[PATCH] dont decrement counters from scsi_kill_request

[PATCH] dont decrement counters from scsi_kill_request

[Mike Christie]

This was debugged by zhenyu.z.wang@intel.com on the open-iscsi lists.

When we call scsi_kill_request() we have not yet incremented the
device and host busy counters so calling __scsi_done from
scsi_kill_request ends up doing extra decrements.

This patch has only been compile tested against scsi-misc. I am not
sure what the correct blk_complete_barrier_rq values should be. For
failures are we supposed to return zero or what is left in the command
and what how are we supposed set to indicate a error?

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>



diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index a7f3f0c..4b6a9fc 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1514,8 +1514,6 @@ static void scsi_kill_request(struct req
 {
 	struct scsi_cmnd *cmd = req->special;
 
-	blkdev_dequeue_request(req);
-
 	if (unlikely(cmd == NULL)) {
 		printk(KERN_CRIT "impossible request in %s.\n",
 				 __FUNCTION__);
@@ -1523,9 +1521,19 @@ static void scsi_kill_request(struct req
 	}
 
 	scsi_init_cmd_errh(cmd);
-	cmd->result = DID_NO_CONNECT << 16;
 	atomic_inc(&cmd->device->iorequest_cnt);
-	__scsi_done(cmd);
+
+	blkdev_dequeue_request(req);
+	req->errors = DID_NO_CONNECT << 16;
+
+	if (blk_complete_barrier_rq(q, req, cmd->bufflen  >> 9)); 
+		return;
+
+	scsi_release_buffers(cmd);
+	while (end_that_request_first(req, 0, req->nr_sectors))
+		;
+	end_that_request_last(req);
+	scsi_put_command(cmd);
 }
 
 /*

Re: [PATCH] dont decrement counters from scsi_kill_request

[Rolf Eike Beer]

[-- Attachment #1: Type: text/plain, Size: 204 bytes --]

Mike Christie wrote:

>+	while (end_that_request_first(req, 0, req->nr_sectors))
>+		;

This should be something like

while (...)
	{}

so it is absolutely clear that the empty body is intentional.

Eike

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [PATCH] dont decrement counters from scsi_kill_request

[Al Viro]

On Wed, Jan 04, 2006 at 08:51:57AM +0100, Rolf Eike Beer wrote:
> Mike Christie wrote:
> 
> >+	while (end_that_request_first(req, 0, req->nr_sectors))
> >+		;
> 
> This should be something like
> 
> while (...)
> 	{}
> 
> so it is absolutely clear that the empty body is intentional.

Please, do not inflict that ugliness on everyone.  FWIW, it's less
idiomatic than normal use of ; and personally I'd parse it as "somebody
had fscked up using editor", not as "wow, they _really_ meant it to be
an empty body".

Re: [PATCH] dont decrement counters from scsi_kill_request

[Rolf Eike Beer]

[-- Attachment #1: Type: text/plain, Size: 847 bytes --]

Al Viro wrote:
>On Wed, Jan 04, 2006 at 08:51:57AM +0100, Rolf Eike Beer wrote:
>> Mike Christie wrote:
>> >+	while (end_that_request_first(req, 0, req->nr_sectors))
>> >+		;
>>
>> This should be something like
>>
>> while (...)
>> 	{}
>>
>> so it is absolutely clear that the empty body is intentional.
>
>Please, do not inflict that ugliness on everyone.  FWIW, it's less
>idiomatic than normal use of ; and personally I'd parse it as "somebody
>had fscked up using editor", not as "wow, they _really_ meant it to be
>an empty body".

Using only ; on a single line looks like some code left over from removing 
code... Ok, so let's agree to disagree. Since coding style says the opening 
brace should be on the same line as the while, would it be better to put 
everthing in one line to make clear what's going on?

Eike

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [PATCH] dont decrement counters from scsi_kill_request

[Christoph Hellwig]

On Wed, Jan 04, 2006 at 09:13:36AM +0100, Rolf Eike Beer wrote:
> Al Viro wrote:
> >On Wed, Jan 04, 2006 at 08:51:57AM +0100, Rolf Eike Beer wrote:
> >> Mike Christie wrote:
> >> >+	while (end_that_request_first(req, 0, req->nr_sectors))
> >> >+		;
> >>
> >> This should be something like
> >>
> >> while (...)
> >> 	{}
> >>
> >> so it is absolutely clear that the empty body is intentional.
> >
> >Please, do not inflict that ugliness on everyone.  FWIW, it's less
> >idiomatic than normal use of ; and personally I'd parse it as "somebody
> >had fscked up using editor", not as "wow, they _really_ meant it to be
> >an empty body".
> 
> Using only ; on a single line looks like some code left over from removing 
> code...

It's a common idiom and above code is copy and pasted from a perfectly
fine indented function.  So could you please stop this silly discussion
now?