Skip to content

Actions: pytorch/FBGEMM

Actions

FBGEMM_GPU-CUDA CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
fbgemm_gpu_ci_cuda.yml
6,819 workflow runs
6,819 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Performance Optimization: Optimized TileShape Configuration for bf16 and Mixed Formats FBGEMM_GPU-CUDA CI #7453: Pull request #3591 opened by MatrixAssembler
January 20, 2025 22:47 Action required MatrixAssembler:opt/tileshape-bf16-mixed
MatrixAssembler:opt/tileshape-bf16-mixed
January 20, 2025 22:47 Action required
FBGEMM_GPU-CUDA CI FBGEMM_GPU-CUDA CI #7452: Scheduled
January 20, 2025 12:54 1h 22m 21s main
main
January 20, 2025 12:54 1h 22m 21s
FBGEMM_GPU-CUDA CI FBGEMM_GPU-CUDA CI #7451: Scheduled
January 19, 2025 12:52 1h 18m 28s main
main
January 19, 2025 12:52 1h 18m 28s
fix data type of block_bucketize_pos in block_bucketize_sparse_features FBGEMM_GPU-CUDA CI #7450: Pull request #3589 synchronize by TroyGarden
January 18, 2025 19:37 1h 20m 52s TroyGarden:export-D49927771
TroyGarden:export-D49927771
January 18, 2025 19:37 1h 20m 52s
FBGEMM_GPU-CUDA CI FBGEMM_GPU-CUDA CI #7449: Scheduled
January 18, 2025 12:52 1h 19m 42s main
main
January 18, 2025 12:52 1h 19m 42s
fix data type of block_bucketize_pos in block_bucketize_sparse_features FBGEMM_GPU-CUDA CI #7448: Pull request #3589 opened by TroyGarden
January 18, 2025 08:20 1h 18m 13s TroyGarden:export-D49927771
TroyGarden:export-D49927771
January 18, 2025 08:20 1h 18m 13s
Return if no data to allreduce FBGEMM_GPU-CUDA CI #7447: Pull request #3586 synchronize by xw285cornell
January 18, 2025 02:24 1h 18m 3s xw285cornell:export-D68318641
xw285cornell:export-D68318641
January 18, 2025 02:24 1h 18m 3s
FBGEMM_GPU-CUDA CI FBGEMM_GPU-CUDA CI #7446: Manually run by q10
January 18, 2025 01:03 2h 43m 11s bm/test-497bad6
bm/test-497bad6
January 18, 2025 01:03 2h 43m 11s
FBGEMM_GPU-CUDA CI FBGEMM_GPU-CUDA CI #7445: Manually run by q10
January 18, 2025 00:21 1h 19m 27s bm/test-d16e2d8
bm/test-d16e2d8
January 18, 2025 00:21 1h 19m 27s
Back out "Enable fast FP8 GEMM for memory bound" (#3588) FBGEMM_GPU-CUDA CI #7444: Commit 2d025dc pushed by facebook-github-bot
January 17, 2025 23:02 45m 19s main
main
January 17, 2025 23:02 45m 19s
Back out "Enable fast FP8 GEMM for memory bound" FBGEMM_GPU-CUDA CI #7443: Pull request #3588 opened by jiawenliu64
January 17, 2025 21:54 1h 18m 53s jiawenliu64:export-D68351266
jiawenliu64:export-D68351266
January 17, 2025 21:54 1h 18m 53s
Add support for int32_t indices in TBE training (2H/N) FBGEMM_GPU-CUDA CI #7442: Pull request #3539 synchronize by q10
January 17, 2025 21:26 3h 7m 7s q10:export-D67784746
q10:export-D67784746
January 17, 2025 21:26 3h 7m 7s
Add support for int32_t indices in TBE training (2K/N) FBGEMM_GPU-CUDA CI #7441: Pull request #3583 synchronize by q10
January 17, 2025 21:26 2h 46m 53s q10:export-D68296454
q10:export-D68296454
January 17, 2025 21:26 2h 46m 53s
Revert D65620886 FBGEMM_GPU-CUDA CI #7440: Pull request #3582 synchronize by q10
January 17, 2025 21:25 43m 5s q10:export-D68291310
q10:export-D68291310
January 17, 2025 21:25 43m 5s
Back out "Optimzed backward pass for ROCm devices (pt 2)" FBGEMM_GPU-CUDA CI #7439: Pull request #3587 opened by q10
January 17, 2025 19:55 2h 59m 47s q10:export-D68341728
q10:export-D68341728
January 17, 2025 19:55 2h 59m 47s
Profile with kineto and warmup for more accurate benchmarking (#3585) FBGEMM_GPU-CUDA CI #7438: Commit 379db5f pushed by facebook-github-bot
January 17, 2025 18:45 26m 51s main
main
January 17, 2025 18:45 26m 51s
Revert D65620886 FBGEMM_GPU-CUDA CI #7437: Pull request #3582 synchronize by q10
January 17, 2025 18:35 2h 45m 16s q10:export-D68291310
q10:export-D68291310
January 17, 2025 18:35 2h 45m 16s
Return if no data to allreduce FBGEMM_GPU-CUDA CI #7436: Pull request #3586 opened by xw285cornell
January 17, 2025 17:51 2h 42m 33s xw285cornell:export-D68318641
xw285cornell:export-D68318641
January 17, 2025 17:51 2h 42m 33s
Support INT4 Dequant onto GPU for Seq INT TBE look up FBGEMM_GPU-CUDA CI #7435: Pull request #3584 synchronize by faran928
January 17, 2025 17:50 2h 41m 35s faran928:export-D68187234
faran928:export-D68187234
January 17, 2025 17:50 2h 41m 35s
FBGEMM_GPU-CUDA CI FBGEMM_GPU-CUDA CI #7434: Scheduled
January 17, 2025 12:54 2h 45m 51s main
main
January 17, 2025 12:54 2h 45m 51s
Profile with kineto and warmup for more accurate benchmarking (#3580) FBGEMM_GPU-CUDA CI #7433: Pull request #3585 synchronize by q10
January 17, 2025 05:57 2h 40m 7s q10:export-D68292871
q10:export-D68292871
January 17, 2025 05:57 2h 40m 7s
Profile with kineto and warmup for more accurate benchmarking (#3580) FBGEMM_GPU-CUDA CI #7432: Pull request #3585 synchronize by q10
January 17, 2025 04:39 1h 18m 59s q10:export-D68292871
q10:export-D68292871
January 17, 2025 04:39 1h 18m 59s
Fix specailization issue in keyed_jagged_index_select_dim1_forward_cu… FBGEMM_GPU-CUDA CI #7431: Commit 21d1260 pushed by facebook-github-bot
January 17, 2025 03:58 2h 42m 58s main
main
January 17, 2025 03:58 2h 42m 58s
Support INT4 Dequant onto GPU for Seq INT TBE look up FBGEMM_GPU-CUDA CI #7430: Pull request #3584 synchronize by faran928
January 17, 2025 01:50 2h 41m 9s faran928:export-D68187234
faran928:export-D68187234
January 17, 2025 01:50 2h 41m 9s
Profile with kineto and warmup for more accurate benchmarking (#3580) FBGEMM_GPU-CUDA CI #7429: Pull request #3585 opened by q10
January 17, 2025 01:47 2h 41m 54s q10:export-D68292871
q10:export-D68292871
January 17, 2025 01:47 2h 41m 54s