RPCS3/llvm
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm.git synced 2025-01-12 23:40:43 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
llvm/test/Transforms/InstCombine/memcpy-from-global.ll

147 lines
7.8 KiB
LLVM
Raw Normal View History

Port the global copy optimization from the SROA pass to InstCombine. This optimization is really just replacing allocas wholesale with globals, there is no scalarization. The underlying motivation for this patch is to simplify the SROA pass and focus it on splitting and promoting allocas. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162271 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-21 08:39:44 +00:00
; RUN: opt < %s -instcombine -S | FileCheck %s
Make opt default to not adding a target data string and update tests that depend on target data to supply it within the test git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@85900 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-03 15:29:06 +00:00
target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64"
new testcase, should be able to eliminate the alloca and memcpy git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@36428 91177308-0d34-0410-b5e6-96231b3b80d8
2007-04-25 06:29:34 +00:00
@C.0.1248 = internal constant [128 x float] [ float -1.000000e+00, float -1.000000e+00, float -1.000000e+00, float 0.000000e+00, float -1.000000e+00, float -1.000000e+00, float 0.000000e+00, float -1.000000e+00, float -1.000000e+00, float -1.000000e+00, float 0.000000e+00, float 1.000000e+00, float -1.000000e+00, float -1.000000e+00, float 1.000000e+00, float 0.000000e+00, float -1.000000e+00, float 0.000000e+00, float -1.000000e+00, float -1.000000e+00, float -1.000000e+00, float 0.000000e+00, float -1.000000e+00, float 1.000000e+00, float -1.000000e+00, float 0.000000e+00, float 1.000000e+00, float -1.000000e+00, float -1.000000e+00, float 0.000000e+00, float 1.000000e+00, float 1.000000e+00, float -1.000000e+00, float 1.000000e+00, float -1.000000e+00, float 0.000000e+00, float -1.000000e+00, float 1.000000e+00, float 0.000000e+00, float -1.000000e+00, float -1.000000e+00, float 1.000000e+00, float 0.000000e+00, float 1.000000e+00, float -1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 0.000000e+00, float 0.000000e+00, float -1.000000e+00, float -1.000000e+00, float -1.000000e+00, float 0.000000e+00, float -1.000000e+00, float -1.000000e+00, float 1.000000e+00, float 0.000000e+00, float -1.000000e+00, float 1.000000e+00, float -1.000000e+00, float 0.000000e+00, float -1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float -1.000000e+00, float -1.000000e+00, float 0.000000e+00, float 1.000000e+00, float -1.000000e+00, float 0.000000e+00, float -1.000000e+00, float 1.000000e+00, float -1.000000e+00, float 0.000000e+00, float 1.000000e+00, float 1.000000e+00, float -1.000000e+00, float 1.000000e+00, float 0.000000e+00, float 1.000000e+00, float 0.000000e+00, float -1.000000e+00, float -1.000000e+00, float 1.000000e+00, float 0.000000e+00, float -1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 0.000000e+00, float 1.000000e+00, float -1.000000e+00, float 1.000000e+00, float 0.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float -1.000000e+00, float 0.000000e+00, float 1.000000e+00, float 1.000000e+00, float 0.000000e+00, float -1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 0.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 0.000000e+00, float 0.000000e+00, float 1.000000e+00, float -1.000000e+00, float -1.000000e+00, float 0.000000e+00, float 1.000000e+00, float -1.000000e+00, float 1.000000e+00, float 0.000000e+00, float 1.000000e+00, float 1.000000e+00, float -1.000000e+00, float 0.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00 ], align 32 ; <[128 x float]*> [#uses=1]
filecheckize git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119681 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:16:43 +00:00
define float @test1(i32 %hash, float %x, float %y, float %z, float %w) {
new testcase, should be able to eliminate the alloca and memcpy git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@36428 91177308-0d34-0410-b5e6-96231b3b80d8
2007-04-25 06:29:34 +00:00
entry:
%lookupTable = alloca [128 x float], align 16 ; <[128 x float]*> [#uses=5]
%lookupTable1 = bitcast [128 x float]* %lookupTable to i8* ; <i8*> [#uses=1]
Port the global copy optimization from the SROA pass to InstCombine. This optimization is really just replacing allocas wholesale with globals, there is no scalarization. The underlying motivation for this patch is to simplify the SROA pass and focus it on splitting and promoting allocas. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162271 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-21 08:39:44 +00:00
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %lookupTable1, i8* bitcast ([128 x float]* @C.0.1248 to i8*), i64 512, i32 16, i1 false)
filecheckize git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119681 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:16:43 +00:00
; CHECK: @test1
; CHECK-NOT: alloca
; CHECK-NOT: call{{.*}}@llvm.memcpy
new testcase, should be able to eliminate the alloca and memcpy git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@36428 91177308-0d34-0410-b5e6-96231b3b80d8
2007-04-25 06:29:34 +00:00
%tmp3 = shl i32 %hash, 2 ; <i32> [#uses=1]
%tmp5 = and i32 %tmp3, 124 ; <i32> [#uses=4]
%tmp753 = getelementptr [128 x float]* %lookupTable, i32 0, i32 %tmp5 ; <float*> [#uses=1]
%tmp9 = load float* %tmp753 ; <float> [#uses=1]
Split the Add, Sub, and Mul instruction opcodes into separate integer and floating-point opcodes, introducing FAdd, FSub, and FMul. For now, the AsmParser, BitcodeReader, and IRBuilder all preserve backwards compatability, and the Core LLVM APIs preserve backwards compatibility for IR producers. Most front-ends won't need to change immediately. This implements the first step of the plan outlined here: http://nondot.org/sabre/LLVMNotes/IntegerOverflow.txt git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72897 91177308-0d34-0410-b5e6-96231b3b80d8
2009-06-04 22:49:04 +00:00
%tmp11 = fmul float %tmp9, %x ; <float> [#uses=1]
%tmp13 = fadd float %tmp11, 0.000000e+00 ; <float> [#uses=1]
new testcase, should be able to eliminate the alloca and memcpy git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@36428 91177308-0d34-0410-b5e6-96231b3b80d8
2007-04-25 06:29:34 +00:00
%tmp17.sum52 = or i32 %tmp5, 1 ; <i32> [#uses=1]
%tmp1851 = getelementptr [128 x float]* %lookupTable, i32 0, i32 %tmp17.sum52 ; <float*> [#uses=1]
%tmp19 = load float* %tmp1851 ; <float> [#uses=1]
Split the Add, Sub, and Mul instruction opcodes into separate integer and floating-point opcodes, introducing FAdd, FSub, and FMul. For now, the AsmParser, BitcodeReader, and IRBuilder all preserve backwards compatability, and the Core LLVM APIs preserve backwards compatibility for IR producers. Most front-ends won't need to change immediately. This implements the first step of the plan outlined here: http://nondot.org/sabre/LLVMNotes/IntegerOverflow.txt git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72897 91177308-0d34-0410-b5e6-96231b3b80d8
2009-06-04 22:49:04 +00:00
%tmp21 = fmul float %tmp19, %y ; <float> [#uses=1]
%tmp23 = fadd float %tmp21, %tmp13 ; <float> [#uses=1]
new testcase, should be able to eliminate the alloca and memcpy git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@36428 91177308-0d34-0410-b5e6-96231b3b80d8
2007-04-25 06:29:34 +00:00
%tmp27.sum50 = or i32 %tmp5, 2 ; <i32> [#uses=1]
%tmp2849 = getelementptr [128 x float]* %lookupTable, i32 0, i32 %tmp27.sum50 ; <float*> [#uses=1]
%tmp29 = load float* %tmp2849 ; <float> [#uses=1]
Split the Add, Sub, and Mul instruction opcodes into separate integer and floating-point opcodes, introducing FAdd, FSub, and FMul. For now, the AsmParser, BitcodeReader, and IRBuilder all preserve backwards compatability, and the Core LLVM APIs preserve backwards compatibility for IR producers. Most front-ends won't need to change immediately. This implements the first step of the plan outlined here: http://nondot.org/sabre/LLVMNotes/IntegerOverflow.txt git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72897 91177308-0d34-0410-b5e6-96231b3b80d8
2009-06-04 22:49:04 +00:00
%tmp31 = fmul float %tmp29, %z ; <float> [#uses=1]
%tmp33 = fadd float %tmp31, %tmp23 ; <float> [#uses=1]
new testcase, should be able to eliminate the alloca and memcpy git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@36428 91177308-0d34-0410-b5e6-96231b3b80d8
2007-04-25 06:29:34 +00:00
%tmp37.sum48 = or i32 %tmp5, 3 ; <i32> [#uses=1]
%tmp3847 = getelementptr [128 x float]* %lookupTable, i32 0, i32 %tmp37.sum48 ; <float*> [#uses=1]
%tmp39 = load float* %tmp3847 ; <float> [#uses=1]
Split the Add, Sub, and Mul instruction opcodes into separate integer and floating-point opcodes, introducing FAdd, FSub, and FMul. For now, the AsmParser, BitcodeReader, and IRBuilder all preserve backwards compatability, and the Core LLVM APIs preserve backwards compatibility for IR producers. Most front-ends won't need to change immediately. This implements the first step of the plan outlined here: http://nondot.org/sabre/LLVMNotes/IntegerOverflow.txt git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72897 91177308-0d34-0410-b5e6-96231b3b80d8
2009-06-04 22:49:04 +00:00
%tmp41 = fmul float %tmp39, %w ; <float> [#uses=1]
%tmp43 = fadd float %tmp41, %tmp33 ; <float> [#uses=1]
new testcase, should be able to eliminate the alloca and memcpy git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@36428 91177308-0d34-0410-b5e6-96231b3b80d8
2007-04-25 06:29:34 +00:00
ret float %tmp43
}
fix a small oversight in the "eliminate memcpy from constant global" optimization. If the alloca that is "memcpy'd from constant" also has a memcpy from *it*, ignore it: it is a load. We now optimize the testcase to: define void @test2() { %B = alloca %T %a = bitcast %T* @G to i8* %b = bitcast %T* %B to i8* call void @llvm.memcpy.p0i8.p0i8.i64(i8* %b, i8* %a, i64 124, i32 4, i1 false) call void @bar(i8* %b) ret void } previously we would generate: define void @test() { %B = alloca %T %b = bitcast %T* %B to i8* %G.0 = getelementptr inbounds %T* @G, i32 0, i32 0 %tmp3 = load i8* %G.0, align 4 %G.1 = getelementptr inbounds %T* @G, i32 0, i32 1 %G.15 = bitcast [123 x i8]* %G.1 to i8* %1 = bitcast [123 x i8]* %G.1 to i984* %srcval = load i984* %1, align 1 %B.0 = getelementptr inbounds %T* %B, i32 0, i32 0 store i8 %tmp3, i8* %B.0, align 4 %B.1 = getelementptr inbounds %T* %B, i32 0, i32 1 %B.12 = bitcast [123 x i8]* %B.1 to i8* %2 = bitcast [123 x i8]* %B.1 to i984* store i984 %srcval, i984* %2, align 1 call void @bar(i8* %b) ret void } git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119682 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:20:47 +00:00
declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
%T = type { i8, [123 x i8] }
Do not eliminate allocas whose alignment exceeds that of the copied-in constant, as a subsequent user may rely on over alignment. Fixes PR12885. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@157134 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-19 22:52:10 +00:00
%U = type { i32, i32, i32, i32, i32 }
fix a small oversight in the "eliminate memcpy from constant global" optimization. If the alloca that is "memcpy'd from constant" also has a memcpy from *it*, ignore it: it is a load. We now optimize the testcase to: define void @test2() { %B = alloca %T %a = bitcast %T* @G to i8* %b = bitcast %T* %B to i8* call void @llvm.memcpy.p0i8.p0i8.i64(i8* %b, i8* %a, i64 124, i32 4, i1 false) call void @bar(i8* %b) ret void } previously we would generate: define void @test() { %B = alloca %T %b = bitcast %T* %B to i8* %G.0 = getelementptr inbounds %T* @G, i32 0, i32 0 %tmp3 = load i8* %G.0, align 4 %G.1 = getelementptr inbounds %T* @G, i32 0, i32 1 %G.15 = bitcast [123 x i8]* %G.1 to i8* %1 = bitcast [123 x i8]* %G.1 to i984* %srcval = load i984* %1, align 1 %B.0 = getelementptr inbounds %T* %B, i32 0, i32 0 store i8 %tmp3, i8* %B.0, align 4 %B.1 = getelementptr inbounds %T* %B, i32 0, i32 1 %B.12 = bitcast [123 x i8]* %B.1 to i8* %2 = bitcast [123 x i8]* %B.1 to i984* store i984 %srcval, i984* %2, align 1 call void @bar(i8* %b) ret void } git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119682 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:20:47 +00:00
@G = constant %T {i8 1, [123 x i8] zeroinitializer }
Do not eliminate allocas whose alignment exceeds that of the copied-in constant, as a subsequent user may rely on over alignment. Fixes PR12885. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@157134 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-19 22:52:10 +00:00
@H = constant [2 x %U] zeroinitializer, align 16
fix a small oversight in the "eliminate memcpy from constant global" optimization. If the alloca that is "memcpy'd from constant" also has a memcpy from *it*, ignore it: it is a load. We now optimize the testcase to: define void @test2() { %B = alloca %T %a = bitcast %T* @G to i8* %b = bitcast %T* %B to i8* call void @llvm.memcpy.p0i8.p0i8.i64(i8* %b, i8* %a, i64 124, i32 4, i1 false) call void @bar(i8* %b) ret void } previously we would generate: define void @test() { %B = alloca %T %b = bitcast %T* %B to i8* %G.0 = getelementptr inbounds %T* @G, i32 0, i32 0 %tmp3 = load i8* %G.0, align 4 %G.1 = getelementptr inbounds %T* @G, i32 0, i32 1 %G.15 = bitcast [123 x i8]* %G.1 to i8* %1 = bitcast [123 x i8]* %G.1 to i984* %srcval = load i984* %1, align 1 %B.0 = getelementptr inbounds %T* %B, i32 0, i32 0 store i8 %tmp3, i8* %B.0, align 4 %B.1 = getelementptr inbounds %T* %B, i32 0, i32 1 %B.12 = bitcast [123 x i8]* %B.1 to i8* %2 = bitcast [123 x i8]* %B.1 to i984* store i984 %srcval, i984* %2, align 1 call void @bar(i8* %b) ret void } git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119682 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:20:47 +00:00
define void @test2() {
%A = alloca %T
%B = alloca %T
%a = bitcast %T* %A to i8*
%b = bitcast %T* %B to i8*
; CHECK: @test2
; %A alloca is deleted
Port the global copy optimization from the SROA pass to InstCombine. This optimization is really just replacing allocas wholesale with globals, there is no scalarization. The underlying motivation for this patch is to simplify the SROA pass and focus it on splitting and promoting allocas. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162271 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-21 08:39:44 +00:00
; CHECK-NEXT: alloca [124 x i8]
; CHECK-NEXT: getelementptr inbounds [124 x i8]*
fix a small oversight in the "eliminate memcpy from constant global" optimization. If the alloca that is "memcpy'd from constant" also has a memcpy from *it*, ignore it: it is a load. We now optimize the testcase to: define void @test2() { %B = alloca %T %a = bitcast %T* @G to i8* %b = bitcast %T* %B to i8* call void @llvm.memcpy.p0i8.p0i8.i64(i8* %b, i8* %a, i64 124, i32 4, i1 false) call void @bar(i8* %b) ret void } previously we would generate: define void @test() { %B = alloca %T %b = bitcast %T* %B to i8* %G.0 = getelementptr inbounds %T* @G, i32 0, i32 0 %tmp3 = load i8* %G.0, align 4 %G.1 = getelementptr inbounds %T* @G, i32 0, i32 1 %G.15 = bitcast [123 x i8]* %G.1 to i8* %1 = bitcast [123 x i8]* %G.1 to i984* %srcval = load i984* %1, align 1 %B.0 = getelementptr inbounds %T* %B, i32 0, i32 0 store i8 %tmp3, i8* %B.0, align 4 %B.1 = getelementptr inbounds %T* %B, i32 0, i32 1 %B.12 = bitcast [123 x i8]* %B.1 to i8* %2 = bitcast [123 x i8]* %B.1 to i984* store i984 %srcval, i984* %2, align 1 call void @bar(i8* %b) ret void } git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119682 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:20:47 +00:00
; use @G instead of %A
Port the global copy optimization from the SROA pass to InstCombine. This optimization is really just replacing allocas wholesale with globals, there is no scalarization. The underlying motivation for this patch is to simplify the SROA pass and focus it on splitting and promoting allocas. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162271 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-21 08:39:44 +00:00
; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.*}}, i8* getelementptr inbounds (%T* @G, i64 0, i32 0)
fix a small oversight in the "eliminate memcpy from constant global" optimization. If the alloca that is "memcpy'd from constant" also has a memcpy from *it*, ignore it: it is a load. We now optimize the testcase to: define void @test2() { %B = alloca %T %a = bitcast %T* @G to i8* %b = bitcast %T* %B to i8* call void @llvm.memcpy.p0i8.p0i8.i64(i8* %b, i8* %a, i64 124, i32 4, i1 false) call void @bar(i8* %b) ret void } previously we would generate: define void @test() { %B = alloca %T %b = bitcast %T* %B to i8* %G.0 = getelementptr inbounds %T* @G, i32 0, i32 0 %tmp3 = load i8* %G.0, align 4 %G.1 = getelementptr inbounds %T* @G, i32 0, i32 1 %G.15 = bitcast [123 x i8]* %G.1 to i8* %1 = bitcast [123 x i8]* %G.1 to i984* %srcval = load i984* %1, align 1 %B.0 = getelementptr inbounds %T* %B, i32 0, i32 0 store i8 %tmp3, i8* %B.0, align 4 %B.1 = getelementptr inbounds %T* %B, i32 0, i32 1 %B.12 = bitcast [123 x i8]* %B.1 to i8* %2 = bitcast [123 x i8]* %B.1 to i984* store i984 %srcval, i984* %2, align 1 call void @bar(i8* %b) ret void } git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119682 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:20:47 +00:00
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%T* @G to i8*), i64 124, i32 4, i1 false)
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %b, i8* %a, i64 124, i32 4, i1 false)
call void @bar(i8* %b)
ret void
}
declare void @bar(i8*)
enhance the "alloca is just a memcpy from constant global" to ignore calls that obviously can't modify the alloca because they are readonly/readnone. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119683 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:26:49 +00:00
;; Should be able to eliminate the alloca.
define void @test3() {
%A = alloca %T
%a = bitcast %T* %A to i8*
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%T* @G to i8*), i64 124, i32 4, i1 false)
call void @bar(i8* %a) readonly
; CHECK: @test3
Port the global copy optimization from the SROA pass to InstCombine. This optimization is really just replacing allocas wholesale with globals, there is no scalarization. The underlying motivation for this patch is to simplify the SROA pass and focus it on splitting and promoting allocas. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162271 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-21 08:39:44 +00:00
; CHECK-NEXT: call void @bar(i8* getelementptr inbounds (%T* @G, i64 0, i32 0))
enhance the "alloca is just a memcpy from constant global" to ignore calls that obviously can't modify the alloca because they are readonly/readnone. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119683 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:26:49 +00:00
ret void
}
allow eliminating an alloca that is just copied from an constant global if it is passed as a byval argument. The byval argument will just be a read, so it is safe to read from the original global instead. This allows us to promote away the %agg.tmp alloca in PR8582 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119686 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:41:51 +00:00
define void @test4() {
%A = alloca %T
%a = bitcast %T* %A to i8*
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%T* @G to i8*), i64 124, i32 4, i1 false)
call void @baz(i8* byval %a)
; CHECK: @test4
Port the global copy optimization from the SROA pass to InstCombine. This optimization is really just replacing allocas wholesale with globals, there is no scalarization. The underlying motivation for this patch is to simplify the SROA pass and focus it on splitting and promoting allocas. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162271 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-21 08:39:44 +00:00
; CHECK-NEXT: call void @baz(i8* byval getelementptr inbounds (%T* @G, i64 0, i32 0))
allow eliminating an alloca that is just copied from an constant global if it is passed as a byval argument. The byval argument will just be a read, so it is safe to read from the original global instead. This allows us to promote away the %agg.tmp alloca in PR8582 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119686 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:41:51 +00:00
ret void
}
Teach one piece of scalarrepl to handle lifetime markers. When transforming an alloca that only holds a copy of a global and we're going to replace the users of the alloca with that global, just nuke the lifetime intrinsics. Part of PR10121. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133905 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-27 05:40:02 +00:00
declare void @llvm.lifetime.start(i64, i8*)
define void @test5() {
%A = alloca %T
%a = bitcast %T* %A to i8*
call void @llvm.lifetime.start(i64 -1, i8* %a)
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%T* @G to i8*), i64 124, i32 4, i1 false)
call void @baz(i8* byval %a)
; CHECK: @test5
Port the global copy optimization from the SROA pass to InstCombine. This optimization is really just replacing allocas wholesale with globals, there is no scalarization. The underlying motivation for this patch is to simplify the SROA pass and focus it on splitting and promoting allocas. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162271 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-21 08:39:44 +00:00
; CHECK-NEXT: call void @baz(i8* byval getelementptr inbounds (%T* @G, i64 0, i32 0))
Teach one piece of scalarrepl to handle lifetime markers. When transforming an alloca that only holds a copy of a global and we're going to replace the users of the alloca with that global, just nuke the lifetime intrinsics. Part of PR10121. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133905 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-27 05:40:02 +00:00
ret void
}
allow eliminating an alloca that is just copied from an constant global if it is passed as a byval argument. The byval argument will just be a read, so it is safe to read from the original global instead. This allows us to promote away the %agg.tmp alloca in PR8582 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119686 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18 06:41:51 +00:00
declare void @baz(i8* byval)
Do not eliminate allocas whose alignment exceeds that of the copied-in constant, as a subsequent user may rely on over alignment. Fixes PR12885. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@157134 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-19 22:52:10 +00:00
define void @test6() {
%A = alloca %U, align 16
%a = bitcast %U* %A to i8*
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast ([2 x %U]* @H to i8*), i64 20, i32 16, i1 false)
call void @bar(i8* %a) readonly
; CHECK: @test6
Port the global copy optimization from the SROA pass to InstCombine. This optimization is really just replacing allocas wholesale with globals, there is no scalarization. The underlying motivation for this patch is to simplify the SROA pass and focus it on splitting and promoting allocas. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162271 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-21 08:39:44 +00:00
; CHECK-NEXT: call void @bar(i8* bitcast ([2 x %U]* @H to i8*))
Do not eliminate allocas whose alignment exceeds that of the copied-in constant, as a subsequent user may rely on over alignment. Fixes PR12885. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@157134 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-19 22:52:10 +00:00
ret void
}
define void @test7() {
%A = alloca %U, align 16
%a = bitcast %U* %A to i8*
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%U* getelementptr ([2 x %U]* @H, i64 0, i32 0) to i8*), i64 20, i32 4, i1 false)
call void @bar(i8* %a) readonly
; CHECK: @test7
Port the global copy optimization from the SROA pass to InstCombine. This optimization is really just replacing allocas wholesale with globals, there is no scalarization. The underlying motivation for this patch is to simplify the SROA pass and focus it on splitting and promoting allocas. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162271 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-21 08:39:44 +00:00
; CHECK-NEXT: call void @bar(i8* bitcast ([2 x %U]* @H to i8*))
Do not eliminate allocas whose alignment exceeds that of the copied-in constant, as a subsequent user may rely on over alignment. Fixes PR12885. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@157134 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-19 22:52:10 +00:00
ret void
}
define void @test8() {
%A = alloca %U, align 16
%a = bitcast %U* %A to i8*
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%U* getelementptr ([2 x %U]* @H, i64 0, i32 1) to i8*), i64 20, i32 4, i1 false)
call void @bar(i8* %a) readonly
; CHECK: @test8
; CHECK: llvm.memcpy
; CHECK: bar
ret void
}
Get rid of the getPointeeAlignment helper function from InstCombineLoadStoreAlloca.cpp, which had many issues. (At least two bugs were noted on llvm-commits, and it was overly conservative.) Instead, use getOrEnforceKnownAlignment. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@168629 91177308-0d34-0410-b5e6-96231b3b80d8
2012-11-26 23:04:53 +00:00
define void @test9() {
%A = alloca %U, align 4
%a = bitcast %U* %A to i8*
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%U* getelementptr ([2 x %U]* @H, i64 0, i32 1) to i8*), i64 20, i32 4, i1 false)
call void @bar(i8* %a) readonly
; CHECK: @test9
; CHECK-NEXT: call void @bar(i8* bitcast (%U* getelementptr inbounds ([2 x %U]* @H, i64 0, i64 1) to i8*))
ret void
}
Reference in New Issue Copy Permalink