Move output tensor allocation out of benchmark function for GEMM #2328
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Anatoly Myachev <[email protected]>
| @@ -256,7 +252,13 @@ def benchmark(B, M, N, K, provider): | |||
| _, min_ms, max_ms, mean_ms, cv = benchmark_suit.do_bench(lambda: torch.matmul(a, b), warmup=10, rep=10, | |||
| quantiles=quantiles, fast_flush=False) | |||
| elif provider == 'triton': | |||
| triton_fn = lambda: matmul(a, b) | |||
| if len(a.shape) == 3 and len(b.shape) == 3: | |||
There was a problem hiding this comment.
how about assert len(a.shape) == len(b.shape)?
There was a problem hiding this comment.
how about assert len(a.shape) == len(b.shape)?
@whitneywhtsang could you elaborate a bit?
how about assert len(a.shape) == len(b.shape)?
Should I add this assert before if?
There was a problem hiding this comment.
something like #2328 (comment)
Co-authored-by: Whitney Tsang <[email protected]>
Signed-off-by: Anatoly Myachev <[email protected]>
whitneywhtsang
approved these changes
Sep 24, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Geomean was incorrectly calculated for GEMM (in #2298), after recalculating I saw a deterioration in the ratio, which may be due to allocations.
Let's see: https://github.com/intel/intel-xpu-backend-for-triton/actions/runs/11019414473
The Triton GEMM adv geomean increased by ~1TFlops.