Skip to content

[mlir][ROCDL] Adds wmma scaled intrinsics for gfx1250 #165915

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
wants to merge 2 commits into
base: main
Choose a base branch
from
+229 −0
Open

[mlir][ROCDL] Adds wmma scaled intrinsics for gfx1250 #165915

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ba086bd
[mlir][ROCDL] adds wmma scaled intrinsics for gfx1250
Muzammiluddin-Syed-ECE Oct 7, 2025
3633a56
adding wmma scaled intrinsics with matrix A size 32x128
Muzammiluddin-Syed-ECE Oct 31, 2025
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
4 changes: 4 additions & 0 deletions mlir/include/mlir/Dialect/LLVMIR/ROCDLOps.td
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -627,6 +627,10 @@ def ROCDL_wmma_f16_16x16x128_fp8_bf8 : ROCDL_Wmma_IntrOp<"wmma.f16.16x16x128.fp8
def ROCDL_wmma_f16_16x16x128_bf8_fp8 : ROCDL_Wmma_IntrOp<"wmma.f16.16x16x128.bf8_fp8", [0]>;
def ROCDL_wmma_f16_16x16x128_bf8_bf8 : ROCDL_Wmma_IntrOp<"wmma.f16.16x16x128.bf8_bf8", [0]>;
def ROCDL_wmma_i32_16x16x64_iu8 : ROCDL_Wmma_IntrOp<"wmma.i32.16x16x64.iu8", [1]>;
def ROCDL_wmma_scale_f32_16x16x128_f8f6f4 : ROCDL_Wmma_IntrOp<"wmma.scale.f32.16x16x128.f8f6f4", [1,3]>;
def ROCDL_wmma_scale16_f32_16x16x128_f8f6f4 : ROCDL_Wmma_IntrOp<"wmma.scale16.f32.16x16x128.f8f6f4", [1,3]>;
def ROCDL_wmma_scale_f32_32x16x128_f4 : ROCDL_Wmma_IntrOp<"wmma.scale.f32.32x16x128.f4", [0,1]>;
def ROCDL_wmma_scale16_f32_32x16x128_f4 : ROCDL_Wmma_IntrOp<"wmma.scale16.f32.32x16x128.f4", [0,1]>;

//===---------------------------------------------------------------------===//
// LDS transpose intrinsics (available in GFX950)
Expand Down
Loading