RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-01 08:23:21 +01:00
Code Issues Projects Releases Wiki Activity
3cda8bc78e
llvm-mirror/test/Transforms/ScalarRepl/memcpy-from-global.ll

97 lines
6.0 KiB
LLVM
Raw Normal View History

filecheckize llvm-svn: 119681
2010-11-18 07:16:43 +01:00
; RUN: opt < %s -scalarrepl -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 llvm-svn: 85900
2009-11-03 16:29:06 +01:00
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 llvm-svn: 36428
2007-04-25 08:29:34 +02:00
@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 llvm-svn: 119681
2010-11-18 07:16:43 +01:00
define float @test1(i32 %hash, float %x, float %y, float %z, float %w) {
new testcase, should be able to eliminate the alloca and memcpy llvm-svn: 36428
2007-04-25 08:29:34 +02:00
entry:
%lookupTable = alloca [128 x float], align 16 ; <[128 x float]*> [#uses=5]
%lookupTable1 = bitcast [128 x float]* %lookupTable to i8* ; <i8*> [#uses=1]
call void @llvm.memcpy.i32( i8* %lookupTable1, i8* bitcast ([128 x float]* @C.0.1248 to i8*), i32 512, i32 16 )
filecheckize llvm-svn: 119681
2010-11-18 07:16:43 +01:00
; CHECK: @test1
; CHECK-NOT: alloca
; CHECK-NOT: call{{.*}}@llvm.memcpy
; CHECK: %lookupTable1 = bitcast [128 x float]* @C.0.1248 to i8*
; CHECK-NOT: call{{.*}}@llvm.memcpy
new testcase, should be able to eliminate the alloca and memcpy llvm-svn: 36428
2007-04-25 08:29:34 +02:00
%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 llvm-svn: 72897
2009-06-05 00:49:04 +02:00
%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 llvm-svn: 36428
2007-04-25 08:29:34 +02:00
%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 llvm-svn: 72897
2009-06-05 00:49:04 +02:00
%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 llvm-svn: 36428
2007-04-25 08:29:34 +02:00
%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 llvm-svn: 72897
2009-06-05 00:49:04 +02:00
%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 llvm-svn: 36428
2007-04-25 08:29:34 +02:00
%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 llvm-svn: 72897
2009-06-05 00:49:04 +02:00
%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 llvm-svn: 36428
2007-04-25 08:29:34 +02:00
ret float %tmp43
}
declare void @llvm.memcpy.i32(i8*, i8*, 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 } llvm-svn: 119682
2010-11-18 07:20:47 +01:00
declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
%T = type { i8, [123 x i8] }
@G = constant %T {i8 1, [123 x i8] zeroinitializer }
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
; CHECK-NEXT: %B = alloca %T
; use @G instead of %A
; CHECK-NEXT: %a = bitcast %T* @G 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 @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. llvm-svn: 119683
2010-11-18 07:26:49 +01:00
;; 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
; CHECK-NEXT: %a = bitcast %T* @G to i8*
; CHECK-NEXT: call void @bar(i8* %a)
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 llvm-svn: 119686
2010-11-18 07:41:51 +01:00
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
; CHECK-NEXT: %a = bitcast %T* @G to i8*
; CHECK-NEXT: call void @baz(i8* byval %a)
ret void
}
declare void @baz(i8* byval)
Reference in New Issue Copy Permalink