From mboxrd@z Thu Jan 1 00:00:00 1970 From: Xose Vazquez Perez Subject: Re: [PATCH 2/2] r8169: allow true forced mode setting Date: Sat, 16 May 2009 21:57:10 +0200 Message-ID: <4A0F1A96.8070208@gmail.com> References: <20090514214605.GB6374@electric-eye.fr.zoreil.com> <200905161450.38281.jdelvare@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Francois Romieu , "David S. Miller" , Vincent Steenhoute , Martin Michlmayr , Ivan Vecera , Edward Hsu , Andrew Morton , netdev@vger.kernel.org To: Jean Delvare Return-path: Received: from mail-ew0-f176.google.com ([209.85.219.176]:65027 "EHLO mail-ew0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752860AbZEPT5N (ORCPT ); Sat, 16 May 2009 15:57:13 -0400 Received: by ewy24 with SMTP id 24so3223910ewy.37 for ; Sat, 16 May 2009 12:57:13 -0700 (PDT) In-Reply-To: <200905161450.38281.jdelvare@suse.de> Sender: netdev-owner@vger.kernel.org List-ID: zOn 05/16/2009 02:50 PM, Jean Delvare wrote: > If I read the code correctly, Realtek's driver (v6.010.00) only > applies the quirk for RTL8169S and RTL8110S chips (what I have) and > not the original RTL8169. Your code applies the quirk to all 3 > chips. Is this on purpose? Or am I missing something? For reference, there it's the latest source code of Realtek drivers r8101/r8168/r8169 : ftp://WebUser:DAx7h9V@202.65.194.212/cn/nic/r8101-1.012.00.tar.bz2 ftp://WebUser:DAx7h9V@202.65.194.212/cn/nic/r8168-8.012.00.tar.bz2 ftp://WebUser:DAx7h9V@202.65.194.212/cn/nic/r8169-6.010.00.tar.bz2 >> (...) >> + if (tp->mac_version <= RTL_GIGA_MAC_VER_03) { >> + if ((speed == SPEED_100) && (autoneg != AUTONEG_ENABLE)) { >> + mdio_write(ioaddr, 0x17, 0x2138); >> + mdio_write(ioaddr, 0x0e, 0x0260); >> + } else { >> + mdio_write(ioaddr, 0x17, 0x2108); >> + mdio_write(ioaddr, 0x0e, 0x0000); >> + } >> + } >> + >> return 0; >> } -thanks- regards, -- Polycommander, Erkowit, Urquiola, Andros Patria, Cason, Aegean Sea, Prestige, ...