Skip to content

[executorch][native-ggml] Compute cuBLAS GEMMs in f32 in the CUDA sample tests #24538

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.

Sign up for GitHub

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

Open
pytorchbot wants to merge 1 commit into
base: gh/digantdesai/189/orig
Choose a base branch
from
Open

[executorch][native-ggml] Compute cuBLAS GEMMs in f32 in the CUDA sample tests #24538

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 6 additions & 1 deletion backends/native/test/runtime/ggml/BUCK
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,6 +45,11 @@ _VARIANTS = [
# Buck does not put the CUDA driver stub's library on the loader path.
# @oss-disable[end= ]: "LD_PRELOAD": "$(location //tools/lazify:lazy_stub_binder[shared])",
},
# ggml-CUDA computes f16 BLAS GEMMs in f16 and f32 ones in TF32 by default.
"sample_env": {
"GGML_CUDA_CUBLAS_COMPUTE_TYPE": "f32",
"NVIDIA_TF32_OVERRIDE": "0",
},
},
{
"suffix": "_hip",
Expand Down Expand Up @@ -101,7 +106,7 @@ _VARIANTS = [
"--packages=$(location //executorch/backends/native/test/samples:op_samples)",
"--device=" + v["device"],
] + flags,
env = v.get("env", {}),
env = v.get("env", {}) | v.get("sample_env", {}),
remote_execution = v.get("remote_execution"),
# TODO: run in CI once they pass.
labels = ci.labels(ci.map(ci.skip_test())),
Expand Down
Loading