[llvm] [AMDGPU] Improve codegen for uniform f16<-->i32 conversions (PR #176833)

Syadus Sefat via llvm-commits llvm-commits at lists.llvm.org
Wed Jan 21 10:17:54 PST 2026


================
@@ -334,4 +302,5 @@ declare half @llvm.trunc.f16(half)
 declare half @llvm.rint.f16(half)
 ;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
 ; GISEL: {{.*}}
-; SDAG: {{.*}}
+; SDAG-GFX11: {{.*}}
+; SDAG-GFX12: {{.*}}
----------------
mssefat wrote:

Removed them.

https://github.com/llvm/llvm-project/pull/176833


More information about the llvm-commits mailing list