From: Rusty Russell <rusty@rustcorp.com.au>
To: Andrew Morton <akpm@digeo.com>
Cc: dipankar@in.ibm.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] kmalloc_percpu 
In-reply-to: Your message of "Mon, 05 May 2003 18:52:48 MST."
             <20030505185248.3c3a478f.akpm@digeo.com> 
Date: Tue, 06 May 2003 14:11:27 +1000
Message-Id: <20030506041213.B80E42C2E0@lists.samba.org>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 961
Lines: 27

In message <20030505185248.3c3a478f.akpm@digeo.com> you write:
> Rusty Russell <rusty@rustcorp.com.au> wrote:
> >
> > > +#define PERCPU_POOL_SIZE 32768
> > > 
> > > What's this?
> > 
> > OK.  It has a size restriction: PERCPU_POOL_SIZE is the maximum total
> > kmalloc_percpu + static DECLARE_PER_CPU you'll get, ever.  This is the
> > main downside.  It's allocated at boot.
> 
> And is subject to fragmentation.

Absolutely.  However, we're looking at an allocation at module insert,
and maybe at each mount (if used for your fuzzy counters).  Until we
see this in practice, I don't think complicating the allocator is
worth it.

Cheers,
Rusty.
--
  Anyone who quotes me in their sig is an idiot. -- Rusty Russell.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/