1commit f82ce27f17e1c706f34a0fdc5cceaf6e42db1476 2Author: Peter Hutterer <peter.hutterer@who-t.net> 3Date: Tue Jan 24 12:01:58 2023 +1000 4 5 Bump the version number to 13.4.0 6 7 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 8 9commit 3ec2fad6ed1a291f050cac4d33ea81e06c469c62 10Author: Ville Skyttä <ville.skytta@iki.fi> 11Date: Sun Dec 15 13:56:04 2013 +0200 12 13 Man page syntax fix. 14 15 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 16 17commit c59c5d241a5f09244be0cdea6dbdc938774c9486 18Author: Ville Skyttä <ville.skytta@iki.fi> 19Date: Sun Dec 15 13:56:03 2013 +0200 20 21 Spelling fixes. 22 23 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 24 25commit f7d50aa328d9488527eb2c97fba6a06c2b445d44 26Author: Alan Coopersmith <alan.coopersmith@oracle.com> 27Date: Sun Jul 24 09:09:00 2022 -0700 28 29 Build xz tarballs instead of bzip2 30 31 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 32 33commit 790d641d1da415ffcda1c251961da63a60c06921 34Author: Alan Coopersmith <alan.coopersmith@oracle.com> 35Date: Mon Jul 18 17:16:33 2022 -0700 36 37 gitlab CI: add a basic build test 38 39 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 40 41commit ff5637a14c52e6ec592592c88938104b087e925c 42Author: Martin Krastev <krastevm@vmware.com> 43Date: Tue Jan 25 15:11:08 2022 +0200 44 45 Garbled XvPutImage output for FOURCC_YV12 when using 3D-accel-texture adaptor 46 47 A helper for the PutImage callback in adaptor 'XA G3D Textured Video' was not 48 taking into account the source data pitches for YV12 format, resulting in 49 garbled frames for misaligned frame widths. Issue reported by Doug Brown. 50 This patch is based off the patch proposed by Doug. 51 52 Repro of the original issue: 53 gst-launch-1.0 videotestsrc ! video/x-raw,format=YV12,width=449,height=240 ! xvimagesink 54 55 Reported-by: Doug Brown <doug@schmorgal.com> 56 Signed-off-by: Martin Krastev <krastevm@vmware.com> 57 Reviewed-by: Zack Rusin <zackr@vmware.com> 58 59commit 77b8183b3395333d5d4c73e25c2d011748f15eda 60Author: Rudi Heitbaum <rudi@heitbaum.com> 61Date: Sun May 8 03:00:10 2022 +0000 62 63 vmwgfx: fix missing array notation 64 65 Fixes error identified by gcc-12.1.0 compiler 66 67 make 68 CC libvmwgfx_la-vmwgfx_tex_video.lo 69 vmwgfx_tex_video.c: In function 'stop_video': 70 vmwgfx_tex_video.c:240:20: error: the comparison will always evaluate as 'true' for the address of 'yuv' will never be NULL [-Werror=address] 71 240 | if (priv->yuv[i]) { 72 | ^~~~ 73 74commit 50ab008cb965dc9e47e0a84a1950e2485bbd3c44 75Author: Martin Krastev <krastevm@vmware.com> 76Date: Tue Nov 30 11:47:39 2021 -0800 77 78 vmwgfx: Change header inclusion order to avoid xorg headers catching stdbool.h 79 80 libdrm commit e641e2a632d779f638ac2ba983b9fceb20b3fac4 added 81 stdbool.h to the library headers which conflicts with xorg headers. 82 83 Signed-off-by: Martin Krastev <krastevm@vmware.com> 84 Reviewed-by: Zack Rusin <zackr@vmware.com> 85 Signed-off-by: Zack Rusin <zackr@vmware.com> 86 87commit 5569f16a14a1b6974051b57cd2850f00882a899c 88Author: Thomas Hellstrom <thellstrom@vmware.com> 89Date: Thu Jan 17 10:19:04 2019 +0100 90 91 vmwgfx: Unify style in scanout_update and present functions 92 93 Since the functions are pretty similar, Make them look more similar in terms 94 of memory allocation methods and variable names 95 96 Reported-by: Brian Paul <brianp@vmware.com> 97 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 98 Reviewed-by: Brian Paul <brianp@vmware.com> 99 100commit 0a7c3570c92d2d831fec34415b3d821af18ba858 101Author: Thomas Hellstrom <thellstrom@vmware.com> 102Date: Wed Jan 9 09:40:03 2019 +0100 103 104 vmwgfx: Limit the number of cliprects in a drm present_readback command v3 105 106 The drm present readback command number of cliprects should not exceed 107 DRM_MODE_FB_DIRTY_MAX_CLIPS. If that number is exceeded, split 108 the command up. 109 110 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 111 Reviewed-by: Brian Paul <brianp@vmware.com> 112 Reviewed-by: Deepak Rawat <drawat@vmware.com> #v1 113 114commit 357ce6e48c2e8f2ba2b2d889d4f2041818947eb3 115Author: Thomas Hellstrom <thellstrom@vmware.com> 116Date: Wed Jan 9 09:20:45 2019 +0100 117 118 vmwgfx: Limit the number of cliprects in a drm present command v3 119 120 The drm present command number of cliprects should not exceed 121 DRM_MODE_FB_DIRTY_MAX_CLIPS. If that number is exceeded, split 122 the command up. 123 124 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 125 Reviewed-by: Brian Paul <brianp@vmware.com> #v2 126 Reviewed-by: Deepak Rawat < drawat@vmware.com> #v1 127 128commit 86189966c3dee0e0c10dd4926cebbe478129c6f0 129Author: Thomas Hellstrom <thellstrom@vmware.com> 130Date: Wed Jan 9 08:27:16 2019 +0100 131 132 vmwgfx: Limit the number of cliprects in a drm dirtyfb command v3 133 134 The drm dirtyfb command would error if the number of cliprects were larger 135 than DRM_MODE_FB_DIRTY_MAX_CLIPS. If that number is exceeded, split the 136 command up. 137 138 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 139 Reviewed-by: Brian Paul <brianp@vmware.com> #v2 140 Reviewed-by: Deepak Rawat <drawat@vmware.com> #v1 141 142commit ba4ef498fd1f66f81aaffc3dac504044b2ad7cc3 143Author: Thomas Hellstrom <thellstrom@vmware.com> 144Date: Wed Jan 9 08:00:00 2019 +0100 145 146 vmwgfx: Don't exceed the device command size limit v3 147 148 With a huge number of DMA clip rects we could exceed the device command 149 buffer command size limit. Fix this by sending multiple DMA commands 150 when we exceed the limit. 151 152 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 153 Reviewed-by: Brian Paul <brianp@vmware.com> 154 Reviewed-by: Deepak Rawat <drawat@vmware.com> #v1 155 156commit d31e8e77e1453c26a02f24b26d96b4660d29e1df 157Author: Thomas Hellstrom <thellstrom@vmware.com> 158Date: Tue Nov 27 08:37:13 2018 +0100 159 160 vmwgfx: Fix invalid memory accesses in CloseScreen 161 162 Some of the CloseScreen callbacks were referencing XA objects so move the 163 destruction of the XA state tracker to the end of drv_close_screen to avoid 164 referencing freed resources. 165 166 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 167 Reviewed-by: Deepak Rawat <drawat@vmware.com> 168 169commit 79d066da48dddcbdfcf41ecbe29cd96aaae2e11c 170Author: Thomas Hellstrom <thellstrom@vmware.com> 171Date: Tue Nov 27 08:29:06 2018 +0100 172 173 saa: Make sure damage destruction happens at the correct location 174 175 Incorrect DestroyPixmap wrapping previously made the destruction of damage 176 objects typically happen in damageDestroyPixmap(), leaving a dangling 177 damage pointer in saa_destroy_pixmap() which was only cleared. However in 178 some cases that caused us to leak damage objects. 179 180 Rework saa initialization somewhat to make sure saa_destroy_pixmap happens 181 before damageDestroyPixmap and destroy the damage object in saa_destroy_pixmap. 182 Also add a damage object destruction notifier callback that clears the 183 saa pixmap damage pointer should the damage object destruction accidentally 184 happen elsewhere. 185 186 This makes sure we don't leak damage objects. 187 188 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 189 Reviewed-by: Deepak Rawat <drawat@vmware.com> 190 191commit 53e87117bbf1cba05a5b5046db9efa961b18fc74 192Author: Thomas Hellstrom <thellstrom@vmware.com> 193Date: Fri Nov 23 09:30:42 2018 +0100 194 195 vmwgfx: Use libdrm to obtain the drm device node name v2 196 197 We were relying on a linux-specific way to do this. Now that the code is 198 used also on FreeBSD and there is functionality in libdrm to do this, 199 Use that functionality. 200 201 v2: 202 Remove unused variable warning in the !VMWGFX_LIBDRM_DEVICENAME case. 203 204 Co-authored-by: Johannes Lundberg <johalun0@gmail.com> 205 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 206 Reviewed-by: Deepak Rawat <drawat@vmware.com> #v1 207 208commit 0b34df288f0e163750962e0e59f5e329642ca457 209Author: Thomas Hellstrom <thellstrom@vmware.com> 210Date: Tue Nov 27 16:36:21 2018 +0100 211 212 vmwgfx: Fix a memory leak 213 214 We were leaking a pointer to a drm encoder. 215 216 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 217 Reviewed-by: Brian Paul <brianp@vmware.com> 218 219commit 4ec3d67da829f4e7a35ab08427002b9d7e4e5f4b 220Author: Thomas Hellstrom <thellstrom@vmware.com> 221Date: Tue Nov 27 16:20:32 2018 +0100 222 223 vmwgfx: Fix XVideo memory leaks 224 225 We were not properly freeing the port privates. 226 In order to access those at CloseScreen time, don't free the adaptor pointers 227 at XV screen init, but hold on to them until CloseScreen. 228 229 Also properly free the new_adaptors pointer. 230 231 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 232 Reviewed-by: Brian Paul <brianp@vmware.com> 233 234commit 71c85a70f1c5ff93e931181d0858af22cba231ec 235Author: Alan Coopersmith <alan.coopersmith@oracle.com> 236Date: Sat Nov 10 13:12:13 2018 -0800 237 238 Remove obsolete B16 & B32 tags in struct definitions 239 240 They were defined as empty strings on all platforms except for the 241 long unsupported Cray systems which needed to use bitfields to define 242 any type smaller than 64-bits. 243 244 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 245 Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com> 246 247commit a5815ba3eba0b222e8082a2e868d68cdc6993d5e 248Author: Thomas Hellstrom <thellstrom@vmware.com> 249Date: Tue May 15 13:57:27 2018 +0200 250 251 Bump version number for release 252 253 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 254 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 255 Reviewed-by: Deepak Rawat <drawat@vmware.com> 256 257commit 7b163c018b6343590cc36ecd58adcdc30852ee28 258Author: Thomas Hellstrom <thellstrom@vmware.com> 259Date: Tue May 15 09:52:53 2018 +0200 260 261 Require at least XA version 2.4 to enable dri3 262 263 The XA version was bumped from 2.3 to 2.4 to signal that there were no 264 significant correctness or performance regressions when running dri3 265 compared to dri2 on the vmware driver stack. 266 267 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 268 Reviewed-by: Brian Paul <brianp@vmware.com> 269 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 270 Reviewed-by: Deepak Rawat <drawat@vmware.com> 271 272commit 6b7dc2f9304298e669e7ac97594b948d1476ae52 273Author: Thomas Hellstrom <thellstrom@vmware.com> 274Date: Thu Mar 22 08:28:57 2018 +0100 275 276 vmwgfx: Clear the DRM mode before use 277 278 Avoid sending partially uninitialized data to the kernel. 279 280 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 281 Reviewed-by: Brian Paul <brianp@vmware.com> 282 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 283 284commit 0f90e2ed52082c1e2af3b47424a6c7c5d4c0c81e 285Author: Adam Jackson <ajax@redhat.com> 286Date: Wed Mar 14 21:49:14 2018 +0100 287 288 saa: Build compatibility with xserver 1.20 289 290 fbGetRotatedPixmap went away with 24bpp support, just treat it as NULL 291 and we'll do the right thing. 292 293 Signed-off-by: Adam Jackson <ajax@redhat.com> 294 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 295 296commit 3cdeaa5db76f5db7ce9482d3e067d3014076f8e4 297Author: Thomas Hellstrom <thellstrom@vmware.com> 298Date: Fri Feb 16 13:55:37 2018 +0100 299 300 Fix a couple of shadowed declaration warnings 301 302 In some enviroments, "index", "y1" and "y2" are defined globally causing 303 warnings about shadowed declarations. Fix this. 304 305 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 306 Reviewed-by: Brian Paul <brianp@vmware.com> 307 308commit 4ea5c4d5fa1dcb125137bcb76b402490ac134b4f 309Author: Thomas Hellstrom <thellstrom@vmware.com> 310Date: Fri Feb 16 13:37:56 2018 +0100 311 312 Build fixes 313 314 A couple of more build fixes for older X servers. 315 316 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 317 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 318 319commit 1899b0896dae4ce86c9871fa491b464f6af88526 320Author: Thomas Hellstrom <thellstrom@vmware.com> 321Date: Wed Feb 14 08:10:41 2018 +0100 322 323 vmwgfx: Fix server termination due to a mesa loader bug 324 325 Some versions of the Gallium loader close our drm file descriptor if 326 xa_tracker_create() fails (typically 2D VMs.) While this is mostly fixed 327 everywhere, we implement a workaround to avoid tracking down the same bug 328 again and again on those setups where this is not fixed in mesa. 329 330 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 331 Reviewed-by: Brian Paul <brianp@vmware.com> 332 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 333 Reviewed-by: Deepak Rawat <drawat@vmware.com> 334 335commit ea77ce3a200bae7d97d6e45becc898765c1aa259 336Author: Thomas Hellstrom <thellstrom@vmware.com> 337Date: Wed Feb 14 11:15:46 2018 +0100 338 339 legacy: Silence a warning about an unused constant 340 341 VMWAREBuildStr was only used in the !LIBPCIACCESS path. 342 343 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 344 Reviewed-by: Brian Paul <brianp@vmware.com> 345 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 346 Reviewed-by: Deepak Rawat <drawat@vmware.com> 347 348commit 7503135a72ab47928f5db838273851693b0dddbb 349Author: Thomas Hellstrom <thellstrom@vmware.com> 350Date: Wed Feb 14 11:12:29 2018 +0100 351 352 legacy: Avoid using the xf86[Un]blockSigio functions 353 354 They have been deprecated since mouse doesn't use SIGIO anymore. 355 356 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 357 Reviewed-by: Brian Paul <brianp@vmware.com> 358 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 359 Reviewed-by: Deepak Rawat <drawat@vmware.com> 360 361commit 69632d06e2bf255b002d39df49aaf99ae86eaf4e 362Author: Thomas Hellstrom <thellstrom@vmware.com> 363Date: Wed Feb 14 10:33:51 2018 +0100 364 365 vmwgfx: Fix potential libudev include and link failures 366 367 We were not including the proper include- and link flags for libudev. 368 369 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 370 Reviewed-by: Brian Paul <brianp@vmware.com> 371 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 372 Reviewed-by: Deepak Rawat <drawat@vmware.com> 373 374commit dc85a6da15768f7a9622677e79c618bfb35ce513 375Author: Thomas Hellstrom <thellstrom@vmware.com> 376Date: Wed Feb 14 10:14:20 2018 +0100 377 378 Fix a number of compilation warnings 379 380 There are a number of compilation warnings caused by const char pointers 381 being either explicitly or implicitly cast to char pointers. There 382 are a number of ABI differences that have hindered this so far, but 383 make a new attempt using the common_compat.h defines. 384 385 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 386 Reviewed-by: Brian Paul <brianp@vmware.com> 387 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 388 Reviewed-by: Deepak Rawat <drawat@vmware.com> 389 390commit 98199b4c0aca89b70d4f0a0e37e1e27ac15a23c7 391Author: Thomas Hellstrom <thellstrom@vmware.com> 392Date: Wed Feb 14 08:24:25 2018 +0100 393 394 vmwgfx: Remove old XWayland and XMir drivers 395 396 Old-style XWayland and XMir aren't available anymore. 397 398 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 399 Reviewed-by: Brian Paul <brianp@vmware.com> 400 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 401 Reviewed-by: Deepak Rawat <drawat@vmware.com> 402 403commit 943548ac765e5dd49d275da2b72ff07b3a64694a 404Author: Thomas Hellstrom <thellstrom@vmware.com> 405Date: Wed Feb 14 08:04:54 2018 +0100 406 407 vmwgfx: Fix compilation failure if dri3 is not available 408 409 Compilation on CentOS failed due to some code not being conditioned on 410 DRI3 headers being present. 411 412 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 413 Reviewed-by: Brian Paul <brianp@vmware.com> 414 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 415 Reviewed-by: Deepak Rawat <drawat@vmware.com> 416 417commit 2d769dc01ea16df65d294f094bea5632e0b2a04e 418Author: Thomas Hellstrom <thellstrom@vmware.com> 419Date: Wed Feb 14 07:59:57 2018 +0100 420 421 saa: Fix an include path 422 423 Fix compilation failure on CentOS 6.5 due to an incorrect include path. 424 425 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 426 Reviewed-by: Brian Paul <brianp@vmware.com> 427 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 428 Reviewed-by: Deepak Rawat <drawat@vmware.com> 429 430commit c0a2f40d978e77287d0cac95254fb6f26b2449a8 431Author: Thomas Hellstrom <thellstrom@vmware.com> 432Date: Wed Jan 3 21:28:56 2018 +0100 433 434 Bump version number for release candidate 435 436 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 437 438commit 1674782e357950ca3a7438ac2a0eefece5566ffd 439Author: Thomas Hellstrom <thellstrom@vmware.com> 440Date: Wed Jan 3 21:26:22 2018 +0100 441 442 Add the common_compat.h source file for distribution 443 444 The common_compat.h source file was missing from Makefile.am causing 445 "make distcheck" to fail. 446 447 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 448 449commit 08ad438e2793248452dab100822cbfcaa05b9938 450Author: Thomas Hellstrom <thellstrom@vmware.com> 451Date: Thu Mar 23 08:57:44 2017 +0100 452 453 vmwgfx: Support DRI3 v2 454 455 Add server-side DRI3 support 456 457 Currently DRI3 introduces extra latency with gnome-shell for 458 the following reasons: 459 1) We enable GLX_EXT_buffer_age. Causes gnome-shell to post fullscreen damage. 460 2) We enable GLX_OML_sync_control. Cases additional slowdown. 461 Not exactly sure why. 462 463 Probably we want to implement workarounds in mesa so that we don't enable 464 these extensions for gnome-shell. That can be done with driconf, using some 465 trickery. 466 467 v2: Verify that sharing an ARGB surface as XRGB works before enabling 468 DRI3. 469 470 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 471 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 472 473commit c8b8b2ee4a7b1085990443bbf1903ec265ff46f0 474Author: Thomas Hellstrom <thellstrom@vmware.com> 475Date: Thu May 11 04:39:57 2017 -0700 476 477 Revert "vmware/vmwgfx: Support multiple dri driver names" 478 479 This reverts commit d5550b7f8375c3d9003731578c1570e014577348. 480 481 The commit was intended to support video drivers, but has the side effect 482 that GLX thinks our driver supports more than it does. 483 484 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 485 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 486 487commit 3aff4ee5ca885d09d3b4ad2b0980b84978597815 488Author: Deepak Singh Rawat <drawat@vmware.com> 489Date: Mon Aug 7 16:02:59 2017 -0700 490 491 vmware/vmwgfx: Wrap deprecated 492 493 Deprecated with 43dbc556f3a4d743b9121d6cfc21961be4a9da56 494 495 Signed-off-by: Deepak Rawat <drawat@vmware.com> 496 Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com> 497 498commit d8c2f8501307a9c0c65601e897e7d919e42ba0b3 499Author: Deepak Singh Rawat <drawat@vmware.com> 500Date: Mon Aug 7 15:57:56 2017 -0700 501 502 vmware: Fix build warnings 503 504 Due to following commit in xserver there were 505 build warnings, as variables now declared const. 506 507 d89b42bda46d36fc0879611cc3b3566957ce36d0 508 e1e01d2e33c632e395d7e396f73fba8ae606b15a 509 510 Added a compat header file. 511 512 Signed-off-by: Deepak Rawat <drawat@vmware.com> 513 Reviewed-by: Brian Paul <brianp@vmware.com> 514 Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>:q 515 516commit ed0fd2f2823917a5dfefa3d3af774d4cdb40f958 517Author: Sinclair Yeh <syeh@vmware.com> 518Date: Thu Jul 13 20:16:19 2017 +0200 519 520 Revert "Revert "vmware/vmwgfx: Support multiple dri driver names"" 521 522 This reverts commit ea94dc570846655bbb8a7c75bf465907e454bef9. 523 524 Temporarily revert this for internal testing. 525 526commit 1388a1592cdab7cc8217546fb1e97da293e70393 527Author: Sinclair Yeh <syeh@vmware.com> 528Date: Thu Jul 13 20:15:52 2017 +0200 529 530 Revert "vmwgfx: Support DRI3 v2" 531 532 This reverts commit d0d5cf3e35fd47b44f32065c5cb64ba62661f03a. 533 534 Temporarily revert this for internal testing. 535 536commit d0d5cf3e35fd47b44f32065c5cb64ba62661f03a 537Author: Thomas Hellstrom <thellstrom@vmware.com> 538Date: Thu Mar 23 08:57:44 2017 +0100 539 540 vmwgfx: Support DRI3 v2 541 542 Add server-side DRI3 support 543 544 Currently DRI3 introduces extra latency with gnome-shell for 545 the following reasons: 546 1) We enable GLX_EXT_buffer_age. Causes gnome-shell to post fullscreen damage. 547 2) We enable GLX_OML_sync_control. Cases additional slowdown. 548 Not exactly sure why. 549 550 Probably we want to implement workarounds in mesa so that we don't enable 551 these extensions for gnome-shell. That can be done with driconf, using some 552 trickery. 553 554 v2: Verify that sharing an ARGB surface as XRGB works before enabling 555 DRI3. 556 557 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 558 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 559 560commit ea94dc570846655bbb8a7c75bf465907e454bef9 561Author: Thomas Hellstrom <thellstrom@vmware.com> 562Date: Thu May 11 04:39:57 2017 -0700 563 564 Revert "vmware/vmwgfx: Support multiple dri driver names" 565 566 This reverts commit d5550b7f8375c3d9003731578c1570e014577348. 567 568 The commit was intended to support video drivers, but has the side effect 569 that GLX thinks our driver supports more than it does. 570 571 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 572 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 573 574commit ed9d2a3fde597f2ff397fc11640c7001b816f3a6 575Author: Mihail Konev <k.mvc@ya.ru> 576Date: Thu Jan 26 14:00:22 2017 +1000 577 578 autogen: add default patch prefix 579 580 Signed-off-by: Mihail Konev <k.mvc@ya.ru> 581 582commit fa4998d1b9e392c47ceff2fe9663950efbb6b134 583Author: Emil Velikov <emil.l.velikov@gmail.com> 584Date: Mon Mar 9 12:00:52 2015 +0000 585 586 autogen.sh: use quoted string variables 587 588 Place quotes around the $srcdir, $ORIGDIR and $0 variables to prevent 589 fall-outs, when they contain space. 590 591 Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> 592 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> 593 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 594 595commit 4a96179cdbeb4a418dfc0b7223d9253a355e91de 596Author: Peter Hutterer <peter.hutterer@who-t.net> 597Date: Tue Jan 24 10:32:07 2017 +1000 598 599 autogen.sh: use exec instead of waiting for configure to finish 600 601 Syncs the invocation of configure with the one from the server. 602 603 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 604 Reviewed-by: Emil Velikov <emil.velikov@collabora.com> 605 606commit 670db8dbf2c8a98df34a7c6592a6277ad9beb15a 607Author: Alan Coopersmith <alan.coopersmith@oracle.com> 608Date: Sun Jun 1 21:03:14 2014 -0700 609 610 autogen.sh: Honor NOCONFIGURE=1 611 612 See http://people.gnome.org/~walters/docs/build-api.txt 613 614 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 615 616commit 91881a66195f4bc41e075a0d573ada7cdc803ba0 617Author: Alan Coopersmith <alan.coopersmith@oracle.com> 618Date: Sun Jun 1 21:03:13 2014 -0700 619 620 configure: Drop AM_MAINTAINER_MODE 621 622 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 623 624commit fade4f46e84e93153edc3472abb73635f87555e6 625Author: Thomas Hellstrom <thellstrom@vmware.com> 626Date: Wed Nov 16 11:35:58 2016 +0700 627 628 vmware/vmwgfx: Add a missing source file for distribution 629 630 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 631 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 632 633commit d64d01c7a599dc3c79125701f076a3e459cefdc2 634Author: Hans de Goede <hdegoede@redhat.com> 635Date: Thu Sep 29 19:37:22 2016 +0200 636 637 Adapt Block/WakeupHandler signature for ABI 23 638 639 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 640 Acked-by: Thomas Hellstrom <thellstrom@vmware.com> 641 642commit d5550b7f8375c3d9003731578c1570e014577348 643Author: Thomas Hellstrom <thellstrom@vmware.com> 644Date: Wed Apr 20 20:10:41 2016 +0200 645 646 vmware/vmwgfx: Support multiple dri driver names 647 648 Support sending multiple driver names and dri2 INFOREC v4. 649 650 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 651 Reviewed-by: Brian Paul <brianp@vmware.com> 652 653commit 0bcd4df1843f00f8e8507750401ec2a2a29330dc 654Author: Thomas Hellstrom <thellstrom@vmware.com> 655Date: Thu Feb 11 09:45:58 2016 +0100 656 657 vmware: Bump minor to signal availability of new features 658 659 While this may never be an official release, bump minor to enable the 660 new resolutionKMS functionality. This signals the availability of 661 autolayout and resolutionKMS support. 662 663 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 664 Reviewed-by: Brian Paul <brianp@vmware.com> 665 666commit 5036e9d11a899d831b1471f64132eefcf459b222 667Author: Thomas Hellstrom <thellstrom@vmware.com> 668Date: Thu Feb 11 09:40:55 2016 +0100 669 670 vmware/vmwgfx: Don't require screen objects to run 671 672 Using the vmwgfx ldu backend works just as fine except that we're limited 673 to implicit layout placement. 674 675 With this test on, we may end up in the odd situation that the vmwgfx kernel 676 module and the vwmare legacy Xorg driver are enabled simultaneously, and that 677 is an unsupported configuration. It will also break resolutionKMS which will 678 be enabled based on vmwgfx version but should be disabled since the legacy 679 Xorg driver runs... 680 681 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 682 Reviewed-by: Brian Paul <brianp@vmware.com> 683 684commit 57d4499cc252bd1c3362a569ca7fd99ba6b645ed 685Author: Thomas Hellstrom <thellstrom@vmware.com> 686Date: Wed Jan 20 19:08:21 2016 +0100 687 688 vmware/vmwgfx: Disable autolayout if we detect an old resolutionset 689 690 While the autolayout feature should really avoid races both with the old 691 resolutionSet RandR12 code and with new compositor layout code, let's 692 disable the autolayout feature if someone tries to set the 693 layout / resolution through the vmwarectrl interface. 694 That's most likely an old resolutionSet. 695 696 Autolayout is turned on on each new screen generation. 697 698 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 699 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 700 701commit 48fb82f13ca290e9fbb8c7f79aba5ee119a3ce7f 702Author: Thomas Hellstrom <thellstrom@vmware.com> 703Date: Wed Jan 20 14:21:40 2016 +0100 704 705 vmware/vmwgfx: Read the implicit output status from KMS if available. 706 707 This information is used to switch to software cursors if we have 708 multiple overlapping explicit outputs and thus might need to display 709 two cursors simultaneously. 710 711 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 712 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 713 714commit a81e359d88151159e3a19ceb95027029fef0a9ba 715Author: Thomas Hellstrom <thellstrom@vmware.com> 716Date: Wed Jan 20 14:04:13 2016 +0100 717 718 vmware/vmwgfx: Avoid filling the log with modeline printouts 719 720 There are specific debug options for that. 721 722 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 723 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 724 725commit 5978597da92898a424837ee89e66f66a8120480c 726Author: Thomas Hellstrom <thellstrom@vmware.com> 727Date: Wed Jan 20 13:01:21 2016 +0100 728 729 vmware/vmwgfx: Add a layout handler v2 730 731 Add a handler that, on hotplug events, scans for a new GUI layout and 732 tries to set that layout using XRandR similar to what the RandR1.2 part of 733 vmware tools resolutionSet module is doing today. 734 735 v2: Address review comments 736 - Keep the old layout in case of screen resizing errors 737 - Fix the vmwgfx_layout handler() declaration. 738 739 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 740 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 741 742commit 885e360b16fd88b48b40930c6277637615aab188 743Author: Thomas Hellstrom <thellstrom@vmware.com> 744Date: Wed Jan 20 12:09:22 2016 +0100 745 746 vmware/vmwgfx: Update RandR output properties on hotplug events v2 747 748 Update also RandR output properties when we receive hotplug events; 749 the RRGetInfo function doesn't do this. 750 751 This makes sure RandR sends out property change events to clients. 752 Also remove some debugging printouts. 753 754 v2: Address review comment from Sinclair Yeh; 755 make sure struct output_private::drm_connector is always valid. 756 757 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 758 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 759 760commit ada41a5988df8414011ed37ec51cbb402c9c8716 761Author: Thomas Hellstrom <thellstrom@vmware.com> 762Date: Wed Jan 20 11:51:46 2016 +0100 763 764 vmware/vmwgfx: Add missing includes of the config.h header. 765 766 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 767 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 768 769commit 755e38f373a511bd774a61c9420a67f630f71037 770Author: Thomas Hellstrom <thellstrom@vmware.com> 771Date: Thu Dec 18 10:46:42 2014 +0100 772 773 vmware/vmwgfx: Expose kernel connector attributes and react to hotplug events 774 775 Much of this code is borrowed from the xf86-video-modesetting driver. 776 777 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 778 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 779 780commit 1e443c4717a8b107c0c9283e222482703ac0f358 781Author: Thomas Hellstrom <thellstrom@vmware.com> 782Date: Wed Feb 25 01:51:11 2015 -0800 783 784 vmware/vmwgfx: Try to use only_hw_present semantics if screen targets are enabled 785 786 If screen targets are enabled and there is a reasonable chance that the vmwgfx 787 drm driver can use the surface backing a pixmap as a screen target surface, 788 then make that surface a modesetting framebuffer rather than the corresponding 789 DMA buffer. In practice this applies when we start scanning out from the 790 origin (0,0) of the pixmap. However, we would also like to apply the constraint 791 that the scanout area is the entire pixmap, since that is the constraint used 792 by the drm driver, but that would currently require drm framebuffer 793 reallocations and possible flicker, so disable that for now. The drm driver 794 will correctly handle the possibly oversized surface handed to it anyway, and 795 the cost we pay for this is an extra hardware copy of the dirtied area when 796 doing a software update of the scanout. 797 798 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 799 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 800 801commit 8f0cf7c03597532feb1a8ccf3bb1cafcd6170f89 802Author: Thomas Klausner <wiz@NetBSD.org> 803Date: Sun Aug 10 11:04:51 2014 +0200 804 805 Add 2560x1440 to supported resolutions. 806 807 Native resolution on iMac 27". 808 Reported missing by Benjamin Lorenz in 809 http://gnats.netbsd.org/49094 810 811 Signed-off-by: Thomas Klausner <wiz@NetBSD.org> 812 Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com> 813 814commit ab5be2312023d8ccc7156a05ccc1b0abe7679244 815Author: Adam Jackson <ajax@redhat.com> 816Date: Thu Dec 4 10:35:04 2014 -0500 817 818 saa: Adapt to GC client clip changes in xserver 1.17 819 820 1.17 always stores the client clip as a region, so there's no longer a 821 clientClipType member to look at. Change the code to just inspect 822 whether the clientClip is non-null, since that works both before and 823 after 1.17. 824 825 Signed-off-by: Adam Jackson <ajax@redhat.com> 826 Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com> 827 828commit 3f7defa8dc595342e2502206ed2006f2be20a884 829Author: Thomas Hellstrom <thellstrom@vmware.com> 830Date: Mon Nov 17 19:18:15 2014 +0100 831 832 vmware/vmwgfx: Make large Xv video blits cheaper 833 834 As screens grow larger, attempt to make large Xv video blits cheaper by 835 performing the color conversion and scaling in two steps: 836 837 1) Color conversion which has a 4x4 matrix multiplication shader is performed 838 to a bounce buffer the size of which is never larger than the source image. 839 2) Scaling is performed as a src composite blit to the destination image with 840 a simple copy shader. 841 842 This split is done only if the destination image is substantially larger than 843 the source image / bounce buffer 844 845 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 846 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 847 848commit 4664412d7a5266d2b392957406b34abc5db95e48 849Author: Stefan Dirsch <sndirsch@suse.de> 850Date: Tue Oct 7 11:29:32 2014 +0200 851 852 xf86xv.h cannot be included without first including xorg-server.h. 853 854 Without this the build fails on systems with the latest glibc, 855 throwing this error: 856 857 In file included from /usr/include/string.h:634:0, 858 from /usr/include/xorg/os.h:53, 859 from /usr/include/xorg/misc.h:115, 860 from /usr/include/xorg/screenint.h:50, 861 from /usr/include/xorg/scrnintstr.h:50, 862 from /usr/include/xorg/xvdix.h:55, 863 from /usr/include/xorg/xf86xv.h:32, 864 from vmwgfx_overlay.c:38: 865 /usr/include/xorg/os.h:579:1: error: expected identifier or '(' before '__extension__' 866 strndup(const char *str, size_t n); 867 868 This is caused by HAVE_STRNDUP not being set (it is set from xorg-server.h), 869 causing os.h to redefine it. 870 871 Signed-off-by: Stefan Dirsch <sndirsch@suse.de> 872 Reviewed-by: Daniel Stone <daniels@collabora.com> 873 Acked-by: Thomas Hellstrom <thellstrom@vmware.com> 874 875commit 8e1c4e5bd521ec2d4f1a21da432f110c606b3fde 876Author: Thomas Hellstrom <thellstrom@vmware.com> 877Date: Mon Sep 22 12:09:13 2014 +0200 878 879 vmware/saa: Fix saa_check_poly_fill_rect_noreadback v2 880 881 We were not clipping the dirty region correctly, Fix this. 882 Also actually do what the comment in the function says: If there are more than 883 an ad-hoc number of rectangles to read back, then revert to the default 884 saa_check_poly_fill_rect function that reads back the whole damage region. 885 886 v2: Fix commit log message. 887 888 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 889 Reviewed-by: Brian Paul <brianp@vmware.com> 890 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 891 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 892 893commit c23fd33d8ce8491891255d93c04e053bf333bd4e 894Author: Thomas Hellstrom <thellstrom@vmware.com> 895Date: Mon Sep 22 11:43:02 2014 +0200 896 897 vmware/saa: Fix saa_check_copy_window dirty region v3 898 899 The saa_check_copy_window could dirty regions that were never touched, since 900 we were using the X server provided damage region rather than the more 901 detailed region actually copied. This would have been OK if we had first done 902 a read-back of the region to be dirtied, but since we want to avoid that, 903 instead compute the detailed destination region and use that for dirtying. 904 905 This fixes rendering glitches seen with motif applications. 906 907 v2: Fix whitespace error. 908 v3: Move dirty region computation. 909 910 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 911 Reviewed-by: Brian Paul <brianp@vmware.com> 912 Reviewed-by: Sinclair Yeh <syeh@vmware.com> 913 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 914 915commit f86434ac69bd288c7fba65d178c22a91a0f4f6c2 916Author: Thomas Hellstrom <thellstrom@vmware.com> 917Date: Mon Sep 15 15:13:53 2014 +0200 918 919 Revert "vmware/saa: Disable the noreadback polyfillrect optimization" 920 921 This reverts commit 88c487cb19aaf2882d582f7f887a52935ef8d92e. 922 923 While the commit made the rendering artefacts less frequent, they didn't 924 disappear completely and are likely caused by something else, so revert this 925 commit. 926 927 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 928 Acked-by: Sinclair Yeh <syeh@vmware.com> 929 930commit 8545a166931617af0d2a9ca8175fd8b11c67e246 931Author: Thomas Hellstrom <thellstrom@vmware.com> 932Date: Tue Sep 9 01:59:07 2014 -0700 933 934 vmware/vmwgfx: Fix cursor problem with multimon 935 936 In a multimon environment, the cursor would sometimes disappear on the 937 newly enabled screen. Fix this. 938 939 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 940 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 941 942commit 32562129750077a23f26f2e69adc8403eb24bf3f 943Author: Thomas Hellstrom <thellstrom@vmware.com> 944Date: Wed Sep 3 07:08:41 2014 -0700 945 946 vmware/vmwgfx: Don't add pixmaps to the pixmap list if they're already on it 947 948 This could cause loops through the list to spin indefinitely. 949 This would most likely occur at VT switches. 950 951 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 952 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 953 954commit 8b6316a312689d83ca8f8de939698bc137206f58 955Author: Thomas Hellstrom <thellstrom@vmware.com> 956Date: Thu Aug 28 15:32:45 2014 +0200 957 958 vmware/legacy: Fix overlay related compile breakage on earlier X servers 959 960 Introduced by commit: 961 vmware/legacy: Apply same fix to auto colrkey fill 962 963 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 964 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 965 966commit 8a47e9a45036005d46a60cf8df366baffcb9a8d1 967Author: Thomas Hellstrom <thellstrom@vmware.com> 968Date: Thu Aug 28 13:43:03 2014 +0200 969 970 vmware: Fix bootstrap related compile errors on earlier X servers 971 972 Introduced by commit 973 Add support for XSERVER_PLATFORM_BUS 974 975 Note that the vmware DriverRec declaration can be cleaned up 976 considerably using C99 designated initializers. Perhaps something for 977 the next release... 978 979 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 980 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 981 982commit f5d8c8f1fe2de727018600f471b6a8829fe492ad 983Author: Thomas Hellstrom <thellstrom@vmware.com> 984Date: Wed Aug 20 13:02:58 2014 +0200 985 986 vmware: Clear the PITCHLOCK register if available before modeset 987 988 In some situations the fbdev driver may have set this register before legacy 989 driver startup causing a weird-looking desktop. Make sure this register 990 is cleared on each modeset. 991 992 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 993 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 994 995commit b1697434348799b0808a6c9e9a899394075b980c 996Author: Jakob Bornecrantz <jakob@vmware.com> 997Date: Thu Aug 14 16:42:05 2014 +0200 998 999 vmware/legacy: Fix out of bound array indexing 1000 1001 Spotted by coverity. 1002 1003 Reported-by: Adam Jackson <ajax@redhat.com> 1004 Reviewed-by: Adam Jackson <ajax@redhat.com> 1005 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1006 1007commit 0825d98744b20a20af7bd81d347334ab4a1de015 1008Author: Jakob Bornecrantz <jakob@vmware.com> 1009Date: Sat Aug 9 00:19:17 2014 +0200 1010 1011 vmware/vmwgfx: Fix compilation issue on older X servers 1012 1013 Introduced in 268307 "Add support for server managed fds" 1014 1015 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1016 Reviewed-by: Brian Paul <brianp@vmware.com> 1017 1018commit 0a212afefd8670a1823c9b6474de8bf26d33bdeb 1019Author: Jakob Bornecrantz <jakob@vmware.com> 1020Date: Fri Aug 8 21:37:38 2014 +0200 1021 1022 vmware/legacy: Fix uninitialized values in video commands 1023 1024 Fixed bug where video stopped working on systems that didn't have the new kms 1025 enabled kernel driver installed. Found on CentOS 6.4. 1026 1027 After updating the register header SVGA_VIDEO_NUM_REGS value got upped by two 1028 in order to support GMR and Screen Objects. Since this path is mostly used 1029 on older hosts that may not support them, don't send them at all. 1030 1031 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1032 Reviewed-by: Brian Paul <brianp@vmware.com> 1033 1034commit 0a596fd0c92fdcf73071869a59029263a2596018 1035Author: Thomas Hellstrom <thellstrom@vmware.com> 1036Date: Wed Jul 2 14:21:45 2014 +0200 1037 1038 vmware/vmwgfx: Fix an error path segfault 1039 1040 Part of freedesktop.org bugzilla bug #80645 1041 If taking a scanout reference on a pixmap fails, the 1042 struct vmwgfx_screen_entry::pixmap pointer must be set to NULL, otherwise 1043 the driver will incorrectly attempt to remove the scanout reference in the 1044 error path, causing a segfault. 1045 1046 This problem is seen in the above-mentioned bug, but it is not the root 1047 cause of the problem. With this patch applied, the server will terminate 1048 cleanly instead of segfaulting. 1049 1050 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1051 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1052 1053commit 4e378a730d2d2deffc915af2a3d013d2a994a1b0 1054Author: Jakob Bornecrantz <jakob@vmware.com> 1055Date: Wed May 21 11:52:29 2014 +0200 1056 1057 vmware/legacy: Apply same fix to auto colorkey fill 1058 1059 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1060 Reviewed-by: Thomas Hellström <thellstrom@vmware.com> 1061 1062commit 82c9b0c1d34459c568d004521e3f35a69f5d7ae3 1063Author: Jakob Bornecrantz <jakob@vmware.com> 1064Date: Wed May 21 11:51:46 2014 +0200 1065 1066 vmware/vmwgfx: Fix auto colorkey fill usage 1067 1068 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1069 Reviewed-by: Thomas Hellström <thellstrom@vmware.com> 1070 1071commit 43e043868fbd74e28cd392a3665b36abdfe87de0 1072Author: Jakob Bornecrantz <jakob@vmware.com> 1073Date: Wed May 21 11:54:48 2014 +0200 1074 1075 vmware/vmwgfx: A bit more error logging when handling cursors 1076 1077 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1078 Reviewed-by: Thomas Hellström <thellstrom@vmware.com> 1079 1080commit aa7309254808e27a4ac5d3dc674be179646280b5 1081Author: Jakob Bornecrantz <jakob@vmware.com> 1082Date: Wed May 21 11:54:24 2014 +0200 1083 1084 vmware/vmwgfx: Don't move cursors without images 1085 1086 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1087 Reviewed-by: Thomas Hellström <thellstrom@vmware.com> 1088 1089commit e8d6facf34adf1c601a3afd0a984ce90ad8fe059 1090Author: Thomas Hellstrom <thellstrom@vmware.com> 1091Date: Tue May 20 06:29:19 2014 -0700 1092 1093 vmware/vmwgfx: Fix xwayland build on newer XA versions 1094 1095 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1096 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1097 1098commit 268307e85c06be27345a6808bcf2b3c9887bf8bc 1099Author: Hans de Goede <hdegoede@redhat.com> 1100Date: Tue May 6 16:33:21 2014 +0200 1101 1102 Add support for server managed fds 1103 1104 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 1105 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1106 1107commit 675fe58eee8acdb302c3501236b86a3289991053 1108Author: Hans de Goede <hdegoede@redhat.com> 1109Date: Tue May 6 16:00:50 2014 +0200 1110 1111 Add support for XSERVER_PLATFORM_BUS 1112 1113 This is a preparation patch for adding support for server managed fds. 1114 1115 Signed-off-by: Hans de Goede <hdegoede@redhat.com> 1116 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1117 1118commit 88c487cb19aaf2882d582f7f887a52935ef8d92e 1119Author: Thomas Hellstrom <thellstrom@vmware.com> 1120Date: Fri May 9 09:35:59 2014 +0200 1121 1122 vmware/saa: Disable the noreadback polyfillrect optimization 1123 1124 It causes rendering artefacts with some motif applications due to 1125 damage area confusion. Until that is sorted out, temporarily disable the 1126 optimization. 1127 1128 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1129 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1130 1131commit f31eed4a578784329a7038aa01acaa10a2d0c246 1132Author: Thomas Hellstrom <thellstrom@vmware.com> 1133Date: Thu May 8 09:24:04 2014 +0200 1134 1135 vmware/vmwgfx: Fix initial dirty region for pixmaps 1136 1137 When storage is attached to pixmaps for the first time the dirty region is 1138 set to cover either the hardware surface or the software buffer based on 1139 the presence of the hardware surface. 1140 1141 However, if the storage was created as part of an accelerated operation, 1142 the dirty region was assigned before the hardware surface was assigned to the 1143 pixmap, causing the dirty region to incorrectly cover the software buffer. 1144 1145 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1146 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1147 1148commit 1ed14687a3f1092db8504fb9cf4f446451801665 1149Author: Thomas Hellstrom <thellstrom@vmware.com> 1150Date: Thu Apr 10 14:05:24 2014 +0200 1151 1152 vmware/vmwgfx: Set desired modes after xMir screen init 1153 1154 While XMir does initial mode configuration, it leaves setting initial 1155 modes to the DDX driver. 1156 1157 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1158 Reviewed-by: Brian Paul <brianp@vmware.com> 1159 1160commit 64e0faf1480663a990e0f6aeef7af0e69845af49 1161Author: Thomas Hellstrom <thellstrom@vmware.com> 1162Date: Wed Apr 2 14:05:23 2014 +0200 1163 1164 vmware/vmwgfx: Don't tell XA that we have a mask when we haven't 1165 1166 In some cases, the X server sends us a composit operation with 1167 mask_pict != NULL, but mask_pix == NULL. Assume there's no mask 1168 involved in that case. 1169 1170 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1171 Reviewed-by: Brian Paul <brianp@vmware.com> 1172 1173commit c062d08a50a26c99a3b8f565a8b2b00e8dc21c3f 1174Author: Thomas Hellstrom <thellstrom@vmware.com> 1175Date: Thu Jan 16 18:27:57 2014 +0100 1176 1177 vmware/vmwgfx: Don't change backing-store of active scanout surfaces 1178 1179 With option "HwPresents" on, the driver would sometimes change backing 1180 store of active scanout surfaces, making the kernel module refuse to 1181 present. This was caused by scanout surfaces not having the RENDERTARGET flag 1182 on by default. So when rendered to, using copies or composites, they 1183 would be reallocated. Fix this by adding the RENDERTARGET flag from start. 1184 1185 Also add code that prints out an error message when we change backing store 1186 of active scanout surfaces 1187 1188 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1189 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1190 1191commit c42e1f750629f18144d065b8bba5e2eaf6cbf675 1192Author: Julien Cristau <jcristau@debian.org> 1193Date: Tue Apr 1 10:11:13 2014 +0200 1194 1195 configure: fix build without xatracker 1196 1197 ACTION-IF-NOT-FOUND is the fourth argument to PKG_CHECK_MODULES, not the 1198 fifth. 1199 1200 Debian bug#743239 <https://bugs.debian.org/743239> 1201 1202 Reviewed-by: Thomas Klausner <wiz@NetBSD.org> 1203 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1204 Signed-off-by: Julien Cristau <jcristau@debian.org> 1205 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1206 1207commit cbd56b3759f01ec18ae0705d023e4c0b2ab38c25 1208Author: Jakob Bornecrantz <jakob@vmware.com> 1209Date: Wed Mar 26 12:45:52 2014 +0100 1210 1211 Bump the version number to 13.0.2 1212 1213 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1214 1215commit 51cdfa6f5102bde92c98630ace6650e4b286f6e2 1216Author: Jakob Bornecrantz <jakob@vmware.com> 1217Date: Fri Feb 21 18:10:07 2014 +0100 1218 1219 vmware: Fix compilation on newer X servers 1220 1221 Found by Bryan Lee. Later versions of X.org turns dispMode pointers 1222 into const upsetting gcc, turn them into size_t's instead. 1223 1224 Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com> 1225 1226commit 8da981712f62050076cff53e1b40ed1e307fcca8 1227Author: Thomas Hellstrom <thellstrom@vmware.com> 1228Date: Wed Jan 15 11:04:05 2014 +0100 1229 1230 vmware/vmwgfx: Always allocate shared hardware surfaces. 1231 1232 Hardware surfaces are all likely to be shared at some point, and we *really* 1233 don't want to change a hardware surface that is bound as a drm framebuffer. 1234 1235 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1236 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1237 1238commit 31bff9f7f3d9a68fd1449532e8ab50065de63857 1239Author: Thomas Hellstrom <thellstrom@vmware.com> 1240Date: Thu Jan 9 13:53:59 2014 +0100 1241 1242 vmware: Require libdrm 2.4.38 to build XMir. 1243 1244 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1245 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1246 1247commit d6b179a5addef6456325adf241eb8dcaf8e4c3d6 1248Author: Thomas Hellstrom <thellstrom@vmware.com> 1249Date: Thu Jan 9 13:28:22 2014 +0100 1250 1251 vmware: Fix build errors and warnings 1252 1253 A previous commit and the hosted merge unfortunately brought in some build 1254 errors / warnings on early X servers. 1255 1256 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1257 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1258 1259commit f37684e7169b9c0ab23ff748d5acacb65fad82f7 1260Author: Thomas Hellstrom <thellstrom@vmware.com> 1261Date: Tue Jan 7 17:55:49 2014 +0100 1262 1263 vmwgfx: Block DMA to prime surfaces for now 1264 1265 Since there is currently no _good_ way to get the surface format of a 1266 prime surface, we block DMA to these surfaces; we don't know if our 1267 software data is compatible with the surface format. 1268 1269 This patch also makes sure that there is a hardware surface backing the 1270 drawable we copy from. 1271 1272 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1273 Reviewed-by: Brian Paul <brianp@vmware.com> 1274 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1275 1276commit 97ce302ddd6e86397ea56ea9089b9549af73c3ac 1277Author: Thomas Hellstrom <thellstrom@vmware.com> 1278Date: Mon Jan 6 12:37:36 2014 +0100 1279 1280 vmwgfx: Enable direct dmas 1281 1282 Enable direct dmas instead of using the xa-provided dma functionality. 1283 This saves a bounce-buffer software copy of all dma'd contents. 1284 1285 This also implies that all drawables with mixed software / hardware contents 1286 will use a kernel buffer for software rendering. 1287 1288 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1289 Reviewed-by: Brian Paul <brianp@vmware.com> 1290 1291commit fd636e39a3f32a6dcd107c370513d8cb9aae26c2 1292Author: Thomas Hellstrom <thellstrom@vmware.com> 1293Date: Mon Dec 16 06:21:09 2013 -0800 1294 1295 vmwgfx: Add support for XWayland 1296 1297 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1298 Reviewed-by: Brian Paul <brianp@vmware.com> 1299 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1300 1301commit c020923597d3bc30dffa89ba0a47f3b9517dd5fb 1302Author: Thomas Hellstrom <thellstrom@vmware.com> 1303Date: Mon Dec 16 06:13:25 2013 -0800 1304 1305 vmwgfx: Add support for XMir v2. 1306 1307 Use the hosted infrastructure to add support for XMir. 1308 Helpers go in vmwgfx_saa.c. 1309 1310 v2: Added comments for the helpers, and added a 1311 vmwgfx_flush_dri2 to be executed when coming back from vt switch. 1312 1313 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1314 Reviewed-by: Brian Paul <brianp@vmware.com> 1315 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1316 1317commit c8e1c498431b1f8758f62c94131c302cf3d07b39 1318Author: Thomas Hellstrom <thellstrom@vmware.com> 1319Date: Mon Dec 16 06:05:35 2013 -0800 1320 1321 vmwgfx: Add an infrastructure to be able to run hosted under a compositor v2 1322 1323 Figure out what's needed both for XMir and XWayland and make a common 1324 driver structure out of it. 1325 1326 v2: Added a lot of comments. No code change. 1327 1328 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1329 Reviewed-by: Brian Paul <brianp@vmware.com> 1330 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1331 1332commit f45a551e9b70ccec8f92df02747d4e2af299cf8c 1333Author: Thomas Hellstrom <thellstrom@vmware.com> 1334Date: Mon Dec 16 10:38:21 2013 +0100 1335 1336 vmwgfx: Fix compile breakage on XA version 1. 1337 1338 A recent revert brought back code that assumed XA version 2. 1339 1340 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1341 Reviewed-by: trivial 1342 1343commit 93228f3dd3355a25583d22dbb96791678b93be3e 1344Author: Thomas Hellstrom <thellstrom@vmware.com> 1345Date: Wed Dec 11 06:54:02 2013 -0800 1346 1347 vmwgfx: Really allow XA version 2. 1348 1349 When XA starts to correctly advertise version 2, we didn't allow it. 1350 Fix this. 1351 1352 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1353 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1354 1355commit 4af6062ce281a7f4aaa50985fe757e9569208ee1 1356Author: Thomas Hellstrom <thellstrom@vmware.com> 1357Date: Wed Dec 11 04:59:49 2013 -0800 1358 1359 Revert "vmwgfx: Get rid of device-specific DMA code" 1360 1361 This reverts commit 45b2457516a9db4bd1d60fbb24a1efbe2d9dd932. 1362 1363 Reverting this because using direct DMA for shared pixmaps should 1364 boost performance. It should be usable both in a composited environment 1365 and when running hosted. Need to retest the DMA flaws I saw before disabling 1366 this code. 1367 1368 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1369 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1370 1371commit c36cbcefb8526944141eb431e1e736dc4485a36e 1372Author: Thomas Hellstrom <thellstrom@vmware.com> 1373Date: Mon Dec 9 00:19:55 2013 -0800 1374 1375 vmwgfx: Remove stray 1 1376 1377 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1378 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1379 1380commit a40cbd7b4fa19219849b4b0a48f07538772c3e22 1381Author: Thomas Hellstrom <thellstrom@vmware.com> 1382Date: Thu Dec 5 02:58:47 2013 -0800 1383 1384 vmwgfx: Support also XA version 1 v2 1385 1386 We need to support also XA version 1, since we want to be backwards 1387 compatible with older mesa releases. 1388 Unfortunately, the intended way of detecting XA major version at 1389 compile-time was broken on mesa 10 so we need a workaround that tests 1390 for XA version 2 at config time. 1391 1392 v2: Update the test for XA version 2. 1393 1394 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1395 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1396 1397commit 7ac45326eccbf3bea9a4a8f95b9662a36ab58eb0 1398Author: Thomas Hellstrom <thellstrom@vmware.com> 1399Date: Thu Nov 28 02:02:20 2013 -0800 1400 1401 vmwgfx: Fix some compilation warnings and indentations 1402 1403 Most of the compilation warnings we're seeing are due to bad xorg headers, 1404 but fix the ones we can fix in the driver. Some indentation fixes as well. 1405 1406 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1407 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1408 1409commit e26dec8f5278df74a102493bf6d2d8444dab5d6d 1410Author: Gaetan Nadon <memsize@videotron.ca> 1411Date: Mon Oct 28 18:21:29 2013 -0400 1412 1413 vmwgfx: handle changes of DamageUnregister API in 1.14.99.2 1414 1415 Fix is inspired from the intel driver. 1416 1417 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1418 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 1419 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1420 1421commit 45b2457516a9db4bd1d60fbb24a1efbe2d9dd932 1422Author: Thomas Hellstrom <thellstrom@vmware.com> 1423Date: Wed Sep 19 21:50:40 2012 +0200 1424 1425 vmwgfx: Get rid of device-specific DMA code 1426 1427 It's rarely used and things seem to work well enough on top of XA. 1428 1429 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1430 Reviewed-by: Zack Rusin <zackr@vmware.com> 1431 1432commit 7192acf9f0bf8e7176ab0b803b861a858623f709 1433Author: Thomas Hellstrom <thellstrom@vmware.com> 1434Date: Wed Sep 19 20:36:57 2012 +0200 1435 1436 vmwgfx: Implement textured video completely on top of XA. 1437 1438 Remove device-specific hacks. This may increase resource usage a little 1439 on old hardware revisions, but we don't need separate code paths on 1440 different hardware revisions. 1441 1442 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1443 Reviewed-by: Zack Rusin <zackr@vmware.com> 1444 1445commit 835ce4698f916ba080f4132988fd4caf898e0b1e 1446Author: Thomas Hellstrom <thellstrom@vmware.com> 1447Date: Thu Sep 26 01:25:33 2013 -0700 1448 1449 vmwgfx: Avoid HW operations when not master 1450 1451 Note that for DRI2, a dri2_copy_region becomes a NOP when not master. 1452 Additionally, all dri2 operations that lead to a potential kernel 1453 access will return FALSE. 1454 1455 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1456 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1457 1458commit 0945bea5fc32eacb7bf42639efbd45dcd43e7ab5 1459Author: Rob Clark <robdclark@gmail.com> 1460Date: Mon Jun 10 13:31:31 2013 -0400 1461 1462 vmwgfx: update for XA API changes 1463 1464 Signed-off-by: Rob Clark <robdclark@gmail.com> 1465 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1466 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1467 Tested-by: Jakob Bornecrantz <jakob@vmware.com> 1468 1469commit b3345886332ee2b9fd1348dc2cd36257dfd71986 1470Author: Jakob Bornecrantz <jakob@vmware.com> 1471Date: Wed Apr 17 12:03:32 2013 +0200 1472 1473 Bump the version number to 13.0.1 1474 1475commit d13a6334fe76f4e869b2cb8c0a6a13c185b1bedd 1476Author: Loïc Yhuel <loic.yhuel@gmail.com> 1477Date: Thu Apr 11 11:29:10 2013 +0200 1478 1479 vmwgfx: Use myGlyphs to fix crashes 1480 1481 Tested-by: Christian Hesse <mail@eworm.de> 1482 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1483 Signed-off-by: Loïc Yhuel <loic.yhuel@gmail.com> 1484 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1485 1486commit 23d87c92b7211cc5249eefe4ab3a2c7f370e279e 1487Author: Zack Rusin <zackr@vmware.com> 1488Date: Thu Feb 14 14:33:01 2013 -0800 1489 1490 Kill mibstore 1491 1492 It was a noop for at least 5 years and it has been removed. 1493 1494 Signed-off-by: Zack Rusin <zackr@vmware.com> 1495 1496commit 686859fba16ed2616c38bbe5d4e38abf4c5b7ae9 1497Author: Zack Rusin <zackr@vmware.com> 1498Date: Tue Jan 15 12:05:45 2013 -0800 1499 1500 Bump the version number to 13.0.0 1501 1502 New abi and various fixes 1503 1504 Signed-off-by: Zack Rusin <zackr@vmware.com> 1505 1506commit adf375f3d400e51b1e445e89de6bd82be0bb4ef3 1507Author: Zack Rusin <zackr@vmware.com> 1508Date: Thu Dec 6 16:40:13 2012 -0500 1509 1510 xf86-video-vmware 12.0.99.901 1511 1512 Internal release for a round of QE testing 1513 1514 Signed-off-by: Zack Rusin <zackr@vmware.com> 1515 1516commit 007bed30270ca9548dfe6e359972b56cbb8e2f13 1517Author: Jakob Bornecrantz <jakob@vmware.com> 1518Date: Tue May 29 23:01:14 2012 +0200 1519 1520 vmwgfx: Guard against null rotate pixmap in shadow destroy 1521 1522 Reviewed-by: Thomas Hellström <thellstrom@vmware.com> 1523 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1524 1525commit e5ac80d8f72963313358e31cfb1ab98583e26f23 1526Author: Dave Airlie <airlied@redhat.com> 1527Date: Tue Jul 3 13:50:30 2012 +0100 1528 1529 vmware: avoid warnings about shadowing 'i' 1530 1531 Signed-off-by: Dave Airlie <airlied@redhat.com> 1532 1533commit 46c0862d205eefa985e36a74391ca392640cb898 1534Author: Dave Airlie <airlied@redhat.com> 1535Date: Sat Jun 16 14:43:52 2012 +0100 1536 1537 vmware: drop infoFromScreen inline in favour of new interface 1538 1539 This also should fix the build regression introduced with the fix 1540 for this macro. 1541 1542 Signed-off-by: Dave Airlie <airlied@redhat.com> 1543 1544commit 50c01998c7800bc02fd6656cc14efa715bf63e08 1545Author: Robert Hooker <sarvatt@ubuntu.com> 1546Date: Fri Jun 15 11:56:59 2012 -0400 1547 1548 vmware: Fix up some warnings after new compat API change. 1549 1550 Fixes multiple instances of these warnings on x86_64: 1551 ../../src/vmware.h:180:5: warning: implicit declaration of function 'xf86ScreenToScrn' [-Wimplicit-function-declaration] 1552 ../../src/vmware.h:180:5: warning: nested extern declaration of 'xf86ScreenToScrn' [-Wnested-externs] 1553 ../../src/vmware.h:180:5: warning: return makes pointer from integer without a cast [enabled by default] 1554 Function `xf86ScreenToScrn' implicitly converted to pointer at ../../src/vmware.h:180 1555 1556 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1557 Signed-off-by: Robert Hooker <sarvatt@ubuntu.com> 1558 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1559 1560commit 61df95a86f4997e342d50d7779b00aba2e8849a6 1561Author: Dave Airlie <airlied@redhat.com> 1562Date: Wed Jun 6 09:58:01 2012 +0100 1563 1564 vmware: port vmware driver to new compat API 1565 1566 This is a port of the vmware driver to the new compat API. 1567 1568 Tested-by: Jakob Bornecrantz <jakob@vmware.com> 1569 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1570 Signed-off-by: Dave Airlie <airlied@redhat.com> 1571 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1572 1573commit 46cdf0ed0ad9df66f36f95e27b209d454f67d526 1574Author: Brian Paul <brianp@vmware.com> 1575Date: Fri Apr 20 14:24:26 2012 -0600 1576 1577 vmwgfx: set the XA_FLAG_SHARED flag for composite dest surfaces 1578 1579 This fixes a failed assertion in the gallium/svga driver in the 1580 svga_texture_get_handle() function. The texture resource wasn't 1581 getting created with the PIPE_BIND_SHARED flag so the !cachable 1582 assertion would fail and the X session would abort. This didn't 1583 happen with release builds. 1584 1585 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1586 1587commit 4d76cfe5e8e5d594ec357b0ef3339f56eeca9c64 1588Author: Jakob Bornecrantz <jakob@vmware.com> 1589Date: Tue Mar 20 22:05:30 2012 +0100 1590 1591 Bump version number for the release 1592 1593 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1594 1595commit c6f27561c7cf7c30f89493d64e3c5694b5f53469 1596Author: Jakob Bornecrantz <jakob@vmware.com> 1597Date: Mon Mar 19 14:53:34 2012 +0100 1598 1599 vmware/legacy: Fix crash on Xserver 1.12.0 1600 1601 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1602 Reviewed-by: Zack Rusin <zackr@vmware.com> 1603 1604commit b70116b907aa8545f0275589a52275235083bb40 1605Author: Zack Rusin <zackr@vmware.com> 1606Date: Thu Mar 15 11:37:26 2012 -0400 1607 1608 Fix a memory corruption due to a redundant free. 1609 1610 We use the empty rect purely as a placeholder and never 1611 initialize it with its own memory so lets not try to free 1612 it. 1613 Spotted by Dave Airlie. 1614 1615 Signed-off-by: Zack Rusin <zackr@vmware.com> 1616 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1617 1618commit 57f1204292a116b35d5930752bc06f9065332879 1619Author: Jakob Bornecrantz <jakob@vmware.com> 1620Date: Tue Mar 13 21:41:11 2012 +0100 1621 1622 Bump version number for the release 1623 1624 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1625 1626commit e585f9ea04bbd17398d2fa804be38a4e75bc60e4 1627Author: Gaetan Nadon <memsize@videotron.ca> 1628Date: Tue Mar 13 09:44:46 2012 -0400 1629 1630 config: move AC_SYS_LARGEFILE in autoconf init section 1631 1632 There is no need for this statement to be conditionally included. 1633 It adds support for large files without taking anything away. 1634 1635 http://www.gnu.org/software/autoconf/manual/autoconf.html 1636 1637 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 1638 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1639 1640commit 04169f3860e4a7d03485631bee40c05fc45b72c6 1641Author: Gaetan Nadon <memsize@videotron.ca> 1642Date: Mon Mar 12 14:51:38 2012 -0400 1643 1644 config: include saa and vmwgfx subdirs in the tarball 1645 1646 Use AM_CONDITIONAL. Automake knows what to distribute. 1647 It needs to be able to navigate down the subdirs to find what 1648 needs to be included in the tarball. 1649 1650 To test reliably, create a tarball and expand it into a separate 1651 directory and build with xatracker. Distcheck will not detect 1652 missing code when such code is configured not to build. 1653 1654 The content of a tarball *must* always be identical, regardless 1655 of the configuration options used or on which platform it was configured. 1656 1657 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 1658 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1659 1660commit e6836ff099e5565dea58a53bba9b2df377241ef7 1661Author: Zack Rusin <zackr@vmware.com> 1662Date: Thu Mar 8 09:55:07 2012 -0500 1663 1664 Bump version number for the release 1665 1666 Signed-off-by: Zack Rusin <zackr@vmware.com> 1667 1668commit c5c38bab8c493985349356be5fadc14d3aee8040 1669Author: Thomas Hellstrom <thellstrom@vmware.com> 1670Date: Wed Feb 8 15:00:20 2012 +0100 1671 1672 vmware/legacy: Fix initial mode size v2 1673 1674 Commit "vmwlegacy: Make the default be a minium of 800x600..." 1675 read the dimensions from the incorrect register pair. 1676 1677 v2: Also treat initial widths and heights larger than the maximum 1678 values as an error and revert back to the minimum. 1679 1680 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1681 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1682 1683commit 49b4cbc5100ce9a4cc40132c2839f9401a58b92e 1684Author: Thomas Hellstrom <thellstrom@vmware.com> 1685Date: Mon Feb 6 19:48:57 2012 +0100 1686 1687 saa: Try to avoid readbacks during polyFillRect 1688 1689 This significantly improves performance of shaped windows on top of 3D 1690 contents 1691 1692 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1693 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1694 Reviewed-by: Zack Rusin <zackr@vmware.com> 1695 1696commit ecf6c5b1081e70ec64132a8188c8cc15945c2290 1697Author: Thomas Hellstrom <thellstrom@vmware.com> 1698Date: Thu Feb 2 10:24:26 2012 +0100 1699 1700 vmware: Kill warnings on Xserver 1.12 1701 1702 Compile-tested on servers 1.0 through 1.12. 1703 1704 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1705 1706commit 3e18d43c7f8ab5dcb94ac285a687eaaa1a578250 1707Author: Thomas Hellstrom <thellstrom@vmware.com> 1708Date: Wed Feb 1 17:30:52 2012 +0100 1709 1710 vmwgfx: Kill a deprecation warning on server 1.12 1711 1712 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1713 1714commit bfb286bea967c26426aebda2a581f35da0d86884 1715Author: Thomas Hellstrom <thellstrom@vmware.com> 1716Date: Wed Feb 1 17:25:19 2012 +0100 1717 1718 vmwgfx: Fix up swap[sl] for server 1.12 1719 1720 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1721 1722commit 8ff19c2b2f288d6851b444dbda39544980837b73 1723Author: Thomas Hellstrom <thellstrom@vmware.com> 1724Date: Wed Feb 1 16:59:13 2012 +0100 1725 1726 vmwgfx: Avoid including a library header and use pixman for type conversion 1727 1728 The PictTransform type is a typedef of a pixman type. 1729 1730 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1731 1732commit 193b5118199028ea0180785bb3f4c8a16359b8b0 1733Author: Thomas Hellstrom <thellstrom@vmware.com> 1734Date: Fri Jan 13 17:01:02 2012 +0100 1735 1736 Bump version number for release 1737 1738 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1739 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1740 1741commit c5af120177949c82cbd0611da565fef9c2c829cb 1742Author: Thomas Hellstrom <thellstrom@vmware.com> 1743Date: Fri Jan 13 17:14:47 2012 +0100 1744 1745 Make the driver build pass "make distcheck" 1746 1747 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1748 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1749 1750commit e6d03aceb1195de0b490f39024f34a33b1a8f6fc 1751Author: Thomas Hellstrom <thellstrom@vmware.com> 1752Date: Thu Jan 12 14:00:41 2012 +0100 1753 1754 vmwgfx: Adapt to new XA version number v2 1755 1756 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1757 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1758 1759commit 59435d2d3af8a9f64ee40eac4f0479470ba6badf 1760Author: Thomas Hellstrom <thellstrom@vmware.com> 1761Date: Thu Jan 12 17:57:37 2012 +0100 1762 1763 vmwgfx,saa: Fix a bunch of compilation warnings 1764 1765 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1766 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1767 1768commit 0628ad20f31a239f99fcbbfe42197c56a5ae8864 1769Author: Jakob Bornecrantz <jakob@vmware.com> 1770Date: Fri Dec 9 07:34:58 2011 +0100 1771 1772 vmwlegacy: Make the default be a minimum of 800x600 1773 1774 This patch and the corresponding vmwgfx patch formalizes both drivers 1775 on the same behviour. The minimum is 800x600 (unless the max size is 1776 smaller). This makes it unnecessary to check against VRAM size since 1777 it is always at least large enough to fit the max values. 1778 1779 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1780 1781commit 8ddbb3216d29b2ecd336d50461216feef0900fd9 1782Author: Thomas Hellstrom <thellstrom@vmware.com> 1783Date: Thu Jan 12 10:00:41 2012 +0100 1784 1785 vmware: Make sure the driver builds on old servers 1786 1787 Move things around a bit so the driver compiles with 1788 servers back to Xserver 1.0. 1789 1790 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1791 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1792 1793commit 194ff6f1517d0a8250e64c3d768a9b7ae9ed3140 1794Author: Thomas Hellström <thellstrom@vmware.com> 1795Date: Wed Jan 11 14:49:56 2012 +0100 1796 1797 vmware: Update Manpage 1798 1799 Update the Manpage with new options and a brief modesetting- and 3D 1800 description. 1801 1802 Signed-off-by: Thomas Hellström <thellstrom@vmware.com> 1803 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1804 1805commit b5546fb1de3f3859151a91e98ab0bd24b6789e2e 1806Merge: abf9cb7 5748c33 1807Author: Thomas Hellstrom <thellstrom@vmware.com> 1808Date: Mon Dec 19 20:25:43 2011 +0100 1809 1810 Merge branch 'vmwgfx_branch' 1811 1812commit 5748c33535bc7d3009b9758653885e6ae0e73002 1813Author: Thomas Hellstrom <thellstrom@vmware.com> 1814Date: Wed Dec 14 12:38:27 2011 +0100 1815 1816 saa, vmwgfx: Fix saa_copy_composite 1817 1818 The traditional accelerated copy methods aren't format aware. 1819 Make saa copy format aware,and pass formats on to the driver copy function 1820 if available. If the driver can't handle format conversions it needs to 1821 return FALSE. 1822 1823 This fixes format confusion in the copy composite fastpath. 1824 1825 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1826 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1827 1828commit f664e31d73bd31594e069d8bcca872687b84d8d9 1829Author: Thomas Hellstrom <thellstrom@vmware.com> 1830Date: Wed Dec 14 10:22:22 2011 +0100 1831 1832 saa: Always read back hw contents on poly_fill_rect 1833 1834 The damage region may be much larger than the actual area 1835 filled, so make sure we read back any hw contents in the damage region 1836 since we use it to mark sw dirty. 1837 Revisit this in the future since it may increase readbacks in an 1838 undesired way. There are workarounds. 1839 1840 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1841 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1842 1843commit 0eef22437093d29f572abea845fff9c03fca5e4e 1844Author: Thomas Hellstrom <thellstrom@vmware.com> 1845Date: Wed Dec 14 10:18:16 2011 +0100 1846 1847 saa: Only try to accelerate pixmaps that are driver pixmaps 1848 1849 Trying to accelerate scratch pixmaps doesn't work currently, 1850 and I'm not sure it's a big benefit either. 1851 1852 This is in line with what EXA does. 1853 1854 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1855 Reviewed-by: Jakob Bornecrants <jakob@vmware.com> 1856 1857commit afd2f828b8a8b2ccc6ebe4d56104db3d03d08493 1858Author: Thomas Hellstrom <thellstrom@vmware.com> 1859Date: Thu Dec 8 13:14:57 2011 +0100 1860 1861 vmwgfx: Reduce system resource usage with textured video 1862 1863 Use U and V textures and corresponding dma buffers that match 1864 the size of the input UV data rather than using textures and DMA buffers 1865 that match the Y data. This simplifies data copy and reduces the amount of 1866 image data transfered to the host by up to 50% (YV12). 1867 1868 It also reduces dma buffer- and surface memory usage correspondingly. 1869 1870 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1871 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1872 1873commit 121dba0093d24eb0aefa3d27a22f05f85ac66f72 1874Author: Thomas Hellstrom <thellstrom@vmware.com> 1875Date: Wed Dec 7 20:19:34 2011 +0100 1876 1877 vmwgfx: Hook up XV color conversion control attributes 1878 1879 This enables the standard XV color conversion control attributes 1880 contrast, scaling, saturation and hue. 1881 1882 Currently the attribute change takes effect on the next image only. 1883 If we ever implement a ReputImage callback, we can use it to update 1884 the image immediately on attribute change. 1885 1886 The attribute ranges follow those used by the proprietary Nvidia driver. 1887 1888 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1889 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1890 1891commit 8bdc6004d34e274d815bb3f3e0aa223085ea848c 1892Author: Thomas Hellstrom <thellstrom@vmware.com> 1893Date: Wed Dec 7 14:17:45 2011 +0100 1894 1895 vmwgfx: Adapt to new interface of xa_yuv_planar_blit 1896 1897 Also make it easier to modify the color conversion matrix based on 1898 yuv encoding properties. 1899 1900 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1901 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1902 1903commit dba7b865d36ebe06b54987176b1a83cdbc2f846a 1904Author: Thomas Hellstrom <thellstrom@vmware.com> 1905Date: Tue Dec 6 10:00:00 2011 +0100 1906 1907 vmwgfx: Fix MessageType members that are incorrectly typed as Bool 1908 1909 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1910 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1911 1912commit 2c438ad8b82f47ac3252403052df16655184acbd 1913Author: Thomas Hellstrom <thellstrom@vmware.com> 1914Date: Tue Dec 6 09:53:16 2011 +0100 1915 1916 vmwgfx: Add an option to enable rendercheck mode 1917 1918 Rendercheck mode is used to enable acceleration of all supported composite 1919 operations, regardless of the current data location. The mode is off by 1920 default, but could be turned on to test the hardware composite operation 1921 implementation. Due to excessive data migration, rendercheck mode will be 1922 slow. 1923 1924 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1925 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1926 1927commit 1736a76246f0a0dab01f5a1066332168d34b2803 1928Author: Thomas Hellstrom <thellstrom@vmware.com> 1929Date: Tue Dec 6 09:21:25 2011 +0100 1930 1931 vmwgfx: Adapt to the dri state tracker sending depth instead of bpp 1932 1933 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1934 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1935 1936commit fb35176958bba2f86d6d466ccb7baadcff806512 1937Author: Thomas Hellstrom <thellstrom@vmware.com> 1938Date: Mon Dec 5 20:58:45 2011 +0100 1939 1940 vmwgfx: Don't register back buffers as "hw_is_dri2_fronts" v2 1941 1942 That is reserved for dri2 front or fake front pixmaps that need 1943 immediate flushing of software contents to hardware. 1944 1945 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1946 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1947 1948commit 3c2486e05e98f2ff8f4e3d2633d91e14140a676e 1949Author: Jakob Bornecrantz <jakob@vmware.com> 1950Date: Thu Dec 1 16:47:41 2011 +0100 1951 1952 vmwgfx: Include drm.h from correct path 1953 1954 Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com> 1955 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 1956 1957commit 995d86ceb0ae8cf2904100405ba6710c89f7f1b7 1958Author: Thomas Hellstrom <thellstrom@vmware.com> 1959Date: Mon Nov 28 14:51:36 2011 +0100 1960 1961 vmwgfx: Make the drm device name persistent 1962 1963 The DRI2 code doesn't copy the device name, but assumes the storage 1964 is persistent. 1965 1966 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1967 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1968 Reviewed-by: Alan Hourihane <alanh@vmware.com> 1969 1970commit 08ca819238f2c2a0494b0bd8452393fc62907cc1 1971Author: Thomas Hellstrom <thellstrom@vmware.com> 1972Date: Fri Nov 25 14:38:45 2011 +0100 1973 1974 vmwgfx: Get the drm device name from the open file descriptor 1975 1976 Rather than hardcoding it. 1977 1978 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1979 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1980 1981commit d7576c2c50dd0f3b7f70f97d9bde4e8b859e89e5 1982Author: Thomas Hellstrom <thellstrom@vmware.com> 1983Date: Fri Nov 25 11:27:26 2011 +0100 1984 1985 vmwgfx: Use config options to optionally set GUI topology 1986 1987 Use option "StaticXinerama" or "GuiLayout" to set initial GUI topology, 1988 similar to how these options are used with the legacy driver. 1989 1990 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 1991 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 1992 1993commit 22d3e58b8810a73c4818d4d3943776a387147b83 1994Author: Thomas Hellstrom <thellstrom@vmware.com> 1995Date: Wed Nov 23 11:02:00 2011 +0100 1996 1997 vmwlegacy: Factor out code that can be common for vmwgfx and vmwlegacy 1998 1999 In this case it's the parsing of the gui / statix xinerama topology. 2000 2001 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2002 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2003 2004commit 2e8aac7031a83f507169f6fe42924c51bdaf7326 2005Author: Thomas Hellstrom <thellstrom@vmware.com> 2006Date: Wed Nov 23 10:49:33 2011 +0100 2007 2008 vmwlegacy: Accept option "GuiLayout" as a synonym for "staticXinerama" 2009 2010 It's more user-friendly and in line with what it's being used for in the 2011 vmwgfx driver. 2012 2013 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2014 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2015 2016commit 31e80b2a609631d85341025d7f2d0fb818f386f5 2017Author: Thomas Hellstrom <thellstrom@vmware.com> 2018Date: Fri Nov 25 09:42:16 2011 +0100 2019 2020 vmware: Update library dependencies 2021 2022 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2023 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2024 2025commit da29c648051208b98704860f851564a80bd3b123 2026Author: Thomas Hellstrom <thellstrom@vmware.com> 2027Date: Tue Nov 22 21:06:24 2011 +0100 2028 2029 vmwgfx: Pre-init fixes 2030 2031 A bunch of fixes that makes the fallback from vmwgfx to vmwlegacy easier. 2032 2033 1) Allow only 16 and 24 bit depths with vmwgfx. We may revisit this later. 2034 2) Try to reclaim some resources if pre_init fails. 2035 3) Print error messages so the user understands what happens on fallback. 2036 2037 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2038 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2039 2040commit e52b1505cfc4a7d0507e14c3b6335a9b9652c50f 2041Author: Thomas Hellstrom <thellstrom@vmware.com> 2042Date: Fri Nov 25 09:15:32 2011 +0100 2043 2044 vmware: Update the vmware register definition 2045 2046 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2047 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2048 2049commit 1c77e3d7434f04c59233ce7e497a949e386e1630 2050Author: Thomas Hellstrom <thellstrom@vmware.com> 2051Date: Tue Nov 22 19:56:20 2011 +0100 2052 2053 vmware: Hook up vmwgfx to be tried before falling back to legacy 2054 2055 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2056 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2057 2058commit 2aa256c96496ed6dd9dbd5880c278f4422500b0f 2059Author: Thomas Hellstrom <thellstrom@vmware.com> 2060Date: Tue Nov 22 18:37:44 2011 +0100 2061 2062 vmwlegacy: Factor out bootstrap functions 2063 2064 Preparing for choosing driver path in preinit instead of using 2065 shim loader. 2066 2067 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2068 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2069 2070commit abf9cb75630cb27bb4741d194cb23014fe3d09b1 2071Author: Thomas Hellstrom <thellstrom@vmware.com> 2072Date: Thu Nov 17 16:38:22 2011 +0100 2073 2074 vmwlegacy: Fix up Xinerama / panoramiX replies. 2075 2076 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2077 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2078 2079commit be23efbc91f47a6a34fbe24f167ee29fdf58f541 2080Author: Thomas Hellstrom <thellstrom@vmware.com> 2081Date: Wed Nov 16 10:23:10 2011 +0100 2082 2083 vmwlegacy: Fix device fifo communication 2084 2085 This fixes two issues with the device fifo communication: 2086 1) Idle the fifo before initializing it. If the fifo is already up and 2087 processing data due to an uncleanly shut down server, and init could 2088 otherwise confuse the device. 2089 2090 2) Mark fifo memory volatile when writing to it and make sure commands 2091 are written before telling the device they are available. 2092 2093 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2094 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2095 2096commit 70487eee975c63b5ed1492927bbc661425278276 2097Author: Thomas Hellstrom <thellstrom@vmware.com> 2098Date: Wed Nov 9 11:34:16 2011 +0100 2099 2100 vmwgfx: Make sure we don't render when VT switched 2101 2102 When compiz is enabled, we may try to DMA sw rendered contents of dri2 2103 drawables to surfaces while VT switched. I'm not sure how this could be 2104 happening, but queue the DMA until the next block handler. 2105 2106 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2107 Reviewed-by: Alan Hourihane <alanh@vmware.com> 2108 2109commit e1810e0b09e07bdab05a57fdef74450a507e9dfa 2110Author: Thomas Hellstrom <thellstrom@vmware.com> 2111Date: Tue Nov 8 15:28:40 2011 +0100 2112 2113 vmwgfx: Fix rendering errors when rotating / unrotating a screen. 2114 2115 The dirty trackers weren't correctly translated to 2116 pending presents / updates when a pixmap was bound as a scanout. 2117 2118 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2119 Reviewed-by: Jakob Bornecrants <jakob@vmware.com> 2120 2121commit d0b493d67afd1b5e8c8397c7938752ecd56c91ba 2122Author: Thomas Hellstrom <thellstrom@vmware.com> 2123Date: Tue Nov 8 14:49:36 2011 +0100 2124 2125 vmwgfx: A number of damage fixes 2126 2127 A damage tracker was sometimes attached to a pixmap in the middle of an 2128 operation. That meant that damage was never registered for that operation 2129 which caused dirty-tracking to fail. Instead, add a damage tracker as 2130 soon as we have storage attached to the pixmap. 2131 2132 Also, when a pixmap was resized, the dirty region trackers could become 2133 stale, and cause DMA operation extending outside of the pixmap with 2134 server crashes as a result. This commit fixes that. 2135 2136 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2137 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2138 2139commit aff799d007c8383321213ad177e5a9deec2b3978 2140Author: Thomas Hellstrom <thellstrom@vmware.com> 2141Date: Tue Nov 1 15:39:07 2011 +0100 2142 2143 vmwgfx: Add an option to use a scanout surface 2144 2145 With this option enabled, we compose all contents (both sw and hw) in 2146 the guest before presenting it. Just like the old xorg state tracker did 2147 when 3D was enabled. It's not as bad as the old xorg state tracker was for 2148 interactivity, but there's a small but noticeable lag when moving windows. 2149 2150 It's mainly intended for testing purposes, like testing the surface 2151 dirty paths in the kernel module. 2152 2153 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2154 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2155 2156commit 5137e8736edfab1aaebef4046baab37361584b2e 2157Author: Thomas Hellstrom <thellstrom@vmware.com> 2158Date: Tue Nov 1 09:59:18 2011 +0100 2159 2160 vmwgfx: Be correct when setting up damage tracking 2161 2162 When setting up damage tracking, previously the code would guess wether 2163 sw or hw was currently dirty. The caller always have that information, 2164 so pass it to the damage tracking setup function. 2165 2166 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2167 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2168 2169commit ed7bba3cda4b454eec771d3bf0ebed4fd998a82a 2170Author: Thomas Hellstrom <thellstrom@vmware.com> 2171Date: Tue Nov 1 08:38:52 2011 +0100 2172 2173 vmwgfx: Add an option for direct presents 2174 2175 With this option set to true, accelerated copies to a scanout pixmap will 2176 be performed as hardware presents instead of copies to an intermediate 2177 3D surface backing the scanout pixmap, followed by a present. 2178 2179 Depending on the application this might be a performance boost, but since 2180 it might trigger device software readbacks in other situations, for example 2181 dri2 copy front->fake_front, it might be a performance hog in other situations, 2182 so disable it by default. (Before this commit it was enabled by default). 2183 2184 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2185 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2186 2187commit b2c4a7e682e14fa8af42415d429d6735117fefcb 2188Author: Thomas Hellstrom <thellstrom@vmware.com> 2189Date: Mon Oct 31 15:14:21 2011 +0100 2190 2191 vmwgfx: Fix acceleration options. 2192 2193 Introduce relevant acceleration options and fix up the log printout. 2194 2195 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2196 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2197 Reviewed-by: Alan Hourihane <alanh@vmware.com> 2198 2199commit da85719e9e373ee8da53388cccc71a0f9c8bbefc 2200Author: Thomas Hellstrom <thellstrom@vmware.com> 2201Date: Mon Oct 31 14:40:55 2011 +0100 2202 2203 vmwgfx: Remove a couple of unused options 2204 2205 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2206 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2207 Reviewed-by: Alan Hourihane <alanh@vmware.com> 2208 2209commit b6faecc7613e8c73f0681d20a1a80e5e0ddd4058 2210Author: Thomas Hellstrom <thellstrom@vmware.com> 2211Date: Mon Oct 31 14:27:29 2011 +0100 2212 2213 vmwgfx: Don't do present readbacks per crtc 2214 2215 The kernel now takes care of doing this the right way; 2216 no need to duplicate that functionality. 2217 2218 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2219 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2220 2221commit cc7c1c961b77c139b95fbb6948204def1b4b908a 2222Author: Thomas Hellstrom <thellstrom@vmware.com> 2223Date: Mon Oct 31 10:50:09 2011 +0100 2224 2225 vmwgfx: Fall back to sw cursors if needed 2226 2227 If there is a risc that we need two simultaneous cursors, 2228 (two outputs showing the same contents, at least one of them explicit), 2229 fall back to sw cursor. 2230 2231 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2232 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2233 2234commit 427064b57c52c8881c7a64d9c9e21411e79e644b 2235Author: Thomas Hellstrom <thellstrom@vmware.com> 2236Date: Wed Oct 26 16:33:30 2011 +0200 2237 2238 vmwgfx: Fix crtc position of transformed fb 2239 2240 The fb allocated for transformed data always matches the scanout region, 2241 so set the crtc origin to (0,0) 2242 2243 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2244 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2245 2246commit 55113dae7cbe99bfb69869831536ff34ed445c73 2247Author: Thomas Hellstrom <thellstrom@vmware.com> 2248Date: Wed Oct 26 15:08:18 2011 +0200 2249 2250 vmwgfx: Hook up vmwarectrl to the gui layout ioctl 2251 2252 Requires drm 2.3. 2253 2254 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2255 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2256 2257commit c73a7f3d991dd95d332829f2ca8a88aabfea5a46 2258Author: Thomas Hellstrom <thellstrom@vmware.com> 2259Date: Wed Oct 26 07:28:31 2011 +0200 2260 2261 vmwgfx: Update the connector type name array 2262 2263 Also map connector types that are not in the array to "Unknown". 2264 2265 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2266 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2267 2268commit a33710b0d16c9c01a1995b60f0b0d4d482f459d3 2269Author: Thomas Hellstrom <thellstrom@vmware.com> 2270Date: Fri Oct 21 11:19:36 2011 +0200 2271 2272 vmwgfx: Save the dri2 requested depth since it may differ from the drawable 2273 depth. 2274 2275 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2276 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2277 2278commit 4c08c2600936efeedb1a33436876ba82a11218e7 2279Author: Jeremy Huddleston <jeremyhu@apple.com> 2280Date: Fri Oct 7 13:15:11 2011 -0700 2281 2282 Build fix for ABI version 12 2283 2284 This results in the vmware driver assuming there is only one PCI domain, 2285 which I think is true for everywhere this is expected to run anyway. 2286 2287 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 2288 2289commit 208ccbc7179b40a570a5943e325dd22740a67577 2290Author: Jeremy Huddleston <jeremyhu@apple.com> 2291Date: Fri Oct 7 13:10:59 2011 -0700 2292 2293 Fix build failures resulting from changes to the swaps and swapl macros in recent xorg-server 2294 2295 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> 2296 2297commit de70a1d0cb79dad9e52baa43306264db80a163da 2298Author: Thomas Hellstrom <thellstrom@vmware.com> 2299Date: Wed Oct 5 13:33:17 2011 +0200 2300 2301 xf86-video-vmware 11.0.99.901 2302 2303 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2304 2305commit 26845eb54a15d43f09288a87c5f74beac8fb6ec7 2306Author: Alan Coopersmith <alan.coopersmith@oracle.com> 2307Date: Tue Sep 27 19:31:22 2011 -0700 2308 2309 Use dixLookupWindow instead of LookupWindow on Xorg 1.5 & later 2310 2311 The LookupWindow function was deprecated in xserver commit ed75b056511ccb4 2312 and removed during the Xorg 1.11 merge window by commit 82a8677d9175732. 2313 2314 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 2315 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2316 2317commit 5c43481aa639053bbbf7fbf5f72f0edecf616731 2318Author: Thomas Hellstrom <thellstrom@vmware.com> 2319Date: Thu Sep 1 13:59:10 2011 +0200 2320 2321 vmwgfx: Adapt to vmwgfx kernel driver 2.1 2322 2323 This introduces fence objects with 2.0, and present / present readback 2324 ioctls with 2.1. 2325 2326 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2327 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2328 2329commit b5438423ad29338429321cd29118a66b715cdcac 2330Author: Matt Turner <mattst88@gmail.com> 2331Date: Sun Sep 25 14:06:47 2011 -0400 2332 2333 Fix wrong-sized swaps 2334 2335 Signed-off-by: Matt Turner <mattst88@gmail.com> 2336 2337commit 3c2f9cc43f7d36952e78d10d224da0ba68ecefea 2338Author: Thomas Hellstrom <thellstrom@vmware.com> 2339Date: Tue Aug 30 12:22:53 2011 +0200 2340 2341 vmwgfx-xorg: Avoid enabling unwanted outputs in initial configuration 2342 2343 Add a hack so that we avoid enabling all connected outputs during the 2344 initial configuration. On older X servers they would be enabled as cloned, 2345 which didn't really cause any problem, but on later X servers they would 2346 initially be enabled next to eachother which looks odd. 2347 2348 A RandR call will still show the disabled outputs as connected, and if there 2349 is a monitor section in the config file for the output in question, 2350 it will also have a connected status, so that it may be explicitly enabled 2351 or disabled from a config file. 2352 2353 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2354 Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> 2355 2356commit f17abaa926fdbedab1e6236e109fa746fcc2320b 2357Author: Jakob Bornecrantz <jakob@vmware.com> 2358Date: Wed Jul 13 16:30:01 2011 +0100 2359 2360 vmwgfx: Just include vmwgfx_drm.h 2361 2362 We do this in mesa as well 2363 2364 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 2365 2366commit b29d71e3adb16f2a8e3e9d534bf8b1c1f8e0d9ef 2367Author: Jakob Bornecrantz <jakob@vmware.com> 2368Date: Wed Jul 13 16:28:52 2011 +0100 2369 2370 saa: Remove superflouos headers includes 2371 2372 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 2373 2374commit 98ff64fd7a13d2824d6f36b0d8f13d95f8be9e48 2375Author: Jakob Bornecrantz <jakob@vmware.com> 2376Date: Mon Jul 11 12:23:32 2011 +0100 2377 2378 vmwgfx: Fix logic when checking for XA 2379 2380 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 2381 2382commit 858dbcdddb35b690a543a103d43dcaeb040260c1 2383Author: Thomas Hellstrom <thellstrom@vmware.com> 2384Date: Sat Jul 9 22:54:41 2011 +0200 2385 2386 vmwgfx: Don't dirty the front buffer as hw when doing a dri2copyregion from it 2387 2388 When copying from a dri2 buffer we usually dirty it as hw, since dri has 2389 been rendering to it, and there can only be hw contents in the buffer. 2390 2391 However for the real front, X has already done the dirty work for us. 2392 2393 Also remove a glxWaitX() debug message. 2394 2395 This should fix piglit read-front. 2396 2397 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2398 2399commit eea96a6f5ccc6a3a1e2bb2737f211076de443462 2400Author: Thomas Hellstrom <thellstrom@vmware.com> 2401Date: Thu Jul 7 22:02:15 2011 +0200 2402 2403 vmwgfx: Add boolean flags for different rendering modes 2404 2405 Basically we support three ways of getting data to the screen: 2406 1) Mixed mode: We mix presents and updates. 2407 2) Mixed present optimized: A version of mixed where copies to the front 2408 buffer end up as presents, saving a blit. 2409 3) 3D surface mode: We only present from a 3D surface. Software contents are 2410 first DMA'd to that surface. 2411 2412 This change adds boolean flags to the saa struct to select which mode to use. 2413 2414 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2415 2416commit 9cff0a84308723d9790c6365c0d4b1c107794995 2417Author: Thomas Hellstrom <thellstrom@vmware.com> 2418Date: Thu Jul 7 09:21:56 2011 +0200 2419 2420 vmwgfx: Various acceleration fixes. 2421 2422 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2423 2424commit 4ae8aac35b7209576eaeb64347470d3145b27832 2425Author: Thomas Hellstrom <thellstrom@vmware.com> 2426Date: Mon Jul 4 15:39:02 2011 +0200 2427 2428 vmwlegacy: Fix server termination due to pitch inconsistency. 2429 2430 Don't require a 32*32 bit pitch alignment when validating modes, since the 2431 requested virtual pitch (pVMWARE->maxWidth*bpp) must be a multiple of that 2432 alignment. If not, the server will terminate with a cryptic error message. 2433 2434 This is only for validating modes. The driver will adjust the pitch to the 2435 host requirement when a mode is set anyway, and hopefully the host won't 2436 require a pitch it doesn't support. 2437 2438 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2439 2440commit 81e7099b6bbd9fdae66a3186dfc3fff2758009cf 2441Author: Thomas Hellstrom <thellstrom@vmware.com> 2442Date: Fri Jul 1 13:23:30 2011 +0200 2443 2444 vmwgfx: Require libxatracker >= 0.4. 2445 2446 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2447 2448commit d2e27ee5c116cd914d98c2ffa4a974704c250be7 2449Author: Thomas Hellstrom <thellstrom@vmware.com> 2450Date: Fri Jul 1 12:29:50 2011 +0200 2451 2452 vmwgfx: Set up XA version checking. 2453 2454 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2455 2456commit 8a9997f060487c0aef07b8e8a95bb3a862de264f 2457Author: Thomas Hellstrom <thellstrom@vmware.com> 2458Date: Fri Jul 1 11:28:52 2011 +0200 2459 2460 vmwgfx: Try to sort out format handling with composite. 2461 2462 Try to catch all cases where we have to do readbacks or format conversions 2463 due to composite formats not being compatible with ordinary accel formats. 2464 2465 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2466 2467commit 8a21da8eaa77bb1ec55ab8b5584554ae6cc41e1e 2468Author: Thomas Hellstrom <thellstrom@vmware.com> 2469Date: Wed Jun 22 22:26:20 2011 +0200 2470 2471 vmwgfx: Support xa composite. 2472 2473 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2474 2475commit c7e8388280fc047b33be4f603afa4b7615ffc585 2476Author: Thomas Hellstrom <thellstrom@vmware.com> 2477Date: Wed Jun 22 11:33:14 2011 +0200 2478 2479 saa: Add driver composite. 2480 2481 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2482 2483commit b2e116971031d1357f614e9ce1d843768489a304 2484Author: Thomas Hellstrom <thellstrom@vmware.com> 2485Date: Wed Jun 22 10:50:35 2011 +0200 2486 2487 saa: Reuse code for computing composite bounds. 2488 2489 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2490 2491commit 3650be74d5da602653b2b1d3cf3a6bb5a9cb3539 2492Author: Thomas Hellstrom <thellstrom@vmware.com> 2493Date: Tue Jun 21 00:13:32 2011 +0200 2494 2495 vmwgfx: Try to match created hardware formats with the dri state tracker. 2496 2497commit 63d7ddb35fe8b6743de182b73e6967f97f6a7eba 2498Author: Thomas Hellstrom <thellstrom@vmware.com> 2499Date: Mon Jun 20 21:57:38 2011 +0200 2500 2501 vmwgfx: Fix dirty present bug 2502 2503 Clear dirty present areas when new contents are drawn to the backing 2504 pixmap. Not when it is actually pushed to the screen. 2505 2506 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2507 2508commit f71928d3f78d9f9e5a0fd240069b9035d4ccac75 2509Author: Thomas Hellstrom <thellstrom@vmware.com> 2510Date: Mon Jun 20 21:48:23 2011 +0200 2511 2512 vmwgfx: Don't call xa_copy_done after a present. 2513 2514 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2515 2516commit 4d5aec4a7e4512145c76014591488cb5a98ff0ea 2517Author: Thomas Hellstrom <thellstrom@vmware.com> 2518Date: Mon Jun 20 14:13:46 2011 +0200 2519 2520 vmwgfx: Make present readbacks not cross scanout bounding box boundaries 2521 2522 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2523 2524commit f6a4690832143b01d419b52a4cb74350cdfa38bf 2525Author: Thomas Hellstrom <thellstrom@vmware.com> 2526Date: Mon Jun 20 09:58:44 2011 +0200 2527 2528 vmwgfx: Make a list of scanout bounding boxes available to each scanout pixmap 2529 2530 This info is needed for present readback. 2531 2532 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2533 2534commit eea0e6e242a604b1f4d85d947cf2b3f9f4977533 2535Author: Thomas Hellstrom <thellstrom@vmware.com> 2536Date: Mon Jun 20 09:05:48 2011 +0200 2537 2538 vmwgfx: Import the WSBM list macros 2539 2540 Use WSBM list handling macros for the glxWaitX flush list. 2541 2542 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2543 2544commit eabf82fad8ad72aab071cd4e3bc9063022b0aeb3 2545Author: Thomas Hellstrom <thellstrom@vmware.com> 2546Date: Fri Jun 17 11:25:53 2011 +0200 2547 2548 vmwgfx: Check for large files 2549 2550 Fixes segfault on 32-bit servers. 2551 2552 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2553 2554commit 7966dfa47c4d2b022ebf0f25a2db3901376e6cbd 2555Author: Thomas Hellstrom <thellstrom@vmware.com> 2556Date: Fri Jun 17 08:18:13 2011 +0200 2557 2558 vmwgfx, saa: Build saa as a convenience library 2559 2560 instead of as a shared object. 2561 2562 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2563 2564commit d28613e557fcdc4fc69f9a138a8c5fc909a719f5 2565Author: Thomas Hellstrom <thellstrom@vmware.com> 2566Date: Thu Jun 16 21:21:25 2011 +0200 2567 2568 vmwgfx: Fix up driver build process for Xserver >= 1.7. 2569 2570 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2571 2572commit 84166d4b457244bcc2f5ace63702d594d602d0c2 2573Author: Thomas Hellstrom <thellstrom@vmware.com> 2574Date: Thu Jun 16 15:55:07 2011 +0200 2575 2576 vmwgfx, saa: Initial import 2577 2578 This imports the vmwgfx driver, based on the Gallium3D Xorg state tracker, 2579 as well as the saa library. A "Shadow Acceleration Architecture", which is 2580 optimized for the case where transfers between system (shadow) and hw memory 2581 is very costly. 2582 2583 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2584 2585commit 0142bb8d10edb153c9ce79a2ea3ff92a7fb15ac5 2586Author: Thomas Hellstrom <thellstrom@vmware.com> 2587Date: Wed Mar 30 12:37:24 2011 +0200 2588 2589 vmwlegacy: Don't prune modes based on the virtual size of the default mode 2590 2591 We might be pruning modes based on the virtual size of the default mode in 2592 some situations. Avoid this by allowing a virtual size equal to the device 2593 max size, unless the user has requested something else. 2594 2595 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2596 Reviewed-by: Alan Hourihane <alanh@vmware.com> 2597 2598commit 04297d435b87e3d04a427ac8ef33d314727220e6 2599Author: Thomas Hellstrom <thellstrom@vmware.com> 2600Date: Wed Mar 30 12:22:43 2011 +0200 2601 2602 vmwlegacy: Fix up default mode VRefresh anc Clock calculation 2603 2604 The VRefresh value was a factor 1000 too high. 2605 Calculate clock based on vrefresh and resolution. 2606 2607 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2608 Reviewed-by: Alan Hourihane <alanh@vmware.com> 2609 2610commit c5ca599361c87ec816db92d4f9841143f8601934 2611Author: Thomas Hellstrom <thellstrom@vmware.com> 2612Date: Wed Mar 30 11:01:06 2011 +0200 2613 2614 vmwlegacy: Fix a bug in screen dimension calculation 2615 2616 This bug resulted in incorrect screen dimensions and DPI being calculated 2617 in some circumstances, leading to among other things bad aspect ratios in xine. 2618 2619 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2620 2621commit c02d5977015983473a9bbae97ec65ce70a5faa3b 2622Author: Thomas Hellstrom <thellstrom@vmware.com> 2623Date: Tue Mar 29 13:03:39 2011 +0200 2624 2625 vmwlegacy: Make sure we draw the colorkey to the right drawable 2626 2627 Don't draw the colorkey to the screen drawable, but to the video drawable 2628 when possible. 2629 2630 Also change the Xv API/ABI test to use the builtin ABI version 2631 functions. 2632 2633 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2634 Reviewed-by: Michel D�nzer <daenzer@vmware.com> 2635 2636commit b234d7d6925b480a9c4fe23e5100e6fe7e5c0eac 2637Author: Thomas Hellstrom <thellstrom@vmware.com> 2638Date: Tue Mar 29 11:18:24 2011 +0200 2639 2640 vmwlegacy: Silence a number of warnings 2641 2642 This potentially also fixes a use of an uninitialized pointer value, which 2643 may cause OOM or segfaults. 2644 2645 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2646 Reviewed-by: Michel D�nzer <daenzer@vmware.com> 2647 2648commit cdb04224db69a6b3beb66c99879e20917b8b0fb2 2649Author: Thomas Hellstrom <thellstrom@vmware.com> 2650Date: Fri Mar 25 19:17:00 2011 +0100 2651 2652 vmwlegacy: Fix segfault and setup of initial mode. 2653 2654 The fact that modes were added without names caused segfaults on older 2655 X servers. At least up to and including Xserver 1.4. 2656 2657 Also, for some reason, at least Xserver 1.4 insists on setting the first mode 2658 in the modelist, even if we set another mode as the current one. 2659 2660 Work around this by inserting a new mode with the current screen dimensions, 2661 and add that modename last to the array of display requested mode names. 2662 This means that if none of the previous mode names are found, we will at least 2663 find the newly inserted mode. 2664 2665 Also, if there are no requested mode names at all, the driver previously 2666 chose the largest mode that fit the timings. Now we will, in that case, 2667 always select the newly inserted mode and thus not change resolution unless 2668 specified. 2669 2670 Also add an option to not add this default mode. The option 2671 "AddDefaultMode" is true by default. 2672 2673 Finally when we restore registers at exit and VT switch, make sure we 2674 reprogram the initial width, height and bpp for the next time we start a 2675 server. 2676 2677 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2678 2679commit 9a8f0af8782fd5e9de2d71dfa86fad1d48d52b64 2680Author: Thomas Hellstrom <thellstrom@vmware.com> 2681Date: Thu Mar 3 09:47:48 2011 +0100 2682 2683 Remove unused variable 2684 2685 Signed-off-by: Cyril Brulebois <kibi@debian.org> 2686 Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com> 2687 2688commit c208a5380b20f910ab49433abc0d39dce6bb5528 2689Author: Alan Hourihane <alanh@vmware.com> 2690Date: Wed Feb 16 18:53:14 2011 +0100 2691 2692 vmwlegacy: Fix mode lookup 2693 2694 In situations where we have trouble finding a specified mode, 2695 use the resolution given by the width and height device registers. 2696 2697 Signed-off-by: Alan Hourihane <alanh@vmware.com> 2698 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2699 2700commit 2e5001b7e93c63c2cd25d7d8ebdee8a420c13b96 2701Author: Thomas Hellstrom <thellstrom@vmware.com> 2702Date: Wed Feb 16 18:47:02 2011 +0100 2703 2704 vmwlegacy: Add a missing include 2705 2706 The test for Xserver >= 1.2 in the affected file was always failing. 2707 2708 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2709 Reviewed-by: Michel Dänzer <daenzer@vmware.com> 2710 2711commit b3c244ebd7899a66aa401d2d7b0ddc66ee56e6d7 2712Author: Thomas Hellstrom <thellstrom@vmware.com> 2713Date: Wed Feb 16 17:29:47 2011 +0100 2714 2715 vmwlegacy: Don't attempt to call RRSendConfigNotify on Xservers < 1.2 2716 2717 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2718 Reviewed-by: Alan Hourihane <alanh@vmware.com> 2719 Reviewed-by: Michel Dänzer <daenzer@vmware.com> 2720 2721commit eacfa46ada8562bfb15d3fc6a8af272d88036d2f 2722Author: Michel Dänzer <daenzer@vmware.com> 2723Date: Thu Jan 27 11:09:39 2011 +0100 2724 2725 vmwlegacy: Send ConfigureNotify events on Xinerama state changes with no mode change 2726 2727 Signed-off-by: Michel D�nzer <daenzer@vmware.com> 2728 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2729 2730commit cc66e4a49a4a9ac90940d7866db5bc7621cb7e16 2731Author: Thomas Hellstrom <thellstrom@vmware.com> 2732Date: Thu Jan 27 11:02:44 2011 +0100 2733 2734 Fix crosscompile build for older Xservers v2 2735 2736 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2737 2738commit d1ac343b8894a9b8a34c1dcf4b6aeb57239034c0 2739Author: Thomas Hellstrom <thellstrom@vmware.com> 2740Date: Fri Jan 21 23:07:10 2011 +0100 2741 2742 Revert "Fix crosscompile build for older Xservers" 2743 2744 This reverts commit 16e16390385d890b3666485a40369f4c690d5033. 2745 2746 The fix was bogus as pointed out by Cyril Bruleboid <kibi@debian.org>, and 2747 Gaetan Nadon <memsize@videotron.ca> 2748 2749commit 16e16390385d890b3666485a40369f4c690d5033 2750Author: Thomas Hellstrom <thellstrom@vmware.com> 2751Date: Thu Jan 13 13:17:17 2011 +0100 2752 2753 Fix crosscompile build for older Xservers 2754 2755 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2756 2757commit 5b30df47117bc65fd5f14667671041428e1d00b5 2758Author: Thomas Hellstrom <thellstrom@vmware.com> 2759Date: Thu Jan 13 12:52:14 2011 +0100 2760 2761 vmwlegacy: Remove Write-Combining setups 2762 2763 The virtual CPUs doesn't support it anyway. 2764 Once suggested by Michel Daenzer. 2765 2766 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2767 2768commit 8f8f71f0075d9d5cbbfca93425462ba03c5b8046 2769Author: Thomas Hellstrom <thellstrom@vmware.com> 2770Date: Tue Nov 9 14:31:30 2010 +0100 2771 2772 Bump package version to 11.0.3 for release 2773 2774 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2775 2776commit 30ac6eaf64aa4831025555b72b65843686a35094 2777Author: Jesse Adkins <jesserayadkins@gmail.com> 2778Date: Tue Sep 28 13:29:52 2010 -0700 2779 2780 Purge cvs tags. 2781 2782 Signed-off-by: Jesse Adkins <jesserayadkins@gmail.com> 2783 Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> 2784 2785commit 083a663bbb186bfb854eda3b9f33d7fc24252ec5 2786Author: Thomas Hellstrom <thellstrom@vmware.com> 2787Date: Thu Sep 16 10:08:12 2010 +0200 2788 2789 Make the modinfo string contain an optional subpatch number 2790 2791 The idea is that the build system assigns this number if needed. 2792 As an example it might be the commit number since the last version tag. 2793 2794 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2795 2796commit 5a2206c328caaed95daf91aff9220d3a51b492bd 2797Author: Thomas Hellstrom <thellstrom@vmware.com> 2798Date: Thu Sep 16 09:58:07 2010 +0200 2799 2800 Make sure we build on 1.4.99. 2801 2802 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2803 2804commit da0f931b90833c7f706ace24b0c9dff2672f1e97 2805Author: Thomas Hellstrom <thellstrom@vmware.com> 2806Date: Wed Sep 8 09:57:35 2010 +0200 2807 2808 vmwlegacy: Fix a segfault in vmwareCursorCloseScreen. 2809 2810 The segfault occured because the named function was using resources that 2811 were already taken down, because VMWARECloseScreen was called very late 2812 in the CloseScreen callchain. 2813 2814 Make sure we wrap the CloseScreen pointers late in ScreenInit to avoid this. 2815 2816 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2817 2818commit 03dca39afa06da04472d56ec4888f010186c6a13 2819Author: Thomas Hellstrom <thellstrom@vmware.com> 2820Date: Tue Sep 7 21:55:28 2010 +0200 2821 2822 vmwlegacy: A number of Xv fixes 2823 2824 Properly init and uninit the port private clipboxes on setup and end. 2825 Clear the port private clipboxes on video stop, to force re-painting 2826 of the colorkey when the stream is resumed. 2827 Make sure the colorkey is painted *before* the overlay is flushed. Some 2828 host implementations seem picky about this. 2829 2830 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2831 2832commit dfb062e024dfd20789ee8d43c7f94640faa5ea9f 2833Author: Thomas Hellstrom <thellstrom@vmware.com> 2834Date: Tue Sep 7 12:17:46 2010 +0200 2835 2836 vmware/vmwlegacy: Silence compilation errors and -warnings. 2837 2838 Silence warnings and errors on various server versions due to incorrect 2839 usage of libc functions. 2840 2841 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2842 2843commit 12e4577c020c6905d09459d833fd1c5313f1647b 2844Author: Thomas Hellstrom <thellstrom@vmware.com> 2845Date: Fri Sep 3 11:19:01 2010 +0200 2846 2847 vmwlegacy: Always scale screen dimensions to the initial DPI on modeswitch. 2848 2849 When we switch mode we usually alter the size in a constant DPI environment 2850 rather than keeping the size constant and alter the DPI. 2851 2852 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2853 2854commit 81a0a1a37e703296d109022bb76a1291e84064fd 2855Author: Thomas Hellstrom <thellstrom@vmware.com> 2856Date: Wed Sep 1 07:58:18 2010 +0200 2857 2858 vmware: Handle chain-loader failures gracefully. 2859 2860 If the vmwlegacy driver wasn't present when expected, that would 2861 leave the X server in a confused state. Thus report loading errors properly 2862 back to the X server. 2863 2864 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2865 2866commit 2a72ea7a1b1519451350e0741d145f513124a208 2867Author: Gaetan Nadon <memsize@videotron.ca> 2868Date: Wed Aug 18 10:57:02 2010 -0400 2869 2870 vmwarectrl: rework standalone package as a subdir of driver 2871 2872 vmwarectrl is a unit test client tool for developers. 2873 By default, it is not built unless --enable-vmwarectrl-client is used. 2874 2875 Acked-by: Thomas Hellstrom <thellstrom@vmware.com> 2876 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 2877 2878commit f51ca2cb756dcd541a27f0326496f0aadab3f214 2879Author: Thomas Hellstrom <thellstrom@vmware.com> 2880Date: Wed Aug 18 11:34:49 2010 +0200 2881 2882 Make the driver build and run without libdrm. 2883 2884 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2885 2886commit 2e6eca3e7bb19ea69db2cc367c9704c7d657db79 2887Author: Thomas Hellstrom <thellstrom@vmware.com> 2888Date: Thu Aug 12 17:43:59 2010 +0200 2889 2890 Only set the first xinerama output as primary. 2891 2892 Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> 2893 2894commit 456b74e8778ab0c44e8a70aa45f447ff5c562be3 2895Author: Adam Jackson <ajax@redhat.com> 2896Date: Tue Aug 10 11:02:05 2010 -0400 2897 2898 xf86-video-vmware 11.0.2 2899 2900 Signed-off-by: Adam Jackson <ajax@redhat.com> 2901 2902commit c29d24b2ca598011b84387006f03a9360910707f 2903Author: Gaetan Nadon <memsize@videotron.ca> 2904Date: Fri Jul 23 21:02:33 2010 -0400 2905 2906 config: add AM_PROG_CC_C_O for per-target compilation flags 2907 2908 The use of per-target compilation flags with C sources requires 2909 that the macro AM_PROG_CC_C_O be called from configure.ac 2910 2911 If the C compiler does not accept the -c and -o options simultaneously, 2912 define NO_MINUS_C_MINUS_O. This macro actually tests both the compiler 2913 found by AC_PROG_CC, and, if different, the first cc in the path. 2914 The test fails if one fails. This macro was created for GNU Make 2915 to choose the default C compilation rule. 2916 2917 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 2918 2919commit dfbef04f1ff9430a01ca0ca682be32a8606e9e13 2920Author: Gaetan Nadon <memsize@videotron.ca> 2921Date: Wed Jul 21 16:49:04 2010 -0400 2922 2923 config: add comments for main statements 2924 2925commit 0dbd97fe7d411549aa343c25dab9c70087532727 2926Author: Gaetan Nadon <memsize@videotron.ca> 2927Date: Wed Jul 21 16:07:00 2010 -0400 2928 2929 config: replace deprecated use of AC_OUTPUT with AC_CONFIG_FILES 2930 2931 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 2932 2933commit dfaa0b6391ebc32f4e81d8add053eabea44d0881 2934Author: Gaetan Nadon <memsize@videotron.ca> 2935Date: Wed Jul 21 14:37:42 2010 -0400 2936 2937 config: replace deprecated AC_HELP_STRING with AS_HELP_STRING 2938 2939 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 2940 2941commit dd0bb4ac5730331070eac2c39880718eeff3760b 2942Author: Gaetan Nadon <memsize@videotron.ca> 2943Date: Wed Jul 21 14:05:23 2010 -0400 2944 2945 config: replace deprecated AM_CONFIG_HEADER with AC_CONFIG_HEADERS 2946 2947 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 2948 2949commit f07d572611ba072c5952fea526e7178202d36685 2950Author: Gaetan Nadon <memsize@videotron.ca> 2951Date: Wed Jul 21 09:27:42 2010 -0400 2952 2953 config: complete AC_INIT m4 quoting 2954 2955 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 2956 2957commit 79956ca804b4ca6227f878cf794edafefa00d25a 2958Author: Gaetan Nadon <memsize@videotron.ca> 2959Date: Tue Jul 20 21:44:58 2010 -0400 2960 2961 config: remove unrequired AC_SUBST([XORG_CFLAGS]) 2962 2963 This macro is called by PKG_CHECK_MODULES 2964 2965 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 2966 2967commit cee7a37a38f83cd5300a19f17493e43eb532e619 2968Author: Gaetan Nadon <memsize@videotron.ca> 2969Date: Tue Jul 20 20:24:42 2010 -0400 2970 2971 config: remove unrequired AC_HEADER_STDC 2972 2973 Autoconf says: 2974 "This macro is obsolescent, as current systems have conforming 2975 header files. New programs need not use this macro". 2976 2977 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 2978 2979commit 8adb672ac3f754d02e377745c16f64151b50522a 2980Author: Gaetan Nadon <memsize@videotron.ca> 2981Date: Tue Jul 20 19:41:31 2010 -0400 2982 2983 config: remove AC_PROG_CC as it overrides AC_PROG_C_C99 2984 2985 XORG_STRICT_OPTION from XORG_DEFAULT_OPTIONS calls 2986 AC_PROG_C_C99. This sets gcc with -std=gnu99. 2987 If AC_PROG_CC macro is called afterwards, it resets CC to gcc. 2988 2989 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 2990 2991commit d867b84e425cdf66542eebaabd8adb95b8121960 2992Author: Gaetan Nadon <memsize@videotron.ca> 2993Date: Tue Jul 20 18:45:19 2010 -0400 2994 2995 config: update AC_PREREQ statement to 2.60 2996 2997 Unrelated to the previous patches, the new value simply reflects 2998 the reality that the minimum level for autoconf to configure 2999 all x.org modules is 2.60 dated June 2006. 3000 3001 ftp://ftp.gnu.org/gnu/autoconf/autoconf-2.60.tar.gz 3002 3003 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 3004 3005commit 190a6b7e032e37e33dda42482038b8fd1fd9d8af 3006Author: Gaetan Nadon <memsize@videotron.ca> 3007Date: Tue Jul 20 16:15:30 2010 -0400 3008 3009 config: upgrade to util-macros 1.8 for additional man page support 3010 3011 Use MAN_SUBST now supplied in XORG_MANPAGE_SECTIONS 3012 The value of MAN_SUBST is the same for all X.Org packages. 3013 3014commit e647511e4b9b924b0617171c56b3bb485674b3ff 3015Author: Julien Cristau <jcristau@debian.org> 3016Date: Tue Jul 13 11:57:15 2010 +0100 3017 3018 Don't crash the server on Xorg -configure 3019 3020commit 0bd99bb142093c0072147ec3bb665864b4cadef5 3021Author: Fernando Carrijo <fcarrijo@yahoo.com.br> 3022Date: Thu Jul 1 07:18:11 2010 -0300 3023 3024 Purge macros NEED_EVENTS and NEED_REPLIES 3025 3026 Signed-off-by: Fernando Carrijo <fcarrijo@yahoo.com.br> 3027 Acked-by: Tiago Vignatti <tiago.vignatti@nokia.com> 3028 Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> 3029 3030commit f307f77a34f3a6777f74458819f6a3519c2c3f2b 3031Author: Gaetan Nadon <memsize@videotron.ca> 3032Date: Sun Jun 13 13:10:55 2010 -0400 3033 3034 COPYING: replace stub file with actual Copyright notices. 3035 3036 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 3037 3038commit 423d8a069a2d86068a8f22926a73bc965670315d 3039Author: Michel Dänzer <daenzer@vmware.com> 3040Date: Mon Mar 29 10:19:04 2010 +0200 3041 3042 Fix some issues pointed out by compiler warnings. 3043 3044 * vmware_chain_module() doesn't return any values. 3045 * Remove an unused local variable. 3046 3047commit cc77a36048863640fb1fa9f82b5b0dbf41326872 3048Author: Jakob Bornecrantz <jakob@vmware.com> 3049Date: Thu Mar 18 15:30:36 2010 +0100 3050 3051 Bump for 11.0.1 release. 3052 3053commit a5fb3698a033bfe7789641fd6719d8d4fb054201 3054Author: Jakob Bornecrantz <jakob@vmware.com> 3055Date: Tue Mar 9 08:37:11 2010 +0000 3056 3057 Rework chainloader code to check for vmwgfx userspace driver as well 3058 3059commit bf18be6f458a4612b2ebdd8d2b5894f8884891e4 3060Author: Roland Scheidegger <sroland@vmware.com> 3061Date: Tue Mar 9 16:10:25 2010 +0100 3062 3063 fix a cursor refcounting bug, leading to segfaults 3064 3065 this is similar to what xf86_use_hw_cursor() does, which is replaced by 3066 vmwareUseHWCursor (otherwise, the refcount could reach zero and hence the 3067 cursor deallocated while xf86CursorEnableDisableFBAccess() could still bring 3068 it back to life from the saved cursor). 3069 It is probably insane to do refcounting here, but this needs a xserver fix, 3070 and even if that's fixed this fix here shouldn't hurt (though would be 3071 unnecessary). 3072 3073commit 257614ae9bea54d6a46e4477496500a84853ee37 3074Author: Roland Scheidegger <sroland@vmware.com> 3075Date: Tue Mar 9 16:03:59 2010 +0100 3076 3077 make DEBUG_LOGGING compile 3078 3079commit 29f6a2dca3e680908e938767256c9b995653ca7f 3080Author: Jakob Bornecrantz <jakob@vmware.com> 3081Date: Sun Mar 7 15:02:39 2010 +0000 3082 3083 Print text about none fatal error message with Fatal in it 3084 3085commit 204504861a2847734a04bd976accb0c730733523 3086Author: Jakob Bornecrantz <jakob@vmware.com> 3087Date: Fri Feb 26 16:27:17 2010 +0100 3088 3089 Don't hardcode the module version in the source 3090 3091 As pointed out by Julien Cristau XORG_RELEASE_VERSION 3092 gives us that info from configure.ac. 3093 3094 Signed-off-by: Jakob Bornecrantz <jakob@vmware.com> 3095 3096commit 667263d9410bd0d5832cf08a2a28841fe6326b84 3097Author: Jakob Bornecrantz <jakob@vmware.com> 3098Date: Thu Feb 18 14:11:32 2010 +0100 3099 3100 Bump major for new chainloading driver 3101 3102commit 048fe839f0d761aeba95c324f14bdc5488e9e634 3103Author: Jakob Bornecrantz <jakob@vmware.com> 3104Date: Mon Feb 22 13:20:06 2010 +0100 3105 3106 Link to drm but on for the vmware chainloader driver 3107 3108commit 30964ffa31f4fdbffe563b7d7d20bfbf1ed22969 3109Author: Jakob Bornecrantz <jakob@vmware.com> 3110Date: Thu Feb 18 15:20:31 2010 +0100 3111 3112 Make sure that modinfo strings are unique 3113 3114commit 0f2b02d72e36dc6390c09d6adba13e4101bf3df8 3115Author: Jakob Bornecrantz <jakob@vmware.com> 3116Date: Thu Feb 11 23:18:37 2010 +0100 3117 3118 Add modinfo for vmware_drv.so 3119 3120commit c4f5bf8ea45b80c2ac4a5eec65ff58f148fb2807 3121Author: Jakob Bornecrantz <jakob@vmware.com> 3122Date: Thu Feb 11 22:04:53 2010 +0100 3123 3124 Add a chain loading module to load new vmwgfx driver if kernel module is loaded 3125 3126commit 0d9d1724dbe113dcc02736a8ca80ab540057cb5e 3127Author: Alan Coopersmith <alan.coopersmith@sun.com> 3128Date: Fri Jan 15 14:13:42 2010 -0800 3129 3130 Update Sun license notices to current X.Org standard form 3131 3132 Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com> 3133 3134commit e0d370ae5f307d96e9dd9384cc0ea66a0f6c06c1 3135Author: Gaetan Nadon <memsize@videotron.ca> 3136Date: Tue Dec 15 22:01:02 2009 -0500 3137 3138 configure.ac: remove unused sdkdir=$(pkg-config...) statement 3139 3140 The sdkdir variable isn't use, so remove the statement. 3141 3142 Acked-by: Dan Nicholson <dbn.lists@gmail.com> 3143 3144 Signed-off-by: Gaetan Nadon <memsize@videotron.ca> 3145 3146commit 991830e39c34d0a4530a5c5023ec92cf0d7395b0 3147Author: Michel Dänzer <daenzer@vmware.com> 3148Date: Tue Dec 29 17:53:35 2009 +0100 3149 3150 Bump for 10.16.9 release. 3151 3152commit 5a36fcb6a5899e5cfdf6d65e05e3f999d5d25b2a 3153Author: Michel Dänzer <daenzer@vmware.com> 3154Date: Tue Dec 29 17:00:45 2009 +0100 3155 3156 VMWAREComposite: Only hide cursor if source picture is backed by drawable. 3157 3158 Should fix https://bugs.freedesktop.org/show_bug.cgi?id=14208 . 3159 3160commit 4bec18ab8dffd90ab84b986de2136a9d848d3a4a 3161Author: Michel Dänzer <daenzer@vmware.com> 3162Date: Tue Dec 29 16:41:16 2009 +0100 3163 3164 vmware: Fix crash with xserver >= 1.7. 3165 3166 Fixes https://bugs.freedesktop.org/show_bug.cgi?id=24541 . 3167 3168commit 74c830e61bdf700c7e422aaa5b1a0136561295dd 3169Author: Gaetan Nadon <memsize@videotron.ca> 3170Date: Mon Nov 23 09:25:06 2009 -0500 3171 3172 Makefile.am: add ChangeLog and INSTALL on MAINTAINERCLEANFILES 3173 3174 Now that the INSTALL file is generated. 3175 Allows running make maintainer-clean. 3176 3177commit e8f5a924226aacc280923aed661b5a821fdaffe8 3178Author: Gaetan Nadon <memsize@videotron.ca> 3179Date: Wed Oct 28 14:41:41 2009 -0400 3180 3181 INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206 3182 3183 Automake 'foreign' option is specified in configure.ac. 3184 Remove from Makefile.am 3185 3186commit edb6fcb49bae735c45d9565f7aa8313396cf3b15 3187Author: Gaetan Nadon <memsize@videotron.ca> 3188Date: Wed Oct 28 14:09:10 2009 -0400 3189 3190 INSTALL, NEWS, README or AUTHORS files are missing/incorrect #24206 3191 3192 Add missing INSTALL file. Use standard GNU file on building tarball 3193 README may have been updated 3194 Remove AUTHORS file as it is empty and no content available yet. 3195 Remove NEWS file as it is empty and no content available yet. 3196 3197commit dcb3d3ed2ba2129b433eee0d18ec86e1e0a6f21f 3198Author: Gaetan Nadon <memsize@videotron.ca> 3199Date: Mon Oct 26 12:54:23 2009 -0400 3200 3201 Several driver modules do not have a ChangeLog target in Makefile.am #23814 3202 3203 The git generated ChangeLog replaces the hand written one. 3204 Update configure.ac to xorg-macros level 1.3. 3205 Use XORG_DEFAULT_OPTIONS which replaces four XORG_* macros 3206 Update Makefile.am to add ChangeLog target if missing 3207 Remove ChangeLog from EXTRA_DIST or *CLEAN variables 3208 This is a pre-req for the INSTALL_CMD 3209 3210commit a9dc11de80966fe00efb6c061a6df00678f80b1f 3211Author: Gaetan Nadon <memsize@videotron.ca> 3212Date: Thu Oct 22 13:06:19 2009 -0400 3213 3214 .gitignore: use common defaults with custom section # 24239 3215 3216 Using common defaults will reduce errors and maintenance. 3217 Only the very small or inexistent custom section need periodic maintenance 3218 when the structure of the component changes. Do not edit defaults. 3219 3220commit 082d1689a8e017d29b108ed3a251141ddf32a35d 3221Author: Gaetan Nadon <memsize@videotron.ca> 3222Date: Thu Oct 22 12:34:18 2009 -0400 3223 3224 .gitignore: use common defaults with custom section # 24239 3225 3226 Using common defaults will reduce errors and maintenance. 3227 Only the very small or inexistent custom section need periodic maintenance 3228 when the structure of the component changes. Do not edit defaults. 3229 3230commit 6c8fc7b280295b823675285ed78b0fb2365bcf07 3231Author: Philip Langdale <philipl@fido2.homeip.net> 3232Date: Tue Sep 8 22:21:33 2009 -0700 3233 3234 Bump fo 10.16.8 Release 3235 3236commit e3e3d8d2488bd289660943e9038bdbe29e11cf77 3237Author: Gaetan Nadon <memsize@videotron.ca> 3238Date: Mon Sep 7 11:12:13 2009 -0400 3239 3240 xf86-video-vmware: Remove unused .cvsignore file #23776 3241 3242 Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> 3243 3244commit cce33db3f95c6f3c8537c782063b6ca9ad68f332 3245Author: Dave Airlie <airlied@redhat.com> 3246Date: Tue Jul 28 15:22:42 2009 +1000 3247 3248 vmware: change to using ABI version check 3249 3250commit 0fddce5511b761206bb5f59dcd17652f15bda5ef 3251Author: Dave Airlie <airlied@redhat.com> 3252Date: Tue Jul 28 14:11:59 2009 +1000 3253 3254 vmware: fix resource removal properly 3255 3256commit 4605e8d3a99ddcd75314051bd00b51d8ecee06a1 3257Author: Dave Airlie <airlied@redhat.com> 3258Date: Tue Jul 28 13:32:41 2009 +1000 3259 3260 vmware: update for resources/RAC API removal 3261 3262commit 382c25f53870da0db0ee74f343c20c312e6b03af 3263Author: Philip Langdale <philipl@fido2.homeip.net> 3264Date: Mon Jul 20 18:03:47 2009 -0700 3265 3266 Bump for 10.16.7 Release. 3267 3268commit c2d9678431a40f6a15dc7f50690507cdda8d11e2 3269Author: Micah Dowty <micah@vmware.com> 3270Date: Mon Jun 1 10:47:51 2009 -0700 3271 3272 An imperfect fix for Xinerama state changes without a mode change 3273 3274 This patch improves behaviour for Xinerama state changes (via the 3275 VMWARE_CTRL) extension that don't have an accompanying mode change. 3276 This will be the case if a new Xinerama monitor layout has a bounding 3277 box with an identical size to that of the previous layout. 3278 3279 Prior to this patch, the behaviour was pretty bad. If you sent two 3280 Xinerama states with the same bounding box, the second state would 3281 be set as pending but no actual mode change would occur, because 3282 the X server would already be in the right video mode. This means 3283 that the pending mode stays pending. 3284 3285 If another Xinerama state comes in after this, we would hit our 3286 "Aborting due to existing pending state" error, and the new state 3287 would be discarded. This means we'd drop the mode switch on the 3288 floor, plus we'd lie to the client and say it worked. 3289 3290 One example of the user-visible symptoms from this: The user has 3291 four monitors of the same size. We'll call them A through D. 3292 The VM goes into full-screen mode, and they set it to use screens 3293 ABC. Now they switch to BCD. These have the same bounding box size, 3294 so no mode change occurs and a topology is still pending. Now they 3295 switch to monitors BC. This mode switch is dropped, so the guest 3296 is still in the ABC topology and the mode is too wide for BC. 3297 3298 This patch is an incomplete fix. If we're setting a new topology 3299 with the same bounding box, we'll flush the Xinerama state 3300 immediately since we know the mode switch will never occur. This 3301 means we don't get stuck with xineramaNextState set when it 3302 shouldn't be, and we don't have the problem with dropping 3303 subsequent mode changes. We also do set the new Xinerama state, 3304 so apps that query it will see the updated state immediately. 3305 3306 But the fix isn't perfect- as far as I can tell, there's no way 3307 to notify applications that the monitor layout changed without 3308 a mode switch. So even though we've set the new topology, most 3309 apps won't notice. There are ways we could hack around this, 3310 but none of them are pretty. 3311 3312commit d10841493c4707f23f928d7580bc5bddb51d22a6 3313Author: Bob Long <blong@vmware.com> 3314Date: Fri May 22 15:48:55 2009 -0700 3315 3316 Fix black screen after resume from hibernate. 3317 3318 The root cause for the black screen and system lock up is 3319 caused by not recovering the SVGA ID register after hibernation. 3320 Incorrect ID register value will invalidate the FIFO memory start 3321 register, and driver will not retrieve correct FIFO memory start 3322 address and the busy read of svga FIFO sync register will lock up 3323 the whole system. 3324 3325 Currently SVGA Xorg driver does not have a kernel module to handle 3326 the power management event, but Xorg will call driver provided 3327 LeaveVT before shutting down system and call EnterVT after resuming 3328 system from hibernation, so these two callback functions are good 3329 entry points to save and restore the ID register value. This patch 3330 saves the ID register value in LeaveVT and restores the value to 3331 SVGA ID register in EnterVT. 3332 3333commit b4ea3052f843c2d1c285252cbf1bed2f4857f96c 3334Author: Philip Langdale <philipl@fido2.homeip.net> 3335Date: Tue May 12 16:48:43 2009 -0700 3336 3337 Bump for 10.16.6 release 3338 3339commit 8e15f6669ff2cb5bf4260ac87a481a4e38044b26 3340Author: Micah Dowty <micah@vmware.com> 3341Date: Tue May 12 16:46:39 2009 -0700 3342 3343 Better cursor size limit and explanation 3344 3345 Increase the cursor size limit to 64x64, and give a 3346 better explanation of the host's cursor size limits. 3347 3348commit bfa3dfc27b05d4a2deff230f8241bd44f72fb7a0 3349Author: Micah Dowty <micah@vmware.com> 3350Date: Tue May 12 16:46:00 2009 -0700 3351 3352 Allow cursor updates while unhidden 3353 3354 This change just adds a flag to our hardware cursor, 3355 telling Xorg that it doesn't need to hide the cursor 3356 when updating its shape. This fixes the cursor flicker 3357 in X11. 3358 3359commit dccc9376a4fb1cba9c35b7617989608497fca7be 3360Author: Micah Dowty <micah@vmware.com> 3361Date: Tue May 12 16:45:29 2009 -0700 3362 3363 Unbreak vmwarectrl setres 3364 3365 The vmwarectrl tool's "setres" command was unusable, 3366 because it looks like someone added the settopology 3367 test without updating the argument indices for setres. 3368 This patch makes setres usable again. 3369 3370commit b7dbdd28764a8f3883833ab818a7b7314632b0b2 3371Author: Micah Dowty <micah@vmware.com> 3372Date: Tue May 12 16:44:42 2009 -0700 3373 3374 Fix dynamic mode edge cases 3375 3376 The VMware Xorg driver supports dynamic modelines that can be set from 3377 userspace via an X extension. These are used to implement VM features 3378 which need to automatically change the resolution of the guest OS. 3379 3380 This driver implements the feature using two modelines. The driver 3381 would alternately update one mode then the other, so that in typical 3382 usage one mode is current and the other is available for the next mode 3383 switch. 3384 3385 This usually worked, but there were many edge cases that could cause 3386 this alternating pattern to get 'out of sync', so we'd end up changing 3387 the resolution of the current video mode. This could end up putting 3388 the X server in a state where the screen resolution has been changed, 3389 but the hardware was never reprogrammed for the new resolution. 3390 3391 This patch fixes the problem by explicitly searching for a dynamic 3392 mode that isn't currently in use. We no longer rely on the alternating 3393 pattern. 3394 3395commit cfe8793180ec633dd7a17d059ad882ef461ed1d9 3396Author: Micah Dowty <micah@vmware.com> 3397Date: Tue May 12 16:43:13 2009 -0700 3398 3399 Update README 3400 3401 Updates the copyright date, and replaces the rather out-of-date 3402 2D documentation with a link to the updated 2D and 3D docs on 3403 Source Forge. 3404 3405commit e3769142d80953d6da484eb979f5274c8a3abeb3 3406Author: Shelley Gong <shelleygong@vmware.com> 3407Date: Thu Apr 16 13:28:47 2009 -0700 3408 3409 Automatically add modelines for the driver's built-in set of modes. 3410 3411 The driver has had a built-in set of modes for a while, but there 3412 was nothing adding modelines to back them up, causing initial modes 3413 to be rejected at startup with certain Xorg versions. 3414 3415 This change adds the actual modelines for sufficiently new versions 3416 of the server (>= 1.2), as the necessary calls were only introduced 3417 at that time. 3418 3419commit 3c223e8f7b03e2d7f8c31faeeeeb37030c461176 3420Author: Alan Coopersmith <alan.coopersmith@sun.com> 3421Date: Fri Jan 9 16:39:07 2009 -0800 3422 3423 Remove xorgconfig & xorgcfg from See Also list in man page 3424 3425commit 1bbef3aa7ab15ee93cd4cd47c3d484ac91f0440d 3426Author: Philip Langdale <philipl@fido2.homeip.net> 3427Date: Tue Aug 19 11:23:44 2008 -0700 3428 3429 Bump version for 10.16.5 release. 3430 3431commit 6ea8e5000519865dd936cfe82d67efc7d107a28c 3432Author: Bankim Bhavsar <bbhavsar@vmware.com> 3433Date: Tue Aug 19 11:23:02 2008 -0700 3434 3435 Add RegionEqual function for older XFree86 versions. 3436 3437 Fixes bug : http://bugzilla.eng.vmware.com/show_bug.cgi?id=312853 3438 3439 When we added AUTOPAINT_COLORKEY capability to our VMware video driver, 3440 region functions were used to keep track of colorkey painting. 3441 3442 REGION_EQUAL was one of them. 3443 3444 Unfortunately REGION_EQUAL was not present in regionstr.h shipped with XFree86 version 3445 4.3.0. 3446 This version is used by TurboLinux 10; causing X server to crash while playing videos. 3447 3448 REGION_EQUAL was added in revision 1.8 of regionstr.h and available for xfree86 version 3449 4.3.99 3450 onwards. 3451 Reference: 3452 http://cvsweb.xfree86.org/cvsweb/xc/programs/Xserver/include/regionstr.h.diff?r1=1.7&r2=1.8 3453 3454 When I compiled the existing code(without my change), I see a warning was generated 3455 indicating REGION_EQUAL is not present. 3456 Too bad we missed it. 3457 3458 This patch includes 3459 1) Slightly modified version of miRegionEqual from miRegion.c 3460 2) Some formating cleanup. 3461 3462commit 0576b87c27d3d885cc698a3cc013bcfa4021942c 3463Author: Philip Langdale <philipl@fido2.homeip.net> 3464Date: Mon Jul 21 15:26:01 2008 -0700 3465 3466 Prepare for 10.16.4 release. 3467 3468commit ad8f17dec71632d0e606af1bea9432ece0bc97db 3469Author: Philip Langdale <philipl@fido2.homeip.net> 3470Date: Mon Jul 21 15:23:50 2008 -0700 3471 3472 Video Overlay: Handle clipping correctly with AutoPaint colorkey. 3473 3474 We recently added XV_AUTOPAINT_COLORKEY attribute to the X video driver 3475 to enable applications like Real player that rely on this attribute to 3476 display video frames in Linux guest. 3477 3478 When this attribute is SET, we paint the colorkey on 1st frame and from 3479 there on only when the video is moved. 3480 3481 This introduced a bug 305202 with clipping. 3482 3483 Consider a case when the video is playing, obscure the player window 3484 with another window, without moving the player window, move the other 3485 window away. The part of the window that was obscured didn't display 3486 the video. 3487 3488 With this patch instead of relying on the target rectangle, we use 3489 the clipBoxes supplied in every frame. 3490 3491commit 71f1ca515cd1b3d92397682f4fcea120c392b26e 3492Author: Philip Langdale <philipl@fido2.homeip.net> 3493Date: Fri Jul 18 14:03:09 2008 -0700 3494 3495 Prepare for 10.16.3 3496 3497 Bump up version numbers. 3498 3499commit e7f404938b7ccd4c7e61c71c6c182ca8ee8d875f 3500Author: Philip Langdale <philipl@fido2.homeip.net> 3501Date: Fri Jul 18 13:59:34 2008 -0700 3502 3503 Video Overlay: Add XV_AUTOPAINT_COLORKEY capability to the Xv extension of Linux guest driver. 3504 3505 Applications use XV_AUTOPAINT_COLORKEY capability to let the driver handle the painting 3506 of colorkey. 3507 3508 Real Player assumes this capability as ON by default and hence prior to this patch videos 3509 didn't play with Real player. 3510 3511 With this patch: 3512 a) If AUTOPAINT_COLORKEY is SET then the driver handles the painting of the colorkey. 3513 Painting is done on the 1st frame and from there on only if the video frame moves. 3514 3515 b) Adds UYVY fourcc format to keep it consistent with Windows SVGA driver and the host backends. 3516 3517 c) Changes the default colorkey from a pungent GREEN to a darker shade. 3518 Real Player sets this colorkey and it looks better when the video is moved around. 3519 3520commit 0705e7093e42f7a5ae909477150fd5bb416f8da0 3521Author: Vinay Bondhugula <vinayb@vmware.com> 3522Date: Wed Jul 2 21:32:44 2008 -0400 3523 3524 Trivial fix: Delete ASSERT() statement 3525 3526 Delete the ASSERT() statement that got introduced in the previous check-in. 3527 3528commit f4b5158960c3402e18a0fcf0e0211b68b0f0f127 3529Author: Vinay Bondhugula <vinayb@vmware.com> 3530Date: Wed Jul 2 18:17:58 2008 -0400 3531 3532 Fix video overlay offscreen allocation bug 3533 3534 When Kaffeine player stops a video, it sets the cleanup flag to FALSE and may 3535 start playing the next video. In its present state the driver does not check or 3536 clean up the offscreen area. This is a bug as the newer video might need more 3537 offscreen space for its frame. The fix is to check for the offscreen frame size 3538 in videoPlay and restart the stream if necessary. 3539 3540commit b53bb2ad4c6f6e378cbf82a98cac917736d62740 3541Author: Vinay Bondhugula <vinayb@vmware.com> 3542Date: Thu Jun 12 17:34:30 2008 -0400 3543 3544 Prepare for 10.16.2 3545 3546 Bump up version numbers. 3547 3548commit 1f7a8390a32852d2ff0039a7d8138a292d17b50a 3549Author: Vinay Bondhugula <vinayb@vmware.com> 3550Date: Tue Apr 29 18:29:11 2008 -0400 3551 3552 Fix VMWARE_CTRL and Xinerama Initialization 3553 3554 Move initialization code for the vmware control and xinerama extensions to 3555 VMWAREScreenInit(), so that auto-resize and multi-mon work fine after a user 3556 logs out and logs in again in a graphical display manager. 3557 3558commit db027200474684c0aa4b9387f3ed53b2e5f24478 3559Author: Vinay Bondhugula <vinayb@vmware.com> 3560Date: Tue Apr 22 14:18:45 2008 -0400 3561 3562 More xorg and xfree86 backwards compatibility 3563 3564 Push most of the xorg-server version checking into the configure script. With 3565 this change, unmodified driver source compiles in the XFree86 4.3 tree. 3566 3567commit 6849c0a5fbf10722ca475d2ffcce7d0d95d2377a 3568Author: Vinay Bondhugula <vinayb@vmware.com> 3569Date: Fri Apr 18 18:00:38 2008 -0400 3570 3571 Change // comments to /* 3572 3573 This will make it easier to compile the driver against ancient versions of X. 3574 3575commit b6c59761e2b984be32493f35224db523dd62bcd8 3576Author: Paulo Cesar Pereira de Andrade <pcpa@mandriva.com.br> 3577Date: Thu Apr 17 23:04:45 2008 -0400 3578 3579 Compile warning fixes 3580 3581 Major problem was prototype vmwareInitVideo not matching implementation 3582 vmwareVideoInit. Remaining are adding an "ansification" of a function without 3583 arguments, and removing/disabling unused variables/functions. 3584 3585commit 7870dc715b1efe173af1e2664778e2066508e066 3586Author: Vinay Bondhugula <vinayb@vmware.com> 3587Date: Thu Apr 17 20:54:23 2008 -0400 3588 3589 Bump version to 10.16.1 in configure.ac 3590 3591 The previous change didn't bump up the version in configure.ac. 3592 3593commit be75c66fa7933c7dfa86e5514c744f47734c5fc8 3594Author: Vinay Bondhugula <vinayb@vmware.com> 3595Date: Thu Apr 17 20:39:02 2008 -0400 3596 3597 Prepare for 10.16.1 3598 3599 Update the modinfo section and make a configure.ac fix. 3600 3601commit 87e9b32618f037d78d7853b48e40a71533b764d2 3602Author: Vinay Bondhugula <vinayb@vmware.com> 3603Date: Thu Apr 17 19:16:48 2008 -0400 3604 3605 Fix build for all modular versions of Xorg 3606 3607 This change makes the video driver compile with Xorg 7.0. There are a couple of 3608 trivial changes that bring down the maximum Xserver version down to 2.0. Hacky, 3609 but good enough for now. 3610 3611commit e4fa13951a7c5dd53005a63392a439e2329fab9c 3612Author: Vinay Bondhugula <vinayb@vmware.com> 3613Date: Wed Mar 26 21:32:47 2008 -0400 3614 3615 Send down guest display topology 3616 3617 Send down the Xinerama topology from the guest down to the host, by setting 3618 display topology registers in the SVGA device. 3619 3620commit 5aced9e39a0bf7590c841824c0b66060eb7d5e03 3621Author: Adam Jackson <ajax@redhat.com> 3622Date: Wed Mar 19 17:03:18 2008 -0400 3623 3624 vmware 10.16.0 3625 3626commit cd8bb3018946e82b98207984efc26b13ac260e5e 3627Author: Adam Jackson <ajax@redhat.com> 3628Date: Wed Mar 19 17:00:41 2008 -0400 3629 3630 Fix distcheck. 3631 3632commit 1314a1cd22823a5a7202dac4ac04da3801f1ec94 3633Author: Bankim Bhavsar <bbhavsar@vmware.com> 3634Date: Fri Mar 14 12:42:40 2008 -0400 3635 3636 Video Overlay: Source video parameters 3637 3638 This patch adds parameters to the source video frame. Applications can request 3639 only a subset of the source video frame to be displayed. These parameters are 3640 srcX, srcY, srcWidth and srcHeight. width and height represent the entire 3641 source video frame. 3642 3643commit fdef3fd5ea62a2df6283d48ba18d5c60300534ef 3644Author: Matthieu Herrb <matthieu.herrb@laas.fr> 3645Date: Sun Mar 9 00:08:32 2008 +0100 3646 3647 Makefile.am: nuke RCS Id 3648 3649commit bfd8398dde2d2de238c600eece9374d24d7d8549 3650Author: Bankim Bhavsar <bbhavsar@vmware.com> 3651Date: Wed Jan 23 22:13:07 2008 -0800 3652 3653 Xv extension for VMware's video driver 3654 3655 This patch implements the Xv extension for VMware's X video driver. 3656 3657 The Xv specification can be found here 3658 http://www.xfree86.org/current/DESIGN16.html 3659 3660 I've written a trivial offscreen memory manager that allocates memory from the 3661 bottom part of the Video RAM and it can handle only 1 video-stream. Eventually 3662 we intend to support upto 32 video-streams (there is already support for 3663 multiple video streams in respective backends). 3664 3665commit a33ab73ffa5d3797d16c3a265ddefc56a5a93b6b 3666Author: Vinay <vinay@vinayb-dev4.(none)> 3667Date: Sat Jan 19 08:39:24 2008 -0800 3668 3669 Trivial libpciaccess related fix 3670 3671 Fix VMWAREUnmapMem to correctly unmap memory for the libpciaccess case. 3672 3673commit 62d898669baccfd4c312f3ed8f228d0c3217d3c3 3674Author: Vinay Bondhugula <vinayb@vmware.com> 3675Date: Mon Oct 8 10:54:04 2007 -0700 3676 3677 Port to libpciaccess. 3678 3679commit 73db59fd82ced7a88f6da35d5088608e163f3d86 3680Author: Philip Langdale <philipl@fido2.homeip.net> 3681Date: Tue Sep 25 16:06:38 2007 -0700 3682 3683 Update for 10.15.2 release 3684 3685commit ee2fe14d37c9f3c8a170f8e3fcbc62403687ab8f 3686Author: Philip Langdale <philipl@fido2.homeip.net> 3687Date: Mon Sep 24 11:48:12 2007 -0700 3688 3689 There are problems when running with old X releases with a 3690 global symbol collision with the mouse driver. 3691 3692commit cfdb0140fa53155ae134ba961ec7d593d5d94713 3693Author: Philip Langdale <philipl@fido2.homeip.net> 3694Date: Mon Sep 10 21:19:38 2007 -0700 3695 3696 Guard modinfo section against gcc. 3697 3698commit e41aef1fb187c1b450fbea0163eb16e0c40f0435 3699Author: Philip Langdale <philipl@fido2.homeip.net> 3700Date: Mon Sep 10 21:16:04 2007 -0700 3701 3702 Ready 10.15.1 release. 3703 3704 To allow for easier detection of driver version by other VMware tools, 3705 we are embedding the version in a .modinfo section so that the Linux 3706 kernel 'modinfo' tool can be (ab)used to check it. 3707 3708commit 2331c7d1629f9841c02cf907ef4fa8b679cc5141 3709Author: James Cloos <cloos@jhcloos.com> 3710Date: Mon Sep 3 05:52:51 2007 -0400 3711 3712 Add *~ to .gitignore to skip patch/emacs droppings 3713 3714commit ec4dc29434bbbcf382de51fe3355679654489ad7 3715Author: Philip Langdale <philipl@fido2.homeip.net> 3716Date: Mon Jul 9 13:45:22 2007 -0700 3717 3718 Update copyright date. 3719 3720commit 0c76b95e7aeac034cc598ea6a407187f76466411 3721Author: Philip Langdale <philipl@fido2.homeip.net> 3722Date: Fri Feb 9 15:06:17 2007 -0800 3723 3724 Add svga_modes.h to Makefile.am 3725 3726commit a283c62c1db9e25ba11c6d72e4d00e0d911d64a2 3727Author: Philip Langdale <philipl@fido2.homeip.net> 3728Date: Fri Feb 9 11:09:17 2007 -0800 3729 3730 Filter out default modes that are larger than the hardware 3731 maxmimum size and add an explicit mode for the hardware 3732 maximum. 3733 3734commit f4540555dc8c823dfec53dddeb82e2b7b0dbfe3d 3735Author: Philip Langdale <philipl@fido2.homeip.net> 3736Date: Fri Feb 9 10:00:32 2007 -0800 3737 3738 Correct the incorrect 1080p resolution. 3739 3740 Should be 1920x1080 and not 1900x1080. Whoops. 3741 3742commit eca6e9f9458440feebd6598aee24f02a3fd7a75f 3743Author: Philip Langdale <philipl@fido2.homeip.net> 3744Date: Thu Feb 8 18:22:16 2007 -0800 3745 3746 Bump version to 10.15.0 3747 3748 This release adds truely usable resizing support by 3749 removing the restriction that the you cannot resize 3750 larger than the initial mode. 3751 3752commit 2ec8c25cc935a318bd07c0d1fba9960dfe8f38d4 3753Author: Philip Langdale <philipl@fido2.homeip.net> 3754Date: Thu Feb 8 18:07:19 2007 -0800 3755 3756 Fix a race condition when switching to a multihead fullscreen mode. 3757 3758 Due to implemtation details on the host side, switching to a multihead 3759 fullscreen mode means multiple mode changes along the way. However, 3760 because rrSetScreenConfig returns before the mode change completes, 3761 we can enter a race, where a subsequent SetDisplayTopology overrides 3762 the current pending one, leading to a mode switch where the resolution 3763 and topology are no longer consistent. 3764 3765 So, we fix this by ignoring any attempts to set the pending topology if 3766 one is already pending. 3767 3768commit a8afe781e108773e4a575ecc3330a5aa83c1cbb9 3769Author: Philip Langdale <philipl@fido2.homeip.net> 3770Date: Thu Feb 8 17:08:08 2007 -0800 3771 3772 Add a bunch of useful modes to the mode list. 3773 3774 The X server prunes modes above the initial mode and 3775 the number of default modes varies with distro and 3776 server version. As it's easy to do, let's add a bunch 3777 of useful versions in the driver so that users don't 3778 have to mess with their config files as often. 3779 3780commit 520f570aa3869090eac3accf6a2b3886ab188590 3781Author: Philip Langdale <philipl@fido2.homeip.net> 3782Date: Thu Feb 8 13:44:57 2007 -0800 3783 3784 Update man page to remove mention of the NoAccel option which 3785 no longer exists as we don't do any 'acceleration'. 3786 3787commit af8c0596a224ba8436d55c5dd6063fff688c7ad1 3788Author: Philip Langdale <philipl@fido2.homeip.net> 3789Date: Thu Feb 8 13:42:31 2007 -0800 3790 3791 Remove XAA 'acceleration'. 3792 3793 We have a lot of stale code to 'accelerate' certain operations 3794 through XAA. However, in practice, this acceleration is 3795 completely unhelpful because whenever we encounter an 3796 unaccelerated action, we have to sync and flush the fifo 3797 which kills any performance gain. 3798 3799 As such, the virtual hardware doesn't even advertise these 3800 acceleration capabilities anymore, so the code is completely 3801 unusued. 3802 3803 In addition, XAA is on the way out, so there's no point 3804 leaving in dead code which will have to go in the fairly 3805 near future. 3806 3807 The one operation we can meaningfully accelerate is a 3808 front-only fill and when we get around to implementing 3809 that, we'll use EXA instead. 3810 3811commit 3dc1ebc7c2c76f575b5150f1890ec13f8bd3e8a1 3812Author: Philip Langdale <philipl@fido2.homeip.net> 3813Date: Thu Feb 8 10:33:14 2007 -0800 3814 3815 Make resizing work once and for all. 3816 3817 This change removes the limitation on resizing larger than the 3818 initial size and removes the need for pitch-locking bu resizing 3819 the screen pixmap. Now the only limit on the screen size is 3820 the maximum width/height reported by the virtual hardware which 3821 is configured on the host side. 3822 3823commit 188c2b44f0f47093092414f9364a52cc2d5016ad 3824Author: Eamon Walsh <ewalsh@tycho.nsa.gov> 3825Date: Mon Dec 18 14:05:19 2006 -0500 3826 3827 Revert "Convert callers of LookupWindow() to dixLookupWindow()." 3828 3829 This reverts commit 1f298c24fe1dde35f0d77cc34518eb9c2b873fc5. 3830 3831commit 1f298c24fe1dde35f0d77cc34518eb9c2b873fc5 3832Author: Eamon Walsh <ewalsh@tycho.nsa.gov> 3833Date: Fri Dec 15 17:41:52 2006 -0500 3834 3835 Convert callers of LookupWindow() to dixLookupWindow(). 3836 3837commit 9d41384616c8f6d00b53a61634f73e728d86d615 3838Author: Philip Langdale <philipl@fido2.homeip.net> 3839Date: Wed Dec 6 16:48:12 2006 -0800 3840 3841 Explicitly initialise buffer to NULL for parsing xinerama topologies. 3842 3843commit ac2a57ea7aea00c695495aa01c27402569a9514a 3844Author: Philip Langdale <philipl@fido2.homeip.net> 3845Date: Tue Oct 31 19:17:39 2006 -0800 3846 3847 Update man page to document "StaticXinerama" 3848 3849commit 4dd038bd73208184344cae4f363c492c7e61524b 3850Author: Philip Langdale <philipl@fido2.homeip.net> 3851Date: Tue Oct 31 19:12:29 2006 -0800 3852 3853 Bump version to 10.14.0 3854 3855commit 73945118a6b3cad60f4a1927b9394206d36e8281 3856Author: Philip Langdale <philipl@fido2.homeip.net> 3857Date: Mon Oct 30 11:52:57 2006 -0800 3858 3859 Fix more warnings. 3860 3861commit e0dca2ff0b695b60830cf8f10a2c0585602a4d40 3862Author: Philip Langdale <philipl@fido2.homeip.net> 3863Date: Mon Oct 30 10:53:18 2006 -0800 3864 3865 Fix some warnings noted when compiling the code against 3866 older Xorg versions. 3867 3868commit f96d12047725becf0515cc4df8572ca4a3956309 3869Author: Philip Langdale <philipl@fido2.homeip.net> 3870Date: Wed Oct 25 18:06:13 2006 -0700 3871 3872 Refactor static Xinerama parsing. 3873 3874commit 57b32570f84d06180be2c9d249174b04f02d9fca 3875Author: Philip Langdale <philipl@fido2.homeip.net> 3876Date: Wed Oct 25 10:49:27 2006 -0700 3877 3878 Add support for a static Xinerama configuration to be specified 3879 in xorg.conf. This static configuration will always be used and 3880 any attempts to update it will be ignored. 3881 3882 Note that all xinerama aware apps that I have seen will not 3883 sanity check the xinerama extents against the screen size, so 3884 if the screen doesn't properly enclose the xinerama extents, 3885 expect your apps to punish you accordingly. 3886 3887commit c62a63b00f7c43e265079d31f1bf7ff447423056 3888Author: Philip Langdale <philipl@fido2.homeip.net> 3889Date: Wed Oct 11 11:47:39 2006 -0700 3890 3891 Update .gitignore 3892 3893commit 9345f8146ae41f7d2269e6fea5e3cba385139960 3894Author: Philip Langdale <philipl@fido2.homeip.net> 3895Date: Wed Oct 11 11:47:25 2006 -0700 3896 3897 Add vmwarectrl sample client. 3898 3899commit b35666fdd567d4cd92574aeecadb8e0d41b32d77 3900Author: Philip Langdale <philipl@fido2.homeip.net> 3901Date: Wed Oct 11 11:36:22 2006 -0700 3902 3903 Add .gitignore files. 3904 3905commit 7631f4d8db2b83878bd886204aaa978dac9c47b5 3906Author: Philip Langdale <philipl@fido2.homeip.net> 3907Date: Wed Oct 11 11:32:18 2006 -0700 3908 3909 Document the 'Xinerama' option and tweak the default to reflect 3910 the capabilites of the underlying hardware. 3911 3912commit 79de175986817cac2fedb267cc9a04438b3694d4 3913Author: Philip Langdale <philipl@fido2.homeip.net> 3914Date: Wed Oct 11 11:26:04 2006 -0700 3915 3916 Check for xinerama in configure.ac 3917 3918commit 34e7264e99ceab5e9e2e022ed9a56531845b0b17 3919Author: Philip Langdale <philipl@fido2.homeip.net> 3920Date: Wed Oct 11 10:57:57 2006 -0700 3921 3922 Add xinerama support to the vmware video driver. (Better late than never). 3923 3924 With this change, the VMWARE_CTRL extension is updated so that it can 3925 receive topology updates at runtime. I will add some sample client code 3926 separately. 3927 3928 I also intend to add support for a static initial topology defined in 3929 xorg.conf but I haven't got around to it yet due to hating to write 3930 string parsing code. 3931 3932commit dd201e23cc50ded2f316dbfa5d265608b80d994d 3933Author: Philip Langdale <philipl@fido2.homeip.net> 3934Date: Wed Sep 20 11:40:54 2006 -0700 3935 3936 Add support for the upcoming PITCHLOCK register. The FIFO register 3937 is deprecated but support is retained so the driver works with 3938 Workstation 5.5 and friends. 3939 3940commit 0850feff708ded63c27dc938ca4b9b8fcbed122b 3941Author: Philip Langdale <philipl@fido2.homeip.net> 3942Date: Sun Sep 3 10:45:22 2006 -0700 3943 3944 Fix https://bugs.freedesktop.org/show_bug.cgi?id=8094 3945 Lazily allocate the dynamic modes used for pixel precise guest 3946 resizing. They will not appear until they are neeed. 3947 3948commit 86aeab28b806642946e59a98573c5b25840c35ed 3949Author: Adam Jackson <ajax@nwnk.net> 3950Date: Fri Apr 7 21:57:24 2006 +0000 3951 3952 Unlibcwrap. Bump server version requirement. Bump to 1.1.0. 3953 3954commit fe6963e09a8197f5c11e8a828c8da4024601a466 3955Author: Adam Jackson <ajax@nwnk.net> 3956Date: Fri Apr 7 21:55:49 2006 +0000 3957 3958 Unlibcwrap. Bump server version requirement. Bump to 10.13.0. 3959 3960commit ac83dff5d6a54b05e0377ceac2b133238293fa0e 3961Author: philipl <philipl> 3962Date: Tue Feb 7 18:51:35 2006 +0000 3963 3964 Add VMWARE_CTRL extension. 3965 3966commit a49f1e65147dc1ab805c74951cb0915b1352e8a9 3967Author: philipl <philipl> 3968Date: Tue Jan 17 00:40:43 2006 +0000 3969 3970 Use DriverFunc to adjust physical size to keep DPI constant. 3971 3972commit ff3f9dc930ff6bd2d6a73d56c70b0284a597cd62 3973Author: Kevin E Martin <kem@kem.org> 3974Date: Wed Dec 21 02:30:04 2005 +0000 3975 3976 Update package version for X11R7 release. 3977 3978commit aa1da8e57939081f15db04fa8c654f363f4e4cb7 3979Author: Adam Jackson <ajax@nwnk.net> 3980Date: Mon Dec 19 16:25:56 2005 +0000 3981 3982 Stub COPYING files 3983 3984commit 835f8000fdbdd35a6bbc5eabf1252d4040a798da 3985Author: Kevin E Martin <kem@kem.org> 3986Date: Thu Dec 15 00:24:21 2005 +0000 3987 3988 Update package version number for final X11R7 release candidate. 3989 3990commit 32a969be355c0c948739b4f18e0caa83c78960cc 3991Author: Kevin E Martin <kem@kem.org> 3992Date: Tue Dec 6 22:48:41 2005 +0000 3993 3994 Change *man_SOURCES ==> *man_PRE to fix autotools warnings. 3995 3996commit 643a290d01553c9fefe9b4e268ae515d49cd1af9 3997Author: Kevin E Martin <kem@kem.org> 3998Date: Sat Dec 3 05:49:37 2005 +0000 3999 4000 Update package version number for X11R7 RC3 release. 4001 4002commit 9d0f12ee6bf340765317214ffef2f70883025e9e 4003Author: Kevin E Martin <kem@kem.org> 4004Date: Fri Dec 2 02:16:12 2005 +0000 4005 4006 Remove extraneous AC_MSG_RESULT. 4007 4008commit e0a08567f46f73005589876d38e3f5ff7543557b 4009Author: Adam Jackson <ajax@nwnk.net> 4010Date: Tue Nov 29 23:30:06 2005 +0000 4011 4012 Only build dlloader modules by default. 4013 4014commit c80ca69d0160b78e8d2bbeae01de52978f8e581f 4015Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 4016Date: Mon Nov 28 22:04:15 2005 +0000 4017 4018 Change *mandir targets to use new *_MAN_DIR variables set by xorg-macros.m4 4019 update to fix bug #5167 (Linux prefers *.1x man pages in man1 subdir) 4020 4021commit 417099c602463ff364f70a938c4a2d2275d8a5b7 4022Author: Eric Anholt <anholt@freebsd.org> 4023Date: Mon Nov 21 10:49:23 2005 +0000 4024 4025 Add .cvsignores for drivers. 4026 4027commit 26a28492f738defa0445059c2dd6bb9cad89d0aa 4028Author: Kevin E Martin <kem@kem.org> 4029Date: Wed Nov 9 21:31:21 2005 +0000 4030 4031 Bump version numbers for X11R6.9/X11R7 RC2. 4032 4033commit 9ba2f5c58f6ffc9533f5d1852b6d7ba25fa4785a 4034Author: Kevin E Martin <kem@kem.org> 4035Date: Wed Nov 9 21:15:23 2005 +0000 4036 4037 Update package version number for X11R7 RC2 release. 4038 4039commit 731860b1fea54f2ce54fbf02ab77d7c01f381f80 4040Author: nolan <nolan> 4041Date: Thu Nov 3 04:26:00 2005 +0000 4042 4043 Fix a bug where the cursor disappears. Committed on behalf of Shelley Gong 4044 (shelleygong@vmware.com). 4045 4046commit 623b3c8297fc0b0d05ce0d6684ef3ab57054b0eb 4047Author: Kevin E Martin <kem@kem.org> 4048Date: Tue Nov 1 15:08:57 2005 +0000 4049 4050 Update pkgcheck depedencies to work with separate build roots. 4051 4052commit 4f1ef5a0bf2172b9d38c6b1b5dd92ae8efe01f42 4053Author: Kevin E Martin <kem@kem.org> 4054Date: Wed Oct 19 02:48:06 2005 +0000 4055 4056 Update package version number for RC1 release. 4057 4058commit 3abe1c93562dbc4bbebfd293f7e90972e90c9dcc 4059Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 4060Date: Tue Oct 18 00:01:55 2005 +0000 4061 4062 Use @DRIVER_MAN_SUFFIX@ instead of $(DRIVER_MAN_SUFFIX) in macro 4063 substitutions to work better with BSD make 4064 4065commit ba3716950db330db08960248827bbf9321d42d71 4066Author: Adam Jackson <ajax@nwnk.net> 4067Date: Mon Oct 17 22:57:38 2005 +0000 4068 4069 More 1.7 braindamage: define EXTRA_DIST in terms of @DRIVER_NAME@ instead 4070 of indirectly 4071 4072commit 18207a8cc41044538520dd9f0036e44d3fe26f94 4073Author: Alan Coopersmith <Alan.Coopersmith@sun.com> 4074Date: Mon Oct 17 00:09:09 2005 +0000 4075 4076 Use sed & cpp to substitute variables in driver man pages 4077 4078commit bce0feda4d05bd8fd4db20476e117026a21f3906 4079Author: nolan <nolan> 4080Date: Fri Sep 23 02:55:25 2005 +0000 4081 4082 Support for XRandR in the VMware driver. Committed on behalf of Phil 4083 Langdale (plangdale@vmware.com) 4084 4085commit f06924363cd77770257db5aac6d4d3c6add89dfa 4086Author: Daniel Stone <daniel@fooishbar.org> 4087Date: Thu Aug 18 09:03:55 2005 +0000 4088 4089 Update autogen.sh to one that does objdir != srcdir. 4090 4091commit 120d7d1c616d1e2e3c7ca0d1d119d1b0731491c6 4092Author: Kevin E Martin <kem@kem.org> 4093Date: Fri Jul 29 21:22:46 2005 +0000 4094 4095 Various changes preparing packages for RC0: 4096 - Verify and update package version numbers as needed 4097 - Implement versioning scheme 4098 - Change bug address to point to bugzilla bug entry form 4099 - Disable loadable i18n in libX11 by default (use --enable-loadable-i18n to 4100 reenable it) 4101 - Fix makedepend to use pkgconfig and pass distcheck 4102 - Update build script to build macros first 4103 - Update modular Xorg version 4104 4105commit cd14003e4bf878e464437143cf18d30803ace999 4106Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> 4107Date: Tue Jul 26 23:57:30 2005 +0000 4108 4109 Build system for vmware 4110 4111commit e5ab935718e3f6aea8edc5edef5c0bd78af0ba14 4112Author: Adam Jackson <ajax@nwnk.net> 4113Date: Mon Jul 11 02:30:07 2005 +0000 4114 4115 Prep for modular builds by adding guarded #include "config.h" everywhere. 4116 4117commit daa2be7a9adf977eb4b63bb601d4b8b0b2881903 4118Author: Daniel Stone <daniel@fooishbar.org> 4119Date: Fri Jul 1 22:43:21 2005 +0000 4120 4121 Change all misc.h and os.h references to <X11/foo.h>. 4122 4123commit ae1ba62c3937e6d1067f850d3aeb0b11d316fb86 4124Author: Adam Jackson <ajax@nwnk.net> 4125Date: Sat Jun 25 21:17:00 2005 +0000 4126 4127 Bug #3626: _X_EXPORT tags for video and input drivers. 4128 4129commit 57d09f92c656203958bbc2040f255040f7d051ca 4130Author: nolan <nolan> 4131Date: Fri Feb 18 23:15:59 2005 +0000 4132 4133 updated copyright string in man page of VMware SVGA driver. 4134 4135commit b4379cd6f99c14a2feaabce4560f6dfd62543c62 4136Author: nolan <nolan> 4137Date: Thu Sep 30 21:44:35 2004 +0000 4138 4139 Bugfix for Alpha Texture acceleration from Constantine Sapuntzakis. 4140 4141commit 325df911a83e336d3463ea2220151af4e9f0ccfa 4142Author: nolan <nolan> 4143Date: Sat Aug 14 01:03:55 2004 +0000 4144 4145 Save the register state before we write to any of them. This makes the 4146 VMware driver coexist with VESA fbcon. 4147 4148commit fed82eea7b64b645926c5102563887bc424245b7 4149Author: nolan <nolan> 4150Date: Tue Jun 29 02:20:25 2004 +0000 4151 4152 Send cursor hotspot through the FIFO. This fixes host cursor. Allow any bpp 4153 when 8BIT_EMULATION is set. Save/restore the ID register. 4154 4155commit a3bdeecbb99c721987a22d52c8adf4b1a96e8eec 4156Author: Eric Anholt <anholt@freebsd.org> 4157Date: Wed Jun 16 09:22:38 2004 +0000 4158 4159 DRI XFree86-4_3_99_12-merge import 4160 4161commit 2e4ba05c72961b611c9c4b93038ce54ab54ab9c3 4162Author: Egbert Eich <eich@suse.de> 4163Date: Fri Apr 23 19:53:57 2004 +0000 4164 4165 Merging XORG-CURRENT into trunk 4166 4167commit b01798791090f20fbf4907902d0d35853c05e3a2 4168Author: Egbert Eich <eich@suse.de> 4169Date: Sun Mar 14 08:33:42 2004 +0000 4170 4171 Importing vendor version xf86-4_4_99_1 on Sun Mar 14 00:26:39 PST 2004 4172 4173commit 2ef56b50d6b4fc44359b2bf181ece342fb569153 4174Author: Egbert Eich <eich@suse.de> 4175Date: Wed Mar 3 12:12:28 2004 +0000 4176 4177 Importing vendor version xf86-4_4_0 on Wed Mar 3 04:09:24 PST 2004 4178 4179commit 7218efdc0d95a5d50b3ddfc6a76921e285cf61b6 4180Author: Egbert Eich <eich@suse.de> 4181Date: Thu Feb 26 13:35:56 2004 +0000 4182 4183 readding XFree86's cvs IDs 4184 4185commit 205e213f8b296c47a80e7c8cfb835449b064fd84 4186Author: Egbert Eich <eich@suse.de> 4187Date: Thu Feb 26 09:23:29 2004 +0000 4188 4189 Importing vendor version xf86-4_3_99_903 on Wed Feb 26 01:21:00 PST 2004 4190 4191commit 2f79bc6606f337d0d892e6c73cdf11fe211ead13 4192Author: Kaleb Keithley <kaleb@freedesktop.org> 4193Date: Tue Nov 25 19:28:43 2003 +0000 4194 4195 XFree86 4.3.99.16 Bring the tree up to date for the Cygwin folks 4196 4197commit 118eb1bdba4191972b1ca182e34f3d3ca456daea 4198Author: Kaleb Keithley <kaleb@freedesktop.org> 4199Date: Fri Nov 14 16:48:55 2003 +0000 4200 4201 XFree86 4.3.0.1 4202 4203commit 0b01e22c9cd727cce23c4a9670806f7a839e1937 4204Author: Kaleb Keithley <kaleb@freedesktop.org> 4205Date: Fri Nov 14 16:48:55 2003 +0000 4206 4207 Initial revision 4208