All of lore.kernel.org
 help / color / mirror / Atom feed
From: Hannes Reinecke <hare@suse.de>
To: Christoph Hellwig <hch@lst.de>
Cc: James Bottomley <jbottomley@parallels.com>, linux-scsi@vger.kernel.org
Subject: Re: [PATCH 14/17] scsi_dh_alua: Use workqueue for RTPG
Date: Mon, 11 May 2015 15:59:16 +0200	[thread overview]
Message-ID: <5550B5B4.4070605@suse.de> (raw)
In-Reply-To: <20150511134914.GA7795@lst.de>

On 05/11/2015 03:49 PM, Christoph Hellwig wrote:
> On Mon, May 04, 2015 at 02:42:20PM +0200, Hannes Reinecke wrote:
>> The current ALUA device_handler has two drawbacks:
>> - We're sending a 'SET TARGET PORT GROUP' command to every LUN,
>>   disregarding the fact that several LUNs might be in a port group
>>   and will be automatically switched whenever _any_ LUN within
>>   that port group receives the command.
>> - Whenever a LUN is in 'transitioning' mode we cannot block I/O
>>   to that LUN, instead the controller has to abort the command.
>>   This leads to increased traffic across the wire and heavy load
>>   on the controller during switchover.
>>
>> With this patch the RTPG handling is moved to a workqueue, which
>> is being run once per port group. This reduces the number of
>> 'REPORT TARGET PORT GROUP' and 'SET TARGET PORT GROUPS' which
>> will be send to the controller. It also allows us to block
>> I/O to any LUN / port group found to be in 'transitioning' ALUA
>> mode, as the workqueue item will be requeued until the controller
>> moves out of transitioning.
> 
> I'm having a hard time understanding the workqueue use here.
> What is the benefit of that one worker function to do everything?
> It seems having a work struct in struct alua_queue_data to just
> run STPG, and a different one to run RPTG in the port group structure
> would be more sensible instead of interwinding them.
> 
Hmm. It _might_ be possible. However, we need to serialize
STPG and RTPG against each other; not sure if that's still possible
when using different workqueue items.

> Also why do you need the sigle threaded workqueue?  That seems
> like a possible limiting factor in a large enough system having
> to deal with a cable disconnect cutting off multiple port groups,
> or just during bootup.
> 
Okay, will be switching to a normal workqueue.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		               zSeries & Storage
hare@suse.de			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

  reply	other threads:[~2015-05-11 13:59 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-04 12:42 [PATCH 00/17] asynchronous ALUA device handler Hannes Reinecke
2015-05-04 12:42 ` [PATCH 01/17] scsi_dh: return individual errors in scsi_dh_activate() Hannes Reinecke
2015-05-07 11:34   ` Bart Van Assche
2015-05-11  6:34   ` Christoph Hellwig
2015-05-04 12:42 ` [PATCH 02/17] scsi_dh_alua: Disable ALUA handling for non-disk devices Hannes Reinecke
2015-05-07 11:34   ` Bart Van Assche
2015-05-11  6:46   ` Christoph Hellwig
2015-05-11 10:25     ` Hannes Reinecke
2015-05-11 11:34       ` Christoph Hellwig
2015-05-11 11:55         ` Hannes Reinecke
2015-05-11 12:19           ` Christoph Hellwig
2015-05-04 12:42 ` [PATCH 03/17] scsi_dh_alua: Use vpd_pg83 information Hannes Reinecke
2015-05-07 11:41   ` Bart Van Assche
2015-05-07 11:50     ` Hannes Reinecke
2015-05-11  6:48   ` Christoph Hellwig
2015-05-11 10:11     ` Hannes Reinecke
2015-05-04 12:42 ` [PATCH 04/17] scsi_dh_alua: Improve error handling Hannes Reinecke
2015-05-07 11:48   ` Bart Van Assche
2015-05-07 11:52     ` Hannes Reinecke
2015-05-11 13:19       ` Hannes Reinecke
2015-05-04 12:42 ` [PATCH 05/17] scsi: remove scsi_show_sense_hdr() Hannes Reinecke
2015-05-07 11:49   ` Bart Van Assche
2015-05-11  6:49   ` Christoph Hellwig
2015-05-04 12:42 ` [PATCH 06/17] scsi_dh_alua: use flag for RTPG extended header Hannes Reinecke
2015-05-07 11:52   ` Bart Van Assche
2015-05-11  6:50   ` Christoph Hellwig
2015-05-04 12:42 ` [PATCH 07/17] scsi_dh_alua: Pass buffer as function argument Hannes Reinecke
2015-05-07 11:57   ` Bart Van Assche
2015-05-11  6:51   ` Christoph Hellwig
2015-05-04 12:42 ` [PATCH 08/17] scsi_dh_alua: Make stpg synchronous Hannes Reinecke
2015-05-07 12:18   ` Bart Van Assche
2015-05-07 13:36     ` Hannes Reinecke
2015-05-11  6:55   ` Christoph Hellwig
2015-05-11  9:59     ` Hannes Reinecke
2015-05-11 13:50       ` Christoph Hellwig
2015-05-11 13:59         ` Hannes Reinecke
2015-05-04 12:42 ` [PATCH 09/17] scsi_dh_alua: switch to scsi_execute() Hannes Reinecke
2015-05-06  9:26   ` Christoph Hellwig
2015-05-06  9:58     ` Hannes Reinecke
2015-05-04 12:42 ` [PATCH 10/17] scsi_dh_alua: Use separate alua_port_group structure Hannes Reinecke
2015-05-07 12:34   ` Bart Van Assche
2015-05-07 13:36     ` Bart Van Assche
2015-05-07 13:46       ` Hannes Reinecke
2015-05-07 13:37     ` Hannes Reinecke
2015-05-11 12:32   ` Christoph Hellwig
2015-05-11 12:36     ` Hannes Reinecke
2015-05-04 12:42 ` [PATCH 11/17] scsi_dh_alua: simplify sense code handling Hannes Reinecke
2015-05-11  6:58   ` Christoph Hellwig
2015-05-11 14:52     ` Hannes Reinecke
2015-05-12  8:20       ` Christoph Hellwig
2015-05-04 12:42 ` [PATCH 12/17] scsi_dh_alua: parse target device id Hannes Reinecke
2015-05-04 12:42 ` [PATCH 13/17] scsi_dh_alua: revert commit a8e5a2d593cbfccf530c3382c2c328d2edaa7b66 Hannes Reinecke
2015-05-11  7:00   ` Christoph Hellwig
2015-05-11 10:00     ` Hannes Reinecke
2015-05-04 12:42 ` [PATCH 14/17] scsi_dh_alua: Use workqueue for RTPG Hannes Reinecke
2015-05-11 13:49   ` Christoph Hellwig
2015-05-11 13:59     ` Hannes Reinecke [this message]
2015-05-12  8:16       ` Christoph Hellwig
2015-05-13  9:10         ` Hannes Reinecke
2015-05-04 12:42 ` [PATCH 15/17] scsi_dh_alua: Recheck state on unit attention Hannes Reinecke
2015-05-04 12:42 ` [PATCH 16/17] scsi_dh_alua: update all port states Hannes Reinecke
2015-05-04 12:42 ` [PATCH 17/17] scsi_dh_alua: Update version to 2.0 Hannes Reinecke

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=5550B5B4.4070605@suse.de \
    --to=hare@suse.de \
    --cc=hch@lst.de \
    --cc=jbottomley@parallels.com \
    --cc=linux-scsi@vger.kernel.org \
    /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.