summaryrefslogtreecommitdiffstats
path: root/drivers/scsi/dtc.c
diff options
context:
space:
mode:
authorMartin Peschke <mp3@de.ibm.com>2007-11-15 13:57:17 +0100
committerJames Bottomley <James.Bottomley@HansenPartnership.com>2007-11-16 13:03:21 -0600
commit86e8dfc5603ed76917eed0a9dd9e85a1e1a8b162 (patch)
tree3d5bf28741aacff3f60df35c0619c7d9cd352f44 /drivers/scsi/dtc.c
parentd0076f7754dce07c7a1d752034561acadd99eafa (diff)
[SCSI] zfcp: fix cleanup of dismissed error recovery actions
Calling zfcp_erp_strategy_check_action() after zfcp_erp_action_to_running() in zfcp_erp_strategy() might cause an unbalanced up() for erp_ready_sem, which makes the zfcp recovery fail somewhere along the way: erp thread processing erp_action: | | someone waking up erp thread for erp_action | | | | someone else dismissing erp_action: | | | V V V write_lock_irqsave(&adapter->erp_lock, flags); ... if (zfcp_erp_action_exists(erp_action) == ZFCP_ERP_ACTION_RUNNING) { zfcp_erp_action_to_ready(erp_action); up(&adapter->erp_ready_sem); /* first up() for erp_action */ } write_unlock_irqrestore(&adapter->erp_lock, flags); write_lock_irqsave(&adapter->erp_lock, flags); ... zfcp_erp_action_to_running(erp_action); write_unlock_restore(&adapter->erp_lock, flags); /* processing erp_action */ write_lock_irqsave(&adapter->erp_lock, flags); ... erp_action->status |= ZFCP_STATUS_ERP_DISMISSED; if (zfcp_erp_action_exists(erp_action) == ZFCP_ERP_ACTION_RUNNING) { zfcp_erp_action_to_ready(erp_action); up(&adapter->erp_ready_sem); /* second, unbalanced up() for erp_action */ } ... write_unlock_restore(&adapter->erp_lock, flags); write_lock_irqsave(&adapter->erp_lock, flags); if (erp_action->status & ZFCP_STATUS_ERP_DISMISSED) { zfcp_erp_action_dequeue(erp_action); retval = ZFCP_ERP_DISMISSED; } ... write_unlock_restore(&adapter->erp_lock, flags); down(&adapter->erp_ready_sem); /* this down() is meant to balance the first up() */ The erp thread must not dismiss an erp_action after moving that action to erp_running_head. Instead it should just go through the down() operation, which balances the first up(), and run through zfcp_erp_strategy one more time for the second up(), which eventually cleans up erp_action. Which is similar to the normal processing of an event for erp_action doing something asynchronously (e.g. waiting for the completion of an fsf_req). This only works if we make sure that a dismissed erp_action is passed to zfcp_erp_strategy() prior to the other action, which caused actions to be dismissed. Therefore the patch implements this rule: running actions go to the head of the ready list; new actions go to the tail of the ready list; the erp thread picks actions to be processed from the ready list's head. Signed-off-by: Martin Peschke <mp3@de.ibm.com> Acked-by: Swen Schillig <swen@vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Diffstat (limited to 'drivers/scsi/dtc.c')
0 files changed, 0 insertions, 0 deletions