[PATCH] D37560: [X86] Lower _mm[256|512]_[mask[z]]_avg_epu[8|16] intrinsics to native llvm IR

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added inline comments.


================
Comment at: test/CodeGen/X86/sse2-intrinsics-fast-isel.ll:277
-}
-declare <8 x i16> @llvm.x86.sse2.pavg.w(<8 x i16>, <8 x i16>) nounwind readnone
-
----------------
-fast-isel.ll tests don't get deleted when you remove the ir  intrinsic - you replace it with the equivalent IR that gets generated from the c intrinsic: see the top of the file:

> ; NOTE: This should use IR equivalent to what is generated by clang/test/CodeGen/sse2-builtins.c


https://reviews.llvm.org/D37560