[PATCH] D67596: AMDGPU/GlobalISel: Allow selection of scalar min/max

[Alexander via Phabricator via llvm-commits]

alex-t added inline comments.


================
Comment at: lib/Target/AMDGPU/SOPInstructions.td:422
 def S_MIN_I32 : SOP2_32 <"s_min_i32",
-  [(set i32:$sdst, (UniformBinFrag<smin> i32:$src0, i32:$src1))]
 >;
----------------
As you remove the "!isDiverent" predicate, I'd like to see the test that ensures that V_MIN/MAX etc are not selected for uniform smin/smax/umin/umax IR operations. The test should check not GlobalIsel but current instruction selection case.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D67596/new/

https://reviews.llvm.org/D67596