[PATCH] D105263: [X86] AVX512FP16 instructions enabling 1/6

[Pengfei Wang via Phabricator via cfe-commits]

pengfei added inline comments.


================
Comment at: clang/test/CodeGen/X86/avx512fp16-complex.c:1
+// RUN: %clang_cc1 %s -O0 -fno-experimental-new-pass-manager -emit-llvm -triple x86_64-unknown-unknown -target-feature +avx512fp16 -o - | FileCheck %s --check-prefix=X86
+
----------------
craig.topper wrote:
> Can we split _Complex out of this patch? This affects other targets that have _Float16 right? So probably needs a different set of reviewers.
Sure. Split to D105331. Do you know someone who is familiar with or may be interested in it?


================
Comment at: llvm/lib/Target/X86/X86ISelLowering.cpp:23042
          Op0.getSimpleValueType().is512BitVector())) {
-      assert(VT.getVectorNumElements() <= 16);
+      assert(VT.getVectorNumElements() <= 16 || Subtarget.hasFP16());
       Opc = IsStrict ? X86ISD::STRICT_CMPM : X86ISD::CMPM;
----------------
craig.topper wrote:
> This should probably include EltVT==MVT::f16 for the FP16 override?
Maybe we can only check `EltVT == MVT::f16` like this?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D105263/new/

https://reviews.llvm.org/D105263