[PATCH] D32769: [X86][LWP] Add llvm support for LWP instructions.

[Filipe Cabecinhas via Phabricator via llvm-commits]

filcab added inline comments.


================
Comment at: lib/Target/X86/X86ISelLowering.cpp:20297
+    case llvm::Intrinsic::x86_flags_write_u32:
+    case llvm::Intrinsic::x86_flags_write_u64: {
       // We need a frame pointer because this will get lowered to a PUSH/POP
----------------
Can you split this refactor, please?


================
Comment at: test/CodeGen/X86/lwp-intrinsics-x86_64.ll:14
+; X64-NEXT:    retq
+  %1 = tail call i8 @llvm.x86.lwpins64(i64 %a0, i32 %a1, i32 81985529216486895)
+  ret i8 %1
----------------
RKSimon wrote:
> craig.topper wrote:
> > That immediate seems way out of range of 32-bit value. Is that intentional?
> Yes, I was trying to check that it truncated correctly. Should I comment it? Add it as a separate test?
I'm not sure this is useful, does it even make it past the IR reader?
Truncating a too large constant seems like it should be tested somewhere else.


Repository:
  rL LLVM

https://reviews.llvm.org/D32769