From mboxrd@z Thu Jan  1 00:00:00 1970
From: Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
Subject: Re: [PATCH 3/6] fs: Introduce a per_cpu last_ino allocator
Date: Thu, 27 Nov 2008 04:46:15 -0500
Message-ID: <20081127094615.GC13860@infradead.org>
References: <Pine.LNX.4.64.0811201727070.9089@quilx.com> <20081121083044.GL16242@elte.hu> <49267694.1030506@cosmosbay.com> <20081121.010508.40225532.davem@davemloft.net> <4926AEDB.10007@cosmosbay.com> <4926D022.5060008@cosmosbay.com> <20081121152148.GA20388@elte.hu> <4926D39D.9050603@cosmosbay.com> <20081121153453.GA23713@elte.hu> <492DDC88.2050305@cosmosbay.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: Ingo Molnar <mingo-X9Un+BFzKDI@public.gmane.org>, David Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>,
	"Rafael J. Wysocki" <rjw-KKrjLPT3xs0@public.gmane.org>, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	kernel-testers-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Mike Galbraith <efault-Mmb7MZpHnFY@public.gmane.org>,
	Peter Zijlstra <a.p.zijlstra-/NLkJaSkS4VmR6Xm/wNWPw@public.gmane.org>,
	Linux Netdev List <netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
	Christoph Lameter <cl-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>,
	Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
To: Eric Dumazet <dada1-fPLkHRcR87vqlBn2x/YWAg@public.gmane.org>
Return-path: <kernel-testers-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
Content-Disposition: inline
In-Reply-To: <492DDC88.2050305-fPLkHRcR87vqlBn2x/YWAg@public.gmane.org>
Sender: kernel-testers-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
List-Id: netdev.vger.kernel.org

On Thu, Nov 27, 2008 at 12:32:24AM +0100, Eric Dumazet wrote:
> new_inode() dirties a contended cache line to get inode numbers.
>
> Solve this problem by providing to each cpu a per_cpu variable,
> feeded by the shared last_ino, but once every 1024 allocations.
>
> This reduce contention on the shared last_ino.
>
> Note : last_ino_get() method must be called with preemption
> disabled on SMP.

Looks a little clumsy.  One idea might be to have a special slab for
synthetic inodes using new_inode and only assign it on the first
allocation and after that re-use it.