[PATCH] D45883: AMDGPU/GlobalISel: Implement select() for 32-bit G_FPTOUI

Matt Arsenault via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Wed Apr 25 13:17:41 PDT 2018


arsenm added inline comments.


================
Comment at: test/CodeGen/AMDGPU/GlobalISel/inst-select-fptoui.mir:20-26
+    ; fptoui s
+    ; GCN: V_CVT_U32_F32_e64
+    %3:vgpr(s32) = G_FPTOUI %0
+
+    ; fptoui v
+    ; GCN: V_CVT_U32_F32_e64
+    %4:vgpr(s32) = G_FPTOUI %1
----------------
tstellar wrote:
> nhaehnle wrote:
> > I personally find the check lines easy to miss when they're indented like that and would prefer them to start at column 0. Admittedly I haven't checked the style elsewhere in LLVM.
> YAML syntax does not allow this unfortunately.
I've just been using update_mir_test_checks for all of these global isel tests


Repository:
  rL LLVM

https://reviews.llvm.org/D45883





More information about the llvm-commits mailing list