Skip to content
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

Buffer Device Address in HLSL #690

Merged

Conversation

devshgraphicsprogramming
Copy link
Member

Description

Testing

TODO list:

nipunG314 added 3 commits May 31, 2024 20:16
-> Use implicit key_t instead of explicit Key template parameter
-> Pass workGroupIndex specifically at Counter creation instead of with
uniform data
-> Remove global memory toroidal access from the histogram shader
sum = inclusive_scan(val, sdata);
histogram.atomicAdd(vid, sum);
if (vid < KeyBucketCount) {
sum = inclusive_scan(sdata.get(vid), sdata);
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

here's the thing, inclusive_scan uses workgroup barriers, so it needs to be outside of non-uniform control flow

the sdata.get() you can do speculatively (OOB read unlikely to mess up your GPU), its only the atomic add (a write) that needs to be protected against OOB

@devshgraphicsprogramming devshgraphicsprogramming merged commit 65d5180 into Devsh-Graphics-Programming:master Jun 1, 2024
1 check failed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants