[PATCH] D32162: Inline asm 0bH conflict

[Andrew V. Tischenko via Phabricator via llvm-commits]

avt77 added inline comments.


================
Comment at: lib/MC/MCParser/AsmParser.cpp:740
   AsmCond StartingCondState = TheCondState;
+  SmallVector<AsmRewrite, 4> AsmStrRewrites;
 
----------------
RKSimon wrote:
> avt77 wrote:
> > craig.topper wrote:
> > > What is this change for?
> > It's used below in "parseStatement". Without it"parseStatement" crashes.
> Should we even allow a ParseStatementInfo default constructor in that case?
From my point of view we should not.


================
Comment at: lib/MC/MCParser/AsmParser.cpp:1654
     // Emit the label.
-    if (!ParsingInlineAsm)
+    if (!getTargetParser().isParsingInlineAsm())
       Out.EmitLabel(Sym, IDLoc);
----------------
RKSimon wrote:
> Please explain why we're using the target parser here and not anywhere else that ParsingInlineAsm is used.
We use "getTargetParser()" below as well. From my point of view we should change the name here. It should be called something like isParsingMSAsm() because this flag really means exactly that.


https://reviews.llvm.org/D32162