Re: nbd locking problems

From: Soren Hansen
Date: Fri Jan 28 2011 - 18:12:11 EST


2011/1/28 Serge E. Hallyn <serge@xxxxxxxxxx>:
> Thanks for looking into this, Soren.

Sure :)

> As Dave just explained to me, BKL is released when you sleep :)

Ah, yes, that makes perfect sense.

>ÂI assume that's how it gets released around the kthread_create(). ÂI
> *think* you're right that the new mutex is superfluous, but I'd like to
> look through the ioctl code and make sure there's no shared state which
> we need to protect.

Yup, I've tried that and I didn't find anything, but this is very new
territory to me, so more eyes sounds great.

> I don't see how there could be, or rather, if there
> were, then it was broken before.

Good point.

> In the meantime, could you please post your patch? Â(Is it the
> three-liner I'd expect?)

Indeed it is. I was holding off on posting it, because I was hoping
Pavel Machek would chime in. Pavel's commit message in [1] suggested
that there was more work to do before the BKL could be vanquished from
this driver (by changing .ioctl to .unlocked_ioctl).

I guess it can't hurt to post it, though. I'll do that in a second.

[1]: 1a2ad21128bb4eb79f3c05e5801edcc5ed3ef1d3

--
Soren Hansen    Â| http://linux2go.dk/
Ubuntu Developer  Â| http://www.ubuntu.com/
OpenStack Developer | http://www.openstack.org/
--
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/