60a32157a5
Previously left these behind due to the required instruction renumbering, drop them now. This more accurately represents opaque pointer input IR. Also drop duplicate opaque pointer check lines in one SROA test.
227 lines
7.1 KiB
LLVM
227 lines
7.1 KiB
LLVM
; RUN: opt < %s -passes=sroa -S | FileCheck %s
|
|
; RUN: opt -passes='debugify,function(sroa)' -S < %s | FileCheck %s -check-prefix DEBUGLOC
|
|
|
|
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-n8:16:32:64"
|
|
|
|
declare void @llvm.memcpy.p0.p0.i32(ptr, ptr, i32, i1)
|
|
|
|
define void @test1(ptr %a, ptr %b) {
|
|
; CHECK-LABEL: @test1(
|
|
; CHECK: %[[gep_a0:.*]] = getelementptr { i8, i8 }, ptr %a, i32 0, i32 0
|
|
; CHECK: %[[gep_b0:.*]] = getelementptr { i8, i8 }, ptr %b, i32 0, i32 0
|
|
; CHECK: %[[a0:.*]] = load i8, ptr %[[gep_a0]], align 16
|
|
; CHECK: %[[gep_a1:.*]] = getelementptr inbounds i8, ptr %[[gep_a0]], i64 1
|
|
; CHECK: %[[a1:.*]] = load i8, ptr %[[gep_a1]], align 1
|
|
; CHECK: store i8 %[[a0]], ptr %[[gep_b0]], align 16
|
|
; CHECK: %[[gep_b1:.*]] = getelementptr inbounds i8, ptr %[[gep_b0]], i64 1
|
|
; CHECK: store i8 %[[a1]], ptr %[[gep_b1]], align 1
|
|
; CHECK: ret void
|
|
|
|
entry:
|
|
%alloca = alloca { i8, i8 }, align 16
|
|
%gep_a = getelementptr { i8, i8 }, ptr %a, i32 0, i32 0
|
|
%gep_alloca = getelementptr { i8, i8 }, ptr %alloca, i32 0, i32 0
|
|
%gep_b = getelementptr { i8, i8 }, ptr %b, i32 0, i32 0
|
|
|
|
store i8 420, ptr %gep_alloca, align 16
|
|
|
|
call void @llvm.memcpy.p0.p0.i32(ptr align 16 %gep_alloca, ptr align 16 %gep_a, i32 2, i1 false)
|
|
call void @llvm.memcpy.p0.p0.i32(ptr align 16 %gep_b, ptr align 16 %gep_alloca, i32 2, i1 false)
|
|
ret void
|
|
}
|
|
|
|
define void @test2() {
|
|
; CHECK-LABEL: @test2(
|
|
; CHECK: alloca i16
|
|
; CHECK: load i8, ptr %{{.*}}
|
|
; CHECK: store i8 42, ptr %{{.*}}
|
|
; CHECK: ret void
|
|
|
|
; Check that when sroa rewrites the alloca partition
|
|
; it preserves the original DebugLocation.
|
|
; DEBUGLOC-LABEL: @test2(
|
|
; DEBUGLOC: {{.*}} = alloca {{.*}} !dbg ![[DbgLoc:[0-9]+]]
|
|
; DEBUGLOC-LABEL: }
|
|
;
|
|
; DEBUGLOC: ![[DbgLoc]] = !DILocation(line: 9,
|
|
|
|
entry:
|
|
%a = alloca { i8, i8, i8, i8 }, align 2 ; "line 9" to -debugify
|
|
%gep1 = getelementptr { i8, i8, i8, i8 }, ptr %a, i32 0, i32 1
|
|
store volatile i16 0, ptr %gep1
|
|
%gep2 = getelementptr { i8, i8, i8, i8 }, ptr %a, i32 0, i32 2
|
|
%result = load i8, ptr %gep2
|
|
store i8 42, ptr %gep2
|
|
ret void
|
|
}
|
|
|
|
define void @PR13920(ptr %a, ptr %b) {
|
|
; Test that alignments on memcpy intrinsics get propagated to loads and stores.
|
|
; CHECK-LABEL: @PR13920(
|
|
; CHECK: load <2 x i64>, ptr %a, align 2
|
|
; CHECK: store <2 x i64> {{.*}}, ptr {{.*}}, align 2
|
|
; CHECK: ret void
|
|
|
|
entry:
|
|
%aa = alloca <2 x i64>, align 16
|
|
call void @llvm.memcpy.p0.p0.i32(ptr align 2 %aa, ptr align 2 %a, i32 16, i1 false)
|
|
call void @llvm.memcpy.p0.p0.i32(ptr align 2 %b, ptr align 2 %aa, i32 16, i1 false)
|
|
ret void
|
|
}
|
|
|
|
define void @test3(ptr %x) {
|
|
; Test that when we promote an alloca to a type with lower ABI alignment, we
|
|
; provide the needed explicit alignment that code using the alloca may be
|
|
; expecting. However, also check that any offset within an alloca can in turn
|
|
; reduce the alignment.
|
|
; CHECK-LABEL: @test3(
|
|
; CHECK: alloca [22 x i8], align 8
|
|
; CHECK: alloca [18 x i8], align 2
|
|
; CHECK: ret void
|
|
|
|
entry:
|
|
%a = alloca { ptr, ptr, ptr }
|
|
%b = alloca { ptr, ptr, ptr }
|
|
call void @llvm.memcpy.p0.p0.i32(ptr align 8 %a, ptr align 8 %x, i32 22, i1 false)
|
|
%b_gep = getelementptr i8, ptr %b, i32 6
|
|
call void @llvm.memcpy.p0.p0.i32(ptr align 2 %b_gep, ptr align 2 %x, i32 18, i1 false)
|
|
ret void
|
|
}
|
|
|
|
define void @test5() {
|
|
; Test that we preserve underaligned loads and stores when splitting. The use
|
|
; of volatile in this test case is just to force the loads and stores to not be
|
|
; split or promoted out of existence.
|
|
;
|
|
; CHECK-LABEL: @test5(
|
|
; CHECK: alloca [9 x i8]
|
|
; CHECK: alloca [9 x i8]
|
|
; CHECK: store volatile double 0.0{{.*}}, ptr %{{.*}}, align 1
|
|
; CHECK: load volatile i16, ptr %{{.*}}, align 1
|
|
; CHECK: load double, ptr %{{.*}}, align 1
|
|
; CHECK: store volatile double %{{.*}}, ptr %{{.*}}, align 1
|
|
; CHECK: load volatile i16, ptr %{{.*}}, align 1
|
|
; CHECK: ret void
|
|
|
|
entry:
|
|
%a = alloca [18 x i8]
|
|
store volatile double 0.0, ptr %a, align 1
|
|
%weird_gep1 = getelementptr inbounds [18 x i8], ptr %a, i32 0, i32 7
|
|
%weird_load1 = load volatile i16, ptr %weird_gep1, align 1
|
|
|
|
%raw2 = getelementptr inbounds [18 x i8], ptr %a, i32 0, i32 9
|
|
%d1 = load double, ptr %a, align 1
|
|
store volatile double %d1, ptr %raw2, align 1
|
|
%weird_gep2 = getelementptr inbounds [18 x i8], ptr %a, i32 0, i32 16
|
|
%weird_load2 = load volatile i16, ptr %weird_gep2, align 1
|
|
|
|
ret void
|
|
}
|
|
|
|
define void @test6() {
|
|
; We should set the alignment on all load and store operations; make sure
|
|
; we choose an appropriate alignment.
|
|
; CHECK-LABEL: @test6(
|
|
; CHECK: alloca double, align 8{{$}}
|
|
; CHECK: alloca double, align 8{{$}}
|
|
; CHECK: store{{.*}}, align 8
|
|
; CHECK: load{{.*}}, align 8
|
|
; CHECK: store{{.*}}, align 8
|
|
; CHECK-NOT: align
|
|
; CHECK: ret void
|
|
|
|
entry:
|
|
%a = alloca [16 x i8]
|
|
store volatile double 0.0, ptr %a, align 1
|
|
|
|
%raw2 = getelementptr inbounds [16 x i8], ptr %a, i32 0, i32 8
|
|
%val = load double, ptr %a, align 1
|
|
store volatile double %val, ptr %raw2, align 1
|
|
|
|
ret void
|
|
}
|
|
|
|
define void @test7(ptr %out) {
|
|
; Test that we properly compute the destination alignment when rewriting
|
|
; memcpys as direct loads or stores.
|
|
; CHECK-LABEL: @test7(
|
|
; CHECK-NOT: alloca
|
|
|
|
entry:
|
|
%a = alloca [16 x i8]
|
|
%raw2 = getelementptr inbounds [16 x i8], ptr %a, i32 0, i32 8
|
|
|
|
call void @llvm.memcpy.p0.p0.i32(ptr %a, ptr %out, i32 16, i1 false)
|
|
; CHECK: %[[val2:.*]] = load double, ptr %{{.*}}, align 1
|
|
; CHECK: %[[val1:.*]] = load double, ptr %{{.*}}, align 1
|
|
|
|
%val1 = load double, ptr %raw2, align 1
|
|
%val2 = load double, ptr %a, align 1
|
|
|
|
store double %val1, ptr %a, align 1
|
|
store double %val2, ptr %raw2, align 1
|
|
|
|
call void @llvm.memcpy.p0.p0.i32(ptr %out, ptr %a, i32 16, i1 false)
|
|
; CHECK: store double %[[val1]], ptr %{{.*}}, align 1
|
|
; CHECK: store double %[[val2]], ptr %{{.*}}, align 1
|
|
|
|
ret void
|
|
; CHECK: ret void
|
|
}
|
|
|
|
define void @test8() {
|
|
; CHECK-LABEL: @test8(
|
|
; CHECK: load i32, {{.*}}, align 1
|
|
; CHECK: load i32, {{.*}}, align 1
|
|
; CHECK: load i32, {{.*}}, align 1
|
|
; CHECK: load i32, {{.*}}, align 1
|
|
; CHECK: load i32, {{.*}}, align 1
|
|
|
|
%ptr = alloca [5 x i32], align 1
|
|
call void @populate(ptr %ptr)
|
|
%val = load [5 x i32], ptr %ptr, align 1
|
|
ret void
|
|
}
|
|
|
|
define void @test9() {
|
|
; CHECK-LABEL: @test9(
|
|
; CHECK: load i32, {{.*}}, align 8
|
|
; CHECK: load i32, {{.*}}, align 4
|
|
; CHECK: load i32, {{.*}}, align 8
|
|
; CHECK: load i32, {{.*}}, align 4
|
|
; CHECK: load i32, {{.*}}, align 8
|
|
|
|
%ptr = alloca [5 x i32], align 8
|
|
call void @populate(ptr %ptr)
|
|
%val = load [5 x i32], ptr %ptr, align 8
|
|
ret void
|
|
}
|
|
|
|
define void @test10() {
|
|
; CHECK-LABEL: @test10(
|
|
; CHECK: load i32, {{.*}}, align 2
|
|
; CHECK: load i8, {{.*}}, align 2
|
|
; CHECK: load i8, {{.*}}, align 1
|
|
; CHECK: load i8, {{.*}}, align 2
|
|
; CHECK: load i16, {{.*}}, align 2
|
|
|
|
%ptr = alloca {i32, i8, i8, {i8, i16}}, align 2
|
|
call void @populate(ptr %ptr)
|
|
%val = load {i32, i8, i8, {i8, i16}}, ptr %ptr, align 2
|
|
ret void
|
|
}
|
|
|
|
%struct = type { i32, i32 }
|
|
define dso_local i32 @pr45010(ptr %A) {
|
|
; CHECK-LABEL: @pr45010
|
|
; CHECK: load atomic volatile i32, {{.*}}, align 4
|
|
|
|
%B = alloca %struct, align 4
|
|
%1 = load i32, ptr %A, align 4
|
|
store atomic volatile i32 %1, ptr %B release, align 4
|
|
%x = load atomic volatile i32, ptr %B acquire, align 4
|
|
ret i32 %x
|
|
}
|
|
|
|
declare void @populate(ptr)
|