From mboxrd@z Thu Jan 1 00:00:00 1970 From: Francois Romieu Subject: Re: [PATCH,RFC] Re: r8169 driver problem with RTL8110SB chip (on iop3xx ARM board) Date: Fri, 8 Sep 2006 22:38:31 +0200 Message-ID: <20060908203831.GB23545@electric-eye.fr.zoreil.com> References: <20060728101109.GA25854@xi.wantstofly.org> <20060818212852.GA5379@xi.wantstofly.org> <20060820213558.GA13188@electric-eye.fr.zoreil.com> <20060904222715.GF3782@xi.wantstofly.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: netdev@vger.kernel.org, tbm@cyrius.com Return-path: Received: from electric-eye.fr.zoreil.com ([213.41.134.224]:17058 "EHLO fr.zoreil.com") by vger.kernel.org with ESMTP id S1751215AbWIHUkg (ORCPT ); Fri, 8 Sep 2006 16:40:36 -0400 To: Lennert Buytenhek Content-Disposition: inline In-Reply-To: <20060904222715.GF3782@xi.wantstofly.org> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Lennert Buytenhek : [...] > I tried your series from step (1) plus my TxDesc change (so I didn't > include the hunk from (2)), and that seems to work fine. I.e. I don't > need to disable error interrupts anymore to keep it working. > > So really the only thing that would need addressing would be the TXDesc > MMIO bug (either by using I/O accesses, or doing the top/bottom writes > the other way round) and we can dump the vendor driver. I'll turn the top/bottom write the other way around for it looks like the less intrusive option so far. The current r8169 serie stood long enough in -mm that I can ask Jeff to pull it. The link management of the 8136 needs more work but it is not a showstopper. -- Ueimor