maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

x86_64 ABI: Tweak merging algorithm so that we always bail early when 

we see a Memory classification.

llvm-svn: 63295
This commit is contained in:
Daniel Dunbar 2009-01-29 08:35:40 +00:00
parent 012468aa7f
commit 7cd7d47c5c
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
clang/lib/CodeGen/CGCall.cpp
View File

@ -548,13 +548,13 @@ void X86_64ABIInfo::classify(QualType Ty,
//
// (f) Otherwise class SSE is used.
if (Target == FieldLo || FieldLo == NoClass) ;
else if (Target == NoClass)
Target = FieldLo;
else if (FieldLo == Memory) {
// Memory is never over-ridden, just bail.
Lo = Memory;
return;
}
else if (Target == NoClass)
Target = FieldLo;
else if (Target == Integer || FieldLo == Integer)
Target = Integer;
else if (FieldLo == X87 || FieldLo == X87Up || FieldLo == ComplexX87) {