From mboxrd@z Thu Jan 1 00:00:00 1970 From: Greg Kroah-Hartman Subject: Re: [PATCH v2] usb: xhci: fix interrupt transfer error happened on MTK platforms Date: Fri, 7 Sep 2018 10:56:45 +0200 Message-ID: <20180907085645.GA17270@kroah.com> References: <20180907074251.GA2079@kroah.com> <1536309826.32173.45.camel@mhfsdcap03> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <1536309826.32173.45.camel@mhfsdcap03> Sender: stable-owner@vger.kernel.org To: Chunfeng Yun Cc: Mathias Nyman , Felipe Balbi , Matthias Brugger , Sean Wang , linux-usb@vger.kernel.org, devicetree@vger.kernel.org, linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-mediatek@lists.infradead.org, stable@vger.kernel.org List-Id: linux-mediatek@lists.infradead.org On Fri, Sep 07, 2018 at 04:43:46PM +0800, Chunfeng Yun wrote: > Hi, > > On Fri, 2018-09-07 at 09:42 +0200, Greg Kroah-Hartman wrote: > > On Fri, Sep 07, 2018 at 03:29:12PM +0800, Chunfeng Yun wrote: > > > The MTK xHCI controller use some reserved bytes in endpoint context for > > > bandwidth scheduling, so need keep them in xhci_endpoint_copy(); > > > > If they are "reserved" shouldn't they be properly named? And by using > > reserved bytes, isn't that a spec violation? > It indeed violates the spec, "they shall be treated by system software > as Reserved and Opaque", and it's a quirk of the MTK xHCI controller. So as the "system software" here, we should just ignore them otherwise we violate the spec? :) Anyway, that's fine, no objection from me for the patch, thanks. greg k-h