This is an archive of the discontinued LLVM Phabricator instance.

Differential D124676

[clang][OpenMP] Local variable alignment incorrect with align clause
ClosedPublic

Authored by ddpagan on Apr 29 2022, 8:23 AM.

Details

Reviewers
jdoerfert
ABataev
Commits
rG37471cf2c3fd: [clang][OpenMP] Local variable alignment incorrect with align clause
Summary

If alignment specified with align clause is less than natural alignment for list item type, the alignment should be set to the natural alignment.

See OMP5.1 specification, page 185, lines 7-10

Diff Detail

Event Timeline

ddpagan created this revision.Apr 29 2022, 8:23 AM
Herald added a project: Restricted Project. · View Herald TranscriptApr 29 2022, 8:23 AM
Herald added subscribers: guansong, yaxunl. · View Herald Transcript
ddpagan requested review of this revision.Apr 29 2022, 8:23 AM
Herald added a reviewer: jdoerfert. · View Herald TranscriptApr 29 2022, 8:23 AM
Herald added a subscriber: sstefan1. · View Herald Transcript
Harbormaster completed remote builds in B161988: Diff 426062.Apr 29 2022, 9:10 AM
ddpagan added a reviewer: ABataev.Apr 29 2022, 12:50 PM
ABataev added inline comments.Apr 29 2022, 1:08 PM
clang/lib/CodeGen/CGOpenMPRuntime.cpp
12161
if (!Alignment)
  return nullptr;
12170–12176
return llvm::ConstantInt::get(CGF.CGM.SizeTy, std::max(UserAlign, NaturalAlign.getQuantity()));
12215–12216

What if the type is a reference?

ddpagan updated this revision to Diff 426453.EditedMay 2 2022, 10:44 AM

Thanks for the review, Alexey. Suggested changes have been made.

ABataev accepted this revision.May 2 2022, 10:56 AM

LG

This revision is now accepted and ready to land.May 2 2022, 10:56 AM
Harbormaster completed remote builds in B162279: Diff 426453.May 2 2022, 11:53 AM
Closed by commit rG37471cf2c3fd: [clang][OpenMP] Local variable alignment incorrect with align clause (authored by ddpagan, committed by mikerice). · Explain WhyMay 3 2022, 1:20 PM
This revision was automatically updated to reflect the committed changes.
mikerice added a commit: rG37471cf2c3fd: [clang][OpenMP] Local variable alignment incorrect with align clause.
Herald added a project: Restricted Project. · View Herald TranscriptMay 3 2022, 1:20 PM
Herald added a subscriber: cfe-commits. · View Herald Transcript

Revision Contents

PathSize
clang/
lib/
CodeGen/
28 lines
test/
OpenMP/
20 lines

Diff 426453

clang/lib/CodeGen/CGOpenMPRuntime.cpp

  • This file is larger than 256 KB, so syntax highlighting is disabled by default.
Show First 20 Lines • Show All 12,147 Lines • ▼ Show 20 Linesstatic llvm::Value *getAllocatorVal(CodeGenFunction &CGF,
} else { } else {
// If no allocator specified, it defaults to the null allocator. // If no allocator specified, it defaults to the null allocator.
AllocVal = llvm::Constant::getNullValue( AllocVal = llvm::Constant::getNullValue(
CGF.CGM.getTypes().ConvertType(CGF.getContext().VoidPtrTy)); CGF.CGM.getTypes().ConvertType(CGF.getContext().VoidPtrTy));
} }
return AllocVal; return AllocVal;
} }
/// Given the allocate directive list item type and align clause value,
/// return appropriate alignment.
static llvm::Value *getAlignmentValue(CodeGenFunction &CGF, QualType ListItemTy,
const Expr *Alignment) {
if (!Alignment)
return nullptr;
ABataevUnsubmitted
Not Done
ReplyInline Actions
if (!Alignment)
  return nullptr;
ABataev: ``` if (!Alignment) return nullptr; ```
unsigned UserAlign =
Alignment->EvaluateKnownConstInt(CGF.getContext()).getExtValue();
CharUnits NaturalAlign = CGF.CGM.getNaturalTypeAlignment(ListItemTy);
// OpenMP5.1 pg 185 lines 7-10
// Each item in the align modifier list must be aligned to the maximum
// of the specified alignment and the type's natural alignment.
//
// If no alignment specified then use the natural alignment.
return llvm::ConstantInt::get(
CGF.CGM.SizeTy,
std::max<unsigned>(UserAlign, NaturalAlign.getQuantity()));
}
ABataevUnsubmitted
Not Done
ReplyInline Actions
return llvm::ConstantInt::get(CGF.CGM.SizeTy, std::max(UserAlign, NaturalAlign.getQuantity()));
ABataev: ``` return llvm::ConstantInt::get(CGF.CGM.SizeTy, std::max(UserAlign, NaturalAlign.getQuantity…
Address CGOpenMPRuntime::getAddressOfLocalVariable(CodeGenFunction &CGF, Address CGOpenMPRuntime::getAddressOfLocalVariable(CodeGenFunction &CGF,
const VarDecl *VD) { const VarDecl *VD) {
if (!VD) if (!VD)
return Address::invalid(); return Address::invalid();
Address UntiedAddr = Address::invalid(); Address UntiedAddr = Address::invalid();
Address UntiedRealAddr = Address::invalid(); Address UntiedRealAddr = Address::invalid();
auto It = FunctionToUntiedTaskStackMap.find(CGF.CurFn); auto It = FunctionToUntiedTaskStackMap.find(CGF.CurFn);
if (It != FunctionToUntiedTaskStackMap.end()) { if (It != FunctionToUntiedTaskStackMap.end()) {
Show All 22 Lines if (CVD->hasAttr<OMPAllocateDeclAttr>()) {
} else { } else {
CharUnits Sz = CGM.getContext().getTypeSizeInChars(CVD->getType()); CharUnits Sz = CGM.getContext().getTypeSizeInChars(CVD->getType());
Size = CGM.getSize(Sz.alignTo(Align)); Size = CGM.getSize(Sz.alignTo(Align));
} }
llvm::Value *ThreadID = getThreadID(CGF, CVD->getBeginLoc()); llvm::Value *ThreadID = getThreadID(CGF, CVD->getBeginLoc());
const auto *AA = CVD->getAttr<OMPAllocateDeclAttr>(); const auto *AA = CVD->getAttr<OMPAllocateDeclAttr>();
const Expr *Allocator = AA->getAllocator(); const Expr *Allocator = AA->getAllocator();
llvm::Value *AllocVal = getAllocatorVal(CGF, Allocator); llvm::Value *AllocVal = getAllocatorVal(CGF, Allocator);
llvm::Value *Alignment = llvm::Value *Alignment = getAlignmentValue(
AA->getAlignment() CGF, VD->getType().getNonReferenceType(), AA->getAlignment());
ABataevUnsubmitted
Not Done
ReplyInline Actions

What if the type is a reference?

ABataev: What if the type is a reference?
? CGF.Builder.CreateIntCast(CGF.EmitScalarExpr(AA->getAlignment()),
CGM.SizeTy, /*isSigned=*/false)
: nullptr;
SmallVector<llvm::Value *, 4> Args; SmallVector<llvm::Value *, 4> Args;
Args.push_back(ThreadID); Args.push_back(ThreadID);
if (Alignment) if (Alignment)
Args.push_back(Alignment); Args.push_back(Alignment);
Args.push_back(Size); Args.push_back(Size);
Args.push_back(AllocVal); Args.push_back(AllocVal);
llvm::omp::RuntimeFunction FnID = llvm::omp::RuntimeFunction FnID =
Alignment ? OMPRTL___kmpc_aligned_alloc : OMPRTL___kmpc_alloc; Alignment ? OMPRTL___kmpc_aligned_alloc : OMPRTL___kmpc_alloc;
▲ Show 20 Lines • Show All 885 Lines • Show Last 20 Lines

clang/test/OpenMP/align_clause_codegen.cpp

Show First 20 Lines • Show All 79 Lines • ▼ Show 20 Linesint template_test() {
return 0; return 0;
} }
#endif #endif
// CHECK-32-LABEL: define {{[^@]+}}@main // CHECK-32-LABEL: define {{[^@]+}}@main
// CHECK-32-SAME: () #[[ATTR0:[0-9]+]] { // CHECK-32-SAME: () #[[ATTR0:[0-9]+]] {
// CHECK-32-NEXT: entry: // CHECK-32-NEXT: entry:
// CHECK-32-NEXT: [[MYALLOC:%.*]] = alloca i32, align 4 // CHECK-32-NEXT: [[MYALLOC:%.*]] = alloca i32, align 4
// CHECK-32-NEXT: [[TMP0:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]]) // CHECK-32-NEXT: [[TMP0:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]])
// CHECK-32-NEXT: [[DOTFOO0__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 1, i32 20, i8* null) // CHECK-32-NEXT: [[DOTFOO0__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 4, i32 20, i8* null)
// CHECK-32-NEXT: [[DOTFOO0__ADDR:%.*]] = bitcast i8* [[DOTFOO0__VOID_ADDR]] to [5 x i32]* // CHECK-32-NEXT: [[DOTFOO0__ADDR:%.*]] = bitcast i8* [[DOTFOO0__VOID_ADDR]] to [5 x i32]*
// CHECK-32-NEXT: [[DOTFOO1__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 2, i32 40, i8* inttoptr (i32 7 to i8*)) // CHECK-32-NEXT: [[DOTFOO1__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 4, i32 40, i8* inttoptr (i32 7 to i8*))
// CHECK-32-NEXT: [[DOTFOO1__ADDR:%.*]] = bitcast i8* [[DOTFOO1__VOID_ADDR]] to [10 x i32]* // CHECK-32-NEXT: [[DOTFOO1__ADDR:%.*]] = bitcast i8* [[DOTFOO1__VOID_ADDR]] to [10 x i32]*
// CHECK-32-NEXT: [[DOTFOO2__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 4, i32 80, i8* inttoptr (i32 6 to i8*)) // CHECK-32-NEXT: [[DOTFOO2__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 4, i32 80, i8* inttoptr (i32 6 to i8*))
// CHECK-32-NEXT: [[DOTFOO2__ADDR:%.*]] = bitcast i8* [[DOTFOO2__VOID_ADDR]] to [20 x i32]* // CHECK-32-NEXT: [[DOTFOO2__ADDR:%.*]] = bitcast i8* [[DOTFOO2__VOID_ADDR]] to [20 x i32]*
// CHECK-32-NEXT: [[DOTFOO3__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 8, i32 120, i8* inttoptr (i32 5 to i8*)) // CHECK-32-NEXT: [[DOTFOO3__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 8, i32 120, i8* inttoptr (i32 5 to i8*))
// CHECK-32-NEXT: [[DOTFOO3__ADDR:%.*]] = bitcast i8* [[DOTFOO3__VOID_ADDR]] to [30 x i32]* // CHECK-32-NEXT: [[DOTFOO3__ADDR:%.*]] = bitcast i8* [[DOTFOO3__VOID_ADDR]] to [30 x i32]*
// CHECK-32-NEXT: [[DOTFOO4__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 16, i32 160, i8* inttoptr (i32 4 to i8*)) // CHECK-32-NEXT: [[DOTFOO4__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 16, i32 160, i8* inttoptr (i32 4 to i8*))
// CHECK-32-NEXT: [[DOTFOO4__ADDR:%.*]] = bitcast i8* [[DOTFOO4__VOID_ADDR]] to [40 x i32]* // CHECK-32-NEXT: [[DOTFOO4__ADDR:%.*]] = bitcast i8* [[DOTFOO4__VOID_ADDR]] to [40 x i32]*
// CHECK-32-NEXT: [[DOTFOO5__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 32, i32 200, i8* inttoptr (i32 3 to i8*)) // CHECK-32-NEXT: [[DOTFOO5__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 32, i32 200, i8* inttoptr (i32 3 to i8*))
Show All 10 Lines
// CHECK-32-NEXT: [[DOTFOO10__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 128, i32 720, i8* null) // CHECK-32-NEXT: [[DOTFOO10__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 128, i32 720, i8* null)
// CHECK-32-NEXT: [[DOTFOO10__ADDR:%.*]] = bitcast i8* [[DOTFOO10__VOID_ADDR]] to [90 x double]* // CHECK-32-NEXT: [[DOTFOO10__ADDR:%.*]] = bitcast i8* [[DOTFOO10__VOID_ADDR]] to [90 x double]*
// CHECK-32-NEXT: [[TMP1:%.*]] = bitcast [90 x double]* [[DOTFOO10__ADDR]] to i8* // CHECK-32-NEXT: [[TMP1:%.*]] = bitcast [90 x double]* [[DOTFOO10__ADDR]] to i8*
// CHECK-32-NEXT: call void @__kmpc_free(i32 [[TMP0]], i8* [[TMP1]], i8* null) // CHECK-32-NEXT: call void @__kmpc_free(i32 [[TMP0]], i8* [[TMP1]], i8* null)
// CHECK-32-NEXT: [[TMP2:%.*]] = bitcast [80 x double]* [[DOTFOO9__ADDR]] to i8* // CHECK-32-NEXT: [[TMP2:%.*]] = bitcast [80 x double]* [[DOTFOO9__ADDR]] to i8*
// CHECK-32-NEXT: call void @__kmpc_free(i32 [[TMP0]], i8* [[TMP2]], i8* inttoptr (i32 8 to i8*)) // CHECK-32-NEXT: call void @__kmpc_free(i32 [[TMP0]], i8* [[TMP2]], i8* inttoptr (i32 8 to i8*))
// CHECK-32-NEXT: [[TMP3:%.*]] = load i32, i32* [[MYALLOC]], align 4 // CHECK-32-NEXT: [[TMP3:%.*]] = load i32, i32* [[MYALLOC]], align 4
// CHECK-32-NEXT: [[CONV:%.*]] = inttoptr i32 [[TMP3]] to i8* // CHECK-32-NEXT: [[CONV:%.*]] = inttoptr i32 [[TMP3]] to i8*
// CHECK-32-NEXT: [[DOTBAR1__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 2, i32 4, i8* [[CONV]]) // CHECK-32-NEXT: [[DOTBAR1__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 4, i32 4, i8* [[CONV]])
// CHECK-32-NEXT: [[DOTBAR1__ADDR:%.*]] = bitcast i8* [[DOTBAR1__VOID_ADDR]] to i32* // CHECK-32-NEXT: [[DOTBAR1__ADDR:%.*]] = bitcast i8* [[DOTBAR1__VOID_ADDR]] to i32*
// CHECK-32-NEXT: [[TMP4:%.*]] = load i32, i32* [[MYALLOC]], align 4 // CHECK-32-NEXT: [[TMP4:%.*]] = load i32, i32* [[MYALLOC]], align 4
// CHECK-32-NEXT: [[CONV1:%.*]] = inttoptr i32 [[TMP4]] to i8* // CHECK-32-NEXT: [[CONV1:%.*]] = inttoptr i32 [[TMP4]] to i8*
// CHECK-32-NEXT: [[DOTBAR2__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 2, i32 40, i8* [[CONV1]]) // CHECK-32-NEXT: [[DOTBAR2__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 4, i32 40, i8* [[CONV1]])
// CHECK-32-NEXT: [[DOTBAR2__ADDR:%.*]] = bitcast i8* [[DOTBAR2__VOID_ADDR]] to [10 x i32]* // CHECK-32-NEXT: [[DOTBAR2__ADDR:%.*]] = bitcast i8* [[DOTBAR2__VOID_ADDR]] to [10 x i32]*
// CHECK-32-NEXT: [[TMP5:%.*]] = load i32, i32* [[MYALLOC]], align 4 // CHECK-32-NEXT: [[TMP5:%.*]] = load i32, i32* [[MYALLOC]], align 4
// CHECK-32-NEXT: [[CONV2:%.*]] = inttoptr i32 [[TMP5]] to i8* // CHECK-32-NEXT: [[CONV2:%.*]] = inttoptr i32 [[TMP5]] to i8*
// CHECK-32-NEXT: [[DOTBAR3__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 2, i32 80, i8* [[CONV2]]) // CHECK-32-NEXT: [[DOTBAR3__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 4, i32 80, i8* [[CONV2]])
// CHECK-32-NEXT: [[DOTBAR3__ADDR:%.*]] = bitcast i8* [[DOTBAR3__VOID_ADDR]] to [20 x i32]* // CHECK-32-NEXT: [[DOTBAR3__ADDR:%.*]] = bitcast i8* [[DOTBAR3__VOID_ADDR]] to [20 x i32]*
// CHECK-32-NEXT: [[DOTBAR4__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 16, i32 4, i8* null) // CHECK-32-NEXT: [[DOTBAR4__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 16, i32 4, i8* null)
// CHECK-32-NEXT: [[DOTBAR4__ADDR:%.*]] = bitcast i8* [[DOTBAR4__VOID_ADDR]] to i32** // CHECK-32-NEXT: [[DOTBAR4__ADDR:%.*]] = bitcast i8* [[DOTBAR4__VOID_ADDR]] to i32**
// CHECK-32-NEXT: [[DOTBAR5__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 16, i32 4, i8* null) // CHECK-32-NEXT: [[DOTBAR5__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 16, i32 4, i8* null)
// CHECK-32-NEXT: [[DOTBAR5__ADDR:%.*]] = bitcast i8* [[DOTBAR5__VOID_ADDR]] to float* // CHECK-32-NEXT: [[DOTBAR5__ADDR:%.*]] = bitcast i8* [[DOTBAR5__VOID_ADDR]] to float*
// CHECK-32-NEXT: [[DOTBAR6__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 16, i32 240, i8* null) // CHECK-32-NEXT: [[DOTBAR6__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i32 16, i32 240, i8* null)
// CHECK-32-NEXT: [[DOTBAR6__ADDR:%.*]] = bitcast i8* [[DOTBAR6__VOID_ADDR]] to [30 x double]* // CHECK-32-NEXT: [[DOTBAR6__ADDR:%.*]] = bitcast i8* [[DOTBAR6__VOID_ADDR]] to [30 x double]*
// CHECK-32-NEXT: [[TMP6:%.*]] = bitcast [30 x double]* [[DOTBAR6__ADDR]] to i8* // CHECK-32-NEXT: [[TMP6:%.*]] = bitcast [30 x double]* [[DOTBAR6__ADDR]] to i8*
▲ Show 20 Lines • Show All 57 Lines • ▼ Show 20 Lines
// CHECK-32-NEXT: ret double [[TMP1]] // CHECK-32-NEXT: ret double [[TMP1]]
// //
// //
// CHECK-LABEL: define {{[^@]+}}@main // CHECK-LABEL: define {{[^@]+}}@main
// CHECK-SAME: () #[[ATTR0:[0-9]+]] { // CHECK-SAME: () #[[ATTR0:[0-9]+]] {
// CHECK-NEXT: entry: // CHECK-NEXT: entry:
// CHECK-NEXT: [[MYALLOC:%.*]] = alloca i64, align 8 // CHECK-NEXT: [[MYALLOC:%.*]] = alloca i64, align 8
// CHECK-NEXT: [[TMP0:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]]) // CHECK-NEXT: [[TMP0:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]])
// CHECK-NEXT: [[DOTFOO0__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 1, i64 32, i8* null) // CHECK-NEXT: [[DOTFOO0__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 4, i64 32, i8* null)
// CHECK-NEXT: [[DOTFOO0__ADDR:%.*]] = bitcast i8* [[DOTFOO0__VOID_ADDR]] to [5 x i32]* // CHECK-NEXT: [[DOTFOO0__ADDR:%.*]] = bitcast i8* [[DOTFOO0__VOID_ADDR]] to [5 x i32]*
// CHECK-NEXT: [[DOTFOO1__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 2, i64 48, i8* inttoptr (i64 7 to i8*)) // CHECK-NEXT: [[DOTFOO1__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 4, i64 48, i8* inttoptr (i64 7 to i8*))
// CHECK-NEXT: [[DOTFOO1__ADDR:%.*]] = bitcast i8* [[DOTFOO1__VOID_ADDR]] to [10 x i32]* // CHECK-NEXT: [[DOTFOO1__ADDR:%.*]] = bitcast i8* [[DOTFOO1__VOID_ADDR]] to [10 x i32]*
// CHECK-NEXT: [[DOTFOO2__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 4, i64 80, i8* inttoptr (i64 6 to i8*)) // CHECK-NEXT: [[DOTFOO2__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 4, i64 80, i8* inttoptr (i64 6 to i8*))
// CHECK-NEXT: [[DOTFOO2__ADDR:%.*]] = bitcast i8* [[DOTFOO2__VOID_ADDR]] to [20 x i32]* // CHECK-NEXT: [[DOTFOO2__ADDR:%.*]] = bitcast i8* [[DOTFOO2__VOID_ADDR]] to [20 x i32]*
// CHECK-NEXT: [[DOTFOO3__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 8, i64 128, i8* inttoptr (i64 5 to i8*)) // CHECK-NEXT: [[DOTFOO3__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 8, i64 128, i8* inttoptr (i64 5 to i8*))
// CHECK-NEXT: [[DOTFOO3__ADDR:%.*]] = bitcast i8* [[DOTFOO3__VOID_ADDR]] to [30 x i32]* // CHECK-NEXT: [[DOTFOO3__ADDR:%.*]] = bitcast i8* [[DOTFOO3__VOID_ADDR]] to [30 x i32]*
// CHECK-NEXT: [[DOTFOO4__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 16, i64 160, i8* inttoptr (i64 4 to i8*)) // CHECK-NEXT: [[DOTFOO4__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 16, i64 160, i8* inttoptr (i64 4 to i8*))
// CHECK-NEXT: [[DOTFOO4__ADDR:%.*]] = bitcast i8* [[DOTFOO4__VOID_ADDR]] to [40 x i32]* // CHECK-NEXT: [[DOTFOO4__ADDR:%.*]] = bitcast i8* [[DOTFOO4__VOID_ADDR]] to [40 x i32]*
// CHECK-NEXT: [[DOTFOO5__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 32, i64 208, i8* inttoptr (i64 3 to i8*)) // CHECK-NEXT: [[DOTFOO5__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 32, i64 208, i8* inttoptr (i64 3 to i8*))
Show All 10 Lines
// CHECK-NEXT: [[DOTFOO10__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 128, i64 720, i8* null) // CHECK-NEXT: [[DOTFOO10__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 128, i64 720, i8* null)
// CHECK-NEXT: [[DOTFOO10__ADDR:%.*]] = bitcast i8* [[DOTFOO10__VOID_ADDR]] to [90 x double]* // CHECK-NEXT: [[DOTFOO10__ADDR:%.*]] = bitcast i8* [[DOTFOO10__VOID_ADDR]] to [90 x double]*
// CHECK-NEXT: [[TMP1:%.*]] = bitcast [90 x double]* [[DOTFOO10__ADDR]] to i8* // CHECK-NEXT: [[TMP1:%.*]] = bitcast [90 x double]* [[DOTFOO10__ADDR]] to i8*
// CHECK-NEXT: call void @__kmpc_free(i32 [[TMP0]], i8* [[TMP1]], i8* null) // CHECK-NEXT: call void @__kmpc_free(i32 [[TMP0]], i8* [[TMP1]], i8* null)
// CHECK-NEXT: [[TMP2:%.*]] = bitcast [80 x double]* [[DOTFOO9__ADDR]] to i8* // CHECK-NEXT: [[TMP2:%.*]] = bitcast [80 x double]* [[DOTFOO9__ADDR]] to i8*
// CHECK-NEXT: call void @__kmpc_free(i32 [[TMP0]], i8* [[TMP2]], i8* inttoptr (i64 8 to i8*)) // CHECK-NEXT: call void @__kmpc_free(i32 [[TMP0]], i8* [[TMP2]], i8* inttoptr (i64 8 to i8*))
// CHECK-NEXT: [[TMP3:%.*]] = load i64, i64* [[MYALLOC]], align 8 // CHECK-NEXT: [[TMP3:%.*]] = load i64, i64* [[MYALLOC]], align 8
// CHECK-NEXT: [[CONV:%.*]] = inttoptr i64 [[TMP3]] to i8* // CHECK-NEXT: [[CONV:%.*]] = inttoptr i64 [[TMP3]] to i8*
// CHECK-NEXT: [[DOTBAR1__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 2, i64 4, i8* [[CONV]]) // CHECK-NEXT: [[DOTBAR1__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 4, i64 4, i8* [[CONV]])
// CHECK-NEXT: [[DOTBAR1__ADDR:%.*]] = bitcast i8* [[DOTBAR1__VOID_ADDR]] to i32* // CHECK-NEXT: [[DOTBAR1__ADDR:%.*]] = bitcast i8* [[DOTBAR1__VOID_ADDR]] to i32*
// CHECK-NEXT: [[TMP4:%.*]] = load i64, i64* [[MYALLOC]], align 8 // CHECK-NEXT: [[TMP4:%.*]] = load i64, i64* [[MYALLOC]], align 8
// CHECK-NEXT: [[CONV1:%.*]] = inttoptr i64 [[TMP4]] to i8* // CHECK-NEXT: [[CONV1:%.*]] = inttoptr i64 [[TMP4]] to i8*
// CHECK-NEXT: [[DOTBAR2__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 2, i64 48, i8* [[CONV1]]) // CHECK-NEXT: [[DOTBAR2__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 4, i64 48, i8* [[CONV1]])
// CHECK-NEXT: [[DOTBAR2__ADDR:%.*]] = bitcast i8* [[DOTBAR2__VOID_ADDR]] to [10 x i32]* // CHECK-NEXT: [[DOTBAR2__ADDR:%.*]] = bitcast i8* [[DOTBAR2__VOID_ADDR]] to [10 x i32]*
// CHECK-NEXT: [[TMP5:%.*]] = load i64, i64* [[MYALLOC]], align 8 // CHECK-NEXT: [[TMP5:%.*]] = load i64, i64* [[MYALLOC]], align 8
// CHECK-NEXT: [[CONV2:%.*]] = inttoptr i64 [[TMP5]] to i8* // CHECK-NEXT: [[CONV2:%.*]] = inttoptr i64 [[TMP5]] to i8*
// CHECK-NEXT: [[DOTBAR3__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 2, i64 80, i8* [[CONV2]]) // CHECK-NEXT: [[DOTBAR3__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 4, i64 80, i8* [[CONV2]])
// CHECK-NEXT: [[DOTBAR3__ADDR:%.*]] = bitcast i8* [[DOTBAR3__VOID_ADDR]] to [20 x i32]* // CHECK-NEXT: [[DOTBAR3__ADDR:%.*]] = bitcast i8* [[DOTBAR3__VOID_ADDR]] to [20 x i32]*
// CHECK-NEXT: [[DOTBAR4__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 16, i64 8, i8* null) // CHECK-NEXT: [[DOTBAR4__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 16, i64 8, i8* null)
// CHECK-NEXT: [[DOTBAR4__ADDR:%.*]] = bitcast i8* [[DOTBAR4__VOID_ADDR]] to i32** // CHECK-NEXT: [[DOTBAR4__ADDR:%.*]] = bitcast i8* [[DOTBAR4__VOID_ADDR]] to i32**
// CHECK-NEXT: [[DOTBAR5__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 16, i64 4, i8* null) // CHECK-NEXT: [[DOTBAR5__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 16, i64 4, i8* null)
// CHECK-NEXT: [[DOTBAR5__ADDR:%.*]] = bitcast i8* [[DOTBAR5__VOID_ADDR]] to float* // CHECK-NEXT: [[DOTBAR5__ADDR:%.*]] = bitcast i8* [[DOTBAR5__VOID_ADDR]] to float*
// CHECK-NEXT: [[DOTBAR6__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 16, i64 240, i8* null) // CHECK-NEXT: [[DOTBAR6__VOID_ADDR:%.*]] = call i8* @__kmpc_aligned_alloc(i32 [[TMP0]], i64 16, i64 240, i8* null)
// CHECK-NEXT: [[DOTBAR6__ADDR:%.*]] = bitcast i8* [[DOTBAR6__VOID_ADDR]] to [30 x double]* // CHECK-NEXT: [[DOTBAR6__ADDR:%.*]] = bitcast i8* [[DOTBAR6__VOID_ADDR]] to [30 x double]*
// CHECK-NEXT: [[TMP6:%.*]] = bitcast [30 x double]* [[DOTBAR6__ADDR]] to i8* // CHECK-NEXT: [[TMP6:%.*]] = bitcast [30 x double]* [[DOTBAR6__ADDR]] to i8*
▲ Show 20 Lines • Show All 59 Lines • Show Last 20 Lines