kc3-lang/angle/src

src

Log

Author Commit Date CI Message
Clemen Deng 95ac7b7c 2019-06-13T11:29:06 Autogen gles2+ declarations Added autogeneration of method declarations for GLES2+ and extensions Bug: angleproject:3526 Change-Id: I9c7263452146098512d9584deae8ea3f15f62c46 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1660949 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Jamie Madill 8118265a 2019-06-14T11:03:14 Add more DrawCallPerf offscreen configs. I thought these might have different performance characteristics when compared to the main configs. It's possible they would also need to forego using SwapBuffers to get different profiles. Right now perf is pretty similar to with the default FBO. Bug: angleproject:3117 Change-Id: Ic03b4ebd8c606af5121854565525e37e66f91435 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1650785 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Jamie Madill 12f38c49 2019-06-13T11:38:59 Vulkan: Use dynamic buffer for driver uniform updates. This should reduce the number of descriptor sets we need to allocate. We only allocate a new descriptor set when the underlying buffer is reallocated. We pass in the buffer offset via the dynamic offsets parameter. Bug: angleproject:3504 Change-Id: I40d031cd7295a8d002caff4f331f4a311f07505c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1646757 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tobin Ehlis <tobine@google.com>
Yuly Novikov 853ebacf 2019-06-13T18:35:25 Suppress Mac 10.13.6 Intel OpenGL dEQP failures Bug: angleproject:3531 Change-Id: I246144beb9c12b2df74612831262c7d263698db4 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1658718 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
James Dong 2806a898 2019-06-07T12:10:46 Vulkan: implement primitive restart Implements ES 3.0 feature GL_PRIMITIVE_RESTART_FIXED_INDEX, which allows the application to use a fixed "restart" index to restart the primitive during a single draw call. This change does't handle GL_LINE_LOOP primitives, which requires a bit of special handling. Bug: angleproject:3215 Change-Id: I2388852683fd17328a6a76c48d70a24d67ce8b67 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1650301 Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Courtney Goeltzenleuchter <courtneygo@google.com>
Shahbaz Youssefi 4316e064 2019-06-13T00:32:09 Vulkan: Output render pass loadOp in graph dump Bug: angleproject:3205 Change-Id: I7ad99bd2f8ddeb899c5fa86fd39296e5fc96d2c6 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1657708 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 88596bea 2019-06-13T14:17:48 Vulkan: Implement a texture descriptor cache. We noticed a significant hotspot in vkAllocateDesctiptorSets. The app was repeatedly cycling through a few combinations of active textures. For each state change in ANGLE we were allocating a new desctiptor set. This in turn would trigger internal driver memory allocation and cause jank. Using a cache avoids allocations entirely since the application is rotating through a stable set of textures. The descriptor cache is stored in each program. It is indexed by a set of 32-bit serials. Each texture generates a unique serial for every combination of VkImage and VkSampler that the texture owns. The texture descriptor is refreshed every time a texture changes or is rebound. The descriptor cache is accessed via an unoredered map with the texture serial sets as the hash key. We also store the maximum active texture index in the cache key so we don't need to hash and memcmp on all 64 active textures. This will currently fail if more than MAX_UINT serials are generated. But that number is high enough that it shouldn't be possible to hit in practice in a practical amount of time. Requires shifting the texture sync to ContextVk so we can get the new serial after the textures are updated. And to make sure to update the image layouts even if the descriptors are not dirty. Improves performance of the T-Rex demo. Also improves the score of the texture state change microbenchmark by about 40%. Bug: angleproject:3117 Change-Id: Ieb9bec1e8c1a7619814afab767a1980b959a8241 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1642226 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
shrekshao 95428331 2019-06-10T18:03:03 Enable floatBlend for D3D9 Although according to https://docs.microsoft.com/en-us/windows/desktop/direct3ddxgi/format-support-for-direct3d-feature-level-9-1-hardware D3D9 doesn't have full blending capability for RGBA32F. But turns out it could provide correct blending result in reality. As a result of some regression reports by client app, we decided to turn floatBlend on for D3D9. I tried to make a native D3D9 program to detect float blending capability. But turns out it needs quite some effort on Win10 for a D3D newbie. Considering the fact that D3D9 backend is not fully passing conformance test anyway. I don't think it worth the effort. Seems to be okay to just turn it on. Bug: chromium:970532 Change-Id: I66bec120be6eaaa62a5472b8521a97b4d23e924a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1652731 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Shrek Shao <shrekshao@google.com>
James Dong 0d3cf708 2019-06-12T10:07:02 Vulkan: fix array size for internal shaders Previously array sizes were calculated as the OR of all flag and enum masks, equal to 2^(number of bits used for flags and enums) - 1, which could be less than the actual needed array size. This fix moves array size calculation to the autogen script and changes the size to one more than the maximum potential value. Bug: angleproject:3524 Change-Id: Ia4eb2a83fc7deea84e18958efcda8b57632035f9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1655954 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Jamie Madill 05ba83a0 2019-06-13T13:26:58 Refactor DrawCallPerfParams. Makes it a bit easier to work with. In prep for adding offscreen configs. Bug: angleproject:3117 Change-Id: Ie2497574b2687592b7b3df2f7b933a19e83b6d16 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1650784 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Jonah Ryan-Davis 78a51911 2019-06-12T14:28:14 Clean up and expose frontend features to egl. gl::Workarounds was used to hold frontend features. Change ownership of this struct from Context to Display, so it can be exposed to egl. Also rename to features and clean up for consistency. Bug: angleproject:1621 Change-Id: I82e98e53873abb7a402c93e60f8a662a7263e0d5 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1655772 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill 444aa0fe 2019-06-08T13:53:17 Vulkan: Prefer immediate present mode for benchmarking. This should give the highest possible scores. Note that immediate is not always available. Bug: angleproject:3163 Change-Id: Ifedabd39819625b57feda71ed83a7bee8fa0a42b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1638904 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Qin Jiajia 3f2b4e7f 2019-06-11T14:47:59 Fix the crash in angle_deqp_gles31_tests This crash only happened on a compute test which following a render test. This root case was that after the render test, mVertexArray11 became an invalid pointer. However, compute test still went to the common path to use mVertexArray11. mVertexArray11 is never updated since compute shader doesn't need to triger DIRTY_BIT_VERTEX_ARRAY_BINDING. This patch adds checking to avoid compute test to use mVertexArray11. Besides, more ASSERT checking are added to convenient debugging. Bug: angleproject:3518, angleproject:1663, angleproject:2619 Change-Id: I446214110d762fc259899cef7635f369fa1f59a7 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1652866 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jiajia Qin <jiajia.qin@intel.com>
Clemen Deng ebddd1a2 2019-06-11T16:32:46 implement core egl image entry points Bug: angleproject:2467 Change-Id: Ica943a3972333e90a516aa6960f333cc9c378be2 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1653714 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Jamie Madill 169ef353 2019-06-12T14:42:13 Vulkan: Fix build with custom secondaries disabled. We should also add a compile-only target that verifies the build works as expected. Bug: angleproject:3117 Change-Id: Ib55969cdd0e670c123cdbe98bd0decf2204e7267 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1650789 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
James Dong 13ac9422 2019-06-10T17:53:09 Vulkan: add support for shadow samplers. Adds support for ES 3.0 shadow samplers. Test: ./angle_deqp_gles3_no_gtest --deqp-egl-display-type=angle-vulkan -n 'dEQP-GLES3.functional.texture.shadow.2d.*' Bug: angleproject:3211 Change-Id: Ic82ecfe19290fb952fcb0ba423691b5bac87c4a3 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1652021 Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Jonah Ryan-Davis fce1e2d1 2019-06-04T15:02:08 Extend eglGetPlatformDisplay to allow feature overrides. Add EGL_FEATURE_OVERRIDES_ENABLED_ANGLE and EGL_FEATURE_OVERRIDES_DISABLED_ANGLE to submit lists of strings naming the features that should be overridden (either enabled or disabled) on display creation. Bug: angleproject:1621 Change-Id: I4bb75c5dbab0e3b701a72069c38f8c60ecfffad2 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1646595 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Clemen Deng 2589cdcc 2019-06-11T11:38:41 GLSLTest stack overflow bug fix Bug: angleproject:3514 Change-Id: Id02bdb230dce0435b0b33db101d196596d6e8c93 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1652435 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 9aa865a6 2019-06-06T17:10:51 Vulkan: Add missing command pool reset. We were still calling destroy in checkCompletedCommands. We can instead call reset. May not significantly impact performance. Bug: angleproject:3489 Change-Id: I61a2cb154faac56dc78a49678dd822d81ce16ad2 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1647029 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tobin Ehlis <tobine@google.com> Commit-Queue: Jamie Madill <jmadill@chromium.org>
James Dong c75d6dc8 2019-06-05T16:12:01 Vulkan: Add support for gl_VertexID/gl_InstanceID Adds support for GLES 3.0 feature which adds gl_VertexID and gl_InstanceID built-in variables to vertex shader. Bug: angleproject:3221 Change-Id: I372d7ac34bed376b506e327725f0eca2513852fa Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1646735 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Kenneth Russell 6064e6ab 2019-06-11T04:01:12 Reland "Temporarily disable creating D3D debug device." This reverts commit 28394c2295335929c17bbfe0ab74f923cd234ff2. Reason for revert: may be the cause of flakiness on Win10 Debug (NVIDIA) per http://crbug.com/972914 . Original change's description: > Revert "Temporarily disable creating D3D debug device." > > This reverts commit 3d544fffe7ff9284b21e3a15f638297c827bbc21. > > Reason for revert: Seems like flakiness is not related. > > Suppress D3D11 test failures uncovered by re-enabling D3D debug device: > ComputeShaderTest.DispatchCompute > Texture2DTestES3.TextureImplPropogatesDirtyBits > D3DTextureTestMS.* > > Also a small fix to ANGLETestBase::checkD3D11SDKLayersMessages > to clear reported messages, otherwise the failures are also reported > for tests following the one with the actual failure. > > Original change's description: > > Temporarily disable creating D3D debug device. > > > > To diagnose whether Windows Intel end2end tests flakiness > > is related to creating debug device and new version of Windows SDK. > > > > Bug: angleproject:3153 > > Change-Id: Ica7eae3fe875e22203b6dfd98b83604d66a97737 > > Reviewed-on: https://chromium-review.googlesource.com/c/1476954 > > Reviewed-by: Geoff Lang <geofflang@chromium.org> > > Commit-Queue: Yuly Novikov <ynovikov@chromium.org> > > Bug: angleproject:3153, angleproject:3493 > Bug: angleproject:3501, angleproject:3512, angleproject:3513 > > Change-Id: I232da72aa09c2b9230cf179dd54e1731f8365f8e > Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1480854 > Reviewed-by: Yuly Novikov <ynovikov@chromium.org> > Reviewed-by: Jamie Madill <jmadill@chromium.org> > Commit-Queue: Yuly Novikov <ynovikov@chromium.org> TBR=ynovikov@chromium.org,geofflang@chromium.org,jmadill@chromium.org Change-Id: I38a5cc9bb6ebcd50e97e40cc972a82679e73732a No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: angleproject:3153, angleproject:3493, angleproject:3501, angleproject:3512, angleproject:3513 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1652784 Reviewed-by: Kenneth Russell <kbr@chromium.org> Commit-Queue: Kenneth Russell <kbr@chromium.org>
Nico Weber 1b5ad312 2019-06-10T18:41:08 Omit "de_DE.ISO-8859-15@euro" from ShCompileTest.DecimalSepLocale on Windows. A new SDK version started throwing on locales that have a code page section (the part after the dot) that's >= 16 characters long. That's a system library bug, but for now we need to work around it. Bug: chromium:972372 Change-Id: I64cee1051e854dac84e8c3c728ce65606b0dc89b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1652016 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Yuly Novikov 28394c22 2019-02-21T03:24:58 Revert "Temporarily disable creating D3D debug device." This reverts commit 3d544fffe7ff9284b21e3a15f638297c827bbc21. Reason for revert: Seems like flakiness is not related. Suppress D3D11 test failures uncovered by re-enabling D3D debug device: ComputeShaderTest.DispatchCompute Texture2DTestES3.TextureImplPropogatesDirtyBits D3DTextureTestMS.* Also a small fix to ANGLETestBase::checkD3D11SDKLayersMessages to clear reported messages, otherwise the failures are also reported for tests following the one with the actual failure. Original change's description: > Temporarily disable creating D3D debug device. > > To diagnose whether Windows Intel end2end tests flakiness > is related to creating debug device and new version of Windows SDK. > > Bug: angleproject:3153 > Change-Id: Ica7eae3fe875e22203b6dfd98b83604d66a97737 > Reviewed-on: https://chromium-review.googlesource.com/c/1476954 > Reviewed-by: Geoff Lang <geofflang@chromium.org> > Commit-Queue: Yuly Novikov <ynovikov@chromium.org> Bug: angleproject:3153, angleproject:3493 Bug: angleproject:3501, angleproject:3512, angleproject:3513 Change-Id: I232da72aa09c2b9230cf179dd54e1731f8365f8e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1480854 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
James Dong c74c3d81 2019-06-05T10:55:40 DEBUG: Log all debug messages to console Logs all debug messages produced by GL and EGL. This is helpful for debugging when the application does not register a debug callback. Validation errors produce log messages with level INFO, and internal errors produce log messages with level WARN. Bug: angleproject:3505 Change-Id: I269055aec49e1d77edeedb482fa6873c8f79f853 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1645817 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tim Van Patten 70642e42 2019-06-05T14:55:49 Vulkan: Implement ES 3.0 rasterizer discard Rasterizer discard is a feature that allows the graphics pipeline to skip the fragment shader stage. Implementing rasterizer discard in Vulkan is as easy as not binding a fragment shader binary. Tests for rasterizer discard live in dEQP-GLES3.functional.rasterizer_discard.*. Bug: angleproject:3214 Test: dEQP-GLES3.functional.rasterizer_discard.* Change-Id: I0815df3b70f6f9cdc6c8c87ec4813bb629d8bd5f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1646692 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tobin Ehlis 35b25fc6 2019-06-07T11:54:36 For WebGL warn on late extension directive A previous change based on ESSL 1.00 spec had made it an error in all cases when an extension directive appeared in a shader after the first non-preprocessor token. However, this is incorrect for WebGL 1.0 so adding warning case for WebGL. BUG=chromium:971660 Change-Id: I026fe60e8b1876de65b001b676f7a0552739a20c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1648661 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Kai Ninomiya <kainino@chromium.org> Commit-Queue: Tobin Ehlis <tobine@google.com>
Shahbaz Youssefi ed65dc43 2019-06-03T11:21:00 Vulkan: implement glSampleCoverage Emulated by applying a mask on top of the mask from glSampleMask. Bug: angleproject:3204 Change-Id: I4c80ab0a3261cbf1b8a43c270fe13c2a247ea663 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1639749 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi b407e1a0 2019-06-03T17:15:51 Vulkan: implement ES3 blit Augment the resolve shaders to be able to stretch and blit too. The UtilsVk resolve function is accordingly expanded to include blit. Bug: angleproject:3200 Change-Id: I30b172a5e388089735ab494f55cbfdc2781a8bf9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1635753 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
James Dong ea49f6f5 2019-06-06T12:58:27 Debug: Add LOG_INFO logging level Will be used for logging GL/EGL debug messages. Bug: angleproject:3505 Change-Id: I10ff32dd4f9227cabac2f3d9425b0428e7af1398 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1648714 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Tim Van Patten 148ecd89 2019-05-28T14:14:42 Vulkan: Handle VK_ERROR_OUT_OF_DATE returned by vkAcquireNextImageKHR() There is some new Android HWUI/SkiaGL code, that (in certain cases) looks up the underlying Vulkan swapchain that ANGLE created, and adds an image to it. This causes the next call to vkAcquireNextImageKHR() to return VK_ERROR_OUT_OF_DATE and requires the swapchain to be recreated. Bug: angleproject:3480 Test: Use ANGLE on Android and verify apps load correctly. Change-Id: I21c86035664878e75d6a9dc769546747aa4c8256 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1632424 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jeff Gilbert 55ca3a51 2019-06-05T17:32:32 Use execv instead of execve w/ environ. Some platforms don't want us messing with environ, and these seem to be equivalent. Bug: angleproject:3506 Change-Id: I66804e8accc30421c1cd32c0eda750500e3ecb60 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1646887 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Courtney Goeltzenleuchter 3f17f929 2019-06-05T13:13:12 Avoid recreating swapchain for preRotation Android will return VK_SUBOPTIMAL_KHR if the orientation of the screen is not optimial for the orientation of the device. We were re-creating the swapchain when this happens but we weren't changing the orientation and would get this return code every time we call vkQueuePresent and re-create the swapchain each time. Apparently recreating the swapchain causes visual artifacts on some devices. Just ignoring VK_SUBOPTIMAL_KHR allows things to work. Bug: angleproject:3497 Need to implement proper preRotation support for longterm fix. Bug: angleproject:3502 Change-Id: I0904eb60b742d24618d502c111510117758a8502 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1641206 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Courtney Goeltzenleuchter <courtneygo@google.com>
James Dong a4a24af3 2019-06-06T10:50:21 Vulkan: fix uint indices causing incomplete buffer Fixes buffer size calculation when uploading element buffer to account for 4-byte (uint) indices. Bug: angleproject:2902 Change-Id: Id70fc0fdffecd0b27995820bd0ad88e4cb61e013 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1648325 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Jamie Madill 266a9e8d 2019-06-05T15:15:10 Vulkan: Move descriptor pools into ProgramVk. Previously ContextVk owned the descriptor pools. We were trying to maximize descriptor reuse to conserve memory. However the default uniforms would have no possible sharing. And because uniform buffers are usually unique to a program it's likely there would be less reuse. Image descriptors could be shared. But with the advent of a descriptor cache in the Program it becomes difficult to manage the cache through descriptor pool recycling. Moving the pools into the Program simplifies the cache management. We could look at adding back more reuse in the future. Also shifts driver uniforms back into the end of the descriptor sets to make indexing into the Program's descriptor pools simpler. Bug: angleproject:3117 Change-Id: I52bb49cf322d944ad7cf08791efdf24b7fe573ce Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1644775 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
James Dong b91f87b6 2019-06-05T11:38:39 Vulkan: enable OES_element_index_uint for Vulkan We already handle 32-bit indices, so this change enables the corresponding extension for ES 2.0. Bug: angleproject:2902 Change-Id: Iccde1d4aac383a9c11edf851778ff315de8d4ad1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1646032 Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: James Dong <dongja@google.com>
Jamie Madill 03bc3db8 2019-06-05T13:48:43 Vulkan: Minor cleanup to UtilsVk. Renames "ContextVk *context" to "ContextVk *contextVk" for consistency. Also adds an "allocateDescriptorSet" helper function. Bug: angleproject:3117 Change-Id: Id03e30432d8e5e35b682350a8e98fd2e62cdf89a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1644776 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi 5643afc0 2019-06-04T15:37:58 Vulkan: add GLSL placeholder for transform feedback output Bug: angleproject:3205 Change-Id: Id3941b6953e03e7b2183acc57b2c6db4a6439352 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1643424 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Tobin Ehlis <tobine@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi ac8bafc0 2019-06-03T14:59:39 Remove unused bindGenericBuffer from TransformFeedback* Bug: angleproject:3205 Change-Id: I6dc29b48155cd3c0ef98248076cf02e70c9310d9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1643423 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Michael Spang 3c219171 2019-05-31T16:23:23 Add ARM vendor id This fixes a crash when run with asserts on ARM GPU: [2089630:1566159375:0531/202059.262200:ERROR:angle_platform_impl.cc(47)] GetVendorString(133): ! Assert failed in GetVendorString (../../third_party/angle/src/libANGLE/renderer/driver_utils.cpp:133): vendorId == 0xba5eba11 Bug: angleproject:3492 Change-Id: I786edc459b1567a3eee0034a0d8ae7050ac31952 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1642331 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill 44063c80 2019-06-04T15:20:30 Vulkan: Store array buffer conversions in BufferVk. The intent of this CL is to call convertVertexBuffer*PU only when we have new data to convert. If the app unbinds and rebinds a vertex buffer without changing the data we can now retrieve the cached vertex buffer info from the BufferVk class. Previously we would always reconvert the data on a rebind. This was slowing down applications and benchmarks. To achieve this we add a conversion cache to BufferVk. Each cache entry stores a key based on the vertex info. Also we store a ring buffer for each cache entry and a flag to indicate if the entry is dirty. The cache is dirtied on a bufffer data update or a map call. Improves performance in the T-Rex benchmark. Bug: angleproject:3495 Change-Id: Ia999c9187510748ba95bc98362eb332e1990d270 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1638903 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com>
Jonah Ryan-Davis e431aaa1 2019-06-04T11:36:43 Rename EGL_ANGLE_workaround_control to EGL_ANGLE_feature_control. For consistency, call these ANGLE "features", a subset of which may be workarounds. Also, whether the feature is enabled/disabled should be publically visible as "status". Bug: angleproject:1621 Change-Id: I0de90a932fbfe1fc9b59138153d616d29fa7268b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1643410 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
Shahbaz Youssefi 071d2a44 2019-06-03T11:49:13 Vulkan: Process dirty framebuffers first The currently bound framebuffer affects some dirty bits. This means that processing those dirty bits temporarily use data from a framebuffer that's about to be changed. At best, that creates unnecessary dirty bits in the Vulkan pipeline description. Bug: angleproject:3204 Change-Id: Ie001d6c8fcc61af2e78cf2cb58a1691d8b735ff3 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1639750 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi de70a713 2019-06-03T17:05:16 Vulkan: Rename Resolve* shaders to BlitResolve In preparation for adding blit functionality. Bug: angleproject:3200 Change-Id: I77b6b6cbe90b26ee23c25edebaa2b130f3c2987a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1641626 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jonah Ryan-Davis 2dd40a44 2019-05-30T16:40:13 Add eglQueryDisplayAttribANGLE to query workarounds count. Add eglQueryDisplayAttribANGLE based on eglQueryDisplayAttribEXT to add behavior for quering the count of all workarounds available. Used externally to build a list of workarounds. Bug: angleproject:1621 Change-Id: I793acedc76111fd018600169d58bf5d8cf4a63ee Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1637817 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill 20d380fa 2019-06-03T15:21:47 Print stack backtrace on critical failure. We reuse code from Skia to walk the stack on Posix platforms. See: https://github.com/google/skia/blob/master/tools/CrashHandler.cpp On Windows we use a BSD-licensed tool called StackWalker. See: https://github.com/JochenKalmbach/StackWalker This allows us to get high quality stack traces on Win/Linux/Mac. Bug: angleproject:3162 Change-Id: I9c50ede2c6a41ed0ee85a0507372df42a487bcef Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1632950 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Jamie Madill 5993d899 2019-06-03T13:05:38 Vulkan: Use storage buffers for index conversion shader. This allows us to use the shader regardless of driver support for texel buffer views. It also allows us to convert indices on the GPU unconditionally. We add a new internal compute shader that converts pairs of indices into a packed single uint value that stores two 16-bit values. In the future we could add support for converting primitive restart indices. Should speed up benchmarks on systems which didn't have R8_UINT support for compute shader buffers. Bug: angleproject:3490 Change-Id: I56ca0cabb094e97f36ab4edc779e6c8ad2d2601e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1639058 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com>
Kimmo Kinnunen 4a757416 2019-06-03T14:04:01 D3D: Fix heap corruption when copying texture arrays Do not use depth dimension when copying normal 2D textures of a 2d texture array. It copies past the end of the buffer obtained via ID3D11DeviceContext::Map(). Fixes a case which can be reproed with the payload disabled: gn args out\debug --args="is_debug=true" ninja -C out\debug angle_end2end_tests out\debug\angle_end2end_tests ^ --gtest_filter=Texture2DArrayCopy.SnormFormats* --gtest_catch_exceptions=0 ^ --gtest_repeat=-1 Bug: angleproject:2865 Change-Id: Id9bc5489fa41749d1b1d2e87992166b1a0af76fc Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1641247 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Anders Leino 12a52423 2019-06-03T10:39:54 Sharpen GL test skip condition for NVIDIA The skip condition for GLSLTest_ES3.VaryingStructNotInitializedInVertexShader is modified to not skip NVIDIA on windows. Bug: angleproject:3413 Change-Id: I126d9fa9e36c3b6c07abfd24bdaf6feb76ac4f6f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1640208 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Qin Jiajia 0730e630 2019-05-31T14:25:37 Fix needStructMapping In previous logic, we didn't check the structure field member type. So when passing the non-struct member of a structure to a function, it would think that struct mapping was needed. In this patch, we add more checking so that struct mapping only happens when there are structure copy or passing a structure to a function. BUG=angleproject:2967 Change-Id: Ic98e884c8f8540e180cdf40a0e036ffef18c1689 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1638227 Commit-Queue: Jiajia Qin <jiajia.qin@intel.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 98f21671 2019-05-31T15:34:39 Vulkan: Refactor index buffer convert functions. This is a prepratory refactor for converting index buffers on the GPU using a more generic compute shader. No functional change. Bug: angleproject:3490 Change-Id: Iadf4b1429314db6850320aee33c4113f38577378 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1639057 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Jamie Madill c104b2d2 2019-05-31T16:25:02 Re-enable WGL perf test configs. These were accidentally dropped in https://crrev.com/c/1574674 . Re-enable so we can gather better perf stats on Windows machines. Bug: angleproject:3494 Change-Id: I92e9ba7e7783fddde1a47c0ffe056da7750d2c3b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1639063 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Cody Northrop f7106d16 2019-05-30T17:34:42 Revert "Revert "Vulkan: Use atomics in SerialFactory."" This reverts commit 35fa85ba2fafa583bd706b64d181f3f95094325d. Reason for revert: This commit was an innocent bystander. The bugs are intermittent, requiring multiple runs to verify. Bug: angleproject:3492 Change-Id: Idc9dcb1631cfca5fea342266c732732c6f2bd2d5 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1636263 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Michael Spang cc9bc655 2019-05-31T12:00:25 Vulkan: Fix synchronization with current surface Since a6242e4d4 ("Vulkan: Support submitting multiple semaphores"), semaphores are not generated because the function generating checks for non-empty command graph right after flushing. Bug: angleproject:3492 Change-Id: I03db4a77fe24fdfdf39ba535b235451ae0275a0e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1638579 Reviewed-by: Cody Northrop <cnorthrop@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Michael Spang <spang@chromium.org>
Jamie Madill bc30e72c 2019-05-30T13:30:33 Vulkan: Recycle command pools with vkResetCommandPool. vkDestroyCommandPool has worse performance than reset. Instead of destroying and creating new command pools for every batch of work we can instead recycle unused command pools. This seems to improve performance on most systems. Bug: angleproject:3489 Change-Id: I7f8fe7f9385ca481e04f226a26426c3e92da19ff Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1636410 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com> Commit-Queue: Courtney Goeltzenleuchter <courtneygo@google.com>
Tim Van Patten 0a56f0e3 2019-05-14T17:07:27 Vulkan: Update front face mode during syncState() When handling DIRTY_BIT_DRAW_FRAMEBUFFER_BINDING in syncState(), the call to update the front face was missing, so culling could be enabled for the wrong front faces. Bug: angleproject:3237 Test: Verify 3D apps render (more) correctly Test: New SimpleStateChangeTest end2end tests Change-Id: I1d94a977bea9e48d90b5346861e5565d2371cadd Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1611753 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Cody Northrop 35fa85ba 2019-05-29T15:59:13 Revert "Vulkan: Use atomics in SerialFactory." This reverts commit 1b0f79ee92b51122634fca6342876db196b85e3c. Reason for revert: Breaks a number of applications on Android Original change's description: > Vulkan: Use atomics in SerialFactory. > > This allows serials to be generated from multiple threads without locking for > very little/no cost. > > BUG=angleproject:2464 > > Change-Id: Id61d170e7a985c3100da0057156859ffcb083dad > Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1516514 > Commit-Queue: Geoff Lang <geofflang@chromium.org> > Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> TBR=geofflang@chromium.org,syoussefi@chromium.org,jmadill@chromium.org Change-Id: I52011f294d68490998869fa898e929fd41eaa51c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1636272 Reviewed-by: Cody Northrop <cnorthrop@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Cody Northrop <cnorthrop@google.com>
Shahbaz Youssefi 6d9b8908 2019-05-29T13:18:31 Vulkan: Merge Color and Depth/Stencil resolve shaders There is a good deal of overlap in the way these shaders calculate offsets, flip etc. This change merges the two shaders together, as well as the UtilsVk functions that invoke them. This is in preparation for adding blit functionality to the same shader. Bug: angleproject:3200 Change-Id: Id7c2cf12e05db1802ae497766fb361fe4bef6479 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1635750 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Shahbaz Youssefi 172e5d37 2019-05-29T10:00:50 Vulkan: Implement universal stencil resolve This is done by resolving stencil into a temporary buffer and copying that into the stencil aspect of the resolved image. Bug: angleproject:3200 Change-Id: I29111b44db2cb093acc4544034fbe61178f055a1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1635709 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com>
Michael Spang ab6a59b2 2019-05-21T21:26:26 Vulkan: Submit semaphores from glWaitSemaphoreEXT & glSignalSemaphoreEXT Implement submission of client semaphores passed to glWaitSemaphoreEXT & glSignalSemaphoreEXT. This also relaxes the expectation that we will not flush() if there are no commands. Signaling semaphores in particular requires queue submission irrespective of whether there are any command buffers to submit. If there are neither commands nor semaphores, we can still skip queue submission. WebGL runs in Chrome with ANGLE & Vulkan interop as of this patch, albeit with incorrect synchronization due to texture barriers not being implemented yet. Quite a few flags are needed to try this: GN args: angle_vulkan_conformant_configs_only=true chrome \ --enable-features=UseSkiaRenderer,UiGpuRasterization \ --enable-gpu-rasterization \ --enable-oop-rasterization \ --enable-vulkan \ --use-gl=angle \ --use-angle=vulkan Bug: angleproject:3289 Change-Id: I3d49c230a2fbf0cd2a2b943b05ded0e4604cc313 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1623815 Commit-Queue: Michael Spang <spang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi d3e7a41b 2019-05-29T14:40:52 Vulkan: fix buffer copy barriers Bug: angleproject:3362 Change-Id: Ibf6520fb73bb1a9b76714cddd786569789c70430 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1635752 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Jonah Ryan-Davis f52f2637 2019-05-23T13:52:52 Add EGL_ANGLE_workaround_control extension. This extension is used to query strings from an array based on index, which will be used to query all the information about workarounds in ANGLE. Bug: angleproject:1621 Change-Id: I27157f278f7f17c92c8b4fd7753e2a5ecd0528f6 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1627723 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 06055b5e 2019-05-29T14:31:42 Update Windows AMD suppression. Should have been specified for OpenGL instead of D3D11. Bug: chromium:967796 Change-Id: I099ab51709f69e7abb7fe698e6010241bdccb4bf Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1635751 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Michael Spang a6242e4d 2019-05-22T15:15:09 Vulkan: Support submitting multiple semaphores With GL_EXT_semaphore, the client can create and submit semaphores to synchronize with an external vulkan instance. We need to be able to submit semaphores from both the current surface as well as the client. In order to support glWaitSemaphoreEXT, make ContextVk hold a vector of semaphores that will be submitted with the next batch. In order to support glSignalSemaphoreEXT, make ContextVk::flush() build a vector of semaphores to submit, rather than just one. Submissions that result from a client signal operation will need to signal both the client semaphore as well as a semaphore from the surface's semaphore chain. Bug: angleproject:3289 Change-Id: I52b07c4c2a7bf52300d3107f028aa64585c6c4db Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1623814 Commit-Queue: Michael Spang <spang@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Michael Spang 2008c8ff 2019-05-21T20:38:59 Vulkan: Reuse flushImpl() from finishImpl() Reuse flushing logic in finish() as there's an increasing amount of logic needed to submit work in order that would otherwise be duplicated. Bug: angleproject:3289 Change-Id: Ia489e285615ebaa48b01d8a202871bebf7de0ac2 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1623813 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Michael Spang <spang@chromium.org>
Shahbaz Youssefi f2a1c384 2019-05-21T16:32:49 Vulkan: Implement multisampled framebuffers Simultaneously implements ANGLE_framebuffer_multisample and ES3 multisampled framebuffers. Additionally, implements ES3 framebuffer blitting where multisampled framebuffers are involved. Bug: angleproject:3203 Bug: angleproject:3204 Bug: angleproject:3200 Change-Id: I5694a30f71168e807688a9568e3742b81d907918 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1622667 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Geoff Lang e0956a16 2019-05-06T10:34:27 Vulkan: Rename getSubmitFence to getNextSubmitFence. Now that getLastSubmittedFence exists, the new name is more specific. BUG=angleproject:2464 Change-Id: I0bce1387b95b7b7b739e311d37e35e7d162d8f2d Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1595435 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Anders Leino 262e2824 2019-05-20T14:24:40 Use D3D11 GetDimensions driver workaround for 3D textures Some NVIDIA D3D11 drivers are buggy and interprets the level passed to GetDimensions as being relative to 0, rather than the SRV's MostDetailedMip. This affects all integer format textures, because the dimensions are used for sample position calculations with integer format textures, which leads to sampling outside texture when the base level is non-zero. Bug: angleproject:3441 Change-Id: Ic54328e3d712e28a40efb4e63b8fce0baeb4ef42 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1619785 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Geoff Lang ded5f903 2019-05-01T13:58:27 Vulkan: Make the Vulkan renderer thread safe. Gate all access to the queue and caches with mutexes. Does not handle sharing of resources in share groups across threads yet. BUG=angleproject:2464 Change-Id: I297f8f1a535b99efca663cf72bac3d90df8b5d97 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1592253 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Geoff Lang 1b0f79ee 2019-02-28T11:08:21 Vulkan: Use atomics in SerialFactory. This allows serials to be generated from multiple threads without locking for very little/no cost. BUG=angleproject:2464 Change-Id: Id61d170e7a985c3100da0057156859ffcb083dad Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1516514 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Jamie Madill 29ac274f 2019-05-28T15:53:00 D3D11: Suppress failing Texture test on AMD. Bug: chromium:967796 Change-Id: I48280909db9c927c12d3c0d97f1ad451dffb849e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1632949 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Geoff Lang ee244c77 2019-05-06T10:30:18 Vulkan: Move command graph and garbage to ContextVk. To support multithreading, contexts should manage their own command graphs and garbage. This allows safe access to vulkan resources such as command pools without thread synchronization. BUG=angleproject:2464 Change-Id: I930149bc9f0793028761ee05ab50b8c0a4dec98a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1516515 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Geoff Lang 5a808b86 2019-05-27T13:22:32 GL: Allow BlitGL to copy from external textures. BUG=967410 Change-Id: I503dda6f2493b56123a8bd369e2f1305971abb4e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1364110 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Commit-Queue: Geoff Lang <geofflang@chromium.org>
Geoff Lang 44462cd6 2019-05-27T17:12:21 D3D: Use the ProgramD3D::GetExecutableTask context for loading program binaries ProgramD3D::GetExecutableTask is a d3d::Context type itself so that errors can be recorded in a thread-safe maner. Use the task context when loading program binaries in ProgramD3D. Fix up incorrect casts in RendererD3D::loadExecutable in both D3D9 and D3D11 backends to stop incorrectly assuming the context type. BUG=962439 Change-Id: I5b29372a7254f709e1bbb67ee322ef4109f73e48 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1630294 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Commit-Queue: Geoff Lang <geofflang@chromium.org>
Shahbaz Youssefi ce9be8c7 2019-05-24T09:35:06 Vulkan: Have a cubemap as 2D-array view handy Previously, only texture copies used a shader that performed texelFetch(). To support cubemaps, a hack was used to temporarily create a 2D array view. With upcoming support for multisample resolve, more shaders will be using texelFetch() all requiring this workaround. This change instead makes sure that a separate view is created for cubemaps for the purpose of being used with these shaders. As a result, we have three logical views on textures and render targets: - Draw: a view that can be used as a color/depth/stencil attachment - Read: a view that can be used to sample from - Fetch: a view that can be used to fetch from The fetch view is generally the same as the read view, except for cube maps. Bug: angleproject:3200 Change-Id: I21547f728c16f0aa8f0fcae152c400b5cc1565da Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1628585 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jonah Ryan-Davis 46fe0e4b 2019-05-22T13:13:50 Cleanup Features headers. Cleanup redundant angle:: prefix in Features headers. Cleanup workaround descriptions. Bug: angleproject:1621 Change-Id: I4325d3603877efee9c6a035cd5b5bbbbc4f5c169 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1625609 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
Anders Leino 69d04939 2019-05-20T14:04:13 Use D3D11 GetDimensions driver workaround for 2D array textures Some NVIDIA D3D11 drivers are buggy and interprets the level passed to GetDimensions as being relative to 0, rather than the SRV's MostDetailedMip. This affects all integer format textures, because the dimensions are used for sample position calculations with integer format textures, which leads to sampling outside texture when the base level is non-zero. Bug: angleproject:3441 Change-Id: Ic012cac94ce0a627221087ecb2ee6aa3665900d1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1617780 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Kimmo Kinnunen FI <kkinnunen@nvidia.com> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Anders Leino 1b6aded6 2019-05-20T12:56:34 Test D3D11 GetDimensions driver workaround for textureProjOffset Some NVIDIA D3D11 drivers are buggy and interprets the level passed to GetDimensions as being relative to 0, rather than the SRV's MostDetailedMip. This affects all integer texture formats because the dimensions are used for sample position calculations on integer format textures, which leads to sampling outside the texture when the base level is non-zero. Bug: angleproject:3441 Change-Id: I9d2f8fa8e66caca704cbc46d4933d3f97aaf93ea Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1617779 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Kimmo Kinnunen FI <kkinnunen@nvidia.com> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Geoff Lang 3fe06eb3 2019-05-21T17:37:25 D3D: Make sure Lod0 functions are never referenced in non-fragment shaders. The lod0 functions would not be declared but could still be referenced by vertex shaders. BUG=angleproject:3471 Change-Id: I635a8465ce68dc22a6f7387b30bf7e93b14dd67d Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1622741 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jiajia Qin <jiajia.qin@intel.com>
Jamie Madill 6722009e 2019-05-20T11:12:53 Vulkan: Handle dirty RTs with state messages. Prior to this CL we were handling dirty state change notifications by flushing the RT Images just prior to use or just after they were changed. This could lead to a few redundant checks in several places. It also meant we needed an owner pointer from the RT to the parent Image. This pointer would be null for Surfaces and Renderbuffers. This cleans up the image flushing logic to be handled by dirty bit notifications. When an app updates an attached Texture with TexSubImage or related calls it will send a notification to the Framebuffer. The Framebuffer then sets a dirty contents bit that is handled in the implementation. In Vulkan this means flushing the dirty bits. Requires adding a flag to the FramebufferImpl class to determine if we need to syncState before we checkStatus. Adding the option allows us to only call syncState for the GL back-end. Not calling syncState allows the robust resource init operation to happen *before* we syncState. Which in turn allows FramebuffeVk to initialize the VkImages in one go. Added new regression tests for Texture updates. This might not cover all cases. I found it was very hard to trigger some of the resource update staging in TextureVk. Bug: angleproject:3427 Change-Id: Idfa177436ba7fcb9d398f2b67922e085f778f82a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1601552 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Jamie Madill 3f3a05d4 2019-05-27T09:08:46 Fix KHR dEQP test data dir. Had an extra "gl_cts" appended. Bug: angleproject:3353 Change-Id: I962cd0196b898e5fdd631790d88bfb1d478d16b9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1630292 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Michael Spang 40757b28 2019-05-22T15:23:22 Use ConvertToGLBoolean instead of ternary operator in Context Clean up Context by replacing foo ? GL_TRUE : GL_FALSE by ConvertToGLBoolean(foo); Change-Id: I6a6ee9c23bb30d8b4ef43371c04a94578758145e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1626309 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Michael Spang <spang@chromium.org>
Qin Jiajia d68aff6c 2019-05-22T13:01:50 Fix the crash when the array size is not big enough. Previously, we used IMPLEMENTATION_MAX_DRAW_BUFFERS(8) as the array size of 'previouslyBound' which is used to record whether more than two different blocks are bound to a single buffer. However, when the underlying driver is d3d11_1, the max allowed shader storage blocks will be 64 which is larger than 8. An assert error will be met in push_back like below: Test case 'dEQP-GLES31.functional.ssbo.layout.random.all_per_block_buffers.45'.. ERR: push_back(250): ! Assert failed in push_back (../../src\common/FixedVector.h:250): mSize < N So we use 64 as the upper limit of max shader storage blocks. Bug: angleproject:3447 Change-Id: Ie76bfd15486c1ed02afca4283d6e70bdb4298cb6 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1623794 Commit-Queue: Jiajia Qin <jiajia.qin@intel.com> Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Minkyu Jeong 431ef2fd 2019-05-21T17:13:08 Add support for EXT_blend_minmax This is to add additional modes (MIN, MAX) to "BlendEquation". - add cases for "GL_MIN" and "GL_MAX" into "PackGLBlendOp()" in vk_cache_utils.cpp - add enabling "blendMinMax" codes into vk_caps_utils.cpp And, AUTHORS and CONTRIBUTORS are updated. BUG=angleproject:2897 Tests: dEQP-GLES3.functional.fragment_ops.blend.*min* dEQP-GLES3.functional.fragment_ops.blend.*max* Change-Id: I13a1d6d28a104b18e21697f9f23d77e4eda2d1a6 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1621582 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Jiawei Shao e794ddc8 2019-05-10T11:05:10 OpenGL: Remove all attachments before executing internal clear This patch fixes a bug in BlitGL::clearRenderbuffer() by removing all the attachments of mScratchFBO before attaching the real target of clear to keep mScratchFBO always being incomplete. Without this fix, the WebGL CTS tests mentioned in issue 2760 will fail on Intel OpenGL drivers when Chromium is using pass through command buffer because when a multisampled depth renderbuffer is attached to mScratchFBO, mScratchFBO will sometimes become incomplete because it may already have a non-multisampled color attachment. Bug: angleproject:2760 Test: angle_end2end_tests Change-Id: Id206fb4b338545ab46aba118e80d288158dcd8ec Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1612305 Commit-Queue: Jiawei Shao <jiawei.shao@intel.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Yuly Novikov d2683459 2019-05-23T16:11:19 Skip Texture2DIntegerTestES3.IntegerTextureNonZeroBaseLevel on Win AMD OpenGL Bug: angleproject:3478 Change-Id: I8f0ffeb0da31d08f4f4f45048d295c847cf095bd Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1626328 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Michael Spang 6bb193c8 2019-05-22T16:32:21 Vulkan: Implement glImportSemaphoreFdEXT Allow importing file descriptors into semaphores on linux. This can be used to synchronize ANGLE's GL renderer with respect to vulkan composition in chromium. Bug: angleproject:3289 Change-Id: I04ba3bbb2e343baa000ff89c21c03ca36163a713 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1623812 Commit-Queue: Michael Spang <spang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jeff Gilbert d9fa0744 2019-04-25T14:57:26 Emit OVR_multiview2 on ESSL/GLSL outputs. Add ARB_shader_viewport_layer_array support to SH_SELECT_VIEW_IN_NV_GLSL_VERTEX_SHADER. Bug: angleproject:3404 Change-Id: Ia89517d0cc92400ce47c9118e8c1abf8285aec41 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1585452 Reviewed-by: Jamie Madill <jmadill@chromium.org>
Geoff Lang 344ecaa6 2019-05-21T11:17:55 D3D: Return Incomplete when failing to load a program binary. When Stop is returned, the ProgramD3D::GetExecutableTask attempts to pop an error that has not been set. Returning Incomplete matches the other error cases in program binary loading. BUG=962439 Change-Id: I70895ddc706045620e3df219681982357b7b63e5 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1620914 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Geoff Lang <geofflang@chromium.org>
Anders Leino 8224a58b 2019-05-20T12:39:29 Normalize fragment color in Texture2DIntegerTestES3 The sampler returns colors with components in the range 0-255. Bug: chromium:963595 Change-Id: I7030ab9da6e0803e3aaf8aef4691ce7276636b10 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1617778 Reviewed-by: Kimmo Kinnunen FI <kkinnunen@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Michael Spang 5093ba67 2019-05-14T17:36:36 Implement resource management for GL_EXT_semaphore This implements glGenSemaphoresEXT, glDeleteSemaphoresEXT, and glIsSemaphoreEXT. It's not possible to do anything useful with them yet. Bug: angleproject:3289 Change-Id: I20ad90dbcd3fc573a4650c8531d6e1b8ccf7ca9b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1623811 Commit-Queue: Michael Spang <spang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Michael Spang 215b2191 2019-05-21T21:35:31 Add packed enum for GL_EXT_semaphore image layouts Add a new packed enum gl::ImageLayout which consists of the following enum values: GL_LAYOUT_GENERAL_EXT GL_LAYOUT_COLOR_ATTACHMENT_EXT GL_LAYOUT_DEPTH_STENCIL_ATTACHMENT_EXT GL_LAYOUT_DEPTH_STENCIL_READ_ONLY_EXT GL_LAYOUT_SHADER_READ_ONLY_EXT GL_LAYOUT_TRANSFER_SRC_EXT GL_LAYOUT_TRANSFER_DST_EXT GL_LAYOUT_DEPTH_READ_ONLY_STENCIL_ATTACHMENT_EXT GL_LAYOUT_DEPTH_ATTACHMENT_STENCIL_READ_ONLY_EXT Bug: angleproject:3289 Change-Id: Idc1615717d54fb1193e2f27c84a24993f6007d7b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1623810 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Michael Spang <spang@chromium.org>
Anders Leino e4452440 2019-05-09T13:29:49 Fix integer cube map sample HLSL calculation This patch fixes two issues in the HLSL generated to sample an integer cube map. The first issue was that an inappropriate major axis was selected when sampling from corners of the cube map. In particular, the added test case demonstrates a situation where a major axis of length 0 was selected, which lead to an infinite LOD being calculated. The fix was to adjust inequalities such that always exactly one of xMajor, yMajor and zMajor is true. The second issue was that the derivative functions ddx and ddy were being used on values that depended on the choice of major axis, which is not continuous at the corners of a cube map. This lead to a finite but incorrect LOD being calculated. The fix was to make sure major axis choice is not included when estimating the scaling factor. Bug: angleproject:3442 Change-Id: Ia3eb8c89f47d1bfdadc18aec989e8cbebc088ec0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1601515 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi fb8e1b25 2019-05-17T13:18:40 Vulkan: Fix multisampled clear UtilsVk::clearFramebuffer was not setting the sample count. Bug: angleproject:3204 Change-Id: Iab7d385ec58b12f21d888c5dfc9e94b7846ac40c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1616858 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Yuly Novikov 01f98cbf 2019-05-21T12:21:36 Skip failing MultisampleTest on Nexus 6P and NVIDIA Shield TV Bug: angleproject:3464, angleproject:3470 Change-Id: Ie864118ceebfc0d72aa95928a02ef0939be5b89a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1620992 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Shahbaz Youssefi b86e73da 2019-05-21T10:23:23 Vulkan: Update glslang_validator binary for Linux The new binary has the ability to optimize SPIR-V code. Bug: angleproject:3432 Change-Id: I5953be6f207f153baaf2efe17380448aff374d60 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1620913 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Jeff Gilbert e8247a57 2019-04-25T18:54:24 Add scripts/export_targets.py. $ export_targets.py <out_dir> <targets...> Flattens all library targets under <targets...> and checks that all includes in 'sources' are known to the build. Firefox can use this as the first half of our ANGLE vendoring process. Included additions to src/libGLESv2.gni were found with: $ export_targets.py out //:libEGL Eventually, it would be nice to have this as a presubmit check. Bug: angleproject:3403 Change-Id: I0f455e8c37c4b0a9e17e9f843843422b6a32f588 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1584652 Commit-Queue: Kenneth Russell <kbr@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Kenneth Russell <kbr@chromium.org>
Tom Anderson 08146a27 2019-05-17T10:40:44 Remove non-source sources from binary targets No behavior changes. BUG=chromium:964411 Change-Id: I843757e65f110882c01514fe6bf4aed28e07dd21 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1617011 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 7a026133 2019-05-20T10:34:08 Fix Android data dir for KHR dEQP tests. Also includes a fix for case list paths on developer machines. Also includes a test expectation update for GLES2. Bug: angleproject:3353 Change-Id: Idb01eb8d806067739253a6676a6d2d8ee1199929 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1616853 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Cody Northrop 70958d15 2019-04-17T13:59:45 Let attribute aliasing logic detect overflow Checking against the max attribute count is not needed as aliasing is implemented correctly below. If not enough locations are available, linking will fail. This works for backends that support aliasing (Vulkan, GLES), but not for D3D11, which asserts. To handle this, add a limitation that informs the frontend of aliased attribute support. Bug: angleproject:3252 Bug: chromium:964404 Change-Id: Ib9ae5d381bbb2e6bf496272fc3d9e88467c17290 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1572817 Commit-Queue: Cody Northrop <cnorthrop@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi b16d69c3 2019-05-13T16:28:27 Vulkan: Add support for surface multisampling A multisample image is created for the surface if multisampling is enabled. Prior to present, this multisample image is resolved into the swapchain image. FramebufferVk::readPixelsImpl similarly has got the ability to resolve the region of interest into a temporary image prior to readback. Tests are added to render a point, line and a triangle on a 4x multisampled surface. Bug: angleproject:3204 Change-Id: I34aca502fa1918b5cbf000ff11521c350372e051 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1610188 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
kmx.io     mail     discord kmxgit v0.4.0