aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
Diffstat
-rw-r--r--clang/docs/ReleaseNotes.rst3
-rw-r--r--clang/include/clang/Sema/SemaConcept.h2
-rw-r--r--clang/lib/AST/ByteCode/InterpBuiltin.cpp4
-rw-r--r--clang/lib/AST/DeclPrinter.cpp4
-rw-r--r--clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp6
-rw-r--r--clang/test/AST/ast-print-record-decl.c13
-rw-r--r--clang/test/CIR/CodeGen/struct.cpp33
-rw-r--r--compiler-rt/test/asan/TestCases/wcscat.cpp20
-rw-r--r--compiler-rt/test/asan/TestCases/wcscpy.cpp20
-rw-r--r--compiler-rt/test/asan/TestCases/wcsncat.cpp20
-rw-r--r--compiler-rt/test/asan/TestCases/wcsncpy.cpp20
-rw-r--r--llvm/lib/Transforms/Vectorize/LoopVectorize.cpp16
-rw-r--r--llvm/lib/Transforms/Vectorize/VPlan.cpp9
-rw-r--r--llvm/lib/Transforms/Vectorize/VPlanHelpers.h16
-rw-r--r--llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp123
-rw-r--r--llvm/test/Transforms/LoopVectorize/X86/replicating-load-store-costs.ll126
-rw-r--r--llvm/tools/llvm-c-test/debuginfo.c2
-rw-r--r--mlir/test/Dialect/Vector/canonicalize.mlir9
18 files changed, 137 insertions, 309 deletions
diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index ab536ad..05379f4 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -123,6 +123,9 @@ AST Dumping Potentially Breaking Changes
``__atomic_test_and_set(p, 0)``
+- Pretty-printing of templates with inherited (i.e. specified in a previous
+ redeclaration) default arguments has been fixed.
+
Clang Frontend Potentially Breaking Changes
-------------------------------------------
- Members of anonymous unions/structs are now injected as ``IndirectFieldDecl``
diff --git a/clang/include/clang/Sema/SemaConcept.h b/clang/include/clang/Sema/SemaConcept.h
index 51ca1e1..bdd997b 100644
--- a/clang/include/clang/Sema/SemaConcept.h
+++ b/clang/include/clang/Sema/SemaConcept.h
@@ -257,6 +257,7 @@ public:
case ConstraintKind::FoldExpanded:
return FoldExpanded.Pattern->getBeginLoc();
}
+ llvm_unreachable("Unknown ConstraintKind enum");
}
SourceLocation getEndLoc() const {
@@ -270,6 +271,7 @@ public:
case ConstraintKind::FoldExpanded:
return FoldExpanded.Pattern->getEndLoc();
}
+ llvm_unreachable("Unknown ConstraintKind enum");
}
SourceRange getSourceRange() const { return {getBeginLoc(), getEndLoc()}; }
diff --git a/clang/lib/AST/ByteCode/InterpBuiltin.cpp b/clang/lib/AST/ByteCode/InterpBuiltin.cpp
index 6053237..68ebfdf 100644
--- a/clang/lib/AST/ByteCode/InterpBuiltin.cpp
+++ b/clang/lib/AST/ByteCode/InterpBuiltin.cpp
@@ -2796,7 +2796,7 @@ static bool interp__builtin_ia32_pshuf(InterpState &S, CodePtr OpPC,
unsigned LaneBase = (Idx / LaneElts) * LaneElts;
unsigned LaneIdx = Idx % LaneElts;
unsigned SrcIdx = Idx;
- unsigned Sel = (Ctl >> (2 * LaneIdx)) & 0x3;
+ unsigned Sel = (Ctl >> (2 * (LaneIdx & 0x3))) & 0x3;
if (ElemBits == 32) {
SrcIdx = LaneBase + Sel;
} else {
@@ -2805,8 +2805,6 @@ static bool interp__builtin_ia32_pshuf(InterpState &S, CodePtr OpPC,
if (!IsShufHW && !InHigh) {
SrcIdx = LaneBase + Sel;
} else if (IsShufHW && InHigh) {
- unsigned Rel = LaneIdx - HalfSize;
- Sel = (Ctl >> (2 * Rel)) & 0x3;
SrcIdx = LaneBase + HalfSize + Sel;
}
}
diff --git a/clang/lib/AST/DeclPrinter.cpp b/clang/lib/AST/DeclPrinter.cpp
index 196057f..7001ade 100644
--- a/clang/lib/AST/DeclPrinter.cpp
+++ b/clang/lib/AST/DeclPrinter.cpp
@@ -1894,7 +1894,7 @@ void DeclPrinter::VisitTemplateTypeParmDecl(const TemplateTypeParmDecl *TTP) {
Out << TTP->getDeclName();
}
- if (TTP->hasDefaultArgument()) {
+ if (TTP->hasDefaultArgument() && !TTP->defaultArgumentWasInherited()) {
Out << " = ";
TTP->getDefaultArgument().getArgument().print(Policy, Out,
/*IncludeType=*/false);
@@ -1909,7 +1909,7 @@ void DeclPrinter::VisitNonTypeTemplateParmDecl(
Policy.CleanUglifiedParameters ? II->deuglifiedName() : II->getName();
printDeclType(NTTP->getType(), Name, NTTP->isParameterPack());
- if (NTTP->hasDefaultArgument()) {
+ if (NTTP->hasDefaultArgument() && !NTTP->defaultArgumentWasInherited()) {
Out << " = ";
NTTP->getDefaultArgument().getArgument().print(Policy, Out,
/*IncludeType=*/false);
diff --git a/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp b/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp
index e51c3fc..60ccf18 100644
--- a/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp
+++ b/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp
@@ -201,8 +201,10 @@ public:
}
void VisitDesignatedInitUpdateExpr(DesignatedInitUpdateExpr *e) {
- cgf.cgm.errorNYI(e->getSourceRange(),
- "AggExprEmitter: VisitDesignatedInitUpdateExpr");
+ AggValueSlot dest = ensureSlot(cgf.getLoc(e->getExprLoc()), e->getType());
+ LValue destLV = cgf.makeAddrLValue(dest.getAddress(), e->getType());
+ emitInitializationToLValue(e->getBase(), destLV);
+ VisitInitListExpr(e->getUpdater());
}
void VisitAbstractConditionalOperator(const AbstractConditionalOperator *e) {
cgf.cgm.errorNYI(e->getSourceRange(),
diff --git a/clang/test/AST/ast-print-record-decl.c b/clang/test/AST/ast-print-record-decl.c
index d3717a4..fd81588 100644
--- a/clang/test/AST/ast-print-record-decl.c
+++ b/clang/test/AST/ast-print-record-decl.c
@@ -290,9 +290,9 @@ KW DeclGroupInMemberList {
// A tag decl group in the tag decl's own member list is exercised in
// defSelfRef above.
+#ifdef __cplusplus
// Check out-of-line record definition
-#ifdef __cplusplus
// PRINT-CXX-NEXT: [[KW]] OutOfLineRecord {
KW OutOfLineRecord {
// PRINT-CXX-NEXT: [[KW]] Inner
@@ -304,4 +304,15 @@ KW OutOfLineRecord {
KW OutOfLineRecord::Inner {
// PRINT-CXX-NEXT: };
};
+
+// PRINT-CXX-NEXT: template <typename, typename = int> [[KW]] SmearedTypeDefArgs;
+template <typename, typename = int> KW SmearedTypeDefArgs;
+// PRINT-CXX-NEXT: template <typename = int, typename> [[KW]] SmearedTypeDefArgs;
+template <typename = int, typename> KW SmearedTypeDefArgs;
+
+// PRINT-CXX-NEXT: template <int, int = 0> [[KW]] SmearedNTTPDefArgs;
+template <int, int = 0> KW SmearedNTTPDefArgs;
+// PRINT-CXX-NEXT: template <int = 0, int> [[KW]] SmearedNTTPDefArgs;
+template <int = 0, int> KW SmearedNTTPDefArgs;
+
#endif
diff --git a/clang/test/CIR/CodeGen/struct.cpp b/clang/test/CIR/CodeGen/struct.cpp
index c68f742..263799f 100644
--- a/clang/test/CIR/CodeGen/struct.cpp
+++ b/clang/test/CIR/CodeGen/struct.cpp
@@ -184,6 +184,39 @@ void generic_selection() {
// OGCG: %[[D_ADDR:.*]] = alloca %struct.CompleteS, align 4
// OGCG: call void @llvm.memcpy.p0.p0.i64(ptr align 4 %[[D_ADDR]], ptr align 4 %[[A_ADDR]], i64 8, i1 false)
+void designated_init_update_expr() {
+ CompleteS a;
+
+ struct Container {
+ CompleteS c;
+ } b = {a, .c.a = 1};
+}
+
+// CIR: %[[A_ADDR:.*]] = cir.alloca !rec_CompleteS, !cir.ptr<!rec_CompleteS>, ["a"]
+// CIR: %[[B_ADDR:.*]] = cir.alloca !rec_Container, !cir.ptr<!rec_Container>, ["b", init]
+// CIR: %[[C_ADDR:.*]] = cir.get_member %[[B_ADDR]][0] {name = "c"} : !cir.ptr<!rec_Container> -> !cir.ptr<!rec_CompleteS>
+// CIR: cir.call @_ZN9CompleteSC1ERKS_(%2, %[[A_ADDR]]) nothrow : (!cir.ptr<!rec_CompleteS>, !cir.ptr<!rec_CompleteS>) -> ()
+// CIR: %[[ELEM_0_PTR:.*]] = cir.get_member %[[C_ADDR]][0] {name = "a"} : !cir.ptr<!rec_CompleteS> -> !cir.ptr<!s32i>
+// CIR: %[[CONST_1:.*]] = cir.const #cir.int<1> : !s32i
+// CIR: cir.store{{.*}} %[[CONST_1]], %[[ELEM_0_PTR]] : !s32i, !cir.ptr<!s32i>
+// CIR: %[[ELEM_1_PTR:.*]] = cir.get_member %[[C_ADDR]][1] {name = "b"} : !cir.ptr<!rec_CompleteS> -> !cir.ptr<!s8i>
+
+// LLVM: %[[A_ADDR:.*]] = alloca %struct.CompleteS, i64 1, align 4
+// LLVM: %[[B_ADDR:.*]] = alloca %struct.Container, i64 1, align 4
+// LLVM: %[[C_ADDR:.*]] = getelementptr %struct.Container, ptr %[[B_ADDR]], i32 0, i32 0
+// LLVM: call void @_ZN9CompleteSC1ERKS_(ptr %[[C_ADDR]], ptr %[[A_ADDR]])
+// LLVM: %[[ELEM_0_PTR:.*]] = getelementptr %struct.CompleteS, ptr %[[C_ADDR]], i32 0, i32 0
+// LLVM: store i32 1, ptr %[[ELEM_0_PTR]], align 4
+// LLVM: %[[ELEM_1_PTR:.*]] = getelementptr %struct.CompleteS, ptr %[[C_ADDR]], i32 0, i32 1
+
+// OGCG: %[[A_ADDR:.*]] = alloca %struct.CompleteS, align 4
+// OGCG: %[[B_ADDR:.*]] = alloca %struct.Container, align 4
+// OGCG: %[[C_ADDR:.*]] = getelementptr inbounds nuw %struct.Container, ptr %[[B_ADDR]], i32 0, i32 0
+// OGCG: call void @llvm.memcpy.p0.p0.i64(ptr align 4 %[[C_ADDR]], ptr align 4 %[[A_ADDR]], i64 8, i1 false)
+// OGCG: %[[ELEM_0_PTR:.*]] = getelementptr inbounds nuw %struct.CompleteS, ptr %[[C_ADDR]], i32 0, i32 0
+// OGCG: store i32 1, ptr %[[ELEM_0_PTR]], align 4
+// OGCG: %[[ELEM_1_PTR:.*]] = getelementptr inbounds nuw %struct.CompleteS, ptr %[[C_ADDR]], i32 0, i32 1
+
void atomic_init() {
_Atomic CompleteS a;
__c11_atomic_init(&a, {});
diff --git a/compiler-rt/test/asan/TestCases/wcscat.cpp b/compiler-rt/test/asan/TestCases/wcscat.cpp
index f0a8ec1..dcdff88 100644
--- a/compiler-rt/test/asan/TestCases/wcscat.cpp
+++ b/compiler-rt/test/asan/TestCases/wcscat.cpp
@@ -1,26 +1,26 @@
-// RUN: %clangxx_asan -O0 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O1 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O2 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O3 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O0 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O1 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O2 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O3 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
#include <stdio.h>
#include <wchar.h>
int main() {
- const wchar_t *start = L"X means ";
- const wchar_t *append = L"dog";
+ wchar_t *start = L"X means ";
+ wchar_t *append = L"dog";
wchar_t goodDst[12];
wcscpy(goodDst, start);
wcscat(goodDst, append);
wchar_t badDst[9];
wcscpy(badDst, start);
- fprintf(stderr, "Good so far.\n");
+ printf("Good so far.\n");
// CHECK: Good so far.
- fflush(stderr);
+ fflush(stdout);
wcscat(badDst, append); // Boom!
// CHECK: ERROR: AddressSanitizer: stack-buffer-overflow on address [[ADDR:0x[0-9a-f]+]] at pc {{0x[0-9a-f]+}} bp {{0x[0-9a-f]+}} sp {{0x[0-9a-f]+}}
- // CHECK: WRITE of size {{[0-9]+}} at [[ADDR]] thread T0
- // CHECK: #0 {{0x[0-9a-f]+}} in wcscat
+ // CHECK: WRITE of size {{[0-9]+}} at [[ADDR:0x[0-9a-f]+]] thread T0
+ // CHECK: #0 [[ADDR:0x[0-9a-f]+]] in wcscat{{.*}}sanitizer_common_interceptors.inc:{{[0-9]+}}
printf("Should have failed with ASAN error.\n");
} \ No newline at end of file
diff --git a/compiler-rt/test/asan/TestCases/wcscpy.cpp b/compiler-rt/test/asan/TestCases/wcscpy.cpp
index a280d29..414d833 100644
--- a/compiler-rt/test/asan/TestCases/wcscpy.cpp
+++ b/compiler-rt/test/asan/TestCases/wcscpy.cpp
@@ -1,23 +1,23 @@
-// RUN: %clangxx_asan -O0 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O1 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O2 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O3 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O0 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O1 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O2 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O3 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
#include <stdio.h>
#include <wchar.h>
int main() {
- const wchar_t *src = L"X means dog";
+ wchar_t *src = L"X means dog";
wchar_t goodDst[12];
wcscpy(goodDst, src);
wchar_t badDst[7];
- fprintf(stderr, "Good so far.\n");
+ printf("Good so far.\n");
// CHECK: Good so far.
- fflush(stderr);
+ fflush(stdout);
wcscpy(badDst, src); // Boom!
- // CHECK: ERROR: AddressSanitizer: stack-buffer-overflow on address [[ADDR:0x[0-9a-f]+]] at pc {{0x[0-9a-f]+}} bp {{0x[0-9a-f]+}} sp {{0x[0-9a-f]+}}
- // CHECK: WRITE of size {{[0-9]+}} at [[ADDR]] thread T0
- // CHECK: #0 {{0x[0-9a-f]+}} in wcscpy
+ // CHECK:ERROR: AddressSanitizer: stack-buffer-overflow on address [[ADDR:0x[0-9a-f]+]] at pc {{0x[0-9a-f]+}} bp {{0x[0-9a-f]+}} sp {{0x[0-9a-f]+}}
+ // CHECK: WRITE of size {{[0-9]+}} at [[ADDR:0x[0-9a-f]+]] thread T0
+ // CHECK: #0 [[ADDR:0x[0-9a-f]+]] in wcscpy{{.*}}asan_interceptors.cpp:{{[0-9]+}}
printf("Should have failed with ASAN error.\n");
} \ No newline at end of file
diff --git a/compiler-rt/test/asan/TestCases/wcsncat.cpp b/compiler-rt/test/asan/TestCases/wcsncat.cpp
index eb7d095..3ab7fc8 100644
--- a/compiler-rt/test/asan/TestCases/wcsncat.cpp
+++ b/compiler-rt/test/asan/TestCases/wcsncat.cpp
@@ -1,14 +1,14 @@
-// RUN: %clangxx_asan -O0 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O1 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O2 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O3 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O0 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O1 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O2 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O3 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
#include <stdio.h>
#include <wchar.h>
int main() {
- const wchar_t *start = L"X means ";
- const wchar_t *append = L"dog";
+ wchar_t *start = L"X means ";
+ wchar_t *append = L"dog";
wchar_t goodDst[15];
wcscpy(goodDst, start);
wcsncat(goodDst, append, 5);
@@ -16,12 +16,12 @@ int main() {
wchar_t badDst[11];
wcscpy(badDst, start);
wcsncat(badDst, append, 1);
- fprintf(stderr, "Good so far.\n");
+ printf("Good so far.\n");
// CHECK: Good so far.
- fflush(stderr);
+ fflush(stdout);
wcsncat(badDst, append, 3); // Boom!
// CHECK: ERROR: AddressSanitizer: stack-buffer-overflow on address [[ADDR:0x[0-9a-f]+]] at pc {{0x[0-9a-f]+}} bp {{0x[0-9a-f]+}} sp {{0x[0-9a-f]+}}
- // CHECK: WRITE of size {{[0-9]+}} at [[ADDR]] thread T0
- // CHECK: #0 {{0x[0-9a-f]+}} in wcsncat
+ // CHECK: WRITE of size {{[0-9]+}} at [[ADDR:0x[0-9a-f]+]] thread T0
+ // CHECK: #0 [[ADDR:0x[0-9a-f]+]] in wcsncat{{.*}}sanitizer_common_interceptors.inc:{{[0-9]+}}
printf("Should have failed with ASAN error.\n");
} \ No newline at end of file
diff --git a/compiler-rt/test/asan/TestCases/wcsncpy.cpp b/compiler-rt/test/asan/TestCases/wcsncpy.cpp
index 1106bf5..6177b72 100644
--- a/compiler-rt/test/asan/TestCases/wcsncpy.cpp
+++ b/compiler-rt/test/asan/TestCases/wcsncpy.cpp
@@ -1,25 +1,25 @@
-// RUN: %clangxx_asan -O0 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O1 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O2 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
-// RUN: %clangxx_asan -O3 %s -o %t && not %env_asan_opts=log_to_stderr=1 %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O0 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O1 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O2 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
+// RUN: %clangxx_asan -O3 %s -o %t && not %run %t 2>&1 | FileCheck %s --check-prefix=CHECK
#include <stdio.h>
#include <wchar.h>
int main() {
- const wchar_t *src = L"X means dog";
+ wchar_t *src = L"X means dog";
wchar_t goodDst[12];
wcsncpy(goodDst, src, 12);
wchar_t badDst[7];
wcsncpy(badDst, src, 7); // This should still work.
- fprintf(stderr, "Good so far.\n");
+ printf("Good so far.\n");
// CHECK: Good so far.
- fflush(stderr);
+ fflush(stdout);
wcsncpy(badDst, src, 15); // Boom!
- // CHECK: ERROR: AddressSanitizer: stack-buffer-overflow on address [[ADDR:0x[0-9a-f]+]] at pc {{0x[0-9a-f]+}} bp {{0x[0-9a-f]+}} sp {{0x[0-9a-f]+}}
- // CHECK: WRITE of size {{[0-9]+}} at [[ADDR]] thread T0
- // CHECK: #0 {{0x[0-9a-f]+}} in wcsncpy
+ // CHECK:ERROR: AddressSanitizer: stack-buffer-overflow on address [[ADDR:0x[0-9a-f]+]] at pc {{0x[0-9a-f]+}} bp {{0x[0-9a-f]+}} sp {{0x[0-9a-f]+}}
+ // CHECK: WRITE of size {{[0-9]+}} at [[ADDR:0x[0-9a-f]+]] thread T0
+ // CHECK: #0 [[ADDR:0x[0-9a-f]+]] in wcsncpy{{.*}}asan_interceptors.cpp:{{[0-9]+}}
printf("Should have failed with ASAN error.\n");
} \ No newline at end of file
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index cb6bfb2..56a3d6d 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -3903,8 +3903,7 @@ void LoopVectorizationPlanner::emitInvalidCostRemarks(
if (VF.isScalar())
continue;
- VPCostContext CostCtx(CM.TTI, *CM.TLI, *Plan, CM, CM.CostKind,
- *CM.PSE.getSE());
+ VPCostContext CostCtx(CM.TTI, *CM.TLI, *Plan, CM, CM.CostKind);
precomputeCosts(*Plan, VF, CostCtx);
auto Iter = vp_depth_first_deep(Plan->getVectorLoopRegion()->getEntry());
for (VPBasicBlock *VPBB : VPBlockUtils::blocksOnly<VPBasicBlock>(Iter)) {
@@ -4161,8 +4160,7 @@ VectorizationFactor LoopVectorizationPlanner::selectVectorizationFactor() {
// Add on other costs that are modelled in VPlan, but not in the legacy
// cost model.
- VPCostContext CostCtx(CM.TTI, *CM.TLI, *P, CM, CM.CostKind,
- *CM.PSE.getSE());
+ VPCostContext CostCtx(CM.TTI, *CM.TLI, *P, CM, CM.CostKind);
VPRegionBlock *VectorRegion = P->getVectorLoopRegion();
assert(VectorRegion && "Expected to have a vector region!");
for (VPBasicBlock *VPBB : VPBlockUtils::blocksOnly<VPBasicBlock>(
@@ -6854,7 +6852,7 @@ LoopVectorizationPlanner::precomputeCosts(VPlan &Plan, ElementCount VF,
InstructionCost LoopVectorizationPlanner::cost(VPlan &Plan,
ElementCount VF) const {
- VPCostContext CostCtx(CM.TTI, *CM.TLI, Plan, CM, CM.CostKind, *PSE.getSE());
+ VPCostContext CostCtx(CM.TTI, *CM.TLI, Plan, CM, CM.CostKind);
InstructionCost Cost = precomputeCosts(Plan, VF, CostCtx);
// Now compute and add the VPlan-based cost.
@@ -7087,8 +7085,7 @@ VectorizationFactor LoopVectorizationPlanner::computeBestVF() {
// simplifications not accounted for in the legacy cost model. If that's the
// case, don't trigger the assertion, as the extra simplifications may cause a
// different VF to be picked by the VPlan-based cost model.
- VPCostContext CostCtx(CM.TTI, *CM.TLI, BestPlan, CM, CM.CostKind,
- *CM.PSE.getSE());
+ VPCostContext CostCtx(CM.TTI, *CM.TLI, BestPlan, CM, CM.CostKind);
precomputeCosts(BestPlan, BestFactor.Width, CostCtx);
// Verify that the VPlan-based and legacy cost models agree, except for VPlans
// with early exits and plans with additional VPlan simplifications. The
@@ -8624,8 +8621,7 @@ VPlanPtr LoopVectorizationPlanner::tryToBuildVPlanWithVPRecipes(
// TODO: Enable following transform when the EVL-version of extended-reduction
// and mulacc-reduction are implemented.
if (!CM.foldTailWithEVL()) {
- VPCostContext CostCtx(CM.TTI, *CM.TLI, *Plan, CM, CM.CostKind,
- *CM.PSE.getSE());
+ VPCostContext CostCtx(CM.TTI, *CM.TLI, *Plan, CM, CM.CostKind);
VPlanTransforms::runPass(VPlanTransforms::convertToAbstractRecipes, *Plan,
CostCtx, Range);
}
@@ -10079,7 +10075,7 @@ bool LoopVectorizePass::processLoop(Loop *L) {
bool ForceVectorization =
Hints.getForce() == LoopVectorizeHints::FK_Enabled;
VPCostContext CostCtx(CM.TTI, *CM.TLI, LVP.getPlanFor(VF.Width), CM,
- CM.CostKind, *CM.PSE.getSE());
+ CM.CostKind);
if (!ForceVectorization &&
!isOutsideLoopWorkProfitable(Checks, VF, L, PSE, CostCtx,
LVP.getPlanFor(VF.Width), SEL,
diff --git a/llvm/lib/Transforms/Vectorize/VPlan.cpp b/llvm/lib/Transforms/Vectorize/VPlan.cpp
index 2555ebe..07b191a 100644
--- a/llvm/lib/Transforms/Vectorize/VPlan.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlan.cpp
@@ -1772,8 +1772,7 @@ VPCostContext::getOperandInfo(VPValue *V) const {
}
InstructionCost VPCostContext::getScalarizationOverhead(
- Type *ResultTy, ArrayRef<const VPValue *> Operands, ElementCount VF,
- bool AlwaysIncludeReplicatingR) {
+ Type *ResultTy, ArrayRef<const VPValue *> Operands, ElementCount VF) {
if (VF.isScalar())
return 0;
@@ -1793,11 +1792,7 @@ InstructionCost VPCostContext::getScalarizationOverhead(
SmallPtrSet<const VPValue *, 4> UniqueOperands;
SmallVector<Type *> Tys;
for (auto *Op : Operands) {
- if (Op->isLiveIn() ||
- (!AlwaysIncludeReplicatingR &&
- isa<VPReplicateRecipe, VPPredInstPHIRecipe>(Op)) ||
- (isa<VPReplicateRecipe>(Op) &&
- cast<VPReplicateRecipe>(Op)->getOpcode() == Instruction::Load) ||
+ if (Op->isLiveIn() || isa<VPReplicateRecipe, VPPredInstPHIRecipe>(Op) ||
!UniqueOperands.insert(Op).second)
continue;
Tys.push_back(toVectorizedTy(Types.inferScalarType(Op), VF));
diff --git a/llvm/lib/Transforms/Vectorize/VPlanHelpers.h b/llvm/lib/Transforms/Vectorize/VPlanHelpers.h
index 1580a3b..fc1a09e 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanHelpers.h
+++ b/llvm/lib/Transforms/Vectorize/VPlanHelpers.h
@@ -349,14 +349,12 @@ struct VPCostContext {
LoopVectorizationCostModel &CM;
SmallPtrSet<Instruction *, 8> SkipCostComputation;
TargetTransformInfo::TargetCostKind CostKind;
- ScalarEvolution &SE;
VPCostContext(const TargetTransformInfo &TTI, const TargetLibraryInfo &TLI,
const VPlan &Plan, LoopVectorizationCostModel &CM,
- TargetTransformInfo::TargetCostKind CostKind,
- ScalarEvolution &SE)
+ TargetTransformInfo::TargetCostKind CostKind)
: TTI(TTI), TLI(TLI), Types(Plan), LLVMCtx(Plan.getContext()), CM(CM),
- CostKind(CostKind), SE(SE) {}
+ CostKind(CostKind) {}
/// Return the cost for \p UI with \p VF using the legacy cost model as
/// fallback until computing the cost of all recipes migrates to VPlan.
@@ -376,12 +374,10 @@ struct VPCostContext {
/// Estimate the overhead of scalarizing a recipe with result type \p ResultTy
/// and \p Operands with \p VF. This is a convenience wrapper for the
- /// type-based getScalarizationOverhead API. If \p AlwaysIncludeReplicatingR
- /// is true, always compute the cost of scalarizing replicating operands.
- InstructionCost
- getScalarizationOverhead(Type *ResultTy, ArrayRef<const VPValue *> Operands,
- ElementCount VF,
- bool AlwaysIncludeReplicatingR = false);
+ /// type-based getScalarizationOverhead API.
+ InstructionCost getScalarizationOverhead(Type *ResultTy,
+ ArrayRef<const VPValue *> Operands,
+ ElementCount VF);
};
/// This class can be used to assign names to VPValues. For VPValues without
diff --git a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
index a88cffc..67b9244 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
@@ -40,7 +40,6 @@
#include <cassert>
using namespace llvm;
-using namespace llvm::VPlanPatternMatch;
using VectorParts = SmallVector<Value *, 2>;
@@ -304,6 +303,7 @@ VPPartialReductionRecipe::computeCost(ElementCount VF,
VPRecipeBase *OpR = Op->getDefiningRecipe();
// If the partial reduction is predicated, a select will be operand 0
+ using namespace llvm::VPlanPatternMatch;
if (match(getOperand(1), m_Select(m_VPValue(), m_VPValue(Op), m_VPValue()))) {
OpR = Op->getDefiningRecipe();
}
@@ -1963,6 +1963,7 @@ InstructionCost VPWidenSelectRecipe::computeCost(ElementCount VF,
Type *VectorTy = toVectorTy(Ctx.Types.inferScalarType(this), VF);
VPValue *Op0, *Op1;
+ using namespace llvm::VPlanPatternMatch;
if (!ScalarCond && ScalarTy->getScalarSizeInBits() == 1 &&
(match(this, m_LogicalAnd(m_VPValue(Op0), m_VPValue(Op1))) ||
match(this, m_LogicalOr(m_VPValue(Op0), m_VPValue(Op1))))) {
@@ -3110,62 +3111,6 @@ bool VPReplicateRecipe::shouldPack() const {
});
}
-/// Returns true if \p Ptr is a pointer computation for which the legacy cost
-/// model computes a SCEV expression when computing the address cost.
-static bool shouldUseAddressAccessSCEV(const VPValue *Ptr) {
- auto *PtrR = Ptr->getDefiningRecipe();
- if (!PtrR || !((isa<VPReplicateRecipe>(PtrR) &&
- cast<VPReplicateRecipe>(PtrR)->getOpcode() ==
- Instruction::GetElementPtr) ||
- isa<VPWidenGEPRecipe>(PtrR) ||
- match(Ptr, m_GetElementPtr(m_VPValue(), m_VPValue()))))
- return false;
-
- // We are looking for a GEP where all indices are either loop invariant or
- // inductions.
- for (VPValue *Opd : drop_begin(PtrR->operands())) {
- if (!Opd->isDefinedOutsideLoopRegions() &&
- !isa<VPScalarIVStepsRecipe, VPWidenIntOrFpInductionRecipe>(Opd))
- return false;
- }
-
- return true;
-}
-
-/// Returns true if \p V is used as part of the address of another load or
-/// store.
-static bool isUsedByLoadStoreAddress(const VPUser *V) {
- SmallPtrSet<const VPUser *, 4> Seen;
- SmallVector<const VPUser *> WorkList = {V};
-
- while (!WorkList.empty()) {
- auto *Cur = dyn_cast<VPSingleDefRecipe>(WorkList.pop_back_val());
- if (!Cur || !Seen.insert(Cur).second)
- continue;
-
- for (VPUser *U : Cur->users()) {
- if (auto *InterleaveR = dyn_cast<VPInterleaveBase>(U))
- if (InterleaveR->getAddr() == Cur)
- return true;
- if (auto *RepR = dyn_cast<VPReplicateRecipe>(U)) {
- if (RepR->getOpcode() == Instruction::Load &&
- RepR->getOperand(0) == Cur)
- return true;
- if (RepR->getOpcode() == Instruction::Store &&
- RepR->getOperand(1) == Cur)
- return true;
- }
- if (auto *MemR = dyn_cast<VPWidenMemoryRecipe>(U)) {
- if (MemR->getAddr() == Cur && MemR->isConsecutive())
- return true;
- }
- }
-
- append_range(WorkList, cast<VPSingleDefRecipe>(Cur)->users());
- }
- return false;
-}
-
InstructionCost VPReplicateRecipe::computeCost(ElementCount VF,
VPCostContext &Ctx) const {
Instruction *UI = cast<Instruction>(getUnderlyingValue());
@@ -3273,59 +3218,21 @@ InstructionCost VPReplicateRecipe::computeCost(ElementCount VF,
}
case Instruction::Load:
case Instruction::Store: {
- if (VF.isScalable() && !isSingleScalar())
- return InstructionCost::getInvalid();
-
+ if (isSingleScalar()) {
+ bool IsLoad = UI->getOpcode() == Instruction::Load;
+ Type *ValTy = Ctx.Types.inferScalarType(IsLoad ? this : getOperand(0));
+ Type *ScalarPtrTy = Ctx.Types.inferScalarType(getOperand(IsLoad ? 0 : 1));
+ const Align Alignment = getLoadStoreAlignment(UI);
+ unsigned AS = getLoadStoreAddressSpace(UI);
+ TTI::OperandValueInfo OpInfo = TTI::getOperandInfo(UI->getOperand(0));
+ InstructionCost ScalarMemOpCost = Ctx.TTI.getMemoryOpCost(
+ UI->getOpcode(), ValTy, Alignment, AS, Ctx.CostKind, OpInfo, UI);
+ return ScalarMemOpCost + Ctx.TTI.getAddressComputationCost(
+ ScalarPtrTy, nullptr, nullptr, Ctx.CostKind);
+ }
// TODO: See getMemInstScalarizationCost for how to handle replicating and
// predicated cases.
- const VPRegionBlock *ParentRegion = getParent()->getParent();
- if (ParentRegion && ParentRegion->isReplicator())
- break;
-
- bool IsLoad = UI->getOpcode() == Instruction::Load;
- const VPValue *PtrOp = getOperand(!IsLoad);
- // TODO: Handle cases where we need to pass a SCEV to
- // getAddressComputationCost.
- if (shouldUseAddressAccessSCEV(PtrOp))
- break;
-
- Type *ValTy = Ctx.Types.inferScalarType(IsLoad ? this : getOperand(0));
- Type *ScalarPtrTy = Ctx.Types.inferScalarType(PtrOp);
- const Align Alignment = getLoadStoreAlignment(UI);
- unsigned AS = getLoadStoreAddressSpace(UI);
- TTI::OperandValueInfo OpInfo = TTI::getOperandInfo(UI->getOperand(0));
- InstructionCost ScalarMemOpCost = Ctx.TTI.getMemoryOpCost(
- UI->getOpcode(), ValTy, Alignment, AS, Ctx.CostKind, OpInfo);
-
- Type *PtrTy = isSingleScalar() ? ScalarPtrTy : toVectorTy(ScalarPtrTy, VF);
- bool UsedByLoadStoreAddress = isUsedByLoadStoreAddress(this);
- InstructionCost ScalarCost =
- ScalarMemOpCost + Ctx.TTI.getAddressComputationCost(
- PtrTy, UsedByLoadStoreAddress ? nullptr : &Ctx.SE,
- nullptr, Ctx.CostKind);
- if (isSingleScalar())
- return ScalarCost;
-
- SmallVector<const VPValue *> OpsToScalarize;
- Type *ResultTy = Type::getVoidTy(PtrTy->getContext());
- // Set ResultTy and OpsToScalarize, if scalarization is needed. Currently we
- // don't assign scalarization overhead in general, if the target prefers
- // vectorized addressing or the loaded value is used as part of an address
- // of another load or store.
- bool PreferVectorizedAddressing = Ctx.TTI.prefersVectorizedAddressing();
- if (PreferVectorizedAddressing || !UsedByLoadStoreAddress) {
- bool EfficientVectorLoadStore =
- Ctx.TTI.supportsEfficientVectorElementLoadStore();
- if (!(IsLoad && !PreferVectorizedAddressing) &&
- !(!IsLoad && EfficientVectorLoadStore))
- append_range(OpsToScalarize, operands());
-
- if (!EfficientVectorLoadStore)
- ResultTy = Ctx.Types.inferScalarType(this);
- }
-
- return (ScalarCost * VF.getFixedValue()) +
- Ctx.getScalarizationOverhead(ResultTy, OpsToScalarize, VF, true);
+ break;
}
}
diff --git a/llvm/test/Transforms/LoopVectorize/X86/replicating-load-store-costs.ll b/llvm/test/Transforms/LoopVectorize/X86/replicating-load-store-costs.ll
index f5329cf..8784873 100644
--- a/llvm/test/Transforms/LoopVectorize/X86/replicating-load-store-costs.ll
+++ b/llvm/test/Transforms/LoopVectorize/X86/replicating-load-store-costs.ll
@@ -454,132 +454,6 @@ exit:
ret void
}
-declare i1 @cond()
-
-define double @test_load_used_by_other_load_scev(ptr %ptr.a, ptr %ptr.b, ptr %ptr.c) {
-; I64-LABEL: define double @test_load_used_by_other_load_scev(
-; I64-SAME: ptr [[PTR_A:%.*]], ptr [[PTR_B:%.*]], ptr [[PTR_C:%.*]]) {
-; I64-NEXT: [[ENTRY:.*]]:
-; I64-NEXT: br label %[[OUTER_LOOP:.*]]
-; I64: [[OUTER_LOOP_LOOPEXIT:.*]]:
-; I64-NEXT: br label %[[OUTER_LOOP]]
-; I64: [[OUTER_LOOP]]:
-; I64-NEXT: [[ACCUM:%.*]] = phi double [ 0.000000e+00, %[[ENTRY]] ], [ [[TMP29:%.*]], %[[OUTER_LOOP_LOOPEXIT]] ]
-; I64-NEXT: [[COND:%.*]] = call i1 @cond()
-; I64-NEXT: br i1 [[COND]], label %[[INNER_LOOP_PREHEADER:.*]], label %[[EXIT:.*]]
-; I64: [[INNER_LOOP_PREHEADER]]:
-; I64-NEXT: br label %[[VECTOR_PH:.*]]
-; I64: [[VECTOR_PH]]:
-; I64-NEXT: br label %[[VECTOR_BODY:.*]]
-; I64: [[VECTOR_BODY]]:
-; I64-NEXT: [[TMP0:%.*]] = add i64 0, 1
-; I64-NEXT: [[TMP1:%.*]] = add i64 1, 1
-; I64-NEXT: [[TMP2:%.*]] = getelementptr i8, ptr [[PTR_C]], i64 [[TMP0]]
-; I64-NEXT: [[TMP3:%.*]] = getelementptr i8, ptr [[PTR_C]], i64 [[TMP1]]
-; I64-NEXT: [[TMP4:%.*]] = getelementptr i64, ptr [[PTR_A]], i64 [[TMP0]]
-; I64-NEXT: [[TMP5:%.*]] = getelementptr i64, ptr [[PTR_A]], i64 [[TMP1]]
-; I64-NEXT: [[TMP6:%.*]] = load i64, ptr [[TMP4]], align 8
-; I64-NEXT: [[TMP7:%.*]] = load i64, ptr [[TMP5]], align 8
-; I64-NEXT: [[TMP8:%.*]] = getelementptr double, ptr [[PTR_B]], i64 [[TMP6]]
-; I64-NEXT: [[TMP9:%.*]] = getelementptr double, ptr [[PTR_B]], i64 [[TMP7]]
-; I64-NEXT: [[TMP10:%.*]] = load double, ptr [[PTR_A]], align 8
-; I64-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x double> poison, double [[TMP10]], i64 0
-; I64-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x double> [[BROADCAST_SPLATINSERT]], <2 x double> poison, <2 x i32> zeroinitializer
-; I64-NEXT: [[TMP11:%.*]] = fadd <2 x double> [[BROADCAST_SPLAT]], zeroinitializer
-; I64-NEXT: [[TMP12:%.*]] = getelementptr i8, ptr [[TMP2]], i64 8
-; I64-NEXT: [[TMP13:%.*]] = getelementptr i8, ptr [[TMP3]], i64 8
-; I64-NEXT: [[TMP14:%.*]] = load double, ptr [[TMP12]], align 8
-; I64-NEXT: [[TMP15:%.*]] = load double, ptr [[TMP13]], align 8
-; I64-NEXT: [[TMP16:%.*]] = insertelement <2 x double> poison, double [[TMP14]], i32 0
-; I64-NEXT: [[TMP17:%.*]] = insertelement <2 x double> [[TMP16]], double [[TMP15]], i32 1
-; I64-NEXT: [[TMP18:%.*]] = fmul <2 x double> [[TMP11]], zeroinitializer
-; I64-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <2 x double> poison, double [[ACCUM]], i64 0
-; I64-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <2 x double> [[BROADCAST_SPLATINSERT1]], <2 x double> poison, <2 x i32> zeroinitializer
-; I64-NEXT: [[TMP19:%.*]] = shufflevector <2 x double> [[BROADCAST_SPLAT2]], <2 x double> [[TMP18]], <2 x i32> <i32 1, i32 2>
-; I64-NEXT: [[TMP20:%.*]] = fmul <2 x double> [[TMP17]], zeroinitializer
-; I64-NEXT: [[TMP21:%.*]] = fadd <2 x double> [[TMP20]], zeroinitializer
-; I64-NEXT: [[TMP22:%.*]] = fadd <2 x double> [[TMP21]], splat (double 1.000000e+00)
-; I64-NEXT: [[TMP23:%.*]] = load double, ptr [[TMP8]], align 8
-; I64-NEXT: [[TMP24:%.*]] = load double, ptr [[TMP9]], align 8
-; I64-NEXT: [[TMP25:%.*]] = insertelement <2 x double> poison, double [[TMP23]], i32 0
-; I64-NEXT: [[TMP26:%.*]] = insertelement <2 x double> [[TMP25]], double [[TMP24]], i32 1
-; I64-NEXT: [[TMP27:%.*]] = fdiv <2 x double> [[TMP26]], [[TMP22]]
-; I64-NEXT: [[TMP28:%.*]] = fsub <2 x double> [[TMP19]], [[TMP27]]
-; I64-NEXT: br label %[[MIDDLE_BLOCK:.*]]
-; I64: [[MIDDLE_BLOCK]]:
-; I64-NEXT: [[TMP29]] = extractelement <2 x double> [[TMP28]], i32 1
-; I64-NEXT: br label %[[OUTER_LOOP_LOOPEXIT]]
-; I64: [[EXIT]]:
-; I64-NEXT: ret double [[ACCUM]]
-;
-; I32-LABEL: define double @test_load_used_by_other_load_scev(
-; I32-SAME: ptr [[PTR_A:%.*]], ptr [[PTR_B:%.*]], ptr [[PTR_C:%.*]]) {
-; I32-NEXT: [[ENTRY:.*]]:
-; I32-NEXT: br label %[[OUTER_LOOP:.*]]
-; I32: [[OUTER_LOOP]]:
-; I32-NEXT: [[ACCUM:%.*]] = phi double [ 0.000000e+00, %[[ENTRY]] ], [ [[RESULT:%.*]], %[[INNER_LOOP:.*]] ]
-; I32-NEXT: [[COND:%.*]] = call i1 @cond()
-; I32-NEXT: br i1 [[COND]], label %[[INNER_LOOP]], label %[[EXIT:.*]]
-; I32: [[INNER_LOOP]]:
-; I32-NEXT: [[IV:%.*]] = phi i64 [ 0, %[[OUTER_LOOP]] ], [ [[IV_NEXT:%.*]], %[[INNER_LOOP]] ]
-; I32-NEXT: [[ACCUM_INNER:%.*]] = phi double [ [[ACCUM]], %[[OUTER_LOOP]] ], [ [[MUL1:%.*]], %[[INNER_LOOP]] ]
-; I32-NEXT: [[IDX_PLUS1:%.*]] = add i64 [[IV]], 1
-; I32-NEXT: [[GEP_C:%.*]] = getelementptr i8, ptr [[PTR_C]], i64 [[IDX_PLUS1]]
-; I32-NEXT: [[GEP_A_I64:%.*]] = getelementptr i64, ptr [[PTR_A]], i64 [[IDX_PLUS1]]
-; I32-NEXT: [[LOAD_IDX:%.*]] = load i64, ptr [[GEP_A_I64]], align 8
-; I32-NEXT: [[GEP_B:%.*]] = getelementptr double, ptr [[PTR_B]], i64 [[LOAD_IDX]]
-; I32-NEXT: [[LOAD_A:%.*]] = load double, ptr [[PTR_A]], align 8
-; I32-NEXT: [[ADD1:%.*]] = fadd double [[LOAD_A]], 0.000000e+00
-; I32-NEXT: [[GEP_C_OFFSET:%.*]] = getelementptr i8, ptr [[GEP_C]], i64 8
-; I32-NEXT: [[LOAD_C:%.*]] = load double, ptr [[GEP_C_OFFSET]], align 8
-; I32-NEXT: [[MUL1]] = fmul double [[ADD1]], 0.000000e+00
-; I32-NEXT: [[MUL2:%.*]] = fmul double [[LOAD_C]], 0.000000e+00
-; I32-NEXT: [[ADD2:%.*]] = fadd double [[MUL2]], 0.000000e+00
-; I32-NEXT: [[ADD3:%.*]] = fadd double [[ADD2]], 1.000000e+00
-; I32-NEXT: [[LOAD_B:%.*]] = load double, ptr [[GEP_B]], align 8
-; I32-NEXT: [[DIV:%.*]] = fdiv double [[LOAD_B]], [[ADD3]]
-; I32-NEXT: [[RESULT]] = fsub double [[ACCUM_INNER]], [[DIV]]
-; I32-NEXT: [[IV_NEXT]] = add i64 [[IV]], 1
-; I32-NEXT: [[EXITCOND:%.*]] = icmp eq i64 [[IV]], 1
-; I32-NEXT: br i1 [[EXITCOND]], label %[[OUTER_LOOP]], label %[[INNER_LOOP]]
-; I32: [[EXIT]]:
-; I32-NEXT: ret double [[ACCUM]]
-;
-entry:
- br label %outer.loop
-
-outer.loop:
- %accum = phi double [ 0.0, %entry ], [ %result, %inner.loop ]
- %cond = call i1 @cond()
- br i1 %cond, label %inner.loop, label %exit
-
-inner.loop:
- %iv = phi i64 [ 0, %outer.loop ], [ %iv.next, %inner.loop ]
- %accum.inner = phi double [ %accum, %outer.loop ], [ %mul1, %inner.loop ]
- %idx.plus1 = add i64 %iv, 1
- %gep.c = getelementptr i8, ptr %ptr.c, i64 %idx.plus1
- %gep.a.i64 = getelementptr i64, ptr %ptr.a, i64 %idx.plus1
- %load.idx = load i64, ptr %gep.a.i64, align 8
- %gep.b = getelementptr double, ptr %ptr.b, i64 %load.idx
- %load.a = load double, ptr %ptr.a, align 8
- %add1 = fadd double %load.a, 0.000000e+00
- %gep.c.offset = getelementptr i8, ptr %gep.c, i64 8
- %load.c = load double, ptr %gep.c.offset, align 8
- %mul1 = fmul double %add1, 0.000000e+00
- %mul2 = fmul double %load.c, 0.000000e+00
- %add2 = fadd double %mul2, 0.000000e+00
- %add3 = fadd double %add2, 1.000000e+00
- %load.b = load double, ptr %gep.b, align 8
- %div = fdiv double %load.b, %add3
- %result = fsub double %accum.inner, %div
- %iv.next = add i64 %iv, 1
- %exitcond = icmp eq i64 %iv, 1
- br i1 %exitcond, label %outer.loop, label %inner.loop
-
-exit:
- ret double %accum
-}
-
attributes #0 = { "target-cpu"="znver2" }
!0 = distinct !{!0, !1}
diff --git a/llvm/tools/llvm-c-test/debuginfo.c b/llvm/tools/llvm-c-test/debuginfo.c
index e376d82..a2f4b3e 100644
--- a/llvm/tools/llvm-c-test/debuginfo.c
+++ b/llvm/tools/llvm-c-test/debuginfo.c
@@ -328,8 +328,10 @@ int llvm_test_dibuilder(void) {
// Test that LLVMGetFirstDbgRecord and LLVMGetLastDbgRecord return NULL for
// instructions without debug info.
LLVMDbgRecordRef Phi1FirstDbgRecord = LLVMGetFirstDbgRecord(Phi1);
+ (void)Phi1FirstDbgRecord;
assert(Phi1FirstDbgRecord == NULL);
LLVMDbgRecordRef Phi1LastDbgRecord = LLVMGetLastDbgRecord(Phi1);
+ (void)Phi1LastDbgRecord;
assert(Phi1LastDbgRecord == NULL);
// Insert a non-phi before the `ret` but not before the debug records to
diff --git a/mlir/test/Dialect/Vector/canonicalize.mlir b/mlir/test/Dialect/Vector/canonicalize.mlir
index 5448976f..bccf5d5b 100644
--- a/mlir/test/Dialect/Vector/canonicalize.mlir
+++ b/mlir/test/Dialect/Vector/canonicalize.mlir
@@ -3377,6 +3377,9 @@ func.func @negative_from_elements_to_constant() -> vector<1x!llvm.ptr> {
// -----
+// `foldFromElementsToConstant` does not support `ub.poison`, so it bails out.
+// Instead, other folders apply here (e.g. `rewriteFromElementsAsBroadcast`).
+
// CHECK-LABEL: @negative_from_elements_poison
// CHECK: %[[VAL:.*]] = ub.poison : vector<2xf32>
// CHECK: return %[[VAL]] : vector<2xf32>
@@ -3388,6 +3391,9 @@ func.func @negative_from_elements_poison_f32() -> vector<2xf32> {
// -----
+// `foldFromElementsToConstant` does not support `ub.poison`, so it bails out.
+// Instead, other folders apply here (e.g. `rewriteFromElementsAsBroadcast`).
+
// CHECK-LABEL: @negative_from_elements_poison_i32
// CHECK: %[[VAL:.*]] = ub.poison : vector<2xi32>
// CHECK: return %[[VAL]] : vector<2xi32>
@@ -3399,6 +3405,9 @@ func.func @negative_from_elements_poison_i32() -> vector<2xi32> {
// -----
+// `foldFromElementsToConstant` does not support `ub.poison`, so it bails out.
+// Instead, other folders apply here (e.g. `rewriteFromElementsAsBroadcast`).
+
// CHECK-LABEL: @negative_from_elements_poison_constant_mix
// CHECK: %[[POISON:.*]] = ub.poison : f32
// CHECK: %[[CONST:.*]] = arith.constant 1.000000e+00 : f32
generated by cgit v1.1 at 2025-10-05 19:57:41 +0000