From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754104AbaCMLN0 (ORCPT ); Thu, 13 Mar 2014 07:13:26 -0400 Received: from smtp02.citrix.com ([66.165.176.63]:44933 "EHLO SMTP02.CITRIX.COM" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753437AbaCMLNY (ORCPT ); Thu, 13 Mar 2014 07:13:24 -0400 X-IronPort-AV: E=Sophos;i="4.97,646,1389744000"; d="scan'208";a="109519158" Date: Thu, 13 Mar 2014 11:13:22 +0000 From: Wei Liu To: Ian Campbell CC: David Vrabel , Zoltan Kiss , , , , , Subject: Re: [Xen-devel] [PATCH net-next v7 4/9] xen-netback: Introduce TX grant mapping Message-ID: <20140313111322.GB16807@zion.uk.xensource.com> References: <1394142511-14827-1-git-send-email-zoltan.kiss@citrix.com> <1394142511-14827-5-git-send-email-zoltan.kiss@citrix.com> <1394706812.25873.28.camel@kazak.uk.xensource.com> <53218EDA.8040904@citrix.com> <1394708555.25873.43.camel@kazak.uk.xensource.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Disposition: inline In-Reply-To: <1394708555.25873.43.camel@kazak.uk.xensource.com> User-Agent: Mutt/1.5.21 (2010-09-15) X-DLP: MIA1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Mar 13, 2014 at 11:02:35AM +0000, Ian Campbell wrote: > On Thu, 2014-03-13 at 10:56 +0000, David Vrabel wrote: > > On 13/03/14 10:33, Ian Campbell wrote: > > > On Thu, 2014-03-06 at 21:48 +0000, Zoltan Kiss wrote: > > >> @@ -135,13 +146,31 @@ struct xenvif { > > >> pending_ring_idx_t pending_cons; > > >> u16 pending_ring[MAX_PENDING_REQS]; > > >> struct pending_tx_info pending_tx_info[MAX_PENDING_REQS]; > > >> + grant_handle_t grant_tx_handle[MAX_PENDING_REQS]; > > >> > > >> /* Coalescing tx requests before copying makes number of grant > > >> * copy ops greater or equal to number of slots required. In > > >> * worst case a tx request consumes 2 gnttab_copy. > > >> */ > > >> struct gnttab_copy tx_copy_ops[2*MAX_PENDING_REQS]; > > >> - > > >> + struct gnttab_map_grant_ref tx_map_ops[MAX_PENDING_REQS]; > > >> + struct gnttab_unmap_grant_ref tx_unmap_ops[MAX_PENDING_REQS]; > > > > > > I wonder if we should break some of these arrays into separate > > > allocations? Wasn't there a problem with sizeof(struct xenvif) at one > > > point? > > > > alloc_netdev() falls back to vmalloc() if the kmalloc failed so there's > > no need to split these structures. > > Is vmalloc space in abundant supply? For some reason I thought it was > limited (maybe that's a 32-bit only limitation?) 32-bit has a limitation of 128MB by default. 64-bit has much larger address space. Wei.