Workaround Intel Mesa driver issue about T2B copy with stencil aspects
In the test DepthStencilCopyTests/FromStencilAspect we are writing into
gl_FragDepth but we disable the depth writing when creating the render
pipeline, which causes Intel Mesa driver hangs when the test is running
on it. In this patch we try to add workaround to this issue by enabling
the depth writing.
BUG=dawn:439
TEST=dawn_end2end_tests
Change-Id: I618246097a26ee62c830f0ac8dc1aa0ab4d224a9
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/27000
Commit-Queue: Austin Eng <enga@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
diff --git a/src/tests/end2end/DepthStencilCopyTests.cpp b/src/tests/end2end/DepthStencilCopyTests.cpp
index 55cb721..6d2ed23 100644
--- a/src/tests/end2end/DepthStencilCopyTests.cpp
+++ b/src/tests/end2end/DepthStencilCopyTests.cpp
@@ -99,9 +99,6 @@
// Test copying the stencil-only aspect into a buffer.
TEST_P(DepthStencilCopyTests, FromStencilAspect) {
- // TODO(enga): Figure out why this fails on Linux Vulkan Intel
- DAWN_SKIP_TEST_IF(IsLinux() && IsVulkan() && IsIntel());
-
// Create a stencil texture
constexpr uint32_t kWidth = 4;
constexpr uint32_t kHeight = 4;
@@ -121,6 +118,7 @@
renderPipelineDesc.vertexStage.module = mVertexModule;
renderPipelineDesc.cFragmentStage.module = mFragmentModule;
renderPipelineDesc.cDepthStencilState.format = texDescriptor.format;
+ renderPipelineDesc.cDepthStencilState.depthWriteEnabled = true;
renderPipelineDesc.cDepthStencilState.stencilFront.passOp =
wgpu::StencilOperation::IncrementClamp;
renderPipelineDesc.depthStencilState = &renderPipelineDesc.cDepthStencilState;
