From mboxrd@z Thu Jan  1 00:00:00 1970
From: hch@infradead.org ('Christoph Hellwig')
Date: Thu, 18 Aug 2016 08:21:07 -0700
Subject: nvme/rdma initiator stuck on reboot
In-Reply-To: <012701d1f958$b4953290$1dbf97b0$@opengridcomputing.com>
References: <043901d1f7f5$fb5f73c0$f21e5b40$@opengridcomputing.com>
 <2202d08c-2b4c-3bd9-6340-d630b8e2f8b5@grimberg.me>
 <073301d1f894$5ddb81d0$19928570$@opengridcomputing.com>
 <7c4827ff-21c9-21e9-5577-1bd374305a0b@grimberg.me>
 <075901d1f899$e5cc6f00$b1654d00$@opengridcomputing.com>
 <e2e04664-c374-3745-ecf3-f49ca7a3addf@grimberg.me>
 <012701d1f958$b4953290$1dbf97b0$@opengridcomputing.com>
Message-ID: <20160818152107.GA17807@infradead.org>

On Thu, Aug 18, 2016@08:59:15AM -0500, Steve Wise wrote:
> Sure, but let me ask a question:  So the bug was that the delete controller
> worker was blocked waiting for the reconnect worker to complete.

Btw, in that case the patch is not actually correct, as even workqueue
with a higher concurrency level MAY deadlock under enough memory
pressure.  We'll need separate workqueues to handle this case I think.

> Yes?  And the
> reconnect worker was never completing?  Why is that?  Here are a few tidbits
> about iWARP connections:  address resolution == neighbor discovery.  So if the
> neighbor is unreachable, it will take a few seconds for the OS to give up and
> fail the resolution.  If the neigh entry is valid and the peer becomes
> unreachable during connection setup, it might take 60 seconds or so for a
> connect operation to give up and fail.  So this is probably slowing the
> reconnect thread down.   But shouldn't the reconnect thread notice that a delete
> is trying to happen and bail out? 

I think we should aim for a state machine that can detect this, but
we'll have to see if that will end up in synchronization overkill.