llvm-mirror/test/CodeGen/X86/fastcc-byval.ll
Dan Gohman dcd4896f12 Fix byval arguments in the fastcc calling convention. The fastcc convention
delegates to the regular x86-32 convention which handles byval, but only
after it handles a few cases, and it's necessary to handle byval before
handling those cases. This fixes PR3122 (and rdar://6400815), llvm-gcc
miscompiling LLVM.

llvm-svn: 60453
2008-12-03 01:28:04 +00:00

21 lines
627 B
LLVM

; RUN: llvm-as < %s | llc | grep {movl\[\[:space:\]\]*8(%esp), %eax} | count 2
; PR3122
; rdar://6400815
; byval requires a copy, even with fastcc.
target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
target triple = "i386-apple-darwin9.5"
%struct.MVT = type { i32 }
define fastcc i32 @bar() nounwind {
%V = alloca %struct.MVT
%a = getelementptr %struct.MVT* %V, i32 0, i32 0
store i32 1, i32* %a
call fastcc void @foo(%struct.MVT* byval %V) nounwind
%t = load i32* %a
ret i32 %t
}
declare fastcc void @foo(%struct.MVT* byval)