[LLVMdev] Exception handling question

[James Williams]

2010/1/22 Duncan Sands <baldrick at free.fr>

> Hi James,
>
>
>     want to send us your testcase code?  Then we can give it a whirl.
>>
>>
>> Test code is at http://giantblob.com/ehtest.tar.gz
>>
>> Thanks for the help. I apologize in advance if it turns out I'm doing
>> something stupid!
>>
>
> I hope you realise that by running llvm-ld without -native you are actually
> executing your program from the JIT.  I did a native compilation as
> follows:
> used llvm-link to link all of the bitcode into "total.bc"; ran llc on
> total.bc,
> producing "total.s"; did "g++ -o total total.s"; ran ./total.  It seems to
> work:
>
> $ ./total
> __l_personality called$
>
> Of course it is probably supposed to work from the JIT too (I don't know
> anything about the JIT), but it clearly doesn't: when I tried I got:
>
> lli: lib/ExecutionEngine/JIT/JIT.cpp:624: void
> llvm::JIT::runJITOnFunctionUnlocked(llvm::Function*, const
> llvm::MutexGuard&): Assertion `!isAlreadyCodeGenerating && "Error: Recursive
> compilation detected!"' failed.
> ...
> 6  lli       0x0000000000d3f5e2
> llvm::JIT::runJITOnFunctionUnlocked(llvm::Function*, llvm::MutexGuard
> const&) + 62
> 7  lli       0x0000000000d3f9c6
> llvm::JIT::getPointerToFunction(llvm::Function*) + 686
> 8  lli       0x0000000000d67c86
> llvm::ExecutionEngine::getPointerToGlobal(llvm::GlobalValue const*) + 70
> 9  lli       0x0000000000d62805
> llvm::JITDwarfEmitter::EmitCommonEHFrame(llvm::Function const*) const + 613
> 10 lli       0x0000000000d60ac3
> llvm::JITDwarfEmitter::EmitDwarfTable(llvm::MachineFunction&,
> llvm::JITCodeEmitter&, unsigned char*, unsigned char*, unsigned char*&) +
> 335
> ...
>
Thanks for looking at this.

Yes, I realise this will link to bitcode and that the result is a script
that runs lli. I kind of just expected JIT to work, particularly since the
example code on the wiki uses JIT.

The JIT is what my project will eventually target (goal is an out of process
compiler will incrementally generate bitcode on disk combined with an
application server that will run resulting bitcode via JIT) and hence I've
only been testing on the JIT.

I'll see if I can work around the recursive compilation problem.

-- James


> Ciao,
>
> Duncan.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20100122/05532e6a/attachment.html>