cutlass icon indicating copy to clipboard operation
cutlass copied to clipboard

[FEA] Specify L2 cache eviction in TMA copy

Open tridao opened this issue 5 months ago • 3 comments

Which component requires the feature?

CuTe DSL

Feature Request

I'd love to be able to control L2 cache eviction when doing TMA load and TMA store (e.g. evict_first, evict_last)

Additional context This is important for some attention kernels, as we used it in FA3, e.g. here: https://github.com/Dao-AILab/flash-attention/blob/413d07e9deef1e3c793c7de59d7146b43ae4d558/hopper/mainloop_fwd_sm90_tma_gmma_ws.hpp#L753

tridao avatar Jul 31 '25 14:07 tridao

thanks for reporting this issue, we will add this feature asap.

vickiw973 avatar Aug 01 '25 09:08 vickiw973

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] avatar Aug 31 '25 10:08 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.

github-actions[bot] avatar Nov 29 '25 10:11 github-actions[bot]