ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-04 21:51:28 +00:00
Code Issues Packages Projects Releases Wiki Activity

Revert "Optimize fmha fwd decode & prefill for gfx950 (#2641)" (#2670)

Browse Source 
This reverts commit b7322a521a.
This commit is contained in:
asleepzzz
2025-08-12 20:27:10 +08:00
committed by GitHub
parent b7322a521a
commit 5b39de4bb6
31 changed files with 639 additions and 3545 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
example/ck_tile/01_fmha/fmha_fwd.cpp
View File

@@ -1135,7 +1135,7 @@ bool run(const ck_tile::ArgParser& arg_parser)
std::cout << std::fixed << ", " << std::setprecision(3) << ave_time << " ms, "
<< std::setprecision(2) << tflops << " TFlops, " << std::setprecision(2) << gb_per_sec
<< " GB/s" << std::flush << std::endl;
<< " GB/s" << std::flush;
if(do_validation == 0)
{