similar to: [LLVMdev] Call Graph Analysis

On Wed, Aug 20, 2008 at 9:47 AM, kapil anand <kapilanand2 at gmail.com> wrote: > Hi, > > I have been working on implementing new analysis for a LLVM module. For > that, I need that Call Graph of the module. Inside the CallGraph file, the > constructor has been declared "protected" hence I can't define its instance > and thus use the CallGraph for further

Hi, I have made my optimization as an extension of ModulePass as suggested in last mail. I included "addRequired<CallGraph>" in getAnalysisUsage function. The application compiles fine but while linking, I get following errors: "undefined reference to llvm::CallGraphLinkVar" "undefined reference to llvm::CallGraph::ID The above error is at point where I refer to

On Aug 21, 2008, at 2:20 PM, kapil anand wrote: > Hi, > > I have made my optimization as an extension of ModulePass as > suggested in last mail. I included "addRequired<CallGraph>" in > getAnalysisUsage function. > The application compiles fine but while linking, I get following > errors: > "undefined reference to llvm::CallGraphLinkVar"

Hi Devang, GlobalModRefPass is also a ModulePass and it uses CallGraph Analysis. So, I think it should not necessary to extend CallGraphSCCPass to use CallGraph information. Module Pass shoule be sufficient... --Kapil On 10/8/08, Devang Patel <dpatel at apple.com> wrote: > Hi Kapil, > > On Oct 8, 2008, at 10:19 AM, kapil anand wrote: > >> Hi all, >> >> I

Hi all, I need a new kind of analysis on LLVM Module, so I made a new pass to do this. This new pass extends the ModulePass class and follows the conventions used in GlobalModRefPass, which is also a Module Pass.I need the CallGraph analysis for this pass, hence I have added (addRequired(CallGraph)) in getAnalysisUsage function of this new pass. I also added it to CallGraph Analysis group through

Hi Kapil, On Oct 8, 2008, at 10:19 AM, kapil anand wrote: > Hi all, > > I need a new kind of analysis on LLVM Module, so I made a new pass > to do this. This new pass extends the ModulePass class and follows > the conventions used in GlobalModRefPass, which is also a Module > Pass.I need the CallGraph analysis for this pass, hence I have added >

On Oct 8, 2008, at 10:59 AM, kapil anand wrote: > Hi Devang, > > GlobalModRefPass is also a ModulePass and it uses CallGraph Analysis. > So, I think it should not necessary to extend CallGraphSCCPass to use > CallGraph information. Module Pass shoule be sufficient... ok, but you're Registering your pass in CallGraph Analysis group. What if you remove

HI Devang, Initially I had not added this pass to the group and I got this error regarding PassManager. I added the pass to group just to check whether that removes the error or not... I have removed it again. Is there any other registration required to configure pass manager for the pass? --Kapil On Wed, Oct 8, 2008 at 2:37 PM, Devang Patel <dpatel at apple.com> wrote: > > On

Hi, I am not making this application part of LLVM library although I do link the libLLVMipa library ( actually I am using llvm-config --libs all option in the Makefile which links all the libraries). Is it necessary to make it part of LLVM library? Thanks for you responses. --Kapil On Fri, Aug 22, 2008 at 3:40 PM, Bill Wendling <isanbard at gmail.com> wrote: > On Aug 21, 2008, at

Hi all, I am implementing a new pass for LLVM which extends Call Graph SCCPass. I need DominatorTree Information when I get to individual function. I have added AU.addrequired<DominatorTree>() and AU.addRequired<DominanceFrontier>() in getAnalysisUsage() function. But, when I get to the pass, Pass Manager gives following runtime error Unable to schedule 'Dominator Tree

On Feb 11, 2009, at 6:05 PM, kapil anand wrote: > Hi all, > > I am implementing a new pass for LLVM which extends Call Graph > SCCPass. I need DominatorTree Information when I get to individual > function. I have added AU.addrequired<DominatorTree>() and > AU.addRequired<DominanceFrontier>() in getAnalysisUsage() function. > > But, when I get to the pass,

Hi all, I have an ARM virtual machine with ubuntu and want to port LLVM over that. In release documents, ARM is not in the list of supported platforms. I tried to compile LLVM on ARM platform and ran into an error. Has anyone tried to port LLVM onto ARM before? Is it going to require lots of changes? Thanks --Kapil -------------- next part -------------- An HTML attachment was scrubbed...

The Compilation error I am getting: "Error while compiling ARMJITInfo.cpp.. Selected processor does not support fstmfdd sp!,{d0,d1,d2,d3,d4,d5,d6,d7}.... On Thu, Sep 24, 2009 at 8:45 PM, kapil anand <kapilanand2 at gmail.com> wrote: > Hi all, > > I have an ARM virtual machine with ubuntu and want to port LLVM over that. > In release documents, ARM is not in the list of

Hi Kapil! LLVM both builds and runs fine on ARM Linux. I am running a public buildbot for the llvm project that are compiling and testing llvm on arm-linux daily and this machine are running Ubuntu Jaunty. http://google1.osuosl.org:8011/builders/llvm-arm-linux I have successfully compiled LLVM on ARM using GCC 4.3.3 on a Ubuntu Jaunty system and GCC 4.4 on a Ubuntu Karmic system. Do you use

My god! I would love a branch predictor! It would simplify many aspects of my register allocator. Second, I am surprised it did not make it into the tree. Since more is being done with register allocation as a while "RegAllocBasic" was just put in, I hope this is looked at again. Do you have a working svn copy? Also, could you send me a copy/link to that '94 paper off the list

Hello Kapil, I have implemented a static profiler for LLVM as a google summer of code project in 2009. I wrote it for the 2.4 branch, but the implementation never made into the tree. I have recently ported it to LLVM 2.8, but I haven't tested it. You can take a look at the code from: http://homepages.dcc.ufmg.br/~rimsa/tools/stprof-llvm.patch The implementation is based on Wu's

You said it was expensive, but if you had to put a big-o estimate on it, what would it be? -Thanks Jeff Kunkel On Tue, Nov 2, 2010 at 8:54 PM, Andrei Alvares <logytech at gmail.com> wrote: > Hello Jeff, > > On Tue, Nov 2, 2010 at 9:17 PM, Jeff Kunkel <jdkunk3 at gmail.com> wrote: > > My god! I would love a branch predictor! It would simplify many aspects > of >

Hi all, Does LLVM infrastructure contain implementation of any static profiling algorithm apart from "Spill-Weight" calculation present in Live Intervals class? The future work page does suggest implementation of some "static profiling" algorithms to make an educated guesses about the relative execution frequencies of various parts of the code. Thanks --Kapil --------------

Thanks Andrei! I haven't read the paper. I would see whether this fulfills my requirements or whether I need to make any changes. --Kapil On Tue, Nov 2, 2010 at 12:43 PM, Andrei Alvares <logytech at gmail.com> wrote: > Hello Kapil, > > I have implemented a static profiler for LLVM as a google summer of > code project in 2009. I wrote it for the 2.4 branch, but the >

Hello Jeff, On Tue, Nov 2, 2010 at 9:17 PM, Jeff Kunkel <jdkunk3 at gmail.com> wrote: > My god! I would love a branch predictor! It would simplify many aspects of > my register allocator. The branch predictor of the implementation is not as accurate as the one from the paper, but it is close enough. Unfortunately, the branch predictor is a very expensive pass, because it relies on