Ben Skeggs <bskeggs@redhat.com> 3598 Date: Mon Jan 17 08:55:04 2011 +1000 3599 3600 fix hardcoding of nvc0 wfb path 3601 3602 commit b414b1f948da18d96309bad84c54948d877efaa6 3603 Author: Ben Skeggs <bskeggs@redhat.com> 3604 Date: Mon Jan 17 08:52:53 2011 +1000 3605 3606 general tidy-ups in preparation for merge to master 3607 3608 commit 73bf71d5de0cb990178a1d2a31e0ea1717c214a4 3609 Merge: fb499a4 b795ca6 3610 Author: Ben Skeggs <bskeggs@redhat.com> 3611 Date: Mon Jan 17 08:47:40 2011 +1000 3612 3613 Merge branch 'master' into nvc0 3614 3615 commit fb499a4e9d95650dc89f4c1820b94d01344733f6 3616 Author: Ben Skeggs <bskeggs@redhat.com> 3617 Date: Mon Dec 20 12:54:43 2010 +1000 3618 3619 nvc0: switch to "standard" RING macros 3620 3621 commit 64f0e1d0e9e29c5437ad4e52d01ccfb0b03a2ae9 3622 Merge: eba7e17 c2092a4 3623 Author: Ben Skeggs <bskeggs@redhat.com> 3624 Date: Mon Dec 20 11:58:42 2010 +1000 3625 3626 Merge branch 'master' into nvc0 3627 3628 Conflicts: 3629 src/nv_dma.c 3630 3631 commit eba7e173a5006ff50b950d0e8b8154564471a9ce 3632 Author: Ben Skeggs <bskeggs@redhat.com> 3633 Date: Fri Dec 17 11:04:40 2010 +1000 3634 3635 nvc0/xv: use mad in nv12 shader, rather than mul+add 3636 3637 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3638 3639 commit b025ceb40a5e94a3657f3c7f76044c49a1a05e8d 3640 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3641 Date: Thu Dec 16 23:34:29 2010 +0100 3642 3643 nvc0: fix offset of 17bc buffer 3644 3645 I had messed it up in 5e4da3451c3c0f645f98fea7fb0ce2629b1bfd2f. 3646 3647 commit 6ff5a9a7f3411ef218f17fa481c583596ffde2cc 3648 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3649 Date: Thu Dec 16 01:07:25 2010 +0100 3650 3651 nvc0/xv: bind textures for the FP, not the TEP 3652 3653 commit 5e4da3451c3c0f645f98fea7fb0ce2629b1bfd2f 3654 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3655 Date: Thu Dec 16 01:02:58 2010 +0100 3656 3657 nvc0: switch to rnn headers 3658 3659 commit 0839041e79dbb0e629326d0f240220b33921db1a 3660 Author: Ben Skeggs <bskeggs@redhat.com> 3661 Date: Wed Dec 15 16:42:41 2010 +1000 3662 3663 nvc0/exa: fix m2mf dfs 3664 3665 commit 88c1c43302331e1cb8b7035bfb4677a5e7da3090 3666 Author: Ben Skeggs <bskeggs@redhat.com> 3667 Date: Mon Dec 13 12:00:23 2010 +1000 3668 3669 nvc0: add Xv implementation 3670 3671 commit ad5b4cfe388a102a0c93f7281cb4bba532b1796b 3672 Merge: 1942a81 b6cb21b 3673 Author: Ben Skeggs <bskeggs@redhat.com> 3674 Date: Wed Dec 15 14:57:04 2010 +1000 3675 3676 Merge branch 'master' into nvc0 3677 3678 commit 1942a81c0eb592972d4a3e8c46b80192ae8d1a59 3679 Author: Ben Skeggs <bskeggs@redhat.com> 3680 Date: Thu Dec 9 12:48:27 2010 +1000 3681 3682 nvc0: call grobj_alloc for all used object classes 3683 3684 Though the hardware has no concept of object handles any more, the nouveau 3685 DRM still uses this call to know what engines are in use. 3686 3687 commit 0c670a6d406fe0a57373fa9e03005be6f28ecc3e 3688 Author: Ben Skeggs <bskeggs@redhat.com> 3689 Date: Thu Dec 9 12:44:41 2010 +1000 3690 3691 nvc0: don't try and create DmaNotifier0 3692 3693 commit 5d3d20d908a50ad7742567d58969c7f12dd944f7 3694 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3695 Date: Thu Dec 9 01:36:32 2010 +0100 3696 3697 nvc0: change tile_mode to contain 0xZYX instead of 0xZY 3698 3699 Warning: the kernel will not like that (EVO) 3700 3701 commit dd7fd8263e0e9f92412fb20086e4c8acf8cab099 3702 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3703 Date: Thu Dec 9 00:29:58 2010 +0100 3704 3705 exa: don't fall back to nv04 m2mf if nvc0 m2mf fails 3706 3707 commit b4cf5a8f2276392945c2530f0d6fe96b5ab4b5da 3708 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3709 Date: Thu Dec 9 00:07:59 2010 +0100 3710 3711 nvc0/exa: remove noisy debug messages 3712 3713 commit 5419ef6d2cfabb7329982cb8dbdb5bbc8e973fe3 3714 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3715 Date: Wed Dec 8 23:55:51 2010 +0100 3716 3717 nvc0/accel: remove unneeded scratch method 0x3420 init 3718 3719 It was intended be read by a MACRO method to get the address for 3720 constants upload. 3721 3722 commit 178356391ff6831599eba257a5912079894641a1 3723 Merge: 8573280 8bb8231 3724 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3725 Date: Thu Dec 9 01:31:40 2010 +0100 3726 3727 Merge remote branch 'origin/master' into nvc0 3728 3729 Conflicts: 3730 src/drmmode_display.c 3731 src/nouveau_exa.c 3732 src/nv_driver.c 3733 src/nv_type.h 3734 3735 commit 8573280871401e29f27fa24d0b7c6ac5e35b2cc1 3736 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3737 Date: Sat Jul 3 14:27:36 2010 +0200 3738 3739 nvc0: initial implementation 3740 3741 EXA should work, Xv not yet done. 3742 3743commit b795ca6e97fae9735843748585401098dae1c3e1 3744Author: Ben Skeggs <bskeggs@redhat.com> 3745Date: Fri Jan 7 13:33:36 2011 +1000 3746 3747 nv50: fix pitch of linear scanout buffers 3748 3749 Un-breaks NoAccel on these chipsets. 3750 3751 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3752 3753commit dc89dac734167bcbc667b39ca6ee2043871a60bf 3754Author: Xavier Chantry <chantry.xavier@gmail.com> 3755Date: Sun Dec 19 23:17:34 2010 +0100 3756 3757 nv04/exa: kill useless WAIT_RING 3758 3759 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3760 3761commit b76b43928bf4dcf5ce3a4b58c7fefe780070c0d3 3762Author: Ben Skeggs <bskeggs@redhat.com> 3763Date: Tue Dec 21 16:25:23 2010 +1000 3764 3765 add back generic include of nouveau_pushbuf.h 3766 3767 So places that FIRE_RING() still work. 3768 3769 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3770 3771commit c2092a458401377f421d9bdfe1d97ce580b99508 3772Author: Ben Skeggs <bskeggs@redhat.com> 3773Date: Mon Dec 20 11:50:57 2010 +1000 3774 3775 include nv04_pushbuf.h for ring macros, rather than nouveau_pushbuf.h 3776 3777 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3778 3779commit aa2821a42706ac7b69703d1869e2d00a4ced9f4b 3780Author: Ben Skeggs <bskeggs@redhat.com> 3781Date: Mon Dec 20 11:53:53 2010 +1000 3782 3783 kill NVSync(), it's rather useless 3784 3785 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3786 3787commit 1fc564fe3494cf0abcc848d0e90bf2232f8fd272 3788Author: Ben Skeggs <bskeggs@redhat.com> 3789Date: Thu Dec 16 09:49:09 2010 +1000 3790 3791 nv50/xv: use mad in nv12 shader, rather than mul+add 3792 3793 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3794 3795commit b6cb21b91f7d0409794a14ddbdd606728d4d49e3 3796Author: Ben Skeggs <bskeggs@redhat.com> 3797Date: Mon Feb 8 12:51:47 2010 +1000 3798 3799 set canDoBGNoneRoot if building against new enough server 3800 3801commit 11510f9ce8c9eab15e7f6d5962b83fc0052c8619 3802Author: Ben Skeggs <bskeggs@redhat.com> 3803Date: Tue Dec 14 15:39:01 2010 +1000 3804 3805 nv50/xv: implement brightness/contrast/saturation/hue controls 3806 3807 Adapted from the code in xf86-video-ati. 3808 3809 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3810 3811commit 214ea264cdb386520a1308a29f204e441f349cb1 3812Author: Ben Skeggs <bskeggs@redhat.com> 3813Date: Tue Dec 14 14:42:24 2010 +1000 3814 3815 nv50/xv: kick csc parameters out into a constant buffer 3816 3817 These were previously inlined into the shader, which is simple, but makes 3818 implementing brightness/contrast adjustments impossible. 3819 3820 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3821 3822commit 8bb8231236a6b877895663aeaa9cef731d67fe68 3823Author: Ben Skeggs <bskeggs@redhat.com> 3824Date: Tue Nov 30 15:27:36 2010 +1000 3825 3826 dri2: fix certain clients blocking forever when sync-to-vblank unavailable 3827 3828 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3829 3830commit f29afe65b67c9b89ceba421d9b3aee590f7ded20 3831Author: Ben Skeggs <bskeggs@redhat.com> 3832Date: Mon Nov 29 09:23:35 2010 +1000 3833 3834 dri2: don't return failure from get_msc if sync-to-vblank isn't available 3835 3836 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3837 3838commit 4b1d26578116f4d1745dd92a7c99413972a2bedf 3839Author: Francisco Jerez <currojerez@riseup.net> 3840Date: Thu Nov 25 21:33:46 2010 +0100 3841 3842 dri2: Don't return failure from schedule_wait if sync-to-vblank is unavailable. 3843 3844 Reported-by: Ben Skeggs <bskeggs@redhat.com> 3845 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3846 3847commit 1143e7a97dce1b741376e178b73b8b766e03849a 3848Author: Jesse Adkins <jesserayadkins@gmail.com> 3849Date: Tue Oct 26 16:05:32 2010 -0700 3850 3851 Purge cvs tags. 3852 3853 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com> 3854 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3855 3856commit 22020457f43eceaf26532ecb50e46d743f4a691f 3857Author: Francisco Jerez <currojerez@riseup.net> 3858Date: Tue Oct 26 00:24:06 2010 +0200 3859 3860 dri2: Fix throttling on occluded drawables. 3861 3862 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3863 3864commit 9737a812aacbc544b93eeaa031cf1c2e8adbd661 3865Author: Ben Skeggs <bskeggs@redhat.com> 3866Date: Mon Oct 25 10:37:08 2010 +1000 3867 3868 kms: initial pass at supporting zaphod-mode 3869 3870 This works ok from some simple testing, there's undoubtedly things that 3871 will need polishing still however. 3872 3873 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3874 3875commit d40c16f5497739c0215f37566db1651a74cb20ba 3876Author: Francisco Jerez <currojerez@riseup.net> 3877Date: Sat Oct 23 02:20:30 2010 +0200 3878 3879 dri2: Fix build on X server 1.8. 3880 3881 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3882 3883commit a4d580bf05d7aac1d486e5d1ba9a2201c026e5f5 3884Author: Francisco Jerez <currojerez@riseup.net> 3885Date: Fri Oct 22 23:09:30 2010 +0200 3886 3887 Bump the required X server version to 1.8. 3888 3889 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3890 3891commit d0f2e16bc35660d01d3cd20a770cf8e613ab5323 3892Author: Francisco Jerez <currojerez@riseup.net> 3893Date: Fri Oct 22 23:06:28 2010 +0200 3894 3895 dri2: Report damage on page-flipped drawables. 3896 3897 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3898 3899commit c61e8d03161adcfbba4162172153814accf9e86b 3900Author: Francisco Jerez <currojerez@riseup.net> 3901Date: Fri Oct 22 23:05:24 2010 +0200 3902 3903 Make the sync-to-vblank helpers take a box as argument. 3904 3905 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3906 3907commit a685a5c3be12f3e95a6ec1140af115e7ef28fcaa 3908Author: Francisco Jerez <currojerez@riseup.net> 3909Date: Wed Oct 13 18:45:04 2010 +0200 3910 3911 dri2: Implement the ScheduleWaitMSC/GetMSC hooks. 3912 3913 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3914 3915commit eb83c830c87bce345748edef3b50660246143db7 3916Author: Francisco Jerez <currojerez@riseup.net> 3917Date: Thu Oct 21 22:57:08 2010 +0200 3918 3919 dri2: Add pageflip/exchange support. 3920 3921 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3922 3923commit c88f13e25b0040c1dd0f93e0ac40f62a6005ce59 3924Author: Francisco Jerez <currojerez@riseup.net> 3925Date: Thu Oct 14 22:25:50 2010 +0200 3926 3927 Allocate a tiled scanout on pre-nv50 too. 3928 3929 Pageflip between tiled and non-tiled surfaces isn't such a good idea. 3930 3931 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3932 3933commit 4642d71bdffc04ce0dd458225541940c65f5443d 3934Author: Francisco Jerez <currojerez@riseup.net> 3935Date: Thu Oct 14 22:17:58 2010 +0200 3936 3937 dri2: Give out buffers ready for scanout. 3938 3939 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3940 3941commit 75daada678d03d2e162f8003a6f2edcbe0525f92 3942Author: Francisco Jerez <currojerez@riseup.net> 3943Date: Thu Oct 14 22:32:54 2010 +0200 3944 3945 Restructure tiled pixmap allocation in a single place. 3946 3947 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3948 3949commit 70f0d2c886ceaa965ca4864788f4dd8e8f757a92 3950Author: Francisco Jerez <currojerez@riseup.net> 3951Date: Thu Oct 21 22:55:02 2010 +0200 3952 3953 dri2: Add sync-to-vblank support. 3954 3955 Note that you need a recent enough kernel for it to work. Use the 3956 "GLXVBlank" option to enable it (it defaults to disabled for now). 3957 3958 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3959 3960commit 72dd4899ff036b09e3680ed2e09b3dbae3a7ba71 3961Author: Francisco Jerez <currojerez@riseup.net> 3962Date: Thu Oct 21 22:31:45 2010 +0200 3963 3964 nv04-nv40/exa: Make the copy hook blit downwards for non-overlapping regions. 3965 3966 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3967 3968commit 21ed642d3f76f286f23cd64ca817038ec94a1bf6 3969Author: Francisco Jerez <currojerez@riseup.net> 3970Date: Wed Oct 13 18:22:22 2010 +0200 3971 3972 Make the sync-to-vblank helpers more consistent. 3973 3974 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3975 3976commit 258e483d47f0f63155be18981c1118261f7675a8 3977Author: Francisco Jerez <currojerez@riseup.net> 3978Date: Thu Oct 21 22:41:24 2010 +0200 3979 3980 kms: Register a wakeup handler to poll vblank/pageflip/udev events. 3981 3982 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3983 3984commit 4f42708c26fa57bac7854701a6e757958c8d0654 3985Author: Francisco Jerez <currojerez@riseup.net> 3986Date: Sun Oct 10 04:39:27 2010 +0200 3987 3988 Update to the last libdrm API. 3989 3990 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3991 3992commit 8c8f15c9d3a31c35bf2ef20faf8c4133295a678f 3993Author: Francisco Jerez <currojerez@riseup.net> 3994Date: Sun Oct 10 04:36:17 2010 +0200 3995 3996 nv10-nv40/exa: Fix tiled pixmap pitch alignment calculation. 3997 3998 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3999 4000commit a47c274f329b3b8bfbdcf155d6d832599d823dc8 4001Author: Francisco Jerez <currojerez@riseup.net> 4002Date: Sun Oct 10 04:43:25 2010 +0200 4003 4004 Fix for the last xserver ABI. 4005 4006 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4007 4008commit 9a5142fae6e1a0a470352f511427916e6fa01b9c 4009Author: Maarten Maathuis <madman2003@gmail.com> 4010Date: Fri Aug 27 21:11:59 2010 +0200 4011 4012 nv50/exa: Stop using wfb by default, but add an option to enable it. 4013 4014 - Some scenarios, such as those heavy in core font rendering might prefer wfb. 4015 - Other (modern) situations are better of with wfb disabled. 4016 4017 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4018 4019commit d5a6f103113985c50bbe749ed341de34af1b47a5 4020Author: Francisco Jerez <currojerez@riseup.net> 4021Date: Tue Sep 28 21:49:48 2010 +0200 4022 4023 dri2: Throttle frontbuffer flushes. 4024 4025 There was no need for throttling until the hardware interchannel sync 4026 kernel changes because the pushbuf IOCTL blocked to sync the GL and X 4027 channels anyway. 4028 4029 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4030 4031commit f1ac413d1d3dec2ccf63d4a8c79b9bd9ea578dcf 4032Author: Ben Skeggs <bskeggs@redhat.com> 4033Date: Mon Sep 20 08:05:00 2010 +1000 4034 4035 nv50: use 0x8697 on NVAF 4036 4037 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4038 4039commit 14c93acfc3fde04415e1c6a2a056465d15c86315 4040Author: Ben Skeggs <bskeggs@redhat.com> 4041Date: Fri Sep 17 13:17:37 2010 +1000 4042 4043 noaccel: fix rotation, as well as an assert during server shutdown 4044 4045 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4046 4047commit 065576d5df0df4da61050d789f548801656b62d5 4048Author: Ben Skeggs <bskeggs@redhat.com> 4049Date: Thu Aug 26 13:22:33 2010 +1000 4050 4051 dri2: remove explit flush from copy_region() 4052 4053 Unnecessary now that we kickoff from within the flush callback chain. 4054 4055 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4056 4057commit 34025f0dc9284873d4bf7e4f547ed69fc562a3c9 4058Author: Ben Skeggs <bskeggs@redhat.com> 4059Date: Thu Aug 26 13:19:39 2010 +1000 4060 4061 kickoff from the flush callback chain 4062 4063 This is a port of intel commit: 69d65f9184006eac790efcff78a0e425160e95aa 4064 4065 Submit batch buffers from flush callback chain 4066 4067 There are a few cases where the server will flush client output buffers 4068 but our block handler only catches the most common (before going into select 4069 If the server flushes client buffers before we submit our batch buffer, 4070 the client may receive a damage event for rendering that hasn't happened yet 4071 4072 Instead, we can hook into the flush callback chain, which the server will 4073 invoke just before flushing output. This lets us submit batch buffers 4074 before sending out events, preserving ordering. 4075 4076 Fixes 28438: [bisected] incorrect character in gnome-terminal under compiz 4077 https://bugs.freedesktop.org/show_bug.cgi?id=28438 4078 4079 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net> 4080 4081 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4082 4083commit 390f1c86d9583f4ceabd407ee47410073c2a1b55 4084Author: Ben Skeggs <bskeggs@redhat.com> 4085Date: Fri Aug 20 09:48:32 2010 +1000 4086 4087 nv50/xv: support UYVY 4088 4089 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4090 4091commit 00d390952c912d4e9fc2c962caaeb90bf563d5b1 4092Author: Maarten Maathuis <madman2003@gmail.com> 4093Date: Fri Aug 13 22:24:26 2010 +0200 4094 4095 nv50/xv: Fix the uncommon codepath to use width/height variable. 4096 4097 - The common codepath should be the correct one, because that is 4098 triggered often, while the other is rare. 4099 - This isn't neccesarily a problem, but let's be on the safe side. 4100 4101 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4102 4103commit f5b486dd237679aec1987985f9ecffc6a4fd3105 4104Author: Maarten Maathuis <madman2003@gmail.com> 4105Date: Fri Aug 13 22:16:51 2010 +0200 4106 4107 nv50/xv: We have 18 relocs, not 16, adjust MARK_RING. 4108 4109 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4110 4111commit 1cb6c91a9afe88d73f8a1180b00f56a9150a16ed 4112Author: Roy Spliet <r.spliet@student.tudelft.nl> 4113Date: Sat Aug 7 20:59:06 2010 +0200 4114 4115 nouveau_exa.c: move line_count logic out of UTS/DFS loop 4116 4117 Signed-off-by: Roy Spliet <r.spliet@student.tudelft.nl> 4118 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4119 4120commit 9f50b62229d5984ee5c16bc54953c7f8e2d5e575 4121Author: Ben Skeggs <bskeggs@redhat.com> 4122Date: Fri Aug 6 08:28:34 2010 +1000 4123 4124 nvc0: recognise these chipsets, and allow NoAccel mode on them 4125 4126 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4127 4128commit b96170a4e51c1d3167601210ca31e98283c02885 4129Author: Francisco Jerez <currojerez@riseup.net> 4130Date: Thu Aug 5 00:40:40 2010 +0200 4131 4132 nv10/exa: Bump maxX/maxY to 4096/4096. 4133 4134 This way we can do accelerated 2D on larger pixmaps, it will still 4135 fall back to software when we hit the 2048x2048 texturing limit in a 4136 composite operation. 4137 4138commit 321eb2df5c8b1da4d6469037dd1d8888b508fabd 4139Author: Francisco Jerez <currojerez@riseup.net> 4140Date: Wed Aug 4 13:11:24 2010 +0200 4141 4142 Revert "nv04-nv40/exa: Match the blob behavior more closely on PrepareCopy." 4143 4144 This reverts commit 9de0d97bd2fc2ee8800d48b5340a3d495525ad3e. Unnecessary 4145 after kernel commit "drm/nouveau: Ack the context switch interrupt before 4146 switching contexts.". 4147 4148commit db98ad23633958e52f84501c5a2061d42d346b64 4149Author: Francisco Jerez <currojerez@riseup.net> 4150Date: Tue Jun 15 15:32:36 2010 +0200 4151 4152 dri2: Fix for the last DRI2InfoRec ABI. 4153 4154 Just pre-fill the whole structure with zeros to avoid future 4155 segfaults. 4156 4157commit 9a3d0ab43f6335ab6cd48be4aa2f4f728386b386 4158Author: Ben Skeggs <bskeggs@redhat.com> 4159Date: Fri Jun 11 21:58:38 2010 +1000 4160 4161 call NVEnterVT from CreateScreenResources 4162 4163 We need to grab DRM master here for things to work, so we may as well call 4164 NVEnterVT as we used to in NVScreenInit up until the previous commit. 4165 4166 Fixes server regeneration. 4167 4168commit 964eeac6dc2209da6aa8861aa8b0bc028700ab3d 4169Author: Ben Skeggs <bskeggs@redhat.com> 4170Date: Thu Jun 10 09:38:19 2010 +1000 4171 4172 fix for server 1.9 4173 4174commit c079408e287ad09e954fa3985f95912c02cf06bc 4175Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 4176Date: Thu Jun 10 01:07:49 2010 +0200 4177 4178 nv50/accel: fix perspective interps in YUV to RGB shader 4179 4180commit 65153490f7d74d45a92adbd9c5e2bbe11c884b00 4181Author: Ben Skeggs <bskeggs@redhat.com> 4182Date: Wed Jun 9 16:12:28 2010 +1000 4183 4184 use libc directly for [cm]alloc/free 4185 4186commit 2958cf464dca9761f0fa21c8dd2d8fa8c8a96791 4187Author: Ben Skeggs <bskeggs@redhat.com> 4188Date: Thu May 27 10:06:47 2010 +1000 4189 4190 nv50/exa: rip out compat defines, we depend on 1.7 anyway 4191 4192commit af40bf0af47ebcb7de2dc418429f0b56f8c04d15 4193Author: Ben Skeggs <bskeggs@redhat.com> 4194Date: Wed May 26 16:11:01 2010 +1000 4195 4196 nv50/exa: new nouveau_class.h has correct blend func defines 4197 4198commit 8950c5b2a0d318e2568bec38a39d744b968c9b85 4199Author: Ben Skeggs <bskeggs@redhat.com> 4200Date: Wed May 26 16:09:48 2010 +1000 4201 4202 update nouveau_class.h 4203 4204commit 4b8f1a082069a17555d4142003213cf83d4e5464 4205Author: Lubomir Rintel <lkundrak@v3.sk> 4206Date: Mon May 10 09:25:55 2010 +0200 4207 4208 Handle startup without monitor attached more gracefully 4209 4210 Refuse to configure a screen which has no valid modes. Otherwise X server 4211 would crash attempting to dereference to modes list later on. 4212 4213 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4214 4215commit ac8f8d14a82ae8ba5b41fa88903edcad67f339c2 4216Author: Maarten Maathuis <madman2003@gmail.com> 4217Date: Fri May 7 20:51:43 2010 +0200 4218 4219 wfb: use 28 bits instead of 32 bits for multiply_factor optimisation 4220 4221 - Avoiding division in a hot code path is noticable even on an otherwise idle 4222 computer, but it's subject to rounding errors. If anyone knows a corner case 4223 and has better ideas, don't hesitate to contact me. For the moment this should 4224 make the problem less likely to trigger. 4225 4226 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4227 4228commit 13c10430ba8f7b0edff3ad8aae4a97672eea4a8f 4229Author: Ben Skeggs <bskeggs@redhat.com> 4230Date: Fri Apr 23 14:09:35 2010 +1000 4231 4232 Bump version number to 0.0.16 4233 4234 Forgotten when the ABI broke a while back.. 4235 4236commit 6ff8ade0c7cd835b4172257ea310a8d88f28e757 4237Author: Ben Skeggs <bskeggs@redhat.com> 4238Date: Fri Apr 16 16:09:51 2010 +1000 4239 4240 Add uevent support 4241 4242 When a hotplug event is received from the kernel we should notify the 4243 client side to reconfigure the display. 4244 4245 Ported to nouveau from radeon driver, based on work by ajax in intel driver. 4246 4247commit 4063616938f76af8028491276039d422c0782b1b 4248Author: Ben Skeggs <bskeggs@redhat.com> 4249Date: Fri Apr 9 20:43:51 2010 +1000 4250 4251 Import latest nouveau_class.h from renouveau 4252 4253commit bfb95cc0ca600135474ff02a99889b09d41b3c80 4254Author: Rico Tzschichholz <ricotz@t-online.de> 4255Date: Thu Apr 1 20:39:23 2010 +0200 4256 4257 Fix "make dist". 4258 4259 - remove README.NV1 reference from Makefile.am 4260 - add nv_rop.h to src/Makefile.am 4261 4262 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4263 4264commit 2462b417fc550b71f021ca9736808f8f2d1cb6f0 4265Author: Ben Skeggs <skeggsb@beleth.localdomain> 4266Date: Mon May 4 17:04:34 2009 +1000 4267 4268 xv: only use bicubic filtering when scaling >=2x 4269 4270commit e2146d3b29a4bea3d584c145e3190c3313692ed9 4271Author: Francisco Jerez <currojerez@riseup.net> 4272Date: Thu Mar 18 14:32:54 2010 +0100 4273 4274 Fix for the latest nouveau_class.h. 4275 4276 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4277 4278commit 7858345b6b384e88902ccc516715156401edec04 4279Author: Ben Skeggs <bskeggs@redhat.com> 4280Date: Tue Mar 16 13:08:37 2010 +1000 4281 4282 nv50: fix texturing from >=4GiB mark 4283 4284commit 6b8b15731b7610c9d8fb3625861797a0c9fd9546 4285Author: Xavier Chantry <chantry.xavier@gmail.com> 4286Date: Fri Mar 5 00:53:47 2010 +0100 4287 4288 Remove leftover in man page and nv_const.h 4289 4290 Signed-off-by: Xavier Chantry <chantry.xavier@gmail.com> 4291 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4292 4293commit c642b9f7a13bdeecd0a83ddcbf6d6d4f2c287501 4294Author: Francisco Jerez <currojerez@riseup.net> 4295Date: Fri Mar 5 14:27:08 2010 +0100 4296 4297 kms: Don't try to copy fbcon contents when there's no framebuffer bound. 4298 4299 drmModeGetFB complains loudly when it's given an invalid framebuffer 4300 id, bail out earlier if that's the case. 4301 4302 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4303 4304commit efd7a526ca1cbac10b33b9d5d4cf3e542bc9b73b 4305Author: Ben Skeggs <bskeggs@redhat.com> 4306Date: Thu Feb 25 09:32:49 2010 +1000 4307 4308 fix typo 4309 4310commit 70d0a48b6c3f1a817bf850acd3bae48d063e56b9 4311Author: Ben Skeggs <bskeggs@redhat.com> 4312Date: Tue Feb 23 15:08:13 2010 +1000 4313 4314 Protect against potential crash for unknown connector types 4315 4316commit 83bbeea0d842094dd647e754608420a404eef6e3 4317Author: Ben Skeggs <bskeggs@redhat.com> 4318Date: Tue Feb 23 15:03:38 2010 +1000 4319 4320 Add connector name for eDP 4321 4322commit 29647021044463768cbfa3eead1416ef1cf27fa1 4323Author: Ben Skeggs <bskeggs@redhat.com> 4324Date: Thu Feb 18 14:17:57 2010 +1000 4325 4326 remove drm patchlevel check, libdrm checks this for us 4327 4328commit 9b4118d6d1fa488db86cd7d2875beea9cdefb096 4329Author: Francisco Jerez <currojerez@riseup.net> 4330Date: Wed Feb 10 18:43:39 2010 +0100 4331 4332 Revert "kms: work around some bong hits with dpms" 4333 4334 This reverts commit 98c9e4edb58374f18249e5c9c53b392fb8b4a1d1. AFAIK 4335 it's no longer needed. 4336 4337 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4338 4339commit f7deabfec0dcec4f203f85cd8494a70fe8fa6ebb 4340Author: Ben Skeggs <bskeggs@redhat.com> 4341Date: Thu Feb 11 08:39:49 2010 +1000 4342 4343 Pass pScreen to drmmode_fbcon_copy() 4344 4345commit 43524fcbb786c5b9e9ec25776560790aec208719 4346Author: Francisco Jerez <currojerez@riseup.net> 4347Date: Wed Feb 10 17:58:24 2010 +0100 4348 4349 dri2: Fix crashes on server regeneration. 4350 4351 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4352 4353commit f50577c117c82eea05e750fecd1ff5653105c648 4354Author: Francisco Jerez <currojerez@riseup.net> 4355Date: Wed Feb 10 13:23:46 2010 +0100 4356 4357 kms: Clean up the set_mode_major hook. 4358 4359 The X server already takes care of restoring the previous CRTC 4360 parameters after a failed mode set. As a bonus this patch fixes 4361 RandR13 projective transforms. 4362 4363 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4364 4365commit b16792b553dee41fc81d74c6e453832807e36368 4366Author: Marcin Kościelnicki <koriakin@0x04.net> 4367Date: Tue Feb 9 20:51:17 2010 +0000 4368 4369 Remove some UMS leftovers. 4370 4371 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4372 4373commit c56bce93f120d0ff18895149d1984896372cdd60 4374Author: Francisco Jerez <currojerez@riseup.net> 4375Date: Wed Feb 10 12:31:54 2010 +0100 4376 4377 kms: Don't pretend it works on xserver-1.5. 4378 4379 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4380 4381commit 0917665d1f2f1e76b6a0e7a4c027512f9f45f41b 4382Author: Francisco Jerez <currojerez@riseup.net> 4383Date: Wed Feb 10 11:48:47 2010 +0100 4384 4385 kms: Kill nv_cursor_convert_cursor. 4386 4387 Well, mostly, the remaining pitch conversion will be unnecessary soon. 4388 4389 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4390 4391commit e75dd23bd28f636b4c1759633d8dbc775e799add 4392Author: Marcin Kościelnicki <koriakin@0x04.net> 4393Date: Fri Feb 5 18:09:55 2010 +0100 4394 4395 nv50/accel: Fix mangled A8+A8 shader. 4396 4397 Turns out we used a misaligned long instruction near the end, and the 4398 shader was getting killed after writing R, A components. This has gone 4399 unnoticed since the remaining G, B outputs aren't actually used. 4400 4401commit 41fc1fdec47fa824c598e1f3d828f4c8c4a2cb71 4402Author: Marcin Kościelnicki <koriakin@0x04.net> 4403Date: Sun Jan 31 19:29:32 2010 +0000 4404 4405 Update to new nouveau_class.h 4406 4407 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4408 4409commit c94544c83ef84e96f8d132ac9ff9296c22c4468e 4410Author: Francisco Jerez <currojerez@riseup.net> 4411Date: Sun Jan 31 18:26:03 2010 +0100 4412 4413 nv10/exa: Fix up nv11 accel. 4414 4415 Fixes an off-by-one error made when UMS was removed. 4416 4417 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4418 4419commit a397b94e32e6b0109096aba09bbe7b5dcfdd8231 4420Author: Francisco Jerez <currojerez@riseup.net> 4421Date: Sun Jan 31 18:11:06 2010 +0100 4422 4423 dri2: Fix for the new DRI2InfoRec ABI. 4424 4425 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4426 4427commit 8b5c67145a9c29cd4c82f411f15e57723d76d0c0 4428Author: Francisco Jerez <currojerez@riseup.net> 4429Date: Sun Jan 31 17:43:54 2010 +0100 4430 4431 nv10/exa: Fix build after the last nouveau_class.h changes. 4432 4433 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4434 4435commit 2630a1553d560ce2c7391bc103d4a543f7d3d0da 4436Author: Marcin Kościelnicki <koriakin@0x04.net> 4437Date: Wed Jan 20 07:46:32 2010 +0000 4438 4439 Remove nv_dripriv.h, nv_local.h 4440 4441commit 2d341f7e0a0dc70a28c6091ef6919fc6c0d60561 4442Author: Ben Skeggs <bskeggs@redhat.com> 4443Date: Thu Jan 21 14:58:37 2010 +1000 4444 4445 xv: use nouveau_bo_new_tile() 4446 4447commit 2a76ff1106ee8c76bd2e16a4954b3c05714d1db3 4448Author: David Heidelberger <d.okias@gmail.com> 4449Date: Mon Jan 18 11:50:19 2010 +0100 4450 4451 Remove deprecated NV1 readme 4452 4453commit fe484a9b635ad56afec7be086a56c692b00dd6d1 4454Author: Ben Skeggs <bskeggs@redhat.com> 4455Date: Wed Jan 13 16:14:41 2010 +1000 4456 4457 exa: remove nouveau_pixmap_offset() 4458 4459 Completely useless now we don't have to try and support non-driver pixmaps. 4460 4461 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4462 4463commit bbf79469a878a8ab5b4c07faadcfdc8d85ceac83 4464Author: Ben Skeggs <bskeggs@redhat.com> 4465Date: Wed Jan 13 16:07:13 2010 +1000 4466 4467 exa: remove nouveau_exa_pixmap_{map,unmap} 4468 4469 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4470 4471commit 18e3a649f7df73bc1e6d199ecf7a62ddcfe37d6e 4472Author: Ben Skeggs <bskeggs@redhat.com> 4473Date: Wed Jan 13 14:26:04 2010 +1000 4474 4475 Check for drm+kms availability in NVPciProbe() 4476 4477 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4478 4479commit 6267b85460ef364aa51a7d33006940bb00a049d9 4480Author: Ben Skeggs <bskeggs@redhat.com> 4481Date: Wed Jan 13 12:03:14 2010 +1000 4482 4483 bump xorg-server requirement to 1.7 4484 4485 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4486 4487commit da5c0dc6040096fd12ddb6a04a22fada89115ccb 4488Author: Ben Skeggs <bskeggs@redhat.com> 4489Date: Mon Jan 11 13:23:06 2010 +1000 4490 4491 remove more unused code 4492 4493 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4494 4495commit 104b3ea842d4cebd993d3e7e9d55ebcadae0cf28 4496Author: Ben Skeggs <bskeggs@redhat.com> 4497Date: Mon Jan 11 12:59:12 2010 +1000 4498 4499 exa: Remove non-driver pixmaps path 4500 4501 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4502 4503commit e8798d91eedc5d8337223b2257ca607da8a197de 4504Author: Ben Skeggs <bskeggs@redhat.com> 4505Date: Mon Jan 11 12:48:00 2010 +1000 4506 4507 Remove pNv->NVArch 4508 4509 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4510 4511commit 7a6f785de1732ec0af429a0b66d5d6ba45427dcb 4512Author: Ben Skeggs <bskeggs@redhat.com> 4513Date: Mon Jan 11 12:00:19 2010 +1000 4514 4515 Remove pNv->Chipset use 4516 4517 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4518 4519commit 68c1f908be5739692c9f0bde9c23f3dc2a7210ea 4520Author: Ben Skeggs <bskeggs@redhat.com> 4521Date: Mon Jan 11 11:56:19 2010 +1000 4522 4523 remove pNv->VRAMPhysical 4524 4525 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4526 4527commit 06499197d6eb07a272a1812e1c8bce886798a1d7 4528Author: Ben Skeggs <bskeggs@redhat.com> 4529Date: Mon Jan 11 11:51:14 2010 +1000 4530 4531 dri1: remove DRI1 support 4532 4533 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4534 4535commit 6975651b0b74dca6edaee759fb1003f65189f1df 4536Author: Ben Skeggs <bskeggs@redhat.com> 4537Date: Mon Jan 11 10:07:11 2010 +1000 4538 4539 xv: ifdef out another problematic piece of overlay support for the moment 4540 4541commit 17485c234ff191cee3dd19e3dd693a80b024e189 4542Author: Ben Skeggs <bskeggs@redhat.com> 4543Date: Wed Jan 6 14:58:15 2010 +1000 4544 4545 Initial pass at removal of non-KMS support. 4546 4547 It's likely there's a lot more that can be cleaned up/removed as a result 4548 of this, we can handle that as we come across it. 4549 4550 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4551 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4552 Acked-by: Maarten Maathuis <madman2003@gmail.com> 4553 4554commit 061cb284b93612848296599759b915a4d66b6d01 4555Author: Maarten Maathuis <madman2003@gmail.com> 4556Date: Sun Jan 3 02:14:36 2010 +0100 4557 4558 exa: Some compat defines for new pixman formats. 4559 4560 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4561 4562commit 8ca5a7383aad0e0557b3c9a1e9edd5431cd509fc 4563Author: Marcin Slusarz <marcin.slusarz@gmail.com> 4564Date: Wed Dec 30 21:46:04 2009 +0100 4565 4566 start using BEGIN_RING_NI 4567 4568commit b259b7ce5e57e1d20111878318b4703a14ec27ec 4569Author: Maarten Maathuis <madman2003@gmail.com> 4570Date: Fri Jan 1 20:16:13 2010 +0100 4571 4572 nv50: always set semaphore handle and offset 4573 4574 - This is in preperation of using semaphores for more than one thing. 4575 4576 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4577 4578commit 8deeed25014b36f8d60f25c7ea2509d88fcb52ae 4579Author: Marcin Slusarz <marcin.slusarz@gmail.com> 4580Date: Mon Dec 28 18:39:16 2009 +0100 4581 4582 nv50: upgrade nv50_texture.h 4583 4584 - copy changes from mesa/src/gallium/drivers/nv50/nv50_texture.h 4585 - remove 2 redefinitions 4586 - fix R<->B rename (see df189c9efc0fbcdce816af483f0147ab635280d1 4587 (nv50: TIC/TSC fixes and additions) in mesa repo) 4588 4589commit 1877ee268774443a58a7bb73244bd70c82d67879 4590Author: Marcin Slusarz <marcin.slusarz@gmail.com> 4591Date: Thu Dec 24 21:29:10 2009 +0100 4592 4593 nv50/exa: add support for more color formats 4594 4595commit 8b3753fc34e3557b91c3bdebe7c8addfb148b3bd 4596Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 4597Date: Wed Dec 30 18:28:25 2009 +0100 4598 4599 nv50: check drawable.bitsPerPixel instead of depth to get format 4600 4601commit 920a346e14cdc40efe9e89c3583d8e573e7e2f87 4602Author: Marcin Kościelnicki <koriakin@0x04.net> 4603Date: Mon Dec 28 16:11:51 2009 +0000 4604 4605 nv50: Dehexify and bring up to date with new method defines. 4606 4607 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4608 4609commit 7c72ac1f47e19af97c997f57012e3af12f941894 4610Author: Francisco Jerez <currojerez@riseup.net> 4611Date: Sat Dec 26 15:56:58 2009 +0100 4612 4613 nv10/exa: Fix for the latest nouveau_class.h changes. 4614 4615 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4616 4617commit 6992d0e7a0cb3c32b16af0b724246e44f7a35d7e 4618Author: Francisco Jerez <currojerez@riseup.net> 4619Date: Thu Dec 17 19:13:08 2009 +0100 4620 4621 dri: Use classic mesa on the fixed pipe cards. 4622 4623 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4624 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4625 4626commit 87b9916dd4b471a9b8b02f166f6553f04f219769 4627Author: Maarten Maathuis <madman2003@gmail.com> 4628Date: Wed Dec 23 21:53:36 2009 +0100 4629 4630 kms: Remove the one-off offset of connector names compared to the kernel. 4631 4632 - Seeing DVI-I-0 in xrandr and DVI-I-1 in the kernel logs can be a source of 4633 confusion. 4634 4635 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4636 4637commit bb1947831d9a4e080b8d1e9dba086af6527ef479 4638Author: Ben Skeggs <bskeggs@redhat.com> 4639Date: Thu Dec 17 13:08:34 2009 +1000 4640 4641 Revert "exa: map the pixmap bo after creation to ensure a useable driver pixmap" 4642 4643 This reverts commit 740c9f3e79729e4e06c693e66826afbad9bd1ce5. 4644 4645 To get the same effect as this commit, update to latest libdrm git. 4646 4647commit 03c7d675f6d12886b6bca8e0375888ecf1b393df 4648Author: Francisco Jerez <currojerez@riseup.net> 4649Date: Fri Dec 11 15:47:05 2009 +0100 4650 4651 dri2: Make the render targets tiled for pre-G80 cards. 4652 4653 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4654 4655commit 963aaeff6b414f371eff3e8ce016f7707c60f7c0 4656Author: Francisco Jerez <currojerez@riseup.net> 4657Date: Fri Dec 11 15:44:47 2009 +0100 4658 4659 exa: Pre-G80 tiling support. 4660 4661 For now, pixmaps will only be tiled if driver pixmaps is being used 4662 and we're told to with the NOUVEAU_CREATE_PIXMAP_TILED usage hint. 4663 4664 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4665 4666commit 41d8d511124fcbe83a3cc68aaf22c3ce0fb4eb8a 4667Author: Francisco Jerez <currojerez@riseup.net> 4668Date: Mon Dec 14 00:51:17 2009 +0100 4669 4670 kms: Allow interlaced/doublescan modes. 4671 4672 After bcafdfbed6 (an X server commit) we have to explicitly state we 4673 can handle them. 4674 4675 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4676 4677commit 740c9f3e79729e4e06c693e66826afbad9bd1ce5 4678Author: Maarten Maathuis <madman2003@gmail.com> 4679Date: Sat Dec 12 22:27:30 2009 +0100 4680 4681 exa: map the pixmap bo after creation to ensure a useable driver pixmap 4682 4683 - Large pixmaps (think of 8000x8000 pixmaps on NV50) will cause crashes 4684 otherwise because exa assumes driver pixmaps to be mappable. Handling this 4685 elsewhere is problematic. 4686 4687 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4688 4689commit 0b5120cf8ef2528a15d88054afdc6fc33eb521f5 4690Author: Ben Skeggs <skeggsb@beleth.(none)> 4691Date: Mon Dec 7 15:31:53 2009 +1000 4692 4693 Allow texturing from GART 4694 4695 Unfortunately NV_CONTEXT_SURFACES_2D on most cards (probably only PCIE cards 4696 work actually, but that's not confirmed) rejects a GART DMA object, so the 4697 usefulness is a bit more limited than I'd hoped. 4698 4699commit 60a394222b410da8bb832935fc525132218b4d94 4700Author: Ben Skeggs <bskeggs@redhat.com> 4701Date: Thu Dec 3 21:14:00 2009 +1000 4702 4703 nv50: avoid touching 0x16b0 on 0x8597, it causes a DATA_ERROR from the GPU 4704 4705 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4706 4707commit 95fc663d5fb653fc0a50aa4d589bbf3b9c79e928 4708Author: Ben Skeggs <bskeggs@redhat.com> 4709Date: Thu Nov 26 13:10:45 2009 +1000 4710 4711 shadowfb: fix segv on fb resize 4712 4713 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4714 4715commit 1cf0cfa926dd78698ff8b8c2d776bcdaa6a867bd 4716Author: Ben Skeggs <bskeggs@redhat.com> 4717Date: Thu Nov 26 10:04:43 2009 +1000 4718 4719 kms: zfill fb if we have no accel to do copy, better than random garbage 4720 4721commit 92578fb4db0feb549bea82d0874ba0e39a2b1ae2 4722Author: Francisco Jerez <currojerez@riseup.net> 4723Date: Thu Nov 5 18:30:28 2009 +0100 4724 4725 nv10/exa: Rename VIEWPORT_SCALE to VIEWPORT_TRANSLATE. 4726 4727 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4728 4729commit 9de0d97bd2fc2ee8800d48b5340a3d495525ad3e 4730Author: Francisco Jerez <currojerez@riseup.net> 4731Date: Thu Nov 5 13:29:09 2009 +0100 4732 4733 nv04-nv40/exa: Match the blob behavior more closely on PrepareCopy. 4734 4735 Fix a somewhat indeterministic corruption problem on nv17 when there 4736 is stuff going on the other fifos (e.g. gallium but I've also 4737 reproduced it with an app just SIFM-ing memory around): in some cases 4738 it made the blits the X server had scheduled fail corrupting the 4739 nearby screen areas. 4740 4741 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4742 Reviewed-by: Ben Skeggs <bskeggs@redhat.com> 4743 4744commit 16ef9ff7e394a693fc74764d8bc2630e784947f3 4745Author: Francisco Jerez <currojerez@riseup.net> 4746Date: Thu Nov 5 03:28:00 2009 +0100 4747 4748 nv10/exa: Spring-cleaning 4749 4750 * Kill the A8+A8 hack. Recent enough X servers (>=1.7) fall back to 4751 ARGB glyphs for drivers not supporting A8 render targets. 4752 4753 * Kill all the global state. It doesn't matter a lot yet but it might 4754 if we get multicard working at some point. 4755 4756 * Other random clean-ups with no functional changes. 4757 4758 Some numbers from x11perf -aa10text -aa24text -comppixwin10 -comppixwin500: 4759 4760 * Before, with A glyphs: 4761 12000000 trep @ 0.0025 msec (394000.0/sec): Char in 80-char aa line (Charter 10) 4762 3200000 trep @ 0.0119 msec ( 84300.0/sec): Char in 30-char aa line (Charter 24) 4763 8000000 trep @ 0.0036 msec (280000.0/sec): Composite 10x10 from pixmap to window 4764 60000 trep @ 0.4813 msec ( 2080.0/sec): Composite 500x500 from pixmap to window 4765 4766 * After, ARGB glyphs: 4767 20000000 trep @ 0.0015 msec (649000.0/sec): Char in 80-char aa line (Charter 10) 4768 4800000 trep @ 0.0057 msec (175000.0/sec): Char in 30-char aa line (Charter 24) 4769 8000000 trep @ 0.0036 msec (280000.0/sec): Composite 10x10 from pixmap to window 4770 60000 trep @ 0.4813 msec ( 2080.0/sec): Composite 500x500 from pixmap to window 4771 4772 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4773 Tested-by: Pekka Paalanen <pq@iki.fi> 4774 4775commit e1c2efd7e7703f5860e5fe99e0413e3be10576d4 4776Author: Ben Skeggs <bskeggs@redhat.com> 4777Date: Thu Nov 5 13:04:17 2009 +1000 4778 4779 kms: don't update drmmode_output->mode_output when vtSema is FALSE 4780 4781 Fixes fast user switching issues. 4782 4783 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4784 4785commit aed811e71b544dccfe84671f962503342acb75ab 4786Author: Ben Skeggs <bskeggs@redhat.com> 4787Date: Wed Nov 4 18:03:12 2009 +1000 4788 4789 exa/nv10: handle reloc failures 4790 4791commit 5315034b7d46980a3527c814526857aed3bc98f7 4792Author: Ben Skeggs <skeggsb@beleth.(none)> 4793Date: Wed Nov 4 16:52:02 2009 +1000 4794 4795 xv/nv04: handle reloc failures + fail on bad surface format 4796 4797commit 99b1bf3d3e6b8c8ef2aa4aae3e33f3908cd00dab 4798Author: Ben Skeggs <skeggsb@beleth.(none)> 4799Date: Wed Nov 4 16:46:35 2009 +1000 4800 4801 xv/nv40: don't even try with unknown surface format 4802 4803commit c1447574a025416f92f8dec2c361859c3945db92 4804Author: Ben Skeggs <skeggsb@beleth.(none)> 4805Date: Wed Nov 4 16:46:18 2009 +1000 4806 4807 xv/nv40: handle reloc failures 4808 4809commit 6ca14ba77104542a2915af8dca56236646be139a 4810Author: Ben Skeggs <skeggsb@beleth.(none)> 4811Date: Wed Nov 4 16:41:58 2009 +1000 4812 4813 xv/nv30: don't even try with an unknown surface format.. 4814 4815commit 53ef7be58a2c46b2b5ffe912dcd408d74705ccbf 4816Author: Ben Skeggs <skeggsb@beleth.(none)> 4817Date: Wed Nov 4 16:41:44 2009 +1000 4818 4819 xv/nv30: handle reloc failures 4820 4821commit 6facc0f765c1130aeb19ddfb76cbaaabaecf3f4a 4822Author: Ben Skeggs <skeggsb@beleth.(none)> 4823Date: Wed Nov 4 16:36:09 2009 +1000 4824 4825 exa/nv40: handle reloc failures 4826 4827commit a97501a4e64d690be4232b01f9a3663fc2338c8a 4828Author: Ben Skeggs <skeggsb@beleth.(none)> 4829Date: Wed Nov 4 16:31:56 2009 +1000 4830 4831 exa/nv30: handle reloc failures 4832 4833commit 222d13a03b41eb610a05063f81f0c47f57e6098e 4834Author: Ben Skeggs <skeggsb@beleth.(none)> 4835Date: Wed Nov 4 16:02:50 2009 +1000 4836 4837 exa/nv04: handle reloc failures 4838 4839commit bd8aa7515f69c412e8f379777c49d59364df32fe 4840Author: Ben Skeggs <bskeggs@redhat.com> 4841Date: Wed Oct 28 12:59:59 2009 +1000 4842 4843 xv: handle reloc failures 4844 4845commit 0214cb8b25a6959573bfc183b259b435f0a728bd 4846Author: Ben Skeggs <bskeggs@redhat.com> 4847Date: Wed Oct 28 12:55:25 2009 +1000 4848 4849 xv/nv50: handle reloc failures 4850 4851commit e84e98d7f9260006a8b16a5785e4d2926d2370d4 4852Author: Ben Skeggs <bskeggs@redhat.com> 4853Date: Wed Oct 28 12:47:49 2009 +1000 4854 4855 exa/nv50: handle reloc failures 4856 4857commit 62c5c875e4c0812815be48f74bdef25d9f046515 4858Author: Ben Skeggs <bskeggs@redhat.com> 4859Date: Wed Oct 28 12:30:24 2009 +1000 4860 4861 exa: handle reloc failures 4862 4863commit c839f6098451aa59c7cb1fd7819e17a966e8ace7 4864Author: Marcin Slusarz <marcin.slusarz@gmail.com> 4865Date: Sun Nov 1 22:16:50 2009 +0100 4866 4867 nv30/exa: support more 16bpp formats 4868 4869 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> 4870 4871commit 1d93e207db6e5a58b702c1812feaeb6b7252bc70 4872Author: Marcin Slusarz <marcin.slusarz@gmail.com> 4873Date: Sat Oct 31 20:24:52 2009 +0100 4874 4875 nv/exa: fix 15/16 bits solid fill 4876 4877 after this change nouveau passes all fill and blend tests of rendercheck 4878 (before: fill - 108/120, blend - 3323868/3569150) 4879 tested on NV34 4880 4881 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> 4882 4883commit df5f17e59a7f42781b388074798d26f7981b2296 4884Author: Arthur HUILLET <arthur.huillet@free.fr> 4885Date: Mon Nov 2 15:56:56 2009 +0100 4886 4887 nv10exa: improve codestyle 4888 4889 Signed-off-by: Arthur HUILLET <arthur.huillet@free.fr> 4890 4891commit 5587f40c1b8af7d178f9a68d0f1fecdfc0ca9749 4892Author: Ben Skeggs <bskeggs@redhat.com> 4893Date: Fri Oct 30 14:01:09 2009 +1000 4894 4895 exa: force the use of sysmem pixmaps on low-mem cards 4896 4897 Very similar effect to forcing MigrationHeuristic "greedy" on classic 4898 EXA. Far better than the migration ping-pong that'd occur otherwise 4899 4900 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4901 4902commit 718a41b9a2b5649ac37c18f456b94eaf99584598 4903Author: Ben Skeggs <bskeggs@redhat.com> 4904Date: Thu Oct 22 12:35:32 2009 +1000 4905 4906 nv50: NVA8's 3D object class is 0x8597 4907 4908commit 7494b18bf4061fb960a2126c745721161baddd01 4909Author: Francisco Jerez <currojerez@riseup.net> 4910Date: Sun Oct 18 19:45:02 2009 +0200 4911 4912 nv50/crtc: Don't perform the LUT expansion twice. 4913 4914 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4915 4916commit b161a7f7fa81b6d499c91c17f988a38ed537ed35 4917Author: Francisco Jerez <currojerez@riseup.net> 4918Date: Sun Oct 18 19:41:11 2009 +0200 4919 4920 nv04-nv40/crtc: Don't perform the LUT expansion twice. 4921 4922 Fixes bug 24595. 4923 4924 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4925 4926commit 9cbc139176ae991150bd9ddf66b61db471b391f2 4927Author: Francisco Jerez <currojerez@riseup.net> 4928Date: Sun Oct 11 23:42:01 2009 +0200 4929 4930 Partially revert "kms: always set gamma on modeset" 4931 4932 6c8d3e39b72abe897b08d9e90cae85e121863d2c breaks gamma setting with 4933 servers older than 1.7. Keep the call to xf86HandleColormaps even with 4934 KMS enabled, as it's a no-op in the cases we don't need it. 4935 4936 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4937 4938commit cef4b61b4d0cf096428d89eff65a78fd16f612d5 4939Author: Francisco Jerez <currojerez@riseup.net> 4940Date: Sun Oct 11 20:23:45 2009 +0200 4941 4942 Fix NVLoadPalette for 15/16 bit depth. 4943 4944 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4945 4946commit cc4f7a733e3504260437d06091490fcb8a4c8063 4947Author: Francisco Jerez <currojerez@riseup.net> 4948Date: Sun Oct 11 20:31:11 2009 +0200 4949 4950 Remove an useless call to NVLeaveVT from NVCloseScreen. 4951 4952 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4953 4954commit 5a4afade51fe5fd149f8726fc243fbb69a660bf3 4955Author: Francisco Jerez <currojerez@riseup.net> 4956Date: Thu Oct 8 03:33:34 2009 +0200 4957 4958 nv10/exa: Replace "celcius" with "celsius". 4959 4960 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4961 Tested-by: Pekka Paalanen <pq@iki.fi> 4962 4963commit 2e744325827d702b564c5d6becaf65f3ad164761 4964Author: Francisco Jerez <currojerez@riseup.net> 4965Date: Sun Oct 11 17:40:56 2009 +0200 4966 4967 nv10/exa: Allow component-alpha masks. 4968 4969 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4970 Tested-by: Pekka Paalanen <pq@iki.fi> 4971 4972commit c3bb746ab6b371e843e1754a919dd25fa7b96b19 4973Author: Francisco Jerez <currojerez@riseup.net> 4974Date: Sun Oct 11 17:38:44 2009 +0200 4975 4976 nv10/exa: Fix PictOpAtopReverse for destination pixmaps with no alpha channel. 4977 4978 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4979 Tested-by: Pekka Paalanen <pq@iki.fi> 4980 4981commit f0a3d9f59817ebf2b9606097785f0ad1ae45da12 4982Author: Francisco Jerez <currojerez@riseup.net> 4983Date: Sun Oct 11 16:15:30 2009 +0200 4984 4985 nv10/exa: Refactor the reg combiner/alpha blending code. 4986 4987 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4988 Tested-by: Pekka Paalanen <pq@iki.fi> 4989 4990commit 3f020b0bf2e8d2bc54eceda771829c4b2535688a 4991Author: Ben Skeggs <bskeggs@redhat.com> 4992Date: Thu Oct 8 16:19:16 2009 +1000 4993 4994 kms: reload cursor on modeset, needed on rotation (rh#524632) 4995 4996commit 38d7520986078fd01c19957a75ebfd9d91665225 4997Author: Ben Skeggs <bskeggs@redhat.com> 4998Date: Thu Oct 8 11:34:20 2009 +1000 4999 5000 kms: switch to CP+MPH instead of GetScratchPixmapHeader 5001 5002commit 6c8d3e39b72abe897b08d9e90cae85e121863d2c 5003Author: Ben Skeggs <bskeggs@redhat.com> 5004Date: Wed Oct 7 15:57:49 2009 +1000 5005 5006 kms: always set gamma on modeset 5007 5008 fixes issues vt-switching between console and X of different depths. 5009 5010commit b13c0509be63f27a4a9c918112d0a0fc84ccce10 5011Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 5012Date: Wed Sep 16 18:17:30 2009 +0200 5013 5014 nv50/exa: use A8 RT format for PICT_a8 5015 5016 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 5017 5018commit 7a63e4c61cd5d082caf074702cde79e1a66354b0 5019Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 5020Date: Tue Oct 6 11:58:39 2009 +0200 5021 5022 nv50/exa: cleanup shader settings 5023 5024 Since we only pass through 2D coordinates in the VP, 5025 no need to occupy extra output registers for HPOS z 5026 and, which can be coded as 0, 1 into VP_RESULT_MAP. 5027 5028 We don't have to fetch all 4 attribute components if 5029 we only read 2 in the VPs. 5030 5031 Using 0x1, 0x1 as terminator instructions looks odd, 5032 just make the last ones that do something long and 5033 set the exit bit. 5034 5035 Setup VP_RESULT_MAP and INTERPOLANT_CTRL correctly. 5036 5037 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 5038 5039commit 60818bdeb975a9f469d3ae890932d4de1cbb59f9 5040Author: Ben Skeggs <bskeggs@redhat.com> 5041Date: Wed Oct 7 10:49:00 2009 +1000 5042 5043 kms: don't copy fbcon if depths don't match 5044 5045commit dd8339fc4e18fa4fd409f2997db741be209e852d 5046Author: Ben Skeggs <bskeggs@redhat.com> 5047Date: Tue Sep 29 10:52:53 2009 +1000 5048 5049 Pass NULL fb pointer to fbScreenInit when using EXA 5050 5051 Fixes 1.7 xserver getting very confused... 5052 5053commit 6dbdc80016c2a81c390d77db613833cd4d15c617 5054Author: Ben Skeggs <bskeggs@redhat.com> 5055Date: Tue Sep 29 09:00:59 2009 +1000 5056 5057 exa: make tiled scanout a separate condition from wfb being used 5058 5059commit 96d025232e292a8b1aebd54eaad3a21dbf6e2f77 5060Author: Maarten Maathuis <madman2003@gmail.com> 5061Date: Fri Sep 25 19:50:11 2009 +0200 5062 5063 nv40/exa: fallback on special case xrgb non-repeat textures 5064 5065commit ed05bd6dab4fdd1289947154df02cd21c0839574 5066Author: Maarten Maathuis <madman2003@gmail.com> 5067Date: Fri Sep 25 19:49:14 2009 +0200 5068 5069 nv30/exa: fallback on special case xrgb non-repeat textures 5070 5071commit dd7b86cb5351a60f21eee9cf81ae0ef61cf45f13 5072Author: Maarten Maathuis <madman2003@gmail.com> 5073Date: Fri Sep 25 19:39:24 2009 +0200 5074 5075 exa: fallback on src pictures without drawable 5076 5077commit de0b095a1deaa1e53e1f555a6eb537482c488638 5078Author: Maarten Maathuis <madman2003@gmail.com> 5079Date: Thu Sep 24 19:21:59 2009 +0200 5080 5081 nv50/exa: avoid fallback on xrgb repeat none when dst has no alpha 5082 5083 - This should fix resizing slowness in evince amongst other things. 5084 5085commit df94ebdbcd89c1678ac243217e7f5b20cbbe857c 5086Author: Ben Skeggs <bskeggs@redhat.com> 5087Date: Mon Sep 21 08:37:27 2009 +1000 5088 5089 kms: use displayWidth for pitch calc in transition, not virtualX (rh#523281) 5090 5091commit 5a6c3c427796f7f3dac0f597a320b09feb25ffc8 5092Author: Ben Skeggs <bskeggs@redhat.com> 5093Date: Wed Sep 16 15:13:08 2009 +1000 5094 5095 dri2: call CreatePixmap rather than yet another custom-built one 5096 5097commit 95086de507df7c5676c7ee712139ac4c91c9f6d2 5098Author: Pekka Paalanen <pq@iki.fi> 5099Date: Tue Sep 15 20:31:33 2009 +0300 5100 5101 Partially revert "kms: wait for fbcon to be copied before continuing" 5102 5103 This partially reverts commit 6c045fc44783454180d7b3d00b5f25436bd5544e. 5104 5105 The nv_driver.c hunk of the above patch breaks the build on at least git 5106 and 1.6.3.901 xorg-servers due to: 5107 nv_driver.c:745: error: 'struct _ScrnInfoRec' has no member named 5108 'canDoBGNoneRoot' 5109 5110 Signed-off-by: Pekka Paalanen <pq@iki.fi> 5111 5112commit 6c045fc44783454180d7b3d00b5f25436bd5544e 5113Author: Ben Skeggs <bskeggs@redhat.com> 5114Date: Mon Sep 14 13:04:12 2009 +1000 5115 5116 kms: wait for fbcon to be copied before continuing 5117 5118commit 22971d5545a2cd03e5c0b5d5d16fb85d7fff2aff 5119Author: Ben Skeggs <bskeggs@redhat.com> 5120Date: Thu Sep 10 16:05:15 2009 +1000 5121 5122 dri2: switch to bo_new_tile() 5123 5124commit 806eaf6b0b36cb05ca9d883ff4572629812a1d48 5125Author: Ben Skeggs <bskeggs@redhat.com> 5126Date: Thu Sep 10 12:00:44 2009 +1000 5127 5128 kms: rework fbcon transition 5129 5130 This does 2 things. Most importantly, it avoids a situation where we get 5131 multiple GEM handles for a single object, making it possible to confuse 5132 the kernel if both appear in the same push buffer. 5133 5134 Secondly, it moves the transition to only on X startup, and not on every 5135 modeset etc. 5136 5137commit f90edc168c4a18b20750043d33c193bdc1b89156 5138Author: Ben Skeggs <bskeggs@redhat.com> 5139Date: Wed Sep 9 09:23:58 2009 +1000 5140 5141 Revert "nv50: use 0x7000 everywhere for now, until we can fix mixed mode alignment" 5142 5143 This reverts commit c80f125a97003c2d0db77f54e22d9de447a06f63. 5144 5145 This was not supposed to sneak in.. 5146 5147commit ff6bdb14ac13168c1a4b7866784e4833a1162dbc 5148Author: Ben Skeggs <bskeggs@redhat.com> 5149Date: Wed Sep 9 09:19:42 2009 +1000 5150 5151 Revert "kms: fetch active encoder during init if there is one" 5152 5153 This reverts commit 4f66fc6e871969448eba25817bfb383f2fdd4b34. 5154 5155 See reason in previous commit. 5156 5157commit 9708e50e820970d526b084e0ab736c1c7a4f50c0 5158Author: Ben Skeggs <bskeggs@redhat.com> 5159Date: Wed Sep 9 09:18:48 2009 +1000 5160 5161 Revert "kms: update encoder config during detect" 5162 5163 This reverts commit 25610704d4b16293f5bc24d5ad18cc76e20896d2. 5164 5165 Very bad thinko, drm_mode_get_connector.encoder_id is *not* the same thing 5166 as our DRM's nouveau_connector.detected_encoder. Need to find a better 5167 way of achieving this. 5168 5169commit c80f125a97003c2d0db77f54e22d9de447a06f63 5170Author: Ben Skeggs <bskeggs@redhat.com> 5171Date: Tue Sep 8 16:20:33 2009 +1000 5172 5173 nv50: use 0x7000 everywhere for now, until we can fix mixed mode alignment 5174 5175commit 25610704d4b16293f5bc24d5ad18cc76e20896d2 5176Author: Ben Skeggs <bskeggs@redhat.com> 5177Date: Tue Sep 8 11:33:51 2009 +1000 5178 5179 kms: update encoder config during detect 5180 5181commit 4f66fc6e871969448eba25817bfb383f2fdd4b34 5182Author: Ben Skeggs <bskeggs@redhat.com> 5183Date: Tue Sep 8 11:26:29 2009 +1000 5184 5185 kms: fetch active encoder during init if there is one 5186 5187commit 2b5ec6ad708164eb9b2df00f453ea84f4898153a 5188Author: Ben Skeggs <bskeggs@redhat.com> 5189Date: Fri Sep 4 10:35:57 2009 +1000 5190 5191 kms: only call drmModeSetCursor when cursor is visible 5192 5193 It seems we get load_cursor_argb() calls even when the cursor is hidden, 5194 and the drmModeSetCursor() call would then decide a cursor is available 5195 and show it on outputs where the cursor shouldn't have been actuve. 5196 5197commit 569a17a98d03d965f9500c37c0cb14e0fb462311 5198Author: Francisco Jerez <currojerez@riseup.net> 5199Date: Thu Aug 20 18:44:38 2009 +0200 5200 5201 kms: No need to upload gamma on set_mode_major in older X servers. 5202 5203 Fixes build for CRTC ABI < 3. 5204 5205 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 5206 5207commit 32fc8a9a3104b38fe89427ffa60d21dad1572c15 5208Author: Francisco Jerez <currojerez@riseup.net> 5209Date: Thu Aug 20 18:43:58 2009 +0200 5210 5211 kms: Fix <nv11 hardware cursor. 5212 5213 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 5214 5215commit e4e6f863b626d77934935e13714c6cbfd014d916 5216Author: Maarten Maathuis <madman2003@gmail.com> 5217Date: Thu Aug 20 12:04:26 2009 +0200 5218 5219 kms: actually call kernel when loading cursor image 5220 5221 - In the past we depended on getting a cursor show to call the kernel and actually make it visible. 5222 5223commit 95f09df23efd0a36d29cb4810faa659f839194f0 5224Author: Ben Skeggs <bskeggs@redhat.com> 5225Date: Thu Aug 20 15:48:26 2009 +1000 5226 5227 kms: prevent X from hiding cursor before updating image 5228 5229 This wasn't noticeable when everything was being done in userspace. However, 5230 with KMS there's more of a delay between the hide<->update<->show steps 5231 and this produces a noticeable flicker. 5232 5233 If there's any good reason we shouldn't be updating the cursor image while 5234 it's being shown, we'll handle it transparently in the DRM as required. 5235 5236commit fe2b5e6c6f81f4c4a76f5e4c5816812459872af3 5237Author: Maarten Maathuis <madman2003@gmail.com> 5238Date: Wed Aug 19 22:06:42 2009 +0200 5239 5240 kms: fix gamma 5241 5242 - crtc->active is needed in order to set the gamma manually 5243 - the initial gamma must be set by us 5244 5245commit 9d270f65ee85dedc6bf9cd9cf24e3d8df062f670 5246Author: Maarten Maathuis <madman2003@gmail.com> 5247Date: Tue Aug 18 21:59:10 2009 +0200 5248 5249 exa: update to latest CreatePixmap2 + do not align width to 64 5250 5251 - Only the pitch needs to be aligned to 64. 5252 5253commit b50110d00e32396e186168271adbb88ace8ba986 5254Author: Francisco Jerez <currojerez@riseup.net> 5255Date: Mon Aug 17 16:54:48 2009 +0200 5256 5257 kms: Some TV-out fixes. 5258 5259commit 523feba8fdc6d285929ac72be0a6e037019f019e 5260Author: Maarten Maathuis <madman2003@gmail.com> 5261Date: Mon Aug 17 00:09:42 2009 +0200 5262 5263 wfb: more correct and more error tolerant 5264 5265 - In a strange world it is possible that a high index is tiled, and a lower index empty. 5266 - This should avoid that loophole and also emit a warning when we run out of indices. 5267 5268commit 3db3e5350e2f6dea50cd2bce0c5fb0437fe08dcf 5269Author: Francisco Jerez <currojerez@riseup.net> 5270Date: Wed Aug 12 02:18:27 2009 +0200 5271 5272 kms: Implement output->get_property when RandR1.3 is available. 5273 5274 This allows getting updated values when the kernel is modifying them 5275 behind our back. 5276 5277commit ba9817b88d31364a5d43d945359e357535aeb09a 5278Author: Francisco Jerez <currojerez@riseup.net> 5279Date: Wed Aug 12 02:18:26 2009 +0200 5280 5281 kms: Don't hardcode the output properties 5282 5283 Replicate any properties the kernel exposes. Mostly taken from the 5284 intel DDX. 5285 5286commit 92f271020a1b9486bd5890a00649aed3678ead62 5287Author: Ben Skeggs <bskeggs@redhat.com> 5288Date: Thu Aug 13 15:40:52 2009 +1000 5289 5290 nv50/xv: provide SYNC_TO_VBLANK attribute 5291 5292commit ada1d36d7097c389c8e21707d36b3cdd6378db94 5293Author: Ben Skeggs <bskeggs@redhat.com> 5294Date: Thu Aug 13 15:38:14 2009 +1000 5295 5296 xv: disable overlay with kms for the moment 5297 5298commit 85b1c86983a39cb921bd68a8601785975c948e2d 5299Author: Ben Skeggs <bskeggs@redhat.com> 5300Date: Mon Aug 10 11:42:15 2009 +1000 5301 5302 dri2: call exaMoveInPixmap() before accessing driver pixmap private 5303 5304commit 292f46a0a0e031fc8a4256aa1eed91e3b2167acd 5305Author: Ben Skeggs <bskeggs@redhat.com> 5306Date: Mon Aug 10 11:36:56 2009 +1000 5307 5308 xv: call exaMoveInPixmap() for driver pixmaps 5309 5310commit 0210fbf01d61fb0541770d541b71396537bcb728 5311Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 5312Date: Sun Aug 9 14:53:28 2009 +0200 5313 5314 nv50/exa: fix pointer increment in UploadSIFC 5315 5316 We always output 'size' dwords to the ring buffer, so the pointer 5317 should be advanced accordingly, independent of cpp. 5318 5319commit 1ac7d83c1e46a164c4a3a3688de1bcde9e5a58e0 5320Author: Maarten Maathuis <madman2003@gmail.com> 5321Date: Sat Aug 8 19:41:02 2009 +0200 5322 5323 wfb: don't default to linear when encountering a pixmap without bo 5324 5325 - With exa "mixed" there are actually very valid cases where this happens. 5326 - This fixes certain types of corruptions, like the yellow and black in a gimp window. 5327 5328commit 64790e950e7f46e6f4f5b8486f8d0bd929cb5504 5329Author: Maarten Maathuis <madman2003@gmail.com> 5330Date: Fri Aug 7 16:12:41 2009 +0200 5331 5332 Revert "nv50: enforce certain allignments on a tiled frontbuffer, as well as overallocate it a bit" 5333 5334 This reverts commit dd72640c8133ee78304f875023134f214d94a630. 5335 5336 Moved to the kernel. 5337 5338commit dd72640c8133ee78304f875023134f214d94a630 5339Author: Maarten Maathuis <madman2003@gmail.com> 5340Date: Fri Aug 7 11:56:29 2009 +0200 5341 5342 nv50: enforce certain allignments on a tiled frontbuffer, as well as overallocate it a bit 5343 5344 - There is reason to believe that with driver modifications (part of) this won't be needed. 5345 - Until then this should fix "tiles" appearing at the upper left or lower right corner. 5346 5347commit bfbe4d33679cc25939db0da729d89ba2a4295764 5348Author: Ben Skeggs <bskeggs@redhat.com> 5349Date: Fri Aug 7 16:02:27 2009 +1000 5350 5351 exa: default to enabling driver pixmaps if everything else is new enough 5352 5353commit 30ddd08bb2739514edf6d6e137f533b34a4369d2 5354Author: Ben Skeggs <bskeggs@redhat.com> 5355Date: Fri Aug 7 16:01:27 2009 +1000 5356 5357 exa: turn on new exa migration mode for driver pixmaps 5358 5359commit 712064e6188d65e2c7d85078b95a0292edcad0e7 5360Author: Ben Skeggs <bskeggs@redhat.com> 5361Date: Wed Aug 5 14:00:51 2009 +1000 5362 5363 nv04-nv40/exa: set cliprect to drawing area for IFC 5364 5365commit 7f63da83d8622176114ecd55048736fecd703322 5366Author: Ben Skeggs <bskeggs@redhat.com> 5367Date: Wed Aug 5 12:04:46 2009 +1000 5368 5369 dri: close before dropping master 5370 5371commit 0cc44b713ec07dfb56a4b4a6d7ae1c1a75c74971 5372Author: Ben Skeggs <bskeggs@redhat.com> 5373Date: Wed Aug 5 11:57:01 2009 +1000 5374 5375 Set/Drop master in Enter/LeaveVT 5376 5377commit 087e9f61b81ea90a1591d8b43e29bda091156405 5378Author: Ben Skeggs <bskeggs@redhat.com> 5379Date: Wed Aug 5 11:51:54 2009 +1000 5380 5381 dri: close before taking down accel 5382 5383 Fixes "failed to destroy server context" errors on exit. 5384 5385commit 96ac8c5a9473d81b737afa116f4fd388e3152f50 5386Author: Ben Skeggs <bskeggs@redhat.com> 5387Date: Wed Aug 5 11:36:39 2009 +1000 5388 5389 nv04-nv40: deal with framebuffer resize in !kms_enable mode 5390 5391commit b8ea3b41b0540635585bc68334691414f7cff53d 5392Author: Ben Skeggs <bskeggs@redhat.com> 5393Date: Wed Aug 5 11:30:51 2009 +1000 5394 5395 nv50: deal with framebuffer resize for !kms_enable 5396 5397commit d0dcc89825acb4b480f3849d8896b17a80df02f4 5398Author: Ben Skeggs <bskeggs@redhat.com> 5399Date: Wed Aug 5 11:28:45 2009 +1000 5400 5401 No wfb without KMS 5402 5403commit 2c7502680e1ab3a4ea4869a78d0317746084dd92 5404Author: Ben Skeggs <bskeggs@redhat.com> 5405Date: Wed Aug 5 08:55:09 2009 +1000 5406 5407 dri2: fix CopyRegion() 5408 5409commit 2a4c93a064353cf61c30dd1e96f93203895beb15 5410Author: Ben Skeggs <bskeggs@redhat.com> 5411Date: Wed Aug 5 08:36:26 2009 +1000 5412 5413 dri2: merge DRI2BufferRec and nouveau_dri2_buffer into a single struct 5414 5415commit 4f386f983fde197e9136ffe6fbf941a343996897 5416Author: Ben Skeggs <bskeggs@redhat.com> 5417Date: Wed Aug 5 08:20:46 2009 +1000 5418 5419 dri2: remove support for < version 3 5420 5421commit 0f691257e6d2c3662f9c76adc3287f5320a8822d 5422Author: Ben Skeggs <bskeggs@redhat.com> 5423Date: Mon Aug 3 11:30:46 2009 +1000 5424 5425 zfill scanout buffer, not offscreen 5426 5427commit d893600356b3480d3df536a04e9860bfada5a159 5428Author: Maarten Maathuis <madman2003@gmail.com> 5429Date: Sun Aug 2 19:43:58 2009 +0200 5430 5431 kms: remove fb in NVCloseScreen otherwise server regeneration fails 5432 5433 - The bo was deleted, but fb_id remained non-zero, so it never picked up the new one. 5434 5435commit ac2a644773e5b160b6794a252b85c0bf8acc5169 5436Author: Ben Skeggs <bskeggs@redhat.com> 5437Date: Sun Aug 2 20:10:35 2009 +1000 5438 5439 xv: really fix offscreen pixmap calculation 5440 5441 sleep.. good.. 5442 5443commit 60635e5073e69c6f690dc3c876772b5478013783 5444Author: Ben Skeggs <bskeggs@redhat.com> 5445Date: Sun Aug 2 19:42:17 2009 +1000 5446 5447 xv: fix check for offscreen pixmaps 5448 5449commit 1ecd4358c851b3d7bebfb00f5f7ac3c7c151bc20 5450Author: Ben Skeggs <bskeggs@redhat.com> 5451Date: Fri Jul 31 15:30:08 2009 +1000 5452 5453 Recalculate displayWidth if we do a late NoAccel fallback 5454 5455 Fixes G80, where we'd have calculated the pitch for tiled scanout. 5456 5457commit 08ef8aaf2bd7b9565cc36679412b721499f9a367 5458Author: Ben Skeggs <bskeggs@redhat.com> 5459Date: Fri Jul 31 15:23:49 2009 +1000 5460 5461 Don't allocate offscreen memory or GART when using NoAccel 5462 5463commit 64a1515fab9764d0ada304426e7f0d7ec995b84e 5464Author: Ben Skeggs <bskeggs@redhat.com> 5465Date: Fri Jul 31 13:49:49 2009 +1000 5466 5467 Fix NoAccel/ShadowFB with fb resize 5468 5469commit e9034a2e8d8285cf45cd1410b9e8933991f82a2b 5470Author: Ben Skeggs <bskeggs@redhat.com> 5471Date: Fri Jul 31 13:29:11 2009 +1000 5472 5473 NoAccel + wfb == bad idea 5474 5475commit 69a40131b700a4fbc4de9432110045870de0806e 5476Author: Ben Skeggs <bskeggs@redhat.com> 5477Date: Fri Jul 31 13:28:02 2009 +1000 5478 5479 Fix up screen pixmap's devPrivate.ptr after resize 5480 5481commit 4ea65c594a1f10f586da2ca32e499de2ea0a2605 5482Author: Ben Skeggs <bskeggs@redhat.com> 5483Date: Fri Jul 31 13:25:45 2009 +1000 5484 5485 Point fb at the correct buffer.. 5486 5487commit c6b98608daec4162a7d23ce8c1d50a4a9c3ca3dc 5488Author: Ben Skeggs <bskeggs@redhat.com> 5489Date: Fri Jul 31 13:09:56 2009 +1000 5490 5491 Switch on can_resize for non-kms 5492 5493commit 4bb007ebb8e60c87c815f9c4ce63b8d1d46e8b75 5494Author: Ben Skeggs <bskeggs@redhat.com> 5495Date: Fri Jul 31 13:03:45 2009 +1000 5496 5497 Fall back to NoAccel if channel/object creation fails 5498 5499commit 641fdacbd1aeb9882caa30da6d568ff9a0bd78f5 5500Author: Ben Skeggs <bskeggs@redhat.com> 5501Date: Fri Jul 31 11:46:47 2009 +1000 5502 5503 nv50/exa: use scissored triangle for composite too 5504 5505commit c6c0ef1b5a837838a0fc2e8d1659b1b4d76d26b0 5506Author: Ben Skeggs <bskeggs@redhat.com> 5507Date: Fri Jul 31 11:23:04 2009 +1000 5508 5509 nv50/xv: enable sync-to-vblank 5510 5511commit b3d6861fa482a92807aadd9222ca2c8c605beaeb 5512Author: Maarten Maathuis <madman2003@gmail.com> 5513Date: Thu Jul 30 17:23:25 2009 +0200 5514 5515 nv50_xv: switch to a scissored triangle to avoid strange artifacts when vsync is enabled 5516 5517commit 9b37bc96d86882b7f9c355e92cbeff517090a98d 5518Author: Ben Skeggs <bskeggs@redhat.com> 5519Date: Thu Jul 30 19:24:19 2009 +1000 5520 5521 more cleanup 5522 5523commit d8c10eb14f8d6b5cad88a39a5da7104a403060cd 5524Author: Ben Skeggs <bskeggs@redhat.com> 5525Date: Thu Jul 30 10:17:38 2009 +1000 5526 5527 non-kms fb resize 5528 5529commit 3b1a861aa56f823e8d26f33b7139e290bf44b79d 5530Author: Ben Skeggs <bskeggs@redhat.com> 5531Date: Thu Jul 30 07:14:17 2009 +1000 5532 5533 free scanout buffer in NVUnmapMem() 5534 5535commit 4d20547df30d630b94b0cfc28dcf7ead564f7369 5536Author: Ben Skeggs <bskeggs@redhat.com> 5537Date: Tue Jul 28 18:14:03 2009 +1000 5538 5539 nv50: tile scanout buffer for "classic" exa if xserver new enough 5540 5541commit 03bc3a604e5c12829f81f2582374d37012da7f36 5542Author: Ben Skeggs <bskeggs@redhat.com> 5543Date: Tue Jul 28 11:45:51 2009 +1000 5544 5545 kms: only use rounded height for allocation 5546 5547 Fixes minor cursor corruption in some circumstances with driver pixmaps. 5548 5549commit e4fb3e7064df14b21a6ea34b919ccf2bf9fa0ca2 5550Author: Ben Skeggs <bskeggs@redhat.com> 5551Date: Tue Jul 28 11:27:09 2009 +1000 5552 5553 More fixing of pitches + use 0x7000 tile_flags for tiled 16bpp on nv50 5554 5555commit 97cca7630441876414da5dfabb9ad9f2b208e29d 5556Author: Ben Skeggs <skeggsb@beleth.keine.ath.cx> 5557Date: Tue Jul 28 10:21:11 2009 +1000 5558 5559 kms: implement rotation support without driver pixmaps 5560 5561commit 786ff2994aa4f614b7cc9d466455eef183697d42 5562Author: Ben Skeggs <skeggsb@beleth.keine.ath.cx> 5563Date: Tue Jul 28 09:43:00 2009 +1000 5564 5565 kms: simplify drmmode_is_rotate_pixmaps args 5566 5567commit ff62da668a468796e4001523958e73eb2060b7f2 5568Author: Ben Skeggs <bskeggs@redhat.com> 5569Date: Mon Jul 27 22:07:38 2009 +1000 5570 5571 dri1: dodgyness to "survive" a framebuffer resize (or rather, move) 5572 5573 This can never work well without a lot of work, which I'm not willing to 5574 put in for DRI1. This will at least let it keep sort-of working, any DRI1 5575 apps running when a resize event occurs will continue rendering into the 5576 old framebuffer. 5577 5578commit b77d580ebcc9b371153774010d29038645effffb 5579Author: Ben Skeggs <bskeggs@redhat.com> 5580Date: Mon Jul 27 21:43:25 2009 +1000 5581 5582 We already map offscreen and store the virtual, no need to keep doing it 5583 5584commit bdf2c9aee02a100b44c083347a2415c100bd2546 5585Author: Ben Skeggs <bskeggs@redhat.com> 5586Date: Mon Jul 27 21:35:19 2009 +1000 5587 5588 kms: implement framebuffer resize without driver pixmaps 5589 5590commit 595b1c74e52f16ef482be6f14e4a33b4649b8bfd 5591Author: Ben Skeggs <bskeggs@redhat.com> 5592Date: Mon Jul 27 07:44:41 2009 +1000 5593 5594 Fix build with 0.0.15 5595 5596commit 9b879b9b38809e7d206c9e78610836bdd9c2d09b 5597Author: Ben Skeggs <bskeggs@redhat.com> 5598Date: Mon Jul 27 21:09:42 2009 +1000 5599 5600 Cleanup 5601 5602commit 0336eb15ebc351ab6900b48972e12f5cc6abf1bb 5603Author: Ben Skeggs <bskeggs@redhat.com> 5604Date: Mon Jul 27 17:52:15 2009 +1000 5605 5606 Switch to using scanout buffer detached from EXA offscreen memory area 5607 5608commit 0403cf3698ecb9a290bb4b7e24d0deedc957ad2a 5609Author: Ben Skeggs <bskeggs@redhat.com> 5610Date: Mon Jul 27 17:45:29 2009 +1000 5611 5612 Allocate scanout buffer independently of EXA offscreen area 5613 5614 This is just wasted memory for the moment... 5615 5616commit f8ac9e35bd20391f923c5c5ccc509251ea23a4c5 5617Author: Ben Skeggs <bskeggs@redhat.com> 5618Date: Mon Jul 27 17:30:54 2009 +1000 5619 5620 Remove NoDRM mode 5621 5622commit e259925bbdcebe4c3d136f476fc54f426eea9261 5623Author: Ben Skeggs <bskeggs@redhat.com> 5624Date: Mon Jul 27 17:14:00 2009 +1000 5625 5626 xv: fix some badness in nv3x/nv4x textured video adaptors 5627 5628commit 3a10daca16752b52ec3187deb8b1c25d6c6fd641 5629Author: Ben Skeggs <bskeggs@redhat.com> 5630Date: Mon Jul 27 16:05:03 2009 +1000 5631 5632 rename pNv->FB to pNv->offscreen 5633 5634 Not strictly true right now, but it's about to be. 5635 5636commit 91bfda81982a77a79c2baee39929711b38b1ae6a 5637Author: Ben Skeggs <bskeggs@redhat.com> 5638Date: Mon Jul 27 11:46:51 2009 +1000 5639 5640 exa: allow acceleration to rotation bo 5641 5642commit d59aaa81590b52b6227b69028f08d8ab62505525 5643Author: Ben Skeggs <bskeggs@redhat.com> 5644Date: Mon Jul 27 11:02:44 2009 +1000 5645 5646 exa: use PixmapIsOffscreen hook for non-driver pixmaps path too 5647 5648commit c442376d4e9a8296760c1171fee5ca155534d011 5649Author: Ben Skeggs <bskeggs@redhat.com> 5650Date: Mon Jul 27 07:44:14 2009 +1000 5651 5652 Fix typo exposed by removal of NV_* card type defines from DRM header 5653 5654commit 15bfc676094d19e57711bbdff40edd5c8e8592d6 5655Author: Ben Skeggs <bskeggs@redhat.com> 5656Date: Mon Jul 27 07:40:20 2009 +1000 5657 5658 Warning fixes 5659 5660commit 6eb790b6113a999727895694d01910b83348647a 5661Author: Ben Skeggs <bskeggs@redhat.com> 5662Date: Fri Jul 24 13:54:49 2009 +1000 5663 5664 kms: keep rotate buffer pitch in crtc struct, rather than recalculating 5665 5666commit 25f29426293fcac2e7a17b986673b2874e7a7264 5667Author: Ben Skeggs <bskeggs@redhat.com> 5668Date: Fri Jul 24 13:18:22 2009 +1000 5669 5670 Wrap CreateScreenResources and set screen pixmap's BO there 5671 5672 Avoids reallocating pNv->FB yet again when we hit the resize hook. 5673 5674commit d1628dc9159343400f09f696b9c4ef9d0fe1acbf 5675Author: Maarten Maathuis <madman2003@gmail.com> 5676Date: Tue Jul 28 10:08:40 2009 +0200 5677 5678 nouveau: remove some more RAC bits 5679 5680 - The assumption is that RAC is dead anyway on post-libpciaccess xservers. 5681 5682commit 75fba5dfd2a00f3b8212d0e8af544273c14ee0c5 5683Author: Dave Airlie <airlied@redhat.com> 5684Date: Tue Jul 28 13:53:06 2009 +1000 5685 5686 nouveau: rip out resources code whats left at least 5687 5688commit 094e4e075f8e48ace534a6f011f95b6ebf1abc3b 5689Author: Ben Skeggs <skeggsb@beleth.keine.ath.cx> 5690Date: Thu Jul 23 11:25:22 2009 +1000 5691 5692 kms: don't skip resize call if we don't have a backing bo for screen pixmap 5693 5694commit bff801941e23dfe3fb7449af5780ee477b10d092 5695Author: Ben Skeggs <skeggsb@beleth.keine.ath.cx> 5696Date: Thu Jul 23 11:23:15 2009 +1000 5697 5698 Fix a DRM vs DDX confusion typo 5699 5700commit e9ebddae840789535acf56ac2281b3fde3cfee96 5701Author: Maarten Maathuis <madman2003@gmail.com> 5702Date: Wed Jul 22 12:51:08 2009 +0200 5703 5704 Fix build without AM_SILENT_RULES. 5705 5706commit f70287cf7cacc03d8390376519165d50a325cec5 5707Author: Peter Henriksson <peter.henriksson@gmail.com> 5708Date: Mon Jul 20 23:56:13 2009 +0200 5709 5710 Add support for AM_SILENT_RULES 5711 5712 Make use of the AM_SILENT_RULES functionality in automake 1.11 5713 Defaults to quiet output. Verbose output can be enabled with 5714 "./configure --disable-silent-rules" or "make V=1" 5715 5716 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 5717 5718commit 69ba1095a9aef29ce8fdb71ecf48f15cb13202c2 5719Author: Ben Skeggs <bskeggs@redhat.com> 5720Date: Tue Jul 21 16:29:13 2009 +1000 5721 5722 nv50/xv: attempt sync-to-vblank if driver pixmaps are being used 5723 5724commit 994c37927289de7a0ab6e951892f0d66880a6210 5725Author: Ben Skeggs <bskeggs@redhat.com> 5726Date: Tue Jul 21 14:05:59 2009 +1000 5727 5728 nv50: fix a number of tiled buffer sizing issues in kms paths 5729 5730commit 9c1eefd221366b9f5efbd30e4a28fba244347d4f 5731Author: Ben Skeggs <bskeggs@redhat.com> 5732Date: Tue Jul 21 13:43:14 2009 +1000 5733 5734 Don't set driver pixmaps enabled with NoAccel 5735 5736commit 52882e82708f79ef198dde51f9bbebee752154be 5737Author: Ben Skeggs <bskeggs@redhat.com> 5738Date: Tue Jul 21 13:19:07 2009 +1000 5739 5740 exa: fix nouveau_exa_pixmap_is_onscreen() for driver pixmaps 5741 5742commit 9fb28d107ad9fcd6a4bbb0a35c1fc0f719d531ae 5743Author: Ben Skeggs <bskeggs@redhat.com> 5744Date: Tue Jul 21 12:52:35 2009 +1000 5745 5746 exa: remove ModifyPixmapHeader hook, it's not required 5747 5748commit 574fa36c73a1b04b67289762a10b5bbda7be083f 5749Author: Ben Skeggs <bskeggs@redhat.com> 5750Date: Tue Jul 21 12:51:26 2009 +1000 5751 5752 exa: only support driver pixmaps on servers with CreatePixmap2 5753 5754commit 365ab69499f466aeebac1437578609a7790f5dc4 5755Author: Ben Skeggs <bskeggs@redhat.com> 5756Date: Fri Jul 17 13:41:07 2009 +1000 5757 5758 remove explicit pin of pNv->FB on creation 5759 5760commit a757db1aadfd7a7d206dfc7968da8647dd91e65f 5761Author: Ben Skeggs <bskeggs@redhat.com> 5762Date: Fri Jul 17 13:38:52 2009 +1000 5763 5764 no need to pin GART bo 5765 5766commit 474ebdeea1266959fddeabf04b24e04aaab8449e 5767Author: Ben Skeggs <bskeggs@redhat.com> 5768Date: Fri Jul 17 13:37:29 2009 +1000 5769 5770 remove pinned creation from nodrm path 5771 5772commit b9333d5d34386d88b58d2267b388bbcc89b73f47 5773Author: Ben Skeggs <bskeggs@redhat.com> 5774Date: Fri Jul 17 13:35:20 2009 +1000 5775 5776 nv04-nv40: explicitly pin scanout buffer when required 5777 5778commit 53b5b68cd168b15c377f83e5e068cee92361f51d 5779Author: Ben Skeggs <bskeggs@redhat.com> 5780Date: Fri Jul 17 13:31:13 2009 +1000 5781 5782 nv50: no need to pin rotate scanout buffers on creation now 5783 5784commit a57a9bb825e04bbe4aa94d9ac1149e2fca6525a4 5785Author: Ben Skeggs <bskeggs@redhat.com> 5786Date: Fri Jul 17 13:28:36 2009 +1000 5787 5788 exa: only allow driver pixmaps when KMS enabled 5789 5790 The pre-KMS code can't deal with driver pixmaps in some circumstances, 5791 and rather than leave it there for people to accidently trip on, 5792 remove support. 5793 5794 The other option would have been to fix the UMS code, but, who really 5795 cares? 5796 5797commit 5d7a0b954c0da879f81a6fa1a9dc8070c58a0fe6 5798Author: Ben Skeggs <bskeggs@redhat.com> 5799Date: Fri Jul 17 13:25:31 2009 +1000 5800 5801 remove remains of prehistoric NOUVEAU_EXA_PIXMAPS stuff 5802 5803commit 784a0337c3165c8d8fb696e4fa30bbdc8cf01b98 5804Author: Ben Skeggs <bskeggs@redhat.com> 5805Date: Fri Jul 17 13:18:06 2009 +1000 5806 5807 nv50: explicitly pin scanout buffer in non-kms path 5808 5809commit 09db6180e4205ef55516c24ac5dfc3bd861d9fd6 5810Author: Ben Skeggs <bskeggs@redhat.com> 5811Date: Fri Jul 17 13:12:25 2009 +1000 5812 5813 nv30: don't pin shader memory 5814 5815commit 7da55d99e3c93cadda50dcad06b09c52daa0cc5b 5816Author: Ben Skeggs <bskeggs@redhat.com> 5817Date: Fri Jul 17 13:10:16 2009 +1000 5818 5819 xv: don't pin buffer unless using the overlay 5820 5821commit e8b9f689b778909e710fc5bd5616e6497b922a09 5822Author: Ben Skeggs <bskeggs@redhat.com> 5823Date: Fri Jul 17 13:07:04 2009 +1000 5824 5825 kms: don't pin scanout buffer, no need, kernel will do it 5826 5827commit b1b233099853b56cfdcff9c4c31d3550b76b949e 5828Author: Ben Skeggs <bskeggs@redhat.com> 5829Date: Fri Jul 17 08:22:02 2009 +1000 5830 5831 Fix build with xextproto 7.1 5832 5833commit c089a71a94e7887c405af7141e188302a5e2d393 5834Author: Ben Skeggs <bskeggs@redhat.com> 5835Date: Thu Jul 16 09:32:59 2009 +1000 5836 5837 dri2: fix for xserver abi change 5838 5839commit ef3a386a6df4569343e63332eb798c1d722fe164 5840Author: Ben Skeggs <bskeggs@redhat.com> 5841Date: Mon Jul 6 20:33:49 2009 +1000 5842 5843 Update manpage + finish removing useless options 5844 5845commit 4b6e0b96188e32cb2a13f8d2bd5973f0287c827b 5846Author: Ben Skeggs <bskeggs@redhat.com> 5847Date: Mon Jul 6 20:25:23 2009 +1000 5848 5849 No need for pci access wrappers now.. 5850 5851commit 1e22673f43acd0f02ba36e189b39e28b635cd50c 5852Author: Ben Skeggs <bskeggs@redhat.com> 5853Date: Mon Jul 6 20:16:02 2009 +1000 5854 5855 Remove non-XSERVER_LIBPCIACCESS paths, we depend on a new xserver anyway 5856 5857commit aafb62f86adebb3418876649e8bd189c1523e961 5858Author: Ben Skeggs <bskeggs@redhat.com> 5859Date: Mon Jul 6 20:08:03 2009 +1000 5860 5861 Some more bits of NVRec to go... 5862 5863commit 0ba588cb75af67a644f51add35778925d4088cf9 5864Author: Ben Skeggs <bskeggs@redhat.com> 5865Date: Mon Jul 6 19:58:15 2009 +1000 5866 5867 Rip out vgaHW + more unused vars from NVRec 5868 5869commit 3e22547b6826ae741b168c4b3391b6ade56b4829 5870Author: Ben Skeggs <bskeggs@redhat.com> 5871Date: Mon Jul 6 19:46:26 2009 +1000 5872 5873 Nuke nv_hw.c, and a heap of unused vars from NVRec 5874 5875commit eea5259d951a5919fa7800877570347c14411a32 5876Author: Ben Skeggs <bskeggs@redhat.com> 5877Date: Mon Jul 6 19:13:44 2009 +1000 5878 5879 Remove nv_dac.c 5880 5881commit a38b91a7a5c3337d405aa217830857832c5399e0 5882Author: Ben Skeggs <bskeggs@redhat.com> 5883Date: Mon Jul 6 19:09:09 2009 +1000 5884 5885 Cull everything in !pNv->randr12_enabled blocks 5886 5887commit 4815d312911cb4cfad4b4a5b194fce41d82f4869 5888Author: Ben Skeggs <bskeggs@redhat.com> 5889Date: Mon Jul 6 11:44:26 2009 +1000 5890 5891 kms: fix CRTC offset when rotation enabled 5892 5893commit 6d143272f497a61c5578752f58261befbfaf911a 5894Author: Ben Skeggs <bskeggs@redhat.com> 5895Date: Wed Jul 1 14:03:06 2009 +1000 5896 5897 nv50: initial support for depth 30 5898 5899commit 9b0919c3f3269b88fb338c043bbad34f20c22f41 5900Author: Ben Skeggs <bskeggs@redhat.com> 5901Date: Tue Jun 30 13:20:15 2009 +1000 5902 5903 kms: don't try transition when we don't have src+dst buffer ids 5904 5905commit c0bf670ac6b58cff60a01ab6b174ece6b1b7d892 5906Author: Ben Skeggs <bskeggs@redhat.com> 5907Date: Thu Jun 25 07:26:54 2009 +1000 5908 5909 wfb: fix stupid thinko + more safety 5910 5911commit a12cb5c8c0a85a601188178c5fc0d84a13abd87e 5912Author: Ben Skeggs <bskeggs@redhat.com> 5913Date: Thu Jun 18 14:28:25 2009 +1000 5914 5915 bios: oops 5916 5917commit e66867e4cf9f0bc8a1971664ccc3d5c56b08b2fb 5918Author: Ben Skeggs <bskeggs@redhat.com> 5919Date: Tue Jun 16 10:28:00 2009 +1000 5920 5921 bios: use image from PRAMIN in preference to PROM on NV50 5922 5923 There's at least one known case (rh#492658) where the DCB table present in 5924 the VBIOS image from PROM is not suitable for use. It contained all 16 5925 entries filled, each entry valid in itself, but contradicting other entries. 5926 5927 The VBIOS image in PRAMIN however, still has all 16 entries filled, but the 5928 first few entries now match what is present on the hardware, and the rest 5929 are set as type 0xf, which we ignore. 5930 5931commit 42c5730a4f6c243b75fbd40f1dc26e8433191c7b 5932Author: Ben Skeggs <bskeggs@redhat.com> 5933Date: Tue Jun 16 09:48:39 2009 +1000 5934 5935 wrap BlockHandler a little earlier 5936 5937commit 09e663aea82403f69a0e44e3b7cd89dfe96970d5 5938Author: Ben Skeggs <bskeggs@redhat.com> 5939Date: Mon Jun 15 16:09:27 2009 +1000 5940 5941 kms: safer fb resize func 5942 5943commit 90be5d5d6a696564bda23e9a2dad14a3a1808868 5944Author: Ben Skeggs <bskeggs@redhat.com> 5945Date: Mon Jun 15 13:43:21 2009 +1000 5946 5947 wfb: use straight memcpy hook if no tiled wraps present 5948 5949commit 86d905d34a5693e3d9b986660d4a2bcd6f30a06a 5950Author: Ben Skeggs <bskeggs@redhat.com> 5951Date: Mon Jun 15 13:41:21 2009 +1000 5952 5953 wfb: be more cautious in a few places 5954 5955commit 1bcbc4b26f71f90345bd82b836fc634e741aced8 5956Author: Ben Skeggs <bskeggs@redhat.com> 5957Date: Mon Jun 15 10:19:53 2009 +1000 5958 5959 kms: small cleanup 5960 5961commit c2d3550aafeb3e7c4df3334f0ac3554a8aa15da3 5962Author: Ben Skeggs <bskeggs@redhat.com> 5963Date: Mon Jun 15 09:16:10 2009 +1000 5964 5965 kms: fix rotation buffer pitch 5966 5967commit 6fd982979e03ce3eace7dc7523c428aca2c30b4a 5968Author: Ben Skeggs <bskeggs@redhat.com> 5969Date: Mon Jun 15 09:00:02 2009 +1000 5970 5971 kms: point rotation pixmap at correct buffer 5972 5973commit 67f8ebe77f047c12d76e508b4d7f6384baed39ac 5974Author: Ben Skeggs <bskeggs@redhat.com> 5975Date: Mon Jun 15 08:52:36 2009 +1000 5976 5977 nv50: create non-linear scanout buffers for rotation 5978 5979commit b7e3306ddc3693699f6f9de7b22913ee22ed31ed 5980Author: Stuart Bennett <stuart@freedesktop.org> 5981Date: Sun May 31 00:26:13 2009 +0100 5982 5983 Add DVI-A output info for Mac card missing bios tables (#21273) 5984 5985 The reporter indicates the card also has DVI-D and ADC, but those can't be 5986 tested to deduce the fake correct data at present 5987 5988commit 01b19c9fc25b147596e1719008ed9d861246ced9 5989Author: Ben Skeggs <bskeggs@redhat.com> 5990Date: Wed Jun 10 18:00:05 2009 +1000 5991 5992 nv50: sigh 5993 5994commit 4d9e63baff079fad727430fc605cb1e43c4303a7 5995Author: Ben Skeggs <bskeggs@redhat.com> 5996Date: Wed Jun 10 11:38:43 2009 +1000 5997 5998 wfb: "mode0" tiles have the same pitch as the others 5999 6000 This appears to be the case now (as in: after making the GPU do some of the 6001 dirty work) even though it didn't in the previous wfb patches. 6002 6003commit 02ed6b66e37e13ab4064974b89db2e73613108ef 6004Author: Ben Skeggs <bskeggs@redhat.com> 6005Date: Wed Jun 10 11:23:37 2009 +1000 6006 6007 wfb: work-around wfb suckage 6008 6009 wfb sucks far more than anticipated: 6010 - can only expose a single set of rd/wr hooks, not per-wrap as wfb doesn't 6011 treat the hooks as per-pixmap. 6012 - the hooks will get called on buffers that aren't even pixmaps 6013 6014commit 669c59ecb69a6f84ef8b8714cda6e87eacb3de9f 6015Author: Ben Skeggs <bskeggs@redhat.com> 6016Date: Wed Jun 10 09:15:37 2009 +1000 6017 6018 wfb: need a 64-bit datatype for multiply_factor 6019 6020commit 86dedb748cb6d047b195e50e921e840c1c403f6f 6021Author: Ben Skeggs <bskeggs@redhat.com> 6022Date: Tue Jun 9 22:25:01 2009 +1000 6023 6024 fix some pitch issues 6025 6026commit 794a27724fb9ab173dc7ea8062730797e1b4de04 6027Author: Ben Skeggs <bskeggs@redhat.com> 6028Date: Tue Jun 9 10:54:54 2009 +1000 6029 6030 nv50: use libwfb for pixmap access when driver pixmaps enabled 6031 6032commit 11d9690dd9e1190d1f6873bb9a5df8f98a39d4f5 6033Author: Ben Skeggs <bskeggs@redhat.com> 6034Date: Tue Jun 9 09:10:57 2009 +1000 6035 6036 exa: use the sane CreatePixmap hook, if available 6037 6038commit 6ee453381f00063522187939cb54ab9447b73ec7 6039Author: Ben Skeggs <bskeggs@redhat.com> 6040Date: Thu Jun 4 14:43:54 2009 +1000 6041 6042 xv: fix some issues with driver pixmaps 6043 6044commit 67c5287f31bf20ab3a38ffcdee410d9d49ce4d2b 6045Author: Ben Skeggs <bskeggs@redhat.com> 6046Date: Thu Jun 4 14:19:14 2009 +1000 6047 6048 exa: wait_marker becomes a NOP with driver pixmaps 6049 6050 mapping/unmapping buffers will sync as necessary 6051 6052commit f44e52852a199196f8428822bf9fb71711b9c4f9 6053Author: Ben Skeggs <bskeggs@redhat.com> 6054Date: Thu Jun 4 14:08:10 2009 +1000 6055 6056 nv50: make use of larger tile sizes 6057 6058commit 30c44ce3142ba44fc7a6c95ceaa9bf028cc112c1 6059Author: Ben Skeggs <bskeggs@redhat.com> 6060Date: Thu Jun 4 13:41:36 2009 +1000 6061 6062 exa: create unacceleratable pixmaps (ie. 1bpp) in system memory 6063 6064commit 1aa22d5894c4e4efe746592bf1c51cdaf0c31c24 6065Author: Ben Skeggs <bskeggs@redhat.com> 6066Date: Thu Jun 4 13:20:37 2009 +1000 6067 6068 nv50: use non-linear scanout buffer when driver pixmaps enabled 6069 6070commit 13d8d49f74fb10f631bcd38b169068e691b9ca99 6071Author: Ben Skeggs <bskeggs@redhat.com> 6072Date: Thu Jun 4 13:11:40 2009 +1000 6073 6074 nv50: use tile_mode from bo 6075 6076 should have no effect currently, but needed later 6077 6078commit bd9f5f2cbe7687aba7bb3a7f397d9b1bf87a8ef3 6079Author: Ben Skeggs <bskeggs@redhat.com> 6080Date: Thu Jun 4 12:11:48 2009 +1000 6081 6082 exa: fix compile against latest libdrm_nouveau 6083 6084commit 5f97afe38095a274938b8c2e9c5ebd17b7733bb6 6085Author: Stuart Bennett <stuart@freedesktop.org> 6086Date: Wed Jun 3 12:16:23 2009 +0100 6087 6088 randr12: split from pre-randr12 structs into new header, de-typedef 6089 6090 New nv_crtc->state shortcut and some renaming included for convenience 6091 6092commit 0c17b8790086a529ee9e4b5aca4b6f19f31900a9 6093Author: Stuart Bennett <stuart@freedesktop.org> 6094Date: Thu May 28 22:40:30 2009 +0100 6095 6096 Move head getting into nouveau_hw.c 6097 6098commit 12314fa6abb2dd6d325a370724c7d16c371c229c 6099Author: Stuart Bennett <stuart@freedesktop.org> 6100Date: Thu May 28 21:45:36 2009 +0100 6101 6102 randr12: improve uniformity of props code 6103 6104commit b541c1c4400942c59be37f62b7a25719dbdaa1a1 6105Author: Stuart Bennett <stuart@freedesktop.org> 6106Date: Thu May 28 18:23:37 2009 +0100 6107 6108 Remove "TMDS table script pointers not stubbed" bios warning 6109 6110 Scripts appear to be init scripts (+7 one for 0x6808b?, +9 for 0x6828b?), 6111 which when the pointer is non-zero (and the script isn't 'q') get called 6112 at POST (at least on my nv4b). 6113 6114 The warning, from a time when the scripts were thought to perhaps be needed 6115 for setting up outputs, seems redundant. 6116 6117commit 8af50289136e562198b5acc851629606d9ce450f 6118Author: Adam Jackson <ajax@redhat.com> 6119Date: Thu May 28 15:08:35 2009 -0400 6120 6121 Remove useless loader symbol lists. 6122 6123commit 317b58166bf9f3189f33e99493cf8a1b40e014b8 6124Author: Stuart Bennett <stuart@freedesktop.org> 6125Date: Wed May 13 19:05:25 2009 +0100 6126 6127 randr12: disable lvds (invalidate all modes) if bios lvds parsing fails 6128 6129commit e897191c7bcf61193576ecf52d23fcddf5b70634 6130Author: Stuart Bennett <stuart@freedesktop.org> 6131Date: Thu May 28 16:42:29 2009 +0100 6132 6133 randr12: better behaviour (avoid crash) when fp native mode can't be found 6134 6135commit 03167483b41f7205dd1efc5b571ec73ec4b3cdb5 6136Author: Stuart Bennett <stuart@freedesktop.org> 6137Date: Thu May 28 02:18:52 2009 +0100 6138 6139 randr12: line length improvements 6140 6141commit b60c16cdcbeff2d02ea0866dfbb742c3880b7ae4 6142Author: Stuart Bennett <stuart@freedesktop.org> 6143Date: Wed May 13 02:21:17 2009 +0100 6144 6145 Simplify tests for digital fp outputs 6146 6147commit 992d4b5d3aa04f88c22e52e25dc2e3e7d9bd776e 6148Author: Stuart Bennett <stuart@freedesktop.org> 6149Date: Thu May 28 15:59:12 2009 +0100 6150 6151 randr12: a bunch of trivial improvements and tidyups 6152 6153commit 6f0a324a503431d757b2247e7232de2eb7c318a5 6154Author: Stuart Bennett <stuart@freedesktop.org> 6155Date: Wed May 27 16:17:38 2009 +0100 6156 6157 Fix oopsy from 81bbdd4e causing broken framebuffer 6158 6159commit c5799186b1f70df0484c6cef63c8485ddf924733 6160Author: Stuart Bennett <stuart@freedesktop.org> 6161Date: Wed May 27 16:01:42 2009 +0100 6162 6163 randr12: fix two colour cursor on second head 6164 6165 Colour caching was preventing cursor from being set correctly on second head 6166 6167commit 81bbdd4ee2e942d91d056a10ef993a8f41651c64 6168Author: Ben Skeggs <bskeggs@redhat.com> 6169Date: Tue May 26 15:48:25 2009 +1000 6170 6171 explicitly mark buffers mappable 6172 6173commit 6c09ad5296820a845571d07cbe42d88bb19560df 6174Author: Stuart Bennett <stuart@freedesktop.org> 6175Date: Wed May 27 00:53:51 2009 +0100 6176 6177 Tolerate missing fp bios table (rh#502371) 6178 6179 Modern desktop IGPs frequently miss it out, and now it seems even some 6180 laptop parts (MXM style) do too 6181 6182commit db9ff958291416dd5b72d20d32d7412ec45c1e60 6183Author: Ben Skeggs <bskeggs@redhat.com> 6184Date: Mon May 25 08:31:07 2009 +1000 6185 6186 kms: clip sw transition dims to smallest common area 6187 6188commit 2688c97169d8ea579ba09b9db049e31704f96626 6189Author: Ben Skeggs <bskeggs@redhat.com> 6190Date: Mon May 25 08:29:47 2009 +1000 6191 6192 don't run vbios parser when kms enabled 6193 6194commit 9656762ba186e91dd2b3b7f3f9427ba520982f9d 6195Author: Ben Skeggs <bskeggs@redhat.com> 6196Date: Thu May 14 10:26:44 2009 +1000 6197 6198 nv50: fix multiple-display hangs when encoders swap crtcs 6199 6200commit d70eed91b2909ba0b20415ab1d5a099fb6b39b08 6201Author: Stuart Bennett <stuart@freedesktop.org> 6202Date: Tue May 12 16:54:33 2009 +0100 6203 6204 randr12: fix digital dpms regression 6205 6206 introduced in 3971dda57004894d5d4fc9420aa00da400815af9 6207 6208 setting two different "blank" DPMS modes (say Suspend, followed by Off) 6209 consecutively would lead to the saved fp control value getting overwritten 6210 with FP_TG_CONTROL_OFF flavoured state, so DPMS on would then never work 6211 6212 also, an improvement to LVDS DPMS state caching 6213 6214commit 10721038b854d9945ae551c6aa382b90f5d743e1 6215Author: Ben Skeggs <skeggsb@gmail.com> 6216Date: Thu May 7 21:39:41 2009 +1000 6217 6218 exa: use exaDriverAlloc() to prevent issues across exa changes 6219 6220commit 8502a8024354b88a4f19c7ad355268bb20772356 6221Author: Stuart Bennett <stuart@freedesktop.org> 6222Date: Wed May 6 18:01:54 2009 +0100 6223 6224 Fix server regeneration again 6225 6226 The DRM is now opened in PreInit, so it must not be closed in CloseScreen, 6227 as the DRM wouldn't then be open in the subsequent ScreenInit 6228 6229commit 7a796a94d0c0e18a77c5c50a0f9ca9c218d7201f 6230Author: Stuart Bennett <stuart@freedesktop.org> 6231Date: Thu Apr 30 16:04:59 2009 +0100 6232 6233 randr12: off-chip lvds for nv28 (rh#487456) 6234 6235 Quite how this possibly works with the off-chip DVI connector the laptop 6236 in question also claims to have is unclear: I suspect that (at least in 6237 nouveau's implementation) it doesn't. Sadly we've no dvi-in-use reg dump 6238 to see what should be done. 6239 6240commit 5959512ef5ce20aa52bbd41d5d987392f27b9e65 6241Author: Stuart Bennett <stuart@freedesktop.org> 6242Date: Wed May 6 14:44:40 2009 +0100 6243 6244 'drmCheckModesettingSupported' implicit function declaration fix 6245 6246commit f62719e2ac8b04d4496dc56d5fa181934bcea9f7 6247Author: Peter Hjalmarsson <xake@rymdraket.net> 6248Date: Wed May 6 14:43:22 2009 +0100 6249 6250 Unused variable warning fixes 6251 6252commit c8260ef52b58234335bdbfab31364d3d25840c21 6253Author: Pierre Pronchery <khorben@defora.org> 6254Date: Wed May 6 14:32:34 2009 +0100 6255 6256 Fix crash with Xinerama enabled 6257 6258 From the mailing list, minor tweaks by Stuart Bennett 6259 6260commit 1a478edbd58bb2eb1a41f22dea893c6d85902318 6261Author: Stuart Bennett <stuart@freedesktop.org> 6262Date: Wed May 6 02:23:42 2009 +0100 6263 6264 randr12: de-magic ramdac general control values 6265 6266commit 6c209a87d2d90b39fddc446378b370bd7fb81f6b 6267Author: Stuart Bennett <stuart@freedesktop.org> 6268Date: Tue Apr 7 14:37:05 2009 +0100 6269 6270 randr12: pre-nv17 load detection 6271 6272 Sampling heuristic as close to nvidia's as mmiotrace-based inference admits 6273 6274 Works on both nv05 and nv11 6275 6276commit 3971dda57004894d5d4fc9420aa00da400815af9 6277Author: Stuart Bennett <stuart@freedesktop.org> 6278Date: Thu Apr 30 19:31:27 2009 +0100 6279 6280 randr12: fix fp_control again again again 6281 6282 As prophesied, 9c991d3a84ad16f893b739282403eb01e02bb57f broke something: 6283 turning off a digital output, then turning on an analogue output on the 6284 same crtc as previously used by the digital output led to FP_TG_CONTROL 6285 being turned on again when the (unmodified by analogue mode_set) fp_control 6286 state was written out 6287 6288 So, belt and braces attempt to get this right, one more time 6289 6290commit 0ef16031c7e27ffabb0299e9463def125bd5e72c 6291Author: Stuart Bennett <stuart@freedesktop.org> 6292Date: Thu Apr 30 15:27:33 2009 +0100 6293 6294 randr12: solve off-chip encoder crtc exclusion in prepare, rather than dpms 6295 6296commit 0447ce06989d5599e268fdb8a4ba7bebf72f5c95 6297Author: Stuart Bennett <stuart@freedesktop.org> 6298Date: Sat Apr 4 20:08:32 2009 +0100 6299 6300 randr12: simplify nv11 digital encoder-crtc binding 6301 6302 Always setting |0x10 on the appropriate crtc for nv11 digital is fine; 6303 it's what the proprietary driver does, even if the bios doesn't 6304 6305 Fix some line lengths in nv_output_mode_set() too 6306 6307commit f69b34aa92050987e8f3da07d92d5698e9333b75 6308Author: Ben Skeggs <skeggsb@gmail.com> 6309Date: Fri May 1 09:07:55 2009 +1000 6310 6311 nv50: return immediately in GetDDCModes if no DDC on connector... 6312 6313commit 45e16187612bc09b234b598dff25bbcc8069fde1 6314Author: Ben Skeggs <skeggsb@gmail.com> 6315Date: Thu Apr 30 16:37:29 2009 +1000 6316 6317 bios: use NV_ARCH_50 to match G8x chips 6318 6319 There's at least a couple of chips where we get 0x77 for chip_version, 6320 and fail to match the chip_version >= 0x80 test.. 6321 6322commit 3d61697c477dd7358e5fd4e5bc608b4fe763b5d3 6323Author: Ben Skeggs <skeggsb@gmail.com> 6324Date: Thu Apr 30 10:23:11 2009 +1000 6325 6326 nv50: remove reading back vbios-programmed lvds native mode 6327 6328commit bade2499db8e02dd2904bc811399c891bf126208 6329Author: Ben Skeggs <bskeggs@redhat.com> 6330Date: Thu Apr 30 09:33:21 2009 +1000 6331 6332 nv50: group encoders into connectors by i2c port 6333 6334commit 9ee2ac19d7adc067a5551e6d6853a19b89946d46 6335Author: Ben Skeggs <skeggsb@gmail.com> 6336Date: Wed Apr 29 16:26:41 2009 +1000 6337 6338 bios: remove dodgy mode-table search for g80 6339 6340 This appears to work in exactly the same way as on earlier chipsets, the 6341 strap value just moved slightly.. 6342 6343commit 6282574dbeeb7075a00ce03940c4f2a597a048b5 6344Author: Ben Skeggs <skeggsb@gmail.com> 6345Date: Wed Apr 29 16:25:43 2009 +1000 6346 6347 bios: modify get_fp_strap() for g80 6348 6349commit 09b832e24b9bd887275ba882beabe9bba56f0207 6350Author: Ben Skeggs <skeggsb@gmail.com> 6351Date: Tue Apr 28 09:46:55 2009 +1000 6352 6353 nv50: merge NV50SorSetClockMode and NV50SorSetClockModeLVDS 6354 6355commit ed2c185caeb75c595dfb2e814748a856a23b2b35 6356Author: Ben Skeggs <skeggsb@gmail.com> 6357Date: Tue Apr 28 08:44:01 2009 +1000 6358 6359 bios/nv50: initial parsing of display script tables, not quite complete 6360 6361commit 3fccc9e83a194fd58c333c6c4793a27cb05ce026 6362Author: Ben Skeggs <skeggsb@gmail.com> 6363Date: Tue Apr 28 09:38:21 2009 +1000 6364 6365 bios: some nv50 init scripts have flags in reg values, deal with (1<<30) 6366 6367commit 2b4b3da73babbe9ebf4acc26238a4b2b8a6e6ee8 6368Author: Ben Skeggs <skeggsb@gmail.com> 6369Date: Wed Apr 29 11:00:17 2009 +1000 6370 6371 bios: dcb location is only 2 bits 6372 6373commit 7b5bfff21916461ce7f6a2adb27dc443ad0b8201 6374Author: Ben Skeggs <skeggsb@gmail.com> 6375Date: Tue Apr 28 08:12:05 2009 +1000 6376 6377 bios: parse BIT U table 6378 6379commit e4260ec4f0303b286f16eab4fe5a1bb66ff89e3d 6380Author: Stuart Bennett <stuart@freedesktop.org> 6381Date: Wed Apr 1 02:56:05 2009 +0100 6382 6383 randr12: more use of logging abstraction in modesetting code 6384 6385commit 4d0fdba9abcd0af736821d54baca5f0c9a8d6ce1 6386Author: Stuart Bennett <stuart@freedesktop.org> 6387Date: Mon Apr 6 01:14:04 2009 +0100 6388 6389 randr12: fix for off-chip dual link digital (enable extra wide interface?) 6390 6391commit bb246d72a01285299599cbf8ff82c67f14682a4f 6392Author: Stuart Bennett <stuart@freedesktop.org> 6393Date: Mon Apr 6 01:00:41 2009 +0100 6394 6395 Partial fix for nv28 lvds bios parsing (part of rh#487456) 6396 6397 For rh#487456, though this is neither a complete fix for that bug, nor a 6398 complete fix for nv25/28 lvds (the EDID matching mentioned in the commit is 6399 omitted until someone appears with a laptop requiring it) 6400 6401commit 2915926702b7ae9ccb1e900c07ca14a2f32e4512 6402Author: Stuart Bennett <sb476@cam.ac.uk> 6403Date: Mon Apr 7 23:20:12 2008 +0100 6404 6405 plls: max_log2p_bias is likely actually max_log2p 6406 6407 As explained in nouveau_bios.h, a separate max_usable_log2p is needed 6408 (lack of change from 6 to 7 tested by setting a stable small mode, say 6409 400x300, then manually tweaking the multiplier up and post divider down). 6410 The max_usable_log2p values hardcoded here are unchanged from those 6411 previously taken from the bios pll calculation code 6412 6413 For reference, the blob will quite happily attempt to set clocks using the 6414 raw parsed max_log2p values (and go up to 7 therefore) when setting modes 6415 around 200x200 (3MHz pxclk), but at such a low clock my monitor won't agree 6416 to give a sensible picture anyway. Reg dumped data using "nvidia" below: 6417 6418 nv40: log2P up to 7 (dual pll mode), at very low clocks driver gives up and sets 0x0006ff0d 0x80001f04 6419 nv43: log2P up to 7 (dual pll mode), at very low clocks driver gives up and sets 0x0006ff0d 0x80001f04 6420 nv11: single pll, therefore goes quite happily down to at least 0.25MHz 6421 nv31: log2P up to 7 (dual pll mode), at very low clocks card locks up 6422 nv34: log2P up to 5 (single pll), goes down to at least 2.25MHz 6423 nv4b: log2P up to 6 (single pll mode), at very low clocks driver gives up and sets 0x0006ff0d 0x80001f04 6424 6425commit 47bb00fee8b9906adc03b372efa1ae813bc9ca9a 6426Author: Ben Skeggs <skeggsb@gmail.com> 6427Date: Tue Apr 21 08:57:07 2009 +1000 6428 6429 nv50: missed a WAIT_RING 6430 6431commit 523ccecee26454fc43a65e9b33d91a7d0ff8487d 6432Author: Ben Skeggs <skeggsb@gmail.com> 6433Date: Fri Apr 17 14:18:54 2009 +1000 6434 6435 nv50: use vbios-programmed lvds mode as native mode if no ddc 6436 6437 Why was this even removed? Its replacement (read VBIOS fp mode table) has 6438 proven useless in the majority of cases I've seen. 6439 6440commit fa2f1115d5f7f3e3767d423a77c6d929a4ae8d0f 6441Author: Ben Skeggs <skeggsb@gmail.com> 6442Date: Fri Apr 17 10:25:57 2009 +1000 6443 6444 Remove pNv->GART conditional for selecting CB_LOCATION in AGP/PCI 6445 6446 Kinda useless, the DRM knows better whether GART is available. 6447 6448commit 7100c06be099bacc0f8bb8898bbf7eb34ff1cc6e 6449Author: Ben Skeggs <skeggsb@gmail.com> 6450Date: Mon Apr 13 20:21:51 2009 +1000 6451 6452 kms: fix displayWidth in resize 6453 6454commit ef2de256dabfde2d619e4d451bd417973c919329 6455Author: Ben Skeggs <skeggsb@gmail.com> 6456Date: Mon Apr 13 19:09:32 2009 +1000 6457 6458 do e-edid for legacy modesetting path 6459 6460commit d8545e669aaa4362a236ba2e7231ca27ffdea1bd 6461Author: Ben Skeggs <skeggsb@gmail.com> 6462Date: Wed Apr 8 16:15:02 2009 +1000 6463 6464 bios: logic typo from earlier commit 6465 6466commit 960a5c82a80272a1577dede6a05ab7dc5db37cc3 6467Author: Ben Skeggs <skeggsb@gmail.com> 6468Date: Wed Apr 8 14:44:06 2009 +1000 6469 6470 nv50: modify ddc<->nvreg assignments again.. 6471 6472commit 3a6c7091d9706778a2e3971b7f36601d79fe7a83 6473Author: Ben Skeggs <skeggsb@gmail.com> 6474Date: Wed Apr 8 14:34:20 2009 +1000 6475 6476 kms: fix setting the randr edid property 6477 6478commit a1194b3690ced5db552d63be118a605508a1b965 6479Author: Ben Skeggs <skeggsb@gmail.com> 6480Date: Tue Apr 7 19:24:25 2009 +1000 6481 6482 bios: fix typo 6483 6484commit 11451cabbd5357dd71ab65e6155d7ca7d9766db7 6485Author: Ben Skeggs <skeggsb@gmail.com> 6486Date: Tue Apr 7 18:34:57 2009 +1000 6487 6488 bios: some G8x don't have PBUS at 0x1800, us 0x88000 unconditionally there 6489 6490commit 62d69bdf807423a644d7c2c493f94948ea11e491 6491Author: Ben Skeggs <skeggsb@gmail.com> 6492Date: Sat Apr 4 01:41:47 2009 +1000 6493 6494 bios: add exception for chipset 0x73 for lack of fp table 6495 6496commit 620d519564fdc6b3a7d55184bd481fa69035371b 6497Author: Ben Skeggs <skeggsb@gmail.com> 6498Date: Fri Apr 3 23:07:56 2009 +1000 6499 6500 consistent connector naming across <nv50, nv50 and kms 6501 6502commit a5d45c80e85611c9e22d8eca27294eef5378a549 6503Author: Stuart Bennett <stuart@freedesktop.org> 6504Date: Tue Apr 7 00:40:31 2009 +0100 6505 6506 randr12: set 1 on cr59 for off-chip digital, 0 otherwise (#21023, rh#492399) 6507 6508 The ubiqitous setting of 1 on cr59 for <nv40 tmds was breaking the 6509 integrated dvi of nv34gl. 6510 Thanks to Bill Nottingham for isolating the problem. 6511 6512commit d63c92435956b23351f1950563bbb92e24e60cea 6513Author: Stuart Bennett <stuart@freedesktop.org> 6514Date: Fri Apr 3 23:39:22 2009 +0100 6515 6516 randr12: unlock CR21 *after* turning off digital output 6517 6518 otherwise the analogue output still fails to come up. 6519 6520 That'll teach me to rearrange things just before committing 6521 6522commit 11be9a982073d66a68cd3db2bfc611eb58d3ea81 6523Author: Ben Skeggs <skeggsb@gmail.com> 6524Date: Fri Apr 3 12:22:04 2009 +1000 6525 6526 nv50: fix i2c port addresses 6527 6528commit e2aa03794420324ae5a682bbc415a79a4d830f7e 6529Author: Stuart Bennett <stuart@freedesktop.org> 6530Date: Fri Apr 3 00:57:19 2009 +0100 6531 6532 randr12: fall back to standard timings when finding native mode (rh#492819) 6533 6534commit 6965663ddcc8dd3330842d3c4c4eb4cbf4e4fc39 6535Author: Stuart Bennett <stuart@freedesktop.org> 6536Date: Wed Apr 1 05:38:29 2009 +0100 6537 6538 randr12: fix hw cursor for fully transparent pixels 6539 6540 ARGB8 pixels of 0x00000000 were getting bumped to 0x01000000, which was 6541 _just_ visible as a box around the cursor 6542 6543commit d12f70cedb0ba598234bbfc26dd6beed23e3dfe7 6544Author: Stuart Bennett <stuart@freedesktop.org> 6545Date: Sun Mar 29 00:51:05 2009 +0000 6546 6547 randr12: pre-nv17 digital fixes 6548 6549 1) set panel regs better (don't do min_front_porch adjustment) 6550 2) scaler programming - nv11 and nv20 don't need / 2 for aspect scaling 6551 6552commit ea567dbc7bc9c8d2ee5d7d9e0dcd33b03c638edb 6553Author: Stuart Bennett <stuart@freedesktop.org> 6554Date: Sat Mar 28 01:58:09 2009 +0000 6555 6556 randr12: fix restore for cards where CR21 is not left unlocked at POST 6557 6558 More nv11 inspired fun. If CR21 is left without a 0xfa mask after boot, 6559 we'd restore the old CR21 value and the subsequent restoration of VGA 6560 CRTC values wouldn't get through. 6561 6562 RE'd doc of CR21 behaviour included 6563 6564commit bd263d83139fcc93b967c71c99fd5e63e1ba6315 6565Author: Stuart Bennett <stuart@freedesktop.org> 6566Date: Fri Mar 27 18:27:18 2009 +0000 6567 6568 randr12: make nv11 dvi work for both crtcs 6569 6570commit 52c287dc64a97a750eb953ac5f725d794cbf0c64 6571Author: Stuart Bennett <stuart@freedesktop.org> 6572Date: Fri Mar 27 00:23:04 2009 +0000 6573 6574 Every card since nv5 can do panels 6575 6576 We'll only try to do ones using scripts atm though 6577 6578 Note that as we don't have a load detect function for pre-nv17 you'll 6579 likely get a 1024x768 mode set up on VGA-0 even if only the DVI output 6580 is plugged. Unclear if this is a bug or feature :) 6581 6582commit 3d371a78bc9522e9a1b2c967957a090bb527a30c 6583Author: Stuart Bennett <stuart@freedesktop.org> 6584Date: Thu Mar 26 22:21:57 2009 +0000 6585 6586 randr12: enable tmds reg access on pre-nv17 6587 6588 I think this nv11 hates me 6589 6590commit 1bb85c8020802719e82c1335a18df9367215a8ff 6591Author: Stuart Bennett <stuart@freedesktop.org> 6592Date: Wed Apr 1 04:09:09 2009 +0100 6593 6594 randr12: reorder ramdac reg access 6595 6596commit 4465fddb2467ad9d923120a2fd2c648222603771 6597Author: Stuart Bennett <stuart@freedesktop.org> 6598Date: Wed Apr 1 04:10:18 2009 +0100 6599 6600 randr12: remove nv30 special case 6601 6602 0x68?89? are just an alternative way to set scaling (set | 0x40 on 0x68.880, 6603 then horiz scaling in 0x68.898, vert in 0x68.89c) and afaict don't have any 6604 special significance on nv30 6605 6606commit 72eced5eeeef438979dff06b6c1cbaa120738cfe 6607Author: Stuart Bennett <stuart@freedesktop.org> 6608Date: Wed Apr 1 03:12:30 2009 +0100 6609 6610 randr12: fix/improve a load of version/feature tests 6611 6612 Main improvement is introduction of flag for nv17+ display architecture 6613 6614 Other version fixes include: 6615 * allow access of fp control regs on all cards (defined since nv4) 6616 * only read crtc_850 and gpio_ext on nv40+, since only written on nv40+ 6617 * set dither state on all versions (only written out on relevant cards) 6618 * PRAMDAC_DACCLK is only defined/usable on nv17+ 6619 * no point saving the lvds/tmds head for restore if there's only one head 6620 * fix a bunch of places the nforce (0x1a) case had been missed 6621 6622commit 37c6916d3eac54cc13812276fca99662162c4bbc 6623Author: Ben Skeggs <skeggsb@gmail.com> 6624Date: Wed Apr 1 15:00:03 2009 +1000 6625 6626 bios: fix chips with pll limits table version 0 6627 6628commit 76c5a05603b0eaa0831e031268e48ab6ffdee3a9 6629Author: Ben Skeggs <skeggsb@gmail.com> 6630Date: Wed Apr 1 10:36:34 2009 +1000 6631 6632 bios: xf86DrvMsg->NV_ERROR 6633 6634commit b1b9bcb111f1f5ec15e19d5a84b7ca576983c4a1 6635Author: Ben Skeggs <skeggsb@gmail.com> 6636Date: Tue Mar 31 11:43:26 2009 +1000 6637 6638 bios: support for pll limits table v3.0 6639 6640commit 8c85b4b7451bd09b0607b1f4ef8ce4b833006328 6641Author: Ben Skeggs <skeggsb@gmail.com> 6642Date: Tue Mar 31 11:11:12 2009 +1000 6643 6644 kms: CONNECTORn -> CONNECTOR-n 6645 6646commit 4e1c3239648510a4024917289127ad1e982ed34d 6647Author: Stuart Bennett <stuart@freedesktop.org> 6648Date: Tue Mar 31 20:59:26 2009 +0100 6649 6650 Restore call to NVDRIGetVersion, to ensure the dri module is loaded 6651 6652commit c973f756cfdf36db3a5c7095f6e2c57ff5841388 6653Author: Stuart Bennett <stuart@freedesktop.org> 6654Date: Tue Mar 31 17:07:26 2009 +0100 6655 6656 Abstract logging in nv_bios.c 6657 6658commit b17bebd811316b5ccca21146b58f8fbf4773f0ad 6659Author: Stuart Bennett <stuart@freedesktop.org> 6660Date: Mon Mar 30 20:52:29 2009 +0100 6661 6662 nForce DIMM check is now in DRM 6663 6664 Since it only at most suggests an optimization, there's no real need to 6665 wait for a DRM version bump 6666 6667commit 9c991d3a84ad16f893b739282403eb01e02bb57f 6668Author: Stuart Bennett <stuart@freedesktop.org> 6669Date: Mon Mar 30 03:31:01 2009 +0100 6670 6671 randr12: FP_TG_CONTROL 2: dpms improvements 6672 6673 For DPMS on digital outputs nvidia does not just OR the 0x68.848 value 6674 with 0x20000022, they mask off 0x10000011 too. 6675 To do the same thing here, don't change in-memory state of programmed reg 6676 value (hopefully nothing breaks) 6677 6678commit 5fb2888d79d39a1c74ab005d8d518686ad823224 6679Author: Stuart Bennett <stuart@freedesktop.org> 6680Date: Fri Mar 27 18:27:18 2009 +0000 6681 6682 randr12: FP_TG_CONTROL 1: turn off digital path when using VGA encoder 6683 6684 On pre-nv17, having a CRTC's digital output path turned on when 6685 programming the CRTC for VGA confuses the chip and often results in the 6686 analogue output not working. Indeed, having it turned on even just before 6687 programming the CRTC seems to break it. 6688 6689 Nvidia solves this by turning off the digital path early in modesetting 6690 and then delaying for 50ms, so (if the digital path was on to start with) 6691 that's reproduced here. 6692 6693 Although this problem has not been seen on nv17+, nvidia uses the same 6694 approach, so we may as well too. 6695 6696 Some tidying of the output_prepare function thrown in as well. 6697 6698commit 8023dc3e2dcfa8a4cc11a5a64750a7505e8aa704 6699Author: Stuart Bennett <stuart@freedesktop.org> 6700Date: Mon Mar 30 02:27:50 2009 +0100 6701 6702 Split DCB 1.5 parsing from 2.0+, get closer to 80 cols 6703 6704commit ea027b35b28c0305aca03edc8c1ff5cd84bbf3ef 6705Author: Stuart Bennett <stuart@freedesktop.org> 6706Date: Fri Mar 27 00:03:38 2009 +0000 6707 6708 The dcb prior to v1.5 is pretty hopeless, just add a crt. 6709 6710 Also, we don't use dcb 1.2 i2c for modesetting due to cards with broken 6711 tables (ala #14821), and I've got an nv11 with duff entries (v1.4), so 6712 don't raise hopes of init_dcb_i2c_entry working on these versions either 6713 6714commit 921fbccd53eae37ed86ecece6dc1fcae4b7bf3c4 6715Author: Stuart Bennett <stuart@freedesktop.org> 6716Date: Thu Mar 26 16:41:02 2009 +0000 6717 6718 randr12: avoid nv11 chip lockup when saving palette regs 6719 6720commit 9d46930f0eec7b787666671681e9a93fc33dce62 6721Author: Ben Skeggs <skeggsb@gmail.com> 6722Date: Mon Mar 30 20:11:03 2009 +1000 6723 6724 xv: oops 6725 6726commit 9213c39d3a8d9a7fc90a44f9427c1fb8218243e2 6727Author: Ben Skeggs <skeggsb@gmail.com> 6728Date: Mon Mar 30 11:14:20 2009 +1000 6729 6730 xv: post damage after we draw (rh#492239) 6731 6732commit d68a0527aa2a3e93c3408526d4bbb93db89e9b3c 6733Author: Ben Skeggs <skeggsb@gmail.com> 6734Date: Mon Mar 30 10:45:59 2009 +1000 6735 6736 kms: small cleanup 6737 6738commit c9cb6a6f7f79233bdab737fa6b6d0916dab45971 6739Author: Stuart Bennett <stuart@freedesktop.org> 6740Date: Fri Mar 27 16:43:27 2009 +0000 6741 6742 rh#492511 has a CRTC_OWNER of 0x7 for some reason 6743 6744 not that this probably affects the bug in any way 6745 6746commit 95bff61597ddf21d6415b40759258802a5f42150 6747Author: Stuart Bennett <stuart@freedesktop.org> 6748Date: Wed Mar 25 02:40:17 2009 +0000 6749 6750 randr12: fix LVDS legitimately disconnected case 6751 6752 DPMS off still needs to work, but there's no native_mode for the pxclk. 6753 Fortunately LVDS_PANEL_OFF doesn't actually need a pxclk, so just use 0 6754 6755commit f86e395231c97155c1a211010835fd02b8e4dc5f 6756Author: Stuart Bennett <stuart@freedesktop.org> 6757Date: Wed Mar 25 02:07:15 2009 +0000 6758 6759 randr12: permit vga outputs to be force enabled in xorg.conf 6760 6761 You need: 6762 Section "Monitor" 6763 Identifier "VGA-0" 6764 Option "Enable" "true" 6765 EndSection 6766 6767commit 8427b39a2f6cb32496bbbe370b793d5ba78f6b6c 6768Author: Stuart Bennett <stuart@freedesktop.org> 6769Date: Tue Mar 24 03:32:24 2009 +0000 6770 6771 randr12: move a load of hw mode programming into nouveau_hw.c 6772 6773 Pull in pll setting (common, no reason to bloat nv_bios with it), state 6774 load/unload from nv_crtc, and useful bits of nv_hw.c (remnants in nv_hw.c 6775 are only used on the non-randr12 path) 6776 6777commit 163bdce98c0f9ca5c1125e33a9a2bf2d2d9a1286 6778Author: Stuart Bennett <stuart@freedesktop.org> 6779Date: Tue Mar 24 03:19:38 2009 +0000 6780 6781 Turn the smaller mmio wrappers into static inline 6782 6783commit 21ee927f28d1ca8d3807acc767ef14783e43f057 6784Author: Stuart Bennett <stuart@freedesktop.org> 6785Date: Mon Mar 23 21:34:26 2009 +0000 6786 6787 Put arbitration and mnp calcs in separate file 6788 6789 Make bios->chip_version public so the pll calcs can use it 6790 6791commit 83dc890f0b1fc88d2fbc7559f5d4decf4c84f340 6792Author: Maarten Maathuis <madman2003@gmail.com> 6793Date: Sat Mar 28 01:25:32 2009 +0100 6794 6795 nv50: use E-EDID when available + apply edid quirks 6796 6797commit 2c130cf71587706a4f1dcd8f5db1de47a10522f5 6798Author: Pekka Paalanen <pq@iki.fi> 6799Date: Fri Mar 27 23:16:59 2009 +0200 6800 6801 nv50: fix xf86GetDefaultModes() call for older servers 6802 6803 Thanks to stillunknown for this build fix. 6804 6805commit f1907dcef8d06d7ee4ef10ba22bb7decef700110 6806Author: Ben Skeggs <skeggsb@gmail.com> 6807Date: Fri Mar 27 15:37:17 2009 +1000 6808 6809 nv50: add default modes to mode pool for LVDS panel 6810 6811 At some point X stopped doing this if the display doesn't support 6812 continuous timing. For a lot of G80 LVDS panels, they only report 6813 their native mode via DDC. 6814 6815 We'll add the default modes ourself, and use the GPU scalers. 6816 6817commit 7da6fdb8b477d2007c83b47e9cbcc2476ae40f22 6818Author: Ben Skeggs <skeggsb@gmail.com> 6819Date: Fri Mar 27 14:17:29 2009 +1000 6820 6821 kms: fix bug which prevented getting edid from the kernel 6822 6823commit f431e204d3a9be92df5ca606e86ed7c6d82103f0 6824Author: Ben Skeggs <skeggsb@gmail.com> 6825Date: Fri Mar 27 11:50:22 2009 +1000 6826 6827 randr: fix crash when rotation requested 6828 6829commit 36dedd04da1b3c89be61a95270414477d284b2ef 6830Author: Ben Skeggs <skeggsb@gmail.com> 6831Date: Fri Mar 27 11:26:09 2009 +1000 6832 6833 kms: check for mm_enabled as an additional test for kms presence 6834 6835commit a923bc1e4840c0386301f8648add2ccbfbf79a88 6836Author: Ben Skeggs <skeggsb@gmail.com> 6837Date: Thu Mar 26 16:24:01 2009 +1000 6838 6839 nv50/xv: correct rendering to partically obscured windows 6840 6841 Won't be exposed by current git, due to lack of accelerated front-buffer 6842 rendering. But should fix rh#492173, rh#492229. 6843 6844commit 01cee2906686897c4b795d308270b3b69a3e286e 6845Author: Ben Skeggs <skeggsb@gmail.com> 6846Date: Thu Mar 26 11:43:09 2009 +1000 6847 6848 nv50: call NVSync() in CloseScreen() before restoring video mode 6849 6850 Without this, if accel is still being done to the front buffer (usually at 6851 offset 0), we'll restore the video mode and then the engine will clobber 6852 the text mode fonts right afterwards leaving a messed up console. 6853 6854commit aa7c0375b51d813be54de5e19b96e958ac183f17 6855Author: Ben Skeggs <skeggsb@gmail.com> 6856Date: Thu Mar 26 06:36:17 2009 +1000 6857 6858 kms: implement AdjustFrame, should fix crash in fdo#24236 6859 6860commit 61879b8079bee1caeedc0bd9c11061423ff27b4a 6861Author: Ben Skeggs <skeggsb@gmail.com> 6862Date: Wed Mar 25 10:09:39 2009 +1000 6863 6864 kms: drm_mode_modeinfo struct changed names at some point 6865 6866commit 79306fc29f6b27fd61fe51855bad5489fc4fd986 6867Author: Ben Skeggs <skeggsb@gmail.com> 6868Date: Wed Mar 25 08:15:41 2009 +1000 6869 6870 dri: fail harder 6871 6872commit da1ba93f1dc2b02615e4f0366450872659176799 6873Author: Ben Skeggs <skeggsb@gmail.com> 6874Date: Wed Mar 25 08:12:26 2009 +1000 6875 6876 device close doesn't belong in DRI close, can have DRM without DRI 6877 6878commit 56b11c773ac9392069c6a589f4f87c5cc79e020e 6879Author: Younes Manton <younes.m@gmail.com> 6880Date: Tue Mar 24 14:27:40 2009 -0400 6881 6882 Fix mouse cursor disappearing when near top/left edge of screen. 6883 6884commit 4067ab466fe3aa817e0323959f70c7dd3494de0a 6885Author: Ben Skeggs <skeggsb@gmail.com> 6886Date: Mon Mar 23 14:43:22 2009 +1000 6887 6888 another ppc fix 6889 6890commit 862dba8b6ca6354d915f2430826a5581f41d1002 6891Author: Ben Skeggs <skeggsb@gmail.com> 6892Date: Mon Mar 23 14:22:15 2009 +1000 6893 6894 more ppc.. 6895 6896commit 30634865617737171835f6ac9437df597e5a8951 6897Author: Ben Skeggs <skeggsb@gmail.com> 6898Date: Mon Mar 23 13:57:52 2009 +1000 6899 6900 fix ppc build 6901 6902commit d80fe7878379aefd3345746a149f2b46d3a84805 6903Author: Francisco Jerez <currojerez@riseup.net> 6904Date: Mon Mar 23 03:27:16 2009 +0100 6905 6906 Fix nouveau_hw_decode_pll on NV30/35. 6907 6908commit 62aa81deca5c63d30407872c0e2b784f5084acb3 6909Author: Stuart Bennett <stuart@freedesktop.org> 6910Date: Sat Mar 21 01:43:14 2009 +0000 6911 6912 Some laptop mode finding fixes resulting from rh#487456 6913 6914 On closer inspection, it seems BMP laptops only depend on 6915 use_straps_for_mode to ban DDC. Therefore use pxclk alone to determine 6916 EDID case in parse_lvds_table, and invert sense of fp ddc flag to get a 6917 default of "false" 6918 6919commit aa6edfaf1c32432b07c298c42b68f2d93f3b337c 6920Author: Stuart Bennett <stuart@freedesktop.org> 6921Date: Sun Mar 15 15:18:16 2009 +0000 6922 6923 randr12: name VGA CRTC fields 6924 6925 Now with added consistent indentation 6926 Use fields in a couple of other cases while we're about it 6927 6928commit 7b7c44c828eda2edfad2fa9bea95f70889c59453 6929Author: Stuart Bennett <stuart@freedesktop.org> 6930Date: Sun Mar 15 03:45:54 2009 +0000 6931 6932 randr12: rename reg state members to follow reg names 6933 6934commit 4ae6c22957a929760e583d5704e350adb4404c3d 6935Author: Stuart Bennett <stuart@freedesktop.org> 6936Date: Sun Mar 15 03:11:40 2009 +0000 6937 6938 randr12: name some FP flags, move some unrelated stuff out from nvreg.h 6939 6940commit b50b49f6f54148ee64ba7500e797f432b198a324 6941Author: Stuart Bennett <stuart@freedesktop.org> 6942Date: Sun Mar 15 04:21:40 2009 +0000 6943 6944 pre-nv50: use nvidia's names for PRAMDAC regs where known 6945 6946 Mostly adding 'P' to the previous NV_RAMDAC regs -- either when known 6947 correct, or when totally unknown (eg NV_PRAMDAC_580); made-up names 6948 untouched. Lack of code changes checked with cpp. 6949 6950 Unused defines removed too. 6951 6952commit a9df3049a7e38b3bf4478459e5f0736e10927247 6953Author: Stuart Bennett <stuart@freedesktop.org> 6954Date: Sun Mar 15 04:08:56 2009 +0000 6955 6956 pre-nv50: use nvidia's names for PCRTC regs where known 6957 6958 Mostly adding 'P' to the previous NV_CTRC regs -- either when known 6959 correct, or when totally unknown (eg NV_PCRTC_850); made-up names 6960 untouched. Lack of code changes checked with cpp. 6961 6962 Unused defines removed too. 6963 6964commit 2c0710d334e3e1dcb04cc4fc0b46ec4c8cdb9ba6 6965Author: Stuart Bennett <stuart@freedesktop.org> 6966Date: Tue Mar 17 18:50:18 2009 +0000 6967 6968 Separate maximum limits for encoder devices and i2c records, DCB_ prefixes 6969 6970 From the comment with pNv->connector in nv_type.h I've assumed nv50 was 6971 using the previous define as an I2C limit, rather than encoder limit 6972 6973commit 0db27a6f1aee023782de14ae2f260fec4e60229d 6974Author: Stuart Bennett <stuart@freedesktop.org> 6975Date: Tue Mar 17 13:43:13 2009 +0000 6976 6977 Add DCB 1.5 entry from a GeForce2 Go reported by Martin Ketzer on the ML 6978 6979commit 3e7fa97a50e42c0255876c79137dc1a40a6052a6 6980Author: Ben Skeggs <skeggsb@gmail.com> 6981Date: Wed Mar 18 09:36:33 2009 +1000 6982 6983 dri: pass object handle instead of offset if using real mm 6984 6985commit 88efe405213a3ea23edd9bed971da1d2492021eb 6986Author: Ben Skeggs <skeggsb@gmail.com> 6987Date: Mon Mar 16 17:45:49 2009 +1000 6988 6989 dri2: initial implementation, only with driver pixmaps 6990 6991 Also, no DRI1 when driver pixmaps enabled. 6992 6993commit 675126e0f0c1c077bde4b25ccb64f87d1d08547b 6994Author: Ben Skeggs <skeggsb@gmail.com> 6995Date: Mon Mar 16 15:06:31 2009 +1000 6996 6997 kms: support framebuffer resize if driver pixmaps enabled 6998 6999commit 8b354b92a859cf277b19e51fc6cb109f4ea44b7a 7000Author: Ben Skeggs <skeggsb@gmail.com> 7001Date: Mon Mar 16 13:13:07 2009 +1000 7002 7003 kms: copy old fb to new fb on modeset 7004 7005 Requires driver pixmaps enabled, which suck on G8x currently. Paves the 7006 way for a nicer transition from console->X. There's still a flash where 7007 the screen goes black before X starts here, but this is on the way to what 7008 we want. 7009 7010commit 6658403cfd91556d06afecdde03feec7384c5398 7011Author: Ben Skeggs <skeggsb@gmail.com> 7012Date: Mon Mar 16 12:34:49 2009 +1000 7013 7014 Call NVEnterVT from NVScreenInit rather than duplicating it. 7015 7016commit f59fef15c94fc5c6f216d0f5dcdcb3d3188c0049 7017Author: Ben Skeggs <skeggsb@gmail.com> 7018Date: Mon Mar 16 09:17:29 2009 +1000 7019 7020 exa: fix multiple prepare/finish_access on a pixmap 7021 7022commit 75a1c4158a19bf0857de49b736fb87a40c095048 7023Author: Stuart Bennett <stuart@freedesktop.org> 7024Date: Sat Mar 14 15:48:39 2009 +0000 7025 7026 Wrap bios endian casting 7027 7028 Also convert some functions to use individual bios pointers, rather than 7029 bios->data[offset + ... 7030 Plus some minor bmp structure documentation and tidying 7031 7032commit 3666d73333ecbdd3a4941a76af2528113db427bb 7033Author: Stuart Bennett <stuart@freedesktop.org> 7034Date: Wed Feb 25 20:21:42 2009 +0000 7035 7036 Switch to using pll_vals struct for passing pll stuff around 7037 7038commit a3dbc06d4f9dc340f5c167b94d8e10ac8386c5b4 7039Author: Stuart Bennett <stuart@freedesktop.org> 7040Date: Fri Apr 11 16:51:30 2008 +0100 7041 7042 randr12: cull old unused pll code 7043 7044commit 71c821b8119cbac8dcd91b27e3b7556d00ceb546 7045Author: Stuart Bennett <stuart@freedesktop.org> 7046Date: Sat Mar 14 19:06:12 2009 +0000 7047 7048 Don't pointlessly overallocate cursor storage 7049 7050commit 7284c07a0ad6551ea99c28454f4e7308699835e6 7051Author: Stuart Bennett <stuart@freedesktop.org> 7052Date: Fri Mar 13 21:23:48 2009 +0000 7053 7054 randr12: nv10 hw cursor fixes/changes 7055 7056 Remove access to PCRTC_GPIO, it's unused and on NV10 overwrites 7057 PCRTC_CURSOR_CONFIG leading to all sorts of cursor weirdness 7058 7059 NV10 hw cursor is now set-up on the assumption it can do 64x64 (we've been 7060 using 64x32 up to now), but if it can't do cursors 64px high it can easily 7061 be fixed to behave like nv0x at 32x32 (64x32's pretty silly) 7062 7063 It's possible that with the PCRTC_GPIO thing resolved the nv10 cursor can 7064 be more featureful: the code would also be simpler if it supported 32bpp 7065 cursors (test by seeing if setting NV_CRTC_CURSOR_CONFIG_32BPP has any 7066 effect). It may even do alpha -- change pNv->alphaCursor to be 7067 (NVArch >= 0x10) to find out. 7068 7069commit bcc3a14b15cdad3647cc7df0f540df1111843be5 7070Author: Stuart Bennett <stuart@freedesktop.org> 7071Date: Fri Mar 13 02:31:35 2009 +0000 7072 7073 randr12: use nv0x cursor colour expansion funcs on later cards, unify argb paths 7074 7075commit b21c807d2ced35ac1b06300f6e68d661710b336a 7076Author: Stuart Bennett <stuart@freedesktop.org> 7077Date: Fri Mar 13 03:52:54 2009 +0000 7078 7079 randr12: store nv0x hw cursor image in VRAM 7080 7081 nv_driver.c is already doing the alloc, nvidia uses the same method, and it saves code :) 7082 7083commit 3ea4b194e83735145d5e9d5c5357f92ae6c6a569 7084Author: Stuart Bennett <stuart@freedesktop.org> 7085Date: Thu Mar 12 15:37:59 2009 +0000 7086 7087 Kill pre-randr12 hwcursor code, and tidy randr12 hwcursor code 7088 7089 If you're not using randr12 you should be. 7090 7091 Misc fixes included: 7092 Ensure PNVM bit set on PCRTC_CURSOR_CONFIG for consistency 7093 Fix NV11 alpha cursors 7094 Add missing bo unref 7095 7096commit b988160dfa5a535fd042062d7318ab99fc7efa5a 7097Author: Stuart Bennett <stuart@freedesktop.org> 7098Date: Thu Mar 12 16:31:24 2009 +0000 7099 7100 Remove calls to load ddc and i2c modules -- they're built-in since server 1.3 7101 7102commit d5893ecf3dd915fc6937c61e6eae7c221edeb6e2 7103Author: Stuart Bennett <stuart@freedesktop.org> 7104Date: Thu Mar 12 21:46:02 2009 +0000 7105 7106 Hack around missing display table in rdivacky's GeForce 7050 PV/nForce 630a 7107 7108commit 369e1efd23c0180e210dbe898318b6b9efa1faad 7109Author: Stuart Bennett <stuart@freedesktop.org> 7110Date: Thu Mar 12 16:09:27 2009 +0000 7111 7112 Fix xserver shutdown for randr12 off 7113 7114commit 8343d6b2c0ebddf1a1a7d146890ed90278559fea 7115Author: Stuart Bennett <stuart@freedesktop.org> 7116Date: Wed Mar 11 01:11:02 2009 +0000 7117 7118 Remove bios lvds parsing pxclk test; ddc_permitted is a sufficient condition 7119 7120commit 8b56c699ad39c64326d76658c741801bb4118c5f 7121Author: Stuart Bennett <stuart@freedesktop.org> 7122Date: Wed Mar 11 00:43:33 2009 +0000 7123 7124 Name some regs/values and reduce crtc reg struct size 7125 7126 TVOUT_LATENCY taken from nv11 .scp file 7127 7128 CRTC[]: even nvidia only use up to 9f, and they know what the regs do 7129 7130commit 79d23d87b489332af521fbda418a13b9ae032b7d 7131Author: Ben Skeggs <skeggsb@gmail.com> 7132Date: Fri Mar 13 11:43:54 2009 +1000 7133 7134 improve drm/dri initialisation code 7135 7136 The main motivation here was to fix the scenario where X is started without 7137 nouveau.ko loaded, checks if kms is enabled, is told no because nouveau.ko 7138 isn't loaded, and then later nouveau.ko gets loaded with kms enabled by 7139 default and starts fighting with the ddx for the display. 7140 7141 One cool side-effect is that we're not depending on DRIScreenInit for our 7142 DRM fd now, which means that if DRIScreenInit fails for a non-fatal reason 7143 (like, NVIDIA's libglx being around instead of xorg's) we can still run 7144 with full 2D acceleration. 7145 7146commit bd1448216715681beebabe82b88c8b4d0e62eced 7147Author: Ben Skeggs <skeggsb@gmail.com> 7148Date: Wed Mar 11 16:07:41 2009 +1000 7149 7150 kms: cleanup set_property 7151 7152commit 7a455929785e00beb2d1e65007a3407427ca693a 7153Author: Ben Skeggs <skeggsb@gmail.com> 7154Date: Wed Mar 11 15:51:59 2009 +1000 7155 7156 kms: get current values for output properties on startup 7157 7158commit 7095e8942c74953fcf1615900e2d7864024ece5c 7159Author: Ben Skeggs <skeggsb@gmail.com> 7160Date: Wed Mar 11 11:30:08 2009 +1000 7161 7162 kms: work around some xserver stupidity with DPMS 7163 7164 The server calls the output DPMS functions with the intention of 7165 turning off outputs before reprogramming them. It probably shouldn't 7166 be doing this in the case of the randr driver using set_mode_major, 7167 and leave it up to the driver. 7168 7169 The DRM modesetting will usually reenable the outputs after a modeset 7170 anyway, which is the behaviour that the xserver currently relies on. 7171 However, there's a shortcut used in the drm to allow far faster fb 7172 swaps without doing a modeset. When this path is hit, the DRM will 7173 rightly not touch any power states and the display will never get 7174 turned back on. 7175 7176commit 8f9a58070c2fc55c922e9d82d0796e32959a8431 7177Author: Stuart Bennett <stuart@freedesktop.org> 7178Date: Tue Mar 10 00:20:35 2009 +0000 7179 7180 Another DCB 1.5 entry (seen in rh#455194) 7181 7182 Plus a few minor PLL tidyings 7183 7184commit 9a719900adcebf47442a30fe1989509f93e13e16 7185Author: Stuart Bennett <stuart@freedesktop.org> 7186Date: Mon Mar 9 15:57:06 2009 +0000 7187 7188 randr12: add output properties for digital vibrance and image sharpening 7189 7190commit 678ddc2e56aea3deb1e95b2309b47e1f2cce0635 7191Author: Stuart Bennett <stuart@freedesktop.org> 7192Date: Mon Mar 9 15:34:03 2009 +0000 7193 7194 randr12: functionate output properties creation 7195 7196 notes: 7197 * setting pending at change time is necessary for props created as pending, 7198 so that PostPending doesn't overwrite the correct current state with the 7199 incorrect pending one 7200 * our call to PostPending can go; relic from when props were updated in 7201 update_output_fields 7202 7203commit ed9bd8848fcfe6a77894bfe520a5aa25a9b3f769 7204Author: Ben Skeggs <skeggsb@gmail.com> 7205Date: Wed Mar 4 15:10:24 2009 +1000 7206 7207 kms: remove early-exit from gamma_set, the kernel side works now 7208 7209commit 98c9e4edb58374f18249e5c9c53b392fb8b4a1d1 7210Author: Ben Skeggs <skeggsb@gmail.com> 7211Date: Wed Mar 4 13:40:49 2009 +1000 7212 7213 kms: work around some bong hits with dpms 7214 7215 This'll need to get fixed elsewhere, but in the meantime, we have working 7216 DPMS *and* can actually see what's happening when X starts! 7217 7218commit 88ae2c92c0efc1919dbaa7e8db4c37b86987046c 7219Author: Ben Skeggs <skeggsb@gmail.com> 7220Date: Wed Mar 4 13:36:36 2009 +1000 7221 7222 kms: implement dpms 7223 7224commit 5269cc8060e0ce5160dfca408b0807a4cf2ff17c 7225Author: Stuart Bennett <stuart@freedesktop.org> 7226Date: Sun Mar 8 20:07:23 2009 +0000 7227 7228 randr12: make props code slightly less awful to behold 7229 7230commit 0b87c4964076101a2d8373fd3addd9355fb7765e 7231Author: Stuart Bennett <stuart@freedesktop.org> 7232Date: Sat Mar 7 21:13:55 2009 +0000 7233 7234 More robust PLL upclocking for old cards 7235 7236commit cb237a34f31f8b8c841974cb64141a2f2e36c53b 7237Author: Stuart Bennett <stuart@freedesktop.org> 7238Date: Sat Mar 7 04:38:12 2009 +0000 7239 7240 bios/randr12: mode validation fixes (includes fix for #20298) 7241 7242 Remove arbitrary 12000kHz mode clock lower bound (#20298) 7243 Use DCB supplied max dac clock (and remove a memcpy) 7244 7245commit 9cfccd7281a715e45a4296226d9d52854230f346 7246Author: Stuart Bennett <stuart@freedesktop.org> 7247Date: Sat Mar 7 04:34:20 2009 +0000 7248 7249 Handle failure to calculate pll values better 7250 7251 Ensure pll limit table get parsed on nv30 and nv35 too 7252 7253commit 1e994400913bc656b34440df67aa105aa2b211c4 7254Author: Stuart Bennett <sb476@cam.ac.uk> 7255Date: Fri Jan 16 22:26:28 2009 +0000 7256 7257 xv: return BadAlloc if destination pixmap is not put into EXA offscreen (#15792) 7258 7259 Lets the player get X errors, instead of the card choking trying to do system 7260 memory addresses 7261 7262 We could use exaDrawableIsOffscreen() here, but that's only in new xservers. 7263 7264commit 77347da64f86822f128e0dae3e89a6b80282c4ba 7265Author: Stuart Bennett <stuart@freedesktop.org> 7266Date: Fri Feb 27 01:22:13 2009 +0000 7267 7268 Remove duplicated DCB parsing 7269 7270 Store stuff in DCB private instead 7271 7272commit 1b6cabeee4c0160335c174bd4e9111dbd9535aaa 7273Author: Stuart Bennett <stuart@freedesktop.org> 7274Date: Tue Mar 3 01:03:45 2009 +0000 7275 7276 Remove VBIOS access in parse_dcb_entry 7277 7278commit 72fd0ae537e4db208bbfa142dd70144901d7f267 7279Author: Stuart Bennett <stuart@freedesktop.org> 7280Date: Tue Mar 3 21:51:46 2009 +0000 7281 7282 randr12: no need to re-detect edid for lvds, it's not going away 7283 7284 Reuse of nv_connector->edid is safe as nv_lvds_output_get_modes doesn't call 7285 nv_output_get_edid_modes after a mode has been found once 7286 7287commit ae981a8075395af40cf530f2aa24ac68d2ca0e77 7288Author: Maarten Maathuis <madman2003@gmail.com> 7289Date: Thu Mar 5 20:21:39 2009 +0100 7290 7291 nv50: some minor modesetting changes 7292 7293commit 42f99e68940a526f804e2046676c1d06d1bbbb42 7294Author: Ben Skeggs <skeggsb@gmail.com> 7295Date: Thu Mar 5 12:11:03 2009 +1100 7296 7297 nv30/xv: restore original viewport/clipping after putimage 7298 7299 Fixes two cases of corruption I seen while using the texture adaptors with 7300 a composite manager. 7301 - desktop getting clipped to video size sometimes 7302 - missing fonts on some menu items after xv+compmgr 7303 7304commit 56bb8f8e81e9d152217df7863e59781009a9bb77 7305Author: Stuart Bennett <stuart@freedesktop.org> 7306Date: Wed Mar 4 19:48:29 2009 +0000 7307 7308 Allow parse_fp_mode_table for x86 cards that find themselves on ppc 7309 7310commit 1c2f33ed33791f391d059778b1490222e8cd1e1d 7311Author: Stuart Bennett <stuart@freedesktop.org> 7312Date: Wed Mar 4 19:40:37 2009 +0000 7313 7314 Don't rely on the bios mobile feature bit for BMP biosen (#19986) 7315 7316 Quadro bios builds also set the mobile bit (not on nv10gl, but for all 7317 observed subsequent Quadros before nv4x (intro of BIT bios structure). 7318 So for BMP cards, use bit 6 of cr4b on head A to detect mobile biosen (all 7319 sampled mobile bios images set this in their init scripts, and non-mobile 7320 don't) 7321 7322commit e436c7ee98c2eda9d5ba45f7d897d68497b52799 7323Author: Stuart Bennett <stuart@freedesktop.org> 7324Date: Wed Mar 4 20:59:42 2009 +0000 7325 7326 randr12: make treatment of CR4B more closely resemble blob's behaviour 7327 7328 Possibly totally unnecessary anyway 7329 7330commit c61102b326455f6c63a55c8c73ed629994281588 7331Author: Stuart Bennett <stuart@freedesktop.org> 7332Date: Wed Feb 18 01:17:29 2009 +0000 7333 7334 Parse bios fp mode during modesetting, removing extra alloc 7335 7336 Also, it seems that the mode parsing was wrong on big endian for years, so 7337 fix that (though Apple laptops have DDC panels anyway) 7338 7339commit 631e15a03b8a790d342aefe677f8c5eb33935de6 7340Author: Stuart Bennett <stuart@freedesktop.org> 7341Date: Tue Mar 3 22:03:00 2009 +0000 7342 7343 Get dual_link and dithering bios info through lvds table parsing function 7344 7345 Now that lvds table is only parsed during modesetting, return the parsed 7346 data in out parameters to make the ordering requirements obvious 7347 7348 Since calling this on nv50 is pointless, that code was commented out 7349 7350commit 2d4377193f5e2166af3df32ddcfb1235be82764e 7351Author: Stuart Bennett <stuart@freedesktop.org> 7352Date: Tue Mar 3 21:52:04 2009 +0000 7353 7354 Call parse_lvds_manufacturer_table during modesetting only (#19986 partial fix) 7355 7356 Also remove SetOwner call now that locking crtcs doesn't change the set head 7357 7358 Moving parse_lvds_manufacturer_table to modesetting means it only gets 7359 called for lvds outputs, removing the dependency on checking for the mobile 7360 bios build. Another benefit is making the strap and EDID FP paths common. 7361 7362 A side effect is that as the lvds dithering toggle depends on the results of 7363 parse_lvds_manufacturer_table, setting that moves to after calling the 7364 parsing function 7365 7366commit ce8f5d71dacbe0d56cfcb17dfe60f1c4c6cf1773 7367Author: Stuart Bennett <stuart@freedesktop.org> 7368Date: Tue Mar 3 17:24:09 2009 +0000 7369 7370 Variety of LVDS script changes 7371 7372 Remove lvds invocation static 7373 No LVDS_INIT for BIT biosen 7374 Merge in one-time call to LVDS_INIT script to main script function 7375 Make sel_clk preservation stuff use regular reg writing functions 7376 7377commit 071d5373df744c0c9578134fa634bce1e202d4f3 7378Author: Stuart Bennett <stuart@freedesktop.org> 7379Date: Tue Feb 17 16:45:50 2009 +0000 7380 7381 Read embedded bios edid only on demand 7382 7383commit f04689cb326da409448d54694e2660cf4b4b7950 7384Author: Ben Skeggs <skeggsb@gmail.com> 7385Date: Tue Mar 3 10:14:26 2009 +1000 7386 7387 move memset of fb after NVSave(), hopefully fix lost VGA fonts 7388 7389commit 6961efb84dd6894aaaf010a9f620231dfcbad004 7390Author: Ben Skeggs <skeggsb@gmail.com> 7391Date: Mon Mar 2 11:35:05 2009 +1000 7392 7393 kms: fix check for xf86drmMode.h 7394 7395commit 76fa6569ffd1a7f61e295753383bd01b246b5348 7396Author: Stuart Bennett <stuart@freedesktop.org> 7397Date: Mon Mar 2 16:43:06 2009 +0000 7398 7399 Detypedef bios_t, rename bios io functions, group various helper functions 7400 7401commit 39c9d0231c0d24bf91dbe9ab262f54fcad1b3b2f 7402Author: Stuart Bennett <stuart@freedesktop.org> 7403Date: Fri Feb 27 18:57:35 2009 +0000 7404 7405 Move parsed DCB to bios structs, split into public and private parts 7406 7407commit 275daaf7be46e54ca59067d3f2a6e6790e1ce7e0 7408Author: Stuart Bennett <stuart@freedesktop.org> 7409Date: Fri Feb 27 17:46:26 2009 +0000 7410 7411 Split bios struct into struct for nv_bios.c, and exported public info struct 7412 7413commit 8b3e966e3230add04aa05a9ded3f8a647afe0d91 7414Author: Stuart Bennett <stuart@freedesktop.org> 7415Date: Sun Mar 1 20:50:15 2009 +0000 7416 7417 Remove never updated bios opcodes 0x31, 0x5A and 0x73 7418 7419 Notices of removal left in case any bios using them ever appears, so they 7420 can be found in git history 7421 7422 Also nuke a few unused reg access defines 7423 7424commit ce9eb05f516899f56e66cc290fc7c4f9d3085882 7425Author: Stuart Bennett <stuart@freedesktop.org> 7426Date: Fri Feb 27 23:14:24 2009 +0000 7427 7428 Trivial bios convenience variable changes 7429 7430commit cfeef1144943e72d62a2c0d9701ddf4a639deeca 7431Author: Maarten Maathuis <madman2003@gmail.com> 7432Date: Fri Feb 27 17:30:40 2009 +0100 7433 7434 exa: forgot to check if EXA_SUPPORTS_PREPARE_AUX was defined 7435 7436commit 37a85dcbd6f6134a43b5c9bb883553fdaf7c0bd9 7437Author: Maarten Maathuis <madman2003@gmail.com> 7438Date: Fri Feb 27 17:23:42 2009 +0100 7439 7440 exa: we support PREPARE_AUX 7441 7442commit 1249fd25a529702a4c6fb4831ef92e801f0b6122 7443Author: Maarten Maathuis <madman2003@gmail.com> 7444Date: Fri Feb 27 17:21:47 2009 +0100 7445 7446 Revert "exa: support major version 3" 7447 7448 This reverts commit b7f99aec7282ca7c81d19ff3ad7d9ff22275d476. 7449 7450commit 1c4a284a80ebed9f9d1e01c47b929481801566b5 7451Author: Stuart Bennett <stuart@freedesktop.org> 7452Date: Fri Feb 20 20:02:38 2009 +0000 7453 7454 Rework PLL decoding in terms of pll_vals, save/restore pll_vals on vt switch 7455 7456 If your nv04-nv4x vt after running nouveau is more broken than it was with 7457 the previous commit, file a bug 7458 7459commit 81b4732162915ab1ac0593e0ef21b56b089dafae 7460Author: Stuart Bennett <stuart@freedesktop.org> 7461Date: Fri Feb 20 19:11:41 2009 +0000 7462 7463 New common struct for pll values 7464 7465 Makes the compiler do our dirty accessing work for us 7466 7467commit adf2e350341b983d27123550041c3b1ddaac4526 7468Author: Stuart Bennett <stuart@freedesktop.org> 7469Date: Wed Feb 25 02:11:27 2009 +0000 7470 7471 Check whether NV_RAMDAC_580 has changed when setting PLLs 7472 7473commit b7f99aec7282ca7c81d19ff3ad7d9ff22275d476 7474Author: Maarten Maathuis <madman2003@gmail.com> 7475Date: Tue Feb 24 19:13:23 2009 +0100 7476 7477 exa: support major version 3 7478 7479commit d91fc78859b11d00c822e5cae70f0fc511b26701 7480Author: Ben Skeggs <skeggsb@gmail.com> 7481Date: Tue Feb 24 08:31:40 2009 +1000 7482 7483 kms: fix resize func 7484 7485 Don't change virtualX/virtualY when we're not resizing the actual framebuffer, 7486 just change displayWidth to account for the adjusted scanout size. 7487 7488 With driver pixmaps, we'll be able to do "proper" framebuffer resize. 7489 7490commit 535498bcfe79858e08ae1f01125025f56bcd504f 7491Author: Ben Skeggs <skeggsb@gmail.com> 7492Date: Mon Feb 23 08:20:34 2009 +1000 7493 7494 clear the scanout buffer on statup to avoid seeing garbage on-screen 7495 7496commit 022767373474a0dacf50ee6fcd6401db15018aca 7497Author: Ben Skeggs <skeggsb@gmail.com> 7498Date: Fri Feb 20 13:59:58 2009 +1000 7499 7500 nv50/exa: properly align offset for mem_tile ioctl 7501 7502 I really really hate this ioctl().. 7503 7504commit e09f50e5f9126c682289b7ba790f16b93a80b876 7505Author: Stuart Bennett <stuart@freedesktop.org> 7506Date: Wed Feb 18 21:04:00 2009 +0000 7507 7508 bios/randr12: fix dual link TMDS (#20006) 7509 7510 TMDS04 needs setting, just as for LVDS, and bit 31 of fp_control should also 7511 be set (although in this instance the scripts ensure it is set) 7512 Furthermore, stop dpms clobbering any changes made to fp_control by scripts 7513 7514 Many thanks to bmaass for testing 7515 7516commit b968db77037790dd5be91d6fafa0552eb7e9520f 7517Author: Stuart Bennett <stuart@freedesktop.org> 7518Date: Mon Feb 16 21:38:31 2009 +0000 7519 7520 Return of "bios: fix create_i2c_device for g80" 7521 7522commit 0a439ba05aa192e7433d430acd91b60719d20caf 7523Author: Stuart Bennett <stuart@freedesktop.org> 7524Date: Wed Feb 11 20:52:43 2009 +0000 7525 7526 Parse DCB before running init 7527 7528commit 870b6f0d2df3e64af5ac921594b1742f82329c49 7529Author: Stuart Bennett <sb476@cam.ac.uk> 7530Date: Tue Jan 20 05:29:38 2009 +0000 7531 7532 Split parsing and init into separate functions 7533 7534commit 0280a0c975830b5c4b2edf4fff8a79fb7d67b9df 7535Author: Stuart Bennett <stuart@freedesktop.org> 7536Date: Wed Feb 11 18:53:56 2009 +0000 7537 7538 Attempt to remove pNv and bios_t use in parsing the DCB 7539 7540 It's meant to parse a section of rom image into useful things, so use of pNv 7541 ought to be unusual. 7542 Access to things contained in pNv and bios_t makes it harder to prove parsing 7543 ordering requirements 7544 7545commit 9179c6667cf859248a525c8d01922b69ed45c17b 7546Author: Stuart Bennett <stuart@freedesktop.org> 7547Date: Wed Feb 11 18:33:47 2009 +0000 7548 7549 Variety of DCB enhancements 7550 7551 * ensure parsing of dcb_table.i2c_default_indices happens 7552 * only merge entries on dcb versions < 2.1, and note consequnces of merging 7553 * correct dcbent->index on post-merge compacted entries 7554 * drop pointless memcpy 7555 * return an error if no outputs found 7556 * define dcb structure in nouveau_bios.h (ought to rename pNv->dcb_table to pNv->dcb one day) 7557 7558commit 317ca477cdbb936f29360984bcee25a0ecda3aa3 7559Author: Stuart Bennett <stuart@freedesktop.org> 7560Date: Mon Feb 16 21:15:51 2009 +0000 7561 7562 Revert "bios: fix create_i2c_device for g80" 7563 7564 This reverts commit ca72ca24784b9da09b6e3e0b39f3aa32b8f7b502. 7565 It'll be back in a minute. 7566 7567commit 7b25a3019beba6342505fa16327a5bcd43669f90 7568Author: Maarten Maathuis <madman2003@gmail.com> 7569Date: Mon Feb 16 21:23:56 2009 +0100 7570 7571 nv04: unbreak copy 7572 7573commit 60c8bb1391a51188a4981b4936c190c149b79b7f 7574Author: Ben Skeggs <skeggsb@gmail.com> 7575Date: Mon Feb 16 13:09:39 2009 +1000 7576 7577 default to autodetecting whether to enable the driver's kms paths 7578 7579commit a31b1d0118dc7b4d07a2c232fcd5aafbf94f69b9 7580Author: Ben Skeggs <skeggsb@gmail.com> 7581Date: Mon Feb 16 09:49:39 2009 +1000 7582 7583 exa: reimplement driver-controlled pixmaps 7584 7585commit d97e993072dc773ecfeb80b5e138b5e4598a1c3d 7586Author: Ben Skeggs <skeggsb@gmail.com> 7587Date: Mon Feb 16 08:50:02 2009 +1000 7588 7589 exa: preparation for reintroducing driver-controlled pixmaps 7590 7591commit 1e00c7d1e6edb7f6b4e6d12ef227d27413948f8f 7592Author: Maarten Maathuis <madman2003@gmail.com> 7593Date: Sun Feb 15 18:52:59 2009 +0100 7594 7595 shadowfb: clip copied regions to frontbuffer size. 7596 7597 - There have been reports of crashed when moving windows outside the visible screen. 7598 7599commit efe27964fa25f71dcbdc7a751210860369ea0e11 7600Author: Maarten Maathuis <madman2003@gmail.com> 7601Date: Sun Feb 15 16:54:18 2009 +0100 7602 7603 nv50: correct a poorly written comment 7604 7605commit 2573c065423af47e14263479c66f2308259f29b8 7606Author: Ben Skeggs <skeggsb@gmail.com> 7607Date: Fri Feb 13 10:46:13 2009 +1000 7608 7609 exa: remove exaWaitSync() call in nouveau_exa_pixmap_map() 7610 7611 unnecessary, nouveau_bo_map() will sync if required. 7612 7613commit 500053517cb5fe058468c8965f43e6c3209ba7b5 7614Author: Ben Skeggs <skeggsb@gmail.com> 7615Date: Thu Feb 12 09:04:43 2009 +1000 7616 7617 exa: don't need exaMarkSync on DFS, the operation will be done already 7618 7619commit a7e7c557dbf9b2fe54db401793608128de08bdf5 7620Author: Ben Skeggs <skeggsb@gmail.com> 7621Date: Thu Feb 12 09:01:49 2009 +1000 7622 7623 xv: remove notifier usage 7624 7625commit 8ecb8d5331bdb910beea416cdb0b662c624c5c96 7626Author: Ben Skeggs <skeggsb@gmail.com> 7627Date: Thu Feb 12 09:00:07 2009 +1000 7628 7629 exa: remove notifier waits in UTS/DFS 7630 7631 nouveau_bo will handle CPU/GPU sync if necessary when bo_map() is called. 7632 7633commit cfb884b816eebbdaa2103dd233c6f975834e536d 7634Author: Maarten Maathuis <madman2003@gmail.com> 7635Date: Fri Feb 13 22:29:33 2009 +0100 7636 7637 bios: add some more valid regs 7638 7639commit ca72ca24784b9da09b6e3e0b39f3aa32b8f7b502 7640Author: Maarten Maathuis <madman2003@gmail.com> 7641Date: Fri Feb 13 22:18:18 2009 +0100 7642 7643 bios: fix create_i2c_device for g80 7644 7645commit bbd7041791eacd357afa775236d4123dd98ab9c9 7646Author: Maarten Maathuis <madman2003@gmail.com> 7647Date: Fri Feb 13 20:30:24 2009 +0100 7648 7649 bios: implement opcode 0x76 (INIT_IO_CONDITION). 7650 7651commit 5dd2cbe454ee8560c814f3fd96d7c28e0fec1dc1 7652Author: Maarten Maathuis <madman2003@gmail.com> 7653Date: Thu Feb 12 17:13:42 2009 +0100 7654 7655 exa: add nouveau_exa_pixmap_is_tiled 7656 7657commit 15ccf721791fa4f61784ad00449630c6d2496328 7658Author: Stuart Bennett <stuart@freedesktop.org> 7659Date: Sat Feb 7 16:40:46 2009 +0000 7660 7661 randr12: anti-crash measures for #19854 7662 7663commit 17e9d4bb8c46d67e201b2a369eff7db805dff07b 7664Author: Stuart Bennett <stuart@freedesktop.org> 7665Date: Fri Apr 4 19:01:20 2008 +0100 7666 7667 randr12: use bios PLL setting routines for modesetting 7668 7669 Only used for setting modes currently, old code remains for restore 7670 7671commit e444e0e348b3aab59671ffdeb605f9a57c46bf70 7672Author: Stuart Bennett <stuart@freedesktop.org> 7673Date: Tue Jan 20 03:23:40 2009 +0000 7674 7675 Readability improvements for nv_hw.c and nvreg.h 7676 7677commit f5ea66c2ef2c726b6a0aff72843c379ab25dbaff 7678Author: Stuart Bennett <stuart@freedesktop.org> 7679Date: Tue Feb 10 19:01:43 2009 +0000 7680 7681 Pull in extras from CRTC PLL code to bios routines 7682 7683 * g7x magic bits 7684 * 0xc040 manipulation for vplls 7685 7686commit 416636509152cf12c801ac537cb8c1bd20c19b5e 7687Author: Stuart Bennett <stuart@freedesktop.org> 7688Date: Tue Jan 20 02:11:08 2009 +0000 7689 7690 Pull out ramdac580 setting and powerctrl_1 shift code to separate functions 7691 7692 And only twiddle powerctrl_1 for vplls on nv3x single stage pll chips 7693 7694commit 8b7041833c11f32feb7fa5edc21648cc9db1a333 7695Author: Stuart Bennett <stuart@freedesktop.org> 7696Date: Tue Feb 10 19:01:05 2009 +0000 7697 7698 Miscellaneous fixes to bios PLL code 7699 7700 * bios variable was only used for chip_version. simplify. 7701 * set plls faster on single pll cards after nv15/16 7702 * allow early return for single stage plls when already set 7703 * single stage pll mode is only for nv41+ 7704 * 80 cols fixes 7705 7706commit 01bb4c1576a45217953632c9f5612aea9173f95f 7707Author: Stuart Bennett <stuart@freedesktop.org> 7708Date: Mon Feb 9 21:14:43 2009 +0000 7709 7710 Make PROM score badly, out of available bios images, if checksum bad 7711 7712 Problem exposed on a Mac following 0387ac32 7713 7714 Having PROM first is good, if the image is fine. But if the read-only PROM 7715 has problems, not a good idea to use it if there are other options available 7716 7717 Also avoid result contamination, if a read method has returned early and not 7718 overwritten previous image in *data 7719 7720commit 30848dd378d05917c646a24df404846e6f9b2352 7721Author: Stuart Bennett <sb476@cam.ac.uk> 7722Date: Tue Jan 6 16:23:33 2009 +0000 7723 7724 randr12: bodge-in continuous timings when the scaler is in use (#19203) 7725 7726 Similar to 3354e660b0744976871683ce226f17e873f26b50 in the intel driver, except 7727 I can't see much benefit in allowing bigger-than-native resolutions, so no sync 7728 range hacking 7729 7730 A pity the xserver has no idea of scalers and so each driver ends up with a hack 7731 7732commit 0387ac32bef6c714f54917d5d36071ae1224458c 7733Author: Stuart Bennett <stuart@freedesktop.org> 7734Date: Sun Feb 8 09:19:53 2009 +0000 7735 7736 Allow reading of PROM and PCI rom on PPC (for people using x86 cards) 7737 7738commit 6ad66c0b6165822dde167fd1700bd14727ce7087 7739Author: Stuart Bennett <sb476@cam.ac.uk> 7740Date: Sun Jan 25 20:09:36 2009 +0000 7741 7742 bios/randr12: be strict about when to use modes from bios vs modes from ddc 7743 7744 Again, the Inspiron 8100 in #14567 is an example of wackiness, having a 7745 non-0xf i2c_index, but using straps 7746 7747 If the previous commit didn't break your DCB 2.0 EDID equipped laptop, this will 7748 7749commit b5391eb909f8b477c3c438f1136d788acfe6bfd8 7750Author: Stuart Bennett <sb476@cam.ac.uk> 7751Date: Sun Jan 25 03:03:29 2009 +0000 7752 7753 Simplify FP strapping semantics and trust bios parsing of use_straps_for_mode 7754 7755 In many ways a spiritual revert of 11de8d 7756 7757 The compound fp.strapping member faithfully implemented what nvidia sticks in 7758 cr58 (for cr57 == 0xf), but this was never directly useful and had differing 7759 overloaded interpretations with bios version, causing confusion. Return to 7760 just using the result of get_fp_strap (g80 code of which unused and so dropped) 7761 7762 As noted in the commit, there have been DCB 2.0 laptops (as in bug 14567) which 7763 don't set use_straps_for_mode when they should, so we set it for all DCB 1.4-2.0 7764 cards. If EDID-ful DCB 1.4-2.0 laptops exist they will be broken, and will need 7765 fixing later (simplistic fix: use_straps_for_mode = (get_fp_strap() == 0xf) ) 7766 7767 Will also break nv40 straps-based laptops whose tables used reserved values, 7768 but that's pretty broken anyway, and I've no evidence they exist 7769 7770commit 34c2b1539bb2d69dd3bd5911c4fac62e46385d20 7771Author: Stuart Bennett <sb476@cam.ac.uk> 7772Date: Sat Jan 24 18:41:31 2009 +0000 7773 7774 bios/randr12: misc (mainly lvds related) clarifications and cleanups 7775 7776 get_fp_strap() moved (& unused g80 bodge removed) in anticipation of next commit 7777 7778commit 98b8cada6c355d437925a92ef0413e96751ed567 7779Author: Maarten Maathuis <madman2003@gmail.com> 7780Date: Sat Feb 7 21:02:31 2009 +0100 7781 7782 Always allocate 2 hw cursors. 7783 7784 - This hopefully fixes the legacy codepath again. 7785 7786commit f1099810bb3cfe451877667a0224eb3d664e442f 7787Author: Maarten Maathuis <madman2003@gmail.com> 7788Date: Sat Feb 7 20:04:54 2009 +0100 7789 7790 Fix a few warnings. 7791 7792commit 691589d8c05b8b0354e359b4942711f6fffa9903 7793Author: Maarten Maathuis <madman2003@gmail.com> 7794Date: Sat Feb 7 19:59:28 2009 +0100 7795 7796 Fix some cursor fallout, mostly in legacy code. 7797 7798commit 945f0cb644b6e4f34cc04927177fe53529aeb5cf 7799Author: Ben Skeggs <skeggsb@gmail.com> 7800Date: Fri Feb 6 00:44:14 2009 +1000 7801 7802 kms: update drmmode_display 7803 7804commit ba1f897159d910a25560796d11c9d0357869e1c0 7805Author: Ben Skeggs <skeggsb@gmail.com> 7806Date: Thu Feb 5 23:15:24 2009 +1000 7807 7808 keep resources around after leavevt, far too problematic currently 7809 7810commit 4dfd0b1b21415c48c4b6e0e8d335f3fc70922d27 7811Author: Ben Skeggs <skeggsb@gmail.com> 7812Date: Wed Feb 4 15:35:53 2009 +1000 7813 7814 oops 7815 7816commit e02813aab0376271ee777a2b8dafc9381f96311a 7817Author: Ben Skeggs <skeggsb@gmail.com> 7818Date: Wed Feb 4 14:50:35 2009 +1000 7819 7820 link against libdrm_nouveau (installed with libdrm) 7821 7822commit 2c06308a6c5878fca3ff4f2f9c969705d966607f 7823Author: Ben Skeggs <skeggsb@gmail.com> 7824Date: Wed Feb 4 14:09:21 2009 +1000 7825 7826 require drm 0.0.12 7827 7828commit 29aa1341fb28578968953f7e7dc3de0ec66f70f5 7829Author: Ben Skeggs <skeggsb@gmail.com> 7830Date: Mon Feb 2 11:13:55 2009 +1000 7831 7832 fix NoAccel from when map/unmap became stricter 7833 7834commit 62adc557560fa82c2086e32cd16bc51f67a28b48 7835Author: Ben Skeggs <skeggsb@gmail.com> 7836Date: Tue Feb 3 08:10:45 2009 +1000 7837 7838 Revert "Improve the NoAccel situation, but it's not working yet." 7839 7840 This reverts commit 285027a180b676ac63f4c7e158e94ffbe22bf11d. 7841 7842 Sorry, fixed this yesterday and forgot to push the commit. 7843 7844commit 285027a180b676ac63f4c7e158e94ffbe22bf11d 7845Author: Maarten Maathuis <madman2003@gmail.com> 7846Date: Mon Feb 2 11:15:02 2009 +0100 7847 7848 Improve the NoAccel situation, but it's not working yet. 7849 7850 - nouveau_bo related functions will cause problems for fake bo's (since they don't have a bo_priv). 7851 - NoAccel still doesn't work for me (the scanout contains random crap). 7852 7853commit 5e63c7866ac93f39fde49af185387d2094110627 7854Author: Ben Skeggs <skeggsb@gmail.com> 7855Date: Mon Feb 2 13:46:01 2009 +1000 7856 7857 Revert "nv50: make entire offscreen area tiled, use extra blits to scanout buffer" 7858 7859 This reverts commit 6c3f8da1bff3c94353748321b40a523add824327. 7860 7861 EXA doesn't use UTS/DFS hooks when it's accessing the "front buffer" for 7862 software fallbacks, so sw rendering to it was garbled because it doesn't 7863 understand tiling. 7864 7865 The Prepare/FinishAccess hooks could be used to point the pixmap at the 7866 linear scanout buffer instead for sw access, but on VT switch for some 7867 reason exa get called to try and do things, hit the PrepareAccess() hook 7868 and die horribly because in LeaveVT() the driver deallocates the buffers. 7869 7870 The other alternative was to use libwfb to wrap the fb accesses and 7871 handle tiling, but exa isn't quite ready for that yet either - it's 7872 being worked on. 7873 7874 Another point made about this patch was that we should figure out the 7875 display engine and get it to scanout the tiled buffer directly, not 7876 commit shoddy workarounds. Even for this "correct" solution, the two 7877 above-mentioned issues still apply. 7878 7879commit 6c3f8da1bff3c94353748321b40a523add824327 7880Author: Ben Skeggs <skeggsb@gmail.com> 7881Date: Mon Feb 2 08:52:18 2009 +1000 7882 7883 nv50: make entire offscreen area tiled, use extra blits to scanout buffer 7884 7885 This fixes some pretty bad performance issues with window resize and desktop 7886 switching where ARGB windows are concerned. It also allows Xv to be used 7887 without a composite manager. 7888 7889 The drawbacks are the use of a little more memory for the shadow scanout 7890 buffer, and a bit more GPU time is used for the extra blits. But it's worth 7891 it. 7892 7893 If you want to get rid of this, figure out how to scanout tiled buffers :) 7894 7895commit 61791028ee9321748b635a85d4e53ed80d143df5 7896Author: Ben Skeggs <skeggsb@gmail.com> 7897Date: Sun Feb 1 18:58:27 2009 +1000 7898 7899 nv10: use the correct 3d object on NV1A 7900 7901commit dfd87ece014402339f82ecc26bc27b2e47e213df 7902Author: Younes Manton <younes.m@gmail.com> 7903Date: Sat Jan 31 16:52:46 2009 -0500 7904 7905 xv: Unmap filter table BO after init so it can be validated (nv30,40) 7906 7907commit b8de749bbaeacbbe677df1939f7fd39d9d3863a7 7908Author: Stuart Bennett <sb476@cam.ac.uk> 7909Date: Sat Jan 24 05:40:02 2009 +0000 7910 7911 Avoid any risk of parsing oscillation on if_is_24bit 7912 7913 Would depend on having a BMP laptop with EDID equipped panel and a bios that 7914 doesn't match current suspicions, but better safe 7915 7916commit 0a0bb720da63e4a323978e4a52a4d6e1e4f4ae27 7917Author: Stuart Bennett <sb476@cam.ac.uk> 7918Date: Wed Jan 21 00:33:55 2009 +0000 7919 7920 Split up parsing the bit structure, complain when necessary tables are missing 7921 7922 Added bios major version 0x60 (g80+) check since it now complains about missing 7923 the BIT "A" table on nv4x roms 7924 Also permits on-demand parsing of individual tables if needed in the future 7925 7926commit 2d86018d52c387d4f2370b7c1ff83798ad6e1cf7 7927Author: Stuart Bennett <sb476@cam.ac.uk> 7928Date: Sun Jan 25 20:12:41 2009 +0000 7929 7930 Swap the order of calling parse_fp_mode_table and parse_lvds_manufacturer_table 7931 7932 Last seen in this order pre 11de8d7a 7933 7934commit 8c6b3a800ae2df35b6475a46aebab32f0577261a 7935Author: Stuart Bennett <sb476@cam.ac.uk> 7936Date: Tue Jan 20 04:04:43 2009 +0000 7937 7938 Make calling of init bios functions common 7939 7940 Bios structure parsing stage should no longer require hw access 7941 This involves serializing BIT hw-accessing functions after BIT structure 7942 parsing, but order is maintained so should cause no issues 7943 7944 Also, set a digital_min_front_porch default on BIT PPC (G5s), as they have no BIT 7945 "display" table (might even fix some wonky display issues like #17571) 7946 7947commit 9b32034eb72e580d4cb5a5e2fa22a76eff0e7f26 7948Author: Stuart Bennett <sb476@cam.ac.uk> 7949Date: Thu Jan 29 04:07:04 2009 +0000 7950 7951 Don't read obviously useless bios images, and drop 5x reading on PROM 7952 7953 If some nv43 actually need the 5x bodge (no pramin image) we can add it back as 7954 a method of last resort 7955 PCIR check taken from mmiotrace of the blob's PROM behaviour 7956 7957commit 63aa210f7aee4674337fed3ebfebd198ff1e699b 7958Author: Stuart Bennett <sb476@cam.ac.uk> 7959Date: Wed Jan 28 16:08:42 2009 +0000 7960 7961 Implement bios I2C opcode 4E (untested) 7962 7963 Courtesy an old trace by RAOF 7964 7965commit a81f1590d384bad214111139db9c988ff7676e83 7966Author: Stuart Bennett <sb476@cam.ac.uk> 7967Date: Wed Jan 28 15:39:24 2009 +0000 7968 7969 Handle I2C device 0xff fallback for I2C bios opcodes 7970 7971 Courtesy an old trace by RAOF 7972 Also, use i2c record entry length when available and make rundigitaloutscript 7973 function name legible 7974 7975commit 34e967e689a186cdde2f6c3ec222a6870b3584ea 7976Author: Stuart Bennett <sb476@cam.ac.uk> 7977Date: Wed Jan 28 15:48:27 2009 +0000 7978 7979 Fix pasta error in bios opcode 4D 7980 7981commit 786dd366474006b10c740c5012630ff5f1cd2a0f 7982Author: Stuart Bennett <sb476@cam.ac.uk> 7983Date: Thu Jan 29 17:56:40 2009 +0000 7984 7985 Fix nv04+ IFC upload and frag prog upload 7986 7987commit 3ab65cf6af0d3f892ade584c29f861f9f8a9d021 7988Author: Ben Skeggs <skeggsb@gmail.com> 7989Date: Thu Jan 29 16:51:53 2009 +1000 7990 7991 add vm_vram_base field to nouveau_device instead of hardcoding 512MiB 7992 7993commit d833c81fadf7dd63ab4ca94b953397fcbe0be581 7994Author: Ben Skeggs <skeggsb@gmail.com> 7995Date: Thu Jan 29 16:25:14 2009 +1000 7996 7997 nv50/xv: previous xv commit was bonged - fix that 7998 7999 It actually still worked just fine, but it was wrong. It's possible we'll 8000 have more than 1 clipbox, in which case the check for enough space in the 8001 pushbuf for a single frame is insufficient. 8002 8003 Check for enough space just before emitting quads, if there's not enough 8004 flush and reemit state - then go again. 8005 8006commit 0f741eb09c556f8d7893628964fedea862322771 8007Author: Ben Skeggs <skeggsb@gmail.com> 8008Date: Thu Jan 29 16:10:18 2009 +1000 8009 8010 exa: small cleanup 8011 8012commit c10c3c9bbabe8682fe1500098dc53f631690b182 8013Author: Ben Skeggs <skeggsb@gmail.com> 8014Date: Thu Jan 29 15:43:51 2009 +1000 8015 8016 exa: mass renaming and movement of nv04 solid/copy code to nv04_exa.c 8017 8018commit 81ace5dd1fa2c2760036fb1371cc61a383b2698e 8019Author: Ben Skeggs <skeggsb@gmail.com> 8020Date: Thu Jan 29 15:29:37 2009 +1000 8021 8022 exa: remove driver pixmaps path that never worked here anyway 8023 8024 Bits and pieces remain (disabled) in various modesetting bits for reference. 8025 Some of it may be useful when it reappears again. 8026 8027commit 4b1829e2c66f92a19a233afb749f3fb946930db5 8028Author: Ben Skeggs <skeggsb@nisroch.(none)> 8029Date: Tue Oct 7 00:44:46 2008 +1100 8030 8031 exa: remove NVCopyROP table.. 8032 8033commit 410788dc896d99b658508e3a7ebcb4940a039f35 8034Author: Ben Skeggs <skeggsb@gmail.com> 8035Date: Thu Jan 29 14:32:13 2009 +1000 8036 8037 blow up horribly if GPU access to mapped buffer is attempted 8038 8039 ...and fix the numerous cases where this was done. The current bufmgr couldn't 8040 care less, and synchronisation is done elsewhere. This won't be the case for 8041 much longer. 8042 8043commit 4c9346bd750797ee0f0485195f7ce8beb2114b04 8044Author: Ben Skeggs <skeggsb@gmail.com> 8045Date: Thu Jan 29 14:24:57 2009 +1000 8046 8047 bo_del->bo_ref, closer to ng api 8048 8049commit c533f86b512b9d84ec698163a8fbc54233648fbb 8050Author: Ben Skeggs <skeggsb@gmail.com> 8051Date: Thu Jan 29 14:13:06 2009 +1000 8052 8053 exa: same for UTS state 8054 8055commit 02677c07f9af01fa3b1eb787d6d320aa43c9140b 8056Author: Ben Skeggs <skeggsb@gmail.com> 8057Date: Thu Jan 29 14:10:51 2009 +1000 8058 8059 exa: ensure DFS state is updated across flushes 8060 8061commit 69d02bc73409c8010d19c9f87780b0c0e7621e43 8062Author: Ben Skeggs <skeggsb@gmail.com> 8063Date: Thu Jan 29 13:52:08 2009 +1000 8064 8065 nv50: remove some redundant init, this stuff is done as required now 8066 8067commit 60aea9057a78255a0ded1f64a9b9efd75017347f 8068Author: Ben Skeggs <skeggsb@gmail.com> 8069Date: Thu Jan 29 13:50:28 2009 +1000 8070 8071 nv50: forgotten buffer access flags during init 8072 8073commit 5a382dddc948c49f3f5159c078875ee6cdf56db7 8074Author: Ben Skeggs <skeggsb@gmail.com> 8075Date: Thu Jan 29 13:04:04 2009 +1000 8076 8077 nv10/exa: resubmit state in composite as needed 8078 8079commit d4ff8db7b0fb34e0494042b128251ad78c19ad09 8080Author: Ben Skeggs <skeggsb@gmail.com> 8081Date: Thu Jan 29 12:58:30 2009 +1000 8082 8083 nv30/exa: resubmit state in composite as needed 8084 8085commit b71ebcb2c8a0c97837f3de09f034c0bbb35f562c 8086Author: Ben Skeggs <skeggsb@gmail.com> 8087Date: Thu Jan 29 12:56:39 2009 +1000 8088 8089 nv40/exa: resubmit state in composite as necessary 8090 8091commit 6ddaddbb7b9442000f1e665bf61994b16d52cd25 8092Author: Ben Skeggs <skeggsb@gmail.com> 8093Date: Thu Jan 29 12:11:24 2009 +1000 8094 8095 exa: rework common pre-nv50 code in the same way nv50 code just was 8096 8097 Still have card-specific composite hooks to go. 8098 8099commit f36fa31401b060b9785d1835845c8ce09f8fe6b1 8100Author: Ben Skeggs <skeggsb@gmail.com> 8101Date: Thu Jan 29 11:25:22 2009 +1000 8102 8103 nv50/xv: ensure entire frame will fit in pushbuf without causing a flush 8104 8105commit 1b48331f6bd077f39797bb2d25b62c3c9df5ccb3 8106Author: Ben Skeggs <skeggsb@gmail.com> 8107Date: Thu Jan 29 10:38:53 2009 +1000 8108 8109 nv50/exa: ensure buffer usage is tracked correctly 8110 8111commit 7e644265267b4e35a05328b1c4cb1f2ef5beea58 8112Author: Ben Skeggs <skeggsb@gmail.com> 8113Date: Thu Jan 29 10:13:57 2009 +1000 8114 8115 remove unconditional FIRE_RING() in DoneComposite() hooks 8116 8117 Far, far too much flushing. Will completely destroy performance with kernel 8118 command submission. 8119 8120commit 34fc4461a6f612e4e00db483effe3a2b9760e70c 8121Author: Ben Skeggs <skeggsb@gmail.com> 8122Date: Thu Jan 29 09:47:21 2009 +1000 8123 8124 hack a flush_notify() hook in to match ng 8125 8126 preparation for rework of accel code so it'll be more suitable for ng-style 8127 command submission. 8128 8129commit d9da090cae1d2dfc539c9db4c923c0e665bce975 8130Author: Ben Skeggs <skeggsb@gmail.com> 8131Date: Thu Jan 29 09:38:39 2009 +1000 8132 8133 Drop GPU resources on LeaveVT, reaquire on EnterVT 8134 8135 This allows GDM's switch-user functionality to work on some systems. 8136 8137commit 1eb8b8778037336e530e113b8bbd61d724b1cc52 8138Author: Ben Skeggs <skeggsb@gmail.com> 8139Date: Wed Jan 28 13:50:55 2009 +1000 8140 8141 Enable ShadowFB for DRM-less mode. 8142 8143commit 487463895390500fd4e899f4e3badbb44f2b5429 8144Author: Ben Skeggs <skeggsb@gmail.com> 8145Date: Wed Jan 28 13:41:44 2009 +1000 8146 8147 Allow NoAccel operation without the DRM module present. 8148 8149commit f78aeed720ef185ccdea2d62e02b2714f00ee49c 8150Author: Ben Skeggs <skeggsb@gmail.com> 8151Date: Wed Jan 28 10:30:43 2009 +1000 8152 8153 if CARD_INIT ioctl fails, report it instead of continuing 8154 8155commit cf65b875ba3467060d8a3806914b75c710504b78 8156Author: Stuart Bennett <sb476@cam.ac.uk> 8157Date: Wed Jan 21 05:22:34 2009 +0000 8158 8159 Fix stupidity introduced in 25de6b867f319099dac05ba84f170da0f0e6c803 8160 8161 Wow. (function executes before bios is parsed, hence version check never works) 8162 8163commit 55e9cdda6e555937e8f5f3998a2e603669b84df4 8164Author: Stuart Bennett <sb476@cam.ac.uk> 8165Date: Tue Jan 20 05:09:22 2009 +0000 8166 8167 Split out a function for finding the panel power sequencer stuff 8168 8169 and add a bios_t convenience member 8170 8171commit 1b3546b0687a1ae8080009f0c58b26074f08d3af 8172Author: Stuart Bennett <sb476@cam.ac.uk> 8173Date: Tue Jan 20 04:11:11 2009 +0000 8174 8175 Move the bios flat-panel pointers back into the bios struct 8176 8177 Last there in 2007, removed in 82fdd3f76e20fc7dd10fcf00cf1c13520dc56364 :-) 8178 8179commit d29c4815b61635ed53ded94361ba52ee61ea5ac7 8180Author: Stuart Bennett <sb476@cam.ac.uk> 8181Date: Tue Jan 20 03:46:05 2009 +0000 8182 8183 Remove a BITism from BMP parsing path, and move old style BMP init into common path 8184 8185commit 3e971f63b6b6859acc19a52a4003bb6041cc5374 8186Author: Stuart Bennett <sb476@cam.ac.uk> 8187Date: Mon Jan 19 23:02:09 2009 +0000 8188 8189 randr12: lock independent crtc base registers separately (#19620) 8190 8191 Thanks to Mjules for tireless testing, surprising nobody else noticed it earlier 8192 8193commit ffbca7272878d440a06a613df95ce5d7be46f288 8194Author: Stuart Bennett <sb476@cam.ac.uk> 8195Date: Sun Jan 18 17:24:47 2009 +0000 8196 8197 randr12: code motion and allow ramdac580 setting on nv40 8198 8199commit 75a037623adecbe460198066b5059c349de427aa 8200Author: Stuart Bennett <sb476@cam.ac.uk> 8201Date: Sun Jan 18 15:41:23 2009 +0000 8202 8203 randr12: better selection of nv4x single pll mode and some notes 8204 8205commit 767fcea420fc21c5b5ec6c9f2db56c1a667db306 8206Author: Stuart Bennett <sb476@cam.ac.uk> 8207Date: Mon Jan 19 01:51:21 2009 +0000 8208 8209 randr12: no mmiotrace indication that the blob ever clears these bits 8210 8211 some hardware drops them on readback, but I don't think that matters 8212 also make defined bits shifts comprehensible 8213 8214commit 985ac728276633d14fa5577eacbc2095582b2587 8215Author: Stuart Bennett <sb476@cam.ac.uk> 8216Date: Sat Jan 17 23:21:19 2009 +0000 8217 8218 randr12: no mmiotrace evidence for nv47 using the special pll bits 8219 8220commit 3ef6b41a1c072d2f3545915180b522e43e0c8754 8221Author: Stuart Bennett <sb476@cam.ac.uk> 8222Date: Sun Jan 18 16:13:57 2009 +0000 8223 8224 twoStagePLL is actually a description of cards using two PLL regs 8225 8226commit c7a423a24ad7da7b510007a0d5a779999bd8cf73 8227Author: Stuart Bennett <sb476@cam.ac.uk> 8228Date: Sat Jan 17 04:34:48 2009 +0000 8229 8230 nv04 blitter: use correct number of dma params for real nv04 8231 8232commit 25de6b867f319099dac05ba84f170da0f0e6c803 8233Author: Stuart Bennett <sb476@cam.ac.uk> 8234Date: Fri Jan 16 20:22:16 2009 +0000 8235 8236 Complain more about weird bios images 8237 8238 also use bios chip version to test for g80 8239 8240commit 4d72981cc7d156708683aba51489d4691106438c 8241Author: Stuart Bennett <sb476@cam.ac.uk> 8242Date: Wed Jan 14 18:15:21 2009 +0000 8243 8244 randr12: avoid classic macro error 8245 8246 Guess there's not many ppc nv11 users, as it's been here since Feb '08 8247 8248commit 133c1a55134853161a25f358432fd05e736718b5 8249Author: Stuart Bennett <sb476@cam.ac.uk> 8250Date: Tue Jan 6 17:47:07 2009 +0000 8251 8252 No need of separate allocation for bios image 8253 8254 nor of manual allocations for short lived I2C devices 8255 8256commit 8a25049abd2b755187f47572f92b11e9072e522a 8257Author: Stuart Bennett <sb476@cam.ac.uk> 8258Date: Tue Jan 6 16:55:42 2009 +0000 8259 8260 Make turning on bios execution logging slightly easier 8261 8262 and add a minor parsing check 8263 8264commit 54d43c4760b573b5d376859281268046b61fb2f4 8265Author: Maarten Maathuis <madman2003@gmail.com> 8266Date: Fri Jan 9 20:02:44 2009 +0100 8267 8268 nv50: small fix 8269 8270 - planemask is rarely used, so it's just a cleanup. 8271 8272commit 5c9f8d3ce8964ad974175f94b858f9efc11756bb 8273Author: Maarten Maathuis <madman2003@gmail.com> 8274Date: Fri Jan 9 17:48:07 2009 +0100 8275 8276 nv50: Add a fallback for non-repeating XRGB pixmaps. 8277 8278 - Other card generations will most likely need this too. 8279 8280commit 0321e6741d9749f9f6b15ac2099139f24e88eef9 8281Author: Ben Skeggs <skeggsb@gmail.com> 8282Date: Thu Jan 8 14:31:39 2009 +1100 8283 8284 nv50: don't segfault if preinit fails before display setup done 8285 8286commit 7dc567d432fc6a3734bf64244e57bcc3612d5ec2 8287Author: Ben Skeggs <skeggsb@gmail.com> 8288Date: Mon Jan 5 10:51:03 2009 +1100 8289 8290 nv50: correct size/layout of uploaded data 8291 8292 This should fix the minor corruption that is present while playing some 8293 videos. 8294 8295commit 159c18cea2ecb92da74f9a9fdff0a9f5ef03df53 8296Author: Maarten Maathuis <madman2003@gmail.com> 8297Date: Sun Dec 28 17:56:55 2008 +0100 8298 8299 Make CHECK_TIMEOUT() a little more correct. 8300 8301 - Decrease the maximum counter value by a factor 16. 8302 - Determine t_start at the start. 8303 8304commit 5d281a2439de1e8c1848b6b700f30476575966e0 8305Author: Stuart Bennett <sb476@cam.ac.uk> 8306Date: Wed Dec 24 23:38:02 2008 +0000 8307 8308 nv04-nv4x: support RandR 1.3 panning 8309 8310commit 3c4705cb4472680b6cdc02fd5152d565d214226f 8311Author: Stuart Bennett <sb476@cam.ac.uk> 8312Date: Mon Dec 15 01:54:12 2008 +0000 8313 8314 Implement bios opcodes 4C and 4D 8315 8316commit 1fd60f202d41d6bb2028955d3bf9e6fc3921e3ff 8317Author: Stuart Bennett <sb476@cam.ac.uk> 8318Date: Mon Dec 15 01:19:26 2008 +0000 8319 8320 Group all DCB I2C info for a given bus in a structure 8321 8322 Furthermore, use the port type field to determine how to drive a bus, rather 8323 than the chip(s) characteristically associated with that port type 8324 8325commit 31c9958779d95ae84993f4772607367927a0e313 8326Author: Maarten Maathuis <madman2003@gmail.com> 8327Date: Fri Dec 19 19:59:53 2008 +0100 8328 8329 nv50: implement get_crtc 8330 8331 - Won't help much right now (pre-KMS), but it doesn't hurt to prototype now. 8332 8333commit f7a757827c1d62ac8a7246fe3b64fea1891a7a16 8334Author: Maarten Maathuis <madman2003@gmail.com> 8335Date: Thu Dec 18 19:57:50 2008 +0100 8336 8337 nv50: implement set_origin (mostly used for panning) 8338 8339commit 74b2e4e92822fb321a9304f0e077c18de57e585f 8340Author: Maarten Maathuis <madman2003@gmail.com> 8341Date: Wed Dec 17 22:42:20 2008 +0100 8342 8343 configure: Avoid compiling drm modesetting code by accident. 8344 8345commit d104f145260ff183b5f7e4c16cebc1680cf771b7 8346Author: Maarten Maathuis <madman2003@gmail.com> 8347Date: Tue Dec 16 23:54:39 2008 +0100 8348 8349 nv50: Improve lut code. 8350 8351commit 20f93a94ae034ee7c744947f42a29a7c0b50ee9d 8352Author: Stuart Bennett <sb476@cam.ac.uk> 8353Date: Wed Nov 26 01:37:38 2008 +0000 8354 8355 randr12: pass cursor visibility changes through mode state structures 8356 8357commit 763e0f88c79105a726ea05219875a25fc0129e1f 8358Author: Stuart Bennett <sb476@cam.ac.uk> 8359Date: Wed Nov 12 20:32:47 2008 +0000 8360 8361 randr12: allow ddc while crtc is locked 8362 8363commit 9d8bd459485392ec21d90c405510f3f097a61921 8364Author: Stuart Bennett <sb476@cam.ac.uk> 8365Date: Sat Nov 22 21:20:07 2008 +0000 8366 8367 Punt all bios structs and defines into a new header 8368 8369commit a34a2c07662b0175e03161cb48abaea9d814c1eb 8370Author: Stuart Bennett <sb476@cam.ac.uk> 8371Date: Thu Nov 20 17:35:59 2008 +0000 8372 8373 Do not attempt to parse FP mode tables for Mac laptops (#18636) 8374 8375 The two varieties of these laptops I've had the misfortune to encounter use DDC, 8376 and the NV34M based one appears to not include a fp mode table at all 8377 8378commit 1aba513a0b5a00996f28462cd4118214dffd87e2 8379Author: Ben Skeggs <skeggsb@gmail.com> 8380Date: Fri Nov 21 13:16:38 2008 +1100 8381 8382 nv50: program physical addresses into CRTCs, not VM addresses. 8383 8384commit 65b956fb407818117ce17ced6bbe66c55b4acb7c 8385Author: Stuart Bennett <sb476@cam.ac.uk> 8386Date: Mon Nov 17 20:02:19 2008 +0000 8387 8388 randr12: fix stupidity from 539f4990 8389 8390 found by AVRS2 8391 8392commit 1666e858dc027bc028083c15618da5cde5a1b6f1 8393Author: Maarten Maathuis <madman2003@gmail.com> 8394Date: Sun Nov 9 23:10:15 2008 +0100 8395 8396 nv50: use bios provided load detect value 8397 8398commit a020a44a2c48ff20bca6188469cb9cbbb5bf6093 8399Author: Maarten Maathuis <madman2003@gmail.com> 8400Date: Sun Nov 9 22:56:30 2008 +0100 8401 8402 bios: support g80 loadval table 8403 8404commit 220cbb6347cb0404bd24f457ac6cb5e39917cdbf 8405Author: Maarten Maathuis <madman2003@gmail.com> 8406Date: Sat Nov 8 18:01:18 2008 +0100 8407 8408 nv50_randr: plug a small memory leak 8409 8410commit cea05e14f94dcb842adbfcec515e17e9f5655089 8411Author: Stuart Bennett <sb476@cam.ac.uk> 8412Date: Sun Oct 26 03:51:36 2008 +0000 8413 8414 randr12: make colour-map setting less complex 8415 8416commit 0e4b01a5e06d8b5323c010ce89723369f17b10b2 8417Author: Stuart Bennett <sb476@cam.ac.uk> 8418Date: Sat Oct 25 02:11:37 2008 +0100 8419 8420 randr12: convenience functions for vga crtc state access 8421 8422commit fe9bfd9b1b5ce6fd0cd0360ec850ae48b4abac41 8423Author: Stuart Bennett <sb476@cam.ac.uk> 8424Date: Fri Oct 24 21:02:00 2008 +0100 8425 8426 Tidy EnterVT, and don't call it from ScreenInit 8427 8428commit 3e1ccff8744d9cabef0cd3901e987deeeb38ec12 8429Author: Stuart Bennett <sb476@cam.ac.uk> 8430Date: Fri Oct 24 02:49:19 2008 +0100 8431 8432 Condense crtc locking 8433 8434 We unlock all crtcs whenever the driver is active, and relock them when 8435 returning to VT (which is what we were doing, but in a more complicated manner) 8436 8437commit 83b541e0a9c454059d0e55a932a2ecdbbf650d60 8438Author: Stuart Bennett <sb476@cam.ac.uk> 8439Date: Mon Oct 27 19:20:22 2008 +0000 8440 8441 Common function for the bios condition table 8442 8443commit 36c6aa449063387b34c31015d618e53d4b6634e0 8444Author: Stuart Bennett <sb476@cam.ac.uk> 8445Date: Mon Oct 27 18:55:28 2008 +0000 8446 8447 Allow more time in INIT_CONDITION_TIME (#18234) 8448 8449commit bf810b32bc959f3821bca18cf158f711dd314a26 8450Author: Stuart Bennett <sb476@cam.ac.uk> 8451Date: Mon Oct 27 18:41:43 2008 +0000 8452 8453 More warnings for C51's magical unaligned regs 8454 8455commit d481b5e2e1dcb7ac5b5a7583bc6542b42567eac2 8456Author: Stuart Bennett <sb476@cam.ac.uk> 8457Date: Sat Oct 25 17:48:35 2008 +0100 8458 8459 Rework bios parser error handling 8460 8461commit 387f30821ff7add22376de33a203744cdc719849 8462Author: Stuart Bennett <sb476@cam.ac.uk> 8463Date: Sat Oct 25 00:38:29 2008 +0100 8464 8465 RivaTuner dumps call the `how much memory have I got?' reg NV_PFB_CSTATUS 8466 8467commit bb20175001748e5d2fab1b9cb2075bbe5d94e5ed 8468Author: Stuart Bennett <sb476@cam.ac.uk> 8469Date: Fri Oct 24 17:42:53 2008 +0100 8470 8471 pedantry fix for nv11 locking, and some minor cleanups 8472 8473commit 9aecc4226ca9067667004150f7d885b2247273b2 8474Author: Stuart Bennett <sb476@cam.ac.uk> 8475Date: Thu Oct 23 23:17:15 2008 +0100 8476 8477 Add a function to lock/unlock all crtcs, use it in nv_bios code 8478 8479 Also split initial head owner determination into separate function 8480 8481commit 48581da4d0e1663456ddcbf7a456029cf961d4cd 8482Author: Stuart Bennett <sb476@cam.ac.uk> 8483Date: Thu Oct 23 02:35:30 2008 +0100 8484 8485 Attempt to make the usage of cr44 rational instead of cargoculted 8486 8487 and a couple of minor cleanups in nv_crtc 8488 8489commit 1c7bee95610a47175064b17afd6aa2a45e311b1a 8490Author: Stuart Bennett <sb476@cam.ac.uk> 8491Date: Thu Oct 23 02:46:03 2008 +0100 8492 8493 Robust nv11 head setting (from mmiotrace) 8494 8495commit cf5162b5ed99666d3dc8db8328115afd6fff6a8e 8496Author: Stuart Bennett <sb476@cam.ac.uk> 8497Date: Tue Oct 21 02:12:24 2008 +0100 8498 8499 randr12: update scrn field on server regen, and emit an error if rotated 8500 8501 (updating the scrn is easier than removing and re-adding outputs) 8502 8503commit 13739cf36893e8ab6d3d4911c76a45f0b8bd5f4c 8504Author: Stuart Bennett <sb476@cam.ac.uk> 8505Date: Mon Oct 20 21:47:07 2008 +0100 8506 8507 randr12: don't change virtualX/virtualY 8508 8509 all the things in the ddx that use virtual* (setting displayWidth, allocating 8510 the FB, EXA init etc) use the values present before this function is called 8511 8512 these prior values are derived from any Virtual line in xorg.conf and are what 8513 we want to use 8514 8515 setting virtual* to the crtc size here, rather than the desired Virtual size, 8516 breaks everything using them in ScreenInit on server regeneration 8517 8518 as the rhd guys say: 8519 /* This is strange... if we set virtualX/virtualY like the intel driver 8520 * does, we limit ourself in the future to this maximum size. 8521 * The check for this is internally in RandR, no idea why the intel driver 8522 * actually works this way... 8523 * Even more curious: if we DON'T update virtual, everything seems to 8524 * work as expected... */ 8525 so I'll take that as precedent. 8526 8527commit 98a4dcb112221fa66d82b87ef9229f5c443daf90 8528Author: Stuart Bennett <sb476@cam.ac.uk> 8529Date: Mon Oct 20 18:59:35 2008 +0100 8530 8531 Delete notifiers and grobjs in CloseScreen to avoid channel member becoming stale on server regeneration 8532 8533commit 12fce00b2ecf4c6be6cc86c8a7374878bb7dfc20 8534Author: Stuart Bennett <sb476@cam.ac.uk> 8535Date: Wed Oct 22 02:47:45 2008 +0100 8536 8537 Remove all object with mmaps in CloseScreen, so that drmClose actually calls the drm release method 8538 8539 (and so subsequently, we can be DRM master on the next drmOpen) 8540 8541commit 8e3f27a30078675f1a7974ba436a68a5b45e0924 8542Author: Stuart Bennett <sb476@cam.ac.uk> 8543Date: Wed Oct 22 03:32:51 2008 +0100 8544 8545 No reason to call AccelCommonInit twice in ScreenInit, but calling it in EnterVT helps resume... 8546 8547commit f495fa9153ee57d3ee0dc735c3e9e836d84a2ccc 8548Author: Stuart Bennett <sb476@cam.ac.uk> 8549Date: Wed Oct 22 02:41:17 2008 +0100 8550 8551 nouveau_channel_free should remove mmaps made in nouveau_channel_alloc 8552 8553commit 1cedb8e49c832890efd12904e78dfd8ab5855691 8554Author: Stuart Bennett <sb476@cam.ac.uk> 8555Date: Wed Oct 22 02:15:06 2008 +0100 8556 8557 Do frag prog allocation and shader upload in TCL init 8558 8559 Similar in spirit to the nv50 inplementation. 8560 Removes many dependencies on state stored in static variables, which in turn 8561 makes reinitializing shaders on resume / server regeneration tractable. 8562 8563commit 9e2c08937b0d1bfed71997b2b16bded91830af71 8564Author: Stuart Bennett <sb476@cam.ac.uk> 8565Date: Wed Oct 22 01:44:10 2008 +0100 8566 8567 Only hackup shaders once 8568 8569commit afc6668d2b4a1971c211c421b6a415fe74dbd2ea 8570Author: Stuart Bennett <sb476@cam.ac.uk> 8571Date: Sat Oct 18 02:23:18 2008 +0100 8572 8573 cr26 is just another view of 0x3c0 8574 8575 remove superfluous crtc unlock too 8576 8577commit 539f4990c8f1a32da57992dc863f277f8ede8194 8578Author: Stuart Bennett <sb476@cam.ac.uk> 8579Date: Wed Oct 15 03:59:06 2008 +0100 8580 8581 Use symbolic define values where known 8582 8583commit 791666a76c68880136ae26d0477bf484cc667f72 8584Author: Stuart Bennett <sb476@cam.ac.uk> 8585Date: Wed Oct 15 00:25:26 2008 +0100 8586 8587 Convert CIO, DIO and VIO use to use defines from nvreg for index and data reg offsets 8588 8589commit a1b7f8d9b20bff59956ea1dc2130e5590a60aa4c 8590Author: Stuart Bennett <sb476@cam.ac.uk> 8591Date: Wed Oct 15 00:12:11 2008 +0100 8592 8593 Rename relevant functions, sizes and offsets to PRM.IO from P.IO, in keeping with the nvidia scheme 8594 8595commit 267c0eef02f5e2a48217235d170a0530de6a94d3 8596Author: Stuart Bennett <sb476@cam.ac.uk> 8597Date: Tue Oct 14 23:58:11 2008 +0100 8598 8599 Name crtc index regs according to rules.xml 8600 8601commit 9261c34e873356b02a0fda00e9911d9795d9f1f2 8602Author: Stuart Bennett <sb476@cam.ac.uk> 8603Date: Tue Oct 21 17:46:37 2008 +0100 8604 8605 SaveGeneration isn't doing a lot 8606 8607 add another close screen thing too 8608 8609commit c29c190036c721dac2e58e66d0c5a70a24377a1d 8610Author: Stuart Bennett <sb476@cam.ac.uk> 8611Date: Tue Oct 21 16:27:00 2008 +0100 8612 8613 Replace a few memsets with initializers 8614 8615commit 3b53f6ceb86458f9df9b7becc18a79e74b21c91f 8616Author: Stuart Bennett <sb476@cam.ac.uk> 8617Date: Sun Oct 12 15:52:43 2008 +0100 8618 8619 Eliminate separate NVRec ctor and dtor 8620 8621commit 1b18db28f124bd5021f61b72389e5c0ce003b212 8622Author: Stuart Bennett <sb476@cam.ac.uk> 8623Date: Sat Oct 11 21:05:57 2008 +0100 8624 8625 randr12: remove pointless debug in nv_output and nv_crtc 8626 8627commit 240d51dc928cea38f6b9d842b59011ad7c5500c0 8628Author: Stuart Bennett <sb476@cam.ac.uk> 8629Date: Sat Oct 11 18:40:35 2008 +0100 8630 8631 randr12: deBoolification and eliminate `override' arg to nv_crtc_load_state_ext 8632 8633commit b95c3e23f0eac7d12d5ee271af8ec5707ac4ef8e 8634Author: Stuart Bennett <sb476@cam.ac.uk> 8635Date: Sat Oct 11 18:29:40 2008 +0100 8636 8637 Make all CR 57/58 access use the proper functions and defines 8638 8639 Use a define for usleep in bios too 8640 8641commit 9988ae594f6f03385b261b0af83b324981f11593 8642Author: Stuart Bennett <sb476@cam.ac.uk> 8643Date: Sat Oct 11 17:56:23 2008 +0100 8644 8645 Redo bios logging 8646 8647commit 0e6a9cc61c88deaf0e0935d5a7d269621c3d6cd0 8648Author: Stuart Bennett <sb476@cam.ac.uk> 8649Date: Sat Oct 11 18:04:58 2008 +0100 8650 8651 Improve some nv_bios messages, remove others 8652 8653commit b175bfcd6485d9dcddb66ab5f79104567d256336 8654Author: Stuart Bennett <sb476@cam.ac.uk> 8655Date: Sat Oct 11 17:59:02 2008 +0100 8656 8657 Bios register list is easier to read when sorted numerically 8658 8659 and remove NULL initialization for calloc'd struct 8660 8661commit e51b49c938483ccbd172ff07d0385a6fb474b57e 8662Author: Stuart Bennett <sb476@cam.ac.uk> 8663Date: Sat Oct 11 02:35:40 2008 +0100 8664 8665 randr12: panels with edid have only a constant mode 8666 8667 prove me wrong. 8668 8669commit 765494ebf9b129bd613121cb714e928b09e19500 8670Author: Stuart Bennett <sb476@cam.ac.uk> 8671Date: Sat Oct 11 02:16:50 2008 +0100 8672 8673 Remove unnecessary "Setting owner" message 8674 8675commit bf585ad38a80f7828b821936296316bcee4658f5 8676Author: Maarten Maathuis <madman2003@gmail.com> 8677Date: Sat Oct 11 01:22:35 2008 +0200 8678 8679 The fifo channel should be removed along with closing of the fd, so only clear the client memory. 8680 8681commit 8b7e42461add708e75d3f7d675245f5bbf3652e8 8682Author: Maarten Maathuis <madman2003@gmail.com> 8683Date: Fri Oct 10 23:37:55 2008 +0200 8684 8685 git-log is no longer valid for git 1.6, switch to git log. 8686 8687commit 933fd803f0aadc934f448937dbeb305256fe8893 8688Author: Maarten Maathuis <madman2003@gmail.com> 8689Date: Fri Oct 10 23:14:57 2008 +0200 8690 8691 Some fixes that bring me closer to surviving to the 2nd X server generation. 8692 8693 - Not quite done, i get a fifo dump when something tries to render. 8694 - Something is keeping an fd, which is preventing a full takedown, but this is not a valid reason for failing (kernel modesetting 8695 for example will never have full takedown). 8696 8697commit daee3c38a02f48f9094f712800174b88b065f0ff 8698Author: Stuart Bennett <sb476@cam.ac.uk> 8699Date: Thu Oct 9 17:09:58 2008 +0100 8700 8701 randr12: separate cursor pixmap location setting function 8702 8703commit caf2e920f18cad1c57e945a4e474e63a65ec6a43 8704Author: Stuart Bennett <sb476@cam.ac.uk> 8705Date: Thu Oct 9 17:07:30 2008 +0100 8706 8707 Move cursor show/hide funcs to nv_hw, document nv40 bug, and set curctl2 before applying nv40 fix 8708 8709commit f76074c7041b1246efdf2c0a4b0c467e2fe0bef1 8710Author: Stuart Bennett <sb476@cam.ac.uk> 8711Date: Thu Oct 9 17:24:23 2008 +0100 8712 8713 Use correct refclk for nv_get_clock, and use bool type as appropriate in arbitration 8714 8715commit bd5b8074b4a250610cf8c9e2f181c2946e902bf8 8716Author: Stuart Bennett <sb476@cam.ac.uk> 8717Date: Thu Oct 9 15:25:16 2008 +0100 8718 8719 Tidy arbitration prototypes and split randr12 cursor pixmap setting from arbitration 8720 8721 Also fix ages old warning in nv_hw 8722 8723commit 5dc41fdcf4ba76d7154d32f030d2e99756681e4a 8724Author: Stuart Bennett <sb476@cam.ac.uk> 8725Date: Fri Oct 10 01:11:46 2008 +0100 8726 8727 Merge nv4, nv10 and nForce UpdateArbitrationSettings functions 8728 8729commit 0ffd012c922d241aec22c8e44416171e39e6bf0c 8730Author: Stuart Bennett <sb476@cam.ac.uk> 8731Date: Fri Oct 10 00:57:45 2008 +0100 8732 8733 Add PCI slot reading defines, and use them for nForce code 8734 8735 Also move nForce dimm checking code to nv_setup, so it is only tested once 8736 8737commit 520a76ef3df3cdc735691cfda329ceaae0ec3123 8738Author: Stuart Bennett <sb476@cam.ac.uk> 8739Date: Thu Oct 9 20:22:34 2008 +0100 8740 8741 Some tidying of arbitration functions 8742 8743commit 3ceb2b60a09379ef51635f8309fdc82060681e8f 8744Author: Stuart Bennett <sb476@cam.ac.uk> 8745Date: Thu Oct 9 19:05:01 2008 +0100 8746 8747 Reindent arbitration stuff 8748 8749commit d7439fc50a924f6f1a6b5b59ba3ec076cf90c52e 8750Author: Stuart Bennett <sb476@cam.ac.uk> 8751Date: Wed Oct 8 23:45:13 2008 +0100 8752 8753 Wrap usleep when called in bios code 8754 8755commit 62cf9f371c420abe7a74ac4bb9fa2da081f1d29b 8756Author: Stuart Bennett <sb476@cam.ac.uk> 8757Date: Fri Oct 10 16:41:29 2008 +0100 8758 8759 randr12: don't use in use (by another vga output) crtc for load detect (avoids flicker) 8760 8761commit f1d0fd0595b40d3a7f555bada88f1f79b517d610 8762Author: Stuart Bennett <sb476@cam.ac.uk> 8763Date: Tue Oct 7 21:39:10 2008 +0100 8764 8765 Better behaviour on allocator fail 8766 8767 also fix randr12 code which doesn't know how to call calloc 8768 and be sure to use the number of actually inited crtcs 8769 8770commit 2b67015e5911cb7e321da72d7d3d5f0fefedd8e0 8771Author: Stuart Bennett <sb476@cam.ac.uk> 8772Date: Tue Oct 7 02:35:44 2008 +0100 8773 8774 randr12: reorder nv_output functions 8775 8776 roughly into order of use 8777 8778commit 9f4eef0165766be2e1aa5516eae409e94c09553b 8779Author: Stuart Bennett <sb476@cam.ac.uk> 8780Date: Thu Sep 18 18:32:36 2008 +0100 8781 8782 randr12: distinguish detected encoder from active encoder 8783 8784 don't swap output props until a mode is set on new encoder either 8785 8786commit 549cda114ee378640dda89621b0ce9c937c473da 8787Author: Stuart Bennett <sb476@cam.ac.uk> 8788Date: Wed Oct 8 01:22:17 2008 +0100 8789 8790 Reduce includes, everything necessary is in nv_include.h 8791 8792commit 30eaa0f08250f72b0171a858f2694ca8a62c3ebd 8793Author: Stuart Bennett <sb476@cam.ac.uk> 8794Date: Sun Sep 21 00:52:56 2008 +0100 8795 8796 Oops. 8797 8798commit 2481e5646ab3ea5e18d3f5ee4042fa20a8dad3de 8799Author: Stuart Bennett <sb476@cam.ac.uk> 8800Date: Sun Sep 21 00:44:50 2008 +0100 8801 8802 Optimizing memory usage for unaccelerated case is not interesting 8803 8804commit 19f0ea63a505e6a840cf07bc554fd8f1787616b4 8805Author: Stuart Bennett <sb476@cam.ac.uk> 8806Date: Sun Sep 21 00:43:57 2008 +0100 8807 8808 randr12: rename mon -> edid 8809 8810commit 0cb081758f52e73c3055758ed7aced12b93c2519 8811Author: Stuart Bennett <sb476@cam.ac.uk> 8812Date: Sun Sep 21 00:31:40 2008 +0100 8813 8814 randr12: crtc destroy 8815 8816 also alleviate world shortage of curly braces 8817 8818commit 37adca1dd68d5791b0371eed4b96f6613fc747c2 8819Author: Stuart Bennett <sb476@cam.ac.uk> 8820Date: Sat Sep 20 23:57:06 2008 +0100 8821 8822 Need to increment dcb entry index when fabricating additional entries 8823 8824commit e6b6d8c9189f3b5ebbc77f55e318246be0fa1bd4 8825Author: Maarten Maathuis <madman2003@gmail.com> 8826Date: Fri Sep 19 21:23:04 2008 +0200 8827 8828 Also remove man page entry. 8829 8830commit cb1e9c1ab288bcf380789f775533e4c8ff8705db 8831Author: Maarten Maathuis <madman2003@gmail.com> 8832Date: Fri Sep 19 21:14:09 2008 +0200 8833 8834 NV50: remove {Prepare,Finish}Access since i realise now it's not so useful 8835 8836 - despite that the xserver code suggests that pixmaps can be offscreen on this call, that is not done as far as i can tell 8837 - not to mention that there is no way to know for sure if a pixmap is offscreen (exaPixmapIsOffscreen is not exported), making the 8838 offset useless 8839 - this will (obviously) become relevant as soon as driver managed pixmaps come into play, but that is for another day 8840 8841commit 3e397f575df6c3fee6eb9ee810fb3fc70d2c0cc8 8842Author: Maarten Maathuis <madman2003@gmail.com> 8843Date: Fri Sep 19 20:41:59 2008 +0200 8844 8845 NV50: The TMDS dual link threshold is meaningless for LVDS 8846 8847commit 599d258948be0505f02947251ad6877e7168d9a8 8848Author: Stuart Bennett <sb476@cam.ac.uk> 8849Date: Mon Aug 25 01:47:45 2008 +0100 8850 8851 randr12: multiple encoders per connector (DVI-I) 8852 8853 that this doesn't break things in subtle and unexpected ways is too much to 8854 hope for. one not-subtle and expected way things will break is that xorg.confs 8855 with specific setups for DVI-A-n and DVI-D-n connections will need renaming (to 8856 DVI-I-n). 8857 8858 as the last patch in the series, this also fixes an issue introduced with 8859 144b04a2 (randr12: fix dpms, detect, destroy, ...), that for a screen connected 8860 to a dvi connector, both the analogue and digital variants were detected as 8861 connected 8862 8863commit 5e8ac848936a1b70babc417454599c0231343399 8864Author: Ben Skeggs <skeggsb@gmail.com> 8865Date: Wed Sep 17 22:12:07 2008 +1000 8866 8867 nv50: 0xa0 family uses class 0x8397 for 3D 8868 8869commit 9c36eefe991daf47d20badffb3009def99473bdd 8870Author: Stuart Bennett <sb476@cam.ac.uk> 8871Date: Sun Sep 14 23:01:49 2008 +0100 8872 8873 randr12: avoid weirdness when tv-out happens to share i2c_index with another output 8874 8875commit 144b04a258eaa8e3b63c48b5e574524b40a59ac7 8876Author: Stuart Bennett <sb476@cam.ac.uk> 8877Date: Sun Aug 24 04:15:20 2008 +0100 8878 8879 randr12: fix dpms, detect, destroy, save and restore for multiple encoders per connector 8880 8881commit 2b30b8914a32049e87205cf896b2e89ab6649bca 8882Author: Stuart Bennett <sb476@cam.ac.uk> 8883Date: Sun Aug 24 04:26:19 2008 +0100 8884 8885 randr12: re-do detection of dvi-a vs vga 8886 8887commit 50bac3fcd24072b8926ba2dbebd571424fbb7d6d 8888Author: Stuart Bennett <sb476@cam.ac.uk> 8889Date: Sun Aug 24 03:56:16 2008 +0100 8890 8891 randr12: unify output funcs for analogue and tmds 8892 8893commit 69e30e6727895a8edeffc9a348b13a353cd0c6d0 8894Author: Stuart Bennett <sb476@cam.ac.uk> 8895Date: Sun Aug 24 03:12:58 2008 +0100 8896 8897 randr12: unified entry path for dpms 8898 8899commit a794a0c864e5b651bea5e40a4e10978aee9da5b6 8900Author: Stuart Bennett <sb476@cam.ac.uk> 8901Date: Sun Aug 24 04:45:48 2008 +0100 8902 8903 randr12: nouveau_output -> nouveau_connector 8904 8905commit 0a5119bd8a927da1f3f4ed6caa2e93acdae29c04 8906Author: Stuart Bennett <sb476@cam.ac.uk> 8907Date: Sun Aug 24 02:45:41 2008 +0100 8908 8909 randr12: separate encoder struct 8910 8911commit 352b8fd98c519f85ade02aa679d342bc64faa508 8912Author: Stuart Bennett <sb476@cam.ac.uk> 8913Date: Sun Sep 14 16:53:12 2008 +0100 8914 8915 Another DCB 1.4/1.5 TV (#17471) 8916 8917 I think we've got enough for TVs to no longer be fatal 8918 8919commit e8ba4e0baa62f02ba85258847963512791d89385 8920Author: Stuart Bennett <sb476@cam.ac.uk> 8921Date: Sun Sep 14 16:44:32 2008 +0100 8922 8923 Only use OF bios image on PPC 8924 8925commit 54b8a7ea43ec20c798dc93d881cdffa5cb1125a0 8926Author: Ben Skeggs <skeggsb@gmail.com> 8927Date: Thu Sep 11 03:29:46 2008 +1000 8928 8929 nv50/exa: work around corruption issues (see detailed commit msg) 8930 8931 It's not possible for us to do this properly without using the exa 8932 pixmap hooks, they require a real memory manager, and is being 8933 worked on in the ng branches, which doesn't currently support nv5x 8934 at all. 8935 8936 For a better solution until then: 8937 http://nouveau.freedesktop.org/wiki/NV50Support 8938 8939commit f8c0af3886142ea59936a77006ead0cd2c4ff795 8940Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 8941Date: Fri Sep 5 23:08:14 2008 +0200 8942 8943 Partly revert ea152819f45b6cf92f6742ed3f9f639cdbbd53f0, alignment is needed. 8944 8945commit ea152819f45b6cf92f6742ed3f9f639cdbbd53f0 8946Author: Maarten Maathuis <madman2003@gmail.com> 8947Date: Fri Sep 5 22:51:27 2008 +0200 8948 8949 exa: align offscreenBase to something sensible 8950 - Aligning virtualY makes no sense whatsoever. 8951 - displayWidth is the "pitch-width" of the framebuffer. 8952 8953commit 95c19b98b38310c543758fe4288a9e47b970fe67 8954Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 8955Date: Thu Sep 4 07:51:03 2008 +0200 8956 8957 Add 8200 detection. 8958 8959commit 6dd8ad4ed3f89c6ed9e61708a587d417c8d984fe 8960Author: Maarten Maathuis <madman2003@gmail.com> 8961Date: Sat Aug 30 17:51:05 2008 +0200 8962 8963 nv50: solid fill shouldn't use a pattern rop + minor changes 8964 8965commit b2b726d59fd8af819a7e464824a49f6e6fc10b84 8966Author: Stuart Bennett <sb476@cam.ac.uk> 8967Date: Sun Aug 24 00:18:40 2008 +0100 8968 8969 randr12: merge output detection functions 8970 8971commit a223e581b06ffb308aecc2365795951e2c021f29 8972Author: Stuart Bennett <sb476@cam.ac.uk> 8973Date: Sun Aug 24 00:08:41 2008 +0100 8974 8975 randr12: remember the edid from detection for getting modes later 8976 8977 also split native mode finding into separate function 8978 8979commit 16d9e89d9cfae1c21999d880c557ae283e14178e 8980Author: Stuart Bennett <sb476@cam.ac.uk> 8981Date: Sun Aug 24 19:56:24 2008 +0100 8982 8983 randr12: detypedef private structs and use a define to get to the privates 8984 8985commit 3b3f3cdbe0f69c57153e070f248ff001cd5a4a17 8986Author: Stuart Bennett <sb476@cam.ac.uk> 8987Date: Sun Aug 24 00:31:35 2008 +0100 8988 8989 Remove unused code 8990 8991commit 1e2b87c459a3a0eacd630f1de8b567c36cc792bd 8992Author: Stuart Bennett <sb476@cam.ac.uk> 8993Date: Mon Aug 25 20:23:05 2008 +0100 8994 8995 randr12: make the nouveau experience less green 8996 8997commit f89af0e6bc5ac88d0755f9402b772ca823518298 8998Author: Maarten Maathuis <madman2003@gmail.com> 8999Date: Sat Aug 23 03:44:28 2008 +0200 9000 9001 nv50: micro cleanup 9002 9003commit 94bf1060cc9d2f4c0a14bc036f8b76c4d9bfc7d7 9004Author: Maarten Maathuis <madman2003@gmail.com> 9005Date: Sat Aug 23 03:39:39 2008 +0200 9006 9007 nv50: do ROPs properly this time 9008 - Table should work for pre-nv50 as well, i just don't have a card plugged in to be sure. 9009 9010commit 6bd14e4ffe1d5fd9420c4ebb46c9c0844d0547f9 9011Author: Maarten Maathuis <madman2003@gmail.com> 9012Date: Fri Aug 22 17:56:31 2008 +0200 9013 9014 nv10: composite is an async operation by default, no need to call exaMarkSync(). 9015 9016commit 9f11d13e68bbf72be814b37fa3d03853eec981b9 9017Author: Maarten Maathuis <madman2003@gmail.com> 9018Date: Fri Aug 22 17:53:11 2008 +0200 9019 9020 exa: FIRE_RING at the end of composite 9021 9022commit 137d099294617adead2cf3d5cc3372d18579eaef 9023Author: Maarten Maathuis <madman2003@gmail.com> 9024Date: Thu Aug 21 23:17:06 2008 +0200 9025 9026 nv50: move VERTEX_{START,END} to {Prepare,Done}Composite 9027 - If PrepareComposite returns TRUE, then everything upto DoneComposite is required. 9028 9029commit 34dc05fbc4835ec68de5a1f70cc094ff6efcf310 9030Author: Maarten Maathuis <madman2003@gmail.com> 9031Date: Thu Aug 21 22:50:49 2008 +0200 9032 9033 nv50: minor tweaks to composite 9034 - Make the blend stuff more consistent with other composite implementations. 9035 - Verified with "rendercheck -f a8r8g8b8,x8r8g8b8,r5g6b5,a8 -t {blend,composite,cacomposite}" for all accelerated ops. 9036 9037commit eca997750b1cc2263f804afba242f16b6d960e02 9038Author: Maarten Maathuis <madman2003@gmail.com> 9039Date: Thu Aug 21 20:01:49 2008 +0200 9040 9041 nv50: handle ROPs better 9042 - Handle ROP's more like pre-NV50 code. 9043 - Special case for 32bpp which seems to work. 9044 - This does NOT fix all rendering issues, just an improvement. 9045 9046commit e468df855abd8d8e4fecd2445c82bb8f824297fb 9047Author: Maarten Maathuis <madman2003@gmail.com> 9048Date: Wed Aug 20 19:14:54 2008 +0200 9049 9050 nv50: clip SIFC and add {Prepare,Finish}Access hooks. 9051 9052 - when w*cpp is not aligned to a dword, then it's possible to end up with crap at the end of a line, so clip it. 9053 - {Prepare,Finish}Access hooks force tiled memory to be migrated upon cpu access. 9054 - Added an option to disable these hooks in case you hit a xserver bug that existed up until very recently. 9055 9056commit 0c0c1ab2ce786def653e8adb95c7c5a8ff9c7eca 9057Author: Stuart Bennett <sb476@cam.ac.uk> 9058Date: Tue Aug 19 00:02:00 2008 +0100 9059 9060 randr12: don't cache pll values (fixes newrestore removal regression) 9061 9062 for 5ms, it's really not worth it 9063 9064commit 544fd3c502287754f2161c21514203ccaab8aed9 9065Author: Stuart Bennett <sb476@cam.ac.uk> 9066Date: Mon Aug 18 21:50:37 2008 +0100 9067 9068 Remove some of the inaccuracies in the manpage 9069 9070commit e3d9cb926e56d28b7cadd0026ae76c076faced75 9071Author: Stuart Bennett <sb476@cam.ac.uk> 9072Date: Mon Aug 18 21:40:36 2008 +0100 9073 9074 NV_ARCH_04 can have rotated output, it's just not accelerated 9075 9076commit 40e920fafb9f22da23ef11c941561bb14966dc47 9077Author: Stuart Bennett <sb476@cam.ac.uk> 9078Date: Mon Aug 18 21:39:18 2008 +0100 9079 9080 Delete ShadowFB rotation 9081 9082 And nothing of value was lost. 9083 9084commit 6352d7db24451b014b5f6b873881d103fc1eb316 9085Author: Stuart Bennett <sb476@cam.ac.uk> 9086Date: Mon Aug 18 00:59:32 2008 +0100 9087 9088 Bye bye 8 bit depth 9089 9090commit 33d52e476b64cd91dc4b01c6826892f2da9499b0 9091Author: Stuart Bennett <sb476@cam.ac.uk> 9092Date: Sun Aug 17 23:52:28 2008 +0100 9093 9094 CrtcNumber option is long gone 9095 9096commit a9393a87d36cc486114842275bb78e7e75de754e 9097Author: Stuart Bennett <sb476@cam.ac.uk> 9098Date: Sun Aug 17 23:17:01 2008 +0100 9099 9100 Micro-tidyups to nv_driver and nouveau_xv 9101 9102commit 8ed30d05afe4a5bb077c088c48c28741414c09d3 9103Author: Stuart Bennett <sb476@cam.ac.uk> 9104Date: Sun Aug 17 16:57:17 2008 +0100 9105 9106 randr12: a few line length improvements 9107 9108commit 666ab70754035c569eaa2550ca77c92fe3eaaff3 9109Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 9110Date: Sun Aug 17 22:44:11 2008 +0200 9111 9112 Kill some 8 bit code. 9113 9114commit bee8450931395b23040a4cb1ddcb09f8a476cb93 9115Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 9116Date: Sun Aug 17 22:25:43 2008 +0200 9117 9118 Fix IFC for 16bpp. Also fix the color accuracy of 16bpp solid fills. 9119 9120commit 609b10ff00c9e8d51fd7c9a5fb62120516780aed 9121Author: Stuart Bennett <sb476@cam.ac.uk> 9122Date: Sun Aug 17 03:51:39 2008 +0100 9123 9124 0x48 exception does not apply on BIT cards 9125 9126commit a9e2d47e2cee7db447063d7c33f3e42a8d865c92 9127Author: Stuart Bennett <sb476@cam.ac.uk> 9128Date: Sun Aug 17 02:48:33 2008 +0100 9129 9130 randr12: some code simplification, rearrangement and tidying 9131 9132 We don't claim to support a bpp of 15 or 24, so some old calculations can be simplified 9133 9134commit 5715f95b37ad182377024f468982608b48eb3409 9135Author: Stuart Bennett <sb476@cam.ac.uk> 9136Date: Sun Aug 17 02:27:42 2008 +0100 9137 9138 NewRestore: remove, due to lack of interest 9139 9140 nobody wants to do text modes manually. for any future masochists, it's here in the history 9141 9142commit d2e924e847273c5b4d342d0ad611de7d106df429 9143Author: Stuart Bennett <sb476@cam.ac.uk> 9144Date: Sun Aug 17 01:37:41 2008 +0100 9145 9146 randr12: remove another field member 9147 9148commit 55f490c8c29b103d7d8853e2126a8366304f4125 9149Author: Stuart Bennett <sb476@cam.ac.uk> 9150Date: Sun Aug 17 01:26:20 2008 +0100 9151 9152 randr12: fpWidth and fpHeight no more 9153 9154commit 4cbbd44db614d250fdaaf648db2d70ec9bc2f6d9 9155Author: Stuart Bennett <sb476@cam.ac.uk> 9156Date: Sun Aug 17 00:38:23 2008 +0100 9157 9158 randr12: improve aspect scaling code 9159 9160commit 8a364682fd67c10c153cb0778c914a4754e82ddd 9161Author: Stuart Bennett <sb476@cam.ac.uk> 9162Date: Sat Aug 16 23:15:29 2008 +0100 9163 9164 randr12: for native scaling, a clock check isn't enough 9165 9166 1280x1024 and 1280x960 can have the same clock 9167 9168commit 8793bb1d7b79632678d7ce9f1e94c79dfd55fa3c 9169Author: Stuart Bennett <sb476@cam.ac.uk> 9170Date: Fri Aug 15 23:02:48 2008 +0100 9171 9172 Apply minimum front porch only when necessary (#15949) 9173 9174 Also, read minimum front porch from bios table 9175 9176commit 83bb26fd9a8bb283c406bc34d1d7ad0b53d7098a 9177Author: Maarten Maathuis <madman2003@gmail.com> 9178Date: Sat Aug 16 17:18:21 2008 +0200 9179 9180 NV50: disallow doublescan modes on TMDS/LVDS 9181 9182commit 7d5b345f69bf63d36207a7f6ac4d01001d62e79c 9183Author: Maarten Maathuis <madman2003@gmail.com> 9184Date: Sat Aug 16 15:33:13 2008 +0200 9185 9186 NV50: extra warning message 9187 9188commit 743c696c8b0e420a67ce1552ac0cec49c0fdc0f2 9189Author: Maarten Maathuis <madman2003@gmail.com> 9190Date: Sat Aug 16 15:28:43 2008 +0200 9191 9192 NV50: a half decent attempt at doing something when LVDS has no DDC 9193 9194commit fcbaa10a945e04d77c6ea8f83b738fb6c9e47dd5 9195Author: Maarten Maathuis <madman2003@gmail.com> 9196Date: Sat Aug 16 14:35:05 2008 +0200 9197 9198 Forgot something. 9199 9200commit 628acddedc7755a16471e9beee3d95d3b8f11150 9201Author: Maarten Maathuis <madman2003@gmail.com> 9202Date: Sat Aug 16 14:24:18 2008 +0200 9203 9204 NV50: support extended i2c ports (4 and 5) 9205 9206commit c845f5ac0a54b17a9911155030234c229e80626c 9207Author: Maarten Maathuis <madman2003@gmail.com> 9208Date: Sat Aug 16 14:02:51 2008 +0200 9209 9210 nv50: fix some of the rendering bugs 9211 9212 - It seems that 32bpp solid fill only works well for for the non-rop case. 9213 - The SRCCOPY_PREMULT comes from nv and seems to work well, in combination with the above fallback. 9214 - This does not affect the strange stripes with font rendering issue. 9215 9216commit 18288208bdb7f872f0a2e227d20b77c3397747c2 9217Author: Stuart Bennett <sb476@cam.ac.uk> 9218Date: Fri Aug 15 01:40:01 2008 +0100 9219 9220 Redo parsing for panels with EDID, and fix nv3x in the process (#17138) 9221 9222 breakage a possibility :) 9223 9224commit c9d4e86022bffd00d89076a921b734645145f7db 9225Author: Stuart Bennett <sb476@cam.ac.uk> 9226Date: Tue Aug 12 14:15:03 2008 +0100 9227 9228 Try harder to load a good vbios image 9229 9230commit ae594786ffc1cdca5bdd8c8638225819f851061b 9231Author: Stuart Bennett <sb476@cam.ac.uk> 9232Date: Wed Aug 6 00:31:39 2008 +0100 9233 9234 Deal with mobile cards that scribble over the fp strap at POST 9235 9236 It's probably safe to assume a mobile card will have been POSTed at boot, so 9237 this should be fine. 9238 Hopefully there are no cards which have bit 2 set at 0x48 and now break... 9239 9240commit 776670613f79cce3891c65d2f688979767cd1e6d 9241Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 9242Date: Fri Aug 15 00:43:02 2008 +0200 9243 9244 Fix a 16bpp issue, also remove old code that mostly worked around core EXA issues in older xservers. 9245 9246commit 6854f819f6bb89d67bfb737fd24c3e8b16e67e46 9247Author: Maarten Maathuis <madman2003@gmail.com> 9248Date: Tue Aug 5 13:08:22 2008 +0200 9249 9250 NV50: Disable messages from hide, show and load cursor, as they clutter the logs. 9251 9252commit 4b24be811b839f9f836b9f8574fc808d2182dd30 9253Author: Ben Skeggs <skeggsb@gmail.com> 9254Date: Mon Aug 4 16:09:28 2008 +1000 9255 9256 xv: fix thinko from when source was reformatted 9257 9258commit 30f54f2a4b81afde129170a26189cdb2bd3c7e9c 9259Author: Younes Manton <younes.m@gmail.com> 9260Date: Sun Aug 3 12:29:57 2008 -0400 9261 9262 g3dvl: Temporarily disable IDCT. 9263 9264commit 89d786435267af58bcecaff206a09b035e666df5 9265Author: Ben Skeggs <skeggsb@gmail.com> 9266Date: Tue Jul 29 23:38:45 2008 +1000 9267 9268 nv50: fix mystery typo 9269 9270commit c0b67f3f3490e826c86eedb32bbcf327592aaaf5 9271Author: Ben Skeggs <skeggsb@gmail.com> 9272Date: Tue Jul 29 01:26:20 2008 +1000 9273 9274 remove use of implicit variables from pushbuf macros 9275 9276 This matches the gallium winsys now. At some point the ddx will get updates 9277 so the "common" code is once again in sync to make it easier to do mm 9278 changes to both the ddx and dri simultaneously. 9279 9280 But, I want to get this part out of the way first as the changes might take 9281 a bit of work - and I don't want to have to deal with merge conflicts 9282 everywhere if someone changes the accel code before I'm done :) 9283 9284commit 80278c9f53f7c357e499bf2b3c5a1dff75fb9651 9285Author: Younes Manton <younes.m@gmail.com> 9286Date: Sun Jul 27 16:48:58 2008 -0400 9287 9288 g3dvl: Basic XvMC initialization. 9289 9290commit 1f8e27fd1d92d1c07dd8387f1a5e751ab3fab54a 9291Author: Maarten Maathuis <madman2003@gmail.com> 9292Date: Sun Jul 27 00:09:02 2008 +0200 9293 9294 NV50: I was proven wrong quicker than i thought, reinstate a slightly prettier hack. 9295 9296commit 41d46f59eb055130d5d415f0573fbede957c66c3 9297Author: Maarten Maathuis <madman2003@gmail.com> 9298Date: Sat Jul 26 21:53:26 2008 +0200 9299 9300 NV50: forgot something 9301 9302commit 1b905249c492af4485a9a2f4e7ae3ad5441c8eb5 9303Author: Maarten Maathuis <madman2003@gmail.com> 9304Date: Sat Jul 26 21:50:34 2008 +0200 9305 9306 NV50: my best guess at the lvds bios table, time will tell if it's correct for more than a few cases 9307 9308commit f3ec6e0e2d84e172cb634fd1348a54aef7256e74 9309Author: Dave Airlie <airlied@redhat.com> 9310Date: Thu Jul 24 15:41:57 2008 +1000 9311 9312 nouveau: fix macro as pointed out my moondrake on irc 9313 9314commit 6652e9c85fe8e807ed3d8ff76ec18c250aef436f 9315Author: Ben Skeggs <skeggsb@nisroch.keine.ath.cx> 9316Date: Thu Jul 24 10:13:05 2008 +1000 9317 9318 nv50: support YUY2 in textured video adaptor 9319 9320commit 3534f40b64496d9f6f95148ab81b9321057f7af1 9321Author: Ben Skeggs <skeggsb@gmail.com> 9322Date: Thu Jul 24 09:39:09 2008 +1000 9323 9324 nv50: remove 32bpp solid fill fallback 9325 9326 It seems to work, but there's probably a good reason it's a fallback in nv, 9327 so may be reincarnated with a more specific fallback if needed. 9328 9329commit 87f7d1f7084032110c347cb73f63e9a929d8c00e 9330Author: Ben Skeggs <skeggsb@gmail.com> 9331Date: Thu Jul 24 09:38:27 2008 +1000 9332 9333 Revert "NV50EXA: read the notes + cleanup + enabled 32bpp solid fill" 9334 9335 This reverts commit 0b6249fed6909af30856e1afc1a0dd4cd98c0b9e. 9336 9337commit ba7c239e7a13c4726dab7c686ee1c7f34d03a593 9338Author: Maarten Maathuis <madman2003@gmail.com> 9339Date: Thu Jul 24 01:30:16 2008 +0200 9340 9341 NV50EXA: serious brain fart of my part, sorry 9342 9343 - prepare copy may set up a rop and will set operation appropriately. 9344 9345commit 0b6249fed6909af30856e1afc1a0dd4cd98c0b9e 9346Author: Maarten Maathuis <madman2003@gmail.com> 9347Date: Wed Jul 23 21:01:57 2008 +0200 9348 9349 NV50EXA: read the notes + cleanup + enabled 32bpp solid fill 9350 9351 - I observe occasional corruption, some of this goes away when using a compositing manager. 9352 - The above statement contradicts my earlier statement in the previous commit message. 9353 - 32bpp solid fills seem to work fine, but i left the original fallback in case something pops up. 9354 - There is definately something wrong somewhere, i just don't know what. 9355 9356commit edbfbd09ac0a05c4125c3217f5d9c8f198d9345d 9357Author: Maarten Maathuis <madman2003@gmail.com> 9358Date: Wed Jul 23 13:34:28 2008 +0200 9359 9360 EXA: put NOUVEAU_FALLBACK in a do { } while (0), fixed opera corruption for me. 9361 9362 - I don't understand why a single line define would need this, but i noticed that turning on the fallback info fixed rendering. 9363 9364commit 36d13082dd2c2ba5d490b3dbc6e82d9ccfed8669 9365Author: Maarten Maathuis <madman2003@gmail.com> 9366Date: Sun Jul 20 15:44:47 2008 +0200 9367 9368 NV50_KMS: some basic scaling and dithering output property support 9369 9370commit 022a9ed5c2657466dd2471a1b29d5e8bc92ca57a 9371Author: Ben Skeggs <skeggsb@gmail.com> 9372Date: Sat Jul 19 22:14:50 2008 +1000 9373 9374 nv50: no more sync in composite() :) 9375 9376 0x502d mthd 0x0110 some kind of 2D cache flush? 9377 9378commit 6d8096c72e1aa160cef6729443ed06305a8ae870 9379Author: Ben Skeggs <skeggsb@gmail.com> 9380Date: Fri Jul 18 13:57:13 2008 +1000 9381 9382 nv50: sync after composite for the moment 9383 9384 This corrects some severe text rendering issues that get *very* bad with 9385 server 1.5. Will look into a proper fix soon. 9386 9387commit c68d8806b309ef4b48d7a98d6fc7aeccb8bb05ac 9388Author: Ben Skeggs <skeggsb@gmail.com> 9389Date: Fri Jul 18 13:39:59 2008 +1000 9390 9391 nv50: punt vertex emission macro out to header 9392 9393commit 4ad74cb03070f5279e0b1de2ed326f1167400d60 9394Author: Ben Skeggs <skeggsb@gmail.com> 9395Date: Thu Jul 17 11:15:59 2008 +1000 9396 9397 exa: missed a reloc delta 9398 9399 Can't actually hit this bug currently, but one day perhaps :) 9400 9401commit 98a9056224a86244f99ef7e844cda363e563a446 9402Author: Ben Skeggs <skeggsb@gmail.com> 9403Date: Thu Jul 17 11:15:08 2008 +1000 9404 9405 nv50: remove acquire/release surface stuff 9406 9407 It's left over from code that never made it to git in the first place, 9408 long overdue removal. 9409 9410commit f0305a29c2b4010894362d17f21cdfec39cd7b92 9411Author: Ben Skeggs <skeggsb@gmail.com> 9412Date: Thu Jul 17 11:14:02 2008 +1000 9413 9414 exa: rename m2mf locals to prevent some confusion 9415 9416commit 140c36ed936f4f28842af88efdd8b824c01bb097 9417Author: Danny <obiwan@mailmij.org> 9418Date: Fri Jul 18 03:07:38 2008 +0100 9419 9420 randr12: fix dithering output property endian issue (#16624) 9421 9422 tweaked and commented by Stuart Bennett 9423 9424commit b36802fe90165cf0fea61728b06605be5ac25d8d 9425Author: Pawel Worach <pawel.worach@gmail.com> 9426Date: Fri Jul 18 01:17:18 2008 +0200 9427 9428 Fix for #14858 - freebsd compile. Someone still needs to fix the kernel part though. 9429 9430commit 6f5e90ae739c64478ddf3e042e2bc9c4ff2b951c 9431Author: Ben Skeggs <skeggsb@gmail.com> 9432Date: Sat Jul 12 02:24:54 2008 +1000 9433 9434 nv30: more header name changes... 9435 9436commit 4b8427af583a8f84a09e8bd8d6ad614575a4f9c7 9437Author: Ben Skeggs <skeggsb@gmail.com> 9438Date: Sat Jul 12 02:15:08 2008 +1000 9439 9440 nv30: more bustage from header update 9441 9442commit 1672a783c4af58c80ed549fb3dc63c07fc75a882 9443Author: Ben Skeggs <skeggsb@gmail.com> 9444Date: Sat Jul 12 02:03:58 2008 +1000 9445 9446 nv30: un-change something that somehow changed in an earlier commit. oops! 9447 9448commit dae59585abec4b514789a8e580d30757f10d0357 9449Author: Ben Skeggs <skeggsb@gmail.com> 9450Date: Tue Jul 8 12:28:18 2008 +1000 9451 9452 nv50: exa/xv share some stuff, punt it out to nv50_accel.[ch] 9453 9454commit 17fbd810748b61b298c1ea9729d94f7d1177253b 9455Author: Ben Skeggs <skeggsb@gmail.com> 9456Date: Tue Jul 8 12:09:09 2008 +1000 9457 9458 nv50: Xv support 9459 9460 The NV12 YCbCr->RGB shader is hand-written just like the composite shaders, 9461 hopefully correct! I'll probably put a disasm next to each opcode at some 9462 point so it doesn't look like just a bunch of magics. 9463 9464 You'll need to use a composite manager if you want to watch videos, we can't 9465 render directly to the screen for the same reasons as mentioned in the 9466 composite commit. Though, a simple workaround is possible in the Xv case, 9467 so that'll probably come at some point. For now, PutImage() will fail with 9468 BadMatch. 9469 9470 The Y and UV plane upload may not be 100% for what NV50 requires, I want to 9471 rework all that at some point anyway. For now, at least there's some form 9472 of accelerated video to work with. 9473 9474 Enjoy :) 9475 9476commit ec452786d9aa3b6db41f1ab3c89aaa20a758f89e 9477Author: Ben Skeggs <skeggsb@gmail.com> 9478Date: Sun Jul 6 20:38:44 2008 +1000 9479 9480 nv50: de-magic things before someone accuses me of being an NVIDIA employee 9481 9482commit ad56c5e27642674d8e435bbbe3d6030097a01ff9 9483Author: Ben Skeggs <skeggsb@gmail.com> 9484Date: Sun Jul 6 01:30:21 2008 +1000 9485 9486 nv50: exa composite 9487 9488commit 33fad279e935a5a134ee5cc1777b8ed7510101d0 9489Author: Ben Skeggs <skeggsb@gmail.com> 9490Date: Sun Jul 6 01:21:39 2008 +1000 9491 9492 nv50: tile offscreen pixmaps 9493 9494 Until someone figures out how to make the display engine scan out from 9495 tiled surfaces things will be a bit painful. Future commits for exa 9496 composite and Xv will require the use of the 3D engine, which will *not* 9497 render to linear surfaces.. a composite manager will probably be a must 9498 for it to be usable. 9499 9500 I really don't want the NOUVEAU_MEM_TILE ioctl in the end, but we'll have 9501 to deal with it until we can drop support for <xorg-server-1.5. EXA driver 9502 handled pixmaps are far far nicer :) 9503 9504commit 5c1deac8384413dc483238f60c366e98253c11d6 9505Author: Ben Skeggs <skeggsb@gmail.com> 9506Date: Sun Jul 6 00:57:08 2008 +1000 9507 9508 exa: pass window coords to m2mf-based UTS/DFS 9509 9510commit b1f316977e2c79fdf5c83092e953bdfc988299e1 9511Author: Ben Skeggs <skeggsb@gmail.com> 9512Date: Mon Jul 7 01:03:02 2008 +1000 9513 9514 xv: remove some disturbing abuses of buffer related things 9515 9516 Also brings to attention a point about the overlay when we use a real 9517 memory manager. We'll need to find a way for the overlay to signal it's 9518 finished with a buffer, or just resort to pinning pPriv->video_mem.. 9519 9520commit 0ce028b6aaa6c20a1578148add2a96b2d37d9108 9521Author: Maarten Maathuis <madman2003@gmail.com> 9522Date: Sun Jul 6 13:45:32 2008 +0200 9523 9524 EXA: add a pixmapIsOffscreen hook. 9525 9526commit 86c70df254e63ebc5c405803d6faa9ae33df4995 9527Author: Maarten Maathuis <madman2003@gmail.com> 9528Date: Sun Jul 6 12:52:16 2008 +0200 9529 9530 NV50_KMS: support gamma changes 9531 9532commit 6de89c8cb219cc82b46e0d0d723fa9d1cefe4ac8 9533Author: Maarten Maathuis <madman2003@gmail.com> 9534Date: Sun Jul 6 12:29:13 2008 +0200 9535 9536 NV50: Unaccelerated rotation support. 9537 9538 - This will need a PixmapIsOffscreen hook before it will be considered for acceleration. 9539 9540commit 788fefa2e78a4abf7bd5c396ab9639bc6ff78451 9541Author: Maarten Maathuis <madman2003@gmail.com> 9542Date: Sun Jul 6 01:03:44 2008 +0200 9543 9544 NV50_KMS: Unaccelerated rotation support. 9545 9546 - This will need a PixmapIsOffscreen hook for it to be ever considered for acceleration. 9547 9548commit 2bc1cb832e0ded6cd1411757bec2a4b03536b7ea 9549Author: Maarten Maathuis <madman2003@gmail.com> 9550Date: Sun Jul 6 01:02:45 2008 +0200 9551 9552 Revert "EXA: Add PixmapIsOffscreen hook for EXA versions that allow it." 9553 9554 This reverts commit 5a5aee10f84fbea70eee69ece17ddf6e797497a6. 9555 This was never supposed to end up here as it's broken. 9556 9557commit 5a5aee10f84fbea70eee69ece17ddf6e797497a6 9558Author: Maarten Maathuis <madman2003@gmail.com> 9559Date: Sat Jul 5 23:07:58 2008 +0200 9560 9561 EXA: Add PixmapIsOffscreen hook for EXA versions that allow it. 9562 9563 - This should allow more flexibility for rotation shadow buffers. 9564 9565commit b9f2358ac35caedbba0339649ef13b03f209f2a7 9566Author: Ben Skeggs <skeggsb@gmail.com> 9567Date: Sun Jul 6 03:50:03 2008 +1000 9568 9569 xv: remove some defines that are in a header now 9570 9571commit 9f53b2a341a81e1bff87fee8bf5c2754cac17849 9572Author: Ben Skeggs <skeggsb@gmail.com> 9573Date: Sun Jul 6 02:28:26 2008 +1000 9574 9575 xv: remove drawable handling from hw-specific PutImage.. it's common... 9576 9577commit a50bba662ddfe49f1b0a7f8faa3668f5e810811f 9578Author: Ben Skeggs <skeggsb@gmail.com> 9579Date: Sun Jul 6 01:57:17 2008 +1000 9580 9581 xv: misc cleanups 9582 9583commit 08e9489b79395b234282b0c03d94601ce57678b8 9584Author: Maarten Maathuis <madman2003@gmail.com> 9585Date: Sat Jul 5 12:47:48 2008 +0200 9586 9587 NV50_KMS: minor change 9588 9589commit ea7e85b493ff0f684f9ccad92f84348cd835719b 9590Author: Maarten Maathuis <madman2003@gmail.com> 9591Date: Fri Jul 4 23:05:20 2008 +0200 9592 9593 Fix some compile warnings. 9594 9595commit 3298249d8267c27235653f17379f8c9f18e4f344 9596Author: Ben Skeggs <skeggsb@nisroch.keine.ath.cx> 9597Date: Thu Jul 3 23:54:13 2008 +1000 9598 9599 xv: reformat source so mere mortals can understand it 9600 9601commit a1e8ccdf85c5a0177df38a8a2d8c99d60890ec96 9602Author: Ben Skeggs <skeggsb@nisroch.(none)> 9603Date: Thu Jul 3 17:07:45 2008 +1000 9604 9605 xv: nv_video.c to nouveau_xv.c 9606 9607commit 3e17781625c0ef0ec34ee23a6e999de7f1fe4524 9608Author: Maarten Maathuis <madman2003@gmail.com> 9609Date: Thu Jul 3 08:30:43 2008 +0200 9610 9611 NV50_KMS: autodetect kernel modesetting when active 9612 9613commit 48c2116fb31872111239f5434215f30c8707091e 9614Author: Maarten Maathuis <madman2003@gmail.com> 9615Date: Wed Jul 2 19:04:58 2008 +0200 9616 9617 NV50_KMS: no int10 please 9618 9619commit 11102af2896b3fdf2364854b2d7939b2802c5b2a 9620Author: Stuart Bennett <sb476@cam.ac.uk> 9621Date: Fri Jun 27 16:21:19 2008 +0100 9622 9623 Backport console font save/restore from nouveau_ms 9624 9625commit ca2c55dfbf65ac268d43ba62db2399976eaf8f0a 9626Author: Stuart Bennett <sb476@cam.ac.uk> 9627Date: Thu Jun 26 14:44:45 2008 +0100 9628 9629 randr12: fix mode detection for when screen size changes 9630 9631 also some mem leaks 9632 9633commit 9c6eb667aa48f87f6bbbdae10cc9f4d237bdb0c6 9634Author: Maarten Maathuis <madman2003@gmail.com> 9635Date: Sat Jun 28 15:47:18 2008 +0200 9636 9637 NV04-NV4E: Save and restore FB_START 9638 9639commit 82fe9a21596ab80157df20b269ce8df41315de83 9640Author: Maarten Maathuis <madman2003@gmail.com> 9641Date: Fri Jun 27 18:49:16 2008 +0200 9642 9643 NV50_KMS: switch to output dpms 9644 9645commit 13ce1f4c1d02e70333a192db2efefb3f6a6765e3 9646Author: Maarten Maathuis <madman2003@gmail.com> 9647Date: Wed Jun 25 19:45:12 2008 +0200 9648 9649 Properly close drm when exiting + misc cleanup. 9650 9651commit 7cb0eed603ea0bd196a7471e43d60fa5e518c7fb 9652Author: Maarten Maathuis <madman2003@gmail.com> 9653Date: Wed Jun 25 19:07:58 2008 +0200 9654 9655 Fixup and add some license statements. 9656 9657 - The changed statements are a response to nv commit 2fdcda8ea62eba1b7885f0a75249884715936247. 9658 9659commit 27c7c88431cf368c3d3a953daf03ce3f9f15bc3a 9660Author: Ben Skeggs <skeggsb@gmail.com> 9661Date: Wed Jun 25 04:01:21 2008 +1000 9662 9663 change hack to work with latest drm changes.. 9664 9665commit 99a4b686d2b1f8e6768b8cd1cc3fcb6cea2bf6d3 9666Author: Maarten Maathuis <madman2003@gmail.com> 9667Date: Tue Jun 24 00:02:25 2008 +0200 9668 9669 NV50: unbreak kms and implement a basic off mode/dpms 9670 9671commit 1a23dbd1878909709fc130e00f6171508dcd41bf 9672Author: Ben Skeggs <skeggsb@nisroch.keine.ath.cx> 9673Date: Tue Jun 24 02:24:42 2008 +1000 9674 9675 nv50: stub out NV50SorSetClockMode for LVDS, it doesn't work there 9676 9677commit 2ccc80bd87a76a9f0498cfec531e8a46cbd2c384 9678Author: Ben Skeggs <skeggsb@gmail.com> 9679Date: Mon Jun 23 02:10:09 2008 +1000 9680 9681 nv50: bring in int10-constole-restore hack from nv driver 9682 9683 Lack of console restore has been annoying me personally lately while working 9684 on 3D, I lack the time/modsetting skills to fix it properly at the moment, 9685 so this'll do. 9686 9687commit 73b4fcc20334ac6cd2076c658a4937f9a98ee06d 9688Author: Ben Skeggs <skeggsb@gmail.com> 9689Date: Mon Jun 23 01:47:55 2008 +1000 9690 9691 fix non-kms build/run 9692 9693commit 75d8947d40e19c5fc039caf4289cf94fd1570eda 9694Author: Maarten Maathuis <madman2003@gmail.com> 9695Date: Sun Jun 22 16:34:22 2008 +0200 9696 9697 NV50: Some basic code to get kernel modesetting going. 9698 9699 - This will get a cleanup later on. 9700 9701commit df52dc4664df11d57fcf8a6c9ae3f87a149e5392 9702Author: Stuart Bennett <sb476@cam.ac.uk> 9703Date: Fri Jun 13 02:11:09 2008 +0100 9704 9705 No need for G80+ LVDS conf bits reports 9706 9707 I've got a few already. 9708 9709commit 1813559a20f0786b4decdb645f7ba3f52bc30739 9710Author: Stuart Bennett <sb476@cam.ac.uk> 9711Date: Fri Jun 13 01:58:41 2008 +0100 9712 9713 More DCB 1.5 TV variants (#16252 and #16342) 9714 9715commit 29c4a58bb2ba476f9558bda31f1082b4685555fc 9716Author: Stuart Bennett <sb476@cam.ac.uk> 9717Date: Fri Jun 13 01:40:31 2008 +0100 9718 9719 Be a little more discerning about the BIT bios signature 9720 9721 Thank you ABIT 9722 9723commit e8c790721d99e171c57b8dd497f5b3e82126c78b 9724Author: Stuart Bennett <sb476@cam.ac.uk> 9725Date: Fri Jun 13 01:25:49 2008 +0100 9726 9727 LVDS_INIT should not always use head A 9728 9729 While the script doesn't need the head variable set correctly, 9730 link_head_and_output still gets called and will therefore break panels expecting 9731 to be on head B 9732 Fixes Apple NV34 console restore 9733 9734commit fc137f23bed7405221b7f483d9151e296b99caf0 9735Author: Maarten Maathuis <madman2003@gmail.com> 9736Date: Fri Jun 13 19:12:38 2008 +0200 9737 9738 NV50: lvds is probably still broken, but it's a step in the right direction 9739 9740 - Just something i had lingering around. 9741 - Note that lvds still doesn't run by default, but you're more than welcome to fix it ;-) 9742 9743commit e034616ab2bbbc486f90c5f64e2bd2de194cf92e 9744Author: Maarten Maathuis <madman2003@gmail.com> 9745Date: Thu May 22 21:08:48 2008 +0200 9746 9747 NV50: A small extra piece of init that may prove useful to someone, somewhere, at some point for something. 9748 9749commit c21594cf452a1263682386d04718531d76a7a9b8 9750Author: Maarten Maathuis <madman2003@gmail.com> 9751Date: Tue May 20 19:19:53 2008 +0200 9752 9753 NV50: Be consistent with rules-ng. 9754 9755commit 9c1d87fa9e6152bffc5c5cdf16b1931cc6e58399 9756Author: Maarten Maathuis <madman2003@gmail.com> 9757Date: Sat May 10 21:58:00 2008 +0200 9758 9759 NV50: Some misc things. 9760 9761commit 5f08db9057f2970171186b20a25d346386dcb0f2 9762Author: Stuart Bennett <sb476@cam.ac.uk> 9763Date: Thu May 8 21:36:05 2008 +0100 9764 9765 Cosmetic g80 laptop dcb parsing enhancement 9766 9767commit 6a460eae941f3e9a7a8a0297efbbd12dca71a9dd 9768Author: Stuart Bennett <sb476@cam.ac.uk> 9769Date: Thu May 8 14:17:38 2008 +0100 9770 9771 randr12: some reordering, tidying &c 9772 9773commit 7455bc89116bfc5e88fbd444d83c1415a6020a7f 9774Author: Stuart Bennett <sb476@cam.ac.uk> 9775Date: Tue Apr 29 19:42:54 2008 +0100 9776 9777 Enable RandR 1.2 code-path by default 9778 9779commit 98a751e773c098382e7675e41ee60ce3d9a67ce0 9780Author: Stuart Bennett <sb476@cam.ac.uk> 9781Date: Wed May 7 21:25:03 2008 +0100 9782 9783 Don't call call_lvds_script with a null dcbent pointer 9784 9785 Stumbled over by hiyuh :) 9786 9787commit 77b24bc93659471bb20254847da124fdaba13a54 9788Author: Stuart Bennett <sb476@cam.ac.uk> 9789Date: Mon May 5 00:56:28 2008 +0100 9790 9791 randr12: tidy up some nv50 remnants in the pre nv50 code 9792 9793commit 8997f83cd2727b76240e9dc31ec7751f6d69cf6b 9794Author: Maarten Maathuis <madman2003@gmail.com> 9795Date: Tue May 6 19:46:52 2008 +0200 9796 9797 NV50: Some misc fixes. 9798 9799 - Even though tv-out is mentioned, it by no means implies that is (nearly) functional. 9800 9801commit ae574ec7d930d8708e4dd9d1d47558666019afe5 9802Author: Maarten Maathuis <madman2003@gmail.com> 9803Date: Mon May 5 11:02:07 2008 +0200 9804 9805 nv50_exa.c had no license statement. 9806 9807commit a7af057ff67cf5d7baf78d852dee5ac69b24e706 9808Author: Maarten Maathuis <madman2003@gmail.com> 9809Date: Mon May 5 01:26:37 2008 +0200 9810 9811 NV50: Rework the modesetting code into a different model. 9812 9813 - There is now interface independent code for most things. 9814 - This structure, although not exclusive, is only used for NV50 atm. 9815 - The change in approach is that the concept of a connector is added to the model (randr12 sucks in this aspect). 9816 - I'm hacking around randr trying to create a single output for each connector, the indices are the bus numbers. 9817 - Limitations: Hot swapping vga and dvi of same monitor will not make randr12 think it changed, so for the moment force a change. 9818 - Concept is largely inspired by the radeonhd driver. 9819 9820commit 5e1b5708d3a7e4298f80b1a1b8bb3fafae0c69bd 9821Author: Stuart Bennett <sb476@cam.ac.uk> 9822Date: Fri May 2 20:36:59 2008 +0100 9823 9824 xv: nv10 double-buffered overlay mode was getting set on nv04 9825 9826commit 85536023ed7050632d121299ab99382e096b9984 9827Author: Stuart Bennett <sb476@cam.ac.uk> 9828Date: Wed Apr 30 00:59:47 2008 +0100 9829 9830 Use the actual number of BIT table entries 9831 9832commit 78588c71ffd19e18602b95a55186bc53eebb51c3 9833Author: Stuart Bennett <sb476@cam.ac.uk> 9834Date: Mon Apr 28 16:51:01 2008 +0100 9835 9836 randr12, xv: manual overlay clipping for randr12 (see #12825) 9837 9838 There could be something to be said for doing this clipping when using the blitter and texture adapters too 9839 9840commit 6f7d00a0cb4ec85b058afb0150a122a124542b5f 9841Author: Stuart Bennett <sb476@cam.ac.uk> 9842Date: Tue Apr 29 22:53:40 2008 +0100 9843 9844 randr12: remove unneeded separate dcb_entry and "or" members 9845 9846 Make off-chip outputs clearer too 9847 9848commit 52e58c7e799697989fcfbf95050ce10a4c3d1f8f 9849Author: Stuart Bennett <sb476@cam.ac.uk> 9850Date: Tue Apr 29 22:08:01 2008 +0100 9851 9852 Pass DCB entry structs in bios functions, rather than index 9853 9854commit 0c1da6907d8e29ed1c058f9655d436c228ea91b6 9855Author: Stuart Bennett <sb476@cam.ac.uk> 9856Date: Tue Apr 29 21:53:22 2008 +0100 9857 9858 Pass output field, rather than dcb indices, for tmds access functions 9859 9860commit 2333bc907a4be3c7e36bdd02f27e22b63a7ee7c3 9861Author: Stuart Bennett <sb476@cam.ac.uk> 9862Date: Tue Apr 29 21:58:17 2008 +0100 9863 9864 Add an index field to dcb entries, and carry pointer to dcb entry in output struct 9865 9866commit 21f062cad184f21d33d7f5b43f86937979291b58 9867Author: Stuart Bennett <sb476@cam.ac.uk> 9868Date: Tue Apr 29 21:27:07 2008 +0100 9869 9870 randr12, xv: oops, transform_in_use doesn't exist on xserver 1.3 9871 9872commit ee78dc0abef626e29f7617a4b324e847bab1abcc 9873Author: Stuart Bennett <sb476@cam.ac.uk> 9874Date: Mon Apr 28 02:12:17 2008 +0100 9875 9876 randr12: C51 I2C 9877 9878commit 9b70dfaf4f0317e622ca64fc0804f74a31158108 9879Author: Stuart Bennett <sb476@cam.ac.uk> 9880Date: Tue Apr 29 18:46:24 2008 +0100 9881 9882 randr12: only set cr59 to 1 for tmds (fixes nv34 issue) 9883 9884commit c8cec9b2a0b18cd1159613d4dd218725695aef7c 9885Author: Stuart Bennett <sb476@cam.ac.uk> 9886Date: Mon Apr 28 17:00:27 2008 +0100 9887 9888 randr12, xv: no overlay when the crtc is transformed 9889 9890commit 3233ebba07e7adab347dc3c6f9d17f02e1a9d32b 9891Author: Stuart Bennett <sb476@cam.ac.uk> 9892Date: Sat Apr 26 02:08:24 2008 +0100 9893 9894 randr12: fix spread spectrum setting for lvds and turn it off during dpms off 9895 9896 Assuming that the bios has set this up correctly for LVDS initially is probably safe 9897 9898commit 467f83aa43bffdf545b513fdb7b11b4fe2205b4f 9899Author: Stuart Bennett <sb476@cam.ac.uk> 9900Date: Sat Apr 26 19:59:58 2008 +0100 9901 9902 randr12: make nv50 code depend on Architecture 9903 9904commit 8b165722d9f9461353abf84b4136e33b816ac842 9905Author: Stuart Bennett <sb476@cam.ac.uk> 9906Date: Sat Apr 26 01:21:50 2008 +0100 9907 9908 CR27 appears to contain the chip revision 9909 9910commit 59ff20738e88d6621f9beaea94810274029a21a3 9911Author: Maarten Maathuis <madman2003@gmail.com> 9912Date: Sun Apr 27 19:52:41 2008 +0200 9913 9914 NV50: A little bit more information, no functional change. 9915 9916commit bb5a2a105f264502d0985ce70e76d5a3d3b6463c 9917Author: Stuart Bennett <sb476@cam.ac.uk> 9918Date: Fri Apr 25 22:09:46 2008 +0100 9919 9920 randr12: sel_clk binding fix 9921 9922commit 5a85626cccbba80ef3d6fb2fb90ae609b7abde15 9923Author: Stuart Bennett <sb476@cam.ac.uk> 9924Date: Fri Apr 25 02:09:18 2008 +0100 9925 9926 randr12: unified mode validation 9927 9928 doublescan/interlace limits applied too 9929 9930commit 38ad350118aa9592bdbcb530fe63b85983fb9f64 9931Author: Stuart Bennett <sb476@cam.ac.uk> 9932Date: Fri Apr 25 00:03:41 2008 +0100 9933 9934 randr12: only set up one spread spectrum clock on pre nv40 9935 9936 avoids inadvertently getting wobbly crt pictures 9937 9938commit dfd8b42f91257f2cb0762b3d9b07612a7edac549 9939Author: Stuart Bennett <sb476@cam.ac.uk> 9940Date: Thu Apr 24 22:23:50 2008 +0100 9941 9942 randr12: deny doublescan modes to digital outputs 9943 9944commit d150ce5c633c90a82411b4b8e44fa4b91647e6a8 9945Author: Stuart Bennett <sb476@cam.ac.uk> 9946Date: Thu Apr 24 01:22:49 2008 +0100 9947 9948 randr12: off-chip dvi dpms and fp_control dpms fixed properly 9949 9950 Also solves output exclusion problem: two crtcs claiming the same output 9951 through cr33 does not work 9952 9953commit dd152ce186fea953968805cf4a1b309a1fd3987b 9954Author: Stuart Bennett <sb476@cam.ac.uk> 9955Date: Thu Apr 24 00:24:35 2008 +0100 9956 9957 randr12: set CRTCX_LCD in nv_output 9958 9959 This makes my nv28 external dvi work. 9960 9961commit b8d401b0a56d23db7199147e6a5dee1edccff4ff 9962Author: Stuart Bennett <sb476@cam.ac.uk> 9963Date: Thu Apr 24 00:01:50 2008 +0100 9964 9965 These made up defines are more misleading than helpful 9966 9967commit fa8c8437798cfc5494a639d649f0f91f0e7275ce 9968Author: Stuart Bennett <sb476@cam.ac.uk> 9969Date: Tue Apr 22 22:42:12 2008 +0100 9970 9971 randr12: remove FPP setting and other FSEL changes 9972 9973commit a083bb1a55d0983991c45d113eb7e9814369fff4 9974Author: Stuart Bennett <sb476@cam.ac.uk> 9975Date: Tue Apr 22 23:25:40 2008 +0100 9976 9977 randr12: oops (fix to 9edf6d60) 9978 9979commit 16f420e339f0dd6d55d7b87763d97c2eb8e77e66 9980Author: Stuart Bennett <sb476@cam.ac.uk> 9981Date: Tue Apr 22 21:34:52 2008 +0100 9982 9983 randr12: RAMDAC_OUTPUT is only for analogue outputs 9984 9985commit 52174047f3f817fa9ce3674440a13687b73ccea8 9986Author: Stuart Bennett <sb476@cam.ac.uk> 9987Date: Tue Apr 22 21:29:48 2008 +0100 9988 9989 Revert "randr12: correct SEL_CLK register beheaviour for outputs with or=2" 9990 9991 This reverts commit 77af887089be6fef7b793e3c3afdb98ac90a2e8c. 9992 9993 External DVI (which is source of or == 2) do not use SEL_CLK 9994 9995commit 61e416153b24516951962aca407ceaa9b7885e22 9996Author: Maarten Maathuis <madman2003@gmail.com> 9997Date: Sun Apr 13 02:40:55 2008 +0200 9998 9999 NV50: Another few regs. 10000 10001commit 5328370d2f7e10c61485572f53031deace69c515 10002Author: Maarten Maathuis <madman2003@gmail.com> 10003Date: Sun Apr 13 02:16:27 2008 +0200 10004 10005 NV50: Some more regs. 10006 10007commit e5511a564564350bd8755b7f18a1448ea7c796de 10008Author: Maarten Maathuis <madman2003@gmail.com> 10009Date: Sun Apr 13 01:21:20 2008 +0200 10010 10011 NV50: I now know what UNK8A8 is, so update the value register as well. 10012 10013commit 2c9027679db1657e6fe02cf6674024ca97f4cd6e 10014Author: Maarten Maathuis <madman2003@gmail.com> 10015Date: Sat Apr 12 02:11:32 2008 +0200 10016 10017 NV50: Re-implement output status caching. 10018 10019commit d7681e8ad70c07ec963ac9425b868a2686d79769 10020Author: Stuart Bennett <sb476@cam.ac.uk> 10021Date: Fri Apr 11 17:28:16 2008 +0100 10022 10023 randr12: separate pll and arbitration stages of calc_state_ext, and call from mode_set directly 10024 10025commit 0bf08d7699d7c3a1748060febded1c8b3950c9bd 10026Author: Stuart Bennett <sb476@cam.ac.uk> 10027Date: Fri Apr 11 16:20:21 2008 +0100 10028 10029 state->config is not used 10030 10031commit 08b8bd8a97b580a57d7bb75b3da7827d65c046f5 10032Author: Stuart Bennett <sb476@cam.ac.uk> 10033Date: Fri Apr 4 18:59:29 2008 +0100 10034 10035 randr12: unite pre nv40 and nv40 pll restore 10036 10037commit 5b443084ec7b03230d0ec7490c5a3378a9f7187e 10038Author: Stuart Bennett <sb476@cam.ac.uk> 10039Date: Tue Apr 8 02:25:34 2008 +0100 10040 10041 Use the correct reference clock when reading (V)PLLs 10042 10043commit 0f533b8cc12f49f5d5ac017838f7330a4f8b47ea 10044Author: Stuart Bennett <sb476@cam.ac.uk> 10045Date: Tue Apr 8 02:25:22 2008 +0100 10046 10047 Various PLL code improvements 10048 10049commit 4e1caa6edfe9499e6c501e70a8fba14bae510462 10050Author: Maarten Maathuis <madman2003@gmail.com> 10051Date: Wed Apr 9 18:55:06 2008 +0200 10052 10053 NV50: Some misc stuff. 10054 10055commit 099128175ad31956fd1c6622166b65f4ab1fe370 10056Author: Dave Airlie <airlied@redhat.com> 10057Date: Tue Apr 8 16:48:37 2008 +1000 10058 10059 nv50: fix distcheck since header removal 10060 10061commit 4af376fa6031bd41dc247e950319f3a151c49f22 10062Author: Stuart Bennett <sb476@cam.ac.uk> 10063Date: Tue Apr 8 00:28:03 2008 +0100 10064 10065 MCP67 does not do the second stage gain tests 10066 10067commit dbca8b65d7b12a1dd2eb7aee3ac37ece0bd4ce65 10068Author: Stuart Bennett <sb476@cam.ac.uk> 10069Date: Tue Apr 8 01:31:14 2008 +0100 10070 10071 PLL stuff 10072 10073commit 39b2e0b842f4a4e1c3de96c94e380aa9a0327c12 10074Author: Stuart Bennett <sb476@cam.ac.uk> 10075Date: Sat Apr 5 05:20:54 2008 +0100 10076 10077 If no PLL coefficients compatible with the constraints exist, don't brick the hardware 10078 10079commit 0b54450e1db7c6411528bc4a3bac4d1a42510aca 10080Author: Maarten Maathuis <madman2003@gmail.com> 10081Date: Mon Apr 7 22:12:44 2008 +0200 10082 10083 NV50: Scaling works fine on VGA connectors, so support that as well. 10084 10085 - The default is SCALE_PANEL instead of SCALE_ASPECT for obvious reasons. 10086 10087commit 891e72f4316c4e7e1b9d06f39e075b1a8a8a8c53 10088Author: Maarten Maathuis <madman2003@gmail.com> 10089Date: Mon Apr 7 21:18:54 2008 +0200 10090 10091 NV50: Deobfuscate some more commands. 10092 10093commit 13c701f9d23a598c281ced726070a219e1b78607 10094Author: Maarten Maathuis <madman2003@gmail.com> 10095Date: Mon Apr 7 19:29:55 2008 +0200 10096 10097 Make some CARD32 die. 10098 10099commit a3a0710604017f3ceb9c552552d71c3f5dcfab0f 10100Author: Maarten Maathuis <madman2003@gmail.com> 10101Date: Mon Apr 7 19:09:35 2008 +0200 10102 10103 NV50: Give a few registers an UNK label. 10104 10105commit be72a7c9104c1c1d6db960b1d80cc4d820d37440 10106Author: Maarten Maathuis <madman2003@gmail.com> 10107Date: Mon Apr 7 18:35:18 2008 +0200 10108 10109 randr12: NV04/05/06 can't accelerate rotation, so it's better to disable it. 10110 10111commit 9cf76c91a661587d9c54845e6a06fb25cf9b55a2 10112Author: Maarten Maathuis <madman2003@gmail.com> 10113Date: Sun Apr 6 23:36:15 2008 +0200 10114 10115 NV50: Some more stuff. 10116 10117commit c725e33078a2fed374fbb52175beba92b0b9fbe8 10118Author: Maarten Maathuis <madman2003@gmail.com> 10119Date: Sun Apr 6 19:49:11 2008 +0200 10120 10121 NV50: minor rename to avoid confusion 10122 10123commit fcda539ce60ca02f2f4589157f8b8c522430571f 10124Author: Maarten Maathuis <madman2003@gmail.com> 10125Date: Sun Apr 6 15:16:40 2008 +0200 10126 10127 NV50: Deobfuscate another register (partially). 10128 10129commit fa93541b257a8055ea8b55c883475d2db717d971 10130Author: Maarten Maathuis <madman2003@gmail.com> 10131Date: Sun Apr 6 15:05:33 2008 +0200 10132 10133 NV50: Fix another deadlock in the clock setting code. 10134 10135 - Swapping my monitor from dvi to vga causes no mode (=clock) change, but still the supervisor expects a clock change. 10136 - This meant it got stuck in stage 2, with no clock that is supposed to change. 10137 10138commit c001fea3b0ae3921c063e7f44c298cafdfef51f6 10139Author: Maarten Maathuis <madman2003@gmail.com> 10140Date: Sat Apr 5 23:33:22 2008 +0200 10141 10142 Forgot a few things. 10143 10144commit 17a8e1743478e5548a90caf8f84a387366020416 10145Author: Maarten Maathuis <madman2003@gmail.com> 10146Date: Sat Apr 5 23:23:22 2008 +0200 10147 10148 NV50: rename randr-1.2 functions to be nv50_something_function 10149 10150commit a71ef4da76e33e84303a8b9181dbcfd655e78aa9 10151Author: Maarten Maathuis <madman2003@gmail.com> 10152Date: Sat Apr 5 23:00:39 2008 +0200 10153 10154 NV50: Some more cleanup. 10155 10156commit aeab563cc89b2fc3cd6cb306aab880b6e28f2b64 10157Author: Maarten Maathuis <madman2003@gmail.com> 10158Date: Sat Apr 5 22:52:58 2008 +0200 10159 10160 NV50: Some reordering. 10161 10162commit b899ab75781a3076898791381604be575c6d8a70 10163Author: Maarten Maathuis <madman2003@gmail.com> 10164Date: Sat Apr 5 21:17:23 2008 +0200 10165 10166 Some minor corrections. 10167 10168commit b0ef158a7e281329825c7b6c10cab6097e627b54 10169Author: Maarten Maathuis <madman2003@gmail.com> 10170Date: Sat Apr 5 21:06:03 2008 +0200 10171 10172 NV50: Document dvi hotplug detection. 10173 10174 - This can only be used once modesetting is moved to the kernel. 10175 10176commit be54b9299b4a261fc707b74407cf4d847fcbb243 10177Author: Maarten Maathuis <madman2003@gmail.com> 10178Date: Sat Apr 5 16:16:26 2008 +0200 10179 10180 NV50: Remove some questionable guesses. 10181 10182commit c06f202cfbee8568fb8358ea1caffec41c71e648 10183Author: Maarten Maathuis <madman2003@gmail.com> 10184Date: Sat Apr 5 01:54:06 2008 +0200 10185 10186 nvbios: Drop a N2/M2 requirement on NV5x hardware, as it prevents getting a mode on 1024x768'ish and lower. 10187 10188 - Limitations are very strict in nv5x bios tables and N2/M2 can drop below 2 (the blob does this too). 10189 10190commit 5ce26b9aec214dd99e5399fe0f864d8faad15d2c 10191Author: Maarten Maathuis <madman2003@gmail.com> 10192Date: Fri Apr 4 22:49:06 2008 +0200 10193 10194 nvbios: Add some valid register ranges for nv50. 10195 10196commit ce86484979d873325ec4c74ed84107be9ad9ba62 10197Author: Maarten Maathuis <madman2003@gmail.com> 10198Date: Fri Apr 4 22:38:30 2008 +0200 10199 10200 NV50: Switch to common pll calculator function. 10201 10202commit 821ad3d643f9564d043c7f3909e0b8be28e94965 10203Author: Stuart Bennett <sb476@cam.ac.uk> 10204Date: Wed Apr 2 16:58:45 2008 +0100 10205 10206 randr12: delete stuff commented out a week ago that no one's complained about 10207 10208commit e90e534728e9cbe65ee6c70faf596db449495665 10209Author: Maarten Maathuis <madman2003@gmail.com> 10210Date: Fri Apr 4 17:13:25 2008 +0200 10211 10212 NV50: Minor comment updates. 10213 10214commit 08550984aba2bfa3eaeeddbbf6af25206819abe8 10215Author: Maarten Maathuis <madman2003@gmail.com> 10216Date: Fri Apr 4 17:04:49 2008 +0200 10217 10218 NV50: Some more stuff. 10219 10220commit ef4339aacf72bf5422b22ab18b73416f2738b284 10221Author: Maarten Maathuis <madman2003@gmail.com> 10222Date: Fri Apr 4 16:49:00 2008 +0200 10223 10224 NV50: Some more registers. 10225 10226commit 3183df1aa61f3c441a93e5a8b50db6628d804465 10227Author: Maarten Maathuis <madman2003@gmail.com> 10228Date: Thu Apr 3 22:42:31 2008 +0200 10229 10230 NV50: Update my knowledge of interrupts. 10231 10232commit 04fd80d29913ed02524de177fa7c87b616147d15 10233Author: Stuart Bennett <sb476@cam.ac.uk> 10234Date: Thu Apr 3 03:00:13 2008 +0100 10235 10236 Small declaration reordering 10237 10238commit 92934e4b7571a578386327ebdb59eb6d10992763 10239Author: Stuart Bennett <sb476@cam.ac.uk> 10240Date: Thu Apr 3 01:03:38 2008 +0100 10241 10242 Move some variable declarations in the bios parser 10243 10244commit 927384005a01a606835452c9d1e26d7df98f81cb 10245Author: Stuart Bennett <sb476@cam.ac.uk> 10246Date: Thu Apr 3 00:11:31 2008 +0100 10247 10248 randr12: some fixes to dithering property 10249 10250commit 52c880c97acf29d5ed8dfeec5810fcc7d8bb7b35 10251Author: Stuart Bennett <sb476@cam.ac.uk> 10252Date: Wed Apr 2 19:45:43 2008 +0100 10253 10254 randr12: better analogue clock limits (taken from pNv->{Min,Max}VClockFreqKHz) 10255 10256commit 29714e7b9ae952aecd036086c053ec030aad94c9 10257Author: Maarten Maathuis <madman2003@gmail.com> 10258Date: Wed Apr 2 21:31:07 2008 +0200 10259 10260 NV50: Some minor changes to the display supervisor (mostly comment). 10261 10262commit 9ac32d3e2bf400f6a6220118b8008804cd11e435 10263Author: Stuart Bennett <sb476@cam.ac.uk> 10264Date: Wed Apr 2 19:03:07 2008 +0100 10265 10266 Some bios parser PLL writing improvements 10267 10268commit 4a192ea6cf9df39d95368f46482c777524471118 10269Author: Maarten Maathuis <madman2003@gmail.com> 10270Date: Wed Apr 2 18:21:01 2008 +0200 10271 10272 NV50: Add my name to some files. 10273 10274commit db6220b8611829a77d354de1e45f6fe29e118b2a 10275Author: Stuart Bennett <sb476@cam.ac.uk> 10276Date: Wed Apr 2 16:16:13 2008 +0100 10277 10278 randr12: since the crtc function records are now mutable, unify pre nv50 templates 10279 10280commit b9762ba272a3ee817278e3967794c55a4fe73961 10281Author: Stuart Bennett <sb476@cam.ac.uk> 10282Date: Wed Apr 2 16:02:43 2008 +0100 10283 10284 randr12: don't provide rotation functions in NoAccel case 10285 10286commit a24ab3d1559d023270915f373527efb6eefacd8d 10287Author: Stuart Bennett <sb476@cam.ac.uk> 10288Date: Wed Apr 2 16:05:30 2008 +0100 10289 10290 Revert "randr12: Fail rotation in NoAccel case." 10291 10292 This reverts commit b33317ed98957db42691f5f9c41c119c10c91be6. 10293 10294commit 33d4c2f2fe6c6c400ce945f4c7ecc09ee08123a5 10295Author: Stuart Bennett <sb476@cam.ac.uk> 10296Date: Wed Apr 2 15:46:50 2008 +0100 10297 10298 Add quirks for Apple *book backlight 10299 10300commit 8a3f051dc9f007c3d801942d32899b8d6c09fd34 10301Author: Maarten Maathuis <madman2003@gmail.com> 10302Date: Wed Apr 2 14:54:16 2008 +0200 10303 10304 NV50: Some more things. 10305 10306commit b33317ed98957db42691f5f9c41c119c10c91be6 10307Author: Maarten Maathuis <madman2003@gmail.com> 10308Date: Wed Apr 2 09:28:45 2008 +0200 10309 10310 randr12: Fail rotation in NoAccel case. 10311 10312commit d017dfa93e7eff988339b8030af1a563347156bc 10313Author: Stuart Bennett <sb476@cam.ac.uk> 10314Date: Tue Apr 1 21:23:35 2008 +0100 10315 10316 NV30/35 setPLL bios support 10317 10318commit 32f550a534dc3e19f7cd3640549336846ad6052d 10319Author: Maarten Maathuis <madman2003@gmail.com> 10320Date: Tue Apr 1 22:33:15 2008 +0200 10321 10322 NV50: Fix a minor cursor issue. 10323 10324commit 27e819dbcd5ce36be71c7dd3be332f777a80caf4 10325Author: Maarten Maathuis <madman2003@gmail.com> 10326Date: Tue Apr 1 22:20:04 2008 +0200 10327 10328 NV50: Add some debugging statements to important functions. 10329 10330commit 99857933e8edc3be2e4af0c8c7e3fe3b1e221de9 10331Author: Maarten Maathuis <madman2003@gmail.com> 10332Date: Tue Apr 1 20:33:16 2008 +0200 10333 10334 NV50: Some i2c stuff. 10335 10336commit 9199307195c51b52979c483b3c0adf49f2301281 10337Author: Stuart Bennett <sb476@cam.ac.uk> 10338Date: Tue Apr 1 05:16:29 2008 +0100 10339 10340 Use new PLL reading code 10341 10342 Of course, having spent some time perfecting the reading of NVPLL and MPLL, 10343 it turns out to be totally unnecessary on >= nv30 :-/ 10344 10345commit a3f82d92ae6cbd779d341fe11687d12c5aac1762 10346Author: Stuart Bennett <sb476@cam.ac.uk> 10347Date: Tue Apr 1 05:03:14 2008 +0100 10348 10349 Unified PLL reading 10350 10351commit 350fb202e113bc03a71fd7e451e2c26ec990a61a 10352Author: Stuart Bennett <sb476@cam.ac.uk> 10353Date: Tue Apr 1 05:29:05 2008 +0100 10354 10355 randr12: get_clock_from_crtc only necessary for TMDS 10356 10357commit f741bf7b724dd0698f71b28c461337526f20fe31 10358Author: Stuart Bennett <sb476@cam.ac.uk> 10359Date: Tue Apr 1 00:23:06 2008 +0100 10360 10361 Better PLL limits for some nv3x 10362 10363commit 94ca1951feda10b9d6ad828816a459af750de286 10364Author: Stuart Bennett <sb476@cam.ac.uk> 10365Date: Tue Apr 1 01:00:16 2008 +0100 10366 10367 randr12: deal with nv30 and nv35 PLLs individually 10368 10369commit 626d8d62eef205dc542e3a7cd82beb329a23cdaf 10370Author: Maarten Maathuis <madman2003@gmail.com> 10371Date: Mon Mar 31 19:27:28 2008 +0200 10372 10373 Init global gamma on all crtc's, even if they are currently not in use. 10374 10375commit f54cea5662fdfc17762711a88892a27aed62f74c 10376Author: Maarten Maathuis <madman2003@gmail.com> 10377Date: Mon Mar 31 19:17:50 2008 +0200 10378 10379 NV50: Some clock related cleanups + minor changes. 10380 10381commit 7393d8fc76f7bbb8aa156587d81ee1a01c17470e 10382Author: Maarten Maathuis <madman2003@gmail.com> 10383Date: Mon Mar 31 14:10:42 2008 +0200 10384 10385 NV50: Some misc cleanup. 10386 10387commit 9edf6d60b2b6698e88edc7798285bd843bfce0f4 10388Author: Stuart Bennett <sb476@cam.ac.uk> 10389Date: Sun Mar 30 21:48:13 2008 +0100 10390 10391 randr12: different handling for sel_clk for pre and post nv40 10392 10393 This won't set the spread spectrum bits for pre nv40 from null initial state, but that's not a major concern right now 10394 10395commit 84ab4be08947ef35900da6b84702bc9acb4121bb 10396Author: Stuart Bennett <sb476@cam.ac.uk> 10397Date: Sat Mar 29 21:26:37 2008 +0000 10398 10399 Minor tidyup to output merging 10400 10401commit 1fe1d8161059afcc6f6ba152b9331e1576ef1f93 10402Author: Maarten Maathuis <madman2003@gmail.com> 10403Date: Sun Mar 30 22:42:47 2008 +0200 10404 10405 Kill a useless register offset. 10406 10407commit 75680d723144270c12a6d500d8e2c2c421e4ebc1 10408Author: Maarten Maathuis <madman2003@gmail.com> 10409Date: Sun Mar 30 15:18:09 2008 +0200 10410 10411 NV50: remove unneeded function 10412 10413commit 263ec94ec0f553f5ab6a22d3fd335bebff2c029b 10414Author: Maarten Maathuis <madman2003@gmail.com> 10415Date: Sun Mar 30 12:21:29 2008 +0200 10416 10417 NV50: Some minor cursor cleanups. 10418 10419commit 86722ad60fb939743f5ab73413f721cd26bf2c3c 10420Author: Maarten Maathuis <madman2003@gmail.com> 10421Date: Sun Mar 30 03:37:16 2008 +0200 10422 10423 NV50: Deobfuscate some of the modesetting, a few things were hidden away in innocent looking variables. 10424 10425 - Some mode variables were overridden and didn't represent their name anymore. 10426 10427commit 633a8f4f0a8ff3bcb5e6afa9bbcf3f6ed44ef905 10428Author: Stuart Bennett <sb476@cam.ac.uk> 10429Date: Sat Mar 29 19:01:30 2008 +0000 10430 10431 randr12: make sel_clk behaviour match comment (which seems to be correct) 10432 10433 If nv40 breaks, this whole thing will need another look 10434 10435commit 7a8bbb7f0890ef99de539f0f7acccd469f33dce9 10436Author: Stuart Bennett <sb476@cam.ac.uk> 10437Date: Sat Mar 29 00:03:27 2008 +0000 10438 10439 Fix Apple backlight code, broken in pci-rework conversion 10440 10441commit 4d0974c48eff8cd51af66f8fded4fad50044958a 10442Author: Stuart Bennett <sb476@cam.ac.uk> 10443Date: Fri Mar 28 01:10:31 2008 +0000 10444 10445 Consistency fix 10446 10447 + constify 10448 10449commit acd007843f00cb99cf26c4a5f885e365c3a8065b 10450Author: Maarten Maathuis <madman2003@gmail.com> 10451Date: Sat Mar 29 21:47:04 2008 +0100 10452 10453 It seems wise to NVSync for all hardware on NVLeaveVT. 10454 10455commit ea27bc38938277599228ba403a082963bf89d702 10456Author: Maarten Maathuis <madman2003@gmail.com> 10457Date: Fri Mar 28 19:56:36 2008 +0100 10458 10459 NV50: Better safe than sorry. 10460 10461commit d4ca83e2d1f1432dc7de290087f03a6ff3553d98 10462Author: Stuart Bennett <sb476@cam.ac.uk> 10463Date: Thu Mar 27 23:55:30 2008 +0000 10464 10465 randr12: shrink pitch alignment stuff and move 10466 10467commit 43ff98cd89e3dc43c281e41f3a36a76bcd3e5a65 10468Author: Stuart Bennett <sb476@cam.ac.uk> 10469Date: Wed Mar 26 18:35:14 2008 +0000 10470 10471 randr12: tidy-up 3/3 - TMDS04 twiddling 10472 10473 Code commented for easy reinstatement should problems arise 10474 10475commit 9491ee3f3fbb8b50244444424ef6148217ace944 10476Author: Stuart Bennett <sb476@cam.ac.uk> 10477Date: Wed Mar 26 18:34:28 2008 +0000 10478 10479 randr12: tidy-up 2/3 - DPMS CR57/58 twiddling 10480 10481 Code commented for easy reinstatement should problems arise 10482 10483commit 2e001c140eacc5d5f02435b9c81b13e5d40db71f 10484Author: Stuart Bennett <sb476@cam.ac.uk> 10485Date: Wed Mar 26 18:33:59 2008 +0000 10486 10487 randr12: tidy-up 1/3 - FP_DEBUG_0 twiddling 10488 10489 Code commented for easy reinstatement should problems arise 10490 10491commit ec9fa3c958d6763ff96d8b884ed36c80d911475a 10492Author: Stuart Bennett <sb476@cam.ac.uk> 10493Date: Thu Mar 27 19:24:04 2008 +0000 10494 10495 randr12: allow doublescan and low clock modes 10496 10497 320x240 ftw. Also improved mode rejection error messages 10498 10499commit 048f31014092c1dccb1944d2b1a84c7a301be48e 10500Author: Stuart Bennett <sb476@cam.ac.uk> 10501Date: Wed Mar 26 18:58:47 2008 +0000 10502 10503 Minor fix to lvds script caching commit 10504 10505commit 3feaa48b53f51c094718708e52f08b546b0b0e27 10506Author: Maarten Maathuis <madman2003@gmail.com> 10507Date: Wed Mar 26 20:01:19 2008 +0100 10508 10509 NV50: Implement per CRTC gamma correction. 10510 10511 - This remains untested, so if you're in the mood to test ;-) 10512 10513commit 77af887089be6fef7b793e3c3afdb98ac90a2e8c 10514Author: Maarten Maathuis <madman2003@gmail.com> 10515Date: Wed Mar 26 10:02:46 2008 +0100 10516 10517 randr12: correct SEL_CLK register beheaviour for outputs with or=2 10518 10519commit cb5d4cdb3dc83f9dbd53eae9097c0df458313fff 10520Author: Stuart Bennett <sb476@cam.ac.uk> 10521Date: Wed Mar 26 02:27:33 2008 +0000 10522 10523 Cache running of LVDS scripts, rather than blocking LVDS_RESET 10524 10525 Fixes a regression on some nv40 where LVDS_RESET is needed to fix some bad behaviour in the off/on scripts 10526 10527commit 6924d1ac900524d0978a02f8f8509faab1809b62 10528Author: Maarten Maathuis <madman2003@gmail.com> 10529Date: Thu Mar 20 20:20:16 2008 +0100 10530 10531 NV50: Although not essential, it's nicer to save a register as uint32_t. 10532 10533commit 1042c2bed3aee8561411fba85438c3fe5ee2cdea 10534Author: Maarten Maathuis <madman2003@gmail.com> 10535Date: Thu Mar 20 20:10:13 2008 +0100 10536 10537 NV50: Some more stuff. 10538 10539commit b58e5e99aa06e1851ce255211d524a88f1fb297a 10540Author: Maarten Maathuis <madman2003@gmail.com> 10541Date: Thu Mar 20 19:59:46 2008 +0100 10542 10543 NV50: Educated guesses at some of the control registers. 10544 10545commit ef25cb9224d37c7937e67824a262937a747c91cc 10546Author: Maarten Maathuis <madman2003@gmail.com> 10547Date: Thu Mar 20 18:49:58 2008 +0100 10548 10549 NV50: Forgot some more stuff. 10550 10551commit 64b3d1ab1a4226fa688722dabc40fe05a0e30d52 10552Author: Maarten Maathuis <madman2003@gmail.com> 10553Date: Thu Mar 20 18:35:20 2008 +0100 10554 10555 NV50: Forgot something. 10556 10557commit 81929c1810fe4825ff9bd5d4bc84996c35c8ceb7 10558Author: Maarten Maathuis <madman2003@gmail.com> 10559Date: Thu Mar 20 17:43:30 2008 +0100 10560 10561 NV50: More dehexing. 10562 10563commit ee1ac0646c7c2f018abdfd4f58d49d51f5340784 10564Author: Maarten Maathuis <madman2003@gmail.com> 10565Date: Thu Mar 20 17:19:37 2008 +0100 10566 10567 NV50: Some more dehexing. 10568 10569commit ca1b8bbebd4b786f0fe8531ba7145aa932e5abcc 10570Author: Maarten Maathuis <madman2003@gmail.com> 10571Date: Thu Mar 20 14:18:18 2008 +0100 10572 10573 NV50: Some more register guesses. 10574 10575commit aa99fb07cd438f436fdfa4e5dff6d87d27950358 10576Author: Stuart Bennett <sb476@cam.ac.uk> 10577Date: Wed Mar 19 18:16:03 2008 +0000 10578 10579 randr12: do DPMS off better 10580 10581 Half the time that DPMS off is done, there is no crtc bound, which meant outputs never got turned off. 10582 It seems xf86DisableUnusedFunctions is always run *before* setting up the new outputs (when using xrandr): 10583 if this is not the case, this commit may not be quite right. 10584 10585commit 063b393ac12bec286e0b28c2b18a98cedd3422c1 10586Author: Stuart Bennett <sb476@cam.ac.uk> 10587Date: Wed Mar 19 02:56:31 2008 +0000 10588 10589 Optimizations on LVDS script code 10590 10591commit c5203647ddf262978e7d6a4912661a9cc448da66 10592Author: Stuart Bennett <sb476@cam.ac.uk> 10593Date: Wed Mar 19 23:12:59 2008 +0000 10594 10595 randr12: unbreak LVDS and primary I2C for < NV50 10596 10597 NV50 check disallowed pre-NV50 cards using I2C on first head, and made garbage get written to CR0 on LVDS 10598 10599 This should work for both pre-nv50 and nv50. 10600 10601commit ebc75dc9e8394ee1d13a853341ba9995e9d9d04f 10602Author: Maarten Maathuis <madman2003@gmail.com> 10603Date: Wed Mar 19 21:31:10 2008 +0100 10604 10605 NV50: fix i2c for real 10606 10607commit c6e3a6f57a7ae67c187aff0ad87153e988961784 10608Author: Maarten Maathuis <madman2003@gmail.com> 10609Date: Wed Mar 19 08:57:33 2008 +0100 10610 10611 NV50: Fix a typo and hopefully fix i2c. 10612 10613commit a989b4caac077f65034fb328a80ab4ae758c1162 10614Author: Maarten Maathuis <madman2003@gmail.com> 10615Date: Mon Mar 17 23:40:16 2008 +0100 10616 10617 NV50: Fix major thinko in "NV50: kill output_resource and use or directly." 10618 10619commit cd5aaeaa319e2b681c7c6ccc135d76b7c027bd0d 10620Author: Maarten Maathuis <madman2003@gmail.com> 10621Date: Mon Mar 17 22:05:12 2008 +0100 10622 10623 NV50: Handle some things that weren't fixed during rebase. 10624 10625commit f73c2361a7bba5aa9abb4fe65697bfcb2a407e99 10626Author: Maarten Maathuis <madman2003@gmail.com> 10627Date: Sun Mar 16 16:09:11 2008 +0100 10628 10629 NV50: kill output_resource and use or directly. 10630 10631commit de0fc0ece55997f8b2ad86070cb940e2afd2d7e8 10632Author: Maarten Maathuis <madman2003@gmail.com> 10633Date: Sun Mar 16 12:46:40 2008 +0100 10634 10635 NV50: Mostly resorting regs. 10636 10637 - Also a dithering register define and rename cursor related reg. 10638 10639commit 849bab73da8da8ad8b2987f80fcf90bf86299171 10640Author: Maarten Maathuis <madman2003@gmail.com> 10641Date: Sun Mar 16 12:33:19 2008 +0100 10642 10643 NV50: Some more educated guesses at register names. 10644 10645commit 0244dc1dd0698be161d7fd629efa10dcb3df106a 10646Author: Maarten Maathuis <madman2003@gmail.com> 10647Date: Sun Mar 16 01:09:48 2008 +0100 10648 10649 NV50: Another subtle difference. 10650 10651commit 5e2d220edf65357f4d7bd8ce97bd829086e0781c 10652Author: Maarten Maathuis <madman2003@gmail.com> 10653Date: Sun Mar 16 01:05:03 2008 +0100 10654 10655 NV50: Fix a potentional bug. 10656 10657commit cef158cee51ecc53c2f5a9910a88c9e802d49980 10658Author: Maarten Maathuis <madman2003@gmail.com> 10659Date: Sun Mar 16 00:40:59 2008 +0100 10660 10661 NV50: Share output properties. 10662 10663commit 31e7a740a1cddc5edfdede05187e6bd9a6f1b993 10664Author: Maarten Maathuis <madman2003@gmail.com> 10665Date: Sun Mar 16 00:27:41 2008 +0100 10666 10667 randr12: Implement dithering output property. 10668 10669commit 05f0464669eef0a3caff2466eeffd9843c18a7b7 10670Author: Maarten Maathuis <madman2003@gmail.com> 10671Date: Sat Mar 15 19:02:55 2008 +0100 10672 10673 NV50: Merge output creation and i2c init. 10674 10675commit 285a13ab41e869544681260b1f2415143456bdc9 10676Author: Maarten Maathuis <madman2003@gmail.com> 10677Date: Sat Mar 15 18:20:34 2008 +0100 10678 10679 NV50: Merge crtc creation. 10680 10681commit c5dac45a199b7f4fa3692727aa6af1cd9a6b8d35 10682Author: Stuart Bennett <sb476@cam.ac.uk> 10683Date: Mon Mar 17 23:20:39 2008 +0000 10684 10685 randr12 offbyone debug commitnoise 10686 10687 semantic version control messages ftw? 10688 10689commit 4cd5e430d164cb5a834e4946d5e5dc79061cadfb 10690Author: Stuart Bennett <sb476@cam.ac.uk> 10691Date: Mon Mar 17 17:21:28 2008 +0000 10692 10693 randr12: remove NVGetOutputFromCRTC 10694 10695 This function had a major problem for the long term: assuming a maximum of 1 10696 output per CRTC, which is simply not valid 10697 10698commit 27c21c0574bac47bbb311094c32f17e7ac321ba5 10699Author: Stuart Bennett <sb476@cam.ac.uk> 10700Date: Mon Mar 17 17:09:23 2008 +0000 10701 10702 randr12: split out fp-only registers 10703 10704commit c09d190ffe2bab1e1df52d8b83009c10fafbc2bc 10705Author: Stuart Bennett <sb476@cam.ac.uk> 10706Date: Mon Mar 17 03:03:14 2008 +0000 10707 10708 randr12: better handling for output types in nv_crtc 10709 10710commit 78ded96a4e8f88dba038e5ef0c452b7c7bf9a2dc 10711Author: Stuart Bennett <sb476@cam.ac.uk> 10712Date: Mon Mar 17 17:16:27 2008 +0000 10713 10714 randr12: remove duplicated regs 10715 10716commit f7013760f9e9ec91bc3bd3739b6cce9018e34bcb 10717Author: Stuart Bennett <sb476@cam.ac.uk> 10718Date: Mon Mar 17 00:59:31 2008 +0000 10719 10720 randr12: fix noscale scaling mode 10721 10722 Also, a bit of consistency missed in the earlier commit 10723 10724commit a9c01eff4bcf3af52faf73269f63a5fd46b3b3b3 10725Author: Stuart Bennett <sb476@cam.ac.uk> 10726Date: Sun Mar 16 14:27:14 2008 +0000 10727 10728 Misc startup info mostly duplicated the state storage stuff 10729 10730commit a02f82e9ce79ae24a4f2eba71a83e51f3b6d9e52 10731Author: Stuart Bennett <sb476@cam.ac.uk> 10732Date: Sun Mar 16 00:51:09 2008 +0000 10733 10734 randr12: make output parsing consistent 10735 10736commit f04c44eff67adb9099eca7b298b10639dc0b1759 10737Author: Stuart Bennett <sb476@cam.ac.uk> 10738Date: Sat Mar 15 01:36:13 2008 +0000 10739 10740 G80 I2C parsing 10741 10742 Changed to on-demand parsing 10743 10744commit 718e42478c02c7497cd3917b8ea78ed692f30f3f 10745Author: Stuart Bennett <sb476@cam.ac.uk> 10746Date: Fri Mar 14 21:22:48 2008 +0000 10747 10748 randr12: make nv_crtc_calc_state_ext definition smaller 10749 10750commit 772bf76b3697d1d1725017787378802c2940093e 10751Author: Stuart Bennett <sb476@cam.ac.uk> 10752Date: Tue Mar 11 03:39:43 2008 +0000 10753 10754 randr12: merge common PLL paths 10755 10756commit 1326e0176582fd21e16359764e2792cd125548e6 10757Author: Stuart Bennett <sb476@cam.ac.uk> 10758Date: Tue Mar 11 02:19:10 2008 +0000 10759 10760 randr12: make plls belong to their CRTCs 10761 10762commit 97b074889feef7b4a3b6401825a46d53321d1c7e 10763Author: Stuart Bennett <sb476@cam.ac.uk> 10764Date: Wed Jan 30 14:38:59 2008 +0000 10765 10766 Use PMC defines 10767 10768commit ee11c6abe875a758bdfa4469ae6326ac25732178 10769Author: Maarten Maathuis <madman2003@gmail.com> 10770Date: Fri Mar 14 11:11:07 2008 +0100 10771 10772 randr12: Let an unneeded quirk die. 10773 10774commit 8851d9f3ef4a8cb9b35e1cb8ed80b3ff213eeee0 10775Author: Stuart Bennett <sb476@cam.ac.uk> 10776Date: Wed Mar 12 16:29:17 2008 +0000 10777 10778 randr12: sel_clk values are really an FP thing, so move calculation to nv_output 10779 10780commit 63527f978d923317a97dec060be6df16c7baf17d 10781Author: Stuart Bennett <sb476@cam.ac.uk> 10782Date: Thu Mar 13 20:26:07 2008 +0000 10783 10784 randr12: tidy up output restore 10785 10786commit 769436b3840e10da00b51b94d72d0b2ef510fecc 10787Author: Stuart Bennett <sb476@cam.ac.uk> 10788Date: Thu Mar 13 19:52:52 2008 +0000 10789 10790 randr12: common tmds access functions 10791 10792commit c455ca0d3df3bb2322154ded40fbe6b86c6a8acd 10793Author: Stuart Bennett <sb476@cam.ac.uk> 10794Date: Wed Mar 12 03:23:21 2008 +0000 10795 10796 Improve bios' handling of ramdac_580 10797 10798commit bd5ad7c4e1f824d57517da540658c97b5283d5d0 10799Author: Maarten Maathuis <madman2003@gmail.com> 10800Date: Thu Mar 13 20:49:53 2008 +0100 10801 10802 NV50: Accept G9X cards. 10803 10804commit 0318b31e19da12ba0d4dc456ad08d6b059d08d1e 10805Author: Maarten Maathuis <madman2003@gmail.com> 10806Date: Thu Mar 13 16:57:24 2008 +0100 10807 10808 NV50: Add back output_resource. 10809 10810commit a4397563c6845668ad1f4b2352b90c872957f8bc 10811Author: Maarten Maathuis <madman2003@gmail.com> 10812Date: Wed Mar 12 23:16:53 2008 +0100 10813 10814 NV50: Unbreak "NV50: Kill the connection status caching (which was broken btw)." 10815 10816commit 63ffb8dce809267706eb621f03da3c3501935bbd 10817Author: Maarten Maathuis <madman2003@gmail.com> 10818Date: Tue Mar 11 23:15:11 2008 +0100 10819 10820 NV50: Merge the two output private structs. 10821 10822commit b4c3bee2f8cc824cd52aca61121634c55d5f5a60 10823Author: Maarten Maathuis <madman2003@gmail.com> 10824Date: Tue Mar 11 23:09:19 2008 +0100 10825 10826 NV50: rename scale to scaling_mode 10827 10828commit b0d2f93d3adbb798f5d9bfa2d188b666e4c5fc16 10829Author: Maarten Maathuis <madman2003@gmail.com> 10830Date: Tue Mar 11 23:06:16 2008 +0100 10831 10832 NV50: Rename nativeMode 10833 10834commit e5b8efad35df21d35a3893a6c8b10e1f02272544 10835Author: Maarten Maathuis <madman2003@gmail.com> 10836Date: Tue Mar 11 23:00:19 2008 +0100 10837 10838 NV50: Rename or to output_resource. 10839 10840commit 19ca2db7683b8cdcfb74f6cc14c9cb6c3876d2c7 10841Author: Maarten Maathuis <madman2003@gmail.com> 10842Date: Tue Mar 11 22:43:33 2008 +0100 10843 10844 NV50: kill set_pclk 10845 10846commit eef3a665c01c530238c3dc02324573d6bcf6c3f2 10847Author: Maarten Maathuis <madman2003@gmail.com> 10848Date: Tue Mar 11 22:24:46 2008 +0100 10849 10850 NV50: Remove the remnants of cached_status. 10851 10852commit dc0bb590a40f3d8b18d2136ed906e8dac10ec598 10853Author: Maarten Maathuis <madman2003@gmail.com> 10854Date: Tue Mar 11 22:22:12 2008 +0100 10855 10856 NV50: kill partner entry 10857 10858commit f66463713a71b0e3090f33c68186884594255cc0 10859Author: Maarten Maathuis <madman2003@gmail.com> 10860Date: Tue Mar 11 22:12:57 2008 +0100 10861 10862 NV50: fix warning 10863 10864commit cb50cd3276825569a6da1f25a2d89bfd417b5034 10865Author: Maarten Maathuis <madman2003@gmail.com> 10866Date: Tue Mar 11 21:33:29 2008 +0100 10867 10868 NV50: Kill the connection status caching (which was broken btw). 10869 10870commit 6e2af71c73263c3cf2d61f66c02363decd354da5 10871Author: Maarten Maathuis <madman2003@gmail.com> 10872Date: Tue Mar 11 21:15:21 2008 +0100 10873 10874 NV50: change i2c to pDDCBus 10875 10876commit d07b3063953c0b17585249dfbc2938a433bea88f 10877Author: Maarten Maathuis <madman2003@gmail.com> 10878Date: Tue Mar 11 19:50:25 2008 +0100 10879 10880 NV50: Switch to other scaling defines. 10881 10882commit 4ddb92b8e936985b5b1429a0a21890879635bcde 10883Author: Maarten Maathuis <madman2003@gmail.com> 10884Date: Tue Mar 11 19:29:11 2008 +0100 10885 10886 NV50: Switch to the same output type as the rest of the code. 10887 10888commit e605fbfc659d3fe1bd71d478c21429b563fbfd80 10889Author: Stuart Bennett <sb476@cam.ac.uk> 10890Date: Thu Mar 13 15:21:49 2008 +0000 10891 10892 randr12: nothing uses output_resource. kill it 10893 10894commit 53925cfc3540b89dedd317bbae4318cf53daada1 10895Author: Stuart Bennett <sb476@cam.ac.uk> 10896Date: Thu Mar 13 15:31:00 2008 +0000 10897 10898 randr12: fix the setting of NV_RAMDAC_OUTPUT 10899 10900 And turn on/off the DAC for DPMS 10901 10902commit 493c79405be23a543cd21c2583c466342399509d 10903Author: Stuart Bennett <sb476@cam.ac.uk> 10904Date: Thu Mar 13 03:53:18 2008 +0000 10905 10906 Simplify TMDS writes a little 10907 10908commit 4cf1bd9ef016a2c9fe7784fdc83f559bec464f9e 10909Author: Stuart Bennett <sb476@cam.ac.uk> 10910Date: Tue Mar 11 17:20:28 2008 +0000 10911 10912 randr12: don't allow scripts to break overlay 10913 10914commit 95f2b8439f433afb5d637a3580f933e2cd44488a 10915Author: Danny Tholen <obiwan@mailmij.org> 10916Date: Tue Mar 11 17:14:54 2008 +0000 10917 10918 Fix video overlays to treat double-scan correctly when using randr12 10919 10920commit 460cb260c30467799fa31aef11946d1fe2dd6523 10921Author: Stuart Bennett <sb476@cam.ac.uk> 10922Date: Mon Mar 10 23:47:43 2008 +0000 10923 10924 Correct RAMIN offset 10925 10926 and some bits missed in previous nv_setup tidying 10927 10928commit e8a5d4638e74b5957fc12b05865f49dd995d0435 10929Author: Stuart Bennett <sb476@cam.ac.uk> 10930Date: Mon Mar 10 21:46:38 2008 +0000 10931 10932 randr12: changing tmds2 for dvi-d seemingly unnecessary 10933 10934commit 30ddbe9c2b82bc19d1480a1b38df6b5ca0dd057b 10935Author: Stuart Bennett <sb476@cam.ac.uk> 10936Date: Mon Mar 10 23:49:08 2008 +0000 10937 10938 randr12: unbreak sel_clk changes 10939 10940commit 03f38725304842b053b7976db5df97e61159d07d 10941Author: Stuart Bennett <sb476@cam.ac.uk> 10942Date: Mon Mar 10 19:50:55 2008 +0000 10943 10944 randr12: allow spread spectrum bits to be updated on pre nv40 10945 10946commit 323b7ef9c1c26996b84fd1f67297b2958f68fcc9 10947Author: Stuart Bennett <sb476@cam.ac.uk> 10948Date: Mon Mar 10 01:18:56 2008 +0000 10949 10950 randr12: quirk for 17" powerbook, and better sel_clk handling 10951 10952 We set sel_clk *before* running the output script, so make sure the set pll->head binding is not altered 10953 10954commit 58f9079cda17bb6a76ba16c14ddc56bd4ac640fc 10955Author: Stuart Bennett <sb476@cam.ac.uk> 10956Date: Sun Mar 9 16:08:46 2008 +0000 10957 10958 nv_type.h mandates XF86DRI being declared 10959 10960commit 0dbffc868a7b60aa99ca550d33b8488ac9de8d00 10961Author: Stuart Bennett <sb476@cam.ac.uk> 10962Date: Sun Mar 9 03:59:57 2008 +0000 10963 10964 Fix up a bunch of prototypes, remove unused wrappers 10965 10966commit e5ded02925e7d39c67e69d0389778ac795bfe743 10967Author: Stuart Bennett <sb476@cam.ac.uk> 10968Date: Sun Mar 9 01:31:47 2008 +0000 10969 10970 Move setting of alphaCursor to nv_setup, and restyle useful portions of nv_setup 10971 10972commit 0b8629a9a29203553397eb764b9b9c87d38d4b9f 10973Author: Stuart Bennett <sb476@cam.ac.uk> 10974Date: Sat Mar 8 22:05:52 2008 +0000 10975 10976 Kill NVFBLayout 10977 10978 Using pScrn->currentMode is wrong for randr12, but it's no worse than what it was 10979 10980commit 29cfb08c1b75e050aa58fa6019f7d9be063f8555 10981Author: Stuart Bennett <sb476@cam.ac.uk> 10982Date: Sat Mar 8 21:02:35 2008 +0000 10983 10984 Warning fixes 10985 10986 And remove a few unused members from nv_type.h 10987 10988commit b87059fcccfac693df0b2dc5ba85cb358ad1eba6 10989Author: Stuart Bennett <sb476@cam.ac.uk> 10990Date: Sat Mar 8 20:53:20 2008 +0000 10991 10992 Drop the NV*VGA defines and adjust users 10993 10994commit a636b02eba03454b8b757f69e6a62aae2c3fd571 10995Author: Maarten Maathuis <madman2003@gmail.com> 10996Date: Sat Mar 8 21:18:34 2008 +0100 10997 10998 randr12: Disable a quirk that has adverse sideeffects on NV36M. 10999 11000 - NV31M owners, please test if you have: 11001 - LVDS on CRTC 1 and or = 3. 11002 - VGA on CRTC 0 and or = 1. 11003 - Testing requires enabling both lvds and vga. 11004 11005commit f4228861abec9df310efb863833c82022136e464 11006Author: Maarten Maathuis <madman2003@gmail.com> 11007Date: Sat Mar 8 18:23:58 2008 +0100 11008 11009 Merge CRTC private structs. 11010 11011commit b3d0d0c01664e874f3b69b36383ad1a7845d7d91 11012Author: Maarten Maathuis <madman2003@gmail.com> 11013Date: Sat Mar 8 18:10:53 2008 +0100 11014 11015 Death to some unnecesary stuff. 11016 11017commit 03a7a1d7065b55f16385fd16e7a2eff938d59cea 11018Author: Stuart Bennett <sb476@cam.ac.uk> 11019Date: Sat Mar 8 01:56:36 2008 +0000 11020 11021 Various minor tidyups 11022 11023commit 624148ca7067f748c142db3cd4d31314aa31ee0f 11024Author: Maarten Maathuis <madman2003@gmail.com> 11025Date: Fri Mar 7 20:59:30 2008 +0100 11026 11027 NV50: Convert back to absolute registers. 11028 11029 - Someone with a known to work setup should test this, as i have no NV5x card. 11030 11031commit 10bdffb65e2d8f59399556fe3224cbf8e8a171a9 11032Author: Stuart Bennett <sb476@cam.ac.uk> 11033Date: Fri Mar 7 19:08:50 2008 +0000 11034 11035 PLL limit table v0x11 11036 11037commit 2e1af24826e7db0e3f53440a2c5912789ade6c97 11038Author: Stuart Bennett <sb476@cam.ac.uk> 11039Date: Fri Mar 7 18:43:16 2008 +0000 11040 11041 Make bios init code work on 6600 by changing pll setting stuff a bit 11042 11043 I suspect we're going to need some card specific checks in here. ugh. 11044 11045commit 5fd2dd7b4c6e15d83cb7e7daafb6ee801fa823e8 11046Author: Maarten Maathuis <madman2003@gmail.com> 11047Date: Fri Mar 7 07:52:09 2008 +0100 11048 11049 It seems libpciaccess release became version 0.10 not version 0.10.0 11050 11051commit 3e4db03a39770e5fa94ca094deb690cba8175196 11052Author: Maarten Maathuis <madman2003@gmail.com> 11053Date: Fri Mar 7 07:36:16 2008 +0100 11054 11055 randr12: Forgot to clean up when switching output resources. 11056 11057commit 8a5c6e8b4d00871b17b7d564d765ea677596f79a 11058Author: Jaime Velasco Juan <jsagarribay@gmail.com> 11059Date: Sat Feb 16 15:01:23 2008 +0000 11060 11061 Do not access beyond source pixmap in NVAccelUploadIFC 11062 11063commit 7afb739418fbac5058d13d05188a69b142529fb4 11064Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11065Date: Thu Mar 6 23:40:08 2008 +0100 11066 11067 Oops. 11068 11069commit 0da8c84cceb178b04ab535edb4e3f0ced204d00a 11070Author: Stuart Bennett <sb476@cam.ac.uk> 11071Date: Thu Mar 6 22:35:30 2008 +0000 11072 11073 randr12: workaround for xserver randr bug, where the hwcursor always gets displayed after a mode change 11074 11075commit c424144e854563709a01ed2d30a0482edc57fbf3 11076Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11077Date: Thu Mar 6 23:15:31 2008 +0100 11078 11079 Fix my nv10 cursor. 11080 11081commit 633047c65708f1a39e371c2c5da5c111027af1af 11082Author: Maarten Maathuis <madman2003@gmail.com> 11083Date: Thu Mar 6 23:05:43 2008 +0100 11084 11085 randr12: A major rework of output resource conflict handling. 11086 11087 - This mostly affects laptops, such as some geforce5 and the famous 7300go laptops. 11088 - Test this if you know to have a strange vga/dvi port of your laptop. 11089 11090commit c5230961d181f15cd965c4569813219d9292681a 11091Author: Maarten Maathuis <madman2003@gmail.com> 11092Date: Thu Mar 6 19:29:50 2008 +0100 11093 11094 randr12: Readd some useful information about how an output is routed. 11095 11096commit 97ac866a4e433c70d0de7a1e8fb66a3994194ffd 11097Author: Maarten Maathuis <madman2003@gmail.com> 11098Date: Thu Mar 6 19:16:36 2008 +0100 11099 11100 randr12: Kill almost all usage of ErrorF + some misc cleanup. 11101 11102commit ea8fba29bcfe8158c9bf17754a20684430027aee 11103Author: Stuart Bennett <sb476@cam.ac.uk> 11104Date: Wed Mar 5 18:53:54 2008 +0000 11105 11106 randr12: don't reset panel unnecessarily for mere backlight control 11107 11108commit 22ec35504fbef83ba5790c87d27cd13e936fccb0 11109Author: Stuart Bennett <sb476@cam.ac.uk> 11110Date: Wed Mar 5 14:40:08 2008 +0000 11111 11112 randr12: remove switchable_crtc stuff 11113 11114commit 3b0a1a11f662bfa705806f4afc5898338234ca65 11115Author: Stuart Bennett <sb476@cam.ac.uk> 11116Date: Wed Mar 5 14:35:41 2008 +0000 11117 11118 randr12: if no ddc, return XF86OutputStatusUnknown, for !twoHeads and nv11 11119 11120commit 424500cf828903e3449d4bca22a083087258cc99 11121Author: Stuart Bennett <sb476@cam.ac.uk> 11122Date: Wed Mar 5 02:15:28 2008 +0000 11123 11124 randr12: do not change CR43 from its initial value 11125 11126commit a764d3ca2741243f49f34e9a4930a59272927902 11127Author: Maarten Maathuis <madman2003@gmail.com> 11128Date: Wed Mar 5 09:44:46 2008 +0100 11129 11130 Give some defines more consistent names. 11131 11132commit 234ce16558e59dff302f13ea3d8eb48f46eedf0d 11133Author: Stuart Bennett <sb476@cam.ac.uk> 11134Date: Wed Mar 5 01:25:20 2008 +0000 11135 11136 Use the BMP structure's I2C info, instead of DCB 1.2's, as marcheu's bios is broken 11137 11138commit bb5429f08b3ffd16d6786bd65d6a4c8db1210ade 11139Author: Stuart Bennett <sb476@cam.ac.uk> 11140Date: Wed Mar 5 00:38:15 2008 +0000 11141 11142 Add another DCB 1.4 entry 11143 11144commit 5998ba0a35bc1c897916f66b0bcfb9a3e0a7c54d 11145Author: Stuart Bennett <sb476@cam.ac.uk> 11146Date: Wed Mar 5 00:34:16 2008 +0000 11147 11148 Don't zero out i2c_read and i2c_write 11149 11150 pNv is calloc'd, so zero inited, and explicit zeroing overwrites the stuff parsed in parse_bmp_structure. Duh. 11151 11152commit 2d74bcb4602bb0cdae18ed1ffdd17520a3b57a78 11153Author: Maarten Maathuis <madman2003@gmail.com> 11154Date: Tue Mar 4 21:18:49 2008 +0100 11155 11156 Put braces around a bitmask. 11157 11158commit c7ab8e84bf8ad8ee6106b21a1c07b6d8470e7f11 11159Author: Stuart Bennett <sb476@cam.ac.uk> 11160Date: Tue Mar 4 13:34:16 2008 +0000 11161 11162 Add an LVDS output setup quirk for Powerbooks 11163 11164 Again, as a result of much testing by moondrake and sbriglie. 11165 Hopefully it even works this time :) 11166 11167commit fd9b37b31896e1731bdd56ec847c86c0502ba0f8 11168Author: Maarten Maathuis <madman2003@gmail.com> 11169Date: Mon Mar 3 19:04:09 2008 +0100 11170 11171 randr12: Only change vpll value when needed. 11172 11173commit 89882b416de3528cf504c989cc147a86d3572053 11174Author: Maarten Maathuis <madman2003@gmail.com> 11175Date: Sun Mar 2 20:25:45 2008 +0100 11176 11177 randr12: Keep the gamma values when setting a new mode, instead of always resetting to default. 11178 11179commit e8c363b7ca4a416be37a1ea1c97bbd99b21d5be4 11180Author: Stuart Bennett <sb476@cam.ac.uk> 11181Date: Sat Mar 1 15:05:09 2008 +0000 11182 11183 Improve link_head_and_output 11184 11185 Thanks to moondrake and sbriglie for their work on this 11186 11187commit 4a16414c77ad77e7612f62f4ca1859e0d6a24cc2 11188Author: Patrice Mandin <pmandin@caramail.com> 11189Date: Fri Feb 29 22:17:30 2008 +0100 11190 11191 nv30: Remove fix that is useless now, and maybe broken in some cases 11192 11193commit 5cb164876b824e8bf52aa44ccf4a797bab310602 11194Author: Stuart Bennett <sb476@cam.ac.uk> 11195Date: Thu Feb 28 00:43:17 2008 +0000 11196 11197 Less magical way of setting dual link on EDID equipped panels 11198 11199 Eldritch coupling ftl 11200 11201commit 1127974ed8517f28ef9c451d81bf1c61a2b0e6ae 11202Author: Maarten Maathuis <madman2003@gmail.com> 11203Date: Wed Feb 27 23:45:34 2008 +0100 11204 11205 NV40EXA: Support extend/repeat type pad. 11206 11207 - Note that all xservers, up to about a minute ago, did not try to accelerate repeat type reflect or pad. 11208 11209commit 6980d0e24a9dae52903e6971c5c8a18240a7dbc1 11210Author: Stuart Bennett <sb476@cam.ac.uk> 11211Date: Wed Feb 27 12:55:54 2008 +0000 11212 11213 Add default dual link transition frequency 11214 11215 For card bios versions before the relevant field was added 11216 Should fix Apple laptops. 11217 11218commit 27e2e55698ab05bf3469e0f68474d4c3f021ca31 11219Author: Stuart Bennett <sb476@cam.ac.uk> 11220Date: Tue Feb 26 18:07:39 2008 +0000 11221 11222 Deal with pre-nv40 laptops with EDID 11223 11224commit 41992eb90828976105cbc8bba87fe7d079d7c06a 11225Author: Patrice Mandin <pmandin@caramail.com> 11226Date: Fri Feb 22 17:29:12 2008 +0100 11227 11228 Revert dummy commit 11229 11230commit 073d852f36df4a18563ea892e89eca269efd55ef 11231Author: Patrice Mandin <pmandin@caramail.com> 11232Date: Fri Feb 22 17:27:51 2008 +0100 11233 11234 Dummy commit 11235 11236commit 5db7920763aee3a1b124139273566f7add5e27df 11237Author: Stuart Bennett <sb476@cam.ac.uk> 11238Date: Sun Feb 10 23:14:52 2008 +0000 11239 11240 Use some of nv_hw's functions in nv_bios.c 11241 11242commit fa8891e78f37edc02537704172156ba77371a307 11243Author: Stuart Bennett <sb476@cam.ac.uk> 11244Date: Wed Feb 20 00:50:48 2008 +0000 11245 11246 Unbreak console restore again, and use a less ambiguous function name for head locking 11247 11248commit 38d42a7857480a930e2025e4f63aa7f5defca5c5 11249Author: Stuart Bennett <sb476@cam.ac.uk> 11250Date: Tue Feb 19 23:59:12 2008 +0000 11251 11252 randr12: rename NV*VGA functions for api consistency 11253 11254 Also nuke NVCrtcSetOwner wrapper 11255 11256commit c77a54768c3fed5939755f23c5c067bb86f49805 11257Author: Ben Skeggs <darktama@beleth.(none)> 11258Date: Thu Feb 21 00:55:39 2008 +1100 11259 11260 xv: don't crash on init 11261 11262commit cf963a128374bf7c2070e8a380e236b5e083adc1 11263Author: Maarten Maathuis <madman2003@gmail.com> 11264Date: Wed Feb 20 09:36:42 2008 +0100 11265 11266 randr12: Only NV11 needs byteswapping + NV11 uses premultiplied cursors i think. 11267 11268commit 87f1676a243cddcaece9eeb0487a85b7e74b0f13 11269Author: Maarten Maathuis <madman2003@gmail.com> 11270Date: Wed Feb 20 00:14:12 2008 +0100 11271 11272 NV30/40TEX: Create two adapters, one with fancy filtering, the default without. 11273 11274 - Also enable sync to vblank by default, strange that it was off. 11275 - Remove the GET_TEXTURE(D)_PRIVATE macros, as one of them pointed to the blitter and one is now ambigious. 11276 11277commit 7ee1bb47d86396cc1a0f7879a98d04588cb31eee 11278Author: Stuart Bennett <sb476@cam.ac.uk> 11279Date: Thu Feb 14 23:16:01 2008 +0000 11280 11281 randr12: make more hw funcs generic 11282 11283commit 11de8d7a6f9a2dfd5a4f93f031865e284268a0f3 11284Author: Stuart Bennett <sb476@cam.ac.uk> 11285Date: Tue Feb 19 21:44:56 2008 +0000 11286 11287 Use fp strapping to determine whether straps are used for flat panel modes 11288 11289commit c1ac6b977a4756e2dbfdb75ab5b53c7c10e76cf4 11290Author: Stuart Bennett <sb476@cam.ac.uk> 11291Date: Tue Feb 19 14:53:41 2008 +0000 11292 11293 Rehabilitate bios opcode 6d 11294 11295commit afb42307cb2810ca2d759df62240fbf1f0cc7f7c 11296Author: Stuart Bennett <sb476@cam.ac.uk> 11297Date: Tue Feb 19 14:23:06 2008 +0000 11298 11299 Spruce up NV04 bios parsing a little 11300 11301commit dfabaab48034bf36c19510fe5de0f1bdaa2157ef 11302Author: Stuart Bennett <sb476@cam.ac.uk> 11303Date: Tue Feb 19 00:37:27 2008 +0000 11304 11305 More DCB v1.5 entries 11306 11307 Courtesy HenneNWH's NV11 11308 11309commit 2a4373d4f3b092600651b7dc8e2f9f9b520c7e0d 11310Author: Stuart Bennett <sb476@cam.ac.uk> 11311Date: Mon Feb 18 17:32:43 2008 +0000 11312 11313 Treat 0x4 and 0x8 bits in LVDS DCB v2.0 configuration as meaning the same thing 11314 11315commit daa557c0dadb3ad5caf39a4e90d3ebd16b12fc96 11316Author: Danny Tholen <obiwan@mailmij.org> 11317Date: Mon Feb 18 05:00:03 2008 +0100 11318 11319 Fix PPC DFS/UTS by using PCIgart. 11320 11321commit f8f469286fb3acc1ee694c61fe21e42123a5456a 11322Author: Ben Skeggs <skeggsb@gmail.com> 11323Date: Sat Feb 16 03:41:38 2008 +1100 11324 11325 dri: fix broken display in some cases 11326 11327commit a293f32b2172837b6e2bcc3f5771c5219fccfb14 11328Author: Stuart Bennett <sb476@cam.ac.uk> 11329Date: Wed Feb 13 00:52:34 2008 +0000 11330 11331 Fix bios pll setting regression 11332 11333 Oops. 11334 11335commit 7121fe9dfb1c9162c8be808e09ed6b4b3ce3999e 11336Author: Stuart Bennett <sb476@cam.ac.uk> 11337Date: Tue Feb 12 18:20:43 2008 +0000 11338 11339 Use a loop to read RAMIN bios image, rather than memcpy 11340 11341commit 3ad7c35740905ad267faf8e248075640d5ee9cea 11342Author: Stuart Bennett <sb476@cam.ac.uk> 11343Date: Tue Feb 12 17:23:38 2008 +0000 11344 11345 randr12: reorder some variable declarations in nv_crtc and nv_output 11346 11347commit ab40a46430f369a7bd9cef8db4e86414125c8437 11348Author: Stuart Bennett <sb476@cam.ac.uk> 11349Date: Tue Feb 12 15:19:58 2008 +0000 11350 11351 randr12: use integer math for scaling 11352 11353 No floats in modesetting path 11354 11355commit 4a44468c29c37b5ad4828439f37d023865a9d457 11356Author: Stuart Bennett <sb476@cam.ac.uk> 11357Date: Tue Feb 12 15:03:46 2008 +0000 11358 11359 randr12: tidy up some mode retrieval stuff 11360 11361commit 50d77c4199a3a5d2f3ead8327341e91ac2ec5684 11362Author: Stuart Bennett <sb476@cam.ac.uk> 11363Date: Mon Feb 11 16:27:27 2008 +0000 11364 11365 Trim useless nv_type members 11366 11367 Also, crtc_active isn't used on randr12 path 11368 11369commit 0c1e48a2192555856894d7e2b3025cd342c2f448 11370Author: Stuart Bennett <sb476@cam.ac.uk> 11371Date: Mon Feb 11 01:42:04 2008 +0000 11372 11373 Remove some DDX cruft 11374 11375commit 5a8e246fcab3e349f48d0ce0c4da812a411767f2 11376Author: Stuart Bennett <sb476@cam.ac.uk> 11377Date: Mon Feb 11 00:34:27 2008 +0000 11378 11379 Use mmio access wrappers and reorder some variable declarations in nv_bios.c 11380 11381commit 2ffcd89bfb0383fb32e8ec41f1cd472764b9b99d 11382Author: Stuart Bennett <sb476@cam.ac.uk> 11383Date: Mon Feb 11 00:03:50 2008 +0000 11384 11385 randr12: Use absolute regs for 8 bit regs 11386 11387commit c9b1f020d1c5be069dae19f99b3c26bb126052ae 11388Author: Stuart Bennett <sb476@cam.ac.uk> 11389Date: Fri Feb 8 19:44:37 2008 +0000 11390 11391 Move non-crtc specific hw funcs from nv_crtc to nv_hw 11392 11393commit 6cb8fb6cc63bf55f18b4dde53c393336e817474e 11394Author: Stuart Bennett <sb476@cam.ac.uk> 11395Date: Thu Feb 7 02:04:20 2008 +0000 11396 11397 Ordering mechanism for BIT table parsing, and (hopefully) automagic laptop detection 11398 11399commit 4f690235d70ae249323439380a69d8309a8c9981 11400Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11401Date: Wed Feb 6 22:18:46 2008 +0100 11402 11403 Replace a const with its name. 11404 11405commit 75c5912fffde88bfe2ddd158a446588b76ef44d9 11406Author: Maarten Maathuis <madman2003@gmail.com> 11407Date: Wed Feb 6 19:06:14 2008 +0100 11408 11409 Fix bad debugging in NVWritePVIO(). 11410 11411commit 50da8b8f5b5feb6787ed50f221529730d25b5467 11412Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11413Date: Wed Feb 6 17:44:52 2008 +0100 11414 11415 Fix UV swapping on PPC. 11416 11417commit 4ff11d13be6e8050e0b726dc7af8e4bc11bd9d0d 11418Author: Stuart Bennett <sb476@cam.ac.uk> 11419Date: Mon Feb 4 22:34:21 2008 +0000 11420 11421 Missed out "DEBUG" part of the name in the define 11422 11423commit 181ee3bcecfc778c9f671853a696ab3224ac4747 11424Author: Stuart Bennett <sb476@cam.ac.uk> 11425Date: Mon Feb 4 22:31:02 2008 +0000 11426 11427 Load correct PLL limits for single stage PLL cards with blank PLL tables 11428 11429commit cbcb1070db6fbb4d13ddcc35739d2d3e49f3766d 11430Author: Maarten Maathuis <madman2003@gmail.com> 11431Date: Mon Feb 4 07:47:06 2008 +0100 11432 11433 Once again kill usage of an exa private function. 11434 11435 - Must have creeped back in during all the texture adapter changes recently. 11436 11437commit f14d27d638e75faa5bbda903f65a4d5479222ead 11438Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11439Date: Sun Feb 3 17:15:50 2008 +0100 11440 11441 Remove include guards & add copyright. 11442 11443commit 7522640ee8d342c5ebb62989b051f73e230f5ce7 11444Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11445Date: Sun Feb 3 17:01:50 2008 +0100 11446 11447 Last round of fixes for nv30/40 bicubic Xv. 11448 11449commit 78553ab9a0309c7d27632fe7184f3233cfde6610 11450Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11451Date: Sun Feb 3 06:07:35 2008 +0100 11452 11453 Fix tearing. 11454 11455commit 7240f861a98c5f8d9779a4f3a743b748af775337 11456Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11457Date: Sun Feb 3 05:15:41 2008 +0100 11458 11459 Understood and fixed some NV30 FP_CONTROL bits. 11460 11461commit bb2a7bd3461eec37f179748a9bdb9375b1c3da04 11462Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11463Date: Sat Feb 2 05:54:47 2008 +0100 11464 11465 Fixes to nv30/nv40 textured video. 11466 11467commit c10d30dfefe93a9911d89eaffe3d29d3cff66a96 11468Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11469Date: Thu Jan 31 06:15:02 2008 +0100 11470 11471 Misc textured video fixes. 11472 11473commit cfdfd5a99bb7ac959a1188994db4c75a97de2e34 11474Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11475Date: Thu Jan 31 06:00:59 2008 +0100 11476 11477 Of course, add the nv30 video texture file. 11478 11479commit 970a8e474c666d6e33351f7ca6a3a5273024baf4 11480Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11481Date: Thu Jan 31 05:59:08 2008 +0100 11482 11483 nv30 video texture. Doesn't work with bicubic yet, only bilinear. 11484 11485commit c1eba822f1579637f59933b04e31591e86b975fc 11486Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11487Date: Wed Jan 30 02:38:14 2008 +0100 11488 11489 Remove (void)ing that's not necessary any more. 11490 11491commit 7356ad90b387d01c2bf25aa653c41ca9365c5c20 11492Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11493Date: Wed Jan 30 02:30:39 2008 +0100 11494 11495 Remove unneeded stuff. 11496 11497commit e06dfd2964f9bd7f0855a101f5e6971020dfa6be 11498Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11499Date: Wed Jan 30 02:19:08 2008 +0100 11500 11501 Separate the shaders from the rest of the code. 11502 11503commit a69c50780cacf0d0f2948208cfa7e0558a8c3644 11504Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11505Date: Wed Jan 30 02:18:26 2008 +0100 11506 11507 Macros for the PME range. 11508 11509commit 47c8ed472f37d68f3cc241a58ee5b7b8643e537d 11510Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11511Date: Wed Jan 30 02:17:22 2008 +0100 11512 11513 Add PME range. 11514 11515commit 63f47c9bf59ccb9f1b6adbefd55deaec1435faf7 11516Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11517Date: Tue Jan 29 21:25:28 2008 +0100 11518 11519 Small cleanups. 11520 11521commit c98e9fa5c572deeafd7f605bb29d6c83cd38de9d 11522Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11523Date: Tue Jan 29 21:24:39 2008 +0100 11524 11525 nv40 video texture: first pass at bicubic video. 11526 11527commit 10967a4dac8c3eaf828359dcba7a6e1272561aa3 11528Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11529Date: Mon Jan 28 04:52:01 2008 +0100 11530 11531 Update to newer nouveau_class.h 11532 11533commit 82728b3df45e8a5c0a45462d2ce847a41fd31f07 11534Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11535Date: Sat Jan 26 19:02:50 2008 +0100 11536 11537 Update to latest nouveau_class.h 11538 11539commit 9d9a5acae1deae8c76b512362634066e0c32a817 11540Author: Jaime Velasco Juan <jsagarribay@gmail.com> 11541Date: Sun Feb 3 15:45:32 2008 +0100 11542 11543 Don't FIRE_RING() when in NoAccel mode. 11544 11545commit 11c1ee1cb0ee9c8096ea371d95d188c538ac3fb2 11546Author: Maarten Maathuis <madman2003@gmail.com> 11547Date: Sun Feb 3 15:43:35 2008 +0100 11548 11549 Some minor reindenting. 11550 11551commit 0db347294e09d3d17200ca130e719df2d22f69b6 11552Author: Ben Skeggs <skeggsb@gmail.com> 11553Date: Mon Feb 4 01:19:23 2008 +1100 11554 11555 exa: those limits were a bit conservative on some hw. 11556 11557 AFAIK nv2x up should be able to handle 4096x4096 textures and 11558 render targets. 11559 11560commit eb72e67641215b710868da7bb66b466f6ccbbe25 11561Author: Ben Skeggs <skeggsb@gmail.com> 11562Date: Mon Feb 4 01:07:07 2008 +1100 11563 11564 exa: more accurate coordinate limits 11565 11566commit 8645d2d9554ee46397240fbd9904fd4f905aefb2 11567Author: Maarten Maathuis <madman2003@gmail.com> 11568Date: Sun Feb 3 14:54:41 2008 +0100 11569 11570 Fix NVExaPixmapMap() for NOUVEAU_EXA_PIXMAPS case. 11571 11572commit dd81efd0cc64a070f7489decbac345cc1b4fa9ce 11573Author: Maarten Maathuis <madman2003@gmail.com> 11574Date: Sun Feb 3 12:03:43 2008 +0100 11575 11576 Fix undefined symbol. 11577 11578commit 2a1116cbf61e84b16366076366f4bbd5f550fb9f 11579Author: Maarten Maathuis <madman2003@gmail.com> 11580Date: Sat Feb 2 20:34:57 2008 +0100 11581 11582 Remove a few leftovers to actually make the last commits work. 11583 11584commit 714bea6449aab47a9f2211718183ee84a3c70129 11585Author: Maarten Maathuis <madman2003@gmail.com> 11586Date: Sat Feb 2 19:53:32 2008 +0100 11587 11588 Kill NVShowHideCursor and use a wrapper. 11589 11590commit 741acded643945c5a2da7bef088ed1ff7f5fa85e 11591Author: Maarten Maathuis <madman2003@gmail.com> 11592Date: Sat Feb 2 19:26:13 2008 +0100 11593 11594 Remove nvRead/WriteVGA in favor of nvRead/WriteCurVGA wrappers. 11595 11596commit d429cabd9ae9865a45cc88f43ab6d9d61b3fcf36 11597Author: Maarten Maathuis <madman2003@gmail.com> 11598Date: Sat Feb 2 19:17:32 2008 +0100 11599 11600 Kill NVLockUnlock and make it redirect to NVLockUnlockHead. 11601 11602commit dc2aee0ef69c85aaffdc3c1da5a228642d049a30 11603Author: Maarten Maathuis <madman2003@gmail.com> 11604Date: Sat Feb 2 15:12:00 2008 +0100 11605 11606 More wrapper death. 11607 11608commit a57fd67a2a6ae66af218fedd704a04d855260a2f 11609Author: Maarten Maathuis <madman2003@gmail.com> 11610Date: Sat Feb 2 15:06:02 2008 +0100 11611 11612 Let some wrappers die. 11613 11614commit b9a4eb6f6ea4f8f0747baa50037e33a0584cfd29 11615Author: Maarten Maathuis <madman2003@gmail.com> 11616Date: Sat Feb 2 12:47:32 2008 +0100 11617 11618 Undo last commit and move it to drm. 11619 11620commit 288b56545a8da1e7b5aa64b1093316ca2f28f868 11621Author: Maarten Maathuis <madman2003@gmail.com> 11622Date: Sat Feb 2 12:28:44 2008 +0100 11623 11624 NV40: Fix longstanding issue with nouveau crashing after using blob. 11625 11626 - Took me a while to track down, but for me this is definately it. 11627 - In my case i got the X cross and then it would freeze and i would get a fifo dump. 11628 - Don't be shy, let me know if this works for you too. 11629 11630commit 0ec1b10a82c411181ff7a8643bb2390cfe08e4e4 11631Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11632Date: Sat Feb 2 04:04:58 2008 +0100 11633 11634 Don't allocate the whole AGP gart as a scratch buffer, since we still need to allocate the fifo afterwards. 11635 11636commit 5b79b0a9e54fe99eaae5b411f46a601b6944b773 11637Author: Maarten Maathuis <madman2003@gmail.com> 11638Date: Fri Feb 1 23:34:32 2008 +0100 11639 11640 Don't use functions that are not supposed to be used in randr12 mode. 11641 11642 - Also add NVLockUnlockHead() and NVSetOwner() functions. 11643 11644commit cedb604ee79cd73a9ec3940c7b056109811ba670 11645Author: Maarten Maathuis <madman2003@gmail.com> 11646Date: Fri Feb 1 22:41:09 2008 +0100 11647 11648 Remove usage of "illegal" wrappers in code that also used for randr12 mode. 11649 11650 - I've got something prettier in mind to clean this up, which will likely follow. 11651 11652commit 1a5247856929877442e1ec00f190c6d497ca64fe 11653Author: Maarten Maathuis <madman2003@gmail.com> 11654Date: Fri Feb 1 19:14:38 2008 +0100 11655 11656 randr12: Remove a hack, confirmed on 7600GS that it is unneeded. 11657 11658 - This was one of the cards that did need it in the past. 11659 11660commit b6972e7bec0455250851c06d142336cb53be1a74 11661Author: Maarten Maathuis <madman2003@gmail.com> 11662Date: Fri Feb 1 14:47:32 2008 +0100 11663 11664 randr12: Some minor fixes. 11665 11666commit 988dce0f452f6b80b2d02e1a0f78f826674cc3c0 11667Author: Arthur Huillet <arthur.huillet@free.fr> 11668Date: Fri Feb 1 10:04:32 2008 +0100 11669 11670 NV10 EXA: limit regcomb state re-emission, fix logic for A8 + A8 yielding a 3% perf improvement :p 11671 11672commit a84852c2e0b19cf848495dc7e4bde4426b7bd199 11673Author: Maarten Maathuis <madman2003@gmail.com> 11674Date: Wed Jan 30 22:57:09 2008 +0100 11675 11676 (hopefully) Fix build on powerpc. 11677 11678commit e8a975e6a8afe5ae9c77d20f4d59d3d10221f7d9 11679Author: Arthur Huillet <arthur.huillet@free.fr> 11680Date: Wed Jan 30 18:37:39 2008 +0100 11681 11682 Xv: NV30 overlay can not scale down by more than 0.5x 11683 11684commit adc26892f6e711f08a5ed189a643e3fe9646e15c 11685Author: Stuart Bennett <sb476@cam.ac.uk> 11686Date: Wed Jan 30 17:43:16 2008 +0000 11687 11688 Fail gracefully for too-big resolutions / colour depths 11689 11690commit 09c9d7c3c50b630e1509f787ff946cc0656f7cba 11691Author: Maarten Maathuis <madman2003@gmail.com> 11692Date: Wed Jan 30 18:50:37 2008 +0100 11693 11694 randr12: Misc change. 11695 11696commit 6a46bf59f15d68404b536d436bf70337cd88e33d 11697Author: Maarten Maathuis <madman2003@gmail.com> 11698Date: Wed Jan 30 18:35:18 2008 +0100 11699 11700 Check if the framebuffer can fit into the offscreen memory. 11701 11702commit 9607d7828b16b0cb4a428bc29e2ce7c9bfd95371 11703Author: Arthur Huillet <arthur.huillet@free.fr> 11704Date: Wed Jan 30 18:24:44 2008 +0100 11705 11706 Xv: NV04 overlay can not scale down at all, as confirmed by DirectFB 11707 11708commit 7edfb473533ce41db9e2b3709d32bb8ca7da85a4 11709Author: Stuart Bennett <sb476@cam.ac.uk> 11710Date: Wed Jan 30 14:44:14 2008 +0000 11711 11712 PBUS define additions 11713 11714commit 6e090591bd2dab051362d8528a3bc76f30418d94 11715Author: Stuart Bennett <sb476@cam.ac.uk> 11716Date: Wed Jan 30 14:35:25 2008 +0000 11717 11718 PCRTC define additions 11719 11720commit e51799382e40c033b2d6d09dd65dd7f7c1e5e4d6 11721Author: Stuart Bennett <sb476@cam.ac.uk> 11722Date: Wed Jan 30 15:03:06 2008 +0000 11723 11724 Missed these in the ->absolute conversion 11725 11726commit ee0d557a26d326af27445bd0b283a5c6ead7c1cd 11727Author: Arthur Huillet <arthur.huillet@free.fr> 11728Date: Tue Jan 29 18:46:16 2008 +0100 11729 11730 Xv: detection of composite for the overlay 11731 11732commit 3a5f92d48e32d251e41bab561e52082033d26d9d 11733Author: Stuart Bennett <sb476@cam.ac.uk> 11734Date: Wed Jan 30 01:31:43 2008 +0000 11735 11736 Make bios port io use PVIO as appropriate, and have harsh validity checks 11737 11738commit 92c0d9630843957212d8ce296fea6bca8d802069 11739Author: Stuart Bennett <sb476@cam.ac.uk> 11740Date: Wed Jan 30 00:36:59 2008 +0000 11741 11742 Make PLL setting quieter 11743 11744commit f58d174bd50943c3dfb2d6c909d0a98dbbd35dae 11745Author: Stuart Bennett <sb476@cam.ac.uk> 11746Date: Tue Jan 29 23:10:27 2008 +0000 11747 11748 Move some regs from nv_bios.c to nvreg.h 11749 11750 And use more symbolic names in nv_bios 11751 11752commit d0022c35f39fde74ee5f3978ea78e053988fb608 11753Author: Stuart Bennett <sb476@cam.ac.uk> 11754Date: Tue Jan 29 22:41:40 2008 +0000 11755 11756 Tighten bios reg checks a little 11757 11758commit b1b1d76eb187f4ab3020bbd69c2b1cc9b39d0d83 11759Author: Stuart Bennett <sb476@cam.ac.uk> 11760Date: Wed Jan 30 00:30:46 2008 +0000 11761 11762 randr12: make nv4x plls less special 11763 11764commit 8f05d66bad8e74129cffa2918f9ecf6824c9f0b9 11765Author: Stuart Bennett <sb476@cam.ac.uk> 11766Date: Tue Jan 29 22:33:34 2008 +0000 11767 11768 Don't attempt to parse BIT D and L tables on non-mobile cards 11769 11770commit 8e834ece6592df13bd3eb7e9f28c01b4eb0c9791 11771Author: Stuart Bennett <sb476@cam.ac.uk> 11772Date: Tue Jan 29 18:57:28 2008 +0000 11773 11774 Add some mobile pci-ids, and reorder numerically 11775 11776commit 0b586ae0a2998339a9d5d3590e91dfa62af9ab6a 11777Author: Maarten Maathuis <madman2003@gmail.com> 11778Date: Tue Jan 29 19:04:25 2008 +0100 11779 11780 randr12: Remove excessive writing of VTOwner. 11781 11782commit c522ebb293a1c37142e29bf02bb82524e99336b0 11783Author: Stuart Bennett <sb476@cam.ac.uk> 11784Date: Tue Jan 29 17:30:32 2008 +0000 11785 11786 Make sure nv04 has some defaults (as the bios parsing is non-existant) 11787 11788commit 62b6c5fe490578990c249042657f20bb0cf49a01 11789Author: Stuart Bennett <sb476@cam.ac.uk> 11790Date: Sun Jan 27 22:11:31 2008 +0000 11791 11792 Use absolute regs in the DDX 11793 11794 Due to having to make a few defines for regs that were hardcoded, there could 11795 be breakage (nv04 overlay, this means you) 11796 Hopefully not. 11797 11798commit f5c9f9910a46cfa41d817ef7ca345bfea278b358 11799Author: Stuart Bennett <sb476@cam.ac.uk> 11800Date: Tue Jan 29 14:14:26 2008 +0000 11801 11802 Reorganise pll limit retrieval 11803 11804commit 26ff67f03be38925905a572dab3ec61653cd8286 11805Author: Maarten Maathuis <madman2003@gmail.com> 11806Date: Tue Jan 29 10:53:14 2008 +0100 11807 11808 NV30EXA: Always init texture origin, please report if your viewport setup needs to be different. 11809 11810commit 38df9a95740e934b9691dff1ed1b597cd2254cfa 11811Author: Maarten Maathuis <madman2003@gmail.com> 11812Date: Mon Jan 28 20:55:47 2008 +0100 11813 11814 Forgot to be paranoid. 11815 11816commit 2ea0314859569c3ae4356272a35f0b316abf4df7 11817Author: Maarten Maathuis <madman2003@gmail.com> 11818Date: Mon Jan 28 20:51:20 2008 +0100 11819 11820 Small improvement to nouveau_dma_wait, at least makes it show up in profiling in a clear way. 11821 11822 - This may also slightly reduce the cost of a very short wait, but not by much. 11823 11824commit dd483e226519bc4d2c088e97490039613014dfac 11825Author: Stuart Bennett <sb476@cam.ac.uk> 11826Date: Sat Jan 26 20:22:08 2008 +0000 11827 11828 randr12: 2nd attempt on nv11 console restore 11829 11830commit 5b0fa02101c82a83e11f7d279d366689d54bfacb 11831Author: Stuart Bennett <sb476@cam.ac.uk> 11832Date: Sat Jan 26 18:09:10 2008 +0000 11833 11834 Use NV_{WR,RD}08 to avoid double logging io 11835 11836 And one less bios FIXME 11837 11838commit a099bbff8bb3766201a7deb9c4a718fff7edbe51 11839Author: Stuart Bennett <sb476@cam.ac.uk> 11840Date: Mon Jan 21 23:54:51 2008 +0000 11841 11842 randr12: remove unused PLL code 11843 11844commit 0eb428e5a8b60e52d18b3bdffed46ae3f07b7bc0 11845Author: Stuart Bennett <sb476@cam.ac.uk> 11846Date: Fri Jan 25 16:05:59 2008 +0000 11847 11848 randr12: attempt to fix brokenness of head setting / getting on nv11 11849 11850commit eea2fd41ebc17f72b43c776a0391df0733b8d6e4 11851Author: Stuart Bennett <sb476@cam.ac.uk> 11852Date: Sat Jan 26 14:20:01 2008 +0000 11853 11854 randr12: don't do load detection on nv11 11855 11856commit a2c0facefbc878815e00c734464b3c87b5cce842 11857Author: Stuart Bennett <sb476@cam.ac.uk> 11858Date: Sat Jan 26 01:46:59 2008 +0000 11859 11860 Assume 18 bit laptop panels by default, add a message that may trigger on 24 bit panels 11861 11862 Revert "Invert 18/24 bit interface flag, so that the default value (0) does not imply dithering" 11863 11864 This reverts commit 5b9dc7ad6eb2f16c4ca8824c2a19c85bbe1f126b. 11865 11866commit ec548d20239a49f8e9eb0399317d4c68df24f4a3 11867Author: Stuart Bennett <sb476@cam.ac.uk> 11868Date: Sat Jan 26 01:36:41 2008 +0000 11869 11870 Detect mobile BMP biosen 11871 11872 I'd really love to know how where this info is in BIT 11873 11874commit 9f932eb684814e2a04c83d5aee172b9e020d82a3 11875Author: Maarten Maathuis <madman2003@gmail.com> 11876Date: Fri Jan 25 23:31:58 2008 +0100 11877 11878 NV30EXA: Fix viewport setup + switch to triangle based composite. 11879 11880commit afca5f2665a62ef2d025c15f07fe1cdc2b40d8d5 11881Author: Stuart Bennett <sb476@cam.ac.uk> 11882Date: Fri Jan 25 01:18:50 2008 +0000 11883 11884 randr12: fix / annotate a couple of ddc thinkos 11885 11886 catch a longstanding source of error too 11887 11888commit 3e6ea79c3292c3ed5bd676653c4c40532772ae5f 11889Author: Stuart Bennett <sb476@cam.ac.uk> 11890Date: Fri Jan 25 00:33:07 2008 +0000 11891 11892 Implement opcodes 0x66, 0x67, and 0x68 11893 11894 Courtesy jwstolk's trace 11895 Names taken from nv11 biosmod scps 11896 11897commit cdae2e26cc432ba572f92522f1b3454264d38f43 11898Author: Maarten Maathuis <madman2003@gmail.com> 11899Date: Fri Jan 25 00:10:08 2008 +0100 11900 11901 randr12: Rework native modes for TMDS. 11902 11903 - Choose a preferred ddc mode when available, otherwise the one with the highest refresh. 11904 - Fall back on cvt modes, only when needed. 11905 11906commit 46636052498d1e33985e3d98a6ce9462639bd4ff 11907Author: Stuart Bennett <sb476@cam.ac.uk> 11908Date: Wed Jan 23 23:02:46 2008 +0000 11909 11910 oops 11911 11912commit e067a0eee42324ec7c01e4d84a33554054817dd9 11913Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11914Date: Wed Jan 23 23:40:51 2008 +0100 11915 11916 Use proper version. 11917 11918commit 324c19acf6b506aae47e3699dd0fcefb72473590 11919Author: Stuart Bennett <sb476@cam.ac.uk> 11920Date: Wed Jan 23 01:13:58 2008 +0000 11921 11922 Fix C51 refclk 11923 11924commit 178b8bbf0e09c190fef8b2f85acddd3172ace629 11925Author: Stuart Bennett <sb476@cam.ac.uk> 11926Date: Wed Jan 23 20:59:50 2008 +0000 11927 11928 randr12: calculate (and set, for nv40) sel_clk before calculating plls 11929 11930commit 8067bd08d9a928c97b83f2b6ebff996b320c5813 11931Author: Maarten Maathuis <madman2003@gmail.com> 11932Date: Wed Jan 23 19:43:42 2008 +0100 11933 11934 Fix big bug in commit "Mostly log message changes". 11935 11936 - Please don't hide such important changes in a commit like that ;-) 11937 11938commit c5329a42c509efe9336573ff7e32978082ba9355 11939Author: Stuart Bennett <sb476@cam.ac.uk> 11940Date: Wed Jan 23 16:37:31 2008 +0000 11941 11942 Redo sel_clk setting and update its documentation 11943 11944 My apologies if this breaks stuff. 11945 11946commit f07ab92e7bb419df5473e1ed843a79ef4ff08f30 11947Author: Stuart Bennett <sb476@cam.ac.uk> 11948Date: Wed Jan 23 02:09:56 2008 +0000 11949 11950 Fail on LVDS DCB entry parsing issues 11951 11952commit 12efa8cdfe68c312458368ab876030b02919a922 11953Author: Stuart Bennett <sb476@cam.ac.uk> 11954Date: Wed Jan 23 00:09:11 2008 +0000 11955 11956 Mostly log message changes 11957 11958commit 0faa15d9b84b2cad23424dfc1679f119c5b45108 11959Author: Ben Skeggs <skeggsb@gmail.com> 11960Date: Wed Jan 23 23:45:37 2008 +1100 11961 11962 not an appropriate place for this type of thing.. 11963 11964commit 9e982df106028edd0f7ce0feedbe8e082f29f2ac 11965Author: Maarten Maathuis <madman2003@gmail.com> 11966Date: Wed Jan 23 11:42:55 2008 +0100 11967 11968 randr12: Increase cvt mode to 72 Hz. 11969 11970 - I've had occasions were i did slightly notice something at 60 Hz, and no problems at a higher setting. 11971 - I must admit i was a little bit tired at the time, but i figure it won't hurt (much). 11972 11973commit 71435dde5b2fd1c197ef5dc31b22ba40abcbca7e 11974Author: Maarten Maathuis <madman2003@gmail.com> 11975Date: Wed Jan 23 11:20:47 2008 +0100 11976 11977 NV40EXA: Use the same approach as the texture adapter for rendering. 11978 11979 - Verified to work for abitrary rotation. 11980 - This should eliminate diagional tearing altogether. 11981 11982commit d9149bddc758cc0644630b26fe10fc563ba38ce9 11983Author: Maarten Maathuis <madman2003@gmail.com> 11984Date: Wed Jan 23 10:28:25 2008 +0100 11985 11986 NV40TEX: Switch to a big triangle and clip it to a quad. 11987 11988 - On some systems (NV43, NV46 are known) an artifact shows when 2 triangles are combines and the size is larger than roughly 512x512. 11989 - The trick is draw a single triangle and use scissors to turn it into a quad. 11990 - Someone once asked me if i ever made drawings, well i did in this case and i've included it for future reference. 11991 11992commit e578d6e0a8c0fb8877d8b4b6ba1046dccf28ee6d 11993Author: Stuart Bennett <sb476@cam.ac.uk> 11994Date: Tue Jan 22 18:33:42 2008 +0000 11995 11996 Fix LVDS with DCB 2.0 11997 11998 Also use the X_ERROR facility for errors, not ErrorF 11999 12000commit fd3195f503420591b522249bb423c817db0709e7 12001Author: Maarten Maathuis <madman2003@gmail.com> 12002Date: Tue Jan 22 19:00:49 2008 +0100 12003 12004 Xv: Remove illegal usage of private exa symbols. 12005 12006commit 3a995b06ddb507655279565e992604bd40036b19 12007Author: Maarten Maathuis <madman2003@gmail.com> 12008Date: Tue Jan 22 15:53:38 2008 +0100 12009 12010 NV40TEX: Use width, not (x2, y2). 12011 12012commit abaa8e822be49b9d2129d31c4996dfcf8034ea2a 12013Author: Arthur Huillet <arthur.huillet@free.fr> 12014Date: Tue Jan 22 11:42:19 2008 +0100 12015 12016 Xv: NV12 overlay copy - removed wrong byteswapping on PPC, reported by moondrake 12017 12018commit 782237e4e942f2e8a9ed3dd111d60c665459963c 12019Author: Stuart Bennett <sb476@cam.ac.uk> 12020Date: Mon Jan 21 23:44:47 2008 +0000 12021 12022 randr12: some tidyup 12023 12024commit 91ca67056cb236e5770d4fb39dd7fbb9818ff093 12025Author: Stuart Bennett <sb476@cam.ac.uk> 12026Date: Mon Jan 21 22:35:58 2008 +0000 12027 12028 randr12: more beautiful dithering 12029 12030commit 1d5dc47324ddd1939db166af60df7a3a0e055810 12031Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12032Date: Mon Jan 21 22:19:24 2008 +0100 12033 12034 oops :) 12035 12036commit 7080f93a83e0d680f53a7962df300d527edd121a 12037Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12038Date: Mon Jan 21 21:43:11 2008 +0100 12039 12040 Quickly hacked support for NV6x. 12041 12042commit 2d399af61424cc4b4e0080ea016423f1b22c02f3 12043Author: Stuart Bennett <sb476@cam.ac.uk> 12044Date: Mon Jan 21 20:16:56 2008 +0000 12045 12046 randr12: fix dual head restore 12047 12048commit 8be20d6fc21a3869786e8c19656bf8b6569fee6e 12049Author: Stuart Bennett <sb476@cam.ac.uk> 12050Date: Mon Jan 21 18:51:52 2008 +0000 12051 12052 randr12: save and restore palette 12053 12054commit d32f310afa05ad85d6675348400880506d98ac1e 12055Author: Stuart Bennett <sb476@cam.ac.uk> 12056Date: Mon Jan 21 15:03:59 2008 +0000 12057 12058 Make LVDS bios script logs easier to understand 12059 12060 and fix a backwards comparison 12061 12062commit 7c7f0194678c7bd7c5ce791c07b38ac8eeeef608 12063Author: Stuart Bennett <sb476@cam.ac.uk> 12064Date: Mon Jan 21 14:14:07 2008 +0000 12065 12066 randr12: avoid SIGFPE 12067 12068commit b9a3aab7a7f6c5c4a5c440e46bbca7d76fd8df68 12069Author: Stuart Bennett <sb476@cam.ac.uk> 12070Date: Mon Jan 21 14:09:46 2008 +0000 12071 12072 debug_modes breaks build on old xservers 12073 12074commit d2e078b4ebf11b286e152e0ea10b17c41bf61b67 12075Author: Stuart Bennett <sb476@cam.ac.uk> 12076Date: Mon Jan 21 14:06:42 2008 +0000 12077 12078 randr12: use correct clock for LVDS modesetting part 2 12079 12080commit 52154e97b5a40b014fbee747fc542f765b4390bc 12081Author: Stuart Bennett <sb476@cam.ac.uk> 12082Date: Mon Jan 21 13:46:15 2008 +0000 12083 12084 randr12: use correct clock for LVDS modesetting 12085 12086commit d54a9f4b76f3a8c29c5fce6dfe7cc3fbc20f3952 12087Author: Stuart Bennett <sb476@cam.ac.uk> 12088Date: Mon Jan 21 12:27:36 2008 +0000 12089 12090 Don't log strange characters 12091 12092commit 07e01061e9b90348ea38b8889ff9e8daf207b820 12093Author: Stuart Bennett <sb476@cam.ac.uk> 12094Date: Sun Jan 20 21:53:30 2008 +0000 12095 12096 Use more imaginative log levels than X_INFO 12097 12098commit 18017632cbc96f36bbc72b454bd5b48070985a6f 12099Author: Stuart Bennett <sb476@cam.ac.uk> 12100Date: Sun Jan 20 20:17:28 2008 +0000 12101 12102 Use stdbool.h types in nv_bios.c 12103 12104 It makes vim highlighting work better and reduces shift key wear. 12105 Coincidentally it's what the drm uses too. 12106 12107 bool, true and false are defined in IEEE 1003.1 (2003) and POSIX since C99 12108 (as is stdint.h, which we use already) 12109 12110commit e1714a4f4329728cf897f8953acc480b47a4e5d7 12111Author: root <root@dpish.hayton.korsenowski> 12112Date: Sun Jan 20 18:38:11 2008 +0000 12113 12114 randr12: fix early cards 12115 12116commit 271844d3a761af65c9625d1564e1c08a9f64d159 12117Author: Maarten Maathuis <madman2003@gmail.com> 12118Date: Sun Jan 20 19:06:49 2008 +0100 12119 12120 NV40TEX: Don't sync when compositing, will have bad sideeffects if the compositing manager tries it too. 12121 12122commit c338a14ea688953b03f274d0e436b579d0411f18 12123Author: Maarten Maathuis <madman2003@gmail.com> 12124Date: Sun Jan 20 17:28:13 2008 +0100 12125 12126 Some misc cleanup. 12127 12128commit 2cd6e157cebac663f1b3c88e3653a4c2c9971ff1 12129Author: Maarten Maathuis <madman2003@gmail.com> 12130Date: Sun Jan 20 17:11:27 2008 +0100 12131 12132 randr12: Fix a major screwup i made during a cleanup. 12133 12134commit c3127a0c9cf4599c40b27e24a411a7fe4969152a 12135Author: Maarten Maathuis <madman2003@gmail.com> 12136Date: Sun Jan 20 17:00:58 2008 +0100 12137 12138 NV40TEX: Implement sync to vblank. 12139 12140 - I don't know why this didn't work the first time i tried, maybe i missed it due to another bug (that also caused tearing). 12141 - This should also work reasonably well for composited situations, but ofcource we can't guarantee that. 12142 12143commit 512f4ba33528dafb505bc2db5f9bb4d1b92c3a7b 12144Author: Stuart Bennett <sb476@cam.ac.uk> 12145Date: Sun Jan 20 14:47:52 2008 +0000 12146 12147 randr12: init some more Xv state 12148 12149commit 3f6a681a38fccbd53dd10aa42cc8d1b9bfe0eb71 12150Author: Stuart Bennett <sb476@cam.ac.uk> 12151Date: Sat Jan 19 14:02:43 2008 +0000 12152 12153 minor debug correction 12154 12155commit efd3cd086036cd5f8817aee771ee630066a92830 12156Author: Stuart Bennett <sb476@cam.ac.uk> 12157Date: Sun Jan 20 01:56:00 2008 +0000 12158 12159 INIT_8E opcode 12160 12161 again, from pq's trace 12162 12163commit f0bb436796a2d4dc693ec92b73c7abb27d428cda 12164Author: Stuart Bennett <sb476@cam.ac.uk> 12165Date: Sat Jan 19 01:37:34 2008 +0000 12166 12167 INIT_IO opcode 12168 12169 Based on pq's g8x trace 12170 12171commit c1b69fdaa089898821338b14c1f72eda354c2caa 12172Author: Maarten Maathuis <madman2003@gmail.com> 12173Date: Sat Jan 19 19:24:04 2008 +0100 12174 12175 randr12: hide cursor before locking + misc cleanup. 12176 12177commit dfa8901e847e1fbcbcb6de0c0726f53531b15b34 12178Author: Maarten Maathuis <madman2003@gmail.com> 12179Date: Sat Jan 19 16:39:26 2008 +0100 12180 12181 randr12: Remove unneeded vgahw stuff. 12182 12183commit 47452756724007318eb384912c40302b15523998 12184Author: Maarten Maathuis <madman2003@gmail.com> 12185Date: Sat Jan 19 16:29:57 2008 +0100 12186 12187 randr12: Move away from some of the nv_hw wrappers. 12188 12189commit 9e3a2d7c8c36d4dc43396be6b742b6137dfef455 12190Author: Maarten Maathuis <madman2003@gmail.com> 12191Date: Sat Jan 19 15:59:48 2008 +0100 12192 12193 randr12: Remove usage of nvRead/WriteVGA0/1. 12194 12195commit 936e9fe8adb2660cf75959265971c60f378cd7d6 12196Author: Maarten Maathuis <madman2003@gmail.com> 12197Date: Sat Jan 19 11:28:41 2008 +0100 12198 12199 randr12: Some cleanup. 12200 12201 - framebuffer consoles can use most of the normal stuff, so change it back. 12202 - some cursor deobfuscation 12203 12204commit a9bae6ba734882308ac29a9fe2403eaa027bd518 12205Author: Stuart Bennett <sb476@cam.ac.uk> 12206Date: Fri Jan 18 20:49:58 2008 +0000 12207 12208 More detailed modesetting debug (both randr12 and oldstyle) 12209 12210commit a28216639c3cf01d8f7cc85030d69fd58be86ed9 12211Author: Stuart Bennett <sb476@cam.ac.uk> 12212Date: Fri Jan 18 18:10:40 2008 +0000 12213 12214 Remove no longer needed panel quirk 12215 12216commit 59fec415ef32c291363408d5c8a24a2041450434 12217Author: Stuart Bennett <sb476@cam.ac.uk> 12218Date: Fri Jan 18 17:02:32 2008 +0000 12219 12220 Use bios PLL functions on nv4x 12221 12222 And a minor completeness correction 12223 12224commit 7efaf3d97a51d2447310954c9224fd7ec888caf9 12225Author: Maarten Maathuis <madman2003@gmail.com> 12226Date: Fri Jan 18 23:45:51 2008 +0100 12227 12228 randr12: Refine CR4B, anyone know what this does? 12229 12230 - There is more to this register. 12231 12232commit 41646042ab9d9d0c3bc8a76b25a4bff45f039922 12233Author: Stuart Bennett <sb476@cam.ac.uk> 12234Date: Fri Jan 18 16:50:50 2008 +0000 12235 12236 Rerefactor getting pll limits for nv4x 12237 12238commit 174d8f7e015250af4c0b27e044f5088f5465d851 12239Author: Stuart Bennett <sb476@cam.ac.uk> 12240Date: Fri Jan 18 14:24:48 2008 +0000 12241 12242 Use pll limits infrastructure on single stage plls 12243 12244 And add default PLL limits on old cards 12245 12246commit b3ca42d8589bed843ce5e772a1807c4c95377b92 12247Author: Stuart Bennett <sb476@cam.ac.uk> 12248Date: Fri Jan 18 01:47:05 2008 +0000 12249 12250 Don't speculate on the unknown limit field, and call the others log2p, not p 12251 12252commit d9db305e2d4bfb017a8542a7407ad4df27c86121 12253Author: Stuart Bennett <sb476@cam.ac.uk> 12254Date: Fri Jan 18 01:18:42 2008 +0000 12255 12256 Better setPLL_single 12257 12258commit 2a94184112cf06225672ef34836e19723761140a 12259Author: Stuart Bennett <sb476@cam.ac.uk> 12260Date: Fri Jan 18 14:33:01 2008 +0000 12261 12262 Bios parsing safety checks 12263 12264commit 6251f2fa774b913f5622ff9d07f56b2a5b9d1888 12265Author: Maarten Maathuis <madman2003@gmail.com> 12266Date: Fri Jan 18 12:17:50 2008 +0100 12267 12268 Some misc fixes. 12269 12270commit b21789f93c7e7df9db320dea093deab0f1056955 12271Author: Maarten Maathuis <madman2003@gmail.com> 12272Date: Fri Jan 18 09:23:39 2008 +0100 12273 12274 randr12: Actually using SavedReg does tend to help ;-) 12275 12276commit f15ea75b88b68fadb64fdeef75a23329e2a3f903 12277Author: Maarten Maathuis <madman2003@gmail.com> 12278Date: Fri Jan 18 09:13:24 2008 +0100 12279 12280 randr12: Fix a huge gaping hole in NewRestore. 12281 12282 - Restore CR5758 in a better place and actually restore something proper (use SavedReg instead of ModeReg). 12283 - Minor other fix. 12284 12285commit cd427da27d69db000f07ec50d27ca56ded97994e 12286Author: Stuart Bennett <sb476@cam.ac.uk> 12287Date: Thu Jan 17 23:37:23 2008 +0000 12288 12289 Invent a DVI-A output for the DVI-D output, on DCB1.4/1.5 12290 12291commit fbcb57827a1c4ac4b3c7ce66ca935f6b095f020d 12292Author: Stuart Bennett <sb476@cam.ac.uk> 12293Date: Thu Jan 17 22:49:37 2008 +0000 12294 12295 And another DCB 1.4 entry 12296 12297commit a355e5b0dcd5f5c78da20bd07f581b5841b3a2ba 12298Author: Maarten Maathuis <madman2003@gmail.com> 12299Date: Thu Jan 17 22:25:12 2008 +0100 12300 12301 randr12: Forgot to clear a few state variables on restore. 12302 12303commit ba0d0e8571b657b5196d0508347f7641cd7f04cb 12304Author: Stuart Bennett <sb476@cam.ac.uk> 12305Date: Thu Jan 17 20:48:16 2008 +0000 12306 12307 Add me 12308 12309commit 9cfc736ea4b089c74daaf1b0fac6393fec459bc0 12310Author: Stuart Bennett <sb476@cam.ac.uk> 12311Date: Thu Jan 17 20:29:30 2008 +0000 12312 12313 That warning wasn't there a moment ago 12314 12315commit 7eded6883404ce0cff83051497518a99c1d8a832 12316Author: Stuart Bennett <sb476@cam.ac.uk> 12317Date: Thu Jan 17 17:07:24 2008 +0000 12318 12319 Fix for situations with fixed second stage gain and hook code in to init scripts 12320 12321 And make sure bad PLL values are easily seen 12322 12323commit 075a14ac2525b364edc583caac50e03d171f67a7 12324Author: Stuart Bennett <sb476@cam.ac.uk> 12325Date: Thu Jan 17 19:59:55 2008 +0000 12326 12327 PLL limits reg off-by-four is dealt with elsewhere 12328 12329 Also some more imaginative pll limit member naming 12330 and minor get_pll_limits debugging corrections 12331 12332commit a3ef2cb34ad9de4567d30625a3bf87720a5ed8da 12333Author: Stuart Bennett <sb476@cam.ac.uk> 12334Date: Thu Jan 17 17:30:19 2008 +0000 12335 12336 setPLL for nv40 lowreg PLLs 12337 12338commit c3bef786b5291a638791b8a5125b3ff60cf9d1b1 12339Author: Stuart Bennett <sb476@cam.ac.uk> 12340Date: Thu Jan 17 17:30:36 2008 +0000 12341 12342 setPLL for nv40 highreg PLLs 12343 12344commit 996b93cdc73e3d6e581a766cbbf4a29bb34b8eb2 12345Author: Maarten Maathuis <madman2003@gmail.com> 12346Date: Thu Jan 17 18:27:51 2008 +0100 12347 12348 randr12: Some misc fixes. 12349 12350commit aa561c487754210ca22a1fc1d17c54d8e199e7a1 12351Author: Maarten Maathuis <madman2003@gmail.com> 12352Date: Wed Jan 16 22:42:18 2008 +0100 12353 12354 Some misc fixes, mostly related to setting the right crtc for overlay (randr12). 12355 12356commit 276d4e35673bb71410c882badf03550bcee48571 12357Author: Maarten Maathuis <madman2003@gmail.com> 12358Date: Wed Jan 16 08:03:51 2008 +0100 12359 12360 randr12: Fix mode privates again, hopefully ;-) 12361 12362commit f89243e4c2ba33a822b15de9136736f44b4c1579 12363Author: Stuart Bennett <sb476@cam.ac.uk> 12364Date: Wed Jan 16 00:25:08 2008 +0000 12365 12366 Minor reg adjustment 12367 12368commit df0f66750d1d893b2d91b0cfc91ecd4f0a1ba80e 12369Author: Stuart Bennett <sb476@cam.ac.uk> 12370Date: Tue Jan 15 16:30:00 2008 +0000 12371 12372 Refactor pll table parsing 12373 12374 And correct table reg matching on version 0x21 tables, which is off-by-4 12375 12376commit 3171b470f00a11d3528067c7d91e751257aa603c 12377Author: Stuart Bennett <sb476@cam.ac.uk> 12378Date: Tue Jan 15 14:59:49 2008 +0000 12379 12380 Use pll_lims in getMNP_double 12381 12382 Also move quirks elsewhere, and optimize 12383 12384commit 7bc921485c63232fecb59e488ef203166c48c6e1 12385Author: Stuart Bennett <sb476@cam.ac.uk> 12386Date: Tue Jan 15 15:14:46 2008 +0000 12387 12388 Store PLL limit frequencies in kHz, since that's what we use them in 12389 12390commit 51a739771b59a4725e75660b2c8edb54aa763ba5 12391Author: Stuart Bennett <sb476@cam.ac.uk> 12392Date: Tue Jan 15 01:32:25 2008 +0000 12393 12394 Two stage PLL limits parsing corrections 12395 12396 And some corrections to getMNP_double 12397 12398commit 7013116596ce8b065df265f16b3fdb9ac9e3df37 12399Author: Stuart Bennett <sb476@cam.ac.uk> 12400Date: Tue Jan 15 09:53:58 2008 +0000 12401 12402 randr12: Oh no! More pNv->twoHeads 12403 12404commit c7081039bf27aad12e17063d1a9becf160e7ee49 12405Author: Maarten Maathuis <madman2003@gmail.com> 12406Date: Tue Jan 15 21:05:08 2008 +0100 12407 12408 randr12: PrivFlags are not so private, switch to a real mode Private. 12409 12410 - This hopefully fixes the obscure issues people have had these last few days. 12411 12412commit cd14cd8c520361a15f5f8d255171c85810d83033 12413Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12414Date: Tue Jan 15 20:44:48 2008 +0100 12415 12416 Order the reg ranges, add some. 12417 12418commit 17fa42166fb4e8102cbe880be6021f1145268d16 12419Author: Patrice Mandin <pmandin@caramail.com> 12420Date: Tue Jan 15 18:45:30 2008 +0100 12421 12422 nv30: set VIEWPORT_TX_ORIGIN to 0 for all cards, except real nv30 12423 12424commit 777a3df246feb364d440282db0864fb9b0908872 12425Author: Maarten Maathuis <madman2003@gmail.com> 12426Date: Tue Jan 15 17:52:19 2008 +0100 12427 12428 NV30EXA: Chipset should be masked by 0xFFF0, switch to NVArch while i'm at it. 12429 12430commit b55608bac3f4f909f1ef8380a1ac69ca2f262ea2 12431Author: Maarten Maathuis <madman2003@gmail.com> 12432Date: Tue Jan 15 08:10:46 2008 +0100 12433 12434 randr12: Reenable cvtmode. 12435 12436commit 0db297bb6199878c0f77935acc447c568a87c900 12437Author: Maarten Maathuis <madman2003@gmail.com> 12438Date: Mon Jan 14 22:54:42 2008 +0100 12439 12440 randr12: Undo some changes of the past + always set vpll on modeset. 12441 12442commit 2879f4518e6830cf0b490cde5b4daca3e931383d 12443Author: Patrice Mandin <pmandin@caramail.com> 12444Date: Mon Jan 14 23:26:14 2008 +0100 12445 12446 nv30: Only change VIEWPOR_TX_ORIGIN on real nv30 12447 12448commit 405fd62ee8b56e3b1d6cc78df7ecfa654cd54dec 12449Author: Maarten Maathuis <madman2003@gmail.com> 12450Date: Mon Jan 14 20:17:05 2008 +0100 12451 12452 randr12: "NewRestore" Always reset some state regs after restore. 12453 12454commit 8093b83c3d1393ba225f9def1ce055b04cba5a8f 12455Author: Maarten Maathuis <madman2003@gmail.com> 12456Date: Mon Jan 14 19:29:28 2008 +0100 12457 12458 randr12: "NewRestore" proper depth vs bpp awareness. 12459 12460commit 8733ea89d9acece8721a2ae2d62a68916746c374 12461Author: Stuart Bennett <sb476@cam.ac.uk> 12462Date: Mon Jan 14 18:21:47 2008 +0000 12463 12464 randr12: yet more pre nv10 fixes 12465 12466commit c0d1117782673771b0490fe4d19abd83c6620033 12467Author: Stuart Bennett <sb476@cam.ac.uk> 12468Date: Mon Jan 14 11:54:28 2008 +0000 12469 12470 Possible fix for C51's misaligned regs 12471 12472commit 6ac824446d65a28519f378b3c01b1a6683a14d1b 12473Author: Stuart Bennett <sb476@cam.ac.uk> 12474Date: Sun Jan 13 21:36:53 2008 +0000 12475 12476 randr12: better load detection (enabled) 12477 12478commit ef28e91b24ee34911b3238e681064a95be7afd3a 12479Author: Maarten Maathuis <madman2003@gmail.com> 12480Date: Mon Jan 14 00:48:25 2008 +0100 12481 12482 randr12: We didn't copy the Flags to the adjusted mode, very bad. 12483 12484commit d41d5ae34a7d7f4658a1d86635fb8c59b80a1355 12485Author: Stuart Bennett <sb476@cam.ac.uk> 12486Date: Sun Jan 13 21:35:33 2008 +0000 12487 12488 Be more flexible on init table length 12489 12490commit 4744e5410c27d8bd3c2f92bee4554801639dba29 12491Author: Stuart Bennett <sb476@cam.ac.uk> 12492Date: Sun Jan 13 19:37:28 2008 +0000 12493 12494 More DCB 1.4/1.5 support 12495 12496commit c043b3fa7cf6297e010dfa6fcff24f263b8a8fd8 12497Author: Maarten Maathuis <madman2003@gmail.com> 12498Date: Sun Jan 13 17:09:04 2008 +0100 12499 12500 randr12: Minor fix. 12501 12502commit 3f7f0cab91416f6f4f4378ee6bdabeb9e75cf230 12503Author: Maarten Maathuis <madman2003@gmail.com> 12504Date: Sun Jan 13 16:05:22 2008 +0100 12505 12506 randr12: Some "NewRestore" fixes for framebuffer modes. 12507 12508 - This is mostly it for framebuffer modes. 12509 12510commit d8e7c5068835004741b74577344bd4defad4d139 12511Author: Maarten Maathuis <madman2003@gmail.com> 12512Date: Sun Jan 13 15:05:15 2008 +0100 12513 12514 nvbios: Fix a bmp bug for malc0. 12515 12516commit 5c01f96caa2b3b7223383f37d6903fbcaf3643cd 12517Author: Maarten Maathuis <madman2003@gmail.com> 12518Date: Sun Jan 13 14:12:00 2008 +0100 12519 12520 randr12: Some "NewRestore" fixes. 12521 12522 - Improvements to restoring framebuffer modes, still a few minor bugs to fix. 12523 - VGA modes work partially, i doubt this will make mainline though, since it's fragile. 12524 - This is a work in progress, only for testing. 12525 - Added some other functions to NOUVEAU_MODESET_TRACE. 12526 12527commit 364497a4234410adea25e4e51e0bdb17ccd3ba4a 12528Author: Stuart Bennett <sb476@cam.ac.uk> 12529Date: Sat Jan 12 00:39:59 2008 +0000 12530 12531 randr12: more fixes aimed at nv0x 12532 12533commit ac546e77fe71ffbbc76d9c48c5ab0b6faac832d7 12534Author: Stuart Bennett <sb476@cam.ac.uk> 12535Date: Fri Jan 11 22:05:11 2008 +0000 12536 12537 Remove some nv_bios TODOs and shut up some warnings 12538 12539commit fc04d71de6ec36158040617fb973218a8bdd0903 12540Author: Maarten Maathuis <madman2003@gmail.com> 12541Date: Fri Jan 11 23:23:43 2008 +0100 12542 12543 randr12: Add new experimental restore mode. 12544 12545 - Option "NewRestore" 12546 - Only tested on my 16bpp 1280x1024 vesafb console. 12547 - It may work on text consoles. 12548 - This should eventually replace the somewhat frail restore system that we currently have. 12549 12550commit 361bbe074a0d674012615b05fd3da81e8e1ce34e 12551Author: Maarten Maathuis <madman2003@gmail.com> 12552Date: Fri Jan 11 22:30:14 2008 +0100 12553 12554 randr12: Intermediate commit. 12555 12556commit 3b6e551e1afefe86338ab830505dc08412b2e80b 12557Author: Maarten Maathuis <madman2003@gmail.com> 12558Date: Fri Jan 11 18:59:53 2008 +0100 12559 12560 randr12: Intermediate commit. 12561 12562commit 5a989b23322e6ba1abe6456246300067c3bf42cb 12563Author: Stuart Bennett <sb476@cam.ac.uk> 12564Date: Fri Jan 11 21:33:47 2008 +0000 12565 12566 Tidy up and correct DCB merging 12567 12568commit ec7889a5a0787929c8c8ae08aaf729929c483035 12569Author: Stuart Bennett <sb476@cam.ac.uk> 12570Date: Thu Jan 10 23:44:59 2008 +0000 12571 12572 Fix bios opcode 0x56 and add reg alignment check 12573 12574commit 9b356d08f9169e35bdbff079c3351c475e53ec9b 12575Author: Stuart Bennett <sb476@cam.ac.uk> 12576Date: Fri Jan 11 16:34:28 2008 +0000 12577 12578 randr12: undo overzealous code disablement 12579 12580commit 44295b59f57be8956a5226415b5fdfbf49e175c1 12581Author: Stuart Bennett <sb476@cam.ac.uk> 12582Date: Fri Jan 11 12:28:16 2008 +0000 12583 12584 randr12: twoHeads and other fixes for pre nv10 (nv_crtc.c) 12585 12586commit 80fa1e1aef1b093b0310c51817ee9685e4d9d251 12587Author: Stuart Bennett <sb476@cam.ac.uk> 12588Date: Fri Jan 11 11:16:21 2008 +0000 12589 12590 randr12: some tidyup to output save/restore functions, and some twoHeads fixes 12591 12592commit 04f64725b5784ecfd7a5c3218d4710c2ad0f7f0f 12593Author: Stuart Bennett <sb476@cam.ac.uk> 12594Date: Fri Jan 11 11:20:45 2008 +0000 12595 12596 randr12: tidy up dpms functions 12597 12598 also, don't set cr57/58 in output_prepare as the dpms funcs do it, and it shouldn't be done on early cards anyway 12599 12600commit cc6794abb637e5ac8c98d7c6385bd36aa43d1ebc 12601Author: Stuart Bennett <sb476@cam.ac.uk> 12602Date: Thu Jan 10 20:25:50 2008 +0000 12603 12604 randr12: use clocks in kHz throughout 12605 12606commit 8337b77b40da9c3bba106837b4553433fa4e8a76 12607Author: Stuart Bennett <sb476@cam.ac.uk> 12608Date: Thu Jan 10 19:09:16 2008 +0000 12609 12610 The bios transition freq is in 10s kHz - so x10 12611 12612commit d65f59144f8a79116bca2f09de41d5786a7a4f84 12613Author: Maarten Maathuis <madman2003@gmail.com> 12614Date: Thu Jan 10 18:50:47 2008 +0100 12615 12616 randr12: Some minor things for supporting vga modes (in the future). 12617 12618commit 0e646b0332b52e23ef13301765f7f1e8f881e165 12619Author: Maarten Maathuis <madman2003@gmail.com> 12620Date: Thu Jan 10 17:41:39 2008 +0100 12621 12622 randr12: Intermediate commit. 12623 12624commit a5122460cfa07d45cf91ae112db9170fecee6966 12625Author: Stuart Bennett <sb476@cam.ac.uk> 12626Date: Thu Jan 10 16:21:49 2008 +0000 12627 12628 Remove weirdo special case handling on bios reg writes 12629 12630 I don't think these ever get used, and the code only seems to be on gf6 anyway 12631 12632commit 31b1a894079ca3983d17ac7a81d6ecd420f7d377 12633Author: Stuart Bennett <sb476@cam.ac.uk> 12634Date: Thu Jan 10 14:14:24 2008 +0000 12635 12636 randr12: dtrt wrt dual link on lvds when !straps 12637 12638commit e02d2cc8b77443fde4e267ce4d9691ed680eef01 12639Author: Stuart Bennett <sb476@cam.ac.uk> 12640Date: Thu Jan 10 12:03:43 2008 +0000 12641 12642 randr12: improve setting of fp_control 12643 12644commit d8eb924a43576c0e360965a43f5363d0ba2aecd0 12645Author: Arthur Huillet <arthur.huillet@free.fr> 12646Date: Thu Jan 10 12:03:30 2008 +0100 12647 12648 Xv: cleanup. 12649 12650commit 1f727083cfafa9097ed98266420a2fd16905ae10 12651Author: Stuart Bennett <sb476@cam.ac.uk> 12652Date: Thu Jan 10 00:12:52 2008 +0000 12653 12654 randr12: enable LVDS and refactor output creation 12655 12656 Heavily based on initial patch by seventhguardian 12657 12658commit c08c6de0f452c7f5cd1bac8bf974aaac872c8e2c 12659Author: Stuart Bennett <sb476@cam.ac.uk> 12660Date: Wed Jan 9 16:18:30 2008 +0000 12661 12662 deifdefication - we requre xserver 1.3, so randr 1.2 is guaranteed 12663 12664commit 1556144a1af3c89eaa14cbfe8bdc9dc966fef8be 12665Author: Stuart Bennett <sb476@cam.ac.uk> 12666Date: Wed Jan 9 15:47:50 2008 +0000 12667 12668 randr12: optimize away excessive LVDS_RESETing 12669 12670commit 55e6750038edd7010ae8304fe3c3b4ac9e7b963c 12671Author: Maarten Maathuis <madman2003@gmail.com> 12672Date: Wed Jan 9 22:08:49 2008 +0100 12673 12674 randr12: Fix minor bug, related to output resource conflicts. 12675 12676commit c1fe3766bd181d9ed273850e6c0d1939c0efc8aa 12677Author: Maarten Maathuis <madman2003@gmail.com> 12678Date: Wed Jan 9 20:20:03 2008 +0100 12679 12680 randr12: Reset crtc when changing output routes. 12681 12682commit 01b9a081c288c69fa0bbc0d3829a0026796e731a 12683Author: Maarten Maathuis <madman2003@gmail.com> 12684Date: Wed Jan 9 19:30:17 2008 +0100 12685 12686 randr12: Don't change vpll's twice on restore. 12687 12688commit 3643b2635dcaba3e6ec969a3cd6eda43623c8992 12689Author: Stuart Bennett <sb476@cam.ac.uk> 12690Date: Wed Jan 9 15:12:30 2008 +0000 12691 12692 Detect an LVDS panel if there's a hardcoded EDID 12693 12694commit 64d8032c419ef8cf14cdf3f83228d3e0f5714909 12695Author: Stuart Bennett <sb476@cam.ac.uk> 12696Date: Wed Jan 9 15:11:24 2008 +0000 12697 12698 If no PROM, and no PRAMIN, use the PCI rom image 12699 12700commit 99a64d31979a8c83ac9d52fabe3114c576c94bba 12701Author: Stuart Bennett <sb476@cam.ac.uk> 12702Date: Tue Jan 8 22:58:37 2008 +0000 12703 12704 fp bios edid oops commit 12705 12706commit b470cf257433dad174d62cdbd4331ada2809334e 12707Author: Maarten Maathuis <madman2003@gmail.com> 12708Date: Tue Jan 8 23:54:51 2008 +0100 12709 12710 randr12: More fixes for strange output routes. 12711 12712commit cfbdbd6ac846af79d9ddf2b4597736cba74d5916 12713Author: Stuart Bennett <sb476@cam.ac.uk> 12714Date: Tue Jan 8 20:57:10 2008 +0000 12715 12716 randr12: try to use bios fp edid, if nothing else works 12717 12718commit d4e4892244c90b481bfd8c6d28d2390fe696e160 12719Author: Stuart Bennett <sb476@cam.ac.uk> 12720Date: Tue Jan 8 20:07:10 2008 +0000 12721 12722 Read EDID from BIOS if appropriate 12723 12724commit 6fab536aa13b8639d9b419a37fccfb30a4313b77 12725Author: Stuart Bennett <sb476@cam.ac.uk> 12726Date: Tue Jan 8 16:18:09 2008 +0000 12727 12728 Darktama's C51 is a mobile card 12729 12730commit 921f2e226ccbe7468cc8e51919872c3e8cf28c4e 12731Author: Stuart Bennett <sb476@cam.ac.uk> 12732Date: Tue Jan 8 15:37:51 2008 +0000 12733 12734 BIT cards use the init scripts to do the power sequencing 12735 12736commit 33ea2cdef8c46ee9837b6fc9809df82d4bdca339 12737Author: Maarten Maathuis <madman2003@gmail.com> 12738Date: Tue Jan 8 22:36:12 2008 +0100 12739 12740 randr12: Small fix for NV3x, more will be needed. 12741 12742commit 175b91d48644bdbfc4e48f4c2fe6c9cf936802b5 12743Author: Maarten Maathuis <madman2003@gmail.com> 12744Date: Tue Jan 8 20:23:38 2008 +0100 12745 12746 randr12: Fix reverse pll reading code for NV30. 12747 12748commit 4add8cfb68404e9e2764451c4e48ab4d0cf04482 12749Author: Maarten Maathuis <madman2003@gmail.com> 12750Date: Tue Jan 8 19:07:50 2008 +0100 12751 12752 randr12: Some minor things for NV30. 12753 12754commit 74b03f7d4f5506222b6a95ccf8b5c9e0ff69e4dc 12755Author: Stuart Bennett <sb476@cam.ac.uk> 12756Date: Tue Jan 8 13:52:00 2008 +0000 12757 12758 Assorted small bios/randr12 bits and pieces 12759 12760 Make bios execution a bit faster 12761 Better self documenting code for DCB dual link criterion 12762 Tidy up some ineffectual LVDS mode code 12763 12764commit 76f3c47736661df86490b0d133ac82c371cab73d 12765Author: Maarten Maathuis <madman2003@gmail.com> 12766Date: Mon Jan 7 23:38:08 2008 +0100 12767 12768 randr12: Handle NV30 vpll's correctly. 12769 12770 - This only concerns the real NV30, not the other NV3x cards. 12771 12772commit 1a25def0a64ac6437c6ba5f51ddb8743d4a88d00 12773Author: Maarten Maathuis <madman2003@gmail.com> 12774Date: Mon Jan 7 19:06:16 2008 +0100 12775 12776 randr12: Clear dpms cache when needed. 12777 12778commit 3f0008128e2afd2b1a1b595829c6324e46a4bf1d 12779Author: Arthur Huillet <arthur.huillet@free.fr> 12780Date: Mon Jan 7 19:03:53 2008 +0100 12781 12782 Xv: overlay BW workaround applied to NV40 as well 12783 12784commit d5dabd8c316371605d81be55a5c6c82e6351b591 12785Author: Arthur Huillet <arthur.huillet@free.fr> 12786Date: Mon Jan 7 19:01:07 2008 +0100 12787 12788 Xv: removed useless #includes 12789 12790commit 510f3575df87733857dbb5bfe398ce04aebadfd4 12791Author: Arthur Huillet <arthur.huillet@free.fr> 12792Date: Mon Jan 7 17:13:43 2008 +0100 12793 12794 Xv: added NV04 overlay brightness attribute. Original patch by Andrew Randrianasulu 12795 12796commit a71de657e1e2718800ef1c319dab5dafcc9ea0cd 12797Author: Ben Skeggs <skeggsb@gmail.com> 12798Date: Tue Jan 8 02:18:59 2008 +1100 12799 12800 avoid symbol clash with 3d driver. 12801 12802commit 0463dc0682fb192ce82037601289f75913f65e3c 12803Author: Ben Skeggs <skeggsb@gmail.com> 12804Date: Tue Jan 8 02:16:26 2008 +1100 12805 12806 nv50: support hostdata UTS 12807 12808commit 9fb186c57bdad07281f1a0bdebac98a4b18dc6e3 12809Author: Ben Skeggs <skeggsb@gmail.com> 12810Date: Mon Jan 7 18:33:03 2008 +1100 12811 12812 nv50: small fix 12813 12814commit b679d6a8539403f075a57d882bb4d0eb33b180d9 12815Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12816Date: Mon Jan 7 02:39:19 2008 +0100 12817 12818 nv40 video texture: sometimes less is more. 12819 12820commit b9a364a985bb23ee30dc094ba13e19ea2f4f6bdf 12821Author: Maarten Maathuis <madman2003@gmail.com> 12822Date: Sun Jan 6 20:44:56 2008 +0100 12823 12824 NV40EXA: Remove stair shaped tearing i get. 12825 12826commit 80ac301acaf8dba6a93f5b04b97f8dca2748b92c 12827Author: Pierre Ossman <drzeus@drzeus.cx> 12828Date: Sun Jan 6 20:04:10 2008 +0100 12829 12830 We do not get a valid CRTC configuration structure if we're not using RandR 1.2, 12831 so make sure we can handle the old mode with some form of grace. 12832 12833 Signed-off-by: Pierre Ossman <drzeus@drzeus.cx> 12834 12835commit f1aaf3fe616f5a0633e685867cf33b37f2c3f07b 12836Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12837Date: Sun Jan 6 19:14:03 2008 +0100 12838 12839 nv40 video texture: bilinear filtering for UV. 12840 12841commit 57fc0631428f29af0d24036c939826349e3018b5 12842Author: Arthur Huillet <arthur.huillet@free.fr> 12843Date: Sun Jan 6 17:41:50 2008 +0100 12844 12845 Xv: overlay auto CRTC switch with randr12, blitter fallback 12846 12847 Now switching the overlay to the right CRTC when displaying video with the overlay and dual head, with a nice fallback 12848 to the blitter when the window is on both CRTCs. 12849 Fixed a bug preventing the overlay from working on CRTC1. 12850 12851commit b6ba53ee1cbda5cb91e43ae6b84e1737ebc367f4 12852Author: Jeremy Kolb <jkolb@glorfindel.hsd1.ma.comcast.net> 12853Date: Sun Jan 6 12:58:45 2008 -0500 12854 12855 port "G80: Reduce load detection false positives." from xf86-video-nv commit d483ecd851bb84014f1fa86f0ada4540b3ad801e 12856 12857commit b4fa9b060308090f4bab3f8716f7e2ecaf234be7 12858Author: Jeremy Kolb <jkolb@glorfindel.hsd1.ma.comcast.net> 12859Date: Sun Jan 6 12:51:55 2008 -0500 12860 12861 Return NULL not FALSE. 12862 12863commit 47dde8283bcc409037682098cad8f0a0c3f9e3b2 12864Author: Pierre Ossman <drzeus@drzeus.cx> 12865Date: Sun Jan 6 17:48:48 2008 +0100 12866 12867 Xv: don't wait for vblank on CRTC0 _and_ CRTC1 12868 12869commit c8a9b4d0c2145d13c74a00cb24e1310e247a2718 12870Author: Pierre Ossman <drzeus@drzeus.cx> 12871Date: Sun Jan 6 17:45:37 2008 +0100 12872 12873 wait for correct crtc vsync 12874 12875 Make sure we wait for the CRTC that will actually display the video 12876 frame we're blitting. 12877 12878commit 25fac91dfc7072405243d2891f9ec1f0cef49a3d 12879Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12880Date: Sun Jan 6 16:20:43 2008 +0100 12881 12882 Switch to latest nouveau_class.h 12883 12884commit bc82548514daab356fe0e0ee942ff75018fbc383 12885Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12886Date: Sun Jan 6 16:19:50 2008 +0100 12887 12888 Use clamp to edge to avoid the thin black lines. 12889 12890commit acaf2501d17030097761b3f8d369f14874261f6d 12891Author: Maarten Maathuis <madman2003@gmail.com> 12892Date: Sun Jan 6 13:17:49 2008 +0100 12893 12894 randr12: Some fixes for NV31M. 12895 12896commit 7b1521c263511016538286806f54221721c60608 12897Author: Maarten Maathuis <madman2003@gmail.com> 12898Date: Sat Jan 5 19:43:08 2008 +0100 12899 12900 randr12: NV11 and NV20 stop at CR52. 12901 12902commit 9441ca0fd579cb648577f107777236d22a2b1ead 12903Author: Maarten Maathuis <madman2003@gmail.com> 12904Date: Sat Jan 5 19:32:32 2008 +0100 12905 12906 randr12: NV11 and NV20 don't have CR57 and CR58. 12907 12908commit 8f520754459c363313353605207f7598e259ea77 12909Author: Maarten Maathuis <madman2003@gmail.com> 12910Date: Sat Jan 5 18:54:14 2008 +0100 12911 12912 Add a missing function to NOUVEAU_MODESET_TRACE. 12913 12914commit d5bf6a0149dc80d5d76ab8857415d20b42188656 12915Author: Stuart Bennett <sb476@cam.ac.uk> 12916Date: Sat Jan 5 17:22:33 2008 +0000 12917 12918 randr12: fix PLL breaking typo 12919 12920commit 21f02f5f248bce5f49e0fcee5594c83155b532b5 12921Author: Stuart Bennett <sb476@cam.ac.uk> 12922Date: Sat Jan 5 03:10:21 2008 +0000 12923 12924 Need to restore EXTDEV before fp table parsing 12925 12926 and a nice hwsq message 12927 12928commit 4e4819f6722a120d3a22eb9829fe73780a0c1087 12929Author: Maarten Maathuis <madman2003@gmail.com> 12930Date: Sat Jan 5 18:30:08 2008 +0100 12931 12932 randr12: Minor cleanup. 12933 12934commit 758e1cf14d61725d69ca75b7268f22e11a32a5ba 12935Author: Maarten Maathuis <madman2003@gmail.com> 12936Date: Sat Jan 5 18:27:10 2008 +0100 12937 12938 randr12: Fix a thinko, generate data before restoring, not at the beginning. 12939 12940commit f315ce85fa33894767892772c1e7a0966150ba6b 12941Author: Maarten Maathuis <madman2003@gmail.com> 12942Date: Sat Jan 5 14:05:55 2008 +0100 12943 12944 randr12: Some NV40 quirks. 12945 12946commit 08b1a2622708be9f714c27fd31372e28277d7c5b 12947Author: Maarten Maathuis <madman2003@gmail.com> 12948Date: Sat Jan 5 13:03:40 2008 +0100 12949 12950 randr12: Small quirk for NV11. 12951 12952commit 73101dbcfac68688914e2a3f46008ddd69d0dc62 12953Author: Stuart Bennett <sb476@cam.ac.uk> 12954Date: Sat Jan 5 02:46:52 2008 +0000 12955 12956 Load NV17 power sequencing microcode 12957 12958commit 2ec6cf4a6e59b861c1672471872fc4ca75cfce27 12959Author: Maarten Maathuis <madman2003@gmail.com> 12960Date: Sat Jan 5 01:22:42 2008 +0100 12961 12962 randr12: Cache dpms calls to avoid unnecesary blinking and such. 12963 12964commit 4f586ac169e7acd0e0bfa4139bd74e718d7598e2 12965Author: Arthur Huillet <arthur.huillet@free.fr> 12966Date: Sat Jan 5 01:38:52 2008 +0200 12967 12968 Xv: linearly interpolate U and V samples when converting from YV12 to YUY2 12969 12970 Original patch by ahuillet, pq did one last fix. 12971 12972commit 5f1a721070c20a4287d17a8797ef1aabe19a579e 12973Author: Stuart Bennett <sb476@cam.ac.uk> 12974Date: Fri Jan 4 17:59:54 2008 +0000 12975 12976 More DCB 1.4/1.5 fixes 12977 12978commit 95485646eb61aaa84a8d3b6550552ab417be8e16 12979Author: Maarten Maathuis <madman2003@gmail.com> 12980Date: Fri Jan 4 18:33:31 2008 +0100 12981 12982 randr12: Forgot to switch to vpll1_a, etc in one place. 12983 12984commit c43bb4990f44d50d91785f8c9860e27c3357c0a3 12985Author: Stuart Bennett <sb476@cam.ac.uk> 12986Date: Fri Jan 4 01:14:40 2008 +0000 12987 12988 randr12: fix BIT bios LVDS dpms 12989 12990commit 5d9e42bbb5878f52cf7fd6d96532253d868b2a27 12991Author: Stuart Bennett <sb476@cam.ac.uk> 12992Date: Fri Jan 4 13:26:38 2008 +0000 12993 12994 Better guesses at DCB 1.4/1.5 12995 12996commit 73da4362cd7e2918bd066528a9feb157019edb5e 12997Author: Maarten Maathuis <madman2003@gmail.com> 12998Date: Fri Jan 4 11:54:27 2008 +0100 12999 13000 randr12: Remove unnecesary usleep. 13001 13002 - I once did this thinking it may solve a bug, but it turned out a palette issue. 13003 13004commit c35f532fdee6141e871560f8a400040c2eb82891 13005Author: Maarten Maathuis <madman2003@gmail.com> 13006Date: Fri Jan 4 11:43:16 2008 +0100 13007 13008 NV40: Improve VCO2 detection in nvGetClocks() 13009 13010commit 9369c03788d7410150044d2e3a83f0bcb876af42 13011Author: Maarten Maathuis <madman2003@gmail.com> 13012Date: Fri Jan 4 11:39:37 2008 +0100 13013 13014 randr12: Improve speed of mode switches. 13015 13016 - Moved pre-NV40 to vpll1_a, vpll1_b, etc variables. 13017 - Only change vpll when needed. 13018 13019commit ec0fe5294fdee7a79e9bfaf21ea42cfd3f52ec75 13020Author: Stuart Bennett <sb476@cam.ac.uk> 13021Date: Fri Jan 4 02:31:21 2008 +0000 13022 13023 Infra for regs stored over suspend and a sort-of implementation of INIT_COMPUTE_MEM 13024 13025commit 786e133577edcf91b519dd8dede24250ec6fdcdf 13026Author: Stuart Bennett <sb476@cam.ac.uk> 13027Date: Thu Jan 3 02:21:35 2008 +0000 13028 13029 Better bios register read function prototype and a ReadEXTDEV purge 13030 13031commit 951f95ba553f301de2a540ef67cf04152cf72344 13032Author: Stuart Bennett <sb476@cam.ac.uk> 13033Date: Thu Jan 3 00:52:12 2008 +0000 13034 13035 Refactor NVParseBios 13036 13037commit 8b82f24e8a526c0483120f9979a3b9370da3b297 13038Author: Stuart Bennett <sb476@cam.ac.uk> 13039Date: Thu Jan 3 00:32:24 2008 +0000 13040 13041 Revert "randr12: LVDS screens use center mode." 13042 13043 This reverts commit 1142b15c56a0b9dab8b929fc196a825815457374. 13044 13045 They don't, necessarily 13046 13047commit 876d7b11c754dd66738bcc26a07e0d707b81a99f 13048Author: Maarten Maathuis <madman2003@gmail.com> 13049Date: Thu Jan 3 22:45:00 2008 +0100 13050 13051 NV40: Handle broken VCO2 values. 13052 13053commit 50eb355e44017a51636c1e7b8c140bc3b48813af 13054Author: Maarten Maathuis <madman2003@gmail.com> 13055Date: Thu Jan 3 22:36:04 2008 +0100 13056 13057 NV40: A better attempt at YUV->RGB conversion constants + remove vblank sync. 13058 13059 - The vblank sync didn't work (made things worse actually), someone needs to reverse engineer the proper way for the 3d engine. 13060 13061commit 22cdfec9b0858c186ab37ff649248a2a825eb513 13062Author: Maarten Maathuis <madman2003@gmail.com> 13063Date: Thu Jan 3 12:48:43 2008 +0100 13064 13065 NV40: Avoid a SIGFPE in nvGetClocks() when VCO2 is off. 13066 13067commit 5015e2cefcf2069ee0054c433c7ae1763d2f95eb 13068Author: Maarten Maathuis <madman2003@gmail.com> 13069Date: Thu Jan 3 12:37:10 2008 +0100 13070 13071 NV40: Actually add the attribute. 13072 13073commit 5b60344486c8ddeba28204dbea6f1e11704cf951 13074Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 13075Date: Thu Jan 3 05:56:23 2008 +0100 13076 13077 Add a vsync attribute to nv40 textured video. It doesn't seem to work so well for me (6200), I suspect because my card is too slow anyway. Possible solutions include reducing the shader instruction # and forcing drawing from top to bottom. 13078 13079commit f7a356250a9a0cb214dbb1871696506c04fab946 13080Author: Maarten Maathuis <madman2003@gmail.com> 13081Date: Wed Jan 2 12:15:21 2008 +0100 13082 13083 randr12: Hopefully these were all the function declerations that need updating. 13084 13085 - Build didn't even show any errors, so that's why i missed them. 13086 13087commit aa608573918ed34968054fb5a93b11fc5c8aa1a8 13088Author: Maarten Maathuis <madman2003@gmail.com> 13089Date: Wed Jan 2 00:19:12 2008 +0100 13090 13091 Dump some CARD8 and CARD32 in favour of uintX_t. 13092 13093commit b44af66323b622411d5ffa4a78a210abefdb1e2f 13094Author: Maarten Maathuis <madman2003@gmail.com> 13095Date: Tue Jan 1 22:38:34 2008 +0100 13096 13097 randr12: Small fix for NV4x cards with strange output routes. 13098 13099commit 6b204d5ef3ff61236ef86eadcc7e137a722aef72 13100Author: Maarten Maathuis <madman2003@gmail.com> 13101Date: Tue Jan 1 21:51:31 2008 +0100 13102 13103 randr12: Improve output routing. 13104 13105commit 64546788e50ae3482698f9505869e06f727c0e50 13106Author: Maarten Maathuis <madman2003@gmail.com> 13107Date: Tue Jan 1 21:07:36 2008 +0100 13108 13109 randr12: Really hopefully fix output resource conflict detection now. 13110 13111commit b71a52ae9689026e7d96d2c2752eec0c9854b0c3 13112Author: Maarten Maathuis <madman2003@gmail.com> 13113Date: Tue Jan 1 20:38:24 2008 +0100 13114 13115 randr12: Hopefully fix output resource conflict detection. 13116 13117commit 9ac2765ff13f82746991c85649f45c13475e851d 13118Author: Maarten Maathuis <madman2003@gmail.com> 13119Date: Tue Jan 1 19:41:07 2008 +0100 13120 13121 Forgot the NOUVEAU_EXA_PIXMAPS case. 13122 13123commit 53565fc3e854abc0b5e7614da251024a60ddfd9e 13124Author: Maarten Maathuis <madman2003@gmail.com> 13125Date: Tue Jan 1 19:36:28 2008 +0100 13126 13127 Stricter alignment requirements for offscreenBase, to avoid framebuffer corruption. 13128 13129commit c9cedb7712719e3dee450245ebeb7db18cd40770 13130Author: Maarten Maathuis <madman2003@gmail.com> 13131Date: Tue Jan 1 17:36:20 2008 +0100 13132 13133 randr12: Time to let some routing insanity die. 13134 13135 - I hope nvidia is just crazy sometimes when setting CRTC1 on both output resources. 13136 - Or is it some override used on load detect? 13137 13138commit 9f1494ebf1ec1957f362366ab4abfc31c2227092 13139Author: Maarten Maathuis <madman2003@gmail.com> 13140Date: Tue Jan 1 17:28:41 2008 +0100 13141 13142 randr12: Always set bit 26 in FP_DEBUG_0 if already present. 13143 13144commit a83c71244b8dceb2cac968195777636bd5c5f639 13145Author: Maarten Maathuis <madman2003@gmail.com> 13146Date: Tue Jan 1 17:01:19 2008 +0100 13147 13148 randr12: Minor tweak to flatpanel regs. 13149 13150commit afe174cd57173c22011718d3290db5d939076ce6 13151Author: Maarten Maathuis <madman2003@gmail.com> 13152Date: Tue Jan 1 14:25:54 2008 +0100 13153 13154 Remove AM_MAINTAINER_MODE from configure.ac 13155 13156commit ef1a613e0f1a1fffb54a50fd14c10069ed9eac4b 13157Author: Maarten Maathuis <madman2003@gmail.com> 13158Date: Mon Dec 31 20:24:41 2007 +0100 13159 13160 randr12: Fix a bug in an earlier commit. 13161 13162 - This was local, but i wanted to keep some things seperate. 13163 13164commit 01313e3dca07615fa3b2a2b09f6d45489cc8d796 13165Author: Maarten Maathuis <madman2003@gmail.com> 13166Date: Mon Dec 31 20:15:03 2007 +0100 13167 13168 randr12: Add a quirk for 1280x800 panel on 7300go. 13169 13170commit 78bf63088d5805762460801c59f0f93b43ec96a7 13171Author: Maarten Maathuis <madman2003@gmail.com> 13172Date: Mon Dec 31 19:34:33 2007 +0100 13173 13174 randr12: Duplicate native modes, to avoid hitting an invalid pointer. 13175 13176commit 83aad1ad6e8a5e48ee91346b4c7037ddc8456a73 13177Author: Maarten Maathuis <madman2003@gmail.com> 13178Date: Mon Dec 31 18:57:49 2007 +0100 13179 13180 randr12: Set the adjusted mode in output code. 13181 13182commit 23d9d05a68cf2eac51b500d48d31bae3b484bbe8 13183Author: Maarten Maathuis <madman2003@gmail.com> 13184Date: Mon Dec 31 14:20:41 2007 +0100 13185 13186 randr12: Deal with output resources in a better way. 13187 13188commit a4d973214158efa90a1a23da879630d2b53ad313 13189Author: Maarten Maathuis <madman2003@gmail.com> 13190Date: Mon Dec 31 13:18:16 2007 +0100 13191 13192 randr12: Disable crt load detect again, until i know how to use it properly. 13193 13194commit 59346e15688186b4c00ee5348413ae8312ac9a30 13195Author: Maarten Maathuis <madman2003@gmail.com> 13196Date: Sun Dec 30 19:11:48 2007 +0100 13197 13198 Revert "randr12: Another quirk for the 7300GO cards." 13199 13200 This reverts commit 2f2921e6c33a0482f1f30e4d69820dc82690310a. 13201 It was unneeded after all. 13202 13203commit 2f2921e6c33a0482f1f30e4d69820dc82690310a 13204Author: Maarten Maathuis <madman2003@gmail.com> 13205Date: Sun Dec 30 17:07:13 2007 +0100 13206 13207 randr12: Another quirk for the 7300GO cards. 13208 13209commit 642baacdab215f6ad7173757b8d7e00f716d6dbe 13210Author: Maarten Maathuis <madman2003@gmail.com> 13211Date: Sun Dec 30 16:44:23 2007 +0100 13212 13213 randr12: Fix quad view on some 7300GO cards. 13214 13215commit df36ea6d9cd2a22b51a0386bbfe4f6760ed98d97 13216Author: Maarten Maathuis <madman2003@gmail.com> 13217Date: Sun Dec 30 15:01:47 2007 +0100 13218 13219 Add some headerfiles to Makefile.am 13220 13221commit 29851581b2420570efd4fca2d4f6a19113ea6947 13222Author: Maarten Maathuis <madman2003@gmail.com> 13223Date: Sun Dec 30 13:35:47 2007 +0100 13224 13225 randr12: Only load palette's when appropriate. 13226 13227commit c3359789d8f288b403aedbcdfa8795391fc13deb 13228Author: Maarten Maathuis <madman2003@gmail.com> 13229Date: Sun Dec 30 13:26:24 2007 +0100 13230 13231 NV40: Some fixes and safeties. 13232 13233commit a7c922b58af72e3a48b92056c2fcd71c22477c2f 13234Author: Maarten Maathuis <madman2003@gmail.com> 13235Date: Sun Dec 30 11:36:10 2007 +0100 13236 13237 NV40: Fix some comments and a damage reporting bug. 13238 13239commit b6cd9964f9c20c40eda4283a93e447601beddb96 13240Author: Maarten Maathuis <madman2003@gmail.com> 13241Date: Sun Dec 30 02:31:38 2007 +0100 13242 13243 NV40: Fix the texturing adapter when not rendering offscreen. 13244 13245 - Now clip restrictions are taken into consideration. 13246 - No streching when clipping the border of a screen anymore. 13247 13248commit 4efdd1b560228aae7f6f5d6ed549461f8fdf0d00 13249Author: Maarten Maathuis <madman2003@gmail.com> 13250Date: Sat Dec 29 22:34:32 2007 +0100 13251 13252 randr12: Minor cleanup. 13253 13254commit 5af6cd8dea2fa51b362201961191a17b0813fa30 13255Author: Maarten Maathuis <madman2003@gmail.com> 13256Date: Sat Dec 29 21:46:18 2007 +0100 13257 13258 randr12: Micro cleanup. 13259 13260commit 4f91dd1ae08d348b2977b8ede0908ddd2add1cb4 13261Author: Maarten Maathuis <madman2003@gmail.com> 13262Date: Sat Dec 29 20:15:25 2007 +0100 13263 13264 NVBIOS: Fix duallink detection. 13265 13266commit 559a640a542a89238afba02a78cf6399b8edc2fc 13267Author: Maarten Maathuis <madman2003@gmail.com> 13268Date: Sat Dec 29 19:34:12 2007 +0100 13269 13270 randr12: Do not remove bits in DITHER register. 13271 13272commit 4b980011742a6fa0bdeb312a7b367d6a56675fcb 13273Author: Maarten Maathuis <madman2003@gmail.com> 13274Date: Sat Dec 29 15:09:08 2007 +0100 13275 13276 NV40: Preliminary Xv Texture Adaptor. 13277 13278 - Only supports YV12. 13279 - No attributes. 13280 - Only the Y component gets bilinear filtering. 13281 - You're welcome to improve it ;-) 13282 13283commit a4c045aa14b471fcfe29cdbec0cd846b2c5be2a6 13284Author: Peter Winters <peter@whoei.org> 13285Date: Sat Dec 29 12:15:40 2007 +0100 13286 13287 Xv: nforce2 overlay can do yv12 13288 13289commit 1142b15c56a0b9dab8b929fc196a825815457374 13290Author: Maarten Maathuis <madman2003@gmail.com> 13291Date: Fri Dec 28 21:11:49 2007 +0100 13292 13293 randr12: LVDS screens use center mode. 13294 13295commit 8215e970445d99312c8fbbc31944466f70f3e186 13296Author: Andrew Randrianasulu <randrianasulu@yahoo.com> 13297Date: Fri Dec 28 20:59:17 2007 +0100 13298 13299 Fix IFC on nv4/5 by using the proper object. 13300 13301commit c6ae3fe9f077c38f616fd3ad07275dcf8cb65bbc 13302Author: Stuart Bennett <sb476@cam.ac.uk> 13303Date: Fri Dec 28 05:45:10 2007 +0000 13304 13305 Make sure the CRTC regs are unlocked when doing bios init 13306 13307 Also make the single stage PLL setter less exploitable by the whims of the hardware 13308 13309commit bb467ad4baa30fc46a05bbce11d6aa4e3dac89a9 13310Author: Stuart Bennett <sb476@cam.ac.uk> 13311Date: Fri Dec 28 08:23:17 2007 +0000 13312 13313 Fake a DCB entry for cards from the pre-DCB era 13314 13315 Also, drop DCB parsing to after the init parsing again; having an INIT_TMDS in 13316 the init scripts would be silly (the original reason for bumping DCB parsing up) 13317 13318commit cdfe4fef9bd7d0a69d9ba19a79779152baef5489 13319Author: Stuart Bennett <sb476@cam.ac.uk> 13320Date: Fri Dec 28 07:36:58 2007 +0000 13321 13322 Madness. Bios parsing for NV05 13323 13324commit d4e12684617d07dc859e774f3eff9b944bc4c89e 13325Author: Maarten Maathuis <madman2003@gmail.com> 13326Date: Thu Dec 27 20:01:14 2007 +0100 13327 13328 randr12: Minor change. 13329 13330commit ab9dc943dd5036545e3396868c0332e9b727f31f 13331Author: Maarten Maathuis <madman2003@gmail.com> 13332Date: Thu Dec 27 19:50:27 2007 +0100 13333 13334 randr12: Small change. 13335 13336commit 66d343ad594f1019eb8f4d23308eb97dd1bbd94d 13337Author: Maarten Maathuis <madman2003@gmail.com> 13338Date: Thu Dec 27 19:25:00 2007 +0100 13339 13340 randr12: minor change 13341 13342commit cf5072e0c1d9dbee4f22ef190082aa1db78dbb0d 13343Author: Maarten Maathuis <madman2003@gmail.com> 13344Date: Thu Dec 27 18:37:30 2007 +0100 13345 13346 randr12: Undo unnecesary change. 13347 13348commit a7c643601073de5a63f718ddcdc9eb9f4873f2bf 13349Author: Maarten Maathuis <madman2003@gmail.com> 13350Date: Thu Dec 27 17:12:17 2007 +0100 13351 13352 randr12: Some fixes for LVDS and an odd 7300GO. 13353 13354commit b17f8ca13604f6b085b474219843d740b31918e8 13355Author: Stuart Bennett <sb476@cam.ac.uk> 13356Date: Wed Dec 26 20:59:36 2007 +0000 13357 13358 Fix the setPLL optional code to actually compile 13359 13360commit 05b929ccd9c7cd50f60b84bf987fc5220ee83008 13361Author: Stuart Bennett <sb476@cam.ac.uk> 13362Date: Wed Dec 26 21:31:16 2007 +0000 13363 13364 Fix up getting modes for both TMDS and LVDS 13365 13366commit ba92eb062aaf8b53c7f5c1858eb876d3e2f9af3e 13367Author: Maarten Maathuis <madman2003@gmail.com> 13368Date: Wed Dec 26 21:21:15 2007 +0100 13369 13370 randr12: Remove this, since it's not essential and can cause problems. 13371 13372commit 150d3362c5e0a300c35f1cb0c7da0c98cdd6b4de 13373Author: Maarten Maathuis <madman2003@gmail.com> 13374Date: Wed Dec 26 20:28:43 2007 +0100 13375 13376 randr12: Some minor fixed for dvi and lvds. 13377 13378commit 076f86dd67d6e69bd7d2a68f372cbbbb35785a3c 13379Author: Maarten Maathuis <madman2003@gmail.com> 13380Date: Wed Dec 26 14:51:49 2007 +0100 13381 13382 randr12: Some assorted fixes for dual link outputs. 13383 13384commit 15489f15e8130ffa74bee7c52086857554a69a6b 13385Author: Stuart Bennett <sb476@cam.ac.uk> 13386Date: Tue Dec 25 23:50:10 2007 +0000 13387 13388 On some chips we shouldn't look for a 27MHz crystal 13389 13390commit 8fa59a6231e293a8b669517bf11d35dcc15f34ff 13391Author: Maarten Maathuis <madman2003@gmail.com> 13392Date: Tue Dec 25 19:43:47 2007 +0100 13393 13394 randr12: Some minor changes. 13395 13396commit 80bdd0e9f4bd664e217722a57a8db79d56968194 13397Author: Maarten Maathuis <madman2003@gmail.com> 13398Date: Tue Dec 25 11:58:09 2007 +0100 13399 13400 randr12: Restrict TMDS register backup to digital outputs. 13401 13402 - The NV11 locks up when accessing these and it has no digital outputs, so ;-) 13403 13404commit 5574e96ae3e4608b95b5dd5f2dd5676386bb4ffe 13405Author: Stuart Bennett <sb476@cam.ac.uk> 13406Date: Mon Dec 24 22:33:21 2007 +0000 13407 13408 Better (not complete) support for DCB1.1/1.2 13409 13410commit 2aa6a2347fa283dbd9b10258053602bd37880c26 13411Author: Stuart Bennett <sb476@cam.ac.uk> 13412Date: Mon Dec 24 21:00:31 2007 +0000 13413 13414 "type1" bios scripts 13415 13416commit e0c5f65c0091b0f0ac7715fbf399a1f6fea60dc6 13417Author: Stuart Bennett <sb476@cam.ac.uk> 13418Date: Mon Dec 24 18:13:47 2007 +0000 13419 13420 Rehabilitate BMP parsing 13421 13422commit 2789da6b34c9495a88d349b73cb838f4626fb152 13423Author: Stuart Bennett <sb476@cam.ac.uk> 13424Date: Mon Dec 24 22:51:51 2007 +0000 13425 13426 Fix BMP length for version 5.16 13427 13428commit 93448b40627d35fd89d8fd5629dae5f487082f43 13429Author: Stuart Bennett <sb476@cam.ac.uk> 13430Date: Mon Dec 24 15:23:41 2007 +0000 13431 13432 randr12: use new PLL code on < NV40 13433 13434commit 3d90e13c26e389f43253a5d2721ce93a628f156d 13435Author: Stuart Bennett <sb476@cam.ac.uk> 13436Date: Mon Dec 24 05:31:37 2007 +0000 13437 13438 setPLL routines for single and two stage plls 13439 13440 plus some parsing fixes 13441 13442commit eabe1e02504467c7425e62673a68f4dc7dd83aab 13443Author: Stuart Bennett <sb476@cam.ac.uk> 13444Date: Mon Dec 24 02:29:54 2007 +0000 13445 13446 Two stage nv31 PLL mnp calculator and bios limits parsing 13447 13448commit c8b50bc7dc146e01ceb67ee3633b5c8958345fcd 13449Author: Stuart Bennett <sb476@cam.ac.uk> 13450Date: Sun Dec 23 03:09:26 2007 +0000 13451 13452 Single stage PLL MNP calculator 13453 13454commit 935f809f4bc9dab37e80c59007141d7b9ad673a8 13455Author: Maarten Maathuis <madman2003@gmail.com> 13456Date: Sun Dec 23 22:05:10 2007 +0100 13457 13458 Reading into the -1 index of array is a bad idea, this should fix subchannel binding issues. 13459 13460commit f357ba26b7f560cdec03f70a2c3b79b65d70172b 13461Author: Maarten Maathuis <madman2003@gmail.com> 13462Date: Sun Dec 23 15:48:24 2007 +0100 13463 13464 randr12: Add modesetting debug option (showing most register writes). 13465 13466 - Contribution of ahuillet. 13467 13468commit 95387b9779173cc3e51a7fffc60975010253b391 13469Author: Maarten Maathuis <madman2003@gmail.com> 13470Date: Sun Dec 23 15:35:24 2007 +0100 13471 13472 NV40EXA: Suport as many of the repeat types as possible + random cleanup(randr12). 13473 13474commit 8e80f7504f872166a811b6ce20663aaf97fd7181 13475Author: Arthur Huillet <arthur.huillet@free.fr> 13476Date: Sun Dec 23 13:49:24 2007 +0100 13477 13478 Revert a manifestation of my stupidity. 13479 13480commit 86a3a5fd40005a4860d11f0046444ba232db0d1e 13481Author: Arthur Huillet <arthur.huillet@free.fr> 13482Date: Sun Dec 23 13:47:08 2007 +0100 13483 13484 Xv: added an attribute to change the CRTC the overlay is on - useful with dualhead 13485 13486commit 129f04612cdbad756dbe213f77d2f12482b62c47 13487Author: Arthur Huillet <arthur.huillet@free.fr> 13488Date: Sat Dec 22 23:32:36 2007 +0100 13489 13490 aunedsoau~ 13491 13492commit 26953280895698a94c1f886cf7c0d2f389167b1c 13493Author: Maarten Maathuis <madman2003@gmail.com> 13494Date: Sun Dec 23 11:46:17 2007 +0100 13495 13496 randr12: Extend dvi clock setting to nv3x also + cleanup. 13497 13498commit 063d1bd50b164e841e3a725b75ac243827aede50 13499Author: Maarten Maathuis <madman2003@gmail.com> 13500Date: Sun Dec 23 10:46:09 2007 +0100 13501 13502 randr12: Actually load the default palette on modeset. 13503 13504commit 810901abb50a54b8f1eb73029f5348805a53531d 13505Author: Maarten Maathuis <madman2003@gmail.com> 13506Date: Sun Dec 23 10:32:18 2007 +0100 13507 13508 randr12: Simplify code a bit by using the more sane code for NV4x also. 13509 13510commit eb8441824ea2a7dc9a1d7a86a4305fae48a26a0c 13511Author: Maarten Maathuis <madman2003@gmail.com> 13512Date: Sun Dec 23 01:02:18 2007 +0100 13513 13514 Minor cleanup. 13515 13516commit 9eef67abaf8beb9ed93eb5b6ecdf7255f5b06e4c 13517Author: Maarten Maathuis <madman2003@gmail.com> 13518Date: Sun Dec 23 00:18:13 2007 +0100 13519 13520 randr12: Always enable MPLL and NVPLL on pre-NV41 cards. 13521 13522commit fa82eb23904bc1a70209778d2aafcbef6f27cfa1 13523Author: Maarten Maathuis <madman2003@gmail.com> 13524Date: Sun Dec 23 00:07:01 2007 +0100 13525 13526 randr12: Put all NV2x (and lower) cards into restricted mode. 13527 13528commit 33d7313d7dca4501b7ac2e078a87f3a10ab1b501 13529Author: Maarten Maathuis <madman2003@gmail.com> 13530Date: Sat Dec 22 23:55:19 2007 +0100 13531 13532 randr12: Correct CRTC selection for pre-NV40 cards. 13533 13534commit d6243fe33223d2934c1a819c58bb58b9ebe15c25 13535Author: Maarten Maathuis <madman2003@gmail.com> 13536Date: Sat Dec 22 23:39:51 2007 +0100 13537 13538 randr12: Older cards do have switchable CRTC's, i was again wrong. 13539 13540commit a96553544ae76239e394ab34b33814a57252d31c 13541Author: Maarten Maathuis <madman2003@gmail.com> 13542Date: Sat Dec 22 23:23:32 2007 +0100 13543 13544 randr12: Minor cleanup to clock writing for pre-NV40 cards. 13545 13546commit 9b63f42bc7da9b45ecbbd32ef81cae55d2a57220 13547Author: Arthur Huillet <arthur.huillet@free.fr> 13548Date: Sat Dec 22 18:44:36 2007 +0100 13549 13550 NV10EXA: disabled A8+A8 hack on PPC machines. 13551 13552commit 308fec0111d182fb17536ff2746e079c1af32d5a 13553Author: Ben Skeggs <skeggsb@gmail.com> 13554Date: Sat Dec 22 23:18:57 2007 +1100 13555 13556 nv30: Don't touch 0x1a4 13557 13558commit 62916c7c8c6b467436e4628c7d08a4e0de1f21f3 13559Author: Maarten Maathuis <madman2003@gmail.com> 13560Date: Sat Dec 22 12:02:31 2007 +0100 13561 13562 randr12: Undo a previous mistake for NV1x/NV2x. 13563 13564commit 6f9eb17637e2c3b8d49a07332f06a0102d2fe016 13565Author: Stuart Bennett <sb476@cam.ac.uk> 13566Date: Fri Dec 21 20:18:57 2007 +0000 13567 13568 Beginnings of PLL limits for BMP bioses 13569 13570 And the correct way to use the debug_modes member (still disabled for now) 13571 13572commit 73bc1db51e9431c5d1861d8a753aa8dd9abe9882 13573Author: Stuart Bennett <sb476@cam.ac.uk> 13574Date: Sat Dec 22 02:57:40 2007 +0000 13575 13576 On demand PLL limit parsing for NV40 13577 13578 Also some nv_bios message formatting consistancy pedantry 13579 13580commit 57b5dbfac4ff65258e94a17717815156a54cc273 13581Author: Maarten Maathuis <madman2003@gmail.com> 13582Date: Sat Dec 22 01:29:38 2007 +0100 13583 13584 NV40EXA: Fix borders of images that recieved arbitrary rotation. 13585 13586 - It was getting padded up to the edge of the smallest square that could fit the rotated image. 13587 - Please try this on older cards as well. 13588 13589commit 39ef0262be49feaed2ba1c6caeab5134181ef0b9 13590Author: Maarten Maathuis <madman2003@gmail.com> 13591Date: Sat Dec 22 01:26:44 2007 +0100 13592 13593 randr12: Some minor changes. 13594 13595commit 371b6e52bef743f14003e3783ea3dbc624050594 13596Author: Maarten Maathuis <madman2003@gmail.com> 13597Date: Fri Dec 21 18:02:33 2007 +0100 13598 13599 randr12: Restrict crtc's on nv1x and nv2x. 13600 13601 - If anyone can show a mmio-trace showing how to switch them (on cards that have a dcb table claiming they can), then please do. 13602 13603commit c2ac36927827439cf92ff53bbf2414bfaf45a6c3 13604Author: Maarten Maathuis <madman2003@gmail.com> 13605Date: Fri Dec 21 00:05:37 2007 +0100 13606 13607 randr12: Detect switchable crtc's. 13608 13609commit 33a2cf86520f42d549d03c46134b818e90d51a41 13610Author: Peter Winters <peter@whoei.org> 13611Date: Thu Dec 20 23:47:11 2007 +0100 13612 13613 NV10 EXA: use the proper variable 13614 13615commit 71c3b8ff09adcadb10d800d3696762c3ef58be3d 13616Author: Stuart Bennett <sb476@cam.ac.uk> 13617Date: Wed Dec 19 23:44:23 2007 +0000 13618 13619 randr12: fix old and new style scaling for LVDS, and use an enum for the scaling modes 13620 13621commit 6bea51b905a461b4e703053082ab45d9e1074112 13622Author: Maarten Maathuis <madman2003@gmail.com> 13623Date: Thu Dec 20 00:07:29 2007 +0100 13624 13625 randr12: Fix secondary (analog) outputs with ffs(or) != 3 on some NV4x cards. 13626 13627commit e0e04eb561f4b127087befc4e3768d8012a82bfe 13628Author: Stuart Bennett <sb476@cam.ac.uk> 13629Date: Wed Dec 19 04:29:34 2007 +0000 13630 13631 randr12: don't use invalid crtc pointer for lvds dpms 13632 13633commit a68ac97979a3178ea210470bbb3d64f3f9325e07 13634Author: Stuart Bennett <sb476@cam.ac.uk> 13635Date: Wed Dec 19 04:00:14 2007 +0000 13636 13637 Turn LVDS panel on again, after resetting, when we're shutting down 13638 13639commit 1cfd659fc1a5f8dc2faba30155fca46bbe31f3a4 13640Author: Stuart Bennett <sb476@cam.ac.uk> 13641Date: Tue Dec 18 16:58:57 2007 +0000 13642 13643 minor link_head_and_output cleanup 13644 13645commit 73a5d09b3e9fa13d04d6696ca38b6402795acc3d 13646Author: Stuart Bennett <sb476@cam.ac.uk> 13647Date: Wed Dec 19 03:43:59 2007 +0000 13648 13649 randr12: CRTC member wasn't big enough 13650 13651 yay gcc. could fix console colour breakage 13652 13653commit 0a63198195f689b32388b99db61ff942e16fbc6d 13654Author: Maarten Maathuis <madman2003@gmail.com> 13655Date: Tue Dec 18 23:43:57 2007 +0100 13656 13657 NV10EXA: Fix rotation. 13658 13659commit f2a3807618cfeac01d7d28e9d76721ad61495f7f 13660Author: Maarten Maathuis <madman2003@gmail.com> 13661Date: Tue Dec 18 23:23:59 2007 +0100 13662 13663 randr12: Add new pll calculation for NV4x cards (from NVClock). 13664 13665commit ebf96c8b651d1a0dc5a19576cf57efcbb989d51d 13666Author: Maarten Maathuis <madman2003@gmail.com> 13667Date: Tue Dec 18 21:38:56 2007 +0100 13668 13669 randr12: Hopefully reset outputs better between mode changes. 13670 13671commit 5561f54f94e0b31394ecaf461e342470ca768281 13672Author: Maarten Maathuis <madman2003@gmail.com> 13673Date: Tue Dec 18 07:49:05 2007 +0100 13674 13675 Some endianess fixes for the previously committed code. 13676 13677commit 11104ce9c499f30fb5abe29e65ba1918c563f4cd 13678Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 13679Date: Tue Dec 18 07:40:08 2007 +0100 13680 13681 Rename NV30TCL to NV34TCL in preparation of the merge. 13682 13683commit 3e6ea7af5b1613b9e0e4e29433910d998def4899 13684Author: Maarten Maathuis <madman2003@gmail.com> 13685Date: Mon Dec 17 23:13:12 2007 +0100 13686 13687 Add pll limits bios detection code for NV4x cards. 13688 13689commit e56512d283f401a75c6165a0a02e8a7a5905fc4c 13690Author: Maarten Maathuis <madman2003@gmail.com> 13691Date: Mon Dec 17 19:45:42 2007 +0100 13692 13693 NV30EXA: Fix rotation. 13694 13695commit 9ff42b7a83da06a3b9b7076ca888ea447eaa16f3 13696Author: Maarten Maathuis <madman2003@gmail.com> 13697Date: Mon Dec 17 17:50:11 2007 +0100 13698 13699 randr12: Fixes + cleanup. 13700 13701commit 83d99bc3e15d11d3476c54053f6bbe0ff9f094be 13702Author: Maarten Maathuis <madman2003@gmail.com> 13703Date: Mon Dec 17 00:40:24 2007 +0100 13704 13705 randr12: Some minor additions. 13706 13707commit 60c08fedec09dd036b4165881b3001874c4d349a 13708Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 13709Date: Sun Dec 16 23:40:26 2007 +0100 13710 13711 nv30 exa : handle gart case. 13712 13713commit efb10849e136048c9c49cce752d4073402219c8c 13714Author: Maarten Maathuis <madman2003@gmail.com> 13715Date: Sun Dec 16 23:26:49 2007 +0100 13716 13717 Undo a bad change. 13718 13719commit dd8beae22883822ea3555122b0352d4e3cf2c266 13720Author: Maarten Maathuis <madman2003@gmail.com> 13721Date: Sun Dec 16 21:35:18 2007 +0100 13722 13723 A little more cleanup. 13724 13725commit 9ff169ba62dcb2eeefce8a020060de731b72cb6c 13726Author: Maarten Maathuis <madman2003@gmail.com> 13727Date: Sun Dec 16 21:32:32 2007 +0100 13728 13729 More cleanup. 13730 13731commit 2d03d4b999a0a37ff2f9fc87bd20959e789a152e 13732Author: Maarten Maathuis <madman2003@gmail.com> 13733Date: Sun Dec 16 21:19:13 2007 +0100 13734 13735 randr12: Select sensible maximum heights and widths for framebuffer. 13736 13737commit 0e47fa4ab35e529926f5fdccdb44f9d054e1886a 13738Author: Maarten Maathuis <madman2003@gmail.com> 13739Date: Sun Dec 16 21:17:04 2007 +0100 13740 13741 Some reindenting and minor cleanups. 13742 13743commit 2be4b73905dc5767e593bf4753607144fe67b7ab 13744Author: Maarten Maathuis <madman2003@gmail.com> 13745Date: Sun Dec 16 20:24:26 2007 +0100 13746 13747 Some cleanup. 13748 13749commit 6686adbea6638dd894eef697e044941352cc2674 13750Author: Maarten Maathuis <madman2003@gmail.com> 13751Date: Sun Dec 16 20:10:39 2007 +0100 13752 13753 randr12: Restrict digital outputs with ffs(or) == 1 to primary crtc + some cleanup. 13754 13755 - The blob does not reveal what the possible way is (so far), and guesswork didn't help either. 13756 13757commit 9c8da2f562ca571a56224d7fecfb9c3fb3a8d3d0 13758Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 13759Date: Sun Dec 16 18:33:52 2007 +0100 13760 13761 And of course, that comment has to go. 13762 13763commit 02e9a01064930a7081569474be9544fa1f20a0f4 13764Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 13765Date: Sun Dec 16 18:32:52 2007 +0100 13766 13767 Fix front buffer alignment issues (probably needed because of tiling). 13768 13769commit b134b8f1552066b6706521a9c85d9241d99b3ccc 13770Author: Maarten Maathuis <madman2003@gmail.com> 13771Date: Sun Dec 16 13:41:38 2007 +0100 13772 13773 randr12: Fix compile error. 13774 13775commit ec0c9cc9a69c00a960e98e796c00d35d996cf360 13776Author: Maarten Maathuis <madman2003@gmail.com> 13777Date: Sun Dec 16 12:55:29 2007 +0100 13778 13779 randr12: A minor change to OUTPUT register. 13780 13781commit 69782a57285bb53df73c7f45d29882ae990f6054 13782Author: Patrice Mandin <pmandin@caramail.com> 13783Date: Sun Dec 16 00:09:45 2007 +0100 13784 13785 nv30: revert viewport clip change, add myself, add unknown command 0x2b8 (viewport origin?) 13786 13787commit 8dd238f3ce2e7eda7151670ea50a75abdb5b6217 13788Author: Maarten Maathuis <madman2003@gmail.com> 13789Date: Sat Dec 15 23:49:58 2007 +0100 13790 13791 randr12: Some minor changes. 13792 13793commit 08a50d878571f7c9579b15703a08c5c2e52924f2 13794Author: Maarten Maathuis <madman2003@gmail.com> 13795Date: Sat Dec 15 19:54:11 2007 +0100 13796 13797 randr12: Some minor reordering of load/save functions + don't lock crtc when leaving X. 13798 13799commit 4209c06800b97e98261c10e0a625a68f7352ca6c 13800Author: Maarten Maathuis <madman2003@gmail.com> 13801Date: Sat Dec 15 14:48:04 2007 +0100 13802 13803 randr12: Also support accelerated rotation on older xservers (and default code). 13804 13805commit 933b40a2911486a78615984fb17e9ea6757fcfa8 13806Author: Ben Skeggs <skeggsb@gmail.com> 13807Date: Sat Dec 15 23:45:03 2007 +1100 13808 13809 nv40: just for stillunknown... 13810 13811 Fix composite transforms. This is getting to a point where it may be 13812 worth doing this in the vertex program perhaps. For now this works, 13813 it'll do. 13814 13815commit 7212400070264e27a53564a7f5fc95dd9c99f82e 13816Author: Ben Skeggs <skeggsb@gmail.com> 13817Date: Sat Dec 15 23:44:17 2007 +1100 13818 13819 Revert "NV40EXA: Fix composite rotation." 13820 13821 This reverts commit 02d4538b207cf318737ace8c3d8ded76a54e5886. 13822 13823commit 1a622257687a499ce74b3a608618d957784dccf0 13824Author: Maarten Maathuis <madman2003@gmail.com> 13825Date: Sat Dec 15 13:39:23 2007 +0100 13826 13827 randr12: Avoid a vram/memory leak in some situations. 13828 13829commit 02d4538b207cf318737ace8c3d8ded76a54e5886 13830Author: Maarten Maathuis <madman2003@gmail.com> 13831Date: Sat Dec 15 13:35:45 2007 +0100 13832 13833 NV40EXA: Fix composite rotation. 13834 13835 - Thanks to radeon (code) for inspiration and sample code. 13836 13837commit 77d47348dfe01ca3972a683d3c478f910301e910 13838Author: Maarten Maathuis <madman2003@gmail.com> 13839Date: Sat Dec 15 12:02:36 2007 +0100 13840 13841 randr12: Add some ifdefs to make compiling better on older xserver's. 13842 13843commit dbb7efc5b81145ad61d8e6502b918a1e909f3cfb 13844Author: Patrice Mandin <pmandin@caramail.com> 13845Date: Sat Dec 15 11:44:41 2007 +0100 13846 13847 nv30: need to set default depth range 13848 13849commit 6ab8436df0865aab7f12cd8527600353b4728df7 13850Author: Patrice Mandin <pmandin@caramail.com> 13851Date: Sat Dec 15 11:37:43 2007 +0100 13852 13853 nv30: only 4 texture units 13854 13855commit 370d97a4578528c26c3f1546a9522caf40aa2151 13856Author: Patrice Mandin <pmandin@caramail.com> 13857Date: Sat Dec 15 10:34:26 2007 +0100 13858 13859 nv30: fix viewport clipping setup 13860 13861commit 1190d046dfef97dea04d72c993f90a203d81d86c 13862Author: Maarten Maathuis <madman2003@gmail.com> 13863Date: Sat Dec 15 02:00:37 2007 +0100 13864 13865 Fix ugly bug, that creeped in. 13866 13867commit 46f321985ae57457edcedf80f82ba23e53fd9cd2 13868Author: Maarten Maathuis <madman2003@gmail.com> 13869Date: Sat Dec 15 01:44:59 2007 +0100 13870 13871 randr12: Preliminary support for rotation, please read notes. 13872 13873 - It works, but it's slow, very slow. 13874 - Each crtc now has it's own cursor map (fixing the bugs you never saw). 13875 - Dualhead with one rotated screen may show minor corruption near the edges, this will disappear if a fullscreen app opens. 13876 - Acceleration is possible, BUT: 13877 - You need a git xserver. 13878 - Enable NOUVEAU_EXA_PIXMAPS in nouveau_local.h (*maybe* this will become default later). 13879 - Most of the composite hooks have problems with rotation, so left and right won't work, inverted does work for me. 13880 - So please fix the composite hooks ;-) 13881 - Note: No slowdowns are apparent when using proper acceleration. 13882 13883commit 0f07ec3f681b87762fbe7328f47b628711bdbaae 13884Author: Stuart Bennett <sb476@cam.ac.uk> 13885Date: Sat Dec 15 00:01:04 2007 +0000 13886 13887 Better restoration of TMDS regs on LVDS 13888 13889commit 34be8d005fbfff373cc7fe52ca2fa09f5cd5ef00 13890Author: Maarten Maathuis <madman2003@gmail.com> 13891Date: Fri Dec 14 23:58:25 2007 +0100 13892 13893 randr12: It seems DVI on laptops wasn't as strange after all. 13894 13895commit d678932a1818c5c5f4a4f3097992626ce9620c0e 13896Author: Stuart Bennett <sb476@cam.ac.uk> 13897Date: Fri Dec 14 01:40:14 2007 +0000 13898 13899 INIT_COPY_NV_REG and INIT_ZM_REG_GROUP_ADDRESS_LATCHED opcodes 13900 13901 Thanks to RAOF for the trace with these 13902 13903 Also correct a doc oopsy, and a bunch of superfluous whitespace 13904 13905commit 5b9dc7ad6eb2f16c4ca8824c2a19c85bbe1f126b 13906Author: Stuart Bennett <sb476@cam.ac.uk> 13907Date: Thu Dec 13 02:30:06 2007 +0000 13908 13909 Invert 18/24 bit interface flag, so that the default value (0) does not imply dithering 13910 13911commit a878b48c7b5a7aafac9aaa877706f5c7570d3057 13912Author: Stuart Bennett <sb476@cam.ac.uk> 13913Date: Thu Dec 13 01:10:12 2007 +0000 13914 13915 Name bios opcodes 4f and 50 13916 13917 Extrapolating from previous names, but marked as non canon 13918 13919commit 59b46e695b1badbf00f2c03bc78ce58f6887f739 13920Author: Stuart Bennett <sb476@cam.ac.uk> 13921Date: Thu Dec 13 01:03:25 2007 +0000 13922 13923 Get VBIOS from pNv (cleanup) 13924 13925 And a minor doc correction 13926 13927commit c45930d0819668612547b4bf1a8d963bfe40e38b 13928Author: Stuart Bennett <sb476@cam.ac.uk> 13929Date: Thu Dec 13 00:15:21 2007 +0000 13930 13931 randr12: some minor cleanups, no actual changes 13932 13933commit 0e778b2d6970f761e5c0b5eb3054e91589cd2c46 13934Author: Stuart Bennett <sb476@cam.ac.uk> 13935Date: Wed Dec 12 15:54:53 2007 +0000 13936 13937 Setup NV40 LVDS 13938 13939commit 269d83cb46075971fd8dc5402589d4da3eb92f22 13940Author: Stuart Bennett <sb476@cam.ac.uk> 13941Date: Wed Dec 12 22:20:43 2007 +0000 13942 13943 INIT_CONDITION_TIME opcode 13944 13945commit 9b3072d195aae383a993526b3ab4350c500c4277 13946Author: Stuart Bennett <sb476@cam.ac.uk> 13947Date: Wed Dec 12 18:08:07 2007 +0000 13948 13949 Opcode 4F 13950 13951commit 8ee100ced7da347904e45ce590350d7b7c47ed36 13952Author: Stuart Bennett <sb476@cam.ac.uk> 13953Date: Mon Dec 10 22:50:40 2007 +0000 13954 13955 FP BIOS mode finding for nv40 13956 13957commit 9131985f5c359bdb077eb52649369a5c06831544 13958Author: Stuart Bennett <sb476@cam.ac.uk> 13959Date: Tue Dec 11 23:23:27 2007 +0000 13960 13961 CR57 doc addition 13962 13963commit d595eb86790432f6a75dec3388080ffc9ffff8d8 13964Author: Maarten Maathuis <madman2003@gmail.com> 13965Date: Wed Dec 12 22:12:33 2007 +0100 13966 13967 randr12: A good guess at the lower bits of SEL_CLK (i hope ;-)). 13968 13969commit c5bc9899c218abe801e024919c89896e8183dff2 13970Author: Maarten Maathuis <madman2003@gmail.com> 13971Date: Wed Dec 12 21:03:34 2007 +0100 13972 13973 randr12: Add some extra safeties (to mode checking). 13974 13975commit 43c05d306485eb5d8c8b9b39f9bd4a3f470f20a2 13976Author: Maarten Maathuis <madman2003@gmail.com> 13977Date: Tue Dec 11 21:43:02 2007 +0100 13978 13979 randr12: Enable dithering on lvds when needed. 13980 13981commit 8752f18d27a6095485e36f4fb20e9252f27a1e1b 13982Author: Maarten Maathuis <madman2003@gmail.com> 13983Date: Tue Dec 11 20:22:23 2007 +0100 13984 13985 randr12: SEL_CLK attempt 10, return to simplicity, which may actually work ;-) 13986 13987commit c1227d3a2b9194d50a4274a48fa0b3bcf2628e36 13988Author: Maarten Maathuis <madman2003@gmail.com> 13989Date: Tue Dec 11 17:52:39 2007 +0100 13990 13991 randr12: (mostly) Some fixes for NV40. 13992 13993 - The NV40 is a strange card, as it partially behaves like an NV3x. 13994 13995commit 34d9e971e17e763aee96b823bb34da54a281339c 13996Author: Maarten Maathuis <madman2003@gmail.com> 13997Date: Tue Dec 11 11:50:28 2007 +0100 13998 13999 randr12: Fix SEL_CLK assignment again (for non-mobile cards) + fix crosswiring. 14000 14001 - I made a good guess for mobile cards, but i need more mmio-traces. 14002 - I accidentally broke crosswiring for my card (and other's like it). 14003 14004commit bf28c3e7d9976f8695198a938397f001de62836b 14005Author: Maarten Maathuis <madman2003@gmail.com> 14006Date: Tue Dec 11 00:04:38 2007 +0100 14007 14008 randr12: Fix an earlier mistake, make a decent guess about dvi on laptops. 14009 14010 - I could use some more mmiotraces of LVDS+DVI on NV4x. 14011 14012commit 58531ebd4a72368b13d3c1564110ef85fb6d6644 14013Author: Maarten Maathuis <madman2003@gmail.com> 14014Date: Mon Dec 10 20:43:26 2007 +0100 14015 14016 randr12: Improve crosswiring for digital outputs + minor changes and cleanup. 14017 14018commit a388cc5fe21f8692e8ee1dbd71caea88ea2ca5df 14019Author: Maarten Maathuis <madman2003@gmail.com> 14020Date: Mon Dec 10 16:31:47 2007 +0100 14021 14022 randr12: Properly set SEL_CLK based on bus, not based on output. 14023 14024 - This hopefully fixes non-working dvi on some desktop cards and on mobile cards. 14025 14026commit f96a6c2d5d251af560de673ccd91b88b3f5cc535 14027Author: Maarten Maathuis <madman2003@gmail.com> 14028Date: Mon Dec 10 00:03:49 2007 +0100 14029 14030 randr12: Add config option for scaling mode. 14031 14032commit fc73ddad60c551289629be5f2f87b39e26f42674 14033Author: Maarten Maathuis <madman2003@gmail.com> 14034Date: Sun Dec 9 22:46:39 2007 +0100 14035 14036 randr12: Some cleanup. 14037 14038commit 2543e301402b04ae94f349e21a7760beb6f5bdc6 14039Author: Stuart Bennett <sb476@cam.ac.uk> 14040Date: Sun Dec 9 19:58:58 2007 +0000 14041 14042 Bios register write function special case and some INIT_RESET improvements 14043 14044 nv32_wr weirdness only seen on NV40+, but it shouldn't hurt for earlier cards 14045 14046commit 4408ce8d2f902a9ace550a045e6022df2936fbce 14047Author: Stuart Bennett <sb476@cam.ac.uk> 14048Date: Sun Dec 9 17:09:03 2007 +0000 14049 14050 Some renaming for bios indexed port io 14051 14052commit 9eb4294e3f5814f675b403078761bf002540868b 14053Author: Maarten Maathuis <madman2003@gmail.com> 14054Date: Sun Dec 9 18:36:26 2007 +0100 14055 14056 randr12: Revert dither value to that used by nv. 14057 14058commit cfbe4bdef7f4ba958462743dc28bf253b456c1a3 14059Author: Maarten Maathuis <madman2003@gmail.com> 14060Date: Sun Dec 9 18:34:10 2007 +0100 14061 14062 randr12: Set FP_CONTROL correctly for NV2x cards. 14063 14064commit d5dd3afc50e491689f765a5e6a72892efa1768a3 14065Author: Maarten Maathuis <madman2003@gmail.com> 14066Date: Sun Dec 9 18:10:29 2007 +0100 14067 14068 randr12: Improve output selection. 14069 14070 - Now detect if there is a dual link capable output at all. 14071 - NV2x cards can't switch crtc on analog outputs. 14072 14073commit fda83a3c489469d2bb4cbd20c205016b19dfdec5 14074Author: Stuart Bennett <sb476@cam.ac.uk> 14075Date: Sun Dec 9 15:51:02 2007 +0000 14076 14077 Additional LVDS bios script sanity check 14078 14079 Avoids current incorrect situation when LVDS attempted on NV40 14080 14081commit 9932e36d4cf2988d1f6f40d615cd247acc21b592 14082Author: Maarten Maathuis <madman2003@gmail.com> 14083Date: Sun Dec 9 16:29:14 2007 +0100 14084 14085 randr12: Fix minor (potentional) bug. 14086 14087commit 3f83d7308f3b0aca4943262b0ba2715a32b61d58 14088Author: Maarten Maathuis <madman2003@gmail.com> 14089Date: Sun Dec 9 12:47:55 2007 +0100 14090 14091 randr12: Add noscale mode and some misc fixes. 14092 14093commit d688008b38609814de9c5c36a2c8bc569455d14e 14094Author: Maarten Maathuis <madman2003@gmail.com> 14095Date: Sat Dec 8 18:34:04 2007 +0100 14096 14097 randr12: Fix typo, it's "our" not "or". 14098 14099commit 0f7c018af0c7f929c2bf8efb12c510d6e3170885 14100Author: Maarten Maathuis <madman2003@gmail.com> 14101Date: Sat Dec 8 18:25:41 2007 +0100 14102 14103 randr12: Assuming an output always has a crtc is a bad idea. 14104 14105 - Don't rely on current crtc to restore mode. 14106 14107commit 86f248cc2e52de652b1768e15a4ba62eefefd81c 14108Author: Maarten Maathuis <madman2003@gmail.com> 14109Date: Sat Dec 8 16:49:21 2007 +0100 14110 14111 randr12: Avoid that two outputs have the same clock selected. 14112 14113commit 6be67d6cf0f047ceba0aa62989dc44bb58b0db87 14114Author: Maarten Maathuis <madman2003@gmail.com> 14115Date: Sat Dec 8 15:50:39 2007 +0100 14116 14117 randr12: Fix bug in sel_clk assignment. 14118 14119commit ddfec2f1c749af227dd57dbde6751320ba72adb6 14120Author: Stuart Bennett <sb476@cam.ac.uk> 14121Date: Sat Dec 8 02:18:41 2007 +0000 14122 14123 Base error :) 14124 14125commit dff95ae101f701d737c8d2134181693f36178beb 14126Author: Stuart Bennett <sb476@cam.ac.uk> 14127Date: Fri Dec 7 16:56:26 2007 +0000 14128 14129 Fix failure to cope with varying table lengths 14130 14131commit 10731d2a8fe06fa8675c26d201615c2a019de737 14132Author: Stuart Bennett <sb476@cam.ac.uk> 14133Date: Fri Dec 7 02:54:36 2007 +0000 14134 14135 Add a new valid register range, which appears in some gf7 biosen. 14136 14137 NVClock has it down as appearing in gf8 as a new PBUS range, 14138 but the use on gf7 looks PBUS-like too 14139 14140commit b25049233613eff99d767a8ec0c97b5140088213 14141Author: Stuart Bennett <sb476@cam.ac.uk> 14142Date: Fri Dec 7 03:13:52 2007 +0000 14143 14144 Some more bios script opcodes 14145 14146 Thanks to swany for the trace for these 14147 14148commit 6da4923088e8ca1d2de464030205b41d893e65b3 14149Author: Stuart Bennett <sb476@cam.ac.uk> 14150Date: Fri Dec 7 02:17:37 2007 +0000 14151 14152 Some M table parsing, and defer I table execution 14153 14154commit 7e5e477b67279e00d1e56736f9c953972bda4dec 14155Author: Stuart Bennett <sb476@cam.ac.uk> 14156Date: Fri Dec 7 00:18:25 2007 +0000 14157 14158 Improve TMDS04 setting for LVDS 14159 14160commit dd757910f546a97bb42dbefdc7eae058980b1ffb 14161Author: Stuart Bennett <sb476@cam.ac.uk> 14162Date: Thu Dec 6 23:37:01 2007 +0000 14163 14164 TMDS regs setting for pre NV40 14165 14166commit 38f139d473eb3201daaac47e655516f65c7528a6 14167Author: Stuart Bennett <sb476@cam.ac.uk> 14168Date: Fri Dec 7 00:30:14 2007 +0000 14169 14170 BIOS version parsing 14171 14172commit 20b6be6c95a71a41456c4c9b5cb1a1e0238f2e70 14173Author: Stuart Bennett <sb476@cam.ac.uk> 14174Date: Thu Dec 6 17:01:49 2007 +0000 14175 14176 Refactor BIT specific details out of TMDS setting 14177 14178commit f665b4737f1fbbd5e4a6495b561e723f17bc6137 14179Author: Stuart Bennett <sb476@cam.ac.uk> 14180Date: Thu Dec 6 18:48:04 2007 +0000 14181 14182 The 'T' table is clearly a TMDS table, call it as such 14183 14184commit 3759cc9d602b0cd933a0ee7e6f64626822383e19 14185Author: Stuart Bennett <sb476@cam.ac.uk> 14186Date: Thu Dec 6 05:03:49 2007 +0000 14187 14188 The TMDS scripts being non-stubs does not seem to matter - they're still not executed 14189 14190 Message left as log clutter in case there's some future case of broken TMDS when the scripts might be useful 14191 14192commit d6e896dac1018fc3190a5a5283a405c4b031d1ba 14193Author: Stuart Bennett <sb476@cam.ac.uk> 14194Date: Thu Dec 6 04:14:56 2007 +0000 14195 14196 Test for LVDS_INIT script 14197 14198commit c50bc602861fdcdb238e1a4db1146b4870ce83ba 14199Author: Maarten Maathuis <madman2003@gmail.com> 14200Date: Thu Dec 6 21:48:13 2007 +0100 14201 14202 randr12: Small fix for dual link dvi. 14203 14204commit 457d4217624cc73df22ca7f7db0134c54ebac5e5 14205Author: Maarten Maathuis <madman2003@gmail.com> 14206Date: Thu Dec 6 19:54:08 2007 +0100 14207 14208 randr12: Improve compatibility with DCB 2.0 14209 14210commit 5046cb7b3d1af89b72327660932ae06c06d0869b 14211Author: Stuart Bennett <sb476@cam.ac.uk> 14212Date: Thu Dec 6 02:28:18 2007 +0000 14213 14214 randr12: oops, lost the restore case 14215 14216commit 54c7bdf0f3e53c23c0853eac8243407d280d0732 14217Author: Stuart Bennett <sb476@cam.ac.uk> 14218Date: Thu Dec 6 01:54:45 2007 +0000 14219 14220 randr12: a common function to fix TMDS04, and new use of LVDS_RESET 14221 14222 Plus some corrections/additions to call_lvds_script, and a minor CR5758 doc update 14223 14224commit a7811521ed5d227a18af23e2e761ea39fb7eb86f 14225Author: Maarten Maathuis <madman2003@gmail.com> 14226Date: Thu Dec 6 00:28:49 2007 +0100 14227 14228 randr12: Fix comment. 14229 14230commit 4771b4bd4b30f404fb087dd1fd6eac0e12fdd041 14231Author: Maarten Maathuis <madman2003@gmail.com> 14232Date: Thu Dec 6 00:25:48 2007 +0100 14233 14234 randr12: Some assorted fixes, which hopefully improve things for dual link dvi. 14235 14236commit 4d3a015631a0d95392daeec6fd8e99c65c83cb53 14237Author: Maarten Maathuis <madman2003@gmail.com> 14238Date: Wed Dec 5 20:41:27 2007 +0100 14239 14240 randr12: replace preferred_ramdac with preferred_output. 14241 14242commit d3036012db63f9a386d4c1fd8e624b86bcd6c48b 14243Author: Maarten Maathuis <madman2003@gmail.com> 14244Date: Wed Dec 5 20:27:57 2007 +0100 14245 14246 randr12: Clean up the unneeded ramdac variable. 14247 14248commit 37f3a06be4f50c34ee8b82617e5e6eca3a4d66fe 14249Author: Maarten Maathuis <madman2003@gmail.com> 14250Date: Wed Dec 5 19:03:52 2007 +0100 14251 14252 randr12: Pick right register set, to correctly "guess" the bios/VT clock. 14253 14254 - A wrong register set leads to improper detection of crosswiring, thus selecting the wrong clock. 14255 - I noticed a 0.675 MHz difference in the clock (which is vpll2), when instead i should have had vpll1. 14256 14257commit b843dbddbd496ab7d374018e6e05ac1a05ea00e2 14258Author: Maarten Maathuis <madman2003@gmail.com> 14259Date: Wed Dec 5 16:51:04 2007 +0100 14260 14261 randr12: Small fix. 14262 14263commit 4b049994f32058e380543cee81b1d0b76658ea9e 14264Author: Maarten Maathuis <madman2003@gmail.com> 14265Date: Wed Dec 5 16:47:07 2007 +0100 14266 14267 randr12: Final commit, restore TMDS regs before everything else. 14268 14269 - Use the bios to program the tmds registers. 14270 - Restore TMDS registers in nv_crtc_restore, to make sure it's early enough (see code note). 14271 - It works for me, let me know if it breaks stuff ;-) 14272 14273commit 6ff8c1cb2014c7984503a8481cf6ad4f9cccf79a 14274Author: Maarten Maathuis <madman2003@gmail.com> 14275Date: Wed Dec 5 07:57:30 2007 +0100 14276 14277 randr12: Intermediate bios based flat panel init commit, still a few minor issues. 14278 14279commit c14c7deb725c2509c804bae72faa584109f3021c 14280Author: Stuart Bennett <sb476@cam.ac.uk> 14281Date: Wed Dec 5 00:35:35 2007 +0000 14282 14283 Add LVDS script function, and use it for DPMS 14284 14285commit c587731d08bbfd58ff5c2725b7aed4e6477e9f7f 14286Author: Stuart Bennett <sb476@cam.ac.uk> 14287Date: Wed Dec 5 00:22:22 2007 +0000 14288 14289 Parse the LVDS Manufacturer table 14290 14291 Use dual link flag to set bit 31 of FP_CONTROL properly 14292 14293commit affcf6988f1af2c3ea8c47a2f1fdf608cbafcfce 14294Author: Stuart Bennett <sb476@cam.ac.uk> 14295Date: Tue Dec 4 20:31:06 2007 +0000 14296 14297 Modify DCB dual link flag, and DCB config parsing for LVDS 14298 14299 The DCB duallink flag indicates that this setup may be used for duallink; 14300 it obviously can't know whether a duallink display is plugged or not 14301 14302 Use LVDS DCB conf flags for mode detection / selection 14303 14304commit c822e32d95a7afa3d5ef076f97179a1843835f12 14305Author: Maarten Maathuis <madman2003@gmail.com> 14306Date: Tue Dec 4 09:06:10 2007 +0100 14307 14308 randr12: Add fullscreen gpu scaling (only as randr property for the moment). 14309 14310commit a1054c7bef3594f6876d611dde1cf3f0528fb8e1 14311Author: Stuart Bennett <sb476@cam.ac.uk> 14312Date: Tue Dec 4 01:31:24 2007 +0000 14313 14314 Better T table stuff. 14315 14316 Hopefully now usable by calling parse_t_table(...) at the appropriate place 14317 14318 Also collapse some messages, and add an array bound check 14319 14320commit edb9a59ac8d3b16cccd52bde24606c43ae7d36d0 14321Author: Stuart Bennett <sb476@cam.ac.uk> 14322Date: Tue Dec 4 01:40:37 2007 +0000 14323 14324 Allow BIOS execution at runtime, if wanted 14325 14326 And comment the opcodes I haven't double checked yet 14327 14328commit 82fdd3f76e20fc7dd10fcf00cf1c13520dc56364 14329Author: Stuart Bennett <sb476@cam.ac.uk> 14330Date: Tue Dec 4 01:31:24 2007 +0000 14331 14332 Make the bios a bit more accessible 14333 14334 And make a few things use the new structure 14335 14336commit d15bfdfb7e09dce85c3817036ced9efa6ac4522c 14337Author: Stuart Bennett <sb476@cam.ac.uk> 14338Date: Tue Dec 4 00:00:40 2007 +0000 14339 14340 randr12: set CR58 appropriately for 0 and 2 cases 14341 14342commit 02a4f498d903bda55ac38a6e19b1beca0d128e8a 14343Author: Stuart Bennett <sb476@cam.ac.uk> 14344Date: Mon Dec 3 23:35:01 2007 +0000 14345 14346 randr12: refactor dcb usage 14347 14348 heads is also a better name than head 14349 14350commit 1571f81210b48407c9d20264437406d79dfcc784 14351Author: Stuart Bennett <sb476@cam.ac.uk> 14352Date: Mon Dec 3 22:26:33 2007 +0000 14353 14354 randr12: some code economy 14355 14356 Only addition is adding call to output2->funcs->commit when stealing a ramdac, 14357 no logical changes otherwise AFAICT 14358 14359commit 7debf566c7a88258ac8ca925d7138187c1ed8755 14360Author: Stuart Bennett <sb476@cam.ac.uk> 14361Date: Tue Dec 4 00:06:05 2007 +0000 14362 14363 randr12: Improve CR57 and CR58 infrastructure and saving/restoration 14364 14365commit dab828c4793fd7e48e4d8e1d7f8d0a2f8198484c 14366Author: Maarten Maathuis <madman2003@gmail.com> 14367Date: Mon Dec 3 20:55:26 2007 +0100 14368 14369 randr12: A few minor comment changes. 14370 14371commit d959d880b6808adc153c752ea01f3a228879fe3c 14372Author: Maarten Maathuis <madman2003@gmail.com> 14373Date: Mon Dec 3 20:51:16 2007 +0100 14374 14375 randr12: Select CRTC 1 on pre-nv40 (stupid bug) + initial support choosing scaling method on tmds. 14376 14377 - The output property SCALING_MODE accepts "gpu" and "panel". 14378 - The first obeys aspect ratio, the second always goes fullscreen. 14379 - This is just a first attemp at exposing an output property. 14380 14381commit 1d2ea797704d55f63cfe8c00cc000a6b13f1093d 14382Author: Maarten Maathuis <madman2003@gmail.com> 14383Date: Sun Dec 2 22:10:56 2007 +0100 14384 14385 Show the id's of unknown bit structures in the bios. 14386 14387commit fc98b11eb70f1dd7dfb4e7c6eb8d888dfebdc890 14388Author: Maarten Maathuis <madman2003@gmail.com> 14389Date: Sun Dec 2 12:02:08 2007 +0100 14390 14391 randr12: dual-dvi cards need extra consideration. 14392 14393commit 2beece0c634c2c60996da1f1d15b24c5f231a6be 14394Author: Stuart Bennett <sb476@cam.ac.uk> 14395Date: Sun Dec 2 02:33:57 2007 +0000 14396 14397 Fix endian bug on BIT biosen 14398 14399commit d3563cccb43355927d4ba355bc31a957ea01587f 14400Author: Stuart Bennett <sb476@cam.ac.uk> 14401Date: Sun Dec 2 02:14:32 2007 +0000 14402 14403 Don't use PROM if the checksum is bad 14404 14405commit 4097af9bad85031fa1da9c3d21b084f587412604 14406Author: Stuart Bennett <sb476@cam.ac.uk> 14407Date: Sat Dec 1 09:55:31 2007 +0000 14408 14409 randr12: print "or" _pre_ ffs-ing 14410 14411commit a8faa53355f9ba3851f47e290c821bc1b9a03e85 14412Author: Stuart Bennett <sb476@cam.ac.uk> 14413Date: Sat Dec 1 09:53:05 2007 +0000 14414 14415 DCB parsed flag for duallink outputs 14416 14417 Light on proof, strong on suspicion with this one 14418 14419commit 63fc1ab4ce597a2f542c9bb7d3edae67551bda29 14420Author: Maarten Maathuis <madman2003@gmail.com> 14421Date: Sun Dec 2 02:08:20 2007 +0100 14422 14423 randr12: Always use vpll2_true on nv4x cards. 14424 14425commit 97a99c42fb24320177bdae884c4c710c5a25ce4f 14426Author: Maarten Maathuis <madman2003@gmail.com> 14427Date: Sun Dec 2 01:48:01 2007 +0100 14428 14429 randr12: A safer bet for TMDS register 0x2e. 14430 14431commit 3358f4eec53845cf90b9f9ef1f2a70c231bac588 14432Author: Maarten Maathuis <madman2003@gmail.com> 14433Date: Sun Dec 2 01:34:49 2007 +0100 14434 14435 randr12: Remove preferred_crtc, since it's probably incorrect. 14436 14437commit 5994655925173a4e8b1c0ac902f7dc6e1ffe64f1 14438Author: Maarten Maathuis <madman2003@gmail.com> 14439Date: Sun Dec 2 01:08:41 2007 +0100 14440 14441 randr12: Improve output register setting. 14442 14443commit df8d6848a004e724e92c3dd63fc608198c7bd940 14444Author: Maarten Maathuis <madman2003@gmail.com> 14445Date: Sat Dec 1 18:40:46 2007 +0100 14446 14447 randr12: Fix up some dvi related issues, thanks to MighMoS. 14448 14449commit 84cc907beffd9e8f91c4333a6db919ee2efa2740 14450Author: Maarten Maathuis <madman2003@gmail.com> 14451Date: Sat Dec 1 11:23:50 2007 +0100 14452 14453 randr12: Fix NV4x dvi clock selection, mostly affects G70 cards. 14454 14455commit c97ad9850da89b1898f125df1796bb1d00d124bb 14456Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14457Date: Sat Dec 1 10:33:53 2007 +0100 14458 14459 Make the debug variable global to all DMA calls. 14460 14461commit 19d75f5018fd442920378a5083dbe1316cc6f62b 14462Author: Stuart Bennett <sb476@cam.ac.uk> 14463Date: Sat Dec 1 06:29:09 2007 +0000 14464 14465 The BIT bios T table. 14466 14467 T might stand for TMDS. 14468 Parsing once on start may not be the correct answer, could be a runtime thing 14469 14470commit 42a02513a1db7ada586dcaa2eaa5b68e97d67892 14471Author: Stuart Bennett <sb476@cam.ac.uk> 14472Date: Sat Dec 1 06:37:17 2007 +0000 14473 14474 Bump DCB table init so that INIT_50 can use it 14475 14476commit ddc4b0946dfd567fc0ea136db0d8b695c28bc3a2 14477Author: Stuart Bennett <sb476@cam.ac.uk> 14478Date: Sat Dec 1 05:35:40 2007 +0000 14479 14480 "or" can be 0xc 14481 14482 Quite what this means, or why Nvidia couldn't give it a more descriptive 14483 variable name in the g80 driver is anyone's guess 14484 14485commit 384115a5abd756c5e3ee79ba50e15636d159f19c 14486Author: Stuart Bennett <sb476@cam.ac.uk> 14487Date: Sat Dec 1 05:17:47 2007 +0000 14488 14489 Use Nvidia's naming for CRTC_INDEX_COLOR 14490 14491commit 2344b5b5a5a0efa439ba64f92d5b08097178dd6c 14492Author: Stuart Bennett <sb476@cam.ac.uk> 14493Date: Sat Dec 1 05:16:20 2007 +0000 14494 14495 Bios script code 0x50 14496 14497commit 78c6ff292f4c2aa0e06bce095108d0716555104a 14498Author: Stuart Bennett <sb476@cam.ac.uk> 14499Date: Sat Dec 1 00:45:32 2007 +0000 14500 14501 Add dcb "location" field. 0 for on-chip, !0 for external 14502 14503commit 312fb3d65b8db6e3c2d123c0b192da82b7de4f71 14504Author: Maarten Maathuis <madman2003@gmail.com> 14505Date: Fri Nov 30 16:47:26 2007 +0100 14506 14507 randr12: Don't unset bit 8-15 in sel_clk. 14508 14509commit ea10d579fb6053e064a5cd14b47e1da55d4120ac 14510Author: Maarten Maathuis <madman2003@gmail.com> 14511Date: Fri Nov 30 16:43:33 2007 +0100 14512 14513 randr12: Big commit, read message. 14514 14515 - Outputs (on nv4x) should now select vpll2 instead of vpll1 when neccesary (this took a long time to figure out). 14516 - Register 0x00680594 is now set correctly to allow analog secondary outputs to work (for me at least). 14517 - I found the bit that triggers non-VGA modesetting in nv10reg, so in the future we could copy the blob in that area. 14518 - Misc fixes to a lot of things. 14519 14520commit a7a8993f4074e942ec4b3ef58ede0162a2d3c415 14521Author: Maarten Maathuis <madman2003@gmail.com> 14522Date: Mon Nov 26 22:11:03 2007 +0100 14523 14524 randr12: Make sure the vpll is properly set when we steal a ramdac from another output. 14525 14526commit c11d3f68720b698736d5a28a82d3e45b55379bc6 14527Author: Maarten Maathuis <madman2003@gmail.com> 14528Date: Mon Nov 26 19:40:59 2007 +0100 14529 14530 randr12: Fixed a minor cursor bug. 14531 14532commit 59d98374e082c40b57737de56b17173f5b54910e 14533Author: Maarten Maathuis <madman2003@gmail.com> 14534Date: Mon Nov 26 19:24:03 2007 +0100 14535 14536 randr12: Set the value of sel_clk correctly (hopefully) for NV4x. 14537 14538commit 8672d7a33fe369645786c21e130f51560a656d3c 14539Author: Maarten Maathuis <madman2003@gmail.com> 14540Date: Sun Nov 25 22:08:23 2007 +0100 14541 14542 randr12: cleanup + double register assignment (vga CRTC) removed. 14543 14544commit 11809882f24e40b1b8c77f4caca27ce2ebd3ccbf 14545Author: Maarten Maathuis <madman2003@gmail.com> 14546Date: Sun Nov 25 18:15:27 2007 +0100 14547 14548 randr12: Convert premultiplied alpha cursors to normal alpha cursors. 14549 14550commit ece41daabeb0cfd727daec14d9734dac08159c7a 14551Author: Maarten Maathuis <madman2003@gmail.com> 14552Date: Sun Nov 25 16:49:01 2007 +0100 14553 14554 randr12: Fix alpha cursor corruption. 14555 14556commit 5abbf55fb5c957c4be425e9ebd2660ca14d44e27 14557Author: Maarten Maathuis <madman2003@gmail.com> 14558Date: Sun Nov 25 14:15:30 2007 +0100 14559 14560 randr12: alpha cursors don't need background changes. 14561 14562commit 37c36a284508e63af152a9315aa19a729e5516cd 14563Author: Maarten Maathuis <madman2003@gmail.com> 14564Date: Sun Nov 25 13:06:10 2007 +0100 14565 14566 Reindent a few cursor functions. 14567 14568commit 1ad212271cc99b95ff52eb719d24ac58d392c0eb 14569Author: Maarten Maathuis <madman2003@gmail.com> 14570Date: Sun Nov 25 03:20:43 2007 +0100 14571 14572 randr12: Improve TMDS and LVDS registers (and enable for LVDS). 14573 14574commit 39d94139910c53f9092f28ea6d742133666ac18d 14575Author: Maarten Maathuis <madman2003@gmail.com> 14576Date: Sun Nov 25 01:31:39 2007 +0100 14577 14578 randr12: Optimize tmds_regs a little bit. 14579 14580commit c6dce92c47dacfc822315e570bff70a8e42f5dc2 14581Author: Stuart Bennett <sb476@cam.ac.uk> 14582Date: Sat Nov 24 21:51:14 2007 +0000 14583 14584 randr12: use bus field to determine dvi pairs 14585 14586commit 328197b4613f5e25eacfb67888701bb0e298c765 14587Author: Stuart Bennett <sb476@cam.ac.uk> 14588Date: Sat Nov 24 20:44:27 2007 +0000 14589 14590 Remove a load of unnecessary casts, and reorder functions by opcode 14591 14592 Plus a micro opt to checksum thing 14593 14594commit 8cee41bc6112a96f43f242b5178a9c968d2b2d9e 14595Author: Stuart Bennett <sb476@cam.ac.uk> 14596Date: Sat Nov 24 20:25:13 2007 +0000 14597 14598 A bunch more cleaned up bios opcodes 14599 14600 Thanks to stillunknown for the bios trace 14601 14602commit 5d6f214db4cd5f0cdcb6a4395ce2ce4acf322257 14603Author: Maarten Maathuis <madman2003@gmail.com> 14604Date: Sat Nov 24 16:22:04 2007 +0100 14605 14606 randr12: Be more considerate of outputs that can only work on ramdac 1. 14607 14608commit 6e1c37e26f630caa03da9061a5cec200708babbd 14609Author: Maarten Maathuis <madman2003@gmail.com> 14610Date: Sat Nov 24 15:47:04 2007 +0100 14611 14612 randr12: The result of another refinement of dvi crosswiring. 14613 14614commit ce57eb9b34095f2776ef36b09249a4b202824a86 14615Author: Maarten Maathuis <madman2003@gmail.com> 14616Date: Sat Nov 24 13:12:49 2007 +0100 14617 14618 randr12: Disable crt_load_detect for ramdac 1 and set some better valued for test_control. 14619 14620commit 525997f86c5fe0f79cf0fa584b4cb617b83a9386 14621Author: Maarten Maathuis <madman2003@gmail.com> 14622Date: Fri Nov 23 23:12:53 2007 +0100 14623 14624 randr12: add bit30 to vpll register for nv4x cards 14625 14626 - This seems the most common situation, any idea what the bits do? 14627 14628commit f249b3b33c025417d11a85a29848db2112450dce 14629Author: Patrice Mandin <pmandin@caramail.com> 14630Date: Fri Nov 23 22:25:47 2007 +0100 14631 14632 nv30: missing command 14633 14634commit b56ca2ef87fe15132dcbc43361344d8af446c8a7 14635Author: Patrice Mandin <pmandin@caramail.com> 14636Date: Fri Nov 23 22:25:06 2007 +0100 14637 14638 nv30: another reference to an object 14639 14640commit a8af1725ef7ae6c1cd19659d0607e2a702dc148a 14641Author: Patrice Mandin <pmandin@caramail.com> 14642Date: Fri Nov 23 22:24:19 2007 +0100 14643 14644 nv30: rename viewport clip 14645 14646commit 14e631691a2e8704ee016828e94c48de7eb96409 14647Author: Patrice Mandin <pmandin@caramail.com> 14648Date: Fri Nov 23 22:03:21 2007 +0100 14649 14650 nv30: was not using gart handle 14651 14652commit 9648fd259f9906f5db854e8c4b57fd2675f4d5c5 14653Author: Maarten Maathuis <madman2003@gmail.com> 14654Date: Fri Nov 23 21:08:46 2007 +0100 14655 14656 randr12: Reenable a properly working crt load detect. 14657 14658commit 3010847c1ccf14f0b5356aeb05f2c8c0c519ce91 14659Author: Maarten Maathuis <madman2003@gmail.com> 14660Date: Fri Nov 23 18:58:49 2007 +0100 14661 14662 randr12: Remove some unneeded stuff. 14663 14664commit 2b092ca5277405a166ea54de418a8114fb6a8cf6 14665Author: Maarten Maathuis <madman2003@gmail.com> 14666Date: Fri Nov 23 17:32:12 2007 +0100 14667 14668 randr12: fix comment typo 14669 14670commit 5ca53ef57ece7ae547cd5b11e83eb0d15fe69033 14671Author: Maarten Maathuis <madman2003@gmail.com> 14672Date: Fri Nov 23 16:40:02 2007 +0100 14673 14674 randr12: Allow arbitrary ramdac and crtc combinations. 14675 14676 - Support for the misterious ramdac 2 still has to be reverse engineered. 14677 14678commit e16fccde3f817e9eab3a09398a4821566967a86d 14679Author: Maarten Maathuis <madman2003@gmail.com> 14680Date: Fri Nov 23 13:25:38 2007 +0100 14681 14682 randr12: Move a lot of registers from output to crtc. 14683 14684commit cd8293954e1d004ef837d339ae87c4689fd45e1f 14685Author: Maarten Maathuis <madman2003@gmail.com> 14686Date: Fri Nov 23 10:32:08 2007 +0100 14687 14688 randr12: There was more logic to sel_clk than previously thought. 14689 14690commit 0b55ba9f9fbd60044bad880f260e0c70e6f577f0 14691Author: Maarten Maathuis <madman2003@gmail.com> 14692Date: Thu Nov 22 22:18:47 2007 +0100 14693 14694 randr12: revert to old vpll regs on nv4x cards, because the others can't be written 14695 14696 - If anyone knows the bit to allow those registers to be written, then please tell. 14697 - This also fixes db1 vclk ratio issues, hopefully. 14698 14699commit 18117df17ec2ce355344953d70caf33a0272697f 14700Author: Stuart Bennett <sb476@cam.ac.uk> 14701Date: Thu Nov 22 00:30:29 2007 +0000 14702 14703 Bios CRTC head chosing fixed 14704 14705 Also, a slightly tighter limit on values for or, as I don't currently think 14706 bit 4 being set makes any sense 14707 14708commit 425b7cd9dcddf1735d59e9ef60e81bf91db1c65c 14709Author: Stuart Bennett <sb476@cam.ac.uk> 14710Date: Wed Nov 21 20:02:17 2007 +0000 14711 14712 randr12: unbreak LVDS console restore by reverting tmds_regs change 14713 14714commit e3f77b1b4332f5c072be713b5ea6dd234c194bb0 14715Author: Stuart Bennett <sb476@cam.ac.uk> 14716Date: Tue Nov 20 19:02:17 2007 +0000 14717 14718 Complete init_index_addr_latched opcode 14719 14720commit 05f40b2dd3f8c14da2464b19b21aa66defec17e3 14721Author: Stuart Bennett <sb476@cam.ac.uk> 14722Date: Wed Nov 21 01:58:12 2007 +0000 14723 14724 More DCB entry output, and a default for DCB 1.4 14725 14726 The table seems to be the same for all DCB 1.4 biosen I've seen, and 14727 POST seems to make no effort to parse it. I wonder if it is in fact used at all 14728 14729commit d96176c2f07efde4224c35b7c03b8380a73e1831 14730Author: Maarten Maathuis <madman2003@gmail.com> 14731Date: Tue Nov 20 20:41:38 2007 +0100 14732 14733 randr12: better guess for ramdac register 670 (nv44 and up) 14734 14735commit 817b8f1a1af70402c18c45edccec5c746063d4e8 14736Author: Maarten Maathuis <madman2003@gmail.com> 14737Date: Tue Nov 20 19:55:15 2007 +0100 14738 14739 randr12: fix serious bug in vclk calculator for nv4x cards. 14740 14741commit 040e5479344b0f794e79e4ffc6c547c58835401a 14742Author: Maarten Maathuis <madman2003@gmail.com> 14743Date: Tue Nov 20 18:49:39 2007 +0100 14744 14745 randr12: Fix a small mistake that could cause blank screens for NV44 and up. 14746 14747commit 1cc05c9e08a7e4f0c27bb0e08d47265d31123bf9 14748Author: Maarten Maathuis <madman2003@gmail.com> 14749Date: Tue Nov 20 12:56:28 2007 +0100 14750 14751 randr12: I hope i really fixed db1 vclk ratio this time (for nv4x). 14752 14753commit 016f7e84c4ef62840e300d61e5e3b2ddfadd5e32 14754Author: Maarten Maathuis <madman2003@gmail.com> 14755Date: Tue Nov 20 12:44:25 2007 +0100 14756 14757 NV50: Make a small mistake. 14758 14759commit 2cbd24ad579d3e68eaef2f8500b22e101fecd936 14760Author: Maarten Maathuis <madman2003@gmail.com> 14761Date: Tue Nov 20 12:25:21 2007 +0100 14762 14763 NV50: Some deobfuscation. 14764 14765commit ccfc6910f8cd8eb8d4b628cc14ef7db417f576ea 14766Author: Maarten Maathuis <madman2003@gmail.com> 14767Date: Tue Nov 20 09:52:47 2007 +0100 14768 14769 randr12: Also enable DB1 vclk ratio if PLL_SELECT had it enabled. 14770 14771commit 5b3d86562afd35c14d2c8c07618b08bbc98ad775 14772Author: Maarten Maathuis <madman2003@gmail.com> 14773Date: Mon Nov 19 23:55:58 2007 +0100 14774 14775 randr12: Proper support for db1 vclk ratio, thanks to Thunderbird for some info. 14776 14777commit 22d8ba5ebf659f6f126a1380190339d08ee824c6 14778Author: Maarten Maathuis <madman2003@gmail.com> 14779Date: Mon Nov 19 09:23:37 2007 +0100 14780 14781 randr12: Add a gamma_set hook. 14782 14783commit cbd70303a949f387d42497cb4ea52186ed938b9c 14784Author: Maarten Maathuis <madman2003@gmail.com> 14785Date: Sun Nov 18 22:03:03 2007 +0100 14786 14787 randr12: A first attempt at some lvds programming, not hooked up. 14788 14789 - This is a start for someone with a laptop to test and refine. 14790 14791commit 58c3d6a886c5d5557def490ee8c3e75c9f8d2cd7 14792Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 14793Date: Sun Nov 18 02:50:33 2007 +0100 14794 14795 Enable IFC uploads. 14796 14797commit 619ae1f7a028f24293f9a4e4f7777cc51ef8d16d 14798Author: Maarten Maathuis <madman2003@gmail.com> 14799Date: Sun Nov 18 00:15:34 2007 +0100 14800 14801 randr12: Final commit, switch to nv40 native vclk regs. 14802 14803 - I've also seperated the save load pll functions, to improve clarity. 14804 14805commit 601adb4f431b638d4b711ba01aa2d4c3408a7bd3 14806Author: Maarten Maathuis <madman2003@gmail.com> 14807Date: Sat Nov 17 23:47:12 2007 +0100 14808 14809 randr12: Intermediate commit (using the native nv40 vpll regs). 14810 14811commit a29eca9dec176388b1c4666ad801ae75360a1219 14812Author: Maarten Maathuis <madman2003@gmail.com> 14813Date: Sat Nov 17 20:08:27 2007 +0100 14814 14815 randr12: Some comment fixes. 14816 14817commit 6f3d335f6e3b9866457730660538c5f073862dfe 14818Author: Maarten Maathuis <madman2003@gmail.com> 14819Date: Sat Nov 17 22:23:32 2007 +0100 14820 14821 randr12: Reinstate m-divider limitations. 14822 14823commit e2b7f57cbe7e1080736e6088d2997cbcf966fc3d 14824Author: Maarten Maathuis <madman2003@gmail.com> 14825Date: Sat Nov 17 18:36:24 2007 +0100 14826 14827 randr12: Some minor changes. 14828 14829commit 2c0c7612bd08938496abb47b34498065e668ec34 14830Author: Maarten Maathuis <madman2003@gmail.com> 14831Date: Sat Nov 17 15:57:20 2007 +0100 14832 14833 randr12: Change vpll handling, note that it still is a work in progress. 14834 14835commit caec3d1ab69c523b6c0a7e0019d1a9f115cf5da3 14836Author: Ben Skeggs <skeggsb@gmail.com> 14837Date: Sat Nov 17 20:59:25 2007 +1100 14838 14839 EXA 1.4 pixmap hooks (disabled) 14840 14841 Really, you don't want to turn this on yet... :) 14842 14843commit b8e7c64476948426d5801367bab5cdf15db5ce84 14844Author: Ben Skeggs <skeggsb@gmail.com> 14845Date: Sat Nov 17 17:17:34 2007 +1100 14846 14847 Add a user callback for when a channel hang is detected. 14848 14849commit 293b32d0f66e965b3a31d54859b91210b3e35b3d 14850Author: Ben Skeggs <skeggsb@gmail.com> 14851Date: Sat Nov 17 17:02:23 2007 +1100 14852 14853 Use handles from grobj, instead of hardcoded vals from nv_dma.h 14854 14855commit 84bc37723c2cabe7652050b4cbaf9f5218fdb071 14856Author: Ben Skeggs <skeggsb@gmail.com> 14857Date: Sat Nov 17 16:43:05 2007 +1100 14858 14859 Turn vram/fb handles into "real" grobjs. 14860 14861commit f3cdd5086f410711103c6b3453629d85ef4c2788 14862Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14863Date: Sat Nov 17 01:22:31 2007 +0100 14864 14865 Put back memory barrier and write posting. 14866 14867commit 762f732c7aaa8aa3dd7d54a8734cbcc54cd10494 14868Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14869Date: Fri Nov 16 23:38:49 2007 +0100 14870 14871 Add verbose debugging code to help with fixing push buffer issues. 14872 14873commit c8f7a9fcdcc356b4dc5030112b1532d2fa3eaf0f 14874Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14875Date: Fri Nov 16 23:37:10 2007 +0100 14876 14877 Oops. Correct either the header or the packet but not both. 14878 14879commit 5ce573875387490eb7feaa87a3354b888beb4410 14880Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14881Date: Fri Nov 16 23:32:07 2007 +0100 14882 14883 Unmatched OUT_RING(0). 14884 14885commit cc0c1af932a96cec6aabc72cbe9ca508fbc80a36 14886Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14887Date: Fri Nov 16 21:59:53 2007 +0100 14888 14889 fix typo. 14890 14891commit 4b7601c297fefc46adf3bc37a463575d3caafedb 14892Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14893Date: Fri Nov 16 21:56:14 2007 +0100 14894 14895 fix 32/64 bit ism. 14896 14897commit bee8bc5ce321245362847253b7ccc7530899e514 14898Author: Maarten Maathuis <madman2003@gmail.com> 14899Date: Fri Nov 16 21:29:20 2007 +0100 14900 14901 randr12: Minor change. 14902 14903commit 735fae4bb0b92f097665f675ba5127a9bb11d8fa 14904Author: Maarten Maathuis <madman2003@gmail.com> 14905Date: Fri Nov 16 19:19:23 2007 +0100 14906 14907 randr12: That wasn't supposed to end up there ;-) 14908 14909commit 2848c3e7a4d25e5ca311ea1a88458eb2a77976ed 14910Author: Maarten Maathuis <madman2003@gmail.com> 14911Date: Fri Nov 16 19:16:08 2007 +0100 14912 14913 randr12: Disable changing ramdac reg580, until i know why this causes issues. 14914 14915commit 75222fa199f1f5bc1abd66a4e80a446b5d597523 14916Author: Ben Skeggs <skeggsb@gmail.com> 14917Date: Sat Nov 17 15:31:49 2007 +1100 14918 14919 nouveau: fix valgrind complaint 14920 14921commit 1a98f1ae8298ca3b069b146a2866f96e934afb74 14922Author: Maarten Maathuis <madman2003@gmail.com> 14923Date: Fri Nov 16 17:00:59 2007 +0100 14924 14925 randr12: Restore old timings + misc changes. 14926 14927commit 1b0cd6602cf9d96f41242054efe2f96cc70eaa96 14928Author: Ben Skeggs <skeggsb@gmail.com> 14929Date: Sat Nov 17 09:08:03 2007 +1100 14930 14931 oops, forgot these... 14932 14933commit 9c46ec6833ccfc29e4a67b0ce4226d8b38b772ab 14934Author: Ben Skeggs <skeggsb@gmail.com> 14935Date: Sat Nov 17 07:23:41 2007 +1100 14936 14937 Initial hacky relocation stuff. 14938 14939 Entire DDX ported across to OUT_RELOCx/OUT_PIXMAPx, workable in it's current 14940 state for the current BO backend.. definitely not for TTM yet :) 14941 14942commit ddda33decb42ba2f5f9586e1d5de4533d1d88674 14943Author: Ben Skeggs <skeggsb@gmail.com> 14944Date: Sat Nov 17 06:11:01 2007 +1100 14945 14946 Initial buffer object work. 14947 14948 Implemented on top of the current MM interface, and currently used in the 14949 exact same way NVAllocate/FreeMemory was used. Can transition to something 14950 more ttm-friendly gradually. 14951 14952 The "old mm" bo backend is extremely primitive. Buffers can only be created 14953 in mappable memory, forcibly pinned, and never moved once they're created. 14954 Some of these things could be improved, but probably not much point. 14955 14956commit 9b9d995a517b48cd074030c567f9c78bee19b129 14957Author: Ben Skeggs <skeggsb@gmail.com> 14958Date: Sat Nov 17 03:36:04 2007 +1100 14959 14960 Report engine hangs again, instead of spinning forever. 14961 14962commit 2af4a28ae1df46686e264bcda417a896b7cfa59a 14963Author: Ben Skeggs <skeggsb@gmail.com> 14964Date: Fri Nov 16 09:18:33 2007 +1100 14965 14966 Hook up DMA funcs to channels/grobjs etc etc 14967 14968commit 6e3d05e8c60fd7bab371af86151e45eec0a65b3d 14969Author: Ben Skeggs <skeggsb@gmail.com> 14970Date: Fri Nov 16 07:08:23 2007 +1100 14971 14972 notifiers.. 14973 14974 Lacking timeouts at the moment so if X would've crashed before, it'll spin 14975 forever now. 14976 14977commit 020740d92b5d623bee65e598ac60020187011b04 14978Author: Ben Skeggs <skeggsb@gmail.com> 14979Date: Fri Nov 16 06:07:56 2007 +1100 14980 14981 And grobjs. 14982 14983commit 929e2b7d53435d097b07c23ed1d0caaa47924f8a 14984Author: Ben Skeggs <skeggsb@gmail.com> 14985Date: Fri Nov 16 05:10:58 2007 +1100 14986 14987 And channels.. 14988 14989commit 22909d96b8ea720169b34326b93c7b29eb3db013 14990Author: Ben Skeggs <skeggsb@gmail.com> 14991Date: Fri Nov 16 04:36:08 2007 +1100 14992 14993 Implement nouveau_device struct. 14994 14995commit b96e813773b0705bbac57f2d64adff3aad845622 14996Author: Stuart Bennett <sb476@cam.ac.uk> 14997Date: Fri Nov 16 03:26:56 2007 +0000 14998 14999 Begone, foul g5-bios-hack 15000 15001 also, fix the DCB termination test; I'm amazed nothing broke 15002 15003commit 7eee9585fa8e8c45f4ab4712a3b990664203d72b 15004Author: Maarten Maathuis <madman2003@gmail.com> 15005Date: Thu Nov 15 19:33:13 2007 +0100 15006 15007 randr12: Only NV4x have two seperate PVIO ranges. 15008 15009commit 276892b069fd045b20f44d7888a7971371d73ad3 15010Author: Maarten Maathuis <madman2003@gmail.com> 15011Date: Thu Nov 15 19:27:12 2007 +0100 15012 15013 randr12: Improved values for FP_CONTROL register. 15014 15015commit 4a16f14061c88ed49b05ef8779a2584f5053c3f1 15016Author: Maarten Maathuis <madman2003@gmail.com> 15017Date: Thu Nov 15 18:44:52 2007 +0100 15018 15019 randr12: Some minor register changes. 15020 15021commit b436a6809c62e0fe0696d2c6bad066c3afcf9166 15022Author: Arthur Huillet <arthur.huillet@free.fr> 15023Date: Thu Nov 15 15:01:47 2007 +0100 15024 15025 NV10 EXA: removed left out debug statement 15026 15027commit 98e223ebb01440810e01e902e904524c50b463cb 15028Author: Arthur Huillet <arthur.huillet@free.fr> 15029Date: Thu Nov 15 12:57:20 2007 +0100 15030 15031 NV1x EXA : refactored A8 + A8 code, enabled on NV2x, fix A8 txfmt for NV2x 15032 15033 more compact and more readable A8 + A8 code along with some codestyle fixes 15034 now enabling this acceleration on NV2x cards, with the A8 format value hack 15035 15036commit 3cdc4b274f371f1f524d140fddd79dad2987b06c 15037Author: Ben Skeggs <skeggsb@gmail.com> 15038Date: Fri Nov 16 03:28:45 2007 +1100 15039 15040 nouveau: fix a host of warnings 15041 15042commit 6eba6c7fab534efd15c8e96daeeb9a33972c2b6d 15043Author: Stuart Bennett <sb476@cam.ac.uk> 15044Date: Thu Nov 15 02:09:45 2007 +0000 15045 15046 Big endian bios fix 15047 15048commit 69a3a849e65a022609db3a5f33277b744463a6ae 15049Author: Patrice Mandin <pmandin@caramail.com> 15050Date: Wed Nov 14 23:03:06 2007 +0100 15051 15052 nv30exa: set link with image blit 15053 15054commit 2435836202efe7ea42a7bece71f125fb46aac3ba 15055Author: Stuart Bennett <sb476@cam.ac.uk> 15056Date: Wed Nov 14 19:02:28 2007 +0000 15057 15058 nv_bios indentation 15059 15060commit 7b0ca260f57e92e45d7f48f16642bfcd67b10eaf 15061Author: Stuart Bennett <sb476@cam.ac.uk> 15062Date: Wed Nov 14 18:47:40 2007 +0000 15063 15064 More code and logging beautification - wrap I/O better 15065 15066 Plus some actual fixes 15067 15068commit d00420a1286e368fbc20551cae5260bf65ed1b15 15069Author: Stuart Bennett <sb476@cam.ac.uk> 15070Date: Wed Nov 14 19:07:06 2007 +0000 15071 15072 IO flag conditions, beginnings of init_idx_addr_latched 15073 15074 plus usual whitespace and pointless renaming 15075 15076commit 4e8af31a89b7d732b52706f87e1dc66e29ab717e 15077Author: Stuart Bennett <sb476@cam.ac.uk> 15078Date: Wed Nov 14 03:01:27 2007 +0000 15079 15080 More opcodes, and some monstrous reindenting 15081 15082 all part of an evil plan to have the most lines in the Xorg log 15083 15084commit 89617cfcdfd1ee19b99888cfbcfec1ee0550d26c 15085Author: Stuart Bennett <sb476@cam.ac.uk> 15086Date: Wed Nov 14 00:36:02 2007 +0000 15087 15088 Updating some BIOS parser opcodes 15089 15090 plus various debug / whitespace fixes 15091 15092commit cc5db05bed630350613c7cbfa418d814ed5a5093 15093Author: Stuart Bennett <sb476@cam.ac.uk> 15094Date: Wed Nov 14 19:27:06 2007 +0000 15095 15096 randr12: More DCB abstraction to lessen parsing nastiness for old versions 15097 15098commit d46b61f88225d3e2b28a6c4a4ac4b8a1fe942a39 15099Author: Stuart Bennett <sb476@cam.ac.uk> 15100Date: Tue Nov 13 18:24:46 2007 +0000 15101 15102 Gratuitous code renaming, moving + warning fixes for bios 15103 15104commit c87e3570a198b5d69046387d111a2709a571d5e3 15105Author: Maarten Maathuis <madman2003@gmail.com> 15106Date: Wed Nov 14 22:30:11 2007 +0100 15107 15108 randr12: Reinstate the "magic factor" until a better solution comes along. 15109 15110commit 5b62e77ded9e64d439b70ce0b7dce27a7ae1df4f 15111Author: Arthur Huillet <arthur.huillet@free.fr> 15112Date: Wed Nov 14 20:54:24 2007 +0100 15113 15114 NV1x EXA : implemented A8 + A8 via a dirty hack 15115 15116 it works, but probably needs to be optimized 15117 15118commit 3db89822ba94a94eeb1ab891488d7f203a0dfff8 15119Author: Maarten Maathuis <madman2003@gmail.com> 15120Date: Wed Nov 14 19:48:53 2007 +0100 15121 15122 randr12: Disable ramdac register 0x900, until i know for certain what it does. 15123 15124commit 842b33e2dc82e70a7da992642de4503d4cb0a1fc 15125Author: Maarten Maathuis <madman2003@gmail.com> 15126Date: Wed Nov 14 16:02:56 2007 +0100 15127 15128 randr12: fix comment typo 15129 15130commit c7fc4b9981bb242830ddb830863fb26301429c74 15131Author: Maarten Maathuis <madman2003@gmail.com> 15132Date: Wed Nov 14 08:34:44 2007 +0100 15133 15134 randr12: Enable all pll on pre-nv40 and improve usage of ramdac_670 register. 15135 15136commit 992d079d64369a4a8445012b438ec980968c11d8 15137Author: Maarten Maathuis <madman2003@gmail.com> 15138Date: Tue Nov 13 21:46:23 2007 +0100 15139 15140 NV50: fix stupid type from commit d5dfb82f7f39304eb39dd4b1c34a2a2801d90a25 15141 15142commit 7ad7b6ad453b6401ef0bf5128f7ac9790df95bb2 15143Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15144Date: Tue Nov 13 21:36:13 2007 +0100 15145 15146 Don't use GART on PPC. Workaround DMA bugs there. 15147 15148commit b451a21f77678b1fb1d31f3df7e3b1aea6984876 15149Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15150Date: Tue Nov 13 20:58:51 2007 +0100 15151 15152 Remove ENABLE_NV30EXA, which is not needed any more. 15153 15154commit 9dc1e1e1fc049c7a535da8e1bc8ce79fd2d620b5 15155Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15156Date: Tue Nov 13 20:54:53 2007 +0100 15157 15158 Enable nv30 exa on PPC. 15159 15160commit 4fb7b608cbdf53084da63526c066eb7f772c6c14 15161Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15162Date: Tue Nov 13 20:53:58 2007 +0100 15163 15164 Fix nv30 exa for big endian 15165 15166commit 875960dacda54268fc1d318995c145b2db09ca5b 15167Author: Maarten Maathuis <madman2003@gmail.com> 15168Date: Tue Nov 13 19:43:41 2007 +0100 15169 15170 randr12: Set a few new registers + misc fixes. 15171 15172commit cf9aaa2b074eca67cf0119172c3d15328418416d 15173Author: Stuart Bennett <sb476@cam.ac.uk> 15174Date: Mon Nov 12 19:24:34 2007 +0000 15175 15176 Some type / warning fixes for nv_bios.c 15177 15178commit f00f6d1df06d2671255fd988f514a6f6ab0f8b18 15179Author: Stuart Bennett <sb476@cam.ac.uk> 15180Date: Sun Nov 11 17:50:55 2007 +0000 15181 15182 Misc bios parsing bits 15183 15184commit fdc7f4a353c4dda07fffd9800b477ae5bc43baac 15185Author: Patrice Mandin <pmandin@caramail.com> 15186Date: Mon Nov 12 23:54:22 2007 +0100 15187 15188 Remove disable switches, big cleanup, requires xorgs server 1.3 again 15189 15190commit e93a5d3abe6d637e00bd9c998e9735c9766eecfb 15191Author: Maarten Maathuis <madman2003@gmail.com> 15192Date: Mon Nov 12 21:05:00 2007 +0100 15193 15194 randr12: Improve vclk setting. 15195 15196 - It now considers crosswired TMDS situations. 15197 15198commit d32420c82b1e63bbc84768c92c336391af624952 15199Author: Maarten Maathuis <madman2003@gmail.com> 15200Date: Mon Nov 12 20:17:04 2007 +0100 15201 15202 randr12: Fix bad comment. 15203 15204commit 8a04babb88111ec8ccec9af56c3591d41afba9b3 15205Author: Maarten Maathuis <madman2003@gmail.com> 15206Date: Mon Nov 12 20:14:10 2007 +0100 15207 15208 randr12: Some minor changes. 15209 15210commit 821d4b83ed780d9a0a8d950a2595a0e15ef3f103 15211Author: Maarten Maathuis <madman2003@gmail.com> 15212Date: Mon Nov 12 16:21:14 2007 +0100 15213 15214 randr12: Set SEL_CLK appropriately based on wether a TMDS is crosswired. 15215 15216commit d017ce4879ca755fc6d56f232fe3162864f4b38e 15217Author: Maarten Maathuis <madman2003@gmail.com> 15218Date: Mon Nov 12 10:34:54 2007 +0100 15219 15220 A little code cleanup. 15221 15222commit 16fb5b92dba9a0380d2fdd1ece1b9b0691d79d3a 15223Author: Maarten Maathuis <madman2003@gmail.com> 15224Date: Mon Nov 12 00:17:52 2007 +0100 15225 15226 NV50: Port an LVDS detection fix from nv. 15227 15228commit b01a87f18d122d80fac6daf1dfb1c404d7f3a814 15229Author: Maarten Maathuis <madman2003@gmail.com> 15230Date: Mon Nov 12 00:03:19 2007 +0100 15231 15232 NV50: Port one of the nv patches. 15233 15234commit f22691661f8b72cedb8aa4fc96b41fcff4171e12 15235Author: Maarten Maathuis <madman2003@gmail.com> 15236Date: Sun Nov 11 21:53:10 2007 +0100 15237 15238 randr12: Some code cleanup. 15239 15240commit aceceb22775bb039cb0e207f39f7cbd93c2db194 15241Author: Maarten Maathuis <madman2003@gmail.com> 15242Date: Sun Nov 11 21:32:31 2007 +0100 15243 15244 NV50: Clean up nv50_output.c a bit. 15245 15246 - I did not make wrappers for the i2c stuff, since i do not know what that range does, besides i2c. 15247 15248commit a0835f7c64cf5f62626332ccba21cfe436584275 15249Author: Maarten Maathuis <madman2003@gmail.com> 15250Date: Sun Nov 11 21:17:40 2007 +0100 15251 15252 NV50: cleanup of nv50_sor.c 15253 15254commit f463a7f6213bede18917191546c8cd61cf4612e8 15255Author: Maarten Maathuis <madman2003@gmail.com> 15256Date: Sun Nov 11 20:51:01 2007 +0100 15257 15258 randr12: Make sure we don't set 2 "special" bits in sel_clk. 15259 15260commit 27cbf73bc572f435870decc2017cbf1c4f38264d 15261Author: Maarten Maathuis <madman2003@gmail.com> 15262Date: Sun Nov 11 19:32:26 2007 +0100 15263 15264 randr12: Overhaul vpll set code + misc fixes. 15265 15266 - Hopefully the problematic G70's work again. 15267 15268commit 78e88d1f4e2f058c487978c01e19a828271f3f0b 15269Author: Stuart Bennett <sb476@cam.ac.uk> 15270Date: Fri Nov 9 03:17:12 2007 +0000 15271 15272 vbios fptable v2 parsing (for BIT biosen). Not that I know how to index it. 15273 15274 Plus some fptable v1 fixes, bmp structure length checks, 15275 and support for a weird bios version that may be in some gf2go 15276 15277commit 15c771a6b3ecb9b36f9725c518141cf889aa953c 15278Author: Stuart Bennett <sb476@cam.ac.uk> 15279Date: Fri Nov 9 01:19:04 2007 +0000 15280 15281 G5 6600 DCB improvements, but I'm sure to be some bswaps short 15282 15283 This will all become so much nicer when I can do leNN_to_cpu()... 15284 15285commit 9408fcbbbbab80608c8923ab2b7757dfa76b9767 15286Author: Maarten Maathuis <madman2003@gmail.com> 15287Date: Sat Nov 10 17:02:38 2007 +0100 15288 15289 NV50: Last part of nv50_dac.c 15290 15291commit 85b5e6bfc17c347a9976155a755637f7c8d9304f 15292Author: Maarten Maathuis <madman2003@gmail.com> 15293Date: Sat Nov 10 16:59:49 2007 +0100 15294 15295 NV50: More cleanup (this time nv50_dac.c). 15296 15297commit 23d9f09a4600abf5426cb6d069dc5d9aa876a771 15298Author: Maarten Maathuis <madman2003@gmail.com> 15299Date: Sat Nov 10 16:32:51 2007 +0100 15300 15301 NV50: That's the initial cleanup for nv50_display.c 15302 15303commit 25cef08b183d6406ce065b2b3cd224dd1840525b 15304Author: Maarten Maathuis <madman2003@gmail.com> 15305Date: Sat Nov 10 16:24:32 2007 +0100 15306 15307 NV50: Should have run make first ;-) 15308 15309commit d5dfb82f7f39304eb39dd4b1c34a2a2801d90a25 15310Author: Maarten Maathuis <madman2003@gmail.com> 15311Date: Sat Nov 10 16:23:48 2007 +0100 15312 15313 NV50: Even more cleanup. 15314 15315commit 391c498e676752ac6bf03b668a1cb00094c1a049 15316Author: Maarten Maathuis <madman2003@gmail.com> 15317Date: Sat Nov 10 16:19:32 2007 +0100 15318 15319 NV50: More cleanup, small commits to make bisecting easier. 15320 15321commit 59788f9d4e6c264a6b666918989ad6b778026bd9 15322Author: Maarten Maathuis <madman2003@gmail.com> 15323Date: Sat Nov 10 16:05:42 2007 +0100 15324 15325 NV50: Some initial cleanup. 15326 15327commit 1a7d7ccb4c12f6a10aa5e8b41daee1aae52efa8a 15328Author: Maarten Maathuis <madman2003@gmail.com> 15329Date: Sat Nov 10 15:39:04 2007 +0100 15330 15331 NV50: Another wrapper. 15332 15333commit 3ca3170c8bbd758da81a1bc5b5a84cf3ca642c07 15334Author: Maarten Maathuis <madman2003@gmail.com> 15335Date: Sat Nov 10 15:29:13 2007 +0100 15336 15337 NV50: A new wrapper + a few functions renamed. 15338 15339 - I'm trying to clean up the code a bit, but lack of the hardware prevents me from testing it. 15340 15341commit 77dc99b2621523e20a956a8eaa01a100c992c458 15342Author: Maarten Maathuis <madman2003@gmail.com> 15343Date: Sat Nov 10 00:35:04 2007 +0100 15344 15345 randr12: This may break stuff for some g70 cards, if it does, then please scream. 15346 15347 - I hope i've got most situations covered. 15348 - If there are issues getting the second head to work on nv4x cards, then please tell. 15349 - nv4x is geforce 6 and 7 series for those who do not know. 15350 15351commit c169480b0bb177e8c5efb292d38c55554dcdfcdd 15352Author: Maarten Maathuis <madman2003@gmail.com> 15353Date: Fri Nov 9 20:41:48 2007 +0100 15354 15355 randr12: Disable load detect again, until we can identify it from a DVI monitor. 15356 15357commit 7237a7014b426acc4857d32967e0f10974877819 15358Author: Maarten Maathuis <madman2003@gmail.com> 15359Date: Fri Nov 9 20:40:46 2007 +0100 15360 15361 randr12: Improve detection of DVI-D/DVI-A pairs (for naming). 15362 15363commit a8125c458fa53544625da2071a69e3fd27d7c084 15364Author: Stuart Bennett <sb476@cam.ac.uk> 15365Date: Fri Nov 9 19:27:37 2007 +0000 15366 15367 randr12: Revert "randr12: lose ffs wrapper", it's wrong 15368 15369commit d83cb471cfcc18973eb4a5f319bc17a06e0a883f 15370Author: Stuart Bennett <sb476@cam.ac.uk> 15371Date: Fri Nov 9 19:02:00 2007 +0000 15372 15373 randr12: lose ffs wrapper 15374 15375commit dedf68e10f4b0cde6f9c49d60df737ed7b4148c0 15376Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15377Date: Wed Oct 31 02:36:41 2007 +0100 15378 15379 More nv30 exa cleaning up. 15380 15381commit d3679e8a5c0ce54ee3ef4de15be5a1fcd353df8e 15382Author: Maarten Maathuis <madman2003@gmail.com> 15383Date: Fri Nov 9 18:44:37 2007 +0100 15384 15385 randr12: Do not mess with ramdac register 580 until it's proven to be neccesary. 15386 15387 - I cannot seem to find out what the conditions are to change this register, except that it's related to VCLK ratio DB1. 15388 15389commit a49d51c638c2004641e58dc0061652553e9a6751 15390Author: Maarten Maathuis <madman2003@gmail.com> 15391Date: Fri Nov 9 18:31:02 2007 +0100 15392 15393 randr12: Fix vertical scaling. 15394 15395commit 2e6f5e18231a327cac7de4a927435585d1eadcee 15396Author: Patrice Mandin <pmandin@caramail.com> 15397Date: Fri Nov 9 18:13:43 2007 +0100 15398 15399 oops, trailing whitespace 15400 15401commit 596f4e6a78900f3cbb3beb4c99da7606dc630179 15402Author: Patrice Mandin <pmandin@caramail.com> 15403Date: Fri Nov 9 18:11:25 2007 +0100 15404 15405 Also disable nv40exa if nv30exa disabled 15406 15407commit d9f80b9d5f4ae1c00e79b9e903c2fc0c23b37a13 15408Author: Patrice Mandin <pmandin@caramail.com> 15409Date: Fri Nov 9 18:10:53 2007 +0100 15410 15411 Reading DFP tables need randr1.2 15412 15413commit 50aaccb5e0d3b9a978701c55ebe6545a6f7b1456 15414Author: Stuart Bennett <sb476@cam.ac.uk> 15415Date: Thu Nov 8 18:59:50 2007 +0000 15416 15417 randr12: invert fpScaler to regain original meaning (see man page) 15418 15419commit 3f6db3ac15b51a58353396fce2126a369b4cfd89 15420Author: Stuart Bennett <sb476@cam.ac.uk> 15421Date: Thu Nov 8 20:30:28 2007 +0000 15422 15423 randr12: Do scaling / native fixup in the fixup function 15424 15425 Nuke the LVDS specific fixup code 15426 Then use adjusted_mode accordingly 15427 15428commit 51a874a59d82678bbb790b708f9ba62985d99344 15429Author: Stuart Bennett <sb476@cam.ac.uk> 15430Date: Wed Nov 7 18:37:57 2007 +0000 15431 15432 randr12: rename DIGITAL and PANEL to TMDS and LVDS 15433 15434 Following on from earlier renaming, this is less confusing, at least to me. 15435 15436commit 61c92f10e1182403fc6fc8a666ed3d8b14795481 15437Author: Stuart Bennett <sb476@cam.ac.uk> 15438Date: Thu Nov 8 17:00:44 2007 +0000 15439 15440 randr12: blank screen before we're ready to show it 15441 15442commit da64328dd7be7f0e7ee1dd0dd774174fc126948b 15443Author: Maarten Maathuis <madman2003@gmail.com> 15444Date: Thu Nov 8 20:11:25 2007 +0100 15445 15446 randr12: Some outputs are only valid on secondary output. 15447 15448commit fe3d49ccb24ff20fba79ee97cceba64850d9387f 15449Author: Maarten Maathuis <madman2003@gmail.com> 15450Date: Thu Nov 8 19:20:41 2007 +0100 15451 15452 randr12: Improve output naming scheme. 15453 15454commit 42542b36d771142d38f72b3365da8b53e9d25c55 15455Author: Maarten Maathuis <madman2003@gmail.com> 15456Date: Thu Nov 8 19:13:03 2007 +0100 15457 15458 randr12: Add ramdac assignment code to the lvds dpms function. 15459 15460commit 2f1bf8138ba970ad18afe1ba9ee10228ea213926 15461Author: Maarten Maathuis <madman2003@gmail.com> 15462Date: Thu Nov 8 18:42:35 2007 +0100 15463 15464 randr12: Misc fixes. 15465 15466commit 8b733ab61e0b9eb375783e986de776eef5fd49ff 15467Author: Ben Skeggs <skeggsb@gmail.com> 15468Date: Fri Nov 9 03:29:21 2007 +1100 15469 15470 get rid of ALLOCATE_LOCAL etc. 15471 15472commit bfb78d2090ee5c189cf00e6af5107a756e864e98 15473Author: Maarten Maathuis <madman2003@gmail.com> 15474Date: Thu Nov 8 00:20:53 2007 +0100 15475 15476 randr12: It seems the situation was a bit easier to handle, luckily. 15477 15478 - I've tied ramdac and crtc number together. 15479 - Switching crtc's doesn't cause a problem anymore (even when crtc0 is not in use). 15480 15481commit d56e6e498ed9acc35bfa540881243acf03981571 15482Author: Maarten Maathuis <madman2003@gmail.com> 15483Date: Wed Nov 7 22:25:40 2007 +0100 15484 15485 randr12: Intermediate commit. 15486 15487commit f13659a8c8e855dc0b871173ccfdd0b70236be71 15488Author: Arthur Huillet <arthur.huillet@free.fr> 15489Date: Wed Nov 7 23:03:54 2007 +0100 15490 15491 NV10EXA: added support for R5G6B5 as source and destination (tested only with rendercheck) 15492 added A8 + A8 via ARGB + ARGB hack, disabled by default because it cannot work 15493 updated instrumentation to dump pixmap sizes and repeat flag 15494 15495commit 3d671bf3b06e64b49254db666980be3a3b3652cc 15496Author: Stuart Bennett <sb476@cam.ac.uk> 15497Date: Wed Nov 7 15:08:48 2007 +0000 15498 15499 randr12: remove some duplication and reduce mem leaks 15500 15501commit cdbec606e4ec398d6bef87d635499100559e3a19 15502Author: Stuart Bennett <sb476@cam.ac.uk> 15503Date: Wed Nov 7 14:54:18 2007 +0000 15504 15505 randr12: refactor ddc stuff for convenience 15506 15507commit ed8473b228579024a4fdc9c2a9ecf4cb71135378 15508Author: Stuart Bennett <sb476@cam.ac.uk> 15509Date: Wed Nov 7 14:33:50 2007 +0000 15510 15511 randr12: micro-optimization on use of "or" 15512 15513commit f6d0b446eda687bb55649599e1aedfa231feaaa9 15514Author: Stuart Bennett <sb476@cam.ac.uk> 15515Date: Tue Nov 6 21:20:10 2007 +0000 15516 15517 randr12: LVDS warning 15518 15519commit 62f8fed79f1617ffd0289123045fda0a3c0e92b2 15520Author: Stuart Bennett <sb476@cam.ac.uk> 15521Date: Tue Nov 6 01:29:33 2007 +0000 15522 15523 randr12: Use NV_RAMDAC_PLL_SELECT_PLL_SOURCE_ALL for LVDS 15524 15525 Using only VPLL seems to be broken for LVDS. Traces show consistent use of ALL. 15526 15527commit 0b958dc66a5533a7db61b7b2f456df876e9905ff 15528Author: Maarten Maathuis <madman2003@gmail.com> 15529Date: Wed Nov 7 09:33:40 2007 +0100 15530 15531 randr12: Also save flatpanel regs on analog monitors. 15532 15533commit c8ac3f0c7cfd6eca85f332e11e280f715e80283e 15534Author: Stuart Bennett <sb476@cam.ac.uk> 15535Date: Tue Nov 6 21:09:51 2007 +0000 15536 15537 Various minor fixes 15538 15539 Now that ramdac selection uses sensible numbers, it turns out writing to TMDS regs is a bad move on LVDS. 15540 Also, change FP_CONTROL handling a bit for LVDS (bit31 needs preserving) 15541 15542commit d7be5bdc27fee0758df33feb3a331c9c79dd94e0 15543Author: Stuart Bennett <sb476@cam.ac.uk> 15544Date: Tue Nov 6 19:53:08 2007 +0000 15545 15546 Start using OUTPUT_PANEL where appropriate 15547 15548commit f1160ac702a251708f2385b78c1657f95c2916ab 15549Author: Stuart Bennett <sb476@cam.ac.uk> 15550Date: Thu Nov 1 00:24:13 2007 +0000 15551 15552 Beautify DCB 1.2 weirdness, hopefully fixup G5 and use dcb_table more simply 15553 15554commit 38829fa5cdac1469a6a71a092a42885ea048a1ba 15555Author: Stuart Bennett <sb476@cam.ac.uk> 15556Date: Tue Nov 6 19:41:47 2007 +0000 15557 15558 Implement lvds_output_{get_modes, mode_valid, mode_fixup} 15559 15560commit 29c578bfe9415aa86910cf6d5cd44787b4fdf44d 15561Author: Stuart Bennett <sb476@cam.ac.uk> 15562Date: Fri Oct 26 03:01:17 2007 +0100 15563 15564 Reindentation, renaming, adding debug, and assume LVDS always connected 15565 15566commit 7aeae6cdc99ab633361f107752ef1bfa3e0492e0 15567Author: Stuart Bennett <sb476@cam.ac.uk> 15568Date: Wed Oct 31 17:29:16 2007 +0000 15569 15570 LVDS VBIOS FP mode finding 15571 15572commit cd806f92e29573ea968ab2fddb454eb1513e37dd 15573Author: Maarten Maathuis <madman2003@gmail.com> 15574Date: Wed Nov 7 00:16:33 2007 +0100 15575 15576 randr12: Fix ugly alligment. 15577 15578commit 88faf35179b9af8060c35a7a95e9a6baaae552be 15579Author: Maarten Maathuis <madman2003@gmail.com> 15580Date: Wed Nov 7 00:11:51 2007 +0100 15581 15582 randr12: Improve hotplugging of outputs. 15583 15584commit 4aa3acbd5410c8dda57a483cf7967258440888dc 15585Author: Peter Winters <peter@whoei.org> 15586Date: Tue Nov 6 23:40:07 2007 +0100 15587 15588 NV10 EXA: Implement a8 1x1 repeat 15589 15590commit 524e555c468fbab32b515ecd2c1e9cccf6c48443 15591Author: Peter Winters <peter@whoei.org> 15592Date: Tue Nov 6 23:29:03 2007 +0100 15593 15594 NV10 EXA: setup viewport like nv40 exa 15595 15596commit 2f2298f0ae3777c775b3ab81356adebf4f52ed77 15597Author: Maarten Maathuis <madman2003@gmail.com> 15598Date: Tue Nov 6 22:15:20 2007 +0100 15599 15600 randr12: Readd crt load detect + misc changes + restrictions (see full log) 15601 15602 - Crt load detect has been readded, but only for the primary output (the other gives false positives). 15603 - Misc changes to ramdac management. 15604 - Until i know how to trigger a full mode change, i cannot allow adding an output which requires ramdac0 during runtime. 15605 15606commit a6244c25a5c7efc633c74f506ca121e10a5642c0 15607Author: Maarten Maathuis <madman2003@gmail.com> 15608Date: Tue Nov 6 16:11:43 2007 +0100 15609 15610 randr12: Reasonably complete TMDS programming, see notes. 15611 15612 - Please file a bug and make a mmio-trace when it fails for you. 15613 - I can now hotplug TMDS, so that makes both types hotpluggable on nv4x. 15614 - I have no clue what happens on pre-nv40 hardware, so please try. 15615 15616commit 5f6912f083048f01a33582d45bc6b22f0ac27d4e 15617Author: Maarten Maathuis <madman2003@gmail.com> 15618Date: Tue Nov 6 10:56:22 2007 +0100 15619 15620 randr12: Associate VPLL's with ramdac instead of crtc. 15621 15622commit 8241710c94f6df0bf683bc3c93f7ea1ca14d118c 15623Author: Maarten Maathuis <madman2003@gmail.com> 15624Date: Tue Nov 6 10:34:56 2007 +0100 15625 15626 randr12: Put TEST_CONTROL into neutral position. 15627 15628commit 19533d8ba549965dedd9083330b217270efaa49b 15629Author: Maarten Maathuis <madman2003@gmail.com> 15630Date: Mon Nov 5 23:32:16 2007 +0100 15631 15632 NV50: Don't hit pre-NV50 codepath. 15633 15634commit 5ca10da0c7dd2fd689cc4369e37ed385a0c10d21 15635Author: Maarten Maathuis <madman2003@gmail.com> 15636Date: Mon Nov 5 21:28:22 2007 +0100 15637 15638 randr12: Remove incorrect (and hopefully redundant) output initialisation check. 15639 15640commit fcacad59d9233a66111d1c434e14cc4b2a08961e 15641Author: Maarten Maathuis <madman2003@gmail.com> 15642Date: Mon Nov 5 21:18:13 2007 +0100 15643 15644 randr12: Do not remove any bits from crtc register 580. 15645 15646commit e58c6735c73cc3a94a7a139e235b1dff9220e507 15647Author: Maarten Maathuis <madman2003@gmail.com> 15648Date: Mon Nov 5 20:30:41 2007 +0100 15649 15650 randr12: Only use VPLL as timing source. 15651 15652commit 1071aeef24b349bac08a2ac091733c23cad1cb44 15653Author: Maarten Maathuis <madman2003@gmail.com> 15654Date: Mon Nov 5 20:04:26 2007 +0100 15655 15656 randr12: Only enable crtc at the very last stage, should avoid mode setting "blinking". 15657 15658commit 1ac7798bdec75d9336afd1cfbe7fa177f5fe1465 15659Author: Maarten Maathuis <madman2003@gmail.com> 15660Date: Mon Nov 5 10:03:15 2007 +0100 15661 15662 randr12: Add a native mode for TMDS and make that preferred + some ddc related fixes. 15663 15664commit bfb75679e504b335be0de4384830cc8129f92c1b 15665Author: Ben Skeggs <skeggsb@gmail.com> 15666Date: Mon Nov 5 11:44:58 2007 +1100 15667 15668 Fix a thinko from commit 62294d9be8ec6f1159e3c2ad273649f937b7aa7e. 15669 15670commit bf8a2dfa7f83956d100316ed7450669795ea47d2 15671Author: Maarten Maathuis <madman2003@gmail.com> 15672Date: Sat Nov 3 22:03:18 2007 +0100 15673 15674 randr12; Forgot a small thing. 15675 15676commit 19b0971d0c1d03896bfaee2b833194ef1e4d7566 15677Author: Maarten Maathuis <madman2003@gmail.com> 15678Date: Sat Nov 3 21:54:48 2007 +0100 15679 15680 randr12: A legitimate DPMS call could reset the ramdac and effectively disable the mouse. 15681 15682commit a29839c718fd6bf3003081299017536dd9f8223e 15683Author: Maarten Maathuis <madman2003@gmail.com> 15684Date: Sat Nov 3 10:00:06 2007 +0100 15685 15686 randr12: Hopefully improve the native flatpanel mode finder. 15687 15688commit bd20ff5b8c9467ca913c00d3ec946e7e975ad3d4 15689Author: Maarten Maathuis <madman2003@gmail.com> 15690Date: Fri Nov 2 23:17:16 2007 +0100 15691 15692 randr12: We cannot reset ramdacs during detect, there may not be a modesetting after it. 15693 15694commit 7054de29c589a7be0a6b027f3067ae2c02e2fc45 15695Author: Maarten Maathuis <madman2003@gmail.com> 15696Date: Fri Nov 2 21:44:40 2007 +0100 15697 15698 randr12: Futher refine crtc timings, hoping to solve the "pink line" issue. 15699 15700commit 39ced4abb536993249e55a72eceeba26d577724c 15701Author: Maarten Maathuis <madman2003@gmail.com> 15702Date: Thu Nov 1 17:54:26 2007 +0100 15703 15704 randr12: Improved ramdac choosing, based on bios information (or value). 15705 15706commit 28ba9ed765d9956b2f07c6a405fca675a71012db 15707Author: Maarten Maathuis <madman2003@gmail.com> 15708Date: Thu Nov 1 17:01:53 2007 +0100 15709 15710 randr12: Remove redundant output setup + don't print full edid. 15711 15712commit 609f7b9c2962ccf04861d2164a0ff12c58e84010 15713Author: Maarten Maathuis <madman2003@gmail.com> 15714Date: Thu Nov 1 13:00:45 2007 +0100 15715 15716 randr12: Cursor fixes(probably only affects nv4x). 15717 15718commit 6945dfdc5e6b2540432805e845cb8540dc00d326 15719Author: Maarten Maathuis <madman2003@gmail.com> 15720Date: Thu Nov 1 11:10:24 2007 +0100 15721 15722 randr12: Improve analog monitor allignment + misc fixes. 15723 15724commit 5bd8ba52788b3b3b8f91ba59c29e154e38482481 15725Author: Ben Skeggs <skeggsb@gmail.com> 15726Date: Thu Nov 1 18:35:33 2007 +1100 15727 15728 nv40: what can I say? oops! 15729 15730commit f59e596bcd90ef824cd41e0c37952e574d6914bb 15731Author: Ben Skeggs <skeggsb@gmail.com> 15732Date: Thu Nov 1 18:17:20 2007 +1100 15733 15734 nv40: completely untested fix for composite on ppc 15735 15736 Thanks go to ps3rsx guys for finding the issue. 15737 15738commit 59c4fdb650ff2e6b6948d1bdd6bb22aa94195a32 15739Author: Maarten Maathuis <madman2003@gmail.com> 15740Date: Wed Oct 31 22:59:54 2007 +0100 15741 15742 randr12: Some misc cleanup to make malc0's patches work. 15743 15744commit f165d4f2ca90916b1334cdc18d62b82ea71d9f62 15745Author: Maarten Maathuis <madman2003@gmail.com> 15746Date: Wed Oct 31 21:30:36 2007 +0100 15747 15748 randr12: The "easy" TMDS registers, please report regressions. 15749 15750commit 860c4bdd38fd259ae51e32e4873a2aa8b1e6883e 15751Author: Maarten Maathuis <madman2003@gmail.com> 15752Date: Wed Oct 31 19:00:28 2007 +0100 15753 15754 randr12: Misc register fixes. 15755 15756commit d758e8f4c1a7d4effe1298a0a41d4e0b96496ab9 15757Author: Maarten Maathuis <madman2003@gmail.com> 15758Date: Wed Oct 31 13:22:29 2007 +0100 15759 15760 randr12: Some changes to VCLK ratio handling. 15761 15762commit 899559921b0e46db010fa1e771203c2edefb7ade 15763Author: Maarten Maathuis <madman2003@gmail.com> 15764Date: Wed Oct 31 09:18:59 2007 +0100 15765 15766 randr12: some blob copying 15767 15768commit b5ecadb73335d934a8cb1c106983d219c9d8d399 15769Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15770Date: Wed Oct 31 02:09:08 2007 +0100 15771 15772 Fix bugs in nv30 exa (kde text seems back) 15773 15774commit f7b553a87e7f67138e06e74877145a1f83e9d188 15775Author: Peter Winters <peter@whoei.org> 15776Date: Tue Oct 30 23:04:28 2007 +0100 15777 15778 NV10 EXA: implement repeat for 1x1 textures 15779 15780commit bf7b0231db1f7e45cee60d7d2e04b1b006aec2b0 15781Author: Peter Winters <peter@whoei.org> 15782Date: Sun Oct 28 23:52:22 2007 +0100 15783 15784 NV10 EXA: Fix viewport setup 15785 15786commit f8f9c8789b89a02886702c46c5feafec169ae78a 15787Author: Maarten Maathuis <madman2003@gmail.com> 15788Date: Tue Oct 30 21:00:54 2007 +0100 15789 15790 randr12: Flatpanel scaling is finally working, read full message. 15791 15792 - The option FPScale is now obeyed and will cause scaling to happen in the monitor. 15793 - The default is scaling is using the GPU. 15794 - When the aspect ratio is very different(>0.10 diff), it will obey aspect ratio and create bands around it. 15795 15796commit e73aa148d965abbf73392c6ffef3993dc26c31fd 15797Author: Maarten Maathuis <madman2003@gmail.com> 15798Date: Tue Oct 30 13:54:25 2007 +0100 15799 15800 randr12: Some copyright stuff. 15801 15802commit c1508e1448e3822e8b6bb2604c09a9d766a64a93 15803Author: Maarten Maathuis <madman2003@gmail.com> 15804Date: Tue Oct 30 13:44:44 2007 +0100 15805 15806 randr12: Disable some panel scaling stuff, because it is broken. 15807 15808commit 3bf8110b61ef854af803a45f0faf1eaffbbcfac9 15809Author: Maarten Maathuis <madman2003@gmail.com> 15810Date: Tue Oct 30 11:44:58 2007 +0100 15811 15812 randr12: Lots of misc fixes. 15813 15814 - This does not fix the issue with -hsync, +vsync modelines (or the other way around) yet. 15815 - There are two PVIO ranges, i'm surprised noone noticed this. 15816 - Some misc fixes. 15817 15818commit 894f9aacce3d685b811f8961275a9f225b964159 15819Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15820Date: Tue Oct 30 01:49:38 2007 +0100 15821 15822 Disable nv30 exa no big endian until we fix the endianness bugs related to the 3D engine. 15823 15824commit 482cbcb9efdfbdd4bbe998c09f7d29033aeca09b 15825Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15826Date: Tue Oct 30 01:37:45 2007 +0100 15827 15828 Use the new header with license. 15829 15830commit a9908dbb70df0d0abb586b9774efdc10adf4d74f 15831Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15832Date: Mon Oct 29 23:00:14 2007 +0100 15833 15834 More copyrights I remember... 15835 15836commit 67e142394edcb553a40c6c8b0ef091bd4dad2bdd 15837Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15838Date: Mon Oct 29 22:57:30 2007 +0100 15839 15840 Add some copyrights... not all files are covered yet, but I did the ones I recalled. 15841 15842commit f75e36a901fd47bbd9164242c408e9f8b0e03e2c 15843Author: Maarten Maathuis <madman2003@gmail.com> 15844Date: Mon Oct 29 18:37:34 2007 +0100 15845 15846 randr12: My suspicion was incorrect, i2c buses are swapped for all nv4x 15847 it seems. 15848 15849commit e741ac76ac45e8ea959792810670abdf2afbc6bd 15850Author: Maarten Maathuis <madman2003@gmail.com> 15851Date: Mon Oct 29 13:40:30 2007 +0100 15852 15853 randr12: Show NV_RAMDAC_FP_DEBUG_0 of both ramdac's when output is refused. 15854 15855commit c53754162087a7b66d40b7c3982cdc5b6d160ea1 15856Author: Maarten Maathuis <madman2003@gmail.com> 15857Date: Mon Oct 29 13:34:20 2007 +0100 15858 15859 randr12: Minor change, turning on bit 18 of NV_RAMDAC_SEL_CLK. 15860 15861commit fef872bd999b430a41bdfae8d408f8dfed0010a2 15862Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15863Date: Fri Oct 26 15:50:40 2007 +0200 15864 15865 nv30 exa : add A8 dst (requires new drm), add some dst formats, handle 1x1 repeat, enable by default. 15866 15867commit 1d7354cbb5875eb13ed88e6aaab69aff615f82b2 15868Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15869Date: Fri Oct 26 04:37:51 2007 +0200 15870 15871 Fix nv30 exa. 15872 15873commit 2cd3d47713cf3eb0d36124902a2c66b689acba51 15874Author: Arthur Huillet <arthur.huillet@free.fr> 15875Date: Thu Oct 25 16:55:18 2007 +0200 15876 15877 s/#define ABC ;/#define ABC do { ; } while(0);/ 15878 15879commit 228b91d3da445fc58bfe6b13e792bbdff3728301 15880Author: Arthur Huillet <arthur.huillet@free.fr> 15881Date: Thu Oct 25 16:40:34 2007 +0200 15882 15883 NV10 EXA: implemented masking via register combiners and added (disabled by default) statistics displaying 15884 15885 masking works for ARGB, XRGB (it's a nop), A8 - no component alpha for now, but it will come soon hopefully 15886 15887commit 7d77ab0692126cea9248b7ee61b001e138b4e47c 15888Author: Arthur Huillet <arthur.huillet@free.fr> 15889Date: Thu Oct 25 10:52:59 2007 +0200 15890 15891 NV10 EXA: fixed A8 as pointed out by p0g 15892 15893commit cdb54271837564bb4a5afd2dc4cf7e7169138975 15894Author: Arthur Huillet <arthur.huillet@free.fr> 15895Date: Thu Oct 25 10:38:42 2007 +0200 15896 15897 NV10 EXA: added support for A8 as source 15898 15899commit 220acbe07e84015e489257a9b9dae8e8f1be8268 15900Author: Peter Winters <peter@whoei.org> 15901Date: Thu Oct 25 10:03:39 2007 +0200 15902 15903 NV10 EXA: use proper XRGB format and use correct RC setup 15904 15905commit 6034387ea96816d34b627c306998342136ecb01b 15906Author: Peter Winters <peter@whoei.org> 15907Date: Thu Oct 25 09:59:01 2007 +0200 15908 15909 NV10 EXA: always use an even width for NPOT_SIZE 15910 15911commit f7f17b789e89aadafa2734cba0f4464c5f317a68 15912Author: Jeremy Kolb <jkolb@brandeis.edu> 15913Date: Wed Oct 24 19:34:49 2007 -0400 15914 15915 nv30: Fix some corruption. Give us some semblance of icons (still messed up though) 15916 15917commit bafa34a7bd7a9136a95c8ced25c5b80556c78c0c 15918Author: Maarten Maathuis <madman2003@gmail.com> 15919Date: Wed Oct 24 20:28:46 2007 +0200 15920 15921 Fix a few NV_IMAGE_BLIT offsets. 15922 15923commit 09f4b66d35543037c4f50910d0265751dded66c0 15924Author: Arthur Huillet <arthur.huillet@free.fr> 15925Date: Wed Oct 24 10:40:17 2007 +0200 15926 15927 NV10EXA : enabled by default for NV1x. Waiting for pq before doing the same for NV2x. 15928 15929commit 6c978164a212ddf4f60f82cdc72b11ebbc96efbb 15930Author: Arthur Huillet <arthur.huillet@free.fr> 15931Date: Wed Oct 24 09:37:34 2007 +0200 15932 15933 Xv: disabled YV12 overlay for NV20 chipsets that do not seem to have it 15934 15935commit 9f48bd410f8cf9aa700e03db807cef2e7cc07e34 15936Author: Maarten Maathuis <madman2003@gmail.com> 15937Date: Wed Oct 24 09:08:41 2007 +0200 15938 15939 randr12: The last step to make lower flatpanel resolutions working. 15940 15941 - Note: the previous randr12 commit also contained some fixes in this area. 15942 15943commit 2bf8bbd1764979d194ff2dbac4a6529cbcbdbc1d 15944Author: Maarten Maathuis <madman2003@gmail.com> 15945Date: Wed Oct 24 00:00:59 2007 +0200 15946 15947 randr12: Misc changes. 15948 15949commit 5841664c82126058e582f215af2323cc80a6150d 15950Author: Arthur Huillet <arthur.huillet@free.fr> 15951Date: Tue Oct 23 22:53:46 2007 +0200 15952 15953 x8r8g8b8 now seems to work (it passes rendercheck) with format 0x380 15954 15955commit 27a713057f0fd2d3c8a77f5eff86393059b60511 15956Author: Arthur Huillet <arthur.huillet@free.fr> 15957Date: Tue Oct 23 22:53:21 2007 +0200 15958 15959 Remove NV10 texture formats that are all wrong. 15960 15961commit 78a0d5a42f96816889f1c9294f11c637dfcba486 15962Author: Arthur Huillet <arthur.huillet@free.fr> 15963Date: Tue Oct 23 19:12:19 2007 +0200 15964 15965 NV10 EXA: A8R8G8B8 without mask works with rendercheck now 15966 15967commit 7543e65cecab510f6bea8b6efc5cdf3cc9d261ae 15968Author: Maarten Maathuis <madman2003@gmail.com> 15969Date: Tue Oct 23 18:41:38 2007 +0200 15970 15971 randr12: actually add scaling factor's 15972 15973 - This was forgotten when i first added this from the haiku driver. 15974 15975commit a11faa794c5dd24cd74d949006c8e86c6c3f0310 15976Author: Maarten Maathuis <madman2003@gmail.com> 15977Date: Tue Oct 23 16:22:56 2007 +0200 15978 15979 randr12: Add a few debug statements that should explain why an output is refused. 15980 15981commit 2d842c39bf210b4b8294a447ec9ac194ec3e2e47 15982Author: Maarten Maathuis <madman2003@gmail.com> 15983Date: Tue Oct 23 16:19:10 2007 +0200 15984 15985 randr12: Check if output is needed before creating it. 15986 15987 - This hopefully fixes a problem some NV30 people have. 15988 15989commit 5c884a188296c9a026f110f272dff5d88d58c0b6 15990Author: Maarten Maathuis <madman2003@gmail.com> 15991Date: Tue Oct 23 15:47:49 2007 +0200 15992 15993 Revert a G70 workaround, this may BREAK your setup. 15994 15995 - Please report if you have issues, but be prepared to help out to find a real fix. 15996 15997commit 0fb93b45378c9d0fbee9bb96bd60d7a995c667b9 15998Author: Arthur Huillet <arthur.huillet@free.fr> 15999Date: Mon Oct 22 19:02:17 2007 +0200 16000 16001 added (disabled) NV10 EXA hooks 16002 16003commit 78af60e798f5b91cc128cd2f84ddae7d0a1523cd 16004Author: Arthur Huillet <arthur.huillet@free.fr> 16005Date: Mon Oct 22 19:00:20 2007 +0200 16006 16007 added NV10 EXA functions to proto header 16008 16009commit cf053adacabaee887ecedaa9967b07b3185095b5 16010Author: Peter Winters <peter@whoei.org> 16011Date: Mon Oct 22 17:04:23 2007 +0200 16012 16013 Fix some blending stuff for NV10 EXA 16014 16015commit b0605fbeef0765e39554a15ea9f871c19258f72c 16016Author: Peter Winters <peter@whoei.org> 16017Date: Mon Oct 22 17:01:26 2007 +0200 16018 16019 Add basic RC setup for NV10 EXA 16020 16021commit 8dea36f6533c5fc35dcb3c24e2fe0c1d009b6693 16022Author: Peter Winters <peter@whoei.org> 16023Date: Mon Oct 22 16:50:47 2007 +0200 16024 16025 Fix NV10 EXA transform 16026 16027commit b9f97583979d7dbbf0058de9f807a97812362c71 16028Author: Peter Winters <peter@whoei.org> 16029Date: Mon Oct 22 16:48:53 2007 +0200 16030 16031 Fix NV10 EXA list of supported formats 16032 16033commit 3dc9778902d9412c469d79e3d437838a6c9da7a5 16034Author: Peter Winters <peter@whoei.org> 16035Date: Mon Oct 22 16:44:08 2007 +0200 16036 16037 Various NV10 EXA fixes 16038 16039commit 391e2dd1216f598538524fdc52a14bfdd4451567 16040Author: Ben Skeggs <skeggsb@gmail.com> 16041Date: Mon Oct 22 22:48:13 2007 +1100 16042 16043 Fix bad s/ 16044 16045commit 162edc70769b24b39f432cb79f5959eff363c564 16046Author: Pekka Paalanen <pq@iki.fi> 16047Date: Sun Oct 21 19:55:41 2007 +0300 16048 16049 Revert "EXA: now caching BLIT_OPERATION and RECT_OPERATION..." 16050 16051 This reverts commit 6d8caf5e0dd915809152c52c2c56a39d76e2ed8c. 16052 16053 Conflicts: 16054 16055 src/nv_exa.c 16056 16057 The above mentioned commit broke EXA copy and solid operations on nv20 16058 after display mode change. Triggered by running freedroid in fullscreen. 16059 (Not freedroidRPG) 16060 16061commit cc04b651348768dc214f59211ba396343aaa6a17 16062Author: Maarten Maathuis <madman2003@gmail.com> 16063Date: Sat Oct 20 20:30:02 2007 +0200 16064 16065 Add back comment. 16066 16067commit 6fd262027c74d5008e484bb5f09edbf5f657470b 16068Author: Maarten Maathuis <madman2003@gmail.com> 16069Date: Sat Oct 20 00:26:02 2007 +0200 16070 16071 randr12: Fix critical bug and minor change. 16072 16073 - The previous code could allow the TMDS[0x4] register of both heads to be set to 0x88. 16074 - This caused problems and freaked out my monitor and scared me as well :-( 16075 - Until further notice, the usage of this swap bit is restricted to very specific situations. 16076 16077commit 377f69172b4935372538d7bbf26852bf8a1fe1c4 16078Author: Maarten Maathuis <madman2003@gmail.com> 16079Date: Fri Oct 19 20:57:58 2007 +0200 16080 16081 randr12: Some changes to crtc regs. 16082 16083commit 85dac1041b38127d41832a6a322d01f87d70f28f 16084Author: Maarten Maathuis <madman2003@gmail.com> 16085Date: Fri Oct 19 08:06:33 2007 +0200 16086 16087 randr12: Some changes to output detection. 16088 16089 - I'm aware that we still can't fully program modes. 16090 16091commit 8320f4f61ca72ee0101c1d552ea3ef28e573d21d 16092Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16093Date: Thu Oct 18 01:56:17 2007 +0200 16094 16095 Remove xaa includes. 16096 16097commit 86855f03945d3bad90da0b28311bece484cbfce7 16098Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16099Date: Thu Oct 18 01:55:06 2007 +0200 16100 16101 Handle -minline-all-stringops properly 16102 16103commit 53ee9726eeda8cf844d32a9563c82b7549ba6c03 16104Author: Ben Skeggs <skeggsb@gmail.com> 16105Date: Wed Oct 17 23:28:50 2007 +1100 16106 16107 re-align nv40exa after ring macro changes 16108 16109commit a5c4786aaeadce6103fcc9b3a6f4e6f25a66730f 16110Author: Ben Skeggs <skeggsb@gmail.com> 16111Date: Wed Oct 17 23:22:05 2007 +1100 16112 16113 Switch to *_RING instead of NVDma* 16114 16115commit 0d1835d67e38ca12347a7965ac43fbfc29063b87 16116Author: Ben Skeggs <skeggsb@gmail.com> 16117Date: Wed Oct 17 16:29:01 2007 +1100 16118 16119 Switch to header generated from XML + other cleanups. 16120 16121 - Renamed object header to nouveau_class.h, so it's not confused with mmio 16122 - Removed usage of method defines in nv_dma.h 16123 - Removed 2D engine EXA composite hook, it was broken, useless and soon to 16124 be obsoleted by card-specific versions. 16125 - Use names in a few places that magic values were previously used. 16126 16127commit b36fbf174849b04928efda2909c461b6c1688cdc 16128Author: Dave Airlie <airlied@optimus.(none)> 16129Date: Tue Oct 16 09:28:58 2007 +1000 16130 16131 nouveau: missing include 16132 16133commit 8425d93f7f932281a1b9b420e676acba2a2b5ecd 16134Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16135Date: Sun Oct 14 17:10:59 2007 +0200 16136 16137 more changes to nv30 exa... displays the pictures at their correct postition and size but doesn't obey the blending parameters (i.e. no blending takes place). 16138 16139commit 212a4cd8884ff9276f64bdd0a0115782d146e36a 16140Author: Maarten Maathuis <madman2003@gmail.com> 16141Date: Sat Oct 13 23:02:27 2007 +0200 16142 16143 randr12: Some fallback defaults for dual-dvi cards, please submit dumps when you can. 16144 16145commit 6d8caf5e0dd915809152c52c2c56a39d76e2ed8c 16146Author: Arthur Huillet <arthur.huillet@free.fr> 16147Date: Sat Oct 13 22:28:39 2007 +0200 16148 16149 EXA: now caching BLIT_OPERATION and RECT_OPERATION (0x2FC) to limit the number of IRQs on NV04 16150 16151commit ceaf8df424651e0f20d63f817864f3e8657f2c21 16152Author: Maarten Maathuis <madman2003@gmail.com> 16153Date: Sat Oct 13 21:45:50 2007 +0200 16154 16155 randr12: Revert a small unnecesary change. 16156 16157commit 7171acbe7f174dcbe8c28c7f1fcb7481f2fafd90 16158Author: Maarten Maathuis <madman2003@gmail.com> 16159Date: Sat Oct 13 21:33:07 2007 +0200 16160 16161 randr12: Hopefully fix panel size detection and allow negative hsync/vsync polarities. 16162 16163commit be08f09ad1ba6aeebf4491727052d246d5568fd2 16164Author: Maarten Maathuis <madman2003@gmail.com> 16165Date: Sat Oct 13 17:26:32 2007 +0200 16166 16167 randr12: Some debug stuff. 16168 16169commit 418e6117a25e51d5b0dbcd87c17f78f9b04fa588 16170Author: Maarten Maathuis <madman2003@gmail.com> 16171Date: Sat Oct 13 12:52:10 2007 +0200 16172 16173 randr12: Remove a double reg and misc changes. 16174 16175commit e30e21ed0ee8b1bd112780badc933dc4552a2e81 16176Author: Maarten Maathuis <madman2003@gmail.com> 16177Date: Fri Oct 12 19:49:32 2007 +0200 16178 16179 randr12: Some misc changes. 16180 16181commit 9425f7a3ea2d80f4df79239e1ca8000bd729c772 16182Author: Jeremy Kolb <jkolb@brandeis.edu> 16183Date: Thu Oct 11 20:32:47 2007 -0400 16184 16185 Remove division in NV30EXATransformCoord. We want coords between 0 and texture width, height. 16186 16187commit 879e653867552f6bb595160670c8325bcb5a0d3b 16188Author: Maarten Maathuis <madman2003@gmail.com> 16189Date: Wed Oct 10 21:40:13 2007 +0200 16190 16191 randr12: Misc fix. 16192 16193commit 99dd9da140a5750b7c4d6030a0e8e1f7fca44033 16194Author: Maarten Maathuis <madman2003@gmail.com> 16195Date: Wed Oct 10 20:08:28 2007 +0200 16196 16197 randr12: Fix switch mode function, so fullscreen apps don't screw up your screen by definition ;-) 16198 16199commit 0c80565e2cb7fa5ff359f57cc3d7016a78a43ac2 16200Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16201Date: Wed Oct 10 14:43:34 2007 +0200 16202 16203 Oops. 16204 16205commit 761ef4bd32f42fb77a3e48985fd6dda81aa081f5 16206Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16207Date: Wed Oct 10 14:42:46 2007 +0200 16208 16209 Try to do something for our desastrous UTS/DFS performance. My PCI card is really hurting. 16210 16211commit 254f9b075034c86e5936954b52f499d573880039 16212Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16213Date: Wed Oct 10 02:34:06 2007 +0200 16214 16215 Still more work on nv30 exa... now starts but nothing works. 16216 16217commit 79dc3ad2c89eef74596a4c8e5714c2c661f56fee 16218Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16219Date: Wed Oct 10 00:48:46 2007 +0200 16220 16221 Don't swap the registers from user space. This is very wrong, and the drm will handle that soon. 16222 16223commit 2c08930f0b2bf06282976d6a357daf7f45405cdc 16224Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16225Date: Wed Oct 10 00:48:06 2007 +0200 16226 16227 Revert back to the other shaders. The terminator instruction doesn't seem to help, and breaks the A8 hack. 16228 16229commit 90d2457125b059916f9b9dd5ddb7cec73a554c55 16230Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16231Date: Tue Oct 9 04:38:29 2007 +0200 16232 16233 Cleanup some nv30 exa issues, add debbuging info to the composite hook. 16234 16235commit e184fc802c7f9103bbc8ee8c036cfe5a9ddd9513 16236Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16237Date: Tue Oct 9 03:28:39 2007 +0200 16238 16239 Some fixes to nv30_exa. Now goes through a couple of composite() calls without crashing. 16240 16241commit 708f28dc8175375713cd9ae4d07b2cfcea3b106c 16242Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16243Date: Tue Oct 9 03:28:11 2007 +0200 16244 16245 Use terminators on the nv30/40 shaders. 16246 16247commit 43d5f747103721b3e0ccf4fe73494729ed34986c 16248Author: Maarten Maathuis <madman2003@gmail.com> 16249Date: Mon Oct 8 19:53:56 2007 +0200 16250 16251 randr12: Small changes makes finding regressions easy ;-) 16252 16253commit 62294d9be8ec6f1159e3c2ad273649f937b7aa7e 16254Author: Maarten Maathuis <madman2003@gmail.com> 16255Date: Sat Oct 6 19:04:51 2007 +0200 16256 16257 EXA: Starting from version 2.4 we can't use devPrivate.ptr anymore. 16258 16259commit 721de8da8fccd6f7b6e5a3044ce82616cdb689e8 16260Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16261Date: Fri Oct 5 23:50:18 2007 +0200 16262 16263 some more work on nv30 exa... 16264 16265commit 80bcbf493b6b2ad927c7a3826cc6e5fa46bca76e 16266Author: Maarten Maathuis <madman2003@gmail.com> 16267Date: Fri Oct 5 22:29:07 2007 +0200 16268 16269 randr12: Misc changes. 16270 16271commit 63ad301baab5803329ba0f192f12c917b435594e 16272Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16273Date: Fri Oct 5 17:20:05 2007 +0200 16274 16275 mandatory oops commit. 16276 16277commit 3773e0e5045a6c347bf26eff71e6bdd4c592cebc 16278Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16279Date: Fri Oct 5 17:18:40 2007 +0200 16280 16281 Split nv30 exa and nv40 exa. 16282 16283commit 37900537c0f8b8dfa3dc80300d6809f1f4ac9423 16284Author: Jeremy Kolb <jkolb@brandeis.edu> 16285Date: Thu Oct 4 23:15:38 2007 -0400 16286 16287 NV30 texture changes. Commented out table that appears to work for nv35 at least. 16288 16289commit 44eb015e88caba82c8d84fee4ea71768c9a6b2a3 16290Author: Jeremy Kolb <jkolb@brandeis.edu> 16291Date: Thu Oct 4 20:37:35 2007 -0400 16292 16293 Fix NV30 arch check. 16294 16295commit 5a76e5a751cd66021565ab92b2fc3c61204c7d73 16296Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16297Date: Thu Oct 4 23:46:43 2007 +0200 16298 16299 fix nv30_exa for nv40. 16300 16301commit d7b767fc951287a1c87965db3c236e05cc94d8f6 16302Author: Maarten Maathuis <madman2003@gmail.com> 16303Date: Thu Oct 4 21:49:33 2007 +0200 16304 16305 NV50: Fix a minor mistake. 16306 16307commit 926e887a4b374768648fe772b1b3d7730096e45f 16308Author: Maarten Maathuis <madman2003@gmail.com> 16309Date: Wed Oct 3 22:57:56 2007 +0200 16310 16311 randr12: Some reindenting. 16312 16313commit 7ee3f670077641e14a0ff024443805564d5be599 16314Author: Patrice Mandin <pmandin@caramail.com> 16315Date: Thu Oct 4 19:54:40 2007 +0200 16316 16317 nv10exa: init link between tcl and image blit for >=nv11 16318 16319commit 6510e75ada6fb921e417588d98afdc58b2c14972 16320Author: Jeremy Kolb <jkolb@brandeis.edu> 16321Date: Wed Oct 3 23:34:23 2007 -0400 16322 16323 nv30_exa: send down pitch correctly 16324 16325commit b7fc7b63eec4a5e23879120284c882e19e095471 16326Author: Jeremy Kolb <jkolb@brandeis.edu> 16327Date: Wed Oct 3 20:54:28 2007 -0400 16328 16329 Fill in some regs. 16330 16331commit 46facb7c29d2737f1cc414655de8e09ee0aea22f 16332Author: Maarten Maathuis <madman2003@gmail.com> 16333Date: Wed Oct 3 22:16:13 2007 +0200 16334 16335 randr12: Actually borrow some timing stuff from Haiku correctly ;-) + new crtc reg 16336 16337commit 9d6b3216ed519e89386480d04070b1c1e4702921 16338Author: Matthieu Castet <castet.matthieu@free.fr> 16339Date: Wed Oct 3 21:21:41 2007 +0200 16340 16341 abort on Init table command not found when parsing bios 16342 16343commit b6351c0efa3629a448557fc7e597ba14fbf057fb 16344Author: Maarten Maathuis <madman2003@gmail.com> 16345Date: Wed Oct 3 19:54:40 2007 +0200 16346 16347 NV50: Fix small mistake. 16348 16349commit f00ea962e549934965100dc6d2806709129995e8 16350Author: Maarten Maathuis <madman2003@gmail.com> 16351Date: Wed Oct 3 19:03:38 2007 +0200 16352 16353 randr12 + NV50: Misc changes. 16354 16355 - Switched everything back to head variable instead of pcio. 16356 - Removed ugly C() macro for NV50. 16357 - Probably some other stuff too. 16358 - Please test NV5x users ;-) 16359 16360commit 5d7f06c2a205004b78692b4dfad477ade3947fd5 16361Author: Maarten Maathuis <madman2003@gmail.com> 16362Date: Tue Oct 2 19:31:10 2007 +0200 16363 16364 NV50: Some infrastructure and minor changes. 16365 16366commit 7680648c9ddb81e2a1c6295431d7074366fb7b9a 16367Author: Maarten Maathuis <madman2003@gmail.com> 16368Date: Tue Oct 2 16:55:07 2007 +0200 16369 16370 NV50: The beginning of moving some code around. 16371 16372 - Please contact me if something breaks, i don't have an NV5x card. 16373 16374commit b812ecaef75216cc76557e6a3a080efa870d189f 16375Author: Maarten Maathuis <madman2003@gmail.com> 16376Date: Tue Oct 2 10:35:37 2007 +0200 16377 16378 NV30EXA: Minor code cleanup. 16379 16380commit 5b974b3bed86d53ccf6915c46aeb310a2bba3519 16381Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16382Date: Mon Oct 1 17:19:41 2007 +0200 16383 16384 Remove those useless casts. 16385 16386commit 734785c05357455a7ae7181fd7326f02e0bd0478 16387Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16388Date: Mon Oct 1 02:01:30 2007 +0200 16389 16390 Some work on nv10 exa. 16391 16392commit 0cfcbcc9d64bfd47b765f4f2ff41d7ad467990bd 16393Author: Matthieu Castet <castet.matthieu@free.fr> 16394Date: Sun Sep 30 22:41:15 2007 +0200 16395 16396 enable bios PINS parsing on post nv10 cards 16397 16398commit 9ac6d0e9d46abdf867fd06f7ca24b718a340bb74 16399Author: Pekka Paalanen <pq@iki.fi> 16400Date: Sun Sep 30 23:09:33 2007 +0300 16401 16402 nv20 inits nv10tcl. 16403 16404commit c3439aebe000a3c6dd716bd4739ce01321bee26b 16405Author: Maarten Maathuis <madman2003@gmail.com> 16406Date: Sun Sep 30 21:43:41 2007 +0200 16407 16408 Fix detection of NV04 and NV05 when using pci-rework. 16409 16410commit 9e070c09f68b0e3e41109360c455b3e2cc1dbb9c 16411Author: Maarten Maathuis <madman2003@gmail.com> 16412Date: Sun Sep 30 19:37:53 2007 +0200 16413 16414 randr12: Switch to proper cursor api + minor changes + small nv50 fix 16415 16416 - Remove the ugly cursor hack. 16417 - A function name on nv50 was still g80, this caused an implicit decleration. 16418 16419commit c7de03556a51009ce69b073d7ecfd01919d24c5f 16420Author: Maarten Maathuis <madman2003@gmail.com> 16421Date: Sun Sep 30 12:40:50 2007 +0200 16422 16423 NV30EXA: Fix rendering issues. 16424 16425 - Full non-composite tests using rendercheck. 16426 - The full test suite of composite has been run (it takes ~15 hours). 16427 - A partial cacomposite has been run (a8, a8r8g8b8, x8r8g8b8). 16428 - All the rendering tests work fine, as does the cairo-knockout demo. 16429 - Thanks to doener for reporting this issue. 16430 16431commit 0574dc5a8bd09216e41656c943c92a437bcf020b 16432Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16433Date: Sat Sep 29 20:20:47 2007 +0200 16434 16435 A little more work on nv10 exa. 16436 16437commit aae155f149af522e1a83168ed4ec5b7b5a40bb73 16438Author: Maarten Maathuis <madman2003@gmail.com> 16439Date: Sat Sep 29 17:28:10 2007 +0200 16440 16441 randr12: Some more CRTC regs. 16442 16443commit 8b03086ee36ca9c3e77f9df9eb4a1805072d2103 16444Author: Maarten Maathuis <madman2003@gmail.com> 16445Date: Sat Sep 29 14:33:38 2007 +0200 16446 16447 randr12: make crtc owner beheaviour consistent with blob + minor changes 16448 16449commit c572315a974746224dba1213e956a1b6189f175d 16450Author: Maarten Maathuis <madman2003@gmail.com> 16451Date: Fri Sep 28 19:58:39 2007 +0200 16452 16453 randr12: Minor changes and more debugging output. 16454 16455commit db36a1df4ba3dddf3669f491bfe6461e4fd490b6 16456Author: Maarten Maathuis <madman2003@gmail.com> 16457Date: Fri Sep 28 18:52:47 2007 +0200 16458 16459 randr12: Some more debugging output and some other minor changes. 16460 16461commit 2d21405c9c0bff36f47e28a5e4086dacae557d17 16462Author: Maarten Maathuis <madman2003@gmail.com> 16463Date: Fri Sep 28 13:02:51 2007 +0200 16464 16465 randr12: Preperation to allow differentation between crtc and pcio in the future. 16466 16467commit f235cf896c1da098ff635d5a882561f16f5fb4f7 16468Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16469Date: Thu Sep 27 18:21:25 2007 +0200 16470 16471 Import nv10 exa code. Does not work yet. 16472 16473commit 74a243d8e188432854c15849ad1ab6e893e48f0a 16474Author: Maarten Maathuis <madman2003@gmail.com> 16475Date: Wed Sep 26 09:04:43 2007 +0200 16476 16477 randr12: Some registers were not constants, so changed it back with a note about the values i had seen. 16478 16479commit a3c71495097f8d745110aad06c9f8e2f6f03baad 16480Author: Maarten Maathuis <madman2003@gmail.com> 16481Date: Tue Sep 25 22:05:36 2007 +0200 16482 16483 randr12: Getting closer, doing a decent attempt at detecting the proper ramdac. 16484 16485 - crtc detection may need to be added 16486 - When i boot up with both analog and dvi attached, my dvi is misprogrammed, maybe some sharing is happening somewhere? 16487 16488commit 4a57ebb9c2098ce5d568a2da6e0640cdcc18d3f0 16489Author: Maarten Maathuis <madman2003@gmail.com> 16490Date: Tue Sep 25 21:06:03 2007 +0200 16491 16492 randr12: detection changes, more to come later 16493 16494 - Detection of crtc's (or a way to program them effectively) still has to be done/fixed. 16495 - It now detects if tmds controls is on another ramdac, for a non-dual dvi setup. 16496 - Disabled TMDS stuff for the moment, since it's incorrect somehow. 16497 - Dumps for single dvi, dual dvi, etc are appreciated. 16498 16499commit 41df1ff6ce348dc6a38e34ce63887c66b0300c09 16500Author: Maarten Maathuis <madman2003@gmail.com> 16501Date: Tue Sep 25 16:22:38 2007 +0200 16502 16503 randr12: Some reindenting. 16504 16505commit 334a9b065c4973876b894eadf05ce86ca439e08f 16506Author: Maarten Maathuis <madman2003@gmail.com> 16507Date: Tue Sep 25 16:17:03 2007 +0200 16508 16509 randr12: simplification of the swapped i2c buses "workaround" for nv4x and some code reindenting 16510 16511commit c40559dacd6fe20f35710111330dee37fb2b9a58 16512Author: Maarten Maathuis <madman2003@gmail.com> 16513Date: Sun Sep 23 01:16:55 2007 +0200 16514 16515 randr12: A frail attempt at some TMDS programming, i have not figured out yet what switches DVI from one RAMDAC to the other. 16516 16517commit 50bc4dbff6df80f5d80f731fe42b06768c429dc0 16518Author: Maarten Maathuis <madman2003@gmail.com> 16519Date: Sun Sep 23 00:57:18 2007 +0200 16520 16521 randr12: Reindenting and minor changes. 16522 16523commit 1a3303a5b6f261470c1c56f84f9b89af950a0b35 16524Author: Maarten Maathuis <madman2003@gmail.com> 16525Date: Sat Sep 22 22:21:37 2007 +0200 16526 16527 randr12: Assign ramdac's in more sane way, otherwise two outputs may end up on the same ramdac. 16528 16529commit 9b62853dfe1664f3624f6be8aa86f6ca15ca8392 16530Author: Maarten Maathuis <madman2003@gmail.com> 16531Date: Sat Sep 22 20:37:32 2007 +0200 16532 16533 randr12: Add some TMDS register infrastructure, so i can try some stuff in the future. 16534 16535commit a445cbcd71ee51d7680a6eb13cd037e06d59b0ba 16536Author: Maarten Maathuis <madman2003@gmail.com> 16537Date: Sat Sep 22 15:55:01 2007 +0200 16538 16539 randr12: hardcoding a base value for debug_0 register. 16540 16541commit f89a3217a18261424e212b355628ce5f5b8b9142 16542Author: Maarten Maathuis <madman2003@gmail.com> 16543Date: Fri Sep 21 23:16:40 2007 +0200 16544 16545 randr12: Some minor changes. 16546 16547commit 5ea59f8f01dbbd4211aee1a0e46a6529cddc016c 16548Author: Maarten Maathuis <madman2003@gmail.com> 16549Date: Fri Sep 21 20:37:50 2007 +0200 16550 16551 randr12: hardcode more stuff + actually use a few previous regs 16552 16553 - I'm going to try and do full modesetting, before tackling the problems that dual monitor setups face (i tried, it wasn't pretty). 16554 16555commit 0f66c725cb8806041faddf94cceb88009f7e9f06 16556Author: Dave Airlie <airlied@redhat.com> 16557Date: Thu Sep 20 08:29:43 2007 +1000 16558 16559 make distcheck pass 16560 16561commit bf3cceeb9c72c16a1cf601c8bd0dedd8c676e3cd 16562Author: Maarten Maathuis <madman2003@gmail.com> 16563Date: Mon Sep 17 18:58:31 2007 +0200 16564 16565 randr12: Add some CRTC regs which the blob sets. 16566 16567commit e3346c17c1ead9619206a5921994b5de7ed72739 16568Author: Maarten Maathuis <madman2003@gmail.com> 16569Date: Mon Sep 17 18:22:53 2007 +0200 16570 16571 randr12: Change the way CRTC register 59 is set. 16572 16573 - Small change, i suspect the original author had a dfp on the second crtc. 16574 - Information taken from a mmio trace of a nv28. 16575 16576commit 6c3aee9d68abdb9303cf03d9024f4f0deb5f1e3e 16577Author: Ben Skeggs <skeggsb@gmail.com> 16578Date: Mon Sep 17 00:01:22 2007 +1000 16579 16580 Make use of exaGetPixmapOffset(). 16581 16582commit 4d9e5d451a516aac64c4b95a4478f252c0c9eafb 16583Author: Ben Skeggs <skeggsb@gmail.com> 16584Date: Sun Sep 16 23:54:14 2007 +1000 16585 16586 Fix randr1.2 with driver-provided modes (ie. LVDS native mode) 16587 16588 Fixes #12368 16589 16590commit 5e58e74435da2071aab9f21386ac6dabdf3f8e7d 16591Author: Ben Skeggs <skeggsb@gmail.com> 16592Date: Sun Sep 16 19:44:42 2007 +1000 16593 16594 NV30EXA: Support PICT_a8 dest surfaces 16595 16596 Quick hack for now, shader handling will get a rework one day and clean all 16597 this up. 16598 16599commit d6d3ddfafaece2d153942ec37e3a8c3127c7c3fa 16600Author: Ben Skeggs <skeggsb@gmail.com> 16601Date: Sun Sep 16 19:01:21 2007 +1000 16602 16603 Bring G8x modesetting into sync with xf86-video-nv. 16604 16605commit 53e35a1d42070fd683eb6d0f169ac3ce482a45b0 16606Author: Ben Skeggs <skeggsb@gmail.com> 16607Date: Sun Sep 16 17:27:22 2007 +1000 16608 16609 Move G80 VBIOS download into nv_bios.c. 16610 16611commit 02d5443d650e628256ebed1f331ad32b0f642a38 16612Author: Arthur Huillet <arthur.huillet@free.fr> 16613Date: Fri Sep 14 21:13:26 2007 +0200 16614 16615 Xv: NV04 overlay now has one "colorkey" attribute 16616 16617 It is technically possible to add other attributes, by copypasting vidix code from mplayer, but it requires a bit of work especially on testing, so I am going 16618 to do it only if someone actually needs it. 16619 16620commit 0fa042e5123066ca2d9f36ff05cf64d0f68289a8 16621Author: Arthur Huillet <arthur.huillet@free.fr> 16622Date: Fri Sep 14 14:17:32 2007 +0200 16623 16624 Xv: NV04 overlay now almost functionnal 16625 16626commit a7ff07f4ab63c5a574e549326f022f84c2b9bab4 16627Author: Arthur Huillet <arthur.huillet@free.fr> 16628Date: Fri Sep 14 12:27:45 2007 +0200 16629 16630 Xv: quick imple of NV04 overlay 16631 16632commit 77076a391647f6111f38b10192d78ea7560ca9e8 16633Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16634Date: Wed Sep 12 13:10:37 2007 +0200 16635 16636 Try to fix nv4/tnt1 16637 16638commit e03fabf7e14599c1619673430ec3d81db90ffd1e 16639Author: Maarten Maathuis <madman2003@gmail.com> 16640Date: Tue Sep 11 20:15:11 2007 +0200 16641 16642 Restore PRAMDAC1 and PCRTC1 offsets, in proper format (0x2000/4) and move them to their proper place. 16643 16644commit 564043d9c1c9b5023f1b69a79d80742338256b83 16645Author: Maarten Maathuis <madman2003@gmail.com> 16646Date: Mon Sep 10 21:14:24 2007 +0200 16647 16648 Fix the memory adresses of the secondary crtc and ramdac + minor change. 16649 16650commit 5672842381a0e8f605cd482563c5d9ea02118647 16651Author: Patrice Mandin <pmandin@caramail.com> 16652Date: Mon Sep 10 18:29:10 2007 +0200 16653 16654 make randr 1.2 disableable for xorg server < 1.3 16655 16656commit 2778d3a9041ec62462585f623f84b1ed05b4ebad 16657Author: Patrice Mandin <pmandin@caramail.com> 16658Date: Mon Sep 10 18:28:03 2007 +0200 16659 16660 make randr 1.2 disableable for xorg server < 1.3 16661 16662commit 9bb76f73f0e648fda79a9c84d6e78469752a4eac 16663Author: Maarten Maathuis <madman2003@gmail.com> 16664Date: Mon Sep 10 07:57:01 2007 +0200 16665 16666 randr12: Apply the same pll restrictions to a two stage design. 16667 16668commit b46bbab81cc9f2093406fa8b53f9ffaf6f7fd0fc 16669Author: Maarten Maathuis <madman2003@gmail.com> 16670Date: Mon Sep 10 07:25:25 2007 +0200 16671 16672 randr12: Some more pll timer restrictions from Haiku. 16673 16674commit 77b424415d31428e113688ec27a405966d5080e6 16675Author: Maarten Maathuis <madman2003@gmail.com> 16676Date: Sun Sep 9 23:11:18 2007 +0200 16677 16678 randr12: Some new pll clock restrictions. 16679 16680commit a1bbfe17cc863497644b58a9206ae3876a654543 16681Author: Maarten Maathuis <madman2003@gmail.com> 16682Date: Sun Sep 9 22:50:19 2007 +0200 16683 16684 Some reindenting. 16685 16686commit dedb80d1237ebacf7531938d627328a20453d5d0 16687Author: Maarten Maathuis <madman2003@gmail.com> 16688Date: Sun Sep 9 22:12:38 2007 +0200 16689 16690 randr12: Modify dcb table for nv40 to make it look more like older hardware. Remove i2c bus reverse. 16691 16692commit 7aa96706ef428f1390d8e938d0bf26b8c046a67d 16693Author: Maarten Maathuis <madman2003@gmail.com> 16694Date: Sun Sep 9 20:40:20 2007 +0200 16695 16696 Removed a check for > 1 active crtc's, which seems really odd, no regressions occured for me. 16697 16698 - I cannot plug in both the vga and dvi connector of my monitor, so i cannot verify that. 16699 - The seperate monitor, on crtc0 and crtc1 work as they did (the afp being slightly shifted). 16700 - Also did a little reindenting. 16701 16702commit 228cb177242be053595b959dedfdfbfa77b1c54f 16703Author: Maarten Maathuis <madman2003@gmail.com> 16704Date: Sun Sep 9 18:49:08 2007 +0200 16705 16706 randr12: Fix a nasty regression from 6ba7e83f8cf2da84dd796ee59cb24c15bc36b746. 16707 16708 - For non-flatpanels the dac was never initialized. 16709 16710commit 0a56cb53d7e2a58ef044c9090f48fd46a2b251d9 16711Author: Maarten Maathuis <madman2003@gmail.com> 16712Date: Sun Sep 9 00:39:41 2007 +0200 16713 16714 randr12: Some missing init. 16715 16716commit 80bfb59a1f1c3968f8e116de611975df4acf5618 16717Author: Maarten Maathuis <madman2003@gmail.com> 16718Date: Sat Sep 8 23:58:39 2007 +0200 16719 16720 Some much needed indenting and readability improvements. 16721 16722commit 20aaacdd69c427f6c75b80f5f206b3187262f3a1 16723Author: Maarten Maathuis <madman2003@gmail.com> 16724Date: Sat Sep 8 21:00:45 2007 +0200 16725 16726 Hopefully fix bridged chip detection, such as the 7800GS AGP. 16727 16728commit b99549b5cb4f4cba2d60292994b2db483f8d4a8c 16729Author: Maarten Maathuis <madman2003@gmail.com> 16730Date: Sat Sep 8 14:25:39 2007 +0200 16731 16732 Remove list that is incorrect and no longer needed. 16733 16734commit b104314a11d3cf3bbb23095ff5805c1d5d8d9775 16735Author: Maarten Maathuis <madman2003@gmail.com> 16736Date: Sat Sep 8 14:24:54 2007 +0200 16737 16738 Fix previous commit, change the way Digital Flat Panels are programmed. 16739 16740 - This is mostly taken from the Haiku driver. 16741 - It works for me, please report any issues you have. 16742 - Added scaling, for running 4:3 resolutions on widescreen for example (untested). 16743 16744commit a867d6912456a5e15a965050bf784687e60634ee 16745Author: Maarten Maathuis <madman2003@gmail.com> 16746Date: Sat Sep 8 00:46:20 2007 +0200 16747 16748 Hopefully this will improve flatpanel beheaviour. 16749 16750 - I looked at Haiku and copied as much as i could. 16751 - I removed the htiming and vtiming hack. 16752 - Let's hope this works. 16753 16754commit 6ba7e83f8cf2da84dd796ee59cb24c15bc36b746 16755Author: Maarten Maathuis <madman2003@gmail.com> 16756Date: Fri Sep 7 23:11:51 2007 +0200 16757 16758 Some more indenting and such. 16759 16760commit d8d835dfb61f437b3eaede0a6ec188835f6eb5ff 16761Author: Maarten Maathuis <madman2003@gmail.com> 16762Date: Fri Sep 7 22:58:56 2007 +0200 16763 16764 Removing a hack in the hopes of finding a better way and some minor changes. 16765 16766commit dd7e568bd101ed34dcdd7d3b6d8169740499d593 16767Author: Maarten Maathuis <madman2003@gmail.com> 16768Date: Fri Sep 7 22:27:18 2007 +0200 16769 16770 Indenting and minor code style changes. 16771 16772commit 8ed256e94dae884b723c172320b15b0d6ae00a1b 16773Author: Maarten Maathuis <madman2003@gmail.com> 16774Date: Fri Sep 7 14:37:40 2007 +0200 16775 16776 randr12: Fix Xv blitter, please see full changelog for other implications. 16777 16778 - Removed the CRTCNumber variable, instead just store if crtc0 and crtc1 are active. 16779 - This coule break old style setups (picking the wrong monitor for example). 16780 - CRTCNumber option temporarily removed, because the piece of code surrounding it is ugly and needs a rework. 16781 16782commit eb595b2e06ec33d5fe2af43bc7ad107f551d99c8 16783Author: Maarten Maathuis <madman2003@gmail.com> 16784Date: Thu Sep 6 23:26:04 2007 +0200 16785 16786 randr-1.2: Only the NV4x have these busses swapped, revert for other architectures. 16787 16788 - If this is also true for NV3x, then please contact me. 16789 16790commit dd146a074fff575e18a41fd386e2a2ce5a1ec847 16791Author: Maarten Maathuis <madman2003@gmail.com> 16792Date: Thu Sep 6 22:55:16 2007 +0200 16793 16794 Update to latest pciaccess api, this now requires libpciaccess 0.10.0. 16795 16796commit 2f37d249605d6f0b80c290683ca35a9431b1078c 16797Author: Maarten Maathuis <madman2003@gmail.com> 16798Date: Thu Sep 6 22:00:30 2007 +0200 16799 16800 NV03 is dead. 16801 16802commit 2b2c99f2581e14d3f1726bc503951b1fe7760c95 16803Author: Maarten Maathuis <madman2003@gmail.com> 16804Date: Thu Sep 6 21:43:55 2007 +0200 16805 16806 randr12: A prettier hack for DFP, that uses several known htiming and vtiming modes. 16807 16808 - I'm fairly sure these modes are tied to resolution. 16809 - You are welcome to supply additional modes. 16810 - A final solution would be even better ;-) 16811 16812commit ab7d37d61a01afea218321d336664896eaec5651 16813Author: Arthur Huillet <arthur.huillet@free.fr> 16814Date: Thu Sep 6 15:09:59 2007 +0200 16815 16816 Xv: work around some overlay bandwidth problems - may break things though 16817 16818 extended nv workaround from bug 4686 in nv_dac.c to NV1x and NV2x cards, as it fixed problems on someone's NV17. 16819 16820commit f46bc9ad5558e9ecf6b174e24658e812811e94ce 16821Author: Maarten Maathuis <madman2003@gmail.com> 16822Date: Wed Sep 5 23:15:57 2007 +0200 16823 16824 Workaround for flatpanels. 16825 16826 - Remove the hardcoded values. 16827 - This should probably work for monitors whose mode is already set. 16828 16829commit 1825031465fc614a2a972ea7a8b44ac32aed8ea3 16830Author: Maarten Maathuis <madman2003@gmail.com> 16831Date: Wed Sep 5 21:23:25 2007 +0200 16832 16833 Fix minor issue. 16834 16835commit c0eefaba3f3706eb62e172bc57ac2795bd16bd56 16836Author: Maarten Maathuis <madman2003@gmail.com> 16837Date: Wed Sep 5 21:22:05 2007 +0200 16838 16839 Fix DPMS, patch by Bernhard Kaindl <bk@suse.de>. 16840 16841commit d4dda6bb972003394dd3326f8042fa5b759fc063 16842Author: Maarten Maathuis <madman2003@gmail.com> 16843Date: Wed Sep 5 21:18:51 2007 +0200 16844 16845 Reverse DDC1 and DDC2. 16846 16847 - I need them reversed. 16848 - I also found this: http://www.mail-archive.com/git-commits-head@vger.kernel.org/msg12009.html 16849 - If you need them otherwise, then please consult and hopefully we can solve it. 16850 16851commit da89d5a4e0263b8a89c549496d10ab43c7f1f230 16852Author: Maarten Maathuis <madman2003@gmail.com> 16853Date: Wed Sep 5 21:16:11 2007 +0200 16854 16855 randr-1.2 clients have their DDC's initialized elsewhere. 16856 16857commit 2b209d9b8907fad3bbaaf7297ff3c340f086bd24 16858Author: Maarten Maathuis <madman2003@gmail.com> 16859Date: Wed Sep 5 21:12:18 2007 +0200 16860 16861 LockUnlock CRTC's when going back to X (EnterVT), patch by Bernhard Kaindl <bk@suse.de>. 16862 16863commit ba7aa5a1525bb3fbc2ae7fbfce3e42a2e945626e 16864Author: Maarten Maathuis <madman2003@gmail.com> 16865Date: Wed Sep 5 18:34:27 2007 +0200 16866 16867 Xv: improved behavior with NoAccel by disabling blitter and overlay 16868 16869commit 42fd07b25d9abece034e972d1413b31fe78eba4c 16870Author: Ben Skeggs <skeggsb@gmail.com> 16871Date: Wed Sep 5 11:58:19 2007 +1000 16872 16873 Avoid segv. 16874 16875commit e74d58cdef0a0ee561e54f3d7d8afdfac030e04a 16876Author: Ben Skeggs <skeggsb@gmail.com> 16877Date: Wed Sep 5 11:53:48 2007 +1000 16878 16879 Add back DCB table parsing from randr-1.2 branch. 16880 16881commit 8a12758a6b7e4313e0269585b6f64eaffe54c74c 16882Author: Ben Skeggs <skeggsb@gmail.com> 16883Date: Wed Sep 5 11:51:35 2007 +1000 16884 16885 Add LVDS panels when found in DCB. 16886 16887commit b01fbbfa2b6be793dbd7695624699ed703ff4424 16888Author: Ben Skeggs <skeggsb@gmail.com> 16889Date: Wed Sep 5 11:34:14 2007 +1000 16890 16891 Nuke "comparison is always false..." warning. 16892 16893commit c53f495ec2994f013406d9738764ab124ed32061 16894Author: Ben Skeggs <skeggsb@gmail.com> 16895Date: Wed Sep 5 10:59:33 2007 +1000 16896 16897 Fix pciid detect on non-8800 cards (they identify as NV8x, not NV5x) 16898 16899 Actually, use 0x88000 to get pciid on all chipsets since NV40. Evidence 16900 suggests that NV4x has both the 0x88xxx regs and 0x18xx regs. 16901 16902commit 17d1a65296ae4fc524a31cd2fe4b9fe9484c0452 16903Author: Ben Skeggs <skeggsb@gmail.com> 16904Date: Wed Sep 5 10:53:32 2007 +1000 16905 16906 Neither overlay or blit adaptor work on G8x. 16907 16908commit 9ce1115cf26a7cf3212fdf446b83907f5795913a 16909Author: Ben Skeggs <skeggsb@gmail.com> 16910Date: Wed Sep 5 10:51:02 2007 +1000 16911 16912 Nuke riva_* files + bits of NV03 that got reincarnated. 16913 16914commit 51e3855104ff7b68559d37e774cde306424399d8 16915Author: Ben Skeggs <skeggsb@gmail.com> 16916Date: Wed Sep 5 10:46:54 2007 +1000 16917 16918 Nuke XAA. 16919 16920commit a0cb03d32b0da3ad2b4e03515926a3d616f487bc 16921Author: Maarten Maathuis <madman2003@gmail.com> 16922Date: Tue Sep 4 23:04:59 2007 +0200 16923 16924 There was some redundant arch checking code left, must have been missed after the randr merge. 16925 16926commit 13aeb83d7e5fcb3f051f95faa48a2ef636d67260 16927Author: Maarten Maathuis <madman2003@gmail.com> 16928Date: Tue Sep 4 22:54:28 2007 +0200 16929 16930 I unknowingly broke CRT's i think. 16931 16932commit a57fe870e49b2e7a3a6caf17fea8c8c878b1024c 16933Author: Maarten Maathuis <madman2003@gmail.com> 16934Date: Tue Sep 4 22:17:52 2007 +0200 16935 16936 Fix several problems. 16937 16938 - Implement a temporary NVSetMode(), suggestion and code by: Bernhard Kaindl <bk@suse.de>. 16939 Note: I've left his comment intact, just a minor change. 16940 - Fix last commit, i forgot to store the last mode. 16941 - Move some header stuff around, which i created without knowing the true purpose of nv_proto.h ;-) 16942 16943commit be468069b752a3a3c595f62eda40e90719496c90 16944Author: Maarten Maathuis <madman2003@gmail.com> 16945Date: Tue Sep 4 20:45:25 2007 +0200 16946 16947 Fix DVI/Digital alligment with some (magic?) values discovered by Matthew Garrett <mjg59@srcf.ucam.org>. 16948 16949commit 9b54aac1641c35ca3edb2d208d8a9a9c55737894 16950Author: Maarten Maathuis <madman2003@gmail.com> 16951Date: Tue Sep 4 20:20:02 2007 +0200 16952 16953 Add suggestion/patch by Bernhard Kaindl <bk@suse.de> and fix NVRestore(). 16954 16955 - When a monitor comes out of DPMS sleep the old state is restored. 16956 - NVRestore() was not ported from randr-1.2 branch, meaning that going from X to console resulted in a unacceptable mode. 16957 16958commit 186e3b551da6e37e0860d06e9265c3c58b763e27 16959Author: Maarten Maathuis <madman2003@gmail.com> 16960Date: Tue Sep 4 18:39:44 2007 +0200 16961 16962 Add pci-rework compatibility and remove the pci-id list. 16963 16964 - Detection is now done by probing registers. 16965 - Nforce/Nforce2 have special "hacks", they have been converted, but are untested. 16966 - NV5x detection should work, but has seen no testing. 16967 16968commit 6f860d6d7637f4eba49636d2bab11ac62da279cf 16969Author: Ben Skeggs <skeggsb@gmail.com> 16970Date: Tue Sep 4 11:38:39 2007 +1000 16971 16972 Badness. 16973 16974commit 54ca35e7379fdb375c0d7be2f8c3ad1513be2a59 16975Author: Ben Skeggs <skeggsb@gmail.com> 16976Date: Mon Sep 3 22:06:27 2007 +1000 16977 16978 Hook up G8x accel code. 16979 16980commit 8c6fd43f63e450acbfd8de563580db715e72f90b 16981Author: Ben Skeggs <skeggsb@gmail.com> 16982Date: Mon Sep 3 22:06:09 2007 +1000 16983 16984 Fix G8x hwcursor 16985 16986commit f6d0e2dde750436090f50f442e3e0f1aa226a7f3 16987Author: Ben Skeggs <skeggsb@gmail.com> 16988Date: Mon Sep 3 21:35:06 2007 +1000 16989 16990 Fix up M2MF code for G8x. 16991 16992commit fe8788da8cd5aca90bbe85e1c975b79ff02682ba 16993Author: Ben Skeggs <skeggsb@gmail.com> 16994Date: Mon Sep 3 21:30:16 2007 +1000 16995 16996 Init grobjs we need on G8x. 16997 16998commit f450a4d0ab9c199db0a99a5e78642a0d66ca2915 16999Author: Ben Skeggs <skeggsb@gmail.com> 17000Date: Mon Sep 3 21:14:59 2007 +1000 17001 17002 Allocate some VRAM for G8x palette. 17003 17004commit dfcfc9a8dac7f1fdf7b31ebdd2bca4c5f33313e0 17005Author: Ben Skeggs <skeggsb@gmail.com> 17006Date: Mon Sep 3 21:06:15 2007 +1000 17007 17008 Don't pound PMC_ENABLE on startup, this is DRM's job. 17009 17010 Fixes resetting parts of G8x that shouldn't be. 17011 17012commit c8a32bab34ffe74a4a323de3885fae87bd916218 17013Author: Ben Skeggs <skeggsb@gmail.com> 17014Date: Mon Sep 3 21:05:08 2007 +1000 17015 17016 Detect unknown G8x chips in NVProbe() + Force randr12 on G8x 17017 17018commit 087fd31d6a5d6a835fc318549a12faa916cf5ed0 17019Author: Dave Airlie <airlied@linux.ie> 17020Date: Mon Sep 3 16:46:34 2007 +1000 17021 17022 fix distcheck after randr-1.2 brought in 17023 17024commit 423f95c02c5b018d6fc78f501ea8b546d73f33fb 17025Author: Dave Airlie <airlied@linux.ie> 17026Date: Mon Sep 3 16:42:32 2007 +1000 17027 17028 merge in the nv50 code from the randr 1.2 branch 17029 17030commit 39e8b6c7c0a159f7a06d786abc0f9c5c30ae6793 17031Author: Dave Airlie <airlied@linux.ie> 17032Date: Mon Sep 3 16:22:47 2007 +1000 17033 17034 randr12: bring over cursor code 17035 17036commit ffd87b9bf2712311e197063a7c4b65b7c06a5ebb 17037Author: Dave Airlie <airlied@linux.ie> 17038Date: Mon Sep 3 16:16:43 2007 +1000 17039 17040 randr12: try and bring over most of the randr12 code into master 17041 17042 This currently shouldn't break master, and randr-1.2 more than likely doesn't actually work 17043 17044commit 708cb3f9d9b66030dd0bfc4beaa657fcdfded784 17045Author: Arthur Huillet <arthur.huillet@free.fr> 17046Date: Fri Aug 31 21:42:51 2007 +0200 17047 17048 Xv: now "powercycling" the overlay using PMC_ENABLE in order to workaround some lock up problems 17049 17050commit 3a5095db9fd33d2feeb4bcfcb8453848998ce04b 17051Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17052Date: Fri Aug 31 01:36:11 2007 +0200 17053 17054 Cut remains of NV03 support. 17055 17056commit 54a922dffaa1a4a7c29c2cf8d73449e6fe9058ed 17057Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17058Date: Fri Aug 31 01:35:52 2007 +0200 17059 17060 Good ol' nv04 doesn't seem to support dithering type for SIFM. 17061 17062commit b680a9652508e0d0e660fd924937870cd4ad3530 17063Author: Arthur Huillet <arthur.huillet@free.fr> 17064Date: Thu Aug 30 23:19:13 2007 +0200 17065 17066 Xv: nv NV30 arbitration patch for DFPs - fixes (some?) bw problems 17067 17068commit f29c9d3fc95f160ded64b911266a479ad0a2fc1a 17069Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17070Date: Wed Aug 29 00:56:05 2007 +0200 17071 17072 Silence more warnings. 17073 17074commit 663bf3cbc881631e54e7d63932f59e1b4289979f 17075Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17076Date: Wed Aug 29 00:20:19 2007 +0200 17077 17078 Silence a warning. 17079 17080commit 5937ace8e2376f0b38249e38d1f97a88f4aca99f 17081Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17082Date: Tue Aug 28 11:18:54 2007 +0200 17083 17084 Some small fixes. 17085 17086commit 30d2b3f0dd4a9b6be5d3ad7a38993e5b84baf90a 17087Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17088Date: Thu Aug 23 22:54:53 2007 +0200 17089 17090 Oops I forgot to enable the nv35 object on nv35... 17091 17092commit 761d25eb9afd4be34f00ff7eff44dac0c2f9d9a2 17093Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17094Date: Thu Aug 23 22:39:34 2007 +0200 17095 17096 Don't enable that non-working nv30 EXA just yet. 17097 17098commit 7e7dec68969db5dfc15a96f6dbe78edbcd1ab5ab 17099Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17100Date: Thu Aug 23 22:33:58 2007 +0200 17101 17102 Add nv30 exa code, doesn't work yet. 17103 17104commit e1789ac879d474015c87c36847f0e2777486d6ad 17105Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17106Date: Wed Aug 22 22:38:08 2007 +0200 17107 17108 Update to the latest nouveau_reg.h file. 17109 17110commit 4fc38aeb54da1cba9d5e231a5151e6429af0ea7e 17111Author: Matthieu Castet <castet.matthieu@free.fr> 17112Date: Tue Aug 21 20:03:15 2007 +0200 17113 17114 init SCALED_IMAGE_FROM_MEMORY_COLOR_CONVERSION to COLOR_CONVERSION_TYPE_DITHER. 17115 riva and blob init it. 17116 17117commit c63b5a3740dfe3ff5c6b77412e3666a79137c120 17118Author: Arthur Huillet <arthur.huillet@free.fr> 17119Date: Tue Aug 21 14:55:07 2007 +0200 17120 17121 Fixed blending in NVComposite by actually creating ContextBeta1 and 4 objects 17122 17123commit 4454b3d0104c14890b9b5ead33823ffd0362315c 17124Author: Arthur Huillet <arthur.huillet@free.fr> 17125Date: Mon Aug 20 23:50:43 2007 +0200 17126 17127 Workaround for NVComposite bug #11949 17128 17129 Accepting only PictOpSrc now, PictOpOver causing crashes. 17130 17131commit 450cf465bd850d70ba00711e7d85e1e8eaeff502 17132Author: Arthur Huillet <arthur.huillet@free.fr> 17133Date: Mon Aug 20 13:18:15 2007 +0200 17134 17135 Xv: fixed wrong fix of big endian NV12 copy, as explained by Daniel Chassot 17136 17137 No sleepless coding ever again. <- big lie 17138 17139commit b8780f3d9a7e1bd596f19cbb4867908dd6be53db 17140Author: Arthur Huillet <arthur.huillet@free.fr> 17141Date: Mon Aug 20 00:46:00 2007 +0200 17142 17143 Xv: patch from Daniel Chassot to do 32bit memory writes in NVCopyNV12ColorPlanes, with minor reworking of mine 17144 17145commit 690b4b7adc02e7170451bd5bc6616bc03823d2e5 17146Author: Arthur Huillet <arthur.huillet@free.fr> 17147Date: Sun Aug 19 21:04:00 2007 +0200 17148 17149 Xv: this time a correct matching ? :) 17150 17151commit 6b6505bc54fc29fc22d2be995f34b24ddb9dbaa0 17152Author: Arthur Huillet <arthur.huillet@free.fr> 17153Date: Sun Aug 19 20:18:49 2007 +0200 17154 17155 Xv: stupid build error. blah. 17156 17157commit 28b66c26f85f7730ba9a54db3f778bfd5b24915d 17158Author: Arthur Huillet <arthur.huillet@free.fr> 17159Date: Sun Aug 19 20:14:38 2007 +0200 17160 17161 Xv: now catching all chips < NV17 17162 17163commit 94ae23e2a1ebe9fe3f4bf2039ad3b2e5f0e3e55e 17164Author: Arthur Huillet <arthur.huillet@free.fr> 17165Date: Sun Aug 19 19:50:57 2007 +0200 17166 17167 Xv: disabled YV12 overlay on < NV17 17168 17169commit 1f83dbfa6ef774276e0da3dc4f74cbabe55f76fa 17170Author: Arthur Huillet <arthur.huillet@free.fr> 17171Date: Fri Aug 17 00:45:47 2007 +0200 17172 17173 Xv: now only uploading the part of source image that is used to workaround bandwidth problems 17174 17175commit ffa62dc9f573448c8a832324973513f7ba985b33 17176Author: Arthur Huillet <arthur.huillet@free.fr> 17177Date: Wed Aug 15 02:27:15 2007 +0200 17178 17179 Cleaner DMA and CPU copy upload of color plane for overlay. Fixes a few bugs and is necessary for the bandwidth problem workaround. 17180 17181commit 72c62c48faf2acffae1bc311051537f46972ae58 17182Author: Patrice Mandin <pmandin@caramail.com> 17183Date: Tue Aug 14 22:26:02 2007 +0200 17184 17185 Can not setup dfp modes with xorg-server-1.1 also 17186 17187commit 54eadd624282ae6d0493a3cb54afda7600c7e12b 17188Author: Patrice Mandin <pmandin@caramail.com> 17189Date: Tue Aug 14 22:12:37 2007 +0200 17190 17191 Make NV30 EXA disablable 17192 17193commit da6522dbfcff10081ce2ad866ba4afdd18b413d7 17194Author: Arthur Huillet <arthur.huillet@free.fr> 17195Date: Tue Aug 14 15:01:58 2007 +0200 17196 17197 Implemented CPU copy fallback for native YV12 overlay as well. 17198 17199commit e417307bfc8610f010728d82bb80fced7d0d7aa7 17200Author: Arthur Huillet <arthur.huillet@free.fr> 17201Date: Tue Aug 14 14:47:44 2007 +0200 17202 17203 Native YV12 overlay support, cleaner Xv code 17204 17205 Added support for a native YV12 overlay for NV10+ cards. 17206 Cleaned up Xv code. 17207 CPU copy fallback does not do native YV12 yet. 17208 Bandwidth problem with YV12. 17209 17210commit b3e4da61e8c846d8d3f01b015cb7155da4e266e9 17211Author: Carlos Martín Nieto <carlos@cmartin.tk> 17212Date: Sat Aug 11 12:27:56 2007 +0200 17213 17214 Don't dereference the uninitialised pNv->FB pointer. 17215 17216 Opps, mea culpa. I accidentally commited a debug statement that 17217 dereferences an uninitialised pointer. 17218 17219commit a7d29fd932fe14ca5b82cda64341704164fb7e09 17220Author: Carlos Martín Nieto <carlos@cmartin.tk> 17221Date: Fri Aug 10 20:01:27 2007 +0200 17222 17223 Move DRI's FB to the higher half of VRAM 17224 17225 If we don't, DRM's push buffer takes the map away from us. 17226 17227 This is just a workaround until that useless DRI mapping goes away. 17228 17229commit f9e7849e8736d1477e867a3a1262f0f82baefbcb 17230Author: Arthur Huillet <arthur.huillet@free.fr> 17231Date: Fri Aug 10 20:59:12 2007 +0200 17232 17233 of course, we don't need Xv debug messages upstream 17234 17235commit ee098308c6aca7615bc6ee597e9ca531520f4119 17236Author: Arthur Huillet <arthur.huillet@free.fr> 17237Date: Fri Aug 10 19:28:22 2007 +0200 17238 17239 "smart" overlay with composite, and a tentative fix of flickering when dropping frames 17240 17241 added "smart overlay" when composite is enabled. not extensively tested 17242 now not changing the buffer when skipping frames - not sure it even has an impact. 17243 17244commit 9cc11cbc9ce10b99f72d7c08584bf61d179cdbda 17245Author: Ben Skeggs <skeggsb@gmail.com> 17246Date: Thu Aug 9 01:46:46 2007 +1000 17247 17248 Fix blit adaptor + NVWaitVBlank() (ie. replace old magic 0x400710 hack) 17249 17250commit 22e965038bbc64c7b5118fc9a03ecf5f467d1665 17251Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17252Date: Wed Aug 8 12:45:59 2007 +0200 17253 17254 And of course, we don't really need these ErrorFs. Not any more than last time. 17255 17256commit 1c5a3e271ca628c61fce0bcef24d0bda15d06d7e 17257Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17258Date: Wed Aug 8 12:38:14 2007 +0200 17259 17260 Oh, and of course we need a new object. 17261 17262commit a9b44880924ee814f0bd4b0015ab152f2d0b9051 17263Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17264Date: Wed Aug 8 12:37:31 2007 +0200 17265 17266 Use IMAGE_FROM_CPU for smaller UTS. This is async and helps a lot with glyphs especially. 17267 17268commit 9cb4c95a4fbf38fcb8249e765ff71b2e24912244 17269Author: Ben Skeggs <skeggsb@gmail.com> 17270Date: Wed Aug 8 11:48:48 2007 +1000 17271 17272 Add forgotten prototype 17273 17274commit 089ff874e0a798c1b2693b5ab01cfa04c939f758 17275Author: Ben Skeggs <skeggsb@gmail.com> 17276Date: Wed Aug 8 10:32:08 2007 +1000 17277 17278 Timeouts got lost somewhere along the way.. 17279 17280commit 15a680a6d715f87910b07133e1b98e2d4be9a108 17281Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17282Date: Tue Aug 7 23:42:09 2007 +0200 17283 17284 Fix some overlooked hardcoded subchannels... 17285 17286commit 3f0a52c44697787d84d4f2cf61872e13d630f62a 17287Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17288Date: Mon Aug 6 23:30:04 2007 +0200 17289 17290 Fix explicit usage of subchannels, and remove subchannel defines altogether. 17291 17292commit 17b65f1c2be3e609d387066d72dd8f5f7e2a6d92 17293Author: Matthieu Castet <castet.matthieu@free.fr> 17294Date: Mon Aug 6 19:55:52 2007 +0200 17295 17296 make XAA work on nv17 17297 17298commit bf5684ba05e50fe48cce1e634caf03725fa907e6 17299Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17300Date: Mon Aug 6 18:27:19 2007 +0200 17301 17302 oops remove the debugging ErrorFs. 17303 17304commit 2eab8453d73b4e7fd6c2d772d595ca5e8fa8c975 17305Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17306Date: Mon Aug 6 18:13:13 2007 +0200 17307 17308 Fix UTS/DFS when gart is smaller than transfer size. 17309 17310commit 71076f7947ec71b7a0f363c23f5a51ef09cdd9c0 17311Author: Ben Skeggs <skeggsb@gmail.com> 17312Date: Tue Aug 7 02:09:30 2007 +1000 17313 17314 NVDmaSetObjectFromSubchannel doesn't exist anymore! 17315 17316commit 4a6c2bd2bedcf9f67c3e498b1411ad58cd66199c 17317Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17318Date: Mon Aug 6 17:29:41 2007 +0200 17319 17320 Add subchannel handling code 17321 17322commit c7b16f6fff96ffcff8049feed2a0a9ef6de4209f 17323Author: Ben Skeggs <skeggsb@gmail.com> 17324Date: Mon Aug 6 21:45:50 2007 +1000 17325 17326 Adapt to drm 0.0.10 17327 17328commit 59d073c9b01a8f61675b8d74f5c55f134ddfb8bf 17329Author: Arthur Huillet <arthur.huillet@free.fr> 17330Date: Mon Aug 6 01:37:17 2007 +0200 17331 17332 Xv: waiting for last transfers to finish before freeing notifiers 17333 17334commit 047aa7e0a6ecce59c9be8d36c51f082c0ddaafe8 17335Author: Ben Skeggs <skeggsb@gmail.com> 17336Date: Mon Aug 6 00:13:35 2007 +1000 17337 17338 Clamp NV_MEMORY_TO_MEMORY_FORMAT_LINE_COUNT to 2047 lines. 17339 17340 At least NV40 can't handle values larger than this. Patch fixes hang 17341 when visiting opengl.org. 17342 17343commit 68070ff7309c8b1cf149490dcf537892b0bf44bc 17344Author: Arthur Huillet <arthur.huillet@free.fr> 17345Date: Mon Jul 30 03:27:55 2007 +0200 17346 17347 Xv now manages a pool of 6 notifiers for its double buffering. Also improved the memory allocation logic. 17348 17349commit bceef7210e44dc643a5f7a5b5683d5c434a07565 17350Author: Arthur Huillet <arthur.huillet@free.fr> 17351Date: Sun Jul 29 01:54:42 2007 +0200 17352 17353 Xv: fixed bugs when displayed only part of the source image (tvtime with overscan) 17354 Xv: fixed bug with overlay behaving badly when window is partly out of screen 17355 Xv: clean up NVPutImage code 17356 17357commit a434cc6dd38ffa658bacf5547ec0384ec5351e9c 17358Author: Dave Airlie <airlied@ppcg5.localdomain> 17359Date: Sat Jul 28 16:34:49 2007 +1000 17360 17361 exa: disable compositing on big endian machines 17362 17363 This at least allows me to see what I'm typing for now :-) 17364 17365commit 175c2b7f7b9e45924e1de7f7ea1130fda76d02fd 17366Author: Arthur Huillet <arthur.huillet@free.fr> 17367Date: Fri Jul 27 19:50:48 2007 +0200 17368 17369 no overlay with composite 17370 17371commit 200c82634971b32bcc308b5e07c113048cf37aa4 17372Author: Arthur Huillet <arthur.huillet@free.fr> 17373Date: Fri Jul 27 16:28:48 2007 +0200 17374 17375 Xv: gotos are evil 17376 17377commit 8544c0ce64780be1bb5f66550d9d6f1e3c87865c 17378Author: Arthur Huillet <arthur.huillet@free.fr> 17379Date: Fri Jul 27 15:11:31 2007 +0200 17380 17381 Removed comments in preparation for upstream push 17382 17383commit 2c60f6e9fc5260eba5b7ddaf67085781f35bfc58 17384Author: Arthur Huillet <arthur.huillet@free.fr> 17385Date: Thu Jul 26 23:00:02 2007 +0200 17386 17387 re-added CPU copy code as a second fallback option (first being to use X GARTScratch) 17388 17389commit c4dfad770d380dacddd3be025e96f867047327cc 17390Author: Arthur Huillet <arthur.huillet@free.fr> 17391Date: Thu Jul 26 16:08:40 2007 +0200 17392 17393 Completely removed notifier wait, when using double buffered stuff. 17394 I may have to re-add a wait in the future, but I'm not sure yet how to use two notifiers at a time. 17395 17396commit 8ba6a9c7494ebd53a3fe775d4d9b402e918a6598 17397Author: Arthur Huillet <arthur.huillet@free.fr> 17398Date: Thu Jul 26 15:21:08 2007 +0200 17399 17400 Preliminary implementation of "host-side double buffering" to improve performance of Xv 17401 17402commit 4f4721f1b8a76a0bf9938e3c614ba7166d5f3dab 17403Author: Arthur Huillet <arthur.huillet@free.fr> 17404Date: Wed Jul 25 22:23:37 2007 +0200 17405 17406 removed pointless sleep in notifier wait 17407 17408commit 290e71edfbc6297c63989d5c21a67717b4ca5f76 17409Author: Arthur Huillet <arthur.huillet@free.fr> 17410Date: Fri Jul 20 17:46:18 2007 +0200 17411 17412 cleaner Xv code - fallback on CPU copy yet to be implemented 17413 17414commit e760235a0596895b726220366c50d4e7c244006e 17415Author: Arthur Huillet <arthur.huillet@free.fr> 17416Date: Fri Jul 20 01:33:49 2007 +0200 17417 17418 Now using GART as the source for the blitter - preliminary implementation, gotta clean it up 17419 17420commit a97181f95a1f5da6180b8a72f3ca307079b92079 17421Author: Arthur Huillet <arthur.huillet@free.fr> 17422Date: Fri Jul 20 01:22:11 2007 +0200 17423 17424 now yielding in the wait for notifier for Xv only 17425 17426commit 81cc744eb31343a509f9ca0d37993d462f162d80 17427Author: Arthur Huillet <arthur.huillet@free.fr> 17428Date: Sun Jul 15 01:00:19 2007 +0200 17429 17430 VERY EXPERIMENTAL now EXA waits for sync before its operations - this removed the text corruption and didn't affect performance here, please report. 17431 17432commit 74ed14e1d27903f0647fe650da666cacb1e2e940 17433Author: Arthur Huillet <arthur.huillet@free.fr> 17434Date: Sun Jul 15 00:35:46 2007 +0200 17435 17436 VERY EXPERIMENTAL moved the notifier wait at the beginning of NvPutImage 17437 17438commit 590e385cb5e4f26532ec5ee380f2aaae1cd69894 17439Author: Arthur Huillet <arthur.huillet@free.fr> 17440Date: Thu Jul 12 03:19:21 2007 +0200 17441 17442 new structure for PutImage, and DMA transfers 17443 17444commit d4abe1fb011d414fda1871f052dda5b860f0820b 17445Author: Ben Skeggs <skeggsb@gmail.com> 17446Date: Tue Jul 17 13:09:34 2007 +1000 17447 17448 Emit NOTIFY+NOP sequence after M2MF. 17449 17450commit 36396aeaeabfd00c7014fa52c17261c3da8b6a82 17451Author: Ben Skeggs <skeggsb@gmail.com> 17452Date: Fri Jul 13 15:28:56 2007 +1000 17453 17454 Play "nuke the typedef" 17455 17456commit 99e0fbc1a34f535abf41c6f4566901c326dfa4b5 17457Author: Ben Skeggs <skeggsb@gmail.com> 17458Date: Thu Jul 12 10:47:53 2007 +1000 17459 17460 Match drm 0.0.9 interface changes. 17461 17462commit e88b20b392c6972088d88f52b2da83b82b821be5 17463Author: Arthur Huillet <arthur.huillet@free.fr> 17464Date: Wed Jul 11 02:42:24 2007 +0200 17465 17466 added support for PCI GART when no AGP is available - requires DRM interface version 8 17467 17468commit dff1cce299df636813e8f85109f4556f520620c0 17469Author: Ben Skeggs <skeggsb@gmail.com> 17470Date: Sun Jul 1 18:22:37 2007 +1000 17471 17472 Always provide UTS, fallback to memcpy if no GART. 17473 17474commit 2e0e7e3856573fbbcb5fe5c96cadcde4d33db68f 17475Author: Ben Skeggs <skeggsb@gmail.com> 17476Date: Sun Jul 1 18:19:44 2007 +1000 17477 17478 Always provide DFS, use memcpy fallback if no GART. 17479 17480commit ba0d3fbba7c1a83ddc509414772199862a4a1d03 17481Author: Ben Skeggs <skeggsb@gmail.com> 17482Date: Sat Jun 30 15:27:07 2007 +1000 17483 17484 Remove some unused cruft 17485 17486commit 3e2677040937fda1cd6e6ea26c238049bd9f99ba 17487Author: Ben Skeggs <skeggsb@gmail.com> 17488Date: Sat Jun 30 04:26:48 2007 +1000 17489 17490 No need to call exaMarkSync() in UTS/DFS. 17491 17492commit e81c9a29e6d1c6573892df8601f0361c2b09c659 17493Author: Ben Skeggs <skeggsb@gmail.com> 17494Date: Thu Jun 28 15:16:35 2007 +1000 17495 17496 Bail if GlxSetVisualConfigs isn't found. 17497 17498commit fce3edda42d10a9c3249c6eae4218a51a981493f 17499Author: Ben Skeggs <skeggsb@gmail.com> 17500Date: Mon Jun 25 12:45:31 2007 +1000 17501 17502 Bump required xorg-server version to 1.3 17503 17504commit 8f1c9af02651404f8b4800b150a86bb47062fd68 17505Author: Ben Skeggs <skeggsb@gmail.com> 17506Date: Mon Jun 25 12:44:12 2007 +1000 17507 17508 Check PRAMIN for BIOS image if not found in PROM. 17509 17510commit 824331e3fce36098e8dec31746204f7c13038abb 17511Author: Ben Skeggs <skeggsb@gmail.com> 17512Date: Sun Jun 24 17:48:33 2007 +1000 17513 17514 Implement NV40 EXA composite hooks with 3D engine. 17515 17516 Named NV30EXA, but won't work on NV30 just yet. This should change at some 17517 point in the future. 17518 17519 Will likely break non-voodoo'd NV4X cards (if the latest drm changes didn't 17520 already break them). Use EXANoComposite xorg.conf option to disable the hook 17521 if that's the case. 17522 17523commit 2e2e95fc3b650ca633294bf4c11c9ba28f45b743 17524Author: Ben Skeggs <skeggsb@gmail.com> 17525Date: Sun Jun 24 17:34:10 2007 +1000 17526 17527 Slightly saner UploadToScreen implementation 17528 17529commit 35806975047aa8c0dbc059709bb823f475ee7368 17530Author: Ben Skeggs <skeggsb@gmail.com> 17531Date: Sun Jun 24 17:27:42 2007 +1000 17532 17533 Wait on notifier instead of PGRAPH_STATUS 17534 17535 When multiple channels are in use, PGRAPH_STATUS may report busy even if the 17536 ddx's channel is idle. 17537 17538commit c39089b3d686eecef11db2b5a2d606ba1ca2dc0c 17539Author: Ben Skeggs <skeggsb@gmail.com> 17540Date: Sun Jun 24 17:21:27 2007 +1000 17541 17542 Match drm 0.0.7 interface changes 17543 17544commit 45dccb99a541a0e1e7be7b4e5aab12c3d8d9f30d 17545Author: Ben Skeggs <skeggsb@gmail.com> 17546Date: Sun Jun 24 16:47:36 2007 +1000 17547 17548 Hide the allocator's use of physical addresses 17549 17550commit d41cf4febd555cca1c65df7aa981be64a6ae4219 17551Author: Ben Skeggs <skeggsb@gmail.com> 17552Date: Sun Jun 24 16:46:40 2007 +1000 17553 17554 Pass pScrn instead of pNv in a few places 17555 17556commit d59fbb4725602c409beee4ab96dca1eab5d73b2a 17557Author: Ben Skeggs <skeggsb@gmail.com> 17558Date: Mon Jun 11 20:36:28 2007 +1000 17559 17560 don't nuke user-specified CFLAGS 17561 17562commit 7af7d95eb8826fa3fb9914a8c6ad4b6f4867095e 17563Author: Henry Zhao <henryz@localhost.localdomain> 17564Date: Mon Apr 23 01:38:09 2007 +0200 17565 17566 Import from nv : Need to inject a mode corresponding panel width/height for validation 17567 17568commit 896fe62233f62d5e1f874ed0eba3a200fa5ef14b 17569Author: Richard Hughes <richard@hughsie.com> 17570Date: Sun Apr 15 23:01:52 2007 +0200 17571 17572 Fix make dist. 17573 17574commit e0e90f427581929fd01c96f1958a1292982dee83 17575Author: Ben Skeggs <skeggsb@gmail.com> 17576Date: Mon Mar 26 01:04:15 2007 +1000 17577 17578 nouveau: rip out card init, everything except modesetting. 17579 17580commit 23fdec37c580c3b5de263dfd48bd80fa9f21bc19 17581Author: Ben Skeggs <skeggsb@gmail.com> 17582Date: Wed Mar 21 17:55:38 2007 +1100 17583 17584 nouveau: bump for drm interface changes (v0.0.5) 17585 17586commit 1fddcfe6680148ed2edd8017217734a1f812298c 17587Author: Matthieu Castet <castet.matthieu@free.fr> 17588Date: Sun Mar 18 11:30:54 2007 +0100 17589 17590 Fix "Fix NoAccel mode" commit : NVAccelCommonInit return a bool 17591 17592commit 113c6837a6515d533bfa15daf49f6cd2c4c95ac4 17593Author: Matthieu Castet <castet.matthieu@free.fr> 17594Date: Sun Mar 18 11:28:04 2007 +0100 17595 17596 Fix NoAccel mode 17597 17598commit d1b75251a3f225890f1e25b11038389d71b67611 17599Author: Ben Skeggs <skeggsb@gmail.com> 17600Date: Wed Mar 7 21:41:22 2007 +1100 17601 17602 Rip out nv30_exa.c, we have other means to play with the 3D engine now. 17603 17604commit adb2df9a9c88bc9646b65a235aa36416c3064c9f 17605Author: Ben Skeggs <skeggsb@gmail.com> 17606Date: Wed Mar 7 16:46:27 2007 +1100 17607 17608 Xv: Make the blit adaptor play nice with EXA 17609 17610 Previously the blit adaptor assumed that all drawing (ie. XAA) would occur to 17611 the screen pixmap, so it didn't bother touching surface offsets. Now, with 17612 EXA this isn't true and the card complained about exceeding the colour buffer 17613 limits when attempting to play video. 17614 17615 This patch kills 2 birds with one stone; instead of pointing the GPU back at 17616 the screen pixmap when playing video, we point it at the destination drawable. 17617 So, Xv will also hopefully play nicely with redirected windows now. 17618 17619commit f28abbedfc7deeb90317dd24c7ec5dfe1ae94fdf 17620Author: Ben Skeggs <skeggsb@gmail.com> 17621Date: Wed Mar 7 13:58:48 2007 +1100 17622 17623 Only use Y32 for Solid(). 17624 17625 It's the only place the workaround is needed at this point, and in some places 17626 we need to use the "proper" ARGB format (Xv for instance). 17627 17628commit 6441253d34d118befe8950781ba012fc58bb5158 17629Author: Ben Skeggs <skeggsb@gmail.com> 17630Date: Wed Mar 7 04:33:44 2007 +1100 17631 17632 Fail configure if xorg-server isn't at least version 1.1 17633 17634 The minimum requirement has been 1.1 for a while now, but would only fail 17635 at compile time when compiling nv_exa.c. This patch makes configure fail, 17636 and removes the remaining conditional support for older servers. 17637 17638 Hopefully the configure.ac is ok, I suck at autofoo. 17639 17640commit e568c2d8e82b1bcf25e495f4a82331803d954234 17641Author: Ben Skeggs <skeggsb@gmail.com> 17642Date: Wed Mar 7 04:22:24 2007 +1100 17643 17644 reindent and move nv_video.c around a bit 17645 17646commit a0a5586836eb87a8c21abb06f9b8afd64eec714a 17647Author: Ben Skeggs <skeggsb@gmail.com> 17648Date: Tue Mar 6 05:42:29 2007 +1100 17649 17650 Use Y32 format instead of A8R8G8B8 for 32-bpp pixmaps 17651 17652 fixes 32-bpp EXA solid-fills, previously alpha was forced to 0xff. 17653 17654commit 2e9976c034b6c310b7ad0fc858250c5e54781a9e 17655Author: Ben Skeggs <skeggsb@gmail.com> 17656Date: Mon Mar 5 14:45:04 2007 +1100 17657 17658 move NVAccel* funcs into nv_accel_common.c 17659 17660commit add6463b848fab274cb5cc190a78aeafc1389203 17661Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17662Date: Sun Mar 4 22:40:48 2007 +0100 17663 17664 Joint patch from Ben Skeggs and me that fixes some (all ?) composite issues. 17665 17666commit 83868bd0a36520da49e40e2b8404e4652de39824 17667Author: Ben Skeggs <darktama@iinet.net.au> 17668Date: Thu Mar 1 08:39:23 2007 +1100 17669 17670 nouveau: avoid attempting an AGP-only alloc 17671 17672 will fail on non-AGP cards, and leave nasty messages in Xorg.log... 17673 17674commit da1dd87acb7061b4772c271dc3c9071a3b160e8a 17675Author: Ben Skeggs <skeggsb@gmail.com> 17676Date: Wed Feb 28 15:30:52 2007 +1100 17677 17678 match drm v4 interface changes 17679 17680 context (gr/dma) object handling: 17681 - remove use of flags/dmaobj when creating objects 17682 - move object creation/setup into nv_accel_common.c 17683 - NV04: make PGRAPH complain a bit more if we program the hardware 17684 incorrectly. 17685 - NV04/NV10: enable PGRAPH_DEBUG_3_CTX_METHODS 17686 This bit seems to enable the use of SET_* methods. 17687 - NV04/NV10/NV40: enable PGRAPH_DEBUG_3_IGNORE_PATCHVALID 17688 Objects no longer get PATCH_CONFIG_VALID set in their context. 17689 Not sure how to get this set other than to intercept the 17690 PATCH_EXCEPTION error and enable the flag. I opted for the 17691 same method NVIDIA use, which makes PGRAPH ignore it. 17692 misc: 17693 - use NV04_SCALED_IMAGE_FROM_MEMORY on all NV_ARCH_04 cards 17694 - fix offScreenBase calculation (EXA) 17695 - move mem alloc/free helpers into nv_mem.c 17696 17697commit 7e5f90b264fdb3081250ff16748164998b6461a6 17698Author: Adam Jackson <ajax@benzedrine.nwnk.net> 17699Date: Tue Feb 27 14:46:20 2007 -0500 17700 17701 Typo fix. 17702 17703commit c85da5d8b346e17e68da68e57d62c3f77029b680 17704Author: Damien Thebault <damien.thebault@laposte.net> 17705Date: Wed Feb 14 00:35:50 2007 +0100 17706 17707 Don't output the chipset list to the log file, silence some warnings. 17708 17709commit f650cb126e042b44126df6a298e996726356ec5b 17710Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17711Date: Wed Feb 14 00:29:52 2007 +0100 17712 17713 Fix the workaround (or workaround the fix). 17714 17715commit ac2784d38634bc6d8aec3a94d609b60cd49b21b6 17716Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17717Date: Wed Jan 24 20:39:40 2007 +0100 17718 17719 Make NV_DMA_DEBUG print the fifo position. 17720 17721commit d54d441df4294bf4fed26635a738f4111087e06d 17722Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17723Date: Wed Jan 24 19:13:05 2007 +0100 17724 17725 Increase the fifo dump size. 17726 17727commit b21dbb170aa49c83ef450c2ab8c34614277e7fbc 17728Author: B. Rathmann <koala_br@users.sourceforge.net> 17729Date: Wed Jan 24 13:46:16 2007 +0100 17730 17731 Fix the G70 issues. 17732 17733commit e683979666a4c07590b7bade2f62cca33c19a3b2 17734Author: Ben Skeggs <darktama@iinet.net.au> 17735Date: Fri Jan 19 15:40:22 2007 +1100 17736 17737 bump drm patchlevel 17738 17739commit feff7cb67ae61987146a21dad535702f3db2264c 17740Author: Jeremy Kolb <jkolb@brandeis.edu> 17741Date: Thu Jan 18 21:28:51 2007 -0500 17742 17743 Fill in some more nv30 magic values. Not sure what they do yet but they don't hurt anything. 17744 17745 Apply names to other registers. 17746 17747commit 31462c30bb79e8245edee0cf57c46b889d2f259b 17748Author: Jeremy Kolb <jkolb@brandeis.edu> 17749Date: Wed Jan 17 08:34:56 2007 -0500 17750 17751 Fix PGRAPH_DEBUG* regs to match what's done on the nv35. Specifically change DEBUG_3 | 0x00de0000 for data/format/dma checking. This is consistent with other cards and dumps so it should be correct. 17752 17753commit 4583312ba72047e6bc65bfc576a0bcf90e04b2de 17754Author: David Jeannot <djeannot24@gmail.com> 17755Date: Tue Jan 16 00:03:56 2007 +0100 17756 17757 Indent nv_bios.c properly. 17758 17759commit 6246fba5a54dc7af9a8b16c5a14131180831562d 17760Author: Patrice Mandin <patrice@manoir.racoon.city> 17761Date: Fri Jan 12 20:25:16 2007 +0100 17762 17763 Add missing include for malloc/free 17764 17765commit 37937980ec7b21c40539daee029886c4f4a0f0cd 17766Author: Jeremy Kolb <jkolb@brandeis.edu> 17767Date: Fri Jan 12 00:26:20 2007 -0500 17768 17769 NV30: Set DEBUG3 reg bit 1. Turns on automatic hw context switching. 17770 17771commit d5617e91ff13f9e19d9e5d421fd081f59dec685c 17772Author: jkolb@brandeis.edu <jkolb@brandeis.edu> 17773Date: Thu Nov 2 20:23:36 2006 -0500 17774 17775 Just a superficial cleaning up. Nothing exciting. 17776 17777commit 1f6239de8d7618de6aee402eaf1d8552524a996e 17778Author: Ben Skeggs <darktama@iinet.net.au> 17779Date: Tue Jan 2 15:08:36 2007 +1100 17780 17781 NV4x changes to work with latest drm. 17782 17783commit f9ba29a07581ede8c409b0eb8b549db3b446f490 17784Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17785Date: Wed Dec 27 21:52:15 2006 +0100 17786 17787 Output the driver date to the logs on startup. 17788 17789commit 2f1d14fe1c03c32ad60776e0d0c2ae41a42ea4be 17790Author: Ben Skeggs <darktama@iinet.net.au> 17791Date: Tue Dec 12 00:13:05 2006 +1100 17792 17793 Remove final remaining PRAMIN access (NV04-specific), the drm handles it now 17794 17795commit de7255d5420c64a2707f5e78a4157d93618e6c17 17796Author: Ben Skeggs <darktama@iinet.net.au> 17797Date: Fri Dec 8 18:09:26 2006 +1100 17798 17799 Provide some visuals with a stencil buffer, and actually call NVDRIInitVisualConfigs. 17800 17801commit e64dcc59b607ae5ad7049c35fed6d6a300cd5f6e 17802Merge: 0f7957d 656dfcd 17803Author: Matthieu Castet <mat@mat-pc.(none)> 17804Date: Wed Dec 6 22:09:23 2006 +0100 17805 17806 Merge branch 'master' of git+ssh://matc@git.freedesktop.org/git/nouveau/xf86-video-nouveau/ 17807 17808commit 0f7957da9ce856ef0accc922ab368ac10171b5e6 17809Author: Matthieu Castet <mat@mat-pc.(none)> 17810Date: Wed Dec 6 22:07:26 2006 +0100 17811 17812 More pgraph regs 17813 17814commit 656dfcd635551245b2f58835c5ea87474d3eb60d 17815Merge: ed307d5 224578b 17816Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17817Date: Sun Dec 3 10:20:04 2006 +0100 17818 17819 Merge branch 'master' of git+ssh://marcheu@git.freedesktop.org/git/nouveau/xf86-video-nouveau 17820 17821commit ed307d559be72e06f64566574beb36200ebe47e2 17822Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17823Date: Sun Dec 3 10:14:04 2006 +0100 17824 17825 Introduce lock step versioning with the drm. 17826 Fix the MEM_ALLOC issue. 17827 17828commit 224578b74645fe3624669dc3521fe6deb1cc4aae 17829Author: Matthieu Castet <mat@mat-pc.(none)> 17830Date: Tue Nov 28 19:12:55 2006 +0100 17831 17832 Don't alloc fifo if it is not needed. 17833 17834commit 146eb74cb7d1da1da55247a04e7a969d98601e42 17835Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17836Date: Mon Nov 27 00:03:11 2006 +0100 17837 17838 Mandatory oops commit. 17839 17840commit bdfa6776128f4bc7e2f3caa57296571acb9bd661 17841Author: Andrew Mitchell <ajmitch@ubuntu.com> 17842Date: Sun Nov 26 23:59:06 2006 +0100 17843 17844 Remove XF86DRI defines from the driver. It's already mandatory anyway. 17845 17846commit 7ebc8c4327d0b50f60c4cd4423414d4cfbb8b7d8 17847Author: airlied <airlied@optimus.localdomain> 17848Date: Fri Nov 17 16:23:38 2006 +1100 17849 17850 add VGA CRTC registers from haiku 17851 17852commit 7150a4357dff0a2e4468190e2fd228ed07d367f9 17853Author: Dave Airlie <airlied@linux.ie> 17854Date: Fri Nov 17 12:55:37 2006 +1100 17855 17856 remove nvvga.h unused 17857 17858commit 64a1c5bb90fd13ce1c191873a4f4afc2a9ede09b 17859Author: Dave Airlie <airlied@linux.ie> 17860Date: Fri Nov 17 12:51:11 2006 +1100 17861 17862 find and version DRM earlier in startup sequence so it is loaded firstup 17863 17864commit 659b5de8200d9022e7616f03b20a306fff2af0fe 17865Author: Matthieu Castet <castet.matthieu@free.fr> 17866Date: Fri Nov 17 00:59:58 2006 +0100 17867 17868 Cleanup of PVIDEO. 17869 17870commit 370b83eb3347d14003f2878737d25de7dacaeec1 17871Author: Matthieu Castet <castet.matthieu@free.fr> 17872Date: Wed Nov 15 22:10:27 2006 +0100 17873 17874 Use objects from nouveau_reg.h. 17875 17876commit 2ffeb45a953a2ff52abe6f78dacfb9c53f873dbb 17877Author: Ben Skeggs <darktama@iinet.net.au> 17878Date: Wed Nov 15 01:22:18 2006 +1100 17879 17880 Set SURFACE_FORMAT in NVExaPrepareCopy. Fixes one of the many issues with xorg-server-1.2 17881 17882commit 85bda095107558846fe6fb803b6df412213b27f4 17883Author: Ben Skeggs <darktama@iinet.net.au> 17884Date: Wed Nov 15 00:47:04 2006 +1100 17885 17886 nv_exa.c tab goodness 17887 17888commit 8b733a25c30fb9e499d678c26b777d072d1ac187 17889Author: Ben Skeggs <darktama@iinet.net.au> 17890Date: Tue Nov 14 08:31:21 2006 +1100 17891 17892 Fix type. 0x84 is DEBUG_1 17893 17894commit 030eb38dd1fe767807aab0447ed4fab4ec8c368a 17895Author: Ben Skeggs <darktama@iinet.net.au> 17896Date: Tue Nov 14 03:09:05 2006 +1100 17897 17898 Fill in front_offset/front_pitch 17899 17900commit 589bb37ff37bb7d4775f5dca0301c5d97ea73b06 17901Author: Ben Skeggs <darktama@iinet.net.au> 17902Date: Sun Nov 12 16:04:32 2006 +1100 17903 17904 Add NVDRIFinishScreenInit. 17905 17906commit d0c09a1bd0e1c26cc591a3d8738b94b1efe576ec 17907Author: Dave Airlie <airlied@linux.ie> 17908Date: Sun Nov 12 14:08:45 2006 +1100 17909 17910 add some names to registers 17911 17912commit 84928fd946b4f02d8a943226481304417475c4f9 17913Author: Ben Skeggs <darktama@iinet.net.au> 17914Date: Sun Nov 12 07:43:24 2006 +1100 17915 17916 Import nouveau_dri.h from the 3D driver as nv_dri.h, and setup DRIInfo->devPrivate. 17917 17918commit cdb9e3acf0aa5cdf6e5aa33fb499ea8e93f7f1f2 17919Author: Ben Skeggs <darktama@iinet.net.au> 17920Date: Sun Nov 12 07:33:53 2006 +1100 17921 17922 Temporary workaround for DRI drmMap problem, not sure of what we actually need to do here yet. 17923 17924commit d3babbfbed898921e1a167df08eef92d381d6aa0 17925Author: Ben Skeggs <darktama@iinet.net.au> 17926Date: Sun Nov 12 07:28:31 2006 +1100 17927 17928 nv_dri.c still had some space-indented functions, not anymore :) 17929 17930commit bad4e6c674ac4fdc0aebbbc6d36fc2854ee8f885 17931Merge: 1b14821 c096565 17932Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17933Date: Mon Nov 6 03:14:07 2006 +0100 17934 17935 Merge branch 'master' of git+ssh://marcheu@git.freedesktop.org/git/nouveau/xf86-video-nouveau 17936 17937commit 1b14821ecd67a62a34a902e62c0b229bb9ed97ca 17938Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17939Date: Mon Nov 6 03:13:50 2006 +0100 17940 17941 Added a DRI visual configs functions. It should work but I can't test it... 17942 17943commit c096565a0f6e6c12255e57eaefb878904ce04971 17944Author: Ben Skeggs <darktama@iinet.net.au> 17945Date: Mon Nov 6 11:54:19 2006 +1100 17946 17947 Fix TEX0_IN_TEX1 fragprog.. I must've been asleep when I did that.. 17948 17949commit 47f32cfe6e2f513f86b7a8f02a8872c134e2327d 17950Author: Dave Airlie <airlied@ppcg5.localdomain> 17951Date: Mon Nov 6 11:45:53 2006 +1100 17952 17953 use pagesize to work out the sarea size 17954 17955commit e6c4d40e247939f677e285db15a78ccc33b92b41 17956Author: Dave Airlie <airlied@ppcg5.localdomain> 17957Date: Mon Nov 6 08:39:49 2006 +1100 17958 17959 don't fail dri module load if loaded already 17960 17961commit 4cc4a1f7ef559f1ce56706f16ee24a3b6879d28d 17962Merge: b95f8aa b4e4307 17963Author: Dave Airlie <airlied@linux.ie> 17964Date: Sun Nov 5 18:00:29 2006 +1100 17965 17966 Merge ssh://10.1.1.200/home/airlied/git/xf86-video-nouveau 17967 17968commit b4e4307af982b50aff191983d209cb855a377f80 17969Author: Dave Airlie <airlied@ppcg5.localdomain> 17970Date: Sun Nov 5 17:59:17 2006 +1100 17971 17972 fix G5 endianness issue with panel size regs 17973 17974commit b95f8aac9e6dcbf4d10a5e26bde6a0f3595608cb 17975Merge: ba71154 e6b1f68 17976Author: Dave Airlie <airlied@linux.ie> 17977Date: Sun Nov 5 17:53:53 2006 +1100 17978 17979 Merge branch 'master' of ssh://git.freedesktop.org/git/nouveau/xf86-video-nouveau 17980 17981commit ba711545a0fd091c05b514f30c1e904370026499 17982Author: Dave Airlie <airlied@ppcg5.localdomain> 17983Date: Sun Nov 5 17:51:10 2006 +1100 17984 17985 set the PPC use BE MMIO 17986 17987commit e6b1f687e3bf127d8dba1871d67eff9d83ac284e 17988Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17989Date: Sat Nov 4 16:57:30 2006 +0100 17990 17991 Remove the context object flags handling from the DDX since it's now in the DRM. 17992 17993commit ead77e0ec1d0c62829fa973db3d62937b44781e5 17994Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17995Date: Sat Nov 4 04:47:57 2006 +0100 17996 17997 Add the NV44A 3D object class. 17998 17999commit 87463666ab23e047f355cfcd7016aa9c7cd8ad76 18000Author: Ben Skeggs <darktama@iinet.net.au> 18001Date: Fri Nov 3 04:00:09 2006 +1100 18002 18003 Use "do { blah; } while(0)" instead of just braces. (suggestion from megari on irc) 18004 18005commit 68682238b36292409ce7f41ed90bd3be860e77f3 18006Author: Ben Skeggs <darktama@iinet.net.au> 18007Date: Thu Nov 2 16:07:26 2006 +1100 18008 18009 Install NV30_TCL EXA hooks if NV_ENABLE_3D is defined. 18010 18011 Ok.. Firstly I don't recommend using this code, it is not ready for general 18012 usage. We probably only want to keep the Composite hook, but the others are 18013 useful to test things with. 18014 18015 1. You will most likely (read: definitely) need to start X with the binary 18016 driver before starting X with nouveau/NV30_TCL.. If you don't PGRAPH will 18017 yell at you and X will crash. 18018 2. Copy() is bad when the src and dest are the same, so scrolling and moving 18019 windows around will fall back to software rendering.. 18020 3. Composite() is slightly better than the one in nv_exa.c, but it's buggy in 18021 a few cases and not complete yet. Particularly, transforms are missing. 18022 4. There are some correctness problems that I haven't bothered to fix yet, so 18023 you'll probably see bits of windows left behind when moving them around.. 18024 5. Don't be surprised if it's horribly broken :) 18025 18026commit c371d90935d6baf2f1dd0073741637d14b506f59 18027Author: Ben Skeggs <darktama@iinet.net.au> 18028Date: Thu Nov 2 15:57:47 2006 +1100 18029 18030 Put braces around NVDmaSetObjectOnSubchannel... 18031 18032commit b45aea22a8da69ad4a2493f7f33fa65186a83ba3 18033Author: Ben Skeggs <darktama@iinet.net.au> 18034Date: Thu Nov 2 14:52:24 2006 +1100 18035 18036 Merge >=NV30-specific EXA code.. hard-disabled pending testing on my C51, and a proper commit message.. 18037 18038commit e79b0befe92110c1cd2e0e1ae69fced1cd5acba3 18039Author: Ben Skeggs <darktama@iinet.net.au> 18040Date: Thu Nov 2 01:29:34 2006 +1100 18041 18042 Reformat NVExaInit 18043 18044commit 6a77e5bf2e58b19df67c029ad7bffd92f8cf652f 18045Author: Ben Skeggs <darktama@iinet.net.au> 18046Date: Thu Nov 2 01:18:14 2006 +1100 18047 18048 First part of preparation for NV30_TCL EXA code merge. 18049 18050commit 1d6e911ab5391277b403ef6f7b095203cfd19a53 18051Author: Ben Skeggs <darktama@iinet.net.au> 18052Date: Thu Nov 2 00:38:53 2006 +1100 18053 18054 Reindent nv_dma.c with tabs.. 18055 18056commit 06748f74f639242fd690a3dda7e685ae4fa1f95e 18057Author: Ben Skeggs <darktama@iinet.net.au> 18058Date: Wed Oct 11 11:06:16 2006 +1100 18059 18060 s/init/alloc/ 18061 18062commit a2d55603db8c01cc4b9f3404c282b1e4963a152c 18063Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 18064Date: Wed Oct 11 00:39:36 2006 +0200 18065 18066 Remove the PFIFO_REINIT hack, and enable the irq by default. This commit requires the matching drm commit, and will probably break stuff. 18067 18068commit 78537b3342bbf1c16dc78f8f06cb3f989ce8f03f 18069Author: Ben Skeggs <darktama@iinet.net.au> 18070Date: Thu Sep 14 12:56:28 2006 +1000 18071 18072 It usually helps to wait on a notifier, and not on a NVAllocRec.. Oops. 18073 18074commit 01dde28611238cf2d81b087b371cc67296d1b581 18075Author: Ben Skeggs <darktama@iinet.net.au> 18076Date: Thu Sep 14 12:17:32 2006 +1000 18077 18078 Allocate notifiers with NVAllocateMemory. 18079 Fix some warnings. 18080 Return success/fail for DMA and Context object creation. 18081 18082commit 3d5c560f509646853cc863e35ed2412db39cd8a3 18083Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 18084Date: Tue Sep 5 07:39:10 2006 +0200 18085 18086 Port some of the least intrusive nv03 code to the nouveau driver. 18087 18088commit 6fb7a63bf5d4f046583f9d14862c16ec5979a318 18089Author: Ben Skeggs <darktama@iinet.net.au> 18090Date: Sun Sep 3 06:41:52 2006 +1000 18091 18092 Config options for cmdbuf location/size. 18093 18094 Will default to putting the cmdbuf in AGP (except for PowerPC) if an AGP card 18095 is detected. Won't currently be detected because something is wrong DRM-size. 18096 18097 If you have an AGP card you can manually specify 'FBLocation "AGP"' in 18098 xorg.conf to make it work. 18099 18100 Needs an updated DRM to have any effect. 18101 18102commit e49d561d2328eeac7691661e28955a882bba2192 18103Author: Ben Skeggs <darktama@iinet.net.au> 18104Date: Sun Sep 3 01:51:23 2006 +1000 18105 18106 Copy+Paste a workaround for a NV40 hardware bug from NVShowHideCursor() into 18107 LoadStateExt(). This should fix the problem of the HW-cursor still being 18108 enabled when VT-switching back to a framebuffer console. 18109 18110commit 69ace0c4388cb1a66ea74f627fe02fe89e7d35c0 18111Author: Ben Skeggs <darktama@iinet.net.au> 18112Date: Sat Sep 2 22:17:55 2006 +1000 18113 18114 Kill off one of the M2MF objects, and use the engine's SET_OBJECT_{IN_OUT} to 18115 swap between AGP->VRAM and VRAM->AGP. 18116 18117commit 72cf79b9e5dc0d97e933f540d063161129bbebf7 18118Author: Ben Skeggs <darktama@iinet.net.au> 18119Date: Sat Sep 2 21:59:26 2006 +1000 18120 18121 Don't create objects that aren't needed. This frees up a couple of subchannels 18122 for EXA, use one of them for MEMORY_TO_MEMORY_FORMAT. 18123 18124commit 6a06d9022f3c0bc845f02091d83faf95f69f81bd 18125Author: Ben Skeggs <darktama@iinet.net.au> 18126Date: Sat Sep 2 18:15:15 2006 +1000 18127 18128 Move code out of nv_xaa.c that doesn't belong there. 18129 18130commit 46ac77d524dd183c6b50a7a0e77990514ff28840 18131Author: Ben Skeggs <darktama@iinet.net.au> 18132Date: Sat Sep 2 17:57:38 2006 +1000 18133 18134 Add back some setup that went missing during PFB reg cleanup. 18135 Should fix corruption on NV40. 18136 18137commit 03b86f08667a10c44420d60787fe9e17b5d20ab8 18138Author: Ben Skeggs <darktama@iinet.net.au> 18139Date: Thu Aug 31 15:53:25 2006 +1000 18140 18141 Rename module from nv_drv.so to nouveau_drv.so to allow both nv and nouveau 18142 to coexist. Also disabled building of riva128.so, as we plan on merging it 18143 with the rest of the nv code. 18144 18145commit 17c871a44987af04d1688798e2e7802a5f5ba522 18146Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18147Date: Wed Aug 30 06:24:23 2006 +1000 18148 18149 clean up FB register names 18150 18151commit 7d3c12424fd79daf831b7c28780586e8bd9c3574 18152Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18153Date: Wed Aug 30 04:23:43 2006 +1000 18154 18155 add some names for the RAMDAC registers 18156 18157commit 0e70296ce9f7a921c4b971919d12c262202d8d48 18158Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18159Date: Wed Aug 30 02:18:02 2006 +1000 18160 18161 add some names for some of the CRTC registers - some names not very meaningful 18162 yet 18163 18164commit 9a5e2c5be96eaced3aadd153fb2f9b1928173af7 18165Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18166Date: Wed Aug 30 01:48:07 2006 +1000 18167 18168 remove select head registers PCIO and PDIO 18169 18170commit dec0691db79b1ded2942c1cbfa97434fa8693f9b 18171Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18172Date: Wed Aug 30 01:13:05 2006 +1000 18173 18174 remove unused pointers after macroisation 18175 18176commit 9444b692607895572c8199a60d6968fbc374d51e 18177Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18178Date: Wed Aug 30 01:02:45 2006 +1000 18179 18180 missed 2x PGRAPHs in XAA 18181 18182commit f73cb8ef4f4fbb588bfe9ca767286153b2565a37 18183Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18184Date: Wed Aug 30 00:59:39 2006 +1000 18185 18186 remove PCRTC via macros 18187 18188commit 6d5b1765e10856a7dfa75c27ad96f547dc26b23a 18189Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18190Date: Wed Aug 30 00:48:23 2006 +1000 18191 18192 remove PEXTDEV and PTIMER using macros 18193 18194commit 6140bdfe6959cd0f6ce8e06d34b6e838d8cef046 18195Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18196Date: Wed Aug 30 00:43:07 2006 +1000 18197 18198 remove PMC using macros 18199 18200commit 28d9befa6308462a6bc1809b894ebe389873fd6a 18201Author: Ben Skeggs <darktama@iinet.net.au> 18202Date: Wed Aug 30 01:26:58 2006 +1000 18203 18204 Make EXA the default 18205 18206commit fa9209f77df3d835ebad43dc82988c000398f650 18207Merge: ab2b550 2f7caf6 18208Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 18209Date: Tue Aug 29 13:42:49 2006 +0200 18210 18211 Merge branch 'master' of ssh://marcheu@git.freedesktop.org/git/nouveau/xf86-video-nouveau 18212 18213commit ab2b5509e82a646ac187b020c91234e33f81065d 18214Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 18215Date: Tue Aug 29 13:42:34 2006 +0200 18216 18217 Make use of NV5_SCALED_IMAGE_FROM_MEMORY on NV05. 18218 18219commit 9911eb726f99166968165710a9e2f540388cd42a 18220Author: Stephane Marchesin marchesin@icps.u-strasbg.fr <steph@tubular-brain.localdomain> 18221Date: Tue Aug 29 03:18:27 2006 +0200 18222 18223 Enable the composite hooks. 18224 18225 After changing NV_SCALED_IMAGE_FROM_MEMORY for NV10_SCALED_IMAGE_FROM_MEMORY it seems the composite hook works. 18226 Also, add a more complete list of desktop pci ids. 18227 18228commit 2f7caf6eb5eb6a3991720fc413bd20723473212f 18229Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18230Date: Tue Aug 29 08:13:59 2006 +1000 18231 18232 remove PGRAPH direct accesses 18233 18234commit 51f0b67d9064fffc6b357d2923188a030e85eace 18235Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 18236Date: Tue Aug 29 07:51:49 2006 +1000 18237 18238 remove PFB usage with macros 18239 18240commit b50736d1583b960ab70f3e113294db8c2da69e31 18241Author: Ben Skeggs <darktama@iinet.net.au> 18242Date: Mon Aug 28 17:07:06 2006 +1000 18243 18244 Use NVAllocateMemory for AGP scratch buffer. 18245 18246commit a5ddd8fcf4278c35d71605794d0d5711e7c0ebb7 18247Author: Dave Airlie <airlied@linux.ie> 18248Date: Mon Aug 28 16:41:20 2006 +1000 18249 18250 remove legacy crap from nvreg.h add cursor reg defines 18251 18252commit f035bc63c34bce49e1c2619f14ab98b4b1922beb 18253Author: Dave Airlie <airlied@linux.ie> 18254Date: Mon Aug 28 16:23:32 2006 +1000 18255 18256 missed some vga routines 18257 18258commit 0b45a9f725442d5bd784ee67b7473929993f6082 18259Author: Dave Airlie <airlied@linux.ie> 18260Date: Mon Aug 28 16:19:56 2006 +1000 18261 18262 add RAMDAC register read/write 18263 18264commit 24267d7e39383930c589bc80e2bbd4e5cf66ec2a 18265Merge: 349496c c0154e6 18266Author: Ben Skeggs <darktama@iinet.net.au> 18267Date: Mon Aug 28 14:46:51 2006 +1000 18268 18269 Merge branch 'master' of git+ssh://git.freedesktop.org/git/nouveau/xf86-video-nouveau 18270 18271commit 349496c97fafed04806c01bd9d4386b0fcaa9fd2 18272Author: Ben Skeggs <darktama@iinet.net.au> 18273Date: Mon Aug 28 14:45:42 2006 +1000 18274 18275 More cleanups from DRM memory alloc changes. 18276 18277commit c0154e615c091c7bb483fcfd07e216b21618d0f2 18278Author: Dave Airlie <airlied@linux.ie> 18279Date: Mon Aug 28 14:07:05 2006 +1000 18280 18281 use nvReadVGA and nvWriteVGA in more places 18282 18283commit e0f1479773c1a00ec69edbece98e5e2b66c0c2c9 18284Author: Ben Skeggs <darktama@iinet.net.au> 18285Date: Sun Aug 27 22:45:36 2006 +1000 18286 18287 Fix ShadowFB for FB alloc changes, should fix RandR rotation also. 18288 18289commit 0de4d0b8b79e2238f2c4b374b5012a25ae985502 18290Author: Ben Skeggs <darktama@iinet.net.au> 18291Date: Sun Aug 27 22:27:17 2006 +1000 18292 18293 Slash UseFBDev and DGA support. 18294 18295commit e49d08ff482755e9f95c4c3b3cf065e6f0d2bc23 18296Author: Dave Airlie <airlied@linux.ie> 18297Date: Sun Aug 27 10:04:50 2006 +1000 18298 18299 add some read/write NV VGA register wrappers 18300 18301commit c64c55b3133221462980b934a308daf2cc0e473d 18302Author: Dave Airlie <airlied@linux.ie> 18303Date: Sun Aug 27 09:56:36 2006 +1000 18304 18305 remove all U0* types in favour of more standard CARD* types 18306 18307commit 9c294b962379a955d2e8e2cb2c54f845fd4c5740 18308Author: Ben Skeggs <darktama@iinet.net.au> 18309Date: Sat Aug 26 00:43:21 2006 +1000 18310 18311 Wow I suck tonight.. another oops. 18312 18313commit b8411ae92b6833c9336f82ed86dc8755ca8b9dcc 18314Author: Ben Skeggs <darktama@iinet.net.au> 18315Date: Sat Aug 26 00:39:36 2006 +1000 18316 18317 Whoa, how did that happen... 18318 18319commit d458cdac152a49eec9acdfe5081f6d426e712bf2 18320Author: Ben Skeggs <darktama@iinet.net.au> 18321Date: Sat Aug 26 00:33:47 2006 +1000 18322 18323 Initial work in porting the DDX to use the DRM for memory allocations. 18324 This is enough to make it work, but there's still a *lot* of cleanups 18325 to do. 18326 18327commit 83ae039b8f8ea08ae9436201014c7e1ba25a4839 18328Author: Dave Airlie <airlied@linux.ie> 18329Date: Fri Aug 25 19:33:08 2006 +1000 18330 18331 fixup compile for git 18332 18333commit ea2bc33d32bf9abeb23b3fe8cb6c9d28ddccef01 18334Author: Dave Airlie <airlied@linux.ie> 18335Date: Fri Aug 25 19:17:51 2006 +1000 18336 18337 updated git tree from changes that got lost on way from CVS 18338 18339commit 07b81112ba4678085a148e5ffaadd4684f74d64d 18340Author: marcheu <marcheu> 18341Date: Thu Aug 24 21:00:42 2006 +0000 18342 18343 Make NVInitAGP fail if there is no AGP area. Maybe that function shouldn't be called for PCI(-e) cards in the first place. 18344 18345commit 109b5abed6c0c701a1fb84ec63ff8f5914eee8e3 18346Author: marcheu <marcheu> 18347Date: Wed Aug 23 21:32:47 2006 +0000 18348 18349 If you stare at nv_hw.c long enough, things start to make sense. I swear ! 18350 18351commit 80ba1cfec290dbb82c4a1042255a21b626512e84 18352Author: marcheu <marcheu> 18353Date: Wed Aug 23 17:11:22 2006 +0000 18354 18355 Comment out NVDmaKickoff() in NVResetGraphics(). NVResetGraphics is called twice on startup, and flushed to the card in between. Since NVResetGraphics resets the fifo pointer, this sometimes ends up in a situation where the CPU writes at the same place of the fifo (at the beggining) the GPU is currently reading from. Obviously this is bad. 18356 18357 Also, add code to dump fifo in case of crashes. 18358 18359commit 9b19daa55c7770a3e66b2374da0212c75ed21260 18360Author: marcheu <marcheu> 18361Date: Tue Aug 22 01:16:18 2006 +0000 18362 18363 Remove debug message. 18364 18365commit b44b54e4c369c612a692eabb8662b3db97758a83 18366Author: marcheu <marcheu> 18367Date: Mon Aug 21 13:19:06 2006 +0000 18368 18369 Fill more fields in the DRIInfo struct. 18370 18371commit aaf13d0e7cc5c76ee81b4234dc2c55bbedac1faa 18372Author: darktama_ <darktama_> 18373Date: Sun Aug 20 15:11:06 2006 +0000 18374 18375 Oops, this shouldn't be here still... 18376 18377commit 00fd86dea8d2eaa561cc89a35046d15bcf92bb82 18378Author: darktama_ <darktama_> 18379Date: Sun Aug 20 14:40:43 2006 +0000 18380 18381 - Allocate notifiers with the DRM 18382 18383 - Rename agpMemory and friends to agpScratch and friends 18384 18385commit d5c68d3f83f84ca67ee9df269b778bbed50851a0 18386Author: darktama_ <darktama_> 18387Date: Sun Aug 20 12:51:49 2006 +0000 18388 18389 Alloc AGP through the DRM 18390 18391commit 6d0924c272954f44b2a9b6b11be8c209dd67de8e 18392Author: marcheu <marcheu> 18393Date: Sat Aug 19 16:04:56 2006 +0000 18394 18395 Begginings of PINS parsing code I had laying around... 18396 18397commit 59e2d852d6f10ba26a394ff7ea5d3e7b30f2be71 18398Author: darktama_ <darktama_> 18399Date: Mon Jul 31 07:32:57 2006 +0000 18400 18401 Need.To.Pay.More.Attention... 18402 18403commit 9a314361fd004afe7129ae2d41f58ddd7fe6f56a 18404Author: darktama_ <darktama_> 18405Date: Mon Jul 31 03:24:17 2006 +0000 18406 18407 Updates for DRM changes, and a few cleanups 18408 18409commit 11fbb14876e7873a863de16b01fb5778e841b4bd 18410Author: erikwaling <erikwaling> 18411Date: Thu Jul 13 04:13:07 2006 +0000 18412 18413 Added more opcodes and some TODOs. 18414 18415commit dd54fea7562b8c9a6d26b645a03c9644c08090cc 18416Author: erikwaling <erikwaling> 18417Date: Thu Jul 13 03:09:11 2006 +0000 18418 18419 Changed INIT_MACRO to haiku-like implementation. 18420 18421commit b7e68083f3bf6b0943b58cfed0b82de61eeaac33 18422Author: marcheu <marcheu> 18423Date: Mon Jul 10 00:36:31 2006 +0000 18424 18425 Add a check that bios writes fall within known areas. Use these areas to setup mappings as well. Add some debug support to the bios code. 18426 18427commit fe2a1360d72385ad0f61f381a296820d02a60e79 18428Author: marcheu <marcheu> 18429Date: Mon Jul 10 00:30:54 2006 +0000 18430 18431 Use -Wall by default 18432 18433commit ca38f1d9d61393915515e1042204b35aeac863ba 18434Author: darktama_ <darktama_> 18435Date: Sat Jul 8 13:53:00 2006 +0000 18436 18437 Remove FIFO and object handling out of the DDX. This commit makes the DDX dependant on the DRM to operate at all.. 18438 18439commit b9b880b83896defe06ce61c9aef2f5ae4eb4f37e 18440Author: lumag <lumag> 18441Date: Mon Jun 12 20:21:11 2006 +0000 18442 18443 Fix exa acceleration by filling versioning fields. 18444 18445commit 0bbc0bcfe33c2625ea6175c3b052d231cc3b7c8f 18446Author: darktama_ <darktama_> 18447Date: Sat Jun 10 18:34:14 2006 +0000 18448 18449 Quick hack to make sure CRTC 0x44 contains it's original value when VT switching back to the console (also on server shutdown). 18450 18451 Should fix the problem of one display having scrolling issues, and losing 18452 it's text-mode cursor. 18453 18454commit 18e65f4e0604c744ea905319348c688141e14bbf 18455Author: marcheu <marcheu> 18456Date: Thu Jun 8 01:10:01 2006 +0000 18457 18458 Remove xf86ReadPciBIOS, use the bios from the reg aperture instead (that will probably work on more archs anyway) 18459 18460commit 2f5fcd131dbea5cb9873cc8a4aad44ae1bcf1340 18461Author: darktama_ <darktama_> 18462Date: Fri Jun 2 21:54:48 2006 +0000 18463 18464 Remove some XFree86LOADER conditionals 18465 18466commit 7889cc0241450aaa8e9dafc65f9f7976df3cd99b 18467Author: darktama_ <darktama_> 18468Date: Sat Apr 22 02:26:53 2006 +0000 18469 18470 Need to include nv_include.h before checking that XF86DRI is definied. XF86DRI lives in xorg-server.h - included from config.h 18471 18472commit 3083263eb40f57e4c529473c5e9f935109a1719f 18473Author: marcheu <marcheu> 18474Date: Fri Apr 21 22:15:59 2006 +0000 18475 18476 Don't include config.h in all files but in the *_include.h files. That's what they're here for after all. 18477 18478commit 864e2c6936a37005112a3c6f3f7db5a16100b96c 18479Author: darktama_ <darktama_> 18480Date: Fri Apr 21 02:46:31 2006 +0000 18481 18482 Include config.h in all source files 18483 18484commit 74ff8fc182c6f728a1b5e8a0415c25b8b7a4af94 18485Author: marcheu <marcheu> 18486Date: Thu Apr 20 18:54:12 2006 +0000 18487 18488 Added a fixme about laptop chips 18489 18490commit ad38154a7e075207d8f462ffb90489efe9cf01f1 18491Author: erikwaling <erikwaling> 18492Date: Thu Apr 20 18:50:11 2006 +0000 18493 18494 Ported nv_exa.c to work with EXA >=2.0.0 18495 18496commit 7c25d338e0c79a288fa192d9a2d4ac6eb59996c1 18497Author: Dave Airlie <airlied@linux.ie> 18498Date: Fri Aug 25 18:51:03 2006 +1000 18499 18500 initial import of nouveau code 18501 18502commit 98372d0f94a9d658a7257d2199ef37a2449b9961 18503Author: Adam Jackson <ajax@benzedrine.nwnk.net> 18504Date: Fri Aug 11 16:02:28 2006 -0400 18505 18506 Replace panel message + MODE_BAD with MODE_PANEL. 18507 18508commit 43ec66396f805aba87017d193d97a03372c29909 18509Author: Aaron Plattner <aplattner@nvidia.com> 18510Date: Fri Jun 30 12:54:44 2006 -0700 18511 18512 Bump version to 1.2.0 for release. 18513 18514commit 6265b77dc8878e36abd76b73e209a2d76d70ba22 18515Author: Mark Vojkovich <markv@xfree86.org> 18516Date: Thu Jun 29 18:41:44 2006 -0700 18517 18518 New chip support and new PCI IDs. 18519 18520commit 5114639dff73c22845451acffed94f1da284a7df 18521Author: Aaron Plattner <aplattner@nvidia.com> 18522Date: Mon Jun 19 09:32:11 2006 -0700 18523 18524 Generate the ChangeLog from git history for dist. 18525 18526commit 04d344dfcbb9928cf33943a880de333557e14be9 18527Author: Aaron Plattner <aplattner@nvidia.com> 18528Date: Fri Jun 16 18:03:21 2006 -0700 18529 18530 Delete the ChangeLog. The git history is just as good. 18531 18532commit eda3bac20dce2d5584ac07cc6b28a08ec55aa798 18533Author: Aaron Plattner <aplattner@nvidia.com> 18534Date: Wed May 24 15:05:22 2006 -0700 18535 18536 Derive the module version numbers from the package version. This idea was 18537 lifted from Michel Dänzer's similar code in xf86-video-ati. 18538 18539 Also, make the riva and nv driver versions match. 18540 18541commit 1a40c4a79b79034d191496726a27af8b0805dba6 18542Author: Aaron Plattner <aplattner@nvidia.com> 18543Date: Thu Apr 27 18:10:37 2006 -0700 18544 18545 Add the libcwrapper back, conditionalized on the server version. 18546 18547commit 425e30d58b7b62fff01fbb58585ee69b2df6fec1 18548Author: Aaron Plattner <aplattner@nvidia.com> 18549Date: Thu Apr 27 10:32:10 2006 -0700 18550 18551 Determine whether PutImage hooks take a DrawablePtr by server version. Remove 18552 dependency on xorg-server >= 1.0.99.901. 18553 18554commit c34a93265cb42168376ebdf0746b3ccded9f126c 18555Author: Aaron Plattner <aplattner@nvidia.com> 18556Date: Tue Apr 25 14:38:43 2006 -0700 18557 18558 Bump to 1.1.2 for new product names. 18559 18560commit f70247ade9eac8fa6115889662abb8e678a182dc 18561Author: Aaron Plattner <aplattner@nvidia.com> 18562Date: Tue Apr 25 14:29:44 2006 -0700 18563 18564 Add names for Quadro FX 5500, 3500, 1500, 560, 550, and 350. 18565 18566commit ca4c8af624e87e5c6a2927092a9c230ae0e60c0e 18567Author: Aaron Plattner <aplattner@nvidia.com> 18568Date: Wed Apr 12 02:04:52 2006 -0700 18569 18570 Add .gitignores. 18571 18572commit fb3faeb6073cef5a6a1b094f3d3bf020ceba9db6 18573Author: Adam Jackson <ajax@nwnk.net> 18574Date: Sat Apr 8 01:38:33 2006 +0000 18575 18576 Bump to 1.1.1 for Xv changes. 18577 18578commit f793182898c4d844e9861a60a945f53ab7071adc 18579Author: Aaron Plattner <aplattner@nvidia.com> 18580Date: Fri Apr 7 23:11:06 2006 +0000 18581 18582 Add a DrawablePtr argument to the XV functions to pave the way for 18583 redirected video. 18584 18585commit cb6b4073df7404822e2b553519c3a101f2ab0524 18586Author: Adam Jackson <ajax@nwnk.net> 18587Date: Fri Apr 7 19:40:34 2006 +0000 18588 18589 Unlibcwrap. Bump server version requirement. Bump to 1.1.0. 18590 18591commit 3f140052f891a4dc674b82e438c9595e59b4676b 18592Author: Aaron Plattner <aplattner@nvidia.com> 18593Date: Mon Mar 27 21:40:24 2006 +0000 18594 18595 Bump version to 1.0.2.0 for release. 18596 18597commit 84c007b426d321786f9aec1d9fda14b5baa35561 18598Author: Aaron Plattner <aplattner@nvidia.com> 18599Date: Fri Mar 10 02:29:03 2006 +0000 18600 18601 Coverity nos. 868 and 869: Free the result of xf86GetEntityInfo. 18602 18603commit 61e237eec7193701ce1236e67ebdfe7465b5db8a 18604Author: Aaron Plattner <aplattner@nvidia.com> 18605Date: Tue Jan 24 16:45:29 2006 +0000 18606 18607 Fix typo in PCI ID (reported by Stefan Dirsch). Patch from Mark Vojkovich 18608 to XFree86 CVS on 2006-01-23. 18609 18610commit 5064f8a903923ca4b819bdf345f8f4c9254bdf3b 18611Author: Aaron Plattner <aplattner@nvidia.com> 18612Date: Mon Jan 23 00:04:41 2006 +0000 18613 18614 New chip support from Mark Vojkovich's 2006-01-20 and 2006-01-21 XFree86 18615 CVS updates. 18616 18617commit e8fd45625f2e09830136bde6bc70d51f25892450 18618Author: Aaron Plattner <aplattner@nvidia.com> 18619Date: Mon Jan 2 01:54:19 2006 +0000 18620 18621 Preliminary RandR rotation support. Bug #4708 18622 <https://bugs.freedesktop.org/show_bug.cgi?id=4708> 18623 18624commit 5ec53e759c5233a26a6c80901b9f31b51153f756 18625Author: Kevin E Martin <kem@redhat.com> 18626Date: Wed Dec 21 02:30:01 2005 +0000 18627 18628 Update package version for X11R7 release. 18629 18630commit b0c8ca8c03c8ad1648b4597f7825bac7531100a7 18631Author: Adam Jackson <ajax@nwnk.net> 18632Date: Mon Dec 19 16:25:54 2005 +0000 18633 18634 Stub COPYING files 18635 18636commit 4fd25919c2a217d355419efdfbf0fe148af6c146 18637Author: Kevin E Martin <kem@redhat.com> 18638Date: Thu Dec 15 00:24:18 2005 +0000 18639 18640 Update package version number for final X11R7 release candidate. 18641 18642commit e4194bab2a790af10581ea2842717623db8998a2 18643Author: Kevin E Martin <kem@redhat.com> 18644Date: Tue Dec 6 22:48:36 2005 +0000 18645 18646 Change *man_SOURCES ==> *man_PRE to fix autotools warnings. 18647 18648commit baf60566b505fdf4387ec25789584b3f6ad72ce8 18649Author: Kevin E Martin <kem@redhat.com> 18650Date: Sat Dec 3 05:49:34 2005 +0000 18651 18652 Update package version number for X11R7 RC3 release. 18653 18654commit 4df200ea93b0a072f29057eeaf7a09e5a3a1dfe2 18655Author: Kevin E Martin <kem@redhat.com> 18656Date: Fri Dec 2 02:16:09 2005 +0000 18657 18658 Remove extraneous AC_MSG_RESULT. 18659 18660commit d4d5d7325fab1ea81f71d1ffc1fbbcc2d2253a65 18661Author: Adam Jackson <ajax@nwnk.net> 18662Date: Tue Nov 29 23:30:00 2005 +0000 18663 18664 Only build dlloader modules by default. 18665 18666commit d76f3f776bdab920d050542da45ec2871d4b84e6 18667Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18668Date: Mon Nov 28 22:04:10 2005 +0000 18669 18670 Change *mandir targets to use new *_MAN_DIR variables set by xorg-macros.m4 18671 update to fix bug #5167 (Linux prefers *.1x man pages in man1 subdir) 18672 18673commit d4e306b585f737b4c1287c520470425b0c11dc32 18674Author: Kevin E Martin <kem@redhat.com> 18675Date: Wed Nov 9 21:15:19 2005 +0000 18676 18677 Update package version number for X11R7 RC2 release. 18678 18679commit dde6dd1c7f95f9dccabfd919ad4ced795fe3fbd1 18680Author: Kevin E Martin <kem@redhat.com> 18681Date: Tue Nov 1 15:08:53 2005 +0000 18682 18683 Update pkgcheck depedencies to work with separate build roots. 18684 18685commit 92b1bfcc8fd0f34e636a1e9955d5236334f78cd3 18686Author: Kevin E Martin <kem@redhat.com> 18687Date: Wed Oct 19 02:48:03 2005 +0000 18688 18689 Update package version number for RC1 release. 18690 18691commit eda1cbc29cb0b2193a5bfb0eb620c740d9aef7c6 18692Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18693Date: Tue Oct 18 00:01:53 2005 +0000 18694 18695 Use @DRIVER_MAN_SUFFIX@ instead of $(DRIVER_MAN_SUFFIX) in macro 18696 substitutions to work better with BSD make 18697 18698commit 6c58a05e405cbc87fbe0497123d9b82d2480bdd3 18699Author: Adam Jackson <ajax@nwnk.net> 18700Date: Mon Oct 17 22:57:34 2005 +0000 18701 18702 More 1.7 braindamage: define EXTRA_DIST in terms of @DRIVER_NAME@ instead 18703 of indirectly 18704 18705commit fc20886c28c465578e6bd3ef2d9fe62bd993382b 18706Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18707Date: Mon Oct 17 00:09:04 2005 +0000 18708 18709 Use sed & cpp to substitute variables in driver man pages 18710 18711commit 1e3c84f332b7b9a4b320b370b8b2ecafbbdd8f8a 18712Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 18713Date: Tue Oct 4 21:36:51 2005 +0000 18714 18715 Link nvidida readme file into nvidia driver, Exclude duplicated neomagic 18716 readme 18717 18718commit d8be62ab175a013f78d4944036da284ffad34d96 18719Author: Aaron Plattner <aplattner@nvidia.com> 18720Date: Thu Sep 29 21:47:29 2005 +0000 18721 18722 Add entries for GeForce Go 7800. (Change went into XFree86 on 2005-09-28). 18723 Don't hang if j is zero. This should never happen, but it's better to be 18724 safe than sorry. 18725 18726commit e39eb8b8c1a79d04f90a195351c4eec3dba0d612 18727Author: Aaron Plattner <aplattner@nvidia.com> 18728Date: Wed Sep 28 17:18:02 2005 +0000 18729 18730 Bug #3333 <https://bugs.freedesktop.org/show_bug.cgi?id=3333> Patch #3417 18731 <https://bugs.freedesktop.org/attachment.cgi?id=3417> 18732 Fix a hang during screen-to-screen blits on certain GeForce 6xxx and Quadro 18733 FX 540 graphics cards. 18734 18735commit dc5acc2e477edb85f468f8d46862ca138426cfd9 18736Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18737Date: Fri Sep 23 16:25:28 2005 +0000 18738 18739 Bug #4558 <https://bugs.freedesktop.org/show_bug.cgi?id=4558> Patch #3551 18740 <https://bugs.freedesktop.org/attachment.cgi?id=3381> 18741 22-Sep-2005 nv driver update from Mark Vojkovich: Fix possible cause of 18742 some acceleration instability on some GeForce6xxx parts. 18743 18744commit b0ea43a51d2a6983559eecdb7dbb73c5e585697c 18745Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18746Date: Sat Sep 17 01:37:18 2005 +0000 18747 18748 Bugzilla #4480 <https://bugs.freedesktop.org/show_bug.cgi?id=4480> Patch 18749 #3305 <https://bugs.freedesktop.org/attachment.cgi?id=3305> 18750 13-Sep-2005 nv driver update from Mark Vojkovich: 18751 - Adds PCI ids for GeForce 6800 XT cards 18752 - Adds names for existing PCI ids for some GeForce 6xxx & 7xxx cards 18753 - Fix a potential problem with pixmap cache corruption on GeForce 6xxx and 18754 7xxx parts. (Mark Vojkovich, NVIDIA) 18755 18756commit cdf20033e2cfc9e8634876dd42613f7719340a5a 18757Author: Eric Anholt <anholt@freebsd.org> 18758Date: Fri Sep 16 01:36:45 2005 +0000 18759 18760 Put the riva_* files in the riva128.so subdriver, like monolithic. 18761 18762commit 9b1b3184cd586f6f8bf65ebd6f3c1afb1ae64d64 18763Author: Daniel Stone <daniel@fooishbar.org> 18764Date: Thu Aug 18 09:03:47 2005 +0000 18765 18766 Update autogen.sh to one that does objdir != srcdir. 18767 18768commit 75a0deb6e9d1a5bc698ea67b338c56bc99c1b2c0 18769Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18770Date: Fri Aug 5 03:52:28 2005 +0000 18771 18772 Bug #3987 <https://bugs.freedesktop.org/show_bug.cgi?id=3987> Patch #3264 18773 <https://bugs.freedesktop.org/attachment.cgi?id=3264> Workaround for 18774 512 Meg video cards (can only map 256 Meg of it). (Mark Vojkovich) 18775 18776commit fb4033bd2662c199d7e4f0e3df7c4a0fb9870984 18777Author: Kevin E Martin <kem@redhat.com> 18778Date: Fri Jul 29 21:22:44 2005 +0000 18779 18780 Various changes preparing packages for RC0: 18781 - Verify and update package version numbers as needed 18782 - Implement versioning scheme 18783 - Change bug address to point to bugzilla bug entry form 18784 - Disable loadable i18n in libX11 by default (use --enable-loadable-i18n to 18785 reenable it) 18786 - Fix makedepend to use pkgconfig and pass distcheck 18787 - Update build script to build macros first 18788 - Update modular Xorg version 18789 18790commit 6ea22f19210448605b9a18bf500c33168a77e77c 18791Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 18792Date: Tue Jul 26 18:40:29 2005 +0000 18793 18794 Build system for nv 18795 18796commit 74772a7dc49e34229497441fbbf2631bfb259b39 18797Author: Adam Jackson <ajax@nwnk.net> 18798Date: Mon Jul 11 02:29:58 2005 +0000 18799 18800 Prep for modular builds by adding guarded #include "config.h" everywhere. 18801 18802commit 80ac36ec4fbbd9a4aae1bde01fe6ebc1088fa8cb 18803Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18804Date: Sat Jul 9 16:51:58 2005 +0000 18805 18806 Bug #3740 <https://bugs.freedesktop.org/show_bug.cgi?id=3740> Patch #3058 18807 <https://bugs.freedesktop.org/attachment.cgi?id=3058> 18808 08-Jul-2005 nv driver updates from Mark Vojkovich: 18809 Change some console restore code for NV11. Hopefully, we can more reliably 18810 restore the console for desktop systems using DVI. This may correct a 18811 recent regression on NV11. 18812 Also, new PCI IDs. 18813 Add new nVidia PCI ids to match nv_driver.c changes. 18814 18815commit 6b75bc13462f1cfb735fe1de359ed9e45b5f4415 18816Author: Daniel Stone <daniel@fooishbar.org> 18817Date: Fri Jul 1 22:43:12 2005 +0000 18818 18819 Change all misc.h and os.h references to <X11/foo.h>. 18820 18821commit 270f5269a08da8ed0186e8a2223b673e0433a771 18822Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18823Date: Wed Jun 29 15:56:23 2005 +0000 18824 18825 Bugzilla #3657 <https://bugs.freedesktop.org/show_bug.cgi?id=3657> Patch 18826 #2992 <https://bugs.freedesktop.org/attachment.cgi?id=2992> 18827 Experimental DPMS support for GeForce4 and later laptops. (Mark 18828 Vojkovich) 18829 18830commit bf73c0716e9e4be0540d2ee7340bd5833831a434 18831Author: Adam Jackson <ajax@nwnk.net> 18832Date: Sun Jun 26 04:38:12 2005 +0000 18833 18834 Bug #3628: Fix video driver submodule visibility. 18835 18836commit 175766a44ceb0a5d4226e9088d8e3e6d73815b4f 18837Author: Adam Jackson <ajax@nwnk.net> 18838Date: Sat Jun 25 21:16:55 2005 +0000 18839 18840 Bug #3626: _X_EXPORT tags for video and input drivers. 18841 18842commit 8759399a3744f0eb1e0cdc83ed23aaaafb9be444 18843Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18844Date: Thu Apr 21 22:25:47 2005 +0000 18845 18846 programs/Xserver/hw/xfree86/drivers/nv/nv_driver.c 18847 programs/Xserver/hw/xfree86/drivers/nv/nv_setup.c 18848 //bugs.freedesktop.org/show_bug.cgi?id=3095> Patch #2501 18849 <https://bugs.freedesktop.org/attachment.cgi?id=2501> nv DPMS panel 18850 fixes (Mark Vojkovich) 18851 18852commit 2db214d7cf5336fbea36d83bc9b089f70a494614 18853Author: Daniel Stone <daniel@fooishbar.org> 18854Date: Wed Apr 20 12:25:29 2005 +0000 18855 18856 Fix includes right throughout the Xserver tree: 18857 change "foo.h" to <X11/foo.h> for core headers, e.g. X.h, Xpoll.h; 18858 change "foo.h", "extensions/foo.h" and "X11/foo.h" to 18859 <X11/extensions/foo.h> for extension headers, e.g. Xv.h; 18860 change "foo.[ch]" to <X11/Xtrans/foo.[ch]> for Xtrans files. 18861 18862commit 3fe983d254f1ffe3c285548013be677ed4a67137 18863Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18864Date: Mon Feb 14 03:37:20 2005 +0000 18865 18866 Bugzilla #2533 (https://bugs.freedesktop.org/show_bug.cgi?id=2533) Feb. 3, 18867 2005 "nv" driver update from Mark Vojkovich (Mainly driver updates for 18868 nVidia cards with ((pNv->Chipset & 0xfff0) == 0x0090) ) 18869 18870commit 8c95fdc889b56a76497f5194bafd06edfc962387 18871Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18872Date: Wed Jan 26 01:36:04 2005 +0000 18873 18874 Jan. 2005 nv chipset update from Mark Vojkovich Bug #2380 18875 <https://bugs.freedesktop.org/show_bug.cgi?id=2380> Patch #1752 18876 <https://bugs.freedesktop.org/attachment.cgi?id=1752> 18877 18878commit 67c89797f62ea53b37927c64e0403e7e7c42de56 18879Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18880Date: Fri Dec 10 03:56:42 2004 +0000 18881 18882 Bugzilla #1985 <https://bugs.freedesktop.org/show_bug.cgi?id=1985> sync to 18883 12/08/04 version of Nvidia driver from Mark Vojkovich, including these 18884 fixes since last sync: Fix some DAC/Graphics memory contention issues 18885 on newer NVIDIA chips (specifically, NV40). More PCI IDs. Also, I've 18886 had trouble getting some panels to work automagically so I expose an 18887 FPTweak option to let the user adjust a troublesome register. 18888 18889commit 7c34d051168eba499beb20c40fd95703375f4c7b 18890Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18891Date: Wed Nov 10 02:53:44 2004 +0000 18892 18893 (Bugzilla #1777) Import Mark Vojkovich's changes to NV driver since last 18894 sync with XFree86, including: Support for new NVIDIA chips. Egbert has 18895 a laptop that needs some DPMS ordering reversed. New chip support and 18896 workaround for Toshiba M30 laptop issue. PCI-Xpress chip support. DPMS 18897 support for DVI. Cursor changes for NV11 so that alpha blended cursors 18898 will work in conjunction with dithering (laptop panels). It previously 18899 did not. HW support for alpha blended stretch blits on NV5 and higher, 18900 though nothing is using them at this point. Add support for VBlank 18901 syncing the Xv Blit adaptor on hardware that supports it (NV11 and 18902 higher). Make console restoration a little more robust. We were making 18903 an assumption that failed on a newer laptop. Workaround for some NV31 18904 DVI artifacts. 18905 Also add nVidia license statement to nv man page to comply with requirement 18906 that it appear in "user documentation" 18907 18908commit fcd6d7b57edb4e396189a0a2683bcf08974f3b59 18909Author: Egbert Eich <eich@suse.de> 18910Date: Tue Oct 12 19:13:44 2004 +0000 18911 18912 Set fbdev mode as the default mode on PPC (Olaf Hering). 18913 Added support for IBM space saver keyboard (Stefan Dirsch). Added support 18914 for Cherry CyMotion Master XPress (Marcus Schaefer). 18915 Change order of SetDisplayDevice(), HWRestore(), UnbindGART() and 18916 RestoreBIOSMemSize() to be exactly opposite to the Save procedure in 18917 EnterVT() (Matthias Hopf, Alan Hourihane). 18918 Fix text mode restauration by removing the assumption that the register 18919 which determines which head is programmed is set. to the active head by 18920 the BIOS (Mark Vojkovich). 18921 When I wrote the resource code 5 years ago I made some assumptions which 18922 turned out to be false: I've assumed that the bus number of the PCI 18923 hostbridge would be the PCI bus the bridge links to. This is not 18924 correct. Fixing this assumption is not easy. However I hope that the 18925 attached patch will make the system work regardless as it 'ignores' 18926 host bridges from which the target bus is not known. This should not 18927 matter at all as we really don't care about host bridges (unless we 18928 have bridge specific code which retrieves information about the 18929 bridge). 18930 Fixed server crash on reset when a structure allocated in PreInit() was 18931 freed on CloseScreen(). 18932 Fixed ring buffer lock ups that happened because the structure that 18933 contained ringbuffer data was not zeroed after allocation. 18934 Fixed numerous warnings due to signed unsigned comparisons. 18935 programs/Xserver/hw/xfree86/drivers/nv/nv_driver.c: 18936 (NVBacklightEnable): Changed the order in which the sequencer registers 18937 and the backlight control registers are written. The sequencer control 18938 register need to be written first otherwise DPMS screen blanking 18939 produces vertical bars on a mobile device. lib/Xau/Imakefile: Build 18940 libXau static library PIC so it can be linked into toolkits that would 18941 like to wrap its functionality. 18942 18943commit 65e13fc61141b36a74c7a0e696d274574641ba20 18944Author: Adam Jackson <ajax@nwnk.net> 18945Date: Fri Sep 17 03:04:52 2004 +0000 18946 18947 Bug #1192: Remove cfb support from drivers where its use is an option. 18948 Delete xf24_32bpp, as s3virge was the last user. Fix up some comments 18949 to refer to fb rather than cfb. 18950 18951commit bbafc6d2fcb21671de3cfb4ff973409405162e7a 18952Author: Adam Jackson <ajax@nwnk.net> 18953Date: Mon Aug 16 09:13:14 2004 +0000 18954 18955 Bug #1087: Make sure all the *Weak symbols are documented in the import 18956 list, so the loader knows not to complain about unresolved symbols when 18957 an optional module isn't loaded. This typically manifests as warnings 18958 about fbdevHW symbols when the user isn't using the framebuffer 18959 console. 18960 18961commit 8b6de7d388b7ae5169b2e38ad1a40981bc406cfc 18962Author: Adam Jackson <ajax@nwnk.net> 18963Date: Sat Jul 31 01:21:19 2004 +0000 18964 18965 Change several LoaderSymbol calls introduced by the bug #400 patch to 18966 *Weak() resolver functions. 18967 18968commit f745157f15befebe01552b103c573ccb94b8f388 18969Author: Adam Jackson <ajax@nwnk.net> 18970Date: Fri Jul 30 20:30:53 2004 +0000 18971 18972 Bug #400 (partial): Driver fixes for the dlloader. When using dlloader, all 18973 framebuffer formats except cfb and the overlay modes should work, and 18974 r128 and radeon need to be loaded from the ati driver (both issues to 18975 be fixed soon). Tested on i740, s3virge, mach64, tdfx, vesa, and vga 18976 drivers. elfloader users shouldn't be affected. 18977 18978commit b9a6776bc04eb19343882bddb371b705677cfc14 18979Author: Eric Anholt <anholt@freebsd.org> 18980Date: Wed Jun 16 09:23:29 2004 +0000 18981 18982 DRI XFree86-4_3_99_12-merge import 18983 18984commit cbf040deb0ab24425468ed60b3f3edd322ec210f 18985Author: Eric Anholt <anholt@freebsd.org> 18986Date: Wed Jun 16 09:23:29 2004 +0000 18987 18988 Initial revision 18989 18990commit 99a7f7ccce4e012bde57021cafcb132c3d599b96 18991Author: Egbert Eich <eich@suse.de> 18992Date: Fri Apr 23 19:42:10 2004 +0000 18993 18994 Merging XORG-CURRENT into trunk 18995 18996commit 708bb2eb0db3808e3edf94a5f22b3e3f5bb6488d 18997Author: Egbert Eich <eich@suse.de> 18998Date: Sun Mar 14 08:33:30 2004 +0000 18999 19000 Importing vendor version xf86-4_4_99_1 on Sun Mar 14 00:26:39 PST 2004 19001 19002commit fe564263d97e370817fc862e5c8ead09bfb70b58 19003Author: Egbert Eich <eich@suse.de> 19004Date: Wed Mar 3 12:12:23 2004 +0000 19005 19006 Importing vendor version xf86-4_4_0 on Wed Mar 3 04:09:24 PST 2004 19007 19008commit 1a0be799ce4c479e39f8af86fcca3d192e83f455 19009Author: Egbert Eich <eich@suse.de> 19010Date: Thu Feb 26 13:35:54 2004 +0000 19011 19012 readding XFree86's cvs IDs 19013 19014commit ee5c9ef51cea88a630ecc652512603c42b39a533 19015Author: Egbert Eich <eich@suse.de> 19016Date: Thu Feb 26 09:23:23 2004 +0000 19017 19018 Importing vendor version xf86-4_3_99_903 on Wed Feb 26 01:21:00 PST 2004 19019 19020commit d176fa338d2d83bcf8d7ec68ff1aec205a258fec 19021Author: Kaleb Keithley <kaleb@freedesktop.org> 19022Date: Mon Feb 23 20:35:04 2004 +0000 19023 19024 Import most of XFree86 4.4RC3. This import excludes files which have the 19025 new license. If we want to, later we can import 4.4RC3 again and pick 19026 up the files that have the new license, but for now the vendor branch 19027 is "pure." 19028 19029commit ea7a90e680a2e9259e471dd6677bcca12aeea79e 19030Author: Egbert Eich <eich@suse.de> 19031Date: Thu Jan 29 08:08:36 2004 +0000 19032 19033 Importing vendor version xf86-012804-2330 on Thu Jan 29 00:06:33 PST 2004 19034 19035commit a6c83bd98c34aacdf7a6b82ea83d6ae775479502 19036Author: Kaleb Keithley <kaleb@freedesktop.org> 19037Date: Fri Dec 19 20:55:14 2003 +0000 19038 19039 XFree86 4.3.99.902 (RC 2) 19040 19041commit 935c2adf58a10e6d3caa75823b4e7a0689555c49 19042Author: Kaleb Keithley <kaleb@freedesktop.org> 19043Date: Tue Nov 25 19:28:38 2003 +0000 19044 19045 XFree86 4.3.99.16 Bring the tree up to date for the Cygwin folks 19046 19047commit d84248580d5b05bb47a2ccdce7badfc884efa6da 19048Author: Kaleb Keithley <kaleb@freedesktop.org> 19049Date: Tue Nov 25 19:28:38 2003 +0000 19050 19051 Initial revision 19052 19053commit dd1a62caaaac6acc35584a2c8c525962d8574c95 19054Author: Kaleb Keithley <kaleb@freedesktop.org> 19055Date: Fri Nov 14 16:48:55 2003 +0000 19056 19057 XFree86 4.3.0.1 19058 19059commit 2f223903fba2bdee1623f3442d7580c809b428cc 19060Author: Kaleb Keithley <kaleb@freedesktop.org> 19061Date: Fri Nov 14 16:48:55 2003 +0000 19062 19063 Initial revision 19064 19065commit 542d57ea12b5461891a13a339e01ca9e1add124d 19066Author: Kaleb Keithley <kaleb@freedesktop.org> 19067Date: Fri Nov 14 15:54:48 2003 +0000 19068 19069 R6.6 is the Xorg base-line 19070