Skip to content

Conversation

@zahiqbal
Copy link

No description provided.

@zahiqbal zahiqbal requested a review from jayfurmanek February 11, 2025 16:49
i. migrating third_party/mkl to @local_xla//xla/tsl/mkl
ii. removal of cuda.h header
iii. replace build_pip_package rule with wheel in build_rocm_python3 script
fi

if [ -f /usertools/rocm.bazelrc ]; then
# Use the bazelrc files in /usertools if available

Choose a reason for hiding this comment

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

Why are these changes needed?

Copy link
Author

@zahiqbal zahiqbal Feb 12, 2025

Choose a reason for hiding this comment

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

Reverted it. I added this to fix CI build, however it is only use for locally build. I may merge in separate PR if required

@zahiqbal zahiqbal force-pushed the gfx-architecture-support branch from c928bb2 to ed6bf74 Compare February 12, 2025 15:33
absl::flat_hash_set<std::string> FP16SupportedDevices = {
{"gfx906"}, {"gfx908"}, {"gfx90a"}, {"gfx910"}, {"gfx940"}, {"gfx941"},
{"gfx942"}, {"gfx1010"}, {"gfx1012"}, {"gfx1030"},
{"gfx942"}, {"gfx950"},

Choose a reason for hiding this comment

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

Is there any gfx that is not listed here at this point? Should we just return true?

if (token == "sramecc+") {
return "+sramecc";
} else if (token == "sramecc-") {
if (gfx == "gfx90a" || gfx == "gfx940" || gfx == "gfx941" ||

Choose a reason for hiding this comment

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

I think we should remove this? This was a workaround for one specific machine. Albeit we might just do it unconditionally.

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.

6 participants