This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: core changes for mep port
- From: "gcc-gpu at adrenalin-junkie dot net" <gcc-gpu at adrenalin-junkie dot net>
- To: "DJ Delorie" <dj at redhat dot com>, "" <iant at google dot com>
- Cc: "" <gcc at gcc dot gnu dot org>
- Date: Fri, 16 Mar 2007 13:17:55 +0000
- Subject: Re: core changes for mep port
- Reply-to: "gcc-gpu at adrenalin-junkie dot net" <gcc-gpu at adrenalin-junkie dot net>
Guys - what branch/tag are you looking at doing this on?
I've been in discussions with Dorit about SPMD utilisation of graphics cards (AMD CTM) which has more inline with parallel asymetric architectures than it does with simple SIMD optimisations.
Cheers,
Nick.
> -------Original Message-------
> From: DJ Delorie <dj@redhat.com>
> Subject: Re: core changes for mep port
> Sent: 16 Mar '07 03:09
>
>
> > Do you mean where is the best place to call these functions?
>
> Yup.
>
> > Look at the calls to cgraph_mark_edge in ipa-inline.c
>
> There is no such function.??I couldn't find anything in ipa-inline
> that (1) had access to both ends of the call edge, (2) was actually
> called, and (3) was called before the inlining happened.
>
> > and look at inline_forbidden_p--there I assume that cfun will hold
> > the caller.
>
> cfun holds the callee.??I found a lot of functions that only had
> access to the callee (no help there).??I also found lots of functions
> that looked promising but were never called.
>