[InstCombine] Improve with.overflow intrinsic tests; NFC

- Do not use unnamed values in saddo tests
- Add tests for canonicalization of a constant arg0

Patch by Dan Robertson.

Differential Revision: https://reviews.llvm.org/D59476

llvm-svn: 356403
This commit is contained in:
Nikita Popov 2019-03-18 20:08:35 +00:00
parent 1d716acf76
commit c1d4fc8a62
2 changed files with 123 additions and 58 deletions

View File

@ -7,96 +7,107 @@ declare { i32, i1 } @llvm.sadd.with.overflow.i32(i32, i32)
declare { i8, i1 } @llvm.sadd.with.overflow.i8(i8, i8)
define { i32, i1 } @simple_fold(i32) {
define { i32, i1 } @simple_fold(i32 %x) {
; CHECK-LABEL: @simple_fold(
; CHECK-NEXT: [[TMP2:%.*]] = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[TMP0:%.*]], i32 20)
; CHECK-NEXT: ret { i32, i1 } [[TMP2]]
; CHECK-NEXT: [[TMP1:%.*]] = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[X:%.*]], i32 20)
; CHECK-NEXT: ret { i32, i1 } [[TMP1]]
;
%2 = add nsw i32 %0, 7
%3 = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 %2, i32 13)
ret { i32, i1 } %3
%a = add nsw i32 %x, 7
%b = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 %a, i32 13)
ret { i32, i1 } %b
}
define { i32, i1 } @fold_mixed_signs(i32) {
define { i32, i1 } @fold_mixed_signs(i32 %x) {
; CHECK-LABEL: @fold_mixed_signs(
; CHECK-NEXT: [[TMP2:%.*]] = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[TMP0:%.*]], i32 6)
; CHECK-NEXT: ret { i32, i1 } [[TMP2]]
; CHECK-NEXT: [[TMP1:%.*]] = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[X:%.*]], i32 6)
; CHECK-NEXT: ret { i32, i1 } [[TMP1]]
;
%2 = add nsw i32 %0, 13
%3 = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 %2, i32 -7)
ret { i32, i1 } %3
%a = add nsw i32 %x, 13
%b = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 %a, i32 -7)
ret { i32, i1 } %b
}
define { i8, i1 } @fold_on_constant_add_no_overflow(i8) {
define { i8, i1 } @fold_on_constant_add_no_overflow(i8 %x) {
; CHECK-LABEL: @fold_on_constant_add_no_overflow(
; CHECK-NEXT: [[TMP2:%.*]] = call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 [[TMP0:%.*]], i8 127)
; CHECK-NEXT: ret { i8, i1 } [[TMP2]]
; CHECK-NEXT: [[TMP1:%.*]] = call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 [[X:%.*]], i8 127)
; CHECK-NEXT: ret { i8, i1 } [[TMP1]]
;
%2 = add nsw i8 %0, 100
%3 = tail call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 %2, i8 27)
ret { i8, i1 } %3
%a = add nsw i8 %x, 100
%b = tail call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 %a, i8 27)
ret { i8, i1 } %b
}
define { i8, i1 } @no_fold_on_constant_add_overflow(i8) {
define { i8, i1 } @no_fold_on_constant_add_overflow(i8 %x) {
; CHECK-LABEL: @no_fold_on_constant_add_overflow(
; CHECK-NEXT: [[TMP2:%.*]] = add nsw i8 [[TMP0:%.*]], 100
; CHECK-NEXT: [[TMP3:%.*]] = tail call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 [[TMP2]], i8 28)
; CHECK-NEXT: ret { i8, i1 } [[TMP3]]
; CHECK-NEXT: [[A:%.*]] = add nsw i8 [[X:%.*]], 100
; CHECK-NEXT: [[B:%.*]] = tail call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 [[A]], i8 28)
; CHECK-NEXT: ret { i8, i1 } [[B]]
;
%2 = add nsw i8 %0, 100
%3 = tail call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 %2, i8 28)
ret { i8, i1 } %3
%a = add nsw i8 %x, 100
%b = tail call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 %a, i8 28)
ret { i8, i1 } %b
}
define { <2 x i32>, <2 x i1> } @fold_simple_splat_constant(<2 x i32>) {
define { <2 x i32>, <2 x i1> } @fold_simple_splat_constant(<2 x i32> %x) {
; CHECK-LABEL: @fold_simple_splat_constant(
; CHECK-NEXT: [[TMP2:%.*]] = call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> [[TMP0:%.*]], <2 x i32> <i32 42, i32 42>)
; CHECK-NEXT: ret { <2 x i32>, <2 x i1> } [[TMP2]]
; CHECK-NEXT: [[TMP1:%.*]] = call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> [[X:%.*]], <2 x i32> <i32 42, i32 42>)
; CHECK-NEXT: ret { <2 x i32>, <2 x i1> } [[TMP1]]
;
%2 = add nsw <2 x i32> %0, <i32 12, i32 12>
%3 = tail call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> %2, <2 x i32> <i32 30, i32 30>)
ret { <2 x i32>, <2 x i1> } %3
%a = add nsw <2 x i32> %x, <i32 12, i32 12>
%b = tail call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> %a, <2 x i32> <i32 30, i32 30>)
ret { <2 x i32>, <2 x i1> } %b
}
define { <2 x i32>, <2 x i1> } @no_fold_splat_undef_constant(<2 x i32>) {
define { <2 x i32>, <2 x i1> } @no_fold_splat_undef_constant(<2 x i32> %x) {
; CHECK-LABEL: @no_fold_splat_undef_constant(
; CHECK-NEXT: [[TMP2:%.*]] = add nsw <2 x i32> [[TMP0:%.*]], <i32 12, i32 undef>
; CHECK-NEXT: [[TMP3:%.*]] = tail call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> [[TMP2]], <2 x i32> <i32 30, i32 30>)
; CHECK-NEXT: ret { <2 x i32>, <2 x i1> } [[TMP3]]
; CHECK-NEXT: [[A:%.*]] = add nsw <2 x i32> [[X:%.*]], <i32 12, i32 undef>
; CHECK-NEXT: [[B:%.*]] = tail call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> [[A]], <2 x i32> <i32 30, i32 30>)
; CHECK-NEXT: ret { <2 x i32>, <2 x i1> } [[B]]
;
%2 = add nsw <2 x i32> %0, <i32 12, i32 undef>
%3 = tail call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> %2, <2 x i32> <i32 30, i32 30>)
ret { <2 x i32>, <2 x i1> } %3
%a = add nsw <2 x i32> %x, <i32 12, i32 undef>
%b = tail call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> %a, <2 x i32> <i32 30, i32 30>)
ret { <2 x i32>, <2 x i1> } %b
}
define { <2 x i32>, <2 x i1> } @no_fold_splat_not_constant(<2 x i32>, <2 x i32>) {
define { <2 x i32>, <2 x i1> } @no_fold_splat_not_constant(<2 x i32> %x, <2 x i32> %y) {
; CHECK-LABEL: @no_fold_splat_not_constant(
; CHECK-NEXT: [[TMP3:%.*]] = add nsw <2 x i32> [[TMP0:%.*]], [[TMP1:%.*]]
; CHECK-NEXT: [[TMP4:%.*]] = tail call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> [[TMP3]], <2 x i32> <i32 30, i32 30>)
; CHECK-NEXT: ret { <2 x i32>, <2 x i1> } [[TMP4]]
; CHECK-NEXT: [[A:%.*]] = add nsw <2 x i32> [[X:%.*]], [[Y:%.*]]
; CHECK-NEXT: [[B:%.*]] = tail call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> [[A]], <2 x i32> <i32 30, i32 30>)
; CHECK-NEXT: ret { <2 x i32>, <2 x i1> } [[B]]
;
%3 = add nsw <2 x i32> %0, %1
%4 = tail call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> %3, <2 x i32> <i32 30, i32 30>)
ret { <2 x i32>, <2 x i1> } %4
%a = add nsw <2 x i32> %x, %y
%b = tail call { <2 x i32>, <2 x i1> } @llvm.sadd.with.overflow.v2i32(<2 x i32> %a, <2 x i32> <i32 30, i32 30>)
ret { <2 x i32>, <2 x i1> } %b
}
define { i32, i1 } @fold_nuwnsw(i32) {
define { i32, i1 } @fold_nuwnsw(i32 %x) {
; CHECK-LABEL: @fold_nuwnsw(
; CHECK-NEXT: [[TMP2:%.*]] = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[TMP0:%.*]], i32 42)
; CHECK-NEXT: ret { i32, i1 } [[TMP2]]
; CHECK-NEXT: [[TMP1:%.*]] = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[X:%.*]], i32 42)
; CHECK-NEXT: ret { i32, i1 } [[TMP1]]
;
%2 = add nuw nsw i32 %0, 12
%3 = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 %2, i32 30)
ret { i32, i1 } %3
%a = add nuw nsw i32 %x, 12
%b = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 %a, i32 30)
ret { i32, i1 } %b
}
define { i32, i1 } @no_fold_nuw(i32) {
define { i32, i1 } @no_fold_nuw(i32 %x) {
; CHECK-LABEL: @no_fold_nuw(
; CHECK-NEXT: [[TMP2:%.*]] = add nuw i32 [[TMP0:%.*]], 12
; CHECK-NEXT: [[TMP3:%.*]] = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[TMP2]], i32 30)
; CHECK-NEXT: ret { i32, i1 } [[TMP3]]
; CHECK-NEXT: [[A:%.*]] = add nuw i32 [[X:%.*]], 12
; CHECK-NEXT: [[B:%.*]] = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[A]], i32 30)
; CHECK-NEXT: ret { i32, i1 } [[B]]
;
%2 = add nuw i32 %0, 12
%3 = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 %2, i32 30)
ret { i32, i1 } %3
%a = add nuw i32 %x, 12
%b = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 %a, i32 30)
ret { i32, i1 } %b
}
define { i32, i1 } @no_fold_wrapped_add(i32 %x) {
; CHECK-LABEL: @no_fold_wrapped_add(
; CHECK-NEXT: [[A:%.*]] = add i32 [[X:%.*]], 12
; CHECK-NEXT: [[B:%.*]] = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[A]], i32 30)
; CHECK-NEXT: ret { i32, i1 } [[B]]
;
%a = add i32 %x, 12
%b = tail call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 30, i32 %a)
ret { i32, i1 } %b
}

View File

@ -468,3 +468,57 @@ define i1 @uadd_res_ult_const_minus_one(i32 %x, i1* %p) nounwind {
%d = icmp ult i32 %c, -1
ret i1 %d
}
define { i32, i1 } @sadd_canonicalize_constant_arg0(i32 %x) nounwind {
; CHECK-LABEL: @sadd_canonicalize_constant_arg0(
; CHECK-NEXT: [[A:%.*]] = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[X:%.*]], i32 42)
; CHECK-NEXT: ret { i32, i1 } [[A]]
;
%a = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 42, i32 %x)
ret { i32, i1 } %a
}
define { i32, i1 } @uadd_canonicalize_constant_arg0(i32 %x) nounwind {
; CHECK-LABEL: @uadd_canonicalize_constant_arg0(
; CHECK-NEXT: [[A:%.*]] = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 [[X:%.*]], i32 42)
; CHECK-NEXT: ret { i32, i1 } [[A]]
;
%a = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 42, i32 %x)
ret { i32, i1 } %a
}
define { i32, i1 } @ssub_no_canonicalize_constant_arg0(i32 %x) nounwind {
; CHECK-LABEL: @ssub_no_canonicalize_constant_arg0(
; CHECK-NEXT: [[A:%.*]] = call { i32, i1 } @llvm.ssub.with.overflow.i32(i32 42, i32 [[X:%.*]])
; CHECK-NEXT: ret { i32, i1 } [[A]]
;
%a = call { i32, i1 } @llvm.ssub.with.overflow.i32(i32 42, i32 %x)
ret { i32, i1 } %a
}
define { i32, i1 } @usub_no_canonicalize_constant_arg0(i32 %x) nounwind {
; CHECK-LABEL: @usub_no_canonicalize_constant_arg0(
; CHECK-NEXT: [[A:%.*]] = call { i32, i1 } @llvm.usub.with.overflow.i32(i32 42, i32 [[X:%.*]])
; CHECK-NEXT: ret { i32, i1 } [[A]]
;
%a = call { i32, i1 } @llvm.usub.with.overflow.i32(i32 42, i32 %x)
ret { i32, i1 } %a
}
define { i32, i1 } @smul_canonicalize_constant_arg0(i32 %x) nounwind {
; CHECK-LABEL: @smul_canonicalize_constant_arg0(
; CHECK-NEXT: [[A:%.*]] = call { i32, i1 } @llvm.smul.with.overflow.i32(i32 [[X:%.*]], i32 42)
; CHECK-NEXT: ret { i32, i1 } [[A]]
;
%a = call { i32, i1 } @llvm.smul.with.overflow.i32(i32 42, i32 %x)
ret { i32, i1 } %a
}
define { i32, i1 } @umul_canonicalize_constant_arg0(i32 %x) nounwind {
; CHECK-LABEL: @umul_canonicalize_constant_arg0(
; CHECK-NEXT: [[A:%.*]] = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 [[X:%.*]], i32 42)
; CHECK-NEXT: ret { i32, i1 } [[A]]
;
%a = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 42, i32 %x)
ret { i32, i1 } %a
}