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.

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

Implicit GEMM optimizations/bug fixes #2499

Merged
merged 16 commits into from
Nov 18, 2024
Merged

Implicit GEMM optimizations/bug fixes #2499

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
16 commits
Select commit Hold shift + click to select a range
27b93a5
Optimize implicit GEMM
wingertge Nov 11, 2024
3bfdfd6
Ensure weight isn't vectorized if it's loaded directly
wingertge Nov 11, 2024
7535046
Merge branch 'main' into opt/implicit_gemm
wingertge Nov 11, 2024
95822d0
Implement tf32 for implicit GEMM
wingertge Nov 11, 2024
d472a17
Fixes
wingertge Nov 14, 2024
efe36e1
Merge branch 'main' into opt/implicit_gemm
wingertge Nov 14, 2024
7543f7b
Make reduce checked since we're still getting segfaults
wingertge Nov 16, 2024
b17713e
Merge branch 'main' into opt/implicit_gemm
wingertge Nov 16, 2024
07bd12f
Make bicubic interp checked
wingertge Nov 16, 2024
ea9e872
Undo direct weight loader because it was backfiring
wingertge Nov 16, 2024
449e464
Undo version change
wingertge Nov 16, 2024
a95b692
Use git version of cubecl
wingertge Nov 16, 2024
6577812
Update cubecl
wingertge Nov 17, 2024
ca64d18
Use select to ensure correctness in bilinear interpolate
wingertge Nov 17, 2024
83adb2f
Disable reduce_dim_subcube if warp size isn't known, to prevent poten…
wingertge Nov 18, 2024
a3e6095
Merge branch 'main' into opt/implicit_gemm
wingertge Nov 18, 2024
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
Prev Previous commit
Next Next commit
Undo version change
  • Loading branch information
@wingertge
wingertge committed Nov 16, 2024
commit 449e4647b675426decc3f3bad9bbac677e9b0663
Loading
Loading