Re: [PATCH 05/17] lpfc: Allow lpfc_plogi_confirm_nport logic to execute for Fabric nodes

[Himanshu Madhani <himanshu.madhani@oracle.com>]

On 1/30/24 16:35, Justin Tee wrote:
> Remove the early return NLP_FABRIC check in lpfc_plogi_confirm_nport
> because it is possible for switch domain controllers to change WWPN.
> 
> As a result, allow lpfc_plogi_confirm_nport to detect that a new ndlp
> should be initialized in such cases.  The old ndlp object will be cleaned
> up when dev_loss_tmo callbk executes.
> 
> Signed-off-by: Justin Tee <justin.tee@broadcom.com>
> ---
>   drivers/scsi/lpfc/lpfc_els.c | 49 +++++++++++++++++++++++-------------
>   1 file changed, 32 insertions(+), 17 deletions(-)
> 
> diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c
> index 4d723200690a..a17c66e31637 100644
> --- a/drivers/scsi/lpfc/lpfc_els.c
> +++ b/drivers/scsi/lpfc/lpfc_els.c
> @@ -1696,18 +1696,13 @@ lpfc_plogi_confirm_nport(struct lpfc_hba *phba, uint32_t *prsp,
>   	struct serv_parm *sp;
>   	uint8_t  name[sizeof(struct lpfc_name)];
>   	uint32_t keepDID = 0, keep_nlp_flag = 0;
> +	int rc;
>   	uint32_t keep_new_nlp_flag = 0;
>   	uint16_t keep_nlp_state;
>   	u32 keep_nlp_fc4_type = 0;
>   	struct lpfc_nvme_rport *keep_nrport = NULL;
>   	unsigned long *active_rrqs_xri_bitmap = NULL;
>   
> -	/* Fabric nodes can have the same WWPN so we don't bother searching
> -	 * by WWPN.  Just return the ndlp that was given to us.
> -	 */
> -	if (ndlp->nlp_type & NLP_FABRIC)
> -		return ndlp;
> -
>   	sp = (struct serv_parm *) ((uint8_t *) prsp + sizeof(uint32_t));
>   	memset(name, 0, sizeof(struct lpfc_name));
>   
> @@ -1717,15 +1712,9 @@ lpfc_plogi_confirm_nport(struct lpfc_hba *phba, uint32_t *prsp,
>   	new_ndlp = lpfc_findnode_wwpn(vport, &sp->portName);
>   
>   	/* return immediately if the WWPN matches ndlp */
> -	if (!new_ndlp || (new_ndlp == ndlp))
> +	if (new_ndlp == ndlp)
>   		return ndlp;
>   
> -	/*
> -	 * Unregister from backend if not done yet. Could have been skipped
> -	 * due to ADISC
> -	 */
> -	lpfc_nlp_unreg_node(vport, new_ndlp);
> -
>   	if (phba->sli_rev == LPFC_SLI_REV4) {
>   		active_rrqs_xri_bitmap = mempool_alloc(phba->active_rrq_pool,
>   						       GFP_KERNEL);
> @@ -1742,11 +1731,37 @@ lpfc_plogi_confirm_nport(struct lpfc_hba *phba, uint32_t *prsp,
>   			 (new_ndlp ? new_ndlp->nlp_flag : 0),
>   			 (new_ndlp ? new_ndlp->nlp_fc4_type : 0));
>   
> -	keepDID = new_ndlp->nlp_DID;
> +	if (!new_ndlp) {
> +		rc = memcmp(&ndlp->nlp_portname, name,
> +			    sizeof(struct lpfc_name));
> +		if (!rc) {
> +			if (active_rrqs_xri_bitmap)
> +				mempool_free(active_rrqs_xri_bitmap,
> +					     phba->active_rrq_pool);
> +			return ndlp;
> +		}
> +		new_ndlp = lpfc_nlp_init(vport, ndlp->nlp_DID);
> +		if (!new_ndlp) {
> +			if (active_rrqs_xri_bitmap)
> +				mempool_free(active_rrqs_xri_bitmap,
> +					     phba->active_rrq_pool);
> +			return ndlp;
> +		}
> +	} else {
> +		if (phba->sli_rev == LPFC_SLI_REV4 &&
> +		    active_rrqs_xri_bitmap)
> +			memcpy(active_rrqs_xri_bitmap,
> +			       new_ndlp->active_rrqs_xri_bitmap,
> +			       phba->cfg_rrq_xri_bitmap_sz);
>   
> -	if (phba->sli_rev == LPFC_SLI_REV4 && active_rrqs_xri_bitmap)
> -		memcpy(active_rrqs_xri_bitmap, new_ndlp->active_rrqs_xri_bitmap,
> -		       phba->cfg_rrq_xri_bitmap_sz);
> +		/*
> +		 * Unregister from backend if not done yet. Could have been
> +		 * skipped due to ADISC
> +		 */
> +		lpfc_nlp_unreg_node(vport, new_ndlp);
> +	}
> +
> +	keepDID = new_ndlp->nlp_DID;
>   
>   	/* At this point in this routine, we know new_ndlp will be
>   	 * returned. however, any previous GID_FTs that were done

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

-- 
Himanshu Madhani                                Oracle Linux Engineering