Skip to content

Actions: pytorch/FBGEMM

Actions

FBGEMM_GPU_TORCHREC-CPU CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
fbgemm_gpu_torchrec_ci_cpu.yml
733 workflow runs
733 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Save some binary size FBGEMM_GPU_TORCHREC-CPU CI #596: Pull request #4900 opened by cthi
30m 20s cthi:export-D82842915
cthi:export-D82842915
30m 20s
Small changes to improve blackwell_fmha_test.py FBGEMM_GPU_TORCHREC-CPU CI #595: Pull request #4896 synchronize by henrylhtsang
39m 54s henrylhtsang:export-D81992869
henrylhtsang:export-D81992869
39m 54s
Add softmax_scale to blackwell cutlass fmha (#4897) FBGEMM_GPU_TORCHREC-CPU CI #594: Commit f82e3b2 pushed by facebook-github-bot
32m 36s main
main
32m 36s
Backward optimizations + rocm and gfx950 build support FBGEMM_GPU_TORCHREC-CPU CI #593: Pull request #4729 synchronize by shbiswas834
1h 4m 59s ROCm:shbiswas/uptream_develop_0815
ROCm:shbiswas/uptream_develop_0815
1h 4m 59s
Add ability to pad the rowwise quantized tensors FBGEMM_GPU_TORCHREC-CPU CI #592: Pull request #4877 synchronize by yyetim
30m 50s yyetim:export-D82486197
yyetim:export-D82486197
30m 50s
Restrict to fp8, and patch 4.2.0 release changes (#4898) FBGEMM_GPU_TORCHREC-CPU CI #591: Commit 1e80f1a pushed by facebook-github-bot
30m 22s main
main
30m 22s
FBGEMM_GPU_TORCHREC-CPU CI FBGEMM_GPU_TORCHREC-CPU CI #590: Scheduled
30m 30s main
main
30m 30s
Migrate GenAI kv cache kernel arguments to use PTA_B (#4899) FBGEMM_GPU_TORCHREC-CPU CI #589: Commit 9b66962 pushed by facebook-github-bot
31m 13s main
main
31m 13s
add blog link (#4888) FBGEMM_GPU_TORCHREC-CPU CI #588: Commit e6d8742 pushed by facebook-github-bot
36m 15s main
main
36m 15s
Migrate GenAI kv cache kernel arguments to use PTA_B FBGEMM_GPU_TORCHREC-CPU CI #587: Pull request #4899 opened by q10
28m 23s q10:export-D82775043
q10:export-D82775043
28m 23s
Migrate GenAI kv cache kernels to FBGEMM_LAUNCH_KERNEL, pt 4 (#4895) FBGEMM_GPU_TORCHREC-CPU CI #586: Commit 944647a pushed by facebook-github-bot
21m 33s main
main
21m 33s
Restrict to fp8, and patch 4.2.0 release changes FBGEMM_GPU_TORCHREC-CPU CI #585: Pull request #4898 opened by henrylhtsang
32m 39s henrylhtsang:export-D82792657
henrylhtsang:export-D82792657
32m 39s
Add softmax_scale to blackwell cutlass fmha FBGEMM_GPU_TORCHREC-CPU CI #584: Pull request #4897 opened by henrylhtsang
29m 15s henrylhtsang:export-D82788784
henrylhtsang:export-D82788784
29m 15s
Small changes to improve blackwell_fmha_test.py FBGEMM_GPU_TORCHREC-CPU CI #583: Pull request #4896 synchronize by henrylhtsang
35m 3s henrylhtsang:export-D81992869
henrylhtsang:export-D81992869
35m 3s
Small changes to improve blackwell_fmha_test.py FBGEMM_GPU_TORCHREC-CPU CI #582: Pull request #4896 synchronize by henrylhtsang
18s henrylhtsang:export-D81992869
henrylhtsang:export-D81992869
18s
Small changes to improve blackwell_fmha_test.py FBGEMM_GPU_TORCHREC-CPU CI #581: Pull request #4896 opened by henrylhtsang
17m 30s henrylhtsang:export-D81992869
henrylhtsang:export-D81992869
17m 30s
Migrate GenAI kv cache kernels to FBGEMM_LAUNCH_KERNEL, pt 4 FBGEMM_GPU_TORCHREC-CPU CI #580: Pull request #4895 opened by q10
33m 38s q10:export-D82773258
q10:export-D82773258
33m 38s
Migrate GenAI gqa attn splitk kernels to FBGEMM_LAUNCH_KERNEL, pt 1 FBGEMM_GPU_TORCHREC-CPU CI #579: Pull request #4894 opened by q10
30m 22s q10:export-D81817544
q10:export-D81817544
30m 22s
Back out "Enable CUTLASS grouped GEMM for pretraining wgrad on GB200 … FBGEMM_GPU_TORCHREC-CPU CI #578: Commit 53f9e51 pushed by facebook-github-bot
33m 26s main
main
33m 26s
upgrade cutlass to 4.2.0 for fbcode FBGEMM_GPU_TORCHREC-CPU CI #577: Pull request #4893 opened by henrylhtsang
35m 32s henrylhtsang:export-D82567751
henrylhtsang:export-D82567751
35m 32s
Back out "Enable CUTLASS grouped GEMM for pretraining wgrad on GB200 and H100" FBGEMM_GPU_TORCHREC-CPU CI #576: Pull request #4892 synchronize by jiawenliu64
28m 6s jiawenliu64:export-D82746070
jiawenliu64:export-D82746070
28m 6s
Back out "Enable CUTLASS grouped GEMM for pretraining wgrad on GB200 and H100" FBGEMM_GPU_TORCHREC-CPU CI #575: Pull request #4892 opened by jiawenliu64
10m 57s jiawenliu64:export-D82746070
jiawenliu64:export-D82746070
10m 57s
add blog link FBGEMM_GPU_TORCHREC-CPU CI #574: Pull request #4888 synchronize by sijiac
31m 37s sijiac:export-D82681043
sijiac:export-D82681043
31m 37s
Add ability to pad the rowwise quantized tensors FBGEMM_GPU_TORCHREC-CPU CI #573: Pull request #4877 synchronize by yyetim
30m 19s yyetim:export-D82486197
yyetim:export-D82486197
30m 19s
dequantize_fp8_cache_kernel: Move D=128 device-side-assertion check to host FBGEMM_GPU_TORCHREC-CPU CI #572: Pull request #4869 synchronize by ColinPeppler
31m 28s ColinPeppler:export-D82320518
ColinPeppler:export-D82320518
31m 28s