AgeCommit message (Collapse)AuthorFilesLines
2020-03-09docs: add sha256sum for 19.3.519.3Eric Engestrom1-1/+1
Signed-off-by: Eric Engestrom <>
2020-03-09VERSION: bump version to 19.3.5mesa-19.3.5Eric Engestrom1-1/+1
Signed-off-by: Eric Engestrom <>
2020-03-09docs: add release notes for 19.3.5Eric Engestrom1-0/+169
Signed-off-by: Eric Engestrom <>
2020-03-09gitlab-ci: update template to fix container build issuesEric Engestrom2-2/+2
Signed-off-by: Eric Engestrom <>
2020-03-07anv: Parse VkPhysicalDeviceFeatures2 in CreateDeviceJason Ekstrand2-11/+46
The client may enable robustBufferAccess2 via either pCreateInfo->pEnabledFeatures or via a chained-in VkPhysicalDeviceFeatures2 struct. We need to parse both. Fixes: 022e5c7e5a5 "anv: Implement VK_KHR_get_physical_device_properties2" Reviewed-by: Lionel Landwerlin <> Part-of: <> (cherry picked from commit 35ca2ad22e20ad3bc3301ee1e9157b8c351d959e)
2020-03-07.pick_status.json: Update to 0103f02acb10dcdea23461ba214307a6827a7772Eric Engestrom1-0/+738
2020-03-07bin/ fix commit list commandEric Engestrom2-2/+2
Fixes: 86079447da1e00d49db0 ("scripts: Add a script") Signed-off-by: Eric Engestrom <> Reviewed-by: Dylan Baker <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit d7a70fbb2305604ce75b1a0dbcd03e2ebe71f92a)
2020-03-06iris: Don't skip fast depth clears if the color changedJason Ekstrand2-2/+6
We depend on BLORP to convert the clear color and write it into the clear color buffer for us. However, we weren't bothering to call blorp in the case where the state is ISL_AUX_STATE_CLEAR. This leads to the clear color not getting properly updated if we have back-to-back clears with different clear colors. Technically, we could go out of our way to set the clear color directly from iris in this case but this is a case we're unlikely to see in the wild so let's not bother. This matches what we already do for color surfaces. Cc: Reported-by: Mark Janes <> Reviewed-by: Nanley Chery <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 9d07d598423e4015bbc7beb7a2fdc4c657d5e0cf)
2020-03-06isl: Set 3DSTATE_DEPTH_BUFFER::Depth correctly for 3D surfacesJason Ekstrand2-2/+20
Cc: Reviewed-by: Nanley Chery <> Part-of: <> (cherry picked from commit 9f5f4269a66eebfcaa3ae5cd7bdf91d88a7fc69c)
2020-03-06Revert "spirv: Use a simpler and more correct implementaiton of tanh()"Kristian H. Kristensen2-15/+14
This reverts commit da1c49171d0df185545cfbbd600e287f7c6160fa. The reduced formula has precision problems on fp16 around 0. Bring back the old formula, but make sure to keep the clamping. Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 9f9432d56c055b9704a76cad44da88d5e12f825c)
2020-03-06Revert "glsl: Use a simpler formula for tanh"Kristian H. Kristensen2-11/+9
This reverts commit 9807f502eb7a023be619a14119388b2a43271b0e. The simplified formula doesn't pass the tanh dEQP tests when we lower to fp16 math. Part-of: <> (cherry picked from commit 986e92f0ea803caf014adc40e900bc774af71da3)
2020-03-06Revert "mesa: check for z=0 in _mesa_Vertex3dv()"Marek Olšák2-5/+2
This reverts commit f04d7439a0ad6e13ff2912ff824553b6bcf511a4. It no longer helps performance and the current vbo implementation is faster anyway. The app that hit this was a CAD program called Spazio3D. It made pretty terrible use of the OpenGL API and we sent them some tips for improvements. I'm assuming they've fixed this by now. Reviewed-by: Mathias Fröhlich <> Reviewed-by: Ian Romanick <> Part-of: <> (cherry picked from commit df3891e74a72d275aceba91adc94a9e7dc9aa029)
2020-03-06Revert "glx: convert glx_config_create_list to one big calloc"Andrii Simiklit2-27/+38
This reverts commit 35fc7bdf0e6ad6547e39099e7060a3d89539b56d. Unfortunately mentioned commit introduced a memory leak because `driwindowsMapConfigs` and `createDriMode` functions allocate small memory portions for each element: 21,576 (232 direct, 21,344 indirect) bytes in 1 blocks are definitely lost in loss record 1,411 of 1,414 at 0x483A7F3: malloc (in /usr/lib/x86_64-linux-gnu/valgrind/ by 0x5D4AA09: createDriMode (dri_common.c:291) by 0x5D4ABF5: driConvertConfigs (dri_common.c:310) by 0x5D58414: dri3_create_screen (dri3_glx.c:945) by 0x5D39829: AllocAndFetchScreenConfigs (glxext.c:815) by 0x5D39C57: __glXInitialize (glxext.c:941) by 0x5D3290A: GetGLXPrivScreenConfig (glxcmds.c:174) by 0x5D34F38: glXQueryExtensionsString (glxcmds.c:1307) by 0x4F83038: glXQueryExtensionsString (in /usr/local/lib/ by 0x4F2EA6B: ??? (in /usr/lib/x86_64-linux-gnu/ by 0x4F2A0D7: waffle_display_connect (in /usr/lib/x86_64-linux-gnu/ by 0x498F42A: wfl_checked_display_connect (piglit-util-waffle.h:74) There is one more thing which disallow us to easily fix it are different element sizes for instance: `glx_config_create_list` allocates memory just for `glx_config`, `driwindowsMapConfigs` for `driwindows_config` and `createDriMode` for `__GLXDRIconfigPrivate`. Yes it is possible but size of such fix will be more big and complex than original one. So it make sense only if the malloc overhead really is a big problem there. Acked-by: Eric Engestrom <> Signed-off-by: Andrii Simiklit <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 311c82e1923f63070b198881d90c1098f4ff7a08)
2020-03-06intel/gen12+: Disable mid thread preemption.Rafael Antognolli4-1/+26
Fixes a GPU hang in Car Chase. Cc: v2: Add comment explaining why (Jason). Reviewed-by: Jason Ekstrand <> Reviewed-by: Lionel Landwerlin <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 5f13996262a6d72ca5b5c235647d5257ae961b66)
2020-03-06intel/device: bdw_gt1 actually has 6 eus per subslicePaulo Zanoni2-2/+2
Found by inspection, I'm not aware of any bugs caused by this typo. According to Lionel, it seems we only use this to generate masks of available EUs for perfromance queries, and it's only used when we can't query the fused parts of the GPU through DRM_IOCTL_I915_QUERY. So this patch should help for the corner case where the Kernel is too old to support the query ioctl. v2: improve commit message, cc stable (Lionel). Cc: Reviewed-by: Jason Ekstrand <> Reviewed-by: Lionel Landwerlin <> Reviewed-by: Kenneth Graunke <> Signed-off-by: Paulo Zanoni <> Part-of: <> (cherry picked from commit aa78801f0a6cfeaf3d16b4333239c0b862f73c10)
2020-03-06mesa/st: fix formats required for EXT_texture_norm16Tapani Pälli2-8/+1
Earlier commit did not take in to account that lists required for rendering and texturing are parsed separately. This commit simply removes formats added to the other list. Fixes: de4eb9a3bb9 ("mesa/st: toggle EXT_texture_norm16 based on format support") Signed-off-by: Tapani Pälli <> Reviewed-by: Eric Anholt <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit fbd61b3fb66bcc3eb0f65da2c869046c24c35dc8)
2020-03-06intel/compiler: Restrict cs_threads to 64Jordan Justen2-2/+4
Our current GPGPU_WALKER code only supports up to 64 threads. On HSW we could use up to 70 and TGL up to 112, but only if the walker is adjusted so the width does not exceed 64. Work to support this is in progress. Previous to this change, we might try to downgrade to SIMD8 if the SIMD16 shader spilled. Since HSW and TGL have the max number of threads above 64, we would then try to emit an invalid GPGPU walker command. Fixes: 932045061b5 ("i965/cs: Emit compute shader code and upload programs") Signed-off-by: Jordan Justen <> Reviewed-by: Jason Ekstrand <> Reviewed-by: Paulo Zanoni <> Tested-by: Paulo Zanoni <> (cherry picked from commit cf12faef614ab7cd9996410f1d161558a3853936)
2020-03-06ac/llvm: flush denorms for nir_op_fmed3 on GFX8 and older gensSamuel Pitoiset2-1/+6
The hardware doesn't flush denorms, exactly like fmin/fmax, so we have to do it manually. This doesn't fix anything known. Fixes: d6a07732c9c ("ac: use llvm.amdgcn.fmed3 intrinsic for nir_op_fmed3") Signed-off-by: Samuel Pitoiset <> Reviewed-by: Bas Nieuwenhuizen <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 9e5d2a73c5fc12841b62758a035b2bdb191b3f86)
2020-03-06ac/llvm: fix 16-bit fmed3 on GFX8 and older gensSamuel Pitoiset2-3/+5
16-bit med3 is only supported on GFX9+. Fixes dEQP-VK.spirv_assembly.instruction.amd_trinary_minmax.mid3.f16.*. Fixes: d6a07732c9c ("ac: use llvm.amdgcn.fmed3 intrinsic for nir_op_fmed3") Signed-off-by: Samuel Pitoiset <> Reviewed-by: Bas Nieuwenhuizen <> Part-of: <> (cherry picked from commit 30ac733680c3dfbfd1300c5498dd1b0c0a680905)
2020-03-06ac/llvm: fix 64-bit fmed3Samuel Pitoiset2-18/+32
Lower 64-bit fmed3 because LLVM doesn't expose an intrinsic. Fixes dEQP-VK.spirv_assembly.instruction.amd_trinary_minmax.mid3.f64.*. Fixes: d6a07732c9c ("ac: use llvm.amdgcn.fmed3 intrinsic for nir_op_fmed3") Signed-off-by: Samuel Pitoiset <> Reviewed-by: Bas Nieuwenhuizen <> Part-of: <> (cherry picked from commit 50b8c2527464dbe18a01ab6412de4465cebf2225)
2020-03-06mesa: Flush vertices before changing the OpenGL state.Mathias Fröhlich2-5/+11
Reviewed-by: Marek Olšák <> CC: <> Signed-off-by: Mathias Fröhlich <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 636656bcd7801c703ebcf9bd4c65197e4e6cbee8)
2020-03-06gallivm/tgsi: fix stream id regressionDave Airlie2-3/+6
This broke TGSI GS shaders with llvmpipe, it wasn't looking at the right immediates and it should be cast to an integer type. Fixes: 163d5fde0669 (gallium/swr: Enable GL_ARB_gpu_shader5: multiple streams) Reviewed-by: Krzysztof Raszkowski <> Acked-by: Jan Zielinski <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 954cf8e86b6e0d52c04098604d2daa4305bf6f70)
2020-03-06mesa/st: toggle EXT_texture_norm16 based on format supportTapani Pälli2-1/+13
Closes: Fixes: 7f467d4f738 ("mesa: GL_EXT_texture_norm16 extension plumbing") Signed-off-by: Tapani Pälli <> Acked-by: Lionel Landwerlin <> Reviewed-by: Alejandro Piñeiro <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit de4eb9a3bb9fb073a5bf5cc157918bfa0f62b394)
2020-03-06egl: Fix A2RGB10 platform_{device,surfaceless} PBuffer configs.Mathias Fröhlich5-15/+82
The __DRI_IMAGE_FORMAT_* part wants to be handled for the *101010 type formats as well. Factor out a common function for that task. That again makes the piglit egl_ext_device_base test work again for hardware drivers. v2: Factor out a common function for that task. v3: dri2_pbuffer_visuals -> dri2_pbuffer_visuals Reviewed-by: Emil Velikov <> Fixes: 9acb94b6236 "egl: Enable 10bpc EGLConfigs for platform_{device,surfaceless}" Signed-off-by: Mathias Fröhlich <> Part-of: <> (cherry picked from commit d32c458de76c9e0cc08c9ee1a7de23c3fca69298)
2020-03-06i965: toggle on EXT_texture_norm16Tapani Pälli2-1/+2
Fixes: 7f467d4f738 ("mesa: GL_EXT_texture_norm16 extension plumbing") Signed-off-by: Tapani Pälli <> Reviewed-by: Lionel Landwerlin <> Part-of: <> (cherry picked from commit 200a83a98394ce292fd1cdbd6e9166502379b5c9)
2020-03-06mesa: introduce boolean toggle for EXT_texture_norm16Tapani Pälli3-2/+3
Fixes: 7f467d4f738 ("mesa: GL_EXT_texture_norm16 extension plumbing") Signed-off-by: Tapani Pälli <> Reviewed-by: Lionel Landwerlin <> Part-of: <> (cherry picked from commit dc531869a918dc75ffc09b38851b750ba62673f8)
2020-03-06.pick_status.json: Update to 74e4cda64b9d114321216eefe536f80644b0f0fdErik Faye-Lund2-3/+1
2020-03-06.pick_status.json: Mark 9fea90ad5170dd64376d22a14ac88c392813c96c as denominatedEric Engestrom1-1/+1
2020-03-06.pick_status.json: Mark b9773631d3e79e2310ed0eb274b4dd9426205066 as denominatedEric Engestrom1-1/+1
2020-03-06.pick_status.json: Mark ddd767387f336ed1578f171a2af4ca33c564d7f3 as denominatedEric Engestrom1-1/+1
2020-03-06.pick_status.json: Mark 34fd894e42ae1ec9d35bf9c4f05364b03dd4a223 as denominatedEric Engestrom1-1/+1
2020-03-06.pick_status.json: Mark 5ea23ba659adc05ff75ca7a4c9d1bd01db889ddd as denominatedEric Engestrom1-1/+1
2020-03-06.pick_status.json: Mark dba71de5c63617677fe44558f995d35fad643413 as denominatedEric Engestrom1-1/+1
2020-03-06.pick_status.json: Update to 74e4cda64b9d114321216eefe536f80644b0f0fdEric Engestrom1-0/+3780
2020-02-25v3d: Sync on last CS when non-compute stage uses resource written by CSJose Maria Casanova Crespo7-26/+77
When a resource is written by a compute shader and then used by a non-compute stage we sync on last compute job to guarantee that the resource has been completely written when the next stage reads resources. In the other cases how flushes are done guarantee the serialization of the writes and reads. To reproduce the failure the following tests should be executed in batch as last test don't fail when run isolated: KHR-GLES31.core.shader_image_load_store.basic-allFormats-load-fs KHR-GLES31.core.shader_image_load_store.basic-allFormats-loadStoreComputeStage KHR-GLES31.core.shader_image_load_store.basic-allTargets-load-cs KHR-GLES31.core.shader_image_load_store.advanced-sync-vertexArray v2: Use fence dep instead of bo_wait (Eric Anholt) v3: Rename struct names (Iago Toral) Document why is not needed on graphics->compute case. (Iago Toral) Follow same code pattern of the other update of in_sync_bcl. v4: Fixed comments style. (Iago Toral) Fixes KHR-GLES31.core.shader_image_load_store.advanced-sync-vertexArray Reviewed-by: Iago Toral Quiroga <> CC: 19.3 20.0 <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 01496e3d1ea0370af03e6645dbd2b864c2ace94c)
2020-02-25.pick_status.json: Update to 01496e3d1ea0370af03e6645dbd2b864c2ace94cDylan Baker1-0/+288
2020-02-24spirv: Remove outdated SPIR-V decoration warningsArcady Goldmints-Orlov2-4/+3
spirv_to_nir warns if it encounters XFB decorations and errors if it encounters a Stream decoration with value other than 0, despite the fact that these decorations are in fact handled correctly. Fixes dEQP-VK.transform_feedback.simple.query_1_* Fixes: cd4a14be06 "spirv: Handle XFB variable decorations" Reviewed-by: Jason Ekstrand <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 5f3cbbd958d14924dded0e0a0908127f6bfa006d)
2020-02-24util: promote u_debug_memory.c to src/utilErik Faye-Lund6-3/+3
When os_memory_debug.h was promoted to src/util, this source-file on which it depends on when the debug-flag is set on windows was left out. So let's move this also. It doesn't seem there's any way of triggering this issue right now, but it seems better to correct this to avoid this from biting us in the ass in the future. Fixes: 88c4680b5a5 ("util: promote u_memory to src/util") Reviewed-by: Dylan Baker <dylan@pnwbakers> Reviewed-by: Jose Fonseca <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 2e3318b151abddd456077ec0eed13f95245ce344)
2020-02-24.pick_status.json: Update to e4baff90812d799d586296fcad992ddcc553c359Dylan Baker1-0/+567
2020-02-20i965: Do not generate D16 B5G6R5_UNORM configs on gen < 8Danylo Piliaiev2-12/+24
We don't support MESA_FORMAT_Z_UNORM16 before Gen8, see intel_screen_init_surface_formats. As a consequence disables B5G6R5_UNORM configs with depth on gen < 6. Closes: CC: <> Signed-off-by: Danylo Piliaiev <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 5bfd363be4c957c1f7b5c1f3069346f2bce2cd5a)
2020-02-20intel/fs: Correctly handle multiply of fsign with a source modifierIan Romanick2-1/+11
The other source of the multiply will be interpreted as a uint32_t in an XOR instruction. Any source modifiers with either not be interpreted at all or will be misinterpreted due to the differing types. If the other operand of the multiplication has a source modifier, just emit an extra move to resolve the source modifiers. The negation source modifier problem is difficult to reproduce due to an algebraic optimization that changes (-a*b) to -(a*b). However, changes in MR !1359 push the negations back down. On Gen7+ it might be possible to do slightly better for an abs() source modifier by using BFI2 as a glorified copysign(). On Gen8+ it might be possible to do slightly better for a neg() source modifier by emitting (~a ^ b). There were no shader-db changes on any Intel platform, so I think we can deal with that problem when it arises. See also piglit!224. Fixes: 06d2c116415 ("intel/fs: Add a scale factor to emit_fsign") Reviewed-by: Matt Turner <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 273b8cd1ca286e2f43b4a464a391fdcaac49f077)
2020-02-20st/vdpau: Only call is_video_format_supported hook if neededMichel Dänzer2-8/+8
Namely only if *is_supported is true, otherwise the hook result can't affect it. Avoids ../src/gallium/state_trackers/vdpau/vdpau_private.h:138: FormatYCBCRToPipe: Assertion `0' failed. with assertions enabled. Fixes: 5d5b414a7b84 "st/vdpau: fix chroma_format handling in VideoSurfaceQueryGetPutBitsYCbCrCapabilities" Reviewed-by: Pierre-Eric Pelloux-Prayer <> Reviewed-by: Christian König <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 7e6010106fb3c4eb5436de869183e857243c1006)
2020-02-20mesa: fix immediate mode with tessellation and varying patch verticesMarek Olšák2-1/+3
Cc: 19.3 20.0 <> Reviewed-by: Mathias Fröhlich <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 2e05a280b6b6d334388e3824bd82472ccbf33252)
2020-02-20intel/gen12: Take into account opcode when decoding SWSBCaio Marcelo de Oliveira Filho3-4/+8
The interpretation of the fields is different depending whether the instruction is a SEND/MATH or not. This fixes the disassembly output for non-SEND/MATH instructions that have both in-order and out-of-order dependencies. Their dependencies were wrongly represented as `@A $B` when the correct would be `@A $B.dst`. Fixes: 6154cdf924f ("intel/eu/gen12: Add auxiliary type to represent SWSB information during codegen.") Fixes: 83612c01271 ("intel/disasm/gen12: Disassemble software scoreboard information.") Acked-by: Francisco Jerez <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 79788b8f7f07460af8467931501380e47b485e36)
2020-02-20.pick_status.json: Update to 8291d728dc997e87b4d2e4e451692643a1dba881Dylan Baker1-0/+522
2020-02-18zink: confused compilation macro usage for zink in target helpers.luc2-2/+2
Fixes: 8d46e35d16e ("zink: introduce opengl over vulkan") Reviewed-by: Erik Faye-Lund <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 692093fbdc93343dbe500128fdd23167d73036d9)
2020-02-18egl: Implement getImage/putImage on pbuffer swrast.Mathias Fröhlich5-6/+97
This change adds getImage/putImage callbacks to the swrast pbuffer loader extension. This fixes a recent crash with Weston as well as a crashing test with classic swrast without an official gitlab issue. v2: Determine bytes per pixel differently and fix non X11 builds. v3: Plug memory leak and fix crash on out of bounds access. (Daniel Stone) v4: Follow the code structure of the wayland get/put image implementation - hopefully being more obvious. Handle 64 bits formats. Use BufferSize directly. (Emil Velikov) v5: Change pixel size computation. (Eric Engestrom) Reviewed-by: Daniel Stone <> Reviewed-by: Emil Velikov <> Closes: Fixes: d6edccee8da "egl: add EGL_platform_device support" Signed-off-by: Mathias Fröhlich <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit c7617d8908a970124321ce731b43d5996c3c5775)
2020-02-18intel/fs/gen12: Fixup/simplify SWSB annotations of SIMD32 scratch writes.Francisco Jerez2-8/+4
Found by inspection. Existing code was trying to avoid assuming that an SBID had been assigned to the virtual instruction, but synchronizing the header setup with respect to the previous SIMD16 SEND by using SYNC.ALLRD doesn't really seem possible unless the SEND instruction had been assigned an SBID. Assert-fail instead if no SBID has been allocated. Fixes: 15e3a0d9d264becc "intel/eu/gen12: Set SWSB annotations in hand-crafted assembly." Reviewed-by: Caio Marcelo de Oliveira Filho <> Cc: 20.0 <> (cherry picked from commit 4e4e8d793f050eac84f2a850ab2e5c24c4c459ac)
2020-02-18radeonsi: don't wait for shader compilation to finish when destroying a contextMarek Olšák2-4/+1
This was a hack for glsl_types deinitialization and it predates the proper fix, which was the addition of glsl_type_singleton_decref. This fixes a crash when the context is destroyed via the atexit handler. Cc: 19.3 20.0 <> Reviewed-by: Timothy Arceri <> Tested-by: Marge Bot <> Part-of: <> (cherry picked from commit 7e2b4bf256610cc016202893d7b4b4ef60b25b53)
2020-02-18.pick_status.json: Update to bee5c9b0dc13dbae0ccf124124eaccebf7f2a435Dylan Baker1-0/+1071