From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: Queue upcall locking (was: [dm-devel] [RFC][PATCH] fix dm_any_congested() to properly sync up with suspend code path) Date: Mon, 10 Nov 2008 15:46:57 +0100 Message-ID: <1226328417.7685.198.camel@twins> References: <1225944008.14830.1101.camel@chandra-ubuntu> <20081110135401.GA11606@infradead.org> <1226327210.7685.173.camel@twins> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Mikulas Patocka Cc: Christoph Hellwig , Chandra Seetharaman , Alasdair G Kergon , dm-devel , linux-kernel@vger.kernel.org, axboe@kernel.dk List-Id: dm-devel.ids On Mon, 2008-11-10 at 09:32 -0500, Mikulas Patocka wrote: > > On Mon, 10 Nov 2008, Peter Zijlstra wrote: > > > On Mon, 2008-11-10 at 09:19 -0500, Mikulas Patocka wrote: > > > On Mon, 10 Nov 2008, Christoph Hellwig wrote: > > > > > > > On Mon, Nov 10, 2008 at 08:11:51AM -0500, Mikulas Patocka wrote: > > > > > For upstream Linux developers: you are holding a spinlock and calling > > > > > bdi*_congested functions that can take indefinite amount of time (there > > > > > are even users reporting having 50 disks in one logical volume or so). I > > > > > think it would be good to move these calls out of spinlocks. > > > > > > > > Umm, they shouldn't block that long, as that completely defeats their > > > > purpose. These functions are mostly used to avoid throwing more I/O at > > > > a congested device if pdflush could do more useful things instead. But > > > > if it blocks in those functions anyway we wouldn't have to bother using > > > > them. Do you have more details about the uses cases when this happens > > > > and where the routines spend so much time? > > > > > > For device mapper, congested_fn asks every device in the tree and make OR > > > of their bits --- so if the user has 50 devices, it asks them all. > > > > > > For md-linear, md-raid0, md-raid1, md-raid10 and md-multipath it does the > > > same --- asking every device. > > > > > > If you have a better idea how to implement congested_fn, say it. > > > > Fix the infrastructure by adding a function call so that you can have > > the individual devices report their congestion state to the aggregate. > > > > Then congestion_fn can return a valid state in O(1) because the state is > > keps up-to-date by the individual state changes. > > > > IOW, add a set_congested_fn() and clear_congested_fn(). > > If you have a physical disk that has many LVM volumes on it, you end up in > a situation when disk congestion state change is reported to all the > volumes. So it will create O(n) problem at the other side. *sigh* I can almost understand why people want to use lvm to combine multiple disks, but why make the partition thing even worse...