[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <6811acef-143d-4104-a814-d3cb0811192b@grimberg.me>
Date: Thu, 7 Mar 2024 12:33:09 +0200
From: Sagi Grimberg <sagi@...mberg.me>
To: Daniel Wagner <dwagner@...e.de>, James Smart <james.smart@...adcom.com>
Cc: Keith Busch <kbusch@...nel.org>, Christoph Hellwig <hch@....de>,
Hannes Reinecke <hare@...e.de>, linux-nvme@...ts.infradead.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 5/5] nvme-fc: wait for initial connect attempt to
finish
On 21/02/2024 15:24, Daniel Wagner wrote:
> The TCP and RDMA transport are doing a synchronous connects, that is the
> syscal returns with the final result. The operation either fails or
> succeeds. The FC transport offloads the connect attempt to a workqueue
> and thus it's an asynchronous operation.
>
> This async connect feature was introduced to mitigate problems with
> transient connect errors and the task to coordinate retries with
> userspace (nvme-cli).
Maybe it is a better idea to let userspace handle this? the whole async
connect (and reconnect)
just seems it is causing more issues than solving... I may have missed
some conversations
about this one...
Powered by blists - more mailing lists