similar to: [LLVMdev] VMKit broken on trunk

Hi all, I am currently working with vmkit (especially on J3) for a student project, I want to made some experimentations on J3. I see in the file lib/j3/VMCore/JavaClass.h that if I want to add some attributes in the class JavaClass I must change a few thing in LLVMRuntime/* and in JnjvmModule.cpp. But JnjvmModule.cpp does not exist anymore. So my first question is : what is the equivalent of

Hi Julien, I also tried to add a field and get the same error as you. In theory, to add a field in JavaClass or JavaCommonClass you have to put your field at the end of the declaration list (in order to minimize index changes in J3Intrinsics) in C++ code and in LLVMRuntime/runtime-single.ll We are working on this issue to see what is going wrong. The problem is the fact that the error is not

Hi Julien, I found out what was missing for adding a field to JavaClass. While precompiling Java code (JavaAOTCompiler) we create an equivalent LLVM Constant object for each class. So modifying the internal representation of a class means that you also have to modify the AOTCompiler. Here is a patch with a tiny example where an int is added to JavaClass. If you want to add more complex

On Oct 1, 2007, at 12:09 AM, Andreas Fredriksson wrote: > On 10/1/07, Evan Cheng <evan.cheng at apple.com> wrote: > >>> If I lower the calling convention stuff manually rather than relying >>> on the generated CC analyzer, how do I determine that a formal >>> argument I'm about to receive or pack together is supposed to be >>> used

I can't be the first person to think of this, can I? But I couldn't locate any reference on this combination. If you know of one, please tell me. Sparse: http://kernel.org/pub/linux/kernel/people/josh/sparse/ LLVM: http://llvm.org/ So, you may expect compile-llvm.c in a few days. :) -- Seo Sanghyeon

How can I find out, in llvm-c API, whether a basic block is well formed? In C++ I could call getTerminator and test for NULL. -- Seo Sanghyeon

On Apr 29, 2008, at 1:27 AM, Gabor Greif wrote: > Hi all, > > I have reported more than enough about the space savings achieved > and the associated costs, here comes the current patch for review. > > Since this one is substantially smaller than the previous one, I did > not cut it in pieces. The front part is about headers and the rest > the .cpp and other files. Hi

Hi, Has anybody had any problems getting digium hardware lately? Regards Greg Cirino ___________________________________ Cirelle Enterprises Inc. 603-425-2221 www.cirelle.com Website Design www.cirelle.net ProSpeed High Speed Dial-up - 5 Times Faster www.cedata.com Web, FTP, Email Hosting Services www.mlsbot.com MLS IDX Services When You Want It Done Well, Just Call Cirelle It's not just a

http://llvm.org/bugs/show_bug.cgi?id=1837 I reduced the above case to this: ; test.ll define i8* @test() { %x1 = malloc i8, i32 16 %x2 = bitcast i8* %x1 to i8** store i8* %x1, i8** %x2 ret i8* %x1 } $ llvm-as test.ll $ opt -instcombine test.bc -o testopt.bc $ llc test.bc $ llc testopt.bc $ diff -u test.s testopt.s -movl $16, (%esp) +movl $8, (%esp) call malloc I'm afraid that I'm not

On 2008-12-19, at 09:18, Seo Sanghyeon wrote: > How can I find out, in llvm-c API, whether a basic block is well > formed? In C++ I could call getTerminator and test for NULL. There's not currently a binding for this. In general, there's incomplete support for inspection and analysis through the C bindings. BasicBlock::getTerminator() is just a convenient way to spell for

On 10/1/07, Evan Cheng <evan.cheng at apple.com> wrote: > > If I lower the calling convention stuff manually rather than relying > > on the generated CC analyzer, how do I determine that a formal > > argument I'm about to receive or pack together is supposed to be used > > as a pointer so I can stick it in an address register? > > I'd suggest custom

The ARM EABI (AAPCS) defines: - i64 values are 8-bytes aligned - "The alignment of an aggregate shall be the alignment of its most-aligned component." So, struct ss { int x; int y; }; void f(int a, struct ss b); r0 <- a r1-r2 <- b void g(int a, long long b); r0 <- a r2-r3 <- b The problem is: llvm-gcc generates the same bytecode for both functions: declare void

Hi all, I have reported more than enough about the space savings achieved and the associated costs, here comes the current patch for review. Since this one is substantially smaller than the previous one, I did not cut it in pieces. The front part is about headers and the rest the .cpp and other files. Cheers, Gabor -------------- next part -------------- An embedded and charset-unspecified

Hola! I'm using asterisk as H.323 -> PRI gateway. First call goes thru ok, second concurrent call fails with: Aug 6 11:52:30 DEBUG[737292]: chan_h323.c:1038 setup_incoming_call: Sending to context [ip2pri] -- Executing Dial("H323/ip$192.168.32.25:60271/984", "Zap/1/9541163107100") in new stack Aug 6 11:52:30 NOTICE[753677]: app_dial.c:554 dial_exec: Unable to

On Sep 29, 2007, at 2:57 PM, Andreas Fredriksson <deplinenoise at gmail.com> wrote: > On 9/25/07, Evan Cheng <evan.cheng at apple.com> wrote: >>> Hi Evan, >>> wouldn't this generate fairly terrible code if each address register >>> use has to be preceded by instructions to make an address register >>> hold the right value? >> >>

Hi there, I am a student considering a compiler design based dissertation with llvm. I am having problems building llvm on OpenBSD-current. I hope to make a port of llvm for OpenBSD once I have figured out how to build it. Observe: llvm[3]: Compiling Deserialize.cpp for Release build In file included from /home/edd/llvm/llvm-2.3/include/llvm/Bitcode/BitstreamRead er.h:18, from

Hello fellow LLVMers and Clangstas, We want to make Clang great, and we need your help! Helping is easy: just build Clang on your platform and start using it as your main compiler for LLVM and Clang development. Much of the Clang team has been living on Clang for at least several weeks already, and we've found it to be quite stable for development. If you run into problems---poor

I'm not sure how this happened, but llvm-ar q doesn't work, even if llvm-ar.cpp has relevant codes. As the developer policy said "you are allowed to commit patches without approval which you think are obvious", I did. The patch follows, just in case. Index: tools/llvm-ar/llvm-ar.cpp =================================================================== ---

Setting "mail_location=maildir:~/Maildir:INDEX=~/index:CONTROL=~/control" (or separating INDEX and CONTROL in general) along with "quota=maildir" (and or having a ~/.dovecot.sieve) appears to expose a small bug in deliver (version 1.1.3, but probably all). While deliver will create a missing INDEX directory, it throws an error if the CONTROL directory is missing, exiting with

All, I've upgraded our doxygen to 1.5.6. Please let me know if you see any problems. Thanks, Tanya