summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2014-01-30Cleanup of debug messages and failed attempt to pull the latestGPUTextureLoadMark Mueller11-46/+14583
MESA_FORMAT changes in. There wasn't enought time to complete this, but it does run, though some cases may be falling back to CPU load.
2014-01-29Code housekeeping.Mark Mueller6-90/+34
2014-01-27minor updates and fixesMark Mueller7-83/+42
2014-01-13Change try_blorp_blit to let brw_blorp_blit_params::test_formats decideMark Mueller1-54/+8
what formats blorp_blit supports. Send blit's which require Alpha processing through since blorp can handle them.
2014-01-13Fix minor issues with non-linear color spaceMark Mueller3-20/+17
2014-01-13Move all format selection to translate_tex_format and cleanup parameterMark Mueller7-115/+89
usage. Add missing MESA_FORMAT to mesa choose tex format.
2014-01-13Fix more sillynessMark Mueller1-1/+2
2014-01-13Fix silly coding.Mark Mueller1-6/+6
2014-01-13Add handler to blorp for basic type conversion. Eliminate blorpMark Mueller2-13/+164
operations on irrelevant color components for performance.
2014-01-13Fix copy pasta errorMark Mueller1-3/+3
2014-01-13somehow this file snuck in - removedMark Mueller1-565/+0
2014-01-13Remove flip sillynessMark Mueller1-27/+5
2014-01-13Piglit fixes for non-linear color formats; formats with samplers thatMark Mueller17-382/+388
return raw pixel data; and lower precision internal formats. Cleaned up Mesa formats based on Mesa Dev feedback.
2014-01-13Fix typoMark Mueller2-1/+566
2014-01-13Resolve last of the piglit regressionsMark Mueller1-4/+4
2014-01-13Clean up method for detecting depth surface formats. Fix issue withMark Mueller4-27/+31
desktop_gl_sel. Add missing MESA_FORMAT_ABGR2101010_UINT render target.
2014-01-13Reverted to previous usage of BRW_SURFACEFORMAT_R32G32_FLOAT_LD.Mark Mueller2-28/+35
Resolved issue where R format is used as a sampler for Alpha, Intensity, and Luminance user textures. Down to 17 failed regressions and 27 skipped regressions
2014-01-13Use render target format as sample format for depth/stencil blts. AvoidMark Mueller4-8/+30
blts when internal fomrat is smaller then target format and source is larger than internal.
2014-01-13Texture arrays aren't current supportedMark Mueller1-0/+4
2014-01-13Housekeeping. BRW_SURFACEFORMAT_R32G32B32A32_SINT fails piglit tests asMark Mueller5-473/+4
a render target and BRW_SURFACEFORMAT_R24_UNORM_X8_TYPELESS works as a render target for both clears and blt's, previously a different format was used for each, which was confusing
2014-01-13Resolved MESA_FORMAT_X8_Z24 regression caused by upstream using the sameMark Mueller1-2/+2
format for sample and render targets. Fix peren error
2014-01-13Change handling of Z16 depth texturesMark Mueller1-16/+16
2014-01-13Housekeeping. Fixed ~400 piglit regressionsMark Mueller4-149/+172
2014-01-13blorp blit can't support widths less then 16 pixels. Fixes forMark Mueller3-37/+44
unsupported render targets
2014-01-13Lots of code cleanupMark Mueller4-142/+147
2014-01-13more useful printfsMark Mueller2-0/+4
2014-01-13printout changesMark Mueller2-2/+5
2014-01-13added printout for redundant format settingMark Mueller1-0/+2
2014-01-13Resolve issues with internal formats that force extra processing.Mark Mueller6-86/+553
2014-01-13Resolving piglit regressionsMark Mueller2-56/+68
2014-01-13glean pixelFormats is passing againMark Mueller5-21/+12
2014-01-13more printoutsMark Mueller3-6/+10
2014-01-13Resovled render and sample target issuesMark Mueller5-414/+426
2014-01-13Added some print statementsMark Mueller1-0/+4
2014-01-13In process with cleaning up of format handlingMark Mueller10-370/+713
2014-01-13Starting to see very positive resultsMark Mueller16-584/+1044
2014-01-13Revert "Fix typo that that was breaking some blorp operations"Mark Mueller1-2/+2
This reverts commit 4336e5c59085b93ff684251c33459f065f974e7a.
2014-01-13Fix typo that that was breaking some blorp operationsMark Mueller1-2/+2
2014-01-13Allow blorp to make decisions about the formats that it supports where it canMark Mueller7-63/+132
2014-01-13Add gl_formats to cover all GLUser provided format/type combinationsMark Mueller4-4/+789
2014-01-13r600g: fix glClearBuffer by handling PIPE_CLEAR_COLORi flags correctlyMarek Olšák1-38/+31
also restructure the code
2014-01-13r600g: handle NULL colorbuffers correctly on R600-R700Marek Olšák1-65/+55
2014-01-13r600g: handle NULL colorbuffers correctly on EvergreenMarek Olšák2-28/+54
2014-01-13radeonsi: handle NULL colorbuffers correctlyMarek Olšák1-9/+11
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
2014-01-13gallium/util: easy fixes for NULL colorbuffersMarek Olšák2-1/+7
Reviewed-by: Brian Paul <brianp@vmware.com>
2014-01-13st/mesa: bind NULL colorbuffers as specified by glDrawBuffersMarek Olšák6-14/+40
An example why it is required: Let's say there's a fragment shader writing to gl_FragData[0..1]. The user calls: glDrawBuffers(2, {GL_NONE, GL_COLOR_ATTACHMENT0}); That means gl_FragData[0] is unused and gl_FragData[1] is written to GL_COLOR_ATTACHMENT0. st/mesa was skipping the GL_NONE draw buffer, therefore gl_FragData[0] was written to GL_COLOR_ATTACHMENT0, which was wrong. This commit fixes it, but drivers must also be fixed not to crash when binding NULL colorbuffers. There is also a new set of piglit tests for this. The MSAA state also had to be fixed not to crash when reading fb->cbufs[0]. Reviewed-by: Brian Paul <brianp@vmware.com>
2014-01-13mesa: handle GL_NONE draw buffers correctly in glClearMarek Olšák1-1/+5
Reviewed-by: Brian Paul <brianp@vmware.com>
2014-01-13st/mesa: use sRGB formats for MSAA resolving if destination is sRGBMarek Olšák1-0/+32
Copied from the i965 driver, including the big comment. Cc: 9.2 10.0 <mesa-stable@lists.freedesktop.org>
2014-01-13st/mesa: check depth and stencil writemask before clearingMarek Olšák1-2/+15
2014-01-13st/mesa: always prefer pipe->clear over clear_with_quad (v2)Marek Olšák1-5/+14
v2: clear depth and stencil together
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-15 10:56:42 +0000