-
Notifications
You must be signed in to change notification settings - Fork 43
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
Update GEMM XeTLA kernel of triton benchmarks #1597
Conversation
Based on the comment #1564 (comment), can you help change the folder name from "xetla_benchmark" to like "triton_kernels_benchmark". Because XeTLA performance is just for reference, we are actually measure Triton perf, current name is a bit confusing. |
737b0bb
to
fad7617
Compare
90ed5c3
to
75b32f1
Compare
f68862f
to
44f3fef
Compare
44f3fef
to
ab6dac4
Compare
ab6dac4
to
e9dbbe8
Compare
UR_L0_IN_ORDER_BARRIER_BY_SIGNAL=0 \ | ||
IGC_DisableLoopUnroll=1 \ | ||
NEO_CACHE_PERSISTENT=0 \ | ||
SYCL_PROGRAM_COMPILE_OPTIONS=" -vc-codegen -vc-disable-indvars-opt -doubleGRF -Xfinalizer ' -printregusage -enableBCR -DPASTokenReduction ' " \ |
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.
Here, it is necessary to specifically introduce SYCL_PROGRAM_COMPILE_OPTIONS
to make XeTLA compile flag taking effect
69994e3
to
5af1f42
Compare
5af1f42
to
b9de4c7
Compare
https://github.com/intel/intel-xpu-backend-for-triton/actions/runs/10055482873/job/27792231590