[008/146] [SCSI] qla2xxx: Fix hang during driver unload when vport is active.
From: Greg KH
Date: Wed Jun 01 2011 - 04:07:10 EST
2.6.38-stable review patch. If anyone has any objections, please let us know.
------------------
From: Arun Easi <arun.easi@xxxxxxxxxx>
commit 43ebf16d762b082663976b679b813e1b546548d1 upstream.
Bumping ref count during fc_vport_terminate() was the cause. vport
delete would wait for ref count to drop to zero and that would never
happen.
Signed-off-by: Arun Easi <arun.easi@xxxxxxxxxx>
Signed-off-by: Madhuranath Iyengar <Madhu.Iyengar@xxxxxxxxxx>
Signed-off-by: James Bottomley <jbottomley@xxxxxxxxxxxxx>
Signed-off-by: James Bottomley <James.Bottomley@xxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxx>
---
drivers/scsi/qla2xxx/qla_os.c | 25 +++++++++++++++----------
1 file changed, 15 insertions(+), 10 deletions(-)
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -2359,21 +2359,26 @@ qla2x00_remove_one(struct pci_dev *pdev)
base_vha = pci_get_drvdata(pdev);
ha = base_vha->hw;
- spin_lock_irqsave(&ha->vport_slock, flags);
- list_for_each_entry(vha, &ha->vp_list, list) {
- atomic_inc(&vha->vref_count);
+ mutex_lock(&ha->vport_lock);
+ while (ha->cur_vport_count) {
+ struct Scsi_Host *scsi_host;
- if (vha->fc_vport) {
- spin_unlock_irqrestore(&ha->vport_slock, flags);
+ spin_lock_irqsave(&ha->vport_slock, flags);
- fc_vport_terminate(vha->fc_vport);
+ BUG_ON(base_vha->list.next == &ha->vp_list);
+ /* This assumes first entry in ha->vp_list is always base vha */
+ vha = list_first_entry(&base_vha->list, scsi_qla_host_t, list);
+ scsi_host = scsi_host_get(vha->host);
- spin_lock_irqsave(&ha->vport_slock, flags);
- }
+ spin_unlock_irqrestore(&ha->vport_slock, flags);
+ mutex_unlock(&ha->vport_lock);
- atomic_dec(&vha->vref_count);
+ fc_vport_terminate(vha->fc_vport);
+ scsi_host_put(vha->host);
+
+ mutex_lock(&ha->vport_lock);
}
- spin_unlock_irqrestore(&ha->vport_slock, flags);
+ mutex_unlock(&ha->vport_lock);
set_bit(UNLOADING, &base_vha->dpc_flags);
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/