similar to: [LLVMdev] ManagedStatic and order of destruction

Right, I'm suggesting we keep llvm_shutdown() for users who want this control, but also destroy still-live ManagedStatic instances if llvm_shutdown() is not called. This helps in the case where there is not a clear time when llvm_shutdown() can be called, especially given that LLVM cannot be resurrected in the same process due to current limitations in the pass registry, and perhaps

The mutex could be made an actual global static instead of a ManagedStatic. This guarantees it would be constructed before main, and live until the end of main. So even in PassRegistry's destructor, which would get call during llvm_shutdown(), it would always have the same mutex. Ideally I'd like to just delete the mutex, as it doesn't seem like anyone is using it in a

I actually had an idea about how to fix this in a relatively painless manner. Although given my experience over the past 4 days, it might not be best to call it painless without first trying :) The idea is to make a StaticPassRegistry. RegisterPass<> only touches the StaticPassRegistry, and nothing else touches the StaticPassRegistry. So once you enter main(), StaticPassRegistry can be

+cc original authors of these changes. Is PassRegistry intended to be thread-safe? The header file explicitly says that PassRegistry is not thread-safe, but there are mutexes and locking used in the code. This is actually creating a problem, because of a subtle bug involving static initialization order and shutdown. In particular, the RegisterPass<> static template will get invoked

2017-05-02 8:42 GMT-07:00 Duncan P. N. Exon Smith <dexonsmith at apple.com>: > +Teresa, Mehdi > > On May 2, 2017, at 08:31, James Henderson <jh7370.2008 at my.bristol.ac.uk> > wrote: > > Hi, > > We have been investigating an issue when running LTO with our proprietary > linker, which links against libLTO dynamically. The issue is that when we > pass

Based on a recent discussion[1], I started trying to remove the functions llvm_start_multithreaded() and llvm_stop_multithreaded() from the codebase. It turns out this is a little bit tricky. Consider the following scenario: During program initialization, a global static object's constructor dereferences a ManagedStatic. During dereferencing of the ManagedStatic, it needs to know whether

Hi, We have been investigating an issue when running LTO with our proprietary linker, which links against libLTO dynamically. The issue is that when we pass -time-passes via the lto_codegen_debug_options function in the LTO C API, no time information is produced during compilation. The reason for this is that time information is stored in state owned by a ManagedStatic instance, and is only

Hi, We use LLVM libraries to compile C++ code and noticed slow downs when multiple threads of a process were compiling at once. *perf *indicated that most of the CPU time was spent in a spin lock, which was being locked/unlocked from llvm::PassRegistry::getPassInfo(). We read the relevant LLVM code and found out that PassRegistry is a ManagedStatic and is shared among all threads in case of a

I am trying to figure out how to free up some memory that seems to be lost when running valgrind under our internal application. The stack traces I get are: ==19966== 4 bytes in 1 blocks are still reachable in loss record 1 of 12 ==19966== at 0x402569A: operator new(unsigned int) (vg_replace_malloc.c:255) ==19966== by 0x5D9BBE8: void* llvm::object_creator<llvm::PassRegistry>()

Hi llvmdev! In our project (Windows, Visual Studio compiler) we've got some frontend which generates LLVM IR. Now I'm observing strange *llvm_shutdown* behavior. If I called it when I was done using the LLVM APIs I saw that destructors of static objects created new ManagedStatic objects, which was never freed. Then I tried to use static global llvm_shutdown_obj to cause destruction

Hello, I have an application that uses Clang and LLVM wrapped in a DLL. Last week I was debugging reported issues with Microsoft Application Verifier<http://www.microsoft.com/download/en/details.aspx?id=20028> that were related to how critical sections are handled. The issues are caused by critical sections that are either still active when heap blocks are deleted or the DLL is unloaded.

On Wed, Dec 25, 2013 at 9:38 PM, dblaikie at gmail.com <dblaikie at gmail.com> wrote: > Its generally been by design that tblgen leaks. Might be nice to fix one day > but it's been that way for a while now so don't expect it to be fixed soon. > > If thats the suppression mechanism of choice (I would've expected a build > flag option) for lsan, I'd say go for

On Sat, Mar 2, 2013 at 12:56 PM, Alexander Potapenko <glider at google.com>wrote: > > On Mar 2, 2013 3:18 AM, "reed kotler" <rkotler at mips.com> wrote: > > > > Does LLVM try and make sure that all memory is freed before compiler > exit? > > > > For example, It seems like the ARM constant pool code that I'm reading > will leave a lot

We have root caused the segfault - it was due to a caching layer we have in our code, which is to avoid duplicate compilations. Basically, llvm::JIT::getPointerToFunction() looks up PassRegistry, but as our change introduces a separate PassRegistry for each thread, this means that the thread that calls llvm::JIT::getPointerToFunction() should have appropriate PassRegistry setup. In our setup, some

I have come across the issue where repeatedly creating and deleting an interpreter can cause incorrect external functions to be called. The issue had to do with the ManagedStatic map Functions in ExecutionEngine/Interpreter/ExternalFunctions.cpp. I was able to fix this by calling llvm_shutdown() when I delete the interpreter. Is it necessary to delete all ManagedStatic objects to delete

Apologies for the noop question in advance (just getting started with LLVM), and I'm not entirely sure if this is the right list to post to. is it? I have some lambda functions as member variables that I want to have my LLVM language make calls to. I've added a mapping to them, but this doesn't seem to enable LLVM to resolve the functions. I asked on stackoverflow but the suggestion

LGTM. I think it is totally reasonable to just disable LSan directly in tablegen at least for now. I would leave some comments on the disabling to clarify: 1) What it does, the reader may not have heard about LSan. 2) Some of the high-level information from this thread as pointers for anyone who wants to go and fix this some day. -Chandler On Wed, Dec 25, 2013 at 2:26 PM, Kostya Serebryany

I write my pass in a mix way of NVPTXAllocaHoisting, NVPTXSplitBBatBar and transforms/Hello. The following is part of the codes: in NVPTXTargetMachine.cpp bool NVPTXPassConfig::addPreEmitPass() { addPass(createTest()); return false; } in NVPTXTest.h namespace llvm{

I try to use INITIALIZE_PASS instead of RegisterPass<> to register my pass, though I don't understand what's their difference and how it works because its documents doesn't exist. But it still doesn't work. Parts of my codes is as follows: in NVPTXTest.h namespace llvm { void initializeNVPTXTestPass(PassRegistry &r); class NVPTXTest : public ModulePass { public:

Is anyone currently working on overhauling the command-line processing code? We're currently having some design issues with this component, and I'd like to start a larger conversation on it. For context, I am talking from an "LLVM as a library" perspective instead of an "LLVM as a set of tools" perspective. In a nut-shell, the problems we are facing are as follows: