Prev: [40/90] mm: hugetlb: fix hugepage memory leak in walk_page_range()
Next: [151/151] implement early_io{re,un}map for ia64
From: Greg KH on 16 Dec 2009 20:50
2.6.31-stable review patch. If anyone has any objections, please let us know.

------------------

From: NeilBrown <neilb(a)suse.de>

commit aa5cbd103887011b4830355f88fb055f9ad2d556 upstream.

A write intent bitmap can be removed from an array while the
array is active.
When this happens, all IO is suspended and flushed before the
bitmap is removed.
However it is possible that bitmap_daemon_work is still running to
clear old bits from the bitmap. If it is, it can dereference the
bitmap after it has been freed.

So introduce a new mutex to protect bitmap_daemon_work and get it
before destroying a bitmap.

This is suitable for any current -stable kernel.

Signed-off-by: NeilBrown <neilb(a)suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh(a)suse.de>

---
drivers/md/bitmap.c | 24 ++++++++++++++++++------
drivers/md/bitmap.h | 2 +-
drivers/md/md.c | 3 ++-
drivers/md/md.h | 1 +
4 files changed, 22 insertions(+), 8 deletions(-)

--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -1077,23 +1077,31 @@ static bitmap_counter_t *bitmap_get_coun
* out to disk
*/

-void bitmap_daemon_work(struct bitmap *bitmap)
+void bitmap_daemon_work(mddev_t *mddev)
{
+ struct bitmap *bitmap;
unsigned long j;
unsigned long flags;
struct page *page = NULL, *lastpage = NULL;
int blocks;
void *paddr;

- if (bitmap == NULL)
+ /* Use a mutex to guard daemon_work against
+ * bitmap_destroy.
+ */
+ mutex_lock(&mddev->bitmap_mutex);
+ bitmap = mddev->bitmap;
+ if (bitmap == NULL) {
+ mutex_unlock(&mddev->bitmap_mutex);
return;
+ }
if (time_before(jiffies, bitmap->daemon_lastrun + bitmap->daemon_sleep*HZ))
goto done;

bitmap->daemon_lastrun = jiffies;
if (bitmap->allclean) {
bitmap->mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
- return;
+ goto done;
}
bitmap->allclean = 1;

@@ -1202,6 +1210,7 @@ void bitmap_daemon_work(struct bitmap *b
done:
if (bitmap->allclean == 0)
bitmap->mddev->thread->timeout = bitmap->daemon_sleep * HZ;
+ mutex_unlock(&mddev->bitmap_mutex);
}

static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
@@ -1538,9 +1547,9 @@ void bitmap_flush(mddev_t *mddev)
*/
sleep = bitmap->daemon_sleep;
bitmap->daemon_sleep = 0;
- bitmap_daemon_work(bitmap);
- bitmap_daemon_work(bitmap);
- bitmap_daemon_work(bitmap);
+ bitmap_daemon_work(mddev);
+ bitmap_daemon_work(mddev);
+ bitmap_daemon_work(mddev);
bitmap->daemon_sleep = sleep;
bitmap_update_sb(bitmap);
}
@@ -1571,6 +1580,7 @@ static void bitmap_free(struct bitmap *b
kfree(bp);
kfree(bitmap);
}
+
void bitmap_destroy(mddev_t *mddev)
{
struct bitmap *bitmap = mddev->bitmap;
@@ -1578,7 +1588,9 @@ void bitmap_destroy(mddev_t *mddev)
if (!bitmap) /* there was no bitmap */
return;

+ mutex_lock(&mddev->bitmap_mutex);
mddev->bitmap = NULL; /* disconnect from the md device */
+ mutex_unlock(&mddev->bitmap_mutex);
if (mddev->thread)
mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;

--- a/drivers/md/bitmap.h
+++ b/drivers/md/bitmap.h
@@ -282,7 +282,7 @@ void bitmap_close_sync(struct bitmap *bi
void bitmap_cond_end_sync(struct bitmap *bitmap, sector_t sector);

void bitmap_unplug(struct bitmap *bitmap);
-void bitmap_daemon_work(struct bitmap *bitmap);
+void bitmap_daemon_work(mddev_t *mddev);
#endif

#endif
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -361,6 +361,7 @@ static mddev_t * mddev_find(dev_t unit)

mutex_init(&new->open_mutex);
mutex_init(&new->reconfig_mutex);
+ mutex_init(&new->bitmap_mutex);
INIT_LIST_HEAD(&new->disks);
INIT_LIST_HEAD(&new->all_mddevs);
init_timer(&new->safemode_timer);
@@ -6595,7 +6596,7 @@ void md_check_recovery(mddev_t *mddev)


if (mddev->bitmap)
- bitmap_daemon_work(mddev->bitmap);
+ bitmap_daemon_work(mddev);

if (mddev->ro)
return;
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -289,6 +289,7 @@ struct mddev_s
* hot-adding a bitmap. It should
* eventually be settable by sysfs.
*/
+ struct mutex bitmap_mutex;

struct list_head all_mddevs;
};


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
 | 
Pages: 1
Prev: [40/90] mm: hugetlb: fix hugepage memory leak in walk_page_range()
Next: [151/151] implement early_io{re,un}map for ia64