[PATCH] D26586: [X86][AVX512][InlineASM][MS][llvm] (I|G)CC Memory adjustments compatibility

[Elena Demikhovsky via llvm-commits]

delena added inline comments.


================
Comment at: lib/Target/X86/AsmParser/X86AsmParser.cpp:2848
+  // Check for the existence of an AVX512 platform
+  if (!getSTI().getFeatureBits()[X86::FeatureAVX512])
+    return 0;
----------------
Put "assert" instead of "if".


================
Comment at: lib/Target/X86/AsmParser/X86AsmParser.cpp:2885
   for (const auto &Op : Operands) {
     X86Operand *X86Op = static_cast<X86Operand *>(Op.get());
+    if (UnsizedMemOp && !UnsizedMemOpNext)
----------------
IA allows only one memory operand per instruction.
if (UnsizedMemOp) {
      UnsizedMemOpNext = X86Op;
      break;
}
UnsizedMemOp = X86Op;



Repository:
  rL LLVM

https://reviews.llvm.org/D26586