search for: dcrowe

cvs co llvm ... cd llvm; ./configure --with-llvmgccdir=<llvmgccdir>; make ... Compiling llvmAsmParser.cpp Linking /home/dcrowe/llvm/llvmcvs/llvm/lib/Debug/asmparser.o make[2]: Leaving directory `/home/dcrowe/llvm/llvmcvs/llvm/lib/AsmParser' make[2]: Entering directory `/home/dcrowe/llvm/llvmcvs/llvm/lib/Bytecode' make[2]: *** No rule to make target `all'. Stop. make[2]: Leaving directory `/home/dcrowe/llvm/llv...

Is there a facility with which we may automagically create a "fake" exit node, one that is the target of all BasicBlocks ending in return. This would be very helpful to IP dataflow analysis... Dave

...ls(); Doesn't appear to work... getCalledFunction() returns 0 Dave On Fri, 6 Dec 2002, Vikram Adve wrote: > P.S. I have also updated the CSIL tree. Just check out > TarjanSCCIterator.h. > > --Vikram > http://www.cs.uiuc.edu/~vadve > > > > From: David Crowe <dcrowe at tremor.crhc.uiuc.edu> > > Subject: [LLVMbugs] Re: [LLVMdev] Tarjan+function_ptrs == trouble ? > > Sender: llvmbugs-admin at cs.uiuc.edu > > Date: Fri, 6 Dec 2002 00:54:41 -0600 (CST) > > > > Thanks, > > > > I've been through the documentation, a...

...h you can use for your analysis (it will even tell you which BB that is too). Note that a function may _not_ have an exit node (if it contains an infinite loop, for example), so be aware of that, but otherwise it should work for you with no problem. -Chris > From: "David Crowe" <dcrowe at tremor.crhc.uiuc.edu> > To: <llvmdev at cs.uiuc.edu> > Sent: Friday, November 29, 2002 4:58 PM > Subject: [LLVMdev] Fake Exit node > > > > Is there a facility with which we may automagically create a "fake" exit > > node, one that is the target of al...

There is a pass "UnifyFunctionExitNodes()" (you can add it to AnalysisUsage of your pass) that does the trick. -anand ----- Original Message ----- From: "David Crowe" <dcrowe at tremor.crhc.uiuc.edu> To: <llvmdev at cs.uiuc.edu> Sent: Friday, November 29, 2002 4:58 PM Subject: [LLVMdev] Fake Exit node > Is there a facility with which we may automagically create a "fake" exit > node, one that is the target of all BasicBlocks ending in return....

...will even tell you which BB that is too). Note that a function may _not_ > have an exit node (if it contains an infinite loop, for example), so be > aware of that, but otherwise it should work for you with no problem. > > -Chris > > > > From: "David Crowe" <dcrowe at tremor.crhc.uiuc.edu> > > To: <llvmdev at cs.uiuc.edu> > > Sent: Friday, November 29, 2002 4:58 PM > > Subject: [LLVMdev] Fake Exit node > > > > > > > Is there a facility with which we may automagically create a "fake" exit > > >...

...d 501) > id 5523617C1CE; Fri, 8 Nov 2002 18:25:04 -0500 (EST) > Received: from localhost (localhost [127.0.0.1]) > by nondot.org (Postfix) with ESMTP > id 488FD24D019; Fri, 8 Nov 2002 17:25:04 -0600 (CST) > From: Chris Lattner <sabre at nondot.org> > To: David Crowe <dcrowe at tremor.crhc.uiuc.edu> > Cc: llvmdev at cs.uiuc.edu > In-Reply-To: <Pine.LNX.4.33.0211081707090.11485-100000 at tremor.crhc.uiuc.edu> > Message-ID: <Pine.LNX.4.44.0211081721560.18180-100000 at nondot.org> > MIME-Version: 1.0 > Subject: [LLVMdev] Iterating on the DSGr...

llvm, What is the best way to implement a traversal of the DS graph, starting at a scalar and processing all nodes to which the scalar allows access? Currently the links vector is not public and there is no apparent way to bound the getLink call (ie a getNumLinks call).... Dave

The following function produces a segv when the begin() is taken from vmap, the ScalarMap... bool StaticMemAnalysis::isSingleMalloc(DSGraph* theGraph,DSNode* theNode) { int count = 0; std::map<Value*, DSNodeHandle>& vmap = theGraph->getScalarMap(); for (std::map<Value*, DSNodeHandle>::iterator i = vmap.begin(); i != vmap.end(); ++i) { if(i->second.getNode() ==

On Thu, 28 Nov 2002, Chris Lattner wrote: > > > The following function produces a segv when the begin() is taken from > > vmap, the ScalarMap... > > int count = 0; > > std::map<Value*, DSNodeHandle>& vmap = theGraph->getScalarMap(); > > Are you sure "theGraph" is not null at this point? yes... that call returns sucessfully. the segfault

The way we are doing it, per Vikram's recommendation, is to provide a mapping between the callers nodes and the callees nodes at the time of the call. Given that the only way they can communicate is through the parameters, globals, and return node, this is fairly easy. You must take into account that the DSGraphs may be different shapes though. Dave

Dear LLVM, Recently I incorporated code into my project such that function pointers were supported, however it seems that the TarjanSCC_iterator no longer works on the call graph... main() is no longer reached while the function pointers are... I can provide code, but I have a feeling there is a simple fix... Has anyone gone through this? Thanks, Dave

LLVM, I am trying to equate two dsnodes across call boundries. On the Caller side we have an operand of form: sbyte * getelementptr ([20 x sbyte]* %.LC0, long 0, long 0) using getNodeForValue() yields a NULL pointer. On the Callee, the parameter is of form: sbyte * S and getNodeForValue works fine. Is there something special that must be done to access DNodes accessed using GEP? Thanks,

LLVM, Is it possible to cross compile on an x86 platform to a sparc binary? Thanks, Dave

LLVM, When we recompile, is it possible to "make" only one of the tools, or at least relink only the tools whose object files have changed? Thanks Dave

LLVM, What is the status of the InductionVariable "semi-pass"? I have tested it out on spec benchmarks, and while it does correctly identify some of the variables, it fails to recognize most. Typically the following scenario arises a_loop: ... %tmp.19 = load int* %bsLive %tmp.20 = add int %tmp.19, -8 store int %tmp.20, int* %bsLive %tmp.5 = setgt int %tmp.20, 0 br bool %tmp.5, label

Thanks, I've been through the documentation, and if I visit main, i think everything will turn out correctly... Printing out the Call graph reveals that main is calling this external node i think you are making reference to. Could this be the problem? Dave On Fri, 6 Dec 2002, Chris Lattner wrote: > > if you take the time to print out the function each non-looping node iter >

Hello LLVM, Can you suggest a good way to use the loops and induction variable passes to map loop exiting BasicBlocks to InductionVariables. That is, can we use these tools to identify the loop condition. What i have tried Function Pass: foreach BB if(terminal is loop exit of containing loop) trace back to instruction producing the cond that the branch branches on -

Test Cases: (attached) Iteration code: (...) typedef TarjanSCC_iterator<CallGraph*> MyTarjan; CallGraph& callGraph = getAnalysis<CallGraph>(); MyTarjan iter = tarj_begin(&callGraph); MyTarjan end = tarj_end(&callGraph); while(iter!=end) iter++; (...) if you take the time to print out the function each non-looping node iter traverses, it never reaches main...