Skip to content

Add subgroup matrix multiplication#80

Open
junjihashimoto wants to merge 2 commits intoAnswerDotAI:devfrom
junjihashimoto:feature/matmul
Open

Add subgroup matrix multiplication#80
junjihashimoto wants to merge 2 commits intoAnswerDotAI:devfrom
junjihashimoto:feature/matmul

Conversation

@junjihashimoto
Copy link
Copy Markdown
Collaborator

@junjihashimoto junjihashimoto commented Sep 5, 2025 •
edited
Loading

Add the shader of subgroup matrix multiplication.

$ sysctl -n machdep.cpu.brand_string
Apple M4 Max

$ for i in 11 12 13 ; do MATMUL_VERSION=$i ./build/matmul  | grep 'Dispatching\|FLOPS' ; done
# Without subgroupMatrix (f16) 
[info] Dispatching Kernel version 11: f16: 2D blocktiling with loop unrolling, vectorization and transpose, 30 iterations ...
25.8 milliseconds / dispatch ~ 10640.35 GFLOPS
# With subgroupMatrix (f16) 
[info] Dispatching Kernel version 12: f16: Subgroup matrix multiply with transpose, 30 iterations ...
20.7 milliseconds / dispatch ~ 13299.45 GFLOPS
# With subgroupMatrix (f32) 
[info] Dispatching Kernel version 13: f32: Subgroup matrix multiply with transpose (default), 30 iterations ...
24.6 milliseconds / dispatch ~ 11185.63 GFLOPS
@junjihashimoto
Copy link
Copy Markdown
Collaborator Author

junjihashimoto commented Sep 5, 2025 •
edited
Loading

The main branch does not seem to output any shader compilation errors.
@junjihashimoto junjihashimoto changed the base branch from main to dev September 22, 2025 09:33
junjihashimoto
junjihashimoto commented Oct 13, 2025
View reviewed changes
cmake/dawn.cmake
# Ensure source present on required commit (idempotent remote setup)
if(NOT DEFINED DAWN_COMMIT OR DAWN_COMMIT STREQUAL "")
set(DAWN_COMMIT "e1d6e12337080cf9f6d8726209e86df449bc6e9a" CACHE STRING "Dawn commit to checkout" FORCE)
set(DAWN_COMMIT "3f79f3aefe0b0a498002564fcfb13eb21ab6c047" CACHE STRING "Dawn commit to checkout" FORCE)
Copy link
Copy Markdown
Collaborator Author

@junjihashimoto junjihashimoto Oct 13, 2025

Choose a reason for hiding this comment

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

google/dawn@d7d27a6
Required to set subgroupsize to 32 on MacOS.
@junjihashimoto junjihashimoto marked this pull request as ready for review October 13, 2025 06:14
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

1 participant

@junjihashimoto