similar to: [LLVMdev] VMKit Boehm MMTk Compilation

For anyone who encounters this issue in the future, my issue was the configuration of llvm-gcc. Configure with --with-llvmgccdir=YOUR/PATH doesn't work, instead you are supposed to use --with-llvmgcc=PATH/TO/llvm-gcc and --with-llvmgxx=/PATH/TO/llvm-g++ That will at least get you past my error. Best, Kyle Quoting nicolas geoffray <nicolas.geoffray at gmail.com>: > I am

Hi, I am working on compiling vmkit but have run into an error which seems like something that maybe the llvm community would have a better idea of how to solve than vmkit's. The issue is when vmkit is built to compile the MMTk garbage collector it must use the llvm-gcc compiler, at which point MODULE_NAME is defined instead of LIBRARYNAME. This results in the following error

Hi Nicolas, I'm trying now to build vmkit with mmtk: ./configure --with-llvmsrc=/home/zoewolk/RESEARCH/llvm --with-llvmobj=/home/zoewolk/RESEARCH/llvm --with-gnu-classpath-glibj=/home/zoewolk/RESEARCH/classpath-0.97.2/lib --with-gnu-classpath-libs=/home/zoewolk/RESEARCH/classpath-0.97.2/lib --with-gc=mmtk I've built llvm-gcc from source and configured llvm so that it can find it.

Hi All, Thanks Nicolas for pointing out that I needed --enable-targets=x86,cpp for my llvm config to get vmkit to build. The process gets a lot farther but fails in different places depending upon the vmkit config (Can you please see the three case below?). My machine is a intel core i7 running ubuntu karmic and gcc 4.4.1. What arch/OS/distro/gcc do you all have VMKit working on? thanks for

Hi, Nicolas Indeed, I did not have ant installed and installing it allowed me to overcome the previous error. However the next one appeared. I have ran make with ENABLE_OPTIMIZED=1 VERBOSE=1 and here is the output (the last part of it) make[2]: Entering directory `/home/svozniuk/llvm/vmkit/mmtk/java' /usr/bin/ant -buildfile /home/svozniuk/llvm/vmkit/mmtk/java/build.xml Buildfile:

Hi Sergii, I think there is a problem in your configure command for VMKit. If you look at what you provide for glibj: --with-gnu-classpath-glibj="/home/svozniuk/llvm/classpath-0.97.2 I don't think it's right. You should change it to: --with-gnu-classpath-glibj="/home/svozniuk/llvm/classpath-0.97.2/lib/glibj.zip Also please make sure that you have created the links for the

Hi Sergii, On Tue, Oct 25, 2011 at 1:15 PM, Vozniuk, Sergii <sergii.vozniuk at epfl.ch>wrote: > Hi everyone > > I wasn't successful in running VMKit on my machine and I'm still struggling > with doing it. > Maybe someone had at least one of my problems and knows how to solve it > (because it seems to me that I ran into every problem that could occur :) ). >

Hi everyone I wasn't successful in running VMKit on my machine and I'm still struggling with doing it. Maybe someone had at least one of my problems and knows how to solve it (because it seems to me that I ran into every problem that could occur :) ). Thanks in advance, Sergii. So here is the story: I have a fresh installation of Ubuntu 11.04 64bit with all important updates. I'll

Hi, I just did a reconfigure and rebuild with the latest svn for vmkit code. The code still seg faults but the stack trace seems to have changed. llvm[2]: Building Debug+Asserts mmtk-vmkit.jar all vmjc: JavaRuntimeJIT.cpp:172: uint32 j3VirtualTableLookup(j3::Class*, uint32, uint32*, j3::JavaObject*): Assertion `dmeth->classDef->isInitializing() && "Class not ready in a virtual

Thanks Nicolas, I do have --with-gnu-classpath-libs and --with-gnu-classpath-glibj pointing to the right thing, but I have gnu class path 0.98 instead of 0.97 and its the non X version. So I do not have the libgtkpeer.so. Maybe that is the reason why it is crashing. Unfortunately, I am not able to compile GNU class path with gtk+ peer right now. Will give it a try on the mac. Thanks, Akshat

This error usually shows up when vmkit cannot find the classpath libs and jar. Make sure --with-gnu-classpath-libs points to the directory of .so files are, and that --with-gnu-classpath-glibj points to glibj.zip. Nicolas On Sat, Aug 27, 2011 at 5:13 AM, Akshat Gupta <g.akshat at gmail.com> wrote: > Hi, > I just did a reconfigure and rebuild with the latest svn for vmkit code. >

Hi Akshat, Please use version 0.97.2. vmkit does not support 0.98. Also, it should be fine to not compile gtk+ peer. Nicolas On Mon, Aug 29, 2011 at 8:58 PM, Akshat Gupta <g.akshat at gmail.com> wrote: > Thanks Nicolas, > I do have --with-gnu-classpath-libs and --with-gnu-classpath-glibj > pointing to the right thing, but I have gnu class path 0.98 instead of 0.97 > and its

Hi Zoe, maybe you've already solved your problem with VMKit configuration, but here you can find a mini-tutorial that I've done to guide people on VMKit installation. http://www2.dcc.ufmg.br/laboratorios/llp/wiki/doku.php?id=afpf Best regards, Alysson On Wed, Jun 30, 2010 at 1:34 PM, nicolas geoffray < nicolas.geoffray at gmail.com> wrote: > Hi Zoe, > > Unfortunately, I

I just noticed there was some .h move around in llvm and fixed the includes in vmkit. Maybe you were just removing the failing includes? In any case, I have updated vmkit for llvm trunk. Let me know if updating helps. Nicolas On Fri, Aug 26, 2011 at 12:00 PM, Akshat Gupta <g.akshat at gmail.com> wrote: > Yep, I am compiling with llvm-gcc. Also I removed the LD flag on the >

Hi all, I tried the step of the Getting Started page (http://vmkit.llvm.org/get_started.html) on Linux/x64 but I had no luck. I could build them. But when I tried to execute J3, it shows like follows, >> j3: JavaRuntimeJIT.cpp:367: void* j3StartJNI(uint32*, uint32**, mvm::KnownFrame*): Assertion `th->getLastSP() == th->lastKnownFrame->currentFP' failed. Does anybody know if

Hi Zoe, Unfortunately, I don't have access to a x64 machine anymore, so I can't test that architecture on vmkit. Could you install a virtual machine on your machine that is x86_32? I know it has been successfully built and ran on x86_32 on different systems. Cheers, Nicolas On Wed, Jun 30, 2010 at 7:49 AM, Zoe Wolk <zoewolk at gmail.com> wrote: > Hi again Nicolas, > >

Hi Yi, The precompiler is in charge of translating the most common classes of the Java runtime library to llvm code (typically, java.lang.Object), in order to natively compile these classes and decrease the bootstrap time. So, if you have a segmentation fault at this stage, it means that you have broken something in the interface between MMTk and VMKit. If it can helps you: * during a first

Hmm, disregard my previous comment (while you might still have issues with 64bit, that's not what's primarily causing your issue)--your particular error is when you don't point vmkit to the classpath libraries properly. Set it with the configure option --with-gnu-classpath-glibj=/path/to/classpath-0.97.2/lib/glibj.zip Good luck! ~Will On Tue, Aug 23, 2011 at 5:49 PM, Will Dietz

Ah yes, thanks, this was one problem, now I hit another: llvm[2]: Building Release+Asserts mmtk-vmkit.jar all Thread 0x110000000 received a SIGSEGV: either the VM code or an external native method is bogus. Aborting... ; 0x5cfeed (0x1100fe410) in _Z14sigsegvHandleriP7siginfoPv ; 0x2b889c232c60 in Unknown method ; 0x561499 (0x1100fe900) in _ZN2j310JavaObject7acquireEPS0_ ; 0x55b464 (0x1100fe920)

Yep, I am compiling with llvm-gcc. Also I removed the LD flag on the configure, it still gives the same segafault. Thanks, Akshat On Fri, Aug 26, 2011 at 4:21 AM, Nicolas Geoffray < nicolas.geoffray at gmail.com> wrote: > On Fri, Aug 26, 2011 at 1:08 AM, Akshat Gupta <g.akshat at gmail.com> wrote: > >> Hi, >> >> Here are my build details. >> OS: >>