2.6.24-rc4-mm1: hostbyte=0x01 driverbyte=0x00 (now bisected)

2.6.24-rc4-mm1: hostbyte=0x01 driverbyte=0x00 (now bisected)

[Alexey Dobriyan]

>  git-scsi-misc.patch

Apologies for not looking into the problem earlier. See
http://marc.info/?t=119628022300005&r=1&w=2
"2.6.24-rc3-mm2: Result: hostbyte=0x01 driverbyte=0x00\nend_request: I/O error"
for previous installment.

I've bisected it to the following patch in git-scsi-misc branch.
Revert on top of 2.6.24-rc4-mm1 also helps.

commit 8655a546c83fc43f0a73416bbd126d02de7ad6c0
Author: Hannes Reinecke <hare@suse.de>
Date:   Tue Nov 6 09:23:40 2007 +0100

    [SCSI] Do not requeue requests if REQ_FAILFAST is set
    
    Any requests with the REQ_FAILFAST flag set should not be requeued
    to the requeust queue, but rather terminated directly.
    Otherwise the multipath failover will stall until the command
    timeout triggers.
    
    Signed-off-by: Hannes Reinecke <hare@suse.de>
    Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 0f44bdb..0da0dd0 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1286,6 +1286,11 @@ int scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
 			 */
 			if (!(req->cmd_flags & REQ_PREEMPT))
 				ret = BLKPREP_DEFER;
+			/*
+			 * Return failfast requests immediately
+			 */
+			if (req->cmd_flags & REQ_FAILFAST)
+				ret = BLKPREP_KILL;
 			break;
 		default:
 			/*
@@ -1414,6 +1419,17 @@ static inline int scsi_host_queue_ready(struct request_queue *q,
 	return 1;
 }
 
+static void __scsi_kill_request(struct request *req)
+{
+	struct scsi_cmnd *cmd = req->special;
+	struct scsi_device *sdev = cmd->device;
+
+	cmd->result = DID_NO_CONNECT << 16;
+	atomic_inc(&cmd->device->iorequest_cnt);
+	sdev->device_busy--;
+	__scsi_done(cmd);
+}
+
 /*
  * Kill a request for a dead device
  */
@@ -1527,8 +1543,16 @@ static void scsi_request_fn(struct request_queue *q)
 		 * accept it.
 		 */
 		req = elv_next_request(q);
-		if (!req || !scsi_dev_queue_ready(q, sdev))
+		if (!req)
+			break;
+
+		if (!scsi_dev_queue_ready(q, sdev)) {
+			if (req->cmd_flags & REQ_FAILFAST) {
+				scsi_kill_request(req, q);
+				continue;
+			}
 			break;
+		}
 
 		if (unlikely(!scsi_device_online(sdev))) {
 			sdev_printk(KERN_ERR, sdev,
@@ -1609,8 +1633,12 @@ static void scsi_request_fn(struct request_queue *q)
 	 * later time.
 	 */
 	spin_lock_irq(q->queue_lock);
-	blk_requeue_request(q, req);
-	sdev->device_busy--;
+	if (unlikely(req->cmd_flags & REQ_FAILFAST))
+		__scsi_kill_request(req);
+	else {
+		blk_requeue_request(q, req);
+		sdev->device_busy--;
+	}
 	if(sdev->device_busy == 0)
 		blk_plug_device(q);
  out:

Re: 2.6.24-rc4-mm1: hostbyte=0x01 driverbyte=0x00 (now bisected)

[Hannes Reinecke]

[-- Attachment #1: Type: text/plain, Size: 3105 bytes --]

Alexey Dobriyan wrote:
>>  git-scsi-misc.patch
> 
> Apologies for not looking into the problem earlier. See
> http://marc.info/?t=119628022300005&r=1&w=2
> "2.6.24-rc3-mm2: Result: hostbyte=0x01 driverbyte=0x00\nend_request: I/O error"
> for previous installment.
> 
> I've bisected it to the following patch in git-scsi-misc branch.
> Revert on top of 2.6.24-rc4-mm1 also helps.
> 
> commit 8655a546c83fc43f0a73416bbd126d02de7ad6c0
> Author: Hannes Reinecke <hare@suse.de>
> Date:   Tue Nov 6 09:23:40 2007 +0100
> 
>     [SCSI] Do not requeue requests if REQ_FAILFAST is set
>     
>     Any requests with the REQ_FAILFAST flag set should not be requeued
>     to the requeust queue, but rather terminated directly.
>     Otherwise the multipath failover will stall until the command
>     timeout triggers.
>     
>     Signed-off-by: Hannes Reinecke <hare@suse.de>
>     Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
> 
> diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
> index 0f44bdb..0da0dd0 100644
> --- a/drivers/scsi/scsi_lib.c
> +++ b/drivers/scsi/scsi_lib.c
> @@ -1286,6 +1286,11 @@ int scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
>  			 */
>  			if (!(req->cmd_flags & REQ_PREEMPT))
>  				ret = BLKPREP_DEFER;
> +			/*
> +			 * Return failfast requests immediately
> +			 */
> +			if (req->cmd_flags & REQ_FAILFAST)
> +				ret = BLKPREP_KILL;
>  			break;
>  		default:
>  			/*
> @@ -1414,6 +1419,17 @@ static inline int scsi_host_queue_ready(struct request_queue *q,
>  	return 1;
>  }
>  
> +static void __scsi_kill_request(struct request *req)
> +{
> +	struct scsi_cmnd *cmd = req->special;
> +	struct scsi_device *sdev = cmd->device;
> +
> +	cmd->result = DID_NO_CONNECT << 16;
> +	atomic_inc(&cmd->device->iorequest_cnt);
> +	sdev->device_busy--;
> +	__scsi_done(cmd);
> +}
> +
>  /*
>   * Kill a request for a dead device
>   */
> @@ -1527,8 +1543,16 @@ static void scsi_request_fn(struct request_queue *q)
>  		 * accept it.
>  		 */
>  		req = elv_next_request(q);
> -		if (!req || !scsi_dev_queue_ready(q, sdev))
> +		if (!req)
> +			break;
> +
> +		if (!scsi_dev_queue_ready(q, sdev)) {
> +			if (req->cmd_flags & REQ_FAILFAST) {
> +				scsi_kill_request(req, q);
> +				continue;
> +			}
>  			break;
> +		}
>  
>  		if (unlikely(!scsi_device_online(sdev))) {
>  			sdev_printk(KERN_ERR, sdev,
> @@ -1609,8 +1633,12 @@ static void scsi_request_fn(struct request_queue *q)
>  	 * later time.
>  	 */
>  	spin_lock_irq(q->queue_lock);
> -	blk_requeue_request(q, req);
> -	sdev->device_busy--;
> +	if (unlikely(req->cmd_flags & REQ_FAILFAST))
> +		__scsi_kill_request(req);
> +	else {
> +		blk_requeue_request(q, req);
> +		sdev->device_busy--;
> +	}
>  	if(sdev->device_busy == 0)
>  		blk_plug_device(q);
>   out:
Yeah, sorry. That patch was bad. Please use the attached one instead.
Andrew, can you replace them?

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		      zSeries & Storage
hare@suse.de			      +49 911 74053 688
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Markus Rex, HRB 16746 (AG Nürnberg)

[-- Attachment #2: tmp --]
[-- Type: text/plain, Size: 758 bytes --]

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 13e7e09..9ec1566 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1284,13 +1284,15 @@ int scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
 			/*
 			 * If the devices is blocked we defer normal commands.
 			 */
-			if (!(req->cmd_flags & REQ_PREEMPT))
-				ret = BLKPREP_DEFER;
-			/*
-			 * Return failfast requests immediately
-			 */
-			if (req->cmd_flags & REQ_FAILFAST)
-				ret = BLKPREP_KILL;
+			if (!(req->cmd_flags & REQ_PREEMPT)) {
+				/*
+				 * Return failfast requests immediately
+				 */
+				if (req->cmd_flags & REQ_FAILFAST)
+					ret = BLKPREP_KILL;
+				else
+					ret = BLKPREP_DEFER;
+			}
 			break;
 		default:
 			/*

Re: 2.6.24-rc4-mm1: hostbyte=0x01 driverbyte=0x00 (now bisected)

[Jens Axboe]

On Thu, Dec 06 2007, Hannes Reinecke wrote:
> Alexey Dobriyan wrote:
> >>  git-scsi-misc.patch
> > 
> > Apologies for not looking into the problem earlier. See
> > http://marc.info/?t=119628022300005&r=1&w=2
> > "2.6.24-rc3-mm2: Result: hostbyte=0x01 driverbyte=0x00\nend_request: I/O error"
> > for previous installment.
> > 
> > I've bisected it to the following patch in git-scsi-misc branch.
> > Revert on top of 2.6.24-rc4-mm1 also helps.
> > 
> > commit 8655a546c83fc43f0a73416bbd126d02de7ad6c0
> > Author: Hannes Reinecke <hare@suse.de>
> > Date:   Tue Nov 6 09:23:40 2007 +0100
> > 
> >     [SCSI] Do not requeue requests if REQ_FAILFAST is set
> >     
> >     Any requests with the REQ_FAILFAST flag set should not be requeued
> >     to the requeust queue, but rather terminated directly.
> >     Otherwise the multipath failover will stall until the command
> >     timeout triggers.
> >     
> >     Signed-off-by: Hannes Reinecke <hare@suse.de>
> >     Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
> > 
> > diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
> > index 0f44bdb..0da0dd0 100644
> > --- a/drivers/scsi/scsi_lib.c
> > +++ b/drivers/scsi/scsi_lib.c
> > @@ -1286,6 +1286,11 @@ int scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
> >  			 */
> >  			if (!(req->cmd_flags & REQ_PREEMPT))
> >  				ret = BLKPREP_DEFER;
> > +			/*
> > +			 * Return failfast requests immediately
> > +			 */
> > +			if (req->cmd_flags & REQ_FAILFAST)
> > +				ret = BLKPREP_KILL;
> >  			break;
> >  		default:
> >  			/*
> > @@ -1414,6 +1419,17 @@ static inline int scsi_host_queue_ready(struct request_queue *q,
> >  	return 1;
> >  }
> >  
> > +static void __scsi_kill_request(struct request *req)
> > +{
> > +	struct scsi_cmnd *cmd = req->special;
> > +	struct scsi_device *sdev = cmd->device;
> > +
> > +	cmd->result = DID_NO_CONNECT << 16;
> > +	atomic_inc(&cmd->device->iorequest_cnt);
> > +	sdev->device_busy--;
> > +	__scsi_done(cmd);
> > +}
> > +
> >  /*
> >   * Kill a request for a dead device
> >   */
> > @@ -1527,8 +1543,16 @@ static void scsi_request_fn(struct request_queue *q)
> >  		 * accept it.
> >  		 */
> >  		req = elv_next_request(q);
> > -		if (!req || !scsi_dev_queue_ready(q, sdev))
> > +		if (!req)
> > +			break;
> > +
> > +		if (!scsi_dev_queue_ready(q, sdev)) {
> > +			if (req->cmd_flags & REQ_FAILFAST) {
> > +				scsi_kill_request(req, q);
> > +				continue;
> > +			}
> >  			break;
> > +		}
> >  
> >  		if (unlikely(!scsi_device_online(sdev))) {
> >  			sdev_printk(KERN_ERR, sdev,
> > @@ -1609,8 +1633,12 @@ static void scsi_request_fn(struct request_queue *q)
> >  	 * later time.
> >  	 */
> >  	spin_lock_irq(q->queue_lock);
> > -	blk_requeue_request(q, req);
> > -	sdev->device_busy--;
> > +	if (unlikely(req->cmd_flags & REQ_FAILFAST))
> > +		__scsi_kill_request(req);
> > +	else {
> > +		blk_requeue_request(q, req);
> > +		sdev->device_busy--;
> > +	}
> >  	if(sdev->device_busy == 0)
> >  		blk_plug_device(q);
> >   out:
> Yeah, sorry. That patch was bad. Please use the attached one instead.
> Andrew, can you replace them?
> 
> Cheers,
> 
> Hannes
> -- 
> Dr. Hannes Reinecke		      zSeries & Storage
> hare@suse.de			      +49 911 74053 688
> SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg
> GF: Markus Rex, HRB 16746 (AG Nürnberg)

> diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
> index 13e7e09..9ec1566 100644
> --- a/drivers/scsi/scsi_lib.c
> +++ b/drivers/scsi/scsi_lib.c
> @@ -1284,13 +1284,15 @@ int scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
>  			/*
>  			 * If the devices is blocked we defer normal commands.
>  			 */
> -			if (!(req->cmd_flags & REQ_PREEMPT))
> -				ret = BLKPREP_DEFER;
> -			/*
> -			 * Return failfast requests immediately
> -			 */
> -			if (req->cmd_flags & REQ_FAILFAST)
> -				ret = BLKPREP_KILL;
> +			if (!(req->cmd_flags & REQ_PREEMPT)) {
> +				/*
> +				 * Return failfast requests immediately
> +				 */
> +				if (req->cmd_flags & REQ_FAILFAST)
> +					ret = BLKPREP_KILL;
> +				else
> +					ret = BLKPREP_DEFER;
> +			}
>  			break;
>  		default:
>  			/*

can we please stick to using blk_noretry_request() consistently, instead
of thwrowing REQ_FAILFAST tests in there?


-- 
Jens Axboe

-
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

Re: 2.6.24-rc4-mm1: hostbyte=0x01 driverbyte=0x00 (now bisected)

[Alexey Dobriyan]

On Thu, Dec 06, 2007 at 08:52:29AM +0100, Hannes Reinecke wrote:
> Alexey Dobriyan wrote:
> >>  git-scsi-misc.patch
> > 
> > Apologies for not looking into the problem earlier. See
> > http://marc.info/?t=119628022300005&r=1&w=2
> > "2.6.24-rc3-mm2: Result: hostbyte=0x01 driverbyte=0x00\nend_request: I/O error"
> > for previous installment.
> > 
> > I've bisected it to the following patch in git-scsi-misc branch.
> > Revert on top of 2.6.24-rc4-mm1 also helps.
> > 
> > commit 8655a546c83fc43f0a73416bbd126d02de7ad6c0
> > Author: Hannes Reinecke <hare@suse.de>
> > Date:   Tue Nov 6 09:23:40 2007 +0100
> > 
> >     [SCSI] Do not requeue requests if REQ_FAILFAST is set
> >     
> >     Any requests with the REQ_FAILFAST flag set should not be requeued
> >     to the requeust queue, but rather terminated directly.
> >     Otherwise the multipath failover will stall until the command
> >     timeout triggers.
> >     
> >     Signed-off-by: Hannes Reinecke <hare@suse.de>
> >     Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
> > 
> > diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
> > index 0f44bdb..0da0dd0 100644
> > --- a/drivers/scsi/scsi_lib.c
> > +++ b/drivers/scsi/scsi_lib.c
> > @@ -1286,6 +1286,11 @@ int scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
> >  			 */
> >  			if (!(req->cmd_flags & REQ_PREEMPT))
> >  				ret = BLKPREP_DEFER;
> > +			/*
> > +			 * Return failfast requests immediately
> > +			 */
> > +			if (req->cmd_flags & REQ_FAILFAST)
> > +				ret = BLKPREP_KILL;
> >  			break;
> >  		default:
> >  			/*
> > @@ -1414,6 +1419,17 @@ static inline int scsi_host_queue_ready(struct request_queue *q,
> >  	return 1;
> >  }
> >  
> > +static void __scsi_kill_request(struct request *req)
> > +{
> > +	struct scsi_cmnd *cmd = req->special;
> > +	struct scsi_device *sdev = cmd->device;
> > +
> > +	cmd->result = DID_NO_CONNECT << 16;
> > +	atomic_inc(&cmd->device->iorequest_cnt);
> > +	sdev->device_busy--;
> > +	__scsi_done(cmd);
> > +}
> > +
> >  /*
> >   * Kill a request for a dead device
> >   */
> > @@ -1527,8 +1543,16 @@ static void scsi_request_fn(struct request_queue *q)
> >  		 * accept it.
> >  		 */
> >  		req = elv_next_request(q);
> > -		if (!req || !scsi_dev_queue_ready(q, sdev))
> > +		if (!req)
> > +			break;
> > +
> > +		if (!scsi_dev_queue_ready(q, sdev)) {
> > +			if (req->cmd_flags & REQ_FAILFAST) {
> > +				scsi_kill_request(req, q);
> > +				continue;
> > +			}
> >  			break;
> > +		}
> >  
> >  		if (unlikely(!scsi_device_online(sdev))) {
> >  			sdev_printk(KERN_ERR, sdev,
> > @@ -1609,8 +1633,12 @@ static void scsi_request_fn(struct request_queue *q)
> >  	 * later time.
> >  	 */
> >  	spin_lock_irq(q->queue_lock);
> > -	blk_requeue_request(q, req);
> > -	sdev->device_busy--;
> > +	if (unlikely(req->cmd_flags & REQ_FAILFAST))
> > +		__scsi_kill_request(req);
> > +	else {
> > +		blk_requeue_request(q, req);
> > +		sdev->device_busy--;
> > +	}
> >  	if(sdev->device_busy == 0)
> >  		blk_plug_device(q);
> >   out:
> Yeah, sorry. That patch was bad. Please use the attached one instead.
> Andrew, can you replace them?

Instead? It won't apply. And it doesn't help on top of git-scsi.
It helps if 3 hunks involving __scsi_kill_request() are ducked.

> --- a/drivers/scsi/scsi_lib.c
> +++ b/drivers/scsi/scsi_lib.c
> @@ -1284,13 +1284,15 @@ int scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
>  			/*
>  			 * If the devices is blocked we defer normal commands.
>  			 */
> -			if (!(req->cmd_flags & REQ_PREEMPT))
> -				ret = BLKPREP_DEFER;
> -			/*
> -			 * Return failfast requests immediately
> -			 */
> -			if (req->cmd_flags & REQ_FAILFAST)
> -				ret = BLKPREP_KILL;
> +			if (!(req->cmd_flags & REQ_PREEMPT)) {
> +				/*
> +				 * Return failfast requests immediately
> +				 */
> +				if (req->cmd_flags & REQ_FAILFAST)
> +					ret = BLKPREP_KILL;
> +				else
> +					ret = BLKPREP_DEFER;
> +			}
>  			break;
>  		default:
>  			/*