From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rob Gardner Subject: Re: [PATCH net,stable] net: usb: Add HP hs2434 device to ZLP exception table Date: Wed, 28 Aug 2013 17:18:12 -0600 Message-ID: <20130828171812.7097dc8334046b464486d4a3@gmail.com> References: <20130825182913.2865aa1a8cefe329ab9c7a85@gmail.com> <20130828.182332.824853790331642993.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org, linux-usb@vger.kernel.org, bjorn@mork.no, oliver@neukum.org To: David Miller Return-path: Received: from mail-pd0-f171.google.com ([209.85.192.171]:44501 "EHLO mail-pd0-f171.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752119Ab3H1XSP convert rfc822-to-8bit (ORCPT ); Wed, 28 Aug 2013 19:18:15 -0400 In-Reply-To: <20130828.182332.824853790331642993.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, 28 Aug 2013 18:23:32 -0400 (EDT) David Miller wrote: > From: Rob Gardner > Date: Sun, 25 Aug 2013 18:29:13 -0600 >=20 > > From 018343ce2e679d97283fb51da25c43aa876d087a Mon Sep 17 00:00:00 2= 001 > > From: Rob Gardner > > Date: Sun, 25 Aug 2013 16:02:23 -0600 > > Subject: [PATCH net,stable] net: usb: Add HP hs2434 device to ZLP e= xception table > >=20 > > This patch adds another entry (HP hs2434 Mobile Broadband) to the l= ist > > of exceptional devices that require a zero length packet in order t= o > > function properly. This list was added in commit 844e88f0. The hs24= 34 > > is manufactured by Sierra Wireless, who also produces the MC7710, > > which the ZLP exception list was created for in the first place. So > > hopefully it is just this one producer's devices that will need thi= s > > workaround. > >=20 > > Tested on a DM1-4310NR HP notebook, which does not function without= this > > change. > >=20 > > Signed-off-by: Rob Gardner >=20 > Applied and queued up for -tablt, thanks. >=20 > On the topic of dealing with detection of this issue, how harmful is = the > workaround on chips that don't need it? Will the driver not work in = that > case? Bj=F8rn says it could have a minor performance impact, but should not cause any loss of functionality. I don't think there's any hard data on what the performance cost might be but I know he would eventually like to get rid of the exception list and live with the possible performance impact if it means functionality for more devices. --=20 Rob Gardner