mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-07-21 18:29:45 +00:00
7c9c6ed761
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
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230794 91177308-0d34-0410-b5e6-96231b3b80d8
101 lines
3.3 KiB
LLVM
101 lines
3.3 KiB
LLVM
; RUN: llc -mtriple=i686-windows -mattr=+sse2 < %s | FileCheck %s
|
|
|
|
target datalayout = "e-m:w-p:32:32-i64:64-f80:32-n8:16:32-S32"
|
|
target triple = "i686-pc-windows-msvc"
|
|
|
|
; There is a conflict between lowering the X86 memory intrinsics and the "base"
|
|
; register used to address stack locals. See X86RegisterInfo::hasBaseRegister
|
|
; for when this is necessary. Typically, we chose ESI for the base register,
|
|
; which all of the X86 string instructions use.
|
|
|
|
; The pattern of vector icmp and extractelement is used in these tests because
|
|
; it forces creation of an aligned stack temporary. Perhaps such temporaries
|
|
; shouldn't be aligned.
|
|
|
|
declare void @escape_vla_and_icmp(i8*, i1 zeroext)
|
|
declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1)
|
|
declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1)
|
|
|
|
define i32 @memcpy_novla_vector(<4 x i32>* %vp0, i8* %a, i8* %b, i32 %n, i1 zeroext %cond) {
|
|
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a, i8* %b, i32 128, i32 4, i1 false)
|
|
br i1 %cond, label %spill_vectors, label %no_vectors
|
|
|
|
no_vectors:
|
|
ret i32 0
|
|
|
|
spill_vectors:
|
|
%vp1 = getelementptr <4 x i32>, <4 x i32>* %vp0, i32 1
|
|
%v0 = load <4 x i32>, <4 x i32>* %vp0
|
|
%v1 = load <4 x i32>, <4 x i32>* %vp1
|
|
%vicmp = icmp slt <4 x i32> %v0, %v1
|
|
%icmp = extractelement <4 x i1> %vicmp, i32 0
|
|
call void @escape_vla_and_icmp(i8* null, i1 zeroext %icmp)
|
|
%r = extractelement <4 x i32> %v0, i32 0
|
|
ret i32 %r
|
|
}
|
|
|
|
; CHECK-LABEL: _memcpy_novla_vector:
|
|
; CHECK: andl $-16, %esp
|
|
; CHECK-DAG: movl $32, %ecx
|
|
; CHECK-DAG: movl {{.*}}, %esi
|
|
; CHECK-DAG: movl {{.*}}, %edi
|
|
; CHECK: rep;movsl
|
|
|
|
define i32 @memcpy_vla_vector(<4 x i32>* %vp0, i8* %a, i8* %b, i32 %n, i1 zeroext %cond) {
|
|
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a, i8* %b, i32 128, i32 4, i1 false)
|
|
br i1 %cond, label %spill_vectors, label %no_vectors
|
|
|
|
no_vectors:
|
|
ret i32 0
|
|
|
|
spill_vectors:
|
|
%vp1 = getelementptr <4 x i32>, <4 x i32>* %vp0, i32 1
|
|
%v0 = load <4 x i32>, <4 x i32>* %vp0
|
|
%v1 = load <4 x i32>, <4 x i32>* %vp1
|
|
%vicmp = icmp slt <4 x i32> %v0, %v1
|
|
%icmp = extractelement <4 x i1> %vicmp, i32 0
|
|
%vla = alloca i8, i32 %n
|
|
call void @escape_vla_and_icmp(i8* %vla, i1 zeroext %icmp)
|
|
%r = extractelement <4 x i32> %v0, i32 0
|
|
ret i32 %r
|
|
}
|
|
|
|
; CHECK-LABEL: _memcpy_vla_vector:
|
|
; CHECK: andl $-16, %esp
|
|
; CHECK: movl %esp, %esi
|
|
; CHECK: pushl $128
|
|
; CHECK: calll _memcpy
|
|
; CHECK: calll __chkstk
|
|
|
|
; stosd doesn't clobber esi, so we can use it.
|
|
|
|
define i32 @memset_vla_vector(<4 x i32>* %vp0, i8* %a, i32 %n, i1 zeroext %cond) {
|
|
call void @llvm.memset.p0i8.i32(i8* %a, i8 42, i32 128, i32 4, i1 false)
|
|
br i1 %cond, label %spill_vectors, label %no_vectors
|
|
|
|
no_vectors:
|
|
ret i32 0
|
|
|
|
spill_vectors:
|
|
%vp1 = getelementptr <4 x i32>, <4 x i32>* %vp0, i32 1
|
|
%v0 = load <4 x i32>, <4 x i32>* %vp0
|
|
%v1 = load <4 x i32>, <4 x i32>* %vp1
|
|
%vicmp = icmp slt <4 x i32> %v0, %v1
|
|
%icmp = extractelement <4 x i1> %vicmp, i32 0
|
|
%vla = alloca i8, i32 %n
|
|
call void @escape_vla_and_icmp(i8* %vla, i1 zeroext %icmp)
|
|
%r = extractelement <4 x i32> %v0, i32 0
|
|
ret i32 %r
|
|
}
|
|
|
|
; CHECK-LABEL: _memset_vla_vector:
|
|
; CHECK: andl $-16, %esp
|
|
; CHECK: movl %esp, %esi
|
|
; CHECK-DAG: movl $707406378, %eax # imm = 0x2A2A2A2A
|
|
; CHECK-DAG: movl $32, %ecx
|
|
; CHECK-DAG: movl {{.*}}, %edi
|
|
; CHECK-NOT: movl {{.*}}, %esi
|
|
; CHECK: rep;stosl
|
|
|
|
; Add a test for memcmp if we ever add a special lowering for it.
|