From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yan Seiner Subject: Re: Processes dying with ENXIO or EINTR - disk related? Date: Tue, 10 Feb 2009 17:41:22 -0800 Message-ID: <49922CC2.8020705@seiner.com> References: <499229DF.50708@kernel.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from flatoutfitness.com ([66.178.130.209]:40711 "EHLO www.seiner.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750920AbZBKBlX (ORCPT ); Tue, 10 Feb 2009 20:41:23 -0500 In-Reply-To: <499229DF.50708@kernel.org> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Tejun Heo Cc: linux-ide@vger.kernel.org Tejun Heo wrote: > Yan Seiner wrote: > >> Recently I've started getting system freezeups and weird segfaults with >> EINTR (Interrupted system call) and ENXIO (No such device or address). >> >> These errors can be in ld, libc, and so on. I'm stumped; the system has >> worked fine for quite a while and just recently has developed this >> behavior. I can start yanking out hardware, but before I do that I wonder >> if it is something to do with disk access. I'm a bit stumped as to what >> else could cause ENXIO on loading an app.... Needless to say there aren't >> any errors at all anywhere that I can find, so I am guessing. >> > > Highly unlikely it's related to IO. > I was experimenting with nouveau, an xorg driver for nvidia hardware. Turns out that driver left the scsi hardware in a weird state that was causing the segfaults. A warm reboot wouldn't cure it; only a total power down and wait 30 seconds cycle did. Machine stable since then. At a guess something in the nouveau driver screwed up the scsi controller in the machine.