From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jim Ramsay Subject: Re: [PATCH 3/3] Add disk hotswap support to libata RESEND #2 Date: Wed, 24 Aug 2005 10:12:31 -0600 Message-ID: <4789af9e05082409121cc6870@mail.gmail.com> References: <355e5e5e05080103021a8239df@mail.gmail.com> <4789af9e050823124140eb924f@mail.gmail.com> <4789af9e050823154364c8e9eb@mail.gmail.com> <430BA990.9090807@mvista.com> <430BCB41.5070206@s5r6.in-berlin.de> <355e5e5e05082407031138120a@mail.gmail.com> <4789af9e05082408111c4a6294@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Return-path: In-Reply-To: <4789af9e05082408111c4a6294@mail.gmail.com> Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org To: Lukasz Kosewski Cc: Stefan Richter , linux-scsi@vger.kernel.org, linux-ide@vger.kernel.org, linux-kernel@vger.kernel.org List-Id: linux-ide@vger.kernel.org On 8/24/05, Jim Ramsay wrote: > On 8/24/05, Lukasz Kosewski wrote: > > On 8/24/05, Stefan Richter wrote: > > > >> Timers appear to operate in an atomic context, so timers should not be > > > >> allowed to call scsi_remove_device, which eventually schedules. > > > >> > > > >> Any suggestions on the best way to fix this? > > > > > > > > Workqueue, perhaps. > > > > Perhaps. Actually, of course :) > > How about the existing ata_wq workqueue? This makes sense. When the > timer expires, it adds a task to this queue. Note to self - No, you cannot use the exsting 'ata_wq' workqueue - The plug-in events need to put other work on the queue during the hotplug event... and of course this deadlocks since you're in the queuethread already. -- Jim Ramsay "Me fail English? That's unpossible!"