Two fairly simple fixes: one is a change that causes us to have a very
low queue depth leading to performance issues and the other is a null
deref occasionally in tapes thanks to use after put.
The patch is available here:
git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi.git scsi-fixes
The short changelog is:
Jens Axboe (1):
scsi: fix host max depth checking for the 'queue_depth' sysfs interface
Seymour, Shane M (1):
st: null pointer dereference panic caused by use after kref_put by st_open
And the diffstat:
drivers/scsi/scsi_sysfs.c | 2 +-
drivers/scsi/st.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
With full diffs below.
James
---
diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c
index 1ac38e7..9ad4116 100644
--- a/drivers/scsi/scsi_sysfs.c
+++ b/drivers/scsi/scsi_sysfs.c
@@ -859,7 +859,7 @@ sdev_store_queue_depth(struct device *dev, struct device_attribute *attr,
depth = simple_strtoul(buf, NULL, 0);
- if (depth < 1 || depth > sht->can_queue)
+ if (depth < 1 || depth > sdev->host->can_queue)
return -EINVAL;
retval = sht->change_queue_depth(sdev, depth);
diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c
index 3f25b8f..871f355 100644
--- a/drivers/scsi/st.c
+++ b/drivers/scsi/st.c
@@ -1329,9 +1329,9 @@ static int st_open(struct inode *inode, struct file *filp)
spin_lock(&st_use_lock);
STp->in_use = 0;
spin_unlock(&st_use_lock);
- scsi_tape_put(STp);
if (resumed)
scsi_autopm_put_device(STp->device);
+ scsi_tape_put(STp);
return retval;
}
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
