search

dillonhuff / clockwork

A polyhedral compiler for hardware accelerators
55 stars 11 forks source link

Merge the recent Aha branch into master #166

Closed joyliu37 closed 3 years ago

joyliu37 commented 3 years ago
  1. Refactor the wide fetch memory mapping
    • Rewrite the memory optimization pass(shift register optimization, banking, bank merging) for wide fetch memory tile.
    • Refactor the ubuffer generation function in coreIR codegen.
    • Rewrite the vectorization pass to handle sliding window vectorization.
  2. Adding ubuffer unit test flow.
  3. Polyhedral Scheduler update
    • Add the schedule padding logic in DNN schedule for wide fetch memory.
    • Relax the delay for upsample loop in stencil pipeline.

dhuff-test and cgra-flow

joyliu37 commented 3 years ago

@dillonhuff Can we merge this?

dillonhuff commented 3 years ago

Sure.

On Thursday, June 3, 2021, Qiaoyi(Joey) Liu @.***> wrote:

@dillonhuff https://github.com/dillonhuff Can we merge this?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dillonhuff/clockwork/pull/166#issuecomment-854226640, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAZ3BMAG32CF6GRMN6PCOUTTQ762FANCNFSM45V35QCQ .