Skip to content

Actions: ROCm/FlyDSL

Actions

Fly DSL test

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
flydsl.yaml
2,500+ workflow runs
2,500+ workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
[Bugfix] Detach requires_grad tensors before DLPack export Fly DSL test #4281: Pull request #1099 opened by xytpai
In progress xyt/detach
xyt/detach
In progress
[Flydsl] Qwen-Image conv3d 3x3 opt Fly DSL test #4280: Pull request #1069 synchronize by huizzhan
In progress feature/qwenimage-vae-conv-tests
feature/qwenimage-vae-conv-tests
In progress
FlashAttention FP8: Added support for asymmetric head dimensions, variable sequence lengths (varlen), and KV splitting Fly DSL test #4279: Pull request #1097 synchronize by binding7012
1h 28m 2s binding_fa_more_feature_support
binding_fa_more_feature_support
1h 28m 2s
[Flydsl] Qwen-Image conv3d 3x3 opt Fly DSL test #4278: Pull request #1069 synchronize by huizzhan
42m 17s feature/qwenimage-vae-conv-tests
feature/qwenimage-vae-conv-tests
42m 17s
FlashAttention FP8: Added support for asymmetric head dimensions, variable sequence lengths (varlen), and KV splitting Fly DSL test #4277: Pull request #1097 synchronize by binding7012
40m 47s binding_fa_more_feature_support
binding_fa_more_feature_support
40m 47s
[Flydsl] Qwen-Image conv3d 3x3 opt Fly DSL test #4276: Pull request #1069 synchronize by huizzhan
1h 2m 22s feature/qwenimage-vae-conv-tests
feature/qwenimage-vae-conv-tests
1h 2m 22s
[Kernel][FA] Support paged FP8 Flash attention with asymmetric K/V head dim on gfx950 Fly DSL test #4275: Pull request #1066 synchronize by sammysun0711
1h 34m 32s sammysun0711:upstream/fa-paged-fp8-asymmetric-v
sammysun0711:upstream/fa-paged-fp8-asymmetric-v
1h 34m 32s
refactor: migrate to extension api Fly DSL test #4274: Pull request #1098 opened by ppppqp
Action required ppppqp:feat/coop-reduce-scan
ppppqp:feat/coop-reduce-scan
Action required
[Kernel][FA] Support paged FP8 Flash attention with asymmetric K/V head dim on gfx950 Fly DSL test #4273: Pull request #1066 synchronize by sammysun0711
2h 14m 16s sammysun0711:upstream/fa-paged-fp8-asymmetric-v
sammysun0711:upstream/fa-paged-fp8-asymmetric-v
2h 14m 16s
[test only] Fly DSL test #4272: Pull request #1095 synchronize by coderfeli
2m 12s update_llvm_0.3.2
update_llvm_0.3.2
2m 12s
[CI] Run aiter CSV MoE and HGEMM in the wheel/PyPI test job Fly DSL test #4271: Pull request #1094 synchronize by coderfeli
1h 36m 42s ci/pypi-moe-csv-hgemm
ci/pypi-moe-csv-hgemm
1h 36m 42s
FlashAttention FP8: Added support for asymmetric head dimensions, variable sequence lengths (varlen), and KV splitting Fly DSL test #4270: Pull request #1097 synchronize by binding7012
1h 24m 16s binding_fa_more_feature_support
binding_fa_more_feature_support
1h 24m 16s
FlashAttention FP8: Added support for asymmetric head dimensions, variable sequence lengths (varlen), and KV splitting Fly DSL test #4269: Pull request #1097 opened by binding7012
1m 33s binding_fa_more_feature_support
binding_fa_more_feature_support
1m 33s
[CI] Run aiter CSV MoE and HGEMM in the wheel/PyPI test job Fly DSL test #4268: Pull request #1094 synchronize by coderfeli
1h 38m 11s ci/pypi-moe-csv-hgemm
ci/pypi-moe-csv-hgemm
1h 38m 11s
[test only] Fly DSL test #4267: Pull request #1095 synchronize by coderfeli
2m 59s update_llvm_0.3.2
update_llvm_0.3.2
2m 59s
[CI] Run aiter CSV MoE and HGEMM in the wheel/PyPI test job Fly DSL test #4266: Pull request #1094 synchronize by coderfeli
1h 42m 20s ci/pypi-moe-csv-hgemm
ci/pypi-moe-csv-hgemm
1h 42m 20s
[test only] Fly DSL test #4265: Pull request #1095 synchronize by coderfeli
2m 14s update_llvm_0.3.2
update_llvm_0.3.2
2m 14s
[test only] Fly DSL test #4264: Pull request #1095 synchronize by coderfeli
2m 14s update_llvm_0.3.2
update_llvm_0.3.2
2m 14s
[CI] Run aiter CSV MoE and HGEMM in the wheel/PyPI test job Fly DSL test #4263: Pull request #1094 synchronize by coderfeli
1h 41m 32s ci/pypi-moe-csv-hgemm
ci/pypi-moe-csv-hgemm
1h 41m 32s
gemm: add fp8 per-tensor grouped GEMM forward (M-grouped/MoE) Fly DSL test #4262: Pull request #887 synchronize by kyle-256
34s dev/kyle/grouped_gemm
dev/kyle/grouped_gemm
34s
[DSL] Add ordered ROCDL memory primitives Fly DSL test #4261: Pull request #1096 synchronize by big-yellow-duck
1h 25m 51s big-yellow-duck:feat/rocdl-ordered-memory-primitives
big-yellow-duck:feat/rocdl-ordered-memory-primitives
1h 25m 51s
[DSL] Add ordered ROCDL memory primitives Fly DSL test #4260: Pull request #1096 opened by big-yellow-duck
Action required big-yellow-duck:feat/rocdl-ordered-memory-primitives
big-yellow-duck:feat/rocdl-ordered-memory-primitives
Action required
[test only] Fly DSL test #4259: Pull request #1095 opened by coderfeli
3m 3s update_llvm_0.3.2
update_llvm_0.3.2
3m 3s
[Kernel] Add dilation, padding_mode and groups to convolution kernel.… Fly DSL test #4258: Commit b81e99d pushed by coderfeli
1h 46m 20s main
main
1h 46m 20s
[FlyToROCDL] Bump LLVM to 941a04e6 and adapt gpu.launch_func lowering Fly DSL test #4257: Pull request #1051 synchronize by coderfeli
1h 58m 11s phil/bump-llvm-941a04e6
phil/bump-llvm-941a04e6
1h 58m 11s