This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [tree-ssa] Avoid nondeterminism in tree-ssanames
- From: Steven Bosscher <s dot bosscher at student dot tudelft dot nl>
- To: Andrew MacLeod <amacleod at redhat dot com>
- Cc: Jeff Law <law at redhat dot com>, Geoff Keating <geoffk at geoffk dot org>,gcc-patches <gcc-patches at gcc dot gnu dot org>, Jan Hubicka <jh at suse dot cz>
- Date: Tue, 2 Dec 2003 00:42:25 +0100
- Subject: Re: [tree-ssa] Avoid nondeterminism in tree-ssanames
Hi,
> So what are we doing to the GC? Are we fixing the zone collector so we
> can allocate pages to specific purposes?
Yes, I'm working on that.
The zone collector isn't broken, it just needs a few power-ups. I've just
commited a patch to mainline that allows you to create zones on the fly. Next
I am going to try and put SSA names and PHI nodes in a separate zone. Once
that works, anything is possible ;-)
> And will there or will there
> not be a free list mainatined within the zone?
Not within the zone collector, only after collecting objects (doh!). You need
to maintain your own objects cache.
For SSA names, I suppose we need a free list until there is a ggc_free.
Recycling SSA names saved a lot of memory...
> I've abstracted out the use and def operands, Im going to make it work
> for vuse and vdef operands as well. These seem like very likely
> candidates for a zone as well, if it includes a free list, and can
> handle small allocations easily.
Hmm I thought ggc-page already handles small allocations quite nicely. Just
not the deallocations :-)
Gr.
Steven