D106720 introduced features that did not work properly as we could add new queries after a fixpoint was reached and which could not be answered by the information gathered up to the fixpoint alone. As an alternative to D110078, which forced eager computation where we want to continue to be lazy, this patch fixes the problem. QueryAAs are AAs that allow lazy queries during their lifetime. They are never fixed if they have no outstanding dependences and always run as part of the updates in an iteration. To determine if we are done, all query AAs are asked if they received new queries, if not, we only need to consider updated AAs, as before. If new queries are present we go for another iteration. Differential Revision: https://reviews.llvm.org/D118669
621 lines
26 KiB
LLVM
621 lines
26 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --function-signature --check-attributes --check-globals
|
|
; RUN: opt -attributor -enable-new-pm=0 -attributor-manifest-internal -attributor-max-iterations-verify -attributor-annotate-decl-cs -attributor-max-iterations=2 -S < %s | FileCheck %s --check-prefixes=CHECK,NOT_CGSCC_NPM,NOT_CGSCC_OPM,NOT_TUNIT_NPM,IS__TUNIT____,IS________OPM,IS__TUNIT_OPM
|
|
; RUN: opt -aa-pipeline=basic-aa -passes=attributor -attributor-manifest-internal -attributor-max-iterations-verify -attributor-annotate-decl-cs -attributor-max-iterations=2 -S < %s | FileCheck %s --check-prefixes=CHECK,NOT_CGSCC_OPM,NOT_CGSCC_NPM,NOT_TUNIT_OPM,IS__TUNIT____,IS________NPM,IS__TUNIT_NPM
|
|
; RUN: opt -attributor-cgscc -enable-new-pm=0 -attributor-manifest-internal -attributor-annotate-decl-cs -S < %s | FileCheck %s --check-prefixes=CHECK,NOT_TUNIT_NPM,NOT_TUNIT_OPM,NOT_CGSCC_NPM,IS__CGSCC____,IS________OPM,IS__CGSCC_OPM
|
|
; RUN: opt -aa-pipeline=basic-aa -passes=attributor-cgscc -attributor-manifest-internal -attributor-annotate-decl-cs -S < %s | FileCheck %s --check-prefixes=CHECK,NOT_TUNIT_NPM,NOT_TUNIT_OPM,NOT_CGSCC_OPM,IS__CGSCC____,IS________NPM,IS__CGSCC_NPM
|
|
|
|
; Determine dereference-ability before unused loads get deleted:
|
|
; https://bugs.llvm.org/show_bug.cgi?id=21780
|
|
|
|
define <4 x double> @PR21780(double* %ptr) {
|
|
; CHECK: Function Attrs: argmemonly nofree norecurse nosync nounwind readonly willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@PR21780
|
|
; CHECK-SAME: (double* nocapture nofree noundef nonnull readonly align 8 dereferenceable(32) [[PTR:%.*]]) #[[ATTR0:[0-9]+]] {
|
|
; CHECK-NEXT: [[ARRAYIDX1:%.*]] = getelementptr inbounds double, double* [[PTR]], i64 1
|
|
; CHECK-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds double, double* [[PTR]], i64 2
|
|
; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds double, double* [[PTR]], i64 3
|
|
; CHECK-NEXT: [[T0:%.*]] = load double, double* [[PTR]], align 8
|
|
; CHECK-NEXT: [[T1:%.*]] = load double, double* [[ARRAYIDX1]], align 8
|
|
; CHECK-NEXT: [[T2:%.*]] = load double, double* [[ARRAYIDX2]], align 8
|
|
; CHECK-NEXT: [[T3:%.*]] = load double, double* [[ARRAYIDX3]], align 8
|
|
; CHECK-NEXT: [[VECINIT0:%.*]] = insertelement <4 x double> undef, double [[T0]], i32 0
|
|
; CHECK-NEXT: [[VECINIT1:%.*]] = insertelement <4 x double> [[VECINIT0]], double [[T1]], i32 1
|
|
; CHECK-NEXT: [[VECINIT2:%.*]] = insertelement <4 x double> [[VECINIT1]], double [[T2]], i32 2
|
|
; CHECK-NEXT: [[VECINIT3:%.*]] = insertelement <4 x double> [[VECINIT2]], double [[T3]], i32 3
|
|
; CHECK-NEXT: [[SHUFFLE:%.*]] = shufflevector <4 x double> [[VECINIT3]], <4 x double> [[VECINIT3]], <4 x i32> <i32 0, i32 0, i32 2, i32 2>
|
|
; CHECK-NEXT: ret <4 x double> [[SHUFFLE]]
|
|
;
|
|
|
|
; GEP of index 0 is simplified away.
|
|
%arrayidx1 = getelementptr inbounds double, double* %ptr, i64 1
|
|
%arrayidx2 = getelementptr inbounds double, double* %ptr, i64 2
|
|
%arrayidx3 = getelementptr inbounds double, double* %ptr, i64 3
|
|
|
|
%t0 = load double, double* %ptr, align 8
|
|
%t1 = load double, double* %arrayidx1, align 8
|
|
%t2 = load double, double* %arrayidx2, align 8
|
|
%t3 = load double, double* %arrayidx3, align 8
|
|
|
|
%vecinit0 = insertelement <4 x double> undef, double %t0, i32 0
|
|
%vecinit1 = insertelement <4 x double> %vecinit0, double %t1, i32 1
|
|
%vecinit2 = insertelement <4 x double> %vecinit1, double %t2, i32 2
|
|
%vecinit3 = insertelement <4 x double> %vecinit2, double %t3, i32 3
|
|
%shuffle = shufflevector <4 x double> %vecinit3, <4 x double> %vecinit3, <4 x i32> <i32 0, i32 0, i32 2, i32 2>
|
|
ret <4 x double> %shuffle
|
|
}
|
|
|
|
|
|
define double @PR21780_only_access3_with_inbounds(double* %ptr) {
|
|
; CHECK: Function Attrs: argmemonly nofree norecurse nosync nounwind readonly willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@PR21780_only_access3_with_inbounds
|
|
; CHECK-SAME: (double* nocapture nofree nonnull readonly align 8 dereferenceable(32) [[PTR:%.*]]) #[[ATTR0]] {
|
|
; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds double, double* [[PTR]], i64 3
|
|
; CHECK-NEXT: [[T3:%.*]] = load double, double* [[ARRAYIDX3]], align 8
|
|
; CHECK-NEXT: ret double [[T3]]
|
|
;
|
|
|
|
%arrayidx3 = getelementptr inbounds double, double* %ptr, i64 3
|
|
%t3 = load double, double* %arrayidx3, align 8
|
|
ret double %t3
|
|
}
|
|
|
|
define double @PR21780_only_access3_without_inbounds(double* %ptr) {
|
|
; CHECK: Function Attrs: argmemonly nofree norecurse nosync nounwind readonly willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@PR21780_only_access3_without_inbounds
|
|
; CHECK-SAME: (double* nocapture nofree readonly align 8 [[PTR:%.*]]) #[[ATTR0]] {
|
|
; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr double, double* [[PTR]], i64 3
|
|
; CHECK-NEXT: [[T3:%.*]] = load double, double* [[ARRAYIDX3]], align 8
|
|
; CHECK-NEXT: ret double [[T3]]
|
|
;
|
|
%arrayidx3 = getelementptr double, double* %ptr, i64 3
|
|
%t3 = load double, double* %arrayidx3, align 8
|
|
ret double %t3
|
|
}
|
|
|
|
define double @PR21780_without_inbounds(double* %ptr) {
|
|
; CHECK: Function Attrs: argmemonly nofree norecurse nosync nounwind readonly willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@PR21780_without_inbounds
|
|
; CHECK-SAME: (double* nocapture nofree noundef nonnull readonly align 8 dereferenceable(32) [[PTR:%.*]]) #[[ATTR0]] {
|
|
; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr double, double* [[PTR]], i64 3
|
|
; CHECK-NEXT: [[T3:%.*]] = load double, double* [[ARRAYIDX3]], align 8
|
|
; CHECK-NEXT: ret double [[T3]]
|
|
;
|
|
|
|
%arrayidx1 = getelementptr double, double* %ptr, i64 1
|
|
%arrayidx2 = getelementptr double, double* %ptr, i64 2
|
|
%arrayidx3 = getelementptr double, double* %ptr, i64 3
|
|
|
|
%t0 = load double, double* %ptr, align 8
|
|
%t1 = load double, double* %arrayidx1, align 8
|
|
%t2 = load double, double* %arrayidx2, align 8
|
|
%t3 = load double, double* %arrayidx3, align 8
|
|
|
|
ret double %t3
|
|
}
|
|
|
|
; Unsimplified, but still valid. Also, throw in some bogus arguments.
|
|
|
|
define void @gep0(i8* %unused, i8* %other, i8* %ptr) {
|
|
; CHECK: Function Attrs: argmemonly nofree norecurse nosync nounwind willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@gep0
|
|
; CHECK-SAME: (i8* nocapture nofree readnone [[UNUSED:%.*]], i8* nocapture nofree noundef nonnull writeonly dereferenceable(1) [[OTHER:%.*]], i8* nocapture nofree nonnull readonly dereferenceable(3) [[PTR:%.*]]) #[[ATTR1:[0-9]+]] {
|
|
; CHECK-NEXT: [[ARRAYIDX2:%.*]] = getelementptr i8, i8* [[PTR]], i64 2
|
|
; CHECK-NEXT: [[T2:%.*]] = load i8, i8* [[ARRAYIDX2]], align 1
|
|
; CHECK-NEXT: store i8 [[T2]], i8* [[OTHER]], align 1
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx0 = getelementptr i8, i8* %ptr, i64 0
|
|
%arrayidx1 = getelementptr i8, i8* %ptr, i64 1
|
|
%arrayidx2 = getelementptr i8, i8* %ptr, i64 2
|
|
%t0 = load i8, i8* %arrayidx0
|
|
%t1 = load i8, i8* %arrayidx1
|
|
%t2 = load i8, i8* %arrayidx2
|
|
store i8 %t2, i8* %other
|
|
ret void
|
|
}
|
|
|
|
; Order of accesses does not change computation.
|
|
; Multiple arguments may be dereferenceable.
|
|
|
|
define void @ordering(i8* %ptr1, i32* %ptr2) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@ordering
|
|
; CHECK-SAME: (i8* nocapture nofree nonnull readnone dereferenceable(3) [[PTR1:%.*]], i32* nocapture nofree nonnull readnone align 4 dereferenceable(8) [[PTR2:%.*]]) #[[ATTR2:[0-9]+]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%a20 = getelementptr i32, i32* %ptr2, i64 0
|
|
%a12 = getelementptr i8, i8* %ptr1, i64 2
|
|
%t12 = load i8, i8* %a12
|
|
%a11 = getelementptr i8, i8* %ptr1, i64 1
|
|
%t20 = load i32, i32* %a20
|
|
%a10 = getelementptr i8, i8* %ptr1, i64 0
|
|
%t10 = load i8, i8* %a10
|
|
%t11 = load i8, i8* %a11
|
|
%a21 = getelementptr i32, i32* %ptr2, i64 1
|
|
%t21 = load i32, i32* %a21
|
|
ret void
|
|
}
|
|
|
|
; Not in entry block.
|
|
|
|
define void @not_entry_but_guaranteed_to_execute(i8* %ptr) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@not_entry_but_guaranteed_to_execute
|
|
; CHECK-SAME: (i8* nocapture nofree nonnull readnone dereferenceable(3) [[PTR:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: br label [[EXIT:%.*]]
|
|
; CHECK: exit:
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
br label %exit
|
|
exit:
|
|
%arrayidx0 = getelementptr i8, i8* %ptr, i64 0
|
|
%arrayidx1 = getelementptr i8, i8* %ptr, i64 1
|
|
%arrayidx2 = getelementptr i8, i8* %ptr, i64 2
|
|
%t0 = load i8, i8* %arrayidx0
|
|
%t1 = load i8, i8* %arrayidx1
|
|
%t2 = load i8, i8* %arrayidx2
|
|
ret void
|
|
}
|
|
|
|
; Not in entry block and not guaranteed to execute.
|
|
|
|
define void @not_entry_not_guaranteed_to_execute(i8* %ptr, i1 %cond) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@not_entry_not_guaranteed_to_execute
|
|
; CHECK-SAME: (i8* nocapture nofree readnone [[PTR:%.*]], i1 [[COND:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: br i1 [[COND]], label [[LOADS:%.*]], label [[EXIT:%.*]]
|
|
; CHECK: loads:
|
|
; CHECK-NEXT: ret void
|
|
; CHECK: exit:
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
br i1 %cond, label %loads, label %exit
|
|
loads:
|
|
%arrayidx0 = getelementptr i8, i8* %ptr, i64 0
|
|
%arrayidx1 = getelementptr i8, i8* %ptr, i64 1
|
|
%arrayidx2 = getelementptr i8, i8* %ptr, i64 2
|
|
%t0 = load i8, i8* %arrayidx0
|
|
%t1 = load i8, i8* %arrayidx1
|
|
%t2 = load i8, i8* %arrayidx2
|
|
ret void
|
|
exit:
|
|
ret void
|
|
}
|
|
|
|
; The last load may not execute, so derefenceable bytes only covers the 1st two loads.
|
|
|
|
define void @partial_in_entry(i16* %ptr, i1 %cond) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@partial_in_entry
|
|
; CHECK-SAME: (i16* nocapture nofree nonnull readnone align 2 dereferenceable(4) [[PTR:%.*]], i1 [[COND:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: br i1 [[COND]], label [[LOADS:%.*]], label [[EXIT:%.*]]
|
|
; CHECK: loads:
|
|
; CHECK-NEXT: ret void
|
|
; CHECK: exit:
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
%arrayidx0 = getelementptr i16, i16* %ptr, i64 0
|
|
%arrayidx1 = getelementptr i16, i16* %ptr, i64 1
|
|
%arrayidx2 = getelementptr i16, i16* %ptr, i64 2
|
|
%t0 = load i16, i16* %arrayidx0
|
|
%t1 = load i16, i16* %arrayidx1
|
|
br i1 %cond, label %loads, label %exit
|
|
loads:
|
|
%t2 = load i16, i16* %arrayidx2
|
|
ret void
|
|
exit:
|
|
ret void
|
|
}
|
|
|
|
; The volatile load can't be used to prove a non-volatile access is allowed.
|
|
; The 2nd and 3rd loads may never execute.
|
|
|
|
define void @volatile_is_not_dereferenceable(i16* %ptr) {
|
|
; CHECK: Function Attrs: argmemonly nofree norecurse nounwind willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@volatile_is_not_dereferenceable
|
|
; CHECK-SAME: (i16* nofree align 2 [[PTR:%.*]]) #[[ATTR3:[0-9]+]] {
|
|
; CHECK-NEXT: [[T0:%.*]] = load volatile i16, i16* [[PTR]], align 2
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx0 = getelementptr i16, i16* %ptr, i64 0
|
|
%arrayidx1 = getelementptr i16, i16* %ptr, i64 1
|
|
%arrayidx2 = getelementptr i16, i16* %ptr, i64 2
|
|
%t0 = load volatile i16, i16* %arrayidx0
|
|
%t1 = load i16, i16* %arrayidx1
|
|
%t2 = load i16, i16* %arrayidx2
|
|
ret void
|
|
}
|
|
|
|
; TODO: We should allow inference for atomic (but not volatile) ops.
|
|
|
|
define void @atomic_is_alright(i16* %ptr) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@atomic_is_alright
|
|
; CHECK-SAME: (i16* nocapture nofree nonnull readnone align 2 dereferenceable(6) [[PTR:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx0 = getelementptr i16, i16* %ptr, i64 0
|
|
%arrayidx1 = getelementptr i16, i16* %ptr, i64 1
|
|
%arrayidx2 = getelementptr i16, i16* %ptr, i64 2
|
|
%t0 = load atomic i16, i16* %arrayidx0 unordered, align 2
|
|
%t1 = load i16, i16* %arrayidx1
|
|
%t2 = load i16, i16* %arrayidx2
|
|
ret void
|
|
}
|
|
|
|
declare void @may_not_return()
|
|
|
|
define void @not_guaranteed_to_transfer_execution(i16* %ptr) {
|
|
; CHECK-LABEL: define {{[^@]+}}@not_guaranteed_to_transfer_execution
|
|
; CHECK-SAME: (i16* nocapture nofree nonnull readnone align 2 dereferenceable(2) [[PTR:%.*]]) {
|
|
; CHECK-NEXT: call void @may_not_return()
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx0 = getelementptr i16, i16* %ptr, i64 0
|
|
%arrayidx1 = getelementptr i16, i16* %ptr, i64 1
|
|
%arrayidx2 = getelementptr i16, i16* %ptr, i64 2
|
|
%t0 = load i16, i16* %arrayidx0
|
|
call void @may_not_return()
|
|
%t1 = load i16, i16* %arrayidx1
|
|
%t2 = load i16, i16* %arrayidx2
|
|
ret void
|
|
}
|
|
|
|
; We must have consecutive accesses.
|
|
|
|
define void @variable_gep_index(i8* %unused, i8* %ptr, i64 %variable_index) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@variable_gep_index
|
|
; CHECK-SAME: (i8* nocapture nofree readnone [[UNUSED:%.*]], i8* nocapture nofree nonnull readnone dereferenceable(1) [[PTR:%.*]], i64 [[VARIABLE_INDEX:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx1 = getelementptr i8, i8* %ptr, i64 %variable_index
|
|
%arrayidx2 = getelementptr i8, i8* %ptr, i64 2
|
|
%t0 = load i8, i8* %ptr
|
|
%t1 = load i8, i8* %arrayidx1
|
|
%t2 = load i8, i8* %arrayidx2
|
|
ret void
|
|
}
|
|
|
|
; Deal with >1 GEP index.
|
|
|
|
define void @multi_index_gep(<4 x i8>* %ptr) {
|
|
; FIXME: %ptr should be dereferenceable(4)
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@multi_index_gep
|
|
; CHECK-SAME: (<4 x i8>* nocapture nofree nonnull readnone dereferenceable(1) [[PTR:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx00 = getelementptr <4 x i8>, <4 x i8>* %ptr, i64 0, i64 0
|
|
%t0 = load i8, i8* %arrayidx00
|
|
ret void
|
|
}
|
|
|
|
; Could round weird bitwidths down?
|
|
|
|
define void @not_byte_multiple(i9* %ptr) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@not_byte_multiple
|
|
; CHECK-SAME: (i9* nocapture nofree nonnull readnone align 2 dereferenceable(2) [[PTR:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx0 = getelementptr i9, i9* %ptr, i64 0
|
|
%t0 = load i9, i9* %arrayidx0
|
|
ret void
|
|
}
|
|
|
|
; Missing direct access from the pointer.
|
|
|
|
define void @no_pointer_deref(i16* %ptr) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@no_pointer_deref
|
|
; CHECK-SAME: (i16* nocapture nofree readnone align 2 [[PTR:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx1 = getelementptr i16, i16* %ptr, i64 1
|
|
%arrayidx2 = getelementptr i16, i16* %ptr, i64 2
|
|
%t1 = load i16, i16* %arrayidx1
|
|
%t2 = load i16, i16* %arrayidx2
|
|
ret void
|
|
}
|
|
|
|
; Out-of-order is ok, but missing access concludes dereferenceable range.
|
|
|
|
define void @non_consecutive(i32* %ptr) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@non_consecutive
|
|
; CHECK-SAME: (i32* nocapture nofree nonnull readnone align 4 dereferenceable(8) [[PTR:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx1 = getelementptr i32, i32* %ptr, i64 1
|
|
%arrayidx0 = getelementptr i32, i32* %ptr, i64 0
|
|
%arrayidx3 = getelementptr i32, i32* %ptr, i64 3
|
|
%t1 = load i32, i32* %arrayidx1
|
|
%t0 = load i32, i32* %arrayidx0
|
|
%t3 = load i32, i32* %arrayidx3
|
|
ret void
|
|
}
|
|
|
|
; Improve on existing dereferenceable attribute.
|
|
|
|
define void @more_bytes(i32* dereferenceable(8) %ptr) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@more_bytes
|
|
; CHECK-SAME: (i32* nocapture nofree nonnull readnone align 4 dereferenceable(16) [[PTR:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx3 = getelementptr i32, i32* %ptr, i64 3
|
|
%arrayidx1 = getelementptr i32, i32* %ptr, i64 1
|
|
%arrayidx0 = getelementptr i32, i32* %ptr, i64 0
|
|
%arrayidx2 = getelementptr i32, i32* %ptr, i64 2
|
|
%t3 = load i32, i32* %arrayidx3
|
|
%t1 = load i32, i32* %arrayidx1
|
|
%t2 = load i32, i32* %arrayidx2
|
|
%t0 = load i32, i32* %arrayidx0
|
|
ret void
|
|
}
|
|
|
|
; Improve on existing dereferenceable_or_null attribute.
|
|
|
|
define void @more_bytes_and_not_null(i32* dereferenceable_or_null(8) %ptr) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@more_bytes_and_not_null
|
|
; CHECK-SAME: (i32* nocapture nofree nonnull readnone align 4 dereferenceable(16) [[PTR:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx3 = getelementptr i32, i32* %ptr, i64 3
|
|
%arrayidx1 = getelementptr i32, i32* %ptr, i64 1
|
|
%arrayidx0 = getelementptr i32, i32* %ptr, i64 0
|
|
%arrayidx2 = getelementptr i32, i32* %ptr, i64 2
|
|
%t3 = load i32, i32* %arrayidx3
|
|
%t1 = load i32, i32* %arrayidx1
|
|
%t2 = load i32, i32* %arrayidx2
|
|
%t0 = load i32, i32* %arrayidx0
|
|
ret void
|
|
}
|
|
|
|
; But don't pessimize existing dereferenceable attribute.
|
|
|
|
define void @better_bytes(i32* dereferenceable(100) %ptr) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@better_bytes
|
|
; CHECK-SAME: (i32* nocapture nofree nonnull readnone align 4 dereferenceable(100) [[PTR:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arrayidx3 = getelementptr i32, i32* %ptr, i64 3
|
|
%arrayidx1 = getelementptr i32, i32* %ptr, i64 1
|
|
%arrayidx0 = getelementptr i32, i32* %ptr, i64 0
|
|
%arrayidx2 = getelementptr i32, i32* %ptr, i64 2
|
|
%t3 = load i32, i32* %arrayidx3
|
|
%t1 = load i32, i32* %arrayidx1
|
|
%t2 = load i32, i32* %arrayidx2
|
|
%t0 = load i32, i32* %arrayidx0
|
|
ret void
|
|
}
|
|
|
|
define void @bitcast(i32* %arg) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@bitcast
|
|
; CHECK-SAME: (i32* nocapture nofree nonnull readnone align 4 dereferenceable(8) [[ARG:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%ptr = bitcast i32* %arg to float*
|
|
%arrayidx0 = getelementptr float, float* %ptr, i64 0
|
|
%arrayidx1 = getelementptr float, float* %ptr, i64 1
|
|
%t0 = load float, float* %arrayidx0
|
|
%t1 = load float, float* %arrayidx1
|
|
ret void
|
|
}
|
|
|
|
define void @bitcast_different_sizes(double* %arg1, i8* %arg2) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@bitcast_different_sizes
|
|
; CHECK-SAME: (double* nocapture nofree nonnull readnone align 4 dereferenceable(12) [[ARG1:%.*]], i8* nocapture nofree nonnull readnone align 4 dereferenceable(16) [[ARG2:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%ptr1 = bitcast double* %arg1 to float*
|
|
%a10 = getelementptr float, float* %ptr1, i64 0
|
|
%a11 = getelementptr float, float* %ptr1, i64 1
|
|
%a12 = getelementptr float, float* %ptr1, i64 2
|
|
%ld10 = load float, float* %a10
|
|
%ld11 = load float, float* %a11
|
|
%ld12 = load float, float* %a12
|
|
|
|
%ptr2 = bitcast i8* %arg2 to i64*
|
|
%a20 = getelementptr i64, i64* %ptr2, i64 0
|
|
%a21 = getelementptr i64, i64* %ptr2, i64 1
|
|
%ld20 = load i64, i64* %a20
|
|
%ld21 = load i64, i64* %a21
|
|
ret void
|
|
}
|
|
|
|
define void @negative_offset(i32* %arg) {
|
|
; CHECK: Function Attrs: nofree norecurse nosync nounwind readnone willreturn
|
|
; CHECK-LABEL: define {{[^@]+}}@negative_offset
|
|
; CHECK-SAME: (i32* nocapture nofree nonnull readnone align 4 dereferenceable(4) [[ARG:%.*]]) #[[ATTR2]] {
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%ptr = bitcast i32* %arg to float*
|
|
%arrayidx0 = getelementptr float, float* %ptr, i64 0
|
|
%arrayidx1 = getelementptr float, float* %ptr, i64 -1
|
|
%t0 = load float, float* %arrayidx0
|
|
%t1 = load float, float* %arrayidx1
|
|
ret void
|
|
}
|
|
|
|
define void @stores(i32* %arg) {
|
|
; CHECK: Function Attrs: argmemonly nofree norecurse nosync nounwind willreturn writeonly
|
|
; CHECK-LABEL: define {{[^@]+}}@stores
|
|
; CHECK-SAME: (i32* nocapture nofree nonnull writeonly align 4 dereferenceable(8) [[ARG:%.*]]) #[[ATTR4:[0-9]+]] {
|
|
; CHECK-NEXT: [[PTR:%.*]] = bitcast i32* [[ARG]] to float*
|
|
; CHECK-NEXT: [[ARRAYIDX0:%.*]] = getelementptr float, float* [[PTR]], i64 0
|
|
; CHECK-NEXT: [[ARRAYIDX1:%.*]] = getelementptr float, float* [[PTR]], i64 1
|
|
; CHECK-NEXT: store float 1.000000e+00, float* [[ARRAYIDX0]], align 4
|
|
; CHECK-NEXT: store float 2.000000e+00, float* [[ARRAYIDX1]], align 4
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%ptr = bitcast i32* %arg to float*
|
|
%arrayidx0 = getelementptr float, float* %ptr, i64 0
|
|
%arrayidx1 = getelementptr float, float* %ptr, i64 1
|
|
store float 1.0, float* %arrayidx0
|
|
store float 2.0, float* %arrayidx1
|
|
ret void
|
|
}
|
|
|
|
define void @load_store(i32* %arg) {
|
|
; CHECK: Function Attrs: argmemonly nofree norecurse nosync nounwind willreturn writeonly
|
|
; CHECK-LABEL: define {{[^@]+}}@load_store
|
|
; CHECK-SAME: (i32* nocapture nofree nonnull writeonly align 4 dereferenceable(8) [[ARG:%.*]]) #[[ATTR4]] {
|
|
; CHECK-NEXT: [[PTR:%.*]] = bitcast i32* [[ARG]] to float*
|
|
; CHECK-NEXT: [[ARRAYIDX1:%.*]] = getelementptr float, float* [[PTR]], i64 1
|
|
; CHECK-NEXT: store float 2.000000e+00, float* [[ARRAYIDX1]], align 4
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%ptr = bitcast i32* %arg to float*
|
|
%arrayidx0 = getelementptr float, float* %ptr, i64 0
|
|
%arrayidx1 = getelementptr float, float* %ptr, i64 1
|
|
%t1 = load float, float* %arrayidx0
|
|
store float 2.0, float* %arrayidx1
|
|
ret void
|
|
}
|
|
|
|
define void @different_size1(i32* %arg) {
|
|
; CHECK: Function Attrs: argmemonly nofree norecurse nosync nounwind willreturn writeonly
|
|
; CHECK-LABEL: define {{[^@]+}}@different_size1
|
|
; CHECK-SAME: (i32* nocapture nofree noundef nonnull writeonly align 8 dereferenceable(8) [[ARG:%.*]]) #[[ATTR4]] {
|
|
; CHECK-NEXT: [[ARG_CAST:%.*]] = bitcast i32* [[ARG]] to double*
|
|
; CHECK-NEXT: store double 0.000000e+00, double* [[ARG_CAST]], align 8
|
|
; CHECK-NEXT: store i32 0, i32* [[ARG]], align 8
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
%arg-cast = bitcast i32* %arg to double*
|
|
store double 0.000000e+00, double* %arg-cast
|
|
store i32 0, i32* %arg
|
|
ret void
|
|
}
|
|
|
|
define void @different_size2(i32* %arg) {
|
|
; CHECK: Function Attrs: argmemonly nofree norecurse nosync nounwind willreturn writeonly
|
|
; CHECK-LABEL: define {{[^@]+}}@different_size2
|
|
; CHECK-SAME: (i32* nocapture nofree noundef nonnull writeonly align 8 dereferenceable(8) [[ARG:%.*]]) #[[ATTR4]] {
|
|
; CHECK-NEXT: store i32 0, i32* [[ARG]], align 8
|
|
; CHECK-NEXT: [[ARG_CAST:%.*]] = bitcast i32* [[ARG]] to double*
|
|
; CHECK-NEXT: store double 0.000000e+00, double* [[ARG_CAST]], align 8
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
store i32 0, i32* %arg
|
|
%arg-cast = bitcast i32* %arg to double*
|
|
store double 0.000000e+00, double* %arg-cast
|
|
ret void
|
|
}
|
|
|
|
; Make use of MustBeExecuted Explorer
|
|
;
|
|
; [CFG]
|
|
; entry
|
|
; / \
|
|
; l1 l2
|
|
; | X |
|
|
; l3 l4
|
|
; \ /
|
|
; l5
|
|
; / \
|
|
; l6 l7
|
|
; \ /
|
|
; end
|
|
; According to the above CFG, we can see that instructions in l5 Block must be executed.
|
|
; Therefore, %p must be dereferenced.
|
|
;
|
|
; ATTRIBUTOR_CGSCC_NPM-LABEL: define i32 @require_cfg_analysis(i32 %c, i32* {{.*}} dereferenceable(4) %p)
|
|
define i32 @require_cfg_analysis(i32 %c, i32* %p) {
|
|
; IS________OPM: Function Attrs: argmemonly nofree norecurse nosync nounwind willreturn writeonly
|
|
; IS________OPM-LABEL: define {{[^@]+}}@require_cfg_analysis
|
|
; IS________OPM-SAME: (i32 [[C:%.*]], i32* nocapture nofree writeonly [[P:%.*]]) #[[ATTR4]] {
|
|
; IS________OPM-NEXT: [[TOBOOL1:%.*]] = icmp eq i32 [[C]], 0
|
|
; IS________OPM-NEXT: br i1 [[TOBOOL1]], label [[L1:%.*]], label [[L2:%.*]]
|
|
; IS________OPM: l1:
|
|
; IS________OPM-NEXT: [[TOBOOL2:%.*]] = icmp eq i32 [[C]], 1
|
|
; IS________OPM-NEXT: br i1 [[TOBOOL2]], label [[L3:%.*]], label [[L4:%.*]]
|
|
; IS________OPM: l2:
|
|
; IS________OPM-NEXT: [[TOBOOL3:%.*]] = icmp eq i32 [[C]], 2
|
|
; IS________OPM-NEXT: br i1 [[TOBOOL3]], label [[L3]], label [[L4]]
|
|
; IS________OPM: l3:
|
|
; IS________OPM-NEXT: br label [[L5:%.*]]
|
|
; IS________OPM: l4:
|
|
; IS________OPM-NEXT: br label [[L5]]
|
|
; IS________OPM: l5:
|
|
; IS________OPM-NEXT: [[TOBOOL4:%.*]] = icmp eq i32 [[C]], 4
|
|
; IS________OPM-NEXT: br i1 [[TOBOOL4]], label [[L6:%.*]], label [[L7:%.*]]
|
|
; IS________OPM: l6:
|
|
; IS________OPM-NEXT: store i32 0, i32* [[P]], align 4
|
|
; IS________OPM-NEXT: br label [[END:%.*]]
|
|
; IS________OPM: l7:
|
|
; IS________OPM-NEXT: store i32 1, i32* [[P]], align 4
|
|
; IS________OPM-NEXT: br label [[END]]
|
|
; IS________OPM: end:
|
|
; IS________OPM-NEXT: ret i32 1
|
|
;
|
|
; IS________NPM: Function Attrs: argmemonly nofree norecurse nosync nounwind willreturn writeonly
|
|
; IS________NPM-LABEL: define {{[^@]+}}@require_cfg_analysis
|
|
; IS________NPM-SAME: (i32 [[C:%.*]], i32* nocapture nofree nonnull writeonly align 4 dereferenceable(4) [[P:%.*]]) #[[ATTR4]] {
|
|
; IS________NPM-NEXT: [[TOBOOL1:%.*]] = icmp eq i32 [[C]], 0
|
|
; IS________NPM-NEXT: br i1 [[TOBOOL1]], label [[L1:%.*]], label [[L2:%.*]]
|
|
; IS________NPM: l1:
|
|
; IS________NPM-NEXT: br label [[L4:%.*]]
|
|
; IS________NPM: l2:
|
|
; IS________NPM-NEXT: [[TOBOOL3:%.*]] = icmp eq i32 [[C]], 2
|
|
; IS________NPM-NEXT: br i1 [[TOBOOL3]], label [[L3:%.*]], label [[L4]]
|
|
; IS________NPM: l3:
|
|
; IS________NPM-NEXT: br label [[L5:%.*]]
|
|
; IS________NPM: l4:
|
|
; IS________NPM-NEXT: br label [[L5]]
|
|
; IS________NPM: l5:
|
|
; IS________NPM-NEXT: [[TOBOOL4:%.*]] = icmp eq i32 [[C]], 4
|
|
; IS________NPM-NEXT: br i1 [[TOBOOL4]], label [[L6:%.*]], label [[L7:%.*]]
|
|
; IS________NPM: l6:
|
|
; IS________NPM-NEXT: store i32 0, i32* [[P]], align 4
|
|
; IS________NPM-NEXT: br label [[END:%.*]]
|
|
; IS________NPM: l7:
|
|
; IS________NPM-NEXT: store i32 1, i32* [[P]], align 4
|
|
; IS________NPM-NEXT: br label [[END]]
|
|
; IS________NPM: end:
|
|
; IS________NPM-NEXT: ret i32 1
|
|
;
|
|
%tobool1 = icmp eq i32 %c, 0
|
|
br i1 %tobool1, label %l1, label %l2
|
|
l1:
|
|
%tobool2 = icmp eq i32 %c, 1
|
|
br i1 %tobool2, label %l3, label %l4
|
|
l2:
|
|
%tobool3 = icmp eq i32 %c, 2
|
|
br i1 %tobool3, label %l3, label %l4
|
|
l3:
|
|
br label %l5
|
|
l4:
|
|
br label %l5
|
|
l5:
|
|
%tobool4 = icmp eq i32 %c, 4
|
|
br i1 %tobool4, label %l6, label %l7
|
|
l6:
|
|
store i32 0, i32* %p
|
|
br label %end
|
|
l7:
|
|
store i32 1, i32* %p
|
|
br label %end
|
|
end:
|
|
ret i32 1
|
|
}
|
|
;.
|
|
; CHECK: attributes #[[ATTR0]] = { argmemonly nofree norecurse nosync nounwind readonly willreturn }
|
|
; CHECK: attributes #[[ATTR1]] = { argmemonly nofree norecurse nosync nounwind willreturn }
|
|
; CHECK: attributes #[[ATTR2]] = { nofree norecurse nosync nounwind readnone willreturn }
|
|
; CHECK: attributes #[[ATTR3]] = { argmemonly nofree norecurse nounwind willreturn }
|
|
; CHECK: attributes #[[ATTR4]] = { argmemonly nofree norecurse nosync nounwind willreturn writeonly }
|
|
;.
|