This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: g++.dg/tree-ssa/pr45453.C time out


> On Wed, Oct 31, 2012 at 3:17 PM, Paolo Carlini <paolo.carlini@oracle.com> wrote:
> > Hi,
> >
> > whoever a few days ago or so broke this test, can please either fix the
> > testcase, the compiler or just xfail for now the testcase itself, to avoid
> > everybody the waste of time?
> >
> > If you want me to do go ahead with option 3 above, just let me know!
> 
> It was honza - we endlessly recurse in estimate_calls_size_and_time.
> 
> Honza, please make fixing this a priority.

I am looing into it now. Did not noticed that - at -O3 it hits the ipa-cp issue
I hope Martin will fix.

The problem here is not infinite recursion, we keep inlning the foo into itself
since indirect inlining somehow misses  the limits on recursion depth and the
function is empty and thus inlining has no cost.

Honza
> 
> Thanks,
> Richard.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]