declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
-declare void @llvm.memset.i64(i8*, i8, i64, i32)
-declare void @llvm.memcpy.i64(i8*, i8*, i64, i32)
declare i8* @llvm.init.trampoline(i8*, i8*, i8*)
define void @test1(i32* %Q, i32* %P) {
define void @test4(i32* %Q) {
%a = load i32* %Q
- volatile store i32 %a, i32* %Q
+ store volatile i32 %a, i32* %Q
ret void
; CHECK: @test4
; CHECK-NEXT: load i32
-; CHECK-NEXT: volatile store
+; CHECK-NEXT: store volatile
; CHECK-NEXT: ret void
}
define void @test5(i32* %Q) {
- %a = volatile load i32* %Q
+ %a = load volatile i32* %Q
store i32 %a, i32* %Q
ret void
; CHECK: @test5
-; CHECK-NEXT: volatile load
+; CHECK-NEXT: load volatile
; CHECK-NEXT: ret void
}
; alias).
define void @test6(i32 *%p, i8 *%q) {
store i32 10, i32* %p, align 4 ;; dead.
- call void @llvm.memset.i64(i8* %q, i8 42, i64 900, i32 1)
+ call void @llvm.memset.p0i8.i64(i8* %q, i8 42, i64 900, i32 1, i1 false)
store i32 30, i32* %p, align 4
ret void
; CHECK: @test6
; alias).
define void @test7(i32 *%p, i8 *%q, i8* noalias %r) {
store i32 10, i32* %p, align 4 ;; dead.
- call void @llvm.memcpy.i64(i8* %q, i8* %r, i64 900, i32 1)
+ call void @llvm.memcpy.p0i8.p0i8.i64(i8* %q, i8* %r, i64 900, i32 1, i1 false)
store i32 30, i32* %p, align 4
ret void
; CHECK: @test7
}
declare noalias i8* @malloc(i32)
-
+declare noalias i8* @calloc(i32, i32)
define void @test14(i32* %Q) {
;; Fully dead overwrite of memcpy.
define void @test15(i8* %P, i8* %Q) nounwind ssp {
- tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 12, i32 1)
- tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 12, i32 1)
+ tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
+ tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
ret void
; CHECK: @test15
; CHECK-NEXT: call void @llvm.memcpy
;; Full overwrite of smaller memcpy.
define void @test16(i8* %P, i8* %Q) nounwind ssp {
- tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 8, i32 1)
- tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 12, i32 1)
+ tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 8, i32 1, i1 false)
+ tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
ret void
; CHECK: @test16
; CHECK-NEXT: call void @llvm.memcpy
}
;; Overwrite of memset by memcpy.
-define void @test17(i8* %P, i8* %Q) nounwind ssp {
- tail call void @llvm.memset.i64(i8* %P, i8 42, i64 8, i32 1)
- tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 12, i32 1)
+define void @test17(i8* %P, i8* noalias %Q) nounwind ssp {
+ tail call void @llvm.memset.p0i8.i64(i8* %P, i8 42, i64 8, i32 1, i1 false)
+ tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
ret void
; CHECK: @test17
; CHECK-NEXT: call void @llvm.memcpy
; CHECK-NEXT: call void @llvm.memcpy
; CHECK-NEXT: ret
}
+
+; PR8728
+; Do not delete instruction where possible situation is:
+; A = B
+; A = A
+define void @test18(i8* %P, i8* %Q, i8* %R) nounwind ssp {
+ tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
+ tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %R, i64 12, i32 1, i1 false)
+ ret void
+; CHECK: @test18
+; CHECK-NEXT: call void @llvm.memcpy
+; CHECK-NEXT: call void @llvm.memcpy
+; CHECK-NEXT: ret
+}
+
+
+; The store here is not dead because the byval call reads it.
+declare void @test19f({i32}* byval align 4 %P)
+
+define void @test19({i32} * nocapture byval align 4 %arg5) nounwind ssp {
+bb:
+ %tmp7 = getelementptr inbounds {i32}* %arg5, i32 0, i32 0
+ store i32 912, i32* %tmp7
+ call void @test19f({i32}* byval align 4 %arg5)
+ ret void
+
+; CHECK: @test19(
+; CHECK: store i32 912
+; CHECK: call void @test19f
+}
+
+define void @test20() {
+ %m = call i8* @malloc(i32 24)
+ store i8 0, i8* %m
+ ret void
+}
+; CHECK: @test20
+; CHECK-NEXT: ret void
+
+; CHECK: @test21
+define void @test21() {
+ %m = call i8* @calloc(i32 9, i32 7)
+ store i8 0, i8* %m
+; CHECK-NEXT: ret void
+ ret void
+}
+
+; CHECK: @test22(
+define void @test22(i1 %i, i32 %k, i32 %m) nounwind {
+ %k.addr = alloca i32
+ %m.addr = alloca i32
+ %k.addr.m.addr = select i1 %i, i32* %k.addr, i32* %m.addr
+ store i32 0, i32* %k.addr.m.addr, align 4
+; CHECK-NEXT: ret void
+ ret void
+}
+
+; PR13547
+; CHECK: @test23
+; CHECK: store i8 97
+; CHECK: store i8 0
+declare noalias i8* @strdup(i8* nocapture) nounwind
+define noalias i8* @test23() nounwind uwtable ssp {
+ %x = alloca [2 x i8], align 1
+ %arrayidx = getelementptr inbounds [2 x i8]* %x, i64 0, i64 0
+ store i8 97, i8* %arrayidx, align 1
+ %arrayidx1 = getelementptr inbounds [2 x i8]* %x, i64 0, i64 1
+ store i8 0, i8* %arrayidx1, align 1
+ %call = call i8* @strdup(i8* %arrayidx) nounwind
+ ret i8* %call
+}
+
+; Make sure same sized store to later element is deleted
+; CHECK: @test24
+; CHECK-NOT: store i32 0
+; CHECK-NOT: store i32 0
+; CHECK: store i32 %b
+; CHECK: store i32 %c
+; CHECK: ret void
+define void @test24([2 x i32]* %a, i32 %b, i32 %c) nounwind {
+ %1 = getelementptr inbounds [2 x i32]* %a, i64 0, i64 0
+ store i32 0, i32* %1, align 4
+ %2 = getelementptr inbounds [2 x i32]* %a, i64 0, i64 1
+ store i32 0, i32* %2, align 4
+ %3 = getelementptr inbounds [2 x i32]* %a, i64 0, i64 0
+ store i32 %b, i32* %3, align 4
+ %4 = getelementptr inbounds [2 x i32]* %a, i64 0, i64 1
+ store i32 %c, i32* %4, align 4
+ ret void
+}
+
+; Check another case like PR13547 where strdup is not like malloc.
+; CHECK: @test25
+; CHECK: load i8
+; CHECK: store i8 0
+; CHECK: store i8 %tmp
+define i8* @test25(i8* %p) nounwind {
+ %p.4 = getelementptr i8* %p, i64 4
+ %tmp = load i8* %p.4, align 1
+ store i8 0, i8* %p.4, align 1
+ %q = call i8* @strdup(i8* %p) nounwind optsize
+ store i8 %tmp, i8* %p.4, align 1
+ ret i8* %q
+}