This is the mail archive of the 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: Reduce complette unrolling & peeling limits

On Sun, 18 Nov 2012, Jan Hubicka wrote:
> > > > this patch reduces max-peeled-insns and max-completely-peeled-insns from 400
> > > > to 100.  The reason why I am doing this is that I want to reduce code bloat
> > > > caused by my cunroll work that enabled a lot more unrolling then previously
> > > > causing considerable code size regression at -O3.
> > >
> > > Did you notice that gcc.c-torture/compile/pr43186.c regressed?  It now again
> > > takes a while to compile, so times out on slow machines:
> >
> > I did not :(.  I am currently on a trip, but will take a look on tuesday.
> > If it seems to disturb testing, please just revert the patch for time being.
> OK, here are multiple issues.
> 1) recursive inlining makes huge loop nest (of 18 loops)
> 2) SCEV is very slow on answering simple_iv tests in this case becuase it walks
>    the nest
> 3) unroller is computing loop body size even when it is clear the body is much larger
>    than the limit (the outer loop has 78000 instructions)
> I will prepare patches to fix those issues.

The recent (well, a week ago) params.def change also regressed
gfortran.dg/reassoc_4.f almost everywhere; see PR55452.

I guess a fix is fairly trivial, I just don't know to what.

brgds, H-P

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