[PATCH] D140846: [NVPTX] Fix NVPTX lowering of frem when denominator is infinite.

[Stephan Herhut via Phabricator via llvm-commits]

herhut added a subscriber: bkramer.
herhut added a comment.

This looks reasonable to me but @tra or @bkramer would be better reviewers.



================
Comment at: llvm/lib/Target/NVPTX/NVPTXInstrInfo.td:680
+
+def TESTINF32r : NVPTXInst<(outs Int1Regs:$p), (ins Float32Regs:$a),
+                           "testp.infinite.f32 \t$p, $a;",
----------------
Maybe give these more canonical names? Something like 'TESTINF_f32r`. The `f` is redundant (there is no inf for int) but is more in line with how things are named where the suffix after _ is the type.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D140846/new/

https://reviews.llvm.org/D140846