From: Evgeniy Polyakov Subject: Re: problems with e1000 and jumboframes Date: Fri, 4 Aug 2006 10:15:13 +0400 Message-ID: <20060804061513.GB413@2ka.mipt.ru> References: <41b516cb0608031334s6e159e99tb749240f44ae608d@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=koi8-r Cc: Chris Leech , arnd@arndnet.de, olel@ans.pl, linux-kernel@vger.kernel.org, netdev@vger.kernel.org Return-path: Received: from relay.2ka.mipt.ru ([194.85.82.65]:20403 "EHLO 2ka.mipt.ru") by vger.kernel.org with ESMTP id S1751251AbWHDGPq (ORCPT ); Fri, 4 Aug 2006 02:15:46 -0400 To: Herbert Xu Content-Disposition: inline In-Reply-To: Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Fri, Aug 04, 2006 at 03:59:37PM +1000, Herbert Xu (herbert@gondor.apana.org.au) wrote: > Chris Leech wrote: > > > > We could try and only use page allocations for older e1000 devices, > > putting headers and payload into skb->frags and copying the headers > > out into the skb->data area as needed for processing. That would do > > away with large allocations, but in Jesse's experiments calling > > alloc_page() is slower than kmalloc(), so there can actually be a > > performance hit from trying to use page allocations all the time. > > Interesting. Could you guys post figures on alloc_page speed vs. kmalloc? They probalby measured kmalloc cache access, which only falls to alloc_pages when cache is refilled, so it will be faster for some short period of time, but in general (especially for such big-sized allocations) it is essencially the same. > Also, getting memory slower is better than not getting them at all :) Sure. > -- > Visit Openswan at http://www.openswan.org/ > Email: Herbert Xu ~{PmV>HI~} > Home Page: http://gondor.apana.org.au/~herbert/ > PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt -- Evgeniy Polyakov