-
Notifications
You must be signed in to change notification settings - Fork 116
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Move LowerGpuRt into LGC #2852
Move LowerGpuRt into LGC #2852
Conversation
With some tweak to adapt the change.
Test summary for commit d15e7f2CTS tests (Failed: 1645/138443)
Ubuntu navi3x, SrdcvkUbuntu navi2x, Srdcvk |
retest this please |
Some comments, apart from that it looks good. Are the Jenkins / CTS failures related? |
The resource mapping seems to be messed up when using CreateReadBuiltInInput out of builder replayer.
Yes, hopefully my latest commit fixes them. |
Test summary for commit b8614d2CTS tests (Failed: 0/138378)
Ubuntu navi3x, SrdcvkUbuntu navi2x, Srdcvk |
@@ -342,6 +342,10 @@ Options PipelineContext::computePipelineOptions() const { | |||
options.enablePrimGeneratedQuery = getPipelineOptions()->enablePrimGeneratedQuery; | |||
options.enableFragColor = getPipelineOptions()->enableFragColor; | |||
|
|||
options.rtBoxSortHeuristicMode = m_rtState.boxSortHeuristicMode; | |||
options.rtStaticPipelineFlags = m_rtState.staticPipelineFlags; | |||
options.rtTriCompressMode = m_rtState.triCompressMode; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Do you need to default-initialize useGpurt here as well?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The = {}
initialization at the top should take care of that, right?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes. My question is based on the fact that we initialize other members with a constant value here as well, so it is merely a question about coding style.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think we don't need to because it is always false (unknown) at here.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Following yesterday's discussion, luckily @mbrkusanin is going to work on moving the descriptor load lowering out of the builder replayer. Once that's done, it should be possible to remove this workaround and move LowerGpuRt to after the BuilderReplayer and in fact after LowerRaytracingPipeline, where it should be in the long run.
Maybe you can double-check Thomas' question, but apart from that this LGTM.
@@ -342,6 +342,10 @@ Options PipelineContext::computePipelineOptions() const { | |||
options.enablePrimGeneratedQuery = getPipelineOptions()->enablePrimGeneratedQuery; | |||
options.enableFragColor = getPipelineOptions()->enableFragColor; | |||
|
|||
options.rtBoxSortHeuristicMode = m_rtState.boxSortHeuristicMode; | |||
options.rtStaticPipelineFlags = m_rtState.staticPipelineFlags; | |||
options.rtTriCompressMode = m_rtState.triCompressMode; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The = {}
initialization at the top should take care of that, right?
With some tweak to adapt the change.
With some tweak to adapt the change.