Skip to content

A
angle
Switch branch/tag
  1. 03 Jun, 2021 6 commits
  3. 02 Jun, 2021 13 commits
  5. 01 Jun, 2021 9 commits
  7. 31 May, 2021 5 commits
  9. 30 May, 2021 1 commit
  11. 28 May, 2021 3 commits
  13. 27 May, 2021 3 commits
    • Gold Tests: Fix batch uploading · ce7d80bd
      Yuly Novikov authored 
      Uploading to Gold used to stop after the first failure in the batch

Bug: angleproject:5966
Change-Id: Iff89c106b18a9aab5a43742a6085c5f9a3c9a188
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2923784Reviewed-by: 's avatarYuly Novikov <ynovikov@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
    • Vulkan: Fine-grained depth stencil layout transition barrier · b6bd039c
      Charlie Lao authored 
      Gfxbench Manhattan uses depth attachment as texture in the same render
pass, which creates a feedback loop. The layout we are using for this
has kAllShadersPipelineStageFlags even though it only used by fragment
shader. This creates an unnecessary vertex/fragment dependency that
hurts performance on some mobile GPUs. This CL split
DepthStencilReadOnly to DSAttachmentReadAndFragmentShaderRead,
DSAttachmentReadAndAllShadersRead, and DepthStencilAttachmentReadOnly,
so that if it is used by fragment shader, we only barrier against
fragment stage. If not used by any shader, we don't block any shader
stages. To make names consistent, this CL also renames
DepthStencilAttachmentAndFragmentShaderRead to
DSAttachmentWriteAndFragmentShaderRead, renames
DepthStencilAttachmentAndAllShadersRead to
DSAttachmentWriteAndAllShadersRead.

Bug: b/169891513
Change-Id: Iede497e8af028861f9166a32d122f26dc7aa864a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2453599
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: 's avatarShahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: 's avatarTim Van Patten <timvp@google.com>
    • Vulkan: Disable shadowBuffers feature · 5e631c5f
      Charlie Lao authored 
      For PixelUnpack, if the feature is enabled, it always create a shadow
CPU mapped memory. Then when mapBuffer is called, it let user write to
the shadow memory. At unmapBuffer time, it memcpy the shadow buffer data
to context's staging buffer and then issue a vkCmdCopyBuffer from
context staging buffer to the PixelUnpack buffer. This involves too
many data copies. The proper way to do it is simply make sure to create
the buffer as CPU map-able and let user directly write to it. If we find
cases that CPU waiting for GPU to finish is causing performance issues,
there are two ways to improve/fix it: 1)Add event at end of each
renderpass or FBO so that we know if GPU access is finished or not
without have to wait for entire frame to finish rendering. 2) Create
multiple buffers and ping-pong between them, at least there will be no
VkCmdCopyBuffer call involved like shadow buffers do. But we will leave
that to future time when we find such need.

Bug: angleproject:5986
Change-Id: Ib8300e46e779d20533c1f7f81624de0ce003788b
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2909758
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: 's avatarTim Van Patten <timvp@google.com>
Reviewed-by: 's avatarShahbaz Youssefi <syoussefi@chromium.org>