From: James Smart Date: Tue, 23 Oct 2018 20:41:08 +0000 (-0700) Subject: scsi: lpfc: Fix odd recovery in duplicate FLOGIs in point-to-point X-Git-Url: http://git.cdn.openwrt.org/?a=commitdiff_plain;h=d496b9a7246cb9813da1fe49e14edbbbf8e232d5;p=openwrt%2Fstaging%2Fblogic.git scsi: lpfc: Fix odd recovery in duplicate FLOGIs in point-to-point Testing a point-to-point topology and a case of re-FLOGI without intervening link bouncing, showed an odd interaction with firmware and a resulting scenario where the driver no longer probed after accepting the new FLOGI. Work around the firmware issue by issuing a link bounce if a FLOGI is received after the link is already up and FLOGI's accepted. While debugging the issue, realized that some debug traces should be clarified to help in the future. Signed-off-by: Dick Kennedy Signed-off-by: James Smart Reviewed-by: Hannes Reinecke Signed-off-by: Martin K. Petersen --- diff --git a/drivers/scsi/lpfc/lpfc.h b/drivers/scsi/lpfc/lpfc.h index c1eb2b00ca7f..95f0cdbb16a6 100644 --- a/drivers/scsi/lpfc/lpfc.h +++ b/drivers/scsi/lpfc/lpfc.h @@ -490,6 +490,7 @@ struct lpfc_vport { struct nvme_fc_local_port *localport; uint8_t nvmei_support; /* driver supports NVME Initiator */ uint32_t last_fcp_wqidx; + uint32_t rcv_flogi_cnt; /* How many unsol FLOGIs ACK'd. */ }; struct hbq_s { diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c index 832e5e00c1c9..a200cdaf34a6 100644 --- a/drivers/scsi/lpfc/lpfc_els.c +++ b/drivers/scsi/lpfc/lpfc_els.c @@ -1057,9 +1057,9 @@ stop_rr_fcf_flogi: goto flogifail; lpfc_printf_vlog(vport, KERN_WARNING, LOG_ELS, - "0150 FLOGI failure Status:x%x/x%x TMO:x%x\n", + "0150 FLOGI failure Status:x%x/x%x xri x%x TMO:x%x\n", irsp->ulpStatus, irsp->un.ulpWord[4], - irsp->ulpTimeout); + cmdiocb->sli4_xritag, irsp->ulpTimeout); /* FLOGI failed, so there is no fabric */ spin_lock_irq(shost->host_lock); @@ -1113,7 +1113,8 @@ stop_rr_fcf_flogi: /* FLOGI completes successfully */ lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS, "0101 FLOGI completes successfully, I/O tag:x%x, " - "Data: x%x x%x x%x x%x x%x x%x\n", cmdiocb->iotag, + "xri x%x Data: x%x x%x x%x x%x x%x %x\n", + cmdiocb->iotag, cmdiocb->sli4_xritag, irsp->un.ulpWord[4], sp->cmn.e_d_tov, sp->cmn.w2.r_a_tov, sp->cmn.edtovResolution, vport->port_state, vport->fc_flag); @@ -4347,14 +4348,6 @@ lpfc_els_rsp_acc(struct lpfc_vport *vport, uint32_t flag, default: return 1; } - /* Xmit ELS ACC response tag */ - lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS, - "0128 Xmit ELS ACC response tag x%x, XRI: x%x, " - "DID: x%x, nlp_flag: x%x nlp_state: x%x RPI: x%x " - "fc_flag x%x\n", - elsiocb->iotag, elsiocb->iocb.ulpContext, - ndlp->nlp_DID, ndlp->nlp_flag, ndlp->nlp_state, - ndlp->nlp_rpi, vport->fc_flag); if (ndlp->nlp_flag & NLP_LOGO_ACC) { spin_lock_irq(shost->host_lock); if (!(ndlp->nlp_flag & NLP_RPI_REGISTERED || @@ -4523,6 +4516,15 @@ lpfc_els_rsp_adisc_acc(struct lpfc_vport *vport, struct lpfc_iocbq *oldiocb, lpfc_els_free_iocb(phba, elsiocb); return 1; } + + /* Xmit ELS ACC response tag */ + lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS, + "0128 Xmit ELS ACC response Status: x%x, IoTag: x%x, " + "XRI: x%x, DID: x%x, nlp_flag: x%x nlp_state: x%x " + "RPI: x%x, fc_flag x%x\n", + rc, elsiocb->iotag, elsiocb->sli4_xritag, + ndlp->nlp_DID, ndlp->nlp_flag, ndlp->nlp_state, + ndlp->nlp_rpi, vport->fc_flag); return 0; } @@ -6533,6 +6535,11 @@ lpfc_els_rcv_flogi(struct lpfc_vport *vport, struct lpfc_iocbq *cmdiocb, port_state = vport->port_state; vport->fc_flag |= FC_PT2PT; vport->fc_flag &= ~(FC_FABRIC | FC_PUBLIC_LOOP); + + /* Acking an unsol FLOGI. Count 1 for link bounce + * work-around. + */ + vport->rcv_flogi_cnt++; spin_unlock_irq(shost->host_lock); lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS, "3311 Rcv Flogi PS x%x new PS x%x " @@ -7930,8 +7937,9 @@ lpfc_els_unsol_buffer(struct lpfc_hba *phba, struct lpfc_sli_ring *pring, struct ls_rjt stat; uint32_t *payload; uint32_t cmd, did, newnode; - uint8_t rjt_exp, rjt_err = 0; + uint8_t rjt_exp, rjt_err = 0, init_link = 0; IOCB_t *icmd = &elsiocb->iocb; + LPFC_MBOXQ_t *mbox; if (!vport || !(elsiocb->context2)) goto dropit; @@ -8080,6 +8088,19 @@ lpfc_els_unsol_buffer(struct lpfc_hba *phba, struct lpfc_sli_ring *pring, did, vport->port_state, ndlp->nlp_flag); phba->fc_stat.elsRcvFLOGI++; + + /* If the driver believes fabric discovery is done and is ready, + * bounce the link. There is some descrepancy. + */ + if (vport->port_state >= LPFC_LOCAL_CFG_LINK && + vport->fc_flag & FC_PT2PT && + vport->rcv_flogi_cnt >= 1) { + rjt_err = LSRJT_LOGICAL_BSY; + rjt_exp = LSEXP_NOTHING_MORE; + init_link++; + goto lsrjt; + } + lpfc_els_rcv_flogi(vport, elsiocb, ndlp); if (newnode) lpfc_nlp_put(ndlp); @@ -8308,6 +8329,27 @@ lsrjt: lpfc_nlp_put(elsiocb->context1); elsiocb->context1 = NULL; + + /* Special case. Driver received an unsolicited command that + * unsupportable given the driver's current state. Reset the + * link and start over. + */ + if (init_link) { + mbox = mempool_alloc(phba->mbox_mem_pool, GFP_KERNEL); + if (!mbox) + return; + lpfc_linkdown(phba); + lpfc_init_link(phba, mbox, + phba->cfg_topology, + phba->cfg_link_speed); + mbox->u.mb.un.varInitLnk.lipsr_AL_PA = 0; + mbox->mbox_cmpl = lpfc_sli_def_mbox_cmpl; + mbox->vport = vport; + if (lpfc_sli_issue_mbox(phba, mbox, MBX_NOWAIT) == + MBX_NOT_FINISHED) + mempool_free(mbox, phba->mbox_mem_pool); + } + return; dropit: diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c index a26db7e1d821..bfc4ac8fc426 100644 --- a/drivers/scsi/lpfc/lpfc_hbadisc.c +++ b/drivers/scsi/lpfc/lpfc_hbadisc.c @@ -947,6 +947,7 @@ lpfc_linkdown(struct lpfc_hba *phba) } spin_lock_irq(shost->host_lock); phba->pport->fc_flag &= ~(FC_PT2PT | FC_PT2PT_PLOGI); + phba->pport->rcv_flogi_cnt = 0; spin_unlock_irq(shost->host_lock); } return 0; @@ -1018,6 +1019,7 @@ lpfc_linkup(struct lpfc_hba *phba) { struct lpfc_vport **vports; int i; + struct Scsi_Host *shost = lpfc_shost_from_vport(phba->pport); phba->link_state = LPFC_LINK_UP; @@ -1031,6 +1033,13 @@ lpfc_linkup(struct lpfc_hba *phba) lpfc_linkup_port(vports[i]); lpfc_destroy_vport_work_array(phba, vports); + /* Clear the pport flogi counter in case the link down was + * absorbed without an ACQE. No lock here - in worker thread + * and discovery is synchronized. + */ + spin_lock_irq(shost->host_lock); + phba->pport->rcv_flogi_cnt = 0; + spin_unlock_irq(shost->host_lock); return 0; }