From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Vrabel Subject: Re: [PATCH net-next v10 1/7] xen-netback: Move grant_copy_op array back into struct xenvif. Date: Wed, 4 Jun 2014 15:52:52 +0100 Message-ID: <538F32C4.8070409@citrix.com> References: <1401874247-3202-1-git-send-email-wei.liu2@citrix.com> <1401874247-3202-2-git-send-email-wei.liu2@citrix.com> <538F3152.80501@cogentembedded.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: paul.durrant@citrix.com, "Andrew J. Bennieston" , ian.campbell@citrix.com, zoltan.kiss@citrix.com, a.j.bennieston@gmail.com To: Sergei Shtylyov , Wei Liu , , Return-path: In-Reply-To: <538F3152.80501@cogentembedded.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org List-Id: netdev.vger.kernel.org On 04/06/14 15:46, Sergei Shtylyov wrote: > Hello. > > On 06/04/2014 01:30 PM, Wei Liu wrote: > >> From: "Andrew J. Bennieston" > >> This array was allocated separately in commit ac3d5ac2 ("xen-netback: >> fix guest-receive-side array sizes") due to it being very large, and a >> struct xenvif is allocated as the netdev_priv part of a struct >> net_device, i.e. via kmalloc() but falling back to vmalloc() if the >> initial alloc. fails. > >> In preparation for the multi-queue patches, where this array becomes >> part of struct xenvif_queue and is always allocated through vzalloc(), >> move this back into the struct xenvif. > > Won't this cause an allocation failure and so break bisection at this > patch? No. As Andrew already said: "...falling back to vmalloc() if the initial alloc. fails." David From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Vrabel Subject: Re: [PATCH net-next v10 1/7] xen-netback: Move grant_copy_op array back into struct xenvif. Date: Wed, 4 Jun 2014 15:52:52 +0100 Message-ID: <538F32C4.8070409@citrix.com> References: <1401874247-3202-1-git-send-email-wei.liu2@citrix.com> <1401874247-3202-2-git-send-email-wei.liu2@citrix.com> <538F3152.80501@cogentembedded.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <538F3152.80501@cogentembedded.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Sergei Shtylyov , Wei Liu , xen-devel@lists.xen.org, netdev@vger.kernel.org Cc: paul.durrant@citrix.com, "Andrew J. Bennieston" , ian.campbell@citrix.com, zoltan.kiss@citrix.com, a.j.bennieston@gmail.com List-Id: xen-devel@lists.xenproject.org On 04/06/14 15:46, Sergei Shtylyov wrote: > Hello. > > On 06/04/2014 01:30 PM, Wei Liu wrote: > >> From: "Andrew J. Bennieston" > >> This array was allocated separately in commit ac3d5ac2 ("xen-netback: >> fix guest-receive-side array sizes") due to it being very large, and a >> struct xenvif is allocated as the netdev_priv part of a struct >> net_device, i.e. via kmalloc() but falling back to vmalloc() if the >> initial alloc. fails. > >> In preparation for the multi-queue patches, where this array becomes >> part of struct xenvif_queue and is always allocated through vzalloc(), >> move this back into the struct xenvif. > > Won't this cause an allocation failure and so break bisection at this > patch? No. As Andrew already said: "...falling back to vmalloc() if the initial alloc. fails." David