On Tue, 2010-08-03 at 21:26 +0200, Ferenc Wagner wrote: > Package: linux-2.6 > Version: 2.6.32-17 > Severity: normal > > Hi, > > I reported this in February to LKML without much reaction, see > http://lkml.org/lkml/2010/2/2/208 > It also happened with the latest sid kernel, so reporting here. > > Now it didn't happen during system bootup, but after an uptime of > several days. The message, as received by our log server: [...] Please try building and installing a kernel package with the attached patches applied. Follow the instructions at <http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s-common-official>. You may also need to rebuild the linux-base package. To do that, run 'fakeroot make -f debian/rules.real install-linux-base' after following those instructions. Ben. -- Ben Hutchings Once a job is fouled up, anything done to improve it makes it worse.
From 6c71dcb28ff9b63b814a0b76a256f5dae08d3e0d Mon Sep 17 00:00:00 2001
From: Hannes Reinecke <hare@suse.de>
Date: Wed, 2 Dec 2009 14:28:48 -0600
Subject: [PATCH 1/2] [SCSI] scsi_dh_emc: fix mode select request setup
This patch fixes the request setup code for mode selects. I got the fixes from
Hannes Reinecke while trying to hunt down some problems and merged it
into one patch. I am sending it because Hannes is busy with other things.
The patch fixes:
- setting of the length for mode selects.
- setting of the data direction for mode select 10.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
---
drivers/scsi/device_handler/scsi_dh_emc.c | 6 ++++--
1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/scsi/device_handler/scsi_dh_emc.c b/drivers/scsi/device_handler/scsi_dh_emc.c
index 6196675..63032ec 100644
--- a/drivers/scsi/device_handler/scsi_dh_emc.c
+++ b/drivers/scsi/device_handler/scsi_dh_emc.c
@@ -272,7 +272,7 @@ static struct request *get_req(struct scsi_device *sdev, int cmd,
int len = 0;
rq = blk_get_request(sdev->request_queue,
- (cmd == MODE_SELECT) ? WRITE : READ, GFP_NOIO);
+ (cmd != INQUIRY) ? WRITE : READ, GFP_NOIO);
if (!rq) {
sdev_printk(KERN_INFO, sdev, "get_req: blk_get_request failed");
return NULL;
@@ -286,14 +286,17 @@ static struct request *get_req(struct scsi_device *sdev, int cmd,
len = sizeof(short_trespass);
rq->cmd_flags |= REQ_RW;
rq->cmd[1] = 0x10;
+ rq->cmd[4] = len;
break;
case MODE_SELECT_10:
len = sizeof(long_trespass);
rq->cmd_flags |= REQ_RW;
rq->cmd[1] = 0x10;
+ rq->cmd[8] = len;
break;
case INQUIRY:
len = CLARIION_BUFFER_SIZE;
+ rq->cmd[4] = len;
memset(buffer, 0, len);
break;
default:
@@ -301,7 +304,6 @@ static struct request *get_req(struct scsi_device *sdev, int cmd,
break;
}
- rq->cmd[4] = len;
rq->cmd_type = REQ_TYPE_BLOCK_PC;
rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
REQ_FAILFAST_DRIVER;
--
1.7.1
From 5738d4449c1baf05e8345684d12371f76296473d Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Fri, 9 Apr 2010 22:07:37 -0500
Subject: [PATCH 2/2] [SCSI] scsi_dh_emc: request flag cleanup
blk_get_request sets the cmd_flags, so we should not and do not
need to set them. If we did set them to a different value then
it can cause a oops in the elevator code.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
---
drivers/scsi/device_handler/scsi_dh_emc.c | 2 --
1 files changed, 0 insertions(+), 2 deletions(-)
diff --git a/drivers/scsi/device_handler/scsi_dh_emc.c b/drivers/scsi/device_handler/scsi_dh_emc.c
index e8a0bc3..6faf472 100644
--- a/drivers/scsi/device_handler/scsi_dh_emc.c
+++ b/drivers/scsi/device_handler/scsi_dh_emc.c
@@ -285,13 +285,11 @@ static struct request *get_req(struct scsi_device *sdev, int cmd,
switch (cmd) {
case MODE_SELECT:
len = sizeof(short_trespass);
- rq->cmd_flags |= REQ_RW;
rq->cmd[1] = 0x10;
rq->cmd[4] = len;
break;
case MODE_SELECT_10:
len = sizeof(long_trespass);
- rq->cmd_flags |= REQ_RW;
rq->cmd[1] = 0x10;
rq->cmd[8] = len;
break;
--
1.7.1
Attachment:
signature.asc
Description: This is a digitally signed message part