[SCSI] libfc: avoid exchanges collision during lport reset

Currently timer delay is large and is using msleep to avoid
avoid exchanges collision across lport reset, so instead
do this by initializing exches pool indexes during
reset also.

Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com>
Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
This commit is contained in:
Vasu Dev
2011-10-28 11:34:17 -07:00
committed by James Bottomley
parent 14fc315fa3
commit b6e3c84034
2 changed files with 5 additions and 9 deletions

View File

@@ -1793,6 +1793,9 @@ restart:
goto restart;
}
}
pool->next_index = 0;
pool->left = FC_XID_UNKNOWN;
pool->right = FC_XID_UNKNOWN;
spin_unlock_bh(&pool->lock);
}
@@ -2281,6 +2284,7 @@ struct fc_exch_mgr *fc_exch_mgr_alloc(struct fc_lport *lport,
goto free_mempool;
for_each_possible_cpu(cpu) {
pool = per_cpu_ptr(mp->pool, cpu);
pool->next_index = 0;
pool->left = FC_XID_UNKNOWN;
pool->right = FC_XID_UNKNOWN;
spin_lock_init(&pool->lock);