mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-22 10:42:39 +01:00
6655b5e078
Summary: A number of optimizations are inhibited by single-use TokenFactors not being merged into the TokenFactor using it. This makes we consider if we can do the merge immediately. Most tests changes here are due to the change in visitation causing minor reorderings and associated reassociation of paired memory operations. CodeGen tests with non-reordering changes: X86/aligned-variadic.ll -- memory-based add folded into stored leaq value. X86/constant-combiners.ll -- Optimizes out overlap between stores. X86/pr40631_deadstore_elision -- folds constant byte store into preceding quad word constant store. Reviewers: RKSimon, craig.topper, spatel, efriedma, courbet Reviewed By: courbet Subscribers: dylanmckay, sdardis, nemanjai, jvesely, nhaehnle, javed.absar, eraman, hiraditya, kbarton, jrtc27, atanasyan, jsji, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D59260 llvm-svn: 356068
94 lines
2.7 KiB
LLVM
94 lines
2.7 KiB
LLVM
; RUN: llc < %s -mtriple=thumbv7-apple-ios -mcpu=cortex-a8 -pre-RA-sched=source -disable-post-ra | FileCheck %s -check-prefix=CHECK-7A
|
|
; RUN: llc < %s -mtriple=thumbv6m -pre-RA-sched=source -disable-post-ra -mattr=+strict-align | FileCheck %s -check-prefix=CHECK-6M
|
|
|
|
define void @t1(i8* nocapture %c) nounwind optsize {
|
|
entry:
|
|
; CHECK-7A-LABEL: t1:
|
|
; CHECK-7A: movs r1, #0
|
|
; CHECK-7A: strd r1, r1, [r0]
|
|
; CHECK-7A: str r1, [r0, #8]
|
|
; CHECK-6M-LABEL: t1:
|
|
; CHECK-6M: movs r1, #0
|
|
; CHECK-6M: str r1, [r0]
|
|
; CHECK-6M: str r1, [r0, #4]
|
|
; CHECK-6M: str r1, [r0, #8]
|
|
call void @llvm.memset.p0i8.i64(i8* align 8 %c, i8 0, i64 12, i1 false)
|
|
ret void
|
|
}
|
|
|
|
define void @t2() nounwind ssp {
|
|
entry:
|
|
; CHECK-7A-LABEL: t2:
|
|
; CHECK-7A: vmov.i32 {{q[0-9]+}}, #0x0
|
|
; CHECK-7A: movs r1, #10
|
|
; CHECK-7A: vst1.64 {d{{[0-9]+}}, d{{[0-9]+}}}, [r2], r1
|
|
; CHECK-7A: vst1.16 {d{{[0-9]+}}, d{{[0-9]+}}}, [r2]
|
|
; CHECK-6M-LABEL: t2:
|
|
; CHECK-6M: movs [[REG:r[0-9]+]], #0
|
|
; CHECK-6M-DAG: str [[REG]], [sp, #20]
|
|
; CHECK-6M-DAG: str [[REG]], [sp, #16]
|
|
; CHECK-6M-DAG: str [[REG]], [sp, #12]
|
|
; CHECK-6M-DAG: str [[REG]], [sp, #8]
|
|
; CHECK-6M-DAG: str [[REG]], [sp, #4]
|
|
; CHECK-6M-DAG: str [[REG]], [sp]
|
|
%buf = alloca [26 x i8], align 1
|
|
%0 = getelementptr inbounds [26 x i8], [26 x i8]* %buf, i32 0, i32 0
|
|
call void @llvm.memset.p0i8.i32(i8* %0, i8 0, i32 26, i1 false)
|
|
call void @something(i8* %0) nounwind
|
|
ret void
|
|
}
|
|
|
|
define void @t3(i8* %p) {
|
|
entry:
|
|
; CHECK-7A-LABEL: t3:
|
|
; CHECK-7A: muls [[REG:r[0-9]+]],
|
|
; CHECK-7A: str [[REG]],
|
|
; CHECK-6M-LABEL: t3:
|
|
; CHECK-6M-NOT: muls
|
|
; CHECK-6M: strb [[REG:r[0-9]+]],
|
|
; CHECK-6M: strb [[REG]],
|
|
; CHECK-6M: strb [[REG]],
|
|
; CHECK-6M: strb [[REG]],
|
|
br label %for.body
|
|
|
|
for.body:
|
|
%i = phi i32 [ 0, %entry ], [ %inc, %for.body ]
|
|
%0 = trunc i32 %i to i8
|
|
call void @llvm.memset.p0i8.i32(i8* %p, i8 %0, i32 4, i1 false)
|
|
call void @something(i8* %p)
|
|
%inc = add nuw nsw i32 %i, 1
|
|
%exitcond = icmp eq i32 %inc, 255
|
|
br i1 %exitcond, label %for.end, label %for.body
|
|
|
|
for.end:
|
|
ret void
|
|
}
|
|
|
|
define void @t4(i8* %p) {
|
|
entry:
|
|
; CHECK-7A-LABEL: t4:
|
|
; CHECK-7A: muls [[REG:r[0-9]+]],
|
|
; CHECK-7A: str [[REG]],
|
|
; CHECK-6M-LABEL: t4:
|
|
; CHECK-6M: muls [[REG:r[0-9]+]],
|
|
; CHECK-6M: strh [[REG]],
|
|
; CHECK-6M: strh [[REG]],
|
|
br label %for.body
|
|
|
|
for.body:
|
|
%i = phi i32 [ 0, %entry ], [ %inc, %for.body ]
|
|
%0 = trunc i32 %i to i8
|
|
call void @llvm.memset.p0i8.i32(i8* align 2 %p, i8 %0, i32 4, i1 false)
|
|
call void @something(i8* %p)
|
|
%inc = add nuw nsw i32 %i, 1
|
|
%exitcond = icmp eq i32 %inc, 255
|
|
br i1 %exitcond, label %for.end, label %for.body
|
|
|
|
for.end:
|
|
ret void
|
|
}
|
|
|
|
declare void @something(i8*) nounwind
|
|
declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i1) nounwind
|
|
declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i1) nounwind
|