summaryrefslogtreecommitdiff
path: root/drivers/ide
diff options
context:
space:
mode:
authorBartlomiej Zolnierkiewicz <bzolnier@gmail.com>2009-01-06 17:20:51 +0100
committerBartlomiej Zolnierkiewicz <bzolnier@gmail.com>2009-01-06 17:20:51 +0100
commitff670e632e818535950d9e31d9d0dcfc7d5b4090 (patch)
tree3ac6ade46bb9d2ae8578ed63fa8946f817749026 /drivers/ide
parentb65fac32cfe3b2f98cd472fef400bd1c1340de23 (diff)
ide: update ide_unregister() documentation
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Diffstat (limited to 'drivers/ide')
-rw-r--r--drivers/ide/ide.c18
1 files changed, 5 insertions, 13 deletions
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 9e403282dfa..6538b63158b 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -155,22 +155,14 @@ EXPORT_SYMBOL_GPL(ide_port_unregister_devices);
* ide_unregister - free an IDE interface
* @hwif: IDE interface
*
- * Perform the final unregister of an IDE interface. At the moment
- * we don't refcount interfaces so this will also get split up.
+ * Perform the final unregister of an IDE interface.
*
* Locking:
- * The caller must not hold the IDE locks
- * The drive present/vanishing is not yet properly locked
- * Take care with the callbacks. These have been split to avoid
- * deadlocking the IDE layer. The shutdown callback is called
- * before we take the lock and free resources. It is up to the
- * caller to be sure there is no pending I/O here, and that
- * the interface will not be reopened (present/vanishing locking
- * isn't yet done BTW). After we commit to the final kill we
- * call the cleanup callback with the ide locks held.
+ * The caller must not hold the IDE locks.
*
- * Unregister restores the hwif structures to the default state.
- * This is raving bonkers.
+ * It is up to the caller to be sure there is no pending I/O here,
+ * and that the interface will not be reopened (present/vanishing
+ * locking isn't yet done BTW).
*/
void ide_unregister(ide_hwif_t *hwif)