similar to: [LLVMdev] misc CVS patches

On Tue, Apr 19, 2005 at 07:01:40AM +0200, Markus F.X.J. Oberhumer wrote: > While trying to hunt down a codegen bug (not yet found) ... Have you considered using bugpoint for your codegen debugging needs? http://llvm.cs.uiuc.edu/docs/Bugpoint.html#codegendebug > I've collected some small patches you might find useful. Sweet! > Please review and apply as you see fit. I've

On Wed, Apr 20, 2005 at 12:12:54PM +0200, Markus F.X.J. Oberhumer wrote: > Misha Brukman wrote: > >On Tue, Apr 19, 2005 at 07:01:40AM +0200, Markus F.X.J. Oberhumer > >wrote: Have you considered using bugpoint for your codegen debugging > >needs? http://llvm.cs.uiuc.edu/docs/Bugpoint.html#codegendebug > > Well, the (critical) bug in question was >

On Fri, Apr 22, 2005 at 01:46:53AM +0200, Markus F.X.J. Oberhumer wrote: > Debugging code generator problem! > <cbe><gcc><program>Warning: While generating reference output, program > exited with > non-zero exit code. This will NOT be treated as a failure. > > *** The C backend cannot match the reference diff, but it is used as the > 'known good'

On Fri, Apr 22, 2005 at 03:13:44AM +0200, Markus F.X.J. Oberhumer wrote: > Misha Brukman wrote: > >On Fri, Apr 22, 2005 at 02:32:25AM +0200, Markus F.X.J. Oberhumer wrote: > >>[...] > >>bugpoint: Unknown command line argument '-instcombine-load-vn'. Try: > >>'bugpoint --help' > > > >You need a space between -instcombine and

On Thu, Apr 21, 2005 at 05:56:43PM +0200, Markus F.X.J. Oberhumer wrote: > Reid Spencer wrote: > >This patch can't be applied. Not all platforms have getgid() and > >getuid() functions. Placing non-portable code outside of lib/System > >is deprecated. We set the values to 1000 by default because in > >general the uid/gid doesn't matter in an archive and the 1000

On Wed, Apr 20, 2005 at 01:48:41PM +0200, Markus F.X.J. Oberhumer wrote: > In case you really prefer outside patches for this issue, here is a > first version on a small subset of the sources. Committed. > But I'd think you would want to handle this "in house". I think you're right, it'll be easier that way. ;) -- Misha Brukman :: http://misha.brukman.net ::

What's different about code that's been mem2reg'd from straight front end code, or anything that mem2reg hasn't been run on? PHINODES! It appears to be crashing when I try to cast a Value* that's really a BB* (from the PHInode operands) to a User*, insteresting since I am dyn_casting. I just caught this on cerr though (printing out what the Value* was each time). Let me

Markus, This patch can't be applied. Not all platforms have getgid() and getuid () functions. Placing non-portable code outside of lib/System is deprecated. We set the values to 1000 by default because in general the uid/gid doesn't matter in an archive and the 1000 value gets you to a safe (non-root, non-system) value. If a specific value for these is needed on a given platform, then we

On Fri, Apr 22, 2005 at 02:32:25AM +0200, Markus F.X.J. Oberhumer wrote: > Ok, after pasting the output from "gccas -debug-pass=Arguments > </dev/null -o - >/dev/null" I get the following: > > bugpoint -verify -lowersetjmp -funcresolve -raiseallocs -simplifycfg > -mem2reg -globalopt -globaldce -ipconstprop -deadargelim -instcombine > -simplifycfg -prune-eh

Nope. Win32 is not (and does not pretend to be) POSIX compliant. These functions cannot be used outside of lib/System/Unix. New methods need to be added to class llvm::sys::Process. Markus F.X.J. Oberhumer wrote: > Jeff Cohen wrote: > >> Misha Brukman wrote: >> >>>> I didn't use get{u,g}id() because that's not portable - I think we >>>> need

On Fri, Apr 22, 2005 at 02:14:23AM +0200, Markus F.X.J. Oberhumer wrote: > Hmm - is there some option so that bugpoint automatically runs all the > passes that are normally run by gccas and gccld ? Nope, that's not built into bugpoint, hence the procedures in the document. -- Misha Brukman :: http://misha.brukman.net :: http://llvm.cs.uiuc.edu

On Wed, Jun 23, 2004 at 03:50:09PM -0500, Patrick Meredith wrote: > MetaSplit is an anlysis I just finished writing. It doesn't alter > anything, all it does is build a set of "program instructions". For > some reason even though if I run it with any other combination of > passes I've found, anytime I run it with mem2reg I get a seg fault in > dyn_cast!

Somehow it fails with operand out of bounds when the number of operands is 2 and I am asking for the second operand. ----- Original Message ----- From: "Chris Lattner" <sabre at nondot.org> To: <llvmdev at cs.uiuc.edu> Sent: Wednesday, June 23, 2004 4:24 PM Subject: Re: [LLVMdev] weird issue with mem2reg, should have guessed > On Wed, 23 Jun 2004, Patrick Meredith

On Wed, 23 Jun 2004, Patrick Meredith wrote: > What's different about code that's been mem2reg'd from straight front end > code, or anything that mem2reg hasn't been run on? PHINODES! Yup, front-ends generally don't produce SSA form. :) > It appears to be crashing when I try to cast a Value* that's really a > BB* (from the PHInode operands) to a User*,

void MetaSplit::handleProgramUses(Value *V){ if(!isa<BasicBlock>(V)) programValues.insert(V); if(User *U = dyn_cast<User>(V)){ User::op_iterator OB = U->op_begin(), OE = U->op_end(); for(; OB != OE; ++OB){ if(CallInst *CI = dyn_cast<CallInst>(*OB)){ Function *F = CI->getCalledFunction(); if(F == ii || F == fi || F == vi || F == di || F == ci

Dear LLVMers, If you live on the bleeding edge (i.e. CVS version), please read! On Wed, Apr 20, 2005 at 12:12:54PM +0200, Markus F.X.J. Oberhumer wrote: > Do you really want external patches for this ? A simple Perl script > that runs on all *.h and *.cpp files, and a local commit from your > side would be much simpler. I'm in the process of doing just this as we speak. What this

On Mon, May 16, 2005 at 05:15:30PM -0700, John Cortes wrote: > If I use any of the regalloc parameters (local, ...) I get an error in > the LiveVariable.cpp file, in the part that I think cheaks for dead > code because a Variable didn't have a defined Instance to a Machine > instruction. > > " llc: LiveVariables.cpp:86: void >

On Wed, 23 Jun 2004, Patrick Meredith wrote: > Somehow it fails with operand out of bounds when the number of operands > is 2 and I am asking for the second operand. Second meaning operand 1. Okay, so you have something like this: if (CallInst *CI = dyn_cast<CallInst>(...)) { ... = CI->getOperand(1); } Can you send in this snippet of code, the assertion, and the

On Thu, Apr 21, 2005 at 10:25:45PM +0200, Markus F.X.J. Oberhumer wrote: > while you're at this, may I also ask you to look at "indenting by tabs" > ? Some (rather few) source files have hard tabs which could be changed > to spaces in this cleanup process. OK, I'll do this in the next pass over the tree. -- Misha Brukman :: http://misha.brukman.net ::

On Thu, Dec 23, 2004 at 06:42:02AM +0100, Markus F.X.J. Oberhumer wrote: > with the small patch attached below the whole llvm toolchain (llvm & > llvm-gcc) will compile and run under AMD64 Linux in native 64-bit LP64 > mode. Sounds great! I'll add it to the list of supported platforms. > This means that compilation, bytecode management and CWriter output > all work as