Add gfx950 FMHA backward kernel as the production dqdkdv path

b448d6c

Select commit

Loading

Failed to load commit list.

Uh oh!

There was an error while loading. Please reload this page.

Meta CodeSync

Import Status

Meta Import Checks

Meta CLA

Meta CLA Check

MSLK Lint on: pull_request 1

run-lint (3.14)

MSLK ROCm CI on: pull_request 10

generate-build-matrix / generate_ci_matrix

build / build_artifact (3.14, x86, linux.24xlarge.memory, default, gcc, 7.1)

build / build_artifact (3.14, x86, linux.24xlarge.memory, default, gcc, 7.2)

build / build_artifact (3.14, x86, linux.24xlarge.memory, default, clang, 7.1)

build / build_artifact (3.14, x86, linux.24xlarge.memory, default, clang, 7.2)

generate-test-matrix / generate_ci_matrix

test / test_and_publish_artifact (3.14, x86, amd-mi350-runner, default, gcc, 7.1)

test / test_and_publish_artifact (3.14, x86, amd-mi350-runner, default, gcc, 7.2)

test / test_and_publish_artifact (3.14, x86, amd-mi350-runner, default, clang, 7.1)

test / test_and_publish_artifact (3.14, x86, amd-mi350-runner, default, clang, 7.2)

MSLK CUDA CI on: pull_request 22

generate-build-matrix / generate_ci_matrix

build / build_artifact (3.10, x86, linux.24xlarge.memory, default, gcc, 13.0.2)

build / build_artifact (3.10, x86, linux.24xlarge.memory, default, clang, 13.0.2)

build / build_artifact (3.11, x86, linux.24xlarge.memory, default, gcc, 13.0.2)

build / build_artifact (3.11, x86, linux.24xlarge.memory, default, clang, 13.0.2)

build / build_artifact (3.12, x86, linux.24xlarge.memory, default, gcc, 13.0.2)

build / build_artifact (3.12, x86, linux.24xlarge.memory, default, clang, 13.0.2)

build / build_artifact (3.13, x86, linux.24xlarge.memory, default, gcc, 13.0.2)

build / build_artifact (3.13, x86, linux.24xlarge.memory, default, clang, 13.0.2)

build / build_artifact (3.14, x86, linux.24xlarge.memory, default, gcc, 13.0.2)

build / build_artifact (3.14, x86, linux.24xlarge.memory, default, clang, 13.0.2)

generate-test-matrix / generate_ci_matrix

test / test_and_publish_artifact (3.10, x86, linux.g5.4xlarge.nvidia.gpu, default, gcc, 13.0.2)

test / test_and_publish_artifact (3.10, x86, linux.g5.4xlarge.nvidia.gpu, default, clang, 13.0.2)

test / test_and_publish_artifact (3.11, x86, linux.g5.4xlarge.nvidia.gpu, default, gcc, 13.0.2)

test / test_and_publish_artifact (3.11, x86, linux.g5.4xlarge.nvidia.gpu, default, clang, 13.0.2)

test / test_and_publish_artifact (3.12, x86, linux.g5.4xlarge.nvidia.gpu, default, gcc, 13.0.2)

test / test_and_publish_artifact (3.12, x86, linux.g5.4xlarge.nvidia.gpu, default, clang, 13.0.2)

test / test_and_publish_artifact (3.13, x86, linux.g5.4xlarge.nvidia.gpu, default, gcc, 13.0.2)

test / test_and_publish_artifact (3.13, x86, linux.g5.4xlarge.nvidia.gpu, default, clang, 13.0.2)

test / test_and_publish_artifact (3.14, x86, linux.g5.4xlarge.nvidia.gpu, default, gcc, 13.0.2)

test / test_and_publish_artifact (3.14, x86, linux.g5.4xlarge.nvidia.gpu, default, clang, 13.0.2)

Build MSLK Windows Wheels on: pull_request 7

generate-matrix / generate

filter-matrix

meta-pytorch/MSLK / build-wheel-py3_10-cuda13_0

meta-pytorch/MSLK / build-wheel-py3_10-cuda13_2

meta-pytorch/MSLK / upload / upload-wheel-py3_10-cuda13_0

meta-pytorch/MSLK / upload / upload-wheel-py3_10-cuda13_2

Build MSLK Aarch64 Linux Wheels on: pull_request 7

generate-matrix / generate

filter-matrix

meta-pytorch/MSLK / build-wheel-py3_10-cuda-aarch6413_0-aarch64

meta-pytorch/MSLK / build-wheel-py3_10-cuda-aarch6413_2-aarch64

meta-pytorch/MSLK / upload / upload-wheel-py3_10-cuda-aarch6413_0-aarch64

meta-pytorch/MSLK / upload / upload-wheel-py3_10-cuda-aarch6413_2-aarch64

Build MSLK x86 Linux Wheels on: pull_request 11

generate-matrix / generate

filter-matrix

meta-pytorch/MSLK / build-manywheel-py3_10-cuda13_0

meta-pytorch/MSLK / build-manywheel-py3_10-cuda13_2

meta-pytorch/MSLK / build-manywheel-py3_10-rocm7_1

meta-pytorch/MSLK / build-manywheel-py3_10-rocm7_2

meta-pytorch/MSLK / upload / upload-manywheel-py3_10-cuda13_0

meta-pytorch/MSLK / upload / upload-manywheel-py3_10-cuda13_2

meta-pytorch/MSLK / upload / upload-manywheel-py3_10-rocm7_1

meta-pytorch/MSLK / upload / upload-manywheel-py3_10-rocm7_2

Facebook GitHub Tools

Facebook Internal - Linter

Facebook Internal - Builds & Tests

Meta CLA / Meta CLA Check succeeded Aug 4, 2026 in 0s

Contributor License Agreement is valid!

View more details on Meta CLA

Loading

Re-running checks...