netdev - Re: [Bugme-new] [Bug 12876] New: irq 18: nobody cared after down-ing an e1000 interface

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Fri, 20 Mar 2009 22:05:30 -0700 (Pacific Daylight Time)
From:	"Brandeburg, Jesse" <jesse.brandeburg@...el.com>
To:	Andrew Morton <akpm@...ux-foundation.org>
cc:	"e1000-devel@...ts.sourceforge.net" 
	<e1000-devel@...ts.sourceforge.net>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	"bugme-daemon@...zilla.kernel.org" <bugme-daemon@...zilla.kernel.org>,
	"lkundrak@...sk" <lkundrak@...sk>
Subject: Re: [Bugme-new] [Bug 12876] New: irq 18: nobody cared after down-ing
 an e1000 interface

On Sun, 15 Mar 2009, Andrew Morton wrote:
> > http://bugzilla.kernel.org/show_bug.cgi?id=12876
> > I get the following OOPS when I down the e1000 network interface:
> > 
> > irq 18: nobody cared (try booting with the "irqpoll" option)
> > Pid: 9014, comm: udevd Tainted: G        W  2.6.29-0.237.rc7.git4.fc11.i586 #1
> > Call Trace:
> >  [<c0471f2b>] __report_bad_irq+0x33/0x74
> >  [<c047205e>] note_interrupt+0xf2/0x150
> >  [<c04711e0>] ? handle_IRQ_event+0x4f/0x58
> >  [<c04725e8>] handle_fasteoi_irq+0x94/0xb7
> >  [<c0472554>] ? handle_fasteoi_irq+0x0/0xb7
> >  <IRQ>  [<c04045ac>] ? common_interrupt+0x2c/0x40
> > handlers:
> > [<c05f3cb6>] (ata_sff_interrupt+0x0/0xb1)
> > [<c06079ec>] (usb_hcd_irq+0x0/0xa3)
> > Disabling IRQ #18
> > 
> > As you can see, the IRQ 18 is shared with ata_sff and usb_hcd.
> > 
> > I've tried to comment out e1000_free_irq() call from e1000_close() and added a
> > printk() to e1000_intr(). After that, the problem didn't occur, but I've not
> > seen the output of the printk in e1000_intr() (It was visible while the
> > interface was up), as if it weren't called. Seems like the way shared
> > interrupts work is beyond my understanding.

Please try the patch attached to the bug, Alan Cox pointed out that it was 
likely the driver was racy during close and was re-enabling interrupts.

I spotted two possible areas where this was possible and put in checks for 
__E1000_DOWN before re-enabling ints in those spots.

I'm offline for a week and I'll reconnect and check back when I return.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html