llvm-6502/test/CodeGen/X86/pr12360.ll

; RUN: llc < %s -mtriple=x86_64-apple-darwin | FileCheck %s

define zeroext i1 @f1(i8* %x) {
entry:
  %0 = load i8* %x, align 1, !range !0
  %tobool = trunc i8 %0 to i1
  ret i1 %tobool
}

; CHECK: f1:
; CHECK: movb	(%rdi), %al
; CHECK-NEXT: ret

!0 = metadata !{i8 0, i8 2}
Add a triple to the test. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@153818 91177308-0d34-0410-b5e6-96231b3b80d8 2012-03-31 18:59:07 +00:00			`; RUN: llc < %s -mtriple=x86_64-apple-darwin \| FileCheck %s`
Teach CodeGen's version of computeMaskedBits to understand the range metadata. This is the CodeGen equivalent of r153747. I tested that there is not noticeable performance difference with any combination of -O0/-O2 /-g when compiling gcc as a single compilation unit. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@153817 91177308-0d34-0410-b5e6-96231b3b80d8 2012-03-31 18:14:00 +00:00
			`define zeroext i1 @f1(i8* %x) {`
			`entry:`
			`%0 = load i8* %x, align 1, !range !0`
			`%tobool = trunc i8 %0 to i1`
			`ret i1 %tobool`
			`}`

			`; CHECK: f1:`
			`; CHECK: movb (%rdi), %al`
			`; CHECK-NEXT: ret`

			`!0 = metadata !{i8 0, i8 2}`