All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ewan Milne <emilne@redhat.com>
To: Hannes Reinecke <hare@suse.de>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>,
	Christoph Hellwig <hch@lst.de>,
	Bart van Assche <bart.vanassche@sandisk.com>,
	James Bottomley <james.bottomley@hansenpartnership.com>,
	linux-scsi@vger.kernel.org
Subject: Re: [PATCH 18/23] scsi_dh_alua: Send TEST UNIT READY to poll for transitioning
Date: Thu, 11 Feb 2016 15:25:18 -0500	[thread overview]
Message-ID: <1455222318.22112.330.camel@localhost.localdomain> (raw)
In-Reply-To: <1454942086-128704-19-git-send-email-hare@suse.de>

On Mon, 2016-02-08 at 15:34 +0100, Hannes Reinecke wrote:
> Sending a 'REPORT TARGET PORT GROUP' command is a costly operation,
> as the array has to gather information about all ports.
> So instead of using RTPG to poll for a status update when a port
> is in transitioning we should be sending a TEST UNIT READY, and
> wait for the sense code to report success.

Note that we may need to add a timeout on this somehow, I have
recently seen a bug report where an array stayed in the ALUA
transitioning state for an extremely long period of time.

That problem would occur with either the current or this new
ALUA code, the question is whether we want to handle it better.

-Ewan

> 
> Signed-off-by: Hannes Reinecke <hare@suse.de>
> Reviewed-by: Ewan Milne <emilne@redhat.com>
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> ---
>  drivers/scsi/device_handler/scsi_dh_alua.c | 38 ++++++++++++++++++++++++++++++
>  1 file changed, 38 insertions(+)
> 
> diff --git a/drivers/scsi/device_handler/scsi_dh_alua.c b/drivers/scsi/device_handler/scsi_dh_alua.c
> index de8e79e..a1db82f 100644
> --- a/drivers/scsi/device_handler/scsi_dh_alua.c
> +++ b/drivers/scsi/device_handler/scsi_dh_alua.c
> @@ -466,6 +466,30 @@ static int alua_check_sense(struct scsi_device *sdev,
>  }
>  
>  /*
> + * alua_tur - Send a TEST UNIT READY
> + * @sdev: device to which the TEST UNIT READY command should be send
> + *
> + * Send a TEST UNIT READY to @sdev to figure out the device state
> + * Returns SCSI_DH_RETRY if the sense code is NOT READY/ALUA TRANSITIONING,
> + * SCSI_DH_OK if no error occurred, and SCSI_DH_IO otherwise.
> + */
> +static int alua_tur(struct scsi_device *sdev)
> +{
> +	struct scsi_sense_hdr sense_hdr;
> +	int retval;
> +
> +	retval = scsi_test_unit_ready(sdev, ALUA_FAILOVER_TIMEOUT * HZ,
> +				      ALUA_FAILOVER_RETRIES, &sense_hdr);
> +	if (sense_hdr.sense_key == NOT_READY &&
> +	    sense_hdr.asc == 0x04 && sense_hdr.ascq == 0x0a)
> +		return SCSI_DH_RETRY;
> +	else if (retval)
> +		return SCSI_DH_IO;
> +	else
> +		return SCSI_DH_OK;
> +}
> +
> +/*
>   * alua_rtpg - Evaluate REPORT TARGET GROUP STATES
>   * @sdev: the device to be evaluated.
>   *
> @@ -735,8 +759,22 @@ static void alua_rtpg_work(struct work_struct *work)
>  		alua_wq = kaluad_sync_wq;
>  	pg->flags |= ALUA_PG_RUNNING;
>  	if (pg->flags & ALUA_PG_RUN_RTPG) {
> +		int state = pg->state;
> +
>  		pg->flags &= ~ALUA_PG_RUN_RTPG;
>  		spin_unlock_irqrestore(&pg->lock, flags);
> +		if (state == TPGS_STATE_TRANSITIONING) {
> +			if (alua_tur(sdev) == SCSI_DH_RETRY) {
> +				spin_lock_irqsave(&pg->lock, flags);
> +				pg->flags &= ~ALUA_PG_RUNNING;
> +				pg->flags |= ALUA_PG_RUN_RTPG;
> +				spin_unlock_irqrestore(&pg->lock, flags);
> +				queue_delayed_work(alua_wq, &pg->rtpg_work,
> +						   pg->interval * HZ);
> +				return;
> +			}
> +			/* Send RTPG on failure or if TUR indicates SUCCESS */
> +		}
>  		err = alua_rtpg(sdev, pg);
>  		spin_lock_irqsave(&pg->lock, flags);
>  		if (err == SCSI_DH_RETRY || pg->flags & ALUA_PG_RUN_RTPG) {



  reply	other threads:[~2016-02-11 20:25 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-08 14:34 [PATCH 00/23] ALUA device handler update, part II Hannes Reinecke
2016-02-08 14:34 ` [PATCH 01/23] scsi_dh_alua: Pass buffer as function argument Hannes Reinecke
2016-02-08 14:34 ` [PATCH 02/23] scsi_dh_alua: separate out alua_stpg() Hannes Reinecke
2016-02-11 23:54   ` Bart Van Assche
2016-02-08 14:34 ` [PATCH 03/23] scsi_dh_alua: Make stpg synchronous Hannes Reinecke
2016-02-08 14:34 ` [PATCH 04/23] scsi_dh_alua: call alua_rtpg() if stpg fails Hannes Reinecke
2016-02-08 14:34 ` [PATCH 05/23] scsi_dh_alua: switch to scsi_execute_req_flags() Hannes Reinecke
2016-02-08 14:34 ` [PATCH 06/23] scsi_dh_alua: allocate RTPG buffer separately Hannes Reinecke
2016-02-08 14:34 ` [PATCH 07/23] scsi_dh_alua: Use separate alua_port_group structure Hannes Reinecke
2016-02-08 14:34 ` [PATCH 08/23] scsi_dh_alua: use unique device id Hannes Reinecke
2016-02-08 14:34 ` [PATCH 09/23] scsi_dh_alua: simplify alua_initialize() Hannes Reinecke
2016-02-08 14:34 ` [PATCH 10/23] revert commit a8e5a2d593cb ("[SCSI] scsi_dh_alua: ALUA handler attach should succeed while TPG is transitioning") Hannes Reinecke
2016-02-08 14:34 ` [PATCH 11/23] scsi_dh_alua: move optimize_stpg evaluation Hannes Reinecke
2016-02-08 14:34 ` [PATCH 12/23] scsi_dh_alua: remove 'rel_port' from alua_dh_data structure Hannes Reinecke
2016-02-08 14:34 ` [PATCH 13/23] scsi_dh_alua: Use workqueue for RTPG Hannes Reinecke
2016-02-09 12:54   ` Christoph Hellwig
2016-02-08 14:34 ` [PATCH 14/23] scsi_dh_alua: Allow workqueue to run synchronously Hannes Reinecke
2016-02-08 14:34 ` [PATCH 15/23] scsi_dh_alua: Add new blacklist flag 'BLIST_SYNC_ALUA' Hannes Reinecke
2016-02-08 14:34 ` [PATCH 16/23] scsi_dh_alua: Recheck state on unit attention Hannes Reinecke
2016-02-08 14:34 ` [PATCH 17/23] scsi_dh_alua: update all port states Hannes Reinecke
2016-02-08 14:34 ` [PATCH 18/23] scsi_dh_alua: Send TEST UNIT READY to poll for transitioning Hannes Reinecke
2016-02-11 20:25   ` Ewan Milne [this message]
2016-02-12  7:12     ` Hannes Reinecke
2016-02-08 14:34 ` [PATCH 19/23] scsi_dh: add 'rescan' callback Hannes Reinecke
2016-02-11 20:27   ` Ewan Milne
2016-02-12  7:14     ` Hannes Reinecke
2016-02-08 14:34 ` [PATCH 20/23] scsi: Add 'access_state' attribute Hannes Reinecke
2016-02-11 20:33   ` Ewan Milne
2016-02-12  7:16     ` Hannes Reinecke
2016-02-08 14:34 ` [PATCH 21/23] scsi_dh_alua: use common definitions for ALUA state Hannes Reinecke
2016-02-08 14:34 ` [PATCH 22/23] scsi_dh_alua: update 'access_state' field Hannes Reinecke
2016-02-09 12:56   ` Christoph Hellwig
2016-02-11 20:37   ` Ewan Milne
2016-02-12  7:20     ` Hannes Reinecke
2016-02-08 14:34 ` [PATCH 23/23] scsi_dh_alua: Update version to 2.0 Hannes Reinecke
2016-02-08 14:37 ` [PATCH 00/23] ALUA device handler update, part II Hannes Reinecke
2016-02-08 19:39   ` Bart Van Assche
2016-02-11  1:37     ` Martin K. Petersen
2016-02-11 17:41       ` Bart Van Assche
2016-02-11 21:24 ` Bart Van Assche

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=1455222318.22112.330.camel@localhost.localdomain \
    --to=emilne@redhat.com \
    --cc=bart.vanassche@sandisk.com \
    --cc=hare@suse.de \
    --cc=hch@lst.de \
    --cc=james.bottomley@hansenpartnership.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    /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.