RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-21 03:53:04 +02:00
Code Issues Projects Releases Wiki Activity
2559048431
llvm-mirror/test/CodeGen/ARM/vargs_align.ll

30 lines
850 B
LLVM
Raw Normal View History

ARM: simplify and extend byval handling The main issue being fixed here is that APCS targets handling a "byval align N" parameter with N > 4 were miscounting what objects were where on the stack, leading to FrameLowering setting the frame pointer incorrectly and clobbering the stack. But byval handling had grown over many years, and had multiple layers of cruft trying to compensate for each other and calculate padding correctly. This only really needs to be done once, in the HandleByVal function. Elsewhere should just do what it's told by that call. I also stripped out unnecessary APCS/AAPCS distinctions (now that Clang emits byvals with the correct C ABI alignment), which simplified HandleByVal. rdar://20095672 llvm-svn: 231959
2015-03-11 19:54:22 +01:00
; RUN: llc < %s -mtriple=armv7-linux-gnueabihf | FileCheck %s -check-prefix=EABI
Convert to FileCheck. llvm-svn: 82710
2009-09-24 22:23:02 +02:00
; RUN: llc < %s -march=arm -mtriple=arm-linux-gnu | FileCheck %s -check-prefix=OABI
Fix stack alignment in functions with varargs. llvm-svn: 34532
2007-02-23 21:32:57 +01:00
define i32 @f(i32 %a, ...) {
entry:
%a_addr = alloca i32 ; <i32*> [#uses=1]
%retval = alloca i32, align 4 ; <i32*> [#uses=2]
%tmp = alloca i32, align 4 ; <i32*> [#uses=2]
store i32 %a, i32* %a_addr
store i32 0, i32* %tmp
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-27 22:17:42 +01:00
%tmp1 = load i32, i32* %tmp ; <i32> [#uses=1]
Fix stack alignment in functions with varargs. llvm-svn: 34532
2007-02-23 21:32:57 +01:00
store i32 %tmp1, i32* %retval
ARM / x86_64 varargs: Don't save regparms in prologue without va_start There's no need to do this if the user doesn't call va_start. In the future, we're going to have thunks that forward these register parameters with musttail calls, and they won't need these spills for handling va_start. Most of the test suite changes are adding va_start calls to existing tests to keep things working. llvm-svn: 216294
2014-08-22 23:59:26 +02:00
call void @llvm.va_start(i8* null)
ARM: simplify and extend byval handling The main issue being fixed here is that APCS targets handling a "byval align N" parameter with N > 4 were miscounting what objects were where on the stack, leading to FrameLowering setting the frame pointer incorrectly and clobbering the stack. But byval handling had grown over many years, and had multiple layers of cruft trying to compensate for each other and calculate padding correctly. This only really needs to be done once, in the HandleByVal function. Elsewhere should just do what it's told by that call. I also stripped out unnecessary APCS/AAPCS distinctions (now that Clang emits byvals with the correct C ABI alignment), which simplified HandleByVal. rdar://20095672 llvm-svn: 231959
2015-03-11 19:54:22 +01:00
call void asm sideeffect "", "~{d8}"()
Fix stack alignment in functions with varargs. llvm-svn: 34532
2007-02-23 21:32:57 +01:00
br label %return
return: ; preds = %entry
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-27 22:17:42 +01:00
%retval2 = load i32, i32* %retval ; <i32> [#uses=1]
Fix stack alignment in functions with varargs. llvm-svn: 34532
2007-02-23 21:32:57 +01:00
ret i32 %retval2
Convert to FileCheck. llvm-svn: 82710
2009-09-24 22:23:02 +02:00
; EABI: add sp, sp, #16
ARM: simplify and extend byval handling The main issue being fixed here is that APCS targets handling a "byval align N" parameter with N > 4 were miscounting what objects were where on the stack, leading to FrameLowering setting the frame pointer incorrectly and clobbering the stack. But byval handling had grown over many years, and had multiple layers of cruft trying to compensate for each other and calculate padding correctly. This only really needs to be done once, in the HandleByVal function. Elsewhere should just do what it's told by that call. I also stripped out unnecessary APCS/AAPCS distinctions (now that Clang emits byvals with the correct C ABI alignment), which simplified HandleByVal. rdar://20095672 llvm-svn: 231959
2015-03-11 19:54:22 +01:00
; EABI: vpop {d8}
; EABI: add sp, sp, #4
; EABI: add sp, sp, #12
Convert to FileCheck. llvm-svn: 82710
2009-09-24 22:23:02 +02:00
; OABI: add sp, sp, #12
; OABI: add sp, sp, #12
Fix stack alignment in functions with varargs. llvm-svn: 34532
2007-02-23 21:32:57 +01:00
}
ARM / x86_64 varargs: Don't save regparms in prologue without va_start There's no need to do this if the user doesn't call va_start. In the future, we're going to have thunks that forward these register parameters with musttail calls, and they won't need these spills for handling va_start. Most of the test suite changes are adding va_start calls to existing tests to keep things working. llvm-svn: 216294
2014-08-22 23:59:26 +02:00
declare void @llvm.va_start(i8*) nounwind
Reference in New Issue Copy Permalink