[llvm] [NFC][AMDGPU] Precommit test for byte lane extension after vector split (PR #228423)

Shilei Tian via llvm-commits llvm-commits at lists.llvm.org
Sat Oct 10 21:15:11 PDT 2026


================
@@ -0,0 +1,745 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
+; RUN: llc -mtriple=amdgpu9.00 < %s | FileCheck -check-prefixes=GFX9 %s
+; RUN: llc -mtriple=amdgpu10.30 < %s | FileCheck -check-prefixes=GFX10 %s
+; RUN: llc -mtriple=amdgpu11.00 -mattr=+real-true16 < %s | FileCheck -check-prefixes=GFX11-TRUE16 %s
+; RUN: llc -mtriple=amdgpu11.00 -mattr=-real-true16 < %s | FileCheck -check-prefixes=GFX11-FAKE16 %s
+; RUN: llc -mtriple=amdgpu12.50 < %s | FileCheck -check-prefixes=GFX1250 %s
----------------
shiltian wrote:

Do we need fake16 variant for 1250 as well?

https://github.com/llvm/llvm-project/pull/228423


More information about the llvm-commits mailing list