similar to: [LLVMdev] writing a pass

On Wed, Jul 09, 2003 at 10:47:14AM -0500, Nicholas Rizzolo wrote: > I'm having a problem with opt. According to the "Writing an LLVM Pass" > tutorial, all I have to do to get the Hello pass to work is: > > cd ${LLVM_HOME}/llvm/lib/Transforms/Hello > gmake > cd $MYTEST > opt -load ${OBJ_ROOT}/lib/Release/libhello.so -hello < something.bc >

Hi John, I have attached my Makefile.config. I noticed that I did'nt have SRC_ROOT and OBJ_ROOT variables defined as such in my environment. But I noticed that this is defined in the Makefile.config by default to the same values. All the same, I tried defining both these variables in my .cshrc, but I still get the same error. The main directory for me is /home/llvm, and I have

On Wed, 9 Jul 2003, Misha Brukman wrote: > > Also, I don't have a Debug subdirectory like the tutorial said I would - I > > have Release instead. > > 1. What version of LLVM are you using (i.e., when did you get a tarball, > or if you are using anonymous CVS, are you up to date)? I use CVS and I updated a couple days ago. > 3. If you are using a version of LLVM

Hopefully the right list … otherwise pointers appreciated (I’ve spent days trying to find a solution but could not find sufficient documentation, could not find a proper list of -flavor darwin commands) I am setting up an environment for cross compiling shared libraries using LLVM. Overall this works perfectly for many unix like environments but MacOS is a pain in the ass despite that I am

Cc'ing people who are working on macOS. On Thu, Apr 27, 2017 at 9:05 AM, Peter Kriens via llvm-dev < llvm-dev at lists.llvm.org> wrote: > Hopefully the right list … otherwise pointers appreciated (I’ve spent days > trying to find a solution but could not find sufficient documentation, > could not find a proper list of -flavor darwin commands) > > I am setting up an

LLD work on Darwin is stalled out at the moment. Patches are welcome (and I'll try to find time to review them), but I'd recommend ld64 for any real-world linking. - Lang. On Thu, Apr 27, 2017 at 11:06 AM, Rui Ueyama <ruiu at google.com> wrote: > Cc'ing people who are working on macOS. > > On Thu, Apr 27, 2017 at 9:05 AM, Peter Kriens via llvm-dev < > llvm-dev

Thanks for the answer. However, this means I need a different linker for MacOS as for Windows/Linux? I guess it is a not a major thing but I was hoping to set up a cross compile env. where only the header files and shared libraries differed :-( I guess we’re not there yet … :-) Kind regards, Peter Kriens > On 6 May 2017, at 19:57, Lang Hames <lhames at gmail.com> wrote: > >

Sounds like a fantastic goal, and enticingly close … Kind regards, Peter Kriens > On 11 May 2017, at 20:53, Andrew Kelley <superjoe30 at gmail.com> wrote: > > Zig author here in the same boat. Looks like I'm going to have to disable cross compiling for MacOS and when compiling native for MacOS add a dependency on the system linker. Really looking forward to Mach-O support

yue wrote: > hi, > about [LLVMdev] another question > > thanks > > yueqiang One other thing you might want to try is to put your object tree in a directory that is *not* inside of your source tree. Currently, we don't support using separate object trees that are subdirectories of the source tree. In other words: Will work: ========== SRC_ROOT=/home/yue/llvm

Hi all, I'm having problems compiling LLVM 3.2 with RTTI enabled. Since my code makes heavy use of the Boost libraries, RTTI is a must. This is the script I use to compile (as root): SRC_ROOT=$PWD OBJ_ROOT=/export/apps/llvm INSTALL_ROOT=/usr/local mkdir -p $OBJ_ROOT cd $OBJ_ROOT $SRC_ROOT/configure --prefix=$INSTALL_ROOT \

Lang - I think there has perhaps been a regression in that case, because I have an example of a trivial IR module that fails to be linked correctly: https://bugs.llvm.org/show_bug.cgi?id=32376 Kevin provided some clues about this but I was not able to find a fix. On Tue, May 16, 2017 at 12:24 PM, Lang Hames <lhames at gmail.com> wrote: > Hi Guys, > > As I left it, lld on Darwin

On Fri, 13 Apr 2007, Devang Patel wrote: >> I think libhello should drop its use of SlowOperationInformer. > > That'll fix Hello example. However, anyone trying to load their custom > pass will likely to run into this again. It is a long-standing issue. The deal is that libsupport (and many others) are .a files. If one of the .o files in the .a file is used by a plugin, but

On Apr 13, 2007, at 12:28 PM, Chris Lattner wrote: > On Fri, 13 Apr 2007, Devang Patel wrote: >> And this brings back CommandLine Error: Arugment blah defined more >> than once! >> >> And without this, opt -load .../LLVMHello.dylib -hello does not work >> on Darwin because dyld is not able to find SlowerOperationInformer ;) > > I think libhello should drop

Hi, i ran the insert-function-profiling pass on my bytecode and when i jit it, i get ' cant resolve 'llvm_start_func_profiling'. How do I link the "instrument" library. ? Thanks -Sriraman.

Hi, Just to let you know, in the html of http://llvm.cs.uiuc.edu/docs/WritingAnLLVMPass.html there are two occurrences of the following: </pre</p> which I'm guessing should be </pre><p> instead. This does affect the way the document looks in Netscape. - Nick

Hi , I got a few for writing a pass. 1) Is it possible to use input parameters in command line ? For example, we got our own pass, ie. HELLO opt -load ../../Debug/lib/libHELLO.so -HELLO < hello.bc From the above command line, could we use some input parameter and we can read those parameter in Pass routine? 2) For splitting BB, the way what I did is to pick up instruction

hi, when i change the OBJ_ROOT and recompile the llvm, it's successful. but run following test, then the error occues: ----------------------------------------------------- [yue at RH9 obj]$ make -C ./test/Programs make: Entering directory `/home/yue/llvm/obj/test/Programs' make[1]: Entering directory `/home/yue/llvm/obj/test/Programs/SingleSource' make[2]: Entering directory

I am just beginning to use LLVM, and find that 'getting started' is the steep part of the learning curve. I am using MingW and msys, which I have not fully grokked yet. Anyway, actually building the libraries from source is turning out to be difficult. I want to work with LLVM as described in the except from the LLVM FAQ below. I am using C++, so the FFI is a non-issue. Can someone

Hi Anton, The problem is in getting the system configured. Below is the crash I get in configure. Your comment regarding gnuwin32 not being needed is interesting - I actually installed gnu bison/flex to get past an earlier crash in configure. Thats why I was suggesting that a precompiled OBJ_ROOT for MingW32 would be a good idea - configure appears to be inherently fragile, and requires more

Dear All, I'm using LLVM2.9 for profiling basic block frequency. I'm using following commands. rdpatel55 at ubuntu:~$ llvm-gcc -emit-llvm -O0 -c -o adpcm.bc adpcm.c rdpatel55 at ubuntu:~$ llvm-gcc -emit-llvm -O0 -c -o rawcaudio.bc rawcaudio.c rdpatel55 at ubuntu:~$ llvm-link -o main.bc rawcaudio.bc adpcm.bc rdpatel55 at ubuntu:~$ opt -q -f -insert-edge-profiling -o main.inst main.bc