Re: [PATCH] USB: fix deadlock in HCD code

[Alan Stern]

On Wed, 21 May 2008, Jiri Kosina wrote:

> On Wed, 21 May 2008, Alan Stern wrote:
> 
> > > Used in the suspend code. Your patch is indeed correct, but I fear 
> > > there might be a second bug caused by wrong calling conditions.
> > The problem in the Novell bugzilla entry was caused by the fact that the 
> > OHCI irq routine was invoked with interrupts enabled, owing to a missing 
> > IRQF_DISABLED flag.  That bug has already been fixed in 2.6.25.
> 
> That indeed is 2.6.25 kernel. I guess you are talking about commit 
> 442258e2ff69 here.

Yes.

> If so, the reporter is definitely using the kernel 
> containing this commit, and the lockups still trigger.
> 
> Seems that my patch is papering over the real bug (someone enabling 
> interrupts somewhere) indeed, but I can't seem to be able to find such 
> codepath.

You could try testing the interrupt-enable flag at various places 
in ohci-hcd (start with finish_urb) and printing an error message if 
interrupts are enabled.

One possibility is that in an earlier call to finish_urb,
usb_hcd_giveback_urb was called with interrupts disabled and returned 
with interrupts enabled.  In other words, some driver's callback 
routine may have enabled interrupts incorrectly.

Alan Stern

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/