From mboxrd@z Thu Jan  1 00:00:00 1970
Message-ID: <2c37a72c526cc8f79c2b35be1c96ec3d@proxima.alt.za>
To: 9fans@9fans.net
From: Lucio De Re <lucio@proxima.alt.za>
Date: Thu, 18 Nov 2010 20:01:20 +0200
In-Reply-To: <ddd5f2dd81a6a03744b4df7806ba55ac@ladd.quanstro.net>
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Subject: Re: [9fans] That deadlock, again
Topicbox-Message-UUID: 851e97cc-ead6-11e9-9d60-3106f5b1d025

> and i'm just wrong.  intentionally or not, devsd does
> qlock things with no up from sdreset().  ether82598
> does too (my fault).

I suggest you fix ether82598: it is OK to call qlock() and qunlock()
without "up", but only if sure that the qlock() will succeed.  If it
has to wait, it will panic.  Given that, why do the locking at all?

++L