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

Fix configuration heuristic. #164

Merged
merged 1 commit into from
Oct 27, 2023
Merged

Fix configuration heuristic. #164

merged 1 commit into from
Oct 27, 2023

Conversation

maleadt
Copy link
Member

@maleadt maleadt commented Oct 27, 2023

No description provided.

@maleadt
Copy link
Member Author

maleadt commented Oct 27, 2023

Benchmark results for commit acb74db (comparing to e78e1e7):

test master PR Δmin
FPU GEMM Float64*Float64=Float64 (128×128) · (128×128) (NN) OP (8, 16, 2), base shape (4, 8, 1) 147.0 μs ± 2.56% (130.0 … 149.0 μs) 148.0 μs ± 0.176% (147.0 … 149.0 μs) +13.4% ❌
FPU GEMM Float64*Float64=Float64 (128×128) · (128×128) (TN) OP (8, 16, 2), base shape (4, 8, 1) 148.0 μs ± 0.286% (147.0 … 149.0 μs) 130.0 μs ± 0.275% (129.0 … 131.0 μs) -11.9% ✅
WMMA Complex GEMM Float16*Float16=Float32 (128×128) · (128×128) (NT) OP (16, 16, 16) 8.42 μs ± 1.88% (8.11 … 8.82 μs) 9.59 μs ± 1.7% (9.3 … 9.78 μs) +14.7% ❌
FPU GEMM Float32*Float32=Float64 (128×128) · (128×128) (NN) OP (8, 16, 2), base shape (4, 8, 1) 174.0 μs ± 3.21% (167.0 … 181.0 μs) 173.0 μs ± 5.95% (157.0 … 181.0 μs) -6.0% ✅
WMMA GEMM Float16*Float16=Float16 (2048×2048) · (2048×2048) (NT) OP (16, 16, 16) 295.0 μs ± 0.66% (290.0 … 302.0 μs) 293.0 μs ± 2.82% (257.0 … 305.0 μs) -11.3% ✅

@codecov
Copy link

codecov bot commented Oct 27, 2023

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (e78e1e7) 33.29% compared to head (acb74db) 33.29%.

Additional details and impacted files
@@           Coverage Diff           @@
##           master     #164   +/-   ##
=======================================
  Coverage   33.29%   33.29%           
=======================================
  Files          11       11           
  Lines         907      907           
=======================================
  Hits          302      302           
  Misses        605      605           
Files Coverage Δ
src/config.jl 88.40% <ø> (ø)

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

@maleadt maleadt merged commit 93ef046 into master Oct 27, 2023
1 check passed
@maleadt maleadt deleted the tb/config_fix branch October 27, 2023 14:19
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.

1 participant