From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sergei Shtylyov Subject: Re: [PATCH net-next 1/2] sh_eth: add sh_eth_cpu_data::no_xdfar flag Date: Sun, 1 Apr 2018 00:28:20 +0300 Message-ID: References: <998029b4-9400-ebc3-51c1-623c3f72b6cc@cogentembedded.com> <233e32d1-947b-2377-3d1f-39d5d83bd3f7@cogentembedded.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Cc: linux-renesas-soc@vger.kernel.org, linux-sh@vger.kernel.org To: netdev@vger.kernel.org Return-path: Received: from mail-lf0-f66.google.com ([209.85.215.66]:38097 "EHLO mail-lf0-f66.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751431AbeCaV2X (ORCPT ); Sat, 31 Mar 2018 17:28:23 -0400 Received: by mail-lf0-f66.google.com with SMTP id u3-v6so16330574lff.5 for ; Sat, 31 Mar 2018 14:28:23 -0700 (PDT) In-Reply-To: <233e32d1-947b-2377-3d1f-39d5d83bd3f7@cogentembedded.com> Content-Language: en-MW Sender: netdev-owner@vger.kernel.org List-ID: On 04/01/2018 12:22 AM, Sergei Shtylyov wrote: > The commit 6ded286555c2 ("sh_eth: Fix RX recovery on R-Car in case of RX > ring underrun") added a check for an bad RDFAR offset in sh_eth_rx(), so > that the code could work on the R-Car Ether controllers which don't have > this register (and TDFAR), then the commit 3365711df02 ("sh_eth: WARN on Oops, only 11 digits! Should have been 3365711df024. :-) > access to a register not implemented in a particular chip") replaced > offset 0 with 0xffff. Adding/checking the 'no_xdfar' bit field in the > 'struct sh_eth_cpu_data' instead results in less object code... > > Signed-off-by: Sergei Shtylyov [...] MBR, Sergei