[PATCH] D88154: Initial support for vectorization using Libmvec (GLIBC vector math library).

[Florian Hahn via Phabricator via llvm-commits]

fhahn added inline comments.


================
Comment at: llvm/test/Transforms/LoopVectorize/X86/libm-vector-calls-finite.ll:82
+
+!1 = distinct !{!1, !2, !3}
+!2 = !{!"llvm.loop.vectorize.width", i32 4}
----------------
fhahn wrote:
> I don't think we need this. You can just pass `-force-vector-width=4` to the command line and avoid the extra metadata duplicated for each test
Apologies, I missed that this was suggested earlier to test different VFs. If we decide to go with the 'essential' check lines approach it might make sense to just invoke opt with different VFs using `force-vector-width`.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D88154/new/

https://reviews.llvm.org/D88154