2010-08-07 22:15:53

by Dan Carpenter

Subject: [patch] lpfc: change spin_lock_irq() to spin_lock()

In lpfc_cleanup_pending_mbox() we already have IRQs disabled so we don't
need to disable them again.

Also in lpfc_sli_intr_handler() there is a typo where it has
spin_unlock_irq() instead of just spin_unlock().

Signed-off-by: Dan Carpenter <[email protected]>

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index e758eae..f772a7e 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -8460,7 +8460,7 @@ lpfc_sli_intr_handler(int irq, void *dev_id)
* If there is deferred error attention, do not check for any interrupt.
*/
if (unlikely(phba->hba_flag & DEFER_ERATT)) {
- spin_unlock_irq(&phba->hbalock);
+ spin_unlock(&phba->hbalock);
return IRQ_NONE;
}

@@ -12813,9 +12813,9 @@ lpfc_cleanup_pending_mbox(struct lpfc_vport *vport)
}
ndlp = (struct lpfc_nodelist *) mb->context2;
if (ndlp) {
- spin_lock_irq(shost->host_lock);
+ spin_lock(shost->host_lock);
ndlp->nlp_flag &= ~NLP_IGNR_REG_CMPL;
- spin_unlock_irq(shost->host_lock);
+ spin_unlock(shost->host_lock);
lpfc_nlp_put(ndlp);
mb->context2 = NULL;
}
@@ -12831,9 +12831,9 @@ lpfc_cleanup_pending_mbox(struct lpfc_vport *vport)
if (mb->u.mb.mbxCommand == MBX_REG_LOGIN64) {
ndlp = (struct lpfc_nodelist *) mb->context2;
if (ndlp) {
- spin_lock_irq(shost->host_lock);
+ spin_lock(shost->host_lock);
ndlp->nlp_flag &= ~NLP_IGNR_REG_CMPL;
- spin_unlock_irq(shost->host_lock);
+ spin_unlock(shost->host_lock);
lpfc_nlp_put(ndlp);
mb->context2 = NULL;
}

2010-08-09 14:09:31

by James Smart

Subject: Re: [patch] lpfc: change spin_lock_irq() to spin_lock()

Dan Carpenter wrote:
> In lpfc_cleanup_pending_mbox() we already have IRQs disabled so we don't
> need to disable them again.
>
> Also in lpfc_sli_intr_handler() there is a typo where it has
> spin_unlock_irq() instead of just spin_unlock().
>
> Signed-off-by: Dan Carpenter <[email protected]>
>
> diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c

Acked-by: James Smart <[email protected]>

Thank You.

-- james s