100 lines
3.7 KiB
LLVM
100 lines
3.7 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
|
|
; RUN: opt < %s -passes=dse -S | FileCheck %s
|
|
|
|
@BUFFER = external local_unnamed_addr global [0 x i8], align 1
|
|
|
|
define void @ArrayTestFullyOverlapping(i64 %0) {
|
|
;
|
|
; The DSE pass will try to kill the store of size i32 using the store of
|
|
; size i64 because they fully overlap, in fact:
|
|
;
|
|
; - they use the same base pointer (in SCEV style '@BUFFER + %0')
|
|
; - the offset between the two stores is 32 bits
|
|
; - the size of the earlier store is 32 bits
|
|
; - the size of the later store is 64 bits
|
|
;
|
|
; CHECK-LABEL: @ArrayTestFullyOverlapping(
|
|
; CHECK-NEXT: [[TMP2:%.*]] = add i64 [[TMP0:%.*]], -8
|
|
; CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds [0 x i8], ptr @BUFFER, i64 0, i64 [[TMP2]]
|
|
; CHECK-NEXT: store i64 0, ptr [[TMP3]], align 4
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%2 = add i64 %0, -8
|
|
%3 = getelementptr inbounds [0 x i8], ptr @BUFFER, i64 0, i64 %2
|
|
%4 = add i64 %0, -4
|
|
%5 = getelementptr inbounds [0 x i8], ptr @BUFFER, i64 0, i64 %4
|
|
store i32 1, ptr %5
|
|
store i64 0, ptr %3
|
|
ret void
|
|
}
|
|
|
|
define void @VectorTestFullyOverlapping(ptr %arg, i32 %i) {
|
|
; CHECK-LABEL: @VectorTestFullyOverlapping(
|
|
; CHECK-NEXT: bb:
|
|
; CHECK-NEXT: [[I2:%.*]] = zext i32 [[I:%.*]] to i64
|
|
; CHECK-NEXT: [[I3:%.*]] = getelementptr inbounds float, ptr [[ARG:%.*]], i64 [[I2]]
|
|
; CHECK-NEXT: store <2 x float> zeroinitializer, ptr [[I3]], align 16
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
bb:
|
|
%i7 = add nuw nsw i32 %i, 1
|
|
%i8 = zext i32 %i7 to i64
|
|
%i9 = getelementptr inbounds float, ptr %arg, i64 %i8
|
|
store float 0.0, ptr %i9, align 4
|
|
%i2 = zext i32 %i to i64
|
|
%i3 = getelementptr inbounds float, ptr %arg, i64 %i2
|
|
store <2 x float> <float 0.0, float 0.0>, ptr %i3, align 16
|
|
ret void
|
|
}
|
|
|
|
define void @ArrayTestPartiallyOverlapping(i64 %0) {
|
|
;
|
|
; The DSE pass will not kill the store because the overlap is partial
|
|
; and won't fully clobber the i32 store.
|
|
;
|
|
; CHECK-LABEL: @ArrayTestPartiallyOverlapping(
|
|
; CHECK-NEXT: [[TMP2:%.*]] = add i64 [[TMP0:%.*]], 10
|
|
; CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds [0 x i8], ptr @BUFFER, i64 0, i64 [[TMP2]]
|
|
; CHECK-NEXT: [[TMP5:%.*]] = add i64 [[TMP0]], 15
|
|
; CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds [0 x i8], ptr @BUFFER, i64 0, i64 [[TMP5]]
|
|
; CHECK-NEXT: store i32 1, ptr [[TMP6]], align 4
|
|
; CHECK-NEXT: store i64 0, ptr [[TMP3]], align 4
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%2 = add i64 %0, 10
|
|
%3 = getelementptr inbounds [0 x i8], ptr @BUFFER, i64 0, i64 %2
|
|
%4 = add i64 %0, 15
|
|
%5 = getelementptr inbounds [0 x i8], ptr @BUFFER, i64 0, i64 %4
|
|
store i32 1, ptr %5
|
|
store i64 0, ptr %3
|
|
ret void
|
|
}
|
|
|
|
define void @VectorTestPartiallyOverlapping(ptr %arg, i32 %i) {
|
|
;
|
|
; The DSE pass will not kill the store because the overlap is partial
|
|
; and won't fully clobber the original store.
|
|
;
|
|
; CHECK-LABEL: @VectorTestPartiallyOverlapping(
|
|
; CHECK-NEXT: bb:
|
|
; CHECK-NEXT: [[I2:%.*]] = zext i32 [[I:%.*]] to i64
|
|
; CHECK-NEXT: [[I3:%.*]] = getelementptr inbounds float, ptr [[ARG:%.*]], i64 [[I2]]
|
|
; CHECK-NEXT: store <2 x float> <float 1.000000e+00, float 1.000000e+00>, ptr [[I3]], align 16
|
|
; CHECK-NEXT: [[I5:%.*]] = add nuw nsw i32 [[I]], 1
|
|
; CHECK-NEXT: [[I6:%.*]] = zext i32 [[I5]] to i64
|
|
; CHECK-NEXT: [[I7:%.*]] = getelementptr inbounds float, ptr [[ARG]], i64 [[I6]]
|
|
; CHECK-NEXT: store <2 x float> zeroinitializer, ptr [[I7]], align 16
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
bb:
|
|
%i2 = zext i32 %i to i64
|
|
%i3 = getelementptr inbounds float, ptr %arg, i64 %i2
|
|
store <2 x float> <float 1.000000e+00, float 1.000000e+00>, ptr %i3, align 16
|
|
%i5 = add nuw nsw i32 %i, 1
|
|
%i6 = zext i32 %i5 to i64
|
|
%i7 = getelementptr inbounds float, ptr %arg, i64 %i6
|
|
store <2 x float> <float 0.0, float 0.0>, ptr %i7, align 16
|
|
ret void
|
|
}
|
|
|