[PATCH] D48346: [X86] Rewrite the add/mul/or/and reduction intrinsics to make better use of other intrinsics and remove undef shuffle indices.

[Simon Pilgrim via Phabricator via cfe-commits]

RKSimon added inline comments.


================
Comment at: test/CodeGen/avx512-reduceIntrin.c:1
 // RUN: %clang_cc1 -ffreestanding %s -O2 -triple=x86_64-apple-darwin -target-cpu skylake-avx512 -emit-llvm -o - -Wall -Werror | FileCheck %s
 
----------------
We're using -O2 here but most codegen tests use -O0 - is that OK? It makes much more recognisable codegen, but I don't know what people expect from these intrinsics without optimization.


Repository:
  rC Clang

https://reviews.llvm.org/D48346