similar to: [LLVMdev] Profile guided optimization

Hi all, I would like to know the status of profile-guided optimization. What enhancements could be done to the current implementation? The ideas for profile-guided transformations from [0] are still available? Many thanks. Gener [0] http://llvm.org/OpenProjects.html#profileguided -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hi, sorry to be so late with my answer, I was busy finishing my master thesis. On 03/21/2010 11:06 PM, Gener Omer wrote: > > I have finally started my project which is supposed to implement some > algorithms that uses profile information in order to optimize programs. > I have a very basic question for you. I have a very simple source code, > I compiled it with LLVM > >

On 13.06.2013, at 19:12, Diego Novillo <dnovillo at google.com> wrote: > On 2013-06-12 18:15 , Chandler Carruth wrote: >> >> On Wed, Jun 12, 2013 at 3:10 PM, Jakob Stoklund Olesen <stoklund at 2pi.dk> wrote: >> It predates the block frequency interface. It just needs to be hooked up, patches welcome. It would also be nice to remove the floating point computations

On Jun 12, 2013, at 2:23 PM, Diego Novillo <dnovillo at google.com> wrote: > In terms of optimizations, our (Google) experience is that > inlining is the key beneficiary of profile information. > Particularly, in big C++ applications. I expect to focus most > of my attention on the inliner. That sounds plausible to me. It seems like we might need a way of representing call

On Wed, Jun 12, 2013 at 2:55 PM, Jakob Stoklund Olesen <stoklund at 2pi.dk>wrote: > That sounds plausible to me. It seems like we might need a way of > representing call graph profiling in addition to the existing branch > probabilities? > Agreed. An important consideration here is WPO vs. LTO vs. TU-at-a-time call graphs. > FWIW, the greedy register allocator’s live range

On Jun 12, 2013, at 3:05 PM, Chandler Carruth <chandlerc at google.com> wrote: > > On Wed, Jun 12, 2013 at 2:55 PM, Jakob Stoklund Olesen <stoklund at 2pi.dk> wrote: > That sounds plausible to me. It seems like we might need a way of representing call graph profiling in addition to the existing branch probabilities? > > Agreed. An important consideration here is WPO

Apple folks are also gearing up to push on the PGO front. We are primarily interested in using instrumentation, rather than sampling, to collect profile info. However, I suspect the way profile ended up being used in the various optimization and codegen passes would be largely similar. There is also some interests in pursuing profile directed specialization. But that can wait. I think it makes

On 2013-06-12 18:15 , Chandler Carruth wrote: > > On Wed, Jun 12, 2013 at 3:10 PM, Jakob Stoklund Olesen > <stoklund at 2pi.dk <mailto:stoklund at 2pi.dk>> wrote: > > It predates the block frequency interface. It just needs to be > hooked up, patches welcome. It would also be nice to remove the > floating point computations from the spill placement

On Jun 18, 2013, at 2:48 PM, Eric Christopher <echristo at gmail.com> wrote: > On Tue, Jun 18, 2013 at 11:19 AM, Bob Wilson <bob.wilson at apple.com> wrote: >> >> On Jun 17, 2013, at 6:54 AM, Diego Novillo <dnovillo at google.com> wrote: >> >>> On 2013-06-15 14:18 , Evan Cheng wrote: >>>> Apple folks are also gearing up to push on the

On Wed, Jun 12, 2013 at 3:10 PM, Jakob Stoklund Olesen <stoklund at 2pi.dk>wrote: > It predates the block frequency interface. It just needs to be hooked up, > patches welcome. It would also be nice to remove the floating point > computations from the spill placement code. Cool, if Diego doesn't beat me to it, I may send you a patch as that seems easy and obviously

hello,everyone recently, I analyze the compiler optimization level, so I have searched the usage of profile-guided optimization of LLVM, but little result in google, therefore, could you please list detailed usage of profile-guided optimization? LLVM must be install with flags '--enable-profile'? A example will be pleasure. thanks von -------------- next part -------------- An HTML

Dear LLVM community, I am very happy to become a member of this great community. At the German Climate Research Center (DKRZ) I am a PhD student and write my thesis in python. The general topic is Compiler Optimization in High Performance Computing. My use cases are climate models, written in C/C++. In Frankfurt I am working at a HPC Cluster Center and my duties are to monitor and to optimize

On Jun 14, 2013, at 10:56 AM, Krzysztof Parzyszek <kparzysz at codeaurora.org> wrote: > > * Function argument value profiling: > - Is the function called with a small set of values for the parameters? > - Is there any value that is particularly frequent for any of the arguments? > - If the function takes pointers, are the pointers usually aligned? > - If the function

> > After the basic profile-based transformations are working, I would like to > add new sources of profile. Mainly, I am thinking of implementing Auto > FDO. > For those who are not familiar with what autoFDO is -- Auto FDO is originally called Sample Based FDO. Its main author is Dehao Chen @google, and Robert Hundt is the one of the main pushers of technology in Google. The

On 2013-06-15 16:39 , Benjamin Kramer wrote: > Do you want to take over this effort or should I poke more at it? Since you've already started, it's easier if you poke more at it. Thanks. I've got a whole bunch of other things to go through. Diego.

On Tue, Jun 18, 2013 at 11:19 AM, Bob Wilson <bob.wilson at apple.com> wrote: > > On Jun 17, 2013, at 6:54 AM, Diego Novillo <dnovillo at google.com> wrote: > >> On 2013-06-15 14:18 , Evan Cheng wrote: >>> Apple folks are also gearing up to push on the PGO front. We are primarily interested in using instrumentation, rather than sampling, to collect profile info.

On 6/12/2013 4:55 PM, Jakob Stoklund Olesen wrote: > > It seems like we might need a way of representing call graph profiling in addition to the existing branch probabilities? > Metadata on function definitions/declarations perhaps? I agree---the call graph profiling is of critical importance. I guess we should first see what types of data we already know may be useful and see how we

Hi, I would like to learn the details regarding what exactly PGO does for basic block layout/code placement optimizations in llvm. Could you please point me to some descriptions? Is it close to this paper (Karl Pettis and Robert C. Hansen. 1990. Profile guided code positioning. PLDI'90) http://perso.ensta-paristech.fr/~bmonsuez/Cours/B6-4/Articles/papers15.pdf? Whether it is purely

On Jun 17, 2013, at 7:03 AM, Benjamin Kramer <benny.kra at gmail.com> wrote: > > On 17.06.2013, at 15:56, Diego Novillo <dnovillo at google.com> wrote: > >> On 2013-06-15 16:39 , Benjamin Kramer wrote: >>> Do you want to take over this effort or should I poke more at it? >> >> Since you've already started, it's easier if you poke more at

On 17.06.2013, at 15:56, Diego Novillo <dnovillo at google.com> wrote: > On 2013-06-15 16:39 , Benjamin Kramer wrote: >> Do you want to take over this effort or should I poke more at it? > > Since you've already started, it's easier if you poke more at it. Thanks. I've got a whole bunch of other things to go through. OK, will do. Jakob any comments on the