From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Jan Beulich" Subject: Re: [Pv-ops][PATCH] Netback multiple tasklet support Date: Wed, 13 Jan 2010 10:17:18 +0000 Message-ID: <4B4DABBE0200007800029A6F@vpn.id2.novell.com> References: <4FA716B1526C7C4DB0375C6DADBC4EA342A7A7E951@LONPMAILBOX01.citrite.net> <4FA716B1526C7C4DB0375C6DADBC4EA342A7A7E95E@LONPMAILBOX01.citrite.net> <4B182D87.6030901@goop.org> <4B187513.80003@goop.org> <4B200727.8040000@goop.org> <1260436078.23698.45463.camel@zakaz.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <1260436078.23698.45463.camel@zakaz.uk.xensource.com> Content-Disposition: inline List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Ian Campbell Cc: Steven Smith , Ian Pratt , "xen-devel@lists.xensource.com" , Jeremy Fitzhardinge , Dongxiao Xu List-Id: xen-devel@lists.xenproject.org >>> Ian Campbell 10.12.09 10:07 >>> >Subject: xen: ensure locking gnttab_copy_grant_page is safe against = interrupts. > >Now that netback processing occurs in a thread instead of a tasklet >gnttab_copy_grant_page needs to be safe against interrupts. > >The code is currently commented out in this tree but on 2.6.18 we = observed a >deadlock where the netback thread called gnttab_copy_grant_page, locked >gnttab_dma_lock for writing, was interrupted and on return from interrupt = the >network stack's TX tasklet ended up calling __gnttab_dma_map_page via the >hardware driver->swiotlb and tries to take gnttab_dma_lock for reading. > >Correct the commented code so we don't get bitten if/when it is re-enabled= . Wouldn't safe-against-softirq be sufficient then? Or if not (i.e. if meant to be generic), wouldn't the irq-safe variant need to be used independent of the new feature (and then also in the 2.6.18 tree)? Jan