oplavsic 715a589ce3 [FA fwd D=128] Reduce LDS usage in epilogue (#340) ...

* rebase onto improve_fwd_fa

* Fixed a leftover from rebase

* rebase onto improve_fa_fwd

* Reduce tuning space

* Disable bwd with D=128

* Add test for d=128

* Fix an issue with get_best_config when there is only one config

* Added better configs for d=128

* Fix typos

---------

Co-authored-by: Lixun Zhang <lixun.zhang@amd.com>

2023-10-25 12:10:34 -05:00

..

_C

History prior to this date belonged to the now deprecated ISAAC project, and was deleted to save space

2021-07-27 12:38:38 -07:00

common

PyTorch triton branch synchronisation (#354)

2023-10-11 15:30:39 +01:00

compiler

[MFMA] Switch between MFMA types (#352)

2023-10-18 16:57:34 +02:00

interpreter

[FRONTEND] expose tl.max_constancy hint (#1951)

2023-07-17 18:30:25 +00:00

language

[MFMA] Switch between MFMA types (#352)

2023-10-18 16:57:34 +02:00

ops

Merge commit '36fc54b6f28168d3644808bfe299f1ba06a36272' into ifu230908-2

2023-10-02 18:01:04 +00:00

runtime

[FA fwd D=128] Reduce LDS usage in epilogue (#340)

2023-10-25 12:10:34 -05:00

third_party

Merge commit '36fc54b6f28168d3644808bfe299f1ba06a36272' into ifu230908-2

2023-10-02 18:01:04 +00:00

tools

Fix merge conflicts

2023-09-01 04:01:32 +00:00

__init__.py

[TOOLS] improved ahead-of-time compiler (#1805)

2023-06-21 01:02:58 -07:00

testing.py

[FRONTEND] Fix benchmark plotting (#2177)

2023-08-24 20:34:04 -07:00