This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Some work on the zone collector
- From: Daniel Berlin <dberlin at dberlin dot org>
- To: Steven Bosscher <steven at gcc dot gnu dot org>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Wed, 26 Nov 2003 19:49:35 -0500 (EST)
- Subject: Re: [PATCH] Some work on the zone collector
- References: <200311270126.18727.steven@gcc.gnu.org>
On Thu, 27 Nov 2003, Steven Bosscher wrote:
> Hi,
>
> This patch for mainline is a missing piece for runtime allocatable
> GC zones. Bootstrapped and regtested with the zone collector on
> i686-pc-linux-gnu.
>
> There are lots of new FAILs compared to mainline with the page collector,
> so I'm re-running the test suite know with a clean tree and the zone
> collector to see if it's my patch that breaks it (though I wouldn't know
> how it possibly could...).
Hmmmm. That's interesting. it was regtested when it was first submitted,
and bootstrapped with GCAC on. If your patch causes afilurse, something
is broken.
Cookie checking usually helps when things break.
>
> The destroy_ggc_zone function may still require some work, I'm not sure
> if there are other things to free/rechain/etc. But it's unused right
> now anyway, so it cannot break anything.
>
> Assuming these PCH fails are unrelated to this patch, OK for mainline?
I never tested the PCH support beyond trivial PCH's (i did, however, test
it trivially on some simple PCH cases) for the zone collector, which has
to do it slightly differently.
Libstdc++ pch using wasn't working at the time, so i had no easy way to
test.
I'm happy to look at PCH related problems for gc-zone (which i'm sure
exist, i'm not arrogant enough to believe i got it completely righ the
first time :P)