Re: [ping patch] Predict for loop exits in short-circuit conditions

2012-10-08 Thread Richard Guenther
On Mon, Oct 8, 2012 at 12:01 PM, Jan Hubicka wrote: >> On Mon, Oct 8, 2012 at 11:04 AM, Jan Hubicka wrote: >> >> On Mon, Oct 8, 2012 at 4:50 AM, Dehao Chen wrote: >> >> > Attached is the updated patch. Yes, if we add a VRP pass before >> >> > profile pass, this patch would be unnecessary. Should

Re: [ping patch] Predict for loop exits in short-circuit conditions

2012-10-08 Thread Jan Hubicka
> On Mon, Oct 8, 2012 at 11:04 AM, Jan Hubicka wrote: > >> On Mon, Oct 8, 2012 at 4:50 AM, Dehao Chen wrote: > >> > Attached is the updated patch. Yes, if we add a VRP pass before > >> > profile pass, this patch would be unnecessary. Should we add a VRP > >> > pass? > >> > >> No, we don't want VR

Re: [ping patch] Predict for loop exits in short-circuit conditions

2012-10-08 Thread Richard Guenther
On Mon, Oct 8, 2012 at 11:04 AM, Jan Hubicka wrote: >> On Mon, Oct 8, 2012 at 4:50 AM, Dehao Chen wrote: >> > Attached is the updated patch. Yes, if we add a VRP pass before >> > profile pass, this patch would be unnecessary. Should we add a VRP >> > pass? >> >> No, we don't want VRP in early opt

Re: [ping patch] Predict for loop exits in short-circuit conditions

2012-10-08 Thread Jan Hubicka
> On Mon, Oct 8, 2012 at 4:50 AM, Dehao Chen wrote: > > Attached is the updated patch. Yes, if we add a VRP pass before > > profile pass, this patch would be unnecessary. Should we add a VRP > > pass? > > No, we don't want VRP in early optimizations. I am not quite sure about that. VRP 1) make

Re: [ping patch] Predict for loop exits in short-circuit conditions

2012-10-08 Thread Richard Guenther
On Mon, Oct 8, 2012 at 4:50 AM, Dehao Chen wrote: > Attached is the updated patch. Yes, if we add a VRP pass before > profile pass, this patch would be unnecessary. Should we add a VRP > pass? No, we don't want VRP in early optimizations. Richard. > Thanks, > Dehao > > On Sat, Oct 6, 2012 at 9:

Re: [ping patch] Predict for loop exits in short-circuit conditions

2012-10-07 Thread Dehao Chen
Attached is the updated patch. Yes, if we add a VRP pass before profile pass, this patch would be unnecessary. Should we add a VRP pass? Thanks, Dehao On Sat, Oct 6, 2012 at 9:38 AM, Jan Hubicka wrote: >> ping^2 >> >> Honza, do you think this patch can make into 4.8 stage 1? > > + if (check

Re: [ping patch] Predict for loop exits in short-circuit conditions

2012-10-06 Thread Jan Hubicka
> ping^2 > > Honza, do you think this patch can make into 4.8 stage 1? + if (check_value_one ^ integer_onep (val)) Probably better as != (especially because GNU coding standard allows predicates to return more than just boolean) +{ + edge e1; + edge_iterator ei; + tree

Re: [ping patch] Predict for loop exits in short-circuit conditions

2012-10-06 Thread Dehao Chen
ping^2 Honza, do you think this patch can make into 4.8 stage 1? Thanks, Dehao On Wed, Sep 26, 2012 at 2:34 PM, Dehao Chen wrote: > http://gcc.gnu.org/ml/gcc-patches/2012-05/msg01975.html > > Thanks, > Dehao

[ping patch] Predict for loop exits in short-circuit conditions

2012-09-26 Thread Dehao Chen
http://gcc.gnu.org/ml/gcc-patches/2012-05/msg01975.html Thanks, Dehao

Re: Predict for loop exits in short-circuit conditions

2012-05-30 Thread Dehao Chen
Hi, I've updated the patch to invoke predict_extra_loop_exits in the right place. Attached is the new patch. Bootstrapped and passed gcc testsuite. Thanks, Dehao Index: testsuite/g++.dg/predict-loop-exit-1.C === --- testsuite/g++.d

Predict for loop exits in short-circuit conditions

2012-05-27 Thread Dehao Chen
Hi, This patch implements static branch prediction for loop exit conditions that cannot be identified by original heuristic (edge->dst is outside of the loop). It can find extra loop exits for short circuit conditions. Bootstrapped and passed gcc testsuite. Is it ok for trunk? Thanks, Dehao gc