llvm-mirror/test/CodeGen/X86/2007-08-10-SignExtSubreg.ll

; RUN: llvm-as < %s | llc -march=x86 | grep {movsbl}

@X = global i32 0               ; <i32*> [#uses=1]

define i8 @_Z3fooi(i32 %x) signext  {
entry:
        store i32 %x, i32* @X, align 4
        %retval67 = trunc i32 %x to i8          ; <i8> [#uses=1]
        ret i8 %retval67
}
Fix test so it passes. llvm-svn: 41012 2007-08-11 00:20:57 +02:00			`; RUN: llvm-as < %s \| llc -march=x86 \| grep {movsbl}`
Increase efficiency of sign_extend_inreg by using subregisters for truncation. As the README suggests sign_extend_subreg is selected to (sext(trunc)). llvm-svn: 41010 2007-08-10 23:48:46 +02:00
			`@X = global i32 0 ; <i32*> [#uses=1]`

			`define i8 @_Z3fooi(i32 %x) signext {`
			`entry:`
			`store i32 %x, i32* @X, align 4`
			`%retval67 = trunc i32 %x to i8 ; <i8> [#uses=1]`
			`ret i8 %retval67`
Convert tests using "\| wc -l \| grep ..." to use the count script. llvm-svn: 41097 2007-08-15 15:36:28 +02:00			`}`