kc3-lang/angle/src

src

Log

Author Commit Date CI Message
Matthew Denton b9cec916 2025-08-12T19:44:18 WGSL: default uniforms gathered in interface block The default uniform struct was being manually output by OutputUniformBlocksAndSamplers(), which did not add the appropriate @align(16) annotations to lay the struct out according to WGSL's uniform address space layout requirements. This CL uses Vulkan's method of gathering the default uniforms into an interface block. The interface block will be output normally by the traverser, including @align() annotations. The variable declaring an instance of the interface block is still output by OutputUniformBlocksAndSamplers() because it needs special @group() and @binding() annotations. Bug: angleproject:376553328 Change-Id: Ib3be7d51ffedefe5ec579a1c9aaf0a535146c694 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6795028 Commit-Queue: Matthew Denton <mpdenton@chromium.org> Reviewed-by: Liza Burakova <liza@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Cody Northrop 17e99163 2025-08-25T21:19:08 Tests: Add Ace Racer trace Test: angle_trace_tests --gtest_filter="*ace_racer*" Bug: b/441170882 Change-Id: I8a244b2db1c9e62bd401e40ac911e868088c7dba Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6882667 Auto-Submit: Cody Northrop <cnorthrop@google.com> Reviewed-by: Charlie Lao <cclao@google.com> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Tom Sepez 25390156 2025-08-21T00:13:19 Suppress unsafe buffers on a file-by-file basis in src/ [1 of N] In this CL, we suppress many files but stop short of actually enabling the warning by not removing the line from the unsafe_buffers_paths.txt file. That will happen in a follow-on CL, along with resolving any stragglers missed here. This is mostly a manual change so as to familiarize myself with the kinds of issues faced by the Angle codebase when applying buffer safety warnings. -- Re-generate affected hashes. -- Clang-format applied to all changed files. -- Add a few missing .reserve() calls to vectors as noticed. -- Fix some mismatches between file names and header comments. -- Be more consistent with header comment format (blank lines and trailing //-only lines when a filename comment adjoins license boilerplate). Bug: b/436880895 Change-Id: I3bde5cc2059acbe8345057289214f1a26f1c34aa Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6869022 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Sungyong Choi 616a341b 2025-08-26T13:56:40 Vulkan: Set EGL_NATIVE_VISUAL_ID for EGL_KHR_platform_gbm Spec requires EGLConfig on GBM platform to report a GBM color format in EGL_NATIVE_VISUAL_ID. Previously this field was left unset. This change ensures compliance by assigning the appropriate GBM color format. Bug: angleproject:7829 Change-Id: I6c5a23374824c444e0de0b6592ea31eb855722e4 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6882716 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Yuxiang Qian 611f9ac0 2025-08-20T16:10:13 Lower the cap of max element index Currently, ANGLE follows vulkan spec and returns UINT_MAX -1 for GL_MAX_ELEMENT_INDEX. However, ANGLE stores vertex index in int, so even if we use index a little less than the limit, ANGLE would still catch the overflow error. Modify the cap to 2^30 - 1 to make it safer. An end2end test is also added. Bug: angleproject:440227026 Change-Id: Ie089c9847f3908f0e8a1b7c4b83c8fea1cde262e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6870270 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Mark Lobodzinski 19483cc6 2025-08-25T19:27:26 Tests: Add Bingo Blitz ANGLE trace Test: angle_trace_tests --gtest_filter=*bingo_blitz Bug: b/441137178 Change-Id: Ia463fbc93090d2c55a3f35f7cd34ca6267273c7f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6882619 Reviewed-by: Cody Northrop <cnorthrop@google.com> Auto-Submit: Mark Łobodziński <mark@lunarg.com> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Shahbaz Youssefi 677d8281 2025-08-25T15:59:46 Vulkan: Pass ContextVk to view-creation functions In preparation for moving the ycbcr conversion cache to the share group. This change is a no-op. Bug: angleproject:440364873 Change-Id: I0c18062259b07813dd04ec02650bb6fab48947ad Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6879204 Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Charlie Lao <cclao@google.com> Commit-Queue: Charlie Lao <cclao@google.com>
kyongjoo.oh 2b45ec65 2025-08-24T12:26:12 CL/VK: Fix default LWS when GWS is less than 64 In the case where LWS is NULL and GWS is smaller than 64, the runtime produces a useless uniform region info. Thus, this change clamps LWS to the GWS to avoid this edge-case. Bug: angleproject:42267067 Change-Id: I418c7d503918f4f082825924b463c266bfd862dc Signed-off-by: kyongjoo.oh <kyongjoo.oh@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6877794 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Austin Annestrand <a.annestrand@samsung.com>
Shahbaz Youssefi fee2bf73 2025-08-25T15:54:50 Vulkan: Limit EGL_NV_context_priority_realtime to Android Device creation fails on Linux when a REALTIME queue is requested, likely due to permission issues. Bug: angleproject:397474813 Change-Id: I0451db1dac9cbf2795439b251274f1a8057cbd6e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6880530 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
hoonee.cho 5fed937e 2025-08-24T12:21:53 CL/VK: Add a compile option "-inline-entry-points" When processing the -cl-std=CL2.0 option, ensure that the processed options string includes both "-cl-std=CL2.0" and "-inline-entry-points". This change addresses a CLSPV's compiler option check failure where the -inline-entry-points flag was not being added when the -cl-std=CL2.0 (using generic address space) was specified. Bug: angleproject:440814992 Change-Id: Ie3e8907e367f21830e58cd5c405de06ba2cbd1da Signed-off-by: hoonee.cho <hoonee.cho@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6877793 Commit-Queue: Austin Annestrand <a.annestrand@samsung.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Yang Gu 943aed8c 2025-08-25T17:37:34 Skip fake device with remote desktop on Windows When connecting with remote desktop on Windows, a fake device with same vendorId, deviceId, and driverVersion may be reported. This commit adds logic to skip such device to fix end2end tests. Bug: angleproject:440892567 Change-Id: Ie3b602385424bf4a77a302793b34eb06b610bbc9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6878724 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Panfeng Hou d65058be 2025-08-18T11:01:11 Vulkan: Add support for GL_EXT_fragment_shading_rate_primitive Add support for built-in GL_EXT_fragment_shading_rate_primitive and tests. Bug: angleproject:397255625 Change-Id: I0f37a87a73e869a91e39402f01e53166b359ec3c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6850353 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Auto-Submit: Panfeng Hou <panfeng.hou@arm.com> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Reviewed-by: Charlie Lao <cclao@google.com> Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Yuxin Hu f63751db 2025-08-22T10:34:45 Fix the UniformSortComparator not following strict weak order If one uniform is lowp and the other uniform is mediump, then the compare(a, b) and compare (b, a) both return true. This violates the strict weak order rule where: If comp(a, b) == true then comp(b, a) == false This change fixes the bug. Bug: angleproject:405795981 Change-Id: I47e2668fc0fa87ecea9d3f9bce511c9823a39929 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6876519 Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Jisun Lee 0af6e4ee 2025-08-13T06:16:01 Vulkan: Check format compatibility to enable mip levels If redefining the texture level is needed while respecifying the image storage, the base level could be incompatible with the max level count. Enabling the whole level count can cause a null pointer dereference when copying an image that does not support `pixelReadFunction`. To handle the problem, this CL checks the format compatibility and enables only the effective base level. Bug: b/435727832 Test: angle_end2end_tests --gtest_filter=DXT1CompressedTextureTest.* Test: angle_end2end_tests --gtest_filter=Texture2DTestES3.* Change-Id: I01060332f282ef9b2856390d0d7296811a464795 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6813835 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Yuxiang Qian 51cffa97 2025-08-13T14:23:51 Implement EGL_NV_context_priority_realtime In this extension, new attribute value EGL_CONTEXT_PRIORITY_REALTIME_NV is accepted for EGL_CONTEXT_PRIORITY_LEVEL_IMG. In ANGLE, this extension should depend on VK_EXT_global_priority and VK_EXT_global_priority_query. The implementation of creating device queue is also modified for this new attribute value. New end2end test is added to check if this new attribute works. Bug: angleproject:397474813 Change-Id: I68316fe371f5a495dbc78b106fc89787be18e086 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6857673 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
angle-autoroll 9280c6b9 2025-08-20T08:04:15 Roll VK-GL-CTS from ad59a18f2ce0 to 0a3dad34ebb4 (60 revisions) Skip dEQP-GLES31.functional.copy_image.non_compressed.viewclass_16_bits.*.texture2d_array_to_texture3d dEQP-GLES31.functional.copy_image.non_compressed.viewclass_32_bits.* dEQP-GLES31.functional.copy_image.non_compressed.viewclass_64_bits.*.texture2d_array_to_texture3d on Pixel 4 Vulkan https://chromium.googlesource.com/external/github.com/KhronosGroup/VK-GL-CTS.git/+log/ad59a18f2ce0..0a3dad34ebb4 2025-08-15 lorenzo@khronosgroup.org Merge "Merge vk-gl-cts/github-main into vk-gl-cts/main" into main 2025-08-15 lorenzo@khronosgroup.org Merge vk-gl-cts/vulkan-cts-1.4.3 into vk-gl-cts/main 2025-08-14 mateusz.bahyrycz@mobica.com Add tests for copying invalid binding 2025-08-14 ziga@lunarg.com Add HIC tests for all formats and reduce other variations 2025-08-14 rgarcia@igalia.com Test buffer and image capture&replay re-usage with descriptor buffer 2025-08-14 mateusz.bahyrycz@mobica.com Add tests for writting unused and invalid binding 2025-08-14 mateusz.bahyrycz@mobica.com Add tests for image to buffer copy for color formats 2025-08-14 rgarcia@igalia.com Modify suspend/resume FDM offset tests to use consistent offsets 2025-08-14 ziga@lunarg.com Fix image barrier issues in render_to_image tests 2025-08-14 ziga@lunarg.com Fix invalid access mask in buffer_view compute tests 2025-08-14 ziga@lunarg.com Fix image layout transitions in image 2d_array_compatible tests 2025-08-14 lorenzo@khronosgroup.org Update SPIR-V sources: replace gl_SemanticsRelaxed in atomicStore when used with non-zero storage class semantics 2025-08-14 piotr.byszewski@mobica.com Fix hlsl_functionality1.counter_buffer on spirv-tools TOT 2025-08-14 syoussefi@google.com GLES: Fix debug tests not clearing the error (if no message) 2025-08-11 lorenzo@khronosgroup.org Merge remote-tracking branch 'vk-gl-cts/dev/VK_KHR_shader_untyped_pointers' into main 2025-08-11 lorenzo@khronosgroup.org Update Vulkan headers to v1.4.325 2025-08-08 ziga@lunarg.com Test presenting image from retired swapchain 2025-08-08 piotr.byszewski@mobica.com Test computing LOD from explicit cubemap derivatives 2025-08-08 gleese@broadcom.com Enable some low-bit-depth bitfield tests without maint9 2025-08-08 ziga@lunarg.com Fix maintenance9 check in query pool get_create_reset tests 2025-08-08 ziga@lunarg.com Add missing barrier in sample locations tests 2025-08-08 ziga@lunarg.com Fix layer count in geometry layered tests 2025-08-08 ziga@lunarg.com Fix reading from image in undefined layout in query pool tests 2025-08-08 david.rosca@amd.com Don't require all modifiers to be compatible with video 2025-08-08 lorenzo@khronosgroup.org Complete automatic JSON generation for all remaining KHR extensions 2025-08-08 lorenzo@khronosgroup.org Merge vk-gl-cts/vulkan-cts-1.4.3 into vk-gl-cts/main 2025-08-05 lorenzo@khronosgroup.org Update Vulkan headers to 1.4.324 2025-08-01 zzoon@igalia.com Replace the customized parser with one in VVS Khronos. 2025-08-01 lorenzo@khronosgroup.org Merge vk-gl-cts/opengl-es-cts-3.2.12 into vk-gl-cts/main 2025-08-01 lorenzo@khronosgroup.org Add JSON files for VK_KHR_present_id2 and VK_KHR_present_wait2 2025-08-01 lorenzo@khronosgroup.org Merge remote-tracking branch 'vk-gl-cts/dev/VK_KHR_present_id2' into main 2025-08-01 abdolrashidi@google.com Shift fixed-point value if between accepted values 2025-08-01 rgarcia@igalia.com Test descriptor buffer with 3-plane image formats 2025-08-01 rgarcia@igalia.com Check baseTriangle is honored in opacity micromap tests 2025-08-01 rgarcia@igalia.com Test large number of compute and task DGC sequences 2025-08-01 rgarcia@igalia.com Check maxMipLevels/maxArrayLayers in compressed image copy tests 2025-08-01 david.rosca@amd.com Don't call vkCmdFillBuffer on unsupported queues 2025-08-01 michal.jakubek@mobica.com Reduce link time and disk space for Capabilities 2025-08-01 michal.jakubek@mobica.com Unwritten attachments with dual-source blending 2025-08-01 lorenzo@khronosgroup.org Merge vk-gl-cts/vulkan-cts-1.4.3 into vk-gl-cts/main 2025-08-01 lorenzo@khronosgroup.org Auto-generate JSON for VK_EXT_fragment_density_map_offset 2025-08-01 lorenzo@khronosgroup.org Merge remote-tracking branch 'vk-gl-cts/dev/VK_EXT_fragment_density_map_offset' into main 2025-07-25 lorenzo@khronosgroup.org Merge vk-gl-cts/opengl-cts-4.6.6 into vk-gl-cts/main 2025-07-25 lorenzo@khronosgroup.org Merge vk-gl-cts/opengl-es-cts-3.2.12 into vk-gl-cts/main 2025-07-25 ziga@lunarg.com Fix bad merge 2025-07-25 ziga@lunarg.com Add tests for ms input attachment feedback loop 2025-07-25 ziga@lunarg.com Test OpImageSampleImplicitLod with RelaxedPrecision and ycbcr image 2025-07-25 nikigupta@google.com Add test for gl_Position not declared inside built-in block in geometry shader 2025-07-25 marcin.zajac@cognizant.com Degenerate triangles in AS update 2025-07-25 david.rosca@amd.com Skip unsupported AV1 encode tiling tests according to maxTiles cap 2025-07-25 ziga@lunarg.com Fix errors in maintenance9 2d_array_compatible image tests 2025-07-25 ziga@lunarg.com Test writes to partially bound ds attachments are discarded 2025-07-25 gleese@broadcom.com Fix identity checking in subgroup scan tests 2025-07-25 lorenzo@khronosgroup.org Add auto-generation for existing KHR extensions (simple ones) 2025-07-25 scerveau@igalia.com VideoEncodeTestsAV1: keep same name for input filename 2025-07-25 lorenzo@khronosgroup.org Merge vk-gl-cts/vulkan-cts-1.4.3 into vk-gl-cts/main 2025-07-24 lorenzo@khronosgroup.org Merge vk-gl-cts/vulkan-cts-1.4.3 into vk-gl-cts/main 2025-07-24 lorenzo@khronosgroup.org Add logging and checking for missing JSONs for VK_KHR extensions 2025-07-24 lorenzo@khronosgroup.org Add VK_KHR_video_encode_intra_refresh JSON 2025-07-23 okuznetsov@google.com Add full image level copy tests to es31fCopyImageTests If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/vk-gl-cts-angle-autoroll Please CC angle-team@google.com,ynovikov@google.com on the revert to ensure that a human is aware of the problem. To file a bug in ANGLE: https://bugs.chromium.org/p/angleproject/issues/entry To report a problem with the AutoRoller itself, please file a bug: https://issues.skia.org/issues/new?component=1389291&template=1850622 Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md Bug: angleproject:440743718 Change-Id: I9313b003dc8ae2a1dd64d264fa2b7847ae2b02ad Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6865062 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
jh.eun 061a43c6 2025-08-13T16:35:20 CL/Vulkan: Implement cl_khr_priority_hints cl::CommandQueue contains queue priority from clCreateCommandQueueWithProperties. Similar to EGL_IMG_context_priority, a queue with a pre-assigned priority is selected when submission. Bug: angleproject:433980937 Tests-Passing: OCLCTS.test_api queue_hint Change-Id: I3a7bcc62e5d8186b51f771ab2c01cc18ddfcd195 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6785088 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Austin Annestrand <a.annestrand@samsung.com>
Mark Lobodzinski 8d43d9b9 2025-08-21T21:15:06 Tests: Add Hunter Assassin ANGLE trace Test: angle_trace_tests --gtest_filter=*hunter_assassin Bug: b/440565036 Change-Id: I502db22ba7967406f61ffa1437865992b05c1a34 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6876526 Auto-Submit: Mark Łobodziński <mark@lunarg.com> Commit-Queue: Cody Northrop <cnorthrop@google.com> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Raja Wasim Husain 8ef76b6a 2025-08-04T12:34:17 CL/Vulkan: cl_khr_external_memory extension (pt.2) - Make this extension visible if Vulkan implementation support features supportsExternalMemoryFd and/or supportsExternalMemoryDmaBuf - Implemented APIs clEnqueueAcquireExternalMemObjectsKHR and clEnqueueReleaseExternalMemObjectsKHR - Updated clCreateBufferWithProperties to handle external memory file descriptor. Bug: angleproject:378017028 Change-Id: I1751982c8e9b2cd07b7e251cc54db5dcd1bcda20 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6843980 Commit-Queue: Austin Annestrand <a.annestrand@samsung.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Yuxin Hu d4d42356 2025-08-22T11:30:42 Remove the duplicated branch There is no code difference in the if and the else branches, remove the duplicated branch. Bug: angleproject:405795981 Change-Id: If593476d2b5cd02567ed1896625072128cf21b0f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6876346 Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Yuxiang Qian 1f83d739 2025-08-01T16:59:11 Expose VK_EXT_global_priority and VK_EXT_global_priority_query To implement EGL_NV_context_priority_realtime, VK_EXT_global_priority and VK_EXT_global_priority_query should be exposed. With these extensions, we can query supported global priority and set global priority for the Vulkan queue. Bug: angleproject:397474813 Change-Id: I83f9f19d46f3c81b02246aae2892f7a9c1a09d1d Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6857671 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Yuxin Hu 01cfb98f 2025-08-14T14:17:28 Add a new property to LinkedUniform struct The new property isTransformedFloat will be used by front end API to determine if we should transform 32-bit float to 16-bit float before writing to memory, and transform 16-bit float to 32-bit float after reading from memory. The property value will be passed from UsedUniform (child struct of ShaderVariable) object in UniformLinkder::getResults(). This CL is no op. Bug: angleproject:405795981 Change-Id: Ic4908712f5fa56cb657206a25d9d618b51ac6322 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6851554 Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Yuxin Hu 20dca0b9 2025-08-14T13:54:37 Add a new TInterfaceBlock class member to hint ANGLE default Uniform In later change, we will transform floats in ANGLE default uniform struct from 32-bit to 16-bit in SPIRV. The new class member mIsDefaultUniformBlock will help us identify this special struct. This change is no op. Bug: angleproject:405795981 Change-Id: I939caf7491ab5d7de130ba6a2fd8b1202958c375 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6852398 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Yuly Novikov ef80e82f 2025-08-22T15:07:59 Skip slow test on Mac Metal AMD GLSLTest_ES3.ManySamplerFieldsInStructComplex/ES3_Metal takes 38 seconds and causes batch timeouts. Bug: angleproject:440536251 Change-Id: I53ad821bc4da4696609f9b8c0f5cbee1ed94dbd1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6874571 Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com> Auto-Submit: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Geoff Lang 86a8d11c 2025-08-11T16:45:39 Metal: Fix potential incorrect format used for texSubImage TextureMtl caches the most recent format used for a texture redefinition in mFormat. During subImage calls, this format may not be the same as the image being uploaded to. Pass ImageDefinitionMtl to these functions which reference mFormat, it contains the format of the image being updated. MacOS OpenGL is skipped because the driver generates errors in glCompressedTexImage2D. Bug: chromium:435683799 Change-Id: Idec6f71870c2d376cad3a5e3628b957009bdced9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6838878 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Quyen Le <lehoangquyen@chromium.org> Commit-Queue: Geoff Lang <geofflang@chromium.org>
Mark Lobodzinski d0ba2fb9 2025-08-20T12:16:07 Tests: Add Onmyoji ANGLE trace Test: angle_trace_tests --gtest_filter=*onmyoji Bug: b/440323747 Change-Id: Ia549dcfa370fac23785364adb717babde8b8ca4d Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6871872 Commit-Queue: Mark Łobodziński <mark@lunarg.com> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Raja Wasim Husain 6c160cad 2025-08-04T12:32:36 CL/Vulkan: cl_khr_external_memory extension (pt.1) - Make this extension visible if Vulkan implementation support features supportsExternalMemoryFd and/or supportsExternalMemoryDmaBuf - Implemented APIs clEnqueueAcquireExternalMemObjectsKHR and clEnqueueReleaseExternalMemObjectsKHR - Updated clCreateBufferWithProperties to handle external memory file descriptor. Bug: angleproject:378017028 Change-Id: Idff08808bc0227b5f98b84c2086b68541665146d Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6785087 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Austin Annestrand <a.annestrand@samsung.com>
Mark Lobodzinski 995c4c4d 2025-08-20T12:14:47 Capture/Replay: Target GLES3.1 to avoid SSO limitations Retracing with get_min_reqs can produce results which limit trace portability. Modify retrace script to detect when GL_EXT_separate_shader_objects is required, and then to increase min_version to 3.1 where this ext was moved to core and remove the ext from the required list. Test: angle_trace_tests --gtest_filter=*onmyoji --use-gl=native Bug: b/440102792 Change-Id: I36258fd7b9b5c5aa3e87ad9908849ef2a900e898 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6866639 Auto-Submit: Mark Łobodziński <mark@lunarg.com> Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Liza Burakova 03068e0c 2025-08-20T17:36:22 [WebGPU] Implement direct copy for images. This CL implements CopyImage specifically in cases where the source is the same size and format as the destination. Bug: angleproject:438268609 Change-Id: Icf5f3685bb2f49874d996883c839515314f1c189 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6842317 Commit-Queue: Liza Burakova <liza@chromium.org> Reviewed-by: Matthew Denton <mpdenton@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tim Van Patten 5d9d9b9b 2025-08-19T17:12:03 test: Use eglGetPlatformDisplay() From the EGL 1.5 spec: Appendix F Version 1.5 EGL version 1.5 was voted out of the Khronos Technical Working Group on January 31, 2014, and formally approved by the Khronos Board of Promoters on March 14, 2014. EGL 1.5 is the sixth release of EGL. It introduces the following new features (the EGL extension(s) each feature is based on are also shown parenthetically): * Platform support: – Providing a mechanism for support of multiple platforms (such as window systems or offscreen rendering frameworks) in a single EGL implementation at runtime (EGL_EXT_platform_base). Until https://crrev.com/c/6552257, many tests used eglGetPlatformDisplayEXT() which is provided by the EGL extension EGL_EXT_platform_base. With the promotion of the EGL_EXT_platform_base functions to core EGL in version 1.5 and ANGLE supporting EGL 1.5 (as of at least 2019), the calls were updated to use eglGetPlatformDisplay(). Unfortunately, EGLContextPassthroughShadersTest was missed. Bug: b/409384875 Test: angle_end2end_tests Change-Id: I5c620bce98c8e76113588f4c94b77d95a5223171 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6862841 Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Tim Van Patten <timvp@google.com>
Raja Wasim Husain 6ff016fb 2025-03-31T09:52:14 CL/Vulkan: Update caps for image1d_buffer max size Updated max number of pixels for a 1D image created from a buffer object to come from VK implementation. Bug: angleproject:406022415 Change-Id: I56e9821895245c3a3655ca0c4afbe49ed89baded Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6386422 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Gowtham Tammana <g.tammana@samsung.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Austin Annestrand <a.annestrand@samsung.com>
Sungyong Choi da5a5628 2025-08-11T10:53:06 bitset_utils: Fix GCC shift-count-overflow warning GCC triggers a -Werror=shift-count-overflow when kDefaultBitSetSize is 64, due to the constructor compiling a branch with `value >>= 64`. Although this path is never executed at runtime, GCC still inspects the code and emits the warning as an error. Refactored the constructor to: - Move common reset() into a single place - Split initialization into initFromValue<S>() overloads enabled only for S < 64 or S >= 64 via std::enable_if_t - Avoid generating any shift code when S >= 64 This change ensures GCC does not instantiate invalid shift operations and keeps the initialization logic cleaner. BUG: angleproject:438226513 Change-Id: I1ed0b42540f1bfce469d5738757045bc6d92ac61 Signed-off-by: Sungyong Choi <sywow.choi@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845475 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Charlie Lao <cclao@google.com>
Sungyong Choi 68baeeeb 2025-08-12T15:51:07 State: avoid GCC -Waddress by using constexpr null-check helper GCC with -Waddress warns that the static_assert comparing `handlers[0] != nullptr` is always true when MakeDirtyObjectHandlers() fully initializes the array. During constant evaluation, handlers[0] is replaced with the specific member function address (&State::syncActiveTextures), which can never be null, making the comparison tautological. This change replaces the direct `handlers[0] != nullptr` check with a constexpr helper that iterates over all handlers and verifies none are null. By comparing each element in a loop, the check still runs at compile time (constexpr), but avoids a direct “function address vs nullptr” pattern that triggers -Waddress. BUG: angleproject:438226513 Change-Id: Iaa8441000f8b2aa28c44d17730ca223fd4d4595b Signed-off-by: Sungyong Choi <sywow.choi@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845477 Reviewed-by: Charlie Lao <cclao@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Sungyong Choi 39265327 2025-08-13T08:05:24 Fix BindDescriptorSetParams bit-field size and -Wpadded warning Replace pipelineBindPoint : 8 with full VkPipelineBindPoint to hold all enum values. Reorder members and adjust bit-field widths to pack within 32-bit units, adding explicit padding to eliminate compiler-inserted padding and silence -Wpadded. BUG: angleproject:438226513 Change-Id: I8950fcb10d5796bd8485791f63980a2905179324 Signed-off-by: Sungyong Choi <sywow.choi@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845479 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Alex Dean 9bd6785d 2025-08-12T15:19:08 Revert "Tests: Skip CapturedTestCL to unblock vulkan-deps roll" This reverts commit 2a5ad67c883a28436d7399d73a2ccdb9ef53c641. Additionally, this modifies CapturedTestCL.cpp so there is no failure coming from the vulkan-deps roll. Bug: angleproject:438220619 Bug: angleproject:438195622 Change-Id: I6cf2d11d7ff77ce4513948b22b8c59d970c7810b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6842222 Reviewed-by: Austin Annestrand <a.annestrand@samsung.com> Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Sungyong Choi 1622d4ad 2025-08-13T07:22:13 Fix -Wparentheses in unlockAndRelease Add parentheses around (valid && owns_lock()) to clarify the intended grouping and silence GCC -Wparentheses warning. BUG: angleproject:438226513 Change-Id: I0d4f3e0153c277c96ec85f255f844bab085f9550 Signed-off-by: Sungyong Choi <sywow.choi@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845478 Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuxin Hu <yuxinhu@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Sungyong Choi 119ffd44 2025-08-13T16:07:52 Fix -Wreturn-type in ShadowSamplerFunctionsTest::FunctionName Add UNREACHABLE() and a fallback return to ensure the non-void function always returns when switch does not cover all enum values. BUG: angleproject:438226513 Change-Id: I3904dd59b323b90930d9251b060ab7067ff412c7 Signed-off-by: Sungyong Choi <sywow.choi@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845486 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuxin Hu <yuxinhu@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Sungyong Choi 3474d4f3 2025-08-13T08:13:31 Guard overlayVk->onPresent call in drawOverlay GCC warns "'this' pointer is null" (-Wnonnull) when overlayVk could be null at the call site. Wrap the call with an explicit null check to avoid dereferencing a null overlay and silence the warning. BUG: angleproject:438226513 Change-Id: I3c719d703277c0a888eac9fb9edaaac77cf76a85 Signed-off-by: Sungyong Choi <sywow.choi@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845481 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuxin Hu <yuxinhu@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Sungyong Choi a06cb368 2025-08-12T11:45:43 Fix GCC -Wdangling-pointer warning in NodeStackGuard The previous implementation stored the address of a local member (oldNodeStack) into ConsList::tail, which triggers the warning even though the lifetime is safe within the guard's scope. This change allocates a copy of the previous nodeStack on the heap using std::make_unique, and stores that pointer in tail. This ensures the referenced ConsList object has a lifetime tied to the guard object itself and is not considered a dangling pointer by GCC's static analysis. BUG: angleproject:438226513 Change-Id: Ifea04515db7b1988dff6ec41c60799158b753128 Signed-off-by: Sungyong Choi <sywow.choi@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845476 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Sungyong Choi 2b8f216c 2025-08-13T08:44:34 Fix GCC -Wdeprecated-copy warning in ImageLoadContext Add an explicit copy assignment operator to ImageLoadContext to replace the implicitly-declared, deprecated version. This makes the intended assignment behavior explicit and silences the GCC warning when assigning a default-initialized value. BUG: angleproject:438226513 Change-Id: I44d9b8bf12c5e3172d30ce5413460d881e61449b Signed-off-by: Sungyong Choi <sywow.choi@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845482 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Sungyong Choi 7febacfe 2025-08-13T08:10:11 Remove unused compressionProp in TextureVk::initImage compressionProp was declared and sType initialized but never used, triggering -Werror=unused-but-set-variable with GCC. The query path uses compressionInfoVar instead, so drop the dead variable and its initialization. BUG: angleproject:438226513 Change-Id: Ib81f418a570b1a7fd858eec4cee4250db49cdd08 Signed-off-by: Sungyong Choi <sywow.choi@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845480 Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Sungyong Choi 6b03372e 2025-08-13T08:58:35 Fix GCC -Wconversion-null error in WriteParamValueReplay Replace insertion of NULL with the string literal "NULL" to prevent GCC -Wconversion-null. BUG: angleproject:438226513 Change-Id: I7048bc43f282d94e72a5e090176c805d2baa5b06 Signed-off-by: Sungyong Choi <sywow.choi@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845483 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Gowtham Tammana 4c834956 2025-04-09T20:30:43 CL/Vulkan: Add fp16/64 to extension list if supported The checks for fp16 and fp64 support are already setup on the vulkan backend and clspv side. Add them to the reported extension list. Also added "-d" flag to bruteforce smoke to toggle off 64-bit precision testing (i.e. just test single precision in smoke for now) Bug: angleproject:433980941 Change-Id: Icbbde7500cf43e5c0a256411f001252bfec5ad24 Signed-off-by: Gowtham Tammana <g.tammana@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6785092 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Austin Annestrand <a.annestrand@samsung.com>
Yuxin Hu 806c80ec 2025-08-14T13:41:58 Add a new ShaderVariable struct property Add a new ShaderVariable struct property isTransformedFP16FloatUniform. This property will be set by compiler when the corresponding shader variable data type in shader code is transformed from 32 bit to 16 bit. The property value be later passed to LinkedUniform so that the frontend knows if it should transform the data from 32 bit to 16 bit before writing to memory, and if it should transform the data from 16 bit to 32 bit after reading from memory. This change is no op. Bug: angleproject:405795981 Change-Id: Iaae15baa8ab9f7c3295071fa1e1ab04752bd69bb Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6852396 Auto-Submit: Yuxin Hu <yuxinhu@google.com> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Panfeng Hou 25f5e2cf 2025-08-13T17:03:19 Vulkan: Add support for built-in gl_PrimitiveShadingRateEXT Add support for built-in gl_PrimitiveShadingRateEXT and relative const value. Bug: angleproject:397255625 Change-Id: I25d51f5805697e137b232175ee790808c45850b7 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6845515 Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Mark Lobodzinski 240f6af8 2025-08-12T14:32:40 Capture/Replay: Pick up API SSBO binding changes GenerateLinkedProgram pulled uniform block bindings from inShaderBindings, but for trace this does not pick up API-side changes to the bindings with glUniformBlockBinding calls for example, resulting in missing geometry in some games. Test: angle_trace_tests --gtest_filter=*onmyoji Bug: b/438231427 Change-Id: I1b95c004c4fd856dc8b3a91cdef3e9251a275ca6 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6842217 Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Mark Łobodziński <mark@lunarg.com>
Yuxin Hu 1d8279a2 2025-08-13T13:35:02 Do not sort ANGLE internal uniforms or built-in uniforms On Mac, for webGL app on chrome, if the uniforms added by ANGLE are not placed at the top of the ANGLE_UserUniforms struct, the user defined uniforms in the same ANGLE_UserUniforms struct seem to be messed up and not read correctly by shader code. This change fixes it by moving the sortUniforms() function call to the place before we add ANGLE internal uniforms. That way ANGLE internal uniforms will be inserted later and stay on top of ANGLE_UserUniforms struct. Bug: chromium:437678149 Change-Id: I34a0d3dcb98c5fe30ad248381ecb5f7f78252275 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6847281 Auto-Submit: Yuxin Hu <yuxinhu@google.com> Commit-Queue: Yuxin Hu <yuxinhu@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Yuxin Hu 9034e29b 2025-08-14T13:25:39 Vulkan: add a feature flag controlling float uniforms 16-bit transform Add a feature flag that will guard whether we want to transform mediump / lowp float uniforms from 32-bit to 16-bit in SpirV. The feature flag is enabled when VK_KHR_16bit_storage extension is supported, and uniformAndStorageBuffer16BitAccess vulkan feature is supported. The feature flag is initialized to false in this change. It will be turned on in a later change when it is being used. Bug: angleproject:405795981 Change-Id: I37c77270fbd391cc97f58890065b66faf77766c1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6852395 Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Matthew Denton 63d8f74c 2025-08-12T19:39:03 WGSL: move DeclareDefaultUniforms() out of SPIR-V WGSL needs the same thing (default uniforms gathered into an interface block). Changes 3 things in the file move: 1. Now capable of adding a named interface block (Vulkan's is nameless) to hold the default uniforms. 2. Removes inactive uniforms 3. Deals with the case where there are no active default uniforms, by not declaring an interface block at all. Bug: angleproject:376553328 Change-Id: Ia25318b4db93068010c3c44414cd00803b783799 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6843929 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Matthew Denton <mpdenton@chromium.org>
Hareesh Karada 7ad67712 2025-08-13T16:33:03 CL/Vulkan: Enable cl_khr_integer_dot_product extension - Enabled extension string and corresponding opencl_c features - Updated packed_cl_enums.json with three new enums - Enabled support for this extension in function clGetDeviceInfo - Add support for CL_DEVICE_OPENCL_C_FEATURES API query through clGetDeviceInfo corresponding integer_dot_product extension - Updated vk_features.json with new ANGLE Feature supports_shader_integer_dot_product - The changes in autogen files are generated by running the run_code_generation.py script Tests-Passing: ocl_cts: test_integer_ops integer_dot_product ocl_cts: test_compiler features_macro VK-GL-CTS:dEQP-VK.spirv_assembly.instruction.compute.opsdotkhr.* dEQP-VK.spirv_assembly.instruction.compute.opudotkhr.* dEQP-VK.spirv_assembly.instruction.compute.opsudotkhr.* dEQP-VK.spirv_assembly.instruction.compute.opsdotaccsatkhr.* dEQP-VK.spirv_assembly.instruction.compute.opudotaccsatkhr.* dEQP-VK.spirv_assembly.instruction.compute.opsudotaccsatkhr.* Bug: angleproject:433980939 Change-Id: Iae7a7753d0556beeb85c254c61a37a0d047df058 Signed-off-by: Hareesh Karada <hareesh.k@samsung.com> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6785089 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Cody Northrop 2eef3701 2025-08-11T20:25:42 FrameCapture: Handle null texture in MEC When interating through a UnsafeResourceMapIter, we need to check for null entries before dereferencing. Test: Fortnite MEC Bug: b/180418810 Change-Id: I0dc234234f9bb219026b2b67febb32e8af19e8b1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6839286 Reviewed-by: Xiang Wang <xwxw@google.com> Auto-Submit: Cody Northrop <cnorthrop@google.com> Commit-Queue: Xiang Wang <xwxw@google.com>
Yuxin Hu edb05627 2025-08-12T12:17:36 Fix sort uniform bug Current sort doesn't take care of the case when both uniforms are struct specfiers. Given that struct types are easily broken by sort, make the sort not reordering uniforms if both are structs. Bug: b/437825940 Change-Id: Idda1810ac4234f7e1547735e4e09658ab0a57eed Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6842936 Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Tim Van Patten 899f3505 2025-08-07T09:59:19 TestSuite: Android: Call onCrashOrTimeout() on test crash Currently, when a test crashes on Android we don't get any results for tests that ran or for the crashing test itself. This makes it difficult for developers to debug the issue, since they don't know which test crashed. Add signal handlers which call onCrashOrTimeout() when a test crashes. Typically, crashes occur within the individual tests rather than in the test runner (TestSuite), so it should be "safe" to attempt to dump the results up to and including the crash to aid with debugging. This is still risky since things are in a bad state, but we expect (essentially) all crashes to be in the individual tests, ANGLE, or the underlying driver code, rather than in the TestSuite instance, so writing the results JSON file should have a good chance of succeeding. Worst case, we just crash again and get nothing (like we have before this change). Bug: b/279980674 Change-Id: Ifd5f3b97d5af731586e7bbf765bb60b3cb4f503b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6827986 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Cody Northrop <cnorthrop@google.com> Reviewed-by: Solti Ho <solti@google.com>
Charlie Lao cfeea900 2025-08-12T17:23:32 Vulkan: Don't require COLOR_ATTACHMENT_BLEND_BIT for YUV format The only way to render to yuv texture is via EXT_yuv_target, but EXT_yuv_target spec https://registry.khronos.org/OpenGL/extensions/EXT/EXT_YUV_target.txt explicitly said you have to disable blend: "Blending applies only if the color buffer has a fixed-point format and non YUV space format color buffer.". Thus when we check if format supports rendering, we should exclude VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BLEND_BIT if its YUV format. Bug: b/432398245 Change-Id: If5bfa58db30465ab707f63f34dd6487d8a03652f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6843926 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Junhua Chen ef4bc113 2025-08-07T14:20:08 Add DIR_METADATA to specific buganizer_public component_id Currently LUCI Analysis file bugs for angle test flaky issue under the top level Chromium component, example: http://crbug.com/401993176 These bugs should filed directly under the Public Trackers > ANGLE, to avoid engineers updating the component manually. Bug: b/437142709 Change-Id: Ie8fa4f02fea10c49cb94d736f674f655fe9f791b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6829151 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuxin Hu <yuxinhu@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Xiang Wang 9d1edb4e 2025-08-08T18:21:03 Fix the assert on texture target to allow GL_TEXTURE_CUBE_MAP_ARRAY Add anend2end test to verify glCopyImageSubData works. Bug: b/419863436 Test: TextureCubeTestES32.CopyImageSubDataCubeMapArray Change-Id: I2a569680ed20f05b4182045cb4bdf08d4c66d8df Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6832639 Reviewed-by: Cody Northrop <cnorthrop@google.com> Reviewed-by: Charlie Lao <cclao@google.com> Commit-Queue: Xiang Wang <xwxw@google.com>
Matthew Denton 7e28089b 2025-07-31T18:44:01 WGSL: support bvecs in uniforms WGSL does not allow booleans in the uniform address space. The last CL uses u32s to represent bools in the uniform address space, and this CL uses vecN<u32> to represent bvecN. Bug: angleproject:376553328 Change-Id: Ibf9f1fbf230ca03dd3d4661ba599e5fb1cb26ac4 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6808955 Reviewed-by: Liza Burakova <liza@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Matthew Denton <mpdenton@chromium.org>
Austin Annestrand f92d6f4b 2025-04-07T17:42:46 CL: Route API trace to logcat for Android For better logging when running Android APKs Bug: angleproject:433980940 Change-Id: Ib924262a92728d0a9711a5b98f051634341859d4 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6785090 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Austin Annestrand <a.annestrand@samsung.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Shufen Ma 5fb77159 2025-08-12T15:21:25 Add stubs for GL_ARM_texture_unnormalized_coordinates Bug: angleproject:352345942 Change-Id: I05e2b0c83f07bfe35eb58fc657b834f6253d4cb9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6839660 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Auto-Submit: Shufen Ma <Shufen.Ma@arm.com> Commit-Queue: Charlie Lao <cclao@google.com> Reviewed-by: Charlie Lao <cclao@google.com>
Shahbaz Youssefi 316e6831 2025-08-08T14:14:20 Vulkan: Support KHR versions of surface/swapchain_maint1 They are straight promotions of the EXT versions of the extensions. Bug: angleproject:430106352 Change-Id: I889dfda99d8ad35de1672914e7cc73b2879be6a2 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6832392 Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Cody Northrop 2a5ad67c 2025-08-12T10:23:10 Tests: Skip CapturedTestCL to unblock vulkan-deps roll Test: angle_end2end_tests --gtest_filter=CapturedTestCL.* Test: capture_tests.py Bug: angleproject:438195622 Change-Id: I9dd0f861ce4325c29305aa8de598f3434ec86ab3 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6842935 Auto-Submit: Cody Northrop <cnorthrop@google.com> Commit-Queue: Geoff Lang <geofflang@chromium.org> Commit-Queue: Cody Northrop <cnorthrop@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Shahbaz Youssefi b2e2766c 2025-08-12T11:08:46 Translator: Reject bool < bool This is not allowed per the GLSL ES spec. Bug: chromium:437983671 Change-Id: I9bd46718a35917dcdf6382cd51253ca2b68d5636 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6842555 Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Austin Annestrand 3643c21d 2025-08-11T15:02:18 Suppress oclcts NonUniform Workgroup failures pt.2 Suppress 2 other non-uniform workgroup smoke tests (as a stop-gap) until we root-cause the issue. Unblocks: https://chromium-review.googlesource.com/c/angle/angle/+/6831861/1 Bug: angleproject:42267085 Change-Id: I956d3706e160f38e33e5789e70abdc343fac8b54 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6838877 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Charlie Lao 555b2328 2025-08-08T13:32:13 BindVertexBuffer should pick up buffer change in other context Buffer is modified in the first context and calling glBindVertexBuffer() of the same buffer on the second context should pick up the buffer changes made in first context. Bug: b/433331119 Change-Id: I694f66b38b11b87e0efa8d9b939f8384120823af Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6830125 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Commit-Queue: Charlie Lao <cclao@google.com>
Cody Northrop 9a23635c 2025-08-09T18:17:33 Tests: Add Frost and Flame trace Test: angle_trace_tests --gtest_filter="*frost_and_flame*" Bug: b/437598600 Change-Id: Icf8c5ce3e0e1f0cadc442aedc8b38a70efb1e3f7 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6834277 Commit-Queue: Cody Northrop <cnorthrop@google.com> Reviewed-by: Mark Łobodziński <mark@lunarg.com>
Charlie Lao 5f5baf86 2025-08-08T10:22:07 Reduce WebGLCompatibilityTest::L32FTextures test time This test is looping formats. Within each loop it compile the same program which makes this test take very long to finish. This CL caches the programs so that it only compile/link once. This reduce this test run time on my local gLinux box from 1633 ms to 326 ms. Bug: angleproject:40096431 Change-Id: I32d7ad4fd25e1e4130766578fff21b1a2058c183 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6830119 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Cody Northrop 9367369d 2025-08-08T16:56:58 Tests: Add Kingshot trace Test: angle_trace_tests --gtest_filter="*kingshot*" Bug: b/437417676 Change-Id: I6c4a7a095266f6f99b0810854acbbfb50024d52f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6832638 Reviewed-by: Solti Ho <solti@google.com> Auto-Submit: Cody Northrop <cnorthrop@google.com> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Matthew Denton ca3d732d 2025-07-24T16:22:31 WGSL: support bool in uniforms WGSL does not allow booleans in the uniform address space. This CL changes the translator to substitute u32 for bool in the uniform address space, and convert it to bool on use. Arrays of bools are obviously arrays of u32s, and those will need special conversion functions to convert the entire array<u32> to array<bool> if necessary. This also includes the optimization of an array<bool> in a uniform--when indexing into it, only the indexed element will be converted to a native bool, instead of converting the entire array and then indexing. Note that substituting u32 for bool matches std140, so this change requires no changes to layout of uniforms. Also note that WGSL really likes explicit casts, so there's not really a way to avoid inserting explicit casts everywhere when using u32. Bug: angleproject:376553328 Change-Id: I8f72e55c6b401c28ff622622df7a450b7032721f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6785609 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Matthew Denton <mpdenton@chromium.org> Reviewed-by: Liza Burakova <liza@chromium.org>
Charlie Lao 41e44587 2025-08-06T20:04:23 BufferBindingMask in VertexArray::onBufferChanged are same buffer The bufferBiningMask bits received from VertexArray::onBufferChanged() must point to the same buffer, in case of one buffer is bound to multiple binding points. We can do some minor optimization here when we loop the BufferBindingMask bits: We only need to calculate hasConflict once when we call updateCachedTransformFeedbackBindingValidation(). Bug: b/433331119 Change-Id: I9f7cfa543deb46cd369915293b97f5d3122e1708 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6826782 Commit-Queue: Charlie Lao <cclao@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Charlie Lao d63cff37 2025-08-06T13:31:02 Fix bug when rebind the same buffer not pick up the change Buffer is modified in one context and rebind the same buffer on the second context should pick up the buffer changes. Bug: b/433331119 Change-Id: I8085fe826c339692881f2d3d6515efb19bbc4a88 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6825767 Commit-Queue: Charlie Lao <cclao@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Cody Northrop 8796f68c 2025-08-07T14:25:19 Tests: Add Top Heroes: Kingdom Saga trace Test: angle_trace_tests --gtest_filter="*top_heroes_kingdom_saga" Bug: b/437162076 Change-Id: Ia690266b34f0da0c23fbd099ed28f20657529644 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6828593 Auto-Submit: Cody Northrop <cnorthrop@google.com> Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Tingwei Guo 7c8e357c 2025-07-15T17:24:25 Fix NULL handling in glGetAttachedShaders and ShaderPrecisionFormat Fix the case where passing NULL to the shaders parameter in getAttachedShaders and to the range and precision parameters in getShaderPrecisionFormat would trigger assertions. Bug: angleproject:431900689 Change-Id: I0d01bbb0d13f9a4704b47066c85ac7c1368534a0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6733749 Auto-Submit: Tingwei Guo <tingwei.guo@arm.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Charlie Lao <cclao@google.com> Commit-Queue: Charlie Lao <cclao@google.com>
Yuly Novikov b778f1ce 2025-08-07T15:31:01 Balance test shards to 10 minutes Shards increased due to googletest roll: angle_end2end_tests on Win ASAN: 10 -> 14 angle_deqp_gles3_tests on Linux ASAN and TSAN: 8 -> 11 Drive-by shards increase for slowness which was not investigated: angle_end2end_tests: on Linux Intel and Windows and Linux NVIDIA: 2 -> 3 on Win Intel experiment (UHD 770, 2 processes): 3 -> 5 on Win Intel (UHD 630, 1 process): 4 -> 10 angle_restricted_trace_gold_tests on Linux Intel and Windows and Linux NVIDIA: 1 -> 3 on Win Intel UHD 770: 1 -> 4 on Win Intel UHD 630: 1 -> 5 angle_trace_perf_vulkan_tests on Win Intel UHD 770: 2 -> 4 on Win Intel UHD 630: 2 -> 3 Skip EGLDisplayTest.ContextLeakAfterTerminate on Linux Intel, which started failing with the new sharding. Bug: angleproject:428697844, angleproject:437259106 Change-Id: I3ba458aa7b68bddf9786bb9fcbc2a3ed7479556c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6827389 Auto-Submit: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Charlie Lao d8dc3cc2 2025-08-02T12:02:20 Remove shared context lock from glVertexAttrib{I}Format This CL removes shared context lock from glVertexAttribFormat() and glVertexAttribIFormat() APIs, since they no longer access anything outside VertexArrayPrivate. The main problem I had run into is validation code. Before this CL, ValidateIntegerVertexFormat() needs context's mStateCache for mCachedVertexAttribTypesValidation and mCachedIntegerVertexAttribTypesValidation. Given these two cached value are constant after initialization, in this CL, I have moved them to PrivateStateCache. PrivateStateCache argument is added to ValidateVertexAttribFormat() and ValidateVertexAttribIFormat() to get them access to mCachedIntegerVertexAttribTypesValidation. Bug: b/433331119 Change-Id: Ifc3fbed32b4d3722c335dd2c393bc6519ed0b544 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6822032 Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Charlie Lao <cclao@google.com>
Mark Lobodzinski 61ddca79 2025-08-07T11:16:29 Tests: Add Gossip Harbor ANGLE trace Test: angle_trace_tests --gtest_filter=*gossip_harbor Bug: b/437087719 Change-Id: Ia0d313b3e34074df8e24ad3eae63b4718ed2bfc7 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6828000 Commit-Queue: Cody Northrop <cnorthrop@google.com> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Charlie Lao c39f1d0f 2025-08-01T15:44:11 Move updateCachedTransformFeedbackBindingVali* to VertexArray Since updateCachedTransformFeedbackBindingValidation needs buffer argument, this is now moved from VertexArrayPrivate to VertexArray class to make it obvious. Also mCachedTransformFeedbackConflictedBindingsMask is per bindingIndex, so I have changed it from AttributeMask to VertexArrayBufferBindingMask. Bug: b/433331119 Change-Id: I321267637ba661b18ef236260a17e6e8f7043eb9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6814158 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Commit-Queue: Charlie Lao <cclao@google.com>
Charlie Lao ae4aead5 2025-07-24T16:35:38 Remove sharedContextLock from glVertexAttrib{Divisor|Binding} With prior CLs all the functions used by glVertexAttribDivisor*, glVertexBindingDivisor and glVertexAttribBinding only access VertexArrayPrivate. This CL removes shared context lock from these APIs. Bug: b/433331119 Change-Id: Ib1632797c53d2cd7a31c21e93c0e69385c71a27f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6814157 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Commit-Queue: Charlie Lao <cclao@google.com>
Charlie Lao 328a205f 2025-08-07T11:02:12 Revert "Fix bug when reattach the same buffer not pick up the change" This reverts commit a93f4c8d29c8d214e4823032e6e2c061d5dab1b1. Reason for revert: https://issues.chromium.org/437092065 Bug: b/433331119 Original change's description: > Fix bug when reattach the same buffer not pick up the change > > Added a test that the same buffer is used in two contexts. Buffer is > modified in one context and reattached the same buffer on the second > context's vertex array should pick up the buffer changes. This test > exposed the bug and it is fixed in this CL. > > Bug: b/433331119 > Change-Id: Ia5675581857a642fe9727f196326da9af43d52cf > Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6821923 > Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> > Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> > Commit-Queue: Charlie Lao <cclao@google.com> Bug: b/433331119 No-Presubmit: true No-Tree-Checks: true No-Try: true Change-Id: Iaa14b5dbbdb02c69bb060d5073072ec6ddd473ce Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6827998 Auto-Submit: Charlie Lao <cclao@google.com> Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com> Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Mark Lobodzinski 1943d505 2025-08-06T10:03:45 Tests: Add Seaside Escape ANGLE trace Test: angle_trace_tests --gtest_filter=*seaside_escape Bug: b/436801950 Change-Id: Iedc4b03317ef02585ccf86b54494f1938379a68a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6823580 Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Cody Northrop <cnorthrop@google.com> Auto-Submit: Mark Łobodziński <mark@lunarg.com>
Charlie Lao a93f4c8d 2025-08-05T16:45:21 Fix bug when reattach the same buffer not pick up the change Added a test that the same buffer is used in two contexts. Buffer is modified in one context and reattached the same buffer on the second context's vertex array should pick up the buffer changes. This test exposed the bug and it is fixed in this CL. Bug: b/433331119 Change-Id: Ia5675581857a642fe9727f196326da9af43d52cf Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6821923 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Commit-Queue: Charlie Lao <cclao@google.com>
Shahbaz Youssefi 7a193b06 2025-08-06T13:50:50 Unthrottle validation error messages for all of Chromium Chromium uses the debug callback to check if there were any errors instead of glGetError(). With the previous change that throttled error messages, some webgl tests failed to correctly recognize that an invalid call did actually fail validation, as well as passthrough command decoder fuzzer tests. Bug: angleproject:423956139 Bug: chromium:436292490 Change-Id: Ida254c9bdd2137afcc9483cc1860254781338dbc Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6823824 Commit-Queue: Cody Northrop <cnorthrop@google.com> Reviewed-by: Cody Northrop <cnorthrop@google.com> Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
Tim Van Patten 88de021d 2025-08-01T17:15:22 tests: Skip PbufferDifferentConfig on system EGL Skip the test EGLBackwardsCompatibleContextTest.PbufferDifferentConfig when testing the system EGL, because not all platforms expose ANGLE-specific extensions. Bug: b/409384875 Test: angle_end2end_tests \ --gtest_filter=EGLBackwardsCompatibleContextTest.PbufferDifferentConfig/ES2_Vulkan_NoFixture Change-Id: I1556d7981d1e6c8c6fc1803cafcb76fa1052dcef Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6814161 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Tim Van Patten bc7e86b1 2025-07-30T14:55:34 tests: RequestedRendererCreated: Get created context version If the driver being tested does not support the ANGLE extension EGL_ANGLE_create_context_backwards_compatible, then the Context version returned may not match the requested version. Update the test RendererTest.RequestedRendererCreated to query the current Context's version if the extension is not supported so the renderer string is checked against the correct value. Continue enforcing the renderer string has the expected value if the extension is supported. Bug: b/409384875 Test: angle_end2end_tests --gtest_filter=RendererTest.RequestedRendererCreated/* Change-Id: Iad1b6d36a31b7d79e8fc5fae9e4cc9769b55edb8 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6803326 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Yuxin Hu <yuxinhu@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Tim Van Patten c2316adb 2025-08-05T10:43:09 Add support for ARM to expectations file Add support for skipping tests on ARM devices with the tag "ARM". Bug: b/279980674 Change-Id: If310955c5dfb2af19493f8af70f4f6c00c97fcc9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6821832 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Shahbaz Youssefi da8462d4 2025-08-06T05:44:28 Unthrottle validation error messages in WebGL Chromium uses the debug callback to check if there were any errors instead of glGetError(). With the previous change that throttled error messages, some webgl tests failed to correctly recognize that an invalid call did actually fail validation. Bug: angleproject:423956139 Bug: chromium:436354527 Change-Id: Ic65761f412d3548fedd63f48d36357c9568c0b53 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6822578 Commit-Queue: Cody Northrop <cnorthrop@google.com> Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Amirali Abdolrashidi f1c6e1a8 2025-08-04T19:55:02 Vulkan: Disable EXT_fragment_shading_rate Due to a few dEQP failures, this extension will be temporarily disabled. It can be reverted once the issues have been resolved. * Disabled fragmentShadingRateEXT for Vulkan. Bug: b/436169197 Bug: angleproject:420310117 Change-Id: I94fbbb8b0a0866465e8c67e34c214b465fc35558 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6820702 Reviewed-by: Charlie Lao <cclao@google.com> Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Charlie Lao 394df705 2025-08-01T14:03:17 Cache per bufferIndex buffer properties in VertexArrayPrivate This is another preparation CL. This CL caches per bindingIndex buffer properties in VertexArrayPrivate: mCachedBufferPropertyMapped and mCachedBufferPropertyMutableOrImpersistent. The only difference between these and mCachedMappedArrayBuffers / mCachedMutableOrImpersistentArrayBuffers is that one indexed by attribIndex and another is indexed by bindingIndex. With this, when attribute binding changes, we no longer need to make buffer access. With this, we can move setVertexAttribBinding and setVertexBindingDivisor from VertexArray into VertexArrayPrivate class and Context argument is also removed. Bug: b/433331119 Change-Id: I666544ee0585727ca92d640f372f5a64d1d85576 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6814156 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com> Commit-Queue: Charlie Lao <cclao@google.com>
Amirali Abdolrashidi 5f42ce7e 2025-08-05T11:57:54 Enable KHR-GLESEXT tests on bots * Enabled several bots to run the khr_glesext tests. * Skipped some failing tests for Vulkan. Bug: b/436389014 Change-Id: If0c43f508eb04c21307f75c6dc62e3ef30b0556e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6821642 Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Yuxin Hu <yuxinhu@google.com> Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Amirali Abdolrashidi 71b7b5ee 2025-08-04T18:44:35 Add KHR-GLESEXT test suite * Added rules to build new dEQP test binary to run KHR-GLESEXT tests: angle_deqp_khr_glesext_tests Bug: b/436389014 Change-Id: I53b74605d3c2a9a5f6a8fece17bc3f7656a6772c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6820701 Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Yuxin Hu <yuxinhu@google.com> Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Cody Northrop f3b4b716 2025-08-05T14:28:04 TracePerfTest: Disable debug callbacks for GLES1 on Android We're seeing nullptr crashes when running GLES1 traces through the Android EGL loader with debug callbacks enabled. The callbacks are only enabled for Debug builds. Current theory is that the loader exposes the extension, but doesn't actually connect the call for GLES1 contexts. For now, expand the existing ARM skip to all Android devices. Test: restricted_trace_perf.py --renderer vulkan -f minetest Bug: b/352386141 Bug: b/436624755 Change-Id: I8ebe8f53c5efdecb2169ce017886753779ef5219 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6821913 Reviewed-by: Tim Van Patten <timvp@google.com> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Amirali Abdolrashidi bafe39e2 2025-07-14T16:07:51 Vulkan: Use BindIndexBuffer2 when available This Vulkan API becomes available with VK_KHR_maintenance5. In addition to the original args, it also takes the index buffer size. * Added the necessary functions to use the following API: vkCmdBindIndexBuffer2KHR() * handleDirtyGraphicsIndexBuffer() now uses this API if maintenance5 is available. Bug: angleproject:394337110 Change-Id: I9a650ac5a033b449da05e0c226e2cfb6df74a75c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6735725 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com> Reviewed-by: Charlie Lao <cclao@google.com>
Charlie Lao eca5244d 2025-07-30T16:59:16 Remove buffer from VertexArrayState::setAttribBinding In later CLs we try to remove shared lock from glVertexAttribBinding(). One of the lockers for that is right now setAttribBinding() references "buffer" for size. This CL caches the per binding index bufferSize in VertexArrayPrivate (i.e, std::vector<size_t> mCachedBufferSize). setAttribBinding() moved from VertexArray to VertexArrayPrivate since it no longer needs buffer access, but uses mCachedBufferSize instead. This CL also changes std::vector to std::array for mVertexArrayBuffers Bug: b/433331119 Change-Id: Ie34aef82379af8f04099cdc20ebdf354d2787a65 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6803592 Commit-Queue: Charlie Lao <cclao@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Amirali Abdolrashidi fc0e0397 2025-07-31T13:15:45 Translator: Fix location validation for inout Without this validation, a shader using an inout variable can write into a memory beyond the limit set by MaxDrawBuffers when shader framebuffer fetch is enabled. * Extended the location check in visitSymbol() to include InOut qualifiers (EvqFragmentInOut) as well (in ValidateOutputsTraverser). * Moved RewritePixelLocalStorage() after ValidateOutputs() to avoid issues related to locations not being specified in the InOut nodes from the PLS implementation. * Added a unit test to test that this case results in a compilation error: GLSLTest_ES3.CompileFSWithInoutLocBeyondMaxDrawBuffers Bug: chromium:435139154 Change-Id: I44c729377a2e1d3126c2d279172217b2ab92b57b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6806230 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuxin Hu <yuxinhu@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi 91fa29be 2025-08-04T11:08:20 Add a feature to force contexts into debug mode Bug: angleproject:423956139 Change-Id: Ie166b40eb9c241140daf4a35df3391829c64f84e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6817467 Commit-Queue: Cody Northrop <cnorthrop@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Aurora Zhang be43e317 2025-07-31T17:40:24 EXT_external_buffer should have DYNAMIC_DRAW usage EXT_buffer_storage requires that BUFFER_USAGE must be DYNAMIC_DRAW. EXT_external_buffer only mentions that the buffer can be sourced from an external EGLClientBuffer. Therefore, the BUFFER_USAGE should still be DYNAMIC_DRAW. Add an end2end test. Bug: angleproject:436046320 Change-Id: Ie79a1b1f8be0df4566bff26800599b192c4f70b1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6817076 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Auto-Submit: Aurora Zhang <Aurora.Zhang@arm.com> Commit-Queue: Yuxin Hu <yuxinhu@google.com> Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Shahbaz Youssefi 35023c8a 2025-07-31T17:40:45 Throttle validation error messages in release builds Some apps make invalid but harmless calls every frame. ANGLE inserts a debug message for each validation failure which comes with a performance cost. This change makes ANGLE stop reporting validation failure messages after a handful have already been reported. This is only done when assertions are disabled and when the context is not a debug context. Bug: angleproject:423956139 Change-Id: Ie3fb121b2686cddf7da98fc23422b5d12097c6fa Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6805901 Reviewed-by: Yuxin Hu <yuxinhu@google.com> Reviewed-by: Charlie Lao <cclao@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com> Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
Charlie Lao fa29f604 2025-07-02T13:23:31 Remove sharedContextLock from {Enable|Disable}VertexAttribArray VertexArray objects are per context objects. In theory they do not need to protected by shared context lock. The reason we are taking locks because all these functions end up accessing Buffer object which are shared. In prior CLs we have removed subject observer usage from VertexArray which means VertexArray no longer accessed from other thread. In prior CLs we also split VertexArray into two classes: VertexArrayPrivate which has no buffer, and VertexArray which is subclass from VertexArrayPrivate and owns buffer. In this CL, glEnableVertexAttribArray and glDisableVertexAttribArray calls no longer take shared context lock. ContextPrivateEnableVertexAttribArray and ContextPrivateDisableVertexAttribArray are called from these two APIs and they only have access to StatePrivate. State Private holds a VertexArrayPrivate pointer, which means they do not have anyway to access buffer objects. The main challenge I run into here is mCachedActiveClientAttribsMask, mCachedActiveBufferedAttribsMask, mCachedActiveDefaultAttribsMask, mCachedHasAnyEnabledClientAttrib, mCachedNonInstancedVertexElementLimit, mCachedInstancedVertexElementLimit. These StateCache variable needs to be updated when these two APIs are called, and calculating these variable needs access to buffer object. The solution here is adding a bool mIsCachedActiveAttribMasksValid in the PrivateStateCache so that instead of immediately update these mCached* variable, we just set mIsCachedActiveAttribMasksValid to false. Then whenever any of these mCached* variable is needed, we will check mIsCachedActiveAttribMasksValid and calculate these cached variables. It adds one if check when accessing these caches, but the other benefit is that we may have avoided duplicated calculation when multiple states changed. Bug: b/433331119 Change-Id: I3227c72bc40501712db93fb3d540b835f07150b5 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4514436 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Charlie Lao <cclao@google.com> Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Panfeng Hou 2ef85c24 2025-07-09T17:13:52 Vulkan: Add support for GL_EXT_fragment_shading_rate Add support for GL_EXT_fragment_shading_rate. Bug: angleproject:420310117 Change-Id: I7b368afc45baf8551c222b2569991269117d385b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6726817 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Auto-Submit: Panfeng Hou <panfeng.hou@arm.com> Reviewed-by: Charlie Lao <cclao@google.com>
Amirali Abdolrashidi 279652e3 2025-07-21T16:21:28 Vulkan: Flush more often due to render pass count Accumulating too much workload before submission can lead to stuttering and reduced quality in rendering. In this change, a threshold is set for the number of render passes in order to submit the command buffer. * Added the following to ContextVk: mRenderPassCountSinceSubmit * It is incremented every time a render pass begins in the command buffer. * When the count reaches the following threshold at the RP closure, there is a submission: kMaxRenderPassCountPerCommandBuffer * (Currently set to 128) Bug: b/426439980 Change-Id: I4cde223fb81e6a27a61f78c924d3c9f2a082e995 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6775626 Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Matthew Denton a1d5d102 2025-07-22T02:15:43 WGSL: Allow matrices as in/out vars in shaders WGSL only supports scalars and vectors in in/out vars in shaders, matrices will need to be broken into column vectors and then put back together at the beginning (or end) of the shader. Arrays also need to be split, which will be done in another CL. Bug: angleproject:42267100 Change-Id: If1ba28c1b687ae0a3a5a554479f0ff0b5d9df39c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6777201 Commit-Queue: Matthew Denton <mpdenton@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Liza Burakova <liza@chromium.org>
kmx.io     mail     discord kmxgit v0.4.0