From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Subject: Re: [PATCH][RFC] network splice receive Date: Mon, 11 Jun 2007 10:00:53 +0200 Message-ID: <20070611080053.GD21969@kernel.dk> References: <20070607145818.GA26491@2ka.mipt.ru> <20070608074823.GG7341@kernel.dk> <20070608.010629.52902577.davem@davemloft.net> <20070608083853.GH7341@kernel.dk> <20070608085620.GC11488@2ka.mipt.ru> <20070608090439.GK7341@kernel.dk> <20070608135819.GA14302@2ka.mipt.ru> <20070608141452.GR7341@kernel.dk> <20070608145724.GA14561@2ka.mipt.ru> <20070608153011.GA2375@2ka.mipt.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: David Miller , netdev@vger.kernel.org To: Evgeniy Polyakov Return-path: Received: from brick.kernel.dk ([80.160.20.94]:25946 "EHLO kernel.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752404AbXFKIC7 (ORCPT ); Mon, 11 Jun 2007 04:02:59 -0400 Content-Disposition: inline In-Reply-To: <20070608153011.GA2375@2ka.mipt.ru> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Fri, Jun 08 2007, Evgeniy Polyakov wrote: > Size of the received file is bigger than file sent, file contains repeated > blocks of data sometimes. Cloned skb usage is likely too big overhead, > although for receiving fast clone is unused in most cases, so there > might be some gain. That was actually a new bug, here: plen -= *offset; poff += *offset; in __skb_slice_bits(), we should only subtract the offset from plen, not add to poff. Then we just create some weird hole without any meaning. So remove those two poff additions in the two loops, and the size issue is resolved at least. -- Jens Axboe