kc3-lang/angle/src/tests

src/tests

Log

Author Commit Date CI Message
Shahbaz Youssefi 7ef73a52 2019-05-08T10:48:15 Vulkan: Support flat and centroid qualifiers Bug: angleproject:3219 Change-Id: I2e0b60bda9c3af2b9eadf2efd3899250925897f4 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1600335 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi 5517ce95 2019-05-07T00:27:06 Vulkan: fix qualifier erase for unused vertex attributes Bug: angleproject:3219 Change-Id: I8f8f5572eed5126ab03702953883b313f022a264 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1598548 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi 9fa248e1 2019-05-06T14:55:18 Vulkan: Implement EXT_draw_buffers In GLES, color attachments are referenced by their indices. These indices match between the API and GLSL. For example, if a shader has: layout(location=0) out color; layout(location=3) out roughness; Then GLES would bind and enable GL_COLOR_ATTACHMENT0 and GL_COLOR_ATTACHMENT3. In Vulkan, the framebuffer object and the corresponding renderpass define the color attachments, and they don't allow gaps in color attachments as GLES does. A render subpass creates the mapping between the color attachments as defined in the framebuffer and the attachments used by the shader (with possible gaps). This change packs the enabled GL color attachments for the sake of the framebuffer, and sets the subpass up in such a way that the shaders continue to use the same color output indices as GLES. In the example above, we have the attachment indices as follows: Status | GLES | GLSL | RenderPass | Subpass enabled 0 0 0 0 disabled 1 - VK_ATTACHMENT_UNUSED disabled 2 - VK_ATTACHMENT_UNUSED enabled 3 3 1 1 That is, the array of color attachments in the Vulkan framebuffer/renderpass is: [0] = GL color attachment 0 [1] = GL color attachment 3 And the array of color attachment references in the Vulkan render subpass is: [0] = 0 (index 0 of the renderpass attachment array) [1] = VK_ATTACHMENT_UNUSED [2] = VK_ATTACHMENT_UNUSED [3] = 1 (index 1 of the renderpass attachment array) Bug: angleproject:2394 Change-Id: Ib6cd2b60882643ea152986eee453270d09cd4aed Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1595442 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Jonah Ryan-Davis 776694cd 2019-05-08T10:28:55 Change all ANGLE workarounds to use struct definition with info. Change each workaround from a simple bool to a struct with info including name, workaround set, description, and bug IDs. This will help with future workaround integration with Chrome. Bug: angleproject:1621 Change-Id: Ia27c180abaf845e280060c803e5994cc3152a057 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1593917 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Mingyu Hu ebab670c 2019-04-19T14:36:45 Adding new extension GL_OVR_multiview GL_OVR_multiview functions exactly the same as GL_OVR_multiview2. All GL_OVR_multiview2 tests now also repeat the same test using GL_OVR_multiview Bug: angleproject:3341 Change-Id: I7e5294fb6bbf7692535174a15da6a42e1b5fc4e2 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1575904 Commit-Queue: Rafael Cintron <rafael.cintron@microsoft.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tobin Ehlis 52e0e65a 2019-05-08T12:54:41 Vulkan:Exclude pragma_vertex test on MAC With recent pragma invariant(all) fix the pragma_vertex test is failing on MAC so excluding it from testing. Bug: angleproject:3285 Bug: angleproject:3433 Bug: angleproject:3434 Change-Id: I59079afdbef2d23a593fe0fe061997624fea5cde Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1601364 Commit-Queue: Tobin Ehlis <tobine@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 5cbaa3f8 2019-05-07T15:49:22 Don't inherit ANGLETest SetUp and TearDown. Instead of inheriting from testing::Test's SetUp and TearDown we add new methods 'testSetUp' and 'testTearDown'. This helps prevent a common error of forgetting to call the base class method. Also add a check in the ANGLETest destructor that SetUp and TearDown have been called. Bug: angleproject:3393 Change-Id: Iab211305cc06ffea9ca649e864ddc9b180f2cba0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1593960 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Ian Elliott 19905aea 2019-04-26T01:48:31 Vulkan: Set the Vulkan scissor to the GLES viewport Some Vulkan drivers are not clipping rendering that is outside of the viewport, and the Vulkan spec has some language that indicates that ANGLE should set the scissor: The application must ensure (using scissor if necessary) that all rendering is contained within the render area. In this case, ANGLE is "the application". Bug: angleproject:3253 Bug: angleproject:3254 Change-Id: I6885a3aa6daed57f6ae1c3d974663d763cad10c9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1553973 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Ian Elliott <ianelliott@google.com>
Tobin Ehlis e9421b2c 2019-04-08T15:16:56 Vulkan:Only apply invariant pragma to output vars The "#pragma STDGL invariant(all)" directive should only be applied to shader output vars. This change also removes the workaround SH_DONT_REMOVE_INVARIANT_FOR_FRAGMENT_INPUT which is no longer needed. This change fixes two tests that were incorrectly assuming that the pragma would be applied to inputs: GLSLTest.InvariantAll[Both|In]. Bug: angleproject:1293 Bug: angleproject:3285 Change-Id: I4eb03fa89fbc7c560150ee0cc32382024b0cb3e3 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1558678 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tobin Ehlis <tobine@google.com> Commit-Queue: Tobin Ehlis <tobine@google.com>
Shahbaz Youssefi 216f73d0 2019-04-12T13:32:30 Vulkan: add uniform buffer object support Support for layout qualifiers in interface blocks are added. All interface blocks are adjusted to either be in std140 or std430. In the Vulkan backend, a new descriptor set is added for UBOs. A dirty bit is added for UBO updating and pipeline layouts and descriptor bindings are updated. Bug: angleproject:3199, angleproject:3220 Change-Id: I271fc34ac2e1e8b76dee75e54a7cff0fe15fe4ee Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1565061 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tim Van Patten e76c5c3c 2019-05-03T15:51:30 Vulkan: Enable drawRangeElements() Enable drawRangeElements() for the Vulkan backend and the associated tests. Test: angle_deqp_gles3_tests Bug: angleproject:3420 Change-Id: I98770631ce1387131bb45e2c1e9df24e911bc692 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1595101 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Anders Leino f6cbe44a 2019-04-18T15:32:07 Use D3D11 GetDimensions driver workaround for 2D integer 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. A test is added which reads from non-zero base level integer texture. When the workaround is not being used, reads outside the first quadrant return black. Bug: chromium:679639 Change-Id: I5282a1ba207b2d553d1836f9460ec09cb5590ea6 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1591594 Commit-Queue: Kimmo Kinnunen FI <kkinnunen@nvidia.com> Reviewed-by: Kimmo Kinnunen FI <kkinnunen@nvidia.com>
Geoff Lang d7d42395 2019-05-06T13:15:35 Format all of ANGLE's python code. BUG=angleproject:3421 Change-Id: I1d7282ac513c046de5d8ed87f7789290780d30a6 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1595440 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Geoff Lang <geofflang@chromium.org>
Shahbaz Youssefi 3e62561c 2019-05-06T11:48:52 Suppress test hitting Nvidia shield driver bug The same test was hitting a bug with the Nvidia driver on windows+gles. Bug: angleproject:3413, angleproject:3417 Change-Id: I39426291eac731bcda068829686e09cf406ff661 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1595438 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Jonah Ryan-Davis 66e4850d 2019-04-25T10:49:05 Use EnumAdapters to properly detect primary GPU on Win EnumDisplayDevicesA returns the card that's connected to the display, but EnumAdapters return the adapter which the desktop primary is displayed at index 0. We can use this to determine the device used for graphics. Also cleans up the discrepancy between platforms on finding "primary" vs "active" GPU. Asserts that the GPU expected to run ANGLE commands is the active GPU, and deprecates the primary GPU to be equal to the active GPU. Bug: angleproject:3383 Change-Id: I422fba1bbe47d85b7c09e378d559eaebf89e2625 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1584360 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Shahbaz Youssefi e25ff9d8 2019-05-01T00:02:05 Vulkan: Fix qualifiers for varyings `out` variables in the vertex shader and `in` variables in the fragment shader were not decorated with `layout(location=?)` as they should according to the Vulkan GLSL spec. This change makes sure these decorations are present regardless of whether the compiler sees these variables as EvqVaryingIn/Out or EvqVertexOut/FragmentIn. Bug: angleproject:3412 Change-Id: I66473e876cb989a60b0c6d6a5850a8b2c763d8e2 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1590694 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 58957f3d 2019-05-03T12:52:22 Add option to run each test config in a separate process. This CL adds a command line option to angle_end2end_tests that will iterate over all test configs. For each config it'll fork a new child process that will run only a single config. This will allow us to isolate each config to a specific child process. Hopefully this will reduce test flakiness due to driver issues with multiple configs. The command line option is "--separate-process-per-config". Note that there are about 25 configs right now. Bug: angleproject:3393 Change-Id: Ia117b371bbe159c1b0d28d82befffeb0f40467a9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1591428 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Yi Xu 80147d11 2019-04-30T16:21:24 Add support for dynamically allocate initial stack for Angle By updating YYLTYPE_IS_TRIVIAL to true, YYSTACK_RELOCATE can be enabled, so we can dynamically allocate stack size based on needs. BUG=angleproject:3028 Change-Id: I1b9cc0768cd0c6955589695e703595c56d43f24a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1593954 Commit-Queue: Yi Xu <yiyix@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill 467f174f 2019-05-03T12:52:21 Remove EGLTest. Consolidate these tests into ANGLETest. Add a new parameter to PlatformParameters that lets us skip normal test setup. This removes the last 'configless' tests from angle_end2end_tests. Bug: angleproject:3393 Change-Id: I87186698ade90f95577534eb8ed1dfd4245f740e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1590467 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill 97f0affb 2019-05-03T12:52:21 Add command line option to filter by test config. This allows us to run angle_end2end_tests with a single config without using gtest_filter. It will also allow us to run each test config in a separate process. Bug: angleproject:3393 Change-Id: I09aaf9cfe55a117b0af8d79ecfd129f3d0f1d7c8 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1591427 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Shahbaz Youssefi 051b0896 2019-04-15T15:39:39 Vulkan: Fix layout substitution for struct varyings If the shader contains code such as the following: struct S { vec4 field; }; out S varStruct; The layout qualifier macro is defined as @@ LAYOUT-varStruct @@. However, the Vulkan backend was replacing @@ LAYOUT-field @@. Bug: angleproject:3220 Change-Id: Iae15003867e0bed2cc939159a6653429c7a431e0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1571389 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Jamie Madill 3e251337 2019-05-02T16:12:54 Removed SizedFloat helper class from upload tests. This helper was redundant with the image utils library. Switch the test to use the existing methods. Doesn't affect test correctness. This removes more 'configless' tests from angle_end2end_tests. Bug: angleproject:3393 Change-Id: I49cfbd95c3713b2b4b25c4716a7a279d84ddf994 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1591426 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Yi Xu 926f1caa 2019-05-01T14:03:05 Add support for GCC Compilers Add support for GCC Compilers, so it can be build on Ubuntu. BUG=angleproject:3392 Change-Id: I2f5ac7355a40f52123bd465f86b1e966c59c943d Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1592062 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Yi Xu <yiyix@chromium.org>
Jamie Madill 69f72b41 2019-05-02T16:12:54 Simplify ANGLE test config names. We don't need to use upper case. Lower case can be more readable. Skip a test on Linux/NV/Vulkan that became flaky after changing the test order. Bug: angleproject:3393 Change-Id: Ie027e08ddc9d47b9484cf5b1aaca72a05dd60517 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1574676 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill ddc4d33a 2019-05-01T15:11:46 Remove several underused platform configs. * Removes D3D11 Warp and Reference configs from tests. * Removes several permutations of OpenGL back-end specific configs. * Removes FL 9_3 since it is no longer supported. * Removes present path "Copy" since it's redundant with normal D3D11. Reduces number of configs from over 60 to 25. Bug: angleproject:3393 Change-Id: Ia5a23de3c4865b17ee50673a4066757b901a4b5a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1574675 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill 3089f92e 2019-05-01T15:11:46 Move EGL Display config into EGLPlatformParameters. This CL moves all of the EGLDisplay configuration into an EGL platform struct. Consolidating display configuration in the struct allows us to move configuration out of the test constructor. Then when we filter test configs we don't need to wait for the individual test setup. Bug: angleproject:3393 Change-Id: I5bd06dcdc9f2867ebc43c1d4984077ada35cafc8 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1574674 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Shahbaz Youssefi fc0be049 2019-04-18T09:44:06 Fix GL_MAX_COMBINED_UNIFORM_BLOCKS query This was previously returning maxCombinedTextureImageUnits instead of maxCombinedUniformBlocks. Fixing that exposed a bug in the GL backend where the combined values were sometimes capped to vertex+fragment values and sometimes not capped at all. The reasoning for such capping was that the combined limits as queried from GL contains stages not available in GLES (such as tessellation). However, the capping failed to take geometry shader limits into account. This change adjusts such capping to vertex+fragment+geometry. It also applies the cappping to all combined limits, rather than a select few. Bug: angleproject:2099 Change-Id: I7231058b5d7f80b1b2452d9f87d4b0ab6e1cdb17 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1572487 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Jamie Madill b8149075 2019-04-30T16:14:44 Clean up ANGLE test extension functions. None of these functions needed to be member functions. Also make the naming more consistent. Bug: angleproject:3393 Change-Id: I7aafe2269a48af703a87bd9a8cf4cfab9e177dd3 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1574673 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Jamie Madill 0659c991 2019-04-30T16:14:43 Set swap interval explicitly. The swap interval can be changed independent of the Surface config. Thus it makes more sense to set it explicitly in test setup. This simplifies the test config. Also updates some of the API for GLWindowBase. Return an explicit error from makeCurrent. Bug: angleproject:3393 Change-Id: Ic62b33018e872bc0e38f2848e2427ed898b60749 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1574672 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
shrekshao 15ce8227 2019-03-18T19:25:21 Fix error report when active color buffer has no fs output Also modify or remove some tests to sync up with the expected behavior stated in spec. Related to https://github.com/KhronosGroup/WebGL/pull/2780 If any draw buffer with an attachment does not have a defined fragment shader output, draws generate INVALID_OPERATION. Also remove Framebuffer masking for inactive outputs. This workaround is no longer necessary as the WebGL spec has changed. It also was never fully working and had bugs with certain orders of calls. Bug: angleproject:2872 Bug: chromium:927908 Bug: chromium:943538 Change-Id: I73715a6ab851ae3db7096f49ea0a9fdd6f576703 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1530018 Commit-Queue: Shrek Shao <shrekshao@google.com> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Jonah Ryan-Davis 1ad7a072 2019-04-26T14:05:25 Clean up unexpected passed/failed test behavior for generate_stats The lists of tests can't be added to the sheets if more than 50,000 lines. Since the goal is to get rid of all unexpected passes/failures, it's reasonable to clip the lists by length. Other formatting changes too. Bug: angleproject:3398 Change-Id: Ice461862fb4266e0ad1280f20de85224fc6d7e97 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1585612 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
Geoff Lang 6f691fbb 2019-04-25T11:01:52 Port the WebGL2 framebuffer-texture-changing-base-level test to native A nearly direct port of the https://www.khronos.org/registry/webgl/sdk/tests/conformance2/rendering/framebuffer-texture-changing-base-level.html test for easier debugging. BUG=angleproject:2291 Change-Id: Ibc8a7e77629f1ea3f15e92ea93f1210fa850c524 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1584361 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Shrek Shao <shrekshao@google.com>
Jonah Ryan-Davis 51ac37ee 2019-04-25T13:32:16 Print unexpected passed/failed tests with formatting for generate_stats The stats generation should pick up a list of unexpected failed/passed tests, so the tests should be printed with the format used by this script. Bug: angleproject:3398 Change-Id: I96a8eba6c2e294f64ca59ec60042cc688d1133c5 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1584363 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
Geoff Lang fb6fbdbc 2019-04-12T15:16:30 Return that polling queries have completed after context loss. KHR_robustness says that certain queries that poll such as GetSynciv with SYNC_STATUS should still generate errors after context loss but also return that the result is available. BUG=angleproject:3379 Change-Id: Ibf61f8481bc7c3d1b4fa8979dc404d0ca3952fcd Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1566142 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Kenneth Russell <kbr@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Geoff Lang 5a0f64b6 2019-04-23T16:06:47 D3D: Require cube completeness to create render targets TextureD3D_Cube::isFaceLevelComplete skips checks for the base level because it assumes that isCubeComplete has already been called to verify the base level. This assumption is not correct for some cases like glCopyTextureCHROMIUM which attempts to create a render target from cube maps that may not be complete. This caused us to create render targets from incomplete cube maps and then fail to upload data to the cube faces because the sizes or formats did not match between the images and render target. TEST=conformance/more/conformance/quickCheckAPI-S_V.html BUG=angleproject:3384 Change-Id: Iceb9415debfa07436fea1eb4bc23d0ec05679839 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1578957 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Xinghua Cao 1b0acbb5 2019-04-23T15:17:56 Add cases with multiple draws and dispatches D3D11 cannot allow the same (sub)-resource bound as both a SRV and an UAV at the same time. If render pipline and compute pipeline access the same texture, must unset conflicting SRVs and UAVs before drawing and dispatching on D3D backend. Currently, these cases could not pass on D3D11 backend. Bug: angleproject:3152 TEST=angle_end2end_tests.ComputeShaderTest.* Change-Id: I3ed19d7586625aac7fe9f46f86341d136d148019 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1578131 Reviewed-by: Jiajia Qin <jiajia.qin@intel.com> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Xinghua Cao <xinghua.cao@intel.com>
Tobin Ehlis 47b5b1c2 2019-04-22T13:32:31 Vulkan:Improve white box test documentation Add target for angle_white_box_perftests to DevSetup.md Android build instructions. Generously comment the test source to clarify various tests and how to run them. Bug: angleproject:3136 Change-Id: Ie4f0ec6951d354ebac861f727c9d550a93ad48c9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1577740 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Tobin Ehlis <tobine@google.com>
Jamie Madill 0631e19b 2019-04-18T16:09:12 Vulkan: Rename Vulkan "Texture" format to "Image" Also adds some comments to vk_format_utils.h. Bug: angleproject:3372 Change-Id: I529b9b189e4cdfd400c3c981a47139727d9954ab Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1565062 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jonah Ryan-Davis 3b775865 2019-04-18T15:59:15 Skipped tests not accumulated in dEQP stats output. The sSkippedTests is not incremented in the kGpuTestSkip check in runTest, when it should be. Bug: angleproject:3390 Change-Id: I4aca73ba1bbe3060fc3c0cd27679c2276a30a741 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1574620 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
Yuly Novikov d18c048a 2019-04-04T19:56:43 Remove Nexus 5X Vulkan test suppressions. SimpleStateChangeTest.DrawArraysThenDrawElements TextureUploadFormatTest.All VertexAttributeTest.UnsignedByteNormalized VertexAttributeTest.ByteNormalized EXTMultisampleCompatibilityTest.DrawAlphaOneAndResolve was skipped on all Android for unspecified reason in https://chromium-review.googlesource.com/c/angle/angle/+/451877 seems to be passing now. Make some suppressions OpenGL ES specific: MultisampleCompatibilityTest.DrawCoverageAndResolve RobustResourceInitTest.TexImageThenSubImage RobustResourceInitTest.ReadingPartiallyInitializedTexture RobustResourceInitTestES3.MultisampledDepthInitializedCorrectly RobustResourceInitTest.MaskedDepthClear RobustResourceInitTestES3.MaskedDepthClearBuffer RobustResourceInitTest.MaskedStencilClear RobustResourceInitTestES3.MaskedStencilClearBuffer TransformFeedbackTest.OffsetResetOnBeginTransformFeedback Bug: angleproject:3262 Bug: angleproject:3124, angleproject:2657, angleproject:2797 Change-Id: I3ed7a3f0ebbb4598bff75e2b3bd8b3b0630779f8 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1572484 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Michael Spang f02a767d 2019-04-09T18:45:23 Vulkan: Implement glTexStorageMem2DEXT This implements support for creating textures that alias vulkan images allocated inside external memory. Bug: angleproject:3289 Change-Id: Iad071f353a217793102ae737647c7cd572f7b0ad Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1552029 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Michael Spang <spang@chromium.org>
Shahbaz Youssefi 422c94bd 2019-04-12T13:37:19 Vulkan: Enable ES3 fragment output attribute locations GLSLTest_ES3 end2end tests are enabled for Vulkan, though with suppressions for other features that are missing. Bug: angleproject:3199 Change-Id: Ie744e160eab2df9a5a4f2c67ea5acf4b865ea5bf Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1565058 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
jchen10 82bcffa7 2019-04-17T10:45:25 Disable parallel compile for d3d11 compute To observe whether the flaky issues still show up with the feature disabled. Bug: angleproject:3349 Change-Id: I0eae861ad76212fb4a65b57c1e1b5e8c583d6a3b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1571027 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jie A Chen <jie.a.chen@intel.com>
Jonah Ryan-Davis 0f7a2330 2019-04-17T11:13:41 Modify test result output from dEQP tests for visualization script. Test results will now be prepended with [TESTSTATS], and there are two new outputs added for unexpected passes and unexpected failures. Bug: angleproject:3369 Change-Id: Iec4dafe695531dcf56d496083036ccf1e286ead7 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1571388 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Geoff Lang 057b76e1 2019-04-12T14:48:59 Implement GL_CHROMIUM_lose_context. This is useful for testing context lost behaviour. BUG=angleproject:3379 Change-Id: If0e1538553b1761e313fc36ccde5138cd495200f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1566141 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Michael Spang 3b2c6bfd 2019-04-16T17:19:50 Vulkan: Implement glImportMemoryFdEXT Allow importing opaque file descriptors into memory objects on linux. Currently this just holds onto the file descriptor rather than calling vkAllocateMemory immediately. The latter will be easier once we have support for suballocation (anglebug.com/2162). Bug: angleproject:3289 Change-Id: Ia80ce07b2a9ec95b9063feb9bfeb24ffe77fa40e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1552028 Commit-Queue: Michael Spang <spang@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jonah Ryan-Davis c832cdd7 2019-04-15T14:53:01 dEQP test_expectations parser should output an error on unused lines. dEQP test_expectations parser should allow nested overrides. When an expectation line is not used for any test, it's likely duplicated or misspelled, or the test was removed. The parser should output these errors to make it easier to keep the expectations up to date. Also necessary to allow longer/more specific expectations to override shorter ones, so the scenario where you FAIL one set of tests across platforms, and SKIP a subset of that group on one platform will still be allowed. Bug: angleproject:3368 Change-Id: If0470cb6fb12f8d3daa97df0c701185fa086668c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1564725 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Mingyu Hu 7d64c486 2019-03-12T14:27:40 GL_ANGLE_multiview has been renamed to GL_OVR_multiview2. changes include: 1) GL_OVR_multiview to GL_OVR_multiview2 extension directive change 2) Removal of all references to side by side. We no longer support multiple views in a single 2DTexture. Only 2DTextureArray's are supported 3) WebGL 2 (ES3) is required for multiview Bug: angleproject:3341 Change-Id: Ie0c1d21d7610f8feebdb2e4d01c6947f57e69328 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1552023 Commit-Queue: Rafael Cintron <rafael.cintron@microsoft.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Corentin Wallez e02ad4d3 2019-04-16T14:12:37 Suppress tests crashing on teardown on Win Intel D3D11 BUG=angleproject:3349 Change-Id: Ib86f8b87d719919c980240d3220521a3c7f8a942 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1569466 Commit-Queue: Corentin Wallez <cwallez@chromium.org> Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Qin Jiajia e7da32f1 2018-12-27T15:31:47 Fix the failure in UpdateImageTextureInUse If we update the texture data between two dispatch calls, ReadPixels can't get right result after the second dispatch call. The failure reason is that after the first dispatch, ReadPixels will sync framebufer state which will update color render target. Finally, TextureD3D::ensureRenderTarget is reached. However, we are in compute pipeline. mTexStorage->isRenderTarget() will be false. That results the current texture will create a new render target storage. But the UAV is still bound with the previous texture storage. If there is no texture dirty bit between these two dispatch calls, applyTexturesForCompute won't be called. After the second dispatch, readPixels will read data from the new texture storage which is not updated. Bug: angleproject:3015 Change-Id: Ib2494ab8bf6e12faefc0a7370719d383526c36ba Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1390710 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jiajia Qin <jiajia.qin@intel.com>
Jonah Ryan-Davis ccd91316 2019-04-12T14:13:22 Use primaryGPUIndex over activeGPUIndex when determining the system GPU. Mac OPENGL/AMD dEQP test suppressions were being ignored because in SystemInfo the activeGPUIndex was used when the primaryGPUIndex should have been used. Bug: angleproject:3373 Change-Id: I2664049b956bf8903c6d7094ec27646766e7ce16 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1565056 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Shahbaz Youssefi d3ec341d 2019-04-10T11:37:07 Fix LineLoopStateChangeTest to actually draw square and hourglass The test claimed to be drawing an hourglass followed by a square. That is: ---- \/ /\ ---- followed by: ---- | | | | ---- With the end result being: ---- |\/| |/\| ---- However, in reality it was drawing two hour glasses with the same result. That is: ---- \/ /\ ---- followed by: |\/| |/\| Bug: angleproject:3361 Change-Id: I59a2930f78e6e448b6ccb2e5f70c73ff0f15a2ea Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1561650 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jonah Ryan-Davis 5646a7cc 2019-04-12T16:42:56 Uniform buffers not dirtied for gl_VertexID workaround on D3D11. Related to: https://chromium-review.googlesource.com/c/angle/angle/+/1520988 The program uniform buffers were dirtied for this edge case, but it should have been the driver uniform buffers. This CL fixes this and adds a tests case which catches this. Bug: 945903 Change-Id: I6142771e200513ed0251fc97cec68d371d39ec1a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1565059 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Shrek Shao <shrekshao@google.com>
Courtney Goeltzenleuchter 8f0210f7 2019-04-10T10:44:10 Always enable NPOT for Vulkan backend Vulkan natively supports non power-of-two textures so indicate that support for front-end. Bug: angle-project:3239 Test: angle_deqp_gles2_tests --gtest_filter=dEQP.GLES2/functional_texture_completeness_2d_npot_t_repeat --use-angle=vulkan Change-Id: Id9058b6e9afd1d43e5d74612f808ce39beafd35f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1561963 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Jamie Madill 23667b7e 2019-04-10T12:49:56 Update test result output from ANGLE dEQP gtests. * prints lines for total, passed, failed, not supported, skipped, and exception results. * prints unexpected passed and failed tests each individually using the dEQP name of the test. * moves skipped test detection into the test loop. skipped tests will now print a message that they've been skipped. Bug: angleproject:3369 Change-Id: Idbfc0154a658eda3f52b5bfd30fbff0860a00133 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1561970 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Jamie Madill 4877ef35 2019-04-12T10:36:00 Ensure deleteTexture preserves correct texture cache. Also adds a test contributed by jgilbert@mozilla.com. Bug: angleproject:3375 Change-Id: Ibd52daa074bf53b2b213193ccf5a612705a89c67 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1565052 Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Yuly Novikov a95940c6 2019-04-12T10:54:05 Suppress flaky dEQP 3.1 test functional.shaders.builtin_functions.uniform.findLSBMinusOne.highp_compute on D3D11 Bug: angleproject:2619 Change-Id: Ia7a83429d03b871967d2d6c5ecd3ce82111b98fd Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1565427 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Tobin Ehlis 9af10e8a 2019-04-11T16:20:57 Incorrectly placed extension directive is an error The ESSL 1.00 spec states "...the extension directives must occur before any non-preprocessor tokens." This change makes the occurance of an extension directive after non-preprocessor tokens a compilation error. Bug: angleproject:3285 Change-Id: I9f8631366151c92fb6d8b4408596917012d9a3e9 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1565411 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Commit-Queue: Tobin Ehlis <tobine@google.com>
Yuly Novikov e29b61e6 2019-04-11T17:02:42 Skip ParallelShaderCompileTestES31.LinkAndDispatchManyPrograms Flaky on Win NVIDIA D3D11 Bug: angleproject:3359 Change-Id: I56a0e4266f1cd0913852a43852666dee055a8e00 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1564562 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Shahbaz Youssefi 2249d4a7 2019-04-05T16:48:55 Vulkan: remove clear depth ability from clearWithDraw This partially reverts the following change: 60ec8f576 Vulkan: break dependency to the depthClamp feature The feature is no longer necessary, and simplifies the usage of utility shaders. Bug: angleproject:2361 Change-Id: I1e87ac8d2517c5a3b50e3d0cddd55c852e0e3e7d Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1555313 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Tobin Ehlis <tobine@google.com>
Shahbaz Youssefi 127990f9 2019-04-04T13:52:04 Vulkan: Use render pass loadOp for scissored clears At this point, every clear is done through render pass loadOp, except masked color or stencil clears. The only fallback is clearWithDraw, that can clear both color and stencil at the same time. Bug: angleproject:2361 Change-Id: I805fc12475e832ad2f573f665cdfeb766e61a6d0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1553740 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tobin Ehlis <tobine@google.com>
Shahbaz Youssefi d581f918 2019-04-11T10:21:05 Vulkan: few more EGL dEQP suppressions on Android Follow up to c02ef98fb199341d644e4751ac682d858595420e. Bug: angleproject:2546 Change-Id: Ibad7befc5bed95544c0cf15a79d25b47ec8a00bd Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1564710 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Yuly Novikov e6b23e45 2019-04-10T17:19:15 Skip Texture2DTestES3.TextureImplPropogatesDirtyBits on Win AMD GL Bug: angleproject:3371 Change-Id: I7601e6b658835b9c948acab98491f5bd01ea9497 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1562517 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Jonah Ryan-Davis 34cbc544 2019-04-05T11:30:10 Use ANGLE's gpu_info_util libraries for test expectations 3. Remove gpu_info.h and gpu_info.cc and use ANGLE's own gpu_info_util to to gather device info instead. - Support collecting info for and parsing expectations of specific Android devices such as Nexus 5X and Pixel 2. - Change parser behavior to more closely follow: bit.ly/chromium-test-list-format Bug: angleproject:2677 Change-Id: I4c0b9342142b718e884484a6bcaac2dff3ac575a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1553822 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Yuly Novikov 893ccb48 2019-04-09T16:19:20 Skip EGLSurfaceTest.SwapInterval on Nexus 5X and 6P GLES Bug: angleproject:3364 Change-Id: I51dd45cfcd1c548266712b71eac0dca60574e69e Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1558960 Commit-Queue: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi c02ef98f 2019-04-09T16:09:44 Vulkan: update GLES3 and EGL expectations Few GLES3 expectations were updated in preparation for running those tests on the bots (one change on Linux, a handful on Android). A number of EGL tests were discovered to crash on Android. They are suppressed as well. Note that most of these tests were previously marked as failing or flaky on the GLES backend as well, and they are marked as SKIP everywhere instead. Bug: angleproject:2341 Bug: angleproject:2546 Bug: angleproject:2950 Bug: angleproject:3186 Change-Id: If52d5faeea994c8b6200b713b3c2f1b1a1aeff92 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1560251 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Michael Spang a0b00e97 2019-04-09T18:45:22 Vulkan: Expose GL_EXT_memory_object_fd & GL_EXT_semaphore_fd If the vulkan driver has support for VK_KHR_external_memory_fd or VK_KHR_external_semaphore_fd, add the GL versions of these to the vulkan renderer's extensions. Bug: angleproject:3289 Change-Id: I7f04b5cf883f93f6ccd579c2b75d6831b854bfd0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1552027 Commit-Queue: Michael Spang <spang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tim Van Patten f39b4f02 2019-04-09T12:04:51 Vulkan Android: Remove workaround to clamp the point size This is to remove the workaround introduced by anglebug.com/2599 since Nexus 5x devices aren't in the farm anymore and the tests pass on Pixel devices. Bug: angleproject:2599 Test: angle_deqp_gles2_tests Test: angle_end2end_tests Change-Id: I14f724494909486a2164ddd734a95b6980429f29 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1559202 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill e4458b7b 2019-04-08T16:26:50 Fix glCopyTexSubImage3D. Two bugs were present in our implementation. We were using the y offset for z in ensureSubImageInitialized. And for our D3D back-end we were potentially reading from the wrong image index. Bug: chromium:947342 Change-Id: If39671a911e08fcc641b9ba6f5910e3a2c16eb5d Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1558671 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Geoff Lang 3702d8c9 2019-04-08T13:44:06 Propogate dirty bit signals from TextureImpl to Texture up to Context. If TextureImpl sets a local dirty bit and signals gl::Texture of it, the dirtyness is not propogated to context. This can result in draw calls with textures that are not synchronized BUG=949985 Change-Id: I9baf82c96598265a6a4850f1fd48e213b5e98ab5 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1556699 Commit-Queue: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Tim Van Patten 7d689863 2019-04-08T12:31:33 Re-Enable MipmapTest.TextureCubeGeneralLevelZero/ES2_VULKAN The test MipmapTest.TextureCubeGeneralLevelZero/ES2_VULKAN was disabled for Vulkan Android due to Nexus 5x and Pixel 2 issues. The Nexus 5x has been removed from the test infrastructure and the Pixel 2 issue seems to have sorted itself out. Bug: angleproject:2817 Test: angle_end2end_tests MipmapTest.TextureCubeGeneralLevelZero/ES2_VULKAN Change-Id: I85bb56707def2906194c34942fcd74a811255abc Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1558293 Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Tim Van Patten <timvp@google.com>
Yuly Novikov 0086a8ac 2019-04-09T09:15:59 Skip LineLoopStateChangeTest.DrawElementsThenDrawArrays on Win Vulkan AMD Bug: angleproject:3361 Change-Id: I39c595e1b8ab73520422e2dd3433cd007a1415a1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1559789 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Shahbaz Youssefi 65d10f3b 2019-03-21T16:30:31 Vulkan: Implement robust resource initialization If a texture or renderbuffer needs to be cleared for robust access or due to having emulated channels, it is immediately cleared. The former relies on a front-end feature that optimizes robust access clears only to levels and layers that are not fully initialized through data upload. Bug: angleproject:2722 Change-Id: Icdab856eb4ffe963f78569b6d80d9ff5cb27ff9b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1535056 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
shrekshao 8413faba 2019-04-04T17:13:18 Fix formsRenderingFeedbackLoopWith check To make it pass the following webgl conformance test https://github.com/KhronosGroup/WebGL/blob/master/sdk/tests/conformance/rendering/rendering-sampling-feedback-loop.html It used to fail due to 1. Didn't check if texture unit is sampler complete 2. Only checked active drawbuffers. But drawbuffer settings shouldn't be taken into account when checking drawing feedback loop. On top of applying these 2 functional fixes, I also tried to do some optimization by unwrapping the nested for loop for program sampler bindings and texture unit in `Program::samplesFromTexture` and putting them outside of the draw buffer loop according to the old comment by Antonie @piman. https://codereview.chromium.org/2461973002/ > ... turning it around (for each texture check if it's also an attachment, instead of for each attachment check if it's a bound texture). In particular, we have an upper bound on the number of attachments, so we can look them up outside the loop into a fixed size buffer on the stack - and the very common case will be to only have 1 of them making the inner loop cheap. But this unwraps sort of breaks the code structure. An alternative way would be passed in a framebuffer pointer into `Program::samplesFromTexture` but that would ends up in tight class coupling. I am a bit unsure here. Would like to hear if think this change is okay in terms of code style. In addition to further speed up this check (as it runs for every draw validation) I added a cache mLastColorAttachmentId indicating the last i of GL_COLORATTACHMENTi that is not GL_NONE to shorten this inner loop. In most scenario we won't have up to max number of color attachments. A side note: this is still failing https://github.com/KhronosGroup/WebGL/blob/master/sdk/tests/conformance2/rendering/depth-stencil-feedback-loop.html But it's because the test case doesn't fit the spec at this moment. I will update this test later. Bug: chromium:660844 Change-Id: I6d718dada71a5d989caac04de03f2454f2377612 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1553963 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Shrek Shao <shrekshao@google.com>
Jamie Madill bfc5df16 2019-04-08T12:54:56 Remove function pointers in CopyTexture3DTest. No longer needed with the ANGLE loader. Bug: None Change-Id: If206c529efaee5b81b68059341dcc6e841126842 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1556698 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi efec568b 2019-04-03T15:19:26 Expand clear tests for more mask combinations The test was previously either masking every three aspect (color, depth and stencil) or none. This was not exercising every clear path in the Vulkan backend. Bug: angleproject:3241 Change-Id: Ief4085ea302ec17bffe30b1f8510ae357fd01290 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1551523 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Shahbaz Youssefi f6c937f8 2019-04-02T17:04:08 Vulkan: fix masked stencil clear Previously, masked stencil clear was done by clearing every stencil bit to the ClearValue & Mask. The correct behavior as implemented in this change is to clear only the bits that are set in Mask. This can only be done through a draw call, with ClearValue as the stencil reference, and Mask as the stencil write mask. Note: this change relies on the depthClamp Vulkan feature which is not available on ARM. Bug: angleproject:3241 Change-Id: I0a181c32f881ee813f144e7bdd6f42c8ea6f1966 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1548442 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Tobin Ehlis <tobine@google.com>
Tim Van Patten 5f388c24 2019-03-14T09:54:23 Vulkan: dEQP-EGL.functional.image.modify.* Determine the destination internalFormat correctly. dEQP-EGL.functional.image.modify.renderbuffer_rgb565_tex_subimage_rgb8 dEQP-EGL.functional.image.modify.renderbuffer_rgba4_tex_subimage_rgba8 dEQP-EGL.functional.image.modify.tex_rgb565_tex_subimage_rgb8 dEQP-EGL.functional.image.modify.tex_rgba4_tex_subimage_rgba8 dEQP-EGL.functional.image.modify.tex_rgba8_tex_subimage_rgba5_a1 dEQP-EGL.functional.image.modify.tex_rgba8_tex_subimage_rgba4 dEQP-EGL.functional.image.modify.tex_rgba5_a1_tex_subimage_rgba8 dEQP-EGL.functional.image.modify.tex_rgba5_a1_tex_subimage_rgba4 dEQP-EGL.functional.image.modify.tex_rgba4_tex_subimage_rgba8 dEQP-EGL.functional.image.modify.tex_rgba4_tex_subimage_rgba5_a1 dEQP-EGL.functional.image.modify.renderbuffer_rgba4_tex_subimage_rgba8 dEQP-EGL.functional.image.modify.renderbuffer_rgba4_tex_subimage_rgba5_a1 dEQP-EGL.functional.image.modify.renderbuffer_rgb5_a1_tex_subimage_rgba8 dEQP-EGL.functional.image.modify.renderbuffer_rgb5_a1_tex_subimage_rgba4 Bug: angleproject:3170 Change-Id: Ic9870390b2d4a0dcbe561efd3fb1597faadb7d79 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1524404 Commit-Queue: Tim Van Patten <timvp@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Michael Spang fb201c5e 2019-04-03T14:57:35 Implement resource management for GL_EXT_memory_object This implements glCreateMemoryObjectsEXT, glDeleteMemoryObjectsEXT, and glIsMemoryObjectEXT. It's not possible to do anything useful with them yet. Bug: angleproject:3289 Change-Id: I8882b657e9de564b5f97f8dea87838f67b1928f8 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1552025 Commit-Queue: Michael Spang <spang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi 43997017 2019-03-30T23:24:01 Vulkan: fix non-float clear with draw Instead of using one draw call that clears all attachments, multiple draw calls are issued that clear a single attachment each. This allows us to have a manageable number of variations for the ImageClear.frag shader, now that non-float format support is introduced. Bug: angleproject:3187 Change-Id: Ic0c1067a396250bd80f31d00cad5a272acff8be8 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1545523 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi f9f18ef0 2019-04-03T10:07:18 Vulkan: Allow vertex-only pipelines This allows issuing draw calls which only manipulate depth/stencil. Bug: angleproject:3241 Change-Id: I62ab18a185ea5b234d3559f30c5b2b8ecb317bbb Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1550900 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 0719a88e 2019-04-03T13:15:17 Fix OOB access for dynamic attribs with offsets. We were not properly adding the offset to compute the right bounds. Bug: chromium:943709 Change-Id: I93e714b46dd366d5833fffa858ea3ab0322ffa92 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1548441 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Shahbaz Youssefi 91524d8c 2019-04-02T15:31:30 Disable Clear and TextureUpload perf tests on D3D11 The tests crash on Nvidia in the driver. Bug: 945415 Change-Id: I0f635eb5d99ecaeeaff055793b1b5fd3a314b055 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1548440 Commit-Queue: Jamie Madill <jmadill@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 24980278 2019-04-03T09:03:51 D3D11: Skip DepthTexturesWithMipmaps on AMD/D3D11. Possible driver bug. Bug: angleproject:3342 Change-Id: Ia171a7ac0d764859c4440b1fe2986aa931ff75c2 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1550799 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
jchen10 89899748 2019-01-16T08:32:25 ParallelCompile: D3D compute This parallelizes the compiling and linking for compute shaders on the D3D backend. Bug: chromium:849576 Change-Id: Idd6b418cb9c2448209c15eab2756599f8ff7af4c Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1415725 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jie A Chen <jie.a.chen@intel.com>
James Darpinian 2889dff6 2019-03-27T16:25:11 Mac: Support using an IOSurface as the default framebuffer Bug: angleproject:2764 Change-Id: I3fdab330b59ed996f68e3063debca29323a66cf0 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1542599 Commit-Queue: James Darpinian <jdarpinian@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi db4ed317 2019-03-29T00:32:45 Vulkan: glClearBuffer* implementation Refactors FramebufferVk::clear such that specific render targets could be cleared, with clear values not necessarily set through glClearColor etc. FramebufferVk::clearWithRenderPassOp is modified so that loadOp and clear values are set after the render pass has been registered in the graph. This allows multiple glClearBuffer calls to coalesce into the same render pass. glClearBuffer calls are then implemented simply as calls to the refactored clear function with the appropriate parameters. Bug: angleproject:3187 Change-Id: I2fdfcbea5bf244f63ec981b91caca47f5ee3cd3a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1545204 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Tim Van Patten <timvp@google.com>
Jonah Ryan-Davis 750935f4 2019-04-02T11:27:19 Ignore flaky dEQP failures on Windows More flaky failures that were being hidden by test retries. Might take a few more patches in order to disable test retries. Bug: angleproject:3271 Bug: angleproject:1729 Change-Id: I97eb02d6e93892a04db8b09992f32c4c20dbd49b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1549314 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
Shahbaz Youssefi d3fd67e0 2019-03-29T23:22:34 Add a test for glClearBuffers of render targets of same FBO This test is to ensure the upcoming Vulkan implementation can coalesce all the calls into a single render pass clear. Bug: angleproject:2361 Change-Id: I816907b3512715c1d3217689413b6c8e9842a37a Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1544973 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Shahbaz Youssefi 1d72f658 2019-03-29T22:42:45 Add a test for glClearColor followed by glClearBuffer This test is to ensure the upcoming Vulkan implementation uses the right colors for each render target. Bug: angleproject:2361 Change-Id: Ibc6f8b474ac598a20d8cc9ac1c7fd18667370cbf Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1544972 Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Shahbaz Youssefi 6cb0cff5 2019-03-28T11:01:22 Add a test for masked/scissored multi-attachment clear Done after noticing a bug in the Vulkan backend where masked (and possibly scissored) color clears were only clearing the first attachment. Bug: angleproject:2361 Change-Id: I471d337b1c3a3a5e17748690ae2222a6351773ac Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1544971 Reviewed-by: Yuly Novikov <ynovikov@chromium.org> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Jonah Ryan-Davis 3b14b786 2019-03-28T12:12:24 Ignore flaky failures from ANGLE's dEQP deps on Win/Linux/Mac. Flaky failures were being hidden by test retries. Should be ignored so we can disable test retries. Bug: angleproject:3271 Change-Id: I17708f96e2299b9828f5f979f68187b3cd99ecb3 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1542259 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
Yuly Novikov d02c17f9 2019-03-28T16:49:39 Mark vertex attributes dirty during context switches gl::State::mDirtyCurrentValues is a second tier of gl::State::DIRTY_BIT_CURRENT_VALUES bit. Marking mDirtyCurrentValues dirty during context switches treats them in the same way as other gl::State::mDirtyBits are treated. This has regressed in https://chromium-review.googlesource.com/c/angle/angle/+/722423 where individual current values bits were squashed into one. Bug: chromium:920033 Change-Id: I778515f11c975f8d88bd2d8c08c49160fd011497 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1545009 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Jamie Madill 9cce3cd9 2019-03-27T15:24:12 Update texure cache after teleting bound texture. The texture cache could become out of sync. And we could end up dereferencing an invalid pointer. Bug: chromium:943538 Change-Id: I6a99a04e80fc551b6177e25b7bee09c6ae226340 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1541718 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com> Commit-Queue: Jamie Madill <jmadill@chromium.org>
Tobin Ehlis 345cb856 2019-03-28T10:04:24 Vulkan:Clamp stencil ref value We were casting stencil ref value instead of clamping it as the spec requires. Fixing that which allows a number of new dEQP tests to pass. Bug: angleproject:3244 Bug: angleproject:3245 Bug: angleproject:3247 Bug: angleproject:3248 Change-Id: I59242d59982b990e82395f5a1e77b9a54254dc2b Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1541751 Commit-Queue: Tobin Ehlis <tobine@google.com> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Courtney Goeltzenleuchter 5a604a5f 2019-03-27T14:22:50 Ensure framebuffer is up-to-date before drawing The dEQP-GLES2.functional.fbo.render.texsubimage.between_render_tex2d* test was failing because a TexSubImage call modifying the framebuffer wasn't being applied before a subsequent DrawElements call and thus getting an incorrect result. Bug: angleproject:3249 Test: angle_deqp_gles2_tests --gtest_filter=dEQP.GLES2/functional_fbo_render_texsubimage_between_render_tex2d_rgb --use-angle=vulkan Change-Id: I90f791957e6536c5c62fb731d52cd486ab5c05c1 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1542361 Commit-Queue: Courtney Goeltzenleuchter <courtneygo@google.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Shahbaz Youssefi 0c128e15 2019-03-25T23:50:14 Vulkan: Use render pass ops to clear images when possible On tiling GPUs, render pass loadOp and stencilLoadOp can be used to very cheaply clear an image as it is being render to. This change uses this feature to clear render targets when possible. Bug: angleproject:2361 Change-Id: Ic4bdc908873f4802760d549f4893f84a47beac0f Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1500576 Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 2e5afde1 2019-03-28T09:39:49 Add suppressions for Mac dEQP [Part 3]. Bug: angleproject:3317 Change-Id: Ie60ad9670d19d0eae9926bcf14f65fc2e84215d8 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1542602 Reviewed-by: Jamie Madill <jmadill@chromium.org> Commit-Queue: Jamie Madill <jmadill@chromium.org>
James Darpinian 5fd08af4 2019-03-13T19:35:36 Sampler state overrides texture state if set The new validation added in http://crbug.com/809237 failed to consider that sampler object state overrides texture object state if a sampler object is bound. State caching makes this complicated to fix. Fixes WebGL conformance test incompatible-texture-type-for-sampler.html https://github.com/KhronosGroup/WebGL/pull/2823 Bug: 940080, 809237 Change-Id: I26b0fb35c5630c36248edae80f0298a0cb7e14b8 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1522364 Commit-Queue: James Darpinian <jdarpinian@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jonah Ryan-Davis 198de612 2019-03-27T14:26:51 Remove EGLThreadTest to fix angle_end2end_tests failures on Android. EGLThreadTest doesn't clean up well and the test framework can't handle it well. It caused test failures related to eglMakeCurrent (ELG_BAD_ACCESS) on Android that were masked by retries. Calling eglGetDisplay within a test can cause it to return a cached display from the test runner. The test is unable to know if it should terminate this display. bug: angleproject:3321 Change-Id: I14a539a00acaed4ee71622e7416a6c7a75596606 Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1541717 Commit-Queue: Jonah Ryan-Davis <jonahr@google.com> Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
kmx.io     mail     discord kmxgit v0.4.0