[PATCH] D93674: [RISCV] Define vmerge/vfmerge intrinsics.

[Craig Topper via Phabricator via llvm-commits]

craig.topper added inline comments.


================
Comment at: llvm/test/CodeGen/RISCV/rvv/vfmerge-rv32.ll:233
+; CHECK:       vsetvli {{.*}}, {{a[0-9]+}}, e16,mf4
+; CHECK:       vfmerge.vfm {{v[0-9]+}}, {{v[0-9]+}}, {{(a|ft)[0-9]+}}, v0
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmerge.nxv1f16.f16(
----------------
why would the scalar argument be an 'a' register? Shouldn't it always be ft or fa?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D93674/new/

https://reviews.llvm.org/D93674