From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Hutchings Subject: Re: [PATCH 1/1] TX throttling bug-fixing patch of AX88179_178A Date: Mon, 22 Jul 2013 20:47:51 +0100 Message-ID: <1374522471.1635.48.camel@bwh-desktop.uk.level5networks.com> References: <1374311809-4155-1-git-send-email-freddy@asix.com.tw> <1374512879.4990.19.camel@edumazet-glaptop> <1374518312.1635.29.camel@bwh-desktop.uk.level5networks.com> <1374518841.4990.26.camel@edumazet-glaptop> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: Grant Grundler , Freddy Xin , David Miller , netdev , , LKML , ASIX Louis =?UTF-8?Q?=5B=E8=98=87=E5=A8=81=E9=99=B8=5D?= , "Allan Chou" To: Eric Dumazet Return-path: In-Reply-To: <1374518841.4990.26.camel@edumazet-glaptop> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Mon, 2013-07-22 at 11:47 -0700, Eric Dumazet wrote: > On Mon, 2013-07-22 at 19:38 +0100, Ben Hutchings wrote: > > On Mon, 2013-07-22 at 11:29 -0700, Grant Grundler wrote: > > > On Mon, Jul 22, 2013 at 10:07 AM, Eric Dumazet wrote: > > > ... > > > > I guess that if a driver does not advertise NETIF_F_SG, this > > > > skb_linearize() call is not needed : All frames reaching your xmit > > > > function should already be linear > > > > > > As Ben Hutchings pointed out, hw_features is still setting this...but > > > I'm not sure how that matters. > > > > > > ax88179_set_features() doesn't allow setting SG or TSO features. But > > > I expect it would be "not too difficult" to add such that ethtool > > > could set those features after boot. > > [...] > > > > It already can. That's what putting feature flags in hw_features does. > > My original concern, that inspired this patch, was to remove SG support, > as this driver does not have SG support at all. > > Linearize a full TSO packet needs order-5 allocations, thats likely to > fail and lead to very slow TCP performance, because it will only rely on > retransmits. The driver could set gso_max_size to reduce that problem. But I rather doubt that TSO followed by skb_linearize() significantly improves throughput or CPU-efficiency. (If the device has a 1G link but is connected to the host through a USB 2.0 port, then USB is the bottleneck and TSO could improve throughput a few percent. But that's a silly configuration.) The real solution would be for someone to add SG support to the usbnet core. Trying to support 1GbE with only linear skbs is not a great idea... and it can only be a matter of time before there is USB ultra speed (or whatever comes after 'super') with 10GbE devices... Ben. -- Ben Hutchings, Staff Engineer, Solarflare Not speaking for my employer; that's the marketing department's job. They asked us to note that Solarflare product names are trademarked.