kc3-lang/angle/src/libANGLE

src/libANGLE

Log

Author Commit Date CI Message
Ian Elliott 6af0c03f 2021-02-25T18:15:21 Vulkan: Always write to swapchain alpha channel This fixes an issue with the "Ragnarok M: Eternal Love" game, which uses a GL_RGB8 format for its window, which is actually backed by a GL_RGBA8 format (a.k.a. "emulated alpha"). The game does no explicit clear per frame. Therefore, ANGLE selects a render pass loadOp=DONT_CARE, which leaves the alpha channel undefined (0.0 on a Pixel 4 XL). That causes SurfaceFlinger to show (blend or alternate vsyncs) what should have been covered up by the game (e.g. the Android launcher and live wallpaper). The solution is to prevent loadOp=DONT_CARE for emulated alpha. Test: ClearTest.ChangeFramebufferAttachmentFromRGBAtoRGB Test: dEQP-GLES2.functional.fbo.render.stencil_clear.tex2d_rgb_stencil_index8 Bug: b/180139027 Change-Id: Ied97b57c93d41326cb3294ff246691e09f316791 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2704949 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Commit-Queue: Ian Elliott <ianelliott@google.com>
Gert Wollny fb7a2445 2021-02-12T09:29:11 libANGLE: Declare copy operator if copy constructor is defined If the copy constructor is defined then the copy assignment must also be defined, but we can use the default here. v2: Don't inline because the chromium-style errors out on this Bug: angleproject:4486 Change-Id: Iec0f63342244249da32b455b49a0b89b6558cab9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2690953 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Cody Northrop a6d46239 2021-01-25T11:40:47 Capture/Replay: Support CaptureGetQueryObject Test: Fortnite MEC Bug: b/180418810 Bug: angleproject:5658 Change-Id: Icc95faeebbe1fa850f25252c90a5ef8b68634ea0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2702153 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Yuly Novikov 4f639a7e 2021-02-27T17:36:25 Fix Ozone backend build flags. Broken by http://crrev.com/c/2405806 ANGLE_USE_GBM must be defined in Display.cpp, not just GL backend. This also requires depending on libdrm for its headers. Bug: angleproject:3943, chromium:1181952 Change-Id: I4650d7806435413184b36c820cc6d06992f7f5cf Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2726575 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Yuly Novikov 73afaf1e 2021-02-26T18:08:08 Add EGL_WINDOW_BIT to all Ozone configs' EGL_SURFACE_TYPE Ozone's EGL windows are emulated, so we can always create one, thus setting EGL_WINDOW_BIT in all configs is safe. This is now required because in http://crrev.com/c/2378922 EGL_WINDOW_BIT becomes a requirement to be able to run tests. Previously Ozone's configs copied native EGL_SURFACE_TYPE, which was either 0 or having just EGL_PBUFFER_BIT. Bug: chromium:1181952, chromium:1034840 Change-Id: Ic2b14b43b3c32b47ea05d9d8da9fba7439f14d19 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2725273 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Shrek Shao 2b55f876 2021-01-29T18:02:21 Reland "Workaround for Mac Intel drawArraysInstanced with first > 0" This is a reland of 027bc47ca5b7b291fbda907173eefa05ad3d45a8 Original change's description: > Workaround for Mac Intel drawArraysInstanced with first > 0 > > Workaround by forcefully set instanced arrays (divisor > 0) > as streaming attributes and apply extra offset at front. Recover > those attribute bindings when first == 0 and other draw calls > (drawElementsInstanced) > > Bug: chromium:1144207, chromium:1144247, chromium:1144373 > Change-Id: Ie7836cc71b45a290513f34f90d49bd15b14ddba8 > Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2661095 > Commit-Queue: Shrek Shao <shrekshao@google.com> > Reviewed-by: Geoff Lang <geofflang@chromium.org> Bug: chromium:1144207 Bug: chromium:1144247 Bug: chromium:1144373 Bug: angleproject:5271 Change-Id: Id0b818b25a605376c98c2366c1f2029e2490c6cb Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2704799 Commit-Queue: Shrek Shao <shrekshao@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Shahbaz Youssefi da4aa8b8 2021-02-25T16:26:53 Vulkan: Move SPIR-V validation function to common/ For future use by the translator. Bug: angleproject:4889 Change-Id: I051ad48e6edb58ce3055c5fd276c18a6e29a66ac Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2717020 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Mohan Maiya 67333608 2021-02-25T14:15:33 Vulkan: Check texture type before calling into feedbackloop check In ContextVk::updateActiveTextures we end up calling this method shouldSwitchToReadOnlyDepthFeedbackLoopMode(...) for every active texture. Since color textures are more numerous than depth, check the texture type beforehand to decrease function stack depth. This removes 0.5% CPU overhead from a Manhattan30 offscreen run Bug: angleproject:5689 Change-Id: I14758b031e58b269392b4f450a5bb1ba8edabb44 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2723493 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
Mohan Maiya b5af8bde 2021-02-24T09:47:43 Vulkan: No need to unmap buffer when adding to mInFlightBuffers When we move a BufferHelper to the mInFlightBuffers list don't try to unmap it there, instead do that when it gets added to mBufferFreeList This removes 0.75% CPU overhead from a Manhattan30 offscreen run Bug: angleproject:5689 Change-Id: I01a52cdf4803fb8364140b01bedbd3d0e6cc61fd Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2723492 Commit-Queue: Mohan Maiya <m.maiya@samsung.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Stephan Hartmann 46df78dc 2021-02-25T20:30:42 GCC: add missing static constexpr member definition C++14 requires to definition of static constexpr members to emit a linker symbol. Bug: chromium:819294 Change-Id: I06b7f1f7a6260982cf07a71c67295789c1f03084 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2720261 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Stephen White a476d16b 2021-02-24T12:58:56 Check that pixel store params are 0 for fastUnpack path on D3D. Bug: angleproject:5679 Change-Id: I4b4d0e2340592383e20849ad6be99c1608c048c7 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2716705 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Stephen White <senorblanco@chromium.org>
Cody Northrop bdd1ac23 2021-02-25T16:31:25 Capture/Replay: Add CubeMapArray support Test: Recapture Lego Legacy: Heroes Unboxed Bug: angleproject:5685 Change-Id: I41a5ceb72724ab28ce3d57a85108d3dd806d976f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2721103 Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Lubosz Sarnecki 4f07f9b8 2021-02-24T13:50:18 gles_ext_params: Implement CaptureGetTexLevelParameterivANGLE. As seen in CaptureGetTexParameteriv_params from capture_gles_2_0_params.cpp. Bug: angleproject:5645 Change-Id: Id687f73fd5d898e64e8a8bebe92ca50a0e096990 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2718044 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Lubosz Sarnecki 33907625 2021-02-24T12:41:07 gl_angle_ext: Add param groups to glGetTexLevelParameterivANGLE. As seen in glGetTexLevelParameteriv from gl.xml. Bug: angleproject:5645 Change-Id: I1d8445f2178632b0edb0652e63984b31a1e34c48 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2718043 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Lubosz Sarnecki d24398b3 2021-02-25T15:07:00 gl.xml: Annotate glDiscardFramebufferEXT with param groups. Remove tests from expectations list, as it now passes on SwiftShader for ES2 and ES3. Bug: angleproject:4401 Change-Id: If05b16e40632154a0ba2c22f1c3bf29e12113389 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2720259 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Shahbaz Youssefi 77637f2d 2021-02-19T15:18:52 Vulkan: Generate xfb support code in SPIR-V for emulation path This change moves the code generation at link time from source code to SPIR-V. As a result, transform feedback extension and emulation paths are more similarly handled before SPIR-V transformation (they both store information identically in the ShaderInterfaceVariableInfoMap). This change gets rid of the @@ XFB-OUT @@ marker. With no source code generation at link time, shader compilation can be moved to glCompileShader time. Bug: angleproject:4888 Change-Id: I8cdb89c22b57ce48cf5d226b8e41622d9d550d46 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2713269 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com>
Lubosz Sarnecki 46a955f8 2021-02-25T16:06:05 gen_vk_internal_shaders: Don't include full path on Python 3.4+. Since Python 3.4 `__file__` represents an absolute file path: https://docs.python.org/3.4/whatsnew/3.4.html#other-language-changes We only want the basename in the generated scripts. Bug: angleproject:5516 Change-Id: Ia4009419ef627aa1ce12f9a944539c054edc3a6b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2720260 Reviewed-by: Cody Northrop <cnorthrop@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Mohan Maiya 9b1c569b 2021-02-24T11:32:39 Vulkan: Handle vertex count and primitive mode error cases in TS When handling tessellation shaders these 2 requirements need to be met - 1. TCS - vertex count should be declared at least once 2. TES - primitive mode should be declared at least once After translating TCS/TES check whether these requirements are met and error out if they are not. Bug: angleproject:5557 Test: dEQP-GLES31 - functional.shaders.linkage.es31.tessellation.varying.rules.unspecified_* KHR-GLES32 - core.tessellation_shader.compilation_and_linking_errors.te_lacking_primitive_mode_declaration Change-Id: I727d548ad446646bbece1fb78197639537b5f395 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2714465 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
Mohan Maiya 550f2a3e 2021-02-24T09:49:42 Vulkan: Shader support for EXT_shader_framebuffer_fetch_non_coherent Translator can accept gl_LastFragData and 'inout' variable to gain access to framebuffer attachment data. The Vulkan translator replaces it with the SubpassInput type variable. Note that this works only for the noncoherent version of the extension. Bug: angleproject:5454 Test: *EXTShaderFramebufferFetchNoncoherent*.* Change-Id: I392f84ee3ad3eb9fbd09d0b7ff83731a9a3f33f6 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2598060 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
Shahbaz Youssefi 3013f27c 2021-02-23T22:37:42 Vulkan: Fix missing visibility barrier for glReadPixels glReadPixels uses a temporary buffer for readback and was missing an onBufferTransferWrite() call. Bug: angleproject:5669 Change-Id: I3c6daba176e332711fa1ab51e279526717486dd0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2717005 Reviewed-by: Jie A Chen <jie.a.chen@intel.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi 4a24786e 2021-02-23T23:32:48 Vulkan: Fix PBO read pixels with unaligned output pitch Vulkan takes the output pitch in texels instead of bytes in VkBufferImageCopy::bufferRowLength. This means that it's impossible to use vkCmdCopyImageToBuffer when the pitch is not a multiple of texel size. This change makes sure the fallback path is taken. Bug: angleproject:5667 Change-Id: I3f2ef312bb8288de8ca3c6730d85a3c14858812e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2717006 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi 999da35e 2021-02-18T17:29:20 Vulkan: Optimize glMemoryBarrier more The implementation assumed that the X_BARRIER_BIT bits implied two barriers; write->X and X->write. However, they only imply write->X, with the exception of SHADER_IMAGE_ACCESS_BARRIER_BIT and SHADER_STORAGE_BARRIER_BIT which handle X->write for images and buffers respectively. As a result, the other bits no longer set the MEMORY_BARRIER dirty bit as they don't guard against X->write usage. Bug: angleproject:5070 Change-Id: Id23904c455a5f56dc45fc6832a74fdfbba6a4827 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2705702 Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Cody Northrop 7ff605a5 2021-02-23T21:45:40 Capture/Replay: Fix compressed texture level caching Before this commit, the following sequence would cause an assert: glCompressedTexImage2D(GL_TEXTURE_2D, 0, GL_COMPRESSED_RGBA_ASTC_8x8, ...); glCompressedTexImage2D(GL_TEXTURE_2D, 1, GL_COMPRESSED_RGBA_ASTC_8x8, ...); This is due to code we added that clears the compressed texture cache when it detects we already have an entry for the texture level. This is allowed in GL, to respecify a texture level by simply calling glCompressedTexImage on it again. The problem is we would clear the entire texture from the cache, not just the level being respecified. Test: Recapture Asphalt 8 Bug: angleproject:5678 Change-Id: I473d6ea1cb8823bfec680ddd78c3457b93ecc431 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2716639 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Shahbaz Youssefi 39d7fc18 2021-02-17T00:18:41 Vulkan: Don't break the render pass on dispatch calls The only reason a dispatch call may need to break the render pass implicitly is for read-after-writes where the write originates from the render pass but is not through a storage buffer/image. There are only two such scenrios possible: - Framebuffer attachment write -> texture sample - Transform feedback write -> ubo read All other uses of the buffers and textures that require breaking the render pass are handled by `glMemoryBarrier`. Bug: angleproject:5070 Change-Id: I92b50d69d8782097ee8ff477ac57da6209c326a1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2698998 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com>
Shahbaz Youssefi df0a92e6 2021-02-22T06:39:32 Assert valid when using SPIR-V ids Useful for debugging when an uninitialized id is used in the SPIR-V transformer. Bug: angleproject:4888 Change-Id: If446187a8a27a06f5958d5df5bc00d02e98d20e3 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2713268 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Lubosz Sarnecki 37752956 2021-02-11T16:12:48 Generators: Use fixed year in license headers. Remove dynamic year generation from generator scripts, as required by the Chromium C++ style guide. The dynamic year values were replaced by the current year at the time the file was created according to git log. The code to dynamically generate the year was removed. This patch also refreshes generated files and hashes. Bug: angleproject:5516 Change-Id: I735028bccb5c83217e92c380538f1abf0a906b2c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2690950 Commit-Queue: Lubosz Sarnecki <lubosz.sarnecki@collabora.com> Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Shahbaz Youssefi c2d14d03 2021-02-18T13:55:23 Vulkan: Refactor prerotation out of SPIR-V transformer The SPIR-V transformer performs multiple transformations in one pass. This is part of a series of refactoring changes that detangle these transformations and encapsulate them, while allowing the transformation to remain single-pass. Bug: angleproject:4888 Change-Id: Ia971da6d814b5d7c4f51132fe63b58675163c15f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2705026 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Charlie Lao <cclao@google.com>
Shahbaz Youssefi 4a95488c 2021-02-18T12:21:26 Vulkan: Refactor xfb codegen out of SPIR-V transformer The SPIR-V transformer performs multiple transformations in one pass. This is part of a series of refactoring changes that detangle these transformations and encapsulate them, while allowing the transformation to remain single-pass. Bug: angleproject:4888 Change-Id: I3e3164803863c4d477364f5b2ed0679ed501ea3c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2705025 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Charlie Lao <cclao@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi bd9a3b29 2021-02-18T11:52:39 Vulkan: Refactor varying precision fix out of SPIR-V transformer The SPIR-V transformer performs multiple transformations in one pass. This is part of a series of refactoring changes that detangle these transformations and encapsulate them, while allowing the transformation to remain single-pass. Bug: angleproject:4888 Change-Id: I587d601e02164c1372a09565d1713aac27cfbebb Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2705024 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi 454194e4 2021-02-17T23:14:30 Vulkan: Refactor inactive varying removal out of SPIR-V transformer The SPIR-V transformer performs multiple transformations in one pass. This is part of a series of refactoring changes that detangle these transformations and encapsulate them, while allowing the transformation to remain single-pass. Bug: angleproject:4888 Change-Id: Ifd52be302c25392d3e8fc1c589795c113e44d354 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2703792 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi c342647f 2021-02-17T22:46:09 Vulkan: Refactor gl_PerVertex trimming out of SPIR-V transformer The SPIR-V transformer performs multiple transformations in one pass. This is part of a series of refactoring changes that detangle these transformations and encapsulate them, while allowing the transformation to remain single-pass. Bug: angleproject:4888 Change-Id: I41ba9165336487819dcf915af0d1ed868f73f724 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2703791 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi 8f65bb39 2021-02-17T17:11:44 Vulkan: Typed return value in SPIR-V transformer transform funcs Use an enum class instead of a bool for readability. Bug: angleproject:4888 Change-Id: I5fddec4ccc1cb7e28940c0566979f5fd5e051d60 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2702652 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Charlie Lao <cclao@google.com>
Shahbaz Youssefi 6e8cdd39 2021-02-17T00:17:17 Vulkan: Don't break the render pass on indirect calls The render pass is now only broken if the indirect buffer was used as transform feedback. Any other write to the indirect buffer is synchronized with `glMemoryBarrier`. Bug: angleproject:5070 Change-Id: I67868ae9a8f08e1ab186440a3cbdc7439c66808e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2698996 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com>
Shahbaz Youssefi d3d30921 2021-02-17T16:33:20 Vulkan: Refactor ID-discovery out of SPIR-V transformer The SPIR-V transformer performs multiple transformations in one pass. This is the first of a series of refactoring changes that detangle these transformations and encapsulate them, while allowing the transformation to remain single-pass. Bug: angleproject:4888 Change-Id: I327f693d99496a0b89028abe627119e4a2b07c42 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2702651 Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi e96d1744 2021-02-12T14:14:02 Vulkan: Optimize glMemoryBarrier Previous to this change, glMemoryBarrier was processed as it is issued. This made it impossible to know whether a draw call would follow or a dispatch call, and what resources it would use. The render pass was conservatively broken due to this limitation. To address this limitation, handling of glMemoryBarrier is deferred until the next draw or dispatch call. Note that glMemoryBarrier acts as two barriers: - An execution+memory barrier: shader writes are made visible to subsequent accesses - Another execution barrier: shader accesses are finished before subsequent writes An important observation is that for most resources, ANGLE actually necessarily has to issue memory barriers automatically to conform with Vulkan. In terms of memory barrier thus, ANGLE already does the right thing except for when there's no binding change. This means WaW hazards (i.e. storage buffer and image writes) with no binding change require a memory barrier as a result of glMemoryBarrier. In all other cases, it's enough for glMemoryBarrier to break the render pass if necessary and ensure that corresponding bindings are marked dirty (for the execution or memory barriers to happen automatically later). Bug: angleproject:5070 Change-Id: Ide359c43362f8a78805ecf797a91de7aa79221f9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2693473 Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Jamie Madill 6af603e1 2021-02-19T11:56:34 Vulkan: Secondary sort shader enums by name. This preserves the primary sort by bits. If bits are equal we then sort by name to produce a more stable sort. Bug: angleproject:5653 Change-Id: I4cdbc21094fcbd1bb7ad17e124074949bdbf0439 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2704825 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Mohan Maiya 4afd0f27 2021-02-17T13:35:57 Vulkan: Add VK_KHR_image_format_list support to initExternal The code for initializing a VkImage with VK_KHR_image_format_list extension support is now internal to initExternal. Also provides a flag to inform callers when VK_KHR_image_format_list support is enabled. Bug: angleproject:5520 Change-Id: Ie2ade93d1403ab3f9cbcf7d80684bbca201d5d8d Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2702159 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
Corentin Wallez 95584459 2021-02-18T09:32:16 Revert "Workaround for Mac Intel drawArraysInstanced with first > 0" This reverts commit 027bc47ca5b7b291fbda907173eefa05ad3d45a8. Reason for revert: InstancingTests fail on Mac FYI Release (Intel UHD 630), see https://ci.chromium.org/ui/p/chromium/builders/ci/Mac%20FYI%20Release%20(Intel%20UHD%20630)/1729/overview Original change's description: > Workaround for Mac Intel drawArraysInstanced with first > 0 > > Workaround by forcefully set instanced arrays (divisor > 0) > as streaming attributes and apply extra offset at front. Recover > those attribute bindings when first == 0 and other draw calls > (drawElementsInstanced) > > Bug: chromium:1144207, chromium:1144247, chromium:1144373 > Change-Id: Ie7836cc71b45a290513f34f90d49bd15b14ddba8 > Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2661095 > Commit-Queue: Shrek Shao <shrekshao@google.com> > Reviewed-by: Geoff Lang <geofflang@chromium.org> Bug: chromium:1144207 Bug: chromium:1144247 Bug: chromium:1144373 Change-Id: Ia4a6026a8c446490346b373c33fe2b1724c1761f No-Presubmit: true No-Tree-Checks: true No-Try: true Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2704052 Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com> Reviewed-by: Corentin Wallez <cwallez@chromium.org> Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Corentin Wallez 2d3402eb 2021-02-18T09:05:02 Revert "EGL: Merge DisplayAndroid into DisplayEGL." This reverts commit 2cb6671534ddc6efe9664a2fb30be2b2fb13c8ef. Reason for revert: Make DisplayGbm fail to compile. See: https://ci.chromium.org/ui/p/chromium/builders/try/linux-chromeos-compile-dbg/636833/overview Original change's description: > EGL: Merge DisplayAndroid into DisplayEGL. > > These two classes have a lot of duplicated code for no reason. > > Bug: angleproject:5563 > Change-Id: Ia9d8efeb50f3e52ffb1df6a6e462efdb20fd4a9a > Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2636678 > Commit-Queue: Geoff Lang <geofflang@chromium.org> > Reviewed-by: Peng Huang <penghuang@chromium.org> > Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Bug: angleproject:5563 Change-Id: Ic39071130727b37cc2a168d4b8e3c50f3e7184a5 No-Presubmit: true No-Tree-Checks: true No-Try: true Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2704051 Reviewed-by: Corentin Wallez <cwallez@chromium.org> Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Shrek Shao 027bc47c 2021-01-29T18:02:21 Workaround for Mac Intel drawArraysInstanced with first > 0 Workaround by forcefully set instanced arrays (divisor > 0) as streaming attributes and apply extra offset at front. Recover those attribute bindings when first == 0 and other draw calls (drawElementsInstanced) Bug: chromium:1144207, chromium:1144247, chromium:1144373 Change-Id: Ie7836cc71b45a290513f34f90d49bd15b14ddba8 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2661095 Commit-Queue: Shrek Shao <shrekshao@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tim Van Patten bd276beb 2021-02-12T10:58:43 Vulkan: Don't wait on unflushed sync objects without a Context The app "Car Parking Multiplayer" issues a ClientWaitSync() command without having already flushed the sync object and without an active context. We should return TIMEOUT immediately rather than attempting to wait on the sync object, since we can't flush it and it'll never be signalled. Bug: angleproject:5613 Bug: angleproject:5656 Test: MultithreadingTest.NoFlushNoContextReturnsTimeout Change-Id: Ieaf675ca9144f9c851c73b9ca399daaf4ed1cd0f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2693375 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Charlie Lao <cclao@google.com>
Jamie Madill 7e990ef4 2021-02-16T18:15:43 Clean ups to generate_entry_points. Switches to Python 3 support. Made a change to run_code_generation to support this. Affects several generators. Also updates the generator to make a few other small cleanups. Bug: angleproject:5653 Change-Id: I045173c9ca85947c4eac22285701032c09f4c8d0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2699187 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Cody Northrop 51c5137a 2021-01-05T16:58:50 Capture/Replay: Add texture buffer support Test: Fortnite MEC Bug: b/180418810 Bug: angleproject:5658 Change-Id: I2025f0b88f7488a827159029e6d04a4cf4e9a1b8 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2611558 Commit-Queue: Cody Northrop <cnorthrop@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com>
Shahbaz Youssefi 7ae8531b 2021-02-17T15:20:07 Vulkan: Fix crash with 0-sized element buffer VertexArray::syncState syncs all dirty bits, including DIRTY_BIT_ELEMENT_ARRAY_BUFFER even for draw calls that don't use this buffer, such as glDrawArrays. If the element buffer is given 0 size, this caused a crash in the Vulkan backend. Bug: chromium:1172577 Change-Id: I02d78c9660c07b896f7403867b648901478251fe Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2701831 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Mohan Maiya 907a3cee 2021-02-17T08:07:45 Vulkan: Add support for EXT_shader_framebuffer_fetch_non_coherent EXT_shader_framebuffer_fetch_non_coherent is implemented using subpass input attachments. The extension will be enabled in a follow up change that adds required changes to the Vulkan translator. Bug: angleproject:5454 Test: FramebufferFetchNonCoherentES31.*Vulkan Change-Id: Ic73c66a476c4a21db5269431166a198841f1dc0c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2598059 Commit-Queue: Mohan Maiya <m.maiya@samsung.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Peng Huang d6e0b0ca 2021-02-17T15:59:52 Remove a debugging code in renderergl_utils.cpp Bug: chromium:1171371 Change-Id: I7cf00edbc337150e40e7afb4f6c5fc88862f0734 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2700332 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Peng Huang <penghuang@chromium.org>
Geoff Lang 2cb66715 2021-01-19T12:42:38 EGL: Merge DisplayAndroid into DisplayEGL. These two classes have a lot of duplicated code for no reason. Bug: angleproject:5563 Change-Id: Ia9d8efeb50f3e52ffb1df6a6e462efdb20fd4a9a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2636678 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Peng Huang <penghuang@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Tim Van Patten 64b5b3d3 2021-02-12T17:31:58 Vulkan: Further restrict enabling LINE_RASTERIZATION_MODE_BRESENHAM The app "Car Parking Multiplayer" enables GL_SAMPLE_ALPHA_TO_COVERAGE, which leads to a VVL error when LINE_RASTERIZATION_MODE_BRESENHAM is also enabled: VUID-VkGraphicsPipelineCreateInfo-lineRasterizationMode-02766: The Vulkan spec states: If the lineRasterizationMode member of a VkPipelineRasterizationLineStateCreateInfoEXT structure included in the pNext chain of pRasterizationState is VK_LINE_RASTERIZATION_MODE_BRESENHAM_EXT or VK_LINE_RASTERIZATION_MODE_RECTANGULAR_SMOOTH_EXT and if rasterization is enabled, then the alphaToCoverageEnable, alphaToOneEnable, and sampleShadingEnable members of pMultisampleState must all be VK_FALSE. This CL adds the additional checking to GraphicsPipelineDesc::initializePipeline() to ensure those conditions are met before setting lineRasterizationMode to VK_LINE_RASTERIZATION_MODE_BRESENHAM_EXT. Bug: angleproject:5613 Test: StateChangeTest.AlphaToCoverageEnable Change-Id: Ie2286078f6916c01a19ae6f932321a86619bd4e1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2694094 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Tim Van Patten <timvp@google.com>
Shahbaz Youssefi 3c28b2a0 2021-02-10T16:49:32 Vulkan: Remove render pass check from setupDraw() ... and move it to handleDirtyGraphicsRenderPass. Bug: angleproject:5528 Change-Id: I416b2fedb1cd924d04fa739aecb65193fd845f6b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2686441 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi e8c0aa81 2021-01-26T23:40:36 Vulkan: Clean up transform feedback extension pause/resume 1. The xfb counter buffer barrier issued was wrong, following a typo in the spec. This barrier is now correctly issued using the usual barrier APIs. 2. A mechanism was added to automatically pause/resume transform feedback when a program pipeline needs to be rebound. This is incorrect as it misses the xfb counter buffer barrier. The render pass is broken instead if transform feedback is active/unpaused and the program pipeline is changed. 3. The transform feedback counter buffers are now disposed of when transform feedback is ended. This avoids an unnecessary barrier that this change would have otherwise incurred (and hence render pass break) in Manhattan which repurposes the same transform feedback object. Bug: angleproject:5528 Change-Id: I1ffe8b4b8975645ba43afd70e9cdbb0765529da5 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2651647 Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Ian Elliott 60b03e62 2021-02-16T14:31:19 Create getRotatedExtents() and related methods Created: - WindowsSurfaceVk::getRotatedWidth() - WindowsSurfaceVk::getRotatedHeight() - RenderTarget::getRotatedExtents() - ImageHelper::getRotatedExtents() - ImageHelper::getRotatedLevelExtents2D() Note: The FramebufferVk class doesn't use any of these methods Bug: b/175793022 Change-Id: I64395688bfdb172d32853763743fc5f266a6b792 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2686102 Commit-Queue: Ian Elliott <ianelliott@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com>
Cody Northrop 741f4039 2021-02-16T17:00:31 Fix spelling in capture owners Bug: None TBR: geofflang@chromium.org Change-Id: I60fd6bc18d8c86087d7640eb1145d575c4008edb Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2697986 Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Mohan Maiya 2b69462e 2021-02-12T15:55:00 Vulkan: Consolidate image format list initialization code HardwareBufferImageSiblingVkAndroid, RenderbufferVk, SurfaceVk and TextureVk all had similar code to initialize VkImage with VK_KHR_image_format_list extension support. Consolidate all of them into a helper. Bug: angleproject:5520 Change-Id: Ib7ac582a35a26c0da15879e688bcaf0c2a92ea87 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2692641 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
Shahbaz Youssefi d7037aa2 2021-02-11T14:35:30 Vulkan: noop glMemoryBarrier(CLIENT_MAPPED_BUFFER_BARRIER_BIT_EXT) CLIENT_MAPPED_BUFFER_BARRIER_BIT_EXT requires a memory barrier: shader buffer write -> host read. According to the spec, the data is only available after a call to glFinish or wait on sync: > The application must call MemoryBarrier with the > CLIENT_MAPPED_BUFFER_BARRIER_BIT_EXT set and then call FenceSync with > SYNC_GPU_COMMANDS_COMPLETE (or Finish). Then the CPU will see the > writes after the sync is complete. When a buffer is written to by the GPU, ANGLE calls onHostVisibleBufferWrite(), which ensures a "memory write -> host read" barrier is issued at the end of the command buffer. Additionally, persistently mapped buffers use VK_MEMORY_PROPERTY_HOST_COHERENT_BIT, so there's no need for a call to vkInvalidateMappedMemoryRanges. As a result, there's nothing necessary in ANGLE to do for this barrier bit. Note that should persistenly mapped buffers start using non-coherent memory, this barrier should imply a call to vkInvalidateMappedMemoryRanges for the persistently mapped buffers. Bug: angleproject:5070 Change-Id: Iaeae019dadfa659a47d2dac41c0c09f1c15e584b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2689380 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Mohan Maiya <m.maiya@samsung.com> Reviewed-by: Charlie Lao <cclao@google.com>
Shahbaz Youssefi 79ae52dd 2021-02-11T14:18:41 Vulkan: Fix missing visibility barrier for host-visible buffer writes See https://chromium-review.googlesource.com/c/angle/angle/+/1661252 for context. This was accidentally broken during the command graph rework. This will eventually be validated by syncval. See https://github.com/KhronosGroup/Vulkan-ValidationLayers/issues/2329. Bug: angleproject:5070 Change-Id: Ic16fa900e554d46e54b42fc3fbe0f96d5327fa0f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2689379 Reviewed-by: Charlie Lao <cclao@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi 30622479 2021-02-16T12:33:40 Vulkan: Fix crash with deferred clears and MSRTT The following scenario was mishandled: - MSRTT draw with an unresolve operation (i.e. has two subpasses) - Deferred clear - Flush deferred clear with MSRTT framebuffer not needing unresolve (i.e. has one subpass) Bug: chromium:1178693 Change-Id: If3548e99897d698d61dfafbe9f86193723d06e5a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2697648 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Geoff Lang e153063d 2021-02-16T18:34:25 Add timvp to ANGLE, vulkan and capture owners. Bug: None Change-Id: Ic23994efa3f4184c471eb21e7f3bb039f6b45057 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2697947 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Geoff Lang <geofflang@chromium.org>
Peng Huang 9c450bee 2021-02-12T14:23:42 Use the default build target vulkan_memory_allocator Bug: chromium:1170339 Change-Id: I00c868e205a16a889e193383c35b2ed42d8a61ef Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2693028 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi ca6573eb 2021-02-13T21:21:26 Vulkan: Fix dirty bit bug when closing RP in setupDraw() Bug: angleproject:5644 Change-Id: I2e1fa8c4db8eba1beb74c83dcb24cbe2e70ddca6 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2694076 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi 031dd5c5 2021-02-11T14:56:43 Vulkan: Add fast path for pack buffer copies When vkCmdCopyImageToBuffer is possible, it is used to implement glReadPixels with PBO instead of a CPU readback. A few tests are added to verify synchronization between pack/unpack and compute shader writes. Bug: angleproject:5070 Change-Id: Ia5da0feb31348f6b52d164b973ff95e2a0510582 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2691466 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Ian Elliott <ianelliott@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Mohan Maiya 2b15062b 2021-02-11T12:52:39 Allow more formats as texture attachments in GLES 1 This change allows all formats in the GLES 1.1 spec, table 3.4 to be used as a texture attachment. Also normalize values passed into glColorPointer if the format is of type GL_UNSIGNED_BYTE. These changes are needed for the android app, Kick the Buddy to render correctly. Bug: angleproject:5599 Tests: *DrawTextureTest.ColorArrayDifferentTypes* *FramebufferObjectTest.TextureObjectDifferentFormats* Change-Id: Ie9d27fc24d94106651262cf9b2080dd3f05af1c5 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2690920 Commit-Queue: Mohan Maiya <m.maiya@samsung.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Shahbaz Youssefi 19e4cf86 2021-02-09T21:49:42 Capture/Replay: Some fixes for capturing WebGL Bug: angleproject:5635 Change-Id: I756471fbad34660c494fb7f9ba1144accd5c08b2 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2686119 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Mohan Maiya 12990d73 2021-02-11T11:37:03 GetBitSet now uses BitSetArray instead of IterableBitSet Remove the now unused IterableBitSet class. Bug: angleproject:3877 Change-Id: I161e5d062c8183e30a7eb9040f3018116fe6e69e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2683494 Commit-Queue: Mohan Maiya <m.maiya@samsung.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Tim Van Patten baf64903 2021-02-03T14:02:55 Allow small compressed mip level copies Update ValidateCopyImageSubDataBase() to take into consideration if the entire mip level is being written, even if the size of the mip level is smaller than the compressed block size. This also uncovered a bug in ImageHelper::initializeNonZeroMemory() where the image extents are not at least as large as the compressed format block size: VUID-vkCmdCopyBufferToImage-imageExtent-00207 This CL adds isAllocateNonZeroMemoryEnabled() to allow skipping the test when that feature is enabled while that bug is chased. Bug: angleproject:5592 Bug: angleproject:5634 Bug: angleproject:5643 Test: Texture2DTestES3.CopyCompressedImageMipMaps Change-Id: I0381e0d3490fb148604b61dc3bae8f96ba8b5f8e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2673069 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Cody Northrop <cnorthrop@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tim Van Patten a9de5d99 2021-02-04T20:44:15 Vulkan: setAllDefaultUniformsDirty after createPipelineLayout The default uniform descriptor set is reset while recreating the pipeline layout during handling of immutable samplers and then is never re-allocated and bound before the next draw. The call stack to allocate the program uniforms descriptor set: ProgramExecutableVk::allocUniformAndXfbDescriptorSet ProgramVk::updateUniforms ContextVk::setupDraw ContextVk::drawArrays Context::drawArrays Unfortunately, this occurs before the pipeline layout is reset (and the descriptor sets are reset) due to the presence of an immutable sampler: ProgramExecutableVk::reset <<---- mDescriptorSets.fill(VK_NULL_HANDLE); ProgramExecutableVk::createPipelineLayout ContextVk::updateActiveTextures ContextVk::invalidateCurrentTextures ContextVk::syncState Context::syncDirtyBits Context::prepareForDraw Context::drawArrays This CL calls setAllDefaultUniformsDirty() for the Program/PPO to ensure the default uniforms descriptor sets are re-allocated and re-bound before the next draw command. Bug: b/178424566 Bug: angleproject:5624 Test: CtsCameraTestCases Change-Id: If54a9f2cc09809a5103bc3eac641c77f56362229 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2677385 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Tim Van Patten <timvp@google.com>
Peng Huang 46203e32 2021-02-10T18:41:06 Workaround context lost for Adreno 42x and 3xx Bug: chromium:1171371 Change-Id: I8c2e13f3f35bf1f780526ad1d9d483226ce5ea34 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2688901 Commit-Queue: Peng Huang <penghuang@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Alexey Knyazev 1094d324 2021-02-12T03:14:38 GL: Add allow_etc_formats workaround Enable ETC2/EAC formats on newer Intel GPUs. Bug: angleproject:1552 Change-Id: I240b07d6d81d48e1e03dfcd56aad6fa9f1d7817f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2690952 Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
Shahbaz Youssefi 8b624c6d 2021-02-05T15:33:52 Use constexpr initializer list for bitsets Allows setting/resetting multiple bits to be coalesced into one operation. Bug: angleproject:5528 Change-Id: Ibf2dff8c81441a75c268d95066d23da1b2a3c810 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2678885 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi e281fbb0 2021-02-11T15:37:43 Revert "GL: Cache the results of glGetString" This reverts commit 27906e9c9f8a6767fb8af22d9f80264d8672243a. Reason for revert: Causes failures in mac_optional_gpu_tests_rel Original change's description: > GL: Cache the results of glGetString > > To reduce the amount of queries to the driver, we can cache the > results of glGetString. On Mac, we need to invalidate this cache > on GPU switch. > > Bug: chromium:1173672 > Change-Id: I039172068aec35034a87881a8804f52c080ce4ce > Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2676882 > Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> > Reviewed-by: Jamie Madill <jmadill@chromium.org> > Reviewed-by: Geoff Lang <geofflang@chromium.org> TBR=geofflang@chromium.org,jonahr@google.com,jmadill@chromium.org Change-Id: I5799c5d287dd01b946446bd66d4c89aef6756d3a Bug: chromium:1173672 Bug: angleproject:5639 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2690145 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Jonah Ryan-Davis 5af5808b 2021-02-10T13:06:36 GL: Do not support ETC1 textures on Desktop GL As pointed out in WebGL issue 3014, ANGLE has been exposing ETC1 texture support on desktop GL when GL_ARB_ES3_compatibility or GL 4.3+ is available. However, it is very likely that none of these platforms actually have hardware support for this format, and are only supporting it for compatiblity with ES3. We shouldn't expose this so that desktop clients can use a better format instead. Bug: chromium:1048244 Change-Id: I53837a05719d0ca3ff390b5b45b61a1f9c72a535 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2687466 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
Le Hoang Quyen 807d1ad7 2021-02-10T23:30:56 Metal: fix incorrect default viewport size with contentsScale>1 The correct metal layer's drawable size should be calculated after mMetalLayer.get().contentsScale is set during initialization to allow initial viewport to be computed properly on front-end side. Bug: angleproject:2634 Change-Id: If278f40c0b6800bbe12fb5ea94986f728381bc00 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2687300 Commit-Queue: Le Hoang Quyen <le.hoang.q@gmail.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi 45a493ea 2021-02-05T13:48:48 Vulkan: Use a dirty bit to start the render pass Some dirty bits need to run before the render pass starts. An upcoming change for example needs to break the render pass when the program pipeline is changed while transform feedback is active. Another upcoming change may need to do the same based on a preceding glMemoryBarrier. This change adds a new dirty bit to start the render pass after some dirty bits have already been processed. Bug: angleproject:5528 Change-Id: I993c9efefed4c8fee268b218a8dd66a582d4e7cd Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2678863 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com>
Tim Van Patten dccec125 2021-02-08T16:21:30 Vulkan: Clear mCurrentGraphicsPipeline after createPipelineLayout() Set mCurrentGraphicsPipeline = nullptr when the pipeline layout is recreated if immutable samplers are used to prevent a use-after-free. For a fuller discussion on why this is necessary, see: Vulkan: clear mCurrentGraphicsPipeline during invalidate | https://chromium-review.googlesource.com/c/angle/angle/+/2647746 Bug: b/178424566 Bug: angleproject:5624 Change-Id: I70679bda85cd587dbd8893a4576ee5d33aee2b70 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2683041 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Geoff Lang 393c44f1 2021-02-01T15:45:24 EGL: Trim requested config parameters to EGL. ANGLE requests a config with several uneccessary parameters. I suspect these parameters are causing issues in the wild where eglChooseConfig fails. Try requesting ES3 then fall back to ES2 configs. Skip config caveats, floating point configs and multisample configs, we provided the default parameters. The resulting config is very similiar to the one Chrome requests in gl_surface_egl.cc Bug: chromium:1173161 Change-Id: Ifa78ff8081e3d86e823036981e1e2b5f8f751fed Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2665888 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Peng Huang <penghuang@chromium.org>
Peng Huang 0e8f7727 2021-02-10T15:03:10 Fix crash in gl::Context::unMakeCurrent() gl::Context's ctor may need to create a native EGL context and it can fail. If it fails, the context->mImplementation will be nullptr, and it will cause crash elsewhere. Fix the problem by checking it after create a gl::Context instance. Bug: chromium:1171371 Change-Id: Ic57f088dcbf9716b85fee92bf54d557eb94642d0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2686439 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Peng Huang <penghuang@chromium.org>
Jamie Madill 6fc10389 2021-02-10T11:20:16 Move Frame Capture to capture/ folder. This will make it easier to trigger the trace tests when these files are modified. Bug: angleproject:5530 Change-Id: I5f0c450595b380cd91b20c1477dc1845bee35dd9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2686120 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Jonah Ryan-Davis 27906e9c 2021-02-04T16:54:56 GL: Cache the results of glGetString To reduce the amount of queries to the driver, we can cache the results of glGetString. On Mac, we need to invalidate this cache on GPU switch. Bug: chromium:1173672 Change-Id: I039172068aec35034a87881a8804f52c080ce4ce Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2676882 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill a8a2a71b 2021-02-01T17:18:18 Vulkan: Support y-flip with no driver support. We can reuse the surface rotation matrix code to do the y-flip. This requires the SPIR-V transformation support. Because not all rotations are encoded into the table we can only support rotation with the driver support for y-flip (currently). Includes some very minimal regression testing. This work is targeted towards supporting vk-portability implementations which are not as up-to-date with Vulkan features. Bug: angleproject:5596 Change-Id: I270fa1efc03267551d28df33ddac9972e1343d60 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2665892 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Ian Elliott <ianelliott@google.com> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Tim Van Patten 80308e20 2021-02-08T16:16:36 Vulkan: Match min/magFilter to chromaFilter VkSamplerCreateInfo.magFilter and minFilter must match VkSamplerYcbcrConversionCreateInfo.chromaFilter. VkSamplerYcbcrConversionCreateInfo.chromaFilter is currently hard-coded to VK_FILTER_NEAREST in HardwareBufferImageSiblingVkAndroid(), so force magFilter and minFilter to VK_FILTER_NEAREST in SamplerDesc::init() if an externalFormat is being used. Bug: b/178424566 Change-Id: Ia1f7fc21e3faa4e57f837ed3ae70608134150aea Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2683038 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi bb062070 2021-02-09T15:30:04 Vulkan: Fix updates to element buffer If glBufferSubData results in a new vk::BufferHelper allocation, VertexArrayVk::mCurrentElementArrayBuffer needs to be updated. VertexArrayVk::syncState was working under the assumption that DIRTY_BIT_ELEMENT_ARRAY_BUFFER_DATA cannot result in a vk::BufferHelper pointer change. This assumption was broken in https://chromium-review.googlesource.com/c/angle/angle/+/2204655. Bug: b/178231226 Change-Id: I969549c5ffec3456bdc08ac3e03a0fa0e7b4593f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2685439 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Tim Van Patten c3887991 2021-02-03T13:52:32 Capture/Replay Don't capture invalid commands Special Forces Group 2 issues invalid OpenGL ES commands, which are currently captured by ANGLE's frame capture. These invalid commands generate errors on both ANGLE and the native driver, causing the test to be marked as "failed". This CL updates FrameCapture::captureCall() to not capture invalid commands, which prevents replay errors. However, the other work related to the call is still performed, including: - maybeOverrideEntryPoint() - maybeCapturePreCallUpdates() - maybeCapturePostCallUpdates() This work is necessary, otherwise calls that update necessary members like FrameCapture::mClientArraySizes may be missed, causing leading to compilation failures related to missing symbols. For example, mClientArraySizes is updated via the following call stack: captureClientArraySnapshot() maybeCaptureDrawArraysClientData() maybeCapturePreCallUpdates() It's necessary for mClientArraySizes to contain non-zero values, so that the variable gClientArrays[] is output by WriteCppReplayIndexFiles(). Otherwise a compilation error will occur when it's referenced by a glVertexAttribPointer() call in the capture. Bug: angleproject:5592 Change-Id: I625a074e035eb75267d8384dc61ce6de1717af91 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2673068 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com>
Mohan Maiya f6784006 2021-02-08T16:39:04 Modify data type of ActiveTextureMask and ImageUnitMask ActiveTextureMask and ImageUnitMask need more than 64bits. We can now use the BitSetArray datatype and avoid the perf penalty of using std::bitset Bug: angleproject:3877 Tests: angle_perftests.exe --gtest_filter=TexturesBenchmark* Change-Id: Ic82e09d1aa2c1904d06ad48ba6200f35896665ac Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2664734 Commit-Queue: Mohan Maiya <m.maiya@samsung.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Mohan Maiya f1cbcaba 2021-02-07T13:37:49 Use const reference when possible When retrieving BitSet masks return a const reference of the mask to avoid invoking the copy-constructor. Bug: angleproject:3877 Change-Id: I9f724a4864f3db17aa6f1ab74b81053c65313764 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2681767 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
Stephen White ad60c22d 2021-02-09T13:22:09 Add an option to use "metal" as ANGLE_DEFAULT_PLATFORM. Bug: angleproject:5632 Change-Id: Id470993a01609ace851372b1bbe01fee8cde274b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2684635 Commit-Queue: Stephen White <senorblanco@chromium.org> Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Mohan Maiya 54242b8f 2021-02-04T12:20:57 Vulkan: Leverage ExtendedDirtyBitType Expand ExtendedDirtyBitType to include bit for clip distance, mipmap generation hint and shader derivative hint. Handle these dirty bits in ContextVk::syncState Bug: angleproject:5611 Change-Id: If8d1646334e737f81ac72cdddb8fe3ba613b4b94 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2676173 Commit-Queue: Mohan Maiya <m.maiya@samsung.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Jonah Ryan-Davis 4968f6f2 2021-02-04T16:40:36 Move getRendererDescription from ContextImpl to DisplayImpl The other backend description strings are in DisplayImpl. This will help with caching the result of glGetString in the GL backend. Also Update the getters to not be const in order to allow caching. Bug: chromium:1173672 Change-Id: I43df35688762b23429f47f169c04482cf4cd089a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2676881 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
Jonah Ryan-Davis e174c358 2021-02-04T15:14:10 GL: Consolidate glGetString calls First step for cacheing the result of glGetString calls is mostly cleanup. Bug: chromium:1173672 Change-Id: I47281a09b9bd5859655d16376977a6eabcfbb3c4 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2676203 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Doug Horn 580961fb 2021-02-02T10:27:48 Micro-optimizations for setUniform* This CL improves the setUniform* call for the case where the niform component size is 4. In that case, we can issue a single memcpy. This reduces the average wall time of setUniform4fv in our test app by ~20%. Test: Run the dEQP shader tests. Bug: b/179160884 Change-Id: I9352f6188bc87449719aac522d1a2323adf7fca5 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2667592 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Doug Horn <doughorn@google.com>
Ian Elliott ce540d8c 2021-02-01T17:38:10 Create MSAA-swapchain windows' VkImages differently The actual VkImage needs to have rotated extents, but the ImageHelper needs to have non-rotated extents in order for the rest of ANGLE's pre-rotation to work. Bug: b/175793022 Change-Id: I6fa25ab8c636886787ac50b194e566111308f30b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2666514 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Ian Elliott <ianelliott@google.com>
Jamie Madill df48c027 2021-02-02T13:34:29 Vulkan: Don't request aliased FB attachments. This produces undefined pixel values in GL. We can also return UNSUPPORTED if required and still be conformant. This fixes an assert in the gfx-rs portability implementation. Bug: angleproject:5601 Change-Id: Ifb84ad9b5e2365360a81cddba766b0bea81f2d51 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2668311 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Alexis Hetu 18a5e7a6 2021-02-04T11:19:21 Enable packing extensions for the Vulkan backend The following extensions are ES2 extensions which become core in ES3: - EXT_unpack_subimage - NV_pack_subimage Since the Vulkan backend already fully supports ES3, it should be ok to enable these for ES2 contexts. This fixes all YUV related viz_unittests failures when using SwANGLE with the passthrough command decoder. Bug: chromium:1060139 Change-Id: I24ffd7a58f576799b93fe91e6be4032ca3175eda Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2676201 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Alexis Hétu <sugoi@chromium.org>
Tim Van Patten 54367a3e 2021-02-02T19:31:33 Capture/Replay: Support glCopyImageSubData with compressed formats Special Forces Group 2 uses glCopyImageSubData() to copy compressed texture data, which requires special handling for ANGLE's frame capture since compressed texture data can't be read back. To support this, when glCopyImageSubData() is called with textures using compressed formats, a new mCachedTextureLevelData entry will be created for the destination texture which is a copy of the source texture data. Note that this implementation only supports copying the full level data. http://anglebug.com/5604 will track adding support for partial level copies if an app requires that in the future. Bug: angleproject:5592 Bug: angleproject:5604 Change-Id: Ibce13f91f838fd691269c796fbeb5b9c6133c6aa Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2669955 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Cody Northrop <cnorthrop@google.com> Commit-Queue: Tim Van Patten <timvp@google.com>
Mohan Maiya 89f50584 2021-02-03T08:51:04 Vulkan: Add ExtendedDirtyBitType bitset ExtendedDirtyBitType qualifies DIRTY_BIT_EXTENDED dirtybit. Clip control code path can now set the appropriate ExtendedDirtyBitType when there is a change in state. Also remove the ClipSpaceOrigin member in the Vulkan backend that cached front-end state. Bug: angleproject:5471 Tests: dEQP-GLES2.functional.clip_control.* Change-Id: I8dbb509ef940e7905439d32483fd67a8fc171a6e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2673062 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
Brandon Schade 6cc845bb 2020-12-07T18:14:50 Vulkan: Add support for EXT_blend_func_extended This implementation utilizes vulkan's dualSrcBlend feature. Expose this extension if the underlying vulkan backend allows the use of this feature. Test: angle_end2end_tests --gtest_filter=EXTBlendFuncExtendedDrawTest* Bug: angleproject:5074 Change-Id: I7d2f611df89d65e5cac35158cb5f41a0ebd58aae Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2593151 Commit-Queue: Brandon Schade <b.schade@samsung.com> Commit-Queue: Mohan Maiya <m.maiya@samsung.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Amy Liu 4a09df13 2021-02-02T17:52:56 Vulkan:Fix wrong outputCount of emulated RGB copies. The output count of shaderParams equals componentCount divided by Ed. To avoid losing the last data,replace division with UnsignedCeilDivide function. Bug: angleproject:5598 Change-Id: I079d72f4eefa3d6b0f5d80ada25829fea673410c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2666869 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Mohan Maiya 7e81056a 2021-02-01T11:16:14 Vulkan: Support integer type incomplete texture Support both signed and unsigned integer type incomplete textures. Bug: angleproject:5502 Bug: angleproject:4432 Tests: IncompleteTextureTestES3.*IntegerType* dEQP.KHR_GLES31/core_sample_variables_mask_rgba8*i_* Change-Id: Ic8c972aac0ca8589b26333b66dd0cc5fb5134043 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2613245 Commit-Queue: Mohan Maiya <m.maiya@samsung.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Kenneth Russell 572afd90 2021-01-29T22:45:29 Disable SH_SCALARIZE_VEC_AND_MAT_CONSTRUCTOR_ARGS workaround. It causes incorrect compilation of some shaders. The original graphics driver bugs that motivated enabling it have been fixed, at least on one of the platforms (macOS). Tested with new WebGL conformance test in https://github.com/KhronosGroup/WebGL/pull/3214 . Bug: chromium:1165751 Change-Id: Iee9747769ca918aab143592d6cf158ce02a75ee0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2661024 Commit-Queue: Kenneth Russell <kbr@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill 4a426e1f 2021-02-01T12:10:46 Vulkan: Add missing variable inits to ContextVk. This was preventing testing with the viewport flip disabled. Bug: angleproject:5596 Change-Id: I5cc39eb2e1b431625f2f1498ab73828961af42ee Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2664249 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Tim Van Patten 0d175f30 2021-01-20T16:34:40 Tests: Add eFootball PES 2021 trace Test: angle_perftests --gtest_filter=TracePerfTest.Run/*efootball_pes_2021* Bug: angleproject:5564 Bug: angleproject:5517 Change-Id: I9738876b6a6c365984cecce01f2f09e0260b41cc Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2656989 Reviewed-by: Cody Northrop <cnorthrop@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Tim Van Patten <timvp@google.com>
Shahbaz Youssefi b15de455 2021-01-27T13:58:42 Remove extension check in validation of ES3.2 entry points Bug: angleproject:3582 Change-Id: I563c6d77b825afd44809db0d08dc4842a5f93b6a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2653914 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
kmx.io     mail     discord kmxgit v0.4.0