[clang-repl] Fix destructor for interpreter for the cuda negation case #138091
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
llvmbot
added
clang
|
@llvm/pr-subscribers-clang Author: Anutosh Bhat (anutosh491) ChangesCheck this error for more context (https://github.com/compiler-research/CppInterOp/actions/runs/14749797085/job/41407625681?pr=491#step:10:531) This fails with Problem :
Full diff: https://github.com/llvm/llvm-project/pull/138091.diff 4 Files Affected:
diff --git a/clang/include/clang/Interpreter/Interpreter.h b/clang/include/clang/Interpreter/Interpreter.h
index 56213f88b9e30..f8663e3193a18 100644
--- a/clang/include/clang/Interpreter/Interpreter.h
+++ b/clang/include/clang/Interpreter/Interpreter.h
@@ -116,6 +116,9 @@ class Interpreter {
/// Compiler instance performing the incremental compilation.
std::unique_ptr<CompilerInstance> CI;
+ /// An optional compiler instance for CUDA offloading
+ std::unique_ptr<CompilerInstance> DeviceCI;
+
protected:
// Derived classes can use an extended interface of the Interpreter.
Interpreter(std::unique_ptr<CompilerInstance> Instance, llvm::Error &Err,
diff --git a/clang/lib/Interpreter/DeviceOffload.cpp b/clang/lib/Interpreter/DeviceOffload.cpp
index 7d0125403ea52..8ab6a61993972 100644
--- a/clang/lib/Interpreter/DeviceOffload.cpp
+++ b/clang/lib/Interpreter/DeviceOffload.cpp
@@ -25,13 +25,13 @@
namespace clang {
IncrementalCUDADeviceParser::IncrementalCUDADeviceParser(
- std::unique_ptr<CompilerInstance> DeviceInstance,
+ CompilerInstance &DeviceInstance,
CompilerInstance &HostInstance,
llvm::IntrusiveRefCntPtr<llvm::vfs::InMemoryFileSystem> FS,
llvm::Error &Err, const std::list<PartialTranslationUnit> &PTUs)
- : IncrementalParser(*DeviceInstance, Err), PTUs(PTUs), VFS(FS),
+ : IncrementalParser(DeviceInstance, Err), PTUs(PTUs), VFS(FS),
CodeGenOpts(HostInstance.getCodeGenOpts()),
- TargetOpts(DeviceInstance->getTargetOpts()) {
+ TargetOpts(DeviceInstance.getTargetOpts()) {
if (Err)
return;
StringRef Arch = TargetOpts.CPU;
@@ -41,7 +41,6 @@ IncrementalCUDADeviceParser::IncrementalCUDADeviceParser(
llvm::inconvertibleErrorCode()));
return;
}
- DeviceCI = std::move(DeviceInstance);
}
llvm::Expected<llvm::StringRef> IncrementalCUDADeviceParser::GeneratePTX() {
diff --git a/clang/lib/Interpreter/DeviceOffload.h b/clang/lib/Interpreter/DeviceOffload.h
index 43645033c4840..c1cb4cd23a4b5 100644
--- a/clang/lib/Interpreter/DeviceOffload.h
+++ b/clang/lib/Interpreter/DeviceOffload.h
@@ -28,7 +28,7 @@ class IncrementalCUDADeviceParser : public IncrementalParser {
public:
IncrementalCUDADeviceParser(
- std::unique_ptr<CompilerInstance> DeviceInstance,
+ CompilerInstance &DeviceInstance,
CompilerInstance &HostInstance,
llvm::IntrusiveRefCntPtr<llvm::vfs::InMemoryFileSystem> VFS,
llvm::Error &Err, const std::list<PartialTranslationUnit> &PTUs);
@@ -42,7 +42,6 @@ class IncrementalCUDADeviceParser : public IncrementalParser {
~IncrementalCUDADeviceParser();
protected:
- std::unique_ptr<CompilerInstance> DeviceCI;
int SMVersion;
llvm::SmallString<1024> PTXCode;
llvm::SmallVector<char, 1024> FatbinContent;
diff --git a/clang/lib/Interpreter/Interpreter.cpp b/clang/lib/Interpreter/Interpreter.cpp
index 80487eb4ef74a..4ca12f6858c9d 100644
--- a/clang/lib/Interpreter/Interpreter.cpp
+++ b/clang/lib/Interpreter/Interpreter.cpp
@@ -416,6 +416,10 @@ Interpreter::Interpreter(std::unique_ptr<CompilerInstance> Instance,
Interpreter::~Interpreter() {
IncrParser.reset();
Act->FinalizeAction();
+ if (DeviceParser)
+ DeviceParser.reset();
+ if (DeviceAct)
+ DeviceAct->FinalizeAction();
if (IncrExecutor) {
if (llvm::Error Err = IncrExecutor->cleanUp())
llvm::report_fatal_error(
@@ -501,8 +505,10 @@ Interpreter::createWithCUDA(std::unique_ptr<CompilerInstance> CI,
DCI->ExecuteAction(*Interp->DeviceAct);
+ Interp->DeviceCI = std::move(DCI);
+
auto DeviceParser = std::make_unique<IncrementalCUDADeviceParser>(
- std::move(DCI), *Interp->getCompilerInstance(), IMVFS, Err, Interp->PTUs);
+ *Interp->DeviceCI, *Interp->getCompilerInstance(), IMVFS, Err, Interp->PTUs);
if (Err)
return std::move(Err);
|
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
|
The cuda tests for the negation case should now work (not segfault because libcudart.so is not found but simply get skipped). And they were anyways working for systems having cuda support |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Check this error for more context (https://github.com/compiler-research/CppInterOp/actions/runs/14749797085/job/41407625681?pr=491#step:10:531)
This fails with
Problem :
The destructor currently handles no clearance for the DeviceParser and the DeviceAct. We currently only have this
llvm-project/clang/lib/Interpreter/Interpreter.cpp
Lines 416 to 419 in 9764938
The ownership for DeviceCI currently is present in IncrementalCudaDeviceParser. But this should be similar to how the combination for hostCI, hostAction and hostParser are managed by the Interpreter. As on master the DeviceAct and DeviceParser are managed by the Interpreter but not DeviceCI. This is problematic because :
IncrementalParser holds a Sema& which points into the DeviceCI. On master, DeviceCI is destroyed before the base class ~IncrementalParser() runs, causing Parser::reset() to access a dangling Sema (and as Sema holds a reference to Preprocessor which owns PragmaNamespace) we see this