From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vitaly Bordug Subject: Re: STRANGE ERROR Date: Sun, 20 May 2007 20:51:24 +0400 Message-ID: <20070520165124.GA12025@server.dev.rtsoft.ru> References: <20070519163402.b087b9d7.akpm@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset=koi8-r Cc: Sasa Ostrouska , linux-kernel@vger.kernel.org, Jeff Garzik , netdev@vger.kernel.org To: Andrew Morton Return-path: Received: from rtsoft2.corbina.net ([85.21.88.2]:56797 "HELO mail.dev.rtsoft.ru" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with SMTP id S1758669AbXETQv3 (ORCPT ); Sun, 20 May 2007 12:51:29 -0400 Content-Disposition: inline In-Reply-To: <20070519163402.b087b9d7.akpm@linux-foundation.org> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Sat, May 19, 2007 at 16:34 -0700, Andrew Morton wrote: > On Sun, 20 May 2007 00:30:55 +0200 "Sasa Ostrouska" wrote: > > > Hi everybody, > > > > I tried today to upgrade the kernel to 2.6.21.1 and i got the same > > error during the boot time. > > Here is the dmesg of the 2.6.20.2, can somebody tell me what this is ? > > > > ... > > > > Marvell 88E1101: Registered new driver > > Fixed PHY: Registered new driver > > driver_bound: device fixed@100:1 already bound > > I don't know what caused that one. > this is because of issue in fixed phy driver initialisation - have a patch but need to test it a bit more. > > Device 'fixed@100:1' does not have a release() function, it is broken > > and must be fixed. > > BUG: at drivers/base/core.c:104 device_release() > > > > Call Trace: > > [] kobject_cleanup+0x53/0x7e > > [] kobject_release+0x0/0x9 > > [] kref_put+0x74/0x81 > > [] fixed_mdio_register_device+0x230/0x265 > > [] fixed_init+0x1f/0x35 > > [] init+0x147/0x2fb > > [] schedule_tail+0x36/0x92 > > [] child_rip+0xa/0x12 > > [] acpi_ds_init_one_object+0x0/0x83 > > [] init+0x0/0x2fb > > [] child_rip+0x0/0x12 > > This appears to have happened because fixed_mdio_register_device() (or > phy_device_create) didn't suitably initialise phy_device.dev. > > But I don't immediately see why this doesn't affect all phy drivers. > Presumably it's the fixed driver which is at fault. Jeff, how is this > supposed to work? > the fixed phy used to have "specific" bus bound stuff but I've reworked this point.