Clean up sgl kernel (#12413)

Co-authored-by: Byron Hsu <byronhsu1230@gmail.com>
2025-10-31 01:13:34 -07:00
parent 2e48584b62
commit c0652d907b
10 changed files with 514 additions and 379 deletions
@@ -13,6 +13,8 @@ set(FLASHMLA_CUDA_FLAGS
    "--expt-relaxed-constexpr"
    "--expt-extended-lambda"
    "--use_fast_math"
+
+    "-Xcudafe=--diag_suppress=177"   # variable was declared but never referenced
 )

 # The FlashMLA kernels only work on hopper and require CUDA 12.4 or later.