From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kara Subject: Re: [PATCH 0/11] RFC quota scalability V1 Date: Wed, 6 Oct 2010 11:44:31 +0200 Message-ID: <20101006094430.GA3676@quack.suse.cz> References: <1286302827-31043-1-git-send-email-dmonakhov@gmail.com> <87aamraje7.fsf@dmon-lap.sw.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-fsdevel@vger.kernel.org, jack@suse.cz, hch@infradead.org To: Dmitry Return-path: Received: from cantor.suse.de ([195.135.220.2]:50419 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751140Ab0JFJp1 (ORCPT ); Wed, 6 Oct 2010 05:45:27 -0400 Content-Disposition: inline In-Reply-To: <87aamraje7.fsf@dmon-lap.sw.ru> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Wed 06-10-10 11:08:00, Dmitry wrote: > On Tue, 5 Oct 2010 22:20:16 +0400, Dmitry Monakhov wrote: > Sorry, seems like my script screw-up subject header. > > This patch set is my first attempt to make quota code scalable. > > Main goal of this patch-set is to split global locking to per-sb basis. > > Please review it and provide your opinion. > > > > Future plans: > > * More scalability for single super_block > > ** Remove dqptr_sem > > ** Redesing dquot ref accounting interface. > > ** Make fast path for charge function lockless > > > > quota: add wrapper function > > quota: Convert dq_state_lock to per-sb dq_state_lock > > quota: add quota format lock > > quota: make dquot lists per-sb > > quota: make per-sb hash array > > quota: remove global dq_list_lock > > quota: rename dq_lock > > quota: make per-sb dq_data_lock > > quota: protect dquot mem info with objects's lock > > quota: drop dq_data_lock where possible > > quota: relax dq_data_lock dq_lock locking consistency So one generic question: I see you split global locks into per-sb ones. Does it bring substantial advantage? Because from my experience common systems have just one or maybe two filesystems which are heavily used... Honza -- Jan Kara SUSE Labs, CR