[GIT PULL] SCSI bug fixes for 2.6.36-rc7

[GIT PULL] SCSI bug fixes for 2.6.36-rc7

[James Bottomley]

This patch set fixes a couple of longstanding bugs in SCSI.  The VPD
patch fix a possible buffer overrun in the VPD code (only tripped if the
device has a huge number of VPD pages).  The other is where the status
return code on our BSG interface is wrongly right shifted one place.

The patch is available here:

master.kernel.org:/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6.git

The short changelog is:

FUJITA Tomonori (1):
      bsg: fix incorrect device_status value

Martin K. Petersen (1):
      Fix VPD inquiry page wrapper


The diffstat is:

 block/bsg.c         |    2 +-
 drivers/scsi/scsi.c |    4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

And the full diff is below.

James

---

diff --git a/block/bsg.c b/block/bsg.c
index 82d5882..5974682 100644
--- a/block/bsg.c
+++ b/block/bsg.c
@@ -426,7 +426,7 @@ static int blk_complete_sgv4_hdr_rq(struct request *rq, struct sg_io_v4 *hdr,
 	/*
 	 * fill in all the output members
 	 */
-	hdr->device_status = status_byte(rq->errors);
+	hdr->device_status = rq->errors && 0xff;
 	hdr->transport_status = host_byte(rq->errors);
 	hdr->driver_status = driver_byte(rq->errors);
 	hdr->info = 0;
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index ad0ed21..348fba0 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -1046,13 +1046,13 @@ int scsi_get_vpd_page(struct scsi_device *sdev, u8 page, unsigned char *buf,
 
 	/* If the user actually wanted this page, we can skip the rest */
 	if (page == 0)
-		return -EINVAL;
+		return 0;
 
 	for (i = 0; i < min((int)buf[3], buf_len - 4); i++)
 		if (buf[i + 4] == page)
 			goto found;
 
-	if (i < buf[3] && i > buf_len)
+	if (i < buf[3] && i >= buf_len - 4)
 		/* ran off the end of the buffer, give us benefit of doubt */
 		goto found;
 	/* The device claims it doesn't support the requested page */

Re: [GIT PULL] SCSI bug fixes for 2.6.36-rc7

[Linus Torvalds]

On Wed, Oct 13, 2010 at 6:45 AM, James Bottomley
<James.Bottomley@suse.de> wrote:
> This patch set fixes a couple of longstanding bugs in SCSI.  The VPD
> patch fix a possible buffer overrun in the VPD code (only tripped if the
> device has a huge number of VPD pages).  The other is where the status
> return code on our BSG interface is wrongly right shifted one place.

Umm. That other patch seems to be utter crap.

> -       hdr->device_status = status_byte(rq->errors);
> +       hdr->device_status = rq->errors && 0xff;

That's pure shit. Please look at it for one second, and don't send me
crap like this.

                        Linus "grumpy" Torvalds

Re: [GIT PULL] SCSI bug fixes for 2.6.36-rc7

[James Bottomley]

On Wed, 2010-10-13 at 16:49 -0700, Linus Torvalds wrote:
> On Wed, Oct 13, 2010 at 6:45 AM, James Bottomley
> <James.Bottomley@suse.de> wrote:
> > This patch set fixes a couple of longstanding bugs in SCSI.  The VPD
> > patch fix a possible buffer overrun in the VPD code (only tripped if the
> > device has a huge number of VPD pages).  The other is where the status
> > return code on our BSG interface is wrongly right shifted one place.
> 
> Umm. That other patch seems to be utter crap.
> 
> > -       hdr->device_status = status_byte(rq->errors);
> > +       hdr->device_status = rq->errors && 0xff;
> 
> That's pure shit. Please look at it for one second, and don't send me
> crap like this.

What's wrong with it? device_status() is an ancient SCSI-1 macro that
dates from the time the status was left shifted by 1, which makes the
value wrong according to the sg interface spec.  We should really remove
it, but there's still a few ancient drivers using it.  the SG interface
requires the full status byte, which is what the bottom most byte of
req_errors is ... hence the change.

James

Re: [GIT PULL] SCSI bug fixes for 2.6.36-rc7

[Harvey Harrison]

On Wed, Oct 13, 2010 at 5:24 PM, James Bottomley
<James.Bottomley@suse.de> wrote:
> On Wed, 2010-10-13 at 16:49 -0700, Linus Torvalds wrote:
>> On Wed, Oct 13, 2010 at 6:45 AM, James Bottomley
>> <James.Bottomley@suse.de> wrote:

>>
>> Umm. That other patch seems to be utter crap.
>>
>> > -       hdr->device_status = status_byte(rq->errors);
>> > +       hdr->device_status = rq->errors && 0xff;
>>
>> That's pure shit. Please look at it for one second, and don't send me
>> crap like this.
>
> What's wrong with it? device_status() is an ancient SCSI-1 macro that
> dates from the time the status was left shifted by 1, which makes the
> value wrong according to the sg interface spec.  We should really remove
> it, but there's still a few ancient drivers using it.  the SG interface
> requires the full status byte, which is what the bottom most byte of
> req_errors is ... hence the change.
>

How about & instead of &&

Harvey

Re: [GIT PULL] SCSI bug fixes for 2.6.36-rc7

[Harvey Harrison]

On Wed, Oct 13, 2010 at 5:24 PM, James Bottomley
<James.Bottomley@suse.de> wrote:
> On Wed, 2010-10-13 at 16:49 -0700, Linus Torvalds wrote:
>> On Wed, Oct 13, 2010 at 6:45 AM, James Bottomley
>> <James.Bottomley@suse.de> wrote:

>>
>> Umm. That other patch seems to be utter crap.
>>
>> > -       hdr->device_status = status_byte(rq->errors);
>> > +       hdr->device_status = rq->errors && 0xff;
>>
>> That's pure shit. Please look at it for one second, and don't send me
>> crap like this.
>
> What's wrong with it? device_status() is an ancient SCSI-1 macro that
> dates from the time the status was left shifted by 1, which makes the
> value wrong according to the sg interface spec.  We should really remove
> it, but there's still a few ancient drivers using it.  the SG interface
> requires the full status byte, which is what the bottom most byte of
> req_errors is ... hence the change.
>

How about & instead of &&

Harvey
--
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: [GIT PULL] SCSI bug fixes for 2.6.36-rc7

[James Bottomley]

On Wed, 2010-10-13 at 20:24 -0400, James Bottomley wrote:
> On Wed, 2010-10-13 at 16:49 -0700, Linus Torvalds wrote:
> > On Wed, Oct 13, 2010 at 6:45 AM, James Bottomley
> > <James.Bottomley@suse.de> wrote:
> > > This patch set fixes a couple of longstanding bugs in SCSI.  The VPD
> > > patch fix a possible buffer overrun in the VPD code (only tripped if the
> > > device has a huge number of VPD pages).  The other is where the status
> > > return code on our BSG interface is wrongly right shifted one place.
> > 
> > Umm. That other patch seems to be utter crap.
> > 
> > > -       hdr->device_status = status_byte(rq->errors);
> > > +       hdr->device_status = rq->errors && 0xff;
> > 
> > That's pure shit. Please look at it for one second, and don't send me
> > crap like this.
> 
> What's wrong with it? device_status() is an ancient SCSI-1 macro that
> dates from the time the status was left shifted by 1, which makes the
> value wrong according to the sg interface spec.  We should really remove
> it, but there's still a few ancient drivers using it.  the SG interface
> requires the full status byte, which is what the bottom most byte of
> req_errors is ... hence the change.

OK, right ... several people pointed out in private email that it was &&
not & ... I'll fix it when I'm slightly less jet lagged.

James

Re: [GIT PULL] SCSI bug fixes for 2.6.36-rc7

[FUJITA Tomonori]

On Wed, 13 Oct 2010 20:34:37 -0400
James Bottomley <James.Bottomley@suse.de> wrote:

> On Wed, 2010-10-13 at 20:24 -0400, James Bottomley wrote:
> > On Wed, 2010-10-13 at 16:49 -0700, Linus Torvalds wrote:
> > > On Wed, Oct 13, 2010 at 6:45 AM, James Bottomley
> > > <James.Bottomley@suse.de> wrote:
> > > > This patch set fixes a couple of longstanding bugs in SCSI.  The VPD
> > > > patch fix a possible buffer overrun in the VPD code (only tripped if the
> > > > device has a huge number of VPD pages).  The other is where the status
> > > > return code on our BSG interface is wrongly right shifted one place.
> > > 
> > > Umm. That other patch seems to be utter crap.
> > > 
> > > > -       hdr->device_status = status_byte(rq->errors);
> > > > +       hdr->device_status = rq->errors && 0xff;
> > > 
> > > That's pure shit. Please look at it for one second, and don't send me
> > > crap like this.
> > 
> > What's wrong with it? device_status() is an ancient SCSI-1 macro that
> > dates from the time the status was left shifted by 1, which makes the
> > value wrong according to the sg interface spec.  We should really remove
> > it, but there's still a few ancient drivers using it.  the SG interface
> > requires the full status byte, which is what the bottom most byte of
> > req_errors is ... hence the change.
> 
> OK, right ... several people pointed out in private email that it was &&
> not & ... I'll fix it when I'm slightly less jet lagged.

Really sorry about that.

=
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Subject: [PATCH] bsg: fix incorrect device_status value

bsg incorrectly returns sg's masked_status value for device_status.

Reported-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
---
 block/bsg.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/block/bsg.c b/block/bsg.c
index 82d5882..0c00870 100644
--- a/block/bsg.c
+++ b/block/bsg.c
@@ -426,7 +426,7 @@ static int blk_complete_sgv4_hdr_rq(struct request *rq, struct sg_io_v4 *hdr,
 	/*
 	 * fill in all the output members
 	 */
-	hdr->device_status = status_byte(rq->errors);
+	hdr->device_status = rq->errors & 0xff;
 	hdr->transport_status = host_byte(rq->errors);
 	hdr->driver_status = driver_byte(rq->errors);
 	hdr->info = 0;
-- 
1.7.1