Skip to content

Commit

Permalink
[SCSI] libfc: exch mgr is freed while lport still retrying sequences
Browse files Browse the repository at this point in the history
When a sequence cannot be delivered to the target, the local
port will schedule retries, While this process is in progress,
if we destroy the FCoE interface, the fcoe_sw_destroy routine is
entered, and the fc_exch_mgr_free(lp->emp) is called.  Thus
if fc_exch_alloc() is called when retrying the sequence,
the mempool_alloc() will fail to allocate the exchange because
the mempool of the exchange manager has already been released.
This patch is to cancel any pending retry work of the local
port before we start to destroy the interface.

Also, when resetting the local port, we should also stop the
scheduled pending retries.

Signed-off-by: Steve Ma <steve.ma@intel.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
  • Loading branch information
Steve Ma authored and James Bottomley committed Mar 6, 2009
1 parent 26d9cab commit f7db2c1
Showing 1 changed file with 2 additions and 0 deletions.
2 changes: 2 additions & 0 deletions drivers/scsi/libfc/fc_lport.c
Original file line number Diff line number Diff line change
Expand Up @@ -619,6 +619,7 @@ int fc_fabric_logoff(struct fc_lport *lport)
mutex_lock(&lport->lp_mutex);
fc_lport_enter_logo(lport);
mutex_unlock(&lport->lp_mutex);
cancel_delayed_work_sync(&lport->retry_work);
return 0;
}
EXPORT_SYMBOL(fc_fabric_logoff);
Expand Down Expand Up @@ -918,6 +919,7 @@ static void fc_lport_recv_req(struct fc_lport *lport, struct fc_seq *sp,
*/
int fc_lport_reset(struct fc_lport *lport)
{
cancel_delayed_work_sync(&lport->retry_work);
mutex_lock(&lport->lp_mutex);
fc_lport_enter_reset(lport);
mutex_unlock(&lport->lp_mutex);
Expand Down

0 comments on commit f7db2c1

Please sign in to comment.