[QST]What does Epilogue do?(Especially in Hopper)

[ziyuhuang123]

trafficstars

I am writing example 48, and I noticed file: https://github.com/NVIDIA/cutlass/blob/06b21349bcf6ddf6a1686a47a137ad1446579db9/include/cutlass/gemm/kernel/sm90_gemm_tma_warpspecialized_cooperative.hpp#L321 I find out with surprise that, epilogue is in producer, but we do not even enter epilogue?! Previously I guess we save result using epilogue, both rearrange it into shared memory and then save into global, but now, storing is done in cousumer...?

[ziyuhuang123]

Maybe because.... we have no enough shared memory to store all results, so out consumer generate a part and we immediately save it into global, so it must be in consumer? (Understandable.... just no need to have a "epilogue" part in producer...)(also, we have 128 threads in producer, but only 32 threads is working, strange...)

[github-actions[bot]]

This issue has been labeled inactive-30d due to no recent activity in the past 30 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed. This issue will be labeled inactive-90d if there is no activity in the next 60 days.

[github-actions[bot]]

This issue has been labeled inactive-90d due to no recent activity in the past 90 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed.