llvm/test/CodeGen/X86/memcpy.ll
Pete Cooper 8b170f7f29 Change memcpy/memset/memmove to have dest and source alignments.
Note, this was reviewed (and more details are in) http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html

These intrinsics currently have an explicit alignment argument which is
required to be a constant integer.  It represents the alignment of the
source and dest, and so must be the minimum of those.

This change allows source and dest to each have their own alignments
by using the alignment attribute on their arguments.  The alignment
argument itself is removed.

There are a few places in the code for which the code needs to be
checked by an expert as to whether using only src/dest alignment is
safe.  For those places, they currently take the minimum of src/dest
alignments which matches the current behaviour.

For example, code which used to read:
  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 500, i32 8, i1 false)
will now read:
  call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 8 %dest, i8* align 8 %src, i32 500, i1 false)

For out of tree owners, I was able to strip alignment from calls using sed by replacing:
  (call.*llvm\.memset.*)i32\ [0-9]*\,\ i1 false\)
with:
  $1i1 false)

and similarly for memmove and memcpy.

I then added back in alignment to test cases which needed it.

A similar commit will be made to clang which actually has many differences in alignment as now
IRBuilder can generate different source/dest alignments on calls.

In IRBuilder itself, a new argument was added.  Instead of calling:
  CreateMemCpy(Dst, Src, getInt64(Size), DstAlign, /* isVolatile */ false)
you now call
  CreateMemCpy(Dst, Src, getInt64(Size), DstAlign, SrcAlign, /* isVolatile */ false)

There is a temporary class (IntegerAlignment) which takes the source alignment and rejects
implicit conversion from bool.  This is to prevent isVolatile here from passing its default
parameter to the source alignment.

Note, changes in future can now be made to codegen.  I didn't change anything here, but this
change should enable better memcpy code sequences.

Reviewed by Hal Finkel.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@253511 91177308-0d34-0410-b5e6-96231b3b80d8
2015-11-18 22:17:24 +00:00

141 lines
3.6 KiB
LLVM

; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=core2 | FileCheck %s -check-prefix=LINUX
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=core2 | FileCheck %s -check-prefix=DARWIN
declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i1) nounwind
; Variable memcpy's should lower to calls.
define i8* @test1(i8* %a, i8* %b, i64 %n) nounwind {
entry:
tail call void @llvm.memcpy.p0i8.p0i8.i64( i8* %a, i8* %b, i64 %n, i1 0 )
ret i8* %a
; LINUX-LABEL: test1:
; LINUX: memcpy
}
; Variable memcpy's should lower to calls.
define i8* @test2(i64* %a, i64* %b, i64 %n) nounwind {
entry:
%tmp14 = bitcast i64* %a to i8*
%tmp25 = bitcast i64* %b to i8*
tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp14, i8* %tmp25, i64 %n, i1 0 )
ret i8* %tmp14
; LINUX-LABEL: test2:
; LINUX: memcpy
}
; Large constant memcpy's should lower to a call when optimizing for size.
; PR6623
; On the other hand, Darwin's definition of -Os is optimizing for size without
; hurting performance so it should just ignore optsize when expanding memcpy.
; rdar://8821501
define void @test3(i8* nocapture %A, i8* nocapture %B) nounwind optsize noredzone {
entry:
tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i1 false)
ret void
; LINUX-LABEL: test3:
; LINUX: memcpy
; DARWIN-LABEL: test3:
; DARWIN-NOT: memcpy
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
; DARWIN: movq
}
define void @test3_minsize(i8* nocapture %A, i8* nocapture %B) nounwind minsize noredzone {
tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i1 false)
ret void
; LINUX-LABEL: test3_minsize:
; LINUX: memcpy
; DARWIN-LABEL: test3_minsize:
; DARWIN: memcpy
}
define void @test3_minsize_optsize(i8* nocapture %A, i8* nocapture %B) nounwind optsize minsize noredzone {
tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i1 false)
ret void
; LINUX-LABEL: test3_minsize_optsize:
; LINUX: memcpy
; DARWIN-LABEL: test3_minsize_optsize:
; DARWIN: memcpy
}
; Large constant memcpy's should be inlined when not optimizing for size.
define void @test4(i8* nocapture %A, i8* nocapture %B) nounwind noredzone {
entry:
tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %A, i8* %B, i64 64, i1 false)
ret void
; LINUX-LABEL: test4:
; LINUX: movq
; LINUX: movq
; LINUX: movq
; LINUX: movq
; LINUX: movq
; LINUX: movq
; LINUX: movq
; LINUX: movq
; LINUX: movq
; LINUX: movq
; LINUX: movq
; LINUX: movq
}
@.str = private unnamed_addr constant [30 x i8] c"\00aaaaaaaaaaaaaaaaaaaaaaaaaaaa\00", align 1
define void @test5(i8* nocapture %C) nounwind uwtable ssp {
entry:
tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %C, i8* getelementptr inbounds ([30 x i8], [30 x i8]* @.str, i64 0, i64 0), i64 16, i1 false)
ret void
; DARWIN-LABEL: test5:
; DARWIN: movabsq $7016996765293437281
; DARWIN: movabsq $7016996765293437184
}
; PR14896
@.str2 = private unnamed_addr constant [2 x i8] c"x\00", align 1
define void @test6() nounwind uwtable {
entry:
; DARWIN: test6
; DARWIN: movw $0, 8
; DARWIN: movq $120, 0
tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* null, i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str2, i64 0, i64 0), i64 10, i1 false)
ret void
}
define void @PR15348(i8* %a, i8* %b) {
; Ensure that alignment of '0' in an @llvm.memcpy intrinsic results in
; unaligned loads and stores.
; LINUX: PR15348
; LINUX: movb
; LINUX: movb
; LINUX: movq
; LINUX: movq
; LINUX: movq
; LINUX: movq
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %b, i64 17, i1 false)
ret void
}