From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Sun, 05 Jun 2005 12:52:49 -0700 (PDT) Message-Id: <20050605.125249.104050648.davem@davemloft.net> Subject: Re: Avoiding external fragmentation with a placement policy Version 12 From: "David S. Miller" In-Reply-To: <42A10ED2.7020205@yahoo.com.au> References: <42A10ED2.7020205@yahoo.com.au> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org From: Nick Piggin Date: Sat, 04 Jun 2005 12:15:46 +1000 Return-Path: To: nickpiggin@yahoo.com.au Cc: herbert@gondor.apana.org.au, mbligh@mbligh.org, jschopp@austin.ibm.com, mel@csn.ul.ie, linux-mm@kvack.org, linux-kernel@vger.kernel.org, akpm@osdl.org List-ID: > Herbert Xu wrote: > > With Dave's latest super-TSO patch, TCP over loopback will only be > > doing order-0 allocations in the common case. UDP and others may > > still do large allocations but that logic is all localised in > > ip_append_data. > > > > So if we wanted we could easily remove most large allocations over > > the loopback device. > > I would be very interested to look into that. I would be > willing to do benchmarks on a range of machines too if > that would be of any use to you. Even without the super-TSO patch, we never do larger than PAGE_SIZE allocations for sendmsg() when the device is scatter-gather capable (as indicated in netdev->flags). Loopback does set this bit. This PAGE_SIZE limit comes from net/ipv4/tcp.c:select_size(). -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: aart@kvack.org