[PATCH] D103427: [X86] Fix handling of maskmovdqu in X32

Thu Mar 24 18:27:37 PDT 2022

skan added subscribers: HaohaiWen, skan.
skan added a comment.
Herald added a subscriber: StephenFan.
Herald added a project: All.

I'm sorry to say that this patch introduced a serious regression for the disassembler. Almost all the VEX instructions w/ address-size prefix can not be decoded due to this change.  The context `IC_64BIT_VEX_OPSIZE_ADSIZE` should never be added b/c we can always add a addr32 prefix on any VEX instruction w/ a memory operand.  I think the best way to support maskmovdqu in X32 is support something like `ExplicitVEXPrefix` in X86InstrFormats.td.

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103427/new/

https://reviews.llvm.org/D103427