All of lore.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>,
	linux-scsi@vger.kernel.org,
	James Bottomley <James.Bottomley@HansenPartnership.com>,
	Robert Elliott <elliott@hp.com>,
	"Martin K. Petersen" <martin.petersen@oracle.com>,
	Bart van Assche <bvanassche@acm.org>, Jens Axboe <axboe@fb.com>,
	Kashyap Desai <kashyap.desai@avagotech.com>,
	Sreekanth Reddy <sreekanth.reddy@avagotech.com>,
	Mike Christie <michaelc@cs.wisc.edu>,
	Guennadi Liakhovetski <g.liakhovetski@gmx.de>,
	usb-storage@lists.one-eyed-alien.net
Subject: Re: [PATCH 09/11] scsi: don't set tagging state from scsi_adjust_queue_depth
Date: Tue, 4 Nov 2014 11:47:39 +0100	[thread overview]
Message-ID: <20141104104739.GE19154@lst.de> (raw)
In-Reply-To: <54589BE1.6070106@suse.de>

> > diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c
> > index a759cb2..596c2f4 100644
> > --- a/drivers/scsi/aacraid/linit.c
> > +++ b/drivers/scsi/aacraid/linit.c
> > @@ -462,9 +462,8 @@ static int aac_slave_configure(struct scsi_device *sdev)
> >  			depth = 256;
> >  		else if (depth < 2)
> >  			depth = 2;
> > -		scsi_adjust_queue_depth(sdev, MSG_ORDERED_TAG, depth);
> > -	} else
> > -		scsi_adjust_queue_depth(sdev, 0, 1);
> > +		scsi_adjust_queue_depth(sdev, depth);
> > +	}
> >  
> >  	return 0;
> >  }
> Why did you omit the 'else' branch?
> cmd_per_lun is set to 256 for aacraid AFAICS ...

Thanks, fixed.

> Hmm. This is actually wrong; it should be set to '1'.
> '2' is a left-over from the (long since removed) internal queueing
> within the aic7xxx driver.

Please send a separate patch for that.  Interestingly your patch that
sets it to 1 for aic79xx sais that's only temporarily.  If you still
have the hardware (either one should be fine) that driver could use
some major cleanup in how it deals with queuing and tags.

> > diff --git a/drivers/scsi/u14-34f.c b/drivers/scsi/u14-34f.c
> > index d8dcf36..aa0f403 100644
> > --- a/drivers/scsi/u14-34f.c
> > +++ b/drivers/scsi/u14-34f.c
> > @@ -696,25 +696,25 @@ static int u14_34f_slave_configure(struct scsi_device *dev) {
> >     if (TLDEV(dev->type) && dev->tagged_supported)
> >  
> >        if (tag_mode == TAG_SIMPLE) {
> > -         scsi_adjust_queue_depth(dev, MSG_SIMPLE_TAG, tqd);
> > +         scsi_adjust_queue_depth(dev, tqd);
> >           tag_suffix = ", simple tags";
> >           }
> >        else if (tag_mode == TAG_ORDERED) {
> > -         scsi_adjust_queue_depth(dev, MSG_ORDERED_TAG, tqd);
> > +         scsi_adjust_queue_depth(dev, tqd);
> >           tag_suffix = ", ordered tags";
> >           }
> >        else {
> > -         scsi_adjust_queue_depth(dev, 0, tqd);
> > +         scsi_adjust_queue_depth(dev, tqd);
> >           tag_suffix = ", no tags";
> >           }
> >  
> >     else if (TLDEV(dev->type) && linked_comm) {
> > -      scsi_adjust_queue_depth(dev, 0, tqd);
> > +      scsi_adjust_queue_depth(dev, tqd);
> >        tag_suffix = ", untagged";
> >        }
> >  
> >     else {
> > -      scsi_adjust_queue_depth(dev, 0, utqd);
> > +      scsi_adjust_queue_depth(dev, utqd);
> >        tag_suffix = "";
> >        }
> >  
> See my comment on the previous patches. As we're not using
> ordered tags that branch is never used.

The tag_mode is a module parameter.  I've kept it for compatibility
(or lazyness..), but there's no effect left, similar to the queue_type
sysfs file that still accept the "ordered" value.

  reply	other threads:[~2014-11-04 10:47 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-04  7:54 tag handling refactor Christoph Hellwig
2014-11-04  7:54 ` [PATCH 01/11] scsi: provide a generic change_queue_type method Christoph Hellwig
2014-11-04  8:26   ` Hannes Reinecke
2014-11-05  2:09   ` Martin K. Petersen
2014-11-06 15:28   ` Bart Van Assche
2014-11-06 15:35     ` Christoph Hellwig
2014-11-04  7:54 ` [PATCH 02/11] scsi: add new scsi-command flag for tagged commands Christoph Hellwig
2014-11-04  8:38   ` Hannes Reinecke
2014-11-04 10:35     ` Christoph Hellwig
2014-11-04 10:44       ` Hannes Reinecke
2014-11-05  2:12   ` Martin K. Petersen
2014-11-04  7:54 ` [PATCH 03/11] scsi: remove ordered_tags scsi_device field Christoph Hellwig
2014-11-05  2:14   ` Martin K. Petersen
2014-11-06 15:44   ` Bart Van Assche
2014-11-04  7:54 ` [PATCH 04/11] scsi: remove ordered_tag host template field Christoph Hellwig
2014-11-04  8:38   ` Hannes Reinecke
2014-11-05  2:15   ` Martin K. Petersen
2014-11-06 15:45   ` Bart Van Assche
2014-11-04  7:54 ` [PATCH 05/11] scsi: remove abuses of scsi_populate_tag Christoph Hellwig
2014-11-04  8:45   ` Hannes Reinecke
2014-11-04 10:37     ` Christoph Hellwig
2014-11-04  7:54 ` [PATCH 06/11] mptfusion: don't change queue type in ->change_queue_depth Christoph Hellwig
2014-11-04  8:46   ` Hannes Reinecke
2014-11-05  2:17   ` Martin K. Petersen
2014-11-04  7:54 ` [PATCH 07/11] scsi: remove use_blk_tcq Scsi_Host field Christoph Hellwig
2014-11-04  8:46   ` Hannes Reinecke
2014-11-05  2:18   ` Martin K. Petersen
2014-11-04  7:54 ` [PATCH 08/11] scsi: always assign block layer tags if enabled Christoph Hellwig
2014-11-04  9:01   ` Hannes Reinecke
2014-11-04 10:42     ` Christoph Hellwig
2014-11-04 10:46       ` Hannes Reinecke
2014-11-05  2:32   ` Martin K. Petersen
2014-11-05  2:34     ` Martin K. Petersen
2014-11-06 16:28   ` Bart Van Assche
2014-11-06 16:31     ` Christoph Hellwig
2014-11-04  7:54 ` [PATCH 09/11] scsi: don't set tagging state from scsi_adjust_queue_depth Christoph Hellwig
2014-11-04  9:26   ` Hannes Reinecke
2014-11-04 10:47     ` Christoph Hellwig [this message]
2014-11-05  2:50   ` Martin K. Petersen
2014-11-04  7:54 ` [PATCH 10/11] scsi: don't force tagged_supported in drivers Christoph Hellwig
2014-11-04  8:16   ` Jack Wang
2014-11-04 10:34     ` Christoph Hellwig
2014-11-04  7:54 ` [PATCH 11/11] ufs: remove spurious scsi_set_tag_type call Christoph Hellwig
2014-11-05 19:26 ` tag handling refactor Christoph Hellwig
  -- strict thread matches above, loose matches on Subject: below --
2014-11-10 15:56 tag handling refactor V2 Christoph Hellwig
2014-11-10 15:56 ` [PATCH 09/11] scsi: don't set tagging state from scsi_adjust_queue_depth Christoph Hellwig
2014-11-11 15:41   ` 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=20141104104739.GE19154@lst.de \
    --to=hch@lst.de \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=axboe@fb.com \
    --cc=bvanassche@acm.org \
    --cc=elliott@hp.com \
    --cc=g.liakhovetski@gmx.de \
    --cc=hare@suse.de \
    --cc=kashyap.desai@avagotech.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=michaelc@cs.wisc.edu \
    --cc=sreekanth.reddy@avagotech.com \
    --cc=usb-storage@lists.one-eyed-alien.net \
    /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.