diff options
| author | James Smart <james.smart@emulex.com> | 2010-07-14 15:31:37 -0400 |
|---|---|---|
| committer | James Bottomley <James.Bottomley@suse.de> | 2010-07-28 09:05:41 -0500 |
| commit | 3804dc84b8c11038ef75d97fd11e43658f623665 (patch) | |
| tree | 03f6a03b3a489efb11ed40548a5b79b7da34034b /drivers/scsi/lpfc/lpfc_init.c | |
| parent | 589a52d6a97e01c5ff6c244ee6c8ea57726c610f (diff) | |
| download | kernel_samsung_smdk4412-3804dc84b8c11038ef75d97fd11e43658f623665.zip kernel_samsung_smdk4412-3804dc84b8c11038ef75d97fd11e43658f623665.tar.gz kernel_samsung_smdk4412-3804dc84b8c11038ef75d97fd11e43658f623665.tar.bz2 | |
[SCSI] lpfc 8.3.15: FCoE Related Fixes
FCoE Related Fixes
- Correct find-next-FCF routine so that it searches at next FCF rather
than current one.
- Enhanced round-robin FCF failover algorithm to re-start on "New FCF"
async event
- Update the manner in which we look at FCFs while they may be in
their discovery state.
- Use LPFC_FCOE_NULL_VID macro when checkinf for valid vlan_id for FCF
Signed-off-by: Alex Iannicelli <alex.iannicelli@emulex.com>
Signed-off-by: James Smart <james.smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Diffstat (limited to 'drivers/scsi/lpfc/lpfc_init.c')
| -rw-r--r-- | drivers/scsi/lpfc/lpfc_init.c | 22 |
1 files changed, 10 insertions, 12 deletions
diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index fb06933..2786ee3 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -3357,22 +3357,14 @@ lpfc_sli4_async_fcoe_evt(struct lpfc_hba *phba, "evt_tag:x%x, fcf_index:x%x\n", acqe_fcoe->event_tag, acqe_fcoe->index); + /* If the FCF discovery is in progress, do nothing. */ spin_lock_irq(&phba->hbalock); - if ((phba->fcf.fcf_flag & FCF_SCAN_DONE) || - (phba->hba_flag & FCF_DISC_INPROGRESS)) { - /* - * If the current FCF is in discovered state or - * FCF discovery is in progress, do nothing. - */ + if (phba->hba_flag & FCF_DISC_INPROGRESS) { spin_unlock_irq(&phba->hbalock); break; } - + /* If fast FCF failover rescan event is pending, do nothing */ if (phba->fcf.fcf_flag & FCF_REDISC_EVT) { - /* - * If fast FCF failover rescan event is pending, - * do nothing. - */ spin_unlock_irq(&phba->hbalock); break; } @@ -3393,7 +3385,13 @@ lpfc_sli4_async_fcoe_evt(struct lpfc_hba *phba, acqe_fcoe->index); rc = lpfc_sli4_read_fcf_rec(phba, acqe_fcoe->index); } - + /* If the FCF has been in discovered state, do nothing. */ + spin_lock_irq(&phba->hbalock); + if (phba->fcf.fcf_flag & FCF_SCAN_DONE) { + spin_unlock_irq(&phba->hbalock); + break; + } + spin_unlock_irq(&phba->hbalock); /* Otherwise, scan the entire FCF table and re-discover SAN */ lpfc_printf_log(phba, KERN_INFO, LOG_FIP | LOG_DISCOVERY, "2770 Start FCF table scan due to new FCF " |