ChangeLog revision cd34e0e1
1commit a7c190e0c230ed2d5e047ed6aba12d71a4866950 2Author: Lyude <lyude@redhat.com> 3Date: Fri Apr 21 14:41:17 2017 -0400 4 5 Bump version to 1.0.15 6 7 Signed-off-by: Lyude <lyude@redhat.com> 8 9commit 3047e99a641d2b8720371aae292750206f5e91fc 10Author: Ben Skeggs <bskeggs@redhat.com> 11Date: Thu Apr 20 15:57:33 2017 +1000 12 13 fix null pointer deref when building against >=libdrm 2.4.78 14 15 A new pointer got added to drmEventContext. 16 17 As a result of us both: 18 - Setting drmEventContext.version to "latest" AND 19 - Not zeroing the struct 20 21 We end up thwarting libdrm's compatibility checks, resulting in 22 libdrm choosing to call through an invalid pointer. 23 24 Fix this by zeroing the struct. 25 26 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 27 28commit e9418e434311336e905b70553a5ed740838d90ad 29Author: Mariusz Bialonczyk <manio@skyboo.net> 30Date: Wed Mar 29 22:07:50 2017 +0200 31 32 Do not register hotplug without RandR 33 34 When using Xinerama, RandR is automatically disabled, and calling RR 35 routines will trigger an assert() because the RR keys/resources are 36 not set, leading to an Xserver abort. 37 38 Hotplug makes little sense without RandR, so no need to install a 39 udev monitor if RandR is not available. 40 41 Ported from xf86-video-intel commit 42 1a489142c8e6a4828348cc9afbd0f430d3b1e2d8, original work by: 43 Chris Wilson <chris@chris-wilson.co.uk> 44 45 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98383 46 Signed-off-by: Mariusz Bialonczyk <manio@skyboo.net> 47 Acked-by: Ilia Mirkin <imirkin@alum.mit.edu> 48 49commit e472b47d15634a864c8c981ed588d882aceaf26b 50Author: Ilia Mirkin <imirkin@alum.mit.edu> 51Date: Tue Mar 21 21:25:33 2017 -0400 52 53 Add Pascal family support, identical to Maxwell 54 55 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 56 Tested-by: Lyude <lyude@redhat.com> 57 58commit b71de83b7fae0abeb311251e6144294d319062cf 59Author: Lyude <lyude@redhat.com> 60Date: Tue Mar 7 18:44:43 2017 -0500 61 62 Bump version to 1.0.14 63 64 Signed-off-by: Lyude <lyude@redhat.com> 65 66commit 924083938c8f209d8f6ff472caf8692a644f7e78 67Author: Lyude <lyude@redhat.com> 68Date: Fri Mar 3 18:27:42 2017 -0500 69 70 Consider CRTCs disabled when DPMS is off 71 72 It turns out there's a difference in X between a CRTC being "disabled" 73 and simply having it's DPMS turned off. This is problematic though, 74 because if DPMS is turned off you can't really use the CRTC as a normal 75 CRTC anyway since page flipping and vblanks will be non-functional. As a 76 result, we've been considering DPMS-on CRTCs as enabled and attempt to 77 perform pageflips, vblank waits, etc. on them which inevitably fails. and 78 usually breaks the display the first time any of the CRTCs have their 79 DPMS turned on. 80 81 This was a problem that didn't really show itself until kernel 4.10 when 82 atomic modesetting was added which caused nouveau to stop trying to 83 fulfill pageflips and vblank waits on disabled CRTCs. I'm not sure how 84 pageflipping disabled CRTCs ever worked in the first place, but since 85 not doing so is the proper behavior anyway I haven't investigated any 86 further. 87 88 So, copy the ms_crtc_on() function from the modesetting driver and add 89 it here as drmmode_crtc_on(), then use that in all of the places where 90 we should be checking for both DPMS off and disabled CRTCs. 91 92 This fixes issues with the X ceasing to function (usually) after the 93 first time a CRTC has it's DPMS turned on. Reproduction recipe: 94 95 - Load up gnome-shell on a machine 96 - Wait for the display to timeout from inactivity and turn itself off 97 - Shake the cursor or press something on the keyboard. Chances are the 98 monitor will come back on, but the display remains black until the 99 next time the X server is restarted. 100 101 Signed-off-by: Lyude <lyude@redhat.com> 102 Reviewed-by: Adam Jackson <ajax@redhat.com> 103 104commit 1516d35b06c9cda399bef01d992805d6e63dcbae 105Author: Ilia Mirkin <imirkin@alum.mit.edu> 106Date: Sun Oct 16 15:10:10 2016 -0400 107 108 recognize and accelerate GM20x 109 110 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 111 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 112 113commit b00b73c3aa0da1d4cee5c9f580ca65a7bd344e0f 114Author: Ilia Mirkin <imirkin@alum.mit.edu> 115Date: Mon Oct 17 23:23:41 2016 -0400 116 117 copy: add maxwell/pascal copy engine classes 118 119 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 120 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 121 122commit 5a3ada3b0ff3d89441faf35d331bb8b6cf5e51c0 123Author: Ilia Mirkin <imirkin@alum.mit.edu> 124Date: Sun Oct 16 15:03:35 2016 -0400 125 126 nvc0: refactor TIC uploads to allow different specifics per generation 127 128 This flips GM10x to using the updated format, which is what I tested 129 with. However GM20x and GP10x also use this TIC format. 130 131 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 132 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 133 134commit 514441f32c5b948ce87668234ab9f01e8ec28dac 135Author: Ilia Mirkin <imirkin@alum.mit.edu> 136Date: Sun Oct 16 12:42:18 2016 -0400 137 138 nvc0: rename BEGIN_IMC0 to IMMED_NVC0 139 140 For consistency with mesa. It wasn't used anywhere previously. 141 142 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 143 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 144 145commit 4459c04a6bc208924fac8082920f6786538bc5cd 146Author: Ilia Mirkin <imirkin@alum.mit.edu> 147Date: Sun Oct 16 12:38:38 2016 -0400 148 149 nvc0: make use of the new hwdefs for TEX_CB_INDEX 150 151 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 152 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 153 154commit ea6fd12ea059c6af832bd2180b26d75e00d4ea71 155Author: Ilia Mirkin <imirkin@alum.mit.edu> 156Date: Sun Oct 16 12:38:02 2016 -0400 157 158 hwdefs: update nvc0_3d, add gm107_texture for new TIC format 159 160 These are copied directly from the mesa repository. 161 162 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 163 Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 164 165commit a24ded627e42d453546c553d609edc073f59cd57 166Author: Ilia Mirkin <imirkin@alum.mit.edu> 167Date: Thu Sep 25 20:15:51 2014 -0400 168 169 exa: add GM10x acceleration support 170 171 rendercheck -f a8r8g8b8 passes as much as on a GK208, and xv appears to 172 work. Very lightly tested. 173 174 Instead of sticking coordinates into pushbufs, the vertex shader is 175 modified to read them from a constbuf, indexed by vertex id. This 176 approach could be used for all nvc0 generations, but I didn't want to 177 rock the boat. 178 179 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 180 Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 181 182commit e6479845ec0db20dc733c621b7967b751840a552 183Author: Ilia Mirkin <imirkin@alum.mit.edu> 184Date: Tue Sep 20 00:31:06 2016 -0400 185 186 Bump version to 1.0.13 187 188 - Fixes for reverse prime offload 189 - ABI 23 support for Xorg 1.19 190 191 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu 192 193commit 6473b68762b0dca2dfccfdfc74100398b7459296 194Author: Keith Packard <keithp@keithp.com> 195Date: Tue Jul 19 08:29:40 2016 -0700 196 197 Use NotifyFd for drm and udev fds 198 199 NotifyFd is available after API 22, and must be used after API 23. 200 201 Signed-off-by: Keith Packard <keithp@keithp.com> 202 203commit a964931e8dc4e06c27749edb4993a656dc261d75 204Author: Adam Jackson <ajax@redhat.com> 205Date: Tue Jul 19 10:03:56 2016 -0400 206 207 Adapt Block/WakeupHandler signature for ABI 23 208 209 Signed-off-by: Adam Jackson <ajax@redhat.com> 210 211commit 12f77348e7a9579b167b41228dec9e6f97b74de8 212Author: Hans de Goede <hdegoede@redhat.com> 213Date: Fri Jun 3 14:46:10 2016 +0200 214 215 Properly cleanup fb for reverse-prime-offload 216 217 drmmode_set_scanout_pixmap(pix) adds drmmod->fb_id through a call 218 to drmmode_xf86crtc_resize(), but on a subsequent 219 drmmode_set_scanout_pixmap(NULL) it would not remove the fb. 220 221 This keeps the crtc marked as busy, which causes the dgpu to not 222 being able to runtime suspend, after an output attached to the dgpu 223 has been used once. Which causes burning through an additional 10W 224 of power and the laptop to run quite hot. 225 226 This commit adds the missing remove fb call, allowing the dgpu to runtime 227 suspend after an external monitor has been plugged into the laptop. 228 229 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 230 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 231 232commit 1da8a937be19e41c51a3d516bd98cee988bca44b 233Author: Ben Skeggs <bskeggs@redhat.com> 234Date: Thu Jun 2 11:23:53 2016 +1000 235 236 exa/nv50-: fix some potential incomplete pushes 237 238 I don't think these should actually be able to happen, given that we 239 already reject the unknown formats in the Check() functions. But, 240 just in case... 241 242 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 243 244commit bb57f9af9d179af0962456dca369953521ea8c2c 245Author: Ben Skeggs <bskeggs@redhat.com> 246Date: Thu Jun 2 10:17:58 2016 +1000 247 248 fix use of out-of-scope data 249 250 Reported by Coverity. 251 252 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 253 254commit b824d36c28124955eda4aced5e637aa75eea4d6c 255Author: Ilia Mirkin <imirkin@alum.mit.edu> 256Date: Wed Jan 13 11:55:06 2016 -0500 257 258 HAS_DIRTYTRACKING_ROTATION also supports multiple CRTCs 259 260 A port of commit ff9a6b6f from xf86-video-ati. 261 262 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93670 263 Reported-by: Thomas D. <thfrdue@gmx.de> 264 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 265 266commit b18bc036bf9997ea65f5cbd824fd057931e604fe 267Author: Ben Skeggs <bskeggs@redhat.com> 268Date: Tue Dec 8 15:52:25 2015 +1000 269 270 Bump version to 1.0.12 271 272 Various random fixes have been pending for a long while now... 273 274 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 275 276commit 6e6d8ac1c7b4ee047a7b40b95dea1e65a7c3211a 277Author: Mario Kleiner <mario.kleiner.de@gmail.com> 278Date: Sun Jun 28 02:33:49 2015 +0200 279 280 Take shift in crtc positions for ZaphodHeads configs into account. 281 282 In multi-x-screen ZaphodHeads configurations, there isn't a 283 one-to-one mapping of kernel provided drmmode crtc index 284 to the index of the corresponding xf86Crtc inside the 285 xf86CrtcConfig crtc array anymore, ie. for kernel provided 286 drmmode->mode_res->crtcs[i], the i'th crtc won't correspond 287 to the xf86Crtc in the i'th slot of the x-screens xf86CrtcConfig 288 anymore, once ZaphodHeads has only selected a subset of all crtcs 289 of a graphics card for a given x-screen, instead of all crtcs. 290 291 This breaks the mapping of bit positions in the bit masks returned 292 in kencoder->possible_crtcs and kencoder->possible_clones. A 1 bit 293 in position i of those masks allows use of the kernels i'th crtc for 294 the given kencoder. The X-Servers dix code checks those bit masks 295 for valid xf86Output -> xf86Crtc assignments, assuming that the i'th 296 slot xf86CrtcConfigPtr config->crtc[i] corresponds to bit i in the 297 xf86Output->possibe_crtcs bitmask, and bails if the bitmask doesn't 298 allow the specified assignment of crtc to output. If ZaphodHeads 299 breaks the assumption of bit i <-> crtc slot i this ends in failure. 300 301 Take this shift of crtc index positions wrt. encoder bitmask bit 302 positions into account by bit-shifting positions accordingly when 303 assigning encoder->possible_crtcs to output->possible_crtcs, so 304 the proper indices match up again for validation by the dix. 305 306 This problem wasn't apparent last year when testing the ZaphodHeads 307 support on some Kepler cards, as apparently the encoder->possible_crtcs 308 bitmasks returned for those cards by the kernel just had all 4 309 lsb bits set for all tested encoders/output, so each of the cards 4 310 crtcs could go with each output and things worked by chance. 311 312 The current code breaks, e.g., on 2010 MacBookPro with nv50, where 313 one crtc is hardwired to the internal lvds panel, and one crtc 314 is hardwired to the external DP connector, resulting in a failure 315 where dual-display on single-x-screen works fine, but assigning 316 each output to a separate x-screen via ZaphodHeads fails due to 317 the mismatched encoder->possible_crtcs bitmasks. 318 319 This patch fixes the problem. 320 321 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 322 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 323 324commit 1ff13a922535924681b91452235b017e43a4c6f6 325Author: Ilia Mirkin <imirkin@alum.mit.edu> 326Date: Sun Sep 13 15:29:15 2015 -0400 327 328 fix build after glamor removal 329 330 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 331 332commit 3bb943b33bbe8e593afc00d13fe867dd7a708c3c 333Author: Ilia Mirkin <imirkin@alum.mit.edu> 334Date: Thu Aug 21 20:36:41 2014 -0400 335 336 xv: use correct max width/height settings 337 338 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=70931 339 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 340 341commit 3e2e0faa2ee1cce9c1bb5c7ad80d0592460f3edc 342Author: Ilia Mirkin <imirkin@alum.mit.edu> 343Date: Sat Jul 11 17:16:27 2015 -0400 344 345 remove maxwell GM10x support for now 346 347 There is no EXA acceleration, and now no GLAMOR either. 348 User is better off with modesetting until EXA makes an appearance. 349 350 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 351 352commit 1f8b2b348526d94e9bde4a285f331a4934c11078 353Author: Ilia Mirkin <imirkin@alum.mit.edu> 354Date: Sat Jul 11 17:08:52 2015 -0400 355 356 glamor: remove 357 358 If you want glamor, just use modesetting instead. It's hooked up 359 incorrectly in nouveau, and there doesn't seem to be any point in fixing 360 it when the modesetting driver does it all properly. 361 362 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 363 364commit 6296145654b78518f3299bb5887f224f0d3810fd 365Author: Mario Kleiner <mario.kleiner.de@gmail.com> 366Date: Wed Jul 29 14:39:02 2015 +0200 367 368 Add Option "DRI" to allow selection of maximum DRI level. (v2) 369 370 Allow user to select the maximum level of DRI implementation 371 to use, DRI2 or DRI3. 372 373 exa accel supports both DRI2 and, if the kernel supports 374 rendernodes, also DRI3. However, DRI3 still seems to have 375 some bugs on current implementations, and additionally it 376 doesn't work well at all for X-Servers older than 1.16.3 377 due to X-Server bugs. Therefore we default to DRI2 on exa, 378 but allow the user to enable DRI3 with this new option. 379 380 nouveau's glamor accel backend currently doesn't work under 381 DRI2 at all, so we continue to use DRI3 whenever it is 382 supported and ignore this new option for now. 383 384 Also add a bit more output about status of Present and 385 DRI3 to aid debugging. 386 387 Note: This was originally meant to be a boolean parameter, 388 to just select between DRI3 on and off, but changed 389 here to a DRI level to make it consistent with the 390 same option in the released Intel-ddx. 391 392 v2: Use fixed up Bool return type of nouveau_present_init(). 393 394 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 395 Cc: Ilia Mirkin <imirkin@alum.mit.edu> 396 Cc: Emil Velikov <emil.l.velikov@gmail.com> 397 Cc: Martin Peres <martin.peres@free.fr> 398 Cc: Ben Skeggs <bskeggs@redhat.com> 399 Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu> 400 401commit 0deb1fcc6d1099db698ff8c512a4884708a458ec 402Author: Mario Kleiner <mario.kleiner.de@gmail.com> 403Date: Wed Jul 29 14:39:01 2015 +0200 404 405 present: Fixup return type of nouveau_present_init() 406 407 Make it a Bool consistently, as declared in header. 408 409 Reported-by: Ilia Mirkin <imirkin@alum.mit.edu> 410 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 411 Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu> 412 413commit 5212dbd7e5eba35fdd6578b30f5b0c8c8543081b 414Author: Ilia Mirkin <imirkin@alum.mit.edu> 415Date: Tue Jul 14 19:23:01 2015 -0400 416 417 Avoid build fail without COMPOSITE 418 419 Reviewed-by: Dave Airlie <airlied@redhat.com> 420 421commit fdc528306e4018b314f2efcb2be4e37672bddd75 422Author: Vadim Rutkovsky <vrutkovs@redhat.com> 423Date: Thu Jul 9 23:23:37 2015 +0200 424 425 Adapt to new dirty tracking api 426 427 Mirrors commit b6d871bf2 (radeon: adopt for new X server dirty tracking 428 APIs.) in xf86-video-ati. 429 430 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91284 431 432commit dfd827cefb940f46b18dc77c129f0a603c7c95fa 433Author: Samuel Pitoiset <samuel.pitoiset@gmail.com> 434Date: Wed May 20 17:08:58 2015 +0200 435 436 fix a wrong use of a logical operator in drmmode_output_dpms() 437 438 This is probably a typo error which has been introduced in 2009... 439 This fixes the following warning detected by Clang : 440 441 drmmode_display.c:907:30: warning: use of logical '&&' with constant operand [-Wconstant-logical-operand] 442 if (props && (props->flags && DRM_MODE_PROP_ENUM)) { 443 444 Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> 445 Reviewed-by: Dave Airlie <airlied@redhat.com> 446 447commit 27234dbe00caf6dfd4efb24fc7bbc026fa0134fc 448Author: Ilia Mirkin <imirkin@alum.mit.edu> 449Date: Wed May 27 20:26:42 2015 -0400 450 451 nv50/exa: add comments showing the code that makes up the programs 452 453 As decoded by envydis. Perhaps these should move into separate files 454 similar to nvc0, but this seems like a simple intermediate solution. 455 456 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 457 458commit 08ef36dc6766080f173ae2b7e660b3182cab9f2d 459Author: Ilia Mirkin <imirkin@alum.mit.edu> 460Date: Wed May 20 18:36:17 2015 -0400 461 462 older hardware prefers copy to clear, so use that 463 464 Doing a solid copy of 0, or a clear, has the same effect. This is a 465 partial revert of commit 5439a9905eab8c. Pre-nv50 hardware succeeds with 466 copies more often. 467 468 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 469 470commit 278ad73475bd137eac8a49ec7a22406bfc2867e7 471Author: Ilia Mirkin <imirkin@alum.mit.edu> 472Date: Tue May 19 16:26:16 2015 -0400 473 474 nv04-nv40: don't attempt to do 32-bit shifts 475 476 A 32-bit shift is a no-op, which will also make the new planemask get 477 or'd with ~0, thus negating the usefulness of the subsequent 478 planemask != ~0 check. Only do this if it's a less-than-32-bit per pixel 479 format, in which case it will have the desired effect of setting the 480 high bits. 481 482 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 483 Reviewed-by: Martin Peres <martin.peres@free.fr> 484 485commit 7583471e1fe181dc7c26b7d0a0d16f70eb10dd21 486Author: Ilia Mirkin <imirkin@alum.mit.edu> 487Date: Tue May 19 16:11:17 2015 -0400 488 489 Check before trying a solid fill 490 491 Pre-nv50 has all sorts of funny requirements for non-copy alu 492 operations, and will bail out of solid fills left and right. Account for 493 that case and fall back to the memset. 494 495 Reported-by: Andrew Randrianasulu <randrianasulu@gmail.com> 496 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 497 Reviewed-by: Martin Peres <martin.peres@free.fr> 498 499commit 212fc5354ed60ba282d5eabf468c0e2dd0d0afe6 500Author: Ilia Mirkin <imirkin@alum.mit.edu> 501Date: Mon Mar 23 20:05:23 2015 -0400 502 503 nv50: explicitly disable ZETA buffer 504 505 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 506 507commit 734b8404c0a1fa9ce8ee62a0d4cc5baae7c8fcdf 508Author: Ilia Mirkin <imirkin@alum.mit.edu> 509Date: Sat Mar 21 11:46:35 2015 -0400 510 511 use defined method names where available 512 513 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 514 515commit 5439a9905eab8c8d0f86e9963df6c893633bdbfd 516Author: Maarten Lankhorst <dev@mblankhorst.nl> 517Date: Mon Mar 23 19:41:38 2015 +0100 518 519 Actually do a clear instead of doing a bogus copy. 520 521 Oops! 522 523commit f0fdc156039eaa79fb1ad82fcc6437f2432306c6 524Author: Maarten Lankhorst <dev@mblankhorst.nl> 525Date: Sat Mar 21 22:30:10 2015 +0100 526 527 Try using EXA ops harder in drmmode_fbcon_copy fallback 528 529 The memcpy in the fallback path fails with SIGBUS on GK20A because 530 this memory is not coherent, try to use a solid fill instead. 531 532commit 7319b54291d59e1cd3cbe4f39ea8cc6dc8bf45bc 533Author: Maarten Lankhorst <dev@mblankhorst.nl> 534Date: Fri Mar 20 23:24:50 2015 +0100 535 536 Simplify nouveau_allocate_surface's pitch calculation. 537 538 Judging from previous commits all linear buffers should have a 256 byte alignment. 539 540commit 91e51a7571e6e8cb9338fc168f817f56438b1901 541Author: Maarten Lankhorst <dev@mblankhorst.nl> 542Date: Fri Mar 20 23:17:13 2015 +0100 543 544 Use drmGetRenderDeviceNameFromFd to obtain the dri3 node. 545 546commit d29d8baad4d845ec9d3d1d1bc8d1c1e1f6f6f086 547Author: Maarten Lankhorst <dev@mblankhorst.nl> 548Date: Sat Mar 14 18:00:00 2015 +0100 549 550 Use the correct class and shader ops for GK20A. 551 552 Signed-off-by: Maarten Lankhorst <dev@mblankhorst.nl> 553 554commit 6829d4c672fc05f4d3d993ac1bfde21d4b6c1200 555Author: Maarten Lankhorst <dev@mblankhorst.nl> 556Date: Sat Mar 14 11:58:12 2015 +0100 557 558 Make the ddx load with platform devices 559 560 Signed-off-by: Maarten Lankhorst <dev@mblankhorst.nl> 561 562commit 3fb97d78202213c1221a1d3ad8a5cebda78fdf44 563Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 564Date: Thu Oct 23 09:29:14 2014 +0200 565 566 fix null pointer deref 567 568commit 90ba0bb6a371a5beebdf6d363bcd73659b7c9f1f 569Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 570Date: Thu Oct 23 08:25:01 2014 +0200 571 572 check memory allocation in nouveau_glamor_set_shared_pixmap_backing 573 574 As reported by Roy Spliet. 575 576commit 4dfa13c77a1f6cee1affea05b5f3a0b26827d366 577Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 578Date: Thu Oct 23 08:21:55 2014 +0200 579 580 rework initialization to prevent some crashes 581 582commit 485902e61b18aff864226db62ba10399aab40128 583Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 584Date: Wed Oct 22 15:52:22 2014 +0200 585 586 fix !DRI3 case 587 588commit 79d438ec4d23280a242d36c2d81f2f43f6e90ada 589Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 590Date: Wed Oct 22 15:42:55 2014 +0200 591 592 fix crash in nouveau_glamor_set_shared_pixmap_backing 593 594commit 217f6577afe44b10f7f8e066eaf55419fb0f5ee1 595Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 596Date: Wed Oct 22 13:16:33 2014 +0200 597 598 move down call to nouveau_present_init until after fb init. 599 600commit 241e7289f25a342a457952b9b0e539c2f0b81d99 601Author: Maarten Lankhorst <maarten.lankhorst@ubuntu.com> 602Date: Wed Oct 22 11:04:16 2014 +0200 603 604 enable dri3 support without glamor 605 606commit 762b22f5a2b9deed1c6b18d7098f8fddf2f4b759 607Author: Ilia Mirkin <imirkin@alum.mit.edu> 608Date: Tue Sep 23 21:44:12 2014 -0400 609 610 kepler: remove unnecessary texbars, add live-only tex for nvf0 611 612 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 613 614commit ce9d6e165cd810f06d158e3cfbb4b9f14b060ded 615Author: Ilia Mirkin <imirkin@alum.mit.edu> 616Date: Tue Sep 23 21:37:46 2014 -0400 617 618 allow path to envyas binary to be specified 619 620 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 621 622commit 02c1aee91ae22b58e777716ffd38397f9df0a087 623Author: Ilia Mirkin <imirkin@alum.mit.edu> 624Date: Thu Sep 4 15:27:53 2014 -0400 625 626 copy: don't bother trying to create copy on pre-nva3 627 628 Only the NVA3/5/8/F tesla chips have a copy engine. Don't bother trying 629 to create one on earlier tesla chips (and thus printing an error on 630 failure). 631 632 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 633 634commit 3cd4c8494c296c7583dfa7f0823a272c9e932e03 635Author: Mario Kleiner <mario.kleiner.de@gmail.com> 636Date: Thu Aug 28 03:57:48 2014 +0200 637 638 Bump version to 1.0.11 for release. 639 640 Highlights: 641 642 - Support for server managed fd's. 643 - Glamor support. 644 - Maxwell support. 645 - DRI3 and initial Present support. 646 - vsync'ed kms pageflip performance fixes when running on Linux 3.13+ 647 - Multi-display vsync, vblank, swap scheduling, timestamping fixes. 648 - Multi x-screen support fixes. 649 - ZaphodHead support on for multiple outputs per x-screen. 650 - EXA nv-10 fixes. 651 - Enable sync of swaps to vblank by default (Option GLXVblank "on"). 652 - Disable pseudo-triplebuffering by default (Option SwapLimit "1"). 653 654 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 655 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 656 657commit 6d49f3f7aaaec872b6f9841f744a0bf09cc98492 658Author: Mario Kleiner <mario.kleiner.de@gmail.com> 659Date: Thu Aug 28 03:49:55 2014 +0200 660 661 dri2: Set default SwapLimit to 1 for double-buffering. 662 663 This used to be 2 for pseudo-triplebuffering on XOrg 1.12+. 664 665 However, it caused problems like unthrottled swapping 666 at up to twice the video refresh rate for redirected 667 windows under desktop composition, which can't be easily 668 fixed under dri2. 669 670 Default to double-buffering. Users can override this, but 671 dri3 provides better solutions for this. 672 673 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 674 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 675 676commit 10729a442be3cf018c5806c95ec4218977917ef3 677Author: Mario Kleiner <mario.kleiner.de@gmail.com> 678Date: Mon Aug 18 09:53:27 2014 +0200 679 680 xv/dri2/dri3: Switch to optimized crtc selection for drawables. 681 682 Make crtc selection consistent with ati and intel ddx. 683 684 Pick the crtc to use for vblank events, swap scheduling and 685 kms-pageflip completion events as the one with maximum pixel area 686 intersection between its viewport and the drawable. If multiple 687 crtc's viewports display the same amount of area of a drawable, 688 select the crtc whose output is the designated "primary output" 689 for the x-screen. The latter allows, e.g., in a clone display 690 setup, to select the important output on which tearing should be 691 minimized. The former makes sure to minimize or avoid tearing on 692 the display which shows the largest part of the drawable. 693 694 This method is transplanted from the ati ddx. 695 696 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 697 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 698 699commit be96a5a62df5e1f286690311a6bc4df4f2a99e57 700Author: Mario Kleiner <mario.kleiner.de@gmail.com> 701Date: Sun Aug 17 02:30:36 2014 +0200 702 703 Add file "compile" to .gitignore 704 705 Some auto-generated file by new make systems to deal with some 706 old compilers. 707 708 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 709 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 710 711commit 16c885ce9c7bebcfa1252a5d500956dd5f551043 712Author: Mario Kleiner <mario.kleiner.de@gmail.com> 713Date: Tue Aug 12 11:08:24 2014 +0200 714 715 Fix event handling on multi-x-screen configs. 716 717 Only register the wakeup handler and event socket 718 once per card fd and server generation, as the fd 719 and device file is shared between all x-screens for 720 a given card during a given server generation. 721 722 Without this fix, vblank and kms-pageflip completion 723 event processing don't work properly, as the server 724 doesn't kick the wakeup handler for gpu events, and 725 therefore the desktop will freeze, unless the user 726 manually kicks the wakeup handler by moving the mouse 727 or hitting the keyboard. 728 729 Add proper reference counting and checks to make it so. 730 This fix is derived from a similar and proven fix in 731 the ati ddx for the same problem. 732 733 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 734 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 735 736commit 7eab6806927ea8938b9bd143d1702e87fde2f2cc 737Author: Mario Kleiner <mario.kleiner.de@gmail.com> 738Date: Wed Aug 13 06:48:10 2014 +0200 739 740 Make new ZaphodHeads work across server regenerations. 741 742 Reset the mask of already assigned crtc's at start 743 of each new server generation. 744 745 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 746 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 747 748commit 6ad053788e804f3a8bd36ee3653ec95e7df62623 749Author: Mario Kleiner <mario.kleiner.de@gmail.com> 750Date: Wed Aug 6 07:37:32 2014 +0200 751 752 dri2 and xv: Fix NVxxSyncToVBlank() for multi-x-screen's. 753 754 Use helper function drmmode_head() to choose proper hw-crtc 755 to sync to. Fixes xv blits, and dri2 Copy-Swaps. 756 757 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 758 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 759 760commit 17de663a71bcad6f69414eb6bc6bf9b76a6bf5f4 761Author: Ben Skeggs <bskeggs@redhat.com> 762Date: Mon Aug 11 09:48:42 2014 +1000 763 764 fix glamor header check 765 766 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 767 768commit 64ca88570b71289a019a2d130b1a30dc474ef52d 769Author: Ben Skeggs <bskeggs@redhat.com> 770Date: Mon Aug 11 08:59:30 2014 +1000 771 772 present: use drmmode_head() where it's appropriate 773 774 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 775 776commit 6e71bdf5a9da15c60694fa8543243e37a32c5faf 777Author: Mario Kleiner <mario.kleiner.de@gmail.com> 778Date: Mon Jun 23 11:25:41 2014 +0200 779 780 Allow/Fix use of multiple ZaphodHead outputs per x-screen. 781 782 Defining multiple ZaphodHead outputs per x-screen in a 783 multiple x-screen's per gpu configuration caused all 784 outputs except one per x-screen to go dark, because 785 there was a fixed mapping x-screen number -> crtc number, 786 limiting the number of crtc's per x-screen to one. 787 788 On a ZaphodHead's setup, be more clever and assign 789 as many crtc's to a given x-screen as there are 790 ZaphodHeads defined for that screen, assuming 791 there are enough unused crtc's available. 792 793 Tested on a triple display setup with different combos 794 of one, two or three ZaphodHeads per one, two or three 795 x-screens. 796 797 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 798 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 799 800commit fcf91a275e96f14e49a93dc3f8dd6798cf7d98f3 801Author: Mario Kleiner <mario.kleiner.de@gmail.com> 802Date: Mon Jun 23 11:13:57 2014 +0200 803 804 dri2: Fix drmWaitVblank on multi-x-screen configs. 805 806 Old mapping of xf86Crtc -> display head index was 807 too simple and only worked on a single x-screen config. 808 809 Introduce a new helper function drmmode_hw_head_index() 810 which returns proper mapping also for multi x-screens per 811 gpu and ZaphodHeads setups and other non-standard multi- 812 display configurations. 813 814 Tested on triple-head setup with various ZaphodHead and 815 server layout combos. 816 817 This helper will be also needed for dri3/present code. 818 819 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 820 821commit 29572e44c24613be9dec0c98ff2731e46ce6203b 822Author: Ilia Mirkin <imirkin@alum.mit.edu> 823Date: Sun Aug 10 02:25:15 2014 -0400 824 825 exa/nv10: src ops should use the SF macro 826 827 Doesn't actually end up mattering. 828 829 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 830 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 831 832commit 35e17905e1c54558059f6b72b8594a3377f1a57c 833Author: Ilia Mirkin <imirkin@alum.mit.edu> 834Date: Sun Aug 10 02:25:14 2014 -0400 835 836 exa/nv10: fix blend function setup to check the correct mask 837 838 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 839 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 840 841commit dfac18c67a17782a0263daf8cb4ee487374abd2c 842Author: Ilia Mirkin <imirkin@alum.mit.edu> 843Date: Sun Aug 10 02:25:13 2014 -0400 844 845 exa/nv10: use color 0 when the pict format has no RGB component 846 847 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 848 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 849 850commit ff7010e06b6da0fe27143b47b8a4ffd2f12894b7 851Author: Ilia Mirkin <imirkin@alum.mit.edu> 852Date: Sun Aug 10 02:25:12 2014 -0400 853 854 exa/nv10: use same clip settings as mesa driver 855 856 The higher 0x800 was getting overwritten by the 0x7ff anyways, so it 857 wasn't doing any good. The mesa driver just uses 0x800 for the low 858 portion and doesn't set the 8 bit in the higher portion, so do the same 859 thing here. 860 861 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 862 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 863 864commit edd1608c397ebcf487b84948031f7807178e5e4d 865Author: Emil Velikov <emil.l.velikov@gmail.com> 866Date: Tue Jul 1 22:31:47 2014 +0100 867 868 xorg_list: define the xorg_list* symbols 869 870 ...when building against pre 1.12 x. 871 The build will fail as earlier versions of X were missing the xorg_ 872 prefix of the symbols - struct and util functions. 873 874 Reported-by: Pali Rohár <pali.rohar@gmail.com> 875 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 876 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 877 878commit f6c1c755d8230494c8115e2c7899a85663f5c176 879Author: Ben Skeggs <bskeggs@redhat.com> 880Date: Tue Jul 1 12:56:49 2014 +1000 881 882 fix a harmless typo 883 884 fdo#80738 885 886 Reported-by: Paulo Sérgio Travaglia <pstglia@gmail.com> 887 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 888 889commit 096fb1ebc10b14ba1b747832efe378369bcf44e6 890Author: Mario Kleiner <mario.kleiner.de@gmail.com> 891Date: Mon May 26 05:39:50 2014 +0200 892 893 dri2: Add support for handling more than 2 crtc's 894 895 Need to use the DRM_VBLANK_HIGH_CRTC bits to allow 896 selecting crtc's with id > 1 on latest gpu's with 897 up to four display engines. 898 899 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 900 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 901 902commit 4771ff5c668047a2dc56e4c6131175eed837e757 903Author: Marcin Slusarz <marcin.slusarz@gmail.com> 904Date: Sun Jun 22 01:32:42 2014 +0200 905 906 present: build only when glamor is enabled 907 908 nouveau_present_flip_exec references glamor_fd_from_pixmap, 909 which do not exist when glamor is disabled 910 911 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> 912 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 913 914commit 41c1c4ee867753422d6486f13bf20b45905a5dd5 915Author: Marcin Slusarz <marcin.slusarz@gmail.com> 916Date: Sun Jun 22 01:31:46 2014 +0200 917 918 glamor: fix build without glamor.h 919 920 xorg-server can be built without glamor, which leads to: 921 922 CC nouveau_xv.lo 923 In file included from nouveau_xv.c:41:0: 924 nouveau_glamor.h:12:20: fatal error: glamor.h: No such file or directory 925 compilation terminated. 926 927 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> 928 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 929 930commit f0fa8313714c2a5b16e784b257b5ff79da3a443b 931Author: Mario Kleiner <mario.kleiner.de@gmail.com> 932Date: Thu Jun 19 17:11:34 2014 +0200 933 934 dri2: Enable sync of bufferswaps to Vblank by default. 935 936 Make this consistent with other drivers default behaviour. 937 938 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 939 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 940 941commit 02d7e647ce3a4a5d48c4ee94c94bebcac2d4827d 942Author: Mario Kleiner <mario.kleiner.de@gmail.com> 943Date: Sat Jun 21 01:26:16 2014 +0200 944 945 dri2: Fix kms pageflip completion timestamping. 946 947 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 948 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 949 950commit 4d92001ebe6bec8c5eaab8156421e62a7e4dedc8 951Author: Mario Kleiner <mario.kleiner.de@gmail.com> 952Date: Fri Mar 14 18:27:53 2014 +0100 953 954 Add kernel async_flip cap detection. Part II of double-sync fix. 955 956 Query if kernel supports the async_flip cap, thereby needs the 957 new sync behavior. Linux 3.13+ nouveau-kms have this cap and 958 need this fix. 959 960 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 961 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 962 963commit 54dfbb2fed393b65d846e6ed9672acb47e0f19de 964Author: Mario Kleiner <mario.kleiner.de@gmail.com> 965Date: Fri Mar 14 18:23:04 2014 +0100 966 967 dri2: Fix double-sync of pageflips on Linux 3.13+ - Part I 968 969 Linux 3.13 and later sync kms pageflips to vblank in the kms 970 driver, so we must not emit a sync to vblank pushbuf in the ddx 971 on such kernels, or maximum framerate will be cut into half! 972 973 A sync-to-vblank-pushbuf is emitted for copyswaps as in the past, 974 also for older kernels which don't support async_pageflip's and 975 don't sync by themselves. 976 977 This adds the implementation, but not the detection logic for 978 async_pageflip support in the kernel. 979 980 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 981 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 982 983commit 303402e00d55b5296311738184cd61f4aadab74d 984Author: Mario Kleiner <mario.kleiner.de@gmail.com> 985Date: Sat Jun 21 00:09:39 2014 +0200 986 987 Fix nouveau_copy_init() 988 989 Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com> 990 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 991 992commit 65a70c4edd1f0df808a3a4dea6e8c9846cc7c920 993Author: Ben Skeggs <bskeggs@redhat.com> 994Date: Fri Jun 20 16:57:18 2014 +1000 995 996 another build fix... 997 998commit 882ebb4000803f6025e1bb3213a31b19f6d1a2c6 999Author: Ben Skeggs <bskeggs@redhat.com> 1000Date: Mon Nov 18 14:42:02 2013 +1000 1001 1002 present: initial support 1003 1004 Until glamor grows its own implementation. 1005 1006 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1007 1008commit 0d94f32fce2759c4b0f4d22b99f3ced09c2aa5a0 1009Author: Ben Skeggs <bskeggs@redhat.com> 1010Date: Fri Jun 20 09:47:39 2014 +1000 1011 1012 kms: implement a more generic event mechanism 1013 1014 Present wants to use this too. 1015 1016 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1017 1018commit 97feb3b2104287eb385ffba8329f304e2808d808 1019Author: Ben Skeggs <bskeggs@redhat.com> 1020Date: Fri Jun 20 09:29:08 2014 +1000 1021 1022 dri2: move page flip handling out of drmmode 1023 1024 Nothing to see here... This is to (hopefully) avoid busting DRI2 while 1025 implementing Present support. 1026 1027 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1028 1029commit 60b876eab3708f1c277027addb5514974452644c 1030Author: Ben Skeggs <bskeggs@redhat.com> 1031Date: Fri Jun 20 09:20:37 2014 +1000 1032 1033 dri2: move away from directly touching drmmode on page flips 1034 1035 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1036 1037commit 36dda1341008dd87a6b5459f0616776097d11d9f 1038Author: Ben Skeggs <bskeggs@redhat.com> 1039Date: Fri Jun 20 09:04:30 2014 +1000 1040 1041 dri2: shuffle some functions around 1042 1043 Nothing to see here... 1044 1045 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1046 1047commit 8bdec8ebd68ba5eaa696d8a666e18774bff3b2ad 1048Author: Ben Skeggs <bskeggs@redhat.com> 1049Date: Fri Jun 20 12:37:33 2014 +1000 1050 1051 glamor: fix randr resize 1052 1053 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1054 1055commit 3c45b590a31157abca140096b141aa83fecb671f 1056Author: Ben Skeggs <bskeggs@redhat.com> 1057Date: Fri Jun 20 08:28:30 2014 +1000 1058 1059 sync: fix build against DRI3-less server 1060 1061 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1062 1063commit 88ae58d786b0c41148e692dbfe8e2646e7533d9e 1064Author: Ilia Mirkin <imirkin@alum.mit.edu> 1065Date: Thu Jun 19 02:37:45 2014 -0400 1066 1067 update man page with new chips, AccelMethod option 1068 1069 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1070 1071commit 5cb36ba57493bf7edf560bf370c87eff79f1d1a6 1072Author: Ben Skeggs <bskeggs@redhat.com> 1073Date: Thu Jun 19 14:54:20 2014 +1000 1074 1075 default to glamor on maxwell 1076 1077 We have no RENDER/Xv acceleration in EXA for Maxwell. 1078 1079 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1080 1081commit 81148bb1dbc7007c021c59411d56cb31cfc74ef2 1082Author: Ben Skeggs <bskeggs@redhat.com> 1083Date: Thu Jun 19 14:54:20 2014 +1000 1084 1085 glamor: provide dri3 support when enabled 1086 1087 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1088 1089commit 86024ceef015ffe31a204cc5bc6c326a19363ff1 1090Author: Ben Skeggs <bskeggs@redhat.com> 1091Date: Thu Jun 19 14:54:20 2014 +1000 1092 1093 glamor: initial support (no dri) 1094 1095 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1096 1097commit fd0ce8839f307693d86c7602dd926ce79e6b777d 1098Author: Ben Skeggs <bskeggs@redhat.com> 1099Date: Thu Jun 19 14:54:20 2014 +1000 1100 1101 add support for maxwell, minus Xv/renderaccel 1102 1103 More extensive acceleration will be via glamor, at least to begin with. It 1104 seems to make sense to jump on the bandwagon now, and deal with any issues 1105 that arise before it it becomes the standard for XWayland. 1106 1107 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1108 1109commit 8ad0465e15bfc502abc259e886d6dcb448e0141a 1110Author: Ben Skeggs <bskeggs@redhat.com> 1111Date: Thu Jun 19 14:54:20 2014 +1000 1112 1113 50/c0/e0 -> tesla/fermi/kepler 1114 1115 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1116 1117commit 4b138ab18f58c6d459e21dc319615f536c8e69c8 1118Author: Ben Skeggs <bskeggs@redhat.com> 1119Date: Thu Jun 19 14:54:20 2014 +1000 1120 1121 merge nv_dma.c into nv_accel_common.c 1122 1123 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1124 1125commit 5c44e6b86195c22e092db5dae06e70208e8b66a0 1126Author: Ben Skeggs <bskeggs@redhat.com> 1127Date: Thu Jun 19 14:54:20 2014 +1000 1128 1129 pull copy engine handling into one place, a place that isn't also exa 1130 1131 As a side-effect, we now allow Kepler to use async copies too. 1132 1133 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1134 1135commit c375bacbdaeabd0288855c13b4717d0694be9e56 1136Author: Ben Skeggs <bskeggs@redhat.com> 1137Date: Thu Jun 19 14:54:20 2014 +1000 1138 1139 exa: move module load into nouveau_exa_init 1140 1141 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1142 1143commit b24cae0bf5db6ece21439d4c6ff3668aed4c78d6 1144Author: Ben Skeggs <bskeggs@redhat.com> 1145Date: Thu Jun 19 14:54:20 2014 +1000 1146 1147 dri2: move "is supported" checks out of nv_driver.c 1148 1149 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1150 1151commit 07c26a0670dd2558acf8e1159f2f68989a7f753d 1152Author: Ben Skeggs <bskeggs@redhat.com> 1153Date: Thu Jun 19 14:54:20 2014 +1000 1154 1155 replace NoAccel with AccelMethod in preparation for glamor support 1156 1157 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1158 1159commit e7cf48d9529160f7add2fef72d2e42be9370594a 1160Author: Ben Skeggs <bskeggs@redhat.com> 1161Date: Thu Jun 19 14:54:20 2014 +1000 1162 1163 silence a warning 1164 1165 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1166 1167commit 4a18dd74c0cc6e75937eb78347e8ea6c81d1373a 1168Author: Dave Airlie <airlied@redhat.com> 1169Date: Mon May 26 17:30:24 2014 +1000 1170 1171 autogen: add support for gnome autobuilder 1172 1173 Signed-off-by: Dave Airlie <airlied@redhat.com> 1174 1175commit 8604a741baa5360164a882bd9f3c8474840744db 1176Author: Hans de Goede <hdegoede@redhat.com> 1177Date: Fri Apr 11 14:20:43 2014 +0200 1178 1179 Fix building on older servers without xf86platformBus.h 1180 1181 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 1182 1183commit 0c13843e300d1a9db2bfed18dc1b5df38c5adfdf 1184Author: Hans de Goede <hdegoede@redhat.com> 1185Date: Mon Mar 17 17:37:31 2014 +0100 1186 1187 Add support for server managed fds 1188 1189 v2 (Martin Peres): 1190 - Do not define "NVEntPtr pNVEnt" if not needed to avoid some warnings 1191 1192 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 1193 Signed-off-by: Martin Peres <martin.peres@free.fr> 1194 1195commit 47f08aa6d713da4e6ae9c35ce691e790ffaed09d 1196Author: Hans de Goede <hdegoede@redhat.com> 1197Date: Mon Mar 17 17:37:30 2014 +0100 1198 1199 Add a NVOpenNouveauDevice helper function 1200 1201 This is a preparation patch for adding support for server managed fds 1202 1203 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 1204 Signed-off-by: Martin Peres <martin.peres@free.fr> 1205 1206commit 480f0998ffed6d9a5c6656dba75182f00fd88a1b 1207Author: Ben Skeggs <bskeggs@redhat.com> 1208Date: Thu Nov 7 14:56:48 2013 +1000 1209 1210 bump to 1.0.10 for release 1211 1212 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1213 1214commit 59395c738ebdc21087109075ada67ea8d7783141 1215Author: Ben Skeggs <bskeggs@redhat.com> 1216Date: Thu Nov 7 14:56:06 2013 +1000 1217 1218 recognise GK208 as a Kepler board 1219 1220 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1221 1222commit 03642323a954d0adb11662a98e92dfa20bee9728 1223Author: Dave Airlie <airlied@gmail.com> 1224Date: Wed Jul 31 11:19:50 2013 +1000 1225 1226 nouveau: fix build against older servers. 1227 1228 Older servers had no GPU support. 1229 1230 Signed-off-by: Dave Airlie <airlied@redhat.com> 1231 1232commit 27323ce68d58fcb82a648ab14abbc99acdc36f44 1233Author: Dave Airlie <airlied@gmail.com> 1234Date: Wed Jul 31 11:01:15 2013 +1000 1235 1236 nouveau: add gpu identifier to connector names for secondary gpus 1237 1238 this avoids clashes with identifiers from other GPUs. 1239 1240 Signed-off-by: Dave Airlie <airlied@redhat.com> 1241 1242commit 1df177f35a05db505577cdc929e63fde906a704b 1243Author: Dave Airlie <airlied@gmail.com> 1244Date: Wed Jul 31 10:51:03 2013 +1000 1245 1246 nouveau: fix build on ppc by wrapping immintrin include. 1247 1248 Signed-off-by: Dave Airlie <airlied@redhat.com> 1249 1250commit 300c5a32d79cf507b5fede96710335bcd316eccb 1251Author: Dave Airlie <airlied@redhat.com> 1252Date: Tue Jul 30 15:28:06 2013 +1000 1253 1254 bump to 1.0.9 for release 1255 1256 Signed-off-by: Dave Airlie <airlied@redhat.com> 1257 1258commit be44e7804862b4c276ed4d4717b1212920f428e6 1259Author: Dave Airlie <airlied@gmail.com> 1260Date: Tue Jul 30 15:26:46 2013 +1000 1261 1262 nouveau: fix crash when xinerama is enabled. 1263 1264 Signed-off-by: Dave Airlie <airlied@redhat.com> 1265 1266commit 588252556d1c84bf6793bc22edd48e6650a0f1b9 1267Author: Dave Airlie <airlied@gmail.com> 1268Date: Tue Jul 30 15:14:22 2013 +1000 1269 1270 fix make distcheck. 1271 1272 add all the nvf0 stuff. 1273 1274 Signed-off-by: Dave Airlie <airlied@redhat.com> 1275 1276commit 29aecda1b5bdb95700894083874492a6d5dfc08a 1277Author: Dave Airlie <airlied@gmail.com> 1278Date: Tue Jul 30 14:10:51 2013 +1000 1279 1280 drmmode: add support for multi-screen reverse optimus 1281 1282 Initial reverse optimus didn't consider multiple screens, so 1283 this overhauls the code to use the new X server interface, 1284 and allows for multiple outputs on the nvidia to be used with 1285 the intel doing the rendering. 1286 1287 Signed-off-by: Dave Airlie <airlied@redhat.com> 1288 1289commit d7ae886260b480da0ea4cad22eb6f7b7d12f07b0 1290Author: Ilia Mirkin <imirkin@alum.mit.edu> 1291Date: Mon Jul 29 21:03:54 2013 -0400 1292 1293 nouveau: use uppercase for chipset name 1294 1295 Right now the log will output e.g. NVaa instead of NVAA, which is how we 1296 usually talk about it. 1297 1298 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1299 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1300 1301commit 4492cf90c2a7fe27badf6f3f0f296755a65e1e83 1302Author: Ilia Mirkin <imirkin@alum.mit.edu> 1303Date: Mon Jul 29 02:40:39 2013 -0400 1304 1305 xv: speed up YV12 -> NV12 conversion using SSE2 if available 1306 1307 memcpy() goes from taking 45% to 66% of total function time, which 1308 translates to a 30% decrease in NVPutImage runtime. 1309 1310 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1311 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1312 1313commit 08c667b06d28756d0d32f82bd736b0bdf2b4448e 1314Author: Ilia Mirkin <imirkin@alum.mit.edu> 1315Date: Mon Jul 29 02:40:38 2013 -0400 1316 1317 xv: fix last pixel for big-endian machines in YV12 -> NV12 conversion 1318 1319 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1320 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1321 1322commit 8c1c5d4f212958559132506d51dcfeeb17f60371 1323Author: Ben Skeggs <bskeggs@redhat.com> 1324Date: Wed May 15 21:34:13 2013 +1000 1325 1326 nvf0: implement EXA and XVideo acceleration 1327 1328 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1329 1330commit 779b8d44b5ff84b87b52600a2f010ac19cd396ad 1331Author: Sven Joachim <svenjoac@gmx.de> 1332Date: Thu Jun 13 20:41:53 2013 +0200 1333 1334 Link against libdrm 1335 1336 The DDX uses symbols from libdrm but currently does not link against 1337 it, but rather relies on indirect linkage from libdrm_nouveau. This 1338 could lead to insufficient dependencies in distribution packages, seen 1339 in http://bugs.debian.org/633786. 1340 1341 Signed-off-by: Sven Joachim <svenjoac@gmx.de> 1342 1343commit e96ae2b129df2a50d1d3839084df4fb1868d914d 1344Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1345Date: Wed Jun 12 11:05:23 2013 +0200 1346 1347 fixup the dri2 check 1348 1349 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1350 1351commit 7f20a6af25ae28452ec9c971bfa2227857e6ee93 1352Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1353Date: Wed Jun 12 10:46:39 2013 +0200 1354 1355 bump to 1.0.8 1356 1357commit 1fda669c0e5a25eac7f9f1e9a743120cae50ff51 1358Author: Emil Velikov <emil.l.velikov@gmail.com> 1359Date: Sat Feb 16 20:48:47 2013 +0000 1360 1361 configure: printout the configuration info 1362 1363 Let the bikeshedding begin 1364 1365 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1366 1367commit 45dbcaac301efce04d0bb492337f4febf8129ed9 1368Author: Emil Velikov <emil.l.velikov@gmail.com> 1369Date: Sat Feb 16 20:48:45 2013 +0000 1370 1371 nouveau: mandate dri2 build 1372 1373 Building nouveau without dri2 is just silly 1374 1375 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1376 1377commit ea8d225fe100ecabb72e3cc0a92372f16afb10b3 1378Author: Emil Velikov <emil.l.velikov@gmail.com> 1379Date: Sat Feb 16 20:48:44 2013 +0000 1380 1381 dri1: purge the final references 1382 1383 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1384 1385commit b5d1a9bd269c9d2758fee37492cdcccc976fcc3b 1386Author: Emil Velikov <emil.l.velikov@gmail.com> 1387Date: Sat Feb 16 20:48:43 2013 +0000 1388 1389 nouveau: Do not load dri {sub,}module 1390 1391 Nouveau does not use dri1, thus loading XFree86-DRI is not needed 1392 On the other hand, we do use dri2 1393 1394 As a side effect, purge the 'set-but-unused' variable pLibDRMVersion 1395 1396 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1397 1398commit 7c3f6cd8d6b14aea561fb877f6006d339d156f35 1399Author: Emil Velikov <emil.l.velikov@gmail.com> 1400Date: Sat Feb 16 20:48:42 2013 +0000 1401 1402 nouveau: stop using dri1 function DRICreatePCIBusID 1403 1404 Step 1 to completely rip out dri1 out of nouveau 1405 1406 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1407 1408commit 3b1a45325d5fd46f0e7fad17168742b4c9a2465a 1409Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1410Date: Wed Jun 12 10:30:32 2013 +0200 1411 1412 update makefiles for make distcheck 1413 1414commit b82aacd3e72b52240181569c8b0abb55c1f716d5 1415Author: Emil Velikov <emil.l.velikov@gmail.com> 1416Date: Sat Feb 16 12:48:46 2013 -0800 1417 1418 configure: require xorg-macros 1.8 1419 1420 Explicitly check for xorg-macros 1.8, as a subtle hint 1421 to update your build deps if building your own ddx 1422 1423 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1424 1425commit 4f667b40f41e311fbe302237efe173bcc24fb8bb 1426Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1427Date: Mon Jun 10 14:29:59 2013 +0200 1428 1429 fix a xfree from previous patch 1430 1431 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1432 1433commit 2e34d11541e1f987c7ca0556592cb16f4d232299 1434Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1435Date: Mon Jun 10 13:06:35 2013 +0200 1436 1437 nouveau: enable support for normal optimus configuration 1438 1439 nvd7 has no display engine, only enable sink output capability 1440 if there is a crtc. 1441 1442 drmmode doesn't work without crtc's, so prevent crashes by skipping 1443 the drmmode code in that case. 1444 1445 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1446 1447commit bf72ae1f6574c540f0afc2d7845d41df43507a8f 1448Author: Ben Skeggs <bskeggs@redhat.com> 1449Date: Wed May 15 22:05:01 2013 +1000 1450 1451 nve0: envyas isn't overly clever.. avoid short-form instructions 1452 1453 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1454 1455commit 41cc2f24c43d2656ac722f594aaebad93e37dfec 1456Author: Ben Skeggs <bskeggs@redhat.com> 1457Date: Wed May 15 21:13:11 2013 +1000 1458 1459 nvc0-nve0: decompile shaders into their source, add basic build system 1460 1461 Far, far easier to maintain this way... 1462 1463 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1464 1465commit 1edb5d7ae450a32fe6607d536653cf733444e7e2 1466Author: Ben Skeggs <bskeggs@redhat.com> 1467Date: Wed May 15 20:47:44 2013 +1000 1468 1469 nvc0-nve0: split shader code from their headers 1470 1471 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1472 1473commit c18439f759bec267a3b67d375dcbfeb8a6fd7b6a 1474Author: Dave Airlie <airlied@redhat.com> 1475Date: Mon May 13 13:55:51 2013 +1000 1476 1477 nouveau: clean up nv entity index declarations. 1478 1479 Signed-off-by: Dave Airlie <airlied@redhat.com> 1480 1481commit d3b52efe959f255784f5ead16d7276ca0fb4cdb1 1482Author: Dave Airlie <airlied@redhat.com> 1483Date: Mon May 13 13:35:12 2013 +1000 1484 1485 nouveau: attempt to fix zaphod since dri1 code removal 1486 1487 j_v on #nouveau bisected b1a630b48210d6a3c44994fce1b73273000ace5c has 1488 breaking zaphod, on review it was trying to open the drm fd a second time 1489 which was unnecessary. 1490 1491 Avoid the problem by storing the nv fd in an entity and have share it between 1492 the two scrn info recs. 1493 1494 Signed-off-by: Dave Airlie <airlied@redhat.com> 1495 1496commit 50426ac83f9f077791e4df8c04fc0ecefc06e8ab 1497Author: Marek Vasut <marex@denx.de> 1498Date: Fri May 3 15:48:55 2013 +0200 1499 1500 nouveau: Allow 3D accelerators with output ports 1501 1502 This patch let's cards with PCI class 0x30200 (3D controller) be 1503 operated by the nouveau driver as well. The nv Quadro NVS 450 is 1504 one such card, where the first GPU has PCI class 0x30000 and the 1505 other GPU has PCI class 0x30200. By ignoring the 1 << 9 bit in 1506 the PCI class mask, displays attached to the other GPU can also 1507 be used now. 1508 1509 Signed-off-by: Marek Vasut <marex@denx.de> 1510 Cc: Dave Airlie <airlied@redhat.com> 1511 Cc: Ben Skeggs <skeggsb@gmail.com> 1512 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1513 1514commit 2fa3397e348161a3394e2b456f065921272a056a 1515Author: Ilia Mirkin <imirkin@alum.mit.edu> 1516Date: Fri May 3 09:59:39 2013 -0400 1517 1518 nouveau_xv: Avoid reading off the end of the source image on NV50+ 1519 1520 The 'w' argument to NVCopyNV12ColorPlanes is used to index into the 1521 source image. line_len is rounded up to 8 on NV50+, so if the source 1522 image (+ left offset) is not rounded to 8, NVCopyNV12ColorPlanes could 1523 read past the end of the array and crash X. This change can cause the 1524 last few horizontal pixels of dst to not be initialized, but they 1525 should be truncated by the renderer anyways. 1526 1527 See https://bugs.freedesktop.org/show_bug.cgi?id=63263 1528 1529 Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> 1530 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1531 1532commit 81f9ab38f0a5de7217227dd17a4d0e50dffd0d28 1533Author: Ben Skeggs <bskeggs@redhat.com> 1534Date: Thu May 2 11:38:07 2013 +1000 1535 1536 nvf0: recognise as supported chipset 1537 1538 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1539 1540commit 653c95d60af0f7c19a1d2f3b76dd17bb169d9c9b 1541Author: Ben Skeggs <bskeggs@redhat.com> 1542Date: Mon Apr 29 10:07:59 2013 +1000 1543 1544 nve0: explicitly ask kernel for copy2 object 1545 1546 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1547 1548commit 6771424d79e541d2fa7253a582db3dc9108fd97d 1549Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1550Date: Wed Mar 27 09:50:03 2013 +0100 1551 1552 bump to 1.0.7 1553 1554 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1555 1556commit 84998320162a74a0861b7be1fcc230e50f08424a 1557Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1558Date: Mon Mar 25 13:55:37 2013 +0100 1559 1560 Clean up some errors on closing. 1561 1562 If forced close happens, all ioctl's will fail. Some of the handlers 1563 also need to be unregistered before the module is unloaded entirely. 1564 1565commit a80785f79268ed3701fc69cbd9194ad12822c82c 1566Author: Dave Airlie <airlied@redhat.com> 1567Date: Mon Feb 11 09:56:25 2013 +1000 1568 1569 nouveau: fix build against old servers (part 2) 1570 1571 Should fix next bit of 1572 https://bugs.freedesktop.org/show_bug.cgi?id=60369 1573 1574 Signed-off-by: Dave Airlie <airlied@redhat.com> 1575 1576commit 05dc9b57896689a3dbd9e6aeb16d57c49515b02e 1577Author: Marcin Slusarz <marcin.slusarz@gmail.com> 1578Date: Sun Feb 3 21:12:46 2013 +0100 1579 1580 nv30/exa: fix shaders on big-endian machines 1581 1582 Direct port of commit d1bc38b6673c54af61196056c489383fba8dced8 1583 "nv40/exa: fix shaders on big-endian machines" to nv30. 1584 1585 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=60050 1586 1587 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1588 1589commit 61a277488f7195914cbdda10355db0d758fc801c 1590Author: Dave Airlie <airlied@redhat.com> 1591Date: Thu Feb 7 10:47:20 2013 +1000 1592 1593 nouveau: fix build on older X servers 1594 1595 should fix https://bugs.freedesktop.org/show_bug.cgi?id=60369 1596 1597 Reported-by: fabio.ped@libero.it 1598 Signed-off-by: Dave Airlie <airlied@redhat.com> 1599 1600commit e8f222fd10e21f39488d3d8c697e4948cd787cd0 1601Author: Dave Airlie <airlied@redhat.com> 1602Date: Wed Feb 6 16:21:50 2013 +1000 1603 1604 nouveau: make prime regression fix more robust. 1605 1606 This makes this fix more robust and less crashy. 1607 1608 Signed-off-by: Dave Airlie <airlied@redhat.com> 1609 1610commit d762631c9306b6580b34db1e7eb57bbcac901390 1611Author: Dave Airlie <airlied@redhat.com> 1612Date: Wed Feb 6 15:52:33 2013 +1000 1613 1614 nouveau: align shared buffers to 256 1615 1616 We were getting 0xa5 error code on 2D transfers from tiled VRAM to untiled 1617 GART when using USB offload devices, this fixes this by upping the alignment 1618 on the 2D engine for shared buffers. 1619 1620 Signed-off-by: Dave Airlie <airlied@redhat.com> 1621 1622commit c6980b2add5752a7c285a846b2c9353c9d18afd4 1623Author: Dave Airlie <airlied@redhat.com> 1624Date: Wed Feb 6 15:24:09 2013 +1000 1625 1626 nouveau: fix regression in PRIME since 58284cf32778d54cde139423450bc33360784503 1627 1628 Signed-off-by: Dave Airlie <airlied@redhat.com> 1629 1630commit 59fb3f0252e9b769c5877cfe83d8dbc51cdff232 1631Author: Dave Airlie <airlied@redhat.com> 1632Date: Wed Feb 6 14:58:08 2013 +1000 1633 1634 nouveau: create shared pixmaps in GART 1635 1636 this creates the shared pixmaps in GART, and makes sure they are untiled. 1637 1638 Signed-off-by: Dave Airlie <airlied@redhat.com> 1639 1640commit 1fdd7db94b55c65ea62cc9eaefff620b20e9e4ea 1641Author: Dave Airlie <airlied@redhat.com> 1642Date: Mon Jan 7 15:28:53 2013 +1000 1643 1644 nouveau: add reverse prime support 1645 1646 This allows the nvidia card to scanout Intel cards rendering. 1647 1648 Signed-off-by: Dave Airlie <airlied@redhat.com> 1649 1650commit 58284cf32778d54cde139423450bc33360784503 1651Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 1652Date: Tue Jan 29 16:38:21 2013 +1000 1653 1654 dri2: fix missing frame throttle on copy_region buffer swaps 1655 1656 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1657 1658commit 912d418fdfd2e99eef1e5c631c76dda1d82cf451 1659Author: Marcin Slusarz <marcin.slusarz@gmail.com> 1660Date: Mon Jan 21 00:15:38 2013 +0100 1661 1662 nouveau: put cursor BO in GART 1663 1664 Keeping it in VRAM wastes CPU time, because cursor_set ioctl reads 1665 handed BO back to RAM, just to write it to actual cursor BO. 1666 1667 Here (nv92/core i7), this patch decreases overall cpu usage of 1668 drmmode_load_cursor_argb from 4.6ms to ~90us and lets nv50_crtc_cursor_set 1669 disappear from perf traces. 1670 1671 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1672 1673commit 8f934fad5d4934936b3265f272ca81e73c60d7dc 1674Author: Dave Airlie <airlied@redhat.com> 1675Date: Mon Jan 7 14:13:10 2013 +1000 1676 1677 bump to 1.0.6 1678 1679 Signed-off-by: Dave Airlie <airlied@redhat.com> 1680 1681commit 00834273e22f7234a3755269100ea3f10b521e35 1682Author: Dave Airlie <airlied@redhat.com> 1683Date: Mon Jan 7 14:04:40 2013 +1000 1684 1685 nouveau: fix dirty redisplay damage handling 1686 1687 This processes the damage in the correct order. 1688 1689 Signed-off-by: Dave Airlie <airlied@redhat.com> 1690 1691commit 389b8a8dd91a48dee524eac96e60813c507f1a82 1692Author: Ben Skeggs <bskeggs@redhat.com> 1693Date: Mon Jan 7 14:05:06 2013 +1000 1694 1695 bump to 1.0.5 1696 1697 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1698 1699commit efe9c57af6ff4c12936e49ac3e188c54b8b2d5bd 1700Author: Ben Skeggs <bskeggs@redhat.com> 1701Date: Mon Jan 7 11:35:56 2013 +1000 1702 1703 nvc0-/exa: fix typo which caused breakage in rendercheck tmcoords test 1704 1705 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1706 1707commit 5cd86445272818ec2f598f4685a5d7f24e403d07 1708Author: Marcin Slusarz <marcin.slusarz@gmail.com> 1709Date: Sun Nov 25 22:59:49 2012 +0100 1710 1711 add some missing newlines in error messages 1712 1713 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1714 1715commit 3b33a0b426b8915f302247dcd40c64fb1a6dcdd2 1716Author: Emil Velikov <emil.l.velikov@gmail.com> 1717Date: Sun Nov 4 18:06:18 2012 +0000 1718 1719 kms: Let the server know if we got more than 1 EDID block 1720 1721 Verbatim copy from xf86-video-radeon commit 809cacb 1722 1723 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=32341 1724 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1725 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1726 1727commit 018ab2d4634f3c8923038337ce9600973e2f0002 1728Author: Ben Skeggs <bskeggs@redhat.com> 1729Date: Fri Nov 9 12:16:00 2012 +1000 1730 1731 bump version to 1.0.4 1732 1733 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1734 1735commit 0d32d964dbba3864af0946aa8894cef858eb1319 1736Author: Dave Airlie <airlied@redhat.com> 1737Date: Wed Nov 7 14:49:03 2012 +1000 1738 1739 nouveau: drop unused nv private members. 1740 1741 these are no longer needed. 1742 1743 Signed-off-by: Dave Airlie <airlied@redhat.com> 1744 1745commit b1a630b48210d6a3c44994fce1b73273000ace5c 1746Author: Dave Airlie <airlied@redhat.com> 1747Date: Wed Nov 7 14:45:14 2012 +1000 1748 1749 nouveau: drop DRI1 device open interface. 1750 1751 this is a DRI1 left over, the driver is meant to open the device itself now. 1752 1753 this old interface did crazy stuff like drmAddMap. 1754 1755 Reviewed-by: Ben Skeggs <bskeggs@redhat.com> 1756 Signed-off-by: Dave Airlie <airlied@redhat.com> 1757 1758commit 27a1a0616304e9b9f0ae842899b7d614f1026578 1759Author: Emil Velikov <emil.l.velikov@gmail.com> 1760Date: Fri Nov 2 03:57:42 2012 +0000 1761 1762 nouveau: NVPlatformProbe, handle shareable entities 1763 1764 During NVPciProbe, xf86ConfigPciEntity() implicitly calls checks and sets 1765 the entity as shared 1766 Here we are only allocating the screen, thus need to explicitly call the 1767 functions 1768 1769 ZaphodHead and Xinerama should finally work 1770 1771 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=56347 1772 1773 Reported-By: Vladimir Chebotarev <exc@rbcmail.ru> 1774 Reported-By: Damian Nowak <nowaker@geozone.pl> 1775 Tested-By: Damian Nowak <nowaker@geozone.pl> 1776 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1777 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1778 1779commit e34cfbd5bd23f7f15372af52d8a39a5715ce7310 1780Author: Emil Velikov <emil.l.velikov@gmail.com> 1781Date: Fri Nov 2 03:57:41 2012 +0000 1782 1783 nouveau: Factor out common code to NVHasKMS() 1784 1785 As the name suggests checks if it has kernel mode setting, 1786 prints out the interface version and checkes if the chipset 1787 is supported 1788 1789 Function is used in NVPciProbe and NVPlatformProbe 1790 1791 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1792 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1793 1794commit 78f24dc3798d635fca2e20856a09012c28bbbb8b 1795Author: Emil Velikov <emil.l.velikov@gmail.com> 1796Date: Fri Nov 2 03:57:40 2012 +0000 1797 1798 nouveau: introduce NVInitScrn() 1799 1800 Common/helper function for NV{Pci,Platform}Probe() 1801 1802 This introduces some the following functions to the 1803 NVPlatformProbe codepath 1804 * xf86SetEntitySharable() 1805 * xf86SetEntityInstanceForScreen() 1806 1807 Necessary for dualhead, xinerama and/or zaphodhead 1808 1809 v2: Rename nouveau_init_scrn() to NVInitScrn() 1810 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1811 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1812 1813commit 47457fbd5006070ea9a06238a5a69ca8167dac77 1814Author: Emil Velikov <emil.l.velikov@gmail.com> 1815Date: Fri Nov 2 03:57:39 2012 +0000 1816 1817 nouveau: Remove unneeded xf86GetEntityInfo() 1818 1819 Retrieving the whole entity information is not needed as we only 1820 use the index - stored already in entity_num 1821 1822 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 1823 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1824 1825commit 20995bb5920021668b8b607f886201c643ee0e9a 1826Author: Dave Airlie <airlied@redhat.com> 1827Date: Fri Nov 2 09:04:43 2012 +1000 1828 1829 nouveau: fix regression in virtualbox drawing hopefully. 1830 1831 Signed-off-by: Dave Airlie <airlied@redhat.com> 1832 1833commit 8c3e1623b0be15f8cc590d893bfd19be87bd079a 1834Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1835Date: Wed Oct 24 11:19:04 2012 +0200 1836 1837 bump version to 1.0.3 1838 1839commit 6644a3a724a11150eba0c0082c8d4fe167b1755e 1840Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1841Date: Tue Oct 23 20:17:29 2012 +0200 1842 1843 exa: remove exaMarkSync from download/upload screen 1844 1845 They're noops on nouveau since sync markers are unused and 1846 download/upload are always synchronous anyway from exa's POV. 1847 1848 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1849 1850commit 2a757b7dffa85c30879653a1e52e33b8f4fa8806 1851Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1852Date: Tue Oct 23 17:04:37 2012 +0200 1853 1854 exa: fix comparison in download_from_screen 1855 1856 Presumably a copy/paste error from upload_to_screen. 1857 1858 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1859 1860commit 72230cd5a7846f7a26f929262a745dc0bdc34a3d 1861Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1862Date: Tue Oct 23 10:37:27 2012 +0200 1863 1864 exa: fix up download_from/upload_to screen 1865 1866 Add a missing return TRUE to download_from_screen. 1867 Add a missing exaMarkSync to memcpy in upload_to_screen. 1868 Add a really loud warning to both if memcpy to/from tiled is done. 1869 1870 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1871 1872commit 1d7890ecb8b3188177ab85608ea232812672d923 1873Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1874Date: Tue Oct 23 09:53:39 2012 +0200 1875 1876 remove unused members from nv_type.h 1877 1878commit 0b2a548316b0ce88ba19d5e05fdca205fc139962 1879Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1880Date: Mon Oct 22 13:12:06 2012 +0200 1881 1882 nvc0/exa: make solid fill actually work 1883 1884 Some investigation showed that the code for accelerating solid fill 1885 didn't actually work because NVC0EXACheckTexture would fail. 1886 1887 Making the check the same as in NV50EXACheckTexture fixes this, and 1888 seems to get acceleration for solid fills working right. 1889 1890 Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1891 1892commit b4231dd715a8a7f86b04519b5f4e8a8d93c2f561 1893Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 1894Date: Tue Oct 9 09:06:59 2012 +0200 1895 1896 dri2: Fix potential race and crash for swap at next vblank. 1897 1898 This fixes a potential race + crash that wasn't properly 1899 handled by commit 248de8cdbd6d0bc062633b49896fa4791148cd3b 1900 and happened at least on one users machine. 1901 1902 That commit wrongly assumed no special action would be needed 1903 for swaps at next vblank while triple-buffering is enabled on 1904 XOrg server 1.12 or later. 1905 1906 Closer inspection of the x-server main dispatch loop shows 1907 it is possible that the client manages to get the server 1908 to dispatch a new DRI2GetBuffersWithFormat() call before 1909 the server calls the vblank event handler and executes 1910 the nouveau_dri2_finish_swap() routine. Such a race would 1911 cause a crash, as described in above commit. 1912 1913 This commit handles the "swap at next vblank" case by 1914 calling nouveau_dri2_finish_swap() immediately without 1915 the roundtrip (queue vblank_event -> kernel -> deliver event 1916 -> x-server processes event -> nouveau vblank event handler), 1917 before control gets returned to the client. 1918 1919 This avoids the race while retaining triple-buffering. As 1920 a bonus, time-critical swaps at next vblank get processed 1921 without roundtrip delay, increasing the chance of not 1922 skipping a frame due to vblank miss while sync to vblank is 1923 on. 1924 1925 Thanks to Anssi for reporting this problem on the nouveau 1926 mailing list at 12th July 2012 and for testing this patch. 1927 1928 Reported-by: Anssi Hannula <anssi.hannula@iki.fi> 1929 Tested-by: Anssi Hannula <anssi.hannula@iki.fi> 1930 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 1931 1932commit 01c9a9cff7d6232ca6eafa59dd60833a0d9cdc5b 1933Author: Ben Skeggs <bskeggs@redhat.com> 1934Date: Thu Oct 18 18:38:21 2012 +1000 1935 1936 shadowfb: fix segfault due to reading outside of shadow buffer 1937 1938 Probably caused by the new libdrm port, the new libdrm sets the bo 1939 size field to the *allocated* size and not the *requested* size, 1940 making the max_height calculation here invalid. 1941 1942 Switched to using virtualX/virtualY as the bounds, which should 1943 hopefully do the right thing.. 1944 1945 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1946 1947commit 297fd0d0755bda698be1d0b30cc60a41d7673c0b 1948Author: Dave Airlie <airlied@redhat.com> 1949Date: Tue Oct 16 16:15:16 2012 +1000 1950 1951 nouveau/dri2: fix pixmap/window offset calcs. 1952 1953 This should fix prime rendering under kwin, and not break it under the 1954 others. 1955 1956 Signed-off-by: Dave Airlie <airlied@redhat.com> 1957 1958commit 5c9379b14cdabc81fd5d4c916dcd93dc77c4f683 1959Author: Sebastian Keller <sebastian-keller@gmx.de> 1960Date: Tue Sep 25 11:35:35 2012 +0200 1961 1962 exa: use CLAMP_TO_EDGE for RepeatPad 1963 1964 This fixes border rendering in some gtk3 themes. 1965 https://bugs.freedesktop.org/show_bug.cgi?id=55310 1966 1967 v2 (Ben Skeggs): implement same fix for fermi/kepler too 1968 1969 Signed-off-by: Sebastian Keller <sebastian-keller@gmx.de> 1970 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 1971 1972commit 302b5da84b4890c067f5f7e8b0f4e689909bbb51 1973Author: Adam Jackson <ajax@redhat.com> 1974Date: Tue Sep 25 10:44:32 2012 -0400 1975 1976 Remove mibstore.h 1977 1978 Signed-off-by: Adam Jackson <ajax@redhat.com> 1979 1980commit 20f3830ddf868c0d1a46c18fa381585c076bfb8d 1981Author: Maarten Lankhorst <maarten.lankhorst@canonical.com> 1982Date: Wed Sep 12 13:42:43 2012 +0200 1983 1984 bump version to 1.0.2 1985 1986commit 239f39977ebb64c9d545c6f3c905de3c0ad840d0 1987Author: Dave Airlie <airlied@redhat.com> 1988Date: Thu Jul 19 13:46:44 2012 +1000 1989 1990 nouveau: enable provider/capabilities 1991 1992 this adds the provider object for nouveau and enables the capabilities 1993 on it. 1994 1995 Signed-off-by: Dave Airlie <airlied@redhat.com> 1996 1997commit 28abba276e36fd0176f416996d44297ddf13f5cd 1998Author: Dave Airlie <airlied@redhat.com> 1999Date: Thu Jul 19 13:46:12 2012 +1000 2000 2001 nouveau/dri2: add support for DRI2 offload 2002 2003 This adds the nouveau interfaces for DRI2 offload, this allows nouveau 2004 to act as an offload slave. 2005 2006 Signed-off-by: Dave Airlie <airlied@redhat.com> 2007 2008commit b0be0ac86bae871c9f1724c40298d50d7f2b08e4 2009Author: Dave Airlie <airlied@redhat.com> 2010Date: Thu Jul 19 13:45:32 2012 +1000 2011 2012 nouveau: add pixmap tracking/updating support. 2013 2014 This should allow nouveau to act as an output master for driving output 2015 slaves. 2016 2017 Signed-off-by: Dave Airlie <airlied@redhat.com> 2018 2019commit baf485b1d304d741073f058d524ca698af3c15cd 2020Author: Dave Airlie <airlied@redhat.com> 2021Date: Thu Jul 19 13:42:13 2012 +1000 2022 2023 nouveau/exa: add support for shared pixmaps. 2024 2025 This adds the pixmap export/import hooks, along with code for nv50/nvc0 2026 to validate shared buffers in GTT space. 2027 2028 Signed-off-by: Dave Airlie <airlied@redhat.com> 2029 2030commit 2f48b8f462a03cb92db9e9a7ae1957eb27473965 2031Author: Dave Airlie <airlied@redhat.com> 2032Date: Thu Jul 19 13:40:23 2012 +1000 2033 2034 nouveau: add platform bus support 2035 2036 This adds the ability to load nouveau via the new platform apis. 2037 2038 Signed-off-by: Dave Airlie <airlied@redhat.com> 2039 2040commit 3aa9a296f508089a247f92b829036ca6a55b7b42 2041Author: Adam Jackson <ajax@redhat.com> 2042Date: Wed Aug 15 13:26:51 2012 -0400 2043 2044 Remove vbe and int10 includes 2045 2046 This driver has never used int10 or vbe, presumably these were just 2047 copypasta from nv. 2048 2049 Signed-off-by: Adam Jackson <ajax@redhat.com> 2050 2051commit 9b9be11e99930136a591059db1bf4ceaca8806e7 2052Author: Adam Jackson <ajax@redhat.com> 2053Date: Thu Jul 19 18:22:44 2012 -0400 2054 2055 Implement ->driverFunc 2056 2057 Copied from fbdev, makes it so we can run without iopl. 2058 2059 Signed-off-by: Adam Jackson <ajax@redhat.com> 2060 2061commit 6629066b87142cdd1d9e881b61ef55c97aedf949 2062Author: Ben Skeggs <bskeggs@redhat.com> 2063Date: Fri Aug 3 17:05:33 2012 +1000 2064 2065 nvc0/nve0: support sync-to-vblank if kernel is new enough 2066 2067 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2068 2069commit d1bc38b6673c54af61196056c489383fba8dced8 2070Author: Ben Skeggs <bskeggs@redhat.com> 2071Date: Fri Jul 6 16:23:50 2012 +1000 2072 2073 nv40/exa: fix shaders on big-endian machines 2074 2075 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2076 2077commit 0d2b414aff634c996e0468bb0df1894d6c28c09e 2078Author: Ben Skeggs <bskeggs@redhat.com> 2079Date: Tue Jun 19 10:12:37 2012 +1000 2080 2081 bump version to 1.0.1 2082 2083commit f84c28c9e6a0079e527655994689cb9c2ba51999 2084Author: Ben Skeggs <bskeggs@redhat.com> 2085Date: Tue Jun 19 09:49:02 2012 +1000 2086 2087 exa: fix remaining signed datatype issues 2088 2089 Wasn't expecting this situation, apparently it happens.. 2090 2091 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2092 2093commit 794141f22179a09ba6b2a094ba72316c298fee8b 2094Author: Maarten Maathuis <madman2003@gmail.com> 2095Date: Tue Jun 19 00:28:59 2012 +0200 2096 2097 exa: fix various issues in transfer bo handling 2098 2099 Thanks to "Christoph Bumiller <e0425955@student.tuwien.ac.at>" for hinting 2100 at the cause of the memory leak. 2101 2102 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 2103 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2104 2105commit 661d0cf42219a063f8faa0518f3490ddd38209ff 2106Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 2107Date: Mon Jun 18 17:13:55 2012 +0200 2108 2109 nve0: fix and improve NV12 shader 2110 2111 Only the 1st set of coordinates is valid. 2112 NVC0 was fixed in 9a3e579f637267b32efc46cfce5d1a36a41323b1. 2113 2114commit a907efe44177298cb1c42a236aab087c14b71657 2115Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 2116Date: Mon Jun 18 13:50:27 2012 +0200 2117 2118 nvc0/accel: oops fix method size in previous patch 2119 2120commit 8805e20bb9c1cfb7477f3eb7def64c5711fc91ea 2121Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 2122Date: Mon Jun 18 13:41:43 2012 +0200 2123 2124 nvc0/accel: use correct (signed) vertex data type in PUSH_VTX1/2s 2125 2126 Could have kept s16 but decided against mask and shift. 2127 2128commit 6dbf4ea12600275775123f9f564469454415da55 2129Author: Ben Skeggs <bskeggs@redhat.com> 2130Date: Mon Jun 18 13:28:07 2012 +1000 2131 2132 bump version to 1.0.0 2133 2134 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2135 2136commit 5e74cee90eeb4e8e638b371a5f1a2c6737be70ff 2137Author: Ben Skeggs <bskeggs@redhat.com> 2138Date: Mon Jun 18 15:36:47 2012 +1000 2139 2140 distcheck fixes 2141 2142commit 383aef5a8882a65efdf34e8d2681cd5ccec51b22 2143Author: Ben Skeggs <bskeggs@redhat.com> 2144Date: Mon Jun 18 13:31:37 2012 +1000 2145 2146 exa: drop massive gart allocation in favour of smaller ones, as needed 2147 2148 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2149 2150commit 52836c15b1c0e4f28d92ca9e6c76872481917fd4 2151Author: Ben Skeggs <bskeggs@redhat.com> 2152Date: Mon Jun 18 13:20:18 2012 +1000 2153 2154 nv50/exa: somehow fix firefox tabs not being drawn properly since vp xfrm 2155 2156 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2157 2158commit 36d3f8c28c55c05ccf4efa985831d2d249bad2fe 2159Author: Ben Skeggs <bskeggs@redhat.com> 2160Date: Wed Jun 13 22:41:58 2012 +1000 2161 2162 dri2: kill debug spam 2163 2164 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2165 2166commit 4dbc132f22721e3da30eb2e7fc97dea5b8458df6 2167Author: Viktor Novotný <noviktor@seznam.cz> 2168Date: Sat May 26 22:15:20 2012 +0200 2169 2170 nv04/exa: Reset destination surface offset in the same call of NV04EXACopy 2171 2172 Fixes FDO bug #48954. 2173 2174 Signed-off-by: Viktor Novotný <noviktor@seznam.cz> 2175 Reviewed-by: Marcin Slusarz <marcin.slusarz@gmail.com> 2176 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2177 2178commit aab58ee9a8025422416458d637c203af5ba008fd 2179Author: Dave Airlie <airlied@redhat.com> 2180Date: Tue Jun 5 12:34:06 2012 +0100 2181 2182 nouveau: i is used inside the function 2183 2184 fixes build, reported by tallica on irc. 2185 2186 Signed-off-by: Dave Airlie <airlied@redhat.com> 2187 2188commit 619e99731f772ff8d9d93cd1d6d83de5d450574b 2189Author: Dave Airlie <airlied@redhat.com> 2190Date: Tue Jun 5 11:02:54 2012 +0100 2191 2192 nouveau: port to compat server API. 2193 2194 This ports to the new server API. 2195 2196 Signed-off-by: Dave Airlie <airlied@redhat.com> 2197 2198commit d2e16c62b5054f34a439e504bfa261fb71f5d7de 2199Author: Dave Airlie <airlied@redhat.com> 2200Date: Tue Jun 5 10:57:30 2012 +0100 2201 2202 nouveau: drop flags arg to adjust frame 2203 2204 Signed-off-by: Dave Airlie <airlied@redhat.com> 2205 2206commit fadf83d7b373282ccbf0fa0c01928a35ff717a5e 2207Author: Ben Skeggs <bskeggs@redhat.com> 2208Date: Thu May 31 15:46:57 2012 +1000 2209 2210 nvc0/exa: add support for async UTS/DFS copies 2211 2212 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2213 2214commit 1600f7f202721c4b29c29e94f7fb69dc313b99dc 2215Author: Ben Skeggs <bskeggs@redhat.com> 2216Date: Thu May 31 15:40:45 2012 +1000 2217 2218 nva3/exa: add support for async UTS/DFS copies 2219 2220 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2221 2222commit ace77b6b1304826f4004bde23809b55d476b0615 2223Author: Ben Skeggs <bskeggs@redhat.com> 2224Date: Tue May 29 21:21:57 2012 +1000 2225 2226 disable fermi accel on 0.0.16 interface 2227 2228 Kepler accel support broke some assumption made by the older kernel 2229 interface, and Fermi shares the same code. It can't work (without 2230 some annoying hacks anyway) with the 0.0.16 kernel anymore. 2231 2232commit 7041e30ab8beb627bbf569367961a658e79c2bdc 2233Author: Dave Airlie <airlied@redhat.com> 2234Date: Wed May 23 14:18:24 2012 +0100 2235 2236 vl_hwmc: add missing compat include. 2237 2238 Reported-by: tallica on irc. 2239 Signed-off-by: Dave Airlie <airlied@redhat.com> 2240 2241commit 2abf8467cfb7a7648ce73ba5bcbbc62219d65d6d 2242Author: Dave Airlie <airlied@redhat.com> 2243Date: Wed May 23 11:29:05 2012 +0100 2244 2245 nouveau: add compat-api.h to makefile. 2246 2247 Signed-off-by: Dave Airlie <airlied@redhat.com> 2248 2249commit 1d861ad716861c57b2b81531d21840d7c8de024b 2250Author: Dave Airlie <airlied@redhat.com> 2251Date: Wed May 23 11:15:06 2012 +0100 2252 2253 nouveau: convert two more xf86Screens access to macros 2254 2255 for some reason the script missed these two, just fix them manually. 2256 2257 Signed-off-by: Dave Airlie <airlied@redhat.com> 2258 2259commit 5625fb84efc699e65da0062ae101915a49f2969b 2260Author: Dave Airlie <airlied@redhat.com> 2261Date: Wed May 23 11:13:30 2012 +0100 2262 2263 nouveau: convert scrn/screen to using new interfaces 2264 2265 This commit was generated with the util/modular/x-driver-screen-scrn-conv.sh 2266 2267 Signed-off-by: Dave Airlie <airlied@redhat.com> 2268 2269commit 9f037f7be516ebf8fa4a67a5c9ff258594ab353e 2270Author: Dave Airlie <airlied@redhat.com> 2271Date: Wed May 23 11:12:59 2012 +0100 2272 2273 nouveau: add initial compat-api header 2274 2275 This just provides the scrn conversion and picture stuff so far. 2276 2277 Signed-off-by: Dave Airlie <airlied@redhat.com> 2278 2279commit 581564466c47ab0435ecca3cf4d82c03e1b30259 2280Author: Francisco Jerez <currojerez@riseup.net> 2281Date: Tue May 8 14:02:44 2012 +0200 2282 2283 dri2: Don't try to page-flip pixmaps (fdo bug 49351). 2284 2285 Reported-by: Marcin Slusarz <marcin.slusarz@gmail.com> 2286 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 2287 2288commit b1efc807392d372e6a232aad9d16ae6e1b1c1916 2289Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2290Date: Sun Apr 15 13:46:42 2012 +0200 2291 2292 nv50: add missing pushbuf space check in NV50SyncToVBlank 2293 2294 Regression from "WIP: port to new libdrm". 2295 2296commit 174f17072005971fee6e4eb76a273c3cc1aa10d3 2297Author: Ben Skeggs <bskeggs@redhat.com> 2298Date: Tue Apr 24 21:55:03 2012 +1000 2299 2300 nv40/exa: return false from gradient picture setup 2301 2302 Not implemented yet. Shouldn't hit this anyway, fallback in check_composite, 2303 but, best be safe :) 2304 2305 Reported-by: Maarten Maathuis <madman2003@gmail.com> 2306 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2307 2308commit 60edf2a87b928f413385443335493cb27da30a48 2309Author: Ben Skeggs <bskeggs@redhat.com> 2310Date: Tue Apr 24 10:54:51 2012 +1000 2311 2312 nv40/exa: support for solid pictures 2313 2314 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2315 2316commit 067b8416ad1ab3e71ebbb8db8bfb82ff6041939b 2317Author: Ben Skeggs <bskeggs@redhat.com> 2318Date: Mon Apr 23 17:49:57 2012 +1000 2319 2320 nv30/exa: implement support for solid pictures using register combiners 2321 2322 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2323 2324commit b09a61780ea6710b679a6a099e35987e7a5f559f 2325Author: Ben Skeggs <bskeggs@redhat.com> 2326Date: Mon Apr 23 14:49:22 2012 +1000 2327 2328 nv10/exa: implement support for solid pictures 2329 2330 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2331 2332commit 9583753a205c585ec844dfb7168513a2c2a157ba 2333Author: Ben Skeggs <bskeggs@redhat.com> 2334Date: Fri Apr 20 20:52:01 2012 +1000 2335 2336 nv10/exa: use fixed-function texture matrix for transforms 2337 2338 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2339 2340commit c21c0ae57717bfb0fceab82505009a5382d1428d 2341Author: Ben Skeggs <bskeggs@redhat.com> 2342Date: Fri Apr 20 20:30:44 2012 +1000 2343 2344 nv30/exa: use fixed-function texture matrix for src/mask transforms 2345 2346 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2347 2348commit 24a5ca723539a940bc2dabcc50aa631f0d9d4219 2349Author: Ben Skeggs <bskeggs@redhat.com> 2350Date: Fri Apr 20 15:24:45 2012 +1000 2351 2352 nv40/exa: do transforms in vertex program 2353 2354 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2355 2356commit 9a3e579f637267b32efc46cfce5d1a36a41323b1 2357Author: Ben Skeggs <bskeggs@redhat.com> 2358Date: Thu Apr 19 16:58:52 2012 +1000 2359 2360 nvc0/exa: port recent nv50 changes (vp transform, solid pictures, etc) 2361 2362 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2363 2364commit 8ea09db1a1e8dff7c341dc3da17edefda7e56e6d 2365Author: Ben Skeggs <bskeggs@redhat.com> 2366Date: Thu Apr 19 13:46:03 2012 +1000 2367 2368 nv50/exa: support solid pictures 2369 2370 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2371 2372commit b48bcc094beecf521899dd63c8fdbccfd534e5cd 2373Author: Ben Skeggs <bskeggs@redhat.com> 2374Date: Thu Apr 19 13:21:27 2012 +1000 2375 2376 nv50/exa: perform texcoord transformations in vertex program 2377 2378 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2379 2380commit 5ac2ca8c56ec8b055878c8ac4cbc8ca74379abda 2381Author: Ben Skeggs <bskeggs@redhat.com> 2382Date: Thu Apr 19 09:52:24 2012 +1000 2383 2384 nv50: implement and use PUSH_DATAu 2385 2386 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2387 2388commit 6c41b3d7a8799daa9d0b34dcfbfc5c891c24fffa 2389Author: Ben Skeggs <bskeggs@redhat.com> 2390Date: Tue Apr 17 11:07:05 2012 +1000 2391 2392 nv30/nv40: upload shaders directly from rankine/curie init 2393 2394 Also removes A8 surface shader hack, and implements proper shaders for 2395 the specific purpose. 2396 2397 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2398 2399commit ec019b3a50475b7ff1b5e0a5dbe14e5ef677bd9e 2400Author: Ben Skeggs <bskeggs@redhat.com> 2401Date: Tue Apr 17 09:30:49 2012 +1000 2402 2403 Keep a single buffer for random accel data, rather than 3 different ones 2404 2405 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2406 2407commit fb3a36b1e5af0f81bb266da894d3442eed8e4e55 2408Author: Ben Skeggs <bskeggs@redhat.com> 2409Date: Wed Mar 28 16:38:30 2012 +1000 2410 2411 nve0: initial exa/xv acceleration for kepler chipsets 2412 2413 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2414 2415commit 95be16d8cd88db1ace0ac4568b4c67b51fd30fd9 2416Author: Ben Skeggs <bskeggs@redhat.com> 2417Date: Thu Mar 29 19:55:27 2012 +1000 2418 2419 nvc0: create scratch buffer before initialising m2mf 2420 2421 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2422 2423commit 4896c87484e1cf1e5421de377b6b83890948664d 2424Author: Ben Skeggs <bskeggs@redhat.com> 2425Date: Thu Mar 29 19:47:21 2012 +1000 2426 2427 nvc0: switch to kepler's subchannel assignments (see note) 2428 2429 This will bust for kernels with nouveau ABI version < 1.0.0... It should 2430 be possible to add compat for 0.0.16 however... 2431 2432 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2433 2434commit 4225b486cb8ab98d27ba23874096ff0879d45575 2435Author: Ben Skeggs <bskeggs@redhat.com> 2436Date: Thu Mar 29 15:56:02 2012 +1000 2437 2438 nvc0: tidy and cull a heap of fermi init (stuff in default grctx) 2439 2440 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2441 2442commit 007dd62a8c5244bc5e71fe8f32520fb35b67a237 2443Author: Ben Skeggs <bskeggs@redhat.com> 2444Date: Thu Mar 29 15:28:03 2012 +1000 2445 2446 nvc0: punt shaders out to a header file 2447 2448 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2449 2450commit 1c7651532975cf738b3aac4ec3c3b46187e24eea 2451Author: Ben Skeggs <bskeggs@redhat.com> 2452Date: Thu Mar 29 14:25:15 2012 +1000 2453 2454 nvc0: do tic/tsc binding only once, at startup 2455 2456 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2457 2458commit 180d8778764913ed061e68520603c0925477faba 2459Author: Ben Skeggs <bskeggs@redhat.com> 2460Date: Thu Mar 29 14:21:25 2012 +1000 2461 2462 nvc0: replace direct inline m2mf with inline helper to do the same 2463 2464 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2465 2466commit e70d801ae9287eab5e82f4d467dc8cd4be1b31a8 2467Author: Ben Skeggs <bskeggs@redhat.com> 2468Date: Fri Dec 9 18:15:32 2011 +1000 2469 2470 WIP: port to new libdrm 2471 2472 calim: nv50,nvc0/xv: fix src_w/h width/height confusion 2473 2474commit ab7291d368ddc66ad21c5ad5caa0faeee42ccaf1 2475Author: Ben Skeggs <bskeggs@redhat.com> 2476Date: Wed Mar 21 14:56:55 2012 +1000 2477 2478 xv: don't try to init textured video without 3d engine object 2479 2480 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2481 2482commit fe7e0425a4a43b2a7d4743a0a90a86724c3b6775 2483Author: Ben Skeggs <bskeggs@redhat.com> 2484Date: Mon Mar 12 11:59:29 2012 +1000 2485 2486 recognise kepler 2487 2488 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2489 2490commit f5d1cd2cb6808838ae1a188cef888eaa9582c76d 2491Author: Ben Skeggs <bskeggs@redhat.com> 2492Date: Tue Mar 6 08:16:51 2012 +1000 2493 2494 remove use of xf86PciInfo.h, deprecated now apparently 2495 2496 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2497 2498commit 248de8cdbd6d0bc062633b49896fa4791148cd3b 2499Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2500Date: Thu Feb 16 00:45:24 2012 +0100 2501 2502 dri2: Fix corner case crash for swaplimit > 1 2503 2504 If a swaplimit > 1 is set on a server which 2505 supports the swaplimit api (XOrg 1.12.0+), 2506 the following can happen: 2507 2508 1. Client calls glXSwapBuffersMscOML() with a 2509 swap target > 1 vblank in the future, or a 2510 client calls glXSwapbuffers() while the swap 2511 interval is set to > 1 (unusual but possible). 2512 2513 2. nouveau_dri2_finish_swap() is therefore called 2514 only at the target vblank, instead of immediately. 2515 2516 3. Because of the deferred execution of 2517 nouveu_dri2_finish_swap(), the OpenGL client 2518 can call x-servers DRI2GetBuffersWithFormat() 2519 before nouveau_dri2_finish_swap() executes and 2520 it deletes pixmaps that would be needed by 2521 nouveau_dri2_finish_swap() --> Segfault --> Crash. 2522 2523 Prevent this: When a swap is scheduled into the 2524 future, we temporarily reduce the swaplimit to 1 2525 until nouveau_dri2_finish_swap() is done, then 2526 restore it to its original value. This throttles 2527 the client inside the server in DRI2ThrottleClient() 2528 before it can call the evil DRI2GetbuffersWithFormat(). 2529 2530 The client will still be released one video refresh 2531 interval before swap completion, so there is still 2532 some potential win. 2533 2534 This doesn't affect the common case of swapping at 2535 the next vblank, where this throttling is not needed 2536 or done. 2537 2538 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2539 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2540 2541commit 247465ec5e21ce7151ef1e6a3e7644bc74d48a5b 2542Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2543Date: Thu Feb 16 00:45:23 2012 +0100 2544 2545 dri2: Reimplement hack for triple-buffering on old X-Servers. 2546 2547 X-Servers before 1.12.0 don't have the DRI2SwapLimit() 2548 API. On these, default to a swaplimit of 1 - double-buffering. 2549 2550 This patch implements support for swap limit of 2, 2551 triple-buffering, on old x-servers via Francisco Jerez 2552 previous hack: 2553 2554 Return DRI2SwapComplete() before the swap has completed, 2555 so clients don't get blocked on the pending swap. This 2556 allows for a "triple-buffering look-alike" behaviour, but 2557 breaks the swap scheduling and timestamping defined 2558 in the OML_sync_control spec, so applications which 2559 rely on conformant behaviour will break with a swap 2560 limit of 2 on pre 1.12.0 x-servers. 2561 2562 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2563 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2564 2565commit a1902ffaab1f6807fdaac0fce49f784e7208d1a5 2566Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2567Date: Thu Feb 16 00:45:22 2012 +0100 2568 2569 dri2: Allow vblank controlled swaps for redirected windows. Part II 2570 2571 This part implements proper throttling for clients. For 2572 vblank synchronized blits, it defers DRI2SwapComplete() 2573 until 1 vblank after the framebuffer blit is submitted to 2574 the gpu. 2575 2576 Rationale: 2577 2578 For unredirected windows, this is the earliest time the 2579 "blit swap" can complete, as blits are submitted one vblank 2580 before the target vblank and synchronized with vblank in the 2581 gpu. This makes swap completion timestamps at least reasonable. 2582 2583 For redirected windows, the compositor will probably pick 2584 up the "blit swapped" frontbuffer pixmap of the window quickly, 2585 but defer its own recomposition to the next vblank, at least 2586 if sync to vblank for the compositor is on. 2587 2588 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2589 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2590 2591commit c8572d329014afc0c79b5765468dcdfb65be0868 2592Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2593Date: Thu Feb 16 00:45:21 2012 +0100 2594 2595 dri2: Allow vblank controlled swaps for redirected windows. Part I 2596 2597 Make sure that swaps for redirected windows under a 2598 compositor are still scheduled via vblank events, to 2599 avoid effects like 2900 fps swaps under a compositor. 2600 2601 See discussion with Francisco Jerez at: 2602 2603 http://lists.freedesktop.org/archives/nouveau/2011-September/009278.html 2604 http://lists.freedesktop.org/archives/nouveau/2011-September/009292.html 2605 2606 This is part I of the agreed upon band-aid, in a separate patch. 2607 2608 It allows to use vblank related functions on redirected 2609 windows and thereby fixes functions from sgi_sync_control 2610 and oml_sync_control extension, e.g., glXWaitForMscOML(), 2611 glXGetSyncValuesOML(), glXWaitVideoSyncSGI, ... 2612 2613 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2614 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2615 2616commit 919643888045b102f3e10dfdeb6210ab14b6ecab 2617Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2618Date: Thu Feb 16 00:45:20 2012 +0100 2619 2620 dri2: Fixes to swap scheduling. 2621 2622 Fix some small off-by-one errors and a mismatch 2623 between 32 bit kernel interfaces for vblank count 2624 and 64 bit dri2 interfaces for target_msc et al. 2625 2626 Return corrected target_msc to swap scheduling in 2627 x-server. 2628 2629 A revised version of the patch discussed here: 2630 http://lists.freedesktop.org/archives/nouveau/2011-September/009143.html 2631 2632 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2633 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2634 2635commit 5ede4538366a352aa9726560f8d42e85df4aec82 2636Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2637Date: Thu Feb 16 00:45:19 2012 +0100 2638 2639 dri2: Update front buffer pixmap and name before exchanging buffers 2640 2641 Buffer exchange assumes that the front buffer pixmap and name 2642 information is accurate. That may not be the case eg. if the window 2643 has been (un)redirected since the buffer was created. 2644 2645 This is a translation to nouveau of a fix that was originally developed 2646 by Ville Syrjala <syrjala@sci.fi> for the ati/radeon ddx to fix the 2647 same bug there. 2648 2649 See thread at: 2650 2651 http://lists.x.org/archives/xorg-devel/2011-May/021908.html 2652 2653 Fixes FDO bug #35452. 2654 2655 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2656 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2657 2658commit 6fdf60b7288d49f889c80705aabc1db3bc327ba3 2659Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2660Date: Thu Feb 16 00:45:18 2012 +0100 2661 2662 dri2: Add support for DRI2SwapLimit() API. 2663 2664 Uses the new DRI2SwapLimit() API of X-Server 1.12+ 2665 to allow to change the maximum number of pending 2666 swaps on a drawable before the OpenGL client is 2667 throttled by the server. 2668 2669 The new optional xorg.conf parameter "SwapLimit" 2670 allows to select a new swap limit >= 1. The default 2671 swap limit is 2 for triple-buffering on XOrg 1.12+, 2672 1 for double-buffering on older servers, as we can't 2673 change the swap limit there. 2674 2675 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2676 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2677 2678commit 820916e722db2997a53128f2f370d14a02d4401c 2679Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2680Date: Thu Feb 16 00:45:17 2012 +0100 2681 2682 dri2: Implement handling of pageflip completion events. 2683 2684 Requests pageflip completion events from the kernel. 2685 Implements pageflip completion handler to finalize 2686 and timestamp swaps. 2687 2688 Completion handler includes a consistency check, and 2689 disambiguation if multiple crtc's are involved in a 2690 pageflip (e.g., clone mode, extendend desktop). Only 2691 the timestamp of the crtc whose vblank event initially 2692 triggered the swap is used, but handler waits for flip 2693 completion on all involved crtc's before completing the 2694 swap and releasing the old framebuffer. 2695 2696 This code is almost identical to the code used in the 2697 ati/radeon ddx and intel ddx. 2698 2699 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2700 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2701 2702commit 4da68cedbefef4f72c00bb088e179071490882c8 2703Author: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2704Date: Thu Mar 1 19:11:14 2012 +0100 2705 2706 dri2: Fix can_exchange() to allow page-flipping on more mesa versions. 2707 2708 can_exchange() returns false and thereby prevents page 2709 flipping on some drawables where page flipping would 2710 work fine. This due to non-matching drawable depths values 2711 between front buffer pixmap and back buffer pixmap, because 2712 front buffer pixmaps inherit the depth of the screen, typically 2713 24 bits, whereas the depth value of back buffer pixmaps for a 2714 given RGB8 or RGBA8 visual depends on the mesa version in use, 2715 either 24 bits or 32 bits. 2716 2717 Use bitsPerPixel instead of depth to decide if drawable 2718 is flippable. This will still catch really incompatible 2719 formats like 32 bpp vs. 16 bpp buffers. 2720 2721 Tested for screen DefaultDepth 24 and also 30 bits 2722 (for RGB10 framebuffers) on NV-50. 2723 2724 The problem was fixed in the same way in the ati & intel ddx. 2725 2726 Signed-off-by: Mario Kleiner <mario.kleiner@tuebingen.mpg.de> 2727 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2728 2729commit b0d7f4ddd94154491d04dcb9698d63d61e9240dc 2730Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 2731Date: Mon Jan 9 15:17:27 2012 +0100 2732 2733 dri2: CreateBuffer format may be 0, use depth in that case 2734 2735commit 3d9f6b32e0ad72dc8c7db07d039d87e83d8fdb1d 2736Author: Ben Skeggs <bskeggs@redhat.com> 2737Date: Fri Dec 9 17:45:31 2011 +1000 2738 2739 nv50/xv: bump max dims to 8k 2740 2741 Untested, but I don't see why this wouldn't work... 2742 2743commit 57c62228e2e67370feeeb421c369fbf940d30ca4 2744Author: Ben Skeggs <bskeggs@redhat.com> 2745Date: Fri Dec 9 17:44:07 2011 +1000 2746 2747 xv: stop using custom m2mf code 2748 2749commit ad402333157c437ccd2c639f5b3f3bcb81e78d6c 2750Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 2751Date: Sat Dec 10 13:32:24 2011 +0100 2752 2753 dri2: fix allocation of Z16 depth attachments 2754 2755commit 3d2a752ca8aafee5e1e94dabfd7deec439890e95 2756Author: Ben Skeggs <bskeggs@redhat.com> 2757Date: Thu Dec 1 11:31:49 2011 +1000 2758 2759 move away from libdrm's BEGIN_RING (and variants) macros 2760 2761 subchan auto-assignment is now gone. 2762 2763commit b6cfdc01aeb5c2ab401ee81fff66bd0560abdfcc 2764Author: Ben Skeggs <bskeggs@redhat.com> 2765Date: Thu Dec 1 11:25:10 2011 +1000 2766 2767 Append _CLASS to object class id definitions 2768 2769 I want to be able to use ring macros of the same name 2770 2771commit 1dadb5826363d0d8ed1a72209be18277550f8345 2772Author: Ben Skeggs <bskeggs@redhat.com> 2773Date: Thu Dec 1 09:06:24 2011 +1000 2774 2775 exa: create general m2mf rect copy for each chipset, use for UTS/DFS 2776 2777 No point having duplicated UTS/DFS logic in each implementation. This also 2778 splits nv50 out into its own code, in preparation for it being made to not 2779 use relocs at all. 2780 2781 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2782 2783commit b5534a15a5f5e1c3bd6c92e3ed8391ff628c09bb 2784Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2785Date: Wed Nov 30 18:56:54 2011 +0100 2786 2787 fix typo in conversion to rnn headers 2788 2789commit 5975ce34a12d1943b2852d0f5fbe09358754c2ad 2790Author: Ben Skeggs <bskeggs@redhat.com> 2791Date: Wed Nov 30 17:40:02 2011 +1000 2792 2793 convert entire ddx to rnn headers 2794 2795commit 9fa0c6c74d0b9f7d66ebe3c99184eaa13abeb1a0 2796Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2797Date: Sun Nov 6 20:32:10 2011 +0100 2798 2799 Default to 16bpp for low memory cards. 2800 2801commit 5aa30ae6554c957609929ef29d1b27c89e9a7f06 2802Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2803Date: Sun Nov 6 20:32:09 2011 +0100 2804 2805 nv50/exa: wait for enough space for at least one composite operation 2806 2807commit 3d54426c4833a7e058e50624452f89139921e64b 2808Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2809Date: Sun Nov 6 20:32:08 2011 +0100 2810 2811 add missing return in nv50_accel_init failure path 2812 2813commit d575a28a5020ff083836d3d1cf50018a99f49c9f 2814Author: Maarten Lankhorst <m.b.lankhorst@gmail.com> 2815Date: Sat Oct 8 10:33:20 2011 +0200 2816 2817 dri2: Add vdpau driver name entry 2818 2819 libvdpau has a driver loading mechanism that looks for a dri2 driver 2820 first before falling back to nvidia, so lets use that. 2821 2822commit 656cfe3bdb8b487827b710a0cea35ae1bcd5044e 2823Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2824Date: Sat Sep 10 23:14:53 2011 +0200 2825 2826 xv: support AI44/IA44 2827 2828 It's needed for osd in mplayer's xvmc output. 2829 2830commit 169512fbe91f0671a90dfee5e280357f0a4ef701 2831Author: Maxim Levitsky <maximlevitsky@gmail.com> 2832Date: Tue Aug 23 13:10:59 2011 +0200 2833 2834 dri2: Disable the "exchange" swapbuffers path for the moment (bug 35930). 2835 2836 Allow page flipping only for scanout buffer for now as simple swapping 2837 between off-screen pixmaps confuses compiz because there is no 2838 syncronization method to tell it about that swap 2839 2840 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 2841 2842commit b806e3f97a73701f057c45b1f45233e69e19f113 2843Author: Bryan Cain <bryancain3@gmail.com> 2844Date: Wed Jul 20 13:48:36 2011 -0500 2845 2846 g3dvl: update for compatibility with latest pipe-video branch in Mesa 2847 2848 Signed-off-by: Bryan Cain <bryancain3@gmail.com> 2849 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2850 2851commit de9d1ba7efeba64f319efa00df183d3cb78f24af 2852Author: Emil Velikov <emil.l.velikov@gmail.com> 2853Date: Tue Jul 19 00:35:38 2011 +0100 2854 2855 drmmode_display: Resolve missing brackets 2856 2857 Correct some missing/misplaced brackets in drmmode_pre_init() 2858 The issue was exposed when trying a 4 monitor desktop using two 2859 cards/gpus 2860 2861 Resolves https://bugs.freedesktop.org/show_bug.cgi?id=39099 2862 2863 Reported-By: Damian Nowak <nowaker@geozone.pl> 2864 Tested-By: Damian Nowak <nowaker@geozone.pl> 2865 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 2866 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2867 2868commit ef4957492956df54fafa6c75d576a07ce678ea0a 2869Author: Ben Skeggs <bskeggs@redhat.com> 2870Date: Mon Jul 11 12:48:34 2011 +1000 2871 2872 kms: fix multiple rotations in noaccel mode 2873 2874commit ab89aa022f0c78154e525b5f5787feee8024171f 2875Author: Ben Skeggs <bskeggs@redhat.com> 2876Date: Fri Jun 17 10:10:09 2011 +1000 2877 2878 nvc0: accept 0xdx chipsets 2879 2880 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2881 2882commit 7dd8987b8b69f56d04a199207122d2e51b9caa6e 2883Author: Ben Skeggs <bskeggs@redhat.com> 2884Date: Fri Jun 17 10:07:36 2011 +1000 2885 2886 nvc0: assume 0x9097's presence on all chipsets 2887 2888 This appears to be the case thus far, and is what the 3D driver does. 2889 2890 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2891 2892commit 98ee2105dbd8685e59c1967a587fccb6135743ac 2893Author: Ben Skeggs <bskeggs@redhat.com> 2894Date: Mon May 30 10:10:58 2011 +1000 2895 2896 kms: clear new fb to black on fb resize 2897 2898commit 8378443bd3b26b57ef2ae424a700e01ead813d33 2899Author: Ben Skeggs <bskeggs@redhat.com> 2900Date: Thu Mar 24 02:13:12 2011 +1000 2901 2902 nv50-nvc0/exa: fix bug causing surface state to not be reemitted after flush 2903 2904 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2905 2906commit 92db2bc192a074d4b7fc3e9c16e9aa62a8deaeb6 2907Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 2908Date: Thu Mar 3 22:17:23 2011 +0100 2909 2910 nvc0/accel: allow nvc1,nvc3,nvc4 in 3D engine init and use 9097 2911 2912commit ace98a492353e6de712f4f717e6d3f562e3591f0 2913Author: Ben Skeggs <bskeggs@redhat.com> 2914Date: Tue Mar 1 13:54:38 2011 +1000 2915 2916 dri2: return an error rather than crashing if we can't allocate a buffer 2917 2918 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2919 2920commit bc5dec2ca7ca7edc340a99bd73946e228117dfd8 2921Author: Ben Skeggs <bskeggs@redhat.com> 2922Date: Thu Feb 24 15:15:00 2011 +1000 2923 2924 dri2: disable page flipping if any crtc is rotated 2925 2926 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2927 2928commit c123618470b5ff32c6377a4cae0f148d4a518a5d 2929Author: Francesco Marella <francesco.marella@gmail.com> 2930Date: Thu Feb 17 03:48:16 2011 +0100 2931 2932 Fix gcc 4.6.0 set but unused variable warnings 2933 2934 Signed-off-by: Francesco Marella <fmarl@paranoici.org> 2935 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2936 2937commit 3531b2aa13e9c78288d96fe5e1492f0b6e899b00 2938Author: Francesco Marella <francesco.marella@gmail.com> 2939Date: Thu Feb 17 03:47:46 2011 +0100 2940 2941 Page flipping man page update 2942 2943 Signed-off-by: Francesco Marella <fmarl@paranoici.org> 2944 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2945 2946commit 85a706df6f7986158c964e3afd0aa8e352ff5cea 2947Author: Francesco Marella <francesco.marella@gmail.com> 2948Date: Thu Feb 17 03:47:08 2011 +0100 2949 2950 Update configure.ac with autoupdate 2951 2952 Signed-off-by: Francesco Marella <fmarl@paranoici.org> 2953 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2954 2955commit 42c16ff33dcca6cd9bae62fe0d71a40452e03581 2956Author: Ben Skeggs <bskeggs@redhat.com> 2957Date: Wed Feb 16 13:51:54 2011 +1000 2958 2959 nv50: attempt to allocate compressed Z buffers 2960 2961 This should automagically fallback to the non-compressed version on kernels 2962 that don't support compression. 2963 2964 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2965 2966commit 46acb7e095e9c0d55b90070b77c64679fcfb9d4e 2967Author: Marcin Slusarz <marcin.slusarz@gmail.com> 2968Date: Sun Feb 13 20:46:42 2011 +0100 2969 2970 fix drmSetMaster error reporting 2971 2972 On error, drmSetMaster returns -1 and sets errno. 2973 2974 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2975 2976commit 54764980411b4b8124c36edb8730612b04f292a9 2977Author: Maarten Maathuis <madman2003@gmail.com> 2978Date: Thu Feb 10 20:12:07 2011 +0100 2979 2980 Also flush from the BlockHandler. 2981 2982 - The fact that commit 02c50db307a4e4eeb87f7db5d401bc53cce3b21f 2983 fixed missing text for some people proves this is needed. 2984 That commit should have only improved latency a bit. 2985 - The FlushCallback is still needed for DRI2 texture_from_pixmap. 2986 - See bugs.fd.o #33977 for discussion. 2987 2988 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 2989 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 2990 2991commit 02c50db307a4e4eeb87f7db5d401bc53cce3b21f 2992Author: Maarten Maathuis <madman2003@gmail.com> 2993Date: Mon Feb 7 19:37:47 2011 +0100 2994 2995 exa: Flush (S)IFC to the frontbuffer immediately. 2996 2997 This avoids high latency while typing with core fonts for example. 2998 2999 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 3000 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3001 3002commit e4fb9fd2f790ed66e4be61ca00d55522525619f4 3003Author: Ben Skeggs <bskeggs@redhat.com> 3004Date: Wed Feb 9 13:46:32 2011 +1000 3005 3006 dri2: allow page flipping to be disabled in xorg.conf 3007 3008 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3009 3010commit 38e8809bb415bae5c182fc79c8fc62992c5e4ed0 3011Author: Ben Skeggs <bskeggs@redhat.com> 3012Date: Mon Jan 17 09:14:31 2011 +1000 3013 3014 nvc0: merge initial acceleration support 3015 3016 Squashed commit of the following: 3017 3018 commit e8ce1ef351ee51bec51211dca6fda88fdbfbefee 3019 Author: Ben Skeggs <bskeggs@redhat.com> 3020 Date: Mon Jan 17 08:55:04 2011 +1000 3021 3022 fix hardcoding of nvc0 wfb path 3023 3024 commit b414b1f948da18d96309bad84c54948d877efaa6 3025 Author: Ben Skeggs <bskeggs@redhat.com> 3026 Date: Mon Jan 17 08:52:53 2011 +1000 3027 3028 general tidy-ups in preparation for merge to master 3029 3030 commit 73bf71d5de0cb990178a1d2a31e0ea1717c214a4 3031 Merge: fb499a4 b795ca6 3032 Author: Ben Skeggs <bskeggs@redhat.com> 3033 Date: Mon Jan 17 08:47:40 2011 +1000 3034 3035 Merge branch 'master' into nvc0 3036 3037 commit fb499a4e9d95650dc89f4c1820b94d01344733f6 3038 Author: Ben Skeggs <bskeggs@redhat.com> 3039 Date: Mon Dec 20 12:54:43 2010 +1000 3040 3041 nvc0: switch to "standard" RING macros 3042 3043 commit 64f0e1d0e9e29c5437ad4e52d01ccfb0b03a2ae9 3044 Merge: eba7e17 c2092a4 3045 Author: Ben Skeggs <bskeggs@redhat.com> 3046 Date: Mon Dec 20 11:58:42 2010 +1000 3047 3048 Merge branch 'master' into nvc0 3049 3050 Conflicts: 3051 src/nv_dma.c 3052 3053 commit eba7e173a5006ff50b950d0e8b8154564471a9ce 3054 Author: Ben Skeggs <bskeggs@redhat.com> 3055 Date: Fri Dec 17 11:04:40 2010 +1000 3056 3057 nvc0/xv: use mad in nv12 shader, rather than mul+add 3058 3059 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3060 3061 commit b025ceb40a5e94a3657f3c7f76044c49a1a05e8d 3062 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3063 Date: Thu Dec 16 23:34:29 2010 +0100 3064 3065 nvc0: fix offset of 17bc buffer 3066 3067 I had messed it up in 5e4da3451c3c0f645f98fea7fb0ce2629b1bfd2f. 3068 3069 commit 6ff5a9a7f3411ef218f17fa481c583596ffde2cc 3070 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3071 Date: Thu Dec 16 01:07:25 2010 +0100 3072 3073 nvc0/xv: bind textures for the FP, not the TEP 3074 3075 commit 5e4da3451c3c0f645f98fea7fb0ce2629b1bfd2f 3076 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3077 Date: Thu Dec 16 01:02:58 2010 +0100 3078 3079 nvc0: switch to rnn headers 3080 3081 commit 0839041e79dbb0e629326d0f240220b33921db1a 3082 Author: Ben Skeggs <bskeggs@redhat.com> 3083 Date: Wed Dec 15 16:42:41 2010 +1000 3084 3085 nvc0/exa: fix m2mf dfs 3086 3087 commit 88c1c43302331e1cb8b7035bfb4677a5e7da3090 3088 Author: Ben Skeggs <bskeggs@redhat.com> 3089 Date: Mon Dec 13 12:00:23 2010 +1000 3090 3091 nvc0: add Xv implementation 3092 3093 commit ad5b4cfe388a102a0c93f7281cb4bba532b1796b 3094 Merge: 1942a81 b6cb21b 3095 Author: Ben Skeggs <bskeggs@redhat.com> 3096 Date: Wed Dec 15 14:57:04 2010 +1000 3097 3098 Merge branch 'master' into nvc0 3099 3100 commit 1942a81c0eb592972d4a3e8c46b80192ae8d1a59 3101 Author: Ben Skeggs <bskeggs@redhat.com> 3102 Date: Thu Dec 9 12:48:27 2010 +1000 3103 3104 nvc0: call grobj_alloc for all used object classes 3105 3106 Though the hardware has no concept of object handles any more, the nouveau 3107 DRM still uses this call to know what engines are in use. 3108 3109 commit 0c670a6d406fe0a57373fa9e03005be6f28ecc3e 3110 Author: Ben Skeggs <bskeggs@redhat.com> 3111 Date: Thu Dec 9 12:44:41 2010 +1000 3112 3113 nvc0: don't try and create DmaNotifier0 3114 3115 commit 5d3d20d908a50ad7742567d58969c7f12dd944f7 3116 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3117 Date: Thu Dec 9 01:36:32 2010 +0100 3118 3119 nvc0: change tile_mode to contain 0xZYX instead of 0xZY 3120 3121 Warning: the kernel will not like that (EVO) 3122 3123 commit dd7fd8263e0e9f92412fb20086e4c8acf8cab099 3124 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3125 Date: Thu Dec 9 00:29:58 2010 +0100 3126 3127 exa: don't fall back to nv04 m2mf if nvc0 m2mf fails 3128 3129 commit b4cf5a8f2276392945c2530f0d6fe96b5ab4b5da 3130 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3131 Date: Thu Dec 9 00:07:59 2010 +0100 3132 3133 nvc0/exa: remove noisy debug messages 3134 3135 commit 5419ef6d2cfabb7329982cb8dbdb5bbc8e973fe3 3136 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3137 Date: Wed Dec 8 23:55:51 2010 +0100 3138 3139 nvc0/accel: remove unneeded scratch method 0x3420 init 3140 3141 It was intended be read by a MACRO method to get the address for 3142 constants upload. 3143 3144 commit 178356391ff6831599eba257a5912079894641a1 3145 Merge: 8573280 8bb8231 3146 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3147 Date: Thu Dec 9 01:31:40 2010 +0100 3148 3149 Merge remote branch 'origin/master' into nvc0 3150 3151 Conflicts: 3152 src/drmmode_display.c 3153 src/nouveau_exa.c 3154 src/nv_driver.c 3155 src/nv_type.h 3156 3157 commit 8573280871401e29f27fa24d0b7c6ac5e35b2cc1 3158 Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3159 Date: Sat Jul 3 14:27:36 2010 +0200 3160 3161 nvc0: initial implementation 3162 3163 EXA should work, Xv not yet done. 3164 3165commit b795ca6e97fae9735843748585401098dae1c3e1 3166Author: Ben Skeggs <bskeggs@redhat.com> 3167Date: Fri Jan 7 13:33:36 2011 +1000 3168 3169 nv50: fix pitch of linear scanout buffers 3170 3171 Un-breaks NoAccel on these chipsets. 3172 3173 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3174 3175commit dc89dac734167bcbc667b39ca6ee2043871a60bf 3176Author: Xavier Chantry <chantry.xavier@gmail.com> 3177Date: Sun Dec 19 23:17:34 2010 +0100 3178 3179 nv04/exa: kill useless WAIT_RING 3180 3181 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3182 3183commit b76b43928bf4dcf5ce3a4b58c7fefe780070c0d3 3184Author: Ben Skeggs <bskeggs@redhat.com> 3185Date: Tue Dec 21 16:25:23 2010 +1000 3186 3187 add back generic include of nouveau_pushbuf.h 3188 3189 So places that FIRE_RING() still work. 3190 3191 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3192 3193commit c2092a458401377f421d9bdfe1d97ce580b99508 3194Author: Ben Skeggs <bskeggs@redhat.com> 3195Date: Mon Dec 20 11:50:57 2010 +1000 3196 3197 include nv04_pushbuf.h for ring macros, rather than nouveau_pushbuf.h 3198 3199 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3200 3201commit aa2821a42706ac7b69703d1869e2d00a4ced9f4b 3202Author: Ben Skeggs <bskeggs@redhat.com> 3203Date: Mon Dec 20 11:53:53 2010 +1000 3204 3205 kill NVSync(), it's rather useless 3206 3207 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3208 3209commit 1fc564fe3494cf0abcc848d0e90bf2232f8fd272 3210Author: Ben Skeggs <bskeggs@redhat.com> 3211Date: Thu Dec 16 09:49:09 2010 +1000 3212 3213 nv50/xv: use mad in nv12 shader, rather than mul+add 3214 3215 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3216 3217commit b6cb21b91f7d0409794a14ddbdd606728d4d49e3 3218Author: Ben Skeggs <bskeggs@redhat.com> 3219Date: Mon Feb 8 12:51:47 2010 +1000 3220 3221 set canDoBGNoneRoot if building against new enough server 3222 3223commit 11510f9ce8c9eab15e7f6d5962b83fc0052c8619 3224Author: Ben Skeggs <bskeggs@redhat.com> 3225Date: Tue Dec 14 15:39:01 2010 +1000 3226 3227 nv50/xv: implement brightness/contrast/saturation/hue controls 3228 3229 Adapted from the code in xf86-video-ati. 3230 3231 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3232 3233commit 214ea264cdb386520a1308a29f204e441f349cb1 3234Author: Ben Skeggs <bskeggs@redhat.com> 3235Date: Tue Dec 14 14:42:24 2010 +1000 3236 3237 nv50/xv: kick csc parameters out into a constant buffer 3238 3239 These were previously inlined into the shader, which is simple, but makes 3240 implementing brightness/contrast adjustments impossible. 3241 3242 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3243 3244commit 8bb8231236a6b877895663aeaa9cef731d67fe68 3245Author: Ben Skeggs <bskeggs@redhat.com> 3246Date: Tue Nov 30 15:27:36 2010 +1000 3247 3248 dri2: fix certain clients blocking forever when sync-to-vblank unavailable 3249 3250 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3251 3252commit f29afe65b67c9b89ceba421d9b3aee590f7ded20 3253Author: Ben Skeggs <bskeggs@redhat.com> 3254Date: Mon Nov 29 09:23:35 2010 +1000 3255 3256 dri2: don't return failure from get_msc if sync-to-vblank isn't available 3257 3258 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3259 3260commit 4b1d26578116f4d1745dd92a7c99413972a2bedf 3261Author: Francisco Jerez <currojerez@riseup.net> 3262Date: Thu Nov 25 21:33:46 2010 +0100 3263 3264 dri2: Don't return failure from schedule_wait if sync-to-vblank is unavailable. 3265 3266 Reported-by: Ben Skeggs <bskeggs@redhat.com> 3267 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3268 3269commit 1143e7a97dce1b741376e178b73b8b766e03849a 3270Author: Jesse Adkins <jesserayadkins@gmail.com> 3271Date: Tue Oct 26 16:05:32 2010 -0700 3272 3273 Purge cvs tags. 3274 3275 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com> 3276 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3277 3278commit 22020457f43eceaf26532ecb50e46d743f4a691f 3279Author: Francisco Jerez <currojerez@riseup.net> 3280Date: Tue Oct 26 00:24:06 2010 +0200 3281 3282 dri2: Fix throttling on occluded drawables. 3283 3284 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3285 3286commit 9737a812aacbc544b93eeaa031cf1c2e8adbd661 3287Author: Ben Skeggs <bskeggs@redhat.com> 3288Date: Mon Oct 25 10:37:08 2010 +1000 3289 3290 kms: initial pass at supporting zaphod-mode 3291 3292 This works ok from some simple testing, there's undoubtedly things that 3293 will need polishing still however. 3294 3295 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3296 3297commit d40c16f5497739c0215f37566db1651a74cb20ba 3298Author: Francisco Jerez <currojerez@riseup.net> 3299Date: Sat Oct 23 02:20:30 2010 +0200 3300 3301 dri2: Fix build on X server 1.8. 3302 3303 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3304 3305commit a4d580bf05d7aac1d486e5d1ba9a2201c026e5f5 3306Author: Francisco Jerez <currojerez@riseup.net> 3307Date: Fri Oct 22 23:09:30 2010 +0200 3308 3309 Bump the required X server version to 1.8. 3310 3311 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3312 3313commit d0f2e16bc35660d01d3cd20a770cf8e613ab5323 3314Author: Francisco Jerez <currojerez@riseup.net> 3315Date: Fri Oct 22 23:06:28 2010 +0200 3316 3317 dri2: Report damage on page-flipped drawables. 3318 3319 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3320 3321commit c61e8d03161adcfbba4162172153814accf9e86b 3322Author: Francisco Jerez <currojerez@riseup.net> 3323Date: Fri Oct 22 23:05:24 2010 +0200 3324 3325 Make the sync-to-vblank helpers take a box as argument. 3326 3327 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3328 3329commit a685a5c3be12f3e95a6ec1140af115e7ef28fcaa 3330Author: Francisco Jerez <currojerez@riseup.net> 3331Date: Wed Oct 13 18:45:04 2010 +0200 3332 3333 dri2: Implement the ScheduleWaitMSC/GetMSC hooks. 3334 3335 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3336 3337commit eb83c830c87bce345748edef3b50660246143db7 3338Author: Francisco Jerez <currojerez@riseup.net> 3339Date: Thu Oct 21 22:57:08 2010 +0200 3340 3341 dri2: Add pageflip/exchange support. 3342 3343 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3344 3345commit c88f13e25b0040c1dd0f93e0ac40f62a6005ce59 3346Author: Francisco Jerez <currojerez@riseup.net> 3347Date: Thu Oct 14 22:25:50 2010 +0200 3348 3349 Allocate a tiled scanout on pre-nv50 too. 3350 3351 Pageflip between tiled and non-tiled surfaces isn't such a good idea. 3352 3353 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3354 3355commit 4642d71bdffc04ce0dd458225541940c65f5443d 3356Author: Francisco Jerez <currojerez@riseup.net> 3357Date: Thu Oct 14 22:17:58 2010 +0200 3358 3359 dri2: Give out buffers ready for scanout. 3360 3361 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3362 3363commit 75daada678d03d2e162f8003a6f2edcbe0525f92 3364Author: Francisco Jerez <currojerez@riseup.net> 3365Date: Thu Oct 14 22:32:54 2010 +0200 3366 3367 Restructure tiled pixmap allocation in a single place. 3368 3369 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3370 3371commit 70f0d2c886ceaa965ca4864788f4dd8e8f757a92 3372Author: Francisco Jerez <currojerez@riseup.net> 3373Date: Thu Oct 21 22:55:02 2010 +0200 3374 3375 dri2: Add sync-to-vblank support. 3376 3377 Note that you need a recent enough kernel for it to work. Use the 3378 "GLXVBlank" option to enable it (it defaults to disabled for now). 3379 3380 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3381 3382commit 72dd4899ff036b09e3680ed2e09b3dbae3a7ba71 3383Author: Francisco Jerez <currojerez@riseup.net> 3384Date: Thu Oct 21 22:31:45 2010 +0200 3385 3386 nv04-nv40/exa: Make the copy hook blit downwards for non-overlapping regions. 3387 3388 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3389 3390commit 21ed642d3f76f286f23cd64ca817038ec94a1bf6 3391Author: Francisco Jerez <currojerez@riseup.net> 3392Date: Wed Oct 13 18:22:22 2010 +0200 3393 3394 Make the sync-to-vblank helpers more consistent. 3395 3396 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3397 3398commit 258e483d47f0f63155be18981c1118261f7675a8 3399Author: Francisco Jerez <currojerez@riseup.net> 3400Date: Thu Oct 21 22:41:24 2010 +0200 3401 3402 kms: Register a wakeup handler to poll vblank/pageflip/udev events. 3403 3404 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3405 3406commit 4f42708c26fa57bac7854701a6e757958c8d0654 3407Author: Francisco Jerez <currojerez@riseup.net> 3408Date: Sun Oct 10 04:39:27 2010 +0200 3409 3410 Update to the last libdrm API. 3411 3412 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3413 3414commit 8c8f15c9d3a31c35bf2ef20faf8c4133295a678f 3415Author: Francisco Jerez <currojerez@riseup.net> 3416Date: Sun Oct 10 04:36:17 2010 +0200 3417 3418 nv10-nv40/exa: Fix tiled pixmap pitch alignment calculation. 3419 3420 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3421 3422commit a47c274f329b3b8bfbdcf155d6d832599d823dc8 3423Author: Francisco Jerez <currojerez@riseup.net> 3424Date: Sun Oct 10 04:43:25 2010 +0200 3425 3426 Fix for the last xserver ABI. 3427 3428 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3429 3430commit 9a5142fae6e1a0a470352f511427916e6fa01b9c 3431Author: Maarten Maathuis <madman2003@gmail.com> 3432Date: Fri Aug 27 21:11:59 2010 +0200 3433 3434 nv50/exa: Stop using wfb by default, but add an option to enable it. 3435 3436 - Some scenarios, such as those heavy in core font rendering might prefer wfb. 3437 - Other (modern) situations are better of with wfb disabled. 3438 3439 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 3440 3441commit d5a6f103113985c50bbe749ed341de34af1b47a5 3442Author: Francisco Jerez <currojerez@riseup.net> 3443Date: Tue Sep 28 21:49:48 2010 +0200 3444 3445 dri2: Throttle frontbuffer flushes. 3446 3447 There was no need for throttling until the hardware interchannel sync 3448 kernel changes because the pushbuf IOCTL blocked to sync the GL and X 3449 channels anyway. 3450 3451 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3452 3453commit f1ac413d1d3dec2ccf63d4a8c79b9bd9ea578dcf 3454Author: Ben Skeggs <bskeggs@redhat.com> 3455Date: Mon Sep 20 08:05:00 2010 +1000 3456 3457 nv50: use 0x8697 on NVAF 3458 3459 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3460 3461commit 14c93acfc3fde04415e1c6a2a056465d15c86315 3462Author: Ben Skeggs <bskeggs@redhat.com> 3463Date: Fri Sep 17 13:17:37 2010 +1000 3464 3465 noaccel: fix rotation, as well as an assert during server shutdown 3466 3467 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3468 3469commit 065576d5df0df4da61050d789f548801656b62d5 3470Author: Ben Skeggs <bskeggs@redhat.com> 3471Date: Thu Aug 26 13:22:33 2010 +1000 3472 3473 dri2: remove explit flush from copy_region() 3474 3475 Unnecessary now that we kickoff from within the flush callback chain. 3476 3477 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3478 3479commit 34025f0dc9284873d4bf7e4f547ed69fc562a3c9 3480Author: Ben Skeggs <bskeggs@redhat.com> 3481Date: Thu Aug 26 13:19:39 2010 +1000 3482 3483 kickoff from the flush callback chain 3484 3485 This is a port of intel commit: 69d65f9184006eac790efcff78a0e425160e95aa 3486 3487 Submit batch buffers from flush callback chain 3488 3489 There are a few cases where the server will flush client output buffers 3490 but our block handler only catches the most common (before going into select 3491 If the server flushes client buffers before we submit our batch buffer, 3492 the client may receive a damage event for rendering that hasn't happened yet 3493 3494 Instead, we can hook into the flush callback chain, which the server will 3495 invoke just before flushing output. This lets us submit batch buffers 3496 before sending out events, preserving ordering. 3497 3498 Fixes 28438: [bisected] incorrect character in gnome-terminal under compiz 3499 https://bugs.freedesktop.org/show_bug.cgi?id=28438 3500 3501 Signed-off-by: Kristian Høgsberg <krh@bitplanet.net> 3502 3503 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3504 3505commit 390f1c86d9583f4ceabd407ee47410073c2a1b55 3506Author: Ben Skeggs <bskeggs@redhat.com> 3507Date: Fri Aug 20 09:48:32 2010 +1000 3508 3509 nv50/xv: support UYVY 3510 3511 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3512 3513commit 00d390952c912d4e9fc2c962caaeb90bf563d5b1 3514Author: Maarten Maathuis <madman2003@gmail.com> 3515Date: Fri Aug 13 22:24:26 2010 +0200 3516 3517 nv50/xv: Fix the uncommon codepath to use width/height variable. 3518 3519 - The common codepath should be the correct one, because that is 3520 triggered often, while the other is rare. 3521 - This isn't neccesarily a problem, but let's be on the safe side. 3522 3523 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 3524 3525commit f5b486dd237679aec1987985f9ecffc6a4fd3105 3526Author: Maarten Maathuis <madman2003@gmail.com> 3527Date: Fri Aug 13 22:16:51 2010 +0200 3528 3529 nv50/xv: We have 18 relocs, not 16, adjust MARK_RING. 3530 3531 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 3532 3533commit 1cb6c91a9afe88d73f8a1180b00f56a9150a16ed 3534Author: Roy Spliet <r.spliet@student.tudelft.nl> 3535Date: Sat Aug 7 20:59:06 2010 +0200 3536 3537 nouveau_exa.c: move line_count logic out of UTS/DFS loop 3538 3539 Signed-off-by: Roy Spliet <r.spliet@student.tudelft.nl> 3540 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 3541 3542commit 9f50b62229d5984ee5c16bc54953c7f8e2d5e575 3543Author: Ben Skeggs <bskeggs@redhat.com> 3544Date: Fri Aug 6 08:28:34 2010 +1000 3545 3546 nvc0: recognise these chipsets, and allow NoAccel mode on them 3547 3548 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3549 3550commit b96170a4e51c1d3167601210ca31e98283c02885 3551Author: Francisco Jerez <currojerez@riseup.net> 3552Date: Thu Aug 5 00:40:40 2010 +0200 3553 3554 nv10/exa: Bump maxX/maxY to 4096/4096. 3555 3556 This way we can do accelerated 2D on larger pixmaps, it will still 3557 fall back to software when we hit the 2048x2048 texturing limit in a 3558 composite operation. 3559 3560commit 321eb2df5c8b1da4d6469037dd1d8888b508fabd 3561Author: Francisco Jerez <currojerez@riseup.net> 3562Date: Wed Aug 4 13:11:24 2010 +0200 3563 3564 Revert "nv04-nv40/exa: Match the blob behavior more closely on PrepareCopy." 3565 3566 This reverts commit 9de0d97bd2fc2ee8800d48b5340a3d495525ad3e. Unnecessary 3567 after kernel commit "drm/nouveau: Ack the context switch interrupt before 3568 switching contexts.". 3569 3570commit db98ad23633958e52f84501c5a2061d42d346b64 3571Author: Francisco Jerez <currojerez@riseup.net> 3572Date: Tue Jun 15 15:32:36 2010 +0200 3573 3574 dri2: Fix for the last DRI2InfoRec ABI. 3575 3576 Just pre-fill the whole structure with zeros to avoid future 3577 segfaults. 3578 3579commit 9a3d0ab43f6335ab6cd48be4aa2f4f728386b386 3580Author: Ben Skeggs <bskeggs@redhat.com> 3581Date: Fri Jun 11 21:58:38 2010 +1000 3582 3583 call NVEnterVT from CreateScreenResources 3584 3585 We need to grab DRM master here for things to work, so we may as well call 3586 NVEnterVT as we used to in NVScreenInit up until the previous commit. 3587 3588 Fixes server regeneration. 3589 3590commit 964eeac6dc2209da6aa8861aa8b0bc028700ab3d 3591Author: Ben Skeggs <bskeggs@redhat.com> 3592Date: Thu Jun 10 09:38:19 2010 +1000 3593 3594 fix for server 1.9 3595 3596commit c079408e287ad09e954fa3985f95912c02cf06bc 3597Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 3598Date: Thu Jun 10 01:07:49 2010 +0200 3599 3600 nv50/accel: fix perspective interps in YUV to RGB shader 3601 3602commit 65153490f7d74d45a92adbd9c5e2bbe11c884b00 3603Author: Ben Skeggs <bskeggs@redhat.com> 3604Date: Wed Jun 9 16:12:28 2010 +1000 3605 3606 use libc directly for [cm]alloc/free 3607 3608commit 2958cf464dca9761f0fa21c8dd2d8fa8c8a96791 3609Author: Ben Skeggs <bskeggs@redhat.com> 3610Date: Thu May 27 10:06:47 2010 +1000 3611 3612 nv50/exa: rip out compat defines, we depend on 1.7 anyway 3613 3614commit af40bf0af47ebcb7de2dc418429f0b56f8c04d15 3615Author: Ben Skeggs <bskeggs@redhat.com> 3616Date: Wed May 26 16:11:01 2010 +1000 3617 3618 nv50/exa: new nouveau_class.h has correct blend func defines 3619 3620commit 8950c5b2a0d318e2568bec38a39d744b968c9b85 3621Author: Ben Skeggs <bskeggs@redhat.com> 3622Date: Wed May 26 16:09:48 2010 +1000 3623 3624 update nouveau_class.h 3625 3626commit 4b8f1a082069a17555d4142003213cf83d4e5464 3627Author: Lubomir Rintel <lkundrak@v3.sk> 3628Date: Mon May 10 09:25:55 2010 +0200 3629 3630 Handle startup without monitor attached more gracefully 3631 3632 Refuse to configure a screen which has no valid modes. Otherwise X server 3633 would crash attempting to dereference to modes list later on. 3634 3635 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3636 3637commit ac8f8d14a82ae8ba5b41fa88903edcad67f339c2 3638Author: Maarten Maathuis <madman2003@gmail.com> 3639Date: Fri May 7 20:51:43 2010 +0200 3640 3641 wfb: use 28 bits instead of 32 bits for multiply_factor optimisation 3642 3643 - Avoiding division in a hot code path is noticable even on an otherwise idle 3644 computer, but it's subject to rounding errors. If anyone knows a corner case 3645 and has better ideas, don't hesitate to contact me. For the moment this should 3646 make the problem less likely to trigger. 3647 3648 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 3649 3650commit 13c10430ba8f7b0edff3ad8aae4a97672eea4a8f 3651Author: Ben Skeggs <bskeggs@redhat.com> 3652Date: Fri Apr 23 14:09:35 2010 +1000 3653 3654 Bump version number to 0.0.16 3655 3656 Forgotten when the ABI broke a while back.. 3657 3658commit 6ff8ade0c7cd835b4172257ea310a8d88f28e757 3659Author: Ben Skeggs <bskeggs@redhat.com> 3660Date: Fri Apr 16 16:09:51 2010 +1000 3661 3662 Add uevent support 3663 3664 When a hotplug event is received from the kernel we should notify the 3665 client side to reconfigure the display. 3666 3667 Ported to nouveau from radeon driver, based on work by ajax in intel driver. 3668 3669commit 4063616938f76af8028491276039d422c0782b1b 3670Author: Ben Skeggs <bskeggs@redhat.com> 3671Date: Fri Apr 9 20:43:51 2010 +1000 3672 3673 Import latest nouveau_class.h from renouveau 3674 3675commit bfb95cc0ca600135474ff02a99889b09d41b3c80 3676Author: Rico Tzschichholz <ricotz@t-online.de> 3677Date: Thu Apr 1 20:39:23 2010 +0200 3678 3679 Fix "make dist". 3680 3681 - remove README.NV1 reference from Makefile.am 3682 - add nv_rop.h to src/Makefile.am 3683 3684 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3685 3686commit 2462b417fc550b71f021ca9736808f8f2d1cb6f0 3687Author: Ben Skeggs <skeggsb@beleth.localdomain> 3688Date: Mon May 4 17:04:34 2009 +1000 3689 3690 xv: only use bicubic filtering when scaling >=2x 3691 3692commit e2146d3b29a4bea3d584c145e3190c3313692ed9 3693Author: Francisco Jerez <currojerez@riseup.net> 3694Date: Thu Mar 18 14:32:54 2010 +0100 3695 3696 Fix for the latest nouveau_class.h. 3697 3698 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3699 3700commit 7858345b6b384e88902ccc516715156401edec04 3701Author: Ben Skeggs <bskeggs@redhat.com> 3702Date: Tue Mar 16 13:08:37 2010 +1000 3703 3704 nv50: fix texturing from >=4GiB mark 3705 3706commit 6b8b15731b7610c9d8fb3625861797a0c9fd9546 3707Author: Xavier Chantry <chantry.xavier@gmail.com> 3708Date: Fri Mar 5 00:53:47 2010 +0100 3709 3710 Remove leftover in man page and nv_const.h 3711 3712 Signed-off-by: Xavier Chantry <chantry.xavier@gmail.com> 3713 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3714 3715commit c642b9f7a13bdeecd0a83ddcbf6d6d4f2c287501 3716Author: Francisco Jerez <currojerez@riseup.net> 3717Date: Fri Mar 5 14:27:08 2010 +0100 3718 3719 kms: Don't try to copy fbcon contents when there's no framebuffer bound. 3720 3721 drmModeGetFB complains loudly when it's given an invalid framebuffer 3722 id, bail out earlier if that's the case. 3723 3724 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3725 3726commit efd7a526ca1cbac10b33b9d5d4cf3e542bc9b73b 3727Author: Ben Skeggs <bskeggs@redhat.com> 3728Date: Thu Feb 25 09:32:49 2010 +1000 3729 3730 fix typo 3731 3732commit 70d0a48b6c3f1a817bf850acd3bae48d063e56b9 3733Author: Ben Skeggs <bskeggs@redhat.com> 3734Date: Tue Feb 23 15:08:13 2010 +1000 3735 3736 Protect against potential crash for unknown connector types 3737 3738commit 83bbeea0d842094dd647e754608420a404eef6e3 3739Author: Ben Skeggs <bskeggs@redhat.com> 3740Date: Tue Feb 23 15:03:38 2010 +1000 3741 3742 Add connector name for eDP 3743 3744commit 29647021044463768cbfa3eead1416ef1cf27fa1 3745Author: Ben Skeggs <bskeggs@redhat.com> 3746Date: Thu Feb 18 14:17:57 2010 +1000 3747 3748 remove drm patchlevel check, libdrm checks this for us 3749 3750commit 9b4118d6d1fa488db86cd7d2875beea9cdefb096 3751Author: Francisco Jerez <currojerez@riseup.net> 3752Date: Wed Feb 10 18:43:39 2010 +0100 3753 3754 Revert "kms: work around some bong hits with dpms" 3755 3756 This reverts commit 98c9e4edb58374f18249e5c9c53b392fb8b4a1d1. AFAIK 3757 it's no longer needed. 3758 3759 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3760 3761commit f7deabfec0dcec4f203f85cd8494a70fe8fa6ebb 3762Author: Ben Skeggs <bskeggs@redhat.com> 3763Date: Thu Feb 11 08:39:49 2010 +1000 3764 3765 Pass pScreen to drmmode_fbcon_copy() 3766 3767commit 43524fcbb786c5b9e9ec25776560790aec208719 3768Author: Francisco Jerez <currojerez@riseup.net> 3769Date: Wed Feb 10 17:58:24 2010 +0100 3770 3771 dri2: Fix crashes on server regeneration. 3772 3773 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3774 3775commit f50577c117c82eea05e750fecd1ff5653105c648 3776Author: Francisco Jerez <currojerez@riseup.net> 3777Date: Wed Feb 10 13:23:46 2010 +0100 3778 3779 kms: Clean up the set_mode_major hook. 3780 3781 The X server already takes care of restoring the previous CRTC 3782 parameters after a failed mode set. As a bonus this patch fixes 3783 RandR13 projective transforms. 3784 3785 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3786 3787commit b16792b553dee41fc81d74c6e453832807e36368 3788Author: Marcin Kościelnicki <koriakin@0x04.net> 3789Date: Tue Feb 9 20:51:17 2010 +0000 3790 3791 Remove some UMS leftovers. 3792 3793 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3794 3795commit c56bce93f120d0ff18895149d1984896372cdd60 3796Author: Francisco Jerez <currojerez@riseup.net> 3797Date: Wed Feb 10 12:31:54 2010 +0100 3798 3799 kms: Don't pretend it works on xserver-1.5. 3800 3801 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3802 3803commit 0917665d1f2f1e76b6a0e7a4c027512f9f45f41b 3804Author: Francisco Jerez <currojerez@riseup.net> 3805Date: Wed Feb 10 11:48:47 2010 +0100 3806 3807 kms: Kill nv_cursor_convert_cursor. 3808 3809 Well, mostly, the remaining pitch conversion will be unnecessary soon. 3810 3811 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3812 3813commit e75dd23bd28f636b4c1759633d8dbc775e799add 3814Author: Marcin Kościelnicki <koriakin@0x04.net> 3815Date: Fri Feb 5 18:09:55 2010 +0100 3816 3817 nv50/accel: Fix mangled A8+A8 shader. 3818 3819 Turns out we used a misaligned long instruction near the end, and the 3820 shader was getting killed after writing R, A components. This has gone 3821 unnoticed since the remaining G, B outputs aren't actually used. 3822 3823commit 41fc1fdec47fa824c598e1f3d828f4c8c4a2cb71 3824Author: Marcin Kościelnicki <koriakin@0x04.net> 3825Date: Sun Jan 31 19:29:32 2010 +0000 3826 3827 Update to new nouveau_class.h 3828 3829 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3830 3831commit c94544c83ef84e96f8d132ac9ff9296c22c4468e 3832Author: Francisco Jerez <currojerez@riseup.net> 3833Date: Sun Jan 31 18:26:03 2010 +0100 3834 3835 nv10/exa: Fix up nv11 accel. 3836 3837 Fixes an off-by-one error made when UMS was removed. 3838 3839 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3840 3841commit a397b94e32e6b0109096aba09bbe7b5dcfdd8231 3842Author: Francisco Jerez <currojerez@riseup.net> 3843Date: Sun Jan 31 18:11:06 2010 +0100 3844 3845 dri2: Fix for the new DRI2InfoRec ABI. 3846 3847 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3848 3849commit 8b5c67145a9c29cd4c82f411f15e57723d76d0c0 3850Author: Francisco Jerez <currojerez@riseup.net> 3851Date: Sun Jan 31 17:43:54 2010 +0100 3852 3853 nv10/exa: Fix build after the last nouveau_class.h changes. 3854 3855 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3856 3857commit 2630a1553d560ce2c7391bc103d4a543f7d3d0da 3858Author: Marcin Kościelnicki <koriakin@0x04.net> 3859Date: Wed Jan 20 07:46:32 2010 +0000 3860 3861 Remove nv_dripriv.h, nv_local.h 3862 3863commit 2d341f7e0a0dc70a28c6091ef6919fc6c0d60561 3864Author: Ben Skeggs <bskeggs@redhat.com> 3865Date: Thu Jan 21 14:58:37 2010 +1000 3866 3867 xv: use nouveau_bo_new_tile() 3868 3869commit 2a76ff1106ee8c76bd2e16a4954b3c05714d1db3 3870Author: David Heidelberger <d.okias@gmail.com> 3871Date: Mon Jan 18 11:50:19 2010 +0100 3872 3873 Remove deprecated NV1 readme 3874 3875commit fe484a9b635ad56afec7be086a56c692b00dd6d1 3876Author: Ben Skeggs <bskeggs@redhat.com> 3877Date: Wed Jan 13 16:14:41 2010 +1000 3878 3879 exa: remove nouveau_pixmap_offset() 3880 3881 Completely useless now we don't have to try and support non-driver pixmaps. 3882 3883 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3884 3885commit bbf79469a878a8ab5b4c07faadcfdc8d85ceac83 3886Author: Ben Skeggs <bskeggs@redhat.com> 3887Date: Wed Jan 13 16:07:13 2010 +1000 3888 3889 exa: remove nouveau_exa_pixmap_{map,unmap} 3890 3891 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3892 3893commit 18e3a649f7df73bc1e6d199ecf7a62ddcfe37d6e 3894Author: Ben Skeggs <bskeggs@redhat.com> 3895Date: Wed Jan 13 14:26:04 2010 +1000 3896 3897 Check for drm+kms availability in NVPciProbe() 3898 3899 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3900 3901commit 6267b85460ef364aa51a7d33006940bb00a049d9 3902Author: Ben Skeggs <bskeggs@redhat.com> 3903Date: Wed Jan 13 12:03:14 2010 +1000 3904 3905 bump xorg-server requirement to 1.7 3906 3907 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3908 3909commit da5c0dc6040096fd12ddb6a04a22fada89115ccb 3910Author: Ben Skeggs <bskeggs@redhat.com> 3911Date: Mon Jan 11 13:23:06 2010 +1000 3912 3913 remove more unused code 3914 3915 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3916 3917commit 104b3ea842d4cebd993d3e7e9d55ebcadae0cf28 3918Author: Ben Skeggs <bskeggs@redhat.com> 3919Date: Mon Jan 11 12:59:12 2010 +1000 3920 3921 exa: Remove non-driver pixmaps path 3922 3923 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3924 3925commit e8798d91eedc5d8337223b2257ca607da8a197de 3926Author: Ben Skeggs <bskeggs@redhat.com> 3927Date: Mon Jan 11 12:48:00 2010 +1000 3928 3929 Remove pNv->NVArch 3930 3931 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3932 3933commit 7a6f785de1732ec0af429a0b66d5d6ba45427dcb 3934Author: Ben Skeggs <bskeggs@redhat.com> 3935Date: Mon Jan 11 12:00:19 2010 +1000 3936 3937 Remove pNv->Chipset use 3938 3939 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3940 3941commit 68c1f908be5739692c9f0bde9c23f3dc2a7210ea 3942Author: Ben Skeggs <bskeggs@redhat.com> 3943Date: Mon Jan 11 11:56:19 2010 +1000 3944 3945 remove pNv->VRAMPhysical 3946 3947 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3948 3949commit 06499197d6eb07a272a1812e1c8bce886798a1d7 3950Author: Ben Skeggs <bskeggs@redhat.com> 3951Date: Mon Jan 11 11:51:14 2010 +1000 3952 3953 dri1: remove DRI1 support 3954 3955 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3956 3957commit 6975651b0b74dca6edaee759fb1003f65189f1df 3958Author: Ben Skeggs <bskeggs@redhat.com> 3959Date: Mon Jan 11 10:07:11 2010 +1000 3960 3961 xv: ifdef out another problematic piece of overlay support for the moment 3962 3963commit 17485c234ff191cee3dd19e3dd693a80b024e189 3964Author: Ben Skeggs <bskeggs@redhat.com> 3965Date: Wed Jan 6 14:58:15 2010 +1000 3966 3967 Initial pass at removal of non-KMS support. 3968 3969 It's likely there's a lot more that can be cleaned up/removed as a result 3970 of this, we can handle that as we come across it. 3971 3972 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 3973 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 3974 Acked-by: Maarten Maathuis <madman2003@gmail.com> 3975 3976commit 061cb284b93612848296599759b915a4d66b6d01 3977Author: Maarten Maathuis <madman2003@gmail.com> 3978Date: Sun Jan 3 02:14:36 2010 +0100 3979 3980 exa: Some compat defines for new pixman formats. 3981 3982 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 3983 3984commit 8ca5a7383aad0e0557b3c9a1e9edd5431cd509fc 3985Author: Marcin Slusarz <marcin.slusarz@gmail.com> 3986Date: Wed Dec 30 21:46:04 2009 +0100 3987 3988 start using BEGIN_RING_NI 3989 3990commit b259b7ce5e57e1d20111878318b4703a14ec27ec 3991Author: Maarten Maathuis <madman2003@gmail.com> 3992Date: Fri Jan 1 20:16:13 2010 +0100 3993 3994 nv50: always set semaphore handle and offset 3995 3996 - This is in preperation of using semaphores for more than one thing. 3997 3998 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 3999 4000commit 8deeed25014b36f8d60f25c7ea2509d88fcb52ae 4001Author: Marcin Slusarz <marcin.slusarz@gmail.com> 4002Date: Mon Dec 28 18:39:16 2009 +0100 4003 4004 nv50: upgrade nv50_texture.h 4005 4006 - copy changes from mesa/src/gallium/drivers/nv50/nv50_texture.h 4007 - remove 2 redefinitions 4008 - fix R<->B rename (see df189c9efc0fbcdce816af483f0147ab635280d1 4009 (nv50: TIC/TSC fixes and additions) in mesa repo) 4010 4011commit 1877ee268774443a58a7bb73244bd70c82d67879 4012Author: Marcin Slusarz <marcin.slusarz@gmail.com> 4013Date: Thu Dec 24 21:29:10 2009 +0100 4014 4015 nv50/exa: add support for more color formats 4016 4017commit 8b3753fc34e3557b91c3bdebe7c8addfb148b3bd 4018Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 4019Date: Wed Dec 30 18:28:25 2009 +0100 4020 4021 nv50: check drawable.bitsPerPixel instead of depth to get format 4022 4023commit 920a346e14cdc40efe9e89c3583d8e573e7e2f87 4024Author: Marcin Kościelnicki <koriakin@0x04.net> 4025Date: Mon Dec 28 16:11:51 2009 +0000 4026 4027 nv50: Dehexify and bring up to date with new method defines. 4028 4029 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4030 4031commit 7c72ac1f47e19af97c997f57012e3af12f941894 4032Author: Francisco Jerez <currojerez@riseup.net> 4033Date: Sat Dec 26 15:56:58 2009 +0100 4034 4035 nv10/exa: Fix for the latest nouveau_class.h changes. 4036 4037 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4038 4039commit 6992d0e7a0cb3c32b16af0b724246e44f7a35d7e 4040Author: Francisco Jerez <currojerez@riseup.net> 4041Date: Thu Dec 17 19:13:08 2009 +0100 4042 4043 dri: Use classic mesa on the fixed pipe cards. 4044 4045 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4046 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4047 4048commit 87b9916dd4b471a9b8b02f166f6553f04f219769 4049Author: Maarten Maathuis <madman2003@gmail.com> 4050Date: Wed Dec 23 21:53:36 2009 +0100 4051 4052 kms: Remove the one-off offset of connector names compared to the kernel. 4053 4054 - Seeing DVI-I-0 in xrandr and DVI-I-1 in the kernel logs can be a source of 4055 confusion. 4056 4057 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4058 4059commit bb1947831d9a4e080b8d1e9dba086af6527ef479 4060Author: Ben Skeggs <bskeggs@redhat.com> 4061Date: Thu Dec 17 13:08:34 2009 +1000 4062 4063 Revert "exa: map the pixmap bo after creation to ensure a useable driver pixmap" 4064 4065 This reverts commit 740c9f3e79729e4e06c693e66826afbad9bd1ce5. 4066 4067 To get the same effect as this commit, update to latest libdrm git. 4068 4069commit 03c7d675f6d12886b6bca8e0375888ecf1b393df 4070Author: Francisco Jerez <currojerez@riseup.net> 4071Date: Fri Dec 11 15:47:05 2009 +0100 4072 4073 dri2: Make the render targets tiled for pre-G80 cards. 4074 4075 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4076 4077commit 963aaeff6b414f371eff3e8ce016f7707c60f7c0 4078Author: Francisco Jerez <currojerez@riseup.net> 4079Date: Fri Dec 11 15:44:47 2009 +0100 4080 4081 exa: Pre-G80 tiling support. 4082 4083 For now, pixmaps will only be tiled if driver pixmaps is being used 4084 and we're told to with the NOUVEAU_CREATE_PIXMAP_TILED usage hint. 4085 4086 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4087 4088commit 41d8d511124fcbe83a3cc68aaf22c3ce0fb4eb8a 4089Author: Francisco Jerez <currojerez@riseup.net> 4090Date: Mon Dec 14 00:51:17 2009 +0100 4091 4092 kms: Allow interlaced/doublescan modes. 4093 4094 After bcafdfbed6 (an X server commit) we have to explicitly state we 4095 can handle them. 4096 4097 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4098 4099commit 740c9f3e79729e4e06c693e66826afbad9bd1ce5 4100Author: Maarten Maathuis <madman2003@gmail.com> 4101Date: Sat Dec 12 22:27:30 2009 +0100 4102 4103 exa: map the pixmap bo after creation to ensure a useable driver pixmap 4104 4105 - Large pixmaps (think of 8000x8000 pixmaps on NV50) will cause crashes 4106 otherwise because exa assumes driver pixmaps to be mappable. Handling this 4107 elsewhere is problematic. 4108 4109 Signed-off-by: Maarten Maathuis <madman2003@gmail.com> 4110 4111commit 0b5120cf8ef2528a15d88054afdc6fc33eb521f5 4112Author: Ben Skeggs <skeggsb@beleth.(none)> 4113Date: Mon Dec 7 15:31:53 2009 +1000 4114 4115 Allow texturing from GART 4116 4117 Unfortunately NV_CONTEXT_SURFACES_2D on most cards (probably only PCIE cards 4118 work actually, but that's not confirmed) rejects a GART DMA object, so the 4119 usefulness is a bit more limited than I'd hoped. 4120 4121commit 60a394222b410da8bb832935fc525132218b4d94 4122Author: Ben Skeggs <bskeggs@redhat.com> 4123Date: Thu Dec 3 21:14:00 2009 +1000 4124 4125 nv50: avoid touching 0x16b0 on 0x8597, it causes a DATA_ERROR from the GPU 4126 4127 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4128 4129commit 95fc663d5fb653fc0a50aa4d589bbf3b9c79e928 4130Author: Ben Skeggs <bskeggs@redhat.com> 4131Date: Thu Nov 26 13:10:45 2009 +1000 4132 4133 shadowfb: fix segv on fb resize 4134 4135 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4136 4137commit 1cf0cfa926dd78698ff8b8c2d776bcdaa6a867bd 4138Author: Ben Skeggs <bskeggs@redhat.com> 4139Date: Thu Nov 26 10:04:43 2009 +1000 4140 4141 kms: zfill fb if we have no accel to do copy, better than random garbage 4142 4143commit 92578fb4db0feb549bea82d0874ba0e39a2b1ae2 4144Author: Francisco Jerez <currojerez@riseup.net> 4145Date: Thu Nov 5 18:30:28 2009 +0100 4146 4147 nv10/exa: Rename VIEWPORT_SCALE to VIEWPORT_TRANSLATE. 4148 4149 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4150 4151commit 9de0d97bd2fc2ee8800d48b5340a3d495525ad3e 4152Author: Francisco Jerez <currojerez@riseup.net> 4153Date: Thu Nov 5 13:29:09 2009 +0100 4154 4155 nv04-nv40/exa: Match the blob behavior more closely on PrepareCopy. 4156 4157 Fix a somewhat indeterministic corruption problem on nv17 when there 4158 is stuff going on the other fifos (e.g. gallium but I've also 4159 reproduced it with an app just SIFM-ing memory around): in some cases 4160 it made the blits the X server had scheduled fail corrupting the 4161 nearby screen areas. 4162 4163 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4164 Reviewed-by: Ben Skeggs <bskeggs@redhat.com> 4165 4166commit 16ef9ff7e394a693fc74764d8bc2630e784947f3 4167Author: Francisco Jerez <currojerez@riseup.net> 4168Date: Thu Nov 5 03:28:00 2009 +0100 4169 4170 nv10/exa: Spring-cleaning 4171 4172 * Kill the A8+A8 hack. Recent enough X servers (>=1.7) fall back to 4173 ARGB glyphs for drivers not supporting A8 render targets. 4174 4175 * Kill all the global state. It doesn't matter a lot yet but it might 4176 if we get multicard working at some point. 4177 4178 * Other random clean-ups with no functional changes. 4179 4180 Some numbers from x11perf -aa10text -aa24text -comppixwin10 -comppixwin500: 4181 4182 * Before, with A glyphs: 4183 12000000 trep @ 0.0025 msec (394000.0/sec): Char in 80-char aa line (Charter 10) 4184 3200000 trep @ 0.0119 msec ( 84300.0/sec): Char in 30-char aa line (Charter 24) 4185 8000000 trep @ 0.0036 msec (280000.0/sec): Composite 10x10 from pixmap to window 4186 60000 trep @ 0.4813 msec ( 2080.0/sec): Composite 500x500 from pixmap to window 4187 4188 * After, ARGB glyphs: 4189 20000000 trep @ 0.0015 msec (649000.0/sec): Char in 80-char aa line (Charter 10) 4190 4800000 trep @ 0.0057 msec (175000.0/sec): Char in 30-char aa line (Charter 24) 4191 8000000 trep @ 0.0036 msec (280000.0/sec): Composite 10x10 from pixmap to window 4192 60000 trep @ 0.4813 msec ( 2080.0/sec): Composite 500x500 from pixmap to window 4193 4194 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4195 Tested-by: Pekka Paalanen <pq@iki.fi> 4196 4197commit e1c2efd7e7703f5860e5fe99e0413e3be10576d4 4198Author: Ben Skeggs <bskeggs@redhat.com> 4199Date: Thu Nov 5 13:04:17 2009 +1000 4200 4201 kms: don't update drmmode_output->mode_output when vtSema is FALSE 4202 4203 Fixes fast user switching issues. 4204 4205 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4206 4207commit aed811e71b544dccfe84671f962503342acb75ab 4208Author: Ben Skeggs <bskeggs@redhat.com> 4209Date: Wed Nov 4 18:03:12 2009 +1000 4210 4211 exa/nv10: handle reloc failures 4212 4213commit 5315034b7d46980a3527c814526857aed3bc98f7 4214Author: Ben Skeggs <skeggsb@beleth.(none)> 4215Date: Wed Nov 4 16:52:02 2009 +1000 4216 4217 xv/nv04: handle reloc failures + fail on bad surface format 4218 4219commit 99b1bf3d3e6b8c8ef2aa4aae3e33f3908cd00dab 4220Author: Ben Skeggs <skeggsb@beleth.(none)> 4221Date: Wed Nov 4 16:46:35 2009 +1000 4222 4223 xv/nv40: don't even try with unknown surface format 4224 4225commit c1447574a025416f92f8dec2c361859c3945db92 4226Author: Ben Skeggs <skeggsb@beleth.(none)> 4227Date: Wed Nov 4 16:46:18 2009 +1000 4228 4229 xv/nv40: handle reloc failures 4230 4231commit 6ca14ba77104542a2915af8dca56236646be139a 4232Author: Ben Skeggs <skeggsb@beleth.(none)> 4233Date: Wed Nov 4 16:41:58 2009 +1000 4234 4235 xv/nv30: don't even try with an unknown surface format.. 4236 4237commit 53ef7be58a2c46b2b5ffe912dcd408d74705ccbf 4238Author: Ben Skeggs <skeggsb@beleth.(none)> 4239Date: Wed Nov 4 16:41:44 2009 +1000 4240 4241 xv/nv30: handle reloc failures 4242 4243commit 6facc0f765c1130aeb19ddfb76cbaaabaecf3f4a 4244Author: Ben Skeggs <skeggsb@beleth.(none)> 4245Date: Wed Nov 4 16:36:09 2009 +1000 4246 4247 exa/nv40: handle reloc failures 4248 4249commit a97501a4e64d690be4232b01f9a3663fc2338c8a 4250Author: Ben Skeggs <skeggsb@beleth.(none)> 4251Date: Wed Nov 4 16:31:56 2009 +1000 4252 4253 exa/nv30: handle reloc failures 4254 4255commit 222d13a03b41eb610a05063f81f0c47f57e6098e 4256Author: Ben Skeggs <skeggsb@beleth.(none)> 4257Date: Wed Nov 4 16:02:50 2009 +1000 4258 4259 exa/nv04: handle reloc failures 4260 4261commit bd8aa7515f69c412e8f379777c49d59364df32fe 4262Author: Ben Skeggs <bskeggs@redhat.com> 4263Date: Wed Oct 28 12:59:59 2009 +1000 4264 4265 xv: handle reloc failures 4266 4267commit 0214cb8b25a6959573bfc183b259b435f0a728bd 4268Author: Ben Skeggs <bskeggs@redhat.com> 4269Date: Wed Oct 28 12:55:25 2009 +1000 4270 4271 xv/nv50: handle reloc failures 4272 4273commit e84e98d7f9260006a8b16a5785e4d2926d2370d4 4274Author: Ben Skeggs <bskeggs@redhat.com> 4275Date: Wed Oct 28 12:47:49 2009 +1000 4276 4277 exa/nv50: handle reloc failures 4278 4279commit 62c5c875e4c0812815be48f74bdef25d9f046515 4280Author: Ben Skeggs <bskeggs@redhat.com> 4281Date: Wed Oct 28 12:30:24 2009 +1000 4282 4283 exa: handle reloc failures 4284 4285commit c839f6098451aa59c7cb1fd7819e17a966e8ace7 4286Author: Marcin Slusarz <marcin.slusarz@gmail.com> 4287Date: Sun Nov 1 22:16:50 2009 +0100 4288 4289 nv30/exa: support more 16bpp formats 4290 4291 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> 4292 4293commit 1d93e207db6e5a58b702c1812feaeb6b7252bc70 4294Author: Marcin Slusarz <marcin.slusarz@gmail.com> 4295Date: Sat Oct 31 20:24:52 2009 +0100 4296 4297 nv/exa: fix 15/16 bits solid fill 4298 4299 after this change nouveau passes all fill and blend tests of rendercheck 4300 (before: fill - 108/120, blend - 3323868/3569150) 4301 tested on NV34 4302 4303 Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> 4304 4305commit df5f17e59a7f42781b388074798d26f7981b2296 4306Author: Arthur HUILLET <arthur.huillet@free.fr> 4307Date: Mon Nov 2 15:56:56 2009 +0100 4308 4309 nv10exa: improve codestyle 4310 4311 Signed-off-by: Arthur HUILLET <arthur.huillet@free.fr> 4312 4313commit 5587f40c1b8af7d178f9a68d0f1fecdfc0ca9749 4314Author: Ben Skeggs <bskeggs@redhat.com> 4315Date: Fri Oct 30 14:01:09 2009 +1000 4316 4317 exa: force the use of sysmem pixmaps on low-mem cards 4318 4319 Very similar effect to forcing MigrationHeuristic "greedy" on classic 4320 EXA. Far better than the migration ping-pong that'd occur otherwise 4321 4322 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4323 4324commit 718a41b9a2b5649ac37c18f456b94eaf99584598 4325Author: Ben Skeggs <bskeggs@redhat.com> 4326Date: Thu Oct 22 12:35:32 2009 +1000 4327 4328 nv50: NVA8's 3D object class is 0x8597 4329 4330commit 7494b18bf4061fb960a2126c745721161baddd01 4331Author: Francisco Jerez <currojerez@riseup.net> 4332Date: Sun Oct 18 19:45:02 2009 +0200 4333 4334 nv50/crtc: Don't perform the LUT expansion twice. 4335 4336 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4337 4338commit b161a7f7fa81b6d499c91c17f988a38ed537ed35 4339Author: Francisco Jerez <currojerez@riseup.net> 4340Date: Sun Oct 18 19:41:11 2009 +0200 4341 4342 nv04-nv40/crtc: Don't perform the LUT expansion twice. 4343 4344 Fixes bug 24595. 4345 4346 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4347 4348commit 9cbc139176ae991150bd9ddf66b61db471b391f2 4349Author: Francisco Jerez <currojerez@riseup.net> 4350Date: Sun Oct 11 23:42:01 2009 +0200 4351 4352 Partially revert "kms: always set gamma on modeset" 4353 4354 6c8d3e39b72abe897b08d9e90cae85e121863d2c breaks gamma setting with 4355 servers older than 1.7. Keep the call to xf86HandleColormaps even with 4356 KMS enabled, as it's a no-op in the cases we don't need it. 4357 4358 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4359 4360commit cef4b61b4d0cf096428d89eff65a78fd16f612d5 4361Author: Francisco Jerez <currojerez@riseup.net> 4362Date: Sun Oct 11 20:23:45 2009 +0200 4363 4364 Fix NVLoadPalette for 15/16 bit depth. 4365 4366 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4367 4368commit cc4f7a733e3504260437d06091490fcb8a4c8063 4369Author: Francisco Jerez <currojerez@riseup.net> 4370Date: Sun Oct 11 20:31:11 2009 +0200 4371 4372 Remove an useless call to NVLeaveVT from NVCloseScreen. 4373 4374 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4375 4376commit 5a4afade51fe5fd149f8726fc243fbb69a660bf3 4377Author: Francisco Jerez <currojerez@riseup.net> 4378Date: Thu Oct 8 03:33:34 2009 +0200 4379 4380 nv10/exa: Replace "celcius" with "celsius". 4381 4382 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4383 Tested-by: Pekka Paalanen <pq@iki.fi> 4384 4385commit 2e744325827d702b564c5d6becaf65f3ad164761 4386Author: Francisco Jerez <currojerez@riseup.net> 4387Date: Sun Oct 11 17:40:56 2009 +0200 4388 4389 nv10/exa: Allow component-alpha masks. 4390 4391 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4392 Tested-by: Pekka Paalanen <pq@iki.fi> 4393 4394commit c3bb746ab6b371e843e1754a919dd25fa7b96b19 4395Author: Francisco Jerez <currojerez@riseup.net> 4396Date: Sun Oct 11 17:38:44 2009 +0200 4397 4398 nv10/exa: Fix PictOpAtopReverse for destination pixmaps with no alpha channel. 4399 4400 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4401 Tested-by: Pekka Paalanen <pq@iki.fi> 4402 4403commit f0a3d9f59817ebf2b9606097785f0ad1ae45da12 4404Author: Francisco Jerez <currojerez@riseup.net> 4405Date: Sun Oct 11 16:15:30 2009 +0200 4406 4407 nv10/exa: Refactor the reg combiner/alpha blending code. 4408 4409 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4410 Tested-by: Pekka Paalanen <pq@iki.fi> 4411 4412commit 3f020b0bf2e8d2bc54eceda771829c4b2535688a 4413Author: Ben Skeggs <bskeggs@redhat.com> 4414Date: Thu Oct 8 16:19:16 2009 +1000 4415 4416 kms: reload cursor on modeset, needed on rotation (rh#524632) 4417 4418commit 38d7520986078fd01c19957a75ebfd9d91665225 4419Author: Ben Skeggs <bskeggs@redhat.com> 4420Date: Thu Oct 8 11:34:20 2009 +1000 4421 4422 kms: switch to CP+MPH instead of GetScratchPixmapHeader 4423 4424commit 6c8d3e39b72abe897b08d9e90cae85e121863d2c 4425Author: Ben Skeggs <bskeggs@redhat.com> 4426Date: Wed Oct 7 15:57:49 2009 +1000 4427 4428 kms: always set gamma on modeset 4429 4430 fixes issues vt-switching between console and X of different depths. 4431 4432commit b13c0509be63f27a4a9c918112d0a0fc84ccce10 4433Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 4434Date: Wed Sep 16 18:17:30 2009 +0200 4435 4436 nv50/exa: use A8 RT format for PICT_a8 4437 4438 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4439 4440commit 7a63e4c61cd5d082caf074702cde79e1a66354b0 4441Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 4442Date: Tue Oct 6 11:58:39 2009 +0200 4443 4444 nv50/exa: cleanup shader settings 4445 4446 Since we only pass through 2D coordinates in the VP, 4447 no need to occupy extra output registers for HPOS z 4448 and, which can be coded as 0, 1 into VP_RESULT_MAP. 4449 4450 We don't have to fetch all 4 attribute components if 4451 we only read 2 in the VPs. 4452 4453 Using 0x1, 0x1 as terminator instructions looks odd, 4454 just make the last ones that do something long and 4455 set the exit bit. 4456 4457 Setup VP_RESULT_MAP and INTERPOLANT_CTRL correctly. 4458 4459 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 4460 4461commit 60818bdeb975a9f469d3ae890932d4de1cbb59f9 4462Author: Ben Skeggs <bskeggs@redhat.com> 4463Date: Wed Oct 7 10:49:00 2009 +1000 4464 4465 kms: don't copy fbcon if depths don't match 4466 4467commit dd8339fc4e18fa4fd409f2997db741be209e852d 4468Author: Ben Skeggs <bskeggs@redhat.com> 4469Date: Tue Sep 29 10:52:53 2009 +1000 4470 4471 Pass NULL fb pointer to fbScreenInit when using EXA 4472 4473 Fixes 1.7 xserver getting very confused... 4474 4475commit 6dbdc80016c2a81c390d77db613833cd4d15c617 4476Author: Ben Skeggs <bskeggs@redhat.com> 4477Date: Tue Sep 29 09:00:59 2009 +1000 4478 4479 exa: make tiled scanout a separate condition from wfb being used 4480 4481commit 96d025232e292a8b1aebd54eaad3a21dbf6e2f77 4482Author: Maarten Maathuis <madman2003@gmail.com> 4483Date: Fri Sep 25 19:50:11 2009 +0200 4484 4485 nv40/exa: fallback on special case xrgb non-repeat textures 4486 4487commit ed05bd6dab4fdd1289947154df02cd21c0839574 4488Author: Maarten Maathuis <madman2003@gmail.com> 4489Date: Fri Sep 25 19:49:14 2009 +0200 4490 4491 nv30/exa: fallback on special case xrgb non-repeat textures 4492 4493commit dd7b86cb5351a60f21eee9cf81ae0ef61cf45f13 4494Author: Maarten Maathuis <madman2003@gmail.com> 4495Date: Fri Sep 25 19:39:24 2009 +0200 4496 4497 exa: fallback on src pictures without drawable 4498 4499commit de0b095a1deaa1e53e1f555a6eb537482c488638 4500Author: Maarten Maathuis <madman2003@gmail.com> 4501Date: Thu Sep 24 19:21:59 2009 +0200 4502 4503 nv50/exa: avoid fallback on xrgb repeat none when dst has no alpha 4504 4505 - This should fix resizing slowness in evince amongst other things. 4506 4507commit df94ebdbcd89c1678ac243217e7f5b20cbbe857c 4508Author: Ben Skeggs <bskeggs@redhat.com> 4509Date: Mon Sep 21 08:37:27 2009 +1000 4510 4511 kms: use displayWidth for pitch calc in transition, not virtualX (rh#523281) 4512 4513commit 5a6c3c427796f7f3dac0f597a320b09feb25ffc8 4514Author: Ben Skeggs <bskeggs@redhat.com> 4515Date: Wed Sep 16 15:13:08 2009 +1000 4516 4517 dri2: call CreatePixmap rather than yet another custom-built one 4518 4519commit 95086de507df7c5676c7ee712139ac4c91c9f6d2 4520Author: Pekka Paalanen <pq@iki.fi> 4521Date: Tue Sep 15 20:31:33 2009 +0300 4522 4523 Partially revert "kms: wait for fbcon to be copied before continuing" 4524 4525 This partially reverts commit 6c045fc44783454180d7b3d00b5f25436bd5544e. 4526 4527 The nv_driver.c hunk of the above patch breaks the build on at least git 4528 and 1.6.3.901 xorg-servers due to: 4529 nv_driver.c:745: error: 'struct _ScrnInfoRec' has no member named 4530 'canDoBGNoneRoot' 4531 4532 Signed-off-by: Pekka Paalanen <pq@iki.fi> 4533 4534commit 6c045fc44783454180d7b3d00b5f25436bd5544e 4535Author: Ben Skeggs <bskeggs@redhat.com> 4536Date: Mon Sep 14 13:04:12 2009 +1000 4537 4538 kms: wait for fbcon to be copied before continuing 4539 4540commit 22971d5545a2cd03e5c0b5d5d16fb85d7fff2aff 4541Author: Ben Skeggs <bskeggs@redhat.com> 4542Date: Thu Sep 10 16:05:15 2009 +1000 4543 4544 dri2: switch to bo_new_tile() 4545 4546commit 806eaf6b0b36cb05ca9d883ff4572629812a1d48 4547Author: Ben Skeggs <bskeggs@redhat.com> 4548Date: Thu Sep 10 12:00:44 2009 +1000 4549 4550 kms: rework fbcon transition 4551 4552 This does 2 things. Most importantly, it avoids a situation where we get 4553 multiple GEM handles for a single object, making it possible to confuse 4554 the kernel if both appear in the same push buffer. 4555 4556 Secondly, it moves the transition to only on X startup, and not on every 4557 modeset etc. 4558 4559commit f90edc168c4a18b20750043d33c193bdc1b89156 4560Author: Ben Skeggs <bskeggs@redhat.com> 4561Date: Wed Sep 9 09:23:58 2009 +1000 4562 4563 Revert "nv50: use 0x7000 everywhere for now, until we can fix mixed mode alignment" 4564 4565 This reverts commit c80f125a97003c2d0db77f54e22d9de447a06f63. 4566 4567 This was not supposed to sneak in.. 4568 4569commit ff6bdb14ac13168c1a4b7866784e4833a1162dbc 4570Author: Ben Skeggs <bskeggs@redhat.com> 4571Date: Wed Sep 9 09:19:42 2009 +1000 4572 4573 Revert "kms: fetch active encoder during init if there is one" 4574 4575 This reverts commit 4f66fc6e871969448eba25817bfb383f2fdd4b34. 4576 4577 See reason in previous commit. 4578 4579commit 9708e50e820970d526b084e0ab736c1c7a4f50c0 4580Author: Ben Skeggs <bskeggs@redhat.com> 4581Date: Wed Sep 9 09:18:48 2009 +1000 4582 4583 Revert "kms: update encoder config during detect" 4584 4585 This reverts commit 25610704d4b16293f5bc24d5ad18cc76e20896d2. 4586 4587 Very bad thinko, drm_mode_get_connector.encoder_id is *not* the same thing 4588 as our DRM's nouveau_connector.detected_encoder. Need to find a better 4589 way of achieving this. 4590 4591commit c80f125a97003c2d0db77f54e22d9de447a06f63 4592Author: Ben Skeggs <bskeggs@redhat.com> 4593Date: Tue Sep 8 16:20:33 2009 +1000 4594 4595 nv50: use 0x7000 everywhere for now, until we can fix mixed mode alignment 4596 4597commit 25610704d4b16293f5bc24d5ad18cc76e20896d2 4598Author: Ben Skeggs <bskeggs@redhat.com> 4599Date: Tue Sep 8 11:33:51 2009 +1000 4600 4601 kms: update encoder config during detect 4602 4603commit 4f66fc6e871969448eba25817bfb383f2fdd4b34 4604Author: Ben Skeggs <bskeggs@redhat.com> 4605Date: Tue Sep 8 11:26:29 2009 +1000 4606 4607 kms: fetch active encoder during init if there is one 4608 4609commit 2b5ec6ad708164eb9b2df00f453ea84f4898153a 4610Author: Ben Skeggs <bskeggs@redhat.com> 4611Date: Fri Sep 4 10:35:57 2009 +1000 4612 4613 kms: only call drmModeSetCursor when cursor is visible 4614 4615 It seems we get load_cursor_argb() calls even when the cursor is hidden, 4616 and the drmModeSetCursor() call would then decide a cursor is available 4617 and show it on outputs where the cursor shouldn't have been actuve. 4618 4619commit 569a17a98d03d965f9500c37c0cb14e0fb462311 4620Author: Francisco Jerez <currojerez@riseup.net> 4621Date: Thu Aug 20 18:44:38 2009 +0200 4622 4623 kms: No need to upload gamma on set_mode_major in older X servers. 4624 4625 Fixes build for CRTC ABI < 3. 4626 4627 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4628 4629commit 32fc8a9a3104b38fe89427ffa60d21dad1572c15 4630Author: Francisco Jerez <currojerez@riseup.net> 4631Date: Thu Aug 20 18:43:58 2009 +0200 4632 4633 kms: Fix <nv11 hardware cursor. 4634 4635 Signed-off-by: Francisco Jerez <currojerez@riseup.net> 4636 4637commit e4e6f863b626d77934935e13714c6cbfd014d916 4638Author: Maarten Maathuis <madman2003@gmail.com> 4639Date: Thu Aug 20 12:04:26 2009 +0200 4640 4641 kms: actually call kernel when loading cursor image 4642 4643 - In the past we depended on getting a cursor show to call the kernel and actually make it visible. 4644 4645commit 95f09df23efd0a36d29cb4810faa659f839194f0 4646Author: Ben Skeggs <bskeggs@redhat.com> 4647Date: Thu Aug 20 15:48:26 2009 +1000 4648 4649 kms: prevent X from hiding cursor before updating image 4650 4651 This wasn't noticeable when everything was being done in userspace. However, 4652 with KMS there's more of a delay between the hide<->update<->show steps 4653 and this produces a noticeable flicker. 4654 4655 If there's any good reason we shouldn't be updating the cursor image while 4656 it's being shown, we'll handle it transparently in the DRM as required. 4657 4658commit fe2b5e6c6f81f4c4a76f5e4c5816812459872af3 4659Author: Maarten Maathuis <madman2003@gmail.com> 4660Date: Wed Aug 19 22:06:42 2009 +0200 4661 4662 kms: fix gamma 4663 4664 - crtc->active is needed in order to set the gamma manually 4665 - the initial gamma must be set by us 4666 4667commit 9d270f65ee85dedc6bf9cd9cf24e3d8df062f670 4668Author: Maarten Maathuis <madman2003@gmail.com> 4669Date: Tue Aug 18 21:59:10 2009 +0200 4670 4671 exa: update to latest CreatePixmap2 + do not align width to 64 4672 4673 - Only the pitch needs to be aligned to 64. 4674 4675commit b50110d00e32396e186168271adbb88ace8ba986 4676Author: Francisco Jerez <currojerez@riseup.net> 4677Date: Mon Aug 17 16:54:48 2009 +0200 4678 4679 kms: Some TV-out fixes. 4680 4681commit 523feba8fdc6d285929ac72be0a6e037019f019e 4682Author: Maarten Maathuis <madman2003@gmail.com> 4683Date: Mon Aug 17 00:09:42 2009 +0200 4684 4685 wfb: more correct and more error tolerant 4686 4687 - In a strange world it is possible that a high index is tiled, and a lower index empty. 4688 - This should avoid that loophole and also emit a warning when we run out of indices. 4689 4690commit 3db3e5350e2f6dea50cd2bce0c5fb0437fe08dcf 4691Author: Francisco Jerez <currojerez@riseup.net> 4692Date: Wed Aug 12 02:18:27 2009 +0200 4693 4694 kms: Implement output->get_property when RandR1.3 is available. 4695 4696 This allows getting updated values when the kernel is modifying them 4697 behind our back. 4698 4699commit ba9817b88d31364a5d43d945359e357535aeb09a 4700Author: Francisco Jerez <currojerez@riseup.net> 4701Date: Wed Aug 12 02:18:26 2009 +0200 4702 4703 kms: Don't hardcode the output properties 4704 4705 Replicate any properties the kernel exposes. Mostly taken from the 4706 intel DDX. 4707 4708commit 92f271020a1b9486bd5890a00649aed3678ead62 4709Author: Ben Skeggs <bskeggs@redhat.com> 4710Date: Thu Aug 13 15:40:52 2009 +1000 4711 4712 nv50/xv: provide SYNC_TO_VBLANK attribute 4713 4714commit ada1d36d7097c389c8e21707d36b3cdd6378db94 4715Author: Ben Skeggs <bskeggs@redhat.com> 4716Date: Thu Aug 13 15:38:14 2009 +1000 4717 4718 xv: disable overlay with kms for the moment 4719 4720commit 85b1c86983a39cb921bd68a8601785975c948e2d 4721Author: Ben Skeggs <bskeggs@redhat.com> 4722Date: Mon Aug 10 11:42:15 2009 +1000 4723 4724 dri2: call exaMoveInPixmap() before accessing driver pixmap private 4725 4726commit 292f46a0a0e031fc8a4256aa1eed91e3b2167acd 4727Author: Ben Skeggs <bskeggs@redhat.com> 4728Date: Mon Aug 10 11:36:56 2009 +1000 4729 4730 xv: call exaMoveInPixmap() for driver pixmaps 4731 4732commit 0210fbf01d61fb0541770d541b71396537bcb728 4733Author: Christoph Bumiller <e0425955@student.tuwien.ac.at> 4734Date: Sun Aug 9 14:53:28 2009 +0200 4735 4736 nv50/exa: fix pointer increment in UploadSIFC 4737 4738 We always output 'size' dwords to the ring buffer, so the pointer 4739 should be advanced accordingly, independent of cpp. 4740 4741commit 1ac7d83c1e46a164c4a3a3688de1bcde9e5a58e0 4742Author: Maarten Maathuis <madman2003@gmail.com> 4743Date: Sat Aug 8 19:41:02 2009 +0200 4744 4745 wfb: don't default to linear when encountering a pixmap without bo 4746 4747 - With exa "mixed" there are actually very valid cases where this happens. 4748 - This fixes certain types of corruptions, like the yellow and black in a gimp window. 4749 4750commit 64790e950e7f46e6f4f5b8486f8d0bd929cb5504 4751Author: Maarten Maathuis <madman2003@gmail.com> 4752Date: Fri Aug 7 16:12:41 2009 +0200 4753 4754 Revert "nv50: enforce certain allignments on a tiled frontbuffer, as well as overallocate it a bit" 4755 4756 This reverts commit dd72640c8133ee78304f875023134f214d94a630. 4757 4758 Moved to the kernel. 4759 4760commit dd72640c8133ee78304f875023134f214d94a630 4761Author: Maarten Maathuis <madman2003@gmail.com> 4762Date: Fri Aug 7 11:56:29 2009 +0200 4763 4764 nv50: enforce certain allignments on a tiled frontbuffer, as well as overallocate it a bit 4765 4766 - There is reason to believe that with driver modifications (part of) this won't be needed. 4767 - Until then this should fix "tiles" appearing at the upper left or lower right corner. 4768 4769commit bfbe4d33679cc25939db0da729d89ba2a4295764 4770Author: Ben Skeggs <bskeggs@redhat.com> 4771Date: Fri Aug 7 16:02:27 2009 +1000 4772 4773 exa: default to enabling driver pixmaps if everything else is new enough 4774 4775commit 30ddd08bb2739514edf6d6e137f533b34a4369d2 4776Author: Ben Skeggs <bskeggs@redhat.com> 4777Date: Fri Aug 7 16:01:27 2009 +1000 4778 4779 exa: turn on new exa migration mode for driver pixmaps 4780 4781commit 712064e6188d65e2c7d85078b95a0292edcad0e7 4782Author: Ben Skeggs <bskeggs@redhat.com> 4783Date: Wed Aug 5 14:00:51 2009 +1000 4784 4785 nv04-nv40/exa: set cliprect to drawing area for IFC 4786 4787commit 7f63da83d8622176114ecd55048736fecd703322 4788Author: Ben Skeggs <bskeggs@redhat.com> 4789Date: Wed Aug 5 12:04:46 2009 +1000 4790 4791 dri: close before dropping master 4792 4793commit 0cc44b713ec07dfb56a4b4a6d7ae1c1a75c74971 4794Author: Ben Skeggs <bskeggs@redhat.com> 4795Date: Wed Aug 5 11:57:01 2009 +1000 4796 4797 Set/Drop master in Enter/LeaveVT 4798 4799commit 087e9f61b81ea90a1591d8b43e29bda091156405 4800Author: Ben Skeggs <bskeggs@redhat.com> 4801Date: Wed Aug 5 11:51:54 2009 +1000 4802 4803 dri: close before taking down accel 4804 4805 Fixes "failed to destroy server context" errors on exit. 4806 4807commit 96ac8c5a9473d81b737afa116f4fd388e3152f50 4808Author: Ben Skeggs <bskeggs@redhat.com> 4809Date: Wed Aug 5 11:36:39 2009 +1000 4810 4811 nv04-nv40: deal with framebuffer resize in !kms_enable mode 4812 4813commit b8ea3b41b0540635585bc68334691414f7cff53d 4814Author: Ben Skeggs <bskeggs@redhat.com> 4815Date: Wed Aug 5 11:30:51 2009 +1000 4816 4817 nv50: deal with framebuffer resize for !kms_enable 4818 4819commit d0dcc89825acb4b480f3849d8896b17a80df02f4 4820Author: Ben Skeggs <bskeggs@redhat.com> 4821Date: Wed Aug 5 11:28:45 2009 +1000 4822 4823 No wfb without KMS 4824 4825commit 2c7502680e1ab3a4ea4869a78d0317746084dd92 4826Author: Ben Skeggs <bskeggs@redhat.com> 4827Date: Wed Aug 5 08:55:09 2009 +1000 4828 4829 dri2: fix CopyRegion() 4830 4831commit 2a4c93a064353cf61c30dd1e96f93203895beb15 4832Author: Ben Skeggs <bskeggs@redhat.com> 4833Date: Wed Aug 5 08:36:26 2009 +1000 4834 4835 dri2: merge DRI2BufferRec and nouveau_dri2_buffer into a single struct 4836 4837commit 4f386f983fde197e9136ffe6fbf941a343996897 4838Author: Ben Skeggs <bskeggs@redhat.com> 4839Date: Wed Aug 5 08:20:46 2009 +1000 4840 4841 dri2: remove support for < version 3 4842 4843commit 0f691257e6d2c3662f9c76adc3287f5320a8822d 4844Author: Ben Skeggs <bskeggs@redhat.com> 4845Date: Mon Aug 3 11:30:46 2009 +1000 4846 4847 zfill scanout buffer, not offscreen 4848 4849commit d893600356b3480d3df536a04e9860bfada5a159 4850Author: Maarten Maathuis <madman2003@gmail.com> 4851Date: Sun Aug 2 19:43:58 2009 +0200 4852 4853 kms: remove fb in NVCloseScreen otherwise server regeneration fails 4854 4855 - The bo was deleted, but fb_id remained non-zero, so it never picked up the new one. 4856 4857commit ac2a644773e5b160b6794a252b85c0bf8acc5169 4858Author: Ben Skeggs <bskeggs@redhat.com> 4859Date: Sun Aug 2 20:10:35 2009 +1000 4860 4861 xv: really fix offscreen pixmap calculation 4862 4863 sleep.. good.. 4864 4865commit 60635e5073e69c6f690dc3c876772b5478013783 4866Author: Ben Skeggs <bskeggs@redhat.com> 4867Date: Sun Aug 2 19:42:17 2009 +1000 4868 4869 xv: fix check for offscreen pixmaps 4870 4871commit 1ecd4358c851b3d7bebfb00f5f7ac3c7c151bc20 4872Author: Ben Skeggs <bskeggs@redhat.com> 4873Date: Fri Jul 31 15:30:08 2009 +1000 4874 4875 Recalculate displayWidth if we do a late NoAccel fallback 4876 4877 Fixes G80, where we'd have calculated the pitch for tiled scanout. 4878 4879commit 08ef8aaf2bd7b9565cc36679412b721499f9a367 4880Author: Ben Skeggs <bskeggs@redhat.com> 4881Date: Fri Jul 31 15:23:49 2009 +1000 4882 4883 Don't allocate offscreen memory or GART when using NoAccel 4884 4885commit 64a1515fab9764d0ada304426e7f0d7ec995b84e 4886Author: Ben Skeggs <bskeggs@redhat.com> 4887Date: Fri Jul 31 13:49:49 2009 +1000 4888 4889 Fix NoAccel/ShadowFB with fb resize 4890 4891commit e9034a2e8d8285cf45cd1410b9e8933991f82a2b 4892Author: Ben Skeggs <bskeggs@redhat.com> 4893Date: Fri Jul 31 13:29:11 2009 +1000 4894 4895 NoAccel + wfb == bad idea 4896 4897commit 69a40131b700a4fbc4de9432110045870de0806e 4898Author: Ben Skeggs <bskeggs@redhat.com> 4899Date: Fri Jul 31 13:28:02 2009 +1000 4900 4901 Fix up screen pixmap's devPrivate.ptr after resize 4902 4903commit 4ea65c594a1f10f586da2ca32e499de2ea0a2605 4904Author: Ben Skeggs <bskeggs@redhat.com> 4905Date: Fri Jul 31 13:25:45 2009 +1000 4906 4907 Point fb at the correct buffer.. 4908 4909commit c6b98608daec4162a7d23ce8c1d50a4a9c3ca3dc 4910Author: Ben Skeggs <bskeggs@redhat.com> 4911Date: Fri Jul 31 13:09:56 2009 +1000 4912 4913 Switch on can_resize for non-kms 4914 4915commit 4bb007ebb8e60c87c815f9c4ce63b8d1d46e8b75 4916Author: Ben Skeggs <bskeggs@redhat.com> 4917Date: Fri Jul 31 13:03:45 2009 +1000 4918 4919 Fall back to NoAccel if channel/object creation fails 4920 4921commit 641fdacbd1aeb9882caa30da6d568ff9a0bd78f5 4922Author: Ben Skeggs <bskeggs@redhat.com> 4923Date: Fri Jul 31 11:46:47 2009 +1000 4924 4925 nv50/exa: use scissored triangle for composite too 4926 4927commit c6c0ef1b5a837838a0fc2e8d1659b1b4d76d26b0 4928Author: Ben Skeggs <bskeggs@redhat.com> 4929Date: Fri Jul 31 11:23:04 2009 +1000 4930 4931 nv50/xv: enable sync-to-vblank 4932 4933commit b3d6861fa482a92807aadd9222ca2c8c605beaeb 4934Author: Maarten Maathuis <madman2003@gmail.com> 4935Date: Thu Jul 30 17:23:25 2009 +0200 4936 4937 nv50_xv: switch to a scissored triangle to avoid strange artifacts when vsync is enabled 4938 4939commit 9b37bc96d86882b7f9c355e92cbeff517090a98d 4940Author: Ben Skeggs <bskeggs@redhat.com> 4941Date: Thu Jul 30 19:24:19 2009 +1000 4942 4943 more cleanup 4944 4945commit d8c10eb14f8d6b5cad88a39a5da7104a403060cd 4946Author: Ben Skeggs <bskeggs@redhat.com> 4947Date: Thu Jul 30 10:17:38 2009 +1000 4948 4949 non-kms fb resize 4950 4951commit 3b1a861aa56f823e8d26f33b7139e290bf44b79d 4952Author: Ben Skeggs <bskeggs@redhat.com> 4953Date: Thu Jul 30 07:14:17 2009 +1000 4954 4955 free scanout buffer in NVUnmapMem() 4956 4957commit 4d20547df30d630b94b0cfc28dcf7ead564f7369 4958Author: Ben Skeggs <bskeggs@redhat.com> 4959Date: Tue Jul 28 18:14:03 2009 +1000 4960 4961 nv50: tile scanout buffer for "classic" exa if xserver new enough 4962 4963commit 03bc3a604e5c12829f81f2582374d37012da7f36 4964Author: Ben Skeggs <bskeggs@redhat.com> 4965Date: Tue Jul 28 11:45:51 2009 +1000 4966 4967 kms: only use rounded height for allocation 4968 4969 Fixes minor cursor corruption in some circumstances with driver pixmaps. 4970 4971commit e4fb3e7064df14b21a6ea34b919ccf2bf9fa0ca2 4972Author: Ben Skeggs <bskeggs@redhat.com> 4973Date: Tue Jul 28 11:27:09 2009 +1000 4974 4975 More fixing of pitches + use 0x7000 tile_flags for tiled 16bpp on nv50 4976 4977commit 97cca7630441876414da5dfabb9ad9f2b208e29d 4978Author: Ben Skeggs <skeggsb@beleth.keine.ath.cx> 4979Date: Tue Jul 28 10:21:11 2009 +1000 4980 4981 kms: implement rotation support without driver pixmaps 4982 4983commit 786ff2994aa4f614b7cc9d466455eef183697d42 4984Author: Ben Skeggs <skeggsb@beleth.keine.ath.cx> 4985Date: Tue Jul 28 09:43:00 2009 +1000 4986 4987 kms: simplify drmmode_is_rotate_pixmaps args 4988 4989commit ff62da668a468796e4001523958e73eb2060b7f2 4990Author: Ben Skeggs <bskeggs@redhat.com> 4991Date: Mon Jul 27 22:07:38 2009 +1000 4992 4993 dri1: dodgyness to "survive" a framebuffer resize (or rather, move) 4994 4995 This can never work well without a lot of work, which I'm not willing to 4996 put in for DRI1. This will at least let it keep sort-of working, any DRI1 4997 apps running when a resize event occurs will continue rendering into the 4998 old framebuffer. 4999 5000commit b77d580ebcc9b371153774010d29038645effffb 5001Author: Ben Skeggs <bskeggs@redhat.com> 5002Date: Mon Jul 27 21:43:25 2009 +1000 5003 5004 We already map offscreen and store the virtual, no need to keep doing it 5005 5006commit bdf2c9aee02a100b44c083347a2415c100bd2546 5007Author: Ben Skeggs <bskeggs@redhat.com> 5008Date: Mon Jul 27 21:35:19 2009 +1000 5009 5010 kms: implement framebuffer resize without driver pixmaps 5011 5012commit 595b1c74e52f16ef482be6f14e4a33b4649b8bfd 5013Author: Ben Skeggs <bskeggs@redhat.com> 5014Date: Mon Jul 27 07:44:41 2009 +1000 5015 5016 Fix build with 0.0.15 5017 5018commit 9b879b9b38809e7d206c9e78610836bdd9c2d09b 5019Author: Ben Skeggs <bskeggs@redhat.com> 5020Date: Mon Jul 27 21:09:42 2009 +1000 5021 5022 Cleanup 5023 5024commit 0336eb15ebc351ab6900b48972e12f5cc6abf1bb 5025Author: Ben Skeggs <bskeggs@redhat.com> 5026Date: Mon Jul 27 17:52:15 2009 +1000 5027 5028 Switch to using scanout buffer detached from EXA offscreen memory area 5029 5030commit 0403cf3698ecb9a290bb4b7e24d0deedc957ad2a 5031Author: Ben Skeggs <bskeggs@redhat.com> 5032Date: Mon Jul 27 17:45:29 2009 +1000 5033 5034 Allocate scanout buffer independently of EXA offscreen area 5035 5036 This is just wasted memory for the moment... 5037 5038commit f8ac9e35bd20391f923c5c5ccc509251ea23a4c5 5039Author: Ben Skeggs <bskeggs@redhat.com> 5040Date: Mon Jul 27 17:30:54 2009 +1000 5041 5042 Remove NoDRM mode 5043 5044commit e259925bbdcebe4c3d136f476fc54f426eea9261 5045Author: Ben Skeggs <bskeggs@redhat.com> 5046Date: Mon Jul 27 17:14:00 2009 +1000 5047 5048 xv: fix some badness in nv3x/nv4x textured video adaptors 5049 5050commit 3a10daca16752b52ec3187deb8b1c25d6c6fd641 5051Author: Ben Skeggs <bskeggs@redhat.com> 5052Date: Mon Jul 27 16:05:03 2009 +1000 5053 5054 rename pNv->FB to pNv->offscreen 5055 5056 Not strictly true right now, but it's about to be. 5057 5058commit 91bfda81982a77a79c2baee39929711b38b1ae6a 5059Author: Ben Skeggs <bskeggs@redhat.com> 5060Date: Mon Jul 27 11:46:51 2009 +1000 5061 5062 exa: allow acceleration to rotation bo 5063 5064commit d59aaa81590b52b6227b69028f08d8ab62505525 5065Author: Ben Skeggs <bskeggs@redhat.com> 5066Date: Mon Jul 27 11:02:44 2009 +1000 5067 5068 exa: use PixmapIsOffscreen hook for non-driver pixmaps path too 5069 5070commit c442376d4e9a8296760c1171fee5ca155534d011 5071Author: Ben Skeggs <bskeggs@redhat.com> 5072Date: Mon Jul 27 07:44:14 2009 +1000 5073 5074 Fix typo exposed by removal of NV_* card type defines from DRM header 5075 5076commit 15bfc676094d19e57711bbdff40edd5c8e8592d6 5077Author: Ben Skeggs <bskeggs@redhat.com> 5078Date: Mon Jul 27 07:40:20 2009 +1000 5079 5080 Warning fixes 5081 5082commit 6eb790b6113a999727895694d01910b83348647a 5083Author: Ben Skeggs <bskeggs@redhat.com> 5084Date: Fri Jul 24 13:54:49 2009 +1000 5085 5086 kms: keep rotate buffer pitch in crtc struct, rather than recalculating 5087 5088commit 25f29426293fcac2e7a17b986673b2874e7a7264 5089Author: Ben Skeggs <bskeggs@redhat.com> 5090Date: Fri Jul 24 13:18:22 2009 +1000 5091 5092 Wrap CreateScreenResources and set screen pixmap's BO there 5093 5094 Avoids reallocating pNv->FB yet again when we hit the resize hook. 5095 5096commit d1628dc9159343400f09f696b9c4ef9d0fe1acbf 5097Author: Maarten Maathuis <madman2003@gmail.com> 5098Date: Tue Jul 28 10:08:40 2009 +0200 5099 5100 nouveau: remove some more RAC bits 5101 5102 - The assumption is that RAC is dead anyway on post-libpciaccess xservers. 5103 5104commit 75fba5dfd2a00f3b8212d0e8af544273c14ee0c5 5105Author: Dave Airlie <airlied@redhat.com> 5106Date: Tue Jul 28 13:53:06 2009 +1000 5107 5108 nouveau: rip out resources code whats left at least 5109 5110commit 094e4e075f8e48ace534a6f011f95b6ebf1abc3b 5111Author: Ben Skeggs <skeggsb@beleth.keine.ath.cx> 5112Date: Thu Jul 23 11:25:22 2009 +1000 5113 5114 kms: don't skip resize call if we don't have a backing bo for screen pixmap 5115 5116commit bff801941e23dfe3fb7449af5780ee477b10d092 5117Author: Ben Skeggs <skeggsb@beleth.keine.ath.cx> 5118Date: Thu Jul 23 11:23:15 2009 +1000 5119 5120 Fix a DRM vs DDX confusion typo 5121 5122commit e9ebddae840789535acf56ac2281b3fde3cfee96 5123Author: Maarten Maathuis <madman2003@gmail.com> 5124Date: Wed Jul 22 12:51:08 2009 +0200 5125 5126 Fix build without AM_SILENT_RULES. 5127 5128commit f70287cf7cacc03d8390376519165d50a325cec5 5129Author: Peter Henriksson <peter.henriksson@gmail.com> 5130Date: Mon Jul 20 23:56:13 2009 +0200 5131 5132 Add support for AM_SILENT_RULES 5133 5134 Make use of the AM_SILENT_RULES functionality in automake 1.11 5135 Defaults to quiet output. Verbose output can be enabled with 5136 "./configure --disable-silent-rules" or "make V=1" 5137 5138 Signed-off-by: Ben Skeggs <bskeggs@redhat.com> 5139 5140commit 69ba1095a9aef29ce8fdb71ecf48f15cb13202c2 5141Author: Ben Skeggs <bskeggs@redhat.com> 5142Date: Tue Jul 21 16:29:13 2009 +1000 5143 5144 nv50/xv: attempt sync-to-vblank if driver pixmaps are being used 5145 5146commit 994c37927289de7a0ab6e951892f0d66880a6210 5147Author: Ben Skeggs <bskeggs@redhat.com> 5148Date: Tue Jul 21 14:05:59 2009 +1000 5149 5150 nv50: fix a number of tiled buffer sizing issues in kms paths 5151 5152commit 9c1eefd221366b9f5efbd30e4a28fba244347d4f 5153Author: Ben Skeggs <bskeggs@redhat.com> 5154Date: Tue Jul 21 13:43:14 2009 +1000 5155 5156 Don't set driver pixmaps enabled with NoAccel 5157 5158commit 52882e82708f79ef198dde51f9bbebee752154be 5159Author: Ben Skeggs <bskeggs@redhat.com> 5160Date: Tue Jul 21 13:19:07 2009 +1000 5161 5162 exa: fix nouveau_exa_pixmap_is_onscreen() for driver pixmaps 5163 5164commit 9fb28d107ad9fcd6a4bbb0a35c1fc0f719d531ae 5165Author: Ben Skeggs <bskeggs@redhat.com> 5166Date: Tue Jul 21 12:52:35 2009 +1000 5167 5168 exa: remove ModifyPixmapHeader hook, it's not required 5169 5170commit 574fa36c73a1b04b67289762a10b5bbda7be083f 5171Author: Ben Skeggs <bskeggs@redhat.com> 5172Date: Tue Jul 21 12:51:26 2009 +1000 5173 5174 exa: only support driver pixmaps on servers with CreatePixmap2 5175 5176commit 365ab69499f466aeebac1437578609a7790f5dc4 5177Author: Ben Skeggs <bskeggs@redhat.com> 5178Date: Fri Jul 17 13:41:07 2009 +1000 5179 5180 remove explicit pin of pNv->FB on creation 5181 5182commit a757db1aadfd7a7d206dfc7968da8647dd91e65f 5183Author: Ben Skeggs <bskeggs@redhat.com> 5184Date: Fri Jul 17 13:38:52 2009 +1000 5185 5186 no need to pin GART bo 5187 5188commit 474ebdeea1266959fddeabf04b24e04aaab8449e 5189Author: Ben Skeggs <bskeggs@redhat.com> 5190Date: Fri Jul 17 13:37:29 2009 +1000 5191 5192 remove pinned creation from nodrm path 5193 5194commit b9333d5d34386d88b58d2267b388bbcc89b73f47 5195Author: Ben Skeggs <bskeggs@redhat.com> 5196Date: Fri Jul 17 13:35:20 2009 +1000 5197 5198 nv04-nv40: explicitly pin scanout buffer when required 5199 5200commit 53b5b68cd168b15c377f83e5e068cee92361f51d 5201Author: Ben Skeggs <bskeggs@redhat.com> 5202Date: Fri Jul 17 13:31:13 2009 +1000 5203 5204 nv50: no need to pin rotate scanout buffers on creation now 5205 5206commit a57a9bb825e04bbe4aa94d9ac1149e2fca6525a4 5207Author: Ben Skeggs <bskeggs@redhat.com> 5208Date: Fri Jul 17 13:28:36 2009 +1000 5209 5210 exa: only allow driver pixmaps when KMS enabled 5211 5212 The pre-KMS code can't deal with driver pixmaps in some circumstances, 5213 and rather than leave it there for people to accidently trip on, 5214 remove support. 5215 5216 The other option would have been to fix the UMS code, but, who really 5217 cares? 5218 5219commit 5d7a0b954c0da879f81a6fa1a9dc8070c58a0fe6 5220Author: Ben Skeggs <bskeggs@redhat.com> 5221Date: Fri Jul 17 13:25:31 2009 +1000 5222 5223 remove remains of prehistoric NOUVEAU_EXA_PIXMAPS stuff 5224 5225commit 784a0337c3165c8d8fb696e4fa30bbdc8cf01b98 5226Author: Ben Skeggs <bskeggs@redhat.com> 5227Date: Fri Jul 17 13:18:06 2009 +1000 5228 5229 nv50: explicitly pin scanout buffer in non-kms path 5230 5231commit 09db6180e4205ef55516c24ac5dfc3bd861d9fd6 5232Author: Ben Skeggs <bskeggs@redhat.com> 5233Date: Fri Jul 17 13:12:25 2009 +1000 5234 5235 nv30: don't pin shader memory 5236 5237commit 7da55d99e3c93cadda50dcad06b09c52daa0cc5b 5238Author: Ben Skeggs <bskeggs@redhat.com> 5239Date: Fri Jul 17 13:10:16 2009 +1000 5240 5241 xv: don't pin buffer unless using the overlay 5242 5243commit e8b9f689b778909e710fc5bd5616e6497b922a09 5244Author: Ben Skeggs <bskeggs@redhat.com> 5245Date: Fri Jul 17 13:07:04 2009 +1000 5246 5247 kms: don't pin scanout buffer, no need, kernel will do it 5248 5249commit b1b233099853b56cfdcff9c4c31d3550b76b949e 5250Author: Ben Skeggs <bskeggs@redhat.com> 5251Date: Fri Jul 17 08:22:02 2009 +1000 5252 5253 Fix build with xextproto 7.1 5254 5255commit c089a71a94e7887c405af7141e188302a5e2d393 5256Author: Ben Skeggs <bskeggs@redhat.com> 5257Date: Thu Jul 16 09:32:59 2009 +1000 5258 5259 dri2: fix for xserver abi change 5260 5261commit ef3a386a6df4569343e63332eb798c1d722fe164 5262Author: Ben Skeggs <bskeggs@redhat.com> 5263Date: Mon Jul 6 20:33:49 2009 +1000 5264 5265 Update manpage + finish removing useless options 5266 5267commit 4b6e0b96188e32cb2a13f8d2bd5973f0287c827b 5268Author: Ben Skeggs <bskeggs@redhat.com> 5269Date: Mon Jul 6 20:25:23 2009 +1000 5270 5271 No need for pci access wrappers now.. 5272 5273commit 1e22673f43acd0f02ba36e189b39e28b635cd50c 5274Author: Ben Skeggs <bskeggs@redhat.com> 5275Date: Mon Jul 6 20:16:02 2009 +1000 5276 5277 Remove non-XSERVER_LIBPCIACCESS paths, we depend on a new xserver anyway 5278 5279commit aafb62f86adebb3418876649e8bd189c1523e961 5280Author: Ben Skeggs <bskeggs@redhat.com> 5281Date: Mon Jul 6 20:08:03 2009 +1000 5282 5283 Some more bits of NVRec to go... 5284 5285commit 0ba588cb75af67a644f51add35778925d4088cf9 5286Author: Ben Skeggs <bskeggs@redhat.com> 5287Date: Mon Jul 6 19:58:15 2009 +1000 5288 5289 Rip out vgaHW + more unused vars from NVRec 5290 5291commit 3e22547b6826ae741b168c4b3391b6ade56b4829 5292Author: Ben Skeggs <bskeggs@redhat.com> 5293Date: Mon Jul 6 19:46:26 2009 +1000 5294 5295 Nuke nv_hw.c, and a heap of unused vars from NVRec 5296 5297commit eea5259d951a5919fa7800877570347c14411a32 5298Author: Ben Skeggs <bskeggs@redhat.com> 5299Date: Mon Jul 6 19:13:44 2009 +1000 5300 5301 Remove nv_dac.c 5302 5303commit a38b91a7a5c3337d405aa217830857832c5399e0 5304Author: Ben Skeggs <bskeggs@redhat.com> 5305Date: Mon Jul 6 19:09:09 2009 +1000 5306 5307 Cull everything in !pNv->randr12_enabled blocks 5308 5309commit 4815d312911cb4cfad4b4a5b194fce41d82f4869 5310Author: Ben Skeggs <bskeggs@redhat.com> 5311Date: Mon Jul 6 11:44:26 2009 +1000 5312 5313 kms: fix CRTC offset when rotation enabled 5314 5315commit 6d143272f497a61c5578752f58261befbfaf911a 5316Author: Ben Skeggs <bskeggs@redhat.com> 5317Date: Wed Jul 1 14:03:06 2009 +1000 5318 5319 nv50: initial support for depth 30 5320 5321commit 9b0919c3f3269b88fb338c043bbad34f20c22f41 5322Author: Ben Skeggs <bskeggs@redhat.com> 5323Date: Tue Jun 30 13:20:15 2009 +1000 5324 5325 kms: don't try transition when we don't have src+dst buffer ids 5326 5327commit c0bf670ac6b58cff60a01ab6b174ece6b1b7d892 5328Author: Ben Skeggs <bskeggs@redhat.com> 5329Date: Thu Jun 25 07:26:54 2009 +1000 5330 5331 wfb: fix stupid thinko + more safety 5332 5333commit a12cb5c8c0a85a601188178c5fc0d84a13abd87e 5334Author: Ben Skeggs <bskeggs@redhat.com> 5335Date: Thu Jun 18 14:28:25 2009 +1000 5336 5337 bios: oops 5338 5339commit e66867e4cf9f0bc8a1971664ccc3d5c56b08b2fb 5340Author: Ben Skeggs <bskeggs@redhat.com> 5341Date: Tue Jun 16 10:28:00 2009 +1000 5342 5343 bios: use image from PRAMIN in preference to PROM on NV50 5344 5345 There's at least one known case (rh#492658) where the DCB table present in 5346 the VBIOS image from PROM is not suitable for use. It contained all 16 5347 entries filled, each entry valid in itself, but contradicting other entries. 5348 5349 The VBIOS image in PRAMIN however, still has all 16 entries filled, but the 5350 first few entries now match what is present on the hardware, and the rest 5351 are set as type 0xf, which we ignore. 5352 5353commit 42c5730a4f6c243b75fbd40f1dc26e8433191c7b 5354Author: Ben Skeggs <bskeggs@redhat.com> 5355Date: Tue Jun 16 09:48:39 2009 +1000 5356 5357 wrap BlockHandler a little earlier 5358 5359commit 09e663aea82403f69a0e44e3b7cd89dfe96970d5 5360Author: Ben Skeggs <bskeggs@redhat.com> 5361Date: Mon Jun 15 16:09:27 2009 +1000 5362 5363 kms: safer fb resize func 5364 5365commit 90be5d5d6a696564bda23e9a2dad14a3a1808868 5366Author: Ben Skeggs <bskeggs@redhat.com> 5367Date: Mon Jun 15 13:43:21 2009 +1000 5368 5369 wfb: use straight memcpy hook if no tiled wraps present 5370 5371commit 86d905d34a5693e3d9b986660d4a2bcd6f30a06a 5372Author: Ben Skeggs <bskeggs@redhat.com> 5373Date: Mon Jun 15 13:41:21 2009 +1000 5374 5375 wfb: be more cautious in a few places 5376 5377commit 1bcbc4b26f71f90345bd82b836fc634e741aced8 5378Author: Ben Skeggs <bskeggs@redhat.com> 5379Date: Mon Jun 15 10:19:53 2009 +1000 5380 5381 kms: small cleanup 5382 5383commit c2d3550aafeb3e7c4df3334f0ac3554a8aa15da3 5384Author: Ben Skeggs <bskeggs@redhat.com> 5385Date: Mon Jun 15 09:16:10 2009 +1000 5386 5387 kms: fix rotation buffer pitch 5388 5389commit 6fd982979e03ce3eace7dc7523c428aca2c30b4a 5390Author: Ben Skeggs <bskeggs@redhat.com> 5391Date: Mon Jun 15 09:00:02 2009 +1000 5392 5393 kms: point rotation pixmap at correct buffer 5394 5395commit 67f8ebe77f047c12d76e508b4d7f6384baed39ac 5396Author: Ben Skeggs <bskeggs@redhat.com> 5397Date: Mon Jun 15 08:52:36 2009 +1000 5398 5399 nv50: create non-linear scanout buffers for rotation 5400 5401commit b7e3306ddc3693699f6f9de7b22913ee22ed31ed 5402Author: Stuart Bennett <stuart@freedesktop.org> 5403Date: Sun May 31 00:26:13 2009 +0100 5404 5405 Add DVI-A output info for Mac card missing bios tables (#21273) 5406 5407 The reporter indicates the card also has DVI-D and ADC, but those can't be 5408 tested to deduce the fake correct data at present 5409 5410commit 01b19c9fc25b147596e1719008ed9d861246ced9 5411Author: Ben Skeggs <bskeggs@redhat.com> 5412Date: Wed Jun 10 18:00:05 2009 +1000 5413 5414 nv50: sigh 5415 5416commit 4d9e63baff079fad727430fc605cb1e43c4303a7 5417Author: Ben Skeggs <bskeggs@redhat.com> 5418Date: Wed Jun 10 11:38:43 2009 +1000 5419 5420 wfb: "mode0" tiles have the same pitch as the others 5421 5422 This appears to be the case now (as in: after making the GPU do some of the 5423 dirty work) even though it didn't in the previous wfb patches. 5424 5425commit 02ed6b66e37e13ab4064974b89db2e73613108ef 5426Author: Ben Skeggs <bskeggs@redhat.com> 5427Date: Wed Jun 10 11:23:37 2009 +1000 5428 5429 wfb: work-around wfb suckage 5430 5431 wfb sucks far more than anticipated: 5432 - can only expose a single set of rd/wr hooks, not per-wrap as wfb doesn't 5433 treat the hooks as per-pixmap. 5434 - the hooks will get called on buffers that aren't even pixmaps 5435 5436commit 669c59ecb69a6f84ef8b8714cda6e87eacb3de9f 5437Author: Ben Skeggs <bskeggs@redhat.com> 5438Date: Wed Jun 10 09:15:37 2009 +1000 5439 5440 wfb: need a 64-bit datatype for multiply_factor 5441 5442commit 86dedb748cb6d047b195e50e921e840c1c403f6f 5443Author: Ben Skeggs <bskeggs@redhat.com> 5444Date: Tue Jun 9 22:25:01 2009 +1000 5445 5446 fix some pitch issues 5447 5448commit 794a27724fb9ab173dc7ea8062730797e1b4de04 5449Author: Ben Skeggs <bskeggs@redhat.com> 5450Date: Tue Jun 9 10:54:54 2009 +1000 5451 5452 nv50: use libwfb for pixmap access when driver pixmaps enabled 5453 5454commit 11d9690dd9e1190d1f6873bb9a5df8f98a39d4f5 5455Author: Ben Skeggs <bskeggs@redhat.com> 5456Date: Tue Jun 9 09:10:57 2009 +1000 5457 5458 exa: use the sane CreatePixmap hook, if available 5459 5460commit 6ee453381f00063522187939cb54ab9447b73ec7 5461Author: Ben Skeggs <bskeggs@redhat.com> 5462Date: Thu Jun 4 14:43:54 2009 +1000 5463 5464 xv: fix some issues with driver pixmaps 5465 5466commit 67c5287f31bf20ab3a38ffcdee410d9d49ce4d2b 5467Author: Ben Skeggs <bskeggs@redhat.com> 5468Date: Thu Jun 4 14:19:14 2009 +1000 5469 5470 exa: wait_marker becomes a NOP with driver pixmaps 5471 5472 mapping/unmapping buffers will sync as necessary 5473 5474commit f44e52852a199196f8428822bf9fb71711b9c4f9 5475Author: Ben Skeggs <bskeggs@redhat.com> 5476Date: Thu Jun 4 14:08:10 2009 +1000 5477 5478 nv50: make use of larger tile sizes 5479 5480commit 30c44ce3142ba44fc7a6c95ceaa9bf028cc112c1 5481Author: Ben Skeggs <bskeggs@redhat.com> 5482Date: Thu Jun 4 13:41:36 2009 +1000 5483 5484 exa: create unacceleratable pixmaps (ie. 1bpp) in system memory 5485 5486commit 1aa22d5894c4e4efe746592bf1c51cdaf0c31c24 5487Author: Ben Skeggs <bskeggs@redhat.com> 5488Date: Thu Jun 4 13:20:37 2009 +1000 5489 5490 nv50: use non-linear scanout buffer when driver pixmaps enabled 5491 5492commit 13d8d49f74fb10f631bcd38b169068e691b9ca99 5493Author: Ben Skeggs <bskeggs@redhat.com> 5494Date: Thu Jun 4 13:11:40 2009 +1000 5495 5496 nv50: use tile_mode from bo 5497 5498 should have no effect currently, but needed later 5499 5500commit bd9f5f2cbe7687aba7bb3a7f397d9b1bf87a8ef3 5501Author: Ben Skeggs <bskeggs@redhat.com> 5502Date: Thu Jun 4 12:11:48 2009 +1000 5503 5504 exa: fix compile against latest libdrm_nouveau 5505 5506commit 5f97afe38095a274938b8c2e9c5ebd17b7733bb6 5507Author: Stuart Bennett <stuart@freedesktop.org> 5508Date: Wed Jun 3 12:16:23 2009 +0100 5509 5510 randr12: split from pre-randr12 structs into new header, de-typedef 5511 5512 New nv_crtc->state shortcut and some renaming included for convenience 5513 5514commit 0c17b8790086a529ee9e4b5aca4b6f19f31900a9 5515Author: Stuart Bennett <stuart@freedesktop.org> 5516Date: Thu May 28 22:40:30 2009 +0100 5517 5518 Move head getting into nouveau_hw.c 5519 5520commit 12314fa6abb2dd6d325a370724c7d16c371c229c 5521Author: Stuart Bennett <stuart@freedesktop.org> 5522Date: Thu May 28 21:45:36 2009 +0100 5523 5524 randr12: improve uniformity of props code 5525 5526commit b541c1c4400942c59be37f62b7a25719dbdaa1a1 5527Author: Stuart Bennett <stuart@freedesktop.org> 5528Date: Thu May 28 18:23:37 2009 +0100 5529 5530 Remove "TMDS table script pointers not stubbed" bios warning 5531 5532 Scripts appear to be init scripts (+7 one for 0x6808b?, +9 for 0x6828b?), 5533 which when the pointer is non-zero (and the script isn't 'q') get called 5534 at POST (at least on my nv4b). 5535 5536 The warning, from a time when the scripts were thought to perhaps be needed 5537 for setting up outputs, seems redundant. 5538 5539commit 8af50289136e562198b5acc851629606d9ce450f 5540Author: Adam Jackson <ajax@redhat.com> 5541Date: Thu May 28 15:08:35 2009 -0400 5542 5543 Remove useless loader symbol lists. 5544 5545commit 317b58166bf9f3189f33e99493cf8a1b40e014b8 5546Author: Stuart Bennett <stuart@freedesktop.org> 5547Date: Wed May 13 19:05:25 2009 +0100 5548 5549 randr12: disable lvds (invalidate all modes) if bios lvds parsing fails 5550 5551commit e897191c7bcf61193576ecf52d23fcddf5b70634 5552Author: Stuart Bennett <stuart@freedesktop.org> 5553Date: Thu May 28 16:42:29 2009 +0100 5554 5555 randr12: better behaviour (avoid crash) when fp native mode can't be found 5556 5557commit 03167483b41f7205dd1efc5b571ec73ec4b3cdb5 5558Author: Stuart Bennett <stuart@freedesktop.org> 5559Date: Thu May 28 02:18:52 2009 +0100 5560 5561 randr12: line length improvements 5562 5563commit b60c16cdcbeff2d02ea0866dfbb742c3880b7ae4 5564Author: Stuart Bennett <stuart@freedesktop.org> 5565Date: Wed May 13 02:21:17 2009 +0100 5566 5567 Simplify tests for digital fp outputs 5568 5569commit 992d4b5d3aa04f88c22e52e25dc2e3e7d9bd776e 5570Author: Stuart Bennett <stuart@freedesktop.org> 5571Date: Thu May 28 15:59:12 2009 +0100 5572 5573 randr12: a bunch of trivial improvements and tidyups 5574 5575commit 6f0a324a503431d757b2247e7232de2eb7c318a5 5576Author: Stuart Bennett <stuart@freedesktop.org> 5577Date: Wed May 27 16:17:38 2009 +0100 5578 5579 Fix oopsy from 81bbdd4e causing broken framebuffer 5580 5581commit c5799186b1f70df0484c6cef63c8485ddf924733 5582Author: Stuart Bennett <stuart@freedesktop.org> 5583Date: Wed May 27 16:01:42 2009 +0100 5584 5585 randr12: fix two colour cursor on second head 5586 5587 Colour caching was preventing cursor from being set correctly on second head 5588 5589commit 81bbdd4ee2e942d91d056a10ef993a8f41651c64 5590Author: Ben Skeggs <bskeggs@redhat.com> 5591Date: Tue May 26 15:48:25 2009 +1000 5592 5593 explicitly mark buffers mappable 5594 5595commit 6c09ad5296820a845571d07cbe42d88bb19560df 5596Author: Stuart Bennett <stuart@freedesktop.org> 5597Date: Wed May 27 00:53:51 2009 +0100 5598 5599 Tolerate missing fp bios table (rh#502371) 5600 5601 Modern desktop IGPs frequently miss it out, and now it seems even some 5602 laptop parts (MXM style) do too 5603 5604commit db9ff958291416dd5b72d20d32d7412ec45c1e60 5605Author: Ben Skeggs <bskeggs@redhat.com> 5606Date: Mon May 25 08:31:07 2009 +1000 5607 5608 kms: clip sw transition dims to smallest common area 5609 5610commit 2688c97169d8ea579ba09b9db049e31704f96626 5611Author: Ben Skeggs <bskeggs@redhat.com> 5612Date: Mon May 25 08:29:47 2009 +1000 5613 5614 don't run vbios parser when kms enabled 5615 5616commit 9656762ba186e91dd2b3b7f3f9427ba520982f9d 5617Author: Ben Skeggs <bskeggs@redhat.com> 5618Date: Thu May 14 10:26:44 2009 +1000 5619 5620 nv50: fix multiple-display hangs when encoders swap crtcs 5621 5622commit d70eed91b2909ba0b20415ab1d5a099fb6b39b08 5623Author: Stuart Bennett <stuart@freedesktop.org> 5624Date: Tue May 12 16:54:33 2009 +0100 5625 5626 randr12: fix digital dpms regression 5627 5628 introduced in 3971dda57004894d5d4fc9420aa00da400815af9 5629 5630 setting two different "blank" DPMS modes (say Suspend, followed by Off) 5631 consecutively would lead to the saved fp control value getting overwritten 5632 with FP_TG_CONTROL_OFF flavoured state, so DPMS on would then never work 5633 5634 also, an improvement to LVDS DPMS state caching 5635 5636commit 10721038b854d9945ae551c6aa382b90f5d743e1 5637Author: Ben Skeggs <skeggsb@gmail.com> 5638Date: Thu May 7 21:39:41 2009 +1000 5639 5640 exa: use exaDriverAlloc() to prevent issues across exa changes 5641 5642commit 8502a8024354b88a4f19c7ad355268bb20772356 5643Author: Stuart Bennett <stuart@freedesktop.org> 5644Date: Wed May 6 18:01:54 2009 +0100 5645 5646 Fix server regeneration again 5647 5648 The DRM is now opened in PreInit, so it must not be closed in CloseScreen, 5649 as the DRM wouldn't then be open in the subsequent ScreenInit 5650 5651commit 7a796a94d0c0e18a77c5c50a0f9ca9c218d7201f 5652Author: Stuart Bennett <stuart@freedesktop.org> 5653Date: Thu Apr 30 16:04:59 2009 +0100 5654 5655 randr12: off-chip lvds for nv28 (rh#487456) 5656 5657 Quite how this possibly works with the off-chip DVI connector the laptop 5658 in question also claims to have is unclear: I suspect that (at least in 5659 nouveau's implementation) it doesn't. Sadly we've no dvi-in-use reg dump 5660 to see what should be done. 5661 5662commit 5959512ef5ce20aa52bbd41d5d987392f27b9e65 5663Author: Stuart Bennett <stuart@freedesktop.org> 5664Date: Wed May 6 14:44:40 2009 +0100 5665 5666 'drmCheckModesettingSupported' implicit function declaration fix 5667 5668commit f62719e2ac8b04d4496dc56d5fa181934bcea9f7 5669Author: Peter Hjalmarsson <xake@rymdraket.net> 5670Date: Wed May 6 14:43:22 2009 +0100 5671 5672 Unused variable warning fixes 5673 5674commit c8260ef52b58234335bdbfab31364d3d25840c21 5675Author: Pierre Pronchery <khorben@defora.org> 5676Date: Wed May 6 14:32:34 2009 +0100 5677 5678 Fix crash with Xinerama enabled 5679 5680 From the mailing list, minor tweaks by Stuart Bennett 5681 5682commit 1a478edbd58bb2eb1a41f22dea893c6d85902318 5683Author: Stuart Bennett <stuart@freedesktop.org> 5684Date: Wed May 6 02:23:42 2009 +0100 5685 5686 randr12: de-magic ramdac general control values 5687 5688commit 6c209a87d2d90b39fddc446378b370bd7fb81f6b 5689Author: Stuart Bennett <stuart@freedesktop.org> 5690Date: Tue Apr 7 14:37:05 2009 +0100 5691 5692 randr12: pre-nv17 load detection 5693 5694 Sampling heuristic as close to nvidia's as mmiotrace-based inference admits 5695 5696 Works on both nv05 and nv11 5697 5698commit 3971dda57004894d5d4fc9420aa00da400815af9 5699Author: Stuart Bennett <stuart@freedesktop.org> 5700Date: Thu Apr 30 19:31:27 2009 +0100 5701 5702 randr12: fix fp_control again again again 5703 5704 As prophesied, 9c991d3a84ad16f893b739282403eb01e02bb57f broke something: 5705 turning off a digital output, then turning on an analogue output on the 5706 same crtc as previously used by the digital output led to FP_TG_CONTROL 5707 being turned on again when the (unmodified by analogue mode_set) fp_control 5708 state was written out 5709 5710 So, belt and braces attempt to get this right, one more time 5711 5712commit 0ef16031c7e27ffabb0299e9463def125bd5e72c 5713Author: Stuart Bennett <stuart@freedesktop.org> 5714Date: Thu Apr 30 15:27:33 2009 +0100 5715 5716 randr12: solve off-chip encoder crtc exclusion in prepare, rather than dpms 5717 5718commit 0447ce06989d5599e268fdb8a4ba7bebf72f5c95 5719Author: Stuart Bennett <stuart@freedesktop.org> 5720Date: Sat Apr 4 20:08:32 2009 +0100 5721 5722 randr12: simplify nv11 digital encoder-crtc binding 5723 5724 Always setting |0x10 on the appropriate crtc for nv11 digital is fine; 5725 it's what the proprietary driver does, even if the bios doesn't 5726 5727 Fix some line lengths in nv_output_mode_set() too 5728 5729commit f69b34aa92050987e8f3da07d92d5698e9333b75 5730Author: Ben Skeggs <skeggsb@gmail.com> 5731Date: Fri May 1 09:07:55 2009 +1000 5732 5733 nv50: return immediately in GetDDCModes if no DDC on connector... 5734 5735commit 45e16187612bc09b234b598dff25bbcc8069fde1 5736Author: Ben Skeggs <skeggsb@gmail.com> 5737Date: Thu Apr 30 16:37:29 2009 +1000 5738 5739 bios: use NV_ARCH_50 to match G8x chips 5740 5741 There's at least a couple of chips where we get 0x77 for chip_version, 5742 and fail to match the chip_version >= 0x80 test.. 5743 5744commit 3d61697c477dd7358e5fd4e5bc608b4fe763b5d3 5745Author: Ben Skeggs <skeggsb@gmail.com> 5746Date: Thu Apr 30 10:23:11 2009 +1000 5747 5748 nv50: remove reading back vbios-programmed lvds native mode 5749 5750commit bade2499db8e02dd2904bc811399c891bf126208 5751Author: Ben Skeggs <bskeggs@redhat.com> 5752Date: Thu Apr 30 09:33:21 2009 +1000 5753 5754 nv50: group encoders into connectors by i2c port 5755 5756commit 9ee2ac19d7adc067a5551e6d6853a19b89946d46 5757Author: Ben Skeggs <skeggsb@gmail.com> 5758Date: Wed Apr 29 16:26:41 2009 +1000 5759 5760 bios: remove dodgy mode-table search for g80 5761 5762 This appears to work in exactly the same way as on earlier chipsets, the 5763 strap value just moved slightly.. 5764 5765commit 6282574dbeeb7075a00ce03940c4f2a597a048b5 5766Author: Ben Skeggs <skeggsb@gmail.com> 5767Date: Wed Apr 29 16:25:43 2009 +1000 5768 5769 bios: modify get_fp_strap() for g80 5770 5771commit 09b832e24b9bd887275ba882beabe9bba56f0207 5772Author: Ben Skeggs <skeggsb@gmail.com> 5773Date: Tue Apr 28 09:46:55 2009 +1000 5774 5775 nv50: merge NV50SorSetClockMode and NV50SorSetClockModeLVDS 5776 5777commit ed2c185caeb75c595dfb2e814748a856a23b2b35 5778Author: Ben Skeggs <skeggsb@gmail.com> 5779Date: Tue Apr 28 08:44:01 2009 +1000 5780 5781 bios/nv50: initial parsing of display script tables, not quite complete 5782 5783commit 3fccc9e83a194fd58c333c6c4793a27cb05ce026 5784Author: Ben Skeggs <skeggsb@gmail.com> 5785Date: Tue Apr 28 09:38:21 2009 +1000 5786 5787 bios: some nv50 init scripts have flags in reg values, deal with (1<<30) 5788 5789commit 2b4b3da73babbe9ebf4acc26238a4b2b8a6e6ee8 5790Author: Ben Skeggs <skeggsb@gmail.com> 5791Date: Wed Apr 29 11:00:17 2009 +1000 5792 5793 bios: dcb location is only 2 bits 5794 5795commit 7b5bfff21916461ce7f6a2adb27dc443ad0b8201 5796Author: Ben Skeggs <skeggsb@gmail.com> 5797Date: Tue Apr 28 08:12:05 2009 +1000 5798 5799 bios: parse BIT U table 5800 5801commit e4260ec4f0303b286f16eab4fe5a1bb66ff89e3d 5802Author: Stuart Bennett <stuart@freedesktop.org> 5803Date: Wed Apr 1 02:56:05 2009 +0100 5804 5805 randr12: more use of logging abstraction in modesetting code 5806 5807commit 4d0fdba9abcd0af736821d54baca5f0c9a8d6ce1 5808Author: Stuart Bennett <stuart@freedesktop.org> 5809Date: Mon Apr 6 01:14:04 2009 +0100 5810 5811 randr12: fix for off-chip dual link digital (enable extra wide interface?) 5812 5813commit bb246d72a01285299599cbf8ff82c67f14682a4f 5814Author: Stuart Bennett <stuart@freedesktop.org> 5815Date: Mon Apr 6 01:00:41 2009 +0100 5816 5817 Partial fix for nv28 lvds bios parsing (part of rh#487456) 5818 5819 For rh#487456, though this is neither a complete fix for that bug, nor a 5820 complete fix for nv25/28 lvds (the EDID matching mentioned in the commit is 5821 omitted until someone appears with a laptop requiring it) 5822 5823commit 2915926702b7ae9ccb1e900c07ca14a2f32e4512 5824Author: Stuart Bennett <sb476@cam.ac.uk> 5825Date: Mon Apr 7 23:20:12 2008 +0100 5826 5827 plls: max_log2p_bias is likely actually max_log2p 5828 5829 As explained in nouveau_bios.h, a separate max_usable_log2p is needed 5830 (lack of change from 6 to 7 tested by setting a stable small mode, say 5831 400x300, then manually tweaking the multiplier up and post divider down). 5832 The max_usable_log2p values hardcoded here are unchanged from those 5833 previously taken from the bios pll calculation code 5834 5835 For reference, the blob will quite happily attempt to set clocks using the 5836 raw parsed max_log2p values (and go up to 7 therefore) when setting modes 5837 around 200x200 (3MHz pxclk), but at such a low clock my monitor won't agree 5838 to give a sensible picture anyway. Reg dumped data using "nvidia" below: 5839 5840 nv40: log2P up to 7 (dual pll mode), at very low clocks driver gives up and sets 0x0006ff0d 0x80001f04 5841 nv43: log2P up to 7 (dual pll mode), at very low clocks driver gives up and sets 0x0006ff0d 0x80001f04 5842 nv11: single pll, therefore goes quite happily down to at least 0.25MHz 5843 nv31: log2P up to 7 (dual pll mode), at very low clocks card locks up 5844 nv34: log2P up to 5 (single pll), goes down to at least 2.25MHz 5845 nv4b: log2P up to 6 (single pll mode), at very low clocks driver gives up and sets 0x0006ff0d 0x80001f04 5846 5847commit 47bb00fee8b9906adc03b372efa1ae813bc9ca9a 5848Author: Ben Skeggs <skeggsb@gmail.com> 5849Date: Tue Apr 21 08:57:07 2009 +1000 5850 5851 nv50: missed a WAIT_RING 5852 5853commit 523ccecee26454fc43a65e9b33d91a7d0ff8487d 5854Author: Ben Skeggs <skeggsb@gmail.com> 5855Date: Fri Apr 17 14:18:54 2009 +1000 5856 5857 nv50: use vbios-programmed lvds mode as native mode if no ddc 5858 5859 Why was this even removed? Its replacement (read VBIOS fp mode table) has 5860 proven useless in the majority of cases I've seen. 5861 5862commit fa2f1115d5f7f3e3767d423a77c6d929a4ae8d0f 5863Author: Ben Skeggs <skeggsb@gmail.com> 5864Date: Fri Apr 17 10:25:57 2009 +1000 5865 5866 Remove pNv->GART conditional for selecting CB_LOCATION in AGP/PCI 5867 5868 Kinda useless, the DRM knows better whether GART is available. 5869 5870commit 7100c06be099bacc0f8bb8898bbf7eb34ff1cc6e 5871Author: Ben Skeggs <skeggsb@gmail.com> 5872Date: Mon Apr 13 20:21:51 2009 +1000 5873 5874 kms: fix displayWidth in resize 5875 5876commit ef2de256dabfde2d619e4d451bd417973c919329 5877Author: Ben Skeggs <skeggsb@gmail.com> 5878Date: Mon Apr 13 19:09:32 2009 +1000 5879 5880 do e-edid for legacy modesetting path 5881 5882commit d8545e669aaa4362a236ba2e7231ca27ffdea1bd 5883Author: Ben Skeggs <skeggsb@gmail.com> 5884Date: Wed Apr 8 16:15:02 2009 +1000 5885 5886 bios: logic typo from earlier commit 5887 5888commit 960a5c82a80272a1577dede6a05ab7dc5db37cc3 5889Author: Ben Skeggs <skeggsb@gmail.com> 5890Date: Wed Apr 8 14:44:06 2009 +1000 5891 5892 nv50: modify ddc<->nvreg assignments again.. 5893 5894commit 3a6c7091d9706778a2e3971b7f36601d79fe7a83 5895Author: Ben Skeggs <skeggsb@gmail.com> 5896Date: Wed Apr 8 14:34:20 2009 +1000 5897 5898 kms: fix setting the randr edid property 5899 5900commit a1194b3690ced5db552d63be118a605508a1b965 5901Author: Ben Skeggs <skeggsb@gmail.com> 5902Date: Tue Apr 7 19:24:25 2009 +1000 5903 5904 bios: fix typo 5905 5906commit 11451cabbd5357dd71ab65e6155d7ca7d9766db7 5907Author: Ben Skeggs <skeggsb@gmail.com> 5908Date: Tue Apr 7 18:34:57 2009 +1000 5909 5910 bios: some G8x don't have PBUS at 0x1800, us 0x88000 unconditionally there 5911 5912commit 62d69bdf807423a644d7c2c493f94948ea11e491 5913Author: Ben Skeggs <skeggsb@gmail.com> 5914Date: Sat Apr 4 01:41:47 2009 +1000 5915 5916 bios: add exception for chipset 0x73 for lack of fp table 5917 5918commit 620d519564fdc6b3a7d55184bd481fa69035371b 5919Author: Ben Skeggs <skeggsb@gmail.com> 5920Date: Fri Apr 3 23:07:56 2009 +1000 5921 5922 consistent connector naming across <nv50, nv50 and kms 5923 5924commit a5d45c80e85611c9e22d8eca27294eef5378a549 5925Author: Stuart Bennett <stuart@freedesktop.org> 5926Date: Tue Apr 7 00:40:31 2009 +0100 5927 5928 randr12: set 1 on cr59 for off-chip digital, 0 otherwise (#21023, rh#492399) 5929 5930 The ubiqitous setting of 1 on cr59 for <nv40 tmds was breaking the 5931 integrated dvi of nv34gl. 5932 Thanks to Bill Nottingham for isolating the problem. 5933 5934commit d63c92435956b23351f1950563bbb92e24e60cea 5935Author: Stuart Bennett <stuart@freedesktop.org> 5936Date: Fri Apr 3 23:39:22 2009 +0100 5937 5938 randr12: unlock CR21 *after* turning off digital output 5939 5940 otherwise the analogue output still fails to come up. 5941 5942 That'll teach me to rearrange things just before committing 5943 5944commit 11be9a982073d66a68cd3db2bfc611eb58d3ea81 5945Author: Ben Skeggs <skeggsb@gmail.com> 5946Date: Fri Apr 3 12:22:04 2009 +1000 5947 5948 nv50: fix i2c port addresses 5949 5950commit e2aa03794420324ae5a682bbc415a79a4d830f7e 5951Author: Stuart Bennett <stuart@freedesktop.org> 5952Date: Fri Apr 3 00:57:19 2009 +0100 5953 5954 randr12: fall back to standard timings when finding native mode (rh#492819) 5955 5956commit 6965663ddcc8dd3330842d3c4c4eb4cbf4e4fc39 5957Author: Stuart Bennett <stuart@freedesktop.org> 5958Date: Wed Apr 1 05:38:29 2009 +0100 5959 5960 randr12: fix hw cursor for fully transparent pixels 5961 5962 ARGB8 pixels of 0x00000000 were getting bumped to 0x01000000, which was 5963 _just_ visible as a box around the cursor 5964 5965commit d12f70cedb0ba598234bbfc26dd6beed23e3dfe7 5966Author: Stuart Bennett <stuart@freedesktop.org> 5967Date: Sun Mar 29 00:51:05 2009 +0000 5968 5969 randr12: pre-nv17 digital fixes 5970 5971 1) set panel regs better (don't do min_front_porch adjustment) 5972 2) scaler programming - nv11 and nv20 don't need / 2 for aspect scaling 5973 5974commit ea567dbc7bc9c8d2ee5d7d9e0dcd33b03c638edb 5975Author: Stuart Bennett <stuart@freedesktop.org> 5976Date: Sat Mar 28 01:58:09 2009 +0000 5977 5978 randr12: fix restore for cards where CR21 is not left unlocked at POST 5979 5980 More nv11 inspired fun. If CR21 is left without a 0xfa mask after boot, 5981 we'd restore the old CR21 value and the subsequent restoration of VGA 5982 CRTC values wouldn't get through. 5983 5984 RE'd doc of CR21 behaviour included 5985 5986commit bd263d83139fcc93b967c71c99fd5e63e1ba6315 5987Author: Stuart Bennett <stuart@freedesktop.org> 5988Date: Fri Mar 27 18:27:18 2009 +0000 5989 5990 randr12: make nv11 dvi work for both crtcs 5991 5992commit 52c287dc64a97a750eb953ac5f725d794cbf0c64 5993Author: Stuart Bennett <stuart@freedesktop.org> 5994Date: Fri Mar 27 00:23:04 2009 +0000 5995 5996 Every card since nv5 can do panels 5997 5998 We'll only try to do ones using scripts atm though 5999 6000 Note that as we don't have a load detect function for pre-nv17 you'll 6001 likely get a 1024x768 mode set up on VGA-0 even if only the DVI output 6002 is plugged. Unclear if this is a bug or feature :) 6003 6004commit 3d371a78bc9522e9a1b2c967957a090bb527a30c 6005Author: Stuart Bennett <stuart@freedesktop.org> 6006Date: Thu Mar 26 22:21:57 2009 +0000 6007 6008 randr12: enable tmds reg access on pre-nv17 6009 6010 I think this nv11 hates me 6011 6012commit 1bb85c8020802719e82c1335a18df9367215a8ff 6013Author: Stuart Bennett <stuart@freedesktop.org> 6014Date: Wed Apr 1 04:09:09 2009 +0100 6015 6016 randr12: reorder ramdac reg access 6017 6018commit 4465fddb2467ad9d923120a2fd2c648222603771 6019Author: Stuart Bennett <stuart@freedesktop.org> 6020Date: Wed Apr 1 04:10:18 2009 +0100 6021 6022 randr12: remove nv30 special case 6023 6024 0x68?89? are just an alternative way to set scaling (set | 0x40 on 0x68.880, 6025 then horiz scaling in 0x68.898, vert in 0x68.89c) and afaict don't have any 6026 special significance on nv30 6027 6028commit 72eced5eeeef438979dff06b6c1cbaa120738cfe 6029Author: Stuart Bennett <stuart@freedesktop.org> 6030Date: Wed Apr 1 03:12:30 2009 +0100 6031 6032 randr12: fix/improve a load of version/feature tests 6033 6034 Main improvement is introduction of flag for nv17+ display architecture 6035 6036 Other version fixes include: 6037 * allow access of fp control regs on all cards (defined since nv4) 6038 * only read crtc_850 and gpio_ext on nv40+, since only written on nv40+ 6039 * set dither state on all versions (only written out on relevant cards) 6040 * PRAMDAC_DACCLK is only defined/usable on nv17+ 6041 * no point saving the lvds/tmds head for restore if there's only one head 6042 * fix a bunch of places the nforce (0x1a) case had been missed 6043 6044commit 37c6916d3eac54cc13812276fca99662162c4bbc 6045Author: Ben Skeggs <skeggsb@gmail.com> 6046Date: Wed Apr 1 15:00:03 2009 +1000 6047 6048 bios: fix chips with pll limits table version 0 6049 6050commit 76c5a05603b0eaa0831e031268e48ab6ffdee3a9 6051Author: Ben Skeggs <skeggsb@gmail.com> 6052Date: Wed Apr 1 10:36:34 2009 +1000 6053 6054 bios: xf86DrvMsg->NV_ERROR 6055 6056commit b1b9bcb111f1f5ec15e19d5a84b7ca576983c4a1 6057Author: Ben Skeggs <skeggsb@gmail.com> 6058Date: Tue Mar 31 11:43:26 2009 +1000 6059 6060 bios: support for pll limits table v3.0 6061 6062commit 8c85b4b7451bd09b0607b1f4ef8ce4b833006328 6063Author: Ben Skeggs <skeggsb@gmail.com> 6064Date: Tue Mar 31 11:11:12 2009 +1000 6065 6066 kms: CONNECTORn -> CONNECTOR-n 6067 6068commit 4e1c3239648510a4024917289127ad1e982ed34d 6069Author: Stuart Bennett <stuart@freedesktop.org> 6070Date: Tue Mar 31 20:59:26 2009 +0100 6071 6072 Restore call to NVDRIGetVersion, to ensure the dri module is loaded 6073 6074commit c973f756cfdf36db3a5c7095f6e2c57ff5841388 6075Author: Stuart Bennett <stuart@freedesktop.org> 6076Date: Tue Mar 31 17:07:26 2009 +0100 6077 6078 Abstract logging in nv_bios.c 6079 6080commit b17bebd811316b5ccca21146b58f8fbf4773f0ad 6081Author: Stuart Bennett <stuart@freedesktop.org> 6082Date: Mon Mar 30 20:52:29 2009 +0100 6083 6084 nForce DIMM check is now in DRM 6085 6086 Since it only at most suggests an optimization, there's no real need to 6087 wait for a DRM version bump 6088 6089commit 9c991d3a84ad16f893b739282403eb01e02bb57f 6090Author: Stuart Bennett <stuart@freedesktop.org> 6091Date: Mon Mar 30 03:31:01 2009 +0100 6092 6093 randr12: FP_TG_CONTROL 2: dpms improvements 6094 6095 For DPMS on digital outputs nvidia does not just OR the 0x68.848 value 6096 with 0x20000022, they mask off 0x10000011 too. 6097 To do the same thing here, don't change in-memory state of programmed reg 6098 value (hopefully nothing breaks) 6099 6100commit 5fb2888d79d39a1c74ab005d8d518686ad823224 6101Author: Stuart Bennett <stuart@freedesktop.org> 6102Date: Fri Mar 27 18:27:18 2009 +0000 6103 6104 randr12: FP_TG_CONTROL 1: turn off digital path when using VGA encoder 6105 6106 On pre-nv17, having a CRTC's digital output path turned on when 6107 programming the CRTC for VGA confuses the chip and often results in the 6108 analogue output not working. Indeed, having it turned on even just before 6109 programming the CRTC seems to break it. 6110 6111 Nvidia solves this by turning off the digital path early in modesetting 6112 and then delaying for 50ms, so (if the digital path was on to start with) 6113 that's reproduced here. 6114 6115 Although this problem has not been seen on nv17+, nvidia uses the same 6116 approach, so we may as well too. 6117 6118 Some tidying of the output_prepare function thrown in as well. 6119 6120commit 8023dc3e2dcfa8a4cc11a5a64750a7505e8aa704 6121Author: Stuart Bennett <stuart@freedesktop.org> 6122Date: Mon Mar 30 02:27:50 2009 +0100 6123 6124 Split DCB 1.5 parsing from 2.0+, get closer to 80 cols 6125 6126commit ea027b35b28c0305aca03edc8c1ff5cd84bbf3ef 6127Author: Stuart Bennett <stuart@freedesktop.org> 6128Date: Fri Mar 27 00:03:38 2009 +0000 6129 6130 The dcb prior to v1.5 is pretty hopeless, just add a crt. 6131 6132 Also, we don't use dcb 1.2 i2c for modesetting due to cards with broken 6133 tables (ala #14821), and I've got an nv11 with duff entries (v1.4), so 6134 don't raise hopes of init_dcb_i2c_entry working on these versions either 6135 6136commit 921fbccd53eae37ed86ecece6dc1fcae4b7bf3c4 6137Author: Stuart Bennett <stuart@freedesktop.org> 6138Date: Thu Mar 26 16:41:02 2009 +0000 6139 6140 randr12: avoid nv11 chip lockup when saving palette regs 6141 6142commit 9d46930f0eec7b787666671681e9a93fc33dce62 6143Author: Ben Skeggs <skeggsb@gmail.com> 6144Date: Mon Mar 30 20:11:03 2009 +1000 6145 6146 xv: oops 6147 6148commit 9213c39d3a8d9a7fc90a44f9427c1fb8218243e2 6149Author: Ben Skeggs <skeggsb@gmail.com> 6150Date: Mon Mar 30 11:14:20 2009 +1000 6151 6152 xv: post damage after we draw (rh#492239) 6153 6154commit d68a0527aa2a3e93c3408526d4bbb93db89e9b3c 6155Author: Ben Skeggs <skeggsb@gmail.com> 6156Date: Mon Mar 30 10:45:59 2009 +1000 6157 6158 kms: small cleanup 6159 6160commit c9cb6a6f7f79233bdab737fa6b6d0916dab45971 6161Author: Stuart Bennett <stuart@freedesktop.org> 6162Date: Fri Mar 27 16:43:27 2009 +0000 6163 6164 rh#492511 has a CRTC_OWNER of 0x7 for some reason 6165 6166 not that this probably affects the bug in any way 6167 6168commit 95bff61597ddf21d6415b40759258802a5f42150 6169Author: Stuart Bennett <stuart@freedesktop.org> 6170Date: Wed Mar 25 02:40:17 2009 +0000 6171 6172 randr12: fix LVDS legitimately disconnected case 6173 6174 DPMS off still needs to work, but there's no native_mode for the pxclk. 6175 Fortunately LVDS_PANEL_OFF doesn't actually need a pxclk, so just use 0 6176 6177commit f86e395231c97155c1a211010835fd02b8e4dc5f 6178Author: Stuart Bennett <stuart@freedesktop.org> 6179Date: Wed Mar 25 02:07:15 2009 +0000 6180 6181 randr12: permit vga outputs to be force enabled in xorg.conf 6182 6183 You need: 6184 Section "Monitor" 6185 Identifier "VGA-0" 6186 Option "Enable" "true" 6187 EndSection 6188 6189commit 8427b39a2f6cb32496bbbe370b793d5ba78f6b6c 6190Author: Stuart Bennett <stuart@freedesktop.org> 6191Date: Tue Mar 24 03:32:24 2009 +0000 6192 6193 randr12: move a load of hw mode programming into nouveau_hw.c 6194 6195 Pull in pll setting (common, no reason to bloat nv_bios with it), state 6196 load/unload from nv_crtc, and useful bits of nv_hw.c (remnants in nv_hw.c 6197 are only used on the non-randr12 path) 6198 6199commit 163bdce98c0f9ca5c1125e33a9a2bf2d2d9a1286 6200Author: Stuart Bennett <stuart@freedesktop.org> 6201Date: Tue Mar 24 03:19:38 2009 +0000 6202 6203 Turn the smaller mmio wrappers into static inline 6204 6205commit 21ee927f28d1ca8d3807acc767ef14783e43f057 6206Author: Stuart Bennett <stuart@freedesktop.org> 6207Date: Mon Mar 23 21:34:26 2009 +0000 6208 6209 Put arbitration and mnp calcs in separate file 6210 6211 Make bios->chip_version public so the pll calcs can use it 6212 6213commit 83dc890f0b1fc88d2fbc7559f5d4decf4c84f340 6214Author: Maarten Maathuis <madman2003@gmail.com> 6215Date: Sat Mar 28 01:25:32 2009 +0100 6216 6217 nv50: use E-EDID when available + apply edid quirks 6218 6219commit 2c130cf71587706a4f1dcd8f5db1de47a10522f5 6220Author: Pekka Paalanen <pq@iki.fi> 6221Date: Fri Mar 27 23:16:59 2009 +0200 6222 6223 nv50: fix xf86GetDefaultModes() call for older servers 6224 6225 Thanks to stillunknown for this build fix. 6226 6227commit f1907dcef8d06d7ee4ef10ba22bb7decef700110 6228Author: Ben Skeggs <skeggsb@gmail.com> 6229Date: Fri Mar 27 15:37:17 2009 +1000 6230 6231 nv50: add default modes to mode pool for LVDS panel 6232 6233 At some point X stopped doing this if the display doesn't support 6234 continuous timing. For a lot of G80 LVDS panels, they only report 6235 their native mode via DDC. 6236 6237 We'll add the default modes ourself, and use the GPU scalers. 6238 6239commit 7da6fdb8b477d2007c83b47e9cbcc2476ae40f22 6240Author: Ben Skeggs <skeggsb@gmail.com> 6241Date: Fri Mar 27 14:17:29 2009 +1000 6242 6243 kms: fix bug which prevented getting edid from the kernel 6244 6245commit f431e204d3a9be92df5ca606e86ed7c6d82103f0 6246Author: Ben Skeggs <skeggsb@gmail.com> 6247Date: Fri Mar 27 11:50:22 2009 +1000 6248 6249 randr: fix crash when rotation requested 6250 6251commit 36dedd04da1b3c89be61a95270414477d284b2ef 6252Author: Ben Skeggs <skeggsb@gmail.com> 6253Date: Fri Mar 27 11:26:09 2009 +1000 6254 6255 kms: check for mm_enabled as an additional test for kms presence 6256 6257commit a923bc1e4840c0386301f8648add2ccbfbf79a88 6258Author: Ben Skeggs <skeggsb@gmail.com> 6259Date: Thu Mar 26 16:24:01 2009 +1000 6260 6261 nv50/xv: correct rendering to partically obscured windows 6262 6263 Won't be exposed by current git, due to lack of accelerated front-buffer 6264 rendering. But should fix rh#492173, rh#492229. 6265 6266commit 01cee2906686897c4b795d308270b3b69a3e286e 6267Author: Ben Skeggs <skeggsb@gmail.com> 6268Date: Thu Mar 26 11:43:09 2009 +1000 6269 6270 nv50: call NVSync() in CloseScreen() before restoring video mode 6271 6272 Without this, if accel is still being done to the front buffer (usually at 6273 offset 0), we'll restore the video mode and then the engine will clobber 6274 the text mode fonts right afterwards leaving a messed up console. 6275 6276commit aa7c0375b51d813be54de5e19b96e958ac183f17 6277Author: Ben Skeggs <skeggsb@gmail.com> 6278Date: Thu Mar 26 06:36:17 2009 +1000 6279 6280 kms: implement AdjustFrame, should fix crash in fdo#24236 6281 6282commit 61879b8079bee1caeedc0bd9c11061423ff27b4a 6283Author: Ben Skeggs <skeggsb@gmail.com> 6284Date: Wed Mar 25 10:09:39 2009 +1000 6285 6286 kms: drm_mode_modeinfo struct changed names at some point 6287 6288commit 79306fc29f6b27fd61fe51855bad5489fc4fd986 6289Author: Ben Skeggs <skeggsb@gmail.com> 6290Date: Wed Mar 25 08:15:41 2009 +1000 6291 6292 dri: fail harder 6293 6294commit da1ba93f1dc2b02615e4f0366450872659176799 6295Author: Ben Skeggs <skeggsb@gmail.com> 6296Date: Wed Mar 25 08:12:26 2009 +1000 6297 6298 device close doesn't belong in DRI close, can have DRM without DRI 6299 6300commit 56b11c773ac9392069c6a589f4f87c5cc79e020e 6301Author: Younes Manton <younes.m@gmail.com> 6302Date: Tue Mar 24 14:27:40 2009 -0400 6303 6304 Fix mouse cursor disappearing when near top/left edge of screen. 6305 6306commit 4067ab466fe3aa817e0323959f70c7dd3494de0a 6307Author: Ben Skeggs <skeggsb@gmail.com> 6308Date: Mon Mar 23 14:43:22 2009 +1000 6309 6310 another ppc fix 6311 6312commit 862dba8b6ca6354d915f2430826a5581f41d1002 6313Author: Ben Skeggs <skeggsb@gmail.com> 6314Date: Mon Mar 23 14:22:15 2009 +1000 6315 6316 more ppc.. 6317 6318commit 30634865617737171835f6ac9437df597e5a8951 6319Author: Ben Skeggs <skeggsb@gmail.com> 6320Date: Mon Mar 23 13:57:52 2009 +1000 6321 6322 fix ppc build 6323 6324commit d80fe7878379aefd3345746a149f2b46d3a84805 6325Author: Francisco Jerez <currojerez@riseup.net> 6326Date: Mon Mar 23 03:27:16 2009 +0100 6327 6328 Fix nouveau_hw_decode_pll on NV30/35. 6329 6330commit 62aa81deca5c63d30407872c0e2b784f5084acb3 6331Author: Stuart Bennett <stuart@freedesktop.org> 6332Date: Sat Mar 21 01:43:14 2009 +0000 6333 6334 Some laptop mode finding fixes resulting from rh#487456 6335 6336 On closer inspection, it seems BMP laptops only depend on 6337 use_straps_for_mode to ban DDC. Therefore use pxclk alone to determine 6338 EDID case in parse_lvds_table, and invert sense of fp ddc flag to get a 6339 default of "false" 6340 6341commit aa6edfaf1c32432b07c298c42b68f2d93f3b337c 6342Author: Stuart Bennett <stuart@freedesktop.org> 6343Date: Sun Mar 15 15:18:16 2009 +0000 6344 6345 randr12: name VGA CRTC fields 6346 6347 Now with added consistent indentation 6348 Use fields in a couple of other cases while we're about it 6349 6350commit 7b7c44c828eda2edfad2fa9bea95f70889c59453 6351Author: Stuart Bennett <stuart@freedesktop.org> 6352Date: Sun Mar 15 03:45:54 2009 +0000 6353 6354 randr12: rename reg state members to follow reg names 6355 6356commit 4ae6c22957a929760e583d5704e350adb4404c3d 6357Author: Stuart Bennett <stuart@freedesktop.org> 6358Date: Sun Mar 15 03:11:40 2009 +0000 6359 6360 randr12: name some FP flags, move some unrelated stuff out from nvreg.h 6361 6362commit b50b49f6f54148ee64ba7500e797f432b198a324 6363Author: Stuart Bennett <stuart@freedesktop.org> 6364Date: Sun Mar 15 04:21:40 2009 +0000 6365 6366 pre-nv50: use nvidia's names for PRAMDAC regs where known 6367 6368 Mostly adding 'P' to the previous NV_RAMDAC regs -- either when known 6369 correct, or when totally unknown (eg NV_PRAMDAC_580); made-up names 6370 untouched. Lack of code changes checked with cpp. 6371 6372 Unused defines removed too. 6373 6374commit a9df3049a7e38b3bf4478459e5f0736e10927247 6375Author: Stuart Bennett <stuart@freedesktop.org> 6376Date: Sun Mar 15 04:08:56 2009 +0000 6377 6378 pre-nv50: use nvidia's names for PCRTC regs where known 6379 6380 Mostly adding 'P' to the previous NV_CTRC regs -- either when known 6381 correct, or when totally unknown (eg NV_PCRTC_850); made-up names 6382 untouched. Lack of code changes checked with cpp. 6383 6384 Unused defines removed too. 6385 6386commit 2c0710d334e3e1dcb04cc4fc0b46ec4c8cdb9ba6 6387Author: Stuart Bennett <stuart@freedesktop.org> 6388Date: Tue Mar 17 18:50:18 2009 +0000 6389 6390 Separate maximum limits for encoder devices and i2c records, DCB_ prefixes 6391 6392 From the comment with pNv->connector in nv_type.h I've assumed nv50 was 6393 using the previous define as an I2C limit, rather than encoder limit 6394 6395commit 0db27a6f1aee023782de14ae2f260fec4e60229d 6396Author: Stuart Bennett <stuart@freedesktop.org> 6397Date: Tue Mar 17 13:43:13 2009 +0000 6398 6399 Add DCB 1.5 entry from a GeForce2 Go reported by Martin Ketzer on the ML 6400 6401commit 3e7fa97a50e42c0255876c79137dc1a40a6052a6 6402Author: Ben Skeggs <skeggsb@gmail.com> 6403Date: Wed Mar 18 09:36:33 2009 +1000 6404 6405 dri: pass object handle instead of offset if using real mm 6406 6407commit 88efe405213a3ea23edd9bed971da1d2492021eb 6408Author: Ben Skeggs <skeggsb@gmail.com> 6409Date: Mon Mar 16 17:45:49 2009 +1000 6410 6411 dri2: initial implementation, only with driver pixmaps 6412 6413 Also, no DRI1 when driver pixmaps enabled. 6414 6415commit 675126e0f0c1c077bde4b25ccb64f87d1d08547b 6416Author: Ben Skeggs <skeggsb@gmail.com> 6417Date: Mon Mar 16 15:06:31 2009 +1000 6418 6419 kms: support framebuffer resize if driver pixmaps enabled 6420 6421commit 8b354b92a859cf277b19e51fc6cb109f4ea44b7a 6422Author: Ben Skeggs <skeggsb@gmail.com> 6423Date: Mon Mar 16 13:13:07 2009 +1000 6424 6425 kms: copy old fb to new fb on modeset 6426 6427 Requires driver pixmaps enabled, which suck on G8x currently. Paves the 6428 way for a nicer transition from console->X. There's still a flash where 6429 the screen goes black before X starts here, but this is on the way to what 6430 we want. 6431 6432commit 6658403cfd91556d06afecdde03feec7384c5398 6433Author: Ben Skeggs <skeggsb@gmail.com> 6434Date: Mon Mar 16 12:34:49 2009 +1000 6435 6436 Call NVEnterVT from NVScreenInit rather than duplicating it. 6437 6438commit f59fef15c94fc5c6f216d0f5dcdcb3d3188c0049 6439Author: Ben Skeggs <skeggsb@gmail.com> 6440Date: Mon Mar 16 09:17:29 2009 +1000 6441 6442 exa: fix multiple prepare/finish_access on a pixmap 6443 6444commit 75a1c4158a19bf0857de49b736fb87a40c095048 6445Author: Stuart Bennett <stuart@freedesktop.org> 6446Date: Sat Mar 14 15:48:39 2009 +0000 6447 6448 Wrap bios endian casting 6449 6450 Also convert some functions to use individual bios pointers, rather than 6451 bios->data[offset + ... 6452 Plus some minor bmp structure documentation and tidying 6453 6454commit 3666d73333ecbdd3a4941a76af2528113db427bb 6455Author: Stuart Bennett <stuart@freedesktop.org> 6456Date: Wed Feb 25 20:21:42 2009 +0000 6457 6458 Switch to using pll_vals struct for passing pll stuff around 6459 6460commit a3dbc06d4f9dc340f5c167b94d8e10ac8386c5b4 6461Author: Stuart Bennett <stuart@freedesktop.org> 6462Date: Fri Apr 11 16:51:30 2008 +0100 6463 6464 randr12: cull old unused pll code 6465 6466commit 71c821b8119cbac8dcd91b27e3b7556d00ceb546 6467Author: Stuart Bennett <stuart@freedesktop.org> 6468Date: Sat Mar 14 19:06:12 2009 +0000 6469 6470 Don't pointlessly overallocate cursor storage 6471 6472commit 7284c07a0ad6551ea99c28454f4e7308699835e6 6473Author: Stuart Bennett <stuart@freedesktop.org> 6474Date: Fri Mar 13 21:23:48 2009 +0000 6475 6476 randr12: nv10 hw cursor fixes/changes 6477 6478 Remove access to PCRTC_GPIO, it's unused and on NV10 overwrites 6479 PCRTC_CURSOR_CONFIG leading to all sorts of cursor weirdness 6480 6481 NV10 hw cursor is now set-up on the assumption it can do 64x64 (we've been 6482 using 64x32 up to now), but if it can't do cursors 64px high it can easily 6483 be fixed to behave like nv0x at 32x32 (64x32's pretty silly) 6484 6485 It's possible that with the PCRTC_GPIO thing resolved the nv10 cursor can 6486 be more featureful: the code would also be simpler if it supported 32bpp 6487 cursors (test by seeing if setting NV_CRTC_CURSOR_CONFIG_32BPP has any 6488 effect). It may even do alpha -- change pNv->alphaCursor to be 6489 (NVArch >= 0x10) to find out. 6490 6491commit bcc3a14b15cdad3647cc7df0f540df1111843be5 6492Author: Stuart Bennett <stuart@freedesktop.org> 6493Date: Fri Mar 13 02:31:35 2009 +0000 6494 6495 randr12: use nv0x cursor colour expansion funcs on later cards, unify argb paths 6496 6497commit b21c807d2ced35ac1b06300f6e68d661710b336a 6498Author: Stuart Bennett <stuart@freedesktop.org> 6499Date: Fri Mar 13 03:52:54 2009 +0000 6500 6501 randr12: store nv0x hw cursor image in VRAM 6502 6503 nv_driver.c is already doing the alloc, nvidia uses the same method, and it saves code :) 6504 6505commit 3ea4b194e83735145d5e9d5c5357f92ae6c6a569 6506Author: Stuart Bennett <stuart@freedesktop.org> 6507Date: Thu Mar 12 15:37:59 2009 +0000 6508 6509 Kill pre-randr12 hwcursor code, and tidy randr12 hwcursor code 6510 6511 If you're not using randr12 you should be. 6512 6513 Misc fixes included: 6514 Ensure PNVM bit set on PCRTC_CURSOR_CONFIG for consistency 6515 Fix NV11 alpha cursors 6516 Add missing bo unref 6517 6518commit b988160dfa5a535fd042062d7318ab99fc7efa5a 6519Author: Stuart Bennett <stuart@freedesktop.org> 6520Date: Thu Mar 12 16:31:24 2009 +0000 6521 6522 Remove calls to load ddc and i2c modules -- they're built-in since server 1.3 6523 6524commit d5893ecf3dd915fc6937c61e6eae7c221edeb6e2 6525Author: Stuart Bennett <stuart@freedesktop.org> 6526Date: Thu Mar 12 21:46:02 2009 +0000 6527 6528 Hack around missing display table in rdivacky's GeForce 7050 PV/nForce 630a 6529 6530commit 369e1efd23c0180e210dbe898318b6b9efa1faad 6531Author: Stuart Bennett <stuart@freedesktop.org> 6532Date: Thu Mar 12 16:09:27 2009 +0000 6533 6534 Fix xserver shutdown for randr12 off 6535 6536commit 8343d6b2c0ebddf1a1a7d146890ed90278559fea 6537Author: Stuart Bennett <stuart@freedesktop.org> 6538Date: Wed Mar 11 01:11:02 2009 +0000 6539 6540 Remove bios lvds parsing pxclk test; ddc_permitted is a sufficient condition 6541 6542commit 8b56c699ad39c64326d76658c741801bb4118c5f 6543Author: Stuart Bennett <stuart@freedesktop.org> 6544Date: Wed Mar 11 00:43:33 2009 +0000 6545 6546 Name some regs/values and reduce crtc reg struct size 6547 6548 TVOUT_LATENCY taken from nv11 .scp file 6549 6550 CRTC[]: even nvidia only use up to 9f, and they know what the regs do 6551 6552commit 79d23d87b489332af521fbda418a13b9ae032b7d 6553Author: Ben Skeggs <skeggsb@gmail.com> 6554Date: Fri Mar 13 11:43:54 2009 +1000 6555 6556 improve drm/dri initialisation code 6557 6558 The main motivation here was to fix the scenario where X is started without 6559 nouveau.ko loaded, checks if kms is enabled, is told no because nouveau.ko 6560 isn't loaded, and then later nouveau.ko gets loaded with kms enabled by 6561 default and starts fighting with the ddx for the display. 6562 6563 One cool side-effect is that we're not depending on DRIScreenInit for our 6564 DRM fd now, which means that if DRIScreenInit fails for a non-fatal reason 6565 (like, NVIDIA's libglx being around instead of xorg's) we can still run 6566 with full 2D acceleration. 6567 6568commit bd1448216715681beebabe82b88c8b4d0e62eced 6569Author: Ben Skeggs <skeggsb@gmail.com> 6570Date: Wed Mar 11 16:07:41 2009 +1000 6571 6572 kms: cleanup set_property 6573 6574commit 7a455929785e00beb2d1e65007a3407427ca693a 6575Author: Ben Skeggs <skeggsb@gmail.com> 6576Date: Wed Mar 11 15:51:59 2009 +1000 6577 6578 kms: get current values for output properties on startup 6579 6580commit 7095e8942c74953fcf1615900e2d7864024ece5c 6581Author: Ben Skeggs <skeggsb@gmail.com> 6582Date: Wed Mar 11 11:30:08 2009 +1000 6583 6584 kms: work around some xserver stupidity with DPMS 6585 6586 The server calls the output DPMS functions with the intention of 6587 turning off outputs before reprogramming them. It probably shouldn't 6588 be doing this in the case of the randr driver using set_mode_major, 6589 and leave it up to the driver. 6590 6591 The DRM modesetting will usually reenable the outputs after a modeset 6592 anyway, which is the behaviour that the xserver currently relies on. 6593 However, there's a shortcut used in the drm to allow far faster fb 6594 swaps without doing a modeset. When this path is hit, the DRM will 6595 rightly not touch any power states and the display will never get 6596 turned back on. 6597 6598commit 8f9a58070c2fc55c922e9d82d0796e32959a8431 6599Author: Stuart Bennett <stuart@freedesktop.org> 6600Date: Tue Mar 10 00:20:35 2009 +0000 6601 6602 Another DCB 1.5 entry (seen in rh#455194) 6603 6604 Plus a few minor PLL tidyings 6605 6606commit 9a719900adcebf47442a30fe1989509f93e13e16 6607Author: Stuart Bennett <stuart@freedesktop.org> 6608Date: Mon Mar 9 15:57:06 2009 +0000 6609 6610 randr12: add output properties for digital vibrance and image sharpening 6611 6612commit 678ddc2e56aea3deb1e95b2309b47e1f2cce0635 6613Author: Stuart Bennett <stuart@freedesktop.org> 6614Date: Mon Mar 9 15:34:03 2009 +0000 6615 6616 randr12: functionate output properties creation 6617 6618 notes: 6619 * setting pending at change time is necessary for props created as pending, 6620 so that PostPending doesn't overwrite the correct current state with the 6621 incorrect pending one 6622 * our call to PostPending can go; relic from when props were updated in 6623 update_output_fields 6624 6625commit ed9bd8848fcfe6a77894bfe520a5aa25a9b3f769 6626Author: Ben Skeggs <skeggsb@gmail.com> 6627Date: Wed Mar 4 15:10:24 2009 +1000 6628 6629 kms: remove early-exit from gamma_set, the kernel side works now 6630 6631commit 98c9e4edb58374f18249e5c9c53b392fb8b4a1d1 6632Author: Ben Skeggs <skeggsb@gmail.com> 6633Date: Wed Mar 4 13:40:49 2009 +1000 6634 6635 kms: work around some bong hits with dpms 6636 6637 This'll need to get fixed elsewhere, but in the meantime, we have working 6638 DPMS *and* can actually see what's happening when X starts! 6639 6640commit 88ae2c92c0efc1919dbaa7e8db4c37b86987046c 6641Author: Ben Skeggs <skeggsb@gmail.com> 6642Date: Wed Mar 4 13:36:36 2009 +1000 6643 6644 kms: implement dpms 6645 6646commit 5269cc8060e0ce5160dfca408b0807a4cf2ff17c 6647Author: Stuart Bennett <stuart@freedesktop.org> 6648Date: Sun Mar 8 20:07:23 2009 +0000 6649 6650 randr12: make props code slightly less awful to behold 6651 6652commit 0b87c4964076101a2d8373fd3addd9355fb7765e 6653Author: Stuart Bennett <stuart@freedesktop.org> 6654Date: Sat Mar 7 21:13:55 2009 +0000 6655 6656 More robust PLL upclocking for old cards 6657 6658commit cb237a34f31f8b8c841974cb64141a2f2e36c53b 6659Author: Stuart Bennett <stuart@freedesktop.org> 6660Date: Sat Mar 7 04:38:12 2009 +0000 6661 6662 bios/randr12: mode validation fixes (includes fix for #20298) 6663 6664 Remove arbitrary 12000kHz mode clock lower bound (#20298) 6665 Use DCB supplied max dac clock (and remove a memcpy) 6666 6667commit 9cfccd7281a715e45a4296226d9d52854230f346 6668Author: Stuart Bennett <stuart@freedesktop.org> 6669Date: Sat Mar 7 04:34:20 2009 +0000 6670 6671 Handle failure to calculate pll values better 6672 6673 Ensure pll limit table get parsed on nv30 and nv35 too 6674 6675commit 1e994400913bc656b34440df67aa105aa2b211c4 6676Author: Stuart Bennett <sb476@cam.ac.uk> 6677Date: Fri Jan 16 22:26:28 2009 +0000 6678 6679 xv: return BadAlloc if destination pixmap is not put into EXA offscreen (#15792) 6680 6681 Lets the player get X errors, instead of the card choking trying to do system 6682 memory addresses 6683 6684 We could use exaDrawableIsOffscreen() here, but that's only in new xservers. 6685 6686commit 77347da64f86822f128e0dae3e89a6b80282c4ba 6687Author: Stuart Bennett <stuart@freedesktop.org> 6688Date: Fri Feb 27 01:22:13 2009 +0000 6689 6690 Remove duplicated DCB parsing 6691 6692 Store stuff in DCB private instead 6693 6694commit 1b6cabeee4c0160335c174bd4e9111dbd9535aaa 6695Author: Stuart Bennett <stuart@freedesktop.org> 6696Date: Tue Mar 3 01:03:45 2009 +0000 6697 6698 Remove VBIOS access in parse_dcb_entry 6699 6700commit 72fd0ae537e4db208bbfa142dd70144901d7f267 6701Author: Stuart Bennett <stuart@freedesktop.org> 6702Date: Tue Mar 3 21:51:46 2009 +0000 6703 6704 randr12: no need to re-detect edid for lvds, it's not going away 6705 6706 Reuse of nv_connector->edid is safe as nv_lvds_output_get_modes doesn't call 6707 nv_output_get_edid_modes after a mode has been found once 6708 6709commit ae981a8075395af40cf530f2aa24ac68d2ca0e77 6710Author: Maarten Maathuis <madman2003@gmail.com> 6711Date: Thu Mar 5 20:21:39 2009 +0100 6712 6713 nv50: some minor modesetting changes 6714 6715commit 42f99e68940a526f804e2046676c1d06d1bbbb42 6716Author: Ben Skeggs <skeggsb@gmail.com> 6717Date: Thu Mar 5 12:11:03 2009 +1100 6718 6719 nv30/xv: restore original viewport/clipping after putimage 6720 6721 Fixes two cases of corruption I seen while using the texture adaptors with 6722 a composite manager. 6723 - desktop getting clipped to video size sometimes 6724 - missing fonts on some menu items after xv+compmgr 6725 6726commit 56bb8f8e81e9d152217df7863e59781009a9bb77 6727Author: Stuart Bennett <stuart@freedesktop.org> 6728Date: Wed Mar 4 19:48:29 2009 +0000 6729 6730 Allow parse_fp_mode_table for x86 cards that find themselves on ppc 6731 6732commit 1c2f33ed33791f391d059778b1490222e8cd1e1d 6733Author: Stuart Bennett <stuart@freedesktop.org> 6734Date: Wed Mar 4 19:40:37 2009 +0000 6735 6736 Don't rely on the bios mobile feature bit for BMP biosen (#19986) 6737 6738 Quadro bios builds also set the mobile bit (not on nv10gl, but for all 6739 observed subsequent Quadros before nv4x (intro of BIT bios structure). 6740 So for BMP cards, use bit 6 of cr4b on head A to detect mobile biosen (all 6741 sampled mobile bios images set this in their init scripts, and non-mobile 6742 don't) 6743 6744commit e436c7ee98c2eda9d5ba45f7d897d68497b52799 6745Author: Stuart Bennett <stuart@freedesktop.org> 6746Date: Wed Mar 4 20:59:42 2009 +0000 6747 6748 randr12: make treatment of CR4B more closely resemble blob's behaviour 6749 6750 Possibly totally unnecessary anyway 6751 6752commit c61102b326455f6c63a55c8c73ed629994281588 6753Author: Stuart Bennett <stuart@freedesktop.org> 6754Date: Wed Feb 18 01:17:29 2009 +0000 6755 6756 Parse bios fp mode during modesetting, removing extra alloc 6757 6758 Also, it seems that the mode parsing was wrong on big endian for years, so 6759 fix that (though Apple laptops have DDC panels anyway) 6760 6761commit 631e15a03b8a790d342aefe677f8c5eb33935de6 6762Author: Stuart Bennett <stuart@freedesktop.org> 6763Date: Tue Mar 3 22:03:00 2009 +0000 6764 6765 Get dual_link and dithering bios info through lvds table parsing function 6766 6767 Now that lvds table is only parsed during modesetting, return the parsed 6768 data in out parameters to make the ordering requirements obvious 6769 6770 Since calling this on nv50 is pointless, that code was commented out 6771 6772commit 2d4377193f5e2166af3df32ddcfb1235be82764e 6773Author: Stuart Bennett <stuart@freedesktop.org> 6774Date: Tue Mar 3 21:52:04 2009 +0000 6775 6776 Call parse_lvds_manufacturer_table during modesetting only (#19986 partial fix) 6777 6778 Also remove SetOwner call now that locking crtcs doesn't change the set head 6779 6780 Moving parse_lvds_manufacturer_table to modesetting means it only gets 6781 called for lvds outputs, removing the dependency on checking for the mobile 6782 bios build. Another benefit is making the strap and EDID FP paths common. 6783 6784 A side effect is that as the lvds dithering toggle depends on the results of 6785 parse_lvds_manufacturer_table, setting that moves to after calling the 6786 parsing function 6787 6788commit ce8f5d71dacbe0d56cfcb17dfe60f1c4c6cf1773 6789Author: Stuart Bennett <stuart@freedesktop.org> 6790Date: Tue Mar 3 17:24:09 2009 +0000 6791 6792 Variety of LVDS script changes 6793 6794 Remove lvds invocation static 6795 No LVDS_INIT for BIT biosen 6796 Merge in one-time call to LVDS_INIT script to main script function 6797 Make sel_clk preservation stuff use regular reg writing functions 6798 6799commit 071d5373df744c0c9578134fa634bce1e202d4f3 6800Author: Stuart Bennett <stuart@freedesktop.org> 6801Date: Tue Feb 17 16:45:50 2009 +0000 6802 6803 Read embedded bios edid only on demand 6804 6805commit f04689cb326da409448d54694e2660cf4b4b7950 6806Author: Ben Skeggs <skeggsb@gmail.com> 6807Date: Tue Mar 3 10:14:26 2009 +1000 6808 6809 move memset of fb after NVSave(), hopefully fix lost VGA fonts 6810 6811commit 6961efb84dd6894aaaf010a9f620231dfcbad004 6812Author: Ben Skeggs <skeggsb@gmail.com> 6813Date: Mon Mar 2 11:35:05 2009 +1000 6814 6815 kms: fix check for xf86drmMode.h 6816 6817commit 76fa6569ffd1a7f61e295753383bd01b246b5348 6818Author: Stuart Bennett <stuart@freedesktop.org> 6819Date: Mon Mar 2 16:43:06 2009 +0000 6820 6821 Detypedef bios_t, rename bios io functions, group various helper functions 6822 6823commit 39c9d0231c0d24bf91dbe9ab262f54fcad1b3b2f 6824Author: Stuart Bennett <stuart@freedesktop.org> 6825Date: Fri Feb 27 18:57:35 2009 +0000 6826 6827 Move parsed DCB to bios structs, split into public and private parts 6828 6829commit 275daaf7be46e54ca59067d3f2a6e6790e1ce7e0 6830Author: Stuart Bennett <stuart@freedesktop.org> 6831Date: Fri Feb 27 17:46:26 2009 +0000 6832 6833 Split bios struct into struct for nv_bios.c, and exported public info struct 6834 6835commit 8b3e966e3230add04aa05a9ded3f8a647afe0d91 6836Author: Stuart Bennett <stuart@freedesktop.org> 6837Date: Sun Mar 1 20:50:15 2009 +0000 6838 6839 Remove never updated bios opcodes 0x31, 0x5A and 0x73 6840 6841 Notices of removal left in case any bios using them ever appears, so they 6842 can be found in git history 6843 6844 Also nuke a few unused reg access defines 6845 6846commit ce9eb05f516899f56e66cc290fc7c4f9d3085882 6847Author: Stuart Bennett <stuart@freedesktop.org> 6848Date: Fri Feb 27 23:14:24 2009 +0000 6849 6850 Trivial bios convenience variable changes 6851 6852commit cfeef1144943e72d62a2c0d9701ddf4a639deeca 6853Author: Maarten Maathuis <madman2003@gmail.com> 6854Date: Fri Feb 27 17:30:40 2009 +0100 6855 6856 exa: forgot to check if EXA_SUPPORTS_PREPARE_AUX was defined 6857 6858commit 37a85dcbd6f6134a43b5c9bb883553fdaf7c0bd9 6859Author: Maarten Maathuis <madman2003@gmail.com> 6860Date: Fri Feb 27 17:23:42 2009 +0100 6861 6862 exa: we support PREPARE_AUX 6863 6864commit 1249fd25a529702a4c6fb4831ef92e801f0b6122 6865Author: Maarten Maathuis <madman2003@gmail.com> 6866Date: Fri Feb 27 17:21:47 2009 +0100 6867 6868 Revert "exa: support major version 3" 6869 6870 This reverts commit b7f99aec7282ca7c81d19ff3ad7d9ff22275d476. 6871 6872commit 1c4a284a80ebed9f9d1e01c47b929481801566b5 6873Author: Stuart Bennett <stuart@freedesktop.org> 6874Date: Fri Feb 20 20:02:38 2009 +0000 6875 6876 Rework PLL decoding in terms of pll_vals, save/restore pll_vals on vt switch 6877 6878 If your nv04-nv4x vt after running nouveau is more broken than it was with 6879 the previous commit, file a bug 6880 6881commit 81b4732162915ab1ac0593e0ef21b56b089dafae 6882Author: Stuart Bennett <stuart@freedesktop.org> 6883Date: Fri Feb 20 19:11:41 2009 +0000 6884 6885 New common struct for pll values 6886 6887 Makes the compiler do our dirty accessing work for us 6888 6889commit adf2e350341b983d27123550041c3b1ddaac4526 6890Author: Stuart Bennett <stuart@freedesktop.org> 6891Date: Wed Feb 25 02:11:27 2009 +0000 6892 6893 Check whether NV_RAMDAC_580 has changed when setting PLLs 6894 6895commit b7f99aec7282ca7c81d19ff3ad7d9ff22275d476 6896Author: Maarten Maathuis <madman2003@gmail.com> 6897Date: Tue Feb 24 19:13:23 2009 +0100 6898 6899 exa: support major version 3 6900 6901commit d91fc78859b11d00c822e5cae70f0fc511b26701 6902Author: Ben Skeggs <skeggsb@gmail.com> 6903Date: Tue Feb 24 08:31:40 2009 +1000 6904 6905 kms: fix resize func 6906 6907 Don't change virtualX/virtualY when we're not resizing the actual framebuffer, 6908 just change displayWidth to account for the adjusted scanout size. 6909 6910 With driver pixmaps, we'll be able to do "proper" framebuffer resize. 6911 6912commit 535498bcfe79858e08ae1f01125025f56bcd504f 6913Author: Ben Skeggs <skeggsb@gmail.com> 6914Date: Mon Feb 23 08:20:34 2009 +1000 6915 6916 clear the scanout buffer on statup to avoid seeing garbage on-screen 6917 6918commit 022767373474a0dacf50ee6fcd6401db15018aca 6919Author: Ben Skeggs <skeggsb@gmail.com> 6920Date: Fri Feb 20 13:59:58 2009 +1000 6921 6922 nv50/exa: properly align offset for mem_tile ioctl 6923 6924 I really really hate this ioctl().. 6925 6926commit e09f50e5f9126c682289b7ba790f16b93a80b876 6927Author: Stuart Bennett <stuart@freedesktop.org> 6928Date: Wed Feb 18 21:04:00 2009 +0000 6929 6930 bios/randr12: fix dual link TMDS (#20006) 6931 6932 TMDS04 needs setting, just as for LVDS, and bit 31 of fp_control should also 6933 be set (although in this instance the scripts ensure it is set) 6934 Furthermore, stop dpms clobbering any changes made to fp_control by scripts 6935 6936 Many thanks to bmaass for testing 6937 6938commit b968db77037790dd5be91d6fafa0552eb7e9520f 6939Author: Stuart Bennett <stuart@freedesktop.org> 6940Date: Mon Feb 16 21:38:31 2009 +0000 6941 6942 Return of "bios: fix create_i2c_device for g80" 6943 6944commit 0a439ba05aa192e7433d430acd91b60719d20caf 6945Author: Stuart Bennett <stuart@freedesktop.org> 6946Date: Wed Feb 11 20:52:43 2009 +0000 6947 6948 Parse DCB before running init 6949 6950commit 870b6f0d2df3e64af5ac921594b1742f82329c49 6951Author: Stuart Bennett <sb476@cam.ac.uk> 6952Date: Tue Jan 20 05:29:38 2009 +0000 6953 6954 Split parsing and init into separate functions 6955 6956commit 0280a0c975830b5c4b2edf4fff8a79fb7d67b9df 6957Author: Stuart Bennett <stuart@freedesktop.org> 6958Date: Wed Feb 11 18:53:56 2009 +0000 6959 6960 Attempt to remove pNv and bios_t use in parsing the DCB 6961 6962 It's meant to parse a section of rom image into useful things, so use of pNv 6963 ought to be unusual. 6964 Access to things contained in pNv and bios_t makes it harder to prove parsing 6965 ordering requirements 6966 6967commit 9179c6667cf859248a525c8d01922b69ed45c17b 6968Author: Stuart Bennett <stuart@freedesktop.org> 6969Date: Wed Feb 11 18:33:47 2009 +0000 6970 6971 Variety of DCB enhancements 6972 6973 * ensure parsing of dcb_table.i2c_default_indices happens 6974 * only merge entries on dcb versions < 2.1, and note consequnces of merging 6975 * correct dcbent->index on post-merge compacted entries 6976 * drop pointless memcpy 6977 * return an error if no outputs found 6978 * define dcb structure in nouveau_bios.h (ought to rename pNv->dcb_table to pNv->dcb one day) 6979 6980commit 317ca477cdbb936f29360984bcee25a0ecda3aa3 6981Author: Stuart Bennett <stuart@freedesktop.org> 6982Date: Mon Feb 16 21:15:51 2009 +0000 6983 6984 Revert "bios: fix create_i2c_device for g80" 6985 6986 This reverts commit ca72ca24784b9da09b6e3e0b39f3aa32b8f7b502. 6987 It'll be back in a minute. 6988 6989commit 7b25a3019beba6342505fa16327a5bcd43669f90 6990Author: Maarten Maathuis <madman2003@gmail.com> 6991Date: Mon Feb 16 21:23:56 2009 +0100 6992 6993 nv04: unbreak copy 6994 6995commit 60c8bb1391a51188a4981b4936c190c149b79b7f 6996Author: Ben Skeggs <skeggsb@gmail.com> 6997Date: Mon Feb 16 13:09:39 2009 +1000 6998 6999 default to autodetecting whether to enable the driver's kms paths 7000 7001commit a31b1d0118dc7b4d07a2c232fcd5aafbf94f69b9 7002Author: Ben Skeggs <skeggsb@gmail.com> 7003Date: Mon Feb 16 09:49:39 2009 +1000 7004 7005 exa: reimplement driver-controlled pixmaps 7006 7007commit d97e993072dc773ecfeb80b5e138b5e4598a1c3d 7008Author: Ben Skeggs <skeggsb@gmail.com> 7009Date: Mon Feb 16 08:50:02 2009 +1000 7010 7011 exa: preparation for reintroducing driver-controlled pixmaps 7012 7013commit 1e00c7d1e6edb7f6b4e6d12ef227d27413948f8f 7014Author: Maarten Maathuis <madman2003@gmail.com> 7015Date: Sun Feb 15 18:52:59 2009 +0100 7016 7017 shadowfb: clip copied regions to frontbuffer size. 7018 7019 - There have been reports of crashed when moving windows outside the visible screen. 7020 7021commit efe27964fa25f71dcbdc7a751210860369ea0e11 7022Author: Maarten Maathuis <madman2003@gmail.com> 7023Date: Sun Feb 15 16:54:18 2009 +0100 7024 7025 nv50: correct a poorly written comment 7026 7027commit 2573c065423af47e14263479c66f2308259f29b8 7028Author: Ben Skeggs <skeggsb@gmail.com> 7029Date: Fri Feb 13 10:46:13 2009 +1000 7030 7031 exa: remove exaWaitSync() call in nouveau_exa_pixmap_map() 7032 7033 unnecessary, nouveau_bo_map() will sync if required. 7034 7035commit 500053517cb5fe058468c8965f43e6c3209ba7b5 7036Author: Ben Skeggs <skeggsb@gmail.com> 7037Date: Thu Feb 12 09:04:43 2009 +1000 7038 7039 exa: don't need exaMarkSync on DFS, the operation will be done already 7040 7041commit a7e7c557dbf9b2fe54db401793608128de08bdf5 7042Author: Ben Skeggs <skeggsb@gmail.com> 7043Date: Thu Feb 12 09:01:49 2009 +1000 7044 7045 xv: remove notifier usage 7046 7047commit 8ecb8d5331bdb910beea416cdb0b662c624c5c96 7048Author: Ben Skeggs <skeggsb@gmail.com> 7049Date: Thu Feb 12 09:00:07 2009 +1000 7050 7051 exa: remove notifier waits in UTS/DFS 7052 7053 nouveau_bo will handle CPU/GPU sync if necessary when bo_map() is called. 7054 7055commit cfb884b816eebbdaa2103dd233c6f975834e536d 7056Author: Maarten Maathuis <madman2003@gmail.com> 7057Date: Fri Feb 13 22:29:33 2009 +0100 7058 7059 bios: add some more valid regs 7060 7061commit ca72ca24784b9da09b6e3e0b39f3aa32b8f7b502 7062Author: Maarten Maathuis <madman2003@gmail.com> 7063Date: Fri Feb 13 22:18:18 2009 +0100 7064 7065 bios: fix create_i2c_device for g80 7066 7067commit bbd7041791eacd357afa775236d4123dd98ab9c9 7068Author: Maarten Maathuis <madman2003@gmail.com> 7069Date: Fri Feb 13 20:30:24 2009 +0100 7070 7071 bios: implement opcode 0x76 (INIT_IO_CONDITION). 7072 7073commit 5dd2cbe454ee8560c814f3fd96d7c28e0fec1dc1 7074Author: Maarten Maathuis <madman2003@gmail.com> 7075Date: Thu Feb 12 17:13:42 2009 +0100 7076 7077 exa: add nouveau_exa_pixmap_is_tiled 7078 7079commit 15ccf721791fa4f61784ad00449630c6d2496328 7080Author: Stuart Bennett <stuart@freedesktop.org> 7081Date: Sat Feb 7 16:40:46 2009 +0000 7082 7083 randr12: anti-crash measures for #19854 7084 7085commit 17e9d4bb8c46d67e201b2a369eff7db805dff07b 7086Author: Stuart Bennett <stuart@freedesktop.org> 7087Date: Fri Apr 4 19:01:20 2008 +0100 7088 7089 randr12: use bios PLL setting routines for modesetting 7090 7091 Only used for setting modes currently, old code remains for restore 7092 7093commit e444e0e348b3aab59671ffdeb605f9a57c46bf70 7094Author: Stuart Bennett <stuart@freedesktop.org> 7095Date: Tue Jan 20 03:23:40 2009 +0000 7096 7097 Readability improvements for nv_hw.c and nvreg.h 7098 7099commit f5ea66c2ef2c726b6a0aff72843c379ab25dbaff 7100Author: Stuart Bennett <stuart@freedesktop.org> 7101Date: Tue Feb 10 19:01:43 2009 +0000 7102 7103 Pull in extras from CRTC PLL code to bios routines 7104 7105 * g7x magic bits 7106 * 0xc040 manipulation for vplls 7107 7108commit 416636509152cf12c801ac537cb8c1bd20c19b5e 7109Author: Stuart Bennett <stuart@freedesktop.org> 7110Date: Tue Jan 20 02:11:08 2009 +0000 7111 7112 Pull out ramdac580 setting and powerctrl_1 shift code to separate functions 7113 7114 And only twiddle powerctrl_1 for vplls on nv3x single stage pll chips 7115 7116commit 8b7041833c11f32feb7fa5edc21648cc9db1a333 7117Author: Stuart Bennett <stuart@freedesktop.org> 7118Date: Tue Feb 10 19:01:05 2009 +0000 7119 7120 Miscellaneous fixes to bios PLL code 7121 7122 * bios variable was only used for chip_version. simplify. 7123 * set plls faster on single pll cards after nv15/16 7124 * allow early return for single stage plls when already set 7125 * single stage pll mode is only for nv41+ 7126 * 80 cols fixes 7127 7128commit 01bb4c1576a45217953632c9f5612aea9173f95f 7129Author: Stuart Bennett <stuart@freedesktop.org> 7130Date: Mon Feb 9 21:14:43 2009 +0000 7131 7132 Make PROM score badly, out of available bios images, if checksum bad 7133 7134 Problem exposed on a Mac following 0387ac32 7135 7136 Having PROM first is good, if the image is fine. But if the read-only PROM 7137 has problems, not a good idea to use it if there are other options available 7138 7139 Also avoid result contamination, if a read method has returned early and not 7140 overwritten previous image in *data 7141 7142commit 30848dd378d05917c646a24df404846e6f9b2352 7143Author: Stuart Bennett <sb476@cam.ac.uk> 7144Date: Tue Jan 6 16:23:33 2009 +0000 7145 7146 randr12: bodge-in continuous timings when the scaler is in use (#19203) 7147 7148 Similar to 3354e660b0744976871683ce226f17e873f26b50 in the intel driver, except 7149 I can't see much benefit in allowing bigger-than-native resolutions, so no sync 7150 range hacking 7151 7152 A pity the xserver has no idea of scalers and so each driver ends up with a hack 7153 7154commit 0387ac32bef6c714f54917d5d36071ae1224458c 7155Author: Stuart Bennett <stuart@freedesktop.org> 7156Date: Sun Feb 8 09:19:53 2009 +0000 7157 7158 Allow reading of PROM and PCI rom on PPC (for people using x86 cards) 7159 7160commit 6ad66c0b6165822dde167fd1700bd14727ce7087 7161Author: Stuart Bennett <sb476@cam.ac.uk> 7162Date: Sun Jan 25 20:09:36 2009 +0000 7163 7164 bios/randr12: be strict about when to use modes from bios vs modes from ddc 7165 7166 Again, the Inspiron 8100 in #14567 is an example of wackiness, having a 7167 non-0xf i2c_index, but using straps 7168 7169 If the previous commit didn't break your DCB 2.0 EDID equipped laptop, this will 7170 7171commit b5391eb909f8b477c3c438f1136d788acfe6bfd8 7172Author: Stuart Bennett <sb476@cam.ac.uk> 7173Date: Sun Jan 25 03:03:29 2009 +0000 7174 7175 Simplify FP strapping semantics and trust bios parsing of use_straps_for_mode 7176 7177 In many ways a spiritual revert of 11de8d 7178 7179 The compound fp.strapping member faithfully implemented what nvidia sticks in 7180 cr58 (for cr57 == 0xf), but this was never directly useful and had differing 7181 overloaded interpretations with bios version, causing confusion. Return to 7182 just using the result of get_fp_strap (g80 code of which unused and so dropped) 7183 7184 As noted in the commit, there have been DCB 2.0 laptops (as in bug 14567) which 7185 don't set use_straps_for_mode when they should, so we set it for all DCB 1.4-2.0 7186 cards. If EDID-ful DCB 1.4-2.0 laptops exist they will be broken, and will need 7187 fixing later (simplistic fix: use_straps_for_mode = (get_fp_strap() == 0xf) ) 7188 7189 Will also break nv40 straps-based laptops whose tables used reserved values, 7190 but that's pretty broken anyway, and I've no evidence they exist 7191 7192commit 34c2b1539bb2d69dd3bd5911c4fac62e46385d20 7193Author: Stuart Bennett <sb476@cam.ac.uk> 7194Date: Sat Jan 24 18:41:31 2009 +0000 7195 7196 bios/randr12: misc (mainly lvds related) clarifications and cleanups 7197 7198 get_fp_strap() moved (& unused g80 bodge removed) in anticipation of next commit 7199 7200commit 98b8cada6c355d437925a92ef0413e96751ed567 7201Author: Maarten Maathuis <madman2003@gmail.com> 7202Date: Sat Feb 7 21:02:31 2009 +0100 7203 7204 Always allocate 2 hw cursors. 7205 7206 - This hopefully fixes the legacy codepath again. 7207 7208commit f1099810bb3cfe451877667a0224eb3d664e442f 7209Author: Maarten Maathuis <madman2003@gmail.com> 7210Date: Sat Feb 7 20:04:54 2009 +0100 7211 7212 Fix a few warnings. 7213 7214commit 691589d8c05b8b0354e359b4942711f6fffa9903 7215Author: Maarten Maathuis <madman2003@gmail.com> 7216Date: Sat Feb 7 19:59:28 2009 +0100 7217 7218 Fix some cursor fallout, mostly in legacy code. 7219 7220commit 945f0cb644b6e4f34cc04927177fe53529aeb5cf 7221Author: Ben Skeggs <skeggsb@gmail.com> 7222Date: Fri Feb 6 00:44:14 2009 +1000 7223 7224 kms: update drmmode_display 7225 7226commit ba1f897159d910a25560796d11c9d0357869e1c0 7227Author: Ben Skeggs <skeggsb@gmail.com> 7228Date: Thu Feb 5 23:15:24 2009 +1000 7229 7230 keep resources around after leavevt, far too problematic currently 7231 7232commit 4dfd0b1b21415c48c4b6e0e8d335f3fc70922d27 7233Author: Ben Skeggs <skeggsb@gmail.com> 7234Date: Wed Feb 4 15:35:53 2009 +1000 7235 7236 oops 7237 7238commit e02813aab0376271ee777a2b8dafc9381f96311a 7239Author: Ben Skeggs <skeggsb@gmail.com> 7240Date: Wed Feb 4 14:50:35 2009 +1000 7241 7242 link against libdrm_nouveau (installed with libdrm) 7243 7244commit 2c06308a6c5878fca3ff4f2f9c969705d966607f 7245Author: Ben Skeggs <skeggsb@gmail.com> 7246Date: Wed Feb 4 14:09:21 2009 +1000 7247 7248 require drm 0.0.12 7249 7250commit 29aa1341fb28578968953f7e7dc3de0ec66f70f5 7251Author: Ben Skeggs <skeggsb@gmail.com> 7252Date: Mon Feb 2 11:13:55 2009 +1000 7253 7254 fix NoAccel from when map/unmap became stricter 7255 7256commit 62adc557560fa82c2086e32cd16bc51f67a28b48 7257Author: Ben Skeggs <skeggsb@gmail.com> 7258Date: Tue Feb 3 08:10:45 2009 +1000 7259 7260 Revert "Improve the NoAccel situation, but it's not working yet." 7261 7262 This reverts commit 285027a180b676ac63f4c7e158e94ffbe22bf11d. 7263 7264 Sorry, fixed this yesterday and forgot to push the commit. 7265 7266commit 285027a180b676ac63f4c7e158e94ffbe22bf11d 7267Author: Maarten Maathuis <madman2003@gmail.com> 7268Date: Mon Feb 2 11:15:02 2009 +0100 7269 7270 Improve the NoAccel situation, but it's not working yet. 7271 7272 - nouveau_bo related functions will cause problems for fake bo's (since they don't have a bo_priv). 7273 - NoAccel still doesn't work for me (the scanout contains random crap). 7274 7275commit 5e63c7866ac93f39fde49af185387d2094110627 7276Author: Ben Skeggs <skeggsb@gmail.com> 7277Date: Mon Feb 2 13:46:01 2009 +1000 7278 7279 Revert "nv50: make entire offscreen area tiled, use extra blits to scanout buffer" 7280 7281 This reverts commit 6c3f8da1bff3c94353748321b40a523add824327. 7282 7283 EXA doesn't use UTS/DFS hooks when it's accessing the "front buffer" for 7284 software fallbacks, so sw rendering to it was garbled because it doesn't 7285 understand tiling. 7286 7287 The Prepare/FinishAccess hooks could be used to point the pixmap at the 7288 linear scanout buffer instead for sw access, but on VT switch for some 7289 reason exa get called to try and do things, hit the PrepareAccess() hook 7290 and die horribly because in LeaveVT() the driver deallocates the buffers. 7291 7292 The other alternative was to use libwfb to wrap the fb accesses and 7293 handle tiling, but exa isn't quite ready for that yet either - it's 7294 being worked on. 7295 7296 Another point made about this patch was that we should figure out the 7297 display engine and get it to scanout the tiled buffer directly, not 7298 commit shoddy workarounds. Even for this "correct" solution, the two 7299 above-mentioned issues still apply. 7300 7301commit 6c3f8da1bff3c94353748321b40a523add824327 7302Author: Ben Skeggs <skeggsb@gmail.com> 7303Date: Mon Feb 2 08:52:18 2009 +1000 7304 7305 nv50: make entire offscreen area tiled, use extra blits to scanout buffer 7306 7307 This fixes some pretty bad performance issues with window resize and desktop 7308 switching where ARGB windows are concerned. It also allows Xv to be used 7309 without a composite manager. 7310 7311 The drawbacks are the use of a little more memory for the shadow scanout 7312 buffer, and a bit more GPU time is used for the extra blits. But it's worth 7313 it. 7314 7315 If you want to get rid of this, figure out how to scanout tiled buffers :) 7316 7317commit 61791028ee9321748b635a85d4e53ed80d143df5 7318Author: Ben Skeggs <skeggsb@gmail.com> 7319Date: Sun Feb 1 18:58:27 2009 +1000 7320 7321 nv10: use the correct 3d object on NV1A 7322 7323commit dfd87ece014402339f82ecc26bc27b2e47e213df 7324Author: Younes Manton <younes.m@gmail.com> 7325Date: Sat Jan 31 16:52:46 2009 -0500 7326 7327 xv: Unmap filter table BO after init so it can be validated (nv30,40) 7328 7329commit b8de749bbaeacbbe677df1939f7fd39d9d3863a7 7330Author: Stuart Bennett <sb476@cam.ac.uk> 7331Date: Sat Jan 24 05:40:02 2009 +0000 7332 7333 Avoid any risk of parsing oscillation on if_is_24bit 7334 7335 Would depend on having a BMP laptop with EDID equipped panel and a bios that 7336 doesn't match current suspicions, but better safe 7337 7338commit 0a0bb720da63e4a323978e4a52a4d6e1e4f4ae27 7339Author: Stuart Bennett <sb476@cam.ac.uk> 7340Date: Wed Jan 21 00:33:55 2009 +0000 7341 7342 Split up parsing the bit structure, complain when necessary tables are missing 7343 7344 Added bios major version 0x60 (g80+) check since it now complains about missing 7345 the BIT "A" table on nv4x roms 7346 Also permits on-demand parsing of individual tables if needed in the future 7347 7348commit 2d86018d52c387d4f2370b7c1ff83798ad6e1cf7 7349Author: Stuart Bennett <sb476@cam.ac.uk> 7350Date: Sun Jan 25 20:12:41 2009 +0000 7351 7352 Swap the order of calling parse_fp_mode_table and parse_lvds_manufacturer_table 7353 7354 Last seen in this order pre 11de8d7a 7355 7356commit 8c6b3a800ae2df35b6475a46aebab32f0577261a 7357Author: Stuart Bennett <sb476@cam.ac.uk> 7358Date: Tue Jan 20 04:04:43 2009 +0000 7359 7360 Make calling of init bios functions common 7361 7362 Bios structure parsing stage should no longer require hw access 7363 This involves serializing BIT hw-accessing functions after BIT structure 7364 parsing, but order is maintained so should cause no issues 7365 7366 Also, set a digital_min_front_porch default on BIT PPC (G5s), as they have no BIT 7367 "display" table (might even fix some wonky display issues like #17571) 7368 7369commit 9b32034eb72e580d4cb5a5e2fa22a76eff0e7f26 7370Author: Stuart Bennett <sb476@cam.ac.uk> 7371Date: Thu Jan 29 04:07:04 2009 +0000 7372 7373 Don't read obviously useless bios images, and drop 5x reading on PROM 7374 7375 If some nv43 actually need the 5x bodge (no pramin image) we can add it back as 7376 a method of last resort 7377 PCIR check taken from mmiotrace of the blob's PROM behaviour 7378 7379commit 63aa210f7aee4674337fed3ebfebd198ff1e699b 7380Author: Stuart Bennett <sb476@cam.ac.uk> 7381Date: Wed Jan 28 16:08:42 2009 +0000 7382 7383 Implement bios I2C opcode 4E (untested) 7384 7385 Courtesy an old trace by RAOF 7386 7387commit a81f1590d384bad214111139db9c988ff7676e83 7388Author: Stuart Bennett <sb476@cam.ac.uk> 7389Date: Wed Jan 28 15:39:24 2009 +0000 7390 7391 Handle I2C device 0xff fallback for I2C bios opcodes 7392 7393 Courtesy an old trace by RAOF 7394 Also, use i2c record entry length when available and make rundigitaloutscript 7395 function name legible 7396 7397commit 34e967e689a186cdde2f6c3ec222a6870b3584ea 7398Author: Stuart Bennett <sb476@cam.ac.uk> 7399Date: Wed Jan 28 15:48:27 2009 +0000 7400 7401 Fix pasta error in bios opcode 4D 7402 7403commit 786dd366474006b10c740c5012630ff5f1cd2a0f 7404Author: Stuart Bennett <sb476@cam.ac.uk> 7405Date: Thu Jan 29 17:56:40 2009 +0000 7406 7407 Fix nv04+ IFC upload and frag prog upload 7408 7409commit 3ab65cf6af0d3f892ade584c29f861f9f8a9d021 7410Author: Ben Skeggs <skeggsb@gmail.com> 7411Date: Thu Jan 29 16:51:53 2009 +1000 7412 7413 add vm_vram_base field to nouveau_device instead of hardcoding 512MiB 7414 7415commit d833c81fadf7dd63ab4ca94b953397fcbe0be581 7416Author: Ben Skeggs <skeggsb@gmail.com> 7417Date: Thu Jan 29 16:25:14 2009 +1000 7418 7419 nv50/xv: previous xv commit was bonged - fix that 7420 7421 It actually still worked just fine, but it was wrong. It's possible we'll 7422 have more than 1 clipbox, in which case the check for enough space in the 7423 pushbuf for a single frame is insufficient. 7424 7425 Check for enough space just before emitting quads, if there's not enough 7426 flush and reemit state - then go again. 7427 7428commit 0f741eb09c556f8d7893628964fedea862322771 7429Author: Ben Skeggs <skeggsb@gmail.com> 7430Date: Thu Jan 29 16:10:18 2009 +1000 7431 7432 exa: small cleanup 7433 7434commit c10c3c9bbabe8682fe1500098dc53f631690b182 7435Author: Ben Skeggs <skeggsb@gmail.com> 7436Date: Thu Jan 29 15:43:51 2009 +1000 7437 7438 exa: mass renaming and movement of nv04 solid/copy code to nv04_exa.c 7439 7440commit 81ace5dd1fa2c2760036fb1371cc61a383b2698e 7441Author: Ben Skeggs <skeggsb@gmail.com> 7442Date: Thu Jan 29 15:29:37 2009 +1000 7443 7444 exa: remove driver pixmaps path that never worked here anyway 7445 7446 Bits and pieces remain (disabled) in various modesetting bits for reference. 7447 Some of it may be useful when it reappears again. 7448 7449commit 4b1829e2c66f92a19a233afb749f3fb946930db5 7450Author: Ben Skeggs <skeggsb@nisroch.(none)> 7451Date: Tue Oct 7 00:44:46 2008 +1100 7452 7453 exa: remove NVCopyROP table.. 7454 7455commit 410788dc896d99b658508e3a7ebcb4940a039f35 7456Author: Ben Skeggs <skeggsb@gmail.com> 7457Date: Thu Jan 29 14:32:13 2009 +1000 7458 7459 blow up horribly if GPU access to mapped buffer is attempted 7460 7461 ...and fix the numerous cases where this was done. The current bufmgr couldn't 7462 care less, and synchronisation is done elsewhere. This won't be the case for 7463 much longer. 7464 7465commit 4c9346bd750797ee0f0485195f7ce8beb2114b04 7466Author: Ben Skeggs <skeggsb@gmail.com> 7467Date: Thu Jan 29 14:24:57 2009 +1000 7468 7469 bo_del->bo_ref, closer to ng api 7470 7471commit c533f86b512b9d84ec698163a8fbc54233648fbb 7472Author: Ben Skeggs <skeggsb@gmail.com> 7473Date: Thu Jan 29 14:13:06 2009 +1000 7474 7475 exa: same for UTS state 7476 7477commit 02677c07f9af01fa3b1eb787d6d320aa43c9140b 7478Author: Ben Skeggs <skeggsb@gmail.com> 7479Date: Thu Jan 29 14:10:51 2009 +1000 7480 7481 exa: ensure DFS state is updated across flushes 7482 7483commit 69d02bc73409c8010d19c9f87780b0c0e7621e43 7484Author: Ben Skeggs <skeggsb@gmail.com> 7485Date: Thu Jan 29 13:52:08 2009 +1000 7486 7487 nv50: remove some redundant init, this stuff is done as required now 7488 7489commit 60aea9057a78255a0ded1f64a9b9efd75017347f 7490Author: Ben Skeggs <skeggsb@gmail.com> 7491Date: Thu Jan 29 13:50:28 2009 +1000 7492 7493 nv50: forgotten buffer access flags during init 7494 7495commit 5a382dddc948c49f3f5159c078875ee6cdf56db7 7496Author: Ben Skeggs <skeggsb@gmail.com> 7497Date: Thu Jan 29 13:04:04 2009 +1000 7498 7499 nv10/exa: resubmit state in composite as needed 7500 7501commit d4ff8db7b0fb34e0494042b128251ad78c19ad09 7502Author: Ben Skeggs <skeggsb@gmail.com> 7503Date: Thu Jan 29 12:58:30 2009 +1000 7504 7505 nv30/exa: resubmit state in composite as needed 7506 7507commit b71ebcb2c8a0c97837f3de09f034c0bbb35f562c 7508Author: Ben Skeggs <skeggsb@gmail.com> 7509Date: Thu Jan 29 12:56:39 2009 +1000 7510 7511 nv40/exa: resubmit state in composite as necessary 7512 7513commit 6ddaddbb7b9442000f1e665bf61994b16d52cd25 7514Author: Ben Skeggs <skeggsb@gmail.com> 7515Date: Thu Jan 29 12:11:24 2009 +1000 7516 7517 exa: rework common pre-nv50 code in the same way nv50 code just was 7518 7519 Still have card-specific composite hooks to go. 7520 7521commit f36fa31401b060b9785d1835845c8ce09f8fe6b1 7522Author: Ben Skeggs <skeggsb@gmail.com> 7523Date: Thu Jan 29 11:25:22 2009 +1000 7524 7525 nv50/xv: ensure entire frame will fit in pushbuf without causing a flush 7526 7527commit 1b48331f6bd077f39797bb2d25b62c3c9df5ccb3 7528Author: Ben Skeggs <skeggsb@gmail.com> 7529Date: Thu Jan 29 10:38:53 2009 +1000 7530 7531 nv50/exa: ensure buffer usage is tracked correctly 7532 7533commit 7e644265267b4e35a05328b1c4cb1f2ef5beea58 7534Author: Ben Skeggs <skeggsb@gmail.com> 7535Date: Thu Jan 29 10:13:57 2009 +1000 7536 7537 remove unconditional FIRE_RING() in DoneComposite() hooks 7538 7539 Far, far too much flushing. Will completely destroy performance with kernel 7540 command submission. 7541 7542commit 34fc4461a6f612e4e00db483effe3a2b9760e70c 7543Author: Ben Skeggs <skeggsb@gmail.com> 7544Date: Thu Jan 29 09:47:21 2009 +1000 7545 7546 hack a flush_notify() hook in to match ng 7547 7548 preparation for rework of accel code so it'll be more suitable for ng-style 7549 command submission. 7550 7551commit d9da090cae1d2dfc539c9db4c923c0e665bce975 7552Author: Ben Skeggs <skeggsb@gmail.com> 7553Date: Thu Jan 29 09:38:39 2009 +1000 7554 7555 Drop GPU resources on LeaveVT, reaquire on EnterVT 7556 7557 This allows GDM's switch-user functionality to work on some systems. 7558 7559commit 1eb8b8778037336e530e113b8bbd61d724b1cc52 7560Author: Ben Skeggs <skeggsb@gmail.com> 7561Date: Wed Jan 28 13:50:55 2009 +1000 7562 7563 Enable ShadowFB for DRM-less mode. 7564 7565commit 487463895390500fd4e899f4e3badbb44f2b5429 7566Author: Ben Skeggs <skeggsb@gmail.com> 7567Date: Wed Jan 28 13:41:44 2009 +1000 7568 7569 Allow NoAccel operation without the DRM module present. 7570 7571commit f78aeed720ef185ccdea2d62e02b2714f00ee49c 7572Author: Ben Skeggs <skeggsb@gmail.com> 7573Date: Wed Jan 28 10:30:43 2009 +1000 7574 7575 if CARD_INIT ioctl fails, report it instead of continuing 7576 7577commit cf65b875ba3467060d8a3806914b75c710504b78 7578Author: Stuart Bennett <sb476@cam.ac.uk> 7579Date: Wed Jan 21 05:22:34 2009 +0000 7580 7581 Fix stupidity introduced in 25de6b867f319099dac05ba84f170da0f0e6c803 7582 7583 Wow. (function executes before bios is parsed, hence version check never works) 7584 7585commit 55e9cdda6e555937e8f5f3998a2e603669b84df4 7586Author: Stuart Bennett <sb476@cam.ac.uk> 7587Date: Tue Jan 20 05:09:22 2009 +0000 7588 7589 Split out a function for finding the panel power sequencer stuff 7590 7591 and add a bios_t convenience member 7592 7593commit 1b3546b0687a1ae8080009f0c58b26074f08d3af 7594Author: Stuart Bennett <sb476@cam.ac.uk> 7595Date: Tue Jan 20 04:11:11 2009 +0000 7596 7597 Move the bios flat-panel pointers back into the bios struct 7598 7599 Last there in 2007, removed in 82fdd3f76e20fc7dd10fcf00cf1c13520dc56364 :-) 7600 7601commit d29c4815b61635ed53ded94361ba52ee61ea5ac7 7602Author: Stuart Bennett <sb476@cam.ac.uk> 7603Date: Tue Jan 20 03:46:05 2009 +0000 7604 7605 Remove a BITism from BMP parsing path, and move old style BMP init into common path 7606 7607commit 3e971f63b6b6859acc19a52a4003bb6041cc5374 7608Author: Stuart Bennett <sb476@cam.ac.uk> 7609Date: Mon Jan 19 23:02:09 2009 +0000 7610 7611 randr12: lock independent crtc base registers separately (#19620) 7612 7613 Thanks to Mjules for tireless testing, surprising nobody else noticed it earlier 7614 7615commit ffbca7272878d440a06a613df95ce5d7be46f288 7616Author: Stuart Bennett <sb476@cam.ac.uk> 7617Date: Sun Jan 18 17:24:47 2009 +0000 7618 7619 randr12: code motion and allow ramdac580 setting on nv40 7620 7621commit 75a037623adecbe460198066b5059c349de427aa 7622Author: Stuart Bennett <sb476@cam.ac.uk> 7623Date: Sun Jan 18 15:41:23 2009 +0000 7624 7625 randr12: better selection of nv4x single pll mode and some notes 7626 7627commit 767fcea420fc21c5b5ec6c9f2db56c1a667db306 7628Author: Stuart Bennett <sb476@cam.ac.uk> 7629Date: Mon Jan 19 01:51:21 2009 +0000 7630 7631 randr12: no mmiotrace indication that the blob ever clears these bits 7632 7633 some hardware drops them on readback, but I don't think that matters 7634 also make defined bits shifts comprehensible 7635 7636commit 985ac728276633d14fa5577eacbc2095582b2587 7637Author: Stuart Bennett <sb476@cam.ac.uk> 7638Date: Sat Jan 17 23:21:19 2009 +0000 7639 7640 randr12: no mmiotrace evidence for nv47 using the special pll bits 7641 7642commit 3ef6b41a1c072d2f3545915180b522e43e0c8754 7643Author: Stuart Bennett <sb476@cam.ac.uk> 7644Date: Sun Jan 18 16:13:57 2009 +0000 7645 7646 twoStagePLL is actually a description of cards using two PLL regs 7647 7648commit c7a423a24ad7da7b510007a0d5a779999bd8cf73 7649Author: Stuart Bennett <sb476@cam.ac.uk> 7650Date: Sat Jan 17 04:34:48 2009 +0000 7651 7652 nv04 blitter: use correct number of dma params for real nv04 7653 7654commit 25de6b867f319099dac05ba84f170da0f0e6c803 7655Author: Stuart Bennett <sb476@cam.ac.uk> 7656Date: Fri Jan 16 20:22:16 2009 +0000 7657 7658 Complain more about weird bios images 7659 7660 also use bios chip version to test for g80 7661 7662commit 4d72981cc7d156708683aba51489d4691106438c 7663Author: Stuart Bennett <sb476@cam.ac.uk> 7664Date: Wed Jan 14 18:15:21 2009 +0000 7665 7666 randr12: avoid classic macro error 7667 7668 Guess there's not many ppc nv11 users, as it's been here since Feb '08 7669 7670commit 133c1a55134853161a25f358432fd05e736718b5 7671Author: Stuart Bennett <sb476@cam.ac.uk> 7672Date: Tue Jan 6 17:47:07 2009 +0000 7673 7674 No need of separate allocation for bios image 7675 7676 nor of manual allocations for short lived I2C devices 7677 7678commit 8a25049abd2b755187f47572f92b11e9072e522a 7679Author: Stuart Bennett <sb476@cam.ac.uk> 7680Date: Tue Jan 6 16:55:42 2009 +0000 7681 7682 Make turning on bios execution logging slightly easier 7683 7684 and add a minor parsing check 7685 7686commit 54d43c4760b573b5d376859281268046b61fb2f4 7687Author: Maarten Maathuis <madman2003@gmail.com> 7688Date: Fri Jan 9 20:02:44 2009 +0100 7689 7690 nv50: small fix 7691 7692 - planemask is rarely used, so it's just a cleanup. 7693 7694commit 5c9f8d3ce8964ad974175f94b858f9efc11756bb 7695Author: Maarten Maathuis <madman2003@gmail.com> 7696Date: Fri Jan 9 17:48:07 2009 +0100 7697 7698 nv50: Add a fallback for non-repeating XRGB pixmaps. 7699 7700 - Other card generations will most likely need this too. 7701 7702commit 0321e6741d9749f9f6b15ac2099139f24e88eef9 7703Author: Ben Skeggs <skeggsb@gmail.com> 7704Date: Thu Jan 8 14:31:39 2009 +1100 7705 7706 nv50: don't segfault if preinit fails before display setup done 7707 7708commit 7dc567d432fc6a3734bf64244e57bcc3612d5ec2 7709Author: Ben Skeggs <skeggsb@gmail.com> 7710Date: Mon Jan 5 10:51:03 2009 +1100 7711 7712 nv50: correct size/layout of uploaded data 7713 7714 This should fix the minor corruption that is present while playing some 7715 videos. 7716 7717commit 159c18cea2ecb92da74f9a9fdff0a9f5ef03df53 7718Author: Maarten Maathuis <madman2003@gmail.com> 7719Date: Sun Dec 28 17:56:55 2008 +0100 7720 7721 Make CHECK_TIMEOUT() a little more correct. 7722 7723 - Decrease the maximum counter value by a factor 16. 7724 - Determine t_start at the start. 7725 7726commit 5d281a2439de1e8c1848b6b700f30476575966e0 7727Author: Stuart Bennett <sb476@cam.ac.uk> 7728Date: Wed Dec 24 23:38:02 2008 +0000 7729 7730 nv04-nv4x: support RandR 1.3 panning 7731 7732commit 3c4705cb4472680b6cdc02fd5152d565d214226f 7733Author: Stuart Bennett <sb476@cam.ac.uk> 7734Date: Mon Dec 15 01:54:12 2008 +0000 7735 7736 Implement bios opcodes 4C and 4D 7737 7738commit 1fd60f202d41d6bb2028955d3bf9e6fc3921e3ff 7739Author: Stuart Bennett <sb476@cam.ac.uk> 7740Date: Mon Dec 15 01:19:26 2008 +0000 7741 7742 Group all DCB I2C info for a given bus in a structure 7743 7744 Furthermore, use the port type field to determine how to drive a bus, rather 7745 than the chip(s) characteristically associated with that port type 7746 7747commit 31c9958779d95ae84993f4772607367927a0e313 7748Author: Maarten Maathuis <madman2003@gmail.com> 7749Date: Fri Dec 19 19:59:53 2008 +0100 7750 7751 nv50: implement get_crtc 7752 7753 - Won't help much right now (pre-KMS), but it doesn't hurt to prototype now. 7754 7755commit f7a757827c1d62ac8a7246fe3b64fea1891a7a16 7756Author: Maarten Maathuis <madman2003@gmail.com> 7757Date: Thu Dec 18 19:57:50 2008 +0100 7758 7759 nv50: implement set_origin (mostly used for panning) 7760 7761commit 74b2e4e92822fb321a9304f0e077c18de57e585f 7762Author: Maarten Maathuis <madman2003@gmail.com> 7763Date: Wed Dec 17 22:42:20 2008 +0100 7764 7765 configure: Avoid compiling drm modesetting code by accident. 7766 7767commit d104f145260ff183b5f7e4c16cebc1680cf771b7 7768Author: Maarten Maathuis <madman2003@gmail.com> 7769Date: Tue Dec 16 23:54:39 2008 +0100 7770 7771 nv50: Improve lut code. 7772 7773commit 20f93a94ae034ee7c744947f42a29a7c0b50ee9d 7774Author: Stuart Bennett <sb476@cam.ac.uk> 7775Date: Wed Nov 26 01:37:38 2008 +0000 7776 7777 randr12: pass cursor visibility changes through mode state structures 7778 7779commit 763e0f88c79105a726ea05219875a25fc0129e1f 7780Author: Stuart Bennett <sb476@cam.ac.uk> 7781Date: Wed Nov 12 20:32:47 2008 +0000 7782 7783 randr12: allow ddc while crtc is locked 7784 7785commit 9d8bd459485392ec21d90c405510f3f097a61921 7786Author: Stuart Bennett <sb476@cam.ac.uk> 7787Date: Sat Nov 22 21:20:07 2008 +0000 7788 7789 Punt all bios structs and defines into a new header 7790 7791commit a34a2c07662b0175e03161cb48abaea9d814c1eb 7792Author: Stuart Bennett <sb476@cam.ac.uk> 7793Date: Thu Nov 20 17:35:59 2008 +0000 7794 7795 Do not attempt to parse FP mode tables for Mac laptops (#18636) 7796 7797 The two varieties of these laptops I've had the misfortune to encounter use DDC, 7798 and the NV34M based one appears to not include a fp mode table at all 7799 7800commit 1aba513a0b5a00996f28462cd4118214dffd87e2 7801Author: Ben Skeggs <skeggsb@gmail.com> 7802Date: Fri Nov 21 13:16:38 2008 +1100 7803 7804 nv50: program physical addresses into CRTCs, not VM addresses. 7805 7806commit 65b956fb407818117ce17ced6bbe66c55b4acb7c 7807Author: Stuart Bennett <sb476@cam.ac.uk> 7808Date: Mon Nov 17 20:02:19 2008 +0000 7809 7810 randr12: fix stupidity from 539f4990 7811 7812 found by AVRS2 7813 7814commit 1666e858dc027bc028083c15618da5cde5a1b6f1 7815Author: Maarten Maathuis <madman2003@gmail.com> 7816Date: Sun Nov 9 23:10:15 2008 +0100 7817 7818 nv50: use bios provided load detect value 7819 7820commit a020a44a2c48ff20bca6188469cb9cbbb5bf6093 7821Author: Maarten Maathuis <madman2003@gmail.com> 7822Date: Sun Nov 9 22:56:30 2008 +0100 7823 7824 bios: support g80 loadval table 7825 7826commit 220cbb6347cb0404bd24f457ac6cb5e39917cdbf 7827Author: Maarten Maathuis <madman2003@gmail.com> 7828Date: Sat Nov 8 18:01:18 2008 +0100 7829 7830 nv50_randr: plug a small memory leak 7831 7832commit cea05e14f94dcb842adbfcec515e17e9f5655089 7833Author: Stuart Bennett <sb476@cam.ac.uk> 7834Date: Sun Oct 26 03:51:36 2008 +0000 7835 7836 randr12: make colour-map setting less complex 7837 7838commit 0e4b01a5e06d8b5323c010ce89723369f17b10b2 7839Author: Stuart Bennett <sb476@cam.ac.uk> 7840Date: Sat Oct 25 02:11:37 2008 +0100 7841 7842 randr12: convenience functions for vga crtc state access 7843 7844commit fe9bfd9b1b5ce6fd0cd0360ec850ae48b4abac41 7845Author: Stuart Bennett <sb476@cam.ac.uk> 7846Date: Fri Oct 24 21:02:00 2008 +0100 7847 7848 Tidy EnterVT, and don't call it from ScreenInit 7849 7850commit 3e1ccff8744d9cabef0cd3901e987deeeb38ec12 7851Author: Stuart Bennett <sb476@cam.ac.uk> 7852Date: Fri Oct 24 02:49:19 2008 +0100 7853 7854 Condense crtc locking 7855 7856 We unlock all crtcs whenever the driver is active, and relock them when 7857 returning to VT (which is what we were doing, but in a more complicated manner) 7858 7859commit 83b541e0a9c454059d0e55a932a2ecdbbf650d60 7860Author: Stuart Bennett <sb476@cam.ac.uk> 7861Date: Mon Oct 27 19:20:22 2008 +0000 7862 7863 Common function for the bios condition table 7864 7865commit 36c6aa449063387b34c31015d618e53d4b6634e0 7866Author: Stuart Bennett <sb476@cam.ac.uk> 7867Date: Mon Oct 27 18:55:28 2008 +0000 7868 7869 Allow more time in INIT_CONDITION_TIME (#18234) 7870 7871commit bf810b32bc959f3821bca18cf158f711dd314a26 7872Author: Stuart Bennett <sb476@cam.ac.uk> 7873Date: Mon Oct 27 18:41:43 2008 +0000 7874 7875 More warnings for C51's magical unaligned regs 7876 7877commit d481b5e2e1dcb7ac5b5a7583bc6542b42567eac2 7878Author: Stuart Bennett <sb476@cam.ac.uk> 7879Date: Sat Oct 25 17:48:35 2008 +0100 7880 7881 Rework bios parser error handling 7882 7883commit 387f30821ff7add22376de33a203744cdc719849 7884Author: Stuart Bennett <sb476@cam.ac.uk> 7885Date: Sat Oct 25 00:38:29 2008 +0100 7886 7887 RivaTuner dumps call the `how much memory have I got?' reg NV_PFB_CSTATUS 7888 7889commit bb20175001748e5d2fab1b9cb2075bbe5d94e5ed 7890Author: Stuart Bennett <sb476@cam.ac.uk> 7891Date: Fri Oct 24 17:42:53 2008 +0100 7892 7893 pedantry fix for nv11 locking, and some minor cleanups 7894 7895commit 9aecc4226ca9067667004150f7d885b2247273b2 7896Author: Stuart Bennett <sb476@cam.ac.uk> 7897Date: Thu Oct 23 23:17:15 2008 +0100 7898 7899 Add a function to lock/unlock all crtcs, use it in nv_bios code 7900 7901 Also split initial head owner determination into separate function 7902 7903commit 48581da4d0e1663456ddcbf7a456029cf961d4cd 7904Author: Stuart Bennett <sb476@cam.ac.uk> 7905Date: Thu Oct 23 02:35:30 2008 +0100 7906 7907 Attempt to make the usage of cr44 rational instead of cargoculted 7908 7909 and a couple of minor cleanups in nv_crtc 7910 7911commit 1c7bee95610a47175064b17afd6aa2a45e311b1a 7912Author: Stuart Bennett <sb476@cam.ac.uk> 7913Date: Thu Oct 23 02:46:03 2008 +0100 7914 7915 Robust nv11 head setting (from mmiotrace) 7916 7917commit cf5162b5ed99666d3dc8db8328115afd6fff6a8e 7918Author: Stuart Bennett <sb476@cam.ac.uk> 7919Date: Tue Oct 21 02:12:24 2008 +0100 7920 7921 randr12: update scrn field on server regen, and emit an error if rotated 7922 7923 (updating the scrn is easier than removing and re-adding outputs) 7924 7925commit 13739cf36893e8ab6d3d4911c76a45f0b8bd5f4c 7926Author: Stuart Bennett <sb476@cam.ac.uk> 7927Date: Mon Oct 20 21:47:07 2008 +0100 7928 7929 randr12: don't change virtualX/virtualY 7930 7931 all the things in the ddx that use virtual* (setting displayWidth, allocating 7932 the FB, EXA init etc) use the values present before this function is called 7933 7934 these prior values are derived from any Virtual line in xorg.conf and are what 7935 we want to use 7936 7937 setting virtual* to the crtc size here, rather than the desired Virtual size, 7938 breaks everything using them in ScreenInit on server regeneration 7939 7940 as the rhd guys say: 7941 /* This is strange... if we set virtualX/virtualY like the intel driver 7942 * does, we limit ourself in the future to this maximum size. 7943 * The check for this is internally in RandR, no idea why the intel driver 7944 * actually works this way... 7945 * Even more curious: if we DON'T update virtual, everything seems to 7946 * work as expected... */ 7947 so I'll take that as precedent. 7948 7949commit 98a4dcb112221fa66d82b87ef9229f5c443daf90 7950Author: Stuart Bennett <sb476@cam.ac.uk> 7951Date: Mon Oct 20 18:59:35 2008 +0100 7952 7953 Delete notifiers and grobjs in CloseScreen to avoid channel member becoming stale on server regeneration 7954 7955commit 12fce00b2ecf4c6be6cc86c8a7374878bb7dfc20 7956Author: Stuart Bennett <sb476@cam.ac.uk> 7957Date: Wed Oct 22 02:47:45 2008 +0100 7958 7959 Remove all object with mmaps in CloseScreen, so that drmClose actually calls the drm release method 7960 7961 (and so subsequently, we can be DRM master on the next drmOpen) 7962 7963commit 8e3f27a30078675f1a7974ba436a68a5b45e0924 7964Author: Stuart Bennett <sb476@cam.ac.uk> 7965Date: Wed Oct 22 03:32:51 2008 +0100 7966 7967 No reason to call AccelCommonInit twice in ScreenInit, but calling it in EnterVT helps resume... 7968 7969commit f495fa9153ee57d3ee0dc735c3e9e836d84a2ccc 7970Author: Stuart Bennett <sb476@cam.ac.uk> 7971Date: Wed Oct 22 02:41:17 2008 +0100 7972 7973 nouveau_channel_free should remove mmaps made in nouveau_channel_alloc 7974 7975commit 1cedb8e49c832890efd12904e78dfd8ab5855691 7976Author: Stuart Bennett <sb476@cam.ac.uk> 7977Date: Wed Oct 22 02:15:06 2008 +0100 7978 7979 Do frag prog allocation and shader upload in TCL init 7980 7981 Similar in spirit to the nv50 inplementation. 7982 Removes many dependencies on state stored in static variables, which in turn 7983 makes reinitializing shaders on resume / server regeneration tractable. 7984 7985commit 9e2c08937b0d1bfed71997b2b16bded91830af71 7986Author: Stuart Bennett <sb476@cam.ac.uk> 7987Date: Wed Oct 22 01:44:10 2008 +0100 7988 7989 Only hackup shaders once 7990 7991commit afc6668d2b4a1971c211c421b6a415fe74dbd2ea 7992Author: Stuart Bennett <sb476@cam.ac.uk> 7993Date: Sat Oct 18 02:23:18 2008 +0100 7994 7995 cr26 is just another view of 0x3c0 7996 7997 remove superfluous crtc unlock too 7998 7999commit 539f4990c8f1a32da57992dc863f277f8ede8194 8000Author: Stuart Bennett <sb476@cam.ac.uk> 8001Date: Wed Oct 15 03:59:06 2008 +0100 8002 8003 Use symbolic define values where known 8004 8005commit 791666a76c68880136ae26d0477bf484cc667f72 8006Author: Stuart Bennett <sb476@cam.ac.uk> 8007Date: Wed Oct 15 00:25:26 2008 +0100 8008 8009 Convert CIO, DIO and VIO use to use defines from nvreg for index and data reg offsets 8010 8011commit a1b7f8d9b20bff59956ea1dc2130e5590a60aa4c 8012Author: Stuart Bennett <sb476@cam.ac.uk> 8013Date: Wed Oct 15 00:12:11 2008 +0100 8014 8015 Rename relevant functions, sizes and offsets to PRM.IO from P.IO, in keeping with the nvidia scheme 8016 8017commit 267c0eef02f5e2a48217235d170a0530de6a94d3 8018Author: Stuart Bennett <sb476@cam.ac.uk> 8019Date: Tue Oct 14 23:58:11 2008 +0100 8020 8021 Name crtc index regs according to rules.xml 8022 8023commit 9261c34e873356b02a0fda00e9911d9795d9f1f2 8024Author: Stuart Bennett <sb476@cam.ac.uk> 8025Date: Tue Oct 21 17:46:37 2008 +0100 8026 8027 SaveGeneration isn't doing a lot 8028 8029 add another close screen thing too 8030 8031commit c29c190036c721dac2e58e66d0c5a70a24377a1d 8032Author: Stuart Bennett <sb476@cam.ac.uk> 8033Date: Tue Oct 21 16:27:00 2008 +0100 8034 8035 Replace a few memsets with initializers 8036 8037commit 3b53f6ceb86458f9df9b7becc18a79e74b21c91f 8038Author: Stuart Bennett <sb476@cam.ac.uk> 8039Date: Sun Oct 12 15:52:43 2008 +0100 8040 8041 Eliminate separate NVRec ctor and dtor 8042 8043commit 1b18db28f124bd5021f61b72389e5c0ce003b212 8044Author: Stuart Bennett <sb476@cam.ac.uk> 8045Date: Sat Oct 11 21:05:57 2008 +0100 8046 8047 randr12: remove pointless debug in nv_output and nv_crtc 8048 8049commit 240d51dc928cea38f6b9d842b59011ad7c5500c0 8050Author: Stuart Bennett <sb476@cam.ac.uk> 8051Date: Sat Oct 11 18:40:35 2008 +0100 8052 8053 randr12: deBoolification and eliminate `override' arg to nv_crtc_load_state_ext 8054 8055commit b95c3e23f0eac7d12d5ee271af8ec5707ac4ef8e 8056Author: Stuart Bennett <sb476@cam.ac.uk> 8057Date: Sat Oct 11 18:29:40 2008 +0100 8058 8059 Make all CR 57/58 access use the proper functions and defines 8060 8061 Use a define for usleep in bios too 8062 8063commit 9988ae594f6f03385b261b0af83b324981f11593 8064Author: Stuart Bennett <sb476@cam.ac.uk> 8065Date: Sat Oct 11 17:56:23 2008 +0100 8066 8067 Redo bios logging 8068 8069commit 0e6a9cc61c88deaf0e0935d5a7d269621c3d6cd0 8070Author: Stuart Bennett <sb476@cam.ac.uk> 8071Date: Sat Oct 11 18:04:58 2008 +0100 8072 8073 Improve some nv_bios messages, remove others 8074 8075commit b175bfcd6485d9dcddb66ab5f79104567d256336 8076Author: Stuart Bennett <sb476@cam.ac.uk> 8077Date: Sat Oct 11 17:59:02 2008 +0100 8078 8079 Bios register list is easier to read when sorted numerically 8080 8081 and remove NULL initialization for calloc'd struct 8082 8083commit e51b49c938483ccbd172ff07d0385a6fb474b57e 8084Author: Stuart Bennett <sb476@cam.ac.uk> 8085Date: Sat Oct 11 02:35:40 2008 +0100 8086 8087 randr12: panels with edid have only a constant mode 8088 8089 prove me wrong. 8090 8091commit 765494ebf9b129bd613121cb714e928b09e19500 8092Author: Stuart Bennett <sb476@cam.ac.uk> 8093Date: Sat Oct 11 02:16:50 2008 +0100 8094 8095 Remove unnecessary "Setting owner" message 8096 8097commit bf585ad38a80f7828b821936296316bcee4658f5 8098Author: Maarten Maathuis <madman2003@gmail.com> 8099Date: Sat Oct 11 01:22:35 2008 +0200 8100 8101 The fifo channel should be removed along with closing of the fd, so only clear the client memory. 8102 8103commit 8b7e42461add708e75d3f7d675245f5bbf3652e8 8104Author: Maarten Maathuis <madman2003@gmail.com> 8105Date: Fri Oct 10 23:37:55 2008 +0200 8106 8107 git-log is no longer valid for git 1.6, switch to git log. 8108 8109commit 933fd803f0aadc934f448937dbeb305256fe8893 8110Author: Maarten Maathuis <madman2003@gmail.com> 8111Date: Fri Oct 10 23:14:57 2008 +0200 8112 8113 Some fixes that bring me closer to surviving to the 2nd X server generation. 8114 8115 - Not quite done, i get a fifo dump when something tries to render. 8116 - Something is keeping an fd, which is preventing a full takedown, but this is not a valid reason for failing (kernel modesetting 8117 for example will never have full takedown). 8118 8119commit daee3c38a02f48f9094f712800174b88b065f0ff 8120Author: Stuart Bennett <sb476@cam.ac.uk> 8121Date: Thu Oct 9 17:09:58 2008 +0100 8122 8123 randr12: separate cursor pixmap location setting function 8124 8125commit caf2e920f18cad1c57e945a4e474e63a65ec6a43 8126Author: Stuart Bennett <sb476@cam.ac.uk> 8127Date: Thu Oct 9 17:07:30 2008 +0100 8128 8129 Move cursor show/hide funcs to nv_hw, document nv40 bug, and set curctl2 before applying nv40 fix 8130 8131commit f76074c7041b1246efdf2c0a4b0c467e2fe0bef1 8132Author: Stuart Bennett <sb476@cam.ac.uk> 8133Date: Thu Oct 9 17:24:23 2008 +0100 8134 8135 Use correct refclk for nv_get_clock, and use bool type as appropriate in arbitration 8136 8137commit bd5b8074b4a250610cf8c9e2f181c2946e902bf8 8138Author: Stuart Bennett <sb476@cam.ac.uk> 8139Date: Thu Oct 9 15:25:16 2008 +0100 8140 8141 Tidy arbitration prototypes and split randr12 cursor pixmap setting from arbitration 8142 8143 Also fix ages old warning in nv_hw 8144 8145commit 5dc41fdcf4ba76d7154d32f030d2e99756681e4a 8146Author: Stuart Bennett <sb476@cam.ac.uk> 8147Date: Fri Oct 10 01:11:46 2008 +0100 8148 8149 Merge nv4, nv10 and nForce UpdateArbitrationSettings functions 8150 8151commit 0ffd012c922d241aec22c8e44416171e39e6bf0c 8152Author: Stuart Bennett <sb476@cam.ac.uk> 8153Date: Fri Oct 10 00:57:45 2008 +0100 8154 8155 Add PCI slot reading defines, and use them for nForce code 8156 8157 Also move nForce dimm checking code to nv_setup, so it is only tested once 8158 8159commit 520a76ef3df3cdc735691cfda329ceaae0ec3123 8160Author: Stuart Bennett <sb476@cam.ac.uk> 8161Date: Thu Oct 9 20:22:34 2008 +0100 8162 8163 Some tidying of arbitration functions 8164 8165commit 3ceb2b60a09379ef51635f8309fdc82060681e8f 8166Author: Stuart Bennett <sb476@cam.ac.uk> 8167Date: Thu Oct 9 19:05:01 2008 +0100 8168 8169 Reindent arbitration stuff 8170 8171commit d7439fc50a924f6f1a6b5b59ba3ec076cf90c52e 8172Author: Stuart Bennett <sb476@cam.ac.uk> 8173Date: Wed Oct 8 23:45:13 2008 +0100 8174 8175 Wrap usleep when called in bios code 8176 8177commit 62cf9f371c420abe7a74ac4bb9fa2da081f1d29b 8178Author: Stuart Bennett <sb476@cam.ac.uk> 8179Date: Fri Oct 10 16:41:29 2008 +0100 8180 8181 randr12: don't use in use (by another vga output) crtc for load detect (avoids flicker) 8182 8183commit f1d0fd0595b40d3a7f555bada88f1f79b517d610 8184Author: Stuart Bennett <sb476@cam.ac.uk> 8185Date: Tue Oct 7 21:39:10 2008 +0100 8186 8187 Better behaviour on allocator fail 8188 8189 also fix randr12 code which doesn't know how to call calloc 8190 and be sure to use the number of actually inited crtcs 8191 8192commit 2b67015e5911cb7e321da72d7d3d5f0fefedd8e0 8193Author: Stuart Bennett <sb476@cam.ac.uk> 8194Date: Tue Oct 7 02:35:44 2008 +0100 8195 8196 randr12: reorder nv_output functions 8197 8198 roughly into order of use 8199 8200commit 9f4eef0165766be2e1aa5516eae409e94c09553b 8201Author: Stuart Bennett <sb476@cam.ac.uk> 8202Date: Thu Sep 18 18:32:36 2008 +0100 8203 8204 randr12: distinguish detected encoder from active encoder 8205 8206 don't swap output props until a mode is set on new encoder either 8207 8208commit 549cda114ee378640dda89621b0ce9c937c473da 8209Author: Stuart Bennett <sb476@cam.ac.uk> 8210Date: Wed Oct 8 01:22:17 2008 +0100 8211 8212 Reduce includes, everything necessary is in nv_include.h 8213 8214commit 30eaa0f08250f72b0171a858f2694ca8a62c3ebd 8215Author: Stuart Bennett <sb476@cam.ac.uk> 8216Date: Sun Sep 21 00:52:56 2008 +0100 8217 8218 Oops. 8219 8220commit 2481e5646ab3ea5e18d3f5ee4042fa20a8dad3de 8221Author: Stuart Bennett <sb476@cam.ac.uk> 8222Date: Sun Sep 21 00:44:50 2008 +0100 8223 8224 Optimizing memory usage for unaccelerated case is not interesting 8225 8226commit 19f0ea63a505e6a840cf07bc554fd8f1787616b4 8227Author: Stuart Bennett <sb476@cam.ac.uk> 8228Date: Sun Sep 21 00:43:57 2008 +0100 8229 8230 randr12: rename mon -> edid 8231 8232commit 0cb081758f52e73c3055758ed7aced12b93c2519 8233Author: Stuart Bennett <sb476@cam.ac.uk> 8234Date: Sun Sep 21 00:31:40 2008 +0100 8235 8236 randr12: crtc destroy 8237 8238 also alleviate world shortage of curly braces 8239 8240commit 37adca1dd68d5791b0371eed4b96f6613fc747c2 8241Author: Stuart Bennett <sb476@cam.ac.uk> 8242Date: Sat Sep 20 23:57:06 2008 +0100 8243 8244 Need to increment dcb entry index when fabricating additional entries 8245 8246commit e6b6d8c9189f3b5ebbc77f55e318246be0fa1bd4 8247Author: Maarten Maathuis <madman2003@gmail.com> 8248Date: Fri Sep 19 21:23:04 2008 +0200 8249 8250 Also remove man page entry. 8251 8252commit cb1e9c1ab288bcf380789f775533e4c8ff8705db 8253Author: Maarten Maathuis <madman2003@gmail.com> 8254Date: Fri Sep 19 21:14:09 2008 +0200 8255 8256 NV50: remove {Prepare,Finish}Access since i realise now it's not so useful 8257 8258 - despite that the xserver code suggests that pixmaps can be offscreen on this call, that is not done as far as i can tell 8259 - not to mention that there is no way to know for sure if a pixmap is offscreen (exaPixmapIsOffscreen is not exported), making the 8260 offset useless 8261 - this will (obviously) become relevant as soon as driver managed pixmaps come into play, but that is for another day 8262 8263commit 3e397f575df6c3fee6eb9ee810fb3fc70d2c0cc8 8264Author: Maarten Maathuis <madman2003@gmail.com> 8265Date: Fri Sep 19 20:41:59 2008 +0200 8266 8267 NV50: The TMDS dual link threshold is meaningless for LVDS 8268 8269commit 599d258948be0505f02947251ad6877e7168d9a8 8270Author: Stuart Bennett <sb476@cam.ac.uk> 8271Date: Mon Aug 25 01:47:45 2008 +0100 8272 8273 randr12: multiple encoders per connector (DVI-I) 8274 8275 that this doesn't break things in subtle and unexpected ways is too much to 8276 hope for. one not-subtle and expected way things will break is that xorg.confs 8277 with specific setups for DVI-A-n and DVI-D-n connections will need renaming (to 8278 DVI-I-n). 8279 8280 as the last patch in the series, this also fixes an issue introduced with 8281 144b04a2 (randr12: fix dpms, detect, destroy, ...), that for a screen connected 8282 to a dvi connector, both the analogue and digital variants were detected as 8283 connected 8284 8285commit 5e8ac848936a1b70babc417454599c0231343399 8286Author: Ben Skeggs <skeggsb@gmail.com> 8287Date: Wed Sep 17 22:12:07 2008 +1000 8288 8289 nv50: 0xa0 family uses class 0x8397 for 3D 8290 8291commit 9c36eefe991daf47d20badffb3009def99473bdd 8292Author: Stuart Bennett <sb476@cam.ac.uk> 8293Date: Sun Sep 14 23:01:49 2008 +0100 8294 8295 randr12: avoid weirdness when tv-out happens to share i2c_index with another output 8296 8297commit 144b04a258eaa8e3b63c48b5e574524b40a59ac7 8298Author: Stuart Bennett <sb476@cam.ac.uk> 8299Date: Sun Aug 24 04:15:20 2008 +0100 8300 8301 randr12: fix dpms, detect, destroy, save and restore for multiple encoders per connector 8302 8303commit 2b30b8914a32049e87205cf896b2e89ab6649bca 8304Author: Stuart Bennett <sb476@cam.ac.uk> 8305Date: Sun Aug 24 04:26:19 2008 +0100 8306 8307 randr12: re-do detection of dvi-a vs vga 8308 8309commit 50bac3fcd24072b8926ba2dbebd571424fbb7d6d 8310Author: Stuart Bennett <sb476@cam.ac.uk> 8311Date: Sun Aug 24 03:56:16 2008 +0100 8312 8313 randr12: unify output funcs for analogue and tmds 8314 8315commit 69e30e6727895a8edeffc9a348b13a353cd0c6d0 8316Author: Stuart Bennett <sb476@cam.ac.uk> 8317Date: Sun Aug 24 03:12:58 2008 +0100 8318 8319 randr12: unified entry path for dpms 8320 8321commit a794a0c864e5b651bea5e40a4e10978aee9da5b6 8322Author: Stuart Bennett <sb476@cam.ac.uk> 8323Date: Sun Aug 24 04:45:48 2008 +0100 8324 8325 randr12: nouveau_output -> nouveau_connector 8326 8327commit 0a5119bd8a927da1f3f4ed6caa2e93acdae29c04 8328Author: Stuart Bennett <sb476@cam.ac.uk> 8329Date: Sun Aug 24 02:45:41 2008 +0100 8330 8331 randr12: separate encoder struct 8332 8333commit 352b8fd98c519f85ade02aa679d342bc64faa508 8334Author: Stuart Bennett <sb476@cam.ac.uk> 8335Date: Sun Sep 14 16:53:12 2008 +0100 8336 8337 Another DCB 1.4/1.5 TV (#17471) 8338 8339 I think we've got enough for TVs to no longer be fatal 8340 8341commit e8ba4e0baa62f02ba85258847963512791d89385 8342Author: Stuart Bennett <sb476@cam.ac.uk> 8343Date: Sun Sep 14 16:44:32 2008 +0100 8344 8345 Only use OF bios image on PPC 8346 8347commit 54b8a7ea43ec20c798dc93d881cdffa5cb1125a0 8348Author: Ben Skeggs <skeggsb@gmail.com> 8349Date: Thu Sep 11 03:29:46 2008 +1000 8350 8351 nv50/exa: work around corruption issues (see detailed commit msg) 8352 8353 It's not possible for us to do this properly without using the exa 8354 pixmap hooks, they require a real memory manager, and is being 8355 worked on in the ng branches, which doesn't currently support nv5x 8356 at all. 8357 8358 For a better solution until then: 8359 http://nouveau.freedesktop.org/wiki/NV50Support 8360 8361commit f8c0af3886142ea59936a77006ead0cd2c4ff795 8362Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 8363Date: Fri Sep 5 23:08:14 2008 +0200 8364 8365 Partly revert ea152819f45b6cf92f6742ed3f9f639cdbbd53f0, alignment is needed. 8366 8367commit ea152819f45b6cf92f6742ed3f9f639cdbbd53f0 8368Author: Maarten Maathuis <madman2003@gmail.com> 8369Date: Fri Sep 5 22:51:27 2008 +0200 8370 8371 exa: align offscreenBase to something sensible 8372 - Aligning virtualY makes no sense whatsoever. 8373 - displayWidth is the "pitch-width" of the framebuffer. 8374 8375commit 95c19b98b38310c543758fe4288a9e47b970fe67 8376Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 8377Date: Thu Sep 4 07:51:03 2008 +0200 8378 8379 Add 8200 detection. 8380 8381commit 6dd8ad4ed3f89c6ed9e61708a587d417c8d984fe 8382Author: Maarten Maathuis <madman2003@gmail.com> 8383Date: Sat Aug 30 17:51:05 2008 +0200 8384 8385 nv50: solid fill shouldn't use a pattern rop + minor changes 8386 8387commit b2b726d59fd8af819a7e464824a49f6e6fc10b84 8388Author: Stuart Bennett <sb476@cam.ac.uk> 8389Date: Sun Aug 24 00:18:40 2008 +0100 8390 8391 randr12: merge output detection functions 8392 8393commit a223e581b06ffb308aecc2365795951e2c021f29 8394Author: Stuart Bennett <sb476@cam.ac.uk> 8395Date: Sun Aug 24 00:08:41 2008 +0100 8396 8397 randr12: remember the edid from detection for getting modes later 8398 8399 also split native mode finding into separate function 8400 8401commit 16d9e89d9cfae1c21999d880c557ae283e14178e 8402Author: Stuart Bennett <sb476@cam.ac.uk> 8403Date: Sun Aug 24 19:56:24 2008 +0100 8404 8405 randr12: detypedef private structs and use a define to get to the privates 8406 8407commit 3b3f3cdbe0f69c57153e070f248ff001cd5a4a17 8408Author: Stuart Bennett <sb476@cam.ac.uk> 8409Date: Sun Aug 24 00:31:35 2008 +0100 8410 8411 Remove unused code 8412 8413commit 1e2b87c459a3a0eacd630f1de8b567c36cc792bd 8414Author: Stuart Bennett <sb476@cam.ac.uk> 8415Date: Mon Aug 25 20:23:05 2008 +0100 8416 8417 randr12: make the nouveau experience less green 8418 8419commit f89af0e6bc5ac88d0755f9402b772ca823518298 8420Author: Maarten Maathuis <madman2003@gmail.com> 8421Date: Sat Aug 23 03:44:28 2008 +0200 8422 8423 nv50: micro cleanup 8424 8425commit 94bf1060cc9d2f4c0a14bc036f8b76c4d9bfc7d7 8426Author: Maarten Maathuis <madman2003@gmail.com> 8427Date: Sat Aug 23 03:39:39 2008 +0200 8428 8429 nv50: do ROPs properly this time 8430 - Table should work for pre-nv50 as well, i just don't have a card plugged in to be sure. 8431 8432commit 6bd14e4ffe1d5fd9420c4ebb46c9c0844d0547f9 8433Author: Maarten Maathuis <madman2003@gmail.com> 8434Date: Fri Aug 22 17:56:31 2008 +0200 8435 8436 nv10: composite is an async operation by default, no need to call exaMarkSync(). 8437 8438commit 9f11d13e68bbf72be814b37fa3d03853eec981b9 8439Author: Maarten Maathuis <madman2003@gmail.com> 8440Date: Fri Aug 22 17:53:11 2008 +0200 8441 8442 exa: FIRE_RING at the end of composite 8443 8444commit 137d099294617adead2cf3d5cc3372d18579eaef 8445Author: Maarten Maathuis <madman2003@gmail.com> 8446Date: Thu Aug 21 23:17:06 2008 +0200 8447 8448 nv50: move VERTEX_{START,END} to {Prepare,Done}Composite 8449 - If PrepareComposite returns TRUE, then everything upto DoneComposite is required. 8450 8451commit 34dc05fbc4835ec68de5a1f70cc094ff6efcf310 8452Author: Maarten Maathuis <madman2003@gmail.com> 8453Date: Thu Aug 21 22:50:49 2008 +0200 8454 8455 nv50: minor tweaks to composite 8456 - Make the blend stuff more consistent with other composite implementations. 8457 - Verified with "rendercheck -f a8r8g8b8,x8r8g8b8,r5g6b5,a8 -t {blend,composite,cacomposite}" for all accelerated ops. 8458 8459commit eca997750b1cc2263f804afba242f16b6d960e02 8460Author: Maarten Maathuis <madman2003@gmail.com> 8461Date: Thu Aug 21 20:01:49 2008 +0200 8462 8463 nv50: handle ROPs better 8464 - Handle ROP's more like pre-NV50 code. 8465 - Special case for 32bpp which seems to work. 8466 - This does NOT fix all rendering issues, just an improvement. 8467 8468commit e468df855abd8d8e4fecd2445c82bb8f824297fb 8469Author: Maarten Maathuis <madman2003@gmail.com> 8470Date: Wed Aug 20 19:14:54 2008 +0200 8471 8472 nv50: clip SIFC and add {Prepare,Finish}Access hooks. 8473 8474 - 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. 8475 - {Prepare,Finish}Access hooks force tiled memory to be migrated upon cpu access. 8476 - Added an option to disable these hooks in case you hit a xserver bug that existed up until very recently. 8477 8478commit 0c0c1ab2ce786def653e8adb95c7c5a8ff9c7eca 8479Author: Stuart Bennett <sb476@cam.ac.uk> 8480Date: Tue Aug 19 00:02:00 2008 +0100 8481 8482 randr12: don't cache pll values (fixes newrestore removal regression) 8483 8484 for 5ms, it's really not worth it 8485 8486commit 544fd3c502287754f2161c21514203ccaab8aed9 8487Author: Stuart Bennett <sb476@cam.ac.uk> 8488Date: Mon Aug 18 21:50:37 2008 +0100 8489 8490 Remove some of the inaccuracies in the manpage 8491 8492commit e3d9cb926e56d28b7cadd0026ae76c076faced75 8493Author: Stuart Bennett <sb476@cam.ac.uk> 8494Date: Mon Aug 18 21:40:36 2008 +0100 8495 8496 NV_ARCH_04 can have rotated output, it's just not accelerated 8497 8498commit 40e920fafb9f22da23ef11c941561bb14966dc47 8499Author: Stuart Bennett <sb476@cam.ac.uk> 8500Date: Mon Aug 18 21:39:18 2008 +0100 8501 8502 Delete ShadowFB rotation 8503 8504 And nothing of value was lost. 8505 8506commit 6352d7db24451b014b5f6b873881d103fc1eb316 8507Author: Stuart Bennett <sb476@cam.ac.uk> 8508Date: Mon Aug 18 00:59:32 2008 +0100 8509 8510 Bye bye 8 bit depth 8511 8512commit 33d52e476b64cd91dc4b01c6826892f2da9499b0 8513Author: Stuart Bennett <sb476@cam.ac.uk> 8514Date: Sun Aug 17 23:52:28 2008 +0100 8515 8516 CrtcNumber option is long gone 8517 8518commit a9393a87d36cc486114842275bb78e7e75de754e 8519Author: Stuart Bennett <sb476@cam.ac.uk> 8520Date: Sun Aug 17 23:17:01 2008 +0100 8521 8522 Micro-tidyups to nv_driver and nouveau_xv 8523 8524commit 8ed30d05afe4a5bb077c088c48c28741414c09d3 8525Author: Stuart Bennett <sb476@cam.ac.uk> 8526Date: Sun Aug 17 16:57:17 2008 +0100 8527 8528 randr12: a few line length improvements 8529 8530commit 666ab70754035c569eaa2550ca77c92fe3eaaff3 8531Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 8532Date: Sun Aug 17 22:44:11 2008 +0200 8533 8534 Kill some 8 bit code. 8535 8536commit bee8450931395b23040a4cb1ddcb09f8a476cb93 8537Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 8538Date: Sun Aug 17 22:25:43 2008 +0200 8539 8540 Fix IFC for 16bpp. Also fix the color accuracy of 16bpp solid fills. 8541 8542commit 609b10ff00c9e8d51fd7c9a5fb62120516780aed 8543Author: Stuart Bennett <sb476@cam.ac.uk> 8544Date: Sun Aug 17 03:51:39 2008 +0100 8545 8546 0x48 exception does not apply on BIT cards 8547 8548commit a9e2d47e2cee7db447063d7c33f3e42a8d865c92 8549Author: Stuart Bennett <sb476@cam.ac.uk> 8550Date: Sun Aug 17 02:48:33 2008 +0100 8551 8552 randr12: some code simplification, rearrangement and tidying 8553 8554 We don't claim to support a bpp of 15 or 24, so some old calculations can be simplified 8555 8556commit 5715f95b37ad182377024f468982608b48eb3409 8557Author: Stuart Bennett <sb476@cam.ac.uk> 8558Date: Sun Aug 17 02:27:42 2008 +0100 8559 8560 NewRestore: remove, due to lack of interest 8561 8562 nobody wants to do text modes manually. for any future masochists, it's here in the history 8563 8564commit d2e924e847273c5b4d342d0ad611de7d106df429 8565Author: Stuart Bennett <sb476@cam.ac.uk> 8566Date: Sun Aug 17 01:37:41 2008 +0100 8567 8568 randr12: remove another field member 8569 8570commit 55f490c8c29b103d7d8853e2126a8366304f4125 8571Author: Stuart Bennett <sb476@cam.ac.uk> 8572Date: Sun Aug 17 01:26:20 2008 +0100 8573 8574 randr12: fpWidth and fpHeight no more 8575 8576commit 4cbbd44db614d250fdaaf648db2d70ec9bc2f6d9 8577Author: Stuart Bennett <sb476@cam.ac.uk> 8578Date: Sun Aug 17 00:38:23 2008 +0100 8579 8580 randr12: improve aspect scaling code 8581 8582commit 8a364682fd67c10c153cb0778c914a4754e82ddd 8583Author: Stuart Bennett <sb476@cam.ac.uk> 8584Date: Sat Aug 16 23:15:29 2008 +0100 8585 8586 randr12: for native scaling, a clock check isn't enough 8587 8588 1280x1024 and 1280x960 can have the same clock 8589 8590commit 8793bb1d7b79632678d7ce9f1e94c79dfd55fa3c 8591Author: Stuart Bennett <sb476@cam.ac.uk> 8592Date: Fri Aug 15 23:02:48 2008 +0100 8593 8594 Apply minimum front porch only when necessary (#15949) 8595 8596 Also, read minimum front porch from bios table 8597 8598commit 83bb26fd9a8bb283c406bc34d1d7ad0b53d7098a 8599Author: Maarten Maathuis <madman2003@gmail.com> 8600Date: Sat Aug 16 17:18:21 2008 +0200 8601 8602 NV50: disallow doublescan modes on TMDS/LVDS 8603 8604commit 7d5b345f69bf63d36207a7f6ac4d01001d62e79c 8605Author: Maarten Maathuis <madman2003@gmail.com> 8606Date: Sat Aug 16 15:33:13 2008 +0200 8607 8608 NV50: extra warning message 8609 8610commit 743c696c8b0e420a67ce1552ac0cec49c0fdc0f2 8611Author: Maarten Maathuis <madman2003@gmail.com> 8612Date: Sat Aug 16 15:28:43 2008 +0200 8613 8614 NV50: a half decent attempt at doing something when LVDS has no DDC 8615 8616commit fcbaa10a945e04d77c6ea8f83b738fb6c9e47dd5 8617Author: Maarten Maathuis <madman2003@gmail.com> 8618Date: Sat Aug 16 14:35:05 2008 +0200 8619 8620 Forgot something. 8621 8622commit 628acddedc7755a16471e9beee3d95d3b8f11150 8623Author: Maarten Maathuis <madman2003@gmail.com> 8624Date: Sat Aug 16 14:24:18 2008 +0200 8625 8626 NV50: support extended i2c ports (4 and 5) 8627 8628commit c845f5ac0a54b17a9911155030234c229e80626c 8629Author: Maarten Maathuis <madman2003@gmail.com> 8630Date: Sat Aug 16 14:02:51 2008 +0200 8631 8632 nv50: fix some of the rendering bugs 8633 8634 - It seems that 32bpp solid fill only works well for for the non-rop case. 8635 - The SRCCOPY_PREMULT comes from nv and seems to work well, in combination with the above fallback. 8636 - This does not affect the strange stripes with font rendering issue. 8637 8638commit 18288208bdb7f872f0a2e227d20b77c3397747c2 8639Author: Stuart Bennett <sb476@cam.ac.uk> 8640Date: Fri Aug 15 01:40:01 2008 +0100 8641 8642 Redo parsing for panels with EDID, and fix nv3x in the process (#17138) 8643 8644 breakage a possibility :) 8645 8646commit c9d4e86022bffd00d89076a921b734645145f7db 8647Author: Stuart Bennett <sb476@cam.ac.uk> 8648Date: Tue Aug 12 14:15:03 2008 +0100 8649 8650 Try harder to load a good vbios image 8651 8652commit ae594786ffc1cdca5bdd8c8638225819f851061b 8653Author: Stuart Bennett <sb476@cam.ac.uk> 8654Date: Wed Aug 6 00:31:39 2008 +0100 8655 8656 Deal with mobile cards that scribble over the fp strap at POST 8657 8658 It's probably safe to assume a mobile card will have been POSTed at boot, so 8659 this should be fine. 8660 Hopefully there are no cards which have bit 2 set at 0x48 and now break... 8661 8662commit 776670613f79cce3891c65d2f688979767cd1e6d 8663Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 8664Date: Fri Aug 15 00:43:02 2008 +0200 8665 8666 Fix a 16bpp issue, also remove old code that mostly worked around core EXA issues in older xservers. 8667 8668commit 6854f819f6bb89d67bfb737fd24c3e8b16e67e46 8669Author: Maarten Maathuis <madman2003@gmail.com> 8670Date: Tue Aug 5 13:08:22 2008 +0200 8671 8672 NV50: Disable messages from hide, show and load cursor, as they clutter the logs. 8673 8674commit 4b24be811b839f9f836b9f8574fc808d2182dd30 8675Author: Ben Skeggs <skeggsb@gmail.com> 8676Date: Mon Aug 4 16:09:28 2008 +1000 8677 8678 xv: fix thinko from when source was reformatted 8679 8680commit 30f54f2a4b81afde129170a26189cdb2bd3c7e9c 8681Author: Younes Manton <younes.m@gmail.com> 8682Date: Sun Aug 3 12:29:57 2008 -0400 8683 8684 g3dvl: Temporarily disable IDCT. 8685 8686commit 89d786435267af58bcecaff206a09b035e666df5 8687Author: Ben Skeggs <skeggsb@gmail.com> 8688Date: Tue Jul 29 23:38:45 2008 +1000 8689 8690 nv50: fix mystery typo 8691 8692commit c0b67f3f3490e826c86eedb32bbcf327592aaaf5 8693Author: Ben Skeggs <skeggsb@gmail.com> 8694Date: Tue Jul 29 01:26:20 2008 +1000 8695 8696 remove use of implicit variables from pushbuf macros 8697 8698 This matches the gallium winsys now. At some point the ddx will get updates 8699 so the "common" code is once again in sync to make it easier to do mm 8700 changes to both the ddx and dri simultaneously. 8701 8702 But, I want to get this part out of the way first as the changes might take 8703 a bit of work - and I don't want to have to deal with merge conflicts 8704 everywhere if someone changes the accel code before I'm done :) 8705 8706commit 80278c9f53f7c357e499bf2b3c5a1dff75fb9651 8707Author: Younes Manton <younes.m@gmail.com> 8708Date: Sun Jul 27 16:48:58 2008 -0400 8709 8710 g3dvl: Basic XvMC initialization. 8711 8712commit 1f8e27fd1d92d1c07dd8387f1a5e751ab3fab54a 8713Author: Maarten Maathuis <madman2003@gmail.com> 8714Date: Sun Jul 27 00:09:02 2008 +0200 8715 8716 NV50: I was proven wrong quicker than i thought, reinstate a slightly prettier hack. 8717 8718commit 41d46f59eb055130d5d415f0573fbede957c66c3 8719Author: Maarten Maathuis <madman2003@gmail.com> 8720Date: Sat Jul 26 21:53:26 2008 +0200 8721 8722 NV50: forgot something 8723 8724commit 1b905249c492af4485a9a2f4e7ae3ad5441c8eb5 8725Author: Maarten Maathuis <madman2003@gmail.com> 8726Date: Sat Jul 26 21:50:34 2008 +0200 8727 8728 NV50: my best guess at the lvds bios table, time will tell if it's correct for more than a few cases 8729 8730commit f3ec6e0e2d84e172cb634fd1348a54aef7256e74 8731Author: Dave Airlie <airlied@redhat.com> 8732Date: Thu Jul 24 15:41:57 2008 +1000 8733 8734 nouveau: fix macro as pointed out my moondrake on irc 8735 8736commit 6652e9c85fe8e807ed3d8ff76ec18c250aef436f 8737Author: Ben Skeggs <skeggsb@nisroch.keine.ath.cx> 8738Date: Thu Jul 24 10:13:05 2008 +1000 8739 8740 nv50: support YUY2 in textured video adaptor 8741 8742commit 3534f40b64496d9f6f95148ab81b9321057f7af1 8743Author: Ben Skeggs <skeggsb@gmail.com> 8744Date: Thu Jul 24 09:39:09 2008 +1000 8745 8746 nv50: remove 32bpp solid fill fallback 8747 8748 It seems to work, but there's probably a good reason it's a fallback in nv, 8749 so may be reincarnated with a more specific fallback if needed. 8750 8751commit 87f7d1f7084032110c347cb73f63e9a929d8c00e 8752Author: Ben Skeggs <skeggsb@gmail.com> 8753Date: Thu Jul 24 09:38:27 2008 +1000 8754 8755 Revert "NV50EXA: read the notes + cleanup + enabled 32bpp solid fill" 8756 8757 This reverts commit 0b6249fed6909af30856e1afc1a0dd4cd98c0b9e. 8758 8759commit ba7c239e7a13c4726dab7c686ee1c7f34d03a593 8760Author: Maarten Maathuis <madman2003@gmail.com> 8761Date: Thu Jul 24 01:30:16 2008 +0200 8762 8763 NV50EXA: serious brain fart of my part, sorry 8764 8765 - prepare copy may set up a rop and will set operation appropriately. 8766 8767commit 0b6249fed6909af30856e1afc1a0dd4cd98c0b9e 8768Author: Maarten Maathuis <madman2003@gmail.com> 8769Date: Wed Jul 23 21:01:57 2008 +0200 8770 8771 NV50EXA: read the notes + cleanup + enabled 32bpp solid fill 8772 8773 - I observe occasional corruption, some of this goes away when using a compositing manager. 8774 - The above statement contradicts my earlier statement in the previous commit message. 8775 - 32bpp solid fills seem to work fine, but i left the original fallback in case something pops up. 8776 - There is definately something wrong somewhere, i just don't know what. 8777 8778commit edbfbd09ac0a05c4125c3217f5d9c8f198d9345d 8779Author: Maarten Maathuis <madman2003@gmail.com> 8780Date: Wed Jul 23 13:34:28 2008 +0200 8781 8782 EXA: put NOUVEAU_FALLBACK in a do { } while (0), fixed opera corruption for me. 8783 8784 - I don't understand why a single line define would need this, but i noticed that turning on the fallback info fixed rendering. 8785 8786commit 36d13082dd2c2ba5d490b3dbc6e82d9ccfed8669 8787Author: Maarten Maathuis <madman2003@gmail.com> 8788Date: Sun Jul 20 15:44:47 2008 +0200 8789 8790 NV50_KMS: some basic scaling and dithering output property support 8791 8792commit 022a9ed5c2657466dd2471a1b29d5e8bc92ca57a 8793Author: Ben Skeggs <skeggsb@gmail.com> 8794Date: Sat Jul 19 22:14:50 2008 +1000 8795 8796 nv50: no more sync in composite() :) 8797 8798 0x502d mthd 0x0110 some kind of 2D cache flush? 8799 8800commit 6d8096c72e1aa160cef6729443ed06305a8ae870 8801Author: Ben Skeggs <skeggsb@gmail.com> 8802Date: Fri Jul 18 13:57:13 2008 +1000 8803 8804 nv50: sync after composite for the moment 8805 8806 This corrects some severe text rendering issues that get *very* bad with 8807 server 1.5. Will look into a proper fix soon. 8808 8809commit c68d8806b309ef4b48d7a98d6fc7aeccb8bb05ac 8810Author: Ben Skeggs <skeggsb@gmail.com> 8811Date: Fri Jul 18 13:39:59 2008 +1000 8812 8813 nv50: punt vertex emission macro out to header 8814 8815commit 4ad74cb03070f5279e0b1de2ed326f1167400d60 8816Author: Ben Skeggs <skeggsb@gmail.com> 8817Date: Thu Jul 17 11:15:59 2008 +1000 8818 8819 exa: missed a reloc delta 8820 8821 Can't actually hit this bug currently, but one day perhaps :) 8822 8823commit 98a9056224a86244f99ef7e844cda363e563a446 8824Author: Ben Skeggs <skeggsb@gmail.com> 8825Date: Thu Jul 17 11:15:08 2008 +1000 8826 8827 nv50: remove acquire/release surface stuff 8828 8829 It's left over from code that never made it to git in the first place, 8830 long overdue removal. 8831 8832commit f0305a29c2b4010894362d17f21cdfec39cd7b92 8833Author: Ben Skeggs <skeggsb@gmail.com> 8834Date: Thu Jul 17 11:14:02 2008 +1000 8835 8836 exa: rename m2mf locals to prevent some confusion 8837 8838commit 140c36ed936f4f28842af88efdd8b824c01bb097 8839Author: Danny <obiwan@mailmij.org> 8840Date: Fri Jul 18 03:07:38 2008 +0100 8841 8842 randr12: fix dithering output property endian issue (#16624) 8843 8844 tweaked and commented by Stuart Bennett 8845 8846commit b36802fe90165cf0fea61728b06605be5ac25d8d 8847Author: Pawel Worach <pawel.worach@gmail.com> 8848Date: Fri Jul 18 01:17:18 2008 +0200 8849 8850 Fix for #14858 - freebsd compile. Someone still needs to fix the kernel part though. 8851 8852commit 6f5e90ae739c64478ddf3e042e2bc9c4ff2b951c 8853Author: Ben Skeggs <skeggsb@gmail.com> 8854Date: Sat Jul 12 02:24:54 2008 +1000 8855 8856 nv30: more header name changes... 8857 8858commit 4b8427af583a8f84a09e8bd8d6ad614575a4f9c7 8859Author: Ben Skeggs <skeggsb@gmail.com> 8860Date: Sat Jul 12 02:15:08 2008 +1000 8861 8862 nv30: more bustage from header update 8863 8864commit 1672a783c4af58c80ed549fb3dc63c07fc75a882 8865Author: Ben Skeggs <skeggsb@gmail.com> 8866Date: Sat Jul 12 02:03:58 2008 +1000 8867 8868 nv30: un-change something that somehow changed in an earlier commit. oops! 8869 8870commit dae59585abec4b514789a8e580d30757f10d0357 8871Author: Ben Skeggs <skeggsb@gmail.com> 8872Date: Tue Jul 8 12:28:18 2008 +1000 8873 8874 nv50: exa/xv share some stuff, punt it out to nv50_accel.[ch] 8875 8876commit 17fbd810748b61b298c1ea9729d94f7d1177253b 8877Author: Ben Skeggs <skeggsb@gmail.com> 8878Date: Tue Jul 8 12:09:09 2008 +1000 8879 8880 nv50: Xv support 8881 8882 The NV12 YCbCr->RGB shader is hand-written just like the composite shaders, 8883 hopefully correct! I'll probably put a disasm next to each opcode at some 8884 point so it doesn't look like just a bunch of magics. 8885 8886 You'll need to use a composite manager if you want to watch videos, we can't 8887 render directly to the screen for the same reasons as mentioned in the 8888 composite commit. Though, a simple workaround is possible in the Xv case, 8889 so that'll probably come at some point. For now, PutImage() will fail with 8890 BadMatch. 8891 8892 The Y and UV plane upload may not be 100% for what NV50 requires, I want to 8893 rework all that at some point anyway. For now, at least there's some form 8894 of accelerated video to work with. 8895 8896 Enjoy :) 8897 8898commit ec452786d9aa3b6db41f1ab3c89aaa20a758f89e 8899Author: Ben Skeggs <skeggsb@gmail.com> 8900Date: Sun Jul 6 20:38:44 2008 +1000 8901 8902 nv50: de-magic things before someone accuses me of being an NVIDIA employee 8903 8904commit ad56c5e27642674d8e435bbbe3d6030097a01ff9 8905Author: Ben Skeggs <skeggsb@gmail.com> 8906Date: Sun Jul 6 01:30:21 2008 +1000 8907 8908 nv50: exa composite 8909 8910commit 33fad279e935a5a134ee5cc1777b8ed7510101d0 8911Author: Ben Skeggs <skeggsb@gmail.com> 8912Date: Sun Jul 6 01:21:39 2008 +1000 8913 8914 nv50: tile offscreen pixmaps 8915 8916 Until someone figures out how to make the display engine scan out from 8917 tiled surfaces things will be a bit painful. Future commits for exa 8918 composite and Xv will require the use of the 3D engine, which will *not* 8919 render to linear surfaces.. a composite manager will probably be a must 8920 for it to be usable. 8921 8922 I really don't want the NOUVEAU_MEM_TILE ioctl in the end, but we'll have 8923 to deal with it until we can drop support for <xorg-server-1.5. EXA driver 8924 handled pixmaps are far far nicer :) 8925 8926commit 5c1deac8384413dc483238f60c366e98253c11d6 8927Author: Ben Skeggs <skeggsb@gmail.com> 8928Date: Sun Jul 6 00:57:08 2008 +1000 8929 8930 exa: pass window coords to m2mf-based UTS/DFS 8931 8932commit b1f316977e2c79fdf5c83092e953bdfc988299e1 8933Author: Ben Skeggs <skeggsb@gmail.com> 8934Date: Mon Jul 7 01:03:02 2008 +1000 8935 8936 xv: remove some disturbing abuses of buffer related things 8937 8938 Also brings to attention a point about the overlay when we use a real 8939 memory manager. We'll need to find a way for the overlay to signal it's 8940 finished with a buffer, or just resort to pinning pPriv->video_mem.. 8941 8942commit 0ce028b6aaa6c20a1578148add2a96b2d37d9108 8943Author: Maarten Maathuis <madman2003@gmail.com> 8944Date: Sun Jul 6 13:45:32 2008 +0200 8945 8946 EXA: add a pixmapIsOffscreen hook. 8947 8948commit 86c70df254e63ebc5c405803d6faa9ae33df4995 8949Author: Maarten Maathuis <madman2003@gmail.com> 8950Date: Sun Jul 6 12:52:16 2008 +0200 8951 8952 NV50_KMS: support gamma changes 8953 8954commit 6de89c8cb219cc82b46e0d0d723fa9d1cefe4ac8 8955Author: Maarten Maathuis <madman2003@gmail.com> 8956Date: Sun Jul 6 12:29:13 2008 +0200 8957 8958 NV50: Unaccelerated rotation support. 8959 8960 - This will need a PixmapIsOffscreen hook before it will be considered for acceleration. 8961 8962commit 788fefa2e78a4abf7bd5c396ab9639bc6ff78451 8963Author: Maarten Maathuis <madman2003@gmail.com> 8964Date: Sun Jul 6 01:03:44 2008 +0200 8965 8966 NV50_KMS: Unaccelerated rotation support. 8967 8968 - This will need a PixmapIsOffscreen hook for it to be ever considered for acceleration. 8969 8970commit 2bc1cb832e0ded6cd1411757bec2a4b03536b7ea 8971Author: Maarten Maathuis <madman2003@gmail.com> 8972Date: Sun Jul 6 01:02:45 2008 +0200 8973 8974 Revert "EXA: Add PixmapIsOffscreen hook for EXA versions that allow it." 8975 8976 This reverts commit 5a5aee10f84fbea70eee69ece17ddf6e797497a6. 8977 This was never supposed to end up here as it's broken. 8978 8979commit 5a5aee10f84fbea70eee69ece17ddf6e797497a6 8980Author: Maarten Maathuis <madman2003@gmail.com> 8981Date: Sat Jul 5 23:07:58 2008 +0200 8982 8983 EXA: Add PixmapIsOffscreen hook for EXA versions that allow it. 8984 8985 - This should allow more flexibility for rotation shadow buffers. 8986 8987commit b9f2358ac35caedbba0339649ef13b03f209f2a7 8988Author: Ben Skeggs <skeggsb@gmail.com> 8989Date: Sun Jul 6 03:50:03 2008 +1000 8990 8991 xv: remove some defines that are in a header now 8992 8993commit 9f53b2a341a81e1bff87fee8bf5c2754cac17849 8994Author: Ben Skeggs <skeggsb@gmail.com> 8995Date: Sun Jul 6 02:28:26 2008 +1000 8996 8997 xv: remove drawable handling from hw-specific PutImage.. it's common... 8998 8999commit a50bba662ddfe49f1b0a7f8faa3668f5e810811f 9000Author: Ben Skeggs <skeggsb@gmail.com> 9001Date: Sun Jul 6 01:57:17 2008 +1000 9002 9003 xv: misc cleanups 9004 9005commit 08e9489b79395b234282b0c03d94601ce57678b8 9006Author: Maarten Maathuis <madman2003@gmail.com> 9007Date: Sat Jul 5 12:47:48 2008 +0200 9008 9009 NV50_KMS: minor change 9010 9011commit ea7e85b493ff0f684f9ccad92f84348cd835719b 9012Author: Maarten Maathuis <madman2003@gmail.com> 9013Date: Fri Jul 4 23:05:20 2008 +0200 9014 9015 Fix some compile warnings. 9016 9017commit 3298249d8267c27235653f17379f8c9f18e4f344 9018Author: Ben Skeggs <skeggsb@nisroch.keine.ath.cx> 9019Date: Thu Jul 3 23:54:13 2008 +1000 9020 9021 xv: reformat source so mere mortals can understand it 9022 9023commit a1e8ccdf85c5a0177df38a8a2d8c99d60890ec96 9024Author: Ben Skeggs <skeggsb@nisroch.(none)> 9025Date: Thu Jul 3 17:07:45 2008 +1000 9026 9027 xv: nv_video.c to nouveau_xv.c 9028 9029commit 3e17781625c0ef0ec34ee23a6e999de7f1fe4524 9030Author: Maarten Maathuis <madman2003@gmail.com> 9031Date: Thu Jul 3 08:30:43 2008 +0200 9032 9033 NV50_KMS: autodetect kernel modesetting when active 9034 9035commit 48c2116fb31872111239f5434215f30c8707091e 9036Author: Maarten Maathuis <madman2003@gmail.com> 9037Date: Wed Jul 2 19:04:58 2008 +0200 9038 9039 NV50_KMS: no int10 please 9040 9041commit 11102af2896b3fdf2364854b2d7939b2802c5b2a 9042Author: Stuart Bennett <sb476@cam.ac.uk> 9043Date: Fri Jun 27 16:21:19 2008 +0100 9044 9045 Backport console font save/restore from nouveau_ms 9046 9047commit ca2c55dfbf65ac268d43ba62db2399976eaf8f0a 9048Author: Stuart Bennett <sb476@cam.ac.uk> 9049Date: Thu Jun 26 14:44:45 2008 +0100 9050 9051 randr12: fix mode detection for when screen size changes 9052 9053 also some mem leaks 9054 9055commit 9c6eb667aa48f87f6bbbdae10cc9f4d237bdb0c6 9056Author: Maarten Maathuis <madman2003@gmail.com> 9057Date: Sat Jun 28 15:47:18 2008 +0200 9058 9059 NV04-NV4E: Save and restore FB_START 9060 9061commit 82fe9a21596ab80157df20b269ce8df41315de83 9062Author: Maarten Maathuis <madman2003@gmail.com> 9063Date: Fri Jun 27 18:49:16 2008 +0200 9064 9065 NV50_KMS: switch to output dpms 9066 9067commit 13ce1f4c1d02e70333a192db2efefb3f6a6765e3 9068Author: Maarten Maathuis <madman2003@gmail.com> 9069Date: Wed Jun 25 19:45:12 2008 +0200 9070 9071 Properly close drm when exiting + misc cleanup. 9072 9073commit 7cb0eed603ea0bd196a7471e43d60fa5e518c7fb 9074Author: Maarten Maathuis <madman2003@gmail.com> 9075Date: Wed Jun 25 19:07:58 2008 +0200 9076 9077 Fixup and add some license statements. 9078 9079 - The changed statements are a response to nv commit 2fdcda8ea62eba1b7885f0a75249884715936247. 9080 9081commit 27c7c88431cf368c3d3a953daf03ce3f9f15bc3a 9082Author: Ben Skeggs <skeggsb@gmail.com> 9083Date: Wed Jun 25 04:01:21 2008 +1000 9084 9085 change hack to work with latest drm changes.. 9086 9087commit 99a4b686d2b1f8e6768b8cd1cc3fcb6cea2bf6d3 9088Author: Maarten Maathuis <madman2003@gmail.com> 9089Date: Tue Jun 24 00:02:25 2008 +0200 9090 9091 NV50: unbreak kms and implement a basic off mode/dpms 9092 9093commit 1a23dbd1878909709fc130e00f6171508dcd41bf 9094Author: Ben Skeggs <skeggsb@nisroch.keine.ath.cx> 9095Date: Tue Jun 24 02:24:42 2008 +1000 9096 9097 nv50: stub out NV50SorSetClockMode for LVDS, it doesn't work there 9098 9099commit 2ccc80bd87a76a9f0498cfec531e8a46cbd2c384 9100Author: Ben Skeggs <skeggsb@gmail.com> 9101Date: Mon Jun 23 02:10:09 2008 +1000 9102 9103 nv50: bring in int10-constole-restore hack from nv driver 9104 9105 Lack of console restore has been annoying me personally lately while working 9106 on 3D, I lack the time/modsetting skills to fix it properly at the moment, 9107 so this'll do. 9108 9109commit 73b4fcc20334ac6cd2076c658a4937f9a98ee06d 9110Author: Ben Skeggs <skeggsb@gmail.com> 9111Date: Mon Jun 23 01:47:55 2008 +1000 9112 9113 fix non-kms build/run 9114 9115commit 75d8947d40e19c5fc039caf4289cf94fd1570eda 9116Author: Maarten Maathuis <madman2003@gmail.com> 9117Date: Sun Jun 22 16:34:22 2008 +0200 9118 9119 NV50: Some basic code to get kernel modesetting going. 9120 9121 - This will get a cleanup later on. 9122 9123commit df52dc4664df11d57fcf8a6c9ae3f87a149e5392 9124Author: Stuart Bennett <sb476@cam.ac.uk> 9125Date: Fri Jun 13 02:11:09 2008 +0100 9126 9127 No need for G80+ LVDS conf bits reports 9128 9129 I've got a few already. 9130 9131commit 1813559a20f0786b4decdb645f7ba3f52bc30739 9132Author: Stuart Bennett <sb476@cam.ac.uk> 9133Date: Fri Jun 13 01:58:41 2008 +0100 9134 9135 More DCB 1.5 TV variants (#16252 and #16342) 9136 9137commit 29c4a58bb2ba476f9558bda31f1082b4685555fc 9138Author: Stuart Bennett <sb476@cam.ac.uk> 9139Date: Fri Jun 13 01:40:31 2008 +0100 9140 9141 Be a little more discerning about the BIT bios signature 9142 9143 Thank you ABIT 9144 9145commit e8c790721d99e171c57b8dd497f5b3e82126c78b 9146Author: Stuart Bennett <sb476@cam.ac.uk> 9147Date: Fri Jun 13 01:25:49 2008 +0100 9148 9149 LVDS_INIT should not always use head A 9150 9151 While the script doesn't need the head variable set correctly, 9152 link_head_and_output still gets called and will therefore break panels expecting 9153 to be on head B 9154 Fixes Apple NV34 console restore 9155 9156commit fc137f23bed7405221b7f483d9151e296b99caf0 9157Author: Maarten Maathuis <madman2003@gmail.com> 9158Date: Fri Jun 13 19:12:38 2008 +0200 9159 9160 NV50: lvds is probably still broken, but it's a step in the right direction 9161 9162 - Just something i had lingering around. 9163 - Note that lvds still doesn't run by default, but you're more than welcome to fix it ;-) 9164 9165commit e034616ab2bbbc486f90c5f64e2bd2de194cf92e 9166Author: Maarten Maathuis <madman2003@gmail.com> 9167Date: Thu May 22 21:08:48 2008 +0200 9168 9169 NV50: A small extra piece of init that may prove useful to someone, somewhere, at some point for something. 9170 9171commit c21594cf452a1263682386d04718531d76a7a9b8 9172Author: Maarten Maathuis <madman2003@gmail.com> 9173Date: Tue May 20 19:19:53 2008 +0200 9174 9175 NV50: Be consistent with rules-ng. 9176 9177commit 9c1d87fa9e6152bffc5c5cdf16b1931cc6e58399 9178Author: Maarten Maathuis <madman2003@gmail.com> 9179Date: Sat May 10 21:58:00 2008 +0200 9180 9181 NV50: Some misc things. 9182 9183commit 5f08db9057f2970171186b20a25d346386dcb0f2 9184Author: Stuart Bennett <sb476@cam.ac.uk> 9185Date: Thu May 8 21:36:05 2008 +0100 9186 9187 Cosmetic g80 laptop dcb parsing enhancement 9188 9189commit 6a460eae941f3e9a7a8a0297efbbd12dca71a9dd 9190Author: Stuart Bennett <sb476@cam.ac.uk> 9191Date: Thu May 8 14:17:38 2008 +0100 9192 9193 randr12: some reordering, tidying &c 9194 9195commit 7455bc89116bfc5e88fbd444d83c1415a6020a7f 9196Author: Stuart Bennett <sb476@cam.ac.uk> 9197Date: Tue Apr 29 19:42:54 2008 +0100 9198 9199 Enable RandR 1.2 code-path by default 9200 9201commit 98a751e773c098382e7675e41ee60ce3d9a67ce0 9202Author: Stuart Bennett <sb476@cam.ac.uk> 9203Date: Wed May 7 21:25:03 2008 +0100 9204 9205 Don't call call_lvds_script with a null dcbent pointer 9206 9207 Stumbled over by hiyuh :) 9208 9209commit 77b24bc93659471bb20254847da124fdaba13a54 9210Author: Stuart Bennett <sb476@cam.ac.uk> 9211Date: Mon May 5 00:56:28 2008 +0100 9212 9213 randr12: tidy up some nv50 remnants in the pre nv50 code 9214 9215commit 8997f83cd2727b76240e9dc31ec7751f6d69cf6b 9216Author: Maarten Maathuis <madman2003@gmail.com> 9217Date: Tue May 6 19:46:52 2008 +0200 9218 9219 NV50: Some misc fixes. 9220 9221 - Even though tv-out is mentioned, it by no means implies that is (nearly) functional. 9222 9223commit ae574ec7d930d8708e4dd9d1d47558666019afe5 9224Author: Maarten Maathuis <madman2003@gmail.com> 9225Date: Mon May 5 11:02:07 2008 +0200 9226 9227 nv50_exa.c had no license statement. 9228 9229commit a7af057ff67cf5d7baf78d852dee5ac69b24e706 9230Author: Maarten Maathuis <madman2003@gmail.com> 9231Date: Mon May 5 01:26:37 2008 +0200 9232 9233 NV50: Rework the modesetting code into a different model. 9234 9235 - There is now interface independent code for most things. 9236 - This structure, although not exclusive, is only used for NV50 atm. 9237 - The change in approach is that the concept of a connector is added to the model (randr12 sucks in this aspect). 9238 - I'm hacking around randr trying to create a single output for each connector, the indices are the bus numbers. 9239 - Limitations: Hot swapping vga and dvi of same monitor will not make randr12 think it changed, so for the moment force a change. 9240 - Concept is largely inspired by the radeonhd driver. 9241 9242commit 5e1b5708d3a7e4298f80b1a1b8bb3fafae0c69bd 9243Author: Stuart Bennett <sb476@cam.ac.uk> 9244Date: Fri May 2 20:36:59 2008 +0100 9245 9246 xv: nv10 double-buffered overlay mode was getting set on nv04 9247 9248commit 85536023ed7050632d121299ab99382e096b9984 9249Author: Stuart Bennett <sb476@cam.ac.uk> 9250Date: Wed Apr 30 00:59:47 2008 +0100 9251 9252 Use the actual number of BIT table entries 9253 9254commit 78588c71ffd19e18602b95a55186bc53eebb51c3 9255Author: Stuart Bennett <sb476@cam.ac.uk> 9256Date: Mon Apr 28 16:51:01 2008 +0100 9257 9258 randr12, xv: manual overlay clipping for randr12 (see #12825) 9259 9260 There could be something to be said for doing this clipping when using the blitter and texture adapters too 9261 9262commit 6f7d00a0cb4ec85b058afb0150a122a124542b5f 9263Author: Stuart Bennett <sb476@cam.ac.uk> 9264Date: Tue Apr 29 22:53:40 2008 +0100 9265 9266 randr12: remove unneeded separate dcb_entry and "or" members 9267 9268 Make off-chip outputs clearer too 9269 9270commit 52e58c7e799697989fcfbf95050ce10a4c3d1f8f 9271Author: Stuart Bennett <sb476@cam.ac.uk> 9272Date: Tue Apr 29 22:08:01 2008 +0100 9273 9274 Pass DCB entry structs in bios functions, rather than index 9275 9276commit 0c1da6907d8e29ed1c058f9655d436c228ea91b6 9277Author: Stuart Bennett <sb476@cam.ac.uk> 9278Date: Tue Apr 29 21:53:22 2008 +0100 9279 9280 Pass output field, rather than dcb indices, for tmds access functions 9281 9282commit 2333bc907a4be3c7e36bdd02f27e22b63a7ee7c3 9283Author: Stuart Bennett <sb476@cam.ac.uk> 9284Date: Tue Apr 29 21:58:17 2008 +0100 9285 9286 Add an index field to dcb entries, and carry pointer to dcb entry in output struct 9287 9288commit 21f062cad184f21d33d7f5b43f86937979291b58 9289Author: Stuart Bennett <sb476@cam.ac.uk> 9290Date: Tue Apr 29 21:27:07 2008 +0100 9291 9292 randr12, xv: oops, transform_in_use doesn't exist on xserver 1.3 9293 9294commit ee78dc0abef626e29f7617a4b324e847bab1abcc 9295Author: Stuart Bennett <sb476@cam.ac.uk> 9296Date: Mon Apr 28 02:12:17 2008 +0100 9297 9298 randr12: C51 I2C 9299 9300commit 9b70dfaf4f0317e622ca64fc0804f74a31158108 9301Author: Stuart Bennett <sb476@cam.ac.uk> 9302Date: Tue Apr 29 18:46:24 2008 +0100 9303 9304 randr12: only set cr59 to 1 for tmds (fixes nv34 issue) 9305 9306commit c8cec9b2a0b18cd1159613d4dd218725695aef7c 9307Author: Stuart Bennett <sb476@cam.ac.uk> 9308Date: Mon Apr 28 17:00:27 2008 +0100 9309 9310 randr12, xv: no overlay when the crtc is transformed 9311 9312commit 3233ebba07e7adab347dc3c6f9d17f02e1a9d32b 9313Author: Stuart Bennett <sb476@cam.ac.uk> 9314Date: Sat Apr 26 02:08:24 2008 +0100 9315 9316 randr12: fix spread spectrum setting for lvds and turn it off during dpms off 9317 9318 Assuming that the bios has set this up correctly for LVDS initially is probably safe 9319 9320commit 467f83aa43bffdf545b513fdb7b11b4fe2205b4f 9321Author: Stuart Bennett <sb476@cam.ac.uk> 9322Date: Sat Apr 26 19:59:58 2008 +0100 9323 9324 randr12: make nv50 code depend on Architecture 9325 9326commit 8b165722d9f9461353abf84b4136e33b816ac842 9327Author: Stuart Bennett <sb476@cam.ac.uk> 9328Date: Sat Apr 26 01:21:50 2008 +0100 9329 9330 CR27 appears to contain the chip revision 9331 9332commit 59ff20738e88d6621f9beaea94810274029a21a3 9333Author: Maarten Maathuis <madman2003@gmail.com> 9334Date: Sun Apr 27 19:52:41 2008 +0200 9335 9336 NV50: A little bit more information, no functional change. 9337 9338commit bb5a2a105f264502d0985ce70e76d5a3d3b6463c 9339Author: Stuart Bennett <sb476@cam.ac.uk> 9340Date: Fri Apr 25 22:09:46 2008 +0100 9341 9342 randr12: sel_clk binding fix 9343 9344commit 5a85626cccbba80ef3d6fb2fb90ae609b7abde15 9345Author: Stuart Bennett <sb476@cam.ac.uk> 9346Date: Fri Apr 25 02:09:18 2008 +0100 9347 9348 randr12: unified mode validation 9349 9350 doublescan/interlace limits applied too 9351 9352commit 38ad350118aa9592bdbcb530fe63b85983fb9f64 9353Author: Stuart Bennett <sb476@cam.ac.uk> 9354Date: Fri Apr 25 00:03:41 2008 +0100 9355 9356 randr12: only set up one spread spectrum clock on pre nv40 9357 9358 avoids inadvertently getting wobbly crt pictures 9359 9360commit dfd8b42f91257f2cb0762b3d9b07612a7edac549 9361Author: Stuart Bennett <sb476@cam.ac.uk> 9362Date: Thu Apr 24 22:23:50 2008 +0100 9363 9364 randr12: deny doublescan modes to digital outputs 9365 9366commit d150ce5c633c90a82411b4b8e44fa4b91647e6a8 9367Author: Stuart Bennett <sb476@cam.ac.uk> 9368Date: Thu Apr 24 01:22:49 2008 +0100 9369 9370 randr12: off-chip dvi dpms and fp_control dpms fixed properly 9371 9372 Also solves output exclusion problem: two crtcs claiming the same output 9373 through cr33 does not work 9374 9375commit dd152ce186fea953968805cf4a1b309a1fd3987b 9376Author: Stuart Bennett <sb476@cam.ac.uk> 9377Date: Thu Apr 24 00:24:35 2008 +0100 9378 9379 randr12: set CRTCX_LCD in nv_output 9380 9381 This makes my nv28 external dvi work. 9382 9383commit b8d401b0a56d23db7199147e6a5dee1edccff4ff 9384Author: Stuart Bennett <sb476@cam.ac.uk> 9385Date: Thu Apr 24 00:01:50 2008 +0100 9386 9387 These made up defines are more misleading than helpful 9388 9389commit fa8c8437798cfc5494a639d649f0f91f0e7275ce 9390Author: Stuart Bennett <sb476@cam.ac.uk> 9391Date: Tue Apr 22 22:42:12 2008 +0100 9392 9393 randr12: remove FPP setting and other FSEL changes 9394 9395commit a083bb1a55d0983991c45d113eb7e9814369fff4 9396Author: Stuart Bennett <sb476@cam.ac.uk> 9397Date: Tue Apr 22 23:25:40 2008 +0100 9398 9399 randr12: oops (fix to 9edf6d60) 9400 9401commit 16f420e339f0dd6d55d7b87763d97c2eb8e77e66 9402Author: Stuart Bennett <sb476@cam.ac.uk> 9403Date: Tue Apr 22 21:34:52 2008 +0100 9404 9405 randr12: RAMDAC_OUTPUT is only for analogue outputs 9406 9407commit 52174047f3f817fa9ce3674440a13687b73ccea8 9408Author: Stuart Bennett <sb476@cam.ac.uk> 9409Date: Tue Apr 22 21:29:48 2008 +0100 9410 9411 Revert "randr12: correct SEL_CLK register beheaviour for outputs with or=2" 9412 9413 This reverts commit 77af887089be6fef7b793e3c3afdb98ac90a2e8c. 9414 9415 External DVI (which is source of or == 2) do not use SEL_CLK 9416 9417commit 61e416153b24516951962aca407ceaa9b7885e22 9418Author: Maarten Maathuis <madman2003@gmail.com> 9419Date: Sun Apr 13 02:40:55 2008 +0200 9420 9421 NV50: Another few regs. 9422 9423commit 5328370d2f7e10c61485572f53031deace69c515 9424Author: Maarten Maathuis <madman2003@gmail.com> 9425Date: Sun Apr 13 02:16:27 2008 +0200 9426 9427 NV50: Some more regs. 9428 9429commit e5511a564564350bd8755b7f18a1448ea7c796de 9430Author: Maarten Maathuis <madman2003@gmail.com> 9431Date: Sun Apr 13 01:21:20 2008 +0200 9432 9433 NV50: I now know what UNK8A8 is, so update the value register as well. 9434 9435commit 2c9027679db1657e6fe02cf6674024ca97f4cd6e 9436Author: Maarten Maathuis <madman2003@gmail.com> 9437Date: Sat Apr 12 02:11:32 2008 +0200 9438 9439 NV50: Re-implement output status caching. 9440 9441commit d7681e8ad70c07ec963ac9425b868a2686d79769 9442Author: Stuart Bennett <sb476@cam.ac.uk> 9443Date: Fri Apr 11 17:28:16 2008 +0100 9444 9445 randr12: separate pll and arbitration stages of calc_state_ext, and call from mode_set directly 9446 9447commit 0bf08d7699d7c3a1748060febded1c8b3950c9bd 9448Author: Stuart Bennett <sb476@cam.ac.uk> 9449Date: Fri Apr 11 16:20:21 2008 +0100 9450 9451 state->config is not used 9452 9453commit 08b8bd8a97b580a57d7bb75b3da7827d65c046f5 9454Author: Stuart Bennett <sb476@cam.ac.uk> 9455Date: Fri Apr 4 18:59:29 2008 +0100 9456 9457 randr12: unite pre nv40 and nv40 pll restore 9458 9459commit 5b443084ec7b03230d0ec7490c5a3378a9f7187e 9460Author: Stuart Bennett <sb476@cam.ac.uk> 9461Date: Tue Apr 8 02:25:34 2008 +0100 9462 9463 Use the correct reference clock when reading (V)PLLs 9464 9465commit 0f533b8cc12f49f5d5ac017838f7330a4f8b47ea 9466Author: Stuart Bennett <sb476@cam.ac.uk> 9467Date: Tue Apr 8 02:25:22 2008 +0100 9468 9469 Various PLL code improvements 9470 9471commit 4e1caa6edfe9499e6c501e70a8fba14bae510462 9472Author: Maarten Maathuis <madman2003@gmail.com> 9473Date: Wed Apr 9 18:55:06 2008 +0200 9474 9475 NV50: Some misc stuff. 9476 9477commit 099128175ad31956fd1c6622166b65f4ab1fe370 9478Author: Dave Airlie <airlied@redhat.com> 9479Date: Tue Apr 8 16:48:37 2008 +1000 9480 9481 nv50: fix distcheck since header removal 9482 9483commit 4af376fa6031bd41dc247e950319f3a151c49f22 9484Author: Stuart Bennett <sb476@cam.ac.uk> 9485Date: Tue Apr 8 00:28:03 2008 +0100 9486 9487 MCP67 does not do the second stage gain tests 9488 9489commit dbca8b65d7b12a1dd2eb7aee3ac37ece0bd4ce65 9490Author: Stuart Bennett <sb476@cam.ac.uk> 9491Date: Tue Apr 8 01:31:14 2008 +0100 9492 9493 PLL stuff 9494 9495commit 39b2e0b842f4a4e1c3de96c94e380aa9a0327c12 9496Author: Stuart Bennett <sb476@cam.ac.uk> 9497Date: Sat Apr 5 05:20:54 2008 +0100 9498 9499 If no PLL coefficients compatible with the constraints exist, don't brick the hardware 9500 9501commit 0b54450e1db7c6411528bc4a3bac4d1a42510aca 9502Author: Maarten Maathuis <madman2003@gmail.com> 9503Date: Mon Apr 7 22:12:44 2008 +0200 9504 9505 NV50: Scaling works fine on VGA connectors, so support that as well. 9506 9507 - The default is SCALE_PANEL instead of SCALE_ASPECT for obvious reasons. 9508 9509commit 891e72f4316c4e7e1b9d06f39e075b1a8a8a8c53 9510Author: Maarten Maathuis <madman2003@gmail.com> 9511Date: Mon Apr 7 21:18:54 2008 +0200 9512 9513 NV50: Deobfuscate some more commands. 9514 9515commit 13c701f9d23a598c281ced726070a219e1b78607 9516Author: Maarten Maathuis <madman2003@gmail.com> 9517Date: Mon Apr 7 19:29:55 2008 +0200 9518 9519 Make some CARD32 die. 9520 9521commit a3a0710604017f3ceb9c552552d71c3f5dcfab0f 9522Author: Maarten Maathuis <madman2003@gmail.com> 9523Date: Mon Apr 7 19:09:35 2008 +0200 9524 9525 NV50: Give a few registers an UNK label. 9526 9527commit be72a7c9104c1c1d6db960b1d80cc4d820d37440 9528Author: Maarten Maathuis <madman2003@gmail.com> 9529Date: Mon Apr 7 18:35:18 2008 +0200 9530 9531 randr12: NV04/05/06 can't accelerate rotation, so it's better to disable it. 9532 9533commit 9cf76c91a661587d9c54845e6a06fb25cf9b55a2 9534Author: Maarten Maathuis <madman2003@gmail.com> 9535Date: Sun Apr 6 23:36:15 2008 +0200 9536 9537 NV50: Some more stuff. 9538 9539commit c725e33078a2fed374fbb52175beba92b0b9fbe8 9540Author: Maarten Maathuis <madman2003@gmail.com> 9541Date: Sun Apr 6 19:49:11 2008 +0200 9542 9543 NV50: minor rename to avoid confusion 9544 9545commit fcda539ce60ca02f2f4589157f8b8c522430571f 9546Author: Maarten Maathuis <madman2003@gmail.com> 9547Date: Sun Apr 6 15:16:40 2008 +0200 9548 9549 NV50: Deobfuscate another register (partially). 9550 9551commit fa93541b257a8055ea8b55c883475d2db717d971 9552Author: Maarten Maathuis <madman2003@gmail.com> 9553Date: Sun Apr 6 15:05:33 2008 +0200 9554 9555 NV50: Fix another deadlock in the clock setting code. 9556 9557 - Swapping my monitor from dvi to vga causes no mode (=clock) change, but still the supervisor expects a clock change. 9558 - This meant it got stuck in stage 2, with no clock that is supposed to change. 9559 9560commit c001fea3b0ae3921c063e7f44c298cafdfef51f6 9561Author: Maarten Maathuis <madman2003@gmail.com> 9562Date: Sat Apr 5 23:33:22 2008 +0200 9563 9564 Forgot a few things. 9565 9566commit 17a8e1743478e5548a90caf8f84a387366020416 9567Author: Maarten Maathuis <madman2003@gmail.com> 9568Date: Sat Apr 5 23:23:22 2008 +0200 9569 9570 NV50: rename randr-1.2 functions to be nv50_something_function 9571 9572commit a71ef4da76e33e84303a8b9181dbcfd655e78aa9 9573Author: Maarten Maathuis <madman2003@gmail.com> 9574Date: Sat Apr 5 23:00:39 2008 +0200 9575 9576 NV50: Some more cleanup. 9577 9578commit aeab563cc89b2fc3cd6cb306aab880b6e28f2b64 9579Author: Maarten Maathuis <madman2003@gmail.com> 9580Date: Sat Apr 5 22:52:58 2008 +0200 9581 9582 NV50: Some reordering. 9583 9584commit b899ab75781a3076898791381604be575c6d8a70 9585Author: Maarten Maathuis <madman2003@gmail.com> 9586Date: Sat Apr 5 21:17:23 2008 +0200 9587 9588 Some minor corrections. 9589 9590commit b0ef158a7e281329825c7b6c10cab6097e627b54 9591Author: Maarten Maathuis <madman2003@gmail.com> 9592Date: Sat Apr 5 21:06:03 2008 +0200 9593 9594 NV50: Document dvi hotplug detection. 9595 9596 - This can only be used once modesetting is moved to the kernel. 9597 9598commit be54b9299b4a261fc707b74407cf4d847fcbb243 9599Author: Maarten Maathuis <madman2003@gmail.com> 9600Date: Sat Apr 5 16:16:26 2008 +0200 9601 9602 NV50: Remove some questionable guesses. 9603 9604commit c06f202cfbee8568fb8358ea1caffec41c71e648 9605Author: Maarten Maathuis <madman2003@gmail.com> 9606Date: Sat Apr 5 01:54:06 2008 +0200 9607 9608 nvbios: Drop a N2/M2 requirement on NV5x hardware, as it prevents getting a mode on 1024x768'ish and lower. 9609 9610 - Limitations are very strict in nv5x bios tables and N2/M2 can drop below 2 (the blob does this too). 9611 9612commit 5ce26b9aec214dd99e5399fe0f864d8faad15d2c 9613Author: Maarten Maathuis <madman2003@gmail.com> 9614Date: Fri Apr 4 22:49:06 2008 +0200 9615 9616 nvbios: Add some valid register ranges for nv50. 9617 9618commit ce86484979d873325ec4c74ed84107be9ad9ba62 9619Author: Maarten Maathuis <madman2003@gmail.com> 9620Date: Fri Apr 4 22:38:30 2008 +0200 9621 9622 NV50: Switch to common pll calculator function. 9623 9624commit 821ad3d643f9564d043c7f3909e0b8be28e94965 9625Author: Stuart Bennett <sb476@cam.ac.uk> 9626Date: Wed Apr 2 16:58:45 2008 +0100 9627 9628 randr12: delete stuff commented out a week ago that no one's complained about 9629 9630commit e90e534728e9cbe65ee6c70faf596db449495665 9631Author: Maarten Maathuis <madman2003@gmail.com> 9632Date: Fri Apr 4 17:13:25 2008 +0200 9633 9634 NV50: Minor comment updates. 9635 9636commit 08550984aba2bfa3eaeeddbbf6af25206819abe8 9637Author: Maarten Maathuis <madman2003@gmail.com> 9638Date: Fri Apr 4 17:04:49 2008 +0200 9639 9640 NV50: Some more stuff. 9641 9642commit ef4339aacf72bf5422b22ab18b73416f2738b284 9643Author: Maarten Maathuis <madman2003@gmail.com> 9644Date: Fri Apr 4 16:49:00 2008 +0200 9645 9646 NV50: Some more registers. 9647 9648commit 3183df1aa61f3c441a93e5a8b50db6628d804465 9649Author: Maarten Maathuis <madman2003@gmail.com> 9650Date: Thu Apr 3 22:42:31 2008 +0200 9651 9652 NV50: Update my knowledge of interrupts. 9653 9654commit 04fd80d29913ed02524de177fa7c87b616147d15 9655Author: Stuart Bennett <sb476@cam.ac.uk> 9656Date: Thu Apr 3 03:00:13 2008 +0100 9657 9658 Small declaration reordering 9659 9660commit 92934e4b7571a578386327ebdb59eb6d10992763 9661Author: Stuart Bennett <sb476@cam.ac.uk> 9662Date: Thu Apr 3 01:03:38 2008 +0100 9663 9664 Move some variable declarations in the bios parser 9665 9666commit 927384005a01a606835452c9d1e26d7df98f81cb 9667Author: Stuart Bennett <sb476@cam.ac.uk> 9668Date: Thu Apr 3 00:11:31 2008 +0100 9669 9670 randr12: some fixes to dithering property 9671 9672commit 52c880c97acf29d5ed8dfeec5810fcc7d8bb7b35 9673Author: Stuart Bennett <sb476@cam.ac.uk> 9674Date: Wed Apr 2 19:45:43 2008 +0100 9675 9676 randr12: better analogue clock limits (taken from pNv->{Min,Max}VClockFreqKHz) 9677 9678commit 29714e7b9ae952aecd036086c053ec030aad94c9 9679Author: Maarten Maathuis <madman2003@gmail.com> 9680Date: Wed Apr 2 21:31:07 2008 +0200 9681 9682 NV50: Some minor changes to the display supervisor (mostly comment). 9683 9684commit 9ac32d3e2bf400f6a6220118b8008804cd11e435 9685Author: Stuart Bennett <sb476@cam.ac.uk> 9686Date: Wed Apr 2 19:03:07 2008 +0100 9687 9688 Some bios parser PLL writing improvements 9689 9690commit 4a192ea6cf9df39d95368f46482c777524471118 9691Author: Maarten Maathuis <madman2003@gmail.com> 9692Date: Wed Apr 2 18:21:01 2008 +0200 9693 9694 NV50: Add my name to some files. 9695 9696commit db6220b8611829a77d354de1e45f6fe29e118b2a 9697Author: Stuart Bennett <sb476@cam.ac.uk> 9698Date: Wed Apr 2 16:16:13 2008 +0100 9699 9700 randr12: since the crtc function records are now mutable, unify pre nv50 templates 9701 9702commit b9762ba272a3ee817278e3967794c55a4fe73961 9703Author: Stuart Bennett <sb476@cam.ac.uk> 9704Date: Wed Apr 2 16:02:43 2008 +0100 9705 9706 randr12: don't provide rotation functions in NoAccel case 9707 9708commit a24ab3d1559d023270915f373527efb6eefacd8d 9709Author: Stuart Bennett <sb476@cam.ac.uk> 9710Date: Wed Apr 2 16:05:30 2008 +0100 9711 9712 Revert "randr12: Fail rotation in NoAccel case." 9713 9714 This reverts commit b33317ed98957db42691f5f9c41c119c10c91be6. 9715 9716commit 33d4c2f2fe6c6c400ce945f4c7ecc09ee08123a5 9717Author: Stuart Bennett <sb476@cam.ac.uk> 9718Date: Wed Apr 2 15:46:50 2008 +0100 9719 9720 Add quirks for Apple *book backlight 9721 9722commit 8a3f051dc9f007c3d801942d32899b8d6c09fd34 9723Author: Maarten Maathuis <madman2003@gmail.com> 9724Date: Wed Apr 2 14:54:16 2008 +0200 9725 9726 NV50: Some more things. 9727 9728commit b33317ed98957db42691f5f9c41c119c10c91be6 9729Author: Maarten Maathuis <madman2003@gmail.com> 9730Date: Wed Apr 2 09:28:45 2008 +0200 9731 9732 randr12: Fail rotation in NoAccel case. 9733 9734commit d017dfa93e7eff988339b8030af1a563347156bc 9735Author: Stuart Bennett <sb476@cam.ac.uk> 9736Date: Tue Apr 1 21:23:35 2008 +0100 9737 9738 NV30/35 setPLL bios support 9739 9740commit 32f550a534dc3e19f7cd3640549336846ad6052d 9741Author: Maarten Maathuis <madman2003@gmail.com> 9742Date: Tue Apr 1 22:33:15 2008 +0200 9743 9744 NV50: Fix a minor cursor issue. 9745 9746commit 27e819dbcd5ce36be71c7dd3be332f777a80caf4 9747Author: Maarten Maathuis <madman2003@gmail.com> 9748Date: Tue Apr 1 22:20:04 2008 +0200 9749 9750 NV50: Add some debugging statements to important functions. 9751 9752commit 99857933e8edc3be2e4af0c8c7e3fe3b1e221de9 9753Author: Maarten Maathuis <madman2003@gmail.com> 9754Date: Tue Apr 1 20:33:16 2008 +0200 9755 9756 NV50: Some i2c stuff. 9757 9758commit 9199307195c51b52979c483b3c0adf49f2301281 9759Author: Stuart Bennett <sb476@cam.ac.uk> 9760Date: Tue Apr 1 05:16:29 2008 +0100 9761 9762 Use new PLL reading code 9763 9764 Of course, having spent some time perfecting the reading of NVPLL and MPLL, 9765 it turns out to be totally unnecessary on >= nv30 :-/ 9766 9767commit a3f82d92ae6cbd779d341fe11687d12c5aac1762 9768Author: Stuart Bennett <sb476@cam.ac.uk> 9769Date: Tue Apr 1 05:03:14 2008 +0100 9770 9771 Unified PLL reading 9772 9773commit 350fb202e113bc03a71fd7e451e2c26ec990a61a 9774Author: Stuart Bennett <sb476@cam.ac.uk> 9775Date: Tue Apr 1 05:29:05 2008 +0100 9776 9777 randr12: get_clock_from_crtc only necessary for TMDS 9778 9779commit f741bf7b724dd0698f71b28c461337526f20fe31 9780Author: Stuart Bennett <sb476@cam.ac.uk> 9781Date: Tue Apr 1 00:23:06 2008 +0100 9782 9783 Better PLL limits for some nv3x 9784 9785commit 94ca1951feda10b9d6ad828816a459af750de286 9786Author: Stuart Bennett <sb476@cam.ac.uk> 9787Date: Tue Apr 1 01:00:16 2008 +0100 9788 9789 randr12: deal with nv30 and nv35 PLLs individually 9790 9791commit 626d8d62eef205dc542e3a7cd82beb329a23cdaf 9792Author: Maarten Maathuis <madman2003@gmail.com> 9793Date: Mon Mar 31 19:27:28 2008 +0200 9794 9795 Init global gamma on all crtc's, even if they are currently not in use. 9796 9797commit f54cea5662fdfc17762711a88892a27aed62f74c 9798Author: Maarten Maathuis <madman2003@gmail.com> 9799Date: Mon Mar 31 19:17:50 2008 +0200 9800 9801 NV50: Some clock related cleanups + minor changes. 9802 9803commit 7393d8fc76f7bbb8aa156587d81ee1a01c17470e 9804Author: Maarten Maathuis <madman2003@gmail.com> 9805Date: Mon Mar 31 14:10:42 2008 +0200 9806 9807 NV50: Some misc cleanup. 9808 9809commit 9edf6d60b2b6698e88edc7798285bd843bfce0f4 9810Author: Stuart Bennett <sb476@cam.ac.uk> 9811Date: Sun Mar 30 21:48:13 2008 +0100 9812 9813 randr12: different handling for sel_clk for pre and post nv40 9814 9815 This won't set the spread spectrum bits for pre nv40 from null initial state, but that's not a major concern right now 9816 9817commit 84ab4be08947ef35900da6b84702bc9acb4121bb 9818Author: Stuart Bennett <sb476@cam.ac.uk> 9819Date: Sat Mar 29 21:26:37 2008 +0000 9820 9821 Minor tidyup to output merging 9822 9823commit 1fe1d8161059afcc6f6ba152b9331e1576ef1f93 9824Author: Maarten Maathuis <madman2003@gmail.com> 9825Date: Sun Mar 30 22:42:47 2008 +0200 9826 9827 Kill a useless register offset. 9828 9829commit 75680d723144270c12a6d500d8e2c2c421e4ebc1 9830Author: Maarten Maathuis <madman2003@gmail.com> 9831Date: Sun Mar 30 15:18:09 2008 +0200 9832 9833 NV50: remove unneeded function 9834 9835commit 263ec94ec0f553f5ab6a22d3fd335bebff2c029b 9836Author: Maarten Maathuis <madman2003@gmail.com> 9837Date: Sun Mar 30 12:21:29 2008 +0200 9838 9839 NV50: Some minor cursor cleanups. 9840 9841commit 86722ad60fb939743f5ab73413f721cd26bf2c3c 9842Author: Maarten Maathuis <madman2003@gmail.com> 9843Date: Sun Mar 30 03:37:16 2008 +0200 9844 9845 NV50: Deobfuscate some of the modesetting, a few things were hidden away in innocent looking variables. 9846 9847 - Some mode variables were overridden and didn't represent their name anymore. 9848 9849commit 633a8f4f0a8ff3bcb5e6afa9bbcf3f6ed44ef905 9850Author: Stuart Bennett <sb476@cam.ac.uk> 9851Date: Sat Mar 29 19:01:30 2008 +0000 9852 9853 randr12: make sel_clk behaviour match comment (which seems to be correct) 9854 9855 If nv40 breaks, this whole thing will need another look 9856 9857commit 7a8bbb7f0890ef99de539f0f7acccd469f33dce9 9858Author: Stuart Bennett <sb476@cam.ac.uk> 9859Date: Sat Mar 29 00:03:27 2008 +0000 9860 9861 Fix Apple backlight code, broken in pci-rework conversion 9862 9863commit 4d0974c48eff8cd51af66f8fded4fad50044958a 9864Author: Stuart Bennett <sb476@cam.ac.uk> 9865Date: Fri Mar 28 01:10:31 2008 +0000 9866 9867 Consistency fix 9868 9869 + constify 9870 9871commit acd007843f00cb99cf26c4a5f885e365c3a8065b 9872Author: Maarten Maathuis <madman2003@gmail.com> 9873Date: Sat Mar 29 21:47:04 2008 +0100 9874 9875 It seems wise to NVSync for all hardware on NVLeaveVT. 9876 9877commit ea27bc38938277599228ba403a082963bf89d702 9878Author: Maarten Maathuis <madman2003@gmail.com> 9879Date: Fri Mar 28 19:56:36 2008 +0100 9880 9881 NV50: Better safe than sorry. 9882 9883commit d4ca83e2d1f1432dc7de290087f03a6ff3553d98 9884Author: Stuart Bennett <sb476@cam.ac.uk> 9885Date: Thu Mar 27 23:55:30 2008 +0000 9886 9887 randr12: shrink pitch alignment stuff and move 9888 9889commit 43ff98cd89e3dc43c281e41f3a36a76bcd3e5a65 9890Author: Stuart Bennett <sb476@cam.ac.uk> 9891Date: Wed Mar 26 18:35:14 2008 +0000 9892 9893 randr12: tidy-up 3/3 - TMDS04 twiddling 9894 9895 Code commented for easy reinstatement should problems arise 9896 9897commit 9491ee3f3fbb8b50244444424ef6148217ace944 9898Author: Stuart Bennett <sb476@cam.ac.uk> 9899Date: Wed Mar 26 18:34:28 2008 +0000 9900 9901 randr12: tidy-up 2/3 - DPMS CR57/58 twiddling 9902 9903 Code commented for easy reinstatement should problems arise 9904 9905commit 2e001c140eacc5d5f02435b9c81b13e5d40db71f 9906Author: Stuart Bennett <sb476@cam.ac.uk> 9907Date: Wed Mar 26 18:33:59 2008 +0000 9908 9909 randr12: tidy-up 1/3 - FP_DEBUG_0 twiddling 9910 9911 Code commented for easy reinstatement should problems arise 9912 9913commit ec9fa3c958d6763ff96d8b884ed36c80d911475a 9914Author: Stuart Bennett <sb476@cam.ac.uk> 9915Date: Thu Mar 27 19:24:04 2008 +0000 9916 9917 randr12: allow doublescan and low clock modes 9918 9919 320x240 ftw. Also improved mode rejection error messages 9920 9921commit 048f31014092c1dccb1944d2b1a84c7a301be48e 9922Author: Stuart Bennett <sb476@cam.ac.uk> 9923Date: Wed Mar 26 18:58:47 2008 +0000 9924 9925 Minor fix to lvds script caching commit 9926 9927commit 3feaa48b53f51c094718708e52f08b546b0b0e27 9928Author: Maarten Maathuis <madman2003@gmail.com> 9929Date: Wed Mar 26 20:01:19 2008 +0100 9930 9931 NV50: Implement per CRTC gamma correction. 9932 9933 - This remains untested, so if you're in the mood to test ;-) 9934 9935commit 77af887089be6fef7b793e3c3afdb98ac90a2e8c 9936Author: Maarten Maathuis <madman2003@gmail.com> 9937Date: Wed Mar 26 10:02:46 2008 +0100 9938 9939 randr12: correct SEL_CLK register beheaviour for outputs with or=2 9940 9941commit cb5d4cdb3dc83f9dbd53eae9097c0df458313fff 9942Author: Stuart Bennett <sb476@cam.ac.uk> 9943Date: Wed Mar 26 02:27:33 2008 +0000 9944 9945 Cache running of LVDS scripts, rather than blocking LVDS_RESET 9946 9947 Fixes a regression on some nv40 where LVDS_RESET is needed to fix some bad behaviour in the off/on scripts 9948 9949commit 6924d1ac900524d0978a02f8f8509faab1809b62 9950Author: Maarten Maathuis <madman2003@gmail.com> 9951Date: Thu Mar 20 20:20:16 2008 +0100 9952 9953 NV50: Although not essential, it's nicer to save a register as uint32_t. 9954 9955commit 1042c2bed3aee8561411fba85438c3fe5ee2cdea 9956Author: Maarten Maathuis <madman2003@gmail.com> 9957Date: Thu Mar 20 20:10:13 2008 +0100 9958 9959 NV50: Some more stuff. 9960 9961commit b58e5e99aa06e1851ce255211d524a88f1fb297a 9962Author: Maarten Maathuis <madman2003@gmail.com> 9963Date: Thu Mar 20 19:59:46 2008 +0100 9964 9965 NV50: Educated guesses at some of the control registers. 9966 9967commit ef25cb9224d37c7937e67824a262937a747c91cc 9968Author: Maarten Maathuis <madman2003@gmail.com> 9969Date: Thu Mar 20 18:49:58 2008 +0100 9970 9971 NV50: Forgot some more stuff. 9972 9973commit 64b3d1ab1a4226fa688722dabc40fe05a0e30d52 9974Author: Maarten Maathuis <madman2003@gmail.com> 9975Date: Thu Mar 20 18:35:20 2008 +0100 9976 9977 NV50: Forgot something. 9978 9979commit 81929c1810fe4825ff9bd5d4bc84996c35c8ceb7 9980Author: Maarten Maathuis <madman2003@gmail.com> 9981Date: Thu Mar 20 17:43:30 2008 +0100 9982 9983 NV50: More dehexing. 9984 9985commit ee1ac0646c7c2f018abdfd4f58d49d51f5340784 9986Author: Maarten Maathuis <madman2003@gmail.com> 9987Date: Thu Mar 20 17:19:37 2008 +0100 9988 9989 NV50: Some more dehexing. 9990 9991commit ca1b8bbebd4b786f0fe8531ba7145aa932e5abcc 9992Author: Maarten Maathuis <madman2003@gmail.com> 9993Date: Thu Mar 20 14:18:18 2008 +0100 9994 9995 NV50: Some more register guesses. 9996 9997commit aa99fb07cd438f436fdfa4e5dff6d87d27950358 9998Author: Stuart Bennett <sb476@cam.ac.uk> 9999Date: Wed Mar 19 18:16:03 2008 +0000 10000 10001 randr12: do DPMS off better 10002 10003 Half the time that DPMS off is done, there is no crtc bound, which meant outputs never got turned off. 10004 It seems xf86DisableUnusedFunctions is always run *before* setting up the new outputs (when using xrandr): 10005 if this is not the case, this commit may not be quite right. 10006 10007commit 063b393ac12bec286e0b28c2b18a98cedd3422c1 10008Author: Stuart Bennett <sb476@cam.ac.uk> 10009Date: Wed Mar 19 02:56:31 2008 +0000 10010 10011 Optimizations on LVDS script code 10012 10013commit c5203647ddf262978e7d6a4912661a9cc448da66 10014Author: Stuart Bennett <sb476@cam.ac.uk> 10015Date: Wed Mar 19 23:12:59 2008 +0000 10016 10017 randr12: unbreak LVDS and primary I2C for < NV50 10018 10019 NV50 check disallowed pre-NV50 cards using I2C on first head, and made garbage get written to CR0 on LVDS 10020 10021 This should work for both pre-nv50 and nv50. 10022 10023commit ebc75dc9e8394ee1d13a853341ba9995e9d9d04f 10024Author: Maarten Maathuis <madman2003@gmail.com> 10025Date: Wed Mar 19 21:31:10 2008 +0100 10026 10027 NV50: fix i2c for real 10028 10029commit c6e3a6f57a7ae67c187aff0ad87153e988961784 10030Author: Maarten Maathuis <madman2003@gmail.com> 10031Date: Wed Mar 19 08:57:33 2008 +0100 10032 10033 NV50: Fix a typo and hopefully fix i2c. 10034 10035commit a989b4caac077f65034fb328a80ab4ae758c1162 10036Author: Maarten Maathuis <madman2003@gmail.com> 10037Date: Mon Mar 17 23:40:16 2008 +0100 10038 10039 NV50: Fix major thinko in "NV50: kill output_resource and use or directly." 10040 10041commit cd5aaeaa319e2b681c7c6ccc135d76b7c027bd0d 10042Author: Maarten Maathuis <madman2003@gmail.com> 10043Date: Mon Mar 17 22:05:12 2008 +0100 10044 10045 NV50: Handle some things that weren't fixed during rebase. 10046 10047commit f73c2361a7bba5aa9abb4fe65697bfcb2a407e99 10048Author: Maarten Maathuis <madman2003@gmail.com> 10049Date: Sun Mar 16 16:09:11 2008 +0100 10050 10051 NV50: kill output_resource and use or directly. 10052 10053commit de0fc0ece55997f8b2ad86070cb940e2afd2d7e8 10054Author: Maarten Maathuis <madman2003@gmail.com> 10055Date: Sun Mar 16 12:46:40 2008 +0100 10056 10057 NV50: Mostly resorting regs. 10058 10059 - Also a dithering register define and rename cursor related reg. 10060 10061commit 849bab73da8da8ad8b2987f80fcf90bf86299171 10062Author: Maarten Maathuis <madman2003@gmail.com> 10063Date: Sun Mar 16 12:33:19 2008 +0100 10064 10065 NV50: Some more educated guesses at register names. 10066 10067commit 0244dc1dd0698be161d7fd629efa10dcb3df106a 10068Author: Maarten Maathuis <madman2003@gmail.com> 10069Date: Sun Mar 16 01:09:48 2008 +0100 10070 10071 NV50: Another subtle difference. 10072 10073commit 5e2d220edf65357f4d7bd8ce97bd829086e0781c 10074Author: Maarten Maathuis <madman2003@gmail.com> 10075Date: Sun Mar 16 01:05:03 2008 +0100 10076 10077 NV50: Fix a potentional bug. 10078 10079commit cef158cee51ecc53c2f5a9910a88c9e802d49980 10080Author: Maarten Maathuis <madman2003@gmail.com> 10081Date: Sun Mar 16 00:40:59 2008 +0100 10082 10083 NV50: Share output properties. 10084 10085commit 31e7a740a1cddc5edfdede05187e6bd9a6f1b993 10086Author: Maarten Maathuis <madman2003@gmail.com> 10087Date: Sun Mar 16 00:27:41 2008 +0100 10088 10089 randr12: Implement dithering output property. 10090 10091commit 05f0464669eef0a3caff2466eeffd9843c18a7b7 10092Author: Maarten Maathuis <madman2003@gmail.com> 10093Date: Sat Mar 15 19:02:55 2008 +0100 10094 10095 NV50: Merge output creation and i2c init. 10096 10097commit 285a13ab41e869544681260b1f2415143456bdc9 10098Author: Maarten Maathuis <madman2003@gmail.com> 10099Date: Sat Mar 15 18:20:34 2008 +0100 10100 10101 NV50: Merge crtc creation. 10102 10103commit c5dac45a199b7f4fa3692727aa6af1cd9a6b8d35 10104Author: Stuart Bennett <sb476@cam.ac.uk> 10105Date: Mon Mar 17 23:20:39 2008 +0000 10106 10107 randr12 offbyone debug commitnoise 10108 10109 semantic version control messages ftw? 10110 10111commit 4cd5e430d164cb5a834e4946d5e5dc79061cadfb 10112Author: Stuart Bennett <sb476@cam.ac.uk> 10113Date: Mon Mar 17 17:21:28 2008 +0000 10114 10115 randr12: remove NVGetOutputFromCRTC 10116 10117 This function had a major problem for the long term: assuming a maximum of 1 10118 output per CRTC, which is simply not valid 10119 10120commit 27c21c0574bac47bbb311094c32f17e7ac321ba5 10121Author: Stuart Bennett <sb476@cam.ac.uk> 10122Date: Mon Mar 17 17:09:23 2008 +0000 10123 10124 randr12: split out fp-only registers 10125 10126commit c09d190ffe2bab1e1df52d8b83009c10fafbc2bc 10127Author: Stuart Bennett <sb476@cam.ac.uk> 10128Date: Mon Mar 17 03:03:14 2008 +0000 10129 10130 randr12: better handling for output types in nv_crtc 10131 10132commit 78ded96a4e8f88dba038e5ef0c452b7c7bf9a2dc 10133Author: Stuart Bennett <sb476@cam.ac.uk> 10134Date: Mon Mar 17 17:16:27 2008 +0000 10135 10136 randr12: remove duplicated regs 10137 10138commit f7013760f9e9ec91bc3bd3739b6cce9018e34bcb 10139Author: Stuart Bennett <sb476@cam.ac.uk> 10140Date: Mon Mar 17 00:59:31 2008 +0000 10141 10142 randr12: fix noscale scaling mode 10143 10144 Also, a bit of consistency missed in the earlier commit 10145 10146commit a9c01eff4bcf3af52faf73269f63a5fd46b3b3b3 10147Author: Stuart Bennett <sb476@cam.ac.uk> 10148Date: Sun Mar 16 14:27:14 2008 +0000 10149 10150 Misc startup info mostly duplicated the state storage stuff 10151 10152commit a02f82e9ce79ae24a4f2eba71a83e51f3b6d9e52 10153Author: Stuart Bennett <sb476@cam.ac.uk> 10154Date: Sun Mar 16 00:51:09 2008 +0000 10155 10156 randr12: make output parsing consistent 10157 10158commit f04c44eff67adb9099eca7b298b10639dc0b1759 10159Author: Stuart Bennett <sb476@cam.ac.uk> 10160Date: Sat Mar 15 01:36:13 2008 +0000 10161 10162 G80 I2C parsing 10163 10164 Changed to on-demand parsing 10165 10166commit 718e42478c02c7497cd3917b8ea78ed692f30f3f 10167Author: Stuart Bennett <sb476@cam.ac.uk> 10168Date: Fri Mar 14 21:22:48 2008 +0000 10169 10170 randr12: make nv_crtc_calc_state_ext definition smaller 10171 10172commit 772bf76b3697d1d1725017787378802c2940093e 10173Author: Stuart Bennett <sb476@cam.ac.uk> 10174Date: Tue Mar 11 03:39:43 2008 +0000 10175 10176 randr12: merge common PLL paths 10177 10178commit 1326e0176582fd21e16359764e2792cd125548e6 10179Author: Stuart Bennett <sb476@cam.ac.uk> 10180Date: Tue Mar 11 02:19:10 2008 +0000 10181 10182 randr12: make plls belong to their CRTCs 10183 10184commit 97b074889feef7b4a3b6401825a46d53321d1c7e 10185Author: Stuart Bennett <sb476@cam.ac.uk> 10186Date: Wed Jan 30 14:38:59 2008 +0000 10187 10188 Use PMC defines 10189 10190commit ee11c6abe875a758bdfa4469ae6326ac25732178 10191Author: Maarten Maathuis <madman2003@gmail.com> 10192Date: Fri Mar 14 11:11:07 2008 +0100 10193 10194 randr12: Let an unneeded quirk die. 10195 10196commit 8851d9f3ef4a8cb9b35e1cb8ed80b3ff213eeee0 10197Author: Stuart Bennett <sb476@cam.ac.uk> 10198Date: Wed Mar 12 16:29:17 2008 +0000 10199 10200 randr12: sel_clk values are really an FP thing, so move calculation to nv_output 10201 10202commit 63527f978d923317a97dec060be6df16c7baf17d 10203Author: Stuart Bennett <sb476@cam.ac.uk> 10204Date: Thu Mar 13 20:26:07 2008 +0000 10205 10206 randr12: tidy up output restore 10207 10208commit 769436b3840e10da00b51b94d72d0b2ef510fecc 10209Author: Stuart Bennett <sb476@cam.ac.uk> 10210Date: Thu Mar 13 19:52:52 2008 +0000 10211 10212 randr12: common tmds access functions 10213 10214commit c455ca0d3df3bb2322154ded40fbe6b86c6a8acd 10215Author: Stuart Bennett <sb476@cam.ac.uk> 10216Date: Wed Mar 12 03:23:21 2008 +0000 10217 10218 Improve bios' handling of ramdac_580 10219 10220commit bd5ad7c4e1f824d57517da540658c97b5283d5d0 10221Author: Maarten Maathuis <madman2003@gmail.com> 10222Date: Thu Mar 13 20:49:53 2008 +0100 10223 10224 NV50: Accept G9X cards. 10225 10226commit 0318b31e19da12ba0d4dc456ad08d6b059d08d1e 10227Author: Maarten Maathuis <madman2003@gmail.com> 10228Date: Thu Mar 13 16:57:24 2008 +0100 10229 10230 NV50: Add back output_resource. 10231 10232commit a4397563c6845668ad1f4b2352b90c872957f8bc 10233Author: Maarten Maathuis <madman2003@gmail.com> 10234Date: Wed Mar 12 23:16:53 2008 +0100 10235 10236 NV50: Unbreak "NV50: Kill the connection status caching (which was broken btw)." 10237 10238commit 63ffb8dce809267706eb621f03da3c3501935bbd 10239Author: Maarten Maathuis <madman2003@gmail.com> 10240Date: Tue Mar 11 23:15:11 2008 +0100 10241 10242 NV50: Merge the two output private structs. 10243 10244commit b4c3bee2f8cc824cd52aca61121634c55d5f5a60 10245Author: Maarten Maathuis <madman2003@gmail.com> 10246Date: Tue Mar 11 23:09:19 2008 +0100 10247 10248 NV50: rename scale to scaling_mode 10249 10250commit b0d2f93d3adbb798f5d9bfa2d188b666e4c5fc16 10251Author: Maarten Maathuis <madman2003@gmail.com> 10252Date: Tue Mar 11 23:06:16 2008 +0100 10253 10254 NV50: Rename nativeMode 10255 10256commit e5b8efad35df21d35a3893a6c8b10e1f02272544 10257Author: Maarten Maathuis <madman2003@gmail.com> 10258Date: Tue Mar 11 23:00:19 2008 +0100 10259 10260 NV50: Rename or to output_resource. 10261 10262commit 19ca2db7683b8cdcfb74f6cc14c9cb6c3876d2c7 10263Author: Maarten Maathuis <madman2003@gmail.com> 10264Date: Tue Mar 11 22:43:33 2008 +0100 10265 10266 NV50: kill set_pclk 10267 10268commit eef3a665c01c530238c3dc02324573d6bcf6c3f2 10269Author: Maarten Maathuis <madman2003@gmail.com> 10270Date: Tue Mar 11 22:24:46 2008 +0100 10271 10272 NV50: Remove the remnants of cached_status. 10273 10274commit dc0bb590a40f3d8b18d2136ed906e8dac10ec598 10275Author: Maarten Maathuis <madman2003@gmail.com> 10276Date: Tue Mar 11 22:22:12 2008 +0100 10277 10278 NV50: kill partner entry 10279 10280commit f66463713a71b0e3090f33c68186884594255cc0 10281Author: Maarten Maathuis <madman2003@gmail.com> 10282Date: Tue Mar 11 22:12:57 2008 +0100 10283 10284 NV50: fix warning 10285 10286commit cb50cd3276825569a6da1f25a2d89bfd417b5034 10287Author: Maarten Maathuis <madman2003@gmail.com> 10288Date: Tue Mar 11 21:33:29 2008 +0100 10289 10290 NV50: Kill the connection status caching (which was broken btw). 10291 10292commit 6e2af71c73263c3cf2d61f66c02363decd354da5 10293Author: Maarten Maathuis <madman2003@gmail.com> 10294Date: Tue Mar 11 21:15:21 2008 +0100 10295 10296 NV50: change i2c to pDDCBus 10297 10298commit d07b3063953c0b17585249dfbc2938a433bea88f 10299Author: Maarten Maathuis <madman2003@gmail.com> 10300Date: Tue Mar 11 19:50:25 2008 +0100 10301 10302 NV50: Switch to other scaling defines. 10303 10304commit 4ddb92b8e936985b5b1429a0a21890879635bcde 10305Author: Maarten Maathuis <madman2003@gmail.com> 10306Date: Tue Mar 11 19:29:11 2008 +0100 10307 10308 NV50: Switch to the same output type as the rest of the code. 10309 10310commit e605fbfc659d3fe1bd71d478c21429b563fbfd80 10311Author: Stuart Bennett <sb476@cam.ac.uk> 10312Date: Thu Mar 13 15:21:49 2008 +0000 10313 10314 randr12: nothing uses output_resource. kill it 10315 10316commit 53925cfc3540b89dedd317bbae4318cf53daada1 10317Author: Stuart Bennett <sb476@cam.ac.uk> 10318Date: Thu Mar 13 15:31:00 2008 +0000 10319 10320 randr12: fix the setting of NV_RAMDAC_OUTPUT 10321 10322 And turn on/off the DAC for DPMS 10323 10324commit 493c79405be23a543cd21c2583c466342399509d 10325Author: Stuart Bennett <sb476@cam.ac.uk> 10326Date: Thu Mar 13 03:53:18 2008 +0000 10327 10328 Simplify TMDS writes a little 10329 10330commit 4cf1bd9ef016a2c9fe7784fdc83f559bec464f9e 10331Author: Stuart Bennett <sb476@cam.ac.uk> 10332Date: Tue Mar 11 17:20:28 2008 +0000 10333 10334 randr12: don't allow scripts to break overlay 10335 10336commit 95f2b8439f433afb5d637a3580f933e2cd44488a 10337Author: Danny Tholen <obiwan@mailmij.org> 10338Date: Tue Mar 11 17:14:54 2008 +0000 10339 10340 Fix video overlays to treat double-scan correctly when using randr12 10341 10342commit 460cb260c30467799fa31aef11946d1fe2dd6523 10343Author: Stuart Bennett <sb476@cam.ac.uk> 10344Date: Mon Mar 10 23:47:43 2008 +0000 10345 10346 Correct RAMIN offset 10347 10348 and some bits missed in previous nv_setup tidying 10349 10350commit e8a5d4638e74b5957fc12b05865f49dd995d0435 10351Author: Stuart Bennett <sb476@cam.ac.uk> 10352Date: Mon Mar 10 21:46:38 2008 +0000 10353 10354 randr12: changing tmds2 for dvi-d seemingly unnecessary 10355 10356commit 30ddbe9c2b82bc19d1480a1b38df6b5ca0dd057b 10357Author: Stuart Bennett <sb476@cam.ac.uk> 10358Date: Mon Mar 10 23:49:08 2008 +0000 10359 10360 randr12: unbreak sel_clk changes 10361 10362commit 03f38725304842b053b7976db5df97e61159d07d 10363Author: Stuart Bennett <sb476@cam.ac.uk> 10364Date: Mon Mar 10 19:50:55 2008 +0000 10365 10366 randr12: allow spread spectrum bits to be updated on pre nv40 10367 10368commit 323b7ef9c1c26996b84fd1f67297b2958f68fcc9 10369Author: Stuart Bennett <sb476@cam.ac.uk> 10370Date: Mon Mar 10 01:18:56 2008 +0000 10371 10372 randr12: quirk for 17" powerbook, and better sel_clk handling 10373 10374 We set sel_clk *before* running the output script, so make sure the set pll->head binding is not altered 10375 10376commit 58f9079cda17bb6a76ba16c14ddc56bd4ac640fc 10377Author: Stuart Bennett <sb476@cam.ac.uk> 10378Date: Sun Mar 9 16:08:46 2008 +0000 10379 10380 nv_type.h mandates XF86DRI being declared 10381 10382commit 0dbffc868a7b60aa99ca550d33b8488ac9de8d00 10383Author: Stuart Bennett <sb476@cam.ac.uk> 10384Date: Sun Mar 9 03:59:57 2008 +0000 10385 10386 Fix up a bunch of prototypes, remove unused wrappers 10387 10388commit e5ded02925e7d39c67e69d0389778ac795bfe743 10389Author: Stuart Bennett <sb476@cam.ac.uk> 10390Date: Sun Mar 9 01:31:47 2008 +0000 10391 10392 Move setting of alphaCursor to nv_setup, and restyle useful portions of nv_setup 10393 10394commit 0b8629a9a29203553397eb764b9b9c87d38d4b9f 10395Author: Stuart Bennett <sb476@cam.ac.uk> 10396Date: Sat Mar 8 22:05:52 2008 +0000 10397 10398 Kill NVFBLayout 10399 10400 Using pScrn->currentMode is wrong for randr12, but it's no worse than what it was 10401 10402commit 29cfb08c1b75e050aa58fa6019f7d9be063f8555 10403Author: Stuart Bennett <sb476@cam.ac.uk> 10404Date: Sat Mar 8 21:02:35 2008 +0000 10405 10406 Warning fixes 10407 10408 And remove a few unused members from nv_type.h 10409 10410commit b87059fcccfac693df0b2dc5ba85cb358ad1eba6 10411Author: Stuart Bennett <sb476@cam.ac.uk> 10412Date: Sat Mar 8 20:53:20 2008 +0000 10413 10414 Drop the NV*VGA defines and adjust users 10415 10416commit a636b02eba03454b8b757f69e6a62aae2c3fd571 10417Author: Maarten Maathuis <madman2003@gmail.com> 10418Date: Sat Mar 8 21:18:34 2008 +0100 10419 10420 randr12: Disable a quirk that has adverse sideeffects on NV36M. 10421 10422 - NV31M owners, please test if you have: 10423 - LVDS on CRTC 1 and or = 3. 10424 - VGA on CRTC 0 and or = 1. 10425 - Testing requires enabling both lvds and vga. 10426 10427commit f4228861abec9df310efb863833c82022136e464 10428Author: Maarten Maathuis <madman2003@gmail.com> 10429Date: Sat Mar 8 18:23:58 2008 +0100 10430 10431 Merge CRTC private structs. 10432 10433commit b3d0d0c01664e874f3b69b36383ad1a7845d7d91 10434Author: Maarten Maathuis <madman2003@gmail.com> 10435Date: Sat Mar 8 18:10:53 2008 +0100 10436 10437 Death to some unnecesary stuff. 10438 10439commit 03a7a1d7065b55f16385fd16e7a2eff938d59cea 10440Author: Stuart Bennett <sb476@cam.ac.uk> 10441Date: Sat Mar 8 01:56:36 2008 +0000 10442 10443 Various minor tidyups 10444 10445commit 624148ca7067f748c142db3cd4d31314aa31ee0f 10446Author: Maarten Maathuis <madman2003@gmail.com> 10447Date: Fri Mar 7 20:59:30 2008 +0100 10448 10449 NV50: Convert back to absolute registers. 10450 10451 - Someone with a known to work setup should test this, as i have no NV5x card. 10452 10453commit 10bdffb65e2d8f59399556fe3224cbf8e8a171a9 10454Author: Stuart Bennett <sb476@cam.ac.uk> 10455Date: Fri Mar 7 19:08:50 2008 +0000 10456 10457 PLL limit table v0x11 10458 10459commit 2e1af24826e7db0e3f53440a2c5912789ade6c97 10460Author: Stuart Bennett <sb476@cam.ac.uk> 10461Date: Fri Mar 7 18:43:16 2008 +0000 10462 10463 Make bios init code work on 6600 by changing pll setting stuff a bit 10464 10465 I suspect we're going to need some card specific checks in here. ugh. 10466 10467commit 5fd2dd7b4c6e15d83cb7e7daafb6ee801fa823e8 10468Author: Maarten Maathuis <madman2003@gmail.com> 10469Date: Fri Mar 7 07:52:09 2008 +0100 10470 10471 It seems libpciaccess release became version 0.10 not version 0.10.0 10472 10473commit 3e4db03a39770e5fa94ca094deb690cba8175196 10474Author: Maarten Maathuis <madman2003@gmail.com> 10475Date: Fri Mar 7 07:36:16 2008 +0100 10476 10477 randr12: Forgot to clean up when switching output resources. 10478 10479commit 8a5c6e8b4d00871b17b7d564d765ea677596f79a 10480Author: Jaime Velasco Juan <jsagarribay@gmail.com> 10481Date: Sat Feb 16 15:01:23 2008 +0000 10482 10483 Do not access beyond source pixmap in NVAccelUploadIFC 10484 10485commit 7afb739418fbac5058d13d05188a69b142529fb4 10486Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10487Date: Thu Mar 6 23:40:08 2008 +0100 10488 10489 Oops. 10490 10491commit 0da8c84cceb178b04ab535edb4e3f0ced204d00a 10492Author: Stuart Bennett <sb476@cam.ac.uk> 10493Date: Thu Mar 6 22:35:30 2008 +0000 10494 10495 randr12: workaround for xserver randr bug, where the hwcursor always gets displayed after a mode change 10496 10497commit c424144e854563709a01ed2d30a0482edc57fbf3 10498Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10499Date: Thu Mar 6 23:15:31 2008 +0100 10500 10501 Fix my nv10 cursor. 10502 10503commit 633047c65708f1a39e371c2c5da5c111027af1af 10504Author: Maarten Maathuis <madman2003@gmail.com> 10505Date: Thu Mar 6 23:05:43 2008 +0100 10506 10507 randr12: A major rework of output resource conflict handling. 10508 10509 - This mostly affects laptops, such as some geforce5 and the famous 7300go laptops. 10510 - Test this if you know to have a strange vga/dvi port of your laptop. 10511 10512commit c5230961d181f15cd965c4569813219d9292681a 10513Author: Maarten Maathuis <madman2003@gmail.com> 10514Date: Thu Mar 6 19:29:50 2008 +0100 10515 10516 randr12: Readd some useful information about how an output is routed. 10517 10518commit 97ac866a4e433c70d0de7a1e8fb66a3994194ffd 10519Author: Maarten Maathuis <madman2003@gmail.com> 10520Date: Thu Mar 6 19:16:36 2008 +0100 10521 10522 randr12: Kill almost all usage of ErrorF + some misc cleanup. 10523 10524commit ea8fba29bcfe8158c9bf17754a20684430027aee 10525Author: Stuart Bennett <sb476@cam.ac.uk> 10526Date: Wed Mar 5 18:53:54 2008 +0000 10527 10528 randr12: don't reset panel unnecessarily for mere backlight control 10529 10530commit 22ec35504fbef83ba5790c87d27cd13e936fccb0 10531Author: Stuart Bennett <sb476@cam.ac.uk> 10532Date: Wed Mar 5 14:40:08 2008 +0000 10533 10534 randr12: remove switchable_crtc stuff 10535 10536commit 3b0a1a11f662bfa705806f4afc5898338234ca65 10537Author: Stuart Bennett <sb476@cam.ac.uk> 10538Date: Wed Mar 5 14:35:41 2008 +0000 10539 10540 randr12: if no ddc, return XF86OutputStatusUnknown, for !twoHeads and nv11 10541 10542commit 424500cf828903e3449d4bca22a083087258cc99 10543Author: Stuart Bennett <sb476@cam.ac.uk> 10544Date: Wed Mar 5 02:15:28 2008 +0000 10545 10546 randr12: do not change CR43 from its initial value 10547 10548commit a764d3ca2741243f49f34e9a4930a59272927902 10549Author: Maarten Maathuis <madman2003@gmail.com> 10550Date: Wed Mar 5 09:44:46 2008 +0100 10551 10552 Give some defines more consistent names. 10553 10554commit 234ce16558e59dff302f13ea3d8eb48f46eedf0d 10555Author: Stuart Bennett <sb476@cam.ac.uk> 10556Date: Wed Mar 5 01:25:20 2008 +0000 10557 10558 Use the BMP structure's I2C info, instead of DCB 1.2's, as marcheu's bios is broken 10559 10560commit bb5429f08b3ffd16d6786bd65d6a4c8db1210ade 10561Author: Stuart Bennett <sb476@cam.ac.uk> 10562Date: Wed Mar 5 00:38:15 2008 +0000 10563 10564 Add another DCB 1.4 entry 10565 10566commit 5998ba0a35bc1c897916f66b0bcfb9a3e0a7c54d 10567Author: Stuart Bennett <sb476@cam.ac.uk> 10568Date: Wed Mar 5 00:34:16 2008 +0000 10569 10570 Don't zero out i2c_read and i2c_write 10571 10572 pNv is calloc'd, so zero inited, and explicit zeroing overwrites the stuff parsed in parse_bmp_structure. Duh. 10573 10574commit 2d74bcb4602bb0cdae18ed1ffdd17520a3b57a78 10575Author: Maarten Maathuis <madman2003@gmail.com> 10576Date: Tue Mar 4 21:18:49 2008 +0100 10577 10578 Put braces around a bitmask. 10579 10580commit c7ab8e84bf8ad8ee6106b21a1c07b6d8470e7f11 10581Author: Stuart Bennett <sb476@cam.ac.uk> 10582Date: Tue Mar 4 13:34:16 2008 +0000 10583 10584 Add an LVDS output setup quirk for Powerbooks 10585 10586 Again, as a result of much testing by moondrake and sbriglie. 10587 Hopefully it even works this time :) 10588 10589commit fd9b37b31896e1731bdd56ec847c86c0502ba0f8 10590Author: Maarten Maathuis <madman2003@gmail.com> 10591Date: Mon Mar 3 19:04:09 2008 +0100 10592 10593 randr12: Only change vpll value when needed. 10594 10595commit 89882b416de3528cf504c989cc147a86d3572053 10596Author: Maarten Maathuis <madman2003@gmail.com> 10597Date: Sun Mar 2 20:25:45 2008 +0100 10598 10599 randr12: Keep the gamma values when setting a new mode, instead of always resetting to default. 10600 10601commit e8c363b7ca4a416be37a1ea1c97bbd99b21d5be4 10602Author: Stuart Bennett <sb476@cam.ac.uk> 10603Date: Sat Mar 1 15:05:09 2008 +0000 10604 10605 Improve link_head_and_output 10606 10607 Thanks to moondrake and sbriglie for their work on this 10608 10609commit 4a16414c77ad77e7612f62f4ca1859e0d6a24cc2 10610Author: Patrice Mandin <pmandin@caramail.com> 10611Date: Fri Feb 29 22:17:30 2008 +0100 10612 10613 nv30: Remove fix that is useless now, and maybe broken in some cases 10614 10615commit 5cb164876b824e8bf52aa44ccf4a797bab310602 10616Author: Stuart Bennett <sb476@cam.ac.uk> 10617Date: Thu Feb 28 00:43:17 2008 +0000 10618 10619 Less magical way of setting dual link on EDID equipped panels 10620 10621 Eldritch coupling ftl 10622 10623commit 1127974ed8517f28ef9c451d81bf1c61a2b0e6ae 10624Author: Maarten Maathuis <madman2003@gmail.com> 10625Date: Wed Feb 27 23:45:34 2008 +0100 10626 10627 NV40EXA: Support extend/repeat type pad. 10628 10629 - Note that all xservers, up to about a minute ago, did not try to accelerate repeat type reflect or pad. 10630 10631commit 6980d0e24a9dae52903e6971c5c8a18240a7dbc1 10632Author: Stuart Bennett <sb476@cam.ac.uk> 10633Date: Wed Feb 27 12:55:54 2008 +0000 10634 10635 Add default dual link transition frequency 10636 10637 For card bios versions before the relevant field was added 10638 Should fix Apple laptops. 10639 10640commit 27e2e55698ab05bf3469e0f68474d4c3f021ca31 10641Author: Stuart Bennett <sb476@cam.ac.uk> 10642Date: Tue Feb 26 18:07:39 2008 +0000 10643 10644 Deal with pre-nv40 laptops with EDID 10645 10646commit 41992eb90828976105cbc8bba87fe7d079d7c06a 10647Author: Patrice Mandin <pmandin@caramail.com> 10648Date: Fri Feb 22 17:29:12 2008 +0100 10649 10650 Revert dummy commit 10651 10652commit 073d852f36df4a18563ea892e89eca269efd55ef 10653Author: Patrice Mandin <pmandin@caramail.com> 10654Date: Fri Feb 22 17:27:51 2008 +0100 10655 10656 Dummy commit 10657 10658commit 5db7920763aee3a1b124139273566f7add5e27df 10659Author: Stuart Bennett <sb476@cam.ac.uk> 10660Date: Sun Feb 10 23:14:52 2008 +0000 10661 10662 Use some of nv_hw's functions in nv_bios.c 10663 10664commit fa8891e78f37edc02537704172156ba77371a307 10665Author: Stuart Bennett <sb476@cam.ac.uk> 10666Date: Wed Feb 20 00:50:48 2008 +0000 10667 10668 Unbreak console restore again, and use a less ambiguous function name for head locking 10669 10670commit 38d42a7857480a930e2025e4f63aa7f5defca5c5 10671Author: Stuart Bennett <sb476@cam.ac.uk> 10672Date: Tue Feb 19 23:59:12 2008 +0000 10673 10674 randr12: rename NV*VGA functions for api consistency 10675 10676 Also nuke NVCrtcSetOwner wrapper 10677 10678commit c77a54768c3fed5939755f23c5c067bb86f49805 10679Author: Ben Skeggs <darktama@beleth.(none)> 10680Date: Thu Feb 21 00:55:39 2008 +1100 10681 10682 xv: don't crash on init 10683 10684commit cf963a128374bf7c2070e8a380e236b5e083adc1 10685Author: Maarten Maathuis <madman2003@gmail.com> 10686Date: Wed Feb 20 09:36:42 2008 +0100 10687 10688 randr12: Only NV11 needs byteswapping + NV11 uses premultiplied cursors i think. 10689 10690commit 87f1676a243cddcaece9eeb0487a85b7e74b0f13 10691Author: Maarten Maathuis <madman2003@gmail.com> 10692Date: Wed Feb 20 00:14:12 2008 +0100 10693 10694 NV30/40TEX: Create two adapters, one with fancy filtering, the default without. 10695 10696 - Also enable sync to vblank by default, strange that it was off. 10697 - Remove the GET_TEXTURE(D)_PRIVATE macros, as one of them pointed to the blitter and one is now ambigious. 10698 10699commit 7ee1bb47d86396cc1a0f7879a98d04588cb31eee 10700Author: Stuart Bennett <sb476@cam.ac.uk> 10701Date: Thu Feb 14 23:16:01 2008 +0000 10702 10703 randr12: make more hw funcs generic 10704 10705commit 11de8d7a6f9a2dfd5a4f93f031865e284268a0f3 10706Author: Stuart Bennett <sb476@cam.ac.uk> 10707Date: Tue Feb 19 21:44:56 2008 +0000 10708 10709 Use fp strapping to determine whether straps are used for flat panel modes 10710 10711commit c1ac6b977a4756e2dbfdb75ab5b53c7c10e76cf4 10712Author: Stuart Bennett <sb476@cam.ac.uk> 10713Date: Tue Feb 19 14:53:41 2008 +0000 10714 10715 Rehabilitate bios opcode 6d 10716 10717commit afb42307cb2810ca2d759df62240fbf1f0cc7f7c 10718Author: Stuart Bennett <sb476@cam.ac.uk> 10719Date: Tue Feb 19 14:23:06 2008 +0000 10720 10721 Spruce up NV04 bios parsing a little 10722 10723commit dfabaab48034bf36c19510fe5de0f1bdaa2157ef 10724Author: Stuart Bennett <sb476@cam.ac.uk> 10725Date: Tue Feb 19 00:37:27 2008 +0000 10726 10727 More DCB v1.5 entries 10728 10729 Courtesy HenneNWH's NV11 10730 10731commit 2a4373d4f3b092600651b7dc8e2f9f9b520c7e0d 10732Author: Stuart Bennett <sb476@cam.ac.uk> 10733Date: Mon Feb 18 17:32:43 2008 +0000 10734 10735 Treat 0x4 and 0x8 bits in LVDS DCB v2.0 configuration as meaning the same thing 10736 10737commit daa557c0dadb3ad5caf39a4e90d3ebd16b12fc96 10738Author: Danny Tholen <obiwan@mailmij.org> 10739Date: Mon Feb 18 05:00:03 2008 +0100 10740 10741 Fix PPC DFS/UTS by using PCIgart. 10742 10743commit f8f469286fb3acc1ee694c61fe21e42123a5456a 10744Author: Ben Skeggs <skeggsb@gmail.com> 10745Date: Sat Feb 16 03:41:38 2008 +1100 10746 10747 dri: fix broken display in some cases 10748 10749commit a293f32b2172837b6e2bcc3f5771c5219fccfb14 10750Author: Stuart Bennett <sb476@cam.ac.uk> 10751Date: Wed Feb 13 00:52:34 2008 +0000 10752 10753 Fix bios pll setting regression 10754 10755 Oops. 10756 10757commit 7121fe9dfb1c9162c8be808e09ed6b4b3ce3999e 10758Author: Stuart Bennett <sb476@cam.ac.uk> 10759Date: Tue Feb 12 18:20:43 2008 +0000 10760 10761 Use a loop to read RAMIN bios image, rather than memcpy 10762 10763commit 3ad7c35740905ad267faf8e248075640d5ee9cea 10764Author: Stuart Bennett <sb476@cam.ac.uk> 10765Date: Tue Feb 12 17:23:38 2008 +0000 10766 10767 randr12: reorder some variable declarations in nv_crtc and nv_output 10768 10769commit ab40a46430f369a7bd9cef8db4e86414125c8437 10770Author: Stuart Bennett <sb476@cam.ac.uk> 10771Date: Tue Feb 12 15:19:58 2008 +0000 10772 10773 randr12: use integer math for scaling 10774 10775 No floats in modesetting path 10776 10777commit 4a44468c29c37b5ad4828439f37d023865a9d457 10778Author: Stuart Bennett <sb476@cam.ac.uk> 10779Date: Tue Feb 12 15:03:46 2008 +0000 10780 10781 randr12: tidy up some mode retrieval stuff 10782 10783commit 50d77c4199a3a5d2f3ead8327341e91ac2ec5684 10784Author: Stuart Bennett <sb476@cam.ac.uk> 10785Date: Mon Feb 11 16:27:27 2008 +0000 10786 10787 Trim useless nv_type members 10788 10789 Also, crtc_active isn't used on randr12 path 10790 10791commit 0c1e48a2192555856894d7e2b3025cd342c2f448 10792Author: Stuart Bennett <sb476@cam.ac.uk> 10793Date: Mon Feb 11 01:42:04 2008 +0000 10794 10795 Remove some DDX cruft 10796 10797commit 5a8e246fcab3e349f48d0ce0c4da812a411767f2 10798Author: Stuart Bennett <sb476@cam.ac.uk> 10799Date: Mon Feb 11 00:34:27 2008 +0000 10800 10801 Use mmio access wrappers and reorder some variable declarations in nv_bios.c 10802 10803commit 2ffcd89bfb0383fb32e8ec41f1cd472764b9b99d 10804Author: Stuart Bennett <sb476@cam.ac.uk> 10805Date: Mon Feb 11 00:03:50 2008 +0000 10806 10807 randr12: Use absolute regs for 8 bit regs 10808 10809commit c9b1f020d1c5be069dae19f99b3c26bb126052ae 10810Author: Stuart Bennett <sb476@cam.ac.uk> 10811Date: Fri Feb 8 19:44:37 2008 +0000 10812 10813 Move non-crtc specific hw funcs from nv_crtc to nv_hw 10814 10815commit 6cb8fb6cc63bf55f18b4dde53c393336e817474e 10816Author: Stuart Bennett <sb476@cam.ac.uk> 10817Date: Thu Feb 7 02:04:20 2008 +0000 10818 10819 Ordering mechanism for BIT table parsing, and (hopefully) automagic laptop detection 10820 10821commit 4f690235d70ae249323439380a69d8309a8c9981 10822Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10823Date: Wed Feb 6 22:18:46 2008 +0100 10824 10825 Replace a const with its name. 10826 10827commit 75c5912fffde88bfe2ddd158a446588b76ef44d9 10828Author: Maarten Maathuis <madman2003@gmail.com> 10829Date: Wed Feb 6 19:06:14 2008 +0100 10830 10831 Fix bad debugging in NVWritePVIO(). 10832 10833commit 50da8b8f5b5feb6787ed50f221529730d25b5467 10834Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10835Date: Wed Feb 6 17:44:52 2008 +0100 10836 10837 Fix UV swapping on PPC. 10838 10839commit 4ff11d13be6e8050e0b726dc7af8e4bc11bd9d0d 10840Author: Stuart Bennett <sb476@cam.ac.uk> 10841Date: Mon Feb 4 22:34:21 2008 +0000 10842 10843 Missed out "DEBUG" part of the name in the define 10844 10845commit 181ee3bcecfc778c9f671853a696ab3224ac4747 10846Author: Stuart Bennett <sb476@cam.ac.uk> 10847Date: Mon Feb 4 22:31:02 2008 +0000 10848 10849 Load correct PLL limits for single stage PLL cards with blank PLL tables 10850 10851commit cbcb1070db6fbb4d13ddcc35739d2d3e49f3766d 10852Author: Maarten Maathuis <madman2003@gmail.com> 10853Date: Mon Feb 4 07:47:06 2008 +0100 10854 10855 Once again kill usage of an exa private function. 10856 10857 - Must have creeped back in during all the texture adapter changes recently. 10858 10859commit f14d27d638e75faa5bbda903f65a4d5479222ead 10860Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10861Date: Sun Feb 3 17:15:50 2008 +0100 10862 10863 Remove include guards & add copyright. 10864 10865commit 7522640ee8d342c5ebb62989b051f73e230f5ce7 10866Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10867Date: Sun Feb 3 17:01:50 2008 +0100 10868 10869 Last round of fixes for nv30/40 bicubic Xv. 10870 10871commit 78553ab9a0309c7d27632fe7184f3233cfde6610 10872Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10873Date: Sun Feb 3 06:07:35 2008 +0100 10874 10875 Fix tearing. 10876 10877commit 7240f861a98c5f8d9779a4f3a743b748af775337 10878Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10879Date: Sun Feb 3 05:15:41 2008 +0100 10880 10881 Understood and fixed some NV30 FP_CONTROL bits. 10882 10883commit bb2a7bd3461eec37f179748a9bdb9375b1c3da04 10884Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10885Date: Sat Feb 2 05:54:47 2008 +0100 10886 10887 Fixes to nv30/nv40 textured video. 10888 10889commit c10d30dfefe93a9911d89eaffe3d29d3cff66a96 10890Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10891Date: Thu Jan 31 06:15:02 2008 +0100 10892 10893 Misc textured video fixes. 10894 10895commit cfdfd5a99bb7ac959a1188994db4c75a97de2e34 10896Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10897Date: Thu Jan 31 06:00:59 2008 +0100 10898 10899 Of course, add the nv30 video texture file. 10900 10901commit 970a8e474c666d6e33351f7ca6a3a5273024baf4 10902Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10903Date: Thu Jan 31 05:59:08 2008 +0100 10904 10905 nv30 video texture. Doesn't work with bicubic yet, only bilinear. 10906 10907commit c1eba822f1579637f59933b04e31591e86b975fc 10908Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10909Date: Wed Jan 30 02:38:14 2008 +0100 10910 10911 Remove (void)ing that's not necessary any more. 10912 10913commit 7356ad90b387d01c2bf25aa653c41ca9365c5c20 10914Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10915Date: Wed Jan 30 02:30:39 2008 +0100 10916 10917 Remove unneeded stuff. 10918 10919commit e06dfd2964f9bd7f0855a101f5e6971020dfa6be 10920Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10921Date: Wed Jan 30 02:19:08 2008 +0100 10922 10923 Separate the shaders from the rest of the code. 10924 10925commit a69c50780cacf0d0f2948208cfa7e0558a8c3644 10926Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10927Date: Wed Jan 30 02:18:26 2008 +0100 10928 10929 Macros for the PME range. 10930 10931commit 47c8ed472f37d68f3cc241a58ee5b7b8643e537d 10932Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10933Date: Wed Jan 30 02:17:22 2008 +0100 10934 10935 Add PME range. 10936 10937commit 63f47c9bf59ccb9f1b6adbefd55deaec1435faf7 10938Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10939Date: Tue Jan 29 21:25:28 2008 +0100 10940 10941 Small cleanups. 10942 10943commit c98e9fa5c572deeafd7f605bb29d6c83cd38de9d 10944Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10945Date: Tue Jan 29 21:24:39 2008 +0100 10946 10947 nv40 video texture: first pass at bicubic video. 10948 10949commit 10967a4dac8c3eaf828359dcba7a6e1272561aa3 10950Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10951Date: Mon Jan 28 04:52:01 2008 +0100 10952 10953 Update to newer nouveau_class.h 10954 10955commit 82728b3df45e8a5c0a45462d2ce847a41fd31f07 10956Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 10957Date: Sat Jan 26 19:02:50 2008 +0100 10958 10959 Update to latest nouveau_class.h 10960 10961commit 9d9a5acae1deae8c76b512362634066e0c32a817 10962Author: Jaime Velasco Juan <jsagarribay@gmail.com> 10963Date: Sun Feb 3 15:45:32 2008 +0100 10964 10965 Don't FIRE_RING() when in NoAccel mode. 10966 10967commit 11c1ee1cb0ee9c8096ea371d95d188c538ac3fb2 10968Author: Maarten Maathuis <madman2003@gmail.com> 10969Date: Sun Feb 3 15:43:35 2008 +0100 10970 10971 Some minor reindenting. 10972 10973commit 0db347294e09d3d17200ca130e719df2d22f69b6 10974Author: Ben Skeggs <skeggsb@gmail.com> 10975Date: Mon Feb 4 01:19:23 2008 +1100 10976 10977 exa: those limits were a bit conservative on some hw. 10978 10979 AFAIK nv2x up should be able to handle 4096x4096 textures and 10980 render targets. 10981 10982commit eb72e67641215b710868da7bb66b466f6ccbbe25 10983Author: Ben Skeggs <skeggsb@gmail.com> 10984Date: Mon Feb 4 01:07:07 2008 +1100 10985 10986 exa: more accurate coordinate limits 10987 10988commit 8645d2d9554ee46397240fbd9904fd4f905aefb2 10989Author: Maarten Maathuis <madman2003@gmail.com> 10990Date: Sun Feb 3 14:54:41 2008 +0100 10991 10992 Fix NVExaPixmapMap() for NOUVEAU_EXA_PIXMAPS case. 10993 10994commit dd81efd0cc64a070f7489decbac345cc1b4fa9ce 10995Author: Maarten Maathuis <madman2003@gmail.com> 10996Date: Sun Feb 3 12:03:43 2008 +0100 10997 10998 Fix undefined symbol. 10999 11000commit 2a1116cbf61e84b16366076366f4bbd5f550fb9f 11001Author: Maarten Maathuis <madman2003@gmail.com> 11002Date: Sat Feb 2 20:34:57 2008 +0100 11003 11004 Remove a few leftovers to actually make the last commits work. 11005 11006commit 714bea6449aab47a9f2211718183ee84a3c70129 11007Author: Maarten Maathuis <madman2003@gmail.com> 11008Date: Sat Feb 2 19:53:32 2008 +0100 11009 11010 Kill NVShowHideCursor and use a wrapper. 11011 11012commit 741acded643945c5a2da7bef088ed1ff7f5fa85e 11013Author: Maarten Maathuis <madman2003@gmail.com> 11014Date: Sat Feb 2 19:26:13 2008 +0100 11015 11016 Remove nvRead/WriteVGA in favor of nvRead/WriteCurVGA wrappers. 11017 11018commit d429cabd9ae9865a45cc88f43ab6d9d61b3fcf36 11019Author: Maarten Maathuis <madman2003@gmail.com> 11020Date: Sat Feb 2 19:17:32 2008 +0100 11021 11022 Kill NVLockUnlock and make it redirect to NVLockUnlockHead. 11023 11024commit dc2aee0ef69c85aaffdc3c1da5a228642d049a30 11025Author: Maarten Maathuis <madman2003@gmail.com> 11026Date: Sat Feb 2 15:12:00 2008 +0100 11027 11028 More wrapper death. 11029 11030commit a57fd67a2a6ae66af218fedd704a04d855260a2f 11031Author: Maarten Maathuis <madman2003@gmail.com> 11032Date: Sat Feb 2 15:06:02 2008 +0100 11033 11034 Let some wrappers die. 11035 11036commit b9a4eb6f6ea4f8f0747baa50037e33a0584cfd29 11037Author: Maarten Maathuis <madman2003@gmail.com> 11038Date: Sat Feb 2 12:47:32 2008 +0100 11039 11040 Undo last commit and move it to drm. 11041 11042commit 288b56545a8da1e7b5aa64b1093316ca2f28f868 11043Author: Maarten Maathuis <madman2003@gmail.com> 11044Date: Sat Feb 2 12:28:44 2008 +0100 11045 11046 NV40: Fix longstanding issue with nouveau crashing after using blob. 11047 11048 - Took me a while to track down, but for me this is definately it. 11049 - In my case i got the X cross and then it would freeze and i would get a fifo dump. 11050 - Don't be shy, let me know if this works for you too. 11051 11052commit 0ec1b10a82c411181ff7a8643bb2390cfe08e4e4 11053Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11054Date: Sat Feb 2 04:04:58 2008 +0100 11055 11056 Don't allocate the whole AGP gart as a scratch buffer, since we still need to allocate the fifo afterwards. 11057 11058commit 5b79b0a9e54fe99eaae5b411f46a601b6944b773 11059Author: Maarten Maathuis <madman2003@gmail.com> 11060Date: Fri Feb 1 23:34:32 2008 +0100 11061 11062 Don't use functions that are not supposed to be used in randr12 mode. 11063 11064 - Also add NVLockUnlockHead() and NVSetOwner() functions. 11065 11066commit cedb604ee79cd73a9ec3940c7b056109811ba670 11067Author: Maarten Maathuis <madman2003@gmail.com> 11068Date: Fri Feb 1 22:41:09 2008 +0100 11069 11070 Remove usage of "illegal" wrappers in code that also used for randr12 mode. 11071 11072 - I've got something prettier in mind to clean this up, which will likely follow. 11073 11074commit 1a5247856929877442e1ec00f190c6d497ca64fe 11075Author: Maarten Maathuis <madman2003@gmail.com> 11076Date: Fri Feb 1 19:14:38 2008 +0100 11077 11078 randr12: Remove a hack, confirmed on 7600GS that it is unneeded. 11079 11080 - This was one of the cards that did need it in the past. 11081 11082commit b6972e7bec0455250851c06d142336cb53be1a74 11083Author: Maarten Maathuis <madman2003@gmail.com> 11084Date: Fri Feb 1 14:47:32 2008 +0100 11085 11086 randr12: Some minor fixes. 11087 11088commit 988dce0f452f6b80b2d02e1a0f78f826674cc3c0 11089Author: Arthur Huillet <arthur.huillet@free.fr> 11090Date: Fri Feb 1 10:04:32 2008 +0100 11091 11092 NV10 EXA: limit regcomb state re-emission, fix logic for A8 + A8 yielding a 3% perf improvement :p 11093 11094commit a84852c2e0b19cf848495dc7e4bde4426b7bd199 11095Author: Maarten Maathuis <madman2003@gmail.com> 11096Date: Wed Jan 30 22:57:09 2008 +0100 11097 11098 (hopefully) Fix build on powerpc. 11099 11100commit e8a975e6a8afe5ae9c77d20f4d59d3d10221f7d9 11101Author: Arthur Huillet <arthur.huillet@free.fr> 11102Date: Wed Jan 30 18:37:39 2008 +0100 11103 11104 Xv: NV30 overlay can not scale down by more than 0.5x 11105 11106commit adc26892f6e711f08a5ed189a643e3fe9646e15c 11107Author: Stuart Bennett <sb476@cam.ac.uk> 11108Date: Wed Jan 30 17:43:16 2008 +0000 11109 11110 Fail gracefully for too-big resolutions / colour depths 11111 11112commit 09c9d7c3c50b630e1509f787ff946cc0656f7cba 11113Author: Maarten Maathuis <madman2003@gmail.com> 11114Date: Wed Jan 30 18:50:37 2008 +0100 11115 11116 randr12: Misc change. 11117 11118commit 6a46bf59f15d68404b536d436bf70337cd88e33d 11119Author: Maarten Maathuis <madman2003@gmail.com> 11120Date: Wed Jan 30 18:35:18 2008 +0100 11121 11122 Check if the framebuffer can fit into the offscreen memory. 11123 11124commit 9607d7828b16b0cb4a428bc29e2ce7c9bfd95371 11125Author: Arthur Huillet <arthur.huillet@free.fr> 11126Date: Wed Jan 30 18:24:44 2008 +0100 11127 11128 Xv: NV04 overlay can not scale down at all, as confirmed by DirectFB 11129 11130commit 7edfb473533ce41db9e2b3709d32bb8ca7da85a4 11131Author: Stuart Bennett <sb476@cam.ac.uk> 11132Date: Wed Jan 30 14:44:14 2008 +0000 11133 11134 PBUS define additions 11135 11136commit 6e090591bd2dab051362d8528a3bc76f30418d94 11137Author: Stuart Bennett <sb476@cam.ac.uk> 11138Date: Wed Jan 30 14:35:25 2008 +0000 11139 11140 PCRTC define additions 11141 11142commit e51799382e40c033b2d6d09dd65dd7f7c1e5e4d6 11143Author: Stuart Bennett <sb476@cam.ac.uk> 11144Date: Wed Jan 30 15:03:06 2008 +0000 11145 11146 Missed these in the ->absolute conversion 11147 11148commit ee0d557a26d326af27445bd0b283a5c6ead7c1cd 11149Author: Arthur Huillet <arthur.huillet@free.fr> 11150Date: Tue Jan 29 18:46:16 2008 +0100 11151 11152 Xv: detection of composite for the overlay 11153 11154commit 3a5f92d48e32d251e41bab561e52082033d26d9d 11155Author: Stuart Bennett <sb476@cam.ac.uk> 11156Date: Wed Jan 30 01:31:43 2008 +0000 11157 11158 Make bios port io use PVIO as appropriate, and have harsh validity checks 11159 11160commit 92c0d9630843957212d8ce296fea6bca8d802069 11161Author: Stuart Bennett <sb476@cam.ac.uk> 11162Date: Wed Jan 30 00:36:59 2008 +0000 11163 11164 Make PLL setting quieter 11165 11166commit f58d174bd50943c3dfb2d6c909d0a98dbbd35dae 11167Author: Stuart Bennett <sb476@cam.ac.uk> 11168Date: Tue Jan 29 23:10:27 2008 +0000 11169 11170 Move some regs from nv_bios.c to nvreg.h 11171 11172 And use more symbolic names in nv_bios 11173 11174commit d0022c35f39fde74ee5f3978ea78e053988fb608 11175Author: Stuart Bennett <sb476@cam.ac.uk> 11176Date: Tue Jan 29 22:41:40 2008 +0000 11177 11178 Tighten bios reg checks a little 11179 11180commit b1b1d76eb187f4ab3020bbd69c2b1cc9b39d0d83 11181Author: Stuart Bennett <sb476@cam.ac.uk> 11182Date: Wed Jan 30 00:30:46 2008 +0000 11183 11184 randr12: make nv4x plls less special 11185 11186commit 8f05d66bad8e74129cffa2918f9ecf6824c9f0b9 11187Author: Stuart Bennett <sb476@cam.ac.uk> 11188Date: Tue Jan 29 22:33:34 2008 +0000 11189 11190 Don't attempt to parse BIT D and L tables on non-mobile cards 11191 11192commit 8e834ece6592df13bd3eb7e9f28c01b4eb0c9791 11193Author: Stuart Bennett <sb476@cam.ac.uk> 11194Date: Tue Jan 29 18:57:28 2008 +0000 11195 11196 Add some mobile pci-ids, and reorder numerically 11197 11198commit 0b586ae0a2998339a9d5d3590e91dfa62af9ab6a 11199Author: Maarten Maathuis <madman2003@gmail.com> 11200Date: Tue Jan 29 19:04:25 2008 +0100 11201 11202 randr12: Remove excessive writing of VTOwner. 11203 11204commit c522ebb293a1c37142e29bf02bb82524e99336b0 11205Author: Stuart Bennett <sb476@cam.ac.uk> 11206Date: Tue Jan 29 17:30:32 2008 +0000 11207 11208 Make sure nv04 has some defaults (as the bios parsing is non-existant) 11209 11210commit 62b6c5fe490578990c249042657f20bb0cf49a01 11211Author: Stuart Bennett <sb476@cam.ac.uk> 11212Date: Sun Jan 27 22:11:31 2008 +0000 11213 11214 Use absolute regs in the DDX 11215 11216 Due to having to make a few defines for regs that were hardcoded, there could 11217 be breakage (nv04 overlay, this means you) 11218 Hopefully not. 11219 11220commit f5c9f9910a46cfa41d817ef7ca345bfea278b358 11221Author: Stuart Bennett <sb476@cam.ac.uk> 11222Date: Tue Jan 29 14:14:26 2008 +0000 11223 11224 Reorganise pll limit retrieval 11225 11226commit 26ff67f03be38925905a572dab3ec61653cd8286 11227Author: Maarten Maathuis <madman2003@gmail.com> 11228Date: Tue Jan 29 10:53:14 2008 +0100 11229 11230 NV30EXA: Always init texture origin, please report if your viewport setup needs to be different. 11231 11232commit 38df9a95740e934b9691dff1ed1b597cd2254cfa 11233Author: Maarten Maathuis <madman2003@gmail.com> 11234Date: Mon Jan 28 20:55:47 2008 +0100 11235 11236 Forgot to be paranoid. 11237 11238commit 2ea0314859569c3ae4356272a35f0b316abf4df7 11239Author: Maarten Maathuis <madman2003@gmail.com> 11240Date: Mon Jan 28 20:51:20 2008 +0100 11241 11242 Small improvement to nouveau_dma_wait, at least makes it show up in profiling in a clear way. 11243 11244 - This may also slightly reduce the cost of a very short wait, but not by much. 11245 11246commit dd483e226519bc4d2c088e97490039613014dfac 11247Author: Stuart Bennett <sb476@cam.ac.uk> 11248Date: Sat Jan 26 20:22:08 2008 +0000 11249 11250 randr12: 2nd attempt on nv11 console restore 11251 11252commit 5b0fa02101c82a83e11f7d279d366689d54bfacb 11253Author: Stuart Bennett <sb476@cam.ac.uk> 11254Date: Sat Jan 26 18:09:10 2008 +0000 11255 11256 Use NV_{WR,RD}08 to avoid double logging io 11257 11258 And one less bios FIXME 11259 11260commit a099bbff8bb3766201a7deb9c4a718fff7edbe51 11261Author: Stuart Bennett <sb476@cam.ac.uk> 11262Date: Mon Jan 21 23:54:51 2008 +0000 11263 11264 randr12: remove unused PLL code 11265 11266commit 0eb428e5a8b60e52d18b3bdffed46ae3f07b7bc0 11267Author: Stuart Bennett <sb476@cam.ac.uk> 11268Date: Fri Jan 25 16:05:59 2008 +0000 11269 11270 randr12: attempt to fix brokenness of head setting / getting on nv11 11271 11272commit eea2fd41ebc17f72b43c776a0391df0733b8d6e4 11273Author: Stuart Bennett <sb476@cam.ac.uk> 11274Date: Sat Jan 26 14:20:01 2008 +0000 11275 11276 randr12: don't do load detection on nv11 11277 11278commit a2c0facefbc878815e00c734464b3c87b5cce842 11279Author: Stuart Bennett <sb476@cam.ac.uk> 11280Date: Sat Jan 26 01:46:59 2008 +0000 11281 11282 Assume 18 bit laptop panels by default, add a message that may trigger on 24 bit panels 11283 11284 Revert "Invert 18/24 bit interface flag, so that the default value (0) does not imply dithering" 11285 11286 This reverts commit 5b9dc7ad6eb2f16c4ca8824c2a19c85bbe1f126b. 11287 11288commit ec548d20239a49f8e9eb0399317d4c68df24f4a3 11289Author: Stuart Bennett <sb476@cam.ac.uk> 11290Date: Sat Jan 26 01:36:41 2008 +0000 11291 11292 Detect mobile BMP biosen 11293 11294 I'd really love to know how where this info is in BIT 11295 11296commit 9f932eb684814e2a04c83d5aee172b9e020d82a3 11297Author: Maarten Maathuis <madman2003@gmail.com> 11298Date: Fri Jan 25 23:31:58 2008 +0100 11299 11300 NV30EXA: Fix viewport setup + switch to triangle based composite. 11301 11302commit afca5f2665a62ef2d025c15f07fe1cdc2b40d8d5 11303Author: Stuart Bennett <sb476@cam.ac.uk> 11304Date: Fri Jan 25 01:18:50 2008 +0000 11305 11306 randr12: fix / annotate a couple of ddc thinkos 11307 11308 catch a longstanding source of error too 11309 11310commit 3e6ea79c3292c3ed5bd676653c4c40532772ae5f 11311Author: Stuart Bennett <sb476@cam.ac.uk> 11312Date: Fri Jan 25 00:33:07 2008 +0000 11313 11314 Implement opcodes 0x66, 0x67, and 0x68 11315 11316 Courtesy jwstolk's trace 11317 Names taken from nv11 biosmod scps 11318 11319commit cdae2e26cc432ba572f92522f1b3454264d38f43 11320Author: Maarten Maathuis <madman2003@gmail.com> 11321Date: Fri Jan 25 00:10:08 2008 +0100 11322 11323 randr12: Rework native modes for TMDS. 11324 11325 - Choose a preferred ddc mode when available, otherwise the one with the highest refresh. 11326 - Fall back on cvt modes, only when needed. 11327 11328commit 46636052498d1e33985e3d98a6ce9462639bd4ff 11329Author: Stuart Bennett <sb476@cam.ac.uk> 11330Date: Wed Jan 23 23:02:46 2008 +0000 11331 11332 oops 11333 11334commit e067a0eee42324ec7c01e4d84a33554054817dd9 11335Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11336Date: Wed Jan 23 23:40:51 2008 +0100 11337 11338 Use proper version. 11339 11340commit 324c19acf6b506aae47e3699dd0fcefb72473590 11341Author: Stuart Bennett <sb476@cam.ac.uk> 11342Date: Wed Jan 23 01:13:58 2008 +0000 11343 11344 Fix C51 refclk 11345 11346commit 178b8bbf0e09c190fef8b2f85acddd3172ace629 11347Author: Stuart Bennett <sb476@cam.ac.uk> 11348Date: Wed Jan 23 20:59:50 2008 +0000 11349 11350 randr12: calculate (and set, for nv40) sel_clk before calculating plls 11351 11352commit 8067bd08d9a928c97b83f2b6ebff996b320c5813 11353Author: Maarten Maathuis <madman2003@gmail.com> 11354Date: Wed Jan 23 19:43:42 2008 +0100 11355 11356 Fix big bug in commit "Mostly log message changes". 11357 11358 - Please don't hide such important changes in a commit like that ;-) 11359 11360commit c5329a42c509efe9336573ff7e32978082ba9355 11361Author: Stuart Bennett <sb476@cam.ac.uk> 11362Date: Wed Jan 23 16:37:31 2008 +0000 11363 11364 Redo sel_clk setting and update its documentation 11365 11366 My apologies if this breaks stuff. 11367 11368commit f07ab92e7bb419df5473e1ed843a79ef4ff08f30 11369Author: Stuart Bennett <sb476@cam.ac.uk> 11370Date: Wed Jan 23 02:09:56 2008 +0000 11371 11372 Fail on LVDS DCB entry parsing issues 11373 11374commit 12efa8cdfe68c312458368ab876030b02919a922 11375Author: Stuart Bennett <sb476@cam.ac.uk> 11376Date: Wed Jan 23 00:09:11 2008 +0000 11377 11378 Mostly log message changes 11379 11380commit 0faa15d9b84b2cad23424dfc1679f119c5b45108 11381Author: Ben Skeggs <skeggsb@gmail.com> 11382Date: Wed Jan 23 23:45:37 2008 +1100 11383 11384 not an appropriate place for this type of thing.. 11385 11386commit 9e982df106028edd0f7ce0feedbe8e082f29f2ac 11387Author: Maarten Maathuis <madman2003@gmail.com> 11388Date: Wed Jan 23 11:42:55 2008 +0100 11389 11390 randr12: Increase cvt mode to 72 Hz. 11391 11392 - I've had occasions were i did slightly notice something at 60 Hz, and no problems at a higher setting. 11393 - I must admit i was a little bit tired at the time, but i figure it won't hurt (much). 11394 11395commit 71435dde5b2fd1c197ef5dc31b22ba40abcbca7e 11396Author: Maarten Maathuis <madman2003@gmail.com> 11397Date: Wed Jan 23 11:20:47 2008 +0100 11398 11399 NV40EXA: Use the same approach as the texture adapter for rendering. 11400 11401 - Verified to work for abitrary rotation. 11402 - This should eliminate diagional tearing altogether. 11403 11404commit d9149bddc758cc0644630b26fe10fc563ba38ce9 11405Author: Maarten Maathuis <madman2003@gmail.com> 11406Date: Wed Jan 23 10:28:25 2008 +0100 11407 11408 NV40TEX: Switch to a big triangle and clip it to a quad. 11409 11410 - On some systems (NV43, NV46 are known) an artifact shows when 2 triangles are combines and the size is larger than roughly 512x512. 11411 - The trick is draw a single triangle and use scissors to turn it into a quad. 11412 - Someone once asked me if i ever made drawings, well i did in this case and i've included it for future reference. 11413 11414commit e578d6e0a8c0fb8877d8b4b6ba1046dccf28ee6d 11415Author: Stuart Bennett <sb476@cam.ac.uk> 11416Date: Tue Jan 22 18:33:42 2008 +0000 11417 11418 Fix LVDS with DCB 2.0 11419 11420 Also use the X_ERROR facility for errors, not ErrorF 11421 11422commit fd3195f503420591b522249bb423c817db0709e7 11423Author: Maarten Maathuis <madman2003@gmail.com> 11424Date: Tue Jan 22 19:00:49 2008 +0100 11425 11426 Xv: Remove illegal usage of private exa symbols. 11427 11428commit 3a995b06ddb507655279565e992604bd40036b19 11429Author: Maarten Maathuis <madman2003@gmail.com> 11430Date: Tue Jan 22 15:53:38 2008 +0100 11431 11432 NV40TEX: Use width, not (x2, y2). 11433 11434commit abaa8e822be49b9d2129d31c4996dfcf8034ea2a 11435Author: Arthur Huillet <arthur.huillet@free.fr> 11436Date: Tue Jan 22 11:42:19 2008 +0100 11437 11438 Xv: NV12 overlay copy - removed wrong byteswapping on PPC, reported by moondrake 11439 11440commit 782237e4e942f2e8a9ed3dd111d60c665459963c 11441Author: Stuart Bennett <sb476@cam.ac.uk> 11442Date: Mon Jan 21 23:44:47 2008 +0000 11443 11444 randr12: some tidyup 11445 11446commit 91ca67056cb236e5770d4fb39dd7fbb9818ff093 11447Author: Stuart Bennett <sb476@cam.ac.uk> 11448Date: Mon Jan 21 22:35:58 2008 +0000 11449 11450 randr12: more beautiful dithering 11451 11452commit 1d5dc47324ddd1939db166af60df7a3a0e055810 11453Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11454Date: Mon Jan 21 22:19:24 2008 +0100 11455 11456 oops :) 11457 11458commit 7080f93a83e0d680f53a7962df300d527edd121a 11459Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11460Date: Mon Jan 21 21:43:11 2008 +0100 11461 11462 Quickly hacked support for NV6x. 11463 11464commit 2d399af61424cc4b4e0080ea016423f1b22c02f3 11465Author: Stuart Bennett <sb476@cam.ac.uk> 11466Date: Mon Jan 21 20:16:56 2008 +0000 11467 11468 randr12: fix dual head restore 11469 11470commit 8be20d6fc21a3869786e8c19656bf8b6569fee6e 11471Author: Stuart Bennett <sb476@cam.ac.uk> 11472Date: Mon Jan 21 18:51:52 2008 +0000 11473 11474 randr12: save and restore palette 11475 11476commit d32f310afa05ad85d6675348400880506d98ac1e 11477Author: Stuart Bennett <sb476@cam.ac.uk> 11478Date: Mon Jan 21 15:03:59 2008 +0000 11479 11480 Make LVDS bios script logs easier to understand 11481 11482 and fix a backwards comparison 11483 11484commit 7c7f0194678c7bd7c5ce791c07b38ac8eeeef608 11485Author: Stuart Bennett <sb476@cam.ac.uk> 11486Date: Mon Jan 21 14:14:07 2008 +0000 11487 11488 randr12: avoid SIGFPE 11489 11490commit b9a3aab7a7f6c5c4a5c440e46bbca7d76fd8df68 11491Author: Stuart Bennett <sb476@cam.ac.uk> 11492Date: Mon Jan 21 14:09:46 2008 +0000 11493 11494 debug_modes breaks build on old xservers 11495 11496commit d2e078b4ebf11b286e152e0ea10b17c41bf61b67 11497Author: Stuart Bennett <sb476@cam.ac.uk> 11498Date: Mon Jan 21 14:06:42 2008 +0000 11499 11500 randr12: use correct clock for LVDS modesetting part 2 11501 11502commit 52154e97b5a40b014fbee747fc542f765b4390bc 11503Author: Stuart Bennett <sb476@cam.ac.uk> 11504Date: Mon Jan 21 13:46:15 2008 +0000 11505 11506 randr12: use correct clock for LVDS modesetting 11507 11508commit d54a9f4b76f3a8c29c5fce6dfe7cc3fbc20f3952 11509Author: Stuart Bennett <sb476@cam.ac.uk> 11510Date: Mon Jan 21 12:27:36 2008 +0000 11511 11512 Don't log strange characters 11513 11514commit 07e01061e9b90348ea38b8889ff9e8daf207b820 11515Author: Stuart Bennett <sb476@cam.ac.uk> 11516Date: Sun Jan 20 21:53:30 2008 +0000 11517 11518 Use more imaginative log levels than X_INFO 11519 11520commit 18017632cbc96f36bbc72b454bd5b48070985a6f 11521Author: Stuart Bennett <sb476@cam.ac.uk> 11522Date: Sun Jan 20 20:17:28 2008 +0000 11523 11524 Use stdbool.h types in nv_bios.c 11525 11526 It makes vim highlighting work better and reduces shift key wear. 11527 Coincidentally it's what the drm uses too. 11528 11529 bool, true and false are defined in IEEE 1003.1 (2003) and POSIX since C99 11530 (as is stdint.h, which we use already) 11531 11532commit e1714a4f4329728cf897f8953acc480b47a4e5d7 11533Author: root <root@dpish.hayton.korsenowski> 11534Date: Sun Jan 20 18:38:11 2008 +0000 11535 11536 randr12: fix early cards 11537 11538commit 271844d3a761af65c9625d1564e1c08a9f64d159 11539Author: Maarten Maathuis <madman2003@gmail.com> 11540Date: Sun Jan 20 19:06:49 2008 +0100 11541 11542 NV40TEX: Don't sync when compositing, will have bad sideeffects if the compositing manager tries it too. 11543 11544commit c338a14ea688953b03f274d0e436b579d0411f18 11545Author: Maarten Maathuis <madman2003@gmail.com> 11546Date: Sun Jan 20 17:28:13 2008 +0100 11547 11548 Some misc cleanup. 11549 11550commit 2cd6e157cebac663f1b3c88e3653a4c2c9971ff1 11551Author: Maarten Maathuis <madman2003@gmail.com> 11552Date: Sun Jan 20 17:11:27 2008 +0100 11553 11554 randr12: Fix a major screwup i made during a cleanup. 11555 11556commit c3127a0c9cf4599c40b27e24a411a7fe4969152a 11557Author: Maarten Maathuis <madman2003@gmail.com> 11558Date: Sun Jan 20 17:00:58 2008 +0100 11559 11560 NV40TEX: Implement sync to vblank. 11561 11562 - 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). 11563 - This should also work reasonably well for composited situations, but ofcource we can't guarantee that. 11564 11565commit 512f4ba33528dafb505bc2db5f9bb4d1b92c3a7b 11566Author: Stuart Bennett <sb476@cam.ac.uk> 11567Date: Sun Jan 20 14:47:52 2008 +0000 11568 11569 randr12: init some more Xv state 11570 11571commit 3f6a681a38fccbd53dd10aa42cc8d1b9bfe0eb71 11572Author: Stuart Bennett <sb476@cam.ac.uk> 11573Date: Sat Jan 19 14:02:43 2008 +0000 11574 11575 minor debug correction 11576 11577commit efd3cd086036cd5f8817aee771ee630066a92830 11578Author: Stuart Bennett <sb476@cam.ac.uk> 11579Date: Sun Jan 20 01:56:00 2008 +0000 11580 11581 INIT_8E opcode 11582 11583 again, from pq's trace 11584 11585commit f0bb436796a2d4dc693ec92b73c7abb27d428cda 11586Author: Stuart Bennett <sb476@cam.ac.uk> 11587Date: Sat Jan 19 01:37:34 2008 +0000 11588 11589 INIT_IO opcode 11590 11591 Based on pq's g8x trace 11592 11593commit c1b69fdaa089898821338b14c1f72eda354c2caa 11594Author: Maarten Maathuis <madman2003@gmail.com> 11595Date: Sat Jan 19 19:24:04 2008 +0100 11596 11597 randr12: hide cursor before locking + misc cleanup. 11598 11599commit dfa8901e847e1fbcbcb6de0c0726f53531b15b34 11600Author: Maarten Maathuis <madman2003@gmail.com> 11601Date: Sat Jan 19 16:39:26 2008 +0100 11602 11603 randr12: Remove unneeded vgahw stuff. 11604 11605commit 47452756724007318eb384912c40302b15523998 11606Author: Maarten Maathuis <madman2003@gmail.com> 11607Date: Sat Jan 19 16:29:57 2008 +0100 11608 11609 randr12: Move away from some of the nv_hw wrappers. 11610 11611commit 9e3a2d7c8c36d4dc43396be6b742b6137dfef455 11612Author: Maarten Maathuis <madman2003@gmail.com> 11613Date: Sat Jan 19 15:59:48 2008 +0100 11614 11615 randr12: Remove usage of nvRead/WriteVGA0/1. 11616 11617commit 936e9fe8adb2660cf75959265971c60f378cd7d6 11618Author: Maarten Maathuis <madman2003@gmail.com> 11619Date: Sat Jan 19 11:28:41 2008 +0100 11620 11621 randr12: Some cleanup. 11622 11623 - framebuffer consoles can use most of the normal stuff, so change it back. 11624 - some cursor deobfuscation 11625 11626commit a9bae6ba734882308ac29a9fe2403eaa027bd518 11627Author: Stuart Bennett <sb476@cam.ac.uk> 11628Date: Fri Jan 18 20:49:58 2008 +0000 11629 11630 More detailed modesetting debug (both randr12 and oldstyle) 11631 11632commit a28216639c3cf01d8f7cc85030d69fd58be86ed9 11633Author: Stuart Bennett <sb476@cam.ac.uk> 11634Date: Fri Jan 18 18:10:40 2008 +0000 11635 11636 Remove no longer needed panel quirk 11637 11638commit 59fec415ef32c291363408d5c8a24a2041450434 11639Author: Stuart Bennett <sb476@cam.ac.uk> 11640Date: Fri Jan 18 17:02:32 2008 +0000 11641 11642 Use bios PLL functions on nv4x 11643 11644 And a minor completeness correction 11645 11646commit 7efaf3d97a51d2447310954c9224fd7ec888caf9 11647Author: Maarten Maathuis <madman2003@gmail.com> 11648Date: Fri Jan 18 23:45:51 2008 +0100 11649 11650 randr12: Refine CR4B, anyone know what this does? 11651 11652 - There is more to this register. 11653 11654commit 41646042ab9d9d0c3bc8a76b25a4bff45f039922 11655Author: Stuart Bennett <sb476@cam.ac.uk> 11656Date: Fri Jan 18 16:50:50 2008 +0000 11657 11658 Rerefactor getting pll limits for nv4x 11659 11660commit 174d8f7e015250af4c0b27e044f5088f5465d851 11661Author: Stuart Bennett <sb476@cam.ac.uk> 11662Date: Fri Jan 18 14:24:48 2008 +0000 11663 11664 Use pll limits infrastructure on single stage plls 11665 11666 And add default PLL limits on old cards 11667 11668commit b3ca42d8589bed843ce5e772a1807c4c95377b92 11669Author: Stuart Bennett <sb476@cam.ac.uk> 11670Date: Fri Jan 18 01:47:05 2008 +0000 11671 11672 Don't speculate on the unknown limit field, and call the others log2p, not p 11673 11674commit d9db305e2d4bfb017a8542a7407ad4df27c86121 11675Author: Stuart Bennett <sb476@cam.ac.uk> 11676Date: Fri Jan 18 01:18:42 2008 +0000 11677 11678 Better setPLL_single 11679 11680commit 2a94184112cf06225672ef34836e19723761140a 11681Author: Stuart Bennett <sb476@cam.ac.uk> 11682Date: Fri Jan 18 14:33:01 2008 +0000 11683 11684 Bios parsing safety checks 11685 11686commit 6251f2fa774b913f5622ff9d07f56b2a5b9d1888 11687Author: Maarten Maathuis <madman2003@gmail.com> 11688Date: Fri Jan 18 12:17:50 2008 +0100 11689 11690 Some misc fixes. 11691 11692commit b21789f93c7e7df9db320dea093deab0f1056955 11693Author: Maarten Maathuis <madman2003@gmail.com> 11694Date: Fri Jan 18 09:23:39 2008 +0100 11695 11696 randr12: Actually using SavedReg does tend to help ;-) 11697 11698commit f15ea75b88b68fadb64fdeef75a23329e2a3f903 11699Author: Maarten Maathuis <madman2003@gmail.com> 11700Date: Fri Jan 18 09:13:24 2008 +0100 11701 11702 randr12: Fix a huge gaping hole in NewRestore. 11703 11704 - Restore CR5758 in a better place and actually restore something proper (use SavedReg instead of ModeReg). 11705 - Minor other fix. 11706 11707commit cd427da27d69db000f07ec50d27ca56ded97994e 11708Author: Stuart Bennett <sb476@cam.ac.uk> 11709Date: Thu Jan 17 23:37:23 2008 +0000 11710 11711 Invent a DVI-A output for the DVI-D output, on DCB1.4/1.5 11712 11713commit fbcb57827a1c4ac4b3c7ce66ca935f6b095f020d 11714Author: Stuart Bennett <sb476@cam.ac.uk> 11715Date: Thu Jan 17 22:49:37 2008 +0000 11716 11717 And another DCB 1.4 entry 11718 11719commit a355e5b0dcd5f5c78da20bd07f581b5841b3a2ba 11720Author: Maarten Maathuis <madman2003@gmail.com> 11721Date: Thu Jan 17 22:25:12 2008 +0100 11722 11723 randr12: Forgot to clear a few state variables on restore. 11724 11725commit ba0d0e8571b657b5196d0508347f7641cd7f04cb 11726Author: Stuart Bennett <sb476@cam.ac.uk> 11727Date: Thu Jan 17 20:48:16 2008 +0000 11728 11729 Add me 11730 11731commit 9cfc736ea4b089c74daaf1b0fac6393fec459bc0 11732Author: Stuart Bennett <sb476@cam.ac.uk> 11733Date: Thu Jan 17 20:29:30 2008 +0000 11734 11735 That warning wasn't there a moment ago 11736 11737commit 7eded6883404ce0cff83051497518a99c1d8a832 11738Author: Stuart Bennett <sb476@cam.ac.uk> 11739Date: Thu Jan 17 17:07:24 2008 +0000 11740 11741 Fix for situations with fixed second stage gain and hook code in to init scripts 11742 11743 And make sure bad PLL values are easily seen 11744 11745commit 075a14ac2525b364edc583caac50e03d171f67a7 11746Author: Stuart Bennett <sb476@cam.ac.uk> 11747Date: Thu Jan 17 19:59:55 2008 +0000 11748 11749 PLL limits reg off-by-four is dealt with elsewhere 11750 11751 Also some more imaginative pll limit member naming 11752 and minor get_pll_limits debugging corrections 11753 11754commit a3ef2cb34ad9de4567d30625a3bf87720a5ed8da 11755Author: Stuart Bennett <sb476@cam.ac.uk> 11756Date: Thu Jan 17 17:30:19 2008 +0000 11757 11758 setPLL for nv40 lowreg PLLs 11759 11760commit c3bef786b5291a638791b8a5125b3ff60cf9d1b1 11761Author: Stuart Bennett <sb476@cam.ac.uk> 11762Date: Thu Jan 17 17:30:36 2008 +0000 11763 11764 setPLL for nv40 highreg PLLs 11765 11766commit 996b93cdc73e3d6e581a766cbbf4a29bb34b8eb2 11767Author: Maarten Maathuis <madman2003@gmail.com> 11768Date: Thu Jan 17 18:27:51 2008 +0100 11769 11770 randr12: Some misc fixes. 11771 11772commit aa561c487754210ca22a1fc1d17c54d8e199e7a1 11773Author: Maarten Maathuis <madman2003@gmail.com> 11774Date: Wed Jan 16 22:42:18 2008 +0100 11775 11776 Some misc fixes, mostly related to setting the right crtc for overlay (randr12). 11777 11778commit 276d4e35673bb71410c882badf03550bcee48571 11779Author: Maarten Maathuis <madman2003@gmail.com> 11780Date: Wed Jan 16 08:03:51 2008 +0100 11781 11782 randr12: Fix mode privates again, hopefully ;-) 11783 11784commit f89243e4c2ba33a822b15de9136736f44b4c1579 11785Author: Stuart Bennett <sb476@cam.ac.uk> 11786Date: Wed Jan 16 00:25:08 2008 +0000 11787 11788 Minor reg adjustment 11789 11790commit df0f66750d1d893b2d91b0cfc91ecd4f0a1ba80e 11791Author: Stuart Bennett <sb476@cam.ac.uk> 11792Date: Tue Jan 15 16:30:00 2008 +0000 11793 11794 Refactor pll table parsing 11795 11796 And correct table reg matching on version 0x21 tables, which is off-by-4 11797 11798commit 3171b470f00a11d3528067c7d91e751257aa603c 11799Author: Stuart Bennett <sb476@cam.ac.uk> 11800Date: Tue Jan 15 14:59:49 2008 +0000 11801 11802 Use pll_lims in getMNP_double 11803 11804 Also move quirks elsewhere, and optimize 11805 11806commit 7bc921485c63232fecb59e488ef203166c48c6e1 11807Author: Stuart Bennett <sb476@cam.ac.uk> 11808Date: Tue Jan 15 15:14:46 2008 +0000 11809 11810 Store PLL limit frequencies in kHz, since that's what we use them in 11811 11812commit 51a739771b59a4725e75660b2c8edb54aa763ba5 11813Author: Stuart Bennett <sb476@cam.ac.uk> 11814Date: Tue Jan 15 01:32:25 2008 +0000 11815 11816 Two stage PLL limits parsing corrections 11817 11818 And some corrections to getMNP_double 11819 11820commit 7013116596ce8b065df265f16b3fdb9ac9e3df37 11821Author: Stuart Bennett <sb476@cam.ac.uk> 11822Date: Tue Jan 15 09:53:58 2008 +0000 11823 11824 randr12: Oh no! More pNv->twoHeads 11825 11826commit c7081039bf27aad12e17063d1a9becf160e7ee49 11827Author: Maarten Maathuis <madman2003@gmail.com> 11828Date: Tue Jan 15 21:05:08 2008 +0100 11829 11830 randr12: PrivFlags are not so private, switch to a real mode Private. 11831 11832 - This hopefully fixes the obscure issues people have had these last few days. 11833 11834commit cd14cd8c520361a15f5f8d255171c85810d83033 11835Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 11836Date: Tue Jan 15 20:44:48 2008 +0100 11837 11838 Order the reg ranges, add some. 11839 11840commit 17fa42166fb4e8102cbe880be6021f1145268d16 11841Author: Patrice Mandin <pmandin@caramail.com> 11842Date: Tue Jan 15 18:45:30 2008 +0100 11843 11844 nv30: set VIEWPORT_TX_ORIGIN to 0 for all cards, except real nv30 11845 11846commit 777a3df246feb364d440282db0864fb9b0908872 11847Author: Maarten Maathuis <madman2003@gmail.com> 11848Date: Tue Jan 15 17:52:19 2008 +0100 11849 11850 NV30EXA: Chipset should be masked by 0xFFF0, switch to NVArch while i'm at it. 11851 11852commit b55608bac3f4f909f1ef8380a1ac69ca2f262ea2 11853Author: Maarten Maathuis <madman2003@gmail.com> 11854Date: Tue Jan 15 08:10:46 2008 +0100 11855 11856 randr12: Reenable cvtmode. 11857 11858commit 0db297bb6199878c0f77935acc447c568a87c900 11859Author: Maarten Maathuis <madman2003@gmail.com> 11860Date: Mon Jan 14 22:54:42 2008 +0100 11861 11862 randr12: Undo some changes of the past + always set vpll on modeset. 11863 11864commit 2879f4518e6830cf0b490cde5b4daca3e931383d 11865Author: Patrice Mandin <pmandin@caramail.com> 11866Date: Mon Jan 14 23:26:14 2008 +0100 11867 11868 nv30: Only change VIEWPOR_TX_ORIGIN on real nv30 11869 11870commit 405fd62ee8b56e3b1d6cc78df7ecfa654cd54dec 11871Author: Maarten Maathuis <madman2003@gmail.com> 11872Date: Mon Jan 14 20:17:05 2008 +0100 11873 11874 randr12: "NewRestore" Always reset some state regs after restore. 11875 11876commit 8093b83c3d1393ba225f9def1ce055b04cba5a8f 11877Author: Maarten Maathuis <madman2003@gmail.com> 11878Date: Mon Jan 14 19:29:28 2008 +0100 11879 11880 randr12: "NewRestore" proper depth vs bpp awareness. 11881 11882commit 8733ea89d9acece8721a2ae2d62a68916746c374 11883Author: Stuart Bennett <sb476@cam.ac.uk> 11884Date: Mon Jan 14 18:21:47 2008 +0000 11885 11886 randr12: yet more pre nv10 fixes 11887 11888commit c0d1117782673771b0490fe4d19abd83c6620033 11889Author: Stuart Bennett <sb476@cam.ac.uk> 11890Date: Mon Jan 14 11:54:28 2008 +0000 11891 11892 Possible fix for C51's misaligned regs 11893 11894commit 6ac824446d65a28519f378b3c01b1a6683a14d1b 11895Author: Stuart Bennett <sb476@cam.ac.uk> 11896Date: Sun Jan 13 21:36:53 2008 +0000 11897 11898 randr12: better load detection (enabled) 11899 11900commit ef28e91b24ee34911b3238e681064a95be7afd3a 11901Author: Maarten Maathuis <madman2003@gmail.com> 11902Date: Mon Jan 14 00:48:25 2008 +0100 11903 11904 randr12: We didn't copy the Flags to the adjusted mode, very bad. 11905 11906commit d41d5ae34a7d7f4658a1d86635fb8c59b80a1355 11907Author: Stuart Bennett <sb476@cam.ac.uk> 11908Date: Sun Jan 13 21:35:33 2008 +0000 11909 11910 Be more flexible on init table length 11911 11912commit 4744e5410c27d8bd3c2f92bee4554801639dba29 11913Author: Stuart Bennett <sb476@cam.ac.uk> 11914Date: Sun Jan 13 19:37:28 2008 +0000 11915 11916 More DCB 1.4/1.5 support 11917 11918commit c043b3fa7cf6297e010dfa6fcff24f263b8a8fd8 11919Author: Maarten Maathuis <madman2003@gmail.com> 11920Date: Sun Jan 13 17:09:04 2008 +0100 11921 11922 randr12: Minor fix. 11923 11924commit 3f7f0cab91416f6f4f4378ee6bdabeb9e75cf230 11925Author: Maarten Maathuis <madman2003@gmail.com> 11926Date: Sun Jan 13 16:05:22 2008 +0100 11927 11928 randr12: Some "NewRestore" fixes for framebuffer modes. 11929 11930 - This is mostly it for framebuffer modes. 11931 11932commit d8e7c5068835004741b74577344bd4defad4d139 11933Author: Maarten Maathuis <madman2003@gmail.com> 11934Date: Sun Jan 13 15:05:15 2008 +0100 11935 11936 nvbios: Fix a bmp bug for malc0. 11937 11938commit 5c01f96caa2b3b7223383f37d6903fbcaf3643cd 11939Author: Maarten Maathuis <madman2003@gmail.com> 11940Date: Sun Jan 13 14:12:00 2008 +0100 11941 11942 randr12: Some "NewRestore" fixes. 11943 11944 - Improvements to restoring framebuffer modes, still a few minor bugs to fix. 11945 - VGA modes work partially, i doubt this will make mainline though, since it's fragile. 11946 - This is a work in progress, only for testing. 11947 - Added some other functions to NOUVEAU_MODESET_TRACE. 11948 11949commit 364497a4234410adea25e4e51e0bdb17ccd3ba4a 11950Author: Stuart Bennett <sb476@cam.ac.uk> 11951Date: Sat Jan 12 00:39:59 2008 +0000 11952 11953 randr12: more fixes aimed at nv0x 11954 11955commit ac546e77fe71ffbbc76d9c48c5ab0b6faac832d7 11956Author: Stuart Bennett <sb476@cam.ac.uk> 11957Date: Fri Jan 11 22:05:11 2008 +0000 11958 11959 Remove some nv_bios TODOs and shut up some warnings 11960 11961commit fc04d71de6ec36158040617fb973218a8bdd0903 11962Author: Maarten Maathuis <madman2003@gmail.com> 11963Date: Fri Jan 11 23:23:43 2008 +0100 11964 11965 randr12: Add new experimental restore mode. 11966 11967 - Option "NewRestore" 11968 - Only tested on my 16bpp 1280x1024 vesafb console. 11969 - It may work on text consoles. 11970 - This should eventually replace the somewhat frail restore system that we currently have. 11971 11972commit 361bbe074a0d674012615b05fd3da81e8e1ce34e 11973Author: Maarten Maathuis <madman2003@gmail.com> 11974Date: Fri Jan 11 22:30:14 2008 +0100 11975 11976 randr12: Intermediate commit. 11977 11978commit 3b6e551e1afefe86338ab830505dc08412b2e80b 11979Author: Maarten Maathuis <madman2003@gmail.com> 11980Date: Fri Jan 11 18:59:53 2008 +0100 11981 11982 randr12: Intermediate commit. 11983 11984commit 5a989b23322e6ba1abe6456246300067c3bf42cb 11985Author: Stuart Bennett <sb476@cam.ac.uk> 11986Date: Fri Jan 11 21:33:47 2008 +0000 11987 11988 Tidy up and correct DCB merging 11989 11990commit ec7889a5a0787929c8c8ae08aaf729929c483035 11991Author: Stuart Bennett <sb476@cam.ac.uk> 11992Date: Thu Jan 10 23:44:59 2008 +0000 11993 11994 Fix bios opcode 0x56 and add reg alignment check 11995 11996commit 9b356d08f9169e35bdbff079c3351c475e53ec9b 11997Author: Stuart Bennett <sb476@cam.ac.uk> 11998Date: Fri Jan 11 16:34:28 2008 +0000 11999 12000 randr12: undo overzealous code disablement 12001 12002commit 44295b59f57be8956a5226415b5fdfbf49e175c1 12003Author: Stuart Bennett <sb476@cam.ac.uk> 12004Date: Fri Jan 11 12:28:16 2008 +0000 12005 12006 randr12: twoHeads and other fixes for pre nv10 (nv_crtc.c) 12007 12008commit 80fa1e1aef1b093b0310c51817ee9685e4d9d251 12009Author: Stuart Bennett <sb476@cam.ac.uk> 12010Date: Fri Jan 11 11:16:21 2008 +0000 12011 12012 randr12: some tidyup to output save/restore functions, and some twoHeads fixes 12013 12014commit 04f64725b5784ecfd7a5c3218d4710c2ad0f7f0f 12015Author: Stuart Bennett <sb476@cam.ac.uk> 12016Date: Fri Jan 11 11:20:45 2008 +0000 12017 12018 randr12: tidy up dpms functions 12019 12020 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 12021 12022commit cc6794abb637e5ac8c98d7c6385bd36aa43d1ebc 12023Author: Stuart Bennett <sb476@cam.ac.uk> 12024Date: Thu Jan 10 20:25:50 2008 +0000 12025 12026 randr12: use clocks in kHz throughout 12027 12028commit 8337b77b40da9c3bba106837b4553433fa4e8a76 12029Author: Stuart Bennett <sb476@cam.ac.uk> 12030Date: Thu Jan 10 19:09:16 2008 +0000 12031 12032 The bios transition freq is in 10s kHz - so x10 12033 12034commit d65f59144f8a79116bca2f09de41d5786a7a4f84 12035Author: Maarten Maathuis <madman2003@gmail.com> 12036Date: Thu Jan 10 18:50:47 2008 +0100 12037 12038 randr12: Some minor things for supporting vga modes (in the future). 12039 12040commit 0e646b0332b52e23ef13301765f7f1e8f881e165 12041Author: Maarten Maathuis <madman2003@gmail.com> 12042Date: Thu Jan 10 17:41:39 2008 +0100 12043 12044 randr12: Intermediate commit. 12045 12046commit a5122460cfa07d45cf91ae112db9170fecee6966 12047Author: Stuart Bennett <sb476@cam.ac.uk> 12048Date: Thu Jan 10 16:21:49 2008 +0000 12049 12050 Remove weirdo special case handling on bios reg writes 12051 12052 I don't think these ever get used, and the code only seems to be on gf6 anyway 12053 12054commit 31b1a894079ca3983d17ac7a81d6ecd420f7d377 12055Author: Stuart Bennett <sb476@cam.ac.uk> 12056Date: Thu Jan 10 14:14:24 2008 +0000 12057 12058 randr12: dtrt wrt dual link on lvds when !straps 12059 12060commit e02d2cc8b77443fde4e267ce4d9691ed680eef01 12061Author: Stuart Bennett <sb476@cam.ac.uk> 12062Date: Thu Jan 10 12:03:43 2008 +0000 12063 12064 randr12: improve setting of fp_control 12065 12066commit d8eb924a43576c0e360965a43f5363d0ba2aecd0 12067Author: Arthur Huillet <arthur.huillet@free.fr> 12068Date: Thu Jan 10 12:03:30 2008 +0100 12069 12070 Xv: cleanup. 12071 12072commit 1f727083cfafa9097ed98266420a2fd16905ae10 12073Author: Stuart Bennett <sb476@cam.ac.uk> 12074Date: Thu Jan 10 00:12:52 2008 +0000 12075 12076 randr12: enable LVDS and refactor output creation 12077 12078 Heavily based on initial patch by seventhguardian 12079 12080commit c08c6de0f452c7f5cd1bac8bf974aaac872c8e2c 12081Author: Stuart Bennett <sb476@cam.ac.uk> 12082Date: Wed Jan 9 16:18:30 2008 +0000 12083 12084 deifdefication - we requre xserver 1.3, so randr 1.2 is guaranteed 12085 12086commit 1556144a1af3c89eaa14cbfe8bdc9dc966fef8be 12087Author: Stuart Bennett <sb476@cam.ac.uk> 12088Date: Wed Jan 9 15:47:50 2008 +0000 12089 12090 randr12: optimize away excessive LVDS_RESETing 12091 12092commit 55e6750038edd7010ae8304fe3c3b4ac9e7b963c 12093Author: Maarten Maathuis <madman2003@gmail.com> 12094Date: Wed Jan 9 22:08:49 2008 +0100 12095 12096 randr12: Fix minor bug, related to output resource conflicts. 12097 12098commit c1fe3766bd181d9ed273850e6c0d1939c0efc8aa 12099Author: Maarten Maathuis <madman2003@gmail.com> 12100Date: Wed Jan 9 20:20:03 2008 +0100 12101 12102 randr12: Reset crtc when changing output routes. 12103 12104commit 01b9a081c288c69fa0bbc0d3829a0026796e731a 12105Author: Maarten Maathuis <madman2003@gmail.com> 12106Date: Wed Jan 9 19:30:17 2008 +0100 12107 12108 randr12: Don't change vpll's twice on restore. 12109 12110commit 3643b2635dcaba3e6ec969a3cd6eda43623c8992 12111Author: Stuart Bennett <sb476@cam.ac.uk> 12112Date: Wed Jan 9 15:12:30 2008 +0000 12113 12114 Detect an LVDS panel if there's a hardcoded EDID 12115 12116commit 64d8032c419ef8cf14cdf3f83228d3e0f5714909 12117Author: Stuart Bennett <sb476@cam.ac.uk> 12118Date: Wed Jan 9 15:11:24 2008 +0000 12119 12120 If no PROM, and no PRAMIN, use the PCI rom image 12121 12122commit 99a64d31979a8c83ac9d52fabe3114c576c94bba 12123Author: Stuart Bennett <sb476@cam.ac.uk> 12124Date: Tue Jan 8 22:58:37 2008 +0000 12125 12126 fp bios edid oops commit 12127 12128commit b470cf257433dad174d62cdbd4331ada2809334e 12129Author: Maarten Maathuis <madman2003@gmail.com> 12130Date: Tue Jan 8 23:54:51 2008 +0100 12131 12132 randr12: More fixes for strange output routes. 12133 12134commit cfbdbd6ac846af79d9ddf2b4597736cba74d5916 12135Author: Stuart Bennett <sb476@cam.ac.uk> 12136Date: Tue Jan 8 20:57:10 2008 +0000 12137 12138 randr12: try to use bios fp edid, if nothing else works 12139 12140commit d4e4892244c90b481bfd8c6d28d2390fe696e160 12141Author: Stuart Bennett <sb476@cam.ac.uk> 12142Date: Tue Jan 8 20:07:10 2008 +0000 12143 12144 Read EDID from BIOS if appropriate 12145 12146commit 6fab536aa13b8639d9b419a37fccfb30a4313b77 12147Author: Stuart Bennett <sb476@cam.ac.uk> 12148Date: Tue Jan 8 16:18:09 2008 +0000 12149 12150 Darktama's C51 is a mobile card 12151 12152commit 921f2e226ccbe7468cc8e51919872c3e8cf28c4e 12153Author: Stuart Bennett <sb476@cam.ac.uk> 12154Date: Tue Jan 8 15:37:51 2008 +0000 12155 12156 BIT cards use the init scripts to do the power sequencing 12157 12158commit 33ea2cdef8c46ee9837b6fc9809df82d4bdca339 12159Author: Maarten Maathuis <madman2003@gmail.com> 12160Date: Tue Jan 8 22:36:12 2008 +0100 12161 12162 randr12: Small fix for NV3x, more will be needed. 12163 12164commit 175b91d48644bdbfc4e48f4c2fe6c9cf936802b5 12165Author: Maarten Maathuis <madman2003@gmail.com> 12166Date: Tue Jan 8 20:23:38 2008 +0100 12167 12168 randr12: Fix reverse pll reading code for NV30. 12169 12170commit 4add8cfb68404e9e2764451c4e48ab4d0cf04482 12171Author: Maarten Maathuis <madman2003@gmail.com> 12172Date: Tue Jan 8 19:07:50 2008 +0100 12173 12174 randr12: Some minor things for NV30. 12175 12176commit 74b03f7d4f5506222b6a95ccf8b5c9e0ff69e4dc 12177Author: Stuart Bennett <sb476@cam.ac.uk> 12178Date: Tue Jan 8 13:52:00 2008 +0000 12179 12180 Assorted small bios/randr12 bits and pieces 12181 12182 Make bios execution a bit faster 12183 Better self documenting code for DCB dual link criterion 12184 Tidy up some ineffectual LVDS mode code 12185 12186commit 76f3c47736661df86490b0d133ac82c371cab73d 12187Author: Maarten Maathuis <madman2003@gmail.com> 12188Date: Mon Jan 7 23:38:08 2008 +0100 12189 12190 randr12: Handle NV30 vpll's correctly. 12191 12192 - This only concerns the real NV30, not the other NV3x cards. 12193 12194commit 1a25def0a64ac6437c6ba5f51ddb8743d4a88d00 12195Author: Maarten Maathuis <madman2003@gmail.com> 12196Date: Mon Jan 7 19:06:16 2008 +0100 12197 12198 randr12: Clear dpms cache when needed. 12199 12200commit 3f0008128e2afd2b1a1b595829c6324e46a4bf1d 12201Author: Arthur Huillet <arthur.huillet@free.fr> 12202Date: Mon Jan 7 19:03:53 2008 +0100 12203 12204 Xv: overlay BW workaround applied to NV40 as well 12205 12206commit d5dabd8c316371605d81be55a5c6c82e6351b591 12207Author: Arthur Huillet <arthur.huillet@free.fr> 12208Date: Mon Jan 7 19:01:07 2008 +0100 12209 12210 Xv: removed useless #includes 12211 12212commit 510f3575df87733857dbb5bfe398ce04aebadfd4 12213Author: Arthur Huillet <arthur.huillet@free.fr> 12214Date: Mon Jan 7 17:13:43 2008 +0100 12215 12216 Xv: added NV04 overlay brightness attribute. Original patch by Andrew Randrianasulu 12217 12218commit a71de657e1e2718800ef1c319dab5dafcc9ea0cd 12219Author: Ben Skeggs <skeggsb@gmail.com> 12220Date: Tue Jan 8 02:18:59 2008 +1100 12221 12222 avoid symbol clash with 3d driver. 12223 12224commit 0463dc0682fb192ce82037601289f75913f65e3c 12225Author: Ben Skeggs <skeggsb@gmail.com> 12226Date: Tue Jan 8 02:16:26 2008 +1100 12227 12228 nv50: support hostdata UTS 12229 12230commit 9fb186c57bdad07281f1a0bdebac98a4b18dc6e3 12231Author: Ben Skeggs <skeggsb@gmail.com> 12232Date: Mon Jan 7 18:33:03 2008 +1100 12233 12234 nv50: small fix 12235 12236commit b679d6a8539403f075a57d882bb4d0eb33b180d9 12237Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12238Date: Mon Jan 7 02:39:19 2008 +0100 12239 12240 nv40 video texture: sometimes less is more. 12241 12242commit b9a364a985bb23ee30dc094ba13e19ea2f4f6bdf 12243Author: Maarten Maathuis <madman2003@gmail.com> 12244Date: Sun Jan 6 20:44:56 2008 +0100 12245 12246 NV40EXA: Remove stair shaped tearing i get. 12247 12248commit 80ac301acaf8dba6a93f5b04b97f8dca2748b92c 12249Author: Pierre Ossman <drzeus@drzeus.cx> 12250Date: Sun Jan 6 20:04:10 2008 +0100 12251 12252 We do not get a valid CRTC configuration structure if we're not using RandR 1.2, 12253 so make sure we can handle the old mode with some form of grace. 12254 12255 Signed-off-by: Pierre Ossman <drzeus@drzeus.cx> 12256 12257commit f1aaf3fe616f5a0633e685867cf33b37f2c3f07b 12258Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12259Date: Sun Jan 6 19:14:03 2008 +0100 12260 12261 nv40 video texture: bilinear filtering for UV. 12262 12263commit 57fc0631428f29af0d24036c939826349e3018b5 12264Author: Arthur Huillet <arthur.huillet@free.fr> 12265Date: Sun Jan 6 17:41:50 2008 +0100 12266 12267 Xv: overlay auto CRTC switch with randr12, blitter fallback 12268 12269 Now switching the overlay to the right CRTC when displaying video with the overlay and dual head, with a nice fallback 12270 to the blitter when the window is on both CRTCs. 12271 Fixed a bug preventing the overlay from working on CRTC1. 12272 12273commit b6ba53ee1cbda5cb91e43ae6b84e1737ebc367f4 12274Author: Jeremy Kolb <jkolb@glorfindel.hsd1.ma.comcast.net> 12275Date: Sun Jan 6 12:58:45 2008 -0500 12276 12277 port "G80: Reduce load detection false positives." from xf86-video-nv commit d483ecd851bb84014f1fa86f0ada4540b3ad801e 12278 12279commit b4fa9b060308090f4bab3f8716f7e2ecaf234be7 12280Author: Jeremy Kolb <jkolb@glorfindel.hsd1.ma.comcast.net> 12281Date: Sun Jan 6 12:51:55 2008 -0500 12282 12283 Return NULL not FALSE. 12284 12285commit 47dde8283bcc409037682098cad8f0a0c3f9e3b2 12286Author: Pierre Ossman <drzeus@drzeus.cx> 12287Date: Sun Jan 6 17:48:48 2008 +0100 12288 12289 Xv: don't wait for vblank on CRTC0 _and_ CRTC1 12290 12291commit c8a9b4d0c2145d13c74a00cb24e1310e247a2718 12292Author: Pierre Ossman <drzeus@drzeus.cx> 12293Date: Sun Jan 6 17:45:37 2008 +0100 12294 12295 wait for correct crtc vsync 12296 12297 Make sure we wait for the CRTC that will actually display the video 12298 frame we're blitting. 12299 12300commit 25fac91dfc7072405243d2891f9ec1f0cef49a3d 12301Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12302Date: Sun Jan 6 16:20:43 2008 +0100 12303 12304 Switch to latest nouveau_class.h 12305 12306commit bc82548514daab356fe0e0ee942ff75018fbc383 12307Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12308Date: Sun Jan 6 16:19:50 2008 +0100 12309 12310 Use clamp to edge to avoid the thin black lines. 12311 12312commit acaf2501d17030097761b3f8d369f14874261f6d 12313Author: Maarten Maathuis <madman2003@gmail.com> 12314Date: Sun Jan 6 13:17:49 2008 +0100 12315 12316 randr12: Some fixes for NV31M. 12317 12318commit 7b1521c263511016538286806f54221721c60608 12319Author: Maarten Maathuis <madman2003@gmail.com> 12320Date: Sat Jan 5 19:43:08 2008 +0100 12321 12322 randr12: NV11 and NV20 stop at CR52. 12323 12324commit 9441ca0fd579cb648577f107777236d22a2b1ead 12325Author: Maarten Maathuis <madman2003@gmail.com> 12326Date: Sat Jan 5 19:32:32 2008 +0100 12327 12328 randr12: NV11 and NV20 don't have CR57 and CR58. 12329 12330commit 8f520754459c363313353605207f7598e259ea77 12331Author: Maarten Maathuis <madman2003@gmail.com> 12332Date: Sat Jan 5 18:54:14 2008 +0100 12333 12334 Add a missing function to NOUVEAU_MODESET_TRACE. 12335 12336commit d5bf6a0149dc80d5d76ab8857415d20b42188656 12337Author: Stuart Bennett <sb476@cam.ac.uk> 12338Date: Sat Jan 5 17:22:33 2008 +0000 12339 12340 randr12: fix PLL breaking typo 12341 12342commit 21f02f5f248bce5f49e0fcee5594c83155b532b5 12343Author: Stuart Bennett <sb476@cam.ac.uk> 12344Date: Sat Jan 5 03:10:21 2008 +0000 12345 12346 Need to restore EXTDEV before fp table parsing 12347 12348 and a nice hwsq message 12349 12350commit 4e4819f6722a120d3a22eb9829fe73780a0c1087 12351Author: Maarten Maathuis <madman2003@gmail.com> 12352Date: Sat Jan 5 18:30:08 2008 +0100 12353 12354 randr12: Minor cleanup. 12355 12356commit 758e1cf14d61725d69ca75b7268f22e11a32a5ba 12357Author: Maarten Maathuis <madman2003@gmail.com> 12358Date: Sat Jan 5 18:27:10 2008 +0100 12359 12360 randr12: Fix a thinko, generate data before restoring, not at the beginning. 12361 12362commit f315ce85fa33894767892772c1e7a0966150ba6b 12363Author: Maarten Maathuis <madman2003@gmail.com> 12364Date: Sat Jan 5 14:05:55 2008 +0100 12365 12366 randr12: Some NV40 quirks. 12367 12368commit 08b1a2622708be9f714c27fd31372e28277d7c5b 12369Author: Maarten Maathuis <madman2003@gmail.com> 12370Date: Sat Jan 5 13:03:40 2008 +0100 12371 12372 randr12: Small quirk for NV11. 12373 12374commit 73101dbcfac68688914e2a3f46008ddd69d0dc62 12375Author: Stuart Bennett <sb476@cam.ac.uk> 12376Date: Sat Jan 5 02:46:52 2008 +0000 12377 12378 Load NV17 power sequencing microcode 12379 12380commit 2ec6cf4a6e59b861c1672471872fc4ca75cfce27 12381Author: Maarten Maathuis <madman2003@gmail.com> 12382Date: Sat Jan 5 01:22:42 2008 +0100 12383 12384 randr12: Cache dpms calls to avoid unnecesary blinking and such. 12385 12386commit 4f586ac169e7acd0e0bfa4139bd74e718d7598e2 12387Author: Arthur Huillet <arthur.huillet@free.fr> 12388Date: Sat Jan 5 01:38:52 2008 +0200 12389 12390 Xv: linearly interpolate U and V samples when converting from YV12 to YUY2 12391 12392 Original patch by ahuillet, pq did one last fix. 12393 12394commit 5f1a721070c20a4287d17a8797ef1aabe19a579e 12395Author: Stuart Bennett <sb476@cam.ac.uk> 12396Date: Fri Jan 4 17:59:54 2008 +0000 12397 12398 More DCB 1.4/1.5 fixes 12399 12400commit 95485646eb61aaa84a8d3b6550552ab417be8e16 12401Author: Maarten Maathuis <madman2003@gmail.com> 12402Date: Fri Jan 4 18:33:31 2008 +0100 12403 12404 randr12: Forgot to switch to vpll1_a, etc in one place. 12405 12406commit c43bb4990f44d50d91785f8c9860e27c3357c0a3 12407Author: Stuart Bennett <sb476@cam.ac.uk> 12408Date: Fri Jan 4 01:14:40 2008 +0000 12409 12410 randr12: fix BIT bios LVDS dpms 12411 12412commit 5d9e42bbb5878f52cf7fd6d96532253d868b2a27 12413Author: Stuart Bennett <sb476@cam.ac.uk> 12414Date: Fri Jan 4 13:26:38 2008 +0000 12415 12416 Better guesses at DCB 1.4/1.5 12417 12418commit 73da4362cd7e2918bd066528a9feb157019edb5e 12419Author: Maarten Maathuis <madman2003@gmail.com> 12420Date: Fri Jan 4 11:54:27 2008 +0100 12421 12422 randr12: Remove unnecesary usleep. 12423 12424 - I once did this thinking it may solve a bug, but it turned out a palette issue. 12425 12426commit c35f532fdee6141e871560f8a400040c2eb82891 12427Author: Maarten Maathuis <madman2003@gmail.com> 12428Date: Fri Jan 4 11:43:16 2008 +0100 12429 12430 NV40: Improve VCO2 detection in nvGetClocks() 12431 12432commit 9369c03788d7410150044d2e3a83f0bcb876af42 12433Author: Maarten Maathuis <madman2003@gmail.com> 12434Date: Fri Jan 4 11:39:37 2008 +0100 12435 12436 randr12: Improve speed of mode switches. 12437 12438 - Moved pre-NV40 to vpll1_a, vpll1_b, etc variables. 12439 - Only change vpll when needed. 12440 12441commit ec0fe5294fdee7a79e9bfaf21ea42cfd3f52ec75 12442Author: Stuart Bennett <sb476@cam.ac.uk> 12443Date: Fri Jan 4 02:31:21 2008 +0000 12444 12445 Infra for regs stored over suspend and a sort-of implementation of INIT_COMPUTE_MEM 12446 12447commit 786e133577edcf91b519dd8dede24250ec6fdcdf 12448Author: Stuart Bennett <sb476@cam.ac.uk> 12449Date: Thu Jan 3 02:21:35 2008 +0000 12450 12451 Better bios register read function prototype and a ReadEXTDEV purge 12452 12453commit 951f95ba553f301de2a540ef67cf04152cf72344 12454Author: Stuart Bennett <sb476@cam.ac.uk> 12455Date: Thu Jan 3 00:52:12 2008 +0000 12456 12457 Refactor NVParseBios 12458 12459commit 8b82f24e8a526c0483120f9979a3b9370da3b297 12460Author: Stuart Bennett <sb476@cam.ac.uk> 12461Date: Thu Jan 3 00:32:24 2008 +0000 12462 12463 Revert "randr12: LVDS screens use center mode." 12464 12465 This reverts commit 1142b15c56a0b9dab8b929fc196a825815457374. 12466 12467 They don't, necessarily 12468 12469commit 876d7b11c754dd66738bcc26a07e0d707b81a99f 12470Author: Maarten Maathuis <madman2003@gmail.com> 12471Date: Thu Jan 3 22:45:00 2008 +0100 12472 12473 NV40: Handle broken VCO2 values. 12474 12475commit 50eb355e44017a51636c1e7b8c140bc3b48813af 12476Author: Maarten Maathuis <madman2003@gmail.com> 12477Date: Thu Jan 3 22:36:04 2008 +0100 12478 12479 NV40: A better attempt at YUV->RGB conversion constants + remove vblank sync. 12480 12481 - The vblank sync didn't work (made things worse actually), someone needs to reverse engineer the proper way for the 3d engine. 12482 12483commit 22cdfec9b0858c186ab37ff649248a2a825eb513 12484Author: Maarten Maathuis <madman2003@gmail.com> 12485Date: Thu Jan 3 12:48:43 2008 +0100 12486 12487 NV40: Avoid a SIGFPE in nvGetClocks() when VCO2 is off. 12488 12489commit 5015e2cefcf2069ee0054c433c7ae1763d2f95eb 12490Author: Maarten Maathuis <madman2003@gmail.com> 12491Date: Thu Jan 3 12:37:10 2008 +0100 12492 12493 NV40: Actually add the attribute. 12494 12495commit 5b60344486c8ddeba28204dbea6f1e11704cf951 12496Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 12497Date: Thu Jan 3 05:56:23 2008 +0100 12498 12499 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. 12500 12501commit f7a356250a9a0cb214dbb1871696506c04fab946 12502Author: Maarten Maathuis <madman2003@gmail.com> 12503Date: Wed Jan 2 12:15:21 2008 +0100 12504 12505 randr12: Hopefully these were all the function declerations that need updating. 12506 12507 - Build didn't even show any errors, so that's why i missed them. 12508 12509commit aa608573918ed34968054fb5a93b11fc5c8aa1a8 12510Author: Maarten Maathuis <madman2003@gmail.com> 12511Date: Wed Jan 2 00:19:12 2008 +0100 12512 12513 Dump some CARD8 and CARD32 in favour of uintX_t. 12514 12515commit b44af66323b622411d5ffa4a78a210abefdb1e2f 12516Author: Maarten Maathuis <madman2003@gmail.com> 12517Date: Tue Jan 1 22:38:34 2008 +0100 12518 12519 randr12: Small fix for NV4x cards with strange output routes. 12520 12521commit 6b204d5ef3ff61236ef86eadcc7e137a722aef72 12522Author: Maarten Maathuis <madman2003@gmail.com> 12523Date: Tue Jan 1 21:51:31 2008 +0100 12524 12525 randr12: Improve output routing. 12526 12527commit 64546788e50ae3482698f9505869e06f727c0e50 12528Author: Maarten Maathuis <madman2003@gmail.com> 12529Date: Tue Jan 1 21:07:36 2008 +0100 12530 12531 randr12: Really hopefully fix output resource conflict detection now. 12532 12533commit b71a52ae9689026e7d96d2c2752eec0c9854b0c3 12534Author: Maarten Maathuis <madman2003@gmail.com> 12535Date: Tue Jan 1 20:38:24 2008 +0100 12536 12537 randr12: Hopefully fix output resource conflict detection. 12538 12539commit 9ac2765ff13f82746991c85649f45c13475e851d 12540Author: Maarten Maathuis <madman2003@gmail.com> 12541Date: Tue Jan 1 19:41:07 2008 +0100 12542 12543 Forgot the NOUVEAU_EXA_PIXMAPS case. 12544 12545commit 53565fc3e854abc0b5e7614da251024a60ddfd9e 12546Author: Maarten Maathuis <madman2003@gmail.com> 12547Date: Tue Jan 1 19:36:28 2008 +0100 12548 12549 Stricter alignment requirements for offscreenBase, to avoid framebuffer corruption. 12550 12551commit c9cedb7712719e3dee450245ebeb7db18cd40770 12552Author: Maarten Maathuis <madman2003@gmail.com> 12553Date: Tue Jan 1 17:36:20 2008 +0100 12554 12555 randr12: Time to let some routing insanity die. 12556 12557 - I hope nvidia is just crazy sometimes when setting CRTC1 on both output resources. 12558 - Or is it some override used on load detect? 12559 12560commit 9f1494ebf1ec1957f362366ab4abfc31c2227092 12561Author: Maarten Maathuis <madman2003@gmail.com> 12562Date: Tue Jan 1 17:28:41 2008 +0100 12563 12564 randr12: Always set bit 26 in FP_DEBUG_0 if already present. 12565 12566commit a83c71244b8dceb2cac968195777636bd5c5f639 12567Author: Maarten Maathuis <madman2003@gmail.com> 12568Date: Tue Jan 1 17:01:19 2008 +0100 12569 12570 randr12: Minor tweak to flatpanel regs. 12571 12572commit afe174cd57173c22011718d3290db5d939076ce6 12573Author: Maarten Maathuis <madman2003@gmail.com> 12574Date: Tue Jan 1 14:25:54 2008 +0100 12575 12576 Remove AM_MAINTAINER_MODE from configure.ac 12577 12578commit ef1a613e0f1a1fffb54a50fd14c10069ed9eac4b 12579Author: Maarten Maathuis <madman2003@gmail.com> 12580Date: Mon Dec 31 20:24:41 2007 +0100 12581 12582 randr12: Fix a bug in an earlier commit. 12583 12584 - This was local, but i wanted to keep some things seperate. 12585 12586commit 01313e3dca07615fa3b2a2b09f6d45489cc8d796 12587Author: Maarten Maathuis <madman2003@gmail.com> 12588Date: Mon Dec 31 20:15:03 2007 +0100 12589 12590 randr12: Add a quirk for 1280x800 panel on 7300go. 12591 12592commit 78bf63088d5805762460801c59f0f93b43ec96a7 12593Author: Maarten Maathuis <madman2003@gmail.com> 12594Date: Mon Dec 31 19:34:33 2007 +0100 12595 12596 randr12: Duplicate native modes, to avoid hitting an invalid pointer. 12597 12598commit 83aad1ad6e8a5e48ee91346b4c7037ddc8456a73 12599Author: Maarten Maathuis <madman2003@gmail.com> 12600Date: Mon Dec 31 18:57:49 2007 +0100 12601 12602 randr12: Set the adjusted mode in output code. 12603 12604commit 23d9d05a68cf2eac51b500d48d31bae3b484bbe8 12605Author: Maarten Maathuis <madman2003@gmail.com> 12606Date: Mon Dec 31 14:20:41 2007 +0100 12607 12608 randr12: Deal with output resources in a better way. 12609 12610commit a4d973214158efa90a1a23da879630d2b53ad313 12611Author: Maarten Maathuis <madman2003@gmail.com> 12612Date: Mon Dec 31 13:18:16 2007 +0100 12613 12614 randr12: Disable crt load detect again, until i know how to use it properly. 12615 12616commit 59346e15688186b4c00ee5348413ae8312ac9a30 12617Author: Maarten Maathuis <madman2003@gmail.com> 12618Date: Sun Dec 30 19:11:48 2007 +0100 12619 12620 Revert "randr12: Another quirk for the 7300GO cards." 12621 12622 This reverts commit 2f2921e6c33a0482f1f30e4d69820dc82690310a. 12623 It was unneeded after all. 12624 12625commit 2f2921e6c33a0482f1f30e4d69820dc82690310a 12626Author: Maarten Maathuis <madman2003@gmail.com> 12627Date: Sun Dec 30 17:07:13 2007 +0100 12628 12629 randr12: Another quirk for the 7300GO cards. 12630 12631commit 642baacdab215f6ad7173757b8d7e00f716d6dbe 12632Author: Maarten Maathuis <madman2003@gmail.com> 12633Date: Sun Dec 30 16:44:23 2007 +0100 12634 12635 randr12: Fix quad view on some 7300GO cards. 12636 12637commit df36ea6d9cd2a22b51a0386bbfe4f6760ed98d97 12638Author: Maarten Maathuis <madman2003@gmail.com> 12639Date: Sun Dec 30 15:01:47 2007 +0100 12640 12641 Add some headerfiles to Makefile.am 12642 12643commit 29851581b2420570efd4fca2d4f6a19113ea6947 12644Author: Maarten Maathuis <madman2003@gmail.com> 12645Date: Sun Dec 30 13:35:47 2007 +0100 12646 12647 randr12: Only load palette's when appropriate. 12648 12649commit c3359789d8f288b403aedbcdfa8795391fc13deb 12650Author: Maarten Maathuis <madman2003@gmail.com> 12651Date: Sun Dec 30 13:26:24 2007 +0100 12652 12653 NV40: Some fixes and safeties. 12654 12655commit a7c922b58af72e3a48b92056c2fcd71c22477c2f 12656Author: Maarten Maathuis <madman2003@gmail.com> 12657Date: Sun Dec 30 11:36:10 2007 +0100 12658 12659 NV40: Fix some comments and a damage reporting bug. 12660 12661commit b6cd9964f9c20c40eda4283a93e447601beddb96 12662Author: Maarten Maathuis <madman2003@gmail.com> 12663Date: Sun Dec 30 02:31:38 2007 +0100 12664 12665 NV40: Fix the texturing adapter when not rendering offscreen. 12666 12667 - Now clip restrictions are taken into consideration. 12668 - No streching when clipping the border of a screen anymore. 12669 12670commit 4efdd1b560228aae7f6f5d6ed549461f8fdf0d00 12671Author: Maarten Maathuis <madman2003@gmail.com> 12672Date: Sat Dec 29 22:34:32 2007 +0100 12673 12674 randr12: Minor cleanup. 12675 12676commit 5af6cd8dea2fa51b362201961191a17b0813fa30 12677Author: Maarten Maathuis <madman2003@gmail.com> 12678Date: Sat Dec 29 21:46:18 2007 +0100 12679 12680 randr12: Micro cleanup. 12681 12682commit 4f91dd1ae08d348b2977b8ede0908ddd2add1cb4 12683Author: Maarten Maathuis <madman2003@gmail.com> 12684Date: Sat Dec 29 20:15:25 2007 +0100 12685 12686 NVBIOS: Fix duallink detection. 12687 12688commit 559a640a542a89238afba02a78cf6399b8edc2fc 12689Author: Maarten Maathuis <madman2003@gmail.com> 12690Date: Sat Dec 29 19:34:12 2007 +0100 12691 12692 randr12: Do not remove bits in DITHER register. 12693 12694commit 4b980011742a6fa0bdeb312a7b367d6a56675fcb 12695Author: Maarten Maathuis <madman2003@gmail.com> 12696Date: Sat Dec 29 15:09:08 2007 +0100 12697 12698 NV40: Preliminary Xv Texture Adaptor. 12699 12700 - Only supports YV12. 12701 - No attributes. 12702 - Only the Y component gets bilinear filtering. 12703 - You're welcome to improve it ;-) 12704 12705commit a4c045aa14b471fcfe29cdbec0cd846b2c5be2a6 12706Author: Peter Winters <peter@whoei.org> 12707Date: Sat Dec 29 12:15:40 2007 +0100 12708 12709 Xv: nforce2 overlay can do yv12 12710 12711commit 1142b15c56a0b9dab8b929fc196a825815457374 12712Author: Maarten Maathuis <madman2003@gmail.com> 12713Date: Fri Dec 28 21:11:49 2007 +0100 12714 12715 randr12: LVDS screens use center mode. 12716 12717commit 8215e970445d99312c8fbbc31944466f70f3e186 12718Author: Andrew Randrianasulu <randrianasulu@yahoo.com> 12719Date: Fri Dec 28 20:59:17 2007 +0100 12720 12721 Fix IFC on nv4/5 by using the proper object. 12722 12723commit c6ae3fe9f077c38f616fd3ad07275dcf8cb65bbc 12724Author: Stuart Bennett <sb476@cam.ac.uk> 12725Date: Fri Dec 28 05:45:10 2007 +0000 12726 12727 Make sure the CRTC regs are unlocked when doing bios init 12728 12729 Also make the single stage PLL setter less exploitable by the whims of the hardware 12730 12731commit bb467ad4baa30fc46a05bbce11d6aa4e3dac89a9 12732Author: Stuart Bennett <sb476@cam.ac.uk> 12733Date: Fri Dec 28 08:23:17 2007 +0000 12734 12735 Fake a DCB entry for cards from the pre-DCB era 12736 12737 Also, drop DCB parsing to after the init parsing again; having an INIT_TMDS in 12738 the init scripts would be silly (the original reason for bumping DCB parsing up) 12739 12740commit cdfe4fef9bd7d0a69d9ba19a79779152baef5489 12741Author: Stuart Bennett <sb476@cam.ac.uk> 12742Date: Fri Dec 28 07:36:58 2007 +0000 12743 12744 Madness. Bios parsing for NV05 12745 12746commit d4e12684617d07dc859e774f3eff9b944bc4c89e 12747Author: Maarten Maathuis <madman2003@gmail.com> 12748Date: Thu Dec 27 20:01:14 2007 +0100 12749 12750 randr12: Minor change. 12751 12752commit ab9dc943dd5036545e3396868c0332e9b727f31f 12753Author: Maarten Maathuis <madman2003@gmail.com> 12754Date: Thu Dec 27 19:50:27 2007 +0100 12755 12756 randr12: Small change. 12757 12758commit 66d343ad594f1019eb8f4d23308eb97dd1bbd94d 12759Author: Maarten Maathuis <madman2003@gmail.com> 12760Date: Thu Dec 27 19:25:00 2007 +0100 12761 12762 randr12: minor change 12763 12764commit cf5072e0c1d9dbee4f22ef190082aa1db78dbb0d 12765Author: Maarten Maathuis <madman2003@gmail.com> 12766Date: Thu Dec 27 18:37:30 2007 +0100 12767 12768 randr12: Undo unnecesary change. 12769 12770commit a7c643601073de5a63f718ddcdc9eb9f4873f2bf 12771Author: Maarten Maathuis <madman2003@gmail.com> 12772Date: Thu Dec 27 17:12:17 2007 +0100 12773 12774 randr12: Some fixes for LVDS and an odd 7300GO. 12775 12776commit b17f8ca13604f6b085b474219843d740b31918e8 12777Author: Stuart Bennett <sb476@cam.ac.uk> 12778Date: Wed Dec 26 20:59:36 2007 +0000 12779 12780 Fix the setPLL optional code to actually compile 12781 12782commit 05b929ccd9c7cd50f60b84bf987fc5220ee83008 12783Author: Stuart Bennett <sb476@cam.ac.uk> 12784Date: Wed Dec 26 21:31:16 2007 +0000 12785 12786 Fix up getting modes for both TMDS and LVDS 12787 12788commit ba92eb062aaf8b53c7f5c1858eb876d3e2f9af3e 12789Author: Maarten Maathuis <madman2003@gmail.com> 12790Date: Wed Dec 26 21:21:15 2007 +0100 12791 12792 randr12: Remove this, since it's not essential and can cause problems. 12793 12794commit 150d3362c5e0a300c35f1cb0c7da0c98cdd6b4de 12795Author: Maarten Maathuis <madman2003@gmail.com> 12796Date: Wed Dec 26 20:28:43 2007 +0100 12797 12798 randr12: Some minor fixed for dvi and lvds. 12799 12800commit 076f86dd67d6e69bd7d2a68f372cbbbb35785a3c 12801Author: Maarten Maathuis <madman2003@gmail.com> 12802Date: Wed Dec 26 14:51:49 2007 +0100 12803 12804 randr12: Some assorted fixes for dual link outputs. 12805 12806commit 15489f15e8130ffa74bee7c52086857554a69a6b 12807Author: Stuart Bennett <sb476@cam.ac.uk> 12808Date: Tue Dec 25 23:50:10 2007 +0000 12809 12810 On some chips we shouldn't look for a 27MHz crystal 12811 12812commit 8fa59a6231e293a8b669517bf11d35dcc15f34ff 12813Author: Maarten Maathuis <madman2003@gmail.com> 12814Date: Tue Dec 25 19:43:47 2007 +0100 12815 12816 randr12: Some minor changes. 12817 12818commit 80bdd0e9f4bd664e217722a57a8db79d56968194 12819Author: Maarten Maathuis <madman2003@gmail.com> 12820Date: Tue Dec 25 11:58:09 2007 +0100 12821 12822 randr12: Restrict TMDS register backup to digital outputs. 12823 12824 - The NV11 locks up when accessing these and it has no digital outputs, so ;-) 12825 12826commit 5574e96ae3e4608b95b5dd5f2dd5676386bb4ffe 12827Author: Stuart Bennett <sb476@cam.ac.uk> 12828Date: Mon Dec 24 22:33:21 2007 +0000 12829 12830 Better (not complete) support for DCB1.1/1.2 12831 12832commit 2aa6a2347fa283dbd9b10258053602bd37880c26 12833Author: Stuart Bennett <sb476@cam.ac.uk> 12834Date: Mon Dec 24 21:00:31 2007 +0000 12835 12836 "type1" bios scripts 12837 12838commit e0c5f65c0091b0f0ac7715fbf399a1f6fea60dc6 12839Author: Stuart Bennett <sb476@cam.ac.uk> 12840Date: Mon Dec 24 18:13:47 2007 +0000 12841 12842 Rehabilitate BMP parsing 12843 12844commit 2789da6b34c9495a88d349b73cb838f4626fb152 12845Author: Stuart Bennett <sb476@cam.ac.uk> 12846Date: Mon Dec 24 22:51:51 2007 +0000 12847 12848 Fix BMP length for version 5.16 12849 12850commit 93448b40627d35fd89d8fd5629dae5f487082f43 12851Author: Stuart Bennett <sb476@cam.ac.uk> 12852Date: Mon Dec 24 15:23:41 2007 +0000 12853 12854 randr12: use new PLL code on < NV40 12855 12856commit 3d90e13c26e389f43253a5d2721ce93a628f156d 12857Author: Stuart Bennett <sb476@cam.ac.uk> 12858Date: Mon Dec 24 05:31:37 2007 +0000 12859 12860 setPLL routines for single and two stage plls 12861 12862 plus some parsing fixes 12863 12864commit eabe1e02504467c7425e62673a68f4dc7dd83aab 12865Author: Stuart Bennett <sb476@cam.ac.uk> 12866Date: Mon Dec 24 02:29:54 2007 +0000 12867 12868 Two stage nv31 PLL mnp calculator and bios limits parsing 12869 12870commit c8b50bc7dc146e01ceb67ee3633b5c8958345fcd 12871Author: Stuart Bennett <sb476@cam.ac.uk> 12872Date: Sun Dec 23 03:09:26 2007 +0000 12873 12874 Single stage PLL MNP calculator 12875 12876commit 935f809f4bc9dab37e80c59007141d7b9ad673a8 12877Author: Maarten Maathuis <madman2003@gmail.com> 12878Date: Sun Dec 23 22:05:10 2007 +0100 12879 12880 Reading into the -1 index of array is a bad idea, this should fix subchannel binding issues. 12881 12882commit f357ba26b7f560cdec03f70a2c3b79b65d70172b 12883Author: Maarten Maathuis <madman2003@gmail.com> 12884Date: Sun Dec 23 15:48:24 2007 +0100 12885 12886 randr12: Add modesetting debug option (showing most register writes). 12887 12888 - Contribution of ahuillet. 12889 12890commit 95387b9779173cc3e51a7fffc60975010253b391 12891Author: Maarten Maathuis <madman2003@gmail.com> 12892Date: Sun Dec 23 15:35:24 2007 +0100 12893 12894 NV40EXA: Suport as many of the repeat types as possible + random cleanup(randr12). 12895 12896commit 8e80f7504f872166a811b6ce20663aaf97fd7181 12897Author: Arthur Huillet <arthur.huillet@free.fr> 12898Date: Sun Dec 23 13:49:24 2007 +0100 12899 12900 Revert a manifestation of my stupidity. 12901 12902commit 86a3a5fd40005a4860d11f0046444ba232db0d1e 12903Author: Arthur Huillet <arthur.huillet@free.fr> 12904Date: Sun Dec 23 13:47:08 2007 +0100 12905 12906 Xv: added an attribute to change the CRTC the overlay is on - useful with dualhead 12907 12908commit 129f04612cdbad756dbe213f77d2f12482b62c47 12909Author: Arthur Huillet <arthur.huillet@free.fr> 12910Date: Sat Dec 22 23:32:36 2007 +0100 12911 12912 aunedsoau~ 12913 12914commit 26953280895698a94c1f886cf7c0d2f389167b1c 12915Author: Maarten Maathuis <madman2003@gmail.com> 12916Date: Sun Dec 23 11:46:17 2007 +0100 12917 12918 randr12: Extend dvi clock setting to nv3x also + cleanup. 12919 12920commit 063d1bd50b164e841e3a725b75ac243827aede50 12921Author: Maarten Maathuis <madman2003@gmail.com> 12922Date: Sun Dec 23 10:46:09 2007 +0100 12923 12924 randr12: Actually load the default palette on modeset. 12925 12926commit 810901abb50a54b8f1eb73029f5348805a53531d 12927Author: Maarten Maathuis <madman2003@gmail.com> 12928Date: Sun Dec 23 10:32:18 2007 +0100 12929 12930 randr12: Simplify code a bit by using the more sane code for NV4x also. 12931 12932commit eb8441824ea2a7dc9a1d7a86a4305fae48a26a0c 12933Author: Maarten Maathuis <madman2003@gmail.com> 12934Date: Sun Dec 23 01:02:18 2007 +0100 12935 12936 Minor cleanup. 12937 12938commit 9eef67abaf8beb9ed93eb5b6ecdf7255f5b06e4c 12939Author: Maarten Maathuis <madman2003@gmail.com> 12940Date: Sun Dec 23 00:18:13 2007 +0100 12941 12942 randr12: Always enable MPLL and NVPLL on pre-NV41 cards. 12943 12944commit fa82eb23904bc1a70209778d2aafcbef6f27cfa1 12945Author: Maarten Maathuis <madman2003@gmail.com> 12946Date: Sun Dec 23 00:07:01 2007 +0100 12947 12948 randr12: Put all NV2x (and lower) cards into restricted mode. 12949 12950commit 33d7313d7dca4501b7ac2e078a87f3a10ab1b501 12951Author: Maarten Maathuis <madman2003@gmail.com> 12952Date: Sat Dec 22 23:55:19 2007 +0100 12953 12954 randr12: Correct CRTC selection for pre-NV40 cards. 12955 12956commit d6243fe33223d2934c1a819c58bb58b9ebe15c25 12957Author: Maarten Maathuis <madman2003@gmail.com> 12958Date: Sat Dec 22 23:39:51 2007 +0100 12959 12960 randr12: Older cards do have switchable CRTC's, i was again wrong. 12961 12962commit a96553544ae76239e394ab34b33814a57252d31c 12963Author: Maarten Maathuis <madman2003@gmail.com> 12964Date: Sat Dec 22 23:23:32 2007 +0100 12965 12966 randr12: Minor cleanup to clock writing for pre-NV40 cards. 12967 12968commit 9b63f42bc7da9b45ecbbd32ef81cae55d2a57220 12969Author: Arthur Huillet <arthur.huillet@free.fr> 12970Date: Sat Dec 22 18:44:36 2007 +0100 12971 12972 NV10EXA: disabled A8+A8 hack on PPC machines. 12973 12974commit 308fec0111d182fb17536ff2746e079c1af32d5a 12975Author: Ben Skeggs <skeggsb@gmail.com> 12976Date: Sat Dec 22 23:18:57 2007 +1100 12977 12978 nv30: Don't touch 0x1a4 12979 12980commit 62916c7c8c6b467436e4628c7d08a4e0de1f21f3 12981Author: Maarten Maathuis <madman2003@gmail.com> 12982Date: Sat Dec 22 12:02:31 2007 +0100 12983 12984 randr12: Undo a previous mistake for NV1x/NV2x. 12985 12986commit 6f9eb17637e2c3b8d49a07332f06a0102d2fe016 12987Author: Stuart Bennett <sb476@cam.ac.uk> 12988Date: Fri Dec 21 20:18:57 2007 +0000 12989 12990 Beginnings of PLL limits for BMP bioses 12991 12992 And the correct way to use the debug_modes member (still disabled for now) 12993 12994commit 73bc1db51e9431c5d1861d8a753aa8dd9abe9882 12995Author: Stuart Bennett <sb476@cam.ac.uk> 12996Date: Sat Dec 22 02:57:40 2007 +0000 12997 12998 On demand PLL limit parsing for NV40 12999 13000 Also some nv_bios message formatting consistancy pedantry 13001 13002commit 57b5dbfac4ff65258e94a17717815156a54cc273 13003Author: Maarten Maathuis <madman2003@gmail.com> 13004Date: Sat Dec 22 01:29:38 2007 +0100 13005 13006 NV40EXA: Fix borders of images that recieved arbitrary rotation. 13007 13008 - It was getting padded up to the edge of the smallest square that could fit the rotated image. 13009 - Please try this on older cards as well. 13010 13011commit 39ef0262be49feaed2ba1c6caeab5134181ef0b9 13012Author: Maarten Maathuis <madman2003@gmail.com> 13013Date: Sat Dec 22 01:26:44 2007 +0100 13014 13015 randr12: Some minor changes. 13016 13017commit 371b6e52bef743f14003e3783ea3dbc624050594 13018Author: Maarten Maathuis <madman2003@gmail.com> 13019Date: Fri Dec 21 18:02:33 2007 +0100 13020 13021 randr12: Restrict crtc's on nv1x and nv2x. 13022 13023 - 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. 13024 13025commit c2ac36927827439cf92ff53bbf2414bfaf45a6c3 13026Author: Maarten Maathuis <madman2003@gmail.com> 13027Date: Fri Dec 21 00:05:37 2007 +0100 13028 13029 randr12: Detect switchable crtc's. 13030 13031commit 33a2cf86520f42d549d03c46134b818e90d51a41 13032Author: Peter Winters <peter@whoei.org> 13033Date: Thu Dec 20 23:47:11 2007 +0100 13034 13035 NV10 EXA: use the proper variable 13036 13037commit 71c3b8ff09adcadb10d800d3696762c3ef58be3d 13038Author: Stuart Bennett <sb476@cam.ac.uk> 13039Date: Wed Dec 19 23:44:23 2007 +0000 13040 13041 randr12: fix old and new style scaling for LVDS, and use an enum for the scaling modes 13042 13043commit 6bea51b905a461b4e703053082ab45d9e1074112 13044Author: Maarten Maathuis <madman2003@gmail.com> 13045Date: Thu Dec 20 00:07:29 2007 +0100 13046 13047 randr12: Fix secondary (analog) outputs with ffs(or) != 3 on some NV4x cards. 13048 13049commit e0e04eb561f4b127087befc4e3768d8012a82bfe 13050Author: Stuart Bennett <sb476@cam.ac.uk> 13051Date: Wed Dec 19 04:29:34 2007 +0000 13052 13053 randr12: don't use invalid crtc pointer for lvds dpms 13054 13055commit a68ac97979a3178ea210470bbb3d64f3f9325e07 13056Author: Stuart Bennett <sb476@cam.ac.uk> 13057Date: Wed Dec 19 04:00:14 2007 +0000 13058 13059 Turn LVDS panel on again, after resetting, when we're shutting down 13060 13061commit 1cfd659fc1a5f8dc2faba30155fca46bbe31f3a4 13062Author: Stuart Bennett <sb476@cam.ac.uk> 13063Date: Tue Dec 18 16:58:57 2007 +0000 13064 13065 minor link_head_and_output cleanup 13066 13067commit 73a5d09b3e9fa13d04d6696ca38b6402795acc3d 13068Author: Stuart Bennett <sb476@cam.ac.uk> 13069Date: Wed Dec 19 03:43:59 2007 +0000 13070 13071 randr12: CRTC member wasn't big enough 13072 13073 yay gcc. could fix console colour breakage 13074 13075commit 0a63198195f689b32388b99db61ff942e16fbc6d 13076Author: Maarten Maathuis <madman2003@gmail.com> 13077Date: Tue Dec 18 23:43:57 2007 +0100 13078 13079 NV10EXA: Fix rotation. 13080 13081commit f2a3807618cfeac01d7d28e9d76721ad61495f7f 13082Author: Maarten Maathuis <madman2003@gmail.com> 13083Date: Tue Dec 18 23:23:59 2007 +0100 13084 13085 randr12: Add new pll calculation for NV4x cards (from NVClock). 13086 13087commit ebf96c8b651d1a0dc5a19576cf57efcbb989d51d 13088Author: Maarten Maathuis <madman2003@gmail.com> 13089Date: Tue Dec 18 21:38:56 2007 +0100 13090 13091 randr12: Hopefully reset outputs better between mode changes. 13092 13093commit 5561f54f94e0b31394ecaf461e342470ca768281 13094Author: Maarten Maathuis <madman2003@gmail.com> 13095Date: Tue Dec 18 07:49:05 2007 +0100 13096 13097 Some endianess fixes for the previously committed code. 13098 13099commit 11104ce9c499f30fb5abe29e65ba1918c563f4cd 13100Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 13101Date: Tue Dec 18 07:40:08 2007 +0100 13102 13103 Rename NV30TCL to NV34TCL in preparation of the merge. 13104 13105commit 3e6ea7af5b1613b9e0e4e29433910d998def4899 13106Author: Maarten Maathuis <madman2003@gmail.com> 13107Date: Mon Dec 17 23:13:12 2007 +0100 13108 13109 Add pll limits bios detection code for NV4x cards. 13110 13111commit e56512d283f401a75c6165a0a02e8a7a5905fc4c 13112Author: Maarten Maathuis <madman2003@gmail.com> 13113Date: Mon Dec 17 19:45:42 2007 +0100 13114 13115 NV30EXA: Fix rotation. 13116 13117commit 9ff42b7a83da06a3b9b7076ca888ea447eaa16f3 13118Author: Maarten Maathuis <madman2003@gmail.com> 13119Date: Mon Dec 17 17:50:11 2007 +0100 13120 13121 randr12: Fixes + cleanup. 13122 13123commit 83d99bc3e15d11d3476c54053f6bbe0ff9f094be 13124Author: Maarten Maathuis <madman2003@gmail.com> 13125Date: Mon Dec 17 00:40:24 2007 +0100 13126 13127 randr12: Some minor additions. 13128 13129commit 60c08fedec09dd036b4165881b3001874c4d349a 13130Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 13131Date: Sun Dec 16 23:40:26 2007 +0100 13132 13133 nv30 exa : handle gart case. 13134 13135commit efb10849e136048c9c49cce752d4073402219c8c 13136Author: Maarten Maathuis <madman2003@gmail.com> 13137Date: Sun Dec 16 23:26:49 2007 +0100 13138 13139 Undo a bad change. 13140 13141commit dd8beae22883822ea3555122b0352d4e3cf2c266 13142Author: Maarten Maathuis <madman2003@gmail.com> 13143Date: Sun Dec 16 21:35:18 2007 +0100 13144 13145 A little more cleanup. 13146 13147commit 9ff169ba62dcb2eeefce8a020060de731b72cb6c 13148Author: Maarten Maathuis <madman2003@gmail.com> 13149Date: Sun Dec 16 21:32:32 2007 +0100 13150 13151 More cleanup. 13152 13153commit 2d03d4b999a0a37ff2f9fc87bd20959e789a152e 13154Author: Maarten Maathuis <madman2003@gmail.com> 13155Date: Sun Dec 16 21:19:13 2007 +0100 13156 13157 randr12: Select sensible maximum heights and widths for framebuffer. 13158 13159commit 0e47fa4ab35e529926f5fdccdb44f9d054e1886a 13160Author: Maarten Maathuis <madman2003@gmail.com> 13161Date: Sun Dec 16 21:17:04 2007 +0100 13162 13163 Some reindenting and minor cleanups. 13164 13165commit 2be4b73905dc5767e593bf4753607144fe67b7ab 13166Author: Maarten Maathuis <madman2003@gmail.com> 13167Date: Sun Dec 16 20:24:26 2007 +0100 13168 13169 Some cleanup. 13170 13171commit 6686adbea6638dd894eef697e044941352cc2674 13172Author: Maarten Maathuis <madman2003@gmail.com> 13173Date: Sun Dec 16 20:10:39 2007 +0100 13174 13175 randr12: Restrict digital outputs with ffs(or) == 1 to primary crtc + some cleanup. 13176 13177 - The blob does not reveal what the possible way is (so far), and guesswork didn't help either. 13178 13179commit 9c8da2f562ca571a56224d7fecfb9c3fb3a8d3d0 13180Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 13181Date: Sun Dec 16 18:33:52 2007 +0100 13182 13183 And of course, that comment has to go. 13184 13185commit 02e9a01064930a7081569474be9544fa1f20a0f4 13186Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 13187Date: Sun Dec 16 18:32:52 2007 +0100 13188 13189 Fix front buffer alignment issues (probably needed because of tiling). 13190 13191commit b134b8f1552066b6706521a9c85d9241d99b3ccc 13192Author: Maarten Maathuis <madman2003@gmail.com> 13193Date: Sun Dec 16 13:41:38 2007 +0100 13194 13195 randr12: Fix compile error. 13196 13197commit ec0c9cc9a69c00a960e98e796c00d35d996cf360 13198Author: Maarten Maathuis <madman2003@gmail.com> 13199Date: Sun Dec 16 12:55:29 2007 +0100 13200 13201 randr12: A minor change to OUTPUT register. 13202 13203commit 69782a57285bb53df73c7f45d29882ae990f6054 13204Author: Patrice Mandin <pmandin@caramail.com> 13205Date: Sun Dec 16 00:09:45 2007 +0100 13206 13207 nv30: revert viewport clip change, add myself, add unknown command 0x2b8 (viewport origin?) 13208 13209commit 8dd238f3ce2e7eda7151670ea50a75abdb5b6217 13210Author: Maarten Maathuis <madman2003@gmail.com> 13211Date: Sat Dec 15 23:49:58 2007 +0100 13212 13213 randr12: Some minor changes. 13214 13215commit 08a50d878571f7c9579b15703a08c5c2e52924f2 13216Author: Maarten Maathuis <madman2003@gmail.com> 13217Date: Sat Dec 15 19:54:11 2007 +0100 13218 13219 randr12: Some minor reordering of load/save functions + don't lock crtc when leaving X. 13220 13221commit 4209c06800b97e98261c10e0a625a68f7352ca6c 13222Author: Maarten Maathuis <madman2003@gmail.com> 13223Date: Sat Dec 15 14:48:04 2007 +0100 13224 13225 randr12: Also support accelerated rotation on older xservers (and default code). 13226 13227commit 933b40a2911486a78615984fb17e9ea6757fcfa8 13228Author: Ben Skeggs <skeggsb@gmail.com> 13229Date: Sat Dec 15 23:45:03 2007 +1100 13230 13231 nv40: just for stillunknown... 13232 13233 Fix composite transforms. This is getting to a point where it may be 13234 worth doing this in the vertex program perhaps. For now this works, 13235 it'll do. 13236 13237commit 7212400070264e27a53564a7f5fc95dd9c99f82e 13238Author: Ben Skeggs <skeggsb@gmail.com> 13239Date: Sat Dec 15 23:44:17 2007 +1100 13240 13241 Revert "NV40EXA: Fix composite rotation." 13242 13243 This reverts commit 02d4538b207cf318737ace8c3d8ded76a54e5886. 13244 13245commit 1a622257687a499ce74b3a608618d957784dccf0 13246Author: Maarten Maathuis <madman2003@gmail.com> 13247Date: Sat Dec 15 13:39:23 2007 +0100 13248 13249 randr12: Avoid a vram/memory leak in some situations. 13250 13251commit 02d4538b207cf318737ace8c3d8ded76a54e5886 13252Author: Maarten Maathuis <madman2003@gmail.com> 13253Date: Sat Dec 15 13:35:45 2007 +0100 13254 13255 NV40EXA: Fix composite rotation. 13256 13257 - Thanks to radeon (code) for inspiration and sample code. 13258 13259commit 77d47348dfe01ca3972a683d3c478f910301e910 13260Author: Maarten Maathuis <madman2003@gmail.com> 13261Date: Sat Dec 15 12:02:36 2007 +0100 13262 13263 randr12: Add some ifdefs to make compiling better on older xserver's. 13264 13265commit dbb7efc5b81145ad61d8e6502b918a1e909f3cfb 13266Author: Patrice Mandin <pmandin@caramail.com> 13267Date: Sat Dec 15 11:44:41 2007 +0100 13268 13269 nv30: need to set default depth range 13270 13271commit 6ab8436df0865aab7f12cd8527600353b4728df7 13272Author: Patrice Mandin <pmandin@caramail.com> 13273Date: Sat Dec 15 11:37:43 2007 +0100 13274 13275 nv30: only 4 texture units 13276 13277commit 370d97a4578528c26c3f1546a9522caf40aa2151 13278Author: Patrice Mandin <pmandin@caramail.com> 13279Date: Sat Dec 15 10:34:26 2007 +0100 13280 13281 nv30: fix viewport clipping setup 13282 13283commit 1190d046dfef97dea04d72c993f90a203d81d86c 13284Author: Maarten Maathuis <madman2003@gmail.com> 13285Date: Sat Dec 15 02:00:37 2007 +0100 13286 13287 Fix ugly bug, that creeped in. 13288 13289commit 46f321985ae57457edcedf80f82ba23e53fd9cd2 13290Author: Maarten Maathuis <madman2003@gmail.com> 13291Date: Sat Dec 15 01:44:59 2007 +0100 13292 13293 randr12: Preliminary support for rotation, please read notes. 13294 13295 - It works, but it's slow, very slow. 13296 - Each crtc now has it's own cursor map (fixing the bugs you never saw). 13297 - Dualhead with one rotated screen may show minor corruption near the edges, this will disappear if a fullscreen app opens. 13298 - Acceleration is possible, BUT: 13299 - You need a git xserver. 13300 - Enable NOUVEAU_EXA_PIXMAPS in nouveau_local.h (*maybe* this will become default later). 13301 - Most of the composite hooks have problems with rotation, so left and right won't work, inverted does work for me. 13302 - So please fix the composite hooks ;-) 13303 - Note: No slowdowns are apparent when using proper acceleration. 13304 13305commit 0f07ec3f681b87762fbe7328f47b628711bdbaae 13306Author: Stuart Bennett <sb476@cam.ac.uk> 13307Date: Sat Dec 15 00:01:04 2007 +0000 13308 13309 Better restoration of TMDS regs on LVDS 13310 13311commit 34be8d005fbfff373cc7fe52ca2fa09f5cd5ef00 13312Author: Maarten Maathuis <madman2003@gmail.com> 13313Date: Fri Dec 14 23:58:25 2007 +0100 13314 13315 randr12: It seems DVI on laptops wasn't as strange after all. 13316 13317commit d678932a1818c5c5f4a4f3097992626ce9620c0e 13318Author: Stuart Bennett <sb476@cam.ac.uk> 13319Date: Fri Dec 14 01:40:14 2007 +0000 13320 13321 INIT_COPY_NV_REG and INIT_ZM_REG_GROUP_ADDRESS_LATCHED opcodes 13322 13323 Thanks to RAOF for the trace with these 13324 13325 Also correct a doc oopsy, and a bunch of superfluous whitespace 13326 13327commit 5b9dc7ad6eb2f16c4ca8824c2a19c85bbe1f126b 13328Author: Stuart Bennett <sb476@cam.ac.uk> 13329Date: Thu Dec 13 02:30:06 2007 +0000 13330 13331 Invert 18/24 bit interface flag, so that the default value (0) does not imply dithering 13332 13333commit a878b48c7b5a7aafac9aaa877706f5c7570d3057 13334Author: Stuart Bennett <sb476@cam.ac.uk> 13335Date: Thu Dec 13 01:10:12 2007 +0000 13336 13337 Name bios opcodes 4f and 50 13338 13339 Extrapolating from previous names, but marked as non canon 13340 13341commit 59b46e695b1badbf00f2c03bc78ce58f6887f739 13342Author: Stuart Bennett <sb476@cam.ac.uk> 13343Date: Thu Dec 13 01:03:25 2007 +0000 13344 13345 Get VBIOS from pNv (cleanup) 13346 13347 And a minor doc correction 13348 13349commit c45930d0819668612547b4bf1a8d963bfe40e38b 13350Author: Stuart Bennett <sb476@cam.ac.uk> 13351Date: Thu Dec 13 00:15:21 2007 +0000 13352 13353 randr12: some minor cleanups, no actual changes 13354 13355commit 0e778b2d6970f761e5c0b5eb3054e91589cd2c46 13356Author: Stuart Bennett <sb476@cam.ac.uk> 13357Date: Wed Dec 12 15:54:53 2007 +0000 13358 13359 Setup NV40 LVDS 13360 13361commit 269d83cb46075971fd8dc5402589d4da3eb92f22 13362Author: Stuart Bennett <sb476@cam.ac.uk> 13363Date: Wed Dec 12 22:20:43 2007 +0000 13364 13365 INIT_CONDITION_TIME opcode 13366 13367commit 9b3072d195aae383a993526b3ab4350c500c4277 13368Author: Stuart Bennett <sb476@cam.ac.uk> 13369Date: Wed Dec 12 18:08:07 2007 +0000 13370 13371 Opcode 4F 13372 13373commit 8ee100ced7da347904e45ce590350d7b7c47ed36 13374Author: Stuart Bennett <sb476@cam.ac.uk> 13375Date: Mon Dec 10 22:50:40 2007 +0000 13376 13377 FP BIOS mode finding for nv40 13378 13379commit 9131985f5c359bdb077eb52649369a5c06831544 13380Author: Stuart Bennett <sb476@cam.ac.uk> 13381Date: Tue Dec 11 23:23:27 2007 +0000 13382 13383 CR57 doc addition 13384 13385commit d595eb86790432f6a75dec3388080ffc9ffff8d8 13386Author: Maarten Maathuis <madman2003@gmail.com> 13387Date: Wed Dec 12 22:12:33 2007 +0100 13388 13389 randr12: A good guess at the lower bits of SEL_CLK (i hope ;-)). 13390 13391commit c5bc9899c218abe801e024919c89896e8183dff2 13392Author: Maarten Maathuis <madman2003@gmail.com> 13393Date: Wed Dec 12 21:03:34 2007 +0100 13394 13395 randr12: Add some extra safeties (to mode checking). 13396 13397commit 43c05d306485eb5d8c8b9b39f9bd4a3f470f20a2 13398Author: Maarten Maathuis <madman2003@gmail.com> 13399Date: Tue Dec 11 21:43:02 2007 +0100 13400 13401 randr12: Enable dithering on lvds when needed. 13402 13403commit 8752f18d27a6095485e36f4fb20e9252f27a1e1b 13404Author: Maarten Maathuis <madman2003@gmail.com> 13405Date: Tue Dec 11 20:22:23 2007 +0100 13406 13407 randr12: SEL_CLK attempt 10, return to simplicity, which may actually work ;-) 13408 13409commit c1227d3a2b9194d50a4274a48fa0b3bcf2628e36 13410Author: Maarten Maathuis <madman2003@gmail.com> 13411Date: Tue Dec 11 17:52:39 2007 +0100 13412 13413 randr12: (mostly) Some fixes for NV40. 13414 13415 - The NV40 is a strange card, as it partially behaves like an NV3x. 13416 13417commit 34d9e971e17e763aee96b823bb34da54a281339c 13418Author: Maarten Maathuis <madman2003@gmail.com> 13419Date: Tue Dec 11 11:50:28 2007 +0100 13420 13421 randr12: Fix SEL_CLK assignment again (for non-mobile cards) + fix crosswiring. 13422 13423 - I made a good guess for mobile cards, but i need more mmio-traces. 13424 - I accidentally broke crosswiring for my card (and other's like it). 13425 13426commit bf28c3e7d9976f8695198a938397f001de62836b 13427Author: Maarten Maathuis <madman2003@gmail.com> 13428Date: Tue Dec 11 00:04:38 2007 +0100 13429 13430 randr12: Fix an earlier mistake, make a decent guess about dvi on laptops. 13431 13432 - I could use some more mmiotraces of LVDS+DVI on NV4x. 13433 13434commit 58531ebd4a72368b13d3c1564110ef85fb6d6644 13435Author: Maarten Maathuis <madman2003@gmail.com> 13436Date: Mon Dec 10 20:43:26 2007 +0100 13437 13438 randr12: Improve crosswiring for digital outputs + minor changes and cleanup. 13439 13440commit a388cc5fe21f8692e8ee1dbd71caea88ea2ca5df 13441Author: Maarten Maathuis <madman2003@gmail.com> 13442Date: Mon Dec 10 16:31:47 2007 +0100 13443 13444 randr12: Properly set SEL_CLK based on bus, not based on output. 13445 13446 - This hopefully fixes non-working dvi on some desktop cards and on mobile cards. 13447 13448commit f96a6c2d5d251af560de673ccd91b88b3f5cc535 13449Author: Maarten Maathuis <madman2003@gmail.com> 13450Date: Mon Dec 10 00:03:49 2007 +0100 13451 13452 randr12: Add config option for scaling mode. 13453 13454commit fc73ddad60c551289629be5f2f87b39e26f42674 13455Author: Maarten Maathuis <madman2003@gmail.com> 13456Date: Sun Dec 9 22:46:39 2007 +0100 13457 13458 randr12: Some cleanup. 13459 13460commit 2543e301402b04ae94f349e21a7760beb6f5bdc6 13461Author: Stuart Bennett <sb476@cam.ac.uk> 13462Date: Sun Dec 9 19:58:58 2007 +0000 13463 13464 Bios register write function special case and some INIT_RESET improvements 13465 13466 nv32_wr weirdness only seen on NV40+, but it shouldn't hurt for earlier cards 13467 13468commit 4408ce8d2f902a9ace550a045e6022df2936fbce 13469Author: Stuart Bennett <sb476@cam.ac.uk> 13470Date: Sun Dec 9 17:09:03 2007 +0000 13471 13472 Some renaming for bios indexed port io 13473 13474commit 9eb4294e3f5814f675b403078761bf002540868b 13475Author: Maarten Maathuis <madman2003@gmail.com> 13476Date: Sun Dec 9 18:36:26 2007 +0100 13477 13478 randr12: Revert dither value to that used by nv. 13479 13480commit cfbe4bdef7f4ba958462743dc28bf253b456c1a3 13481Author: Maarten Maathuis <madman2003@gmail.com> 13482Date: Sun Dec 9 18:34:10 2007 +0100 13483 13484 randr12: Set FP_CONTROL correctly for NV2x cards. 13485 13486commit d5dd3afc50e491689f765a5e6a72892efa1768a3 13487Author: Maarten Maathuis <madman2003@gmail.com> 13488Date: Sun Dec 9 18:10:29 2007 +0100 13489 13490 randr12: Improve output selection. 13491 13492 - Now detect if there is a dual link capable output at all. 13493 - NV2x cards can't switch crtc on analog outputs. 13494 13495commit fda83a3c489469d2bb4cbd20c205016b19dfdec5 13496Author: Stuart Bennett <sb476@cam.ac.uk> 13497Date: Sun Dec 9 15:51:02 2007 +0000 13498 13499 Additional LVDS bios script sanity check 13500 13501 Avoids current incorrect situation when LVDS attempted on NV40 13502 13503commit 9932e36d4cf2988d1f6f40d615cd247acc21b592 13504Author: Maarten Maathuis <madman2003@gmail.com> 13505Date: Sun Dec 9 16:29:14 2007 +0100 13506 13507 randr12: Fix minor (potentional) bug. 13508 13509commit 3f83d7308f3b0aca4943262b0ba2715a32b61d58 13510Author: Maarten Maathuis <madman2003@gmail.com> 13511Date: Sun Dec 9 12:47:55 2007 +0100 13512 13513 randr12: Add noscale mode and some misc fixes. 13514 13515commit d688008b38609814de9c5c36a2c8bc569455d14e 13516Author: Maarten Maathuis <madman2003@gmail.com> 13517Date: Sat Dec 8 18:34:04 2007 +0100 13518 13519 randr12: Fix typo, it's "our" not "or". 13520 13521commit 0f7c018af0c7f929c2bf8efb12c510d6e3170885 13522Author: Maarten Maathuis <madman2003@gmail.com> 13523Date: Sat Dec 8 18:25:41 2007 +0100 13524 13525 randr12: Assuming an output always has a crtc is a bad idea. 13526 13527 - Don't rely on current crtc to restore mode. 13528 13529commit 86f248cc2e52de652b1768e15a4ba62eefefd81c 13530Author: Maarten Maathuis <madman2003@gmail.com> 13531Date: Sat Dec 8 16:49:21 2007 +0100 13532 13533 randr12: Avoid that two outputs have the same clock selected. 13534 13535commit 6be67d6cf0f047ceba0aa62989dc44bb58b0db87 13536Author: Maarten Maathuis <madman2003@gmail.com> 13537Date: Sat Dec 8 15:50:39 2007 +0100 13538 13539 randr12: Fix bug in sel_clk assignment. 13540 13541commit ddfec2f1c749af227dd57dbde6751320ba72adb6 13542Author: Stuart Bennett <sb476@cam.ac.uk> 13543Date: Sat Dec 8 02:18:41 2007 +0000 13544 13545 Base error :) 13546 13547commit dff95ae101f701d737c8d2134181693f36178beb 13548Author: Stuart Bennett <sb476@cam.ac.uk> 13549Date: Fri Dec 7 16:56:26 2007 +0000 13550 13551 Fix failure to cope with varying table lengths 13552 13553commit 10731d2a8fe06fa8675c26d201615c2a019de737 13554Author: Stuart Bennett <sb476@cam.ac.uk> 13555Date: Fri Dec 7 02:54:36 2007 +0000 13556 13557 Add a new valid register range, which appears in some gf7 biosen. 13558 13559 NVClock has it down as appearing in gf8 as a new PBUS range, 13560 but the use on gf7 looks PBUS-like too 13561 13562commit b25049233613eff99d767a8ec0c97b5140088213 13563Author: Stuart Bennett <sb476@cam.ac.uk> 13564Date: Fri Dec 7 03:13:52 2007 +0000 13565 13566 Some more bios script opcodes 13567 13568 Thanks to swany for the trace for these 13569 13570commit 6da4923088e8ca1d2de464030205b41d893e65b3 13571Author: Stuart Bennett <sb476@cam.ac.uk> 13572Date: Fri Dec 7 02:17:37 2007 +0000 13573 13574 Some M table parsing, and defer I table execution 13575 13576commit 7e5e477b67279e00d1e56736f9c953972bda4dec 13577Author: Stuart Bennett <sb476@cam.ac.uk> 13578Date: Fri Dec 7 00:18:25 2007 +0000 13579 13580 Improve TMDS04 setting for LVDS 13581 13582commit dd757910f546a97bb42dbefdc7eae058980b1ffb 13583Author: Stuart Bennett <sb476@cam.ac.uk> 13584Date: Thu Dec 6 23:37:01 2007 +0000 13585 13586 TMDS regs setting for pre NV40 13587 13588commit 38f139d473eb3201daaac47e655516f65c7528a6 13589Author: Stuart Bennett <sb476@cam.ac.uk> 13590Date: Fri Dec 7 00:30:14 2007 +0000 13591 13592 BIOS version parsing 13593 13594commit 20b6be6c95a71a41456c4c9b5cb1a1e0238f2e70 13595Author: Stuart Bennett <sb476@cam.ac.uk> 13596Date: Thu Dec 6 17:01:49 2007 +0000 13597 13598 Refactor BIT specific details out of TMDS setting 13599 13600commit f665b4737f1fbbd5e4a6495b561e723f17bc6137 13601Author: Stuart Bennett <sb476@cam.ac.uk> 13602Date: Thu Dec 6 18:48:04 2007 +0000 13603 13604 The 'T' table is clearly a TMDS table, call it as such 13605 13606commit 3759cc9d602b0cd933a0ee7e6f64626822383e19 13607Author: Stuart Bennett <sb476@cam.ac.uk> 13608Date: Thu Dec 6 05:03:49 2007 +0000 13609 13610 The TMDS scripts being non-stubs does not seem to matter - they're still not executed 13611 13612 Message left as log clutter in case there's some future case of broken TMDS when the scripts might be useful 13613 13614commit d6e896dac1018fc3190a5a5283a405c4b031d1ba 13615Author: Stuart Bennett <sb476@cam.ac.uk> 13616Date: Thu Dec 6 04:14:56 2007 +0000 13617 13618 Test for LVDS_INIT script 13619 13620commit c50bc602861fdcdb238e1a4db1146b4870ce83ba 13621Author: Maarten Maathuis <madman2003@gmail.com> 13622Date: Thu Dec 6 21:48:13 2007 +0100 13623 13624 randr12: Small fix for dual link dvi. 13625 13626commit 457d4217624cc73df22ca7f7db0134c54ebac5e5 13627Author: Maarten Maathuis <madman2003@gmail.com> 13628Date: Thu Dec 6 19:54:08 2007 +0100 13629 13630 randr12: Improve compatibility with DCB 2.0 13631 13632commit 5046cb7b3d1af89b72327660932ae06c06d0869b 13633Author: Stuart Bennett <sb476@cam.ac.uk> 13634Date: Thu Dec 6 02:28:18 2007 +0000 13635 13636 randr12: oops, lost the restore case 13637 13638commit 54c7bdf0f3e53c23c0853eac8243407d280d0732 13639Author: Stuart Bennett <sb476@cam.ac.uk> 13640Date: Thu Dec 6 01:54:45 2007 +0000 13641 13642 randr12: a common function to fix TMDS04, and new use of LVDS_RESET 13643 13644 Plus some corrections/additions to call_lvds_script, and a minor CR5758 doc update 13645 13646commit a7811521ed5d227a18af23e2e761ea39fb7eb86f 13647Author: Maarten Maathuis <madman2003@gmail.com> 13648Date: Thu Dec 6 00:28:49 2007 +0100 13649 13650 randr12: Fix comment. 13651 13652commit 4771b4bd4b30f404fb087dd1fd6eac0e12fdd041 13653Author: Maarten Maathuis <madman2003@gmail.com> 13654Date: Thu Dec 6 00:25:48 2007 +0100 13655 13656 randr12: Some assorted fixes, which hopefully improve things for dual link dvi. 13657 13658commit 4d3a015631a0d95392daeec6fd8e99c65c83cb53 13659Author: Maarten Maathuis <madman2003@gmail.com> 13660Date: Wed Dec 5 20:41:27 2007 +0100 13661 13662 randr12: replace preferred_ramdac with preferred_output. 13663 13664commit d3036012db63f9a386d4c1fd8e624b86bcd6c48b 13665Author: Maarten Maathuis <madman2003@gmail.com> 13666Date: Wed Dec 5 20:27:57 2007 +0100 13667 13668 randr12: Clean up the unneeded ramdac variable. 13669 13670commit 37f3a06be4f50c34ee8b82617e5e6eca3a4d66fe 13671Author: Maarten Maathuis <madman2003@gmail.com> 13672Date: Wed Dec 5 19:03:52 2007 +0100 13673 13674 randr12: Pick right register set, to correctly "guess" the bios/VT clock. 13675 13676 - A wrong register set leads to improper detection of crosswiring, thus selecting the wrong clock. 13677 - I noticed a 0.675 MHz difference in the clock (which is vpll2), when instead i should have had vpll1. 13678 13679commit b843dbddbd496ab7d374018e6e05ac1a05ea00e2 13680Author: Maarten Maathuis <madman2003@gmail.com> 13681Date: Wed Dec 5 16:51:04 2007 +0100 13682 13683 randr12: Small fix. 13684 13685commit 4b049994f32058e380543cee81b1d0b76658ea9e 13686Author: Maarten Maathuis <madman2003@gmail.com> 13687Date: Wed Dec 5 16:47:07 2007 +0100 13688 13689 randr12: Final commit, restore TMDS regs before everything else. 13690 13691 - Use the bios to program the tmds registers. 13692 - Restore TMDS registers in nv_crtc_restore, to make sure it's early enough (see code note). 13693 - It works for me, let me know if it breaks stuff ;-) 13694 13695commit 6ff8c1cb2014c7984503a8481cf6ad4f9cccf79a 13696Author: Maarten Maathuis <madman2003@gmail.com> 13697Date: Wed Dec 5 07:57:30 2007 +0100 13698 13699 randr12: Intermediate bios based flat panel init commit, still a few minor issues. 13700 13701commit c14c7deb725c2509c804bae72faa584109f3021c 13702Author: Stuart Bennett <sb476@cam.ac.uk> 13703Date: Wed Dec 5 00:35:35 2007 +0000 13704 13705 Add LVDS script function, and use it for DPMS 13706 13707commit c587731d08bbfd58ff5c2725b7aed4e6477e9f7f 13708Author: Stuart Bennett <sb476@cam.ac.uk> 13709Date: Wed Dec 5 00:22:22 2007 +0000 13710 13711 Parse the LVDS Manufacturer table 13712 13713 Use dual link flag to set bit 31 of FP_CONTROL properly 13714 13715commit affcf6988f1af2c3ea8c47a2f1fdf608cbafcfce 13716Author: Stuart Bennett <sb476@cam.ac.uk> 13717Date: Tue Dec 4 20:31:06 2007 +0000 13718 13719 Modify DCB dual link flag, and DCB config parsing for LVDS 13720 13721 The DCB duallink flag indicates that this setup may be used for duallink; 13722 it obviously can't know whether a duallink display is plugged or not 13723 13724 Use LVDS DCB conf flags for mode detection / selection 13725 13726commit c822e32d95a7afa3d5ef076f97179a1843835f12 13727Author: Maarten Maathuis <madman2003@gmail.com> 13728Date: Tue Dec 4 09:06:10 2007 +0100 13729 13730 randr12: Add fullscreen gpu scaling (only as randr property for the moment). 13731 13732commit a1054c7bef3594f6876d611dde1cf3f0528fb8e1 13733Author: Stuart Bennett <sb476@cam.ac.uk> 13734Date: Tue Dec 4 01:31:24 2007 +0000 13735 13736 Better T table stuff. 13737 13738 Hopefully now usable by calling parse_t_table(...) at the appropriate place 13739 13740 Also collapse some messages, and add an array bound check 13741 13742commit edb9a59ac8d3b16cccd52bde24606c43ae7d36d0 13743Author: Stuart Bennett <sb476@cam.ac.uk> 13744Date: Tue Dec 4 01:40:37 2007 +0000 13745 13746 Allow BIOS execution at runtime, if wanted 13747 13748 And comment the opcodes I haven't double checked yet 13749 13750commit 82fdd3f76e20fc7dd10fcf00cf1c13520dc56364 13751Author: Stuart Bennett <sb476@cam.ac.uk> 13752Date: Tue Dec 4 01:31:24 2007 +0000 13753 13754 Make the bios a bit more accessible 13755 13756 And make a few things use the new structure 13757 13758commit d15bfdfb7e09dce85c3817036ced9efa6ac4522c 13759Author: Stuart Bennett <sb476@cam.ac.uk> 13760Date: Tue Dec 4 00:00:40 2007 +0000 13761 13762 randr12: set CR58 appropriately for 0 and 2 cases 13763 13764commit 02a4f498d903bda55ac38a6e19b1beca0d128e8a 13765Author: Stuart Bennett <sb476@cam.ac.uk> 13766Date: Mon Dec 3 23:35:01 2007 +0000 13767 13768 randr12: refactor dcb usage 13769 13770 heads is also a better name than head 13771 13772commit 1571f81210b48407c9d20264437406d79dfcc784 13773Author: Stuart Bennett <sb476@cam.ac.uk> 13774Date: Mon Dec 3 22:26:33 2007 +0000 13775 13776 randr12: some code economy 13777 13778 Only addition is adding call to output2->funcs->commit when stealing a ramdac, 13779 no logical changes otherwise AFAICT 13780 13781commit 7debf566c7a88258ac8ca925d7138187c1ed8755 13782Author: Stuart Bennett <sb476@cam.ac.uk> 13783Date: Tue Dec 4 00:06:05 2007 +0000 13784 13785 randr12: Improve CR57 and CR58 infrastructure and saving/restoration 13786 13787commit dab828c4793fd7e48e4d8e1d7f8d0a2f8198484c 13788Author: Maarten Maathuis <madman2003@gmail.com> 13789Date: Mon Dec 3 20:55:26 2007 +0100 13790 13791 randr12: A few minor comment changes. 13792 13793commit d959d880b6808adc153c752ea01f3a228879fe3c 13794Author: Maarten Maathuis <madman2003@gmail.com> 13795Date: Mon Dec 3 20:51:16 2007 +0100 13796 13797 randr12: Select CRTC 1 on pre-nv40 (stupid bug) + initial support choosing scaling method on tmds. 13798 13799 - The output property SCALING_MODE accepts "gpu" and "panel". 13800 - The first obeys aspect ratio, the second always goes fullscreen. 13801 - This is just a first attemp at exposing an output property. 13802 13803commit 1d2ea797704d55f63cfe8c00cc000a6b13f1093d 13804Author: Maarten Maathuis <madman2003@gmail.com> 13805Date: Sun Dec 2 22:10:56 2007 +0100 13806 13807 Show the id's of unknown bit structures in the bios. 13808 13809commit fc98b11eb70f1dd7dfb4e7c6eb8d888dfebdc890 13810Author: Maarten Maathuis <madman2003@gmail.com> 13811Date: Sun Dec 2 12:02:08 2007 +0100 13812 13813 randr12: dual-dvi cards need extra consideration. 13814 13815commit 2beece0c634c2c60996da1f1d15b24c5f231a6be 13816Author: Stuart Bennett <sb476@cam.ac.uk> 13817Date: Sun Dec 2 02:33:57 2007 +0000 13818 13819 Fix endian bug on BIT biosen 13820 13821commit d3563cccb43355927d4ba355bc31a957ea01587f 13822Author: Stuart Bennett <sb476@cam.ac.uk> 13823Date: Sun Dec 2 02:14:32 2007 +0000 13824 13825 Don't use PROM if the checksum is bad 13826 13827commit 4097af9bad85031fa1da9c3d21b084f587412604 13828Author: Stuart Bennett <sb476@cam.ac.uk> 13829Date: Sat Dec 1 09:55:31 2007 +0000 13830 13831 randr12: print "or" _pre_ ffs-ing 13832 13833commit a8faa53355f9ba3851f47e290c821bc1b9a03e85 13834Author: Stuart Bennett <sb476@cam.ac.uk> 13835Date: Sat Dec 1 09:53:05 2007 +0000 13836 13837 DCB parsed flag for duallink outputs 13838 13839 Light on proof, strong on suspicion with this one 13840 13841commit 63fc1ab4ce597a2f542c9bb7d3edae67551bda29 13842Author: Maarten Maathuis <madman2003@gmail.com> 13843Date: Sun Dec 2 02:08:20 2007 +0100 13844 13845 randr12: Always use vpll2_true on nv4x cards. 13846 13847commit 97a99c42fb24320177bdae884c4c710c5a25ce4f 13848Author: Maarten Maathuis <madman2003@gmail.com> 13849Date: Sun Dec 2 01:48:01 2007 +0100 13850 13851 randr12: A safer bet for TMDS register 0x2e. 13852 13853commit 3358f4eec53845cf90b9f9ef1f2a70c231bac588 13854Author: Maarten Maathuis <madman2003@gmail.com> 13855Date: Sun Dec 2 01:34:49 2007 +0100 13856 13857 randr12: Remove preferred_crtc, since it's probably incorrect. 13858 13859commit 5994655925173a4e8b1c0ac902f7dc6e1ffe64f1 13860Author: Maarten Maathuis <madman2003@gmail.com> 13861Date: Sun Dec 2 01:08:41 2007 +0100 13862 13863 randr12: Improve output register setting. 13864 13865commit df8d6848a004e724e92c3dd63fc608198c7bd940 13866Author: Maarten Maathuis <madman2003@gmail.com> 13867Date: Sat Dec 1 18:40:46 2007 +0100 13868 13869 randr12: Fix up some dvi related issues, thanks to MighMoS. 13870 13871commit 84cc907beffd9e8f91c4333a6db919ee2efa2740 13872Author: Maarten Maathuis <madman2003@gmail.com> 13873Date: Sat Dec 1 11:23:50 2007 +0100 13874 13875 randr12: Fix NV4x dvi clock selection, mostly affects G70 cards. 13876 13877commit c97ad9850da89b1898f125df1796bb1d00d124bb 13878Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 13879Date: Sat Dec 1 10:33:53 2007 +0100 13880 13881 Make the debug variable global to all DMA calls. 13882 13883commit 19d75f5018fd442920378a5083dbe1316cc6f62b 13884Author: Stuart Bennett <sb476@cam.ac.uk> 13885Date: Sat Dec 1 06:29:09 2007 +0000 13886 13887 The BIT bios T table. 13888 13889 T might stand for TMDS. 13890 Parsing once on start may not be the correct answer, could be a runtime thing 13891 13892commit 42a02513a1db7ada586dcaa2eaa5b68e97d67892 13893Author: Stuart Bennett <sb476@cam.ac.uk> 13894Date: Sat Dec 1 06:37:17 2007 +0000 13895 13896 Bump DCB table init so that INIT_50 can use it 13897 13898commit ddc4b0946dfd567fc0ea136db0d8b695c28bc3a2 13899Author: Stuart Bennett <sb476@cam.ac.uk> 13900Date: Sat Dec 1 05:35:40 2007 +0000 13901 13902 "or" can be 0xc 13903 13904 Quite what this means, or why Nvidia couldn't give it a more descriptive 13905 variable name in the g80 driver is anyone's guess 13906 13907commit 384115a5abd756c5e3ee79ba50e15636d159f19c 13908Author: Stuart Bennett <sb476@cam.ac.uk> 13909Date: Sat Dec 1 05:17:47 2007 +0000 13910 13911 Use Nvidia's naming for CRTC_INDEX_COLOR 13912 13913commit 2344b5b5a5a0efa439ba64f92d5b08097178dd6c 13914Author: Stuart Bennett <sb476@cam.ac.uk> 13915Date: Sat Dec 1 05:16:20 2007 +0000 13916 13917 Bios script code 0x50 13918 13919commit 78c6ff292f4c2aa0e06bce095108d0716555104a 13920Author: Stuart Bennett <sb476@cam.ac.uk> 13921Date: Sat Dec 1 00:45:32 2007 +0000 13922 13923 Add dcb "location" field. 0 for on-chip, !0 for external 13924 13925commit 312fb3d65b8db6e3c2d123c0b192da82b7de4f71 13926Author: Maarten Maathuis <madman2003@gmail.com> 13927Date: Fri Nov 30 16:47:26 2007 +0100 13928 13929 randr12: Don't unset bit 8-15 in sel_clk. 13930 13931commit ea10d579fb6053e064a5cd14b47e1da55d4120ac 13932Author: Maarten Maathuis <madman2003@gmail.com> 13933Date: Fri Nov 30 16:43:33 2007 +0100 13934 13935 randr12: Big commit, read message. 13936 13937 - Outputs (on nv4x) should now select vpll2 instead of vpll1 when neccesary (this took a long time to figure out). 13938 - Register 0x00680594 is now set correctly to allow analog secondary outputs to work (for me at least). 13939 - I found the bit that triggers non-VGA modesetting in nv10reg, so in the future we could copy the blob in that area. 13940 - Misc fixes to a lot of things. 13941 13942commit a7a8993f4074e942ec4b3ef58ede0162a2d3c415 13943Author: Maarten Maathuis <madman2003@gmail.com> 13944Date: Mon Nov 26 22:11:03 2007 +0100 13945 13946 randr12: Make sure the vpll is properly set when we steal a ramdac from another output. 13947 13948commit c11d3f68720b698736d5a28a82d3e45b55379bc6 13949Author: Maarten Maathuis <madman2003@gmail.com> 13950Date: Mon Nov 26 19:40:59 2007 +0100 13951 13952 randr12: Fixed a minor cursor bug. 13953 13954commit 59d98374e082c40b57737de56b17173f5b54910e 13955Author: Maarten Maathuis <madman2003@gmail.com> 13956Date: Mon Nov 26 19:24:03 2007 +0100 13957 13958 randr12: Set the value of sel_clk correctly (hopefully) for NV4x. 13959 13960commit 8672d7a33fe369645786c21e130f51560a656d3c 13961Author: Maarten Maathuis <madman2003@gmail.com> 13962Date: Sun Nov 25 22:08:23 2007 +0100 13963 13964 randr12: cleanup + double register assignment (vga CRTC) removed. 13965 13966commit 11809882f24e40b1b8c77f4caca27ce2ebd3ccbf 13967Author: Maarten Maathuis <madman2003@gmail.com> 13968Date: Sun Nov 25 18:15:27 2007 +0100 13969 13970 randr12: Convert premultiplied alpha cursors to normal alpha cursors. 13971 13972commit ece41daabeb0cfd727daec14d9734dac08159c7a 13973Author: Maarten Maathuis <madman2003@gmail.com> 13974Date: Sun Nov 25 16:49:01 2007 +0100 13975 13976 randr12: Fix alpha cursor corruption. 13977 13978commit 5abbf55fb5c957c4be425e9ebd2660ca14d44e27 13979Author: Maarten Maathuis <madman2003@gmail.com> 13980Date: Sun Nov 25 14:15:30 2007 +0100 13981 13982 randr12: alpha cursors don't need background changes. 13983 13984commit 37c36a284508e63af152a9315aa19a729e5516cd 13985Author: Maarten Maathuis <madman2003@gmail.com> 13986Date: Sun Nov 25 13:06:10 2007 +0100 13987 13988 Reindent a few cursor functions. 13989 13990commit 1ad212271cc99b95ff52eb719d24ac58d392c0eb 13991Author: Maarten Maathuis <madman2003@gmail.com> 13992Date: Sun Nov 25 03:20:43 2007 +0100 13993 13994 randr12: Improve TMDS and LVDS registers (and enable for LVDS). 13995 13996commit 39d94139910c53f9092f28ea6d742133666ac18d 13997Author: Maarten Maathuis <madman2003@gmail.com> 13998Date: Sun Nov 25 01:31:39 2007 +0100 13999 14000 randr12: Optimize tmds_regs a little bit. 14001 14002commit c6dce92c47dacfc822315e570bff70a8e42f5dc2 14003Author: Stuart Bennett <sb476@cam.ac.uk> 14004Date: Sat Nov 24 21:51:14 2007 +0000 14005 14006 randr12: use bus field to determine dvi pairs 14007 14008commit 328197b4613f5e25eacfb67888701bb0e298c765 14009Author: Stuart Bennett <sb476@cam.ac.uk> 14010Date: Sat Nov 24 20:44:27 2007 +0000 14011 14012 Remove a load of unnecessary casts, and reorder functions by opcode 14013 14014 Plus a micro opt to checksum thing 14015 14016commit 8cee41bc6112a96f43f242b5178a9c968d2b2d9e 14017Author: Stuart Bennett <sb476@cam.ac.uk> 14018Date: Sat Nov 24 20:25:13 2007 +0000 14019 14020 A bunch more cleaned up bios opcodes 14021 14022 Thanks to stillunknown for the bios trace 14023 14024commit 5d6f214db4cd5f0cdcb6a4395ce2ce4acf322257 14025Author: Maarten Maathuis <madman2003@gmail.com> 14026Date: Sat Nov 24 16:22:04 2007 +0100 14027 14028 randr12: Be more considerate of outputs that can only work on ramdac 1. 14029 14030commit 6e1c37e26f630caa03da9061a5cec200708babbd 14031Author: Maarten Maathuis <madman2003@gmail.com> 14032Date: Sat Nov 24 15:47:04 2007 +0100 14033 14034 randr12: The result of another refinement of dvi crosswiring. 14035 14036commit ce57eb9b34095f2776ef36b09249a4b202824a86 14037Author: Maarten Maathuis <madman2003@gmail.com> 14038Date: Sat Nov 24 13:12:49 2007 +0100 14039 14040 randr12: Disable crt_load_detect for ramdac 1 and set some better valued for test_control. 14041 14042commit 525997f86c5fe0f79cf0fa584b4cb617b83a9386 14043Author: Maarten Maathuis <madman2003@gmail.com> 14044Date: Fri Nov 23 23:12:53 2007 +0100 14045 14046 randr12: add bit30 to vpll register for nv4x cards 14047 14048 - This seems the most common situation, any idea what the bits do? 14049 14050commit f249b3b33c025417d11a85a29848db2112450dce 14051Author: Patrice Mandin <pmandin@caramail.com> 14052Date: Fri Nov 23 22:25:47 2007 +0100 14053 14054 nv30: missing command 14055 14056commit b56ca2ef87fe15132dcbc43361344d8af446c8a7 14057Author: Patrice Mandin <pmandin@caramail.com> 14058Date: Fri Nov 23 22:25:06 2007 +0100 14059 14060 nv30: another reference to an object 14061 14062commit a8af1725ef7ae6c1cd19659d0607e2a702dc148a 14063Author: Patrice Mandin <pmandin@caramail.com> 14064Date: Fri Nov 23 22:24:19 2007 +0100 14065 14066 nv30: rename viewport clip 14067 14068commit 14e631691a2e8704ee016828e94c48de7eb96409 14069Author: Patrice Mandin <pmandin@caramail.com> 14070Date: Fri Nov 23 22:03:21 2007 +0100 14071 14072 nv30: was not using gart handle 14073 14074commit 9648fd259f9906f5db854e8c4b57fd2675f4d5c5 14075Author: Maarten Maathuis <madman2003@gmail.com> 14076Date: Fri Nov 23 21:08:46 2007 +0100 14077 14078 randr12: Reenable a properly working crt load detect. 14079 14080commit 3010847c1ccf14f0b5356aeb05f2c8c0c519ce91 14081Author: Maarten Maathuis <madman2003@gmail.com> 14082Date: Fri Nov 23 18:58:49 2007 +0100 14083 14084 randr12: Remove some unneeded stuff. 14085 14086commit 2b092ca5277405a166ea54de418a8114fb6a8cf6 14087Author: Maarten Maathuis <madman2003@gmail.com> 14088Date: Fri Nov 23 17:32:12 2007 +0100 14089 14090 randr12: fix comment typo 14091 14092commit 5ca53ef57ece7ae547cd5b11e83eb0d15fe69033 14093Author: Maarten Maathuis <madman2003@gmail.com> 14094Date: Fri Nov 23 16:40:02 2007 +0100 14095 14096 randr12: Allow arbitrary ramdac and crtc combinations. 14097 14098 - Support for the misterious ramdac 2 still has to be reverse engineered. 14099 14100commit e16fccde3f817e9eab3a09398a4821566967a86d 14101Author: Maarten Maathuis <madman2003@gmail.com> 14102Date: Fri Nov 23 13:25:38 2007 +0100 14103 14104 randr12: Move a lot of registers from output to crtc. 14105 14106commit cd8293954e1d004ef837d339ae87c4689fd45e1f 14107Author: Maarten Maathuis <madman2003@gmail.com> 14108Date: Fri Nov 23 10:32:08 2007 +0100 14109 14110 randr12: There was more logic to sel_clk than previously thought. 14111 14112commit 0b55ba9f9fbd60044bad880f260e0c70e6f577f0 14113Author: Maarten Maathuis <madman2003@gmail.com> 14114Date: Thu Nov 22 22:18:47 2007 +0100 14115 14116 randr12: revert to old vpll regs on nv4x cards, because the others can't be written 14117 14118 - If anyone knows the bit to allow those registers to be written, then please tell. 14119 - This also fixes db1 vclk ratio issues, hopefully. 14120 14121commit 18117df17ec2ce355344953d70caf33a0272697f 14122Author: Stuart Bennett <sb476@cam.ac.uk> 14123Date: Thu Nov 22 00:30:29 2007 +0000 14124 14125 Bios CRTC head chosing fixed 14126 14127 Also, a slightly tighter limit on values for or, as I don't currently think 14128 bit 4 being set makes any sense 14129 14130commit 425b7cd9dcddf1735d59e9ef60e81bf91db1c65c 14131Author: Stuart Bennett <sb476@cam.ac.uk> 14132Date: Wed Nov 21 20:02:17 2007 +0000 14133 14134 randr12: unbreak LVDS console restore by reverting tmds_regs change 14135 14136commit e3f77b1b4332f5c072be713b5ea6dd234c194bb0 14137Author: Stuart Bennett <sb476@cam.ac.uk> 14138Date: Tue Nov 20 19:02:17 2007 +0000 14139 14140 Complete init_index_addr_latched opcode 14141 14142commit 05f40b2dd3f8c14da2464b19b21aa66defec17e3 14143Author: Stuart Bennett <sb476@cam.ac.uk> 14144Date: Wed Nov 21 01:58:12 2007 +0000 14145 14146 More DCB entry output, and a default for DCB 1.4 14147 14148 The table seems to be the same for all DCB 1.4 biosen I've seen, and 14149 POST seems to make no effort to parse it. I wonder if it is in fact used at all 14150 14151commit d96176c2f07efde4224c35b7c03b8380a73e1831 14152Author: Maarten Maathuis <madman2003@gmail.com> 14153Date: Tue Nov 20 20:41:38 2007 +0100 14154 14155 randr12: better guess for ramdac register 670 (nv44 and up) 14156 14157commit 817b8f1a1af70402c18c45edccec5c746063d4e8 14158Author: Maarten Maathuis <madman2003@gmail.com> 14159Date: Tue Nov 20 19:55:15 2007 +0100 14160 14161 randr12: fix serious bug in vclk calculator for nv4x cards. 14162 14163commit 040e5479344b0f794e79e4ffc6c547c58835401a 14164Author: Maarten Maathuis <madman2003@gmail.com> 14165Date: Tue Nov 20 18:49:39 2007 +0100 14166 14167 randr12: Fix a small mistake that could cause blank screens for NV44 and up. 14168 14169commit 1cc05c9e08a7e4f0c27bb0e08d47265d31123bf9 14170Author: Maarten Maathuis <madman2003@gmail.com> 14171Date: Tue Nov 20 12:56:28 2007 +0100 14172 14173 randr12: I hope i really fixed db1 vclk ratio this time (for nv4x). 14174 14175commit 016f7e84c4ef62840e300d61e5e3b2ddfadd5e32 14176Author: Maarten Maathuis <madman2003@gmail.com> 14177Date: Tue Nov 20 12:44:25 2007 +0100 14178 14179 NV50: Make a small mistake. 14180 14181commit 2cbd24ad579d3e68eaef2f8500b22e101fecd936 14182Author: Maarten Maathuis <madman2003@gmail.com> 14183Date: Tue Nov 20 12:25:21 2007 +0100 14184 14185 NV50: Some deobfuscation. 14186 14187commit ccfc6910f8cd8eb8d4b628cc14ef7db417f576ea 14188Author: Maarten Maathuis <madman2003@gmail.com> 14189Date: Tue Nov 20 09:52:47 2007 +0100 14190 14191 randr12: Also enable DB1 vclk ratio if PLL_SELECT had it enabled. 14192 14193commit 5b3d86562afd35c14d2c8c07618b08bbc98ad775 14194Author: Maarten Maathuis <madman2003@gmail.com> 14195Date: Mon Nov 19 23:55:58 2007 +0100 14196 14197 randr12: Proper support for db1 vclk ratio, thanks to Thunderbird for some info. 14198 14199commit 22d8ba5ebf659f6f126a1380190339d08ee824c6 14200Author: Maarten Maathuis <madman2003@gmail.com> 14201Date: Mon Nov 19 09:23:37 2007 +0100 14202 14203 randr12: Add a gamma_set hook. 14204 14205commit cbd70303a949f387d42497cb4ea52186ed938b9c 14206Author: Maarten Maathuis <madman2003@gmail.com> 14207Date: Sun Nov 18 22:03:03 2007 +0100 14208 14209 randr12: A first attempt at some lvds programming, not hooked up. 14210 14211 - This is a start for someone with a laptop to test and refine. 14212 14213commit 58c3d6a886c5d5557def490ee8c3e75c9f8d2cd7 14214Author: Benjamin Herrenschmidt <benh@kernel.crashing.org> 14215Date: Sun Nov 18 02:50:33 2007 +0100 14216 14217 Enable IFC uploads. 14218 14219commit 619ae1f7a028f24293f9a4e4f7777cc51ef8d16d 14220Author: Maarten Maathuis <madman2003@gmail.com> 14221Date: Sun Nov 18 00:15:34 2007 +0100 14222 14223 randr12: Final commit, switch to nv40 native vclk regs. 14224 14225 - I've also seperated the save load pll functions, to improve clarity. 14226 14227commit 601adb4f431b638d4b711ba01aa2d4c3408a7bd3 14228Author: Maarten Maathuis <madman2003@gmail.com> 14229Date: Sat Nov 17 23:47:12 2007 +0100 14230 14231 randr12: Intermediate commit (using the native nv40 vpll regs). 14232 14233commit a29eca9dec176388b1c4666ad801ae75360a1219 14234Author: Maarten Maathuis <madman2003@gmail.com> 14235Date: Sat Nov 17 20:08:27 2007 +0100 14236 14237 randr12: Some comment fixes. 14238 14239commit 6f3d335f6e3b9866457730660538c5f073862dfe 14240Author: Maarten Maathuis <madman2003@gmail.com> 14241Date: Sat Nov 17 22:23:32 2007 +0100 14242 14243 randr12: Reinstate m-divider limitations. 14244 14245commit e2b7f57cbe7e1080736e6088d2997cbcf966fc3d 14246Author: Maarten Maathuis <madman2003@gmail.com> 14247Date: Sat Nov 17 18:36:24 2007 +0100 14248 14249 randr12: Some minor changes. 14250 14251commit 2c0c7612bd08938496abb47b34498065e668ec34 14252Author: Maarten Maathuis <madman2003@gmail.com> 14253Date: Sat Nov 17 15:57:20 2007 +0100 14254 14255 randr12: Change vpll handling, note that it still is a work in progress. 14256 14257commit caec3d1ab69c523b6c0a7e0019d1a9f115cf5da3 14258Author: Ben Skeggs <skeggsb@gmail.com> 14259Date: Sat Nov 17 20:59:25 2007 +1100 14260 14261 EXA 1.4 pixmap hooks (disabled) 14262 14263 Really, you don't want to turn this on yet... :) 14264 14265commit b8e7c64476948426d5801367bab5cdf15db5ce84 14266Author: Ben Skeggs <skeggsb@gmail.com> 14267Date: Sat Nov 17 17:17:34 2007 +1100 14268 14269 Add a user callback for when a channel hang is detected. 14270 14271commit 293b32d0f66e965b3a31d54859b91210b3e35b3d 14272Author: Ben Skeggs <skeggsb@gmail.com> 14273Date: Sat Nov 17 17:02:23 2007 +1100 14274 14275 Use handles from grobj, instead of hardcoded vals from nv_dma.h 14276 14277commit 84bc37723c2cabe7652050b4cbaf9f5218fdb071 14278Author: Ben Skeggs <skeggsb@gmail.com> 14279Date: Sat Nov 17 16:43:05 2007 +1100 14280 14281 Turn vram/fb handles into "real" grobjs. 14282 14283commit f3cdd5086f410711103c6b3453629d85ef4c2788 14284Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14285Date: Sat Nov 17 01:22:31 2007 +0100 14286 14287 Put back memory barrier and write posting. 14288 14289commit 762f732c7aaa8aa3dd7d54a8734cbcc54cd10494 14290Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14291Date: Fri Nov 16 23:38:49 2007 +0100 14292 14293 Add verbose debugging code to help with fixing push buffer issues. 14294 14295commit c8f7a9fcdcc356b4dc5030112b1532d2fa3eaf0f 14296Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14297Date: Fri Nov 16 23:37:10 2007 +0100 14298 14299 Oops. Correct either the header or the packet but not both. 14300 14301commit 5ce573875387490eb7feaa87a3354b888beb4410 14302Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14303Date: Fri Nov 16 23:32:07 2007 +0100 14304 14305 Unmatched OUT_RING(0). 14306 14307commit cc0c1af932a96cec6aabc72cbe9ca508fbc80a36 14308Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14309Date: Fri Nov 16 21:59:53 2007 +0100 14310 14311 fix typo. 14312 14313commit 4b7601c297fefc46adf3bc37a463575d3caafedb 14314Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14315Date: Fri Nov 16 21:56:14 2007 +0100 14316 14317 fix 32/64 bit ism. 14318 14319commit bee8bc5ce321245362847253b7ccc7530899e514 14320Author: Maarten Maathuis <madman2003@gmail.com> 14321Date: Fri Nov 16 21:29:20 2007 +0100 14322 14323 randr12: Minor change. 14324 14325commit 735fae4bb0b92f097665f675ba5127a9bb11d8fa 14326Author: Maarten Maathuis <madman2003@gmail.com> 14327Date: Fri Nov 16 19:19:23 2007 +0100 14328 14329 randr12: That wasn't supposed to end up there ;-) 14330 14331commit 2848c3e7a4d25e5ca311ea1a88458eb2a77976ed 14332Author: Maarten Maathuis <madman2003@gmail.com> 14333Date: Fri Nov 16 19:16:08 2007 +0100 14334 14335 randr12: Disable changing ramdac reg580, until i know why this causes issues. 14336 14337commit 75222fa199f1f5bc1abd66a4e80a446b5d597523 14338Author: Ben Skeggs <skeggsb@gmail.com> 14339Date: Sat Nov 17 15:31:49 2007 +1100 14340 14341 nouveau: fix valgrind complaint 14342 14343commit 1a98f1ae8298ca3b069b146a2866f96e934afb74 14344Author: Maarten Maathuis <madman2003@gmail.com> 14345Date: Fri Nov 16 17:00:59 2007 +0100 14346 14347 randr12: Restore old timings + misc changes. 14348 14349commit 1b0cd6602cf9d96f41242054efe2f96cc70eaa96 14350Author: Ben Skeggs <skeggsb@gmail.com> 14351Date: Sat Nov 17 09:08:03 2007 +1100 14352 14353 oops, forgot these... 14354 14355commit 9c46ec6833ccfc29e4a67b0ce4226d8b38b772ab 14356Author: Ben Skeggs <skeggsb@gmail.com> 14357Date: Sat Nov 17 07:23:41 2007 +1100 14358 14359 Initial hacky relocation stuff. 14360 14361 Entire DDX ported across to OUT_RELOCx/OUT_PIXMAPx, workable in it's current 14362 state for the current BO backend.. definitely not for TTM yet :) 14363 14364commit ddda33decb42ba2f5f9586e1d5de4533d1d88674 14365Author: Ben Skeggs <skeggsb@gmail.com> 14366Date: Sat Nov 17 06:11:01 2007 +1100 14367 14368 Initial buffer object work. 14369 14370 Implemented on top of the current MM interface, and currently used in the 14371 exact same way NVAllocate/FreeMemory was used. Can transition to something 14372 more ttm-friendly gradually. 14373 14374 The "old mm" bo backend is extremely primitive. Buffers can only be created 14375 in mappable memory, forcibly pinned, and never moved once they're created. 14376 Some of these things could be improved, but probably not much point. 14377 14378commit 9b9d995a517b48cd074030c567f9c78bee19b129 14379Author: Ben Skeggs <skeggsb@gmail.com> 14380Date: Sat Nov 17 03:36:04 2007 +1100 14381 14382 Report engine hangs again, instead of spinning forever. 14383 14384commit 2af4a28ae1df46686e264bcda417a896b7cfa59a 14385Author: Ben Skeggs <skeggsb@gmail.com> 14386Date: Fri Nov 16 09:18:33 2007 +1100 14387 14388 Hook up DMA funcs to channels/grobjs etc etc 14389 14390commit 6e3d05e8c60fd7bab371af86151e45eec0a65b3d 14391Author: Ben Skeggs <skeggsb@gmail.com> 14392Date: Fri Nov 16 07:08:23 2007 +1100 14393 14394 notifiers.. 14395 14396 Lacking timeouts at the moment so if X would've crashed before, it'll spin 14397 forever now. 14398 14399commit 020740d92b5d623bee65e598ac60020187011b04 14400Author: Ben Skeggs <skeggsb@gmail.com> 14401Date: Fri Nov 16 06:07:56 2007 +1100 14402 14403 And grobjs. 14404 14405commit 929e2b7d53435d097b07c23ed1d0caaa47924f8a 14406Author: Ben Skeggs <skeggsb@gmail.com> 14407Date: Fri Nov 16 05:10:58 2007 +1100 14408 14409 And channels.. 14410 14411commit 22909d96b8ea720169b34326b93c7b29eb3db013 14412Author: Ben Skeggs <skeggsb@gmail.com> 14413Date: Fri Nov 16 04:36:08 2007 +1100 14414 14415 Implement nouveau_device struct. 14416 14417commit b96e813773b0705bbac57f2d64adff3aad845622 14418Author: Stuart Bennett <sb476@cam.ac.uk> 14419Date: Fri Nov 16 03:26:56 2007 +0000 14420 14421 Begone, foul g5-bios-hack 14422 14423 also, fix the DCB termination test; I'm amazed nothing broke 14424 14425commit 7eee9585fa8e8c45f4ab4712a3b990664203d72b 14426Author: Maarten Maathuis <madman2003@gmail.com> 14427Date: Thu Nov 15 19:33:13 2007 +0100 14428 14429 randr12: Only NV4x have two seperate PVIO ranges. 14430 14431commit 276892b069fd045b20f44d7888a7971371d73ad3 14432Author: Maarten Maathuis <madman2003@gmail.com> 14433Date: Thu Nov 15 19:27:12 2007 +0100 14434 14435 randr12: Improved values for FP_CONTROL register. 14436 14437commit 4a16f14061c88ed49b05ef8779a2584f5053c3f1 14438Author: Maarten Maathuis <madman2003@gmail.com> 14439Date: Thu Nov 15 18:44:52 2007 +0100 14440 14441 randr12: Some minor register changes. 14442 14443commit b436a6809c62e0fe0696d2c6bad066c3afcf9166 14444Author: Arthur Huillet <arthur.huillet@free.fr> 14445Date: Thu Nov 15 15:01:47 2007 +0100 14446 14447 NV10 EXA: removed left out debug statement 14448 14449commit 98e223ebb01440810e01e902e904524c50b463cb 14450Author: Arthur Huillet <arthur.huillet@free.fr> 14451Date: Thu Nov 15 12:57:20 2007 +0100 14452 14453 NV1x EXA : refactored A8 + A8 code, enabled on NV2x, fix A8 txfmt for NV2x 14454 14455 more compact and more readable A8 + A8 code along with some codestyle fixes 14456 now enabling this acceleration on NV2x cards, with the A8 format value hack 14457 14458commit 3cdc4b274f371f1f524d140fddd79dad2987b06c 14459Author: Ben Skeggs <skeggsb@gmail.com> 14460Date: Fri Nov 16 03:28:45 2007 +1100 14461 14462 nouveau: fix a host of warnings 14463 14464commit 6eba6c7fab534efd15c8e96daeeb9a33972c2b6d 14465Author: Stuart Bennett <sb476@cam.ac.uk> 14466Date: Thu Nov 15 02:09:45 2007 +0000 14467 14468 Big endian bios fix 14469 14470commit 69a3a849e65a022609db3a5f33277b744463a6ae 14471Author: Patrice Mandin <pmandin@caramail.com> 14472Date: Wed Nov 14 23:03:06 2007 +0100 14473 14474 nv30exa: set link with image blit 14475 14476commit 2435836202efe7ea42a7bece71f125fb46aac3ba 14477Author: Stuart Bennett <sb476@cam.ac.uk> 14478Date: Wed Nov 14 19:02:28 2007 +0000 14479 14480 nv_bios indentation 14481 14482commit 7b0ca260f57e92e45d7f48f16642bfcd67b10eaf 14483Author: Stuart Bennett <sb476@cam.ac.uk> 14484Date: Wed Nov 14 18:47:40 2007 +0000 14485 14486 More code and logging beautification - wrap I/O better 14487 14488 Plus some actual fixes 14489 14490commit d00420a1286e368fbc20551cae5260bf65ed1b15 14491Author: Stuart Bennett <sb476@cam.ac.uk> 14492Date: Wed Nov 14 19:07:06 2007 +0000 14493 14494 IO flag conditions, beginnings of init_idx_addr_latched 14495 14496 plus usual whitespace and pointless renaming 14497 14498commit 4e8af31a89b7d732b52706f87e1dc66e29ab717e 14499Author: Stuart Bennett <sb476@cam.ac.uk> 14500Date: Wed Nov 14 03:01:27 2007 +0000 14501 14502 More opcodes, and some monstrous reindenting 14503 14504 all part of an evil plan to have the most lines in the Xorg log 14505 14506commit 89617cfcdfd1ee19b99888cfbcfec1ee0550d26c 14507Author: Stuart Bennett <sb476@cam.ac.uk> 14508Date: Wed Nov 14 00:36:02 2007 +0000 14509 14510 Updating some BIOS parser opcodes 14511 14512 plus various debug / whitespace fixes 14513 14514commit cc5db05bed630350613c7cbfa418d814ed5a5093 14515Author: Stuart Bennett <sb476@cam.ac.uk> 14516Date: Wed Nov 14 19:27:06 2007 +0000 14517 14518 randr12: More DCB abstraction to lessen parsing nastiness for old versions 14519 14520commit d46b61f88225d3e2b28a6c4a4ac4b8a1fe942a39 14521Author: Stuart Bennett <sb476@cam.ac.uk> 14522Date: Tue Nov 13 18:24:46 2007 +0000 14523 14524 Gratuitous code renaming, moving + warning fixes for bios 14525 14526commit c87e3570a198b5d69046387d111a2709a571d5e3 14527Author: Maarten Maathuis <madman2003@gmail.com> 14528Date: Wed Nov 14 22:30:11 2007 +0100 14529 14530 randr12: Reinstate the "magic factor" until a better solution comes along. 14531 14532commit 5b62e77ded9e64d439b70ce0b7dce27a7ae1df4f 14533Author: Arthur Huillet <arthur.huillet@free.fr> 14534Date: Wed Nov 14 20:54:24 2007 +0100 14535 14536 NV1x EXA : implemented A8 + A8 via a dirty hack 14537 14538 it works, but probably needs to be optimized 14539 14540commit 3db89822ba94a94eeb1ab891488d7f203a0dfff8 14541Author: Maarten Maathuis <madman2003@gmail.com> 14542Date: Wed Nov 14 19:48:53 2007 +0100 14543 14544 randr12: Disable ramdac register 0x900, until i know for certain what it does. 14545 14546commit 842b33e2dc82e70a7da992642de4503d4cb0a1fc 14547Author: Maarten Maathuis <madman2003@gmail.com> 14548Date: Wed Nov 14 16:02:56 2007 +0100 14549 14550 randr12: fix comment typo 14551 14552commit c7fc4b9981bb242830ddb830863fb26301429c74 14553Author: Maarten Maathuis <madman2003@gmail.com> 14554Date: Wed Nov 14 08:34:44 2007 +0100 14555 14556 randr12: Enable all pll on pre-nv40 and improve usage of ramdac_670 register. 14557 14558commit 992d079d64369a4a8445012b438ec980968c11d8 14559Author: Maarten Maathuis <madman2003@gmail.com> 14560Date: Tue Nov 13 21:46:23 2007 +0100 14561 14562 NV50: fix stupid type from commit d5dfb82f7f39304eb39dd4b1c34a2a2801d90a25 14563 14564commit 7ad7b6ad453b6401ef0bf5128f7ac9790df95bb2 14565Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14566Date: Tue Nov 13 21:36:13 2007 +0100 14567 14568 Don't use GART on PPC. Workaround DMA bugs there. 14569 14570commit b451a21f77678b1fb1d31f3df7e3b1aea6984876 14571Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14572Date: Tue Nov 13 20:58:51 2007 +0100 14573 14574 Remove ENABLE_NV30EXA, which is not needed any more. 14575 14576commit 9dc1e1e1fc049c7a535da8e1bc8ce79fd2d620b5 14577Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14578Date: Tue Nov 13 20:54:53 2007 +0100 14579 14580 Enable nv30 exa on PPC. 14581 14582commit 4fb7b608cbdf53084da63526c066eb7f772c6c14 14583Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14584Date: Tue Nov 13 20:53:58 2007 +0100 14585 14586 Fix nv30 exa for big endian 14587 14588commit 875960dacda54268fc1d318995c145b2db09ca5b 14589Author: Maarten Maathuis <madman2003@gmail.com> 14590Date: Tue Nov 13 19:43:41 2007 +0100 14591 14592 randr12: Set a few new registers + misc fixes. 14593 14594commit cf9aaa2b074eca67cf0119172c3d15328418416d 14595Author: Stuart Bennett <sb476@cam.ac.uk> 14596Date: Mon Nov 12 19:24:34 2007 +0000 14597 14598 Some type / warning fixes for nv_bios.c 14599 14600commit f00f6d1df06d2671255fd988f514a6f6ab0f8b18 14601Author: Stuart Bennett <sb476@cam.ac.uk> 14602Date: Sun Nov 11 17:50:55 2007 +0000 14603 14604 Misc bios parsing bits 14605 14606commit fdc7f4a353c4dda07fffd9800b477ae5bc43baac 14607Author: Patrice Mandin <pmandin@caramail.com> 14608Date: Mon Nov 12 23:54:22 2007 +0100 14609 14610 Remove disable switches, big cleanup, requires xorgs server 1.3 again 14611 14612commit e93a5d3abe6d637e00bd9c998e9735c9766eecfb 14613Author: Maarten Maathuis <madman2003@gmail.com> 14614Date: Mon Nov 12 21:05:00 2007 +0100 14615 14616 randr12: Improve vclk setting. 14617 14618 - It now considers crosswired TMDS situations. 14619 14620commit d32420c82b1e63bbc84768c92c336391af624952 14621Author: Maarten Maathuis <madman2003@gmail.com> 14622Date: Mon Nov 12 20:17:04 2007 +0100 14623 14624 randr12: Fix bad comment. 14625 14626commit 8a04babb88111ec8ccec9af56c3591d41afba9b3 14627Author: Maarten Maathuis <madman2003@gmail.com> 14628Date: Mon Nov 12 20:14:10 2007 +0100 14629 14630 randr12: Some minor changes. 14631 14632commit 821d4b83ed780d9a0a8d950a2595a0e15ef3f103 14633Author: Maarten Maathuis <madman2003@gmail.com> 14634Date: Mon Nov 12 16:21:14 2007 +0100 14635 14636 randr12: Set SEL_CLK appropriately based on wether a TMDS is crosswired. 14637 14638commit d017ce4879ca755fc6d56f232fe3162864f4b38e 14639Author: Maarten Maathuis <madman2003@gmail.com> 14640Date: Mon Nov 12 10:34:54 2007 +0100 14641 14642 A little code cleanup. 14643 14644commit 16fb5b92dba9a0380d2fdd1ece1b9b0691d79d3a 14645Author: Maarten Maathuis <madman2003@gmail.com> 14646Date: Mon Nov 12 00:17:52 2007 +0100 14647 14648 NV50: Port an LVDS detection fix from nv. 14649 14650commit b01a87f18d122d80fac6daf1dfb1c404d7f3a814 14651Author: Maarten Maathuis <madman2003@gmail.com> 14652Date: Mon Nov 12 00:03:19 2007 +0100 14653 14654 NV50: Port one of the nv patches. 14655 14656commit f22691661f8b72cedb8aa4fc96b41fcff4171e12 14657Author: Maarten Maathuis <madman2003@gmail.com> 14658Date: Sun Nov 11 21:53:10 2007 +0100 14659 14660 randr12: Some code cleanup. 14661 14662commit aceceb22775bb039cb0e207f39f7cbd93c2db194 14663Author: Maarten Maathuis <madman2003@gmail.com> 14664Date: Sun Nov 11 21:32:31 2007 +0100 14665 14666 NV50: Clean up nv50_output.c a bit. 14667 14668 - I did not make wrappers for the i2c stuff, since i do not know what that range does, besides i2c. 14669 14670commit a0835f7c64cf5f62626332ccba21cfe436584275 14671Author: Maarten Maathuis <madman2003@gmail.com> 14672Date: Sun Nov 11 21:17:40 2007 +0100 14673 14674 NV50: cleanup of nv50_sor.c 14675 14676commit f463a7f6213bede18917191546c8cd61cf4612e8 14677Author: Maarten Maathuis <madman2003@gmail.com> 14678Date: Sun Nov 11 20:51:01 2007 +0100 14679 14680 randr12: Make sure we don't set 2 "special" bits in sel_clk. 14681 14682commit 27cbf73bc572f435870decc2017cbf1c4f38264d 14683Author: Maarten Maathuis <madman2003@gmail.com> 14684Date: Sun Nov 11 19:32:26 2007 +0100 14685 14686 randr12: Overhaul vpll set code + misc fixes. 14687 14688 - Hopefully the problematic G70's work again. 14689 14690commit 78e88d1f4e2f058c487978c01e19a828271f3f0b 14691Author: Stuart Bennett <sb476@cam.ac.uk> 14692Date: Fri Nov 9 03:17:12 2007 +0000 14693 14694 vbios fptable v2 parsing (for BIT biosen). Not that I know how to index it. 14695 14696 Plus some fptable v1 fixes, bmp structure length checks, 14697 and support for a weird bios version that may be in some gf2go 14698 14699commit 15c771a6b3ecb9b36f9725c518141cf889aa953c 14700Author: Stuart Bennett <sb476@cam.ac.uk> 14701Date: Fri Nov 9 01:19:04 2007 +0000 14702 14703 G5 6600 DCB improvements, but I'm sure to be some bswaps short 14704 14705 This will all become so much nicer when I can do leNN_to_cpu()... 14706 14707commit 9408fcbbbbab80608c8923ab2b7757dfa76b9767 14708Author: Maarten Maathuis <madman2003@gmail.com> 14709Date: Sat Nov 10 17:02:38 2007 +0100 14710 14711 NV50: Last part of nv50_dac.c 14712 14713commit 85b5e6bfc17c347a9976155a755637f7c8d9304f 14714Author: Maarten Maathuis <madman2003@gmail.com> 14715Date: Sat Nov 10 16:59:49 2007 +0100 14716 14717 NV50: More cleanup (this time nv50_dac.c). 14718 14719commit 23d9f09a4600abf5426cb6d069dc5d9aa876a771 14720Author: Maarten Maathuis <madman2003@gmail.com> 14721Date: Sat Nov 10 16:32:51 2007 +0100 14722 14723 NV50: That's the initial cleanup for nv50_display.c 14724 14725commit 25cef08b183d6406ce065b2b3cd224dd1840525b 14726Author: Maarten Maathuis <madman2003@gmail.com> 14727Date: Sat Nov 10 16:24:32 2007 +0100 14728 14729 NV50: Should have run make first ;-) 14730 14731commit d5dfb82f7f39304eb39dd4b1c34a2a2801d90a25 14732Author: Maarten Maathuis <madman2003@gmail.com> 14733Date: Sat Nov 10 16:23:48 2007 +0100 14734 14735 NV50: Even more cleanup. 14736 14737commit 391c498e676752ac6bf03b668a1cb00094c1a049 14738Author: Maarten Maathuis <madman2003@gmail.com> 14739Date: Sat Nov 10 16:19:32 2007 +0100 14740 14741 NV50: More cleanup, small commits to make bisecting easier. 14742 14743commit 59788f9d4e6c264a6b666918989ad6b778026bd9 14744Author: Maarten Maathuis <madman2003@gmail.com> 14745Date: Sat Nov 10 16:05:42 2007 +0100 14746 14747 NV50: Some initial cleanup. 14748 14749commit 1a7d7ccb4c12f6a10aa5e8b41daee1aae52efa8a 14750Author: Maarten Maathuis <madman2003@gmail.com> 14751Date: Sat Nov 10 15:39:04 2007 +0100 14752 14753 NV50: Another wrapper. 14754 14755commit 3ca3170c8bbd758da81a1bc5b5a84cf3ca642c07 14756Author: Maarten Maathuis <madman2003@gmail.com> 14757Date: Sat Nov 10 15:29:13 2007 +0100 14758 14759 NV50: A new wrapper + a few functions renamed. 14760 14761 - I'm trying to clean up the code a bit, but lack of the hardware prevents me from testing it. 14762 14763commit 77dc99b2621523e20a956a8eaa01a100c992c458 14764Author: Maarten Maathuis <madman2003@gmail.com> 14765Date: Sat Nov 10 00:35:04 2007 +0100 14766 14767 randr12: This may break stuff for some g70 cards, if it does, then please scream. 14768 14769 - I hope i've got most situations covered. 14770 - If there are issues getting the second head to work on nv4x cards, then please tell. 14771 - nv4x is geforce 6 and 7 series for those who do not know. 14772 14773commit c169480b0bb177e8c5efb292d38c55554dcdfcdd 14774Author: Maarten Maathuis <madman2003@gmail.com> 14775Date: Fri Nov 9 20:41:48 2007 +0100 14776 14777 randr12: Disable load detect again, until we can identify it from a DVI monitor. 14778 14779commit 7237a7014b426acc4857d32967e0f10974877819 14780Author: Maarten Maathuis <madman2003@gmail.com> 14781Date: Fri Nov 9 20:40:46 2007 +0100 14782 14783 randr12: Improve detection of DVI-D/DVI-A pairs (for naming). 14784 14785commit a8125c458fa53544625da2071a69e3fd27d7c084 14786Author: Stuart Bennett <sb476@cam.ac.uk> 14787Date: Fri Nov 9 19:27:37 2007 +0000 14788 14789 randr12: Revert "randr12: lose ffs wrapper", it's wrong 14790 14791commit d83cb471cfcc18973eb4a5f319bc17a06e0a883f 14792Author: Stuart Bennett <sb476@cam.ac.uk> 14793Date: Fri Nov 9 19:02:00 2007 +0000 14794 14795 randr12: lose ffs wrapper 14796 14797commit dedf68e10f4b0cde6f9c49d60df737ed7b4148c0 14798Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 14799Date: Wed Oct 31 02:36:41 2007 +0100 14800 14801 More nv30 exa cleaning up. 14802 14803commit d3679e8a5c0ce54ee3ef4de15be5a1fcd353df8e 14804Author: Maarten Maathuis <madman2003@gmail.com> 14805Date: Fri Nov 9 18:44:37 2007 +0100 14806 14807 randr12: Do not mess with ramdac register 580 until it's proven to be neccesary. 14808 14809 - I cannot seem to find out what the conditions are to change this register, except that it's related to VCLK ratio DB1. 14810 14811commit a49d51c638c2004641e58dc0061652553e9a6751 14812Author: Maarten Maathuis <madman2003@gmail.com> 14813Date: Fri Nov 9 18:31:02 2007 +0100 14814 14815 randr12: Fix vertical scaling. 14816 14817commit 2e6f5e18231a327cac7de4a927435585d1eadcee 14818Author: Patrice Mandin <pmandin@caramail.com> 14819Date: Fri Nov 9 18:13:43 2007 +0100 14820 14821 oops, trailing whitespace 14822 14823commit 596f4e6a78900f3cbb3beb4c99da7606dc630179 14824Author: Patrice Mandin <pmandin@caramail.com> 14825Date: Fri Nov 9 18:11:25 2007 +0100 14826 14827 Also disable nv40exa if nv30exa disabled 14828 14829commit d9f80b9d5f4ae1c00e79b9e903c2fc0c23b37a13 14830Author: Patrice Mandin <pmandin@caramail.com> 14831Date: Fri Nov 9 18:10:53 2007 +0100 14832 14833 Reading DFP tables need randr1.2 14834 14835commit 50aaccb5e0d3b9a978701c55ebe6545a6f7b1456 14836Author: Stuart Bennett <sb476@cam.ac.uk> 14837Date: Thu Nov 8 18:59:50 2007 +0000 14838 14839 randr12: invert fpScaler to regain original meaning (see man page) 14840 14841commit 3f6db3ac15b51a58353396fce2126a369b4cfd89 14842Author: Stuart Bennett <sb476@cam.ac.uk> 14843Date: Thu Nov 8 20:30:28 2007 +0000 14844 14845 randr12: Do scaling / native fixup in the fixup function 14846 14847 Nuke the LVDS specific fixup code 14848 Then use adjusted_mode accordingly 14849 14850commit 51a874a59d82678bbb790b708f9ba62985d99344 14851Author: Stuart Bennett <sb476@cam.ac.uk> 14852Date: Wed Nov 7 18:37:57 2007 +0000 14853 14854 randr12: rename DIGITAL and PANEL to TMDS and LVDS 14855 14856 Following on from earlier renaming, this is less confusing, at least to me. 14857 14858commit 61c92f10e1182403fc6fc8a666ed3d8b14795481 14859Author: Stuart Bennett <sb476@cam.ac.uk> 14860Date: Thu Nov 8 17:00:44 2007 +0000 14861 14862 randr12: blank screen before we're ready to show it 14863 14864commit da64328dd7be7f0e7ee1dd0dd774174fc126948b 14865Author: Maarten Maathuis <madman2003@gmail.com> 14866Date: Thu Nov 8 20:11:25 2007 +0100 14867 14868 randr12: Some outputs are only valid on secondary output. 14869 14870commit fe3d49ccb24ff20fba79ee97cceba64850d9387f 14871Author: Maarten Maathuis <madman2003@gmail.com> 14872Date: Thu Nov 8 19:20:41 2007 +0100 14873 14874 randr12: Improve output naming scheme. 14875 14876commit 42542b36d771142d38f72b3365da8b53e9d25c55 14877Author: Maarten Maathuis <madman2003@gmail.com> 14878Date: Thu Nov 8 19:13:03 2007 +0100 14879 14880 randr12: Add ramdac assignment code to the lvds dpms function. 14881 14882commit 2f1bf8138ba970ad18afe1ba9ee10228ea213926 14883Author: Maarten Maathuis <madman2003@gmail.com> 14884Date: Thu Nov 8 18:42:35 2007 +0100 14885 14886 randr12: Misc fixes. 14887 14888commit 8b733ab61e0b9eb375783e986de776eef5fd49ff 14889Author: Ben Skeggs <skeggsb@gmail.com> 14890Date: Fri Nov 9 03:29:21 2007 +1100 14891 14892 get rid of ALLOCATE_LOCAL etc. 14893 14894commit bfb78d2090ee5c189cf00e6af5107a756e864e98 14895Author: Maarten Maathuis <madman2003@gmail.com> 14896Date: Thu Nov 8 00:20:53 2007 +0100 14897 14898 randr12: It seems the situation was a bit easier to handle, luckily. 14899 14900 - I've tied ramdac and crtc number together. 14901 - Switching crtc's doesn't cause a problem anymore (even when crtc0 is not in use). 14902 14903commit d56e6e498ed9acc35bfa540881243acf03981571 14904Author: Maarten Maathuis <madman2003@gmail.com> 14905Date: Wed Nov 7 22:25:40 2007 +0100 14906 14907 randr12: Intermediate commit. 14908 14909commit f13659a8c8e855dc0b871173ccfdd0b70236be71 14910Author: Arthur Huillet <arthur.huillet@free.fr> 14911Date: Wed Nov 7 23:03:54 2007 +0100 14912 14913 NV10EXA: added support for R5G6B5 as source and destination (tested only with rendercheck) 14914 added A8 + A8 via ARGB + ARGB hack, disabled by default because it cannot work 14915 updated instrumentation to dump pixmap sizes and repeat flag 14916 14917commit 3d671bf3b06e64b49254db666980be3a3b3652cc 14918Author: Stuart Bennett <sb476@cam.ac.uk> 14919Date: Wed Nov 7 15:08:48 2007 +0000 14920 14921 randr12: remove some duplication and reduce mem leaks 14922 14923commit cdbec606e4ec398d6bef87d635499100559e3a19 14924Author: Stuart Bennett <sb476@cam.ac.uk> 14925Date: Wed Nov 7 14:54:18 2007 +0000 14926 14927 randr12: refactor ddc stuff for convenience 14928 14929commit ed8473b228579024a4fdc9c2a9ecf4cb71135378 14930Author: Stuart Bennett <sb476@cam.ac.uk> 14931Date: Wed Nov 7 14:33:50 2007 +0000 14932 14933 randr12: micro-optimization on use of "or" 14934 14935commit f6d0b446eda687bb55649599e1aedfa231feaaa9 14936Author: Stuart Bennett <sb476@cam.ac.uk> 14937Date: Tue Nov 6 21:20:10 2007 +0000 14938 14939 randr12: LVDS warning 14940 14941commit 62f8fed79f1617ffd0289123045fda0a3c0e92b2 14942Author: Stuart Bennett <sb476@cam.ac.uk> 14943Date: Tue Nov 6 01:29:33 2007 +0000 14944 14945 randr12: Use NV_RAMDAC_PLL_SELECT_PLL_SOURCE_ALL for LVDS 14946 14947 Using only VPLL seems to be broken for LVDS. Traces show consistent use of ALL. 14948 14949commit 0b958dc66a5533a7db61b7b2f456df876e9905ff 14950Author: Maarten Maathuis <madman2003@gmail.com> 14951Date: Wed Nov 7 09:33:40 2007 +0100 14952 14953 randr12: Also save flatpanel regs on analog monitors. 14954 14955commit c8ac3f0c7cfd6eca85f332e11e280f715e80283e 14956Author: Stuart Bennett <sb476@cam.ac.uk> 14957Date: Tue Nov 6 21:09:51 2007 +0000 14958 14959 Various minor fixes 14960 14961 Now that ramdac selection uses sensible numbers, it turns out writing to TMDS regs is a bad move on LVDS. 14962 Also, change FP_CONTROL handling a bit for LVDS (bit31 needs preserving) 14963 14964commit d7be5bdc27fee0758df33feb3a331c9c79dd94e0 14965Author: Stuart Bennett <sb476@cam.ac.uk> 14966Date: Tue Nov 6 19:53:08 2007 +0000 14967 14968 Start using OUTPUT_PANEL where appropriate 14969 14970commit f1160ac702a251708f2385b78c1657f95c2916ab 14971Author: Stuart Bennett <sb476@cam.ac.uk> 14972Date: Thu Nov 1 00:24:13 2007 +0000 14973 14974 Beautify DCB 1.2 weirdness, hopefully fixup G5 and use dcb_table more simply 14975 14976commit 38829fa5cdac1469a6a71a092a42885ea048a1ba 14977Author: Stuart Bennett <sb476@cam.ac.uk> 14978Date: Tue Nov 6 19:41:47 2007 +0000 14979 14980 Implement lvds_output_{get_modes, mode_valid, mode_fixup} 14981 14982commit 29c578bfe9415aa86910cf6d5cd44787b4fdf44d 14983Author: Stuart Bennett <sb476@cam.ac.uk> 14984Date: Fri Oct 26 03:01:17 2007 +0100 14985 14986 Reindentation, renaming, adding debug, and assume LVDS always connected 14987 14988commit 7aeae6cdc99ab633361f107752ef1bfa3e0492e0 14989Author: Stuart Bennett <sb476@cam.ac.uk> 14990Date: Wed Oct 31 17:29:16 2007 +0000 14991 14992 LVDS VBIOS FP mode finding 14993 14994commit cd806f92e29573ea968ab2fddb454eb1513e37dd 14995Author: Maarten Maathuis <madman2003@gmail.com> 14996Date: Wed Nov 7 00:16:33 2007 +0100 14997 14998 randr12: Fix ugly alligment. 14999 15000commit 88faf35179b9af8060c35a7a95e9a6baaae552be 15001Author: Maarten Maathuis <madman2003@gmail.com> 15002Date: Wed Nov 7 00:11:51 2007 +0100 15003 15004 randr12: Improve hotplugging of outputs. 15005 15006commit 4aa3acbd5410c8dda57a483cf7967258440888dc 15007Author: Peter Winters <peter@whoei.org> 15008Date: Tue Nov 6 23:40:07 2007 +0100 15009 15010 NV10 EXA: Implement a8 1x1 repeat 15011 15012commit 524e555c468fbab32b515ecd2c1e9cccf6c48443 15013Author: Peter Winters <peter@whoei.org> 15014Date: Tue Nov 6 23:29:03 2007 +0100 15015 15016 NV10 EXA: setup viewport like nv40 exa 15017 15018commit 2f2298f0ae3777c775b3ab81356adebf4f52ed77 15019Author: Maarten Maathuis <madman2003@gmail.com> 15020Date: Tue Nov 6 22:15:20 2007 +0100 15021 15022 randr12: Readd crt load detect + misc changes + restrictions (see full log) 15023 15024 - Crt load detect has been readded, but only for the primary output (the other gives false positives). 15025 - Misc changes to ramdac management. 15026 - Until i know how to trigger a full mode change, i cannot allow adding an output which requires ramdac0 during runtime. 15027 15028commit a6244c25a5c7efc633c74f506ca121e10a5642c0 15029Author: Maarten Maathuis <madman2003@gmail.com> 15030Date: Tue Nov 6 16:11:43 2007 +0100 15031 15032 randr12: Reasonably complete TMDS programming, see notes. 15033 15034 - Please file a bug and make a mmio-trace when it fails for you. 15035 - I can now hotplug TMDS, so that makes both types hotpluggable on nv4x. 15036 - I have no clue what happens on pre-nv40 hardware, so please try. 15037 15038commit 5f6912f083048f01a33582d45bc6b22f0ac27d4e 15039Author: Maarten Maathuis <madman2003@gmail.com> 15040Date: Tue Nov 6 10:56:22 2007 +0100 15041 15042 randr12: Associate VPLL's with ramdac instead of crtc. 15043 15044commit 8241710c94f6df0bf683bc3c93f7ea1ca14d118c 15045Author: Maarten Maathuis <madman2003@gmail.com> 15046Date: Tue Nov 6 10:34:56 2007 +0100 15047 15048 randr12: Put TEST_CONTROL into neutral position. 15049 15050commit 19533d8ba549965dedd9083330b217270efaa49b 15051Author: Maarten Maathuis <madman2003@gmail.com> 15052Date: Mon Nov 5 23:32:16 2007 +0100 15053 15054 NV50: Don't hit pre-NV50 codepath. 15055 15056commit 5ca10da0c7dd2fd689cc4369e37ed385a0c10d21 15057Author: Maarten Maathuis <madman2003@gmail.com> 15058Date: Mon Nov 5 21:28:22 2007 +0100 15059 15060 randr12: Remove incorrect (and hopefully redundant) output initialisation check. 15061 15062commit fcacad59d9233a66111d1c434e14cc4b2a08961e 15063Author: Maarten Maathuis <madman2003@gmail.com> 15064Date: Mon Nov 5 21:18:13 2007 +0100 15065 15066 randr12: Do not remove any bits from crtc register 580. 15067 15068commit e58c6735c73cc3a94a7a139e235b1dff9220e507 15069Author: Maarten Maathuis <madman2003@gmail.com> 15070Date: Mon Nov 5 20:30:41 2007 +0100 15071 15072 randr12: Only use VPLL as timing source. 15073 15074commit 1071aeef24b349bac08a2ac091733c23cad1cb44 15075Author: Maarten Maathuis <madman2003@gmail.com> 15076Date: Mon Nov 5 20:04:26 2007 +0100 15077 15078 randr12: Only enable crtc at the very last stage, should avoid mode setting "blinking". 15079 15080commit 1ac7798bdec75d9336afd1cfbe7fa177f5fe1465 15081Author: Maarten Maathuis <madman2003@gmail.com> 15082Date: Mon Nov 5 10:03:15 2007 +0100 15083 15084 randr12: Add a native mode for TMDS and make that preferred + some ddc related fixes. 15085 15086commit bfb75679e504b335be0de4384830cc8129f92c1b 15087Author: Ben Skeggs <skeggsb@gmail.com> 15088Date: Mon Nov 5 11:44:58 2007 +1100 15089 15090 Fix a thinko from commit 62294d9be8ec6f1159e3c2ad273649f937b7aa7e. 15091 15092commit bf8a2dfa7f83956d100316ed7450669795ea47d2 15093Author: Maarten Maathuis <madman2003@gmail.com> 15094Date: Sat Nov 3 22:03:18 2007 +0100 15095 15096 randr12; Forgot a small thing. 15097 15098commit 19b0971d0c1d03896bfaee2b833194ef1e4d7566 15099Author: Maarten Maathuis <madman2003@gmail.com> 15100Date: Sat Nov 3 21:54:48 2007 +0100 15101 15102 randr12: A legitimate DPMS call could reset the ramdac and effectively disable the mouse. 15103 15104commit a29839c718fd6bf3003081299017536dd9f8223e 15105Author: Maarten Maathuis <madman2003@gmail.com> 15106Date: Sat Nov 3 10:00:06 2007 +0100 15107 15108 randr12: Hopefully improve the native flatpanel mode finder. 15109 15110commit bd20ff5b8c9467ca913c00d3ec946e7e975ad3d4 15111Author: Maarten Maathuis <madman2003@gmail.com> 15112Date: Fri Nov 2 23:17:16 2007 +0100 15113 15114 randr12: We cannot reset ramdacs during detect, there may not be a modesetting after it. 15115 15116commit 7054de29c589a7be0a6b027f3067ae2c02e2fc45 15117Author: Maarten Maathuis <madman2003@gmail.com> 15118Date: Fri Nov 2 21:44:40 2007 +0100 15119 15120 randr12: Futher refine crtc timings, hoping to solve the "pink line" issue. 15121 15122commit 39ced4abb536993249e55a72eceeba26d577724c 15123Author: Maarten Maathuis <madman2003@gmail.com> 15124Date: Thu Nov 1 17:54:26 2007 +0100 15125 15126 randr12: Improved ramdac choosing, based on bios information (or value). 15127 15128commit 28ba9ed765d9956b2f07c6a405fca675a71012db 15129Author: Maarten Maathuis <madman2003@gmail.com> 15130Date: Thu Nov 1 17:01:53 2007 +0100 15131 15132 randr12: Remove redundant output setup + don't print full edid. 15133 15134commit 609f7b9c2962ccf04861d2164a0ff12c58e84010 15135Author: Maarten Maathuis <madman2003@gmail.com> 15136Date: Thu Nov 1 13:00:45 2007 +0100 15137 15138 randr12: Cursor fixes(probably only affects nv4x). 15139 15140commit 6945dfdc5e6b2540432805e845cb8540dc00d326 15141Author: Maarten Maathuis <madman2003@gmail.com> 15142Date: Thu Nov 1 11:10:24 2007 +0100 15143 15144 randr12: Improve analog monitor allignment + misc fixes. 15145 15146commit 5bd8ba52788b3b3b8f91ba59c29e154e38482481 15147Author: Ben Skeggs <skeggsb@gmail.com> 15148Date: Thu Nov 1 18:35:33 2007 +1100 15149 15150 nv40: what can I say? oops! 15151 15152commit f59e596bcd90ef824cd41e0c37952e574d6914bb 15153Author: Ben Skeggs <skeggsb@gmail.com> 15154Date: Thu Nov 1 18:17:20 2007 +1100 15155 15156 nv40: completely untested fix for composite on ppc 15157 15158 Thanks go to ps3rsx guys for finding the issue. 15159 15160commit 59c4fdb650ff2e6b6948d1bdd6bb22aa94195a32 15161Author: Maarten Maathuis <madman2003@gmail.com> 15162Date: Wed Oct 31 22:59:54 2007 +0100 15163 15164 randr12: Some misc cleanup to make malc0's patches work. 15165 15166commit f165d4f2ca90916b1334cdc18d62b82ea71d9f62 15167Author: Maarten Maathuis <madman2003@gmail.com> 15168Date: Wed Oct 31 21:30:36 2007 +0100 15169 15170 randr12: The "easy" TMDS registers, please report regressions. 15171 15172commit 860c4bdd38fd259ae51e32e4873a2aa8b1e6883e 15173Author: Maarten Maathuis <madman2003@gmail.com> 15174Date: Wed Oct 31 19:00:28 2007 +0100 15175 15176 randr12: Misc register fixes. 15177 15178commit d758e8f4c1a7d4effe1298a0a41d4e0b96496ab9 15179Author: Maarten Maathuis <madman2003@gmail.com> 15180Date: Wed Oct 31 13:22:29 2007 +0100 15181 15182 randr12: Some changes to VCLK ratio handling. 15183 15184commit 899559921b0e46db010fa1e771203c2edefb7ade 15185Author: Maarten Maathuis <madman2003@gmail.com> 15186Date: Wed Oct 31 09:18:59 2007 +0100 15187 15188 randr12: some blob copying 15189 15190commit b5ecadb73335d934a8cb1c106983d219c9d8d399 15191Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15192Date: Wed Oct 31 02:09:08 2007 +0100 15193 15194 Fix bugs in nv30 exa (kde text seems back) 15195 15196commit f7b553a87e7f67138e06e74877145a1f83e9d188 15197Author: Peter Winters <peter@whoei.org> 15198Date: Tue Oct 30 23:04:28 2007 +0100 15199 15200 NV10 EXA: implement repeat for 1x1 textures 15201 15202commit bf7b0231db1f7e45cee60d7d2e04b1b006aec2b0 15203Author: Peter Winters <peter@whoei.org> 15204Date: Sun Oct 28 23:52:22 2007 +0100 15205 15206 NV10 EXA: Fix viewport setup 15207 15208commit f8f9c8789b89a02886702c46c5feafec169ae78a 15209Author: Maarten Maathuis <madman2003@gmail.com> 15210Date: Tue Oct 30 21:00:54 2007 +0100 15211 15212 randr12: Flatpanel scaling is finally working, read full message. 15213 15214 - The option FPScale is now obeyed and will cause scaling to happen in the monitor. 15215 - The default is scaling is using the GPU. 15216 - When the aspect ratio is very different(>0.10 diff), it will obey aspect ratio and create bands around it. 15217 15218commit e73aa148d965abbf73392c6ffef3993dc26c31fd 15219Author: Maarten Maathuis <madman2003@gmail.com> 15220Date: Tue Oct 30 13:54:25 2007 +0100 15221 15222 randr12: Some copyright stuff. 15223 15224commit c1508e1448e3822e8b6bb2604c09a9d766a64a93 15225Author: Maarten Maathuis <madman2003@gmail.com> 15226Date: Tue Oct 30 13:44:44 2007 +0100 15227 15228 randr12: Disable some panel scaling stuff, because it is broken. 15229 15230commit 3bf8110b61ef854af803a45f0faf1eaffbbcfac9 15231Author: Maarten Maathuis <madman2003@gmail.com> 15232Date: Tue Oct 30 11:44:58 2007 +0100 15233 15234 randr12: Lots of misc fixes. 15235 15236 - This does not fix the issue with -hsync, +vsync modelines (or the other way around) yet. 15237 - There are two PVIO ranges, i'm surprised noone noticed this. 15238 - Some misc fixes. 15239 15240commit 894f9aacce3d685b811f8961275a9f225b964159 15241Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15242Date: Tue Oct 30 01:49:38 2007 +0100 15243 15244 Disable nv30 exa no big endian until we fix the endianness bugs related to the 3D engine. 15245 15246commit 482cbcb9efdfbdd4bbe998c09f7d29033aeca09b 15247Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15248Date: Tue Oct 30 01:37:45 2007 +0100 15249 15250 Use the new header with license. 15251 15252commit a9908dbb70df0d0abb586b9774efdc10adf4d74f 15253Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15254Date: Mon Oct 29 23:00:14 2007 +0100 15255 15256 More copyrights I remember... 15257 15258commit 67e142394edcb553a40c6c8b0ef091bd4dad2bdd 15259Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15260Date: Mon Oct 29 22:57:30 2007 +0100 15261 15262 Add some copyrights... not all files are covered yet, but I did the ones I recalled. 15263 15264commit f75e36a901fd47bbd9164242c408e9f8b0e03e2c 15265Author: Maarten Maathuis <madman2003@gmail.com> 15266Date: Mon Oct 29 18:37:34 2007 +0100 15267 15268 randr12: My suspicion was incorrect, i2c buses are swapped for all nv4x 15269 it seems. 15270 15271commit e741ac76ac45e8ea959792810670abdf2afbc6bd 15272Author: Maarten Maathuis <madman2003@gmail.com> 15273Date: Mon Oct 29 13:40:30 2007 +0100 15274 15275 randr12: Show NV_RAMDAC_FP_DEBUG_0 of both ramdac's when output is refused. 15276 15277commit c53754162087a7b66d40b7c3982cdc5b6d160ea1 15278Author: Maarten Maathuis <madman2003@gmail.com> 15279Date: Mon Oct 29 13:34:20 2007 +0100 15280 15281 randr12: Minor change, turning on bit 18 of NV_RAMDAC_SEL_CLK. 15282 15283commit fef872bd999b430a41bdfae8d408f8dfed0010a2 15284Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15285Date: Fri Oct 26 15:50:40 2007 +0200 15286 15287 nv30 exa : add A8 dst (requires new drm), add some dst formats, handle 1x1 repeat, enable by default. 15288 15289commit 1d7354cbb5875eb13ed88e6aaab69aff615f82b2 15290Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15291Date: Fri Oct 26 04:37:51 2007 +0200 15292 15293 Fix nv30 exa. 15294 15295commit 2cd3d47713cf3eb0d36124902a2c66b689acba51 15296Author: Arthur Huillet <arthur.huillet@free.fr> 15297Date: Thu Oct 25 16:55:18 2007 +0200 15298 15299 s/#define ABC ;/#define ABC do { ; } while(0);/ 15300 15301commit 228b91d3da445fc58bfe6b13e792bbdff3728301 15302Author: Arthur Huillet <arthur.huillet@free.fr> 15303Date: Thu Oct 25 16:40:34 2007 +0200 15304 15305 NV10 EXA: implemented masking via register combiners and added (disabled by default) statistics displaying 15306 15307 masking works for ARGB, XRGB (it's a nop), A8 - no component alpha for now, but it will come soon hopefully 15308 15309commit 7d77ab0692126cea9248b7ee61b001e138b4e47c 15310Author: Arthur Huillet <arthur.huillet@free.fr> 15311Date: Thu Oct 25 10:52:59 2007 +0200 15312 15313 NV10 EXA: fixed A8 as pointed out by p0g 15314 15315commit cdb54271837564bb4a5afd2dc4cf7e7169138975 15316Author: Arthur Huillet <arthur.huillet@free.fr> 15317Date: Thu Oct 25 10:38:42 2007 +0200 15318 15319 NV10 EXA: added support for A8 as source 15320 15321commit 220acbe07e84015e489257a9b9dae8e8f1be8268 15322Author: Peter Winters <peter@whoei.org> 15323Date: Thu Oct 25 10:03:39 2007 +0200 15324 15325 NV10 EXA: use proper XRGB format and use correct RC setup 15326 15327commit 6034387ea96816d34b627c306998342136ecb01b 15328Author: Peter Winters <peter@whoei.org> 15329Date: Thu Oct 25 09:59:01 2007 +0200 15330 15331 NV10 EXA: always use an even width for NPOT_SIZE 15332 15333commit f7f17b789e89aadafa2734cba0f4464c5f317a68 15334Author: Jeremy Kolb <jkolb@brandeis.edu> 15335Date: Wed Oct 24 19:34:49 2007 -0400 15336 15337 nv30: Fix some corruption. Give us some semblance of icons (still messed up though) 15338 15339commit bafa34a7bd7a9136a95c8ced25c5b80556c78c0c 15340Author: Maarten Maathuis <madman2003@gmail.com> 15341Date: Wed Oct 24 20:28:46 2007 +0200 15342 15343 Fix a few NV_IMAGE_BLIT offsets. 15344 15345commit 09f4b66d35543037c4f50910d0265751dded66c0 15346Author: Arthur Huillet <arthur.huillet@free.fr> 15347Date: Wed Oct 24 10:40:17 2007 +0200 15348 15349 NV10EXA : enabled by default for NV1x. Waiting for pq before doing the same for NV2x. 15350 15351commit 6c978164a212ddf4f60f82cdc72b11ebbc96efbb 15352Author: Arthur Huillet <arthur.huillet@free.fr> 15353Date: Wed Oct 24 09:37:34 2007 +0200 15354 15355 Xv: disabled YV12 overlay for NV20 chipsets that do not seem to have it 15356 15357commit 9f48bd410f8cf9aa700e03db807cef2e7cc07e34 15358Author: Maarten Maathuis <madman2003@gmail.com> 15359Date: Wed Oct 24 09:08:41 2007 +0200 15360 15361 randr12: The last step to make lower flatpanel resolutions working. 15362 15363 - Note: the previous randr12 commit also contained some fixes in this area. 15364 15365commit 2bf8bbd1764979d194ff2dbac4a6529cbcbdbc1d 15366Author: Maarten Maathuis <madman2003@gmail.com> 15367Date: Wed Oct 24 00:00:59 2007 +0200 15368 15369 randr12: Misc changes. 15370 15371commit 5841664c82126058e582f215af2323cc80a6150d 15372Author: Arthur Huillet <arthur.huillet@free.fr> 15373Date: Tue Oct 23 22:53:46 2007 +0200 15374 15375 x8r8g8b8 now seems to work (it passes rendercheck) with format 0x380 15376 15377commit 27a713057f0fd2d3c8a77f5eff86393059b60511 15378Author: Arthur Huillet <arthur.huillet@free.fr> 15379Date: Tue Oct 23 22:53:21 2007 +0200 15380 15381 Remove NV10 texture formats that are all wrong. 15382 15383commit 78a0d5a42f96816889f1c9294f11c637dfcba486 15384Author: Arthur Huillet <arthur.huillet@free.fr> 15385Date: Tue Oct 23 19:12:19 2007 +0200 15386 15387 NV10 EXA: A8R8G8B8 without mask works with rendercheck now 15388 15389commit 7543e65cecab510f6bea8b6efc5cdf3cc9d261ae 15390Author: Maarten Maathuis <madman2003@gmail.com> 15391Date: Tue Oct 23 18:41:38 2007 +0200 15392 15393 randr12: actually add scaling factor's 15394 15395 - This was forgotten when i first added this from the haiku driver. 15396 15397commit a11faa794c5dd24cd74d949006c8e86c6c3f0310 15398Author: Maarten Maathuis <madman2003@gmail.com> 15399Date: Tue Oct 23 16:22:56 2007 +0200 15400 15401 randr12: Add a few debug statements that should explain why an output is refused. 15402 15403commit 2d842c39bf210b4b8294a447ec9ac194ec3e2e47 15404Author: Maarten Maathuis <madman2003@gmail.com> 15405Date: Tue Oct 23 16:19:10 2007 +0200 15406 15407 randr12: Check if output is needed before creating it. 15408 15409 - This hopefully fixes a problem some NV30 people have. 15410 15411commit 5c884a188296c9a026f110f272dff5d88d58c0b6 15412Author: Maarten Maathuis <madman2003@gmail.com> 15413Date: Tue Oct 23 15:47:49 2007 +0200 15414 15415 Revert a G70 workaround, this may BREAK your setup. 15416 15417 - Please report if you have issues, but be prepared to help out to find a real fix. 15418 15419commit 0fb93b45378c9d0fbee9bb96bd60d7a995c667b9 15420Author: Arthur Huillet <arthur.huillet@free.fr> 15421Date: Mon Oct 22 19:02:17 2007 +0200 15422 15423 added (disabled) NV10 EXA hooks 15424 15425commit 78af60e798f5b91cc128cd2f84ddae7d0a1523cd 15426Author: Arthur Huillet <arthur.huillet@free.fr> 15427Date: Mon Oct 22 19:00:20 2007 +0200 15428 15429 added NV10 EXA functions to proto header 15430 15431commit cf053adacabaee887ecedaa9967b07b3185095b5 15432Author: Peter Winters <peter@whoei.org> 15433Date: Mon Oct 22 17:04:23 2007 +0200 15434 15435 Fix some blending stuff for NV10 EXA 15436 15437commit b0605fbeef0765e39554a15ea9f871c19258f72c 15438Author: Peter Winters <peter@whoei.org> 15439Date: Mon Oct 22 17:01:26 2007 +0200 15440 15441 Add basic RC setup for NV10 EXA 15442 15443commit 8dea36f6533c5fc35dcb3c24e2fe0c1d009b6693 15444Author: Peter Winters <peter@whoei.org> 15445Date: Mon Oct 22 16:50:47 2007 +0200 15446 15447 Fix NV10 EXA transform 15448 15449commit b9f97583979d7dbbf0058de9f807a97812362c71 15450Author: Peter Winters <peter@whoei.org> 15451Date: Mon Oct 22 16:48:53 2007 +0200 15452 15453 Fix NV10 EXA list of supported formats 15454 15455commit 3dc9778902d9412c469d79e3d437838a6c9da7a5 15456Author: Peter Winters <peter@whoei.org> 15457Date: Mon Oct 22 16:44:08 2007 +0200 15458 15459 Various NV10 EXA fixes 15460 15461commit 391e2dd1216f598538524fdc52a14bfdd4451567 15462Author: Ben Skeggs <skeggsb@gmail.com> 15463Date: Mon Oct 22 22:48:13 2007 +1100 15464 15465 Fix bad s/ 15466 15467commit 162edc70769b24b39f432cb79f5959eff363c564 15468Author: Pekka Paalanen <pq@iki.fi> 15469Date: Sun Oct 21 19:55:41 2007 +0300 15470 15471 Revert "EXA: now caching BLIT_OPERATION and RECT_OPERATION..." 15472 15473 This reverts commit 6d8caf5e0dd915809152c52c2c56a39d76e2ed8c. 15474 15475 Conflicts: 15476 15477 src/nv_exa.c 15478 15479 The above mentioned commit broke EXA copy and solid operations on nv20 15480 after display mode change. Triggered by running freedroid in fullscreen. 15481 (Not freedroidRPG) 15482 15483commit cc04b651348768dc214f59211ba396343aaa6a17 15484Author: Maarten Maathuis <madman2003@gmail.com> 15485Date: Sat Oct 20 20:30:02 2007 +0200 15486 15487 Add back comment. 15488 15489commit 6fd262027c74d5008e484bb5f09edbf5f657470b 15490Author: Maarten Maathuis <madman2003@gmail.com> 15491Date: Sat Oct 20 00:26:02 2007 +0200 15492 15493 randr12: Fix critical bug and minor change. 15494 15495 - The previous code could allow the TMDS[0x4] register of both heads to be set to 0x88. 15496 - This caused problems and freaked out my monitor and scared me as well :-( 15497 - Until further notice, the usage of this swap bit is restricted to very specific situations. 15498 15499commit 377f69172b4935372538d7bbf26852bf8a1fe1c4 15500Author: Maarten Maathuis <madman2003@gmail.com> 15501Date: Fri Oct 19 20:57:58 2007 +0200 15502 15503 randr12: Some changes to crtc regs. 15504 15505commit 85dac1041b38127d41832a6a322d01f87d70f28f 15506Author: Maarten Maathuis <madman2003@gmail.com> 15507Date: Fri Oct 19 08:06:33 2007 +0200 15508 15509 randr12: Some changes to output detection. 15510 15511 - I'm aware that we still can't fully program modes. 15512 15513commit 8320f4f61ca72ee0101c1d552ea3ef28e573d21d 15514Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15515Date: Thu Oct 18 01:56:17 2007 +0200 15516 15517 Remove xaa includes. 15518 15519commit 86855f03945d3bad90da0b28311bece484cbfce7 15520Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15521Date: Thu Oct 18 01:55:06 2007 +0200 15522 15523 Handle -minline-all-stringops properly 15524 15525commit 53ee9726eeda8cf844d32a9563c82b7549ba6c03 15526Author: Ben Skeggs <skeggsb@gmail.com> 15527Date: Wed Oct 17 23:28:50 2007 +1100 15528 15529 re-align nv40exa after ring macro changes 15530 15531commit a5c4786aaeadce6103fcc9b3a6f4e6f25a66730f 15532Author: Ben Skeggs <skeggsb@gmail.com> 15533Date: Wed Oct 17 23:22:05 2007 +1100 15534 15535 Switch to *_RING instead of NVDma* 15536 15537commit 0d1835d67e38ca12347a7965ac43fbfc29063b87 15538Author: Ben Skeggs <skeggsb@gmail.com> 15539Date: Wed Oct 17 16:29:01 2007 +1100 15540 15541 Switch to header generated from XML + other cleanups. 15542 15543 - Renamed object header to nouveau_class.h, so it's not confused with mmio 15544 - Removed usage of method defines in nv_dma.h 15545 - Removed 2D engine EXA composite hook, it was broken, useless and soon to 15546 be obsoleted by card-specific versions. 15547 - Use names in a few places that magic values were previously used. 15548 15549commit b36fbf174849b04928efda2909c461b6c1688cdc 15550Author: Dave Airlie <airlied@optimus.(none)> 15551Date: Tue Oct 16 09:28:58 2007 +1000 15552 15553 nouveau: missing include 15554 15555commit 8425d93f7f932281a1b9b420e676acba2a2b5ecd 15556Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15557Date: Sun Oct 14 17:10:59 2007 +0200 15558 15559 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). 15560 15561commit 212a4cd8884ff9276f64bdd0a0115782d146e36a 15562Author: Maarten Maathuis <madman2003@gmail.com> 15563Date: Sat Oct 13 23:02:27 2007 +0200 15564 15565 randr12: Some fallback defaults for dual-dvi cards, please submit dumps when you can. 15566 15567commit 6d8caf5e0dd915809152c52c2c56a39d76e2ed8c 15568Author: Arthur Huillet <arthur.huillet@free.fr> 15569Date: Sat Oct 13 22:28:39 2007 +0200 15570 15571 EXA: now caching BLIT_OPERATION and RECT_OPERATION (0x2FC) to limit the number of IRQs on NV04 15572 15573commit ceaf8df424651e0f20d63f817864f3e8657f2c21 15574Author: Maarten Maathuis <madman2003@gmail.com> 15575Date: Sat Oct 13 21:45:50 2007 +0200 15576 15577 randr12: Revert a small unnecesary change. 15578 15579commit 7171acbe7f174dcbe8c28c7f1fcb7481f2fafd90 15580Author: Maarten Maathuis <madman2003@gmail.com> 15581Date: Sat Oct 13 21:33:07 2007 +0200 15582 15583 randr12: Hopefully fix panel size detection and allow negative hsync/vsync polarities. 15584 15585commit be08f09ad1ba6aeebf4491727052d246d5568fd2 15586Author: Maarten Maathuis <madman2003@gmail.com> 15587Date: Sat Oct 13 17:26:32 2007 +0200 15588 15589 randr12: Some debug stuff. 15590 15591commit 418e6117a25e51d5b0dbcd87c17f78f9b04fa588 15592Author: Maarten Maathuis <madman2003@gmail.com> 15593Date: Sat Oct 13 12:52:10 2007 +0200 15594 15595 randr12: Remove a double reg and misc changes. 15596 15597commit e30e21ed0ee8b1bd112780badc933dc4552a2e81 15598Author: Maarten Maathuis <madman2003@gmail.com> 15599Date: Fri Oct 12 19:49:32 2007 +0200 15600 15601 randr12: Some misc changes. 15602 15603commit 9425f7a3ea2d80f4df79239e1ca8000bd729c772 15604Author: Jeremy Kolb <jkolb@brandeis.edu> 15605Date: Thu Oct 11 20:32:47 2007 -0400 15606 15607 Remove division in NV30EXATransformCoord. We want coords between 0 and texture width, height. 15608 15609commit 879e653867552f6bb595160670c8325bcb5a0d3b 15610Author: Maarten Maathuis <madman2003@gmail.com> 15611Date: Wed Oct 10 21:40:13 2007 +0200 15612 15613 randr12: Misc fix. 15614 15615commit 99dd9da140a5750b7c4d6030a0e8e1f7fca44033 15616Author: Maarten Maathuis <madman2003@gmail.com> 15617Date: Wed Oct 10 20:08:28 2007 +0200 15618 15619 randr12: Fix switch mode function, so fullscreen apps don't screw up your screen by definition ;-) 15620 15621commit 0c80565e2cb7fa5ff359f57cc3d7016a78a43ac2 15622Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15623Date: Wed Oct 10 14:43:34 2007 +0200 15624 15625 Oops. 15626 15627commit 761ef4bd32f42fb77a3e48985fd6dda81aa081f5 15628Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15629Date: Wed Oct 10 14:42:46 2007 +0200 15630 15631 Try to do something for our desastrous UTS/DFS performance. My PCI card is really hurting. 15632 15633commit 254f9b075034c86e5936954b52f499d573880039 15634Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15635Date: Wed Oct 10 02:34:06 2007 +0200 15636 15637 Still more work on nv30 exa... now starts but nothing works. 15638 15639commit 79dc3ad2c89eef74596a4c8e5714c2c661f56fee 15640Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15641Date: Wed Oct 10 00:48:46 2007 +0200 15642 15643 Don't swap the registers from user space. This is very wrong, and the drm will handle that soon. 15644 15645commit 2c08930f0b2bf06282976d6a357daf7f45405cdc 15646Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15647Date: Wed Oct 10 00:48:06 2007 +0200 15648 15649 Revert back to the other shaders. The terminator instruction doesn't seem to help, and breaks the A8 hack. 15650 15651commit 90d2457125b059916f9b9dd5ddb7cec73a554c55 15652Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15653Date: Tue Oct 9 04:38:29 2007 +0200 15654 15655 Cleanup some nv30 exa issues, add debbuging info to the composite hook. 15656 15657commit e184fc802c7f9103bbc8ee8c036cfe5a9ddd9513 15658Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15659Date: Tue Oct 9 03:28:39 2007 +0200 15660 15661 Some fixes to nv30_exa. Now goes through a couple of composite() calls without crashing. 15662 15663commit 708f28dc8175375713cd9ae4d07b2cfcea3b106c 15664Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15665Date: Tue Oct 9 03:28:11 2007 +0200 15666 15667 Use terminators on the nv30/40 shaders. 15668 15669commit 43d5f747103721b3e0ccf4fe73494729ed34986c 15670Author: Maarten Maathuis <madman2003@gmail.com> 15671Date: Mon Oct 8 19:53:56 2007 +0200 15672 15673 randr12: Small changes makes finding regressions easy ;-) 15674 15675commit 62294d9be8ec6f1159e3c2ad273649f937b7aa7e 15676Author: Maarten Maathuis <madman2003@gmail.com> 15677Date: Sat Oct 6 19:04:51 2007 +0200 15678 15679 EXA: Starting from version 2.4 we can't use devPrivate.ptr anymore. 15680 15681commit 721de8da8fccd6f7b6e5a3044ce82616cdb689e8 15682Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15683Date: Fri Oct 5 23:50:18 2007 +0200 15684 15685 some more work on nv30 exa... 15686 15687commit 80bcbf493b6b2ad927c7a3826cc6e5fa46bca76e 15688Author: Maarten Maathuis <madman2003@gmail.com> 15689Date: Fri Oct 5 22:29:07 2007 +0200 15690 15691 randr12: Misc changes. 15692 15693commit 63ad301baab5803329ba0f192f12c917b435594e 15694Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15695Date: Fri Oct 5 17:20:05 2007 +0200 15696 15697 mandatory oops commit. 15698 15699commit 3773e0e5045a6c347bf26eff71e6bdd4c592cebc 15700Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15701Date: Fri Oct 5 17:18:40 2007 +0200 15702 15703 Split nv30 exa and nv40 exa. 15704 15705commit 37900537c0f8b8dfa3dc80300d6809f1f4ac9423 15706Author: Jeremy Kolb <jkolb@brandeis.edu> 15707Date: Thu Oct 4 23:15:38 2007 -0400 15708 15709 NV30 texture changes. Commented out table that appears to work for nv35 at least. 15710 15711commit 44eb015e88caba82c8d84fee4ea71768c9a6b2a3 15712Author: Jeremy Kolb <jkolb@brandeis.edu> 15713Date: Thu Oct 4 20:37:35 2007 -0400 15714 15715 Fix NV30 arch check. 15716 15717commit 5a76e5a751cd66021565ab92b2fc3c61204c7d73 15718Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15719Date: Thu Oct 4 23:46:43 2007 +0200 15720 15721 fix nv30_exa for nv40. 15722 15723commit d7b767fc951287a1c87965db3c236e05cc94d8f6 15724Author: Maarten Maathuis <madman2003@gmail.com> 15725Date: Thu Oct 4 21:49:33 2007 +0200 15726 15727 NV50: Fix a minor mistake. 15728 15729commit 926e887a4b374768648fe772b1b3d7730096e45f 15730Author: Maarten Maathuis <madman2003@gmail.com> 15731Date: Wed Oct 3 22:57:56 2007 +0200 15732 15733 randr12: Some reindenting. 15734 15735commit 7ee3f670077641e14a0ff024443805564d5be599 15736Author: Patrice Mandin <pmandin@caramail.com> 15737Date: Thu Oct 4 19:54:40 2007 +0200 15738 15739 nv10exa: init link between tcl and image blit for >=nv11 15740 15741commit 6510e75ada6fb921e417588d98afdc58b2c14972 15742Author: Jeremy Kolb <jkolb@brandeis.edu> 15743Date: Wed Oct 3 23:34:23 2007 -0400 15744 15745 nv30_exa: send down pitch correctly 15746 15747commit b7fc7b63eec4a5e23879120284c882e19e095471 15748Author: Jeremy Kolb <jkolb@brandeis.edu> 15749Date: Wed Oct 3 20:54:28 2007 -0400 15750 15751 Fill in some regs. 15752 15753commit 46facb7c29d2737f1cc414655de8e09ee0aea22f 15754Author: Maarten Maathuis <madman2003@gmail.com> 15755Date: Wed Oct 3 22:16:13 2007 +0200 15756 15757 randr12: Actually borrow some timing stuff from Haiku correctly ;-) + new crtc reg 15758 15759commit 9d6b3216ed519e89386480d04070b1c1e4702921 15760Author: Matthieu Castet <castet.matthieu@free.fr> 15761Date: Wed Oct 3 21:21:41 2007 +0200 15762 15763 abort on Init table command not found when parsing bios 15764 15765commit b6351c0efa3629a448557fc7e597ba14fbf057fb 15766Author: Maarten Maathuis <madman2003@gmail.com> 15767Date: Wed Oct 3 19:54:40 2007 +0200 15768 15769 NV50: Fix small mistake. 15770 15771commit f00ea962e549934965100dc6d2806709129995e8 15772Author: Maarten Maathuis <madman2003@gmail.com> 15773Date: Wed Oct 3 19:03:38 2007 +0200 15774 15775 randr12 + NV50: Misc changes. 15776 15777 - Switched everything back to head variable instead of pcio. 15778 - Removed ugly C() macro for NV50. 15779 - Probably some other stuff too. 15780 - Please test NV5x users ;-) 15781 15782commit 5d7f06c2a205004b78692b4dfad477ade3947fd5 15783Author: Maarten Maathuis <madman2003@gmail.com> 15784Date: Tue Oct 2 19:31:10 2007 +0200 15785 15786 NV50: Some infrastructure and minor changes. 15787 15788commit 7680648c9ddb81e2a1c6295431d7074366fb7b9a 15789Author: Maarten Maathuis <madman2003@gmail.com> 15790Date: Tue Oct 2 16:55:07 2007 +0200 15791 15792 NV50: The beginning of moving some code around. 15793 15794 - Please contact me if something breaks, i don't have an NV5x card. 15795 15796commit b812ecaef75216cc76557e6a3a080efa870d189f 15797Author: Maarten Maathuis <madman2003@gmail.com> 15798Date: Tue Oct 2 10:35:37 2007 +0200 15799 15800 NV30EXA: Minor code cleanup. 15801 15802commit 5b974b3bed86d53ccf6915c46aeb310a2bba3519 15803Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15804Date: Mon Oct 1 17:19:41 2007 +0200 15805 15806 Remove those useless casts. 15807 15808commit 734785c05357455a7ae7181fd7326f02e0bd0478 15809Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15810Date: Mon Oct 1 02:01:30 2007 +0200 15811 15812 Some work on nv10 exa. 15813 15814commit 0cfcbcc9d64bfd47b765f4f2ff41d7ad467990bd 15815Author: Matthieu Castet <castet.matthieu@free.fr> 15816Date: Sun Sep 30 22:41:15 2007 +0200 15817 15818 enable bios PINS parsing on post nv10 cards 15819 15820commit 9ac6d0e9d46abdf867fd06f7ca24b718a340bb74 15821Author: Pekka Paalanen <pq@iki.fi> 15822Date: Sun Sep 30 23:09:33 2007 +0300 15823 15824 nv20 inits nv10tcl. 15825 15826commit c3439aebe000a3c6dd716bd4739ce01321bee26b 15827Author: Maarten Maathuis <madman2003@gmail.com> 15828Date: Sun Sep 30 21:43:41 2007 +0200 15829 15830 Fix detection of NV04 and NV05 when using pci-rework. 15831 15832commit 9e070c09f68b0e3e41109360c455b3e2cc1dbb9c 15833Author: Maarten Maathuis <madman2003@gmail.com> 15834Date: Sun Sep 30 19:37:53 2007 +0200 15835 15836 randr12: Switch to proper cursor api + minor changes + small nv50 fix 15837 15838 - Remove the ugly cursor hack. 15839 - A function name on nv50 was still g80, this caused an implicit decleration. 15840 15841commit c7de03556a51009ce69b073d7ecfd01919d24c5f 15842Author: Maarten Maathuis <madman2003@gmail.com> 15843Date: Sun Sep 30 12:40:50 2007 +0200 15844 15845 NV30EXA: Fix rendering issues. 15846 15847 - Full non-composite tests using rendercheck. 15848 - The full test suite of composite has been run (it takes ~15 hours). 15849 - A partial cacomposite has been run (a8, a8r8g8b8, x8r8g8b8). 15850 - All the rendering tests work fine, as does the cairo-knockout demo. 15851 - Thanks to doener for reporting this issue. 15852 15853commit 0574dc5a8bd09216e41656c943c92a437bcf020b 15854Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15855Date: Sat Sep 29 20:20:47 2007 +0200 15856 15857 A little more work on nv10 exa. 15858 15859commit aae155f149af522e1a83168ed4ec5b7b5a40bb73 15860Author: Maarten Maathuis <madman2003@gmail.com> 15861Date: Sat Sep 29 17:28:10 2007 +0200 15862 15863 randr12: Some more CRTC regs. 15864 15865commit 8b03086ee36ca9c3e77f9df9eb4a1805072d2103 15866Author: Maarten Maathuis <madman2003@gmail.com> 15867Date: Sat Sep 29 14:33:38 2007 +0200 15868 15869 randr12: make crtc owner beheaviour consistent with blob + minor changes 15870 15871commit c572315a974746224dba1213e956a1b6189f175d 15872Author: Maarten Maathuis <madman2003@gmail.com> 15873Date: Fri Sep 28 19:58:39 2007 +0200 15874 15875 randr12: Minor changes and more debugging output. 15876 15877commit db36a1df4ba3dddf3669f491bfe6461e4fd490b6 15878Author: Maarten Maathuis <madman2003@gmail.com> 15879Date: Fri Sep 28 18:52:47 2007 +0200 15880 15881 randr12: Some more debugging output and some other minor changes. 15882 15883commit 2d21405c9c0bff36f47e28a5e4086dacae557d17 15884Author: Maarten Maathuis <madman2003@gmail.com> 15885Date: Fri Sep 28 13:02:51 2007 +0200 15886 15887 randr12: Preperation to allow differentation between crtc and pcio in the future. 15888 15889commit f235cf896c1da098ff635d5a882561f16f5fb4f7 15890Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 15891Date: Thu Sep 27 18:21:25 2007 +0200 15892 15893 Import nv10 exa code. Does not work yet. 15894 15895commit 74a243d8e188432854c15849ad1ab6e893e48f0a 15896Author: Maarten Maathuis <madman2003@gmail.com> 15897Date: Wed Sep 26 09:04:43 2007 +0200 15898 15899 randr12: Some registers were not constants, so changed it back with a note about the values i had seen. 15900 15901commit a3c71495097f8d745110aad06c9f8e2f6f03baad 15902Author: Maarten Maathuis <madman2003@gmail.com> 15903Date: Tue Sep 25 22:05:36 2007 +0200 15904 15905 randr12: Getting closer, doing a decent attempt at detecting the proper ramdac. 15906 15907 - crtc detection may need to be added 15908 - When i boot up with both analog and dvi attached, my dvi is misprogrammed, maybe some sharing is happening somewhere? 15909 15910commit 4a57ebb9c2098ce5d568a2da6e0640cdcc18d3f0 15911Author: Maarten Maathuis <madman2003@gmail.com> 15912Date: Tue Sep 25 21:06:03 2007 +0200 15913 15914 randr12: detection changes, more to come later 15915 15916 - Detection of crtc's (or a way to program them effectively) still has to be done/fixed. 15917 - It now detects if tmds controls is on another ramdac, for a non-dual dvi setup. 15918 - Disabled TMDS stuff for the moment, since it's incorrect somehow. 15919 - Dumps for single dvi, dual dvi, etc are appreciated. 15920 15921commit 41df1ff6ce348dc6a38e34ce63887c66b0300c09 15922Author: Maarten Maathuis <madman2003@gmail.com> 15923Date: Tue Sep 25 16:22:38 2007 +0200 15924 15925 randr12: Some reindenting. 15926 15927commit 334a9b065c4973876b894eadf05ce86ca439e08f 15928Author: Maarten Maathuis <madman2003@gmail.com> 15929Date: Tue Sep 25 16:17:03 2007 +0200 15930 15931 randr12: simplification of the swapped i2c buses "workaround" for nv4x and some code reindenting 15932 15933commit c40559dacd6fe20f35710111330dee37fb2b9a58 15934Author: Maarten Maathuis <madman2003@gmail.com> 15935Date: Sun Sep 23 01:16:55 2007 +0200 15936 15937 randr12: A frail attempt at some TMDS programming, i have not figured out yet what switches DVI from one RAMDAC to the other. 15938 15939commit 50bc4dbff6df80f5d80f731fe42b06768c429dc0 15940Author: Maarten Maathuis <madman2003@gmail.com> 15941Date: Sun Sep 23 00:57:18 2007 +0200 15942 15943 randr12: Reindenting and minor changes. 15944 15945commit 1a3303a5b6f261470c1c56f84f9b89af950a0b35 15946Author: Maarten Maathuis <madman2003@gmail.com> 15947Date: Sat Sep 22 22:21:37 2007 +0200 15948 15949 randr12: Assign ramdac's in more sane way, otherwise two outputs may end up on the same ramdac. 15950 15951commit 9b62853dfe1664f3624f6be8aa86f6ca15ca8392 15952Author: Maarten Maathuis <madman2003@gmail.com> 15953Date: Sat Sep 22 20:37:32 2007 +0200 15954 15955 randr12: Add some TMDS register infrastructure, so i can try some stuff in the future. 15956 15957commit a445cbcd71ee51d7680a6eb13cd037e06d59b0ba 15958Author: Maarten Maathuis <madman2003@gmail.com> 15959Date: Sat Sep 22 15:55:01 2007 +0200 15960 15961 randr12: hardcoding a base value for debug_0 register. 15962 15963commit f89a3217a18261424e212b355628ce5f5b8b9142 15964Author: Maarten Maathuis <madman2003@gmail.com> 15965Date: Fri Sep 21 23:16:40 2007 +0200 15966 15967 randr12: Some minor changes. 15968 15969commit 5ea59f8f01dbbd4211aee1a0e46a6529cddc016c 15970Author: Maarten Maathuis <madman2003@gmail.com> 15971Date: Fri Sep 21 20:37:50 2007 +0200 15972 15973 randr12: hardcode more stuff + actually use a few previous regs 15974 15975 - I'm going to try and do full modesetting, before tackling the problems that dual monitor setups face (i tried, it wasn't pretty). 15976 15977commit 0f66c725cb8806041faddf94cceb88009f7e9f06 15978Author: Dave Airlie <airlied@redhat.com> 15979Date: Thu Sep 20 08:29:43 2007 +1000 15980 15981 make distcheck pass 15982 15983commit bf3cceeb9c72c16a1cf601c8bd0dedd8c676e3cd 15984Author: Maarten Maathuis <madman2003@gmail.com> 15985Date: Mon Sep 17 18:58:31 2007 +0200 15986 15987 randr12: Add some CRTC regs which the blob sets. 15988 15989commit e3346c17c1ead9619206a5921994b5de7ed72739 15990Author: Maarten Maathuis <madman2003@gmail.com> 15991Date: Mon Sep 17 18:22:53 2007 +0200 15992 15993 randr12: Change the way CRTC register 59 is set. 15994 15995 - Small change, i suspect the original author had a dfp on the second crtc. 15996 - Information taken from a mmio trace of a nv28. 15997 15998commit 6c3aee9d68abdb9303cf03d9024f4f0deb5f1e3e 15999Author: Ben Skeggs <skeggsb@gmail.com> 16000Date: Mon Sep 17 00:01:22 2007 +1000 16001 16002 Make use of exaGetPixmapOffset(). 16003 16004commit 4d9e5d451a516aac64c4b95a4478f252c0c9eafb 16005Author: Ben Skeggs <skeggsb@gmail.com> 16006Date: Sun Sep 16 23:54:14 2007 +1000 16007 16008 Fix randr1.2 with driver-provided modes (ie. LVDS native mode) 16009 16010 Fixes #12368 16011 16012commit 5e58e74435da2071aab9f21386ac6dabdf3f8e7d 16013Author: Ben Skeggs <skeggsb@gmail.com> 16014Date: Sun Sep 16 19:44:42 2007 +1000 16015 16016 NV30EXA: Support PICT_a8 dest surfaces 16017 16018 Quick hack for now, shader handling will get a rework one day and clean all 16019 this up. 16020 16021commit d6d3ddfafaece2d153942ec37e3a8c3127c7c3fa 16022Author: Ben Skeggs <skeggsb@gmail.com> 16023Date: Sun Sep 16 19:01:21 2007 +1000 16024 16025 Bring G8x modesetting into sync with xf86-video-nv. 16026 16027commit 53e35a1d42070fd683eb6d0f169ac3ce482a45b0 16028Author: Ben Skeggs <skeggsb@gmail.com> 16029Date: Sun Sep 16 17:27:22 2007 +1000 16030 16031 Move G80 VBIOS download into nv_bios.c. 16032 16033commit 02d5443d650e628256ebed1f331ad32b0f642a38 16034Author: Arthur Huillet <arthur.huillet@free.fr> 16035Date: Fri Sep 14 21:13:26 2007 +0200 16036 16037 Xv: NV04 overlay now has one "colorkey" attribute 16038 16039 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 16040 to do it only if someone actually needs it. 16041 16042commit 0fa042e5123066ca2d9f36ff05cf64d0f68289a8 16043Author: Arthur Huillet <arthur.huillet@free.fr> 16044Date: Fri Sep 14 14:17:32 2007 +0200 16045 16046 Xv: NV04 overlay now almost functionnal 16047 16048commit a7ff07f4ab63c5a574e549326f022f84c2b9bab4 16049Author: Arthur Huillet <arthur.huillet@free.fr> 16050Date: Fri Sep 14 12:27:45 2007 +0200 16051 16052 Xv: quick imple of NV04 overlay 16053 16054commit 77076a391647f6111f38b10192d78ea7560ca9e8 16055Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16056Date: Wed Sep 12 13:10:37 2007 +0200 16057 16058 Try to fix nv4/tnt1 16059 16060commit e03fabf7e14599c1619673430ec3d81db90ffd1e 16061Author: Maarten Maathuis <madman2003@gmail.com> 16062Date: Tue Sep 11 20:15:11 2007 +0200 16063 16064 Restore PRAMDAC1 and PCRTC1 offsets, in proper format (0x2000/4) and move them to their proper place. 16065 16066commit 564043d9c1c9b5023f1b69a79d80742338256b83 16067Author: Maarten Maathuis <madman2003@gmail.com> 16068Date: Mon Sep 10 21:14:24 2007 +0200 16069 16070 Fix the memory adresses of the secondary crtc and ramdac + minor change. 16071 16072commit 5672842381a0e8f605cd482563c5d9ea02118647 16073Author: Patrice Mandin <pmandin@caramail.com> 16074Date: Mon Sep 10 18:29:10 2007 +0200 16075 16076 make randr 1.2 disableable for xorg server < 1.3 16077 16078commit 2778d3a9041ec62462585f623f84b1ed05b4ebad 16079Author: Patrice Mandin <pmandin@caramail.com> 16080Date: Mon Sep 10 18:28:03 2007 +0200 16081 16082 make randr 1.2 disableable for xorg server < 1.3 16083 16084commit 9bb76f73f0e648fda79a9c84d6e78469752a4eac 16085Author: Maarten Maathuis <madman2003@gmail.com> 16086Date: Mon Sep 10 07:57:01 2007 +0200 16087 16088 randr12: Apply the same pll restrictions to a two stage design. 16089 16090commit b46bbab81cc9f2093406fa8b53f9ffaf6f7fd0fc 16091Author: Maarten Maathuis <madman2003@gmail.com> 16092Date: Mon Sep 10 07:25:25 2007 +0200 16093 16094 randr12: Some more pll timer restrictions from Haiku. 16095 16096commit 77b424415d31428e113688ec27a405966d5080e6 16097Author: Maarten Maathuis <madman2003@gmail.com> 16098Date: Sun Sep 9 23:11:18 2007 +0200 16099 16100 randr12: Some new pll clock restrictions. 16101 16102commit a1bbfe17cc863497644b58a9206ae3876a654543 16103Author: Maarten Maathuis <madman2003@gmail.com> 16104Date: Sun Sep 9 22:50:19 2007 +0200 16105 16106 Some reindenting. 16107 16108commit dedb80d1237ebacf7531938d627328a20453d5d0 16109Author: Maarten Maathuis <madman2003@gmail.com> 16110Date: Sun Sep 9 22:12:38 2007 +0200 16111 16112 randr12: Modify dcb table for nv40 to make it look more like older hardware. Remove i2c bus reverse. 16113 16114commit 7aa96706ef428f1390d8e938d0bf26b8c046a67d 16115Author: Maarten Maathuis <madman2003@gmail.com> 16116Date: Sun Sep 9 20:40:20 2007 +0200 16117 16118 Removed a check for > 1 active crtc's, which seems really odd, no regressions occured for me. 16119 16120 - I cannot plug in both the vga and dvi connector of my monitor, so i cannot verify that. 16121 - The seperate monitor, on crtc0 and crtc1 work as they did (the afp being slightly shifted). 16122 - Also did a little reindenting. 16123 16124commit 228cb177242be053595b959dedfdfbfa77b1c54f 16125Author: Maarten Maathuis <madman2003@gmail.com> 16126Date: Sun Sep 9 18:49:08 2007 +0200 16127 16128 randr12: Fix a nasty regression from 6ba7e83f8cf2da84dd796ee59cb24c15bc36b746. 16129 16130 - For non-flatpanels the dac was never initialized. 16131 16132commit 0a56cb53d7e2a58ef044c9090f48fd46a2b251d9 16133Author: Maarten Maathuis <madman2003@gmail.com> 16134Date: Sun Sep 9 00:39:41 2007 +0200 16135 16136 randr12: Some missing init. 16137 16138commit 80bfb59a1f1c3968f8e116de611975df4acf5618 16139Author: Maarten Maathuis <madman2003@gmail.com> 16140Date: Sat Sep 8 23:58:39 2007 +0200 16141 16142 Some much needed indenting and readability improvements. 16143 16144commit 20aaacdd69c427f6c75b80f5f206b3187262f3a1 16145Author: Maarten Maathuis <madman2003@gmail.com> 16146Date: Sat Sep 8 21:00:45 2007 +0200 16147 16148 Hopefully fix bridged chip detection, such as the 7800GS AGP. 16149 16150commit b99549b5cb4f4cba2d60292994b2db483f8d4a8c 16151Author: Maarten Maathuis <madman2003@gmail.com> 16152Date: Sat Sep 8 14:25:39 2007 +0200 16153 16154 Remove list that is incorrect and no longer needed. 16155 16156commit b104314a11d3cf3bbb23095ff5805c1d5d8d9775 16157Author: Maarten Maathuis <madman2003@gmail.com> 16158Date: Sat Sep 8 14:24:54 2007 +0200 16159 16160 Fix previous commit, change the way Digital Flat Panels are programmed. 16161 16162 - This is mostly taken from the Haiku driver. 16163 - It works for me, please report any issues you have. 16164 - Added scaling, for running 4:3 resolutions on widescreen for example (untested). 16165 16166commit a867d6912456a5e15a965050bf784687e60634ee 16167Author: Maarten Maathuis <madman2003@gmail.com> 16168Date: Sat Sep 8 00:46:20 2007 +0200 16169 16170 Hopefully this will improve flatpanel beheaviour. 16171 16172 - I looked at Haiku and copied as much as i could. 16173 - I removed the htiming and vtiming hack. 16174 - Let's hope this works. 16175 16176commit 6ba7e83f8cf2da84dd796ee59cb24c15bc36b746 16177Author: Maarten Maathuis <madman2003@gmail.com> 16178Date: Fri Sep 7 23:11:51 2007 +0200 16179 16180 Some more indenting and such. 16181 16182commit d8d835dfb61f437b3eaede0a6ec188835f6eb5ff 16183Author: Maarten Maathuis <madman2003@gmail.com> 16184Date: Fri Sep 7 22:58:56 2007 +0200 16185 16186 Removing a hack in the hopes of finding a better way and some minor changes. 16187 16188commit dd7e568bd101ed34dcdd7d3b6d8169740499d593 16189Author: Maarten Maathuis <madman2003@gmail.com> 16190Date: Fri Sep 7 22:27:18 2007 +0200 16191 16192 Indenting and minor code style changes. 16193 16194commit 8ed256e94dae884b723c172320b15b0d6ae00a1b 16195Author: Maarten Maathuis <madman2003@gmail.com> 16196Date: Fri Sep 7 14:37:40 2007 +0200 16197 16198 randr12: Fix Xv blitter, please see full changelog for other implications. 16199 16200 - Removed the CRTCNumber variable, instead just store if crtc0 and crtc1 are active. 16201 - This coule break old style setups (picking the wrong monitor for example). 16202 - CRTCNumber option temporarily removed, because the piece of code surrounding it is ugly and needs a rework. 16203 16204commit eb595b2e06ec33d5fe2af43bc7ad107f551d99c8 16205Author: Maarten Maathuis <madman2003@gmail.com> 16206Date: Thu Sep 6 23:26:04 2007 +0200 16207 16208 randr-1.2: Only the NV4x have these busses swapped, revert for other architectures. 16209 16210 - If this is also true for NV3x, then please contact me. 16211 16212commit dd146a074fff575e18a41fd386e2a2ce5a1ec847 16213Author: Maarten Maathuis <madman2003@gmail.com> 16214Date: Thu Sep 6 22:55:16 2007 +0200 16215 16216 Update to latest pciaccess api, this now requires libpciaccess 0.10.0. 16217 16218commit 2f37d249605d6f0b80c290683ca35a9431b1078c 16219Author: Maarten Maathuis <madman2003@gmail.com> 16220Date: Thu Sep 6 22:00:30 2007 +0200 16221 16222 NV03 is dead. 16223 16224commit 2b2c99f2581e14d3f1726bc503951b1fe7760c95 16225Author: Maarten Maathuis <madman2003@gmail.com> 16226Date: Thu Sep 6 21:43:55 2007 +0200 16227 16228 randr12: A prettier hack for DFP, that uses several known htiming and vtiming modes. 16229 16230 - I'm fairly sure these modes are tied to resolution. 16231 - You are welcome to supply additional modes. 16232 - A final solution would be even better ;-) 16233 16234commit ab7d37d61a01afea218321d336664896eaec5651 16235Author: Arthur Huillet <arthur.huillet@free.fr> 16236Date: Thu Sep 6 15:09:59 2007 +0200 16237 16238 Xv: work around some overlay bandwidth problems - may break things though 16239 16240 extended nv workaround from bug 4686 in nv_dac.c to NV1x and NV2x cards, as it fixed problems on someone's NV17. 16241 16242commit f46bc9ad5558e9ecf6b174e24658e812811e94ce 16243Author: Maarten Maathuis <madman2003@gmail.com> 16244Date: Wed Sep 5 23:15:57 2007 +0200 16245 16246 Workaround for flatpanels. 16247 16248 - Remove the hardcoded values. 16249 - This should probably work for monitors whose mode is already set. 16250 16251commit 1825031465fc614a2a972ea7a8b44ac32aed8ea3 16252Author: Maarten Maathuis <madman2003@gmail.com> 16253Date: Wed Sep 5 21:23:25 2007 +0200 16254 16255 Fix minor issue. 16256 16257commit c0eefaba3f3706eb62e172bc57ac2795bd16bd56 16258Author: Maarten Maathuis <madman2003@gmail.com> 16259Date: Wed Sep 5 21:22:05 2007 +0200 16260 16261 Fix DPMS, patch by Bernhard Kaindl <bk@suse.de>. 16262 16263commit d4dda6bb972003394dd3326f8042fa5b759fc063 16264Author: Maarten Maathuis <madman2003@gmail.com> 16265Date: Wed Sep 5 21:18:51 2007 +0200 16266 16267 Reverse DDC1 and DDC2. 16268 16269 - I need them reversed. 16270 - I also found this: http://www.mail-archive.com/git-commits-head@vger.kernel.org/msg12009.html 16271 - If you need them otherwise, then please consult and hopefully we can solve it. 16272 16273commit da89d5a4e0263b8a89c549496d10ab43c7f1f230 16274Author: Maarten Maathuis <madman2003@gmail.com> 16275Date: Wed Sep 5 21:16:11 2007 +0200 16276 16277 randr-1.2 clients have their DDC's initialized elsewhere. 16278 16279commit 2b209d9b8907fad3bbaaf7297ff3c340f086bd24 16280Author: Maarten Maathuis <madman2003@gmail.com> 16281Date: Wed Sep 5 21:12:18 2007 +0200 16282 16283 LockUnlock CRTC's when going back to X (EnterVT), patch by Bernhard Kaindl <bk@suse.de>. 16284 16285commit ba7aa5a1525bb3fbc2ae7fbfce3e42a2e945626e 16286Author: Maarten Maathuis <madman2003@gmail.com> 16287Date: Wed Sep 5 18:34:27 2007 +0200 16288 16289 Xv: improved behavior with NoAccel by disabling blitter and overlay 16290 16291commit 42fd07b25d9abece034e972d1413b31fe78eba4c 16292Author: Ben Skeggs <skeggsb@gmail.com> 16293Date: Wed Sep 5 11:58:19 2007 +1000 16294 16295 Avoid segv. 16296 16297commit e74d58cdef0a0ee561e54f3d7d8afdfac030e04a 16298Author: Ben Skeggs <skeggsb@gmail.com> 16299Date: Wed Sep 5 11:53:48 2007 +1000 16300 16301 Add back DCB table parsing from randr-1.2 branch. 16302 16303commit 8a12758a6b7e4313e0269585b6f64eaffe54c74c 16304Author: Ben Skeggs <skeggsb@gmail.com> 16305Date: Wed Sep 5 11:51:35 2007 +1000 16306 16307 Add LVDS panels when found in DCB. 16308 16309commit b01fbbfa2b6be793dbd7695624699ed703ff4424 16310Author: Ben Skeggs <skeggsb@gmail.com> 16311Date: Wed Sep 5 11:34:14 2007 +1000 16312 16313 Nuke "comparison is always false..." warning. 16314 16315commit c53f495ec2994f013406d9738764ab124ed32061 16316Author: Ben Skeggs <skeggsb@gmail.com> 16317Date: Wed Sep 5 10:59:33 2007 +1000 16318 16319 Fix pciid detect on non-8800 cards (they identify as NV8x, not NV5x) 16320 16321 Actually, use 0x88000 to get pciid on all chipsets since NV40. Evidence 16322 suggests that NV4x has both the 0x88xxx regs and 0x18xx regs. 16323 16324commit 17d1a65296ae4fc524a31cd2fe4b9fe9484c0452 16325Author: Ben Skeggs <skeggsb@gmail.com> 16326Date: Wed Sep 5 10:53:32 2007 +1000 16327 16328 Neither overlay or blit adaptor work on G8x. 16329 16330commit 9ce1115cf26a7cf3212fdf446b83907f5795913a 16331Author: Ben Skeggs <skeggsb@gmail.com> 16332Date: Wed Sep 5 10:51:02 2007 +1000 16333 16334 Nuke riva_* files + bits of NV03 that got reincarnated. 16335 16336commit 51e3855104ff7b68559d37e774cde306424399d8 16337Author: Ben Skeggs <skeggsb@gmail.com> 16338Date: Wed Sep 5 10:46:54 2007 +1000 16339 16340 Nuke XAA. 16341 16342commit a0cb03d32b0da3ad2b4e03515926a3d616f487bc 16343Author: Maarten Maathuis <madman2003@gmail.com> 16344Date: Tue Sep 4 23:04:59 2007 +0200 16345 16346 There was some redundant arch checking code left, must have been missed after the randr merge. 16347 16348commit 13aeb83d7e5fcb3f051f95faa48a2ef636d67260 16349Author: Maarten Maathuis <madman2003@gmail.com> 16350Date: Tue Sep 4 22:54:28 2007 +0200 16351 16352 I unknowingly broke CRT's i think. 16353 16354commit a57fe870e49b2e7a3a6caf17fea8c8c878b1024c 16355Author: Maarten Maathuis <madman2003@gmail.com> 16356Date: Tue Sep 4 22:17:52 2007 +0200 16357 16358 Fix several problems. 16359 16360 - Implement a temporary NVSetMode(), suggestion and code by: Bernhard Kaindl <bk@suse.de>. 16361 Note: I've left his comment intact, just a minor change. 16362 - Fix last commit, i forgot to store the last mode. 16363 - Move some header stuff around, which i created without knowing the true purpose of nv_proto.h ;-) 16364 16365commit be468069b752a3a3c595f62eda40e90719496c90 16366Author: Maarten Maathuis <madman2003@gmail.com> 16367Date: Tue Sep 4 20:45:25 2007 +0200 16368 16369 Fix DVI/Digital alligment with some (magic?) values discovered by Matthew Garrett <mjg59@srcf.ucam.org>. 16370 16371commit 9b54aac1641c35ca3edb2d208d8a9a9c55737894 16372Author: Maarten Maathuis <madman2003@gmail.com> 16373Date: Tue Sep 4 20:20:02 2007 +0200 16374 16375 Add suggestion/patch by Bernhard Kaindl <bk@suse.de> and fix NVRestore(). 16376 16377 - When a monitor comes out of DPMS sleep the old state is restored. 16378 - NVRestore() was not ported from randr-1.2 branch, meaning that going from X to console resulted in a unacceptable mode. 16379 16380commit 186e3b551da6e37e0860d06e9265c3c58b763e27 16381Author: Maarten Maathuis <madman2003@gmail.com> 16382Date: Tue Sep 4 18:39:44 2007 +0200 16383 16384 Add pci-rework compatibility and remove the pci-id list. 16385 16386 - Detection is now done by probing registers. 16387 - Nforce/Nforce2 have special "hacks", they have been converted, but are untested. 16388 - NV5x detection should work, but has seen no testing. 16389 16390commit 6f860d6d7637f4eba49636d2bab11ac62da279cf 16391Author: Ben Skeggs <skeggsb@gmail.com> 16392Date: Tue Sep 4 11:38:39 2007 +1000 16393 16394 Badness. 16395 16396commit 54ca35e7379fdb375c0d7be2f8c3ad1513be2a59 16397Author: Ben Skeggs <skeggsb@gmail.com> 16398Date: Mon Sep 3 22:06:27 2007 +1000 16399 16400 Hook up G8x accel code. 16401 16402commit 8c6fd43f63e450acbfd8de563580db715e72f90b 16403Author: Ben Skeggs <skeggsb@gmail.com> 16404Date: Mon Sep 3 22:06:09 2007 +1000 16405 16406 Fix G8x hwcursor 16407 16408commit f6d0e2dde750436090f50f442e3e0f1aa226a7f3 16409Author: Ben Skeggs <skeggsb@gmail.com> 16410Date: Mon Sep 3 21:35:06 2007 +1000 16411 16412 Fix up M2MF code for G8x. 16413 16414commit fe8788da8cd5aca90bbe85e1c975b79ff02682ba 16415Author: Ben Skeggs <skeggsb@gmail.com> 16416Date: Mon Sep 3 21:30:16 2007 +1000 16417 16418 Init grobjs we need on G8x. 16419 16420commit f450a4d0ab9c199db0a99a5e78642a0d66ca2915 16421Author: Ben Skeggs <skeggsb@gmail.com> 16422Date: Mon Sep 3 21:14:59 2007 +1000 16423 16424 Allocate some VRAM for G8x palette. 16425 16426commit dfcfc9a8dac7f1fdf7b31ebdd2bca4c5f33313e0 16427Author: Ben Skeggs <skeggsb@gmail.com> 16428Date: Mon Sep 3 21:06:15 2007 +1000 16429 16430 Don't pound PMC_ENABLE on startup, this is DRM's job. 16431 16432 Fixes resetting parts of G8x that shouldn't be. 16433 16434commit c8a32bab34ffe74a4a323de3885fae87bd916218 16435Author: Ben Skeggs <skeggsb@gmail.com> 16436Date: Mon Sep 3 21:05:08 2007 +1000 16437 16438 Detect unknown G8x chips in NVProbe() + Force randr12 on G8x 16439 16440commit 087fd31d6a5d6a835fc318549a12faa916cf5ed0 16441Author: Dave Airlie <airlied@linux.ie> 16442Date: Mon Sep 3 16:46:34 2007 +1000 16443 16444 fix distcheck after randr-1.2 brought in 16445 16446commit 423f95c02c5b018d6fc78f501ea8b546d73f33fb 16447Author: Dave Airlie <airlied@linux.ie> 16448Date: Mon Sep 3 16:42:32 2007 +1000 16449 16450 merge in the nv50 code from the randr 1.2 branch 16451 16452commit 39e8b6c7c0a159f7a06d786abc0f9c5c30ae6793 16453Author: Dave Airlie <airlied@linux.ie> 16454Date: Mon Sep 3 16:22:47 2007 +1000 16455 16456 randr12: bring over cursor code 16457 16458commit ffd87b9bf2712311e197063a7c4b65b7c06a5ebb 16459Author: Dave Airlie <airlied@linux.ie> 16460Date: Mon Sep 3 16:16:43 2007 +1000 16461 16462 randr12: try and bring over most of the randr12 code into master 16463 16464 This currently shouldn't break master, and randr-1.2 more than likely doesn't actually work 16465 16466commit 708cb3f9d9b66030dd0bfc4beaa657fcdfded784 16467Author: Arthur Huillet <arthur.huillet@free.fr> 16468Date: Fri Aug 31 21:42:51 2007 +0200 16469 16470 Xv: now "powercycling" the overlay using PMC_ENABLE in order to workaround some lock up problems 16471 16472commit 3a5095db9fd33d2feeb4bcfcb8453848998ce04b 16473Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16474Date: Fri Aug 31 01:36:11 2007 +0200 16475 16476 Cut remains of NV03 support. 16477 16478commit 54a922dffaa1a4a7c29c2cf8d73449e6fe9058ed 16479Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16480Date: Fri Aug 31 01:35:52 2007 +0200 16481 16482 Good ol' nv04 doesn't seem to support dithering type for SIFM. 16483 16484commit b680a9652508e0d0e660fd924937870cd4ad3530 16485Author: Arthur Huillet <arthur.huillet@free.fr> 16486Date: Thu Aug 30 23:19:13 2007 +0200 16487 16488 Xv: nv NV30 arbitration patch for DFPs - fixes (some?) bw problems 16489 16490commit f29c9d3fc95f160ded64b911266a479ad0a2fc1a 16491Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16492Date: Wed Aug 29 00:56:05 2007 +0200 16493 16494 Silence more warnings. 16495 16496commit 663bf3cbc881631e54e7d63932f59e1b4289979f 16497Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16498Date: Wed Aug 29 00:20:19 2007 +0200 16499 16500 Silence a warning. 16501 16502commit 5937ace8e2376f0b38249e38d1f97a88f4aca99f 16503Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16504Date: Tue Aug 28 11:18:54 2007 +0200 16505 16506 Some small fixes. 16507 16508commit 30d2b3f0dd4a9b6be5d3ad7a38993e5b84baf90a 16509Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16510Date: Thu Aug 23 22:54:53 2007 +0200 16511 16512 Oops I forgot to enable the nv35 object on nv35... 16513 16514commit 761d25eb9afd4be34f00ff7eff44dac0c2f9d9a2 16515Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16516Date: Thu Aug 23 22:39:34 2007 +0200 16517 16518 Don't enable that non-working nv30 EXA just yet. 16519 16520commit 7e7dec68969db5dfc15a96f6dbe78edbcd1ab5ab 16521Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16522Date: Thu Aug 23 22:33:58 2007 +0200 16523 16524 Add nv30 exa code, doesn't work yet. 16525 16526commit e1789ac879d474015c87c36847f0e2777486d6ad 16527Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16528Date: Wed Aug 22 22:38:08 2007 +0200 16529 16530 Update to the latest nouveau_reg.h file. 16531 16532commit 4fc38aeb54da1cba9d5e231a5151e6429af0ea7e 16533Author: Matthieu Castet <castet.matthieu@free.fr> 16534Date: Tue Aug 21 20:03:15 2007 +0200 16535 16536 init SCALED_IMAGE_FROM_MEMORY_COLOR_CONVERSION to COLOR_CONVERSION_TYPE_DITHER. 16537 riva and blob init it. 16538 16539commit c63b5a3740dfe3ff5c6b77412e3666a79137c120 16540Author: Arthur Huillet <arthur.huillet@free.fr> 16541Date: Tue Aug 21 14:55:07 2007 +0200 16542 16543 Fixed blending in NVComposite by actually creating ContextBeta1 and 4 objects 16544 16545commit 4454b3d0104c14890b9b5ead33823ffd0362315c 16546Author: Arthur Huillet <arthur.huillet@free.fr> 16547Date: Mon Aug 20 23:50:43 2007 +0200 16548 16549 Workaround for NVComposite bug #11949 16550 16551 Accepting only PictOpSrc now, PictOpOver causing crashes. 16552 16553commit 450cf465bd850d70ba00711e7d85e1e8eaeff502 16554Author: Arthur Huillet <arthur.huillet@free.fr> 16555Date: Mon Aug 20 13:18:15 2007 +0200 16556 16557 Xv: fixed wrong fix of big endian NV12 copy, as explained by Daniel Chassot 16558 16559 No sleepless coding ever again. <- big lie 16560 16561commit b8780f3d9a7e1bd596f19cbb4867908dd6be53db 16562Author: Arthur Huillet <arthur.huillet@free.fr> 16563Date: Mon Aug 20 00:46:00 2007 +0200 16564 16565 Xv: patch from Daniel Chassot to do 32bit memory writes in NVCopyNV12ColorPlanes, with minor reworking of mine 16566 16567commit 690b4b7adc02e7170451bd5bc6616bc03823d2e5 16568Author: Arthur Huillet <arthur.huillet@free.fr> 16569Date: Sun Aug 19 21:04:00 2007 +0200 16570 16571 Xv: this time a correct matching ? :) 16572 16573commit 6b6505bc54fc29fc22d2be995f34b24ddb9dbaa0 16574Author: Arthur Huillet <arthur.huillet@free.fr> 16575Date: Sun Aug 19 20:18:49 2007 +0200 16576 16577 Xv: stupid build error. blah. 16578 16579commit 28b66c26f85f7730ba9a54db3f778bfd5b24915d 16580Author: Arthur Huillet <arthur.huillet@free.fr> 16581Date: Sun Aug 19 20:14:38 2007 +0200 16582 16583 Xv: now catching all chips < NV17 16584 16585commit 94ae23e2a1ebe9fe3f4bf2039ad3b2e5f0e3e55e 16586Author: Arthur Huillet <arthur.huillet@free.fr> 16587Date: Sun Aug 19 19:50:57 2007 +0200 16588 16589 Xv: disabled YV12 overlay on < NV17 16590 16591commit 1f83dbfa6ef774276e0da3dc4f74cbabe55f76fa 16592Author: Arthur Huillet <arthur.huillet@free.fr> 16593Date: Fri Aug 17 00:45:47 2007 +0200 16594 16595 Xv: now only uploading the part of source image that is used to workaround bandwidth problems 16596 16597commit ffa62dc9f573448c8a832324973513f7ba985b33 16598Author: Arthur Huillet <arthur.huillet@free.fr> 16599Date: Wed Aug 15 02:27:15 2007 +0200 16600 16601 Cleaner DMA and CPU copy upload of color plane for overlay. Fixes a few bugs and is necessary for the bandwidth problem workaround. 16602 16603commit 72c62c48faf2acffae1bc311051537f46972ae58 16604Author: Patrice Mandin <pmandin@caramail.com> 16605Date: Tue Aug 14 22:26:02 2007 +0200 16606 16607 Can not setup dfp modes with xorg-server-1.1 also 16608 16609commit 54eadd624282ae6d0493a3cb54afda7600c7e12b 16610Author: Patrice Mandin <pmandin@caramail.com> 16611Date: Tue Aug 14 22:12:37 2007 +0200 16612 16613 Make NV30 EXA disablable 16614 16615commit da6522dbfcff10081ce2ad866ba4afdd18b413d7 16616Author: Arthur Huillet <arthur.huillet@free.fr> 16617Date: Tue Aug 14 15:01:58 2007 +0200 16618 16619 Implemented CPU copy fallback for native YV12 overlay as well. 16620 16621commit e417307bfc8610f010728d82bb80fced7d0d7aa7 16622Author: Arthur Huillet <arthur.huillet@free.fr> 16623Date: Tue Aug 14 14:47:44 2007 +0200 16624 16625 Native YV12 overlay support, cleaner Xv code 16626 16627 Added support for a native YV12 overlay for NV10+ cards. 16628 Cleaned up Xv code. 16629 CPU copy fallback does not do native YV12 yet. 16630 Bandwidth problem with YV12. 16631 16632commit b3e4da61e8c846d8d3f01b015cb7155da4e266e9 16633Author: Carlos Martín Nieto <carlos@cmartin.tk> 16634Date: Sat Aug 11 12:27:56 2007 +0200 16635 16636 Don't dereference the uninitialised pNv->FB pointer. 16637 16638 Opps, mea culpa. I accidentally commited a debug statement that 16639 dereferences an uninitialised pointer. 16640 16641commit a7d29fd932fe14ca5b82cda64341704164fb7e09 16642Author: Carlos Martín Nieto <carlos@cmartin.tk> 16643Date: Fri Aug 10 20:01:27 2007 +0200 16644 16645 Move DRI's FB to the higher half of VRAM 16646 16647 If we don't, DRM's push buffer takes the map away from us. 16648 16649 This is just a workaround until that useless DRI mapping goes away. 16650 16651commit f9e7849e8736d1477e867a3a1262f0f82baefbcb 16652Author: Arthur Huillet <arthur.huillet@free.fr> 16653Date: Fri Aug 10 20:59:12 2007 +0200 16654 16655 of course, we don't need Xv debug messages upstream 16656 16657commit ee098308c6aca7615bc6ee597e9ca531520f4119 16658Author: Arthur Huillet <arthur.huillet@free.fr> 16659Date: Fri Aug 10 19:28:22 2007 +0200 16660 16661 "smart" overlay with composite, and a tentative fix of flickering when dropping frames 16662 16663 added "smart overlay" when composite is enabled. not extensively tested 16664 now not changing the buffer when skipping frames - not sure it even has an impact. 16665 16666commit 9cc11cbc9ce10b99f72d7c08584bf61d179cdbda 16667Author: Ben Skeggs <skeggsb@gmail.com> 16668Date: Thu Aug 9 01:46:46 2007 +1000 16669 16670 Fix blit adaptor + NVWaitVBlank() (ie. replace old magic 0x400710 hack) 16671 16672commit 22e965038bbc64c7b5118fc9a03ecf5f467d1665 16673Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16674Date: Wed Aug 8 12:45:59 2007 +0200 16675 16676 And of course, we don't really need these ErrorFs. Not any more than last time. 16677 16678commit 1c5a3e271ca628c61fce0bcef24d0bda15d06d7e 16679Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16680Date: Wed Aug 8 12:38:14 2007 +0200 16681 16682 Oh, and of course we need a new object. 16683 16684commit a9b44880924ee814f0bd4b0015ab152f2d0b9051 16685Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16686Date: Wed Aug 8 12:37:31 2007 +0200 16687 16688 Use IMAGE_FROM_CPU for smaller UTS. This is async and helps a lot with glyphs especially. 16689 16690commit 9cb4c95a4fbf38fcb8249e765ff71b2e24912244 16691Author: Ben Skeggs <skeggsb@gmail.com> 16692Date: Wed Aug 8 11:48:48 2007 +1000 16693 16694 Add forgotten prototype 16695 16696commit 089ff874e0a798c1b2693b5ab01cfa04c939f758 16697Author: Ben Skeggs <skeggsb@gmail.com> 16698Date: Wed Aug 8 10:32:08 2007 +1000 16699 16700 Timeouts got lost somewhere along the way.. 16701 16702commit 15a680a6d715f87910b07133e1b98e2d4be9a108 16703Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16704Date: Tue Aug 7 23:42:09 2007 +0200 16705 16706 Fix some overlooked hardcoded subchannels... 16707 16708commit 3f0a52c44697787d84d4f2cf61872e13d630f62a 16709Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16710Date: Mon Aug 6 23:30:04 2007 +0200 16711 16712 Fix explicit usage of subchannels, and remove subchannel defines altogether. 16713 16714commit 17b65f1c2be3e609d387066d72dd8f5f7e2a6d92 16715Author: Matthieu Castet <castet.matthieu@free.fr> 16716Date: Mon Aug 6 19:55:52 2007 +0200 16717 16718 make XAA work on nv17 16719 16720commit bf5684ba05e50fe48cce1e634caf03725fa907e6 16721Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16722Date: Mon Aug 6 18:27:19 2007 +0200 16723 16724 oops remove the debugging ErrorFs. 16725 16726commit 2eab8453d73b4e7fd6c2d772d595ca5e8fa8c975 16727Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16728Date: Mon Aug 6 18:13:13 2007 +0200 16729 16730 Fix UTS/DFS when gart is smaller than transfer size. 16731 16732commit 71076f7947ec71b7a0f363c23f5a51ef09cdd9c0 16733Author: Ben Skeggs <skeggsb@gmail.com> 16734Date: Tue Aug 7 02:09:30 2007 +1000 16735 16736 NVDmaSetObjectFromSubchannel doesn't exist anymore! 16737 16738commit 4a6c2bd2bedcf9f67c3e498b1411ad58cd66199c 16739Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 16740Date: Mon Aug 6 17:29:41 2007 +0200 16741 16742 Add subchannel handling code 16743 16744commit c7b16f6fff96ffcff8049feed2a0a9ef6de4209f 16745Author: Ben Skeggs <skeggsb@gmail.com> 16746Date: Mon Aug 6 21:45:50 2007 +1000 16747 16748 Adapt to drm 0.0.10 16749 16750commit 59d073c9b01a8f61675b8d74f5c55f134ddfb8bf 16751Author: Arthur Huillet <arthur.huillet@free.fr> 16752Date: Mon Aug 6 01:37:17 2007 +0200 16753 16754 Xv: waiting for last transfers to finish before freeing notifiers 16755 16756commit 047aa7e0a6ecce59c9be8d36c51f082c0ddaafe8 16757Author: Ben Skeggs <skeggsb@gmail.com> 16758Date: Mon Aug 6 00:13:35 2007 +1000 16759 16760 Clamp NV_MEMORY_TO_MEMORY_FORMAT_LINE_COUNT to 2047 lines. 16761 16762 At least NV40 can't handle values larger than this. Patch fixes hang 16763 when visiting opengl.org. 16764 16765commit 68070ff7309c8b1cf149490dcf537892b0bf44bc 16766Author: Arthur Huillet <arthur.huillet@free.fr> 16767Date: Mon Jul 30 03:27:55 2007 +0200 16768 16769 Xv now manages a pool of 6 notifiers for its double buffering. Also improved the memory allocation logic. 16770 16771commit bceef7210e44dc643a5f7a5b5683d5c434a07565 16772Author: Arthur Huillet <arthur.huillet@free.fr> 16773Date: Sun Jul 29 01:54:42 2007 +0200 16774 16775 Xv: fixed bugs when displayed only part of the source image (tvtime with overscan) 16776 Xv: fixed bug with overlay behaving badly when window is partly out of screen 16777 Xv: clean up NVPutImage code 16778 16779commit a434cc6dd38ffa658bacf5547ec0384ec5351e9c 16780Author: Dave Airlie <airlied@ppcg5.localdomain> 16781Date: Sat Jul 28 16:34:49 2007 +1000 16782 16783 exa: disable compositing on big endian machines 16784 16785 This at least allows me to see what I'm typing for now :-) 16786 16787commit 175c2b7f7b9e45924e1de7f7ea1130fda76d02fd 16788Author: Arthur Huillet <arthur.huillet@free.fr> 16789Date: Fri Jul 27 19:50:48 2007 +0200 16790 16791 no overlay with composite 16792 16793commit 200c82634971b32bcc308b5e07c113048cf37aa4 16794Author: Arthur Huillet <arthur.huillet@free.fr> 16795Date: Fri Jul 27 16:28:48 2007 +0200 16796 16797 Xv: gotos are evil 16798 16799commit 8544c0ce64780be1bb5f66550d9d6f1e3c87865c 16800Author: Arthur Huillet <arthur.huillet@free.fr> 16801Date: Fri Jul 27 15:11:31 2007 +0200 16802 16803 Removed comments in preparation for upstream push 16804 16805commit 2c60f6e9fc5260eba5b7ddaf67085781f35bfc58 16806Author: Arthur Huillet <arthur.huillet@free.fr> 16807Date: Thu Jul 26 23:00:02 2007 +0200 16808 16809 re-added CPU copy code as a second fallback option (first being to use X GARTScratch) 16810 16811commit c4dfad770d380dacddd3be025e96f867047327cc 16812Author: Arthur Huillet <arthur.huillet@free.fr> 16813Date: Thu Jul 26 16:08:40 2007 +0200 16814 16815 Completely removed notifier wait, when using double buffered stuff. 16816 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. 16817 16818commit 8ba6a9c7494ebd53a3fe775d4d9b402e918a6598 16819Author: Arthur Huillet <arthur.huillet@free.fr> 16820Date: Thu Jul 26 15:21:08 2007 +0200 16821 16822 Preliminary implementation of "host-side double buffering" to improve performance of Xv 16823 16824commit 4f4721f1b8a76a0bf9938e3c614ba7166d5f3dab 16825Author: Arthur Huillet <arthur.huillet@free.fr> 16826Date: Wed Jul 25 22:23:37 2007 +0200 16827 16828 removed pointless sleep in notifier wait 16829 16830commit 290e71edfbc6297c63989d5c21a67717b4ca5f76 16831Author: Arthur Huillet <arthur.huillet@free.fr> 16832Date: Fri Jul 20 17:46:18 2007 +0200 16833 16834 cleaner Xv code - fallback on CPU copy yet to be implemented 16835 16836commit e760235a0596895b726220366c50d4e7c244006e 16837Author: Arthur Huillet <arthur.huillet@free.fr> 16838Date: Fri Jul 20 01:33:49 2007 +0200 16839 16840 Now using GART as the source for the blitter - preliminary implementation, gotta clean it up 16841 16842commit a97181f95a1f5da6180b8a72f3ca307079b92079 16843Author: Arthur Huillet <arthur.huillet@free.fr> 16844Date: Fri Jul 20 01:22:11 2007 +0200 16845 16846 now yielding in the wait for notifier for Xv only 16847 16848commit 81cc744eb31343a509f9ca0d37993d462f162d80 16849Author: Arthur Huillet <arthur.huillet@free.fr> 16850Date: Sun Jul 15 01:00:19 2007 +0200 16851 16852 VERY EXPERIMENTAL now EXA waits for sync before its operations - this removed the text corruption and didn't affect performance here, please report. 16853 16854commit 74ed14e1d27903f0647fe650da666cacb1e2e940 16855Author: Arthur Huillet <arthur.huillet@free.fr> 16856Date: Sun Jul 15 00:35:46 2007 +0200 16857 16858 VERY EXPERIMENTAL moved the notifier wait at the beginning of NvPutImage 16859 16860commit 590e385cb5e4f26532ec5ee380f2aaae1cd69894 16861Author: Arthur Huillet <arthur.huillet@free.fr> 16862Date: Thu Jul 12 03:19:21 2007 +0200 16863 16864 new structure for PutImage, and DMA transfers 16865 16866commit d4abe1fb011d414fda1871f052dda5b860f0820b 16867Author: Ben Skeggs <skeggsb@gmail.com> 16868Date: Tue Jul 17 13:09:34 2007 +1000 16869 16870 Emit NOTIFY+NOP sequence after M2MF. 16871 16872commit 36396aeaeabfd00c7014fa52c17261c3da8b6a82 16873Author: Ben Skeggs <skeggsb@gmail.com> 16874Date: Fri Jul 13 15:28:56 2007 +1000 16875 16876 Play "nuke the typedef" 16877 16878commit 99e0fbc1a34f535abf41c6f4566901c326dfa4b5 16879Author: Ben Skeggs <skeggsb@gmail.com> 16880Date: Thu Jul 12 10:47:53 2007 +1000 16881 16882 Match drm 0.0.9 interface changes. 16883 16884commit e88b20b392c6972088d88f52b2da83b82b821be5 16885Author: Arthur Huillet <arthur.huillet@free.fr> 16886Date: Wed Jul 11 02:42:24 2007 +0200 16887 16888 added support for PCI GART when no AGP is available - requires DRM interface version 8 16889 16890commit dff1cce299df636813e8f85109f4556f520620c0 16891Author: Ben Skeggs <skeggsb@gmail.com> 16892Date: Sun Jul 1 18:22:37 2007 +1000 16893 16894 Always provide UTS, fallback to memcpy if no GART. 16895 16896commit 2e0e7e3856573fbbcb5fe5c96cadcde4d33db68f 16897Author: Ben Skeggs <skeggsb@gmail.com> 16898Date: Sun Jul 1 18:19:44 2007 +1000 16899 16900 Always provide DFS, use memcpy fallback if no GART. 16901 16902commit ba0d3fbba7c1a83ddc509414772199862a4a1d03 16903Author: Ben Skeggs <skeggsb@gmail.com> 16904Date: Sat Jun 30 15:27:07 2007 +1000 16905 16906 Remove some unused cruft 16907 16908commit 3e2677040937fda1cd6e6ea26c238049bd9f99ba 16909Author: Ben Skeggs <skeggsb@gmail.com> 16910Date: Sat Jun 30 04:26:48 2007 +1000 16911 16912 No need to call exaMarkSync() in UTS/DFS. 16913 16914commit e81c9a29e6d1c6573892df8601f0361c2b09c659 16915Author: Ben Skeggs <skeggsb@gmail.com> 16916Date: Thu Jun 28 15:16:35 2007 +1000 16917 16918 Bail if GlxSetVisualConfigs isn't found. 16919 16920commit fce3edda42d10a9c3249c6eae4218a51a981493f 16921Author: Ben Skeggs <skeggsb@gmail.com> 16922Date: Mon Jun 25 12:45:31 2007 +1000 16923 16924 Bump required xorg-server version to 1.3 16925 16926commit 8f1c9af02651404f8b4800b150a86bb47062fd68 16927Author: Ben Skeggs <skeggsb@gmail.com> 16928Date: Mon Jun 25 12:44:12 2007 +1000 16929 16930 Check PRAMIN for BIOS image if not found in PROM. 16931 16932commit 824331e3fce36098e8dec31746204f7c13038abb 16933Author: Ben Skeggs <skeggsb@gmail.com> 16934Date: Sun Jun 24 17:48:33 2007 +1000 16935 16936 Implement NV40 EXA composite hooks with 3D engine. 16937 16938 Named NV30EXA, but won't work on NV30 just yet. This should change at some 16939 point in the future. 16940 16941 Will likely break non-voodoo'd NV4X cards (if the latest drm changes didn't 16942 already break them). Use EXANoComposite xorg.conf option to disable the hook 16943 if that's the case. 16944 16945commit 2e2e95fc3b650ca633294bf4c11c9ba28f45b743 16946Author: Ben Skeggs <skeggsb@gmail.com> 16947Date: Sun Jun 24 17:34:10 2007 +1000 16948 16949 Slightly saner UploadToScreen implementation 16950 16951commit 35806975047aa8c0dbc059709bb823f475ee7368 16952Author: Ben Skeggs <skeggsb@gmail.com> 16953Date: Sun Jun 24 17:27:42 2007 +1000 16954 16955 Wait on notifier instead of PGRAPH_STATUS 16956 16957 When multiple channels are in use, PGRAPH_STATUS may report busy even if the 16958 ddx's channel is idle. 16959 16960commit c39089b3d686eecef11db2b5a2d606ba1ca2dc0c 16961Author: Ben Skeggs <skeggsb@gmail.com> 16962Date: Sun Jun 24 17:21:27 2007 +1000 16963 16964 Match drm 0.0.7 interface changes 16965 16966commit 45dccb99a541a0e1e7be7b4e5aab12c3d8d9f30d 16967Author: Ben Skeggs <skeggsb@gmail.com> 16968Date: Sun Jun 24 16:47:36 2007 +1000 16969 16970 Hide the allocator's use of physical addresses 16971 16972commit d41cf4febd555cca1c65df7aa981be64a6ae4219 16973Author: Ben Skeggs <skeggsb@gmail.com> 16974Date: Sun Jun 24 16:46:40 2007 +1000 16975 16976 Pass pScrn instead of pNv in a few places 16977 16978commit d59fbb4725602c409beee4ab96dca1eab5d73b2a 16979Author: Ben Skeggs <skeggsb@gmail.com> 16980Date: Mon Jun 11 20:36:28 2007 +1000 16981 16982 don't nuke user-specified CFLAGS 16983 16984commit 7af7d95eb8826fa3fb9914a8c6ad4b6f4867095e 16985Author: Henry Zhao <henryz@localhost.localdomain> 16986Date: Mon Apr 23 01:38:09 2007 +0200 16987 16988 Import from nv : Need to inject a mode corresponding panel width/height for validation 16989 16990commit 896fe62233f62d5e1f874ed0eba3a200fa5ef14b 16991Author: Richard Hughes <richard@hughsie.com> 16992Date: Sun Apr 15 23:01:52 2007 +0200 16993 16994 Fix make dist. 16995 16996commit e0e90f427581929fd01c96f1958a1292982dee83 16997Author: Ben Skeggs <skeggsb@gmail.com> 16998Date: Mon Mar 26 01:04:15 2007 +1000 16999 17000 nouveau: rip out card init, everything except modesetting. 17001 17002commit 23fdec37c580c3b5de263dfd48bd80fa9f21bc19 17003Author: Ben Skeggs <skeggsb@gmail.com> 17004Date: Wed Mar 21 17:55:38 2007 +1100 17005 17006 nouveau: bump for drm interface changes (v0.0.5) 17007 17008commit 1fddcfe6680148ed2edd8017217734a1f812298c 17009Author: Matthieu Castet <castet.matthieu@free.fr> 17010Date: Sun Mar 18 11:30:54 2007 +0100 17011 17012 Fix "Fix NoAccel mode" commit : NVAccelCommonInit return a bool 17013 17014commit 113c6837a6515d533bfa15daf49f6cd2c4c95ac4 17015Author: Matthieu Castet <castet.matthieu@free.fr> 17016Date: Sun Mar 18 11:28:04 2007 +0100 17017 17018 Fix NoAccel mode 17019 17020commit d1b75251a3f225890f1e25b11038389d71b67611 17021Author: Ben Skeggs <skeggsb@gmail.com> 17022Date: Wed Mar 7 21:41:22 2007 +1100 17023 17024 Rip out nv30_exa.c, we have other means to play with the 3D engine now. 17025 17026commit adb2df9a9c88bc9646b65a235aa36416c3064c9f 17027Author: Ben Skeggs <skeggsb@gmail.com> 17028Date: Wed Mar 7 16:46:27 2007 +1100 17029 17030 Xv: Make the blit adaptor play nice with EXA 17031 17032 Previously the blit adaptor assumed that all drawing (ie. XAA) would occur to 17033 the screen pixmap, so it didn't bother touching surface offsets. Now, with 17034 EXA this isn't true and the card complained about exceeding the colour buffer 17035 limits when attempting to play video. 17036 17037 This patch kills 2 birds with one stone; instead of pointing the GPU back at 17038 the screen pixmap when playing video, we point it at the destination drawable. 17039 So, Xv will also hopefully play nicely with redirected windows now. 17040 17041commit f28abbedfc7deeb90317dd24c7ec5dfe1ae94fdf 17042Author: Ben Skeggs <skeggsb@gmail.com> 17043Date: Wed Mar 7 13:58:48 2007 +1100 17044 17045 Only use Y32 for Solid(). 17046 17047 It's the only place the workaround is needed at this point, and in some places 17048 we need to use the "proper" ARGB format (Xv for instance). 17049 17050commit 6441253d34d118befe8950781ba012fc58bb5158 17051Author: Ben Skeggs <skeggsb@gmail.com> 17052Date: Wed Mar 7 04:33:44 2007 +1100 17053 17054 Fail configure if xorg-server isn't at least version 1.1 17055 17056 The minimum requirement has been 1.1 for a while now, but would only fail 17057 at compile time when compiling nv_exa.c. This patch makes configure fail, 17058 and removes the remaining conditional support for older servers. 17059 17060 Hopefully the configure.ac is ok, I suck at autofoo. 17061 17062commit e568c2d8e82b1bcf25e495f4a82331803d954234 17063Author: Ben Skeggs <skeggsb@gmail.com> 17064Date: Wed Mar 7 04:22:24 2007 +1100 17065 17066 reindent and move nv_video.c around a bit 17067 17068commit a0a5586836eb87a8c21abb06f9b8afd64eec714a 17069Author: Ben Skeggs <skeggsb@gmail.com> 17070Date: Tue Mar 6 05:42:29 2007 +1100 17071 17072 Use Y32 format instead of A8R8G8B8 for 32-bpp pixmaps 17073 17074 fixes 32-bpp EXA solid-fills, previously alpha was forced to 0xff. 17075 17076commit 2e9976c034b6c310b7ad0fc858250c5e54781a9e 17077Author: Ben Skeggs <skeggsb@gmail.com> 17078Date: Mon Mar 5 14:45:04 2007 +1100 17079 17080 move NVAccel* funcs into nv_accel_common.c 17081 17082commit add6463b848fab274cb5cc190a78aeafc1389203 17083Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17084Date: Sun Mar 4 22:40:48 2007 +0100 17085 17086 Joint patch from Ben Skeggs and me that fixes some (all ?) composite issues. 17087 17088commit 83868bd0a36520da49e40e2b8404e4652de39824 17089Author: Ben Skeggs <darktama@iinet.net.au> 17090Date: Thu Mar 1 08:39:23 2007 +1100 17091 17092 nouveau: avoid attempting an AGP-only alloc 17093 17094 will fail on non-AGP cards, and leave nasty messages in Xorg.log... 17095 17096commit da1dd87acb7061b4772c271dc3c9071a3b160e8a 17097Author: Ben Skeggs <skeggsb@gmail.com> 17098Date: Wed Feb 28 15:30:52 2007 +1100 17099 17100 match drm v4 interface changes 17101 17102 context (gr/dma) object handling: 17103 - remove use of flags/dmaobj when creating objects 17104 - move object creation/setup into nv_accel_common.c 17105 - NV04: make PGRAPH complain a bit more if we program the hardware 17106 incorrectly. 17107 - NV04/NV10: enable PGRAPH_DEBUG_3_CTX_METHODS 17108 This bit seems to enable the use of SET_* methods. 17109 - NV04/NV10/NV40: enable PGRAPH_DEBUG_3_IGNORE_PATCHVALID 17110 Objects no longer get PATCH_CONFIG_VALID set in their context. 17111 Not sure how to get this set other than to intercept the 17112 PATCH_EXCEPTION error and enable the flag. I opted for the 17113 same method NVIDIA use, which makes PGRAPH ignore it. 17114 misc: 17115 - use NV04_SCALED_IMAGE_FROM_MEMORY on all NV_ARCH_04 cards 17116 - fix offScreenBase calculation (EXA) 17117 - move mem alloc/free helpers into nv_mem.c 17118 17119commit 7e5f90b264fdb3081250ff16748164998b6461a6 17120Author: Adam Jackson <ajax@benzedrine.nwnk.net> 17121Date: Tue Feb 27 14:46:20 2007 -0500 17122 17123 Typo fix. 17124 17125commit c85da5d8b346e17e68da68e57d62c3f77029b680 17126Author: Damien Thebault <damien.thebault@laposte.net> 17127Date: Wed Feb 14 00:35:50 2007 +0100 17128 17129 Don't output the chipset list to the log file, silence some warnings. 17130 17131commit f650cb126e042b44126df6a298e996726356ec5b 17132Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17133Date: Wed Feb 14 00:29:52 2007 +0100 17134 17135 Fix the workaround (or workaround the fix). 17136 17137commit ac2784d38634bc6d8aec3a94d609b60cd49b21b6 17138Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17139Date: Wed Jan 24 20:39:40 2007 +0100 17140 17141 Make NV_DMA_DEBUG print the fifo position. 17142 17143commit d54d441df4294bf4fed26635a738f4111087e06d 17144Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17145Date: Wed Jan 24 19:13:05 2007 +0100 17146 17147 Increase the fifo dump size. 17148 17149commit b21dbb170aa49c83ef450c2ab8c34614277e7fbc 17150Author: B. Rathmann <koala_br@users.sourceforge.net> 17151Date: Wed Jan 24 13:46:16 2007 +0100 17152 17153 Fix the G70 issues. 17154 17155commit e683979666a4c07590b7bade2f62cca33c19a3b2 17156Author: Ben Skeggs <darktama@iinet.net.au> 17157Date: Fri Jan 19 15:40:22 2007 +1100 17158 17159 bump drm patchlevel 17160 17161commit feff7cb67ae61987146a21dad535702f3db2264c 17162Author: Jeremy Kolb <jkolb@brandeis.edu> 17163Date: Thu Jan 18 21:28:51 2007 -0500 17164 17165 Fill in some more nv30 magic values. Not sure what they do yet but they don't hurt anything. 17166 17167 Apply names to other registers. 17168 17169commit 31462c30bb79e8245edee0cf57c46b889d2f259b 17170Author: Jeremy Kolb <jkolb@brandeis.edu> 17171Date: Wed Jan 17 08:34:56 2007 -0500 17172 17173 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. 17174 17175commit 4583312ba72047e6bc65bfc576a0bcf90e04b2de 17176Author: David Jeannot <djeannot24@gmail.com> 17177Date: Tue Jan 16 00:03:56 2007 +0100 17178 17179 Indent nv_bios.c properly. 17180 17181commit 6246fba5a54dc7af9a8b16c5a14131180831562d 17182Author: Patrice Mandin <patrice@manoir.racoon.city> 17183Date: Fri Jan 12 20:25:16 2007 +0100 17184 17185 Add missing include for malloc/free 17186 17187commit 37937980ec7b21c40539daee029886c4f4a0f0cd 17188Author: Jeremy Kolb <jkolb@brandeis.edu> 17189Date: Fri Jan 12 00:26:20 2007 -0500 17190 17191 NV30: Set DEBUG3 reg bit 1. Turns on automatic hw context switching. 17192 17193commit d5617e91ff13f9e19d9e5d421fd081f59dec685c 17194Author: jkolb@brandeis.edu <jkolb@brandeis.edu> 17195Date: Thu Nov 2 20:23:36 2006 -0500 17196 17197 Just a superficial cleaning up. Nothing exciting. 17198 17199commit 1f6239de8d7618de6aee402eaf1d8552524a996e 17200Author: Ben Skeggs <darktama@iinet.net.au> 17201Date: Tue Jan 2 15:08:36 2007 +1100 17202 17203 NV4x changes to work with latest drm. 17204 17205commit f9ba29a07581ede8c409b0eb8b549db3b446f490 17206Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17207Date: Wed Dec 27 21:52:15 2006 +0100 17208 17209 Output the driver date to the logs on startup. 17210 17211commit 2f1d14fe1c03c32ad60776e0d0c2ae41a42ea4be 17212Author: Ben Skeggs <darktama@iinet.net.au> 17213Date: Tue Dec 12 00:13:05 2006 +1100 17214 17215 Remove final remaining PRAMIN access (NV04-specific), the drm handles it now 17216 17217commit de7255d5420c64a2707f5e78a4157d93618e6c17 17218Author: Ben Skeggs <darktama@iinet.net.au> 17219Date: Fri Dec 8 18:09:26 2006 +1100 17220 17221 Provide some visuals with a stencil buffer, and actually call NVDRIInitVisualConfigs. 17222 17223commit e64dcc59b607ae5ad7049c35fed6d6a300cd5f6e 17224Merge: 0f7957d 656dfcd 17225Author: Matthieu Castet <mat@mat-pc.(none)> 17226Date: Wed Dec 6 22:09:23 2006 +0100 17227 17228 Merge branch 'master' of git+ssh://matc@git.freedesktop.org/git/nouveau/xf86-video-nouveau/ 17229 17230commit 0f7957da9ce856ef0accc922ab368ac10171b5e6 17231Author: Matthieu Castet <mat@mat-pc.(none)> 17232Date: Wed Dec 6 22:07:26 2006 +0100 17233 17234 More pgraph regs 17235 17236commit 656dfcd635551245b2f58835c5ea87474d3eb60d 17237Merge: ed307d5 224578b 17238Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17239Date: Sun Dec 3 10:20:04 2006 +0100 17240 17241 Merge branch 'master' of git+ssh://marcheu@git.freedesktop.org/git/nouveau/xf86-video-nouveau 17242 17243commit ed307d559be72e06f64566574beb36200ebe47e2 17244Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17245Date: Sun Dec 3 10:14:04 2006 +0100 17246 17247 Introduce lock step versioning with the drm. 17248 Fix the MEM_ALLOC issue. 17249 17250commit 224578b74645fe3624669dc3521fe6deb1cc4aae 17251Author: Matthieu Castet <mat@mat-pc.(none)> 17252Date: Tue Nov 28 19:12:55 2006 +0100 17253 17254 Don't alloc fifo if it is not needed. 17255 17256commit 146eb74cb7d1da1da55247a04e7a969d98601e42 17257Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17258Date: Mon Nov 27 00:03:11 2006 +0100 17259 17260 Mandatory oops commit. 17261 17262commit bdfa6776128f4bc7e2f3caa57296571acb9bd661 17263Author: Andrew Mitchell <ajmitch@ubuntu.com> 17264Date: Sun Nov 26 23:59:06 2006 +0100 17265 17266 Remove XF86DRI defines from the driver. It's already mandatory anyway. 17267 17268commit 7ebc8c4327d0b50f60c4cd4423414d4cfbb8b7d8 17269Author: airlied <airlied@optimus.localdomain> 17270Date: Fri Nov 17 16:23:38 2006 +1100 17271 17272 add VGA CRTC registers from haiku 17273 17274commit 7150a4357dff0a2e4468190e2fd228ed07d367f9 17275Author: Dave Airlie <airlied@linux.ie> 17276Date: Fri Nov 17 12:55:37 2006 +1100 17277 17278 remove nvvga.h unused 17279 17280commit 64a1c5bb90fd13ce1c191873a4f4afc2a9ede09b 17281Author: Dave Airlie <airlied@linux.ie> 17282Date: Fri Nov 17 12:51:11 2006 +1100 17283 17284 find and version DRM earlier in startup sequence so it is loaded firstup 17285 17286commit 659b5de8200d9022e7616f03b20a306fff2af0fe 17287Author: Matthieu Castet <castet.matthieu@free.fr> 17288Date: Fri Nov 17 00:59:58 2006 +0100 17289 17290 Cleanup of PVIDEO. 17291 17292commit 370b83eb3347d14003f2878737d25de7dacaeec1 17293Author: Matthieu Castet <castet.matthieu@free.fr> 17294Date: Wed Nov 15 22:10:27 2006 +0100 17295 17296 Use objects from nouveau_reg.h. 17297 17298commit 2ffeb45a953a2ff52abe6f78dacfb9c53f873dbb 17299Author: Ben Skeggs <darktama@iinet.net.au> 17300Date: Wed Nov 15 01:22:18 2006 +1100 17301 17302 Set SURFACE_FORMAT in NVExaPrepareCopy. Fixes one of the many issues with xorg-server-1.2 17303 17304commit 85bda095107558846fe6fb803b6df412213b27f4 17305Author: Ben Skeggs <darktama@iinet.net.au> 17306Date: Wed Nov 15 00:47:04 2006 +1100 17307 17308 nv_exa.c tab goodness 17309 17310commit 8b733a25c30fb9e499d678c26b777d072d1ac187 17311Author: Ben Skeggs <darktama@iinet.net.au> 17312Date: Tue Nov 14 08:31:21 2006 +1100 17313 17314 Fix type. 0x84 is DEBUG_1 17315 17316commit 030eb38dd1fe767807aab0447ed4fab4ec8c368a 17317Author: Ben Skeggs <darktama@iinet.net.au> 17318Date: Tue Nov 14 03:09:05 2006 +1100 17319 17320 Fill in front_offset/front_pitch 17321 17322commit 589bb37ff37bb7d4775f5dca0301c5d97ea73b06 17323Author: Ben Skeggs <darktama@iinet.net.au> 17324Date: Sun Nov 12 16:04:32 2006 +1100 17325 17326 Add NVDRIFinishScreenInit. 17327 17328commit d0c09a1bd0e1c26cc591a3d8738b94b1efe576ec 17329Author: Dave Airlie <airlied@linux.ie> 17330Date: Sun Nov 12 14:08:45 2006 +1100 17331 17332 add some names to registers 17333 17334commit 84928fd946b4f02d8a943226481304417475c4f9 17335Author: Ben Skeggs <darktama@iinet.net.au> 17336Date: Sun Nov 12 07:43:24 2006 +1100 17337 17338 Import nouveau_dri.h from the 3D driver as nv_dri.h, and setup DRIInfo->devPrivate. 17339 17340commit cdb9e3acf0aa5cdf6e5aa33fb499ea8e93f7f1f2 17341Author: Ben Skeggs <darktama@iinet.net.au> 17342Date: Sun Nov 12 07:33:53 2006 +1100 17343 17344 Temporary workaround for DRI drmMap problem, not sure of what we actually need to do here yet. 17345 17346commit d3babbfbed898921e1a167df08eef92d381d6aa0 17347Author: Ben Skeggs <darktama@iinet.net.au> 17348Date: Sun Nov 12 07:28:31 2006 +1100 17349 17350 nv_dri.c still had some space-indented functions, not anymore :) 17351 17352commit bad4e6c674ac4fdc0aebbbc6d36fc2854ee8f885 17353Merge: 1b14821 c096565 17354Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17355Date: Mon Nov 6 03:14:07 2006 +0100 17356 17357 Merge branch 'master' of git+ssh://marcheu@git.freedesktop.org/git/nouveau/xf86-video-nouveau 17358 17359commit 1b14821ecd67a62a34a902e62c0b229bb9ed97ca 17360Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17361Date: Mon Nov 6 03:13:50 2006 +0100 17362 17363 Added a DRI visual configs functions. It should work but I can't test it... 17364 17365commit c096565a0f6e6c12255e57eaefb878904ce04971 17366Author: Ben Skeggs <darktama@iinet.net.au> 17367Date: Mon Nov 6 11:54:19 2006 +1100 17368 17369 Fix TEX0_IN_TEX1 fragprog.. I must've been asleep when I did that.. 17370 17371commit 47f32cfe6e2f513f86b7a8f02a8872c134e2327d 17372Author: Dave Airlie <airlied@ppcg5.localdomain> 17373Date: Mon Nov 6 11:45:53 2006 +1100 17374 17375 use pagesize to work out the sarea size 17376 17377commit e6c4d40e247939f677e285db15a78ccc33b92b41 17378Author: Dave Airlie <airlied@ppcg5.localdomain> 17379Date: Mon Nov 6 08:39:49 2006 +1100 17380 17381 don't fail dri module load if loaded already 17382 17383commit 4cc4a1f7ef559f1ce56706f16ee24a3b6879d28d 17384Merge: b95f8aa b4e4307 17385Author: Dave Airlie <airlied@linux.ie> 17386Date: Sun Nov 5 18:00:29 2006 +1100 17387 17388 Merge ssh://10.1.1.200/home/airlied/git/xf86-video-nouveau 17389 17390commit b4e4307af982b50aff191983d209cb855a377f80 17391Author: Dave Airlie <airlied@ppcg5.localdomain> 17392Date: Sun Nov 5 17:59:17 2006 +1100 17393 17394 fix G5 endianness issue with panel size regs 17395 17396commit b95f8aac9e6dcbf4d10a5e26bde6a0f3595608cb 17397Merge: ba71154 e6b1f68 17398Author: Dave Airlie <airlied@linux.ie> 17399Date: Sun Nov 5 17:53:53 2006 +1100 17400 17401 Merge branch 'master' of ssh://git.freedesktop.org/git/nouveau/xf86-video-nouveau 17402 17403commit ba711545a0fd091c05b514f30c1e904370026499 17404Author: Dave Airlie <airlied@ppcg5.localdomain> 17405Date: Sun Nov 5 17:51:10 2006 +1100 17406 17407 set the PPC use BE MMIO 17408 17409commit e6b1f687e3bf127d8dba1871d67eff9d83ac284e 17410Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17411Date: Sat Nov 4 16:57:30 2006 +0100 17412 17413 Remove the context object flags handling from the DDX since it's now in the DRM. 17414 17415commit ead77e0ec1d0c62829fa973db3d62937b44781e5 17416Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17417Date: Sat Nov 4 04:47:57 2006 +0100 17418 17419 Add the NV44A 3D object class. 17420 17421commit 87463666ab23e047f355cfcd7016aa9c7cd8ad76 17422Author: Ben Skeggs <darktama@iinet.net.au> 17423Date: Fri Nov 3 04:00:09 2006 +1100 17424 17425 Use "do { blah; } while(0)" instead of just braces. (suggestion from megari on irc) 17426 17427commit 68682238b36292409ce7f41ed90bd3be860e77f3 17428Author: Ben Skeggs <darktama@iinet.net.au> 17429Date: Thu Nov 2 16:07:26 2006 +1100 17430 17431 Install NV30_TCL EXA hooks if NV_ENABLE_3D is defined. 17432 17433 Ok.. Firstly I don't recommend using this code, it is not ready for general 17434 usage. We probably only want to keep the Composite hook, but the others are 17435 useful to test things with. 17436 17437 1. You will most likely (read: definitely) need to start X with the binary 17438 driver before starting X with nouveau/NV30_TCL.. If you don't PGRAPH will 17439 yell at you and X will crash. 17440 2. Copy() is bad when the src and dest are the same, so scrolling and moving 17441 windows around will fall back to software rendering.. 17442 3. Composite() is slightly better than the one in nv_exa.c, but it's buggy in 17443 a few cases and not complete yet. Particularly, transforms are missing. 17444 4. There are some correctness problems that I haven't bothered to fix yet, so 17445 you'll probably see bits of windows left behind when moving them around.. 17446 5. Don't be surprised if it's horribly broken :) 17447 17448commit c371d90935d6baf2f1dd0073741637d14b506f59 17449Author: Ben Skeggs <darktama@iinet.net.au> 17450Date: Thu Nov 2 15:57:47 2006 +1100 17451 17452 Put braces around NVDmaSetObjectOnSubchannel... 17453 17454commit b45aea22a8da69ad4a2493f7f33fa65186a83ba3 17455Author: Ben Skeggs <darktama@iinet.net.au> 17456Date: Thu Nov 2 14:52:24 2006 +1100 17457 17458 Merge >=NV30-specific EXA code.. hard-disabled pending testing on my C51, and a proper commit message.. 17459 17460commit e79b0befe92110c1cd2e0e1ae69fced1cd5acba3 17461Author: Ben Skeggs <darktama@iinet.net.au> 17462Date: Thu Nov 2 01:29:34 2006 +1100 17463 17464 Reformat NVExaInit 17465 17466commit 6a77e5bf2e58b19df67c029ad7bffd92f8cf652f 17467Author: Ben Skeggs <darktama@iinet.net.au> 17468Date: Thu Nov 2 01:18:14 2006 +1100 17469 17470 First part of preparation for NV30_TCL EXA code merge. 17471 17472commit 1d6e911ab5391277b403ef6f7b095203cfd19a53 17473Author: Ben Skeggs <darktama@iinet.net.au> 17474Date: Thu Nov 2 00:38:53 2006 +1100 17475 17476 Reindent nv_dma.c with tabs.. 17477 17478commit 06748f74f639242fd690a3dda7e685ae4fa1f95e 17479Author: Ben Skeggs <darktama@iinet.net.au> 17480Date: Wed Oct 11 11:06:16 2006 +1100 17481 17482 s/init/alloc/ 17483 17484commit a2d55603db8c01cc4b9f3404c282b1e4963a152c 17485Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17486Date: Wed Oct 11 00:39:36 2006 +0200 17487 17488 Remove the PFIFO_REINIT hack, and enable the irq by default. This commit requires the matching drm commit, and will probably break stuff. 17489 17490commit 78537b3342bbf1c16dc78f8f06cb3f989ce8f03f 17491Author: Ben Skeggs <darktama@iinet.net.au> 17492Date: Thu Sep 14 12:56:28 2006 +1000 17493 17494 It usually helps to wait on a notifier, and not on a NVAllocRec.. Oops. 17495 17496commit 01dde28611238cf2d81b087b371cc67296d1b581 17497Author: Ben Skeggs <darktama@iinet.net.au> 17498Date: Thu Sep 14 12:17:32 2006 +1000 17499 17500 Allocate notifiers with NVAllocateMemory. 17501 Fix some warnings. 17502 Return success/fail for DMA and Context object creation. 17503 17504commit 3d5c560f509646853cc863e35ed2412db39cd8a3 17505Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17506Date: Tue Sep 5 07:39:10 2006 +0200 17507 17508 Port some of the least intrusive nv03 code to the nouveau driver. 17509 17510commit 6fb7a63bf5d4f046583f9d14862c16ec5979a318 17511Author: Ben Skeggs <darktama@iinet.net.au> 17512Date: Sun Sep 3 06:41:52 2006 +1000 17513 17514 Config options for cmdbuf location/size. 17515 17516 Will default to putting the cmdbuf in AGP (except for PowerPC) if an AGP card 17517 is detected. Won't currently be detected because something is wrong DRM-size. 17518 17519 If you have an AGP card you can manually specify 'FBLocation "AGP"' in 17520 xorg.conf to make it work. 17521 17522 Needs an updated DRM to have any effect. 17523 17524commit e49d561d2328eeac7691661e28955a882bba2192 17525Author: Ben Skeggs <darktama@iinet.net.au> 17526Date: Sun Sep 3 01:51:23 2006 +1000 17527 17528 Copy+Paste a workaround for a NV40 hardware bug from NVShowHideCursor() into 17529 LoadStateExt(). This should fix the problem of the HW-cursor still being 17530 enabled when VT-switching back to a framebuffer console. 17531 17532commit 69ace0c4388cb1a66ea74f627fe02fe89e7d35c0 17533Author: Ben Skeggs <darktama@iinet.net.au> 17534Date: Sat Sep 2 22:17:55 2006 +1000 17535 17536 Kill off one of the M2MF objects, and use the engine's SET_OBJECT_{IN_OUT} to 17537 swap between AGP->VRAM and VRAM->AGP. 17538 17539commit 72cf79b9e5dc0d97e933f540d063161129bbebf7 17540Author: Ben Skeggs <darktama@iinet.net.au> 17541Date: Sat Sep 2 21:59:26 2006 +1000 17542 17543 Don't create objects that aren't needed. This frees up a couple of subchannels 17544 for EXA, use one of them for MEMORY_TO_MEMORY_FORMAT. 17545 17546commit 6a06d9022f3c0bc845f02091d83faf95f69f81bd 17547Author: Ben Skeggs <darktama@iinet.net.au> 17548Date: Sat Sep 2 18:15:15 2006 +1000 17549 17550 Move code out of nv_xaa.c that doesn't belong there. 17551 17552commit 46ac77d524dd183c6b50a7a0e77990514ff28840 17553Author: Ben Skeggs <darktama@iinet.net.au> 17554Date: Sat Sep 2 17:57:38 2006 +1000 17555 17556 Add back some setup that went missing during PFB reg cleanup. 17557 Should fix corruption on NV40. 17558 17559commit 03b86f08667a10c44420d60787fe9e17b5d20ab8 17560Author: Ben Skeggs <darktama@iinet.net.au> 17561Date: Thu Aug 31 15:53:25 2006 +1000 17562 17563 Rename module from nv_drv.so to nouveau_drv.so to allow both nv and nouveau 17564 to coexist. Also disabled building of riva128.so, as we plan on merging it 17565 with the rest of the nv code. 17566 17567commit 17c871a44987af04d1688798e2e7802a5f5ba522 17568Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17569Date: Wed Aug 30 06:24:23 2006 +1000 17570 17571 clean up FB register names 17572 17573commit 7d3c12424fd79daf831b7c28780586e8bd9c3574 17574Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17575Date: Wed Aug 30 04:23:43 2006 +1000 17576 17577 add some names for the RAMDAC registers 17578 17579commit 0e70296ce9f7a921c4b971919d12c262202d8d48 17580Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17581Date: Wed Aug 30 02:18:02 2006 +1000 17582 17583 add some names for some of the CRTC registers - some names not very meaningful 17584 yet 17585 17586commit 9a5e2c5be96eaced3aadd153fb2f9b1928173af7 17587Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17588Date: Wed Aug 30 01:48:07 2006 +1000 17589 17590 remove select head registers PCIO and PDIO 17591 17592commit dec0691db79b1ded2942c1cbfa97434fa8693f9b 17593Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17594Date: Wed Aug 30 01:13:05 2006 +1000 17595 17596 remove unused pointers after macroisation 17597 17598commit 9444b692607895572c8199a60d6968fbc374d51e 17599Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17600Date: Wed Aug 30 01:02:45 2006 +1000 17601 17602 missed 2x PGRAPHs in XAA 17603 17604commit f73cb8ef4f4fbb588bfe9ca767286153b2565a37 17605Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17606Date: Wed Aug 30 00:59:39 2006 +1000 17607 17608 remove PCRTC via macros 17609 17610commit 6d5b1765e10856a7dfa75c27ad96f547dc26b23a 17611Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17612Date: Wed Aug 30 00:48:23 2006 +1000 17613 17614 remove PEXTDEV and PTIMER using macros 17615 17616commit 6140bdfe6959cd0f6ce8e06d34b6e838d8cef046 17617Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17618Date: Wed Aug 30 00:43:07 2006 +1000 17619 17620 remove PMC using macros 17621 17622commit 28d9befa6308462a6bc1809b894ebe389873fd6a 17623Author: Ben Skeggs <darktama@iinet.net.au> 17624Date: Wed Aug 30 01:26:58 2006 +1000 17625 17626 Make EXA the default 17627 17628commit fa9209f77df3d835ebad43dc82988c000398f650 17629Merge: ab2b550 2f7caf6 17630Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17631Date: Tue Aug 29 13:42:49 2006 +0200 17632 17633 Merge branch 'master' of ssh://marcheu@git.freedesktop.org/git/nouveau/xf86-video-nouveau 17634 17635commit ab2b5509e82a646ac187b020c91234e33f81065d 17636Author: Stephane Marchesin <marchesin@icps.u-strasbg.fr> 17637Date: Tue Aug 29 13:42:34 2006 +0200 17638 17639 Make use of NV5_SCALED_IMAGE_FROM_MEMORY on NV05. 17640 17641commit 9911eb726f99166968165710a9e2f540388cd42a 17642Author: Stephane Marchesin marchesin@icps.u-strasbg.fr <steph@tubular-brain.localdomain> 17643Date: Tue Aug 29 03:18:27 2006 +0200 17644 17645 Enable the composite hooks. 17646 17647 After changing NV_SCALED_IMAGE_FROM_MEMORY for NV10_SCALED_IMAGE_FROM_MEMORY it seems the composite hook works. 17648 Also, add a more complete list of desktop pci ids. 17649 17650commit 2f7caf6eb5eb6a3991720fc413bd20723473212f 17651Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17652Date: Tue Aug 29 08:13:59 2006 +1000 17653 17654 remove PGRAPH direct accesses 17655 17656commit 51f0b67d9064fffc6b357d2923188a030e85eace 17657Author: Dave Airlie <airlied@starflyer-ubuntu.(none)> 17658Date: Tue Aug 29 07:51:49 2006 +1000 17659 17660 remove PFB usage with macros 17661 17662commit b50736d1583b960ab70f3e113294db8c2da69e31 17663Author: Ben Skeggs <darktama@iinet.net.au> 17664Date: Mon Aug 28 17:07:06 2006 +1000 17665 17666 Use NVAllocateMemory for AGP scratch buffer. 17667 17668commit a5ddd8fcf4278c35d71605794d0d5711e7c0ebb7 17669Author: Dave Airlie <airlied@linux.ie> 17670Date: Mon Aug 28 16:41:20 2006 +1000 17671 17672 remove legacy crap from nvreg.h add cursor reg defines 17673 17674commit f035bc63c34bce49e1c2619f14ab98b4b1922beb 17675Author: Dave Airlie <airlied@linux.ie> 17676Date: Mon Aug 28 16:23:32 2006 +1000 17677 17678 missed some vga routines 17679 17680commit 0b45a9f725442d5bd784ee67b7473929993f6082 17681Author: Dave Airlie <airlied@linux.ie> 17682Date: Mon Aug 28 16:19:56 2006 +1000 17683 17684 add RAMDAC register read/write 17685 17686commit 24267d7e39383930c589bc80e2bbd4e5cf66ec2a 17687Merge: 349496c c0154e6 17688Author: Ben Skeggs <darktama@iinet.net.au> 17689Date: Mon Aug 28 14:46:51 2006 +1000 17690 17691 Merge branch 'master' of git+ssh://git.freedesktop.org/git/nouveau/xf86-video-nouveau 17692 17693commit 349496c97fafed04806c01bd9d4386b0fcaa9fd2 17694Author: Ben Skeggs <darktama@iinet.net.au> 17695Date: Mon Aug 28 14:45:42 2006 +1000 17696 17697 More cleanups from DRM memory alloc changes. 17698 17699commit c0154e615c091c7bb483fcfd07e216b21618d0f2 17700Author: Dave Airlie <airlied@linux.ie> 17701Date: Mon Aug 28 14:07:05 2006 +1000 17702 17703 use nvReadVGA and nvWriteVGA in more places 17704 17705commit e0f1479773c1a00ec69edbece98e5e2b66c0c2c9 17706Author: Ben Skeggs <darktama@iinet.net.au> 17707Date: Sun Aug 27 22:45:36 2006 +1000 17708 17709 Fix ShadowFB for FB alloc changes, should fix RandR rotation also. 17710 17711commit 0de4d0b8b79e2238f2c4b374b5012a25ae985502 17712Author: Ben Skeggs <darktama@iinet.net.au> 17713Date: Sun Aug 27 22:27:17 2006 +1000 17714 17715 Slash UseFBDev and DGA support. 17716 17717commit e49d08ff482755e9f95c4c3b3cf065e6f0d2bc23 17718Author: Dave Airlie <airlied@linux.ie> 17719Date: Sun Aug 27 10:04:50 2006 +1000 17720 17721 add some read/write NV VGA register wrappers 17722 17723commit c64c55b3133221462980b934a308daf2cc0e473d 17724Author: Dave Airlie <airlied@linux.ie> 17725Date: Sun Aug 27 09:56:36 2006 +1000 17726 17727 remove all U0* types in favour of more standard CARD* types 17728 17729commit 9c294b962379a955d2e8e2cb2c54f845fd4c5740 17730Author: Ben Skeggs <darktama@iinet.net.au> 17731Date: Sat Aug 26 00:43:21 2006 +1000 17732 17733 Wow I suck tonight.. another oops. 17734 17735commit b8411ae92b6833c9336f82ed86dc8755ca8b9dcc 17736Author: Ben Skeggs <darktama@iinet.net.au> 17737Date: Sat Aug 26 00:39:36 2006 +1000 17738 17739 Whoa, how did that happen... 17740 17741commit d458cdac152a49eec9acdfe5081f6d426e712bf2 17742Author: Ben Skeggs <darktama@iinet.net.au> 17743Date: Sat Aug 26 00:33:47 2006 +1000 17744 17745 Initial work in porting the DDX to use the DRM for memory allocations. 17746 This is enough to make it work, but there's still a *lot* of cleanups 17747 to do. 17748 17749commit 83ae039b8f8ea08ae9436201014c7e1ba25a4839 17750Author: Dave Airlie <airlied@linux.ie> 17751Date: Fri Aug 25 19:33:08 2006 +1000 17752 17753 fixup compile for git 17754 17755commit ea2bc33d32bf9abeb23b3fe8cb6c9d28ddccef01 17756Author: Dave Airlie <airlied@linux.ie> 17757Date: Fri Aug 25 19:17:51 2006 +1000 17758 17759 updated git tree from changes that got lost on way from CVS 17760 17761commit 07b81112ba4678085a148e5ffaadd4684f74d64d 17762Author: marcheu <marcheu> 17763Date: Thu Aug 24 21:00:42 2006 +0000 17764 17765 Make NVInitAGP fail if there is no AGP area. Maybe that function shouldn't be called for PCI(-e) cards in the first place. 17766 17767commit 109b5abed6c0c701a1fb84ec63ff8f5914eee8e3 17768Author: marcheu <marcheu> 17769Date: Wed Aug 23 21:32:47 2006 +0000 17770 17771 If you stare at nv_hw.c long enough, things start to make sense. I swear ! 17772 17773commit 80ba1cfec290dbb82c4a1042255a21b626512e84 17774Author: marcheu <marcheu> 17775Date: Wed Aug 23 17:11:22 2006 +0000 17776 17777 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. 17778 17779 Also, add code to dump fifo in case of crashes. 17780 17781commit 9b19daa55c7770a3e66b2374da0212c75ed21260 17782Author: marcheu <marcheu> 17783Date: Tue Aug 22 01:16:18 2006 +0000 17784 17785 Remove debug message. 17786 17787commit b44b54e4c369c612a692eabb8662b3db97758a83 17788Author: marcheu <marcheu> 17789Date: Mon Aug 21 13:19:06 2006 +0000 17790 17791 Fill more fields in the DRIInfo struct. 17792 17793commit aaf13d0e7cc5c76ee81b4234dc2c55bbedac1faa 17794Author: darktama_ <darktama_> 17795Date: Sun Aug 20 15:11:06 2006 +0000 17796 17797 Oops, this shouldn't be here still... 17798 17799commit 00fd86dea8d2eaa561cc89a35046d15bcf92bb82 17800Author: darktama_ <darktama_> 17801Date: Sun Aug 20 14:40:43 2006 +0000 17802 17803 - Allocate notifiers with the DRM 17804 17805 - Rename agpMemory and friends to agpScratch and friends 17806 17807commit d5c68d3f83f84ca67ee9df269b778bbed50851a0 17808Author: darktama_ <darktama_> 17809Date: Sun Aug 20 12:51:49 2006 +0000 17810 17811 Alloc AGP through the DRM 17812 17813commit 6d0924c272954f44b2a9b6b11be8c209dd67de8e 17814Author: marcheu <marcheu> 17815Date: Sat Aug 19 16:04:56 2006 +0000 17816 17817 Begginings of PINS parsing code I had laying around... 17818 17819commit 59e2d852d6f10ba26a394ff7ea5d3e7b30f2be71 17820Author: darktama_ <darktama_> 17821Date: Mon Jul 31 07:32:57 2006 +0000 17822 17823 Need.To.Pay.More.Attention... 17824 17825commit 9a314361fd004afe7129ae2d41f58ddd7fe6f56a 17826Author: darktama_ <darktama_> 17827Date: Mon Jul 31 03:24:17 2006 +0000 17828 17829 Updates for DRM changes, and a few cleanups 17830 17831commit 11fbb14876e7873a863de16b01fb5778e841b4bd 17832Author: erikwaling <erikwaling> 17833Date: Thu Jul 13 04:13:07 2006 +0000 17834 17835 Added more opcodes and some TODOs. 17836 17837commit dd54fea7562b8c9a6d26b645a03c9644c08090cc 17838Author: erikwaling <erikwaling> 17839Date: Thu Jul 13 03:09:11 2006 +0000 17840 17841 Changed INIT_MACRO to haiku-like implementation. 17842 17843commit b7e68083f3bf6b0943b58cfed0b82de61eeaac33 17844Author: marcheu <marcheu> 17845Date: Mon Jul 10 00:36:31 2006 +0000 17846 17847 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. 17848 17849commit fe2a1360d72385ad0f61f381a296820d02a60e79 17850Author: marcheu <marcheu> 17851Date: Mon Jul 10 00:30:54 2006 +0000 17852 17853 Use -Wall by default 17854 17855commit ca38f1d9d61393915515e1042204b35aeac863ba 17856Author: darktama_ <darktama_> 17857Date: Sat Jul 8 13:53:00 2006 +0000 17858 17859 Remove FIFO and object handling out of the DDX. This commit makes the DDX dependant on the DRM to operate at all.. 17860 17861commit b9b880b83896defe06ce61c9aef2f5ae4eb4f37e 17862Author: lumag <lumag> 17863Date: Mon Jun 12 20:21:11 2006 +0000 17864 17865 Fix exa acceleration by filling versioning fields. 17866 17867commit 0bbc0bcfe33c2625ea6175c3b052d231cc3b7c8f 17868Author: darktama_ <darktama_> 17869Date: Sat Jun 10 18:34:14 2006 +0000 17870 17871 Quick hack to make sure CRTC 0x44 contains it's original value when VT switching back to the console (also on server shutdown). 17872 17873 Should fix the problem of one display having scrolling issues, and losing 17874 it's text-mode cursor. 17875 17876commit 18e65f4e0604c744ea905319348c688141e14bbf 17877Author: marcheu <marcheu> 17878Date: Thu Jun 8 01:10:01 2006 +0000 17879 17880 Remove xf86ReadPciBIOS, use the bios from the reg aperture instead (that will probably work on more archs anyway) 17881 17882commit 2f5fcd131dbea5cb9873cc8a4aad44ae1bcf1340 17883Author: darktama_ <darktama_> 17884Date: Fri Jun 2 21:54:48 2006 +0000 17885 17886 Remove some XFree86LOADER conditionals 17887 17888commit 7889cc0241450aaa8e9dafc65f9f7976df3cd99b 17889Author: darktama_ <darktama_> 17890Date: Sat Apr 22 02:26:53 2006 +0000 17891 17892 Need to include nv_include.h before checking that XF86DRI is definied. XF86DRI lives in xorg-server.h - included from config.h 17893 17894commit 3083263eb40f57e4c529473c5e9f935109a1719f 17895Author: marcheu <marcheu> 17896Date: Fri Apr 21 22:15:59 2006 +0000 17897 17898 Don't include config.h in all files but in the *_include.h files. That's what they're here for after all. 17899 17900commit 864e2c6936a37005112a3c6f3f7db5a16100b96c 17901Author: darktama_ <darktama_> 17902Date: Fri Apr 21 02:46:31 2006 +0000 17903 17904 Include config.h in all source files 17905 17906commit 74ff8fc182c6f728a1b5e8a0415c25b8b7a4af94 17907Author: marcheu <marcheu> 17908Date: Thu Apr 20 18:54:12 2006 +0000 17909 17910 Added a fixme about laptop chips 17911 17912commit ad38154a7e075207d8f462ffb90489efe9cf01f1 17913Author: erikwaling <erikwaling> 17914Date: Thu Apr 20 18:50:11 2006 +0000 17915 17916 Ported nv_exa.c to work with EXA >=2.0.0 17917 17918commit 7c25d338e0c79a288fa192d9a2d4ac6eb59996c1 17919Author: Dave Airlie <airlied@linux.ie> 17920Date: Fri Aug 25 18:51:03 2006 +1000 17921 17922 initial import of nouveau code 17923 17924commit 98372d0f94a9d658a7257d2199ef37a2449b9961 17925Author: Adam Jackson <ajax@benzedrine.nwnk.net> 17926Date: Fri Aug 11 16:02:28 2006 -0400 17927 17928 Replace panel message + MODE_BAD with MODE_PANEL. 17929 17930commit 43ec66396f805aba87017d193d97a03372c29909 17931Author: Aaron Plattner <aplattner@nvidia.com> 17932Date: Fri Jun 30 12:54:44 2006 -0700 17933 17934 Bump version to 1.2.0 for release. 17935 17936commit 6265b77dc8878e36abd76b73e209a2d76d70ba22 17937Author: Mark Vojkovich <markv@xfree86.org> 17938Date: Thu Jun 29 18:41:44 2006 -0700 17939 17940 New chip support and new PCI IDs. 17941 17942commit 5114639dff73c22845451acffed94f1da284a7df 17943Author: Aaron Plattner <aplattner@nvidia.com> 17944Date: Mon Jun 19 09:32:11 2006 -0700 17945 17946 Generate the ChangeLog from git history for dist. 17947 17948commit 04d344dfcbb9928cf33943a880de333557e14be9 17949Author: Aaron Plattner <aplattner@nvidia.com> 17950Date: Fri Jun 16 18:03:21 2006 -0700 17951 17952 Delete the ChangeLog. The git history is just as good. 17953 17954commit eda3bac20dce2d5584ac07cc6b28a08ec55aa798 17955Author: Aaron Plattner <aplattner@nvidia.com> 17956Date: Wed May 24 15:05:22 2006 -0700 17957 17958 Derive the module version numbers from the package version. This idea was 17959 lifted from Michel Dänzer's similar code in xf86-video-ati. 17960 17961 Also, make the riva and nv driver versions match. 17962 17963commit 1a40c4a79b79034d191496726a27af8b0805dba6 17964Author: Aaron Plattner <aplattner@nvidia.com> 17965Date: Thu Apr 27 18:10:37 2006 -0700 17966 17967 Add the libcwrapper back, conditionalized on the server version. 17968 17969commit 425e30d58b7b62fff01fbb58585ee69b2df6fec1 17970Author: Aaron Plattner <aplattner@nvidia.com> 17971Date: Thu Apr 27 10:32:10 2006 -0700 17972 17973 Determine whether PutImage hooks take a DrawablePtr by server version. Remove 17974 dependency on xorg-server >= 1.0.99.901. 17975 17976commit c34a93265cb42168376ebdf0746b3ccded9f126c 17977Author: Aaron Plattner <aplattner@nvidia.com> 17978Date: Tue Apr 25 14:38:43 2006 -0700 17979 17980 Bump to 1.1.2 for new product names. 17981 17982commit f70247ade9eac8fa6115889662abb8e678a182dc 17983Author: Aaron Plattner <aplattner@nvidia.com> 17984Date: Tue Apr 25 14:29:44 2006 -0700 17985 17986 Add names for Quadro FX 5500, 3500, 1500, 560, 550, and 350. 17987 17988commit ca4c8af624e87e5c6a2927092a9c230ae0e60c0e 17989Author: Aaron Plattner <aplattner@nvidia.com> 17990Date: Wed Apr 12 02:04:52 2006 -0700 17991 17992 Add .gitignores. 17993 17994commit fb3faeb6073cef5a6a1b094f3d3bf020ceba9db6 17995Author: Adam Jackson <ajax@nwnk.net> 17996Date: Sat Apr 8 01:38:33 2006 +0000 17997 17998 Bump to 1.1.1 for Xv changes. 17999 18000commit f793182898c4d844e9861a60a945f53ab7071adc 18001Author: Aaron Plattner <aplattner@nvidia.com> 18002Date: Fri Apr 7 23:11:06 2006 +0000 18003 18004 Add a DrawablePtr argument to the XV functions to pave the way for 18005 redirected video. 18006 18007commit cb6b4073df7404822e2b553519c3a101f2ab0524 18008Author: Adam Jackson <ajax@nwnk.net> 18009Date: Fri Apr 7 19:40:34 2006 +0000 18010 18011 Unlibcwrap. Bump server version requirement. Bump to 1.1.0. 18012 18013commit 3f140052f891a4dc674b82e438c9595e59b4676b 18014Author: Aaron Plattner <aplattner@nvidia.com> 18015Date: Mon Mar 27 21:40:24 2006 +0000 18016 18017 Bump version to 1.0.2.0 for release. 18018 18019commit 84c007b426d321786f9aec1d9fda14b5baa35561 18020Author: Aaron Plattner <aplattner@nvidia.com> 18021Date: Fri Mar 10 02:29:03 2006 +0000 18022 18023 Coverity nos. 868 and 869: Free the result of xf86GetEntityInfo. 18024 18025commit 61e237eec7193701ce1236e67ebdfe7465b5db8a 18026Author: Aaron Plattner <aplattner@nvidia.com> 18027Date: Tue Jan 24 16:45:29 2006 +0000 18028 18029 Fix typo in PCI ID (reported by Stefan Dirsch). Patch from Mark Vojkovich 18030 to XFree86 CVS on 2006-01-23. 18031 18032commit 5064f8a903923ca4b819bdf345f8f4c9254bdf3b 18033Author: Aaron Plattner <aplattner@nvidia.com> 18034Date: Mon Jan 23 00:04:41 2006 +0000 18035 18036 New chip support from Mark Vojkovich's 2006-01-20 and 2006-01-21 XFree86 18037 CVS updates. 18038 18039commit e8fd45625f2e09830136bde6bc70d51f25892450 18040Author: Aaron Plattner <aplattner@nvidia.com> 18041Date: Mon Jan 2 01:54:19 2006 +0000 18042 18043 Preliminary RandR rotation support. Bug #4708 18044 <https://bugs.freedesktop.org/show_bug.cgi?id=4708> 18045 18046commit 5ec53e759c5233a26a6c80901b9f31b51153f756 18047Author: Kevin E Martin <kem@redhat.com> 18048Date: Wed Dec 21 02:30:01 2005 +0000 18049 18050 Update package version for X11R7 release. 18051 18052commit b0c8ca8c03c8ad1648b4597f7825bac7531100a7 18053Author: Adam Jackson <ajax@nwnk.net> 18054Date: Mon Dec 19 16:25:54 2005 +0000 18055 18056 Stub COPYING files 18057 18058commit 4fd25919c2a217d355419efdfbf0fe148af6c146 18059Author: Kevin E Martin <kem@redhat.com> 18060Date: Thu Dec 15 00:24:18 2005 +0000 18061 18062 Update package version number for final X11R7 release candidate. 18063 18064commit e4194bab2a790af10581ea2842717623db8998a2 18065Author: Kevin E Martin <kem@redhat.com> 18066Date: Tue Dec 6 22:48:36 2005 +0000 18067 18068 Change *man_SOURCES ==> *man_PRE to fix autotools warnings. 18069 18070commit baf60566b505fdf4387ec25789584b3f6ad72ce8 18071Author: Kevin E Martin <kem@redhat.com> 18072Date: Sat Dec 3 05:49:34 2005 +0000 18073 18074 Update package version number for X11R7 RC3 release. 18075 18076commit 4df200ea93b0a072f29057eeaf7a09e5a3a1dfe2 18077Author: Kevin E Martin <kem@redhat.com> 18078Date: Fri Dec 2 02:16:09 2005 +0000 18079 18080 Remove extraneous AC_MSG_RESULT. 18081 18082commit d4d5d7325fab1ea81f71d1ffc1fbbcc2d2253a65 18083Author: Adam Jackson <ajax@nwnk.net> 18084Date: Tue Nov 29 23:30:00 2005 +0000 18085 18086 Only build dlloader modules by default. 18087 18088commit d76f3f776bdab920d050542da45ec2871d4b84e6 18089Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18090Date: Mon Nov 28 22:04:10 2005 +0000 18091 18092 Change *mandir targets to use new *_MAN_DIR variables set by xorg-macros.m4 18093 update to fix bug #5167 (Linux prefers *.1x man pages in man1 subdir) 18094 18095commit d4e306b585f737b4c1287c520470425b0c11dc32 18096Author: Kevin E Martin <kem@redhat.com> 18097Date: Wed Nov 9 21:15:19 2005 +0000 18098 18099 Update package version number for X11R7 RC2 release. 18100 18101commit dde6dd1c7f95f9dccabfd919ad4ced795fe3fbd1 18102Author: Kevin E Martin <kem@redhat.com> 18103Date: Tue Nov 1 15:08:53 2005 +0000 18104 18105 Update pkgcheck depedencies to work with separate build roots. 18106 18107commit 92b1bfcc8fd0f34e636a1e9955d5236334f78cd3 18108Author: Kevin E Martin <kem@redhat.com> 18109Date: Wed Oct 19 02:48:03 2005 +0000 18110 18111 Update package version number for RC1 release. 18112 18113commit eda1cbc29cb0b2193a5bfb0eb620c740d9aef7c6 18114Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18115Date: Tue Oct 18 00:01:53 2005 +0000 18116 18117 Use @DRIVER_MAN_SUFFIX@ instead of $(DRIVER_MAN_SUFFIX) in macro 18118 substitutions to work better with BSD make 18119 18120commit 6c58a05e405cbc87fbe0497123d9b82d2480bdd3 18121Author: Adam Jackson <ajax@nwnk.net> 18122Date: Mon Oct 17 22:57:34 2005 +0000 18123 18124 More 1.7 braindamage: define EXTRA_DIST in terms of @DRIVER_NAME@ instead 18125 of indirectly 18126 18127commit fc20886c28c465578e6bd3ef2d9fe62bd993382b 18128Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18129Date: Mon Oct 17 00:09:04 2005 +0000 18130 18131 Use sed & cpp to substitute variables in driver man pages 18132 18133commit 1e3c84f332b7b9a4b320b370b8b2ecafbbdd8f8a 18134Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 18135Date: Tue Oct 4 21:36:51 2005 +0000 18136 18137 Link nvidida readme file into nvidia driver, Exclude duplicated neomagic 18138 readme 18139 18140commit d8be62ab175a013f78d4944036da284ffad34d96 18141Author: Aaron Plattner <aplattner@nvidia.com> 18142Date: Thu Sep 29 21:47:29 2005 +0000 18143 18144 Add entries for GeForce Go 7800. (Change went into XFree86 on 2005-09-28). 18145 Don't hang if j is zero. This should never happen, but it's better to be 18146 safe than sorry. 18147 18148commit e39eb8b8c1a79d04f90a195351c4eec3dba0d612 18149Author: Aaron Plattner <aplattner@nvidia.com> 18150Date: Wed Sep 28 17:18:02 2005 +0000 18151 18152 Bug #3333 <https://bugs.freedesktop.org/show_bug.cgi?id=3333> Patch #3417 18153 <https://bugs.freedesktop.org/attachment.cgi?id=3417> 18154 Fix a hang during screen-to-screen blits on certain GeForce 6xxx and Quadro 18155 FX 540 graphics cards. 18156 18157commit dc5acc2e477edb85f468f8d46862ca138426cfd9 18158Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18159Date: Fri Sep 23 16:25:28 2005 +0000 18160 18161 Bug #4558 <https://bugs.freedesktop.org/show_bug.cgi?id=4558> Patch #3551 18162 <https://bugs.freedesktop.org/attachment.cgi?id=3381> 18163 22-Sep-2005 nv driver update from Mark Vojkovich: Fix possible cause of 18164 some acceleration instability on some GeForce6xxx parts. 18165 18166commit b0ea43a51d2a6983559eecdb7dbb73c5e585697c 18167Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18168Date: Sat Sep 17 01:37:18 2005 +0000 18169 18170 Bugzilla #4480 <https://bugs.freedesktop.org/show_bug.cgi?id=4480> Patch 18171 #3305 <https://bugs.freedesktop.org/attachment.cgi?id=3305> 18172 13-Sep-2005 nv driver update from Mark Vojkovich: 18173 - Adds PCI ids for GeForce 6800 XT cards 18174 - Adds names for existing PCI ids for some GeForce 6xxx & 7xxx cards 18175 - Fix a potential problem with pixmap cache corruption on GeForce 6xxx and 18176 7xxx parts. (Mark Vojkovich, NVIDIA) 18177 18178commit cdf20033e2cfc9e8634876dd42613f7719340a5a 18179Author: Eric Anholt <anholt@freebsd.org> 18180Date: Fri Sep 16 01:36:45 2005 +0000 18181 18182 Put the riva_* files in the riva128.so subdriver, like monolithic. 18183 18184commit 9b1b3184cd586f6f8bf65ebd6f3c1afb1ae64d64 18185Author: Daniel Stone <daniel@fooishbar.org> 18186Date: Thu Aug 18 09:03:47 2005 +0000 18187 18188 Update autogen.sh to one that does objdir != srcdir. 18189 18190commit 75a0deb6e9d1a5bc698ea67b338c56bc99c1b2c0 18191Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18192Date: Fri Aug 5 03:52:28 2005 +0000 18193 18194 Bug #3987 <https://bugs.freedesktop.org/show_bug.cgi?id=3987> Patch #3264 18195 <https://bugs.freedesktop.org/attachment.cgi?id=3264> Workaround for 18196 512 Meg video cards (can only map 256 Meg of it). (Mark Vojkovich) 18197 18198commit fb4033bd2662c199d7e4f0e3df7c4a0fb9870984 18199Author: Kevin E Martin <kem@redhat.com> 18200Date: Fri Jul 29 21:22:44 2005 +0000 18201 18202 Various changes preparing packages for RC0: 18203 - Verify and update package version numbers as needed 18204 - Implement versioning scheme 18205 - Change bug address to point to bugzilla bug entry form 18206 - Disable loadable i18n in libX11 by default (use --enable-loadable-i18n to 18207 reenable it) 18208 - Fix makedepend to use pkgconfig and pass distcheck 18209 - Update build script to build macros first 18210 - Update modular Xorg version 18211 18212commit 6ea22f19210448605b9a18bf500c33168a77e77c 18213Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 18214Date: Tue Jul 26 18:40:29 2005 +0000 18215 18216 Build system for nv 18217 18218commit 74772a7dc49e34229497441fbbf2631bfb259b39 18219Author: Adam Jackson <ajax@nwnk.net> 18220Date: Mon Jul 11 02:29:58 2005 +0000 18221 18222 Prep for modular builds by adding guarded #include "config.h" everywhere. 18223 18224commit 80ac36ec4fbbd9a4aae1bde01fe6ebc1088fa8cb 18225Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18226Date: Sat Jul 9 16:51:58 2005 +0000 18227 18228 Bug #3740 <https://bugs.freedesktop.org/show_bug.cgi?id=3740> Patch #3058 18229 <https://bugs.freedesktop.org/attachment.cgi?id=3058> 18230 08-Jul-2005 nv driver updates from Mark Vojkovich: 18231 Change some console restore code for NV11. Hopefully, we can more reliably 18232 restore the console for desktop systems using DVI. This may correct a 18233 recent regression on NV11. 18234 Also, new PCI IDs. 18235 Add new nVidia PCI ids to match nv_driver.c changes. 18236 18237commit 6b75bc13462f1cfb735fe1de359ed9e45b5f4415 18238Author: Daniel Stone <daniel@fooishbar.org> 18239Date: Fri Jul 1 22:43:12 2005 +0000 18240 18241 Change all misc.h and os.h references to <X11/foo.h>. 18242 18243commit 270f5269a08da8ed0186e8a2223b673e0433a771 18244Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18245Date: Wed Jun 29 15:56:23 2005 +0000 18246 18247 Bugzilla #3657 <https://bugs.freedesktop.org/show_bug.cgi?id=3657> Patch 18248 #2992 <https://bugs.freedesktop.org/attachment.cgi?id=2992> 18249 Experimental DPMS support for GeForce4 and later laptops. (Mark 18250 Vojkovich) 18251 18252commit bf73c0716e9e4be0540d2ee7340bd5833831a434 18253Author: Adam Jackson <ajax@nwnk.net> 18254Date: Sun Jun 26 04:38:12 2005 +0000 18255 18256 Bug #3628: Fix video driver submodule visibility. 18257 18258commit 175766a44ceb0a5d4226e9088d8e3e6d73815b4f 18259Author: Adam Jackson <ajax@nwnk.net> 18260Date: Sat Jun 25 21:16:55 2005 +0000 18261 18262 Bug #3626: _X_EXPORT tags for video and input drivers. 18263 18264commit 8759399a3744f0eb1e0cdc83ed23aaaafb9be444 18265Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18266Date: Thu Apr 21 22:25:47 2005 +0000 18267 18268 programs/Xserver/hw/xfree86/drivers/nv/nv_driver.c 18269 programs/Xserver/hw/xfree86/drivers/nv/nv_setup.c 18270 //bugs.freedesktop.org/show_bug.cgi?id=3095> Patch #2501 18271 <https://bugs.freedesktop.org/attachment.cgi?id=2501> nv DPMS panel 18272 fixes (Mark Vojkovich) 18273 18274commit 2db214d7cf5336fbea36d83bc9b089f70a494614 18275Author: Daniel Stone <daniel@fooishbar.org> 18276Date: Wed Apr 20 12:25:29 2005 +0000 18277 18278 Fix includes right throughout the Xserver tree: 18279 change "foo.h" to <X11/foo.h> for core headers, e.g. X.h, Xpoll.h; 18280 change "foo.h", "extensions/foo.h" and "X11/foo.h" to 18281 <X11/extensions/foo.h> for extension headers, e.g. Xv.h; 18282 change "foo.[ch]" to <X11/Xtrans/foo.[ch]> for Xtrans files. 18283 18284commit 3fe983d254f1ffe3c285548013be677ed4a67137 18285Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18286Date: Mon Feb 14 03:37:20 2005 +0000 18287 18288 Bugzilla #2533 (https://bugs.freedesktop.org/show_bug.cgi?id=2533) Feb. 3, 18289 2005 "nv" driver update from Mark Vojkovich (Mainly driver updates for 18290 nVidia cards with ((pNv->Chipset & 0xfff0) == 0x0090) ) 18291 18292commit 8c95fdc889b56a76497f5194bafd06edfc962387 18293Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18294Date: Wed Jan 26 01:36:04 2005 +0000 18295 18296 Jan. 2005 nv chipset update from Mark Vojkovich Bug #2380 18297 <https://bugs.freedesktop.org/show_bug.cgi?id=2380> Patch #1752 18298 <https://bugs.freedesktop.org/attachment.cgi?id=1752> 18299 18300commit 67c89797f62ea53b37927c64e0403e7e7c42de56 18301Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18302Date: Fri Dec 10 03:56:42 2004 +0000 18303 18304 Bugzilla #1985 <https://bugs.freedesktop.org/show_bug.cgi?id=1985> sync to 18305 12/08/04 version of Nvidia driver from Mark Vojkovich, including these 18306 fixes since last sync: Fix some DAC/Graphics memory contention issues 18307 on newer NVIDIA chips (specifically, NV40). More PCI IDs. Also, I've 18308 had trouble getting some panels to work automagically so I expose an 18309 FPTweak option to let the user adjust a troublesome register. 18310 18311commit 7c34d051168eba499beb20c40fd95703375f4c7b 18312Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 18313Date: Wed Nov 10 02:53:44 2004 +0000 18314 18315 (Bugzilla #1777) Import Mark Vojkovich's changes to NV driver since last 18316 sync with XFree86, including: Support for new NVIDIA chips. Egbert has 18317 a laptop that needs some DPMS ordering reversed. New chip support and 18318 workaround for Toshiba M30 laptop issue. PCI-Xpress chip support. DPMS 18319 support for DVI. Cursor changes for NV11 so that alpha blended cursors 18320 will work in conjunction with dithering (laptop panels). It previously 18321 did not. HW support for alpha blended stretch blits on NV5 and higher, 18322 though nothing is using them at this point. Add support for VBlank 18323 syncing the Xv Blit adaptor on hardware that supports it (NV11 and 18324 higher). Make console restoration a little more robust. We were making 18325 an assumption that failed on a newer laptop. Workaround for some NV31 18326 DVI artifacts. 18327 Also add nVidia license statement to nv man page to comply with requirement 18328 that it appear in "user documentation" 18329 18330commit fcd6d7b57edb4e396189a0a2683bcf08974f3b59 18331Author: Egbert Eich <eich@suse.de> 18332Date: Tue Oct 12 19:13:44 2004 +0000 18333 18334 Set fbdev mode as the default mode on PPC (Olaf Hering). 18335 Added support for IBM space saver keyboard (Stefan Dirsch). Added support 18336 for Cherry CyMotion Master XPress (Marcus Schaefer). 18337 Change order of SetDisplayDevice(), HWRestore(), UnbindGART() and 18338 RestoreBIOSMemSize() to be exactly opposite to the Save procedure in 18339 EnterVT() (Matthias Hopf, Alan Hourihane). 18340 Fix text mode restauration by removing the assumption that the register 18341 which determines which head is programmed is set. to the active head by 18342 the BIOS (Mark Vojkovich). 18343 When I wrote the resource code 5 years ago I made some assumptions which 18344 turned out to be false: I've assumed that the bus number of the PCI 18345 hostbridge would be the PCI bus the bridge links to. This is not 18346 correct. Fixing this assumption is not easy. However I hope that the 18347 attached patch will make the system work regardless as it 'ignores' 18348 host bridges from which the target bus is not known. This should not 18349 matter at all as we really don't care about host bridges (unless we 18350 have bridge specific code which retrieves information about the 18351 bridge). 18352 Fixed server crash on reset when a structure allocated in PreInit() was 18353 freed on CloseScreen(). 18354 Fixed ring buffer lock ups that happened because the structure that 18355 contained ringbuffer data was not zeroed after allocation. 18356 Fixed numerous warnings due to signed unsigned comparisons. 18357 programs/Xserver/hw/xfree86/drivers/nv/nv_driver.c: 18358 (NVBacklightEnable): Changed the order in which the sequencer registers 18359 and the backlight control registers are written. The sequencer control 18360 register need to be written first otherwise DPMS screen blanking 18361 produces vertical bars on a mobile device. lib/Xau/Imakefile: Build 18362 libXau static library PIC so it can be linked into toolkits that would 18363 like to wrap its functionality. 18364 18365commit 65e13fc61141b36a74c7a0e696d274574641ba20 18366Author: Adam Jackson <ajax@nwnk.net> 18367Date: Fri Sep 17 03:04:52 2004 +0000 18368 18369 Bug #1192: Remove cfb support from drivers where its use is an option. 18370 Delete xf24_32bpp, as s3virge was the last user. Fix up some comments 18371 to refer to fb rather than cfb. 18372 18373commit bbafc6d2fcb21671de3cfb4ff973409405162e7a 18374Author: Adam Jackson <ajax@nwnk.net> 18375Date: Mon Aug 16 09:13:14 2004 +0000 18376 18377 Bug #1087: Make sure all the *Weak symbols are documented in the import 18378 list, so the loader knows not to complain about unresolved symbols when 18379 an optional module isn't loaded. This typically manifests as warnings 18380 about fbdevHW symbols when the user isn't using the framebuffer 18381 console. 18382 18383commit 8b6de7d388b7ae5169b2e38ad1a40981bc406cfc 18384Author: Adam Jackson <ajax@nwnk.net> 18385Date: Sat Jul 31 01:21:19 2004 +0000 18386 18387 Change several LoaderSymbol calls introduced by the bug #400 patch to 18388 *Weak() resolver functions. 18389 18390commit f745157f15befebe01552b103c573ccb94b8f388 18391Author: Adam Jackson <ajax@nwnk.net> 18392Date: Fri Jul 30 20:30:53 2004 +0000 18393 18394 Bug #400 (partial): Driver fixes for the dlloader. When using dlloader, all 18395 framebuffer formats except cfb and the overlay modes should work, and 18396 r128 and radeon need to be loaded from the ati driver (both issues to 18397 be fixed soon). Tested on i740, s3virge, mach64, tdfx, vesa, and vga 18398 drivers. elfloader users shouldn't be affected. 18399 18400commit b9a6776bc04eb19343882bddb371b705677cfc14 18401Author: Eric Anholt <anholt@freebsd.org> 18402Date: Wed Jun 16 09:23:29 2004 +0000 18403 18404 DRI XFree86-4_3_99_12-merge import 18405 18406commit cbf040deb0ab24425468ed60b3f3edd322ec210f 18407Author: Eric Anholt <anholt@freebsd.org> 18408Date: Wed Jun 16 09:23:29 2004 +0000 18409 18410 Initial revision 18411 18412commit 99a7f7ccce4e012bde57021cafcb132c3d599b96 18413Author: Egbert Eich <eich@suse.de> 18414Date: Fri Apr 23 19:42:10 2004 +0000 18415 18416 Merging XORG-CURRENT into trunk 18417 18418commit 708bb2eb0db3808e3edf94a5f22b3e3f5bb6488d 18419Author: Egbert Eich <eich@suse.de> 18420Date: Sun Mar 14 08:33:30 2004 +0000 18421 18422 Importing vendor version xf86-4_4_99_1 on Sun Mar 14 00:26:39 PST 2004 18423 18424commit fe564263d97e370817fc862e5c8ead09bfb70b58 18425Author: Egbert Eich <eich@suse.de> 18426Date: Wed Mar 3 12:12:23 2004 +0000 18427 18428 Importing vendor version xf86-4_4_0 on Wed Mar 3 04:09:24 PST 2004 18429 18430commit 1a0be799ce4c479e39f8af86fcca3d192e83f455 18431Author: Egbert Eich <eich@suse.de> 18432Date: Thu Feb 26 13:35:54 2004 +0000 18433 18434 readding XFree86's cvs IDs 18435 18436commit ee5c9ef51cea88a630ecc652512603c42b39a533 18437Author: Egbert Eich <eich@suse.de> 18438Date: Thu Feb 26 09:23:23 2004 +0000 18439 18440 Importing vendor version xf86-4_3_99_903 on Wed Feb 26 01:21:00 PST 2004 18441 18442commit d176fa338d2d83bcf8d7ec68ff1aec205a258fec 18443Author: Kaleb Keithley <kaleb@freedesktop.org> 18444Date: Mon Feb 23 20:35:04 2004 +0000 18445 18446 Import most of XFree86 4.4RC3. This import excludes files which have the 18447 new license. If we want to, later we can import 4.4RC3 again and pick 18448 up the files that have the new license, but for now the vendor branch 18449 is "pure." 18450 18451commit ea7a90e680a2e9259e471dd6677bcca12aeea79e 18452Author: Egbert Eich <eich@suse.de> 18453Date: Thu Jan 29 08:08:36 2004 +0000 18454 18455 Importing vendor version xf86-012804-2330 on Thu Jan 29 00:06:33 PST 2004 18456 18457commit a6c83bd98c34aacdf7a6b82ea83d6ae775479502 18458Author: Kaleb Keithley <kaleb@freedesktop.org> 18459Date: Fri Dec 19 20:55:14 2003 +0000 18460 18461 XFree86 4.3.99.902 (RC 2) 18462 18463commit 935c2adf58a10e6d3caa75823b4e7a0689555c49 18464Author: Kaleb Keithley <kaleb@freedesktop.org> 18465Date: Tue Nov 25 19:28:38 2003 +0000 18466 18467 XFree86 4.3.99.16 Bring the tree up to date for the Cygwin folks 18468 18469commit d84248580d5b05bb47a2ccdce7badfc884efa6da 18470Author: Kaleb Keithley <kaleb@freedesktop.org> 18471Date: Tue Nov 25 19:28:38 2003 +0000 18472 18473 Initial revision 18474 18475commit dd1a62caaaac6acc35584a2c8c525962d8574c95 18476Author: Kaleb Keithley <kaleb@freedesktop.org> 18477Date: Fri Nov 14 16:48:55 2003 +0000 18478 18479 XFree86 4.3.0.1 18480 18481commit 2f223903fba2bdee1623f3442d7580c809b428cc 18482Author: Kaleb Keithley <kaleb@freedesktop.org> 18483Date: Fri Nov 14 16:48:55 2003 +0000 18484 18485 Initial revision 18486 18487commit 542d57ea12b5461891a13a339e01ca9e1add124d 18488Author: Kaleb Keithley <kaleb@freedesktop.org> 18489Date: Fri Nov 14 15:54:48 2003 +0000 18490 18491 R6.6 is the Xorg base-line 18492