From: sagi@lightbits.io (Sagi Grimberg)
Subject: [PATCH 0/2] Get rid of transport layer retry count config parameter
Date: Wed, 22 Jun 2016 19:31:59 +0300 [thread overview]
Message-ID: <576ABD7F.1030204@grimberg.me> (raw)
In-Reply-To: <20160622161559.GA18361@infradead.org>
>> This parameter was added in order to support a proper timeout for
>> error recovery before the spec defined a periodic keep-alive.
>>
>> Now that we have periodic keep-alive, we don't need a user configurable
>> transport layer retry count, the keep-alive timeout is sufficient,
>> transports can retry for as long as they see fit.
>
> Isn't there some IB protocol level rationale for a low retry count
> in various fabric setups?
None that I know of... The QP retry count determines the time it would
take to fail a send/read/write.. The retry_count value is multiplied
with the packet timeout (which is a result of an IB specific
computation - managed by the CM).
It's useful when one needs to limit the time until a send fails in order
to kick error recovery (useful for srp which doesn't implement periodic
keep-alive), but since nvme does, I don't see the reason why RDMA or any
other transport should expose this configuration as the keep-alive
timeout exists for that.
WARNING: multiple messages have this Message-ID (diff)
From: Sagi Grimberg <sagi-ImC7XgPzLAfvYQKSrp0J2Q@public.gmane.org>
To: Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
Cc: linux-nvme-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH 0/2] Get rid of transport layer retry count config parameter
Date: Wed, 22 Jun 2016 19:31:59 +0300 [thread overview]
Message-ID: <576ABD7F.1030204@grimberg.me> (raw)
In-Reply-To: <20160622161559.GA18361-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
>> This parameter was added in order to support a proper timeout for
>> error recovery before the spec defined a periodic keep-alive.
>>
>> Now that we have periodic keep-alive, we don't need a user configurable
>> transport layer retry count, the keep-alive timeout is sufficient,
>> transports can retry for as long as they see fit.
>
> Isn't there some IB protocol level rationale for a low retry count
> in various fabric setups?
None that I know of... The QP retry count determines the time it would
take to fail a send/read/write.. The retry_count value is multiplied
with the packet timeout (which is a result of an IB specific
computation - managed by the CM).
It's useful when one needs to limit the time until a send fails in order
to kick error recovery (useful for srp which doesn't implement periodic
keep-alive), but since nvme does, I don't see the reason why RDMA or any
other transport should expose this configuration as the keep-alive
timeout exists for that.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2016-06-22 16:31 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-22 12:05 [PATCH 0/2] Get rid of transport layer retry count config parameter Sagi Grimberg
2016-06-22 12:06 ` [PATCH 1/2] nvme-rdma: Don't use tl_retry_count Sagi Grimberg
2016-06-22 12:06 ` [PATCH 2/2] nvme-fabrics: Remove tl_retry_count Sagi Grimberg
2016-06-22 16:15 ` [PATCH 0/2] Get rid of transport layer retry count config parameter Christoph Hellwig
2016-06-22 16:15 ` Christoph Hellwig
2016-06-22 16:31 ` Sagi Grimberg [this message]
2016-06-22 16:31 ` Sagi Grimberg
2016-06-22 20:31 ` Jason Gunthorpe
2016-06-22 20:31 ` Jason Gunthorpe
2016-06-23 7:09 ` Sagi Grimberg
2016-06-23 7:09 ` Sagi Grimberg
2016-06-24 7:13 ` Christoph Hellwig
2016-06-24 7:13 ` Christoph Hellwig
2016-06-26 15:48 ` Sagi Grimberg
2016-06-26 15:48 ` Sagi Grimberg
2016-07-17 11:52 ` Sagi Grimberg
2016-07-17 11:52 ` Sagi Grimberg
2016-07-18 4:09 ` Christoph Hellwig
2016-07-18 4:09 ` Christoph Hellwig
2016-07-18 4:09 ` Christoph Hellwig
2016-07-18 4:09 ` Christoph Hellwig
2016-07-18 8:01 ` Sagi Grimberg
2016-07-18 8:01 ` Sagi Grimberg
2016-07-18 15:20 ` Bart Van Assche
2016-07-18 15:20 ` Bart Van Assche
2016-07-20 8:42 ` Sagi Grimberg
2016-07-20 8:42 ` Sagi Grimberg
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=576ABD7F.1030204@grimberg.me \
--to=sagi@lightbits.io \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.