similar to: [LLVMdev] Get Analysis from PassManager

Hi LLVMers, I am fairly new to the LLVM pass framework. My goal is to extend the CFGPrinter analysis pass to label the edges of the graph with their edge-counts from profile data. I can generate the CFGs using 'analyze', but I am having trouble loading the profile data. I added the line AU.addRequired<ProfileInfo>(); to the getAnalysisUsage. It is returning zero for all the edge

Shuguang Feng wrote: >> What does "llc -debug-pass=Structure" say? Is the ProfileLoaderPass >> really the last pass to touch the ProfileInfo before you are using it? > > Below is the sequence of passes that I see. Although the > NoProfileInfo pass is being run, it should be subsequently overridden > by ProfileInfoLoaderPass (LoaderPass) correct? Yes. >

On Wed, 16 Mar 2005, Eric Zimmerman wrote: > Hi LLVMers, > > I am fairly new to the LLVM pass framework. My goal is to extend the > CFGPrinter analysis pass to label the edges of the graph with their > edge-counts from profile data. > > I can generate the CFGs using 'analyze', but I am having trouble > loading the profile data. I added the line >

> What does "llc -debug-pass=Structure" say? Is the ProfileLoaderPass > really the last pass to touch the ProfileInfo before you are using it? Below is the sequence of passes that I see. Although the NoProfileInfo pass is being run, it should be subsequently overridden by ProfileInfoLoaderPass (LoaderPass) correct? Target Data Layout Create Garbage Collector Module Metadata

-------------- next part -------------- An embedded message was scrubbed... From: ambika <ambika at cse.iitb.ac.in> Subject: [LLVMdev] Can someone help me with error while i make my own pass Date: Thu, 08 Apr 2010 21:40:25 +0530 Size: 4901 URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20100411/33876147/attachment.eml>

Hi, How can I use profile information into my pass for performing some analysis. I tried something like: PassManager PassMgr = PassManager(); cl::opt<std::string> ProfileDataFile(cl::Positional, cl::desc("<llvmprof.out file>"), cl::Optional, cl::init("llvmprof.out")); PassMgr.add(createProfileLoaderPass(ProfileDataFile));

> It *is* allowed to access ModulePass analysis information from an > FunctionPass? BasicBlockPlacement (a FunctionPass) also accesses the profile information and I assumed it worked (but haven't independently verified this). > Can you try to manually override the PI value in the > MyCodeGenPass::runOnMachineFunction() to the value seen in llc and then > access the

Try doing a clean build, then doing 'make ENABLE_PIC=1' -Chris On Apr 11, 2010, at 9:13 AM, ambika wrote: > > > From: ambika <ambika at cse.iitb.ac.in> > Date: April 8, 2010 9:10:25 AM PDT > To: llvmdev at cs.uiuc.edu > Subject: [LLVMdev] Can someone help me with error while i make my own pass > > > Hi, > > I have added LoaderInterface pass in

Tried that but still no success Chris Lattner wrote: > Try doing a clean build, then doing 'make ENABLE_PIC=1' > > -Chris > > On Apr 11, 2010, at 9:13 AM, ambika wrote: > >> >> >> *From: *ambika <ambika at cse.iitb.ac.in <mailto:ambika at cse.iitb.ac.in>> >> *Date: *April 8, 2010 9:10:25 AM PDT >> *To: *llvmdev at cs.uiuc.edu

I have not made a separate pass. I have done it in single pass ie added ProfileLoaderPass , got information in PI and now I am planning to perform the analysis here itself. I was trying to print information using ProfileInfoPrinter pass but was unable to do it. namespace { class MyAna : public ModulePass { ProfileInfo *PI; public: static char ID; // Class identification, replacement

Chris Lattner wrote: > On Feb 15, 2008, at 3:01 PM, John Criswell wrote: > > >>> The convention is that 'remove' unlinks the object from the IR, and >>> 'erase' does remove and deletes the result. >>> >>> If someone wants to clarify this behavior in http://llvm.org/docs/ProgrammersManual.html#common >>> , it would be nice :)

Ah BTW... On 25.02.2010, at 17:33, ambika wrote: > ProfileInfo *PI; > PI = &getAnalysis<ProfileInfo>(); If this _in_ your pass, then you have to register the usage of this in the getAnalysisUsage() method: void getAnalysisUsage(AnalysisUsage &AU) const { AU.addRequired<ProfileInfo>(); } Also you have to ensure that the pass you are using is executed right

Yes, I've tried and it did not work On Tue, Jun 12, 2012 at 4:03 PM, Villmow, Micah <Micah.Villmow at amd.com>wrote: > Have you tried "opt -stats bitecode.bc > test.txt"?**** > > ** ** > > *From:* llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at cs.uiuc.edu] *On > Behalf Of *Rafael Parizi > *Sent:* Tuesday, June 12, 2012 11:36 AM > *To:*

On Thursday 09 August 2007 16:29, Devang Patel wrote: > On Aug 9, 2007, at 2:22 PM, David Greene wrote: > > A better question to ask is, how do I do this with llvm-gcc? My > > perusal of > > the code doesn't turn up any obvious places for options for alias > > analysis > > are registered. > > As you've figured out, not all options are exposed by

Hi, Shuguang Feng wrote: > Thanks for such a rapid response! > >> Don't know about Passes in the backend, but this could be a problem of >> an FunctionPassManager trying to use a ModulePass. > > I manually applied the patch you provided for llc (I'm using the 2.5 > release of LLVM not ToT) and it fixed my compilation error. When your > patch replaced the

>> I'm already working on this myself. Would you be interested in having >> this work contributed back to LLVM? > > If this were to better support target independent languages, it would > be very useful. If you're just trying to *reduce* the endianness > assumptions that leak through, I don't think it's a good approach. > There is just no way to solve

>> If you write your code as an llvm pass, you should automatically get >> notified of this stuff in a debug build. If you can't do that, >> manually interfacing to the LeakDetector can help. >> > It is an LLVM pass, it's a debug build, and the only assertion I got > was > the one in ~Value(). > > Should there have been an earlier assertion? You

Chris Lattner wrote: >>> If you write your code as an llvm pass, you should automatically get >>> notified of this stuff in a debug build. If you can't do that, >>> manually interfacing to the LeakDetector can help. >>> >>> >> It is an LLVM pass, it's a debug build, and the only assertion I got >> was >> the one in

On Oct 27, 2008, at 3:14 AM, Jay Foad wrote: >>> I'm already working on this myself. Would you be interested in >>> having >>> this work contributed back to LLVM? >> >> If this were to better support target independent languages, it would >> be very useful. If you're just trying to *reduce* the endianness >> assumptions that leak

Hi, this is the first in a series of patches to cleanup and improve the LLVM Profiling Infrastructure. First and foremost this patch removes duplicate functionality from ProfileInfoLoader and ProfileInfo: The ProfileInfoLoader performed not only the loading of the profile information but also some synthesis of block and function execution counts from edge profiling information. Since the