similar to: [LLVMdev] Removing redundant type checks from Java

Hi all, I wrote you some days ago about one project that I want to do on vmkit: I want to remove redundant instanceof tests. I am right now looking at the LLVM code that vmkit produces for java files, but I am finding it very difficult to identify the code that is produced by each instanceof. Would it be possible for you guys to give me some pointers on how to attack this problem? Should I

I don't know if there is something to do with my problem, but I'm using VMKit front-end to generate the .bc files. So, after generating the example.bc, I remove some redundant instructions with my pass with opt. On Sun, Jul 4, 2010 at 8:32 PM, Alysson <aishofpf at gmail.com> wrote: > Thank you for the answer Reid. > I've been already using the format `opt -load MyPass.so

Thank you for the answer Reid. I've been already using the format `opt -load MyPass.so -my-pass input.bc -o output.bc`. But my problem is to execute the .s generated file. As long as I know, gcc compiler is capable to generate target machine code from assembly. So I tried to compile it with gcc to generate machine code, but there was many problems. Is there any way to compile and run .s files?

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

Thanks Alysson. Unfortunately, those instructions don't work for me (I followed them exactly using a clean system) -- probably because of the recent changes to the svn heads. After fixing EscapeAnalysis in vmkit to use the updated CallInst.getArgOperand interface I hit the error below. I was interested in using mmtk with the system (as recommended), but that requires a llvm-gcc build (and

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, > >

Dear all, I am trying to remove some instanceof tests from bc code that is produced by vmkit. For instance: 1 if (o instanceof String) { 2 String s = (String)o; 3 } Vmkit seems to be inserting two tests into the bc code: one for the test in line 1, and another, implicit, that would lunch perhaps JavaClassCastException if the cast fails at runtime. Would it be possible for some of you

How are you getting LLVM to run your pass? There's plenty of information here if you haven't found this document: http://llvm.org/docs/WritingAnLLVMPass.html I'm going to assume you're using `opt -load MyPass.so -my-pass input.bc -o output.bc` in which case you can just run llc on output.bc to get a .s file. Reid On Thu, Jul 1, 2010 at 8:16 PM, Alysson <aishofpf at

Hi all, I´ve been working on a transformation pass which performs the remotion of some redundant instructions. So now it´s finished I´d like to run the optimized programs generated by my pass. But I´ve got some problems to do this, because I didn´t find a way to generate assembly from the code and then call the gcc compiler to produce machine-dependent code. Please, does anybody know how to run

Hi, Just to comfirm that I have just clean built the current LLVM and VMKIT revisions and it failed too. The problem seems to be the same one (about escape analysis in LLVM). My configuration is Ubuntu 10.04 and GCC 4.4.3 on x86. Cheers, Chanwit Konrad Hinsen-2 wrote: > > On 6 Jul 2010, at 22:25, Zoe Wolk wrote: > >> If anyone is able to build vmkit and get it to run without

Hello Nicolas, Thank you for quick reply. Is LLVM revision 107641 a good one to build with latest VMKIT? Thanks, Chanwit On Thu, Jul 8, 2010 at 00:06, nicolas geoffray <nicolas.geoffray at gmail.com> wrote: > Hi all, > Yes, there was a change in the LLVM CallInst API, and I haven't updated > vmkit yet. It should be fixed real soon. > Cheers, > Nicolas > > On Wed,

On 6 Jul 2010, at 22:25, Zoe Wolk wrote: > If anyone is able to build vmkit and get it to run without a > segfault (for any > distro/arch), using mmtk or the default collector/installation (as > described in the vmkit page or > Alysson's writeup) using the svn head of llvm and vmkit, please let > me know. I had a working installation of VMKit dating from April,

Latest version of vmkit should now build fine with latest llvm. Nicolas On Wed, Jul 7, 2010 at 11:02 AM, nicolas geoffray < nicolas.geoffray at gmail.com> wrote: > The one that I have currently that works with vmkit is 107120. > > > On Wed, Jul 7, 2010 at 10:58 AM, Chanwit Kaewkasi <chanwit at gmail.com>wrote: > >> Hello Nicolas, >> >> Thank you for

Hi all, Yes, there was a change in the LLVM CallInst API, and I haven't updated vmkit yet. It should be fixed real soon. Cheers, Nicolas On Wed, Jul 7, 2010 at 9:50 AM, chanwit <chanwit at gmail.com> wrote: > > Hi, > > Just to comfirm that I have just clean built the current LLVM and VMKIT > revisions and it failed too. > The problem seems to be the same one (about

The one that I have currently that works with vmkit is 107120. On Wed, Jul 7, 2010 at 10:58 AM, Chanwit Kaewkasi <chanwit at gmail.com> wrote: > Hello Nicolas, > > Thank you for quick reply. > Is LLVM revision 107641 a good one to build with latest VMKIT? > > Thanks, > > Chanwit > > On Thu, Jul 8, 2010 at 00:06, nicolas geoffray > <nicolas.geoffray at

Thank you, Nicolas. But I still have no luck compiling the latest revision. However, vmkit-0.27 is compiled and running fine on my machine. Thanks, Chanwit On Thu, Jul 8, 2010 at 05:18, nicolas geoffray <nicolas.geoffray at gmail.com> wrote: > Latest version of vmkit should now build fine with latest llvm. > Nicolas > > On Wed, Jul 7, 2010 at 11:02 AM, nicolas geoffray >

Hi Chanwit, What problems did you have for compiling the latest revision? Could you give me a log? Thanks, Nicolas On Thu, Jul 8, 2010 at 10:05 PM, Chanwit Kaewkasi <chanwit at gmail.com> wrote: > Thank you, Nicolas. > But I still have no luck compiling the latest revision. However, > vmkit-0.27 is compiled and running fine on my machine. > > Thanks, > > Chanwit >

Hi, I'm new to VMKit and I'm trying to use its AOT Java compiler but I'm immediately getting an error message which I don't understand. Everything was compiler as suggested in the Getting Started tutorial, using GNU Classpath. I try to run: llcj ../../tools/trainer/Release+Asserts/HelloWorld.class and get the following output: Can not materiale a function in AOT mode.0

Hi all, I'm interested in using VMKit rather extensively in a project of mine. I was hoping to get feedback on a few vmkit-related questions from those more familiar: --Is there a 'best' version to use? I'd be inclined to use the 2.9 release since it's a stationary target, but curious if I should try to follow ToT.  Is there a recommended best practice? --What features of

Hi Will, On Fri, May 13, 2011 at 7:30 PM, Will Dietz <willdtz at gmail.com> wrote: > Hi all, > > I'm interested in using VMKit rather extensively in a project of mine. > > Great! > I was hoping to get feedback on a few vmkit-related questions from > those more familiar: > > --Is there a 'best' version to use? I'd be inclined to use the 2.9 >