maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/X86/stack-align-memcpy.ll

44 lines
1.3 KiB
LLVM
Raw Normal View History

X86: Disable generation of rep;movsl when %esi is used as a base pointer. This happens when there is both stack realignment and a dynamic alloca in the function. If we overwrite %esi (rep;movsl uses fixed registers) we'll lose the base pointer and the next register spill will write into oblivion. Fixes PR15249 and unbreaks firefox on i386/freebsd. Mozilla uses dynamic allocas and freebsd a 4 byte stack alignment. llvm-svn: 175057
2013-02-13 21:40:35 +08:00
; RUN: llc < %s -force-align-stack -mtriple i386-apple-darwin -mcpu=i486 | FileCheck %s
%struct.foo = type { [88 x i8] }
Check for dynamic allocas and inline asm that clobbers sp before building selection dag (PR19012) In X86SelectionDagInfo::EmitTargetCodeForMemcpy we check with MachineFrameInfo to make sure that ESI isn't used as a base pointer register before we choose to emit rep movs (which clobbers esi). The problem is that MachineFrameInfo wouldn't know about dynamic allocas or inline asm that clobbers the stack pointer until SelectionDAGBuilder has encountered them. This patch fixes the problem by checking for such things when building the FunctionLoweringInfo. Differential Revision: http://llvm-reviews.chandlerc.com/D2954 llvm-svn: 202930
2014-03-05 10:43:26 +08:00
declare void @bar(i8* nocapture, %struct.foo* align 4 byval) nounwind
declare void @baz(i8*) nounwind
X86: Disable generation of rep;movsl when %esi is used as a base pointer. This happens when there is both stack realignment and a dynamic alloca in the function. If we overwrite %esi (rep;movsl uses fixed registers) we'll lose the base pointer and the next register spill will write into oblivion. Fixes PR15249 and unbreaks firefox on i386/freebsd. Mozilla uses dynamic allocas and freebsd a 4 byte stack alignment. llvm-svn: 175057
2013-02-13 21:40:35 +08:00
; PR15249
; We can't use rep;movsl here because it clobbers the base pointer in %esi.
define void @test1(%struct.foo* nocapture %x, i32 %y) nounwind {
%dynalloc = alloca i8, i32 %y, align 1
call void @bar(i8* %dynalloc, %struct.foo* align 4 byval %x)
ret void
Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion. This was done with the following sed invocation to catch label lines demarking function boundaries: sed -i '' "s/^;\( *\)\([A-Z0-9_]*\):\( *\)test\([A-Za-z0-9_-]*\):\( *\)$/;\1\2-LABEL:\3test\4:\5/g" test/CodeGen/*/*.ll which was written conservatively to avoid false positives rather than false negatives. I scanned through all the changes and everything looks correct. llvm-svn: 186258
2013-07-14 04:38:47 +08:00
; CHECK-LABEL: test1:
X86: Disable generation of rep;movsl when %esi is used as a base pointer. This happens when there is both stack realignment and a dynamic alloca in the function. If we overwrite %esi (rep;movsl uses fixed registers) we'll lose the base pointer and the next register spill will write into oblivion. Fixes PR15249 and unbreaks firefox on i386/freebsd. Mozilla uses dynamic allocas and freebsd a 4 byte stack alignment. llvm-svn: 175057
2013-02-13 21:40:35 +08:00
; CHECK: andl $-16, %esp
; CHECK: movl %esp, %esi
Revert "X86 memcpy lowering: use "rep movs" even when esi is used as base pointer" (r204174) > For functions where esi is used as base pointer, we would previously fall ba > from lowering memcpy with "rep movs" because that clobbers esi. > > With this patch, we just store esi in another physical register, and restore > it afterwards. This adds a little bit of register preassure, but the more > efficient memcpy should be worth it. > > Differential Revision: http://llvm-reviews.chandlerc.com/D2968 This didn't work. I was ending up with code like this: lea edi,[esi+38h] mov ecx,0Fh mov edx,esi mov esi,ebx rep movs dword ptr es:[edi],dword ptr [esi] lea ecx,[esi+74h] <-- Ooops, we're now using esi before restoring it from edx. add ebx,3Ch mov esi,edx I guess if we want to do this we need stronger glue or something, or doing the expansion much later. llvm-svn: 204829
2014-03-27 00:30:54 +08:00
; CHECK-NOT: rep;movsl
X86: Disable generation of rep;movsl when %esi is used as a base pointer. This happens when there is both stack realignment and a dynamic alloca in the function. If we overwrite %esi (rep;movsl uses fixed registers) we'll lose the base pointer and the next register spill will write into oblivion. Fixes PR15249 and unbreaks firefox on i386/freebsd. Mozilla uses dynamic allocas and freebsd a 4 byte stack alignment. llvm-svn: 175057
2013-02-13 21:40:35 +08:00
}
Check for dynamic allocas and inline asm that clobbers sp before building selection dag (PR19012) In X86SelectionDagInfo::EmitTargetCodeForMemcpy we check with MachineFrameInfo to make sure that ESI isn't used as a base pointer register before we choose to emit rep movs (which clobbers esi). The problem is that MachineFrameInfo wouldn't know about dynamic allocas or inline asm that clobbers the stack pointer until SelectionDAGBuilder has encountered them. This patch fixes the problem by checking for such things when building the FunctionLoweringInfo. Differential Revision: http://llvm-reviews.chandlerc.com/D2954 llvm-svn: 202930
2014-03-05 10:43:26 +08:00
; PR19012
; Also don't clobber %esi if the dynamic alloca comes after the memcpy.
define void @test2(%struct.foo* nocapture %x, i32 %y, i8* %z) nounwind {
call void @bar(i8* %z, %struct.foo* align 4 byval %x)
%dynalloc = alloca i8, i32 %y, align 1
call void @baz(i8* %dynalloc)
ret void
; CHECK-LABEL: test2:
; CHECK: movl %esp, %esi
Revert "X86 memcpy lowering: use "rep movs" even when esi is used as base pointer" (r204174) > For functions where esi is used as base pointer, we would previously fall ba > from lowering memcpy with "rep movs" because that clobbers esi. > > With this patch, we just store esi in another physical register, and restore > it afterwards. This adds a little bit of register preassure, but the more > efficient memcpy should be worth it. > > Differential Revision: http://llvm-reviews.chandlerc.com/D2968 This didn't work. I was ending up with code like this: lea edi,[esi+38h] mov ecx,0Fh mov edx,esi mov esi,ebx rep movs dword ptr es:[edi],dword ptr [esi] lea ecx,[esi+74h] <-- Ooops, we're now using esi before restoring it from edx. add ebx,3Ch mov esi,edx I guess if we want to do this we need stronger glue or something, or doing the expansion much later. llvm-svn: 204829
2014-03-27 00:30:54 +08:00
; CHECK-NOT: rep;movsl
Check for dynamic allocas and inline asm that clobbers sp before building selection dag (PR19012) In X86SelectionDagInfo::EmitTargetCodeForMemcpy we check with MachineFrameInfo to make sure that ESI isn't used as a base pointer register before we choose to emit rep movs (which clobbers esi). The problem is that MachineFrameInfo wouldn't know about dynamic allocas or inline asm that clobbers the stack pointer until SelectionDAGBuilder has encountered them. This patch fixes the problem by checking for such things when building the FunctionLoweringInfo. Differential Revision: http://llvm-reviews.chandlerc.com/D2954 llvm-svn: 202930
2014-03-05 10:43:26 +08:00
}
; Check that we do use rep movs if we make the alloca static.
define void @test3(%struct.foo* nocapture %x, i32 %y, i8* %z) nounwind {
call void @bar(i8* %z, %struct.foo* align 4 byval %x)
%statalloc = alloca i8, i32 8, align 1
call void @baz(i8* %statalloc)
ret void
; CHECK-LABEL: test3:
; CHECK: rep;movsl
}