From 65a3df63e7ff5addafc75ad8bc5ef5856db55429 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:58:48 -0700
Subject: scsi: lpfc: Fix premature re-enabling of interrupts in
 lpfc_sli_host_down

Use of spin_lock_irq may re-enable interrupts prematurely.

Convert to spin_lock. Note: code is under the phba->hba_lock which has been
locked with irqsave.

Link: https://lore.kernel.org/r/20190922035906.10977-3-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index a0c6945b8139..3b3ae2182e59 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -10678,14 +10678,14 @@ lpfc_sli_host_down(struct lpfc_vport *vport)
 				set_bit(LPFC_DATA_READY, &phba->data_flags);
 			}
 			prev_pring_flag = pring->flag;
-			spin_lock_irq(&pring->ring_lock);
+			spin_lock(&pring->ring_lock);
 			list_for_each_entry_safe(iocb, next_iocb,
 						 &pring->txq, list) {
 				if (iocb->vport != vport)
 					continue;
 				list_move_tail(&iocb->list, &completions);
 			}
-			spin_unlock_irq(&pring->ring_lock);
+			spin_unlock(&pring->ring_lock);
 			list_for_each_entry_safe(iocb, next_iocb,
 						 &pring->txcmplq, list) {
 				if (iocb->vport != vport)
-- 
cgit v1.2.3-59-g8ed1b


From a5f7337f5a82fc4b13b4481a7e56977656cbe7d1 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:58:50 -0700
Subject: scsi: lpfc: Fix NVME io abort failures causing hangs

The nvme-fc transport may call to abort an io on controller reset. If the
driver is out of resources to issue an abort command, it just gives up and
does nothing. The transport expects the lldd to always be able to terminate
an io it has issued.  At that point, the controller hangs waiting for
aborted ios to be returned.  Note: flaged by "6136" and "6176" error
messages.

Root issue was the adapter mis-allocated the number resources it allocated
for command entries for the adapter.

Convert the driver to allocate command resources based on the number of
xris supported by the FC port - 1 resource for the original command and 1
resource for the abort request.

Link: https://lore.kernel.org/r/20190922035906.10977-5-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc.h      |  1 -
 drivers/scsi/lpfc/lpfc_attr.c |  5 -----
 drivers/scsi/lpfc/lpfc_init.c |  9 +++------
 drivers/scsi/lpfc/lpfc_sli.c  | 17 +++++++++++------
 4 files changed, 14 insertions(+), 18 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc.h b/drivers/scsi/lpfc/lpfc.h
index 691acbdcc46d..291335e16806 100644
--- a/drivers/scsi/lpfc/lpfc.h
+++ b/drivers/scsi/lpfc/lpfc.h
@@ -872,7 +872,6 @@ struct lpfc_hba {
 	uint32_t cfg_aer_support;
 	uint32_t cfg_sriov_nr_virtfn;
 	uint32_t cfg_request_firmware_upgrade;
-	uint32_t cfg_iocb_cnt;
 	uint32_t cfg_suppress_link_up;
 	uint32_t cfg_rrq_xri_bitmap_sz;
 	uint32_t cfg_delay_discovery;
diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c
index 41cc91d3c77e..79a192479755 100644
--- a/drivers/scsi/lpfc/lpfc_attr.c
+++ b/drivers/scsi/lpfc/lpfc_attr.c
@@ -3582,9 +3582,6 @@ lpfc_txcmplq_hw_show(struct device *dev, struct device_attribute *attr,
 static DEVICE_ATTR(txcmplq_hw, S_IRUGO,
 			 lpfc_txcmplq_hw_show, NULL);
 
-LPFC_ATTR_R(iocb_cnt, 2, 1, 5,
-	"Number of IOCBs alloc for ELS, CT, and ABTS: 1k to 5k IOCBs");
-
 /*
 # lpfc_nodev_tmo: If set, it will hold all I/O errors on devices that disappear
 # until the timer expires. Value range is [0,255]. Default value is 30.
@@ -6073,7 +6070,6 @@ struct device_attribute *lpfc_hba_attrs[] = {
 	&dev_attr_lpfc_sriov_nr_virtfn,
 	&dev_attr_lpfc_req_fw_upgrade,
 	&dev_attr_lpfc_suppress_link_up,
-	&dev_attr_lpfc_iocb_cnt,
 	&dev_attr_iocb_hw,
 	&dev_attr_txq_hw,
 	&dev_attr_txcmplq_hw,
@@ -7212,7 +7208,6 @@ lpfc_get_cfgparam(struct lpfc_hba *phba)
 	lpfc_sriov_nr_virtfn_init(phba, lpfc_sriov_nr_virtfn);
 	lpfc_request_firmware_upgrade_init(phba, lpfc_req_fw_upgrade);
 	lpfc_suppress_link_up_init(phba, lpfc_suppress_link_up);
-	lpfc_iocb_cnt_init(phba, lpfc_iocb_cnt);
 	lpfc_delay_discovery_init(phba, lpfc_delay_discovery);
 	lpfc_sli_mode_init(phba, lpfc_sli_mode);
 	phba->cfg_enable_dss = 1;
diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c
index e91377a4cafe..bb84d2a20e76 100644
--- a/drivers/scsi/lpfc/lpfc_init.c
+++ b/drivers/scsi/lpfc/lpfc_init.c
@@ -7126,7 +7126,7 @@ lpfc_init_iocb_list(struct lpfc_hba *phba, int iocb_count)
 		if (iocbq_entry == NULL) {
 			printk(KERN_ERR "%s: only allocated %d iocbs of "
 				"expected %d count. Unloading driver.\n",
-				__func__, i, LPFC_IOCB_LIST_CNT);
+				__func__, i, iocb_count);
 			goto out_free_iocbq;
 		}
 
@@ -11591,13 +11591,10 @@ fcponly:
 	}
 
 	/* If the NVME FC4 type is enabled, scale the sg_seg_cnt to
-	 * accommodate 512K and 1M IOs in a single nvme buf and supply
-	 * enough NVME LS iocb buffers for larger connectivity counts.
+	 * accommodate 512K and 1M IOs in a single nvme buf.
 	 */
-	if (phba->cfg_enable_fc4_type & LPFC_ENABLE_NVME) {
+	if (phba->cfg_enable_fc4_type & LPFC_ENABLE_NVME)
 		phba->cfg_sg_seg_cnt = LPFC_MAX_NVME_SEG_CNT;
-		phba->cfg_iocb_cnt = 5;
-	}
 
 	/* Only embed PBDE for if_type 6, PBDE support requires xib be set */
 	if ((bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) !=
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 3b3ae2182e59..827406f58b1e 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -7523,9 +7523,11 @@ lpfc_sli4_hba_setup(struct lpfc_hba *phba)
 		}
 		phba->sli4_hba.nvmet_xri_cnt = rc;
 
-		cnt = phba->cfg_iocb_cnt * 1024;
-		/* We need 1 iocbq for every SGL, for IO processing */
-		cnt += phba->sli4_hba.nvmet_xri_cnt;
+		/* We allocate an iocbq for every receive context SGL.
+		 * The additional allocation is for abort and ls handling.
+		 */
+		cnt = phba->sli4_hba.nvmet_xri_cnt +
+			phba->sli4_hba.max_cfg_param.max_xri;
 	} else {
 		/* update host common xri-sgl sizes and mappings */
 		rc = lpfc_sli4_io_sgl_update(phba);
@@ -7547,14 +7549,17 @@ lpfc_sli4_hba_setup(struct lpfc_hba *phba)
 			rc = -ENODEV;
 			goto out_destroy_queue;
 		}
-		cnt = phba->cfg_iocb_cnt * 1024;
+		/* Each lpfc_io_buf job structure has an iocbq element.
+		 * This cnt provides for abort, els, ct and ls requests.
+		 */
+		cnt = phba->sli4_hba.max_cfg_param.max_xri;
 	}
 
 	if (!phba->sli.iocbq_lookup) {
 		/* Initialize and populate the iocb list per host */
 		lpfc_printf_log(phba, KERN_INFO, LOG_INIT,
-				"2821 initialize iocb list %d total %d\n",
-				phba->cfg_iocb_cnt, cnt);
+				"2821 initialize iocb list with %d entries\n",
+				cnt);
 		rc = lpfc_init_iocb_list(phba, cnt);
 		if (rc) {
 			lpfc_printf_log(phba, KERN_ERR, LOG_INIT,
-- 
cgit v1.2.3-59-g8ed1b


From 97acd0019d5dadd9c0e111c2083c889bfe548f25 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:58:51 -0700
Subject: scsi: lpfc: Fix rpi release when deleting vport

A prior use-after-free mailbox fix solved it's problem by null'ing a ndlp
pointer.  However, further testing has shown that this change causes a
later state change to occasionally be skipped, which results in a reference
count never being decremented thus the rpi is never released, which causes
a vport delete to never succeed.

Revise the fix in the prior patch to no longer null the ndlp. Instead the
RELEASE_RPI flag is set which will drive the release of the rpi.

Given the new code was added at a deep indentation level, refactor the code
block using a new routine that avoids the indentation issues.

Fixes: 	9b1640686470 ("scsi: lpfc: Fix use-after-free mailbox cmd completion")
Link: https://lore.kernel.org/r/20190922035906.10977-6-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_hbadisc.c | 88 +++++++++++++++++++++++++++-------------
 drivers/scsi/lpfc/lpfc_sli.c     |  2 +
 2 files changed, 61 insertions(+), 29 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c
index 749286acdc17..9df6f0cabab0 100644
--- a/drivers/scsi/lpfc/lpfc_hbadisc.c
+++ b/drivers/scsi/lpfc/lpfc_hbadisc.c
@@ -4840,6 +4840,44 @@ lpfc_nlp_logo_unreg(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 	}
 }
 
+/*
+ * Sets the mailbox completion handler to be used for the
+ * unreg_rpi command. The handler varies based on the state of
+ * the port and what will be happening to the rpi next.
+ */
+static void
+lpfc_set_unreg_login_mbx_cmpl(struct lpfc_hba *phba, struct lpfc_vport *vport,
+	struct lpfc_nodelist *ndlp, LPFC_MBOXQ_t *mbox)
+{
+	unsigned long iflags;
+
+	if (ndlp->nlp_flag & NLP_ISSUE_LOGO) {
+		mbox->ctx_ndlp = ndlp;
+		mbox->mbox_cmpl = lpfc_nlp_logo_unreg;
+
+	} else if (phba->sli_rev == LPFC_SLI_REV4 &&
+		   (!(vport->load_flag & FC_UNLOADING)) &&
+		    (bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) >=
+				      LPFC_SLI_INTF_IF_TYPE_2) &&
+		    (kref_read(&ndlp->kref) > 0)) {
+		mbox->ctx_ndlp = lpfc_nlp_get(ndlp);
+		mbox->mbox_cmpl = lpfc_sli4_unreg_rpi_cmpl_clr;
+	} else {
+		if (vport->load_flag & FC_UNLOADING) {
+			if (phba->sli_rev == LPFC_SLI_REV4) {
+				spin_lock_irqsave(&vport->phba->ndlp_lock,
+						  iflags);
+				ndlp->nlp_flag |= NLP_RELEASE_RPI;
+				spin_unlock_irqrestore(&vport->phba->ndlp_lock,
+						       iflags);
+			}
+			lpfc_nlp_get(ndlp);
+		}
+		mbox->ctx_ndlp = ndlp;
+		mbox->mbox_cmpl = lpfc_sli_def_mbox_cmpl;
+	}
+}
+
 /*
  * Free rpi associated with LPFC_NODELIST entry.
  * This routine is called from lpfc_freenode(), when we are removing
@@ -4890,33 +4928,12 @@ lpfc_unreg_rpi(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
 
 			lpfc_unreg_login(phba, vport->vpi, rpi, mbox);
 			mbox->vport = vport;
-			if (ndlp->nlp_flag & NLP_ISSUE_LOGO) {
-				mbox->ctx_ndlp = ndlp;
-				mbox->mbox_cmpl = lpfc_nlp_logo_unreg;
-			} else {
-				if (phba->sli_rev == LPFC_SLI_REV4 &&
-				    (!(vport->load_flag & FC_UNLOADING)) &&
-				    (bf_get(lpfc_sli_intf_if_type,
-				     &phba->sli4_hba.sli_intf) >=
-				      LPFC_SLI_INTF_IF_TYPE_2) &&
-				    (kref_read(&ndlp->kref) > 0)) {
-					mbox->ctx_ndlp = lpfc_nlp_get(ndlp);
-					mbox->mbox_cmpl =
-						lpfc_sli4_unreg_rpi_cmpl_clr;
-					/*
-					 * accept PLOGIs after unreg_rpi_cmpl
-					 */
-					acc_plogi = 0;
-				} else if (vport->load_flag & FC_UNLOADING) {
-					mbox->ctx_ndlp = NULL;
-					mbox->mbox_cmpl =
-						lpfc_sli_def_mbox_cmpl;
-				} else {
-					mbox->ctx_ndlp = ndlp;
-					mbox->mbox_cmpl =
-						lpfc_sli_def_mbox_cmpl;
-				}
-			}
+			lpfc_set_unreg_login_mbx_cmpl(phba, vport, ndlp, mbox);
+			if (mbox->mbox_cmpl == lpfc_sli4_unreg_rpi_cmpl_clr)
+				/*
+				 * accept PLOGIs after unreg_rpi_cmpl
+				 */
+				acc_plogi = 0;
 			if (((ndlp->nlp_DID & Fabric_DID_MASK) !=
 			    Fabric_DID_MASK) &&
 			    (!(vport->fc_flag & FC_OFFLINE_MODE)))
@@ -5057,6 +5074,7 @@ lpfc_cleanup_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
 	struct lpfc_hba  *phba = vport->phba;
 	LPFC_MBOXQ_t *mb, *nextmb;
 	struct lpfc_dmabuf *mp;
+	unsigned long iflags;
 
 	/* Cleanup node for NPort <nlp_DID> */
 	lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE,
@@ -5138,8 +5156,20 @@ lpfc_cleanup_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
 	lpfc_cleanup_vports_rrqs(vport, ndlp);
 	if (phba->sli_rev == LPFC_SLI_REV4)
 		ndlp->nlp_flag |= NLP_RELEASE_RPI;
-	lpfc_unreg_rpi(vport, ndlp);
-
+	if (!lpfc_unreg_rpi(vport, ndlp)) {
+		/* Clean up unregistered and non freed rpis */
+		if ((ndlp->nlp_flag & NLP_RELEASE_RPI) &&
+		    !(ndlp->nlp_rpi == LPFC_RPI_ALLOC_ERROR)) {
+			lpfc_sli4_free_rpi(vport->phba,
+					   ndlp->nlp_rpi);
+			spin_lock_irqsave(&vport->phba->ndlp_lock,
+					  iflags);
+			ndlp->nlp_flag &= ~NLP_RELEASE_RPI;
+			ndlp->nlp_rpi = LPFC_RPI_ALLOC_ERROR;
+			spin_unlock_irqrestore(&vport->phba->ndlp_lock,
+					       iflags);
+		}
+	}
 	return 0;
 }
 
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 827406f58b1e..f764012ba0a6 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -2526,6 +2526,8 @@ lpfc_sli_def_mbox_cmpl(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 			} else {
 				__lpfc_sli_rpi_release(vport, ndlp);
 			}
+			if (vport->load_flag & FC_UNLOADING)
+				lpfc_nlp_put(ndlp);
 			pmb->ctx_ndlp = NULL;
 		}
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 0f154226d699fefe651ccc4db773efc05a820b56 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:58:52 -0700
Subject: scsi: lpfc: Fix device recovery errors after PLOGI failures

When target-side fault injections are made, the driver isn't reconnecting
to the remote port. The driver is logging "2753" error messages which
state:

"PLOGI failure DID:1B2400 Status:x3/xf0240008"

The failures status is indicating a Illegal field error, which points to
the Temporary RPI field being used for the ELS. This error typically means
the driver used an RPI that was already registered (shouldn't be registered
if using it in this context).

Study has found that if the driver were in discovery attempts and
encountered an error, it wouldn't flag the temporary rpi in error.  Yet the
rpi was released for reallocation in these error paths and another ELS
could allocate the rpi. In the failure situation a retry was done on an ELS
that had encountered an error, and as the rpi wasn't marked in error, the
ELS reused the rpi it originally allocated. But that rpi had been allocated
by a different ELS issued after the original error and before the retry
attempt. The different ELS had succeeded and the RPI was registered.

Fix by marking the rpi state for the node to be in error, aka as needing
reallocation, upon an error in the els processing.  Error state marking is
always done prior to release back to the internal rpi free list, which the
driver wasn't doing in cases prior.

Also enhanced some of the logging to help in the next case of problem
troubleshooting.

Link: https://lore.kernel.org/r/20190922035906.10977-7-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_hbadisc.c | 44 ++++++++++++++++++++++++----------------
 drivers/scsi/lpfc/lpfc_init.c    | 40 ++++++++++++++++++++----------------
 drivers/scsi/lpfc/lpfc_sli.c     |  8 +++++---
 3 files changed, 55 insertions(+), 37 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c
index 9df6f0cabab0..144786947b63 100644
--- a/drivers/scsi/lpfc/lpfc_hbadisc.c
+++ b/drivers/scsi/lpfc/lpfc_hbadisc.c
@@ -4046,7 +4046,7 @@ out:
 	ndlp->nlp_flag |= NLP_RPI_REGISTERED;
 	ndlp->nlp_type |= NLP_FABRIC;
 	lpfc_nlp_set_state(vport, ndlp, NLP_STE_UNMAPPED_NODE);
-	lpfc_printf_vlog(vport, KERN_INFO, LOG_SLI,
+	lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE | LOG_DISCOVERY,
 			 "0003 rpi:%x DID:%x flg:%x %d map%x x%px\n",
 			 ndlp->nlp_rpi, ndlp->nlp_DID, ndlp->nlp_flag,
 			 kref_read(&ndlp->kref),
@@ -4575,8 +4575,10 @@ lpfc_enable_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp,
 	return ndlp;
 
 free_rpi:
-	if (phba->sli_rev == LPFC_SLI_REV4)
+	if (phba->sli_rev == LPFC_SLI_REV4) {
 		lpfc_sli4_free_rpi(vport->phba, rpi);
+		ndlp->nlp_rpi = LPFC_RPI_ALLOC_ERROR;
+	}
 	return NULL;
 }
 
@@ -4835,6 +4837,7 @@ lpfc_nlp_logo_unreg(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 		if (ndlp->nlp_flag & NLP_RELEASE_RPI) {
 			lpfc_sli4_free_rpi(vport->phba, ndlp->nlp_rpi);
 			ndlp->nlp_flag &= ~NLP_RELEASE_RPI;
+			ndlp->nlp_rpi = LPFC_RPI_ALLOC_ERROR;
 		}
 		ndlp->nlp_flag &= ~NLP_UNREG_INP;
 	}
@@ -4898,7 +4901,8 @@ lpfc_unreg_rpi(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
 	if (ndlp->nlp_flag & NLP_RPI_REGISTERED ||
 	    ndlp->nlp_flag & NLP_REG_LOGIN_SEND) {
 		if (ndlp->nlp_flag & NLP_REG_LOGIN_SEND)
-			lpfc_printf_vlog(vport, KERN_INFO, LOG_SLI,
+			lpfc_printf_vlog(vport, KERN_INFO,
+					 LOG_NODE | LOG_DISCOVERY,
 					 "3366 RPI x%x needs to be "
 					 "unregistered nlp_flag x%x "
 					 "did x%x\n",
@@ -4909,7 +4913,8 @@ lpfc_unreg_rpi(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
 		 * no need to queue up another one.
 		 */
 		if (ndlp->nlp_flag & NLP_UNREG_INP) {
-			lpfc_printf_vlog(vport, KERN_INFO, LOG_DISCOVERY,
+			lpfc_printf_vlog(vport, KERN_INFO,
+					 LOG_NODE | LOG_DISCOVERY,
 					 "1436 unreg_rpi SKIP UNREG x%x on "
 					 "NPort x%x deferred x%x  flg x%x "
 					 "Data: x%px\n",
@@ -4939,7 +4944,8 @@ lpfc_unreg_rpi(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
 			    (!(vport->fc_flag & FC_OFFLINE_MODE)))
 				ndlp->nlp_flag |= NLP_UNREG_INP;
 
-			lpfc_printf_vlog(vport, KERN_INFO, LOG_DISCOVERY,
+			lpfc_printf_vlog(vport, KERN_INFO,
+					 LOG_NODE | LOG_DISCOVERY,
 					 "1433 unreg_rpi UNREG x%x on "
 					 "NPort x%x deferred flg x%x "
 					 "Data:x%px\n",
@@ -5195,8 +5201,10 @@ lpfc_nlp_remove(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
 		/* For this case we need to cleanup the default rpi
 		 * allocated by the firmware.
 		 */
-		lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE,
-				 "0005 rpi:%x DID:%x flg:%x %d map:%x x%px\n",
+		lpfc_printf_vlog(vport, KERN_INFO,
+				 LOG_NODE | LOG_DISCOVERY,
+				 "0005 Cleanup Default rpi:x%x DID:x%x flg:x%x "
+				 "ref %d map:x%x ndlp x%px\n",
 				 ndlp->nlp_rpi, ndlp->nlp_DID, ndlp->nlp_flag,
 				 kref_read(&ndlp->kref),
 				 ndlp->nlp_usg_map, ndlp);
@@ -5233,8 +5241,9 @@ lpfc_nlp_remove(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
 		 */
 		lpfc_printf_vlog(vport, KERN_WARNING, LOG_NODE,
 				"0940 removed node x%px DID x%x "
-				" rport not null x%px\n",
-				ndlp, ndlp->nlp_DID, ndlp->rport);
+				"rpi %d rport not null x%px\n",
+				 ndlp, ndlp->nlp_DID, ndlp->nlp_rpi,
+				 ndlp->rport);
 		rport = ndlp->rport;
 		rdata = rport->dd_data;
 		rdata->pnode = NULL;
@@ -6026,7 +6035,7 @@ lpfc_mbx_cmpl_fdmi_reg_login(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 	ndlp->nlp_flag |= NLP_RPI_REGISTERED;
 	ndlp->nlp_type |= NLP_FABRIC;
 	lpfc_nlp_set_state(vport, ndlp, NLP_STE_UNMAPPED_NODE);
-	lpfc_printf_vlog(vport, KERN_INFO, LOG_SLI,
+	lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE | LOG_DISCOVERY,
 			 "0004 rpi:%x DID:%x flg:%x %d map:%x x%px\n",
 			 ndlp->nlp_rpi, ndlp->nlp_DID, ndlp->nlp_flag,
 			 kref_read(&ndlp->kref),
@@ -6215,12 +6224,12 @@ lpfc_nlp_init(struct lpfc_vport *vport, uint32_t did)
 	INIT_LIST_HEAD(&ndlp->nlp_listp);
 	if (vport->phba->sli_rev == LPFC_SLI_REV4) {
 		ndlp->nlp_rpi = rpi;
-		lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE,
-				 "0007 rpi:%x DID:%x flg:%x refcnt:%d "
-				 "map:%x x%px\n", ndlp->nlp_rpi, ndlp->nlp_DID,
-				 ndlp->nlp_flag,
-				 kref_read(&ndlp->kref),
-				 ndlp->nlp_usg_map, ndlp);
+		lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE | LOG_DISCOVERY,
+				 "0007 Init New ndlp x%px, rpi:x%x DID:%x "
+				 "flg:x%x refcnt:%d map:x%x\n",
+				 ndlp, ndlp->nlp_rpi, ndlp->nlp_DID,
+				 ndlp->nlp_flag, kref_read(&ndlp->kref),
+				 ndlp->nlp_usg_map);
 
 		ndlp->active_rrqs_xri_bitmap =
 				mempool_alloc(vport->phba->active_rrq_pool,
@@ -6449,7 +6458,8 @@ lpfc_fcf_inuse(struct lpfc_hba *phba)
 				goto out;
 			} else if (ndlp->nlp_flag & NLP_RPI_REGISTERED) {
 				ret = 1;
-				lpfc_printf_log(phba, KERN_INFO, LOG_ELS,
+				lpfc_printf_log(phba, KERN_INFO,
+						LOG_NODE | LOG_DISCOVERY,
 						"2624 RPI %x DID %x flag %x "
 						"still logged in\n",
 						ndlp->nlp_rpi, ndlp->nlp_DID,
diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c
index bb84d2a20e76..12885b01fa27 100644
--- a/drivers/scsi/lpfc/lpfc_init.c
+++ b/drivers/scsi/lpfc/lpfc_init.c
@@ -3053,11 +3053,12 @@ lpfc_sli4_node_prep(struct lpfc_hba *phba)
 				continue;
 			}
 			ndlp->nlp_rpi = rpi;
-			lpfc_printf_vlog(ndlp->vport, KERN_INFO, LOG_NODE,
-					 "0009 rpi:%x DID:%x "
-					 "flg:%x map:%x x%px\n", ndlp->nlp_rpi,
-					 ndlp->nlp_DID, ndlp->nlp_flag,
-					 ndlp->nlp_usg_map, ndlp);
+			lpfc_printf_vlog(ndlp->vport, KERN_INFO,
+					 LOG_NODE | LOG_DISCOVERY,
+					 "0009 Assign RPI x%x to ndlp x%px "
+					 "DID:x%06x flg:x%x map:x%x\n",
+					 ndlp->nlp_rpi, ndlp, ndlp->nlp_DID,
+					 ndlp->nlp_flag, ndlp->nlp_usg_map);
 		}
 	}
 	lpfc_destroy_vport_work_array(phba, vports);
@@ -3453,10 +3454,15 @@ lpfc_offline_prep(struct lpfc_hba *phba, int mbx_action)
 			list_for_each_entry_safe(ndlp, next_ndlp,
 						 &vports[i]->fc_nodes,
 						 nlp_listp) {
-				if (!NLP_CHK_NODE_ACT(ndlp))
-					continue;
-				if (ndlp->nlp_state == NLP_STE_UNUSED_NODE)
+				if ((!NLP_CHK_NODE_ACT(ndlp)) ||
+				    ndlp->nlp_state == NLP_STE_UNUSED_NODE) {
+					/* Driver must assume RPI is invalid for
+					 * any unused or inactive node.
+					 */
+					ndlp->nlp_rpi = LPFC_RPI_ALLOC_ERROR;
 					continue;
+				}
+
 				if (ndlp->nlp_type & NLP_FABRIC) {
 					lpfc_disc_state_machine(vports[i], ndlp,
 						NULL, NLP_EVT_DEVICE_RECOVERY);
@@ -3472,16 +3478,16 @@ lpfc_offline_prep(struct lpfc_hba *phba, int mbx_action)
 				 * comes back online.
 				 */
 				if (phba->sli_rev == LPFC_SLI_REV4) {
-					lpfc_printf_vlog(ndlp->vport,
-							 KERN_INFO, LOG_NODE,
-							 "0011 lpfc_offline: "
-							 "ndlp:x%px did %x "
-							 "usgmap:x%x rpi:%x\n",
-							 ndlp, ndlp->nlp_DID,
-							 ndlp->nlp_usg_map,
-							 ndlp->nlp_rpi);
-
+					lpfc_printf_vlog(ndlp->vport, KERN_INFO,
+						 LOG_NODE | LOG_DISCOVERY,
+						 "0011 Free RPI x%x on "
+						 "ndlp:x%px did x%x "
+						 "usgmap:x%x\n",
+						 ndlp->nlp_rpi, ndlp,
+						 ndlp->nlp_DID,
+						 ndlp->nlp_usg_map);
 					lpfc_sli4_free_rpi(phba, ndlp->nlp_rpi);
+					ndlp->nlp_rpi = LPFC_RPI_ALLOC_ERROR;
 				}
 				lpfc_unreg_rpi(vports[i], ndlp);
 			}
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index f764012ba0a6..24d6779a99f8 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -18131,8 +18131,9 @@ lpfc_sli4_alloc_rpi(struct lpfc_hba *phba)
 		phba->sli4_hba.max_cfg_param.rpi_used++;
 		phba->sli4_hba.rpi_count++;
 	}
-	lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
-			"0001 rpi:%x max:%x lim:%x\n",
+	lpfc_printf_log(phba, KERN_INFO,
+			LOG_NODE | LOG_DISCOVERY,
+			"0001 Allocated rpi:x%x max:x%x lim:x%x\n",
 			(int) rpi, max_rpi, rpi_limit);
 
 	/*
@@ -18192,7 +18193,8 @@ __lpfc_sli4_free_rpi(struct lpfc_hba *phba, int rpi)
 		phba->sli4_hba.rpi_count--;
 		phba->sli4_hba.max_cfg_param.rpi_used--;
 	} else {
-		lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
+		lpfc_printf_log(phba, KERN_INFO,
+				LOG_NODE | LOG_DISCOVERY,
 				"2016 rpi %x not inuse\n",
 				rpi);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 07b8582430370097238b589f4e24da7613ca6dd3 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:58:53 -0700
Subject: scsi: lpfc: Fix locking on mailbox command completion

Symptoms were seen of the driver not having valid data for mailbox
commands. After debugging, the following sequence was found:

The driver maintains a port-wide pointer of the mailbox command that is
currently in execution. Once finished, the port-wide pointer is cleared
(done in lpfc_sli4_mq_release()). The next mailbox command issued will set
the next pointer and so on.

The mailbox response data is only copied if there is a valid port-wide
pointer.

In the failing case, it was seen that a new mailbox command was being
attempted in parallel with the completion.  The parallel path was seeing
the mailbox no long in use (flag check under lock) and thus set the port
pointer.  The completion path had cleared the active flag under lock, but
had not touched the port pointer.  The port pointer is cleared after the
lock is released. In this case, the completion path cleared the just-set
value by the parallel path.

Fix by making the calls that clear mbox state/port pointer while under
lock.  Also slightly cleaned up the error path.

Link: https://lore.kernel.org/r/20190922035906.10977-8-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 24d6779a99f8..313441a3c4cf 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -13165,13 +13165,19 @@ send_current_mbox:
 	phba->sli.sli_flag &= ~LPFC_SLI_MBOX_ACTIVE;
 	/* Setting active mailbox pointer need to be in sync to flag clear */
 	phba->sli.mbox_active = NULL;
+	if (bf_get(lpfc_trailer_consumed, mcqe))
+		lpfc_sli4_mq_release(phba->sli4_hba.mbx_wq);
 	spin_unlock_irqrestore(&phba->hbalock, iflags);
 	/* Wake up worker thread to post the next pending mailbox command */
 	lpfc_worker_wake_up(phba);
+	return workposted;
+
 out_no_mqe_complete:
+	spin_lock_irqsave(&phba->hbalock, iflags);
 	if (bf_get(lpfc_trailer_consumed, mcqe))
 		lpfc_sli4_mq_release(phba->sli4_hba.mbx_wq);
-	return workposted;
+	spin_unlock_irqrestore(&phba->hbalock, iflags);
+	return false;
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 43bfea1bffb6b01089c2fe483ede1b036e166579 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:58:57 -0700
Subject: scsi: lpfc: Fix coverity errors on NULL pointer checks

Coverity flagged several scenarios where checking of null pointer values
wasn't consistent.

Fix the code to that be consistent on checking.

Link: https://lore.kernel.org/r/20190922035906.10977-12-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_els.c   |  5 +++++
 drivers/scsi/lpfc/lpfc_nvmet.c | 19 +++++++++++++++----
 drivers/scsi/lpfc/lpfc_scsi.c  |  2 +-
 drivers/scsi/lpfc/lpfc_sli.c   |  9 ++++++---
 4 files changed, 27 insertions(+), 8 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c
index d5303994bfd6..55ab37572e92 100644
--- a/drivers/scsi/lpfc/lpfc_els.c
+++ b/drivers/scsi/lpfc/lpfc_els.c
@@ -4291,6 +4291,11 @@ lpfc_cmpl_els_rsp(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb,
 
 	irsp = &rspiocb->iocb;
 
+	if (!vport) {
+		lpfc_printf_vlog(vport, KERN_ERR, LOG_ELS,
+				 "3177 ELS response failed\n");
+		goto out;
+	}
 	if (cmdiocb->context_un.mbox)
 		mbox = cmdiocb->context_un.mbox;
 
diff --git a/drivers/scsi/lpfc/lpfc_nvmet.c b/drivers/scsi/lpfc/lpfc_nvmet.c
index 121dbdcbb583..c9481a618b85 100644
--- a/drivers/scsi/lpfc/lpfc_nvmet.c
+++ b/drivers/scsi/lpfc/lpfc_nvmet.c
@@ -1958,12 +1958,10 @@ lpfc_nvmet_unsol_ls_buffer(struct lpfc_hba *phba, struct lpfc_sli_ring *pring,
 	uint32_t *payload;
 	uint32_t size, oxid, sid, rc;
 
-	fc_hdr = (struct fc_frame_header *)(nvmebuf->hbuf.virt);
-	oxid = be16_to_cpu(fc_hdr->fh_ox_id);
 
-	if (!phba->targetport) {
+	if (!nvmebuf || !phba->targetport) {
 		lpfc_printf_log(phba, KERN_ERR, LOG_NVME_IOERR,
-				"6154 LS Drop IO x%x\n", oxid);
+				"6154 LS Drop IO\n");
 		oxid = 0;
 		size = 0;
 		sid = 0;
@@ -1971,6 +1969,9 @@ lpfc_nvmet_unsol_ls_buffer(struct lpfc_hba *phba, struct lpfc_sli_ring *pring,
 		goto dropit;
 	}
 
+	fc_hdr = (struct fc_frame_header *)(nvmebuf->hbuf.virt);
+	oxid = be16_to_cpu(fc_hdr->fh_ox_id);
+
 	tgtp = (struct lpfc_nvmet_tgtport *)phba->targetport->private;
 	payload = (uint32_t *)(nvmebuf->dbuf.virt);
 	size = bf_get(lpfc_rcqe_length,  &nvmebuf->cq_event.cqe.rcqe_cmpl);
@@ -2401,6 +2402,11 @@ lpfc_nvmet_unsol_ls_event(struct lpfc_hba *phba, struct lpfc_sli_ring *pring,
 	d_buf = piocb->context2;
 	nvmebuf = container_of(d_buf, struct hbq_dmabuf, dbuf);
 
+	if (!nvmebuf) {
+		lpfc_printf_log(phba, KERN_ERR, LOG_NVME_IOERR,
+				"3015 LS Drop IO\n");
+		return;
+	}
 	if (phba->nvmet_support == 0) {
 		lpfc_in_buf_free(phba, &nvmebuf->dbuf);
 		return;
@@ -2429,6 +2435,11 @@ lpfc_nvmet_unsol_fcp_event(struct lpfc_hba *phba,
 			   uint64_t isr_timestamp,
 			   uint8_t cqflag)
 {
+	if (!nvmebuf) {
+		lpfc_printf_log(phba, KERN_ERR, LOG_NVME_IOERR,
+				"3167 NVMET FCP Drop IO\n");
+		return;
+	}
 	if (phba->nvmet_support == 0) {
 		lpfc_rq_buf_free(phba, &nvmebuf->hbuf);
 		return;
diff --git a/drivers/scsi/lpfc/lpfc_scsi.c b/drivers/scsi/lpfc/lpfc_scsi.c
index c2773c07657d..f06f63e58596 100644
--- a/drivers/scsi/lpfc/lpfc_scsi.c
+++ b/drivers/scsi/lpfc/lpfc_scsi.c
@@ -3814,7 +3814,7 @@ lpfc_scsi_cmd_iocb_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *pIocbIn,
 
 	/* Sanity check on return of outstanding command */
 	cmd = lpfc_cmd->pCmd;
-	if (!cmd) {
+	if (!cmd || !phba) {
 		lpfc_printf_vlog(vport, KERN_ERR, LOG_INIT,
 				 "2621 IO completion: Not an active IO\n");
 		spin_unlock(&lpfc_cmd->buf_lock);
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 313441a3c4cf..939efee6b5dd 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -2674,7 +2674,8 @@ lpfc_sli_handle_mb_event(struct lpfc_hba *phba)
 			lpfc_printf_log(phba, KERN_ERR, LOG_MBOX | LOG_SLI,
 					"(%d):0323 Unknown Mailbox command "
 					"x%x (x%x/x%x) Cmpl\n",
-					pmb->vport ? pmb->vport->vpi : 0,
+					pmb->vport ? pmb->vport->vpi :
+					LPFC_VPORT_UNKNOWN,
 					pmbox->mbxCommand,
 					lpfc_sli_config_mbox_subsys_get(phba,
 									pmb),
@@ -2695,7 +2696,8 @@ lpfc_sli_handle_mb_event(struct lpfc_hba *phba)
 					"(%d):0305 Mbox cmd cmpl "
 					"error - RETRYing Data: x%x "
 					"(x%x/x%x) x%x x%x x%x\n",
-					pmb->vport ? pmb->vport->vpi : 0,
+					pmb->vport ? pmb->vport->vpi :
+					LPFC_VPORT_UNKNOWN,
 					pmbox->mbxCommand,
 					lpfc_sli_config_mbox_subsys_get(phba,
 									pmb),
@@ -2703,7 +2705,8 @@ lpfc_sli_handle_mb_event(struct lpfc_hba *phba)
 									pmb),
 					pmbox->mbxStatus,
 					pmbox->un.varWords[0],
-					pmb->vport->port_state);
+					pmb->vport ? pmb->vport->port_state :
+					LPFC_VPORT_UNKNOWN);
 				pmbox->mbxStatus = 0;
 				pmbox->mbxOwner = OWN_HOST;
 				rc = lpfc_sli_issue_mbox(phba, pmb, MBX_NOWAIT);
-- 
cgit v1.2.3-59-g8ed1b


From 24c7c0a6d3de68b6e15532d18749e561d260c160 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:58:58 -0700
Subject: scsi: lpfc: Fix host hang at boot or slow boot

Scenarios were seen where a host hung when the system booted or the host
was very slow in booting. The link would not come up and no luns were
visible to the host.

After investigation, this was found to be due to the introduction of a new
ACQE that adapter may generate to report a adapter hw warning. The ACQE was
delivered to the driver very early in adapter initialization, when the
driver did not expect command completion. As part of handling this
unexpected interrupt the an EQEs are consumed and discarded and the EQ
rearmed. The issue is the CQ that cause the EQE and thus the interrupt was
not processed and the CQ was left unarmed. Meaning it would no longer
generate a new interrupt condition. Subsequent mailbox commands used to
initialize the adapter use the same CQ, and as there was no completion
interrupt generated, the driver never saw the mailbox commands complete and
it would wait long command timeouts.

Fix by having the early flush routine also process the related CQ and rearm
the CQ.

Link: https://lore.kernel.org/r/20190922035906.10977-13-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 42 ++++++++++++++++++++++++++++++++++++------
 1 file changed, 36 insertions(+), 6 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 939efee6b5dd..412cd8c56d90 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -87,6 +87,10 @@ static void lpfc_sli4_hba_handle_eqe(struct lpfc_hba *phba,
 				     struct lpfc_eqe *eqe);
 static bool lpfc_sli4_mbox_completions_pending(struct lpfc_hba *phba);
 static bool lpfc_sli4_process_missed_mbox_completions(struct lpfc_hba *phba);
+static struct lpfc_cqe *lpfc_sli4_cq_get(struct lpfc_queue *q);
+static void __lpfc_sli4_consume_cqe(struct lpfc_hba *phba,
+				    struct lpfc_queue *cq,
+				    struct lpfc_cqe *cqe);
 
 static IOCB_t *
 lpfc_get_iocb_from_iocbq(struct lpfc_iocbq *iocbq)
@@ -467,21 +471,47 @@ __lpfc_sli4_consume_eqe(struct lpfc_hba *phba, struct lpfc_queue *eq,
 }
 
 static void
-lpfc_sli4_eq_flush(struct lpfc_hba *phba, struct lpfc_queue *eq)
+lpfc_sli4_eqcq_flush(struct lpfc_hba *phba, struct lpfc_queue *eq)
 {
-	struct lpfc_eqe *eqe;
-	uint32_t count = 0;
+	struct lpfc_eqe *eqe = NULL;
+	u32 eq_count = 0, cq_count = 0;
+	struct lpfc_cqe *cqe = NULL;
+	struct lpfc_queue *cq = NULL, *childq = NULL;
+	int cqid = 0;
 
 	/* walk all the EQ entries and drop on the floor */
 	eqe = lpfc_sli4_eq_get(eq);
 	while (eqe) {
+		/* Get the reference to the corresponding CQ */
+		cqid = bf_get_le32(lpfc_eqe_resource_id, eqe);
+		cq = NULL;
+
+		list_for_each_entry(childq, &eq->child_list, list) {
+			if (childq->queue_id == cqid) {
+				cq = childq;
+				break;
+			}
+		}
+		/* If CQ is valid, iterate through it and drop all the CQEs */
+		if (cq) {
+			cqe = lpfc_sli4_cq_get(cq);
+			while (cqe) {
+				__lpfc_sli4_consume_cqe(phba, cq, cqe);
+				cq_count++;
+				cqe = lpfc_sli4_cq_get(cq);
+			}
+			/* Clear and re-arm the CQ */
+			phba->sli4_hba.sli4_write_cq_db(phba, cq, cq_count,
+			    LPFC_QUEUE_REARM);
+			cq_count = 0;
+		}
 		__lpfc_sli4_consume_eqe(phba, eq, eqe);
-		count++;
+		eq_count++;
 		eqe = lpfc_sli4_eq_get(eq);
 	}
 
 	/* Clear and re-arm the EQ */
-	phba->sli4_hba.sli4_write_eq_db(phba, eq, count, LPFC_QUEUE_REARM);
+	phba->sli4_hba.sli4_write_eq_db(phba, eq, eq_count, LPFC_QUEUE_REARM);
 }
 
 static int
@@ -14236,7 +14266,7 @@ lpfc_sli4_hba_intr_handler(int irq, void *dev_id)
 		spin_lock_irqsave(&phba->hbalock, iflag);
 		if (phba->link_state < LPFC_LINK_DOWN)
 			/* Flush, clear interrupt, and rearm the EQ */
-			lpfc_sli4_eq_flush(phba, fpeq);
+			lpfc_sli4_eqcq_flush(phba, fpeq);
 		spin_unlock_irqrestore(&phba->hbalock, iflag);
 		return IRQ_NONE;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 15498dc1a55b7aaea4b51ff03e3ff0f662e73f44 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:58:59 -0700
Subject: scsi: lpfc: Fix list corruption in lpfc_sli_get_iocbq

After study, it was determined there was a double free of a CT iocb during
execution of lpfc_offline_prep and lpfc_offline.  The prep routine issued
an abort for some CT iocbs, but the aborts did not complete fast enough for
a subsequent routine that waits for completion. Thus the driver proceeded
to lpfc_offline, which releases any pending iocbs. Unfortunately, the
completions for the aborts were then received which re-released the ct
iocbs.

Turns out the issue for why the aborts didn't complete fast enough was not
their time on the wire/in the adapter. It was the lpfc_work_done routine,
which requires the adapter state to be UP before it calls
lpfc_sli_handle_slow_ring_event() to process the completions. The issue is
the prep routine takes the link down as part of it's processing.

To fix, the following was performed:

 - Prevent the offline routine from releasing iocbs that have had aborts
   issued on them. Defer to the abort completions. Also means the driver
   fully waits for the completions.  Given this change, the recognition of
   "driver-generated" status which then releases the iocb is no longer
   valid. As such, the change made in the commit 296012285c90 is reverted.
   As recognition of "driver-generated" status is no longer valid, this
   patch reverts the changes made in
   commit 296012285c90 ("scsi: lpfc: Fix leak of ELS completions on adapter reset")

 - Modify lpfc_work_done to allow slow path completions so that the abort
   completions aren't ignored.

 - Updated the fdmi path to recognize a CT request that fails due to the
   port being unusable. This stops FDMI retries. FDMI will be restarted on
   next link up.

Link: https://lore.kernel.org/r/20190922035906.10977-14-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_ct.c      | 6 ++++++
 drivers/scsi/lpfc/lpfc_els.c     | 3 +++
 drivers/scsi/lpfc/lpfc_hbadisc.c | 5 ++++-
 drivers/scsi/lpfc/lpfc_sli.c     | 3 ---
 4 files changed, 13 insertions(+), 4 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_ct.c b/drivers/scsi/lpfc/lpfc_ct.c
index 25e86706e207..f883fac2d2b1 100644
--- a/drivers/scsi/lpfc/lpfc_ct.c
+++ b/drivers/scsi/lpfc/lpfc_ct.c
@@ -1868,6 +1868,12 @@ lpfc_cmpl_ct_disc_fdmi(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb,
 		if (irsp->ulpStatus == IOSTAT_LOCAL_REJECT) {
 			switch ((irsp->un.ulpWord[4] & IOERR_PARAM_MASK)) {
 			case IOERR_SLI_ABORTED:
+			case IOERR_SLI_DOWN:
+				/* Driver aborted this IO.  No retry as error
+				 * is likely Offline->Online or some adapter
+				 * error.  Recovery will try again.
+				 */
+				break;
 			case IOERR_ABORT_IN_PROGRESS:
 			case IOERR_SEQUENCE_TIMEOUT:
 			case IOERR_ILLEGAL_FRAME:
diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c
index 55ab37572e92..bd8109b2a083 100644
--- a/drivers/scsi/lpfc/lpfc_els.c
+++ b/drivers/scsi/lpfc/lpfc_els.c
@@ -8019,6 +8019,9 @@ lpfc_els_flush_cmd(struct lpfc_vport *vport)
 		if (piocb->vport != vport)
 			continue;
 
+		if (piocb->iocb_flag & LPFC_DRIVER_ABORTED)
+			continue;
+
 		/* On the ELS ring we can have ELS_REQUESTs or
 		 * GEN_REQUESTs waiting for a response.
 		 */
diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c
index f483b3aea22b..808ad666bb1b 100644
--- a/drivers/scsi/lpfc/lpfc_hbadisc.c
+++ b/drivers/scsi/lpfc/lpfc_hbadisc.c
@@ -700,7 +700,10 @@ lpfc_work_done(struct lpfc_hba *phba)
 			if (!(phba->hba_flag & HBA_SP_QUEUE_EVT))
 				set_bit(LPFC_DATA_READY, &phba->data_flags);
 		} else {
-			if (phba->link_state >= LPFC_LINK_UP ||
+			/* Driver could have abort request completed in queue
+			 * when link goes down.  Allow for this transition.
+			 */
+			if (phba->link_state >= LPFC_LINK_DOWN ||
 			    phba->link_flag & LS_MDS_LOOPBACK) {
 				pring->flag &= ~LPFC_DEFERRED_RING_EVENT;
 				lpfc_sli_handle_slow_ring_event(phba, pring,
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 412cd8c56d90..ff261c0c738a 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -11090,9 +11090,6 @@ lpfc_sli_abort_els_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb,
 				irsp->ulpStatus, irsp->un.ulpWord[4]);
 
 		spin_unlock_irq(&phba->hbalock);
-		if (irsp->ulpStatus == IOSTAT_LOCAL_REJECT &&
-		    irsp->un.ulpWord[4] == IOERR_SLI_ABORTED)
-			lpfc_sli_release_iocbq(phba, abort_iocb);
 	}
 release_iocb:
 	lpfc_sli_release_iocbq(phba, cmdiocb);
-- 
cgit v1.2.3-59-g8ed1b


From a4c21acca2be6729ecbe72eda9b08092725b0a77 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:59:01 -0700
Subject: scsi: lpfc: Fix hdwq sgl locks and irq handling

Many of the sgl-per-hdwq paths are locking with spin_lock_irq() and
spin_unlock_irq() and may unwittingly raising irq when it shouldn't. Hard
deadlocks were seen around lpfc_scsi_prep_cmnd().

Fix by converting the locks to irqsave/irqrestore.

Fixes: d79c9e9d4b3d ("scsi: lpfc: Support dynamic unbounded SGL lists on G7 hardware.")
Link: https://lore.kernel.org/r/20190922035906.10977-16-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 38 ++++++++++++++++++++++----------------
 1 file changed, 22 insertions(+), 16 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index ff261c0c738a..6d89dd3dd532 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -20444,8 +20444,9 @@ lpfc_get_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 	struct sli4_hybrid_sgl *allocated_sgl = NULL;
 	struct lpfc_sli4_hdw_queue *hdwq = lpfc_buf->hdwq;
 	struct list_head *buf_list = &hdwq->sgl_list;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	if (likely(!list_empty(buf_list))) {
 		/* break off 1 chunk from the sgl_list */
@@ -20457,7 +20458,7 @@ lpfc_get_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 		}
 	} else {
 		/* allocate more */
-		spin_unlock_irq(&hdwq->hdwq_lock);
+		spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 		tmp = kmalloc_node(sizeof(*tmp), GFP_ATOMIC,
 				   cpu_to_node(smp_processor_id()));
 		if (!tmp) {
@@ -20479,7 +20480,7 @@ lpfc_get_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 			return NULL;
 		}
 
-		spin_lock_irq(&hdwq->hdwq_lock);
+		spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 		list_add_tail(&tmp->list_node, &lpfc_buf->dma_sgl_xtra_list);
 	}
 
@@ -20487,7 +20488,7 @@ lpfc_get_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 					struct sli4_hybrid_sgl,
 					list_node);
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 
 	return allocated_sgl;
 }
@@ -20511,8 +20512,9 @@ lpfc_put_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 	struct sli4_hybrid_sgl *tmp = NULL;
 	struct lpfc_sli4_hdw_queue *hdwq = lpfc_buf->hdwq;
 	struct list_head *buf_list = &hdwq->sgl_list;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	if (likely(!list_empty(&lpfc_buf->dma_sgl_xtra_list))) {
 		list_for_each_entry_safe(list_entry, tmp,
@@ -20525,7 +20527,7 @@ lpfc_put_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 		rc = -EINVAL;
 	}
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 	return rc;
 }
 
@@ -20546,8 +20548,9 @@ lpfc_free_sgl_per_hdwq(struct lpfc_hba *phba,
 	struct list_head *buf_list = &hdwq->sgl_list;
 	struct sli4_hybrid_sgl *list_entry = NULL;
 	struct sli4_hybrid_sgl *tmp = NULL;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	/* Free sgl pool */
 	list_for_each_entry_safe(list_entry, tmp,
@@ -20559,7 +20562,7 @@ lpfc_free_sgl_per_hdwq(struct lpfc_hba *phba,
 		kfree(list_entry);
 	}
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 }
 
 /**
@@ -20583,8 +20586,9 @@ lpfc_get_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 	struct fcp_cmd_rsp_buf *allocated_buf = NULL;
 	struct lpfc_sli4_hdw_queue *hdwq = lpfc_buf->hdwq;
 	struct list_head *buf_list = &hdwq->cmd_rsp_buf_list;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	if (likely(!list_empty(buf_list))) {
 		/* break off 1 chunk from the list */
@@ -20597,7 +20601,7 @@ lpfc_get_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 		}
 	} else {
 		/* allocate more */
-		spin_unlock_irq(&hdwq->hdwq_lock);
+		spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 		tmp = kmalloc_node(sizeof(*tmp), GFP_ATOMIC,
 				   cpu_to_node(smp_processor_id()));
 		if (!tmp) {
@@ -20624,7 +20628,7 @@ lpfc_get_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 		tmp->fcp_rsp = (struct fcp_rsp *)((uint8_t *)tmp->fcp_cmnd +
 				sizeof(struct fcp_cmnd));
 
-		spin_lock_irq(&hdwq->hdwq_lock);
+		spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 		list_add_tail(&tmp->list_node, &lpfc_buf->dma_cmd_rsp_list);
 	}
 
@@ -20632,7 +20636,7 @@ lpfc_get_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 					struct fcp_cmd_rsp_buf,
 					list_node);
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 
 	return allocated_buf;
 }
@@ -20657,8 +20661,9 @@ lpfc_put_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 	struct fcp_cmd_rsp_buf *tmp = NULL;
 	struct lpfc_sli4_hdw_queue *hdwq = lpfc_buf->hdwq;
 	struct list_head *buf_list = &hdwq->cmd_rsp_buf_list;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	if (likely(!list_empty(&lpfc_buf->dma_cmd_rsp_list))) {
 		list_for_each_entry_safe(list_entry, tmp,
@@ -20671,7 +20676,7 @@ lpfc_put_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 		rc = -EINVAL;
 	}
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 	return rc;
 }
 
@@ -20692,8 +20697,9 @@ lpfc_free_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 	struct list_head *buf_list = &hdwq->cmd_rsp_buf_list;
 	struct fcp_cmd_rsp_buf *list_entry = NULL;
 	struct fcp_cmd_rsp_buf *tmp = NULL;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	/* Free cmd_rsp buf pool */
 	list_for_each_entry_safe(list_entry, tmp,
@@ -20706,5 +20712,5 @@ lpfc_free_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 		kfree(list_entry);
 	}
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 35a635af54ce79881eb35ba20b64dcb1e81b0389 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:59:02 -0700
Subject: scsi: lpfc: Fix list corruption detected in lpfc_put_sgl_per_hdwq

In lpfc_release_io_buf, an lpfc_io_buf is returned to the 'available' pool
before any associated sgl or cmd and rsp buffers are returned via their
respective 'put' routines.  If xri rebalancing occurs and an lpfc_io_buf
structure is reused quickly, there may be a race condition between release
of old and association of new resources.

Re-ordered lpfc_release_io_buf to release sgl and cmd/rsp
buffer lists before releasing the lpfc_io_buf structure for re-use.

Fixes: d79c9e9d4b3d ("scsi: lpfc: Support dynamic unbounded SGL lists on G7 hardware.")
Link: https://lore.kernel.org/r/20190922035906.10977-17-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 6d89dd3dd532..09e275e3bcd8 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -20138,6 +20138,13 @@ void lpfc_release_io_buf(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_ncmd,
 	lpfc_ncmd->cur_iocbq.wqe_cmpl = NULL;
 	lpfc_ncmd->cur_iocbq.iocb_cmpl = NULL;
 
+	if (phba->cfg_xpsgl && !phba->nvmet_support &&
+	    !list_empty(&lpfc_ncmd->dma_sgl_xtra_list))
+		lpfc_put_sgl_per_hdwq(phba, lpfc_ncmd);
+
+	if (!list_empty(&lpfc_ncmd->dma_cmd_rsp_list))
+		lpfc_put_cmd_rsp_buf_per_hdwq(phba, lpfc_ncmd);
+
 	if (phba->cfg_xri_rebalancing) {
 		if (lpfc_ncmd->expedite) {
 			/* Return to expedite pool */
@@ -20202,13 +20209,6 @@ void lpfc_release_io_buf(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_ncmd,
 		spin_unlock_irqrestore(&qp->io_buf_list_put_lock,
 				       iflag);
 	}
-
-	if (phba->cfg_xpsgl && !phba->nvmet_support &&
-	    !list_empty(&lpfc_ncmd->dma_sgl_xtra_list))
-		lpfc_put_sgl_per_hdwq(phba, lpfc_ncmd);
-
-	if (!list_empty(&lpfc_ncmd->dma_cmd_rsp_list))
-		lpfc_put_cmd_rsp_buf_per_hdwq(phba, lpfc_ncmd);
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From ff349bca17716f310697b619b8cf9b926e852ba9 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Sat, 21 Sep 2019 20:59:05 -0700
Subject: scsi: lpfc: cleanup: remove unused fcp_txcmlpq_cnt

Local variable fcp_txcmplq_cnt is initialized to 0 and then displayed in
lpfc driver message 0387.

Presumed residual (or unused) code from previous commit.

Removed fcp_txcmplq_cnt.

Link: https://lore.kernel.org/r/20190922035906.10977-20-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 09e275e3bcd8..379c37451645 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -13260,7 +13260,6 @@ lpfc_sli4_sp_handle_els_wcqe(struct lpfc_hba *phba, struct lpfc_queue *cq,
 	struct lpfc_sli_ring *pring = cq->pring;
 	int txq_cnt = 0;
 	int txcmplq_cnt = 0;
-	int fcp_txcmplq_cnt = 0;
 
 	/* Check for response status */
 	if (unlikely(bf_get(lpfc_wcqe_c_status, wcqe))) {
@@ -13282,9 +13281,8 @@ lpfc_sli4_sp_handle_els_wcqe(struct lpfc_hba *phba, struct lpfc_queue *cq,
 			txcmplq_cnt++;
 		lpfc_printf_log(phba, KERN_ERR, LOG_SLI,
 			"0387 NO IOCBQ data: txq_cnt=%d iocb_cnt=%d "
-			"fcp_txcmplq_cnt=%d, els_txcmplq_cnt=%d\n",
+			"els_txcmplq_cnt=%d\n",
 			txq_cnt, phba->iocb_cnt,
-			fcp_txcmplq_cnt,
 			txcmplq_cnt);
 		return false;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 27f3efd637ce4859a44a7ca730c72392b4111c26 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Fri, 18 Oct 2019 14:18:19 -0700
Subject: scsi: lpfc: Fix lockdep errors in sli_ringtx_put

Fix lockdep error in __lpfc_sli_ringtx_put(): The hbalock is valid for
sli3, but not for sli4.  Change lockdep to look at ring lock if sli4.

Also update comment in __lpfc_sli_issue_iocb_s4() to reflect proper
lock. Note: lockdep check is already correct.

Link: https://lore.kernel.org/r/20191018211832.7917-4-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 379c37451645..3a6520187ee5 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -9004,7 +9004,8 @@ lpfc_mbox_api_table_setup(struct lpfc_hba *phba, uint8_t dev_grp)
  * @pring: Pointer to driver SLI ring object.
  * @piocb: Pointer to address of newly added command iocb.
  *
- * This function is called with hbalock held to add a command
+ * This function is called with hbalock held for SLI3 ports or
+ * the ring lock held for SLI4 ports to add a command
  * iocb to the txq when SLI layer cannot submit the command iocb
  * to the ring.
  **/
@@ -9012,7 +9013,10 @@ void
 __lpfc_sli_ringtx_put(struct lpfc_hba *phba, struct lpfc_sli_ring *pring,
 		    struct lpfc_iocbq *piocb)
 {
-	lockdep_assert_held(&phba->hbalock);
+	if (phba->sli_rev == LPFC_SLI_REV4)
+		lockdep_assert_held(&pring->ring_lock);
+	else
+		lockdep_assert_held(&phba->hbalock);
 	/* Insert the caller's iocb in the txq tail for later processing. */
 	list_add_tail(&piocb->list, &pring->txq);
 }
@@ -9903,7 +9907,7 @@ lpfc_sli4_iocb2wqe(struct lpfc_hba *phba, struct lpfc_iocbq *iocbq,
  * __lpfc_sli_issue_iocb_s4 is used by other functions in the driver to issue
  * an iocb command to an HBA with SLI-4 interface spec.
  *
- * This function is called with hbalock held. The function will return success
+ * This function is called with ringlock held. The function will return success
  * after it successfully submit the iocb to firmware or after adding to the
  * txq.
  **/
-- 
cgit v1.2.3-59-g8ed1b


From 324e1c402069e8d277d2a2b18ce40bde1265b96a Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Fri, 18 Oct 2019 14:18:21 -0700
Subject: scsi: lpfc: Fix bad ndlp ptr in xri aborted handling

In cases where I/O may be aborted, such as driver unload or link bounces,
the system will crash based on a bad ndlp pointer.

Example:
  RIP: 0010:lpfc_sli4_abts_err_handler+0x15/0x140 [lpfc]
  ...
  lpfc_sli4_io_xri_aborted+0x20d/0x270 [lpfc]
  lpfc_sli4_sp_handle_abort_xri_wcqe.isra.54+0x84/0x170 [lpfc]
  lpfc_sli4_fp_handle_cqe+0xc2/0x480 [lpfc]
  __lpfc_sli4_process_cq+0xc6/0x230 [lpfc]
  __lpfc_sli4_hba_process_cq+0x29/0xc0 [lpfc]
  process_one_work+0x14c/0x390

Crash was caused by a bad ndlp address passed to I/O indicated by the XRI
aborted CQE.  The address was not NULL so the routine deferenced the ndlp
ptr. The bad ndlp also caused the lpfc_sli4_io_xri_aborted to call an
erroneous io handler.  Root cause for the bad ndlp was an lpfc_ncmd that
was aborted, put on the abort_io list, completed, taken off the abort_io
list, sent to lpfc_release_nvme_buf where it was put back on the abort_io
list because the lpfc_ncmd->flags setting LPFC_SBUF_XBUSY was not cleared
on the final completion.

Rework the exchange busy handling to ensure the flags are properly set for
both scsi and nvme.

Fixes: c490850a0947 ("scsi: lpfc: Adapt partitioned XRI lists to efficient sharing")
Cc: <stable@vger.kernel.org> # v5.1+
Link: https://lore.kernel.org/r/20191018211832.7917-6-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_scsi.c | 11 +++++++----
 drivers/scsi/lpfc/lpfc_sli.c  |  5 ++++-
 drivers/scsi/lpfc/lpfc_sli.h  |  3 +--
 3 files changed, 12 insertions(+), 7 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_scsi.c b/drivers/scsi/lpfc/lpfc_scsi.c
index f06f63e58596..13e3e14b43f9 100644
--- a/drivers/scsi/lpfc/lpfc_scsi.c
+++ b/drivers/scsi/lpfc/lpfc_scsi.c
@@ -526,7 +526,7 @@ lpfc_sli4_io_xri_aborted(struct lpfc_hba *phba,
 		&qp->lpfc_abts_io_buf_list, list) {
 		if (psb->cur_iocbq.sli4_xritag == xri) {
 			list_del_init(&psb->list);
-			psb->exch_busy = 0;
+			psb->flags &= ~LPFC_SBUF_XBUSY;
 			psb->status = IOSTAT_SUCCESS;
 #ifdef BUILD_NVME
 			if (psb->cur_iocbq.iocb_flag == LPFC_IO_NVME) {
@@ -568,7 +568,7 @@ lpfc_sli4_io_xri_aborted(struct lpfc_hba *phba,
 		if (iocbq->sli4_xritag != xri)
 			continue;
 		psb = container_of(iocbq, struct lpfc_io_buf, cur_iocbq);
-		psb->exch_busy = 0;
+		psb->flags &= ~LPFC_SBUF_XBUSY;
 		spin_unlock_irqrestore(&phba->hbalock, iflag);
 		if (!list_empty(&pring->txq))
 			lpfc_worker_wake_up(phba);
@@ -788,7 +788,7 @@ lpfc_release_scsi_buf_s4(struct lpfc_hba *phba, struct lpfc_io_buf *psb)
 	psb->prot_seg_cnt = 0;
 
 	qp = psb->hdwq;
-	if (psb->exch_busy) {
+	if (psb->flags & LPFC_SBUF_XBUSY) {
 		spin_lock_irqsave(&qp->abts_io_buf_list_lock, iflag);
 		psb->pCmd = NULL;
 		list_add_tail(&psb->list, &qp->lpfc_abts_io_buf_list);
@@ -3837,7 +3837,10 @@ lpfc_scsi_cmd_iocb_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *pIocbIn,
 	lpfc_cmd->result = (pIocbOut->iocb.un.ulpWord[4] & IOERR_PARAM_MASK);
 	lpfc_cmd->status = pIocbOut->iocb.ulpStatus;
 	/* pick up SLI4 exhange busy status from HBA */
-	lpfc_cmd->exch_busy = pIocbOut->iocb_flag & LPFC_EXCHANGE_BUSY;
+	if (pIocbOut->iocb_flag & LPFC_EXCHANGE_BUSY)
+		lpfc_cmd->flags |= LPFC_SBUF_XBUSY;
+	else
+		lpfc_cmd->flags &= ~LPFC_SBUF_XBUSY;
 
 #ifdef CONFIG_SCSI_LPFC_DEBUG_FS
 	if (lpfc_cmd->prot_data_type) {
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 3a6520187ee5..bb0e155eb32c 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -11777,7 +11777,10 @@ lpfc_sli_wake_iocb_wait(struct lpfc_hba *phba,
 		!(cmdiocbq->iocb_flag & LPFC_IO_LIBDFC)) {
 		lpfc_cmd = container_of(cmdiocbq, struct lpfc_io_buf,
 			cur_iocbq);
-		lpfc_cmd->exch_busy = rspiocbq->iocb_flag & LPFC_EXCHANGE_BUSY;
+		if (rspiocbq && (rspiocbq->iocb_flag & LPFC_EXCHANGE_BUSY))
+			lpfc_cmd->flags |= LPFC_SBUF_XBUSY;
+		else
+			lpfc_cmd->flags &= ~LPFC_SBUF_XBUSY;
 	}
 
 	pdone_q = cmdiocbq->context_un.wait_queue;
diff --git a/drivers/scsi/lpfc/lpfc_sli.h b/drivers/scsi/lpfc/lpfc_sli.h
index 37fbcb46387e..7bcf922a8be2 100644
--- a/drivers/scsi/lpfc/lpfc_sli.h
+++ b/drivers/scsi/lpfc/lpfc_sli.h
@@ -384,14 +384,13 @@ struct lpfc_io_buf {
 
 	struct lpfc_nodelist *ndlp;
 	uint32_t timeout;
-	uint16_t flags;  /* TBD convert exch_busy to flags */
+	uint16_t flags;
 #define LPFC_SBUF_XBUSY		0x1	/* SLI4 hba reported XB on WCQE cmpl */
 #define LPFC_SBUF_BUMP_QDEPTH	0x2	/* bumped queue depth counter */
 					/* External DIF device IO conversions */
 #define LPFC_SBUF_NORMAL_DIF	0x4	/* normal mode to insert/strip */
 #define LPFC_SBUF_PASS_DIF	0x8	/* insert/strip mode to passthru */
 #define LPFC_SBUF_NOT_POSTED    0x10    /* SGL failed post to FW. */
-	uint16_t exch_busy;     /* SLI4 hba reported XB on complete WCQE */
 	uint16_t status;	/* From IOCB Word 7- ulpStatus */
 	uint32_t result;	/* From IOCB Word 4. */
 
-- 
cgit v1.2.3-59-g8ed1b


From 8156d378c4cbf8ca19df5d8f0c610ce6923b61e2 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Fri, 18 Oct 2019 14:18:26 -0700
Subject: scsi: lpfc: Revise interrupt coalescing for missing scenarios

The existing "auto eq delay" mechanism was sometimes skipping over an EQ,
not ramping the coalescing down under light load fast enough, and in other
cases never kicked in as cpu sharing by multiple vectors didn't quite add
up right.

Tweak the interrupt mechanism such that:

 - Add a flag to the EQ to force checking for colaescing values when being
   serviced in the interrupt handler.  The flag will be set by any CQ bound
   to the EQ whenever the number of CQ elements process in a single scan
   meets or exceeds the hardware queue notify level. E.g. there's a
   significant number of completions happening.

 - In the heartbeat work item that checks coalescing:

   - Replace the structure that was counting the number of EQs that
     interrupted on a single cpu with a new structure that looks at the EQ
     to see whether EQ currently has a coalescing value (thus it should be
     re-evaluate) or was marked by the new flag indicating heavy
     completions.

   - When a cpu, which may be servicing multiple vectors, had at least 1 EQ
     that should be checked, a new coalescing delay is calculated based on
     the number of interrupts that occurred on the cpu.

   - The new coalescing value is then applied to the EQs that had
     interrupted on the cpu.

Link: https://lore.kernel.org/r/20191018211832.7917-11-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_hw4.h  |  1 -
 drivers/scsi/lpfc/lpfc_init.c | 51 ++++++++++++++++++-------------------------
 drivers/scsi/lpfc/lpfc_sli.c  |  3 ++-
 drivers/scsi/lpfc/lpfc_sli4.h |  1 +
 4 files changed, 24 insertions(+), 32 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_hw4.h b/drivers/scsi/lpfc/lpfc_hw4.h
index 1cd3016f7783..ac86b80230e7 100644
--- a/drivers/scsi/lpfc/lpfc_hw4.h
+++ b/drivers/scsi/lpfc/lpfc_hw4.h
@@ -210,7 +210,6 @@ struct lpfc_sli_intf {
 #define LPFC_MAX_IMAX          5000000
 #define LPFC_DEF_IMAX          0
 
-#define LPFC_IMAX_THRESHOLD    1000
 #define LPFC_MAX_AUTO_EQ_DELAY 120
 #define LPFC_EQ_DELAY_STEP     15
 #define LPFC_EQD_ISR_TRIGGER   20000
diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c
index 8292b66e4b07..316a2c2beb0c 100644
--- a/drivers/scsi/lpfc/lpfc_init.c
+++ b/drivers/scsi/lpfc/lpfc_init.c
@@ -1235,10 +1235,9 @@ lpfc_hb_eq_delay_work(struct work_struct *work)
 					     struct lpfc_hba, eq_delay_work);
 	struct lpfc_eq_intr_info *eqi, *eqi_new;
 	struct lpfc_queue *eq, *eq_next;
-	unsigned char *eqcnt = NULL;
+	unsigned char *ena_delay = NULL;
 	uint32_t usdelay;
 	int i;
-	bool update = false;
 
 	if (!phba->cfg_auto_imax || phba->pport->load_flag & FC_UNLOADING)
 		return;
@@ -1247,44 +1246,36 @@ lpfc_hb_eq_delay_work(struct work_struct *work)
 	    phba->pport->fc_flag & FC_OFFLINE_MODE)
 		goto requeue;
 
-	eqcnt = kcalloc(num_possible_cpus(), sizeof(unsigned char),
-			GFP_KERNEL);
-	if (!eqcnt)
+	ena_delay = kcalloc(phba->sli4_hba.num_possible_cpu, sizeof(*ena_delay),
+			    GFP_KERNEL);
+	if (!ena_delay)
 		goto requeue;
 
-	if (phba->cfg_irq_chann > 1) {
-		/* Loop thru all IRQ vectors */
-		for (i = 0; i < phba->cfg_irq_chann; i++) {
-			/* Get the EQ corresponding to the IRQ vector */
-			eq = phba->sli4_hba.hba_eq_hdl[i].eq;
-			if (!eq)
-				continue;
-			if (eq->q_mode) {
-				update = true;
-				break;
-			}
-			if (eqcnt[eq->last_cpu] < 2)
-				eqcnt[eq->last_cpu]++;
+	for (i = 0; i < phba->cfg_irq_chann; i++) {
+		/* Get the EQ corresponding to the IRQ vector */
+		eq = phba->sli4_hba.hba_eq_hdl[i].eq;
+		if (!eq)
+			continue;
+		if (eq->q_mode || eq->q_flag & HBA_EQ_DELAY_CHK) {
+			eq->q_flag &= ~HBA_EQ_DELAY_CHK;
+			ena_delay[eq->last_cpu] = 1;
 		}
-	} else
-		update = true;
+	}
 
 	for_each_present_cpu(i) {
 		eqi = per_cpu_ptr(phba->sli4_hba.eq_info, i);
-		if (!update && eqcnt[i] < 2) {
-			eqi->icnt = 0;
-			continue;
+		if (ena_delay[i]) {
+			usdelay = (eqi->icnt >> 10) * LPFC_EQ_DELAY_STEP;
+			if (usdelay > LPFC_MAX_AUTO_EQ_DELAY)
+				usdelay = LPFC_MAX_AUTO_EQ_DELAY;
+		} else {
+			usdelay = 0;
 		}
 
-		usdelay = (eqi->icnt / LPFC_IMAX_THRESHOLD) *
-			   LPFC_EQ_DELAY_STEP;
-		if (usdelay > LPFC_MAX_AUTO_EQ_DELAY)
-			usdelay = LPFC_MAX_AUTO_EQ_DELAY;
-
 		eqi->icnt = 0;
 
 		list_for_each_entry_safe(eq, eq_next, &eqi->list, cpu_list) {
-			if (eq->last_cpu != i) {
+			if (unlikely(eq->last_cpu != i)) {
 				eqi_new = per_cpu_ptr(phba->sli4_hba.eq_info,
 						      eq->last_cpu);
 				list_move_tail(&eq->cpu_list, &eqi_new->list);
@@ -1296,7 +1287,7 @@ lpfc_hb_eq_delay_work(struct work_struct *work)
 		}
 	}
 
-	kfree(eqcnt);
+	kfree(ena_delay);
 
 requeue:
 	queue_delayed_work(phba->wq, &phba->eq_delay_work,
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index bb0e155eb32c..0e6674bd15c6 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -13640,6 +13640,7 @@ __lpfc_sli4_process_cq(struct lpfc_hba *phba, struct lpfc_queue *cq,
 			phba->sli4_hba.sli4_write_cq_db(phba, cq, consumed,
 						LPFC_QUEUE_NOARM);
 			consumed = 0;
+			cq->assoc_qp->q_flag |= HBA_EQ_DELAY_CHK;
 		}
 
 		if (count == LPFC_NVMET_CQ_NOTIFY)
@@ -14278,7 +14279,7 @@ lpfc_sli4_hba_intr_handler(int irq, void *dev_id)
 	fpeq->last_cpu = raw_smp_processor_id();
 
 	if (icnt > LPFC_EQD_ISR_TRIGGER &&
-	    phba->cfg_irq_chann == 1 &&
+	    fpeq->q_flag & HBA_EQ_DELAY_CHK &&
 	    phba->cfg_auto_imax &&
 	    fpeq->q_mode != LPFC_MAX_AUTO_EQ_DELAY &&
 	    phba->sli.sli_flag & LPFC_SLI_USE_EQDR)
diff --git a/drivers/scsi/lpfc/lpfc_sli4.h b/drivers/scsi/lpfc/lpfc_sli4.h
index 0d4882a9e634..c9e068ca0fec 100644
--- a/drivers/scsi/lpfc/lpfc_sli4.h
+++ b/drivers/scsi/lpfc/lpfc_sli4.h
@@ -199,6 +199,7 @@ struct lpfc_queue {
 	uint8_t q_flag;
 #define HBA_NVMET_WQFULL	0x1 /* We hit WQ Full condition for NVMET */
 #define HBA_NVMET_CQ_NOTIFY	0x1 /* LPFC_NVMET_CQ_NOTIFY CQEs this EQE */
+#define HBA_EQ_DELAY_CHK	0x2 /* EQ is a candidate for coalescing */
 #define LPFC_NVMET_CQ_NOTIFY	4
 	void __iomem *db_regaddr;
 	uint16_t dpp_enable;
-- 
cgit v1.2.3-59-g8ed1b


From 95bfc6d8ad86a76c89f62bb466f740b0fc05a667 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Fri, 18 Oct 2019 14:18:27 -0700
Subject: scsi: lpfc: Make FW logging dynamically configurable

Currently, the FW logging facility is a load/boot time parameter which
requires the driver to be unloaded/reloaded or the system rebooted in order
to change its configuration.

Convert the logging facility to allow dynamic enablement and configuration.
Specifically:

 - Convert the feature so that it can be enabled dynamically via an
   attribute.  Additionally, the size of the buffer can be configured
   dynamically.

 - Add locks around states that now may be changing.

 - Tie the feature into debugfs so that the logs can be read at any time.

Link: https://lore.kernel.org/r/20191018211832.7917-12-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc.h         |   9 ++-
 drivers/scsi/lpfc/lpfc_attr.c    |  48 +++++++++++++++-
 drivers/scsi/lpfc/lpfc_bsg.c     |  18 ++++--
 drivers/scsi/lpfc/lpfc_debugfs.c | 117 ++++++++++++++++++++++++++++++++++++++-
 drivers/scsi/lpfc/lpfc_sli.c     |  22 +++++++-
 5 files changed, 204 insertions(+), 10 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc.h b/drivers/scsi/lpfc/lpfc.h
index dabb5eac5fed..884d6076d7aa 100644
--- a/drivers/scsi/lpfc/lpfc.h
+++ b/drivers/scsi/lpfc/lpfc.h
@@ -605,6 +605,12 @@ struct lpfc_epd_pool {
 	spinlock_t lock;	/* lock for expedite pool */
 };
 
+enum ras_state {
+	INACTIVE,
+	REG_INPROGRESS,
+	ACTIVE
+};
+
 struct lpfc_ras_fwlog {
 	uint8_t *fwlog_buff;
 	uint32_t fw_buffcount; /* Buffer size posted to FW */
@@ -621,7 +627,7 @@ struct lpfc_ras_fwlog {
 	bool ras_enabled;   /* Ras Enabled for the function */
 #define LPFC_RAS_DISABLE_LOGGING 0x00
 #define LPFC_RAS_ENABLE_LOGGING 0x01
-	bool ras_active;    /* RAS logging running state */
+	enum ras_state state;    /* RAS logging running state */
 };
 
 struct lpfc_hba {
@@ -1053,6 +1059,7 @@ struct lpfc_hba {
 #ifdef LPFC_HDWQ_LOCK_STAT
 	struct dentry *debug_lockstat;
 #endif
+	struct dentry *debug_ras_log;
 	atomic_t nvmeio_trc_cnt;
 	uint32_t nvmeio_trc_size;
 	uint32_t nvmeio_trc_output_idx;
diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c
index 266a71b01c47..ba504cc6e8ed 100644
--- a/drivers/scsi/lpfc/lpfc_attr.c
+++ b/drivers/scsi/lpfc/lpfc_attr.c
@@ -5932,7 +5932,53 @@ LPFC_ATTR_RW(enable_mds_diags, 0, 0, 1, "Enable MDS Diagnostics");
  *	[1-4] = Multiple of 1/4th Mb of host memory for FW logging
  * Value range [0..4]. Default value is 0
  */
-LPFC_ATTR_RW(ras_fwlog_buffsize, 0, 0, 4, "Host memory for FW logging");
+LPFC_ATTR(ras_fwlog_buffsize, 0, 0, 4, "Host memory for FW logging");
+lpfc_param_show(ras_fwlog_buffsize);
+
+static ssize_t
+lpfc_ras_fwlog_buffsize_set(struct lpfc_hba  *phba, uint val)
+{
+	int ret = 0;
+	enum ras_state state;
+
+	if (!lpfc_rangecheck(val, 0, 4))
+		return -EINVAL;
+
+	if (phba->cfg_ras_fwlog_buffsize == val)
+		return 0;
+
+	if (phba->cfg_ras_fwlog_func == PCI_FUNC(phba->pcidev->devfn))
+		return -EINVAL;
+
+	spin_lock_irq(&phba->hbalock);
+	state = phba->ras_fwlog.state;
+	spin_unlock_irq(&phba->hbalock);
+
+	if (state == REG_INPROGRESS) {
+		lpfc_printf_log(phba, KERN_ERR, LOG_SLI, "6147 RAS Logging "
+				"registration is in progress\n");
+		return -EBUSY;
+	}
+
+	/* For disable logging: stop the logs and free the DMA.
+	 * For ras_fwlog_buffsize size change we still need to free and
+	 * reallocate the DMA in lpfc_sli4_ras_fwlog_init.
+	 */
+	phba->cfg_ras_fwlog_buffsize = val;
+	if (state == ACTIVE) {
+		lpfc_ras_stop_fwlog(phba);
+		lpfc_sli4_ras_dma_free(phba);
+	}
+
+	lpfc_sli4_ras_init(phba);
+	if (phba->ras_fwlog.ras_enabled)
+		ret = lpfc_sli4_ras_fwlog_init(phba, phba->cfg_ras_fwlog_level,
+					       LPFC_RAS_ENABLE_LOGGING);
+	return ret;
+}
+
+lpfc_param_store(ras_fwlog_buffsize);
+static DEVICE_ATTR_RW(lpfc_ras_fwlog_buffsize);
 
 /*
  * lpfc_ras_fwlog_level: Firmware logging verbosity level
diff --git a/drivers/scsi/lpfc/lpfc_bsg.c b/drivers/scsi/lpfc/lpfc_bsg.c
index 39a736b887b1..d4e1b120cc9e 100644
--- a/drivers/scsi/lpfc/lpfc_bsg.c
+++ b/drivers/scsi/lpfc/lpfc_bsg.c
@@ -5435,10 +5435,12 @@ lpfc_bsg_get_ras_config(struct bsg_job *job)
 		bsg_reply->reply_data.vendor_reply.vendor_rsp;
 
 	/* Current logging state */
-	if (ras_fwlog->ras_active == true)
+	spin_lock_irq(&phba->hbalock);
+	if (ras_fwlog->state == ACTIVE)
 		ras_reply->state = LPFC_RASLOG_STATE_RUNNING;
 	else
 		ras_reply->state = LPFC_RASLOG_STATE_STOPPED;
+	spin_unlock_irq(&phba->hbalock);
 
 	ras_reply->log_level = phba->ras_fwlog.fw_loglevel;
 	ras_reply->log_buff_sz = phba->cfg_ras_fwlog_buffsize;
@@ -5495,10 +5497,13 @@ lpfc_bsg_set_ras_config(struct bsg_job *job)
 
 	if (action == LPFC_RASACTION_STOP_LOGGING) {
 		/* Check if already disabled */
-		if (ras_fwlog->ras_active == false) {
+		spin_lock_irq(&phba->hbalock);
+		if (ras_fwlog->state != ACTIVE) {
+			spin_unlock_irq(&phba->hbalock);
 			rc = -ESRCH;
 			goto ras_job_error;
 		}
+		spin_unlock_irq(&phba->hbalock);
 
 		/* Disable logging */
 		lpfc_ras_stop_fwlog(phba);
@@ -5509,8 +5514,10 @@ lpfc_bsg_set_ras_config(struct bsg_job *job)
 		 * FW-logging with new log-level. Return status
 		 * "Logging already Running" to caller.
 		 **/
-		if (ras_fwlog->ras_active)
+		spin_lock_irq(&phba->hbalock);
+		if (ras_fwlog->state != INACTIVE)
 			action_status = -EINPROGRESS;
+		spin_unlock_irq(&phba->hbalock);
 
 		/* Enable logging */
 		rc = lpfc_sli4_ras_fwlog_init(phba, log_level,
@@ -5626,10 +5633,13 @@ lpfc_bsg_get_ras_fwlog(struct bsg_job *job)
 		goto ras_job_error;
 
 	/* Logging to be stopped before reading */
-	if (ras_fwlog->ras_active == true) {
+	spin_lock_irq(&phba->hbalock);
+	if (ras_fwlog->state == ACTIVE) {
+		spin_unlock_irq(&phba->hbalock);
 		rc = -EINPROGRESS;
 		goto ras_job_error;
 	}
+	spin_unlock_irq(&phba->hbalock);
 
 	if (job->request_len <
 	    sizeof(struct fc_bsg_request) +
diff --git a/drivers/scsi/lpfc/lpfc_debugfs.c b/drivers/scsi/lpfc/lpfc_debugfs.c
index 8d34be60d379..ab124f7d50d6 100644
--- a/drivers/scsi/lpfc/lpfc_debugfs.c
+++ b/drivers/scsi/lpfc/lpfc_debugfs.c
@@ -2078,6 +2078,96 @@ lpfc_debugfs_lockstat_write(struct file *file, const char __user *buf,
 }
 #endif
 
+int
+lpfc_debugfs_ras_log_data(struct lpfc_hba *phba, char *buffer, int size)
+{
+	int copied = 0;
+	struct lpfc_dmabuf *dmabuf, *next;
+
+	spin_lock_irq(&phba->hbalock);
+	if (phba->ras_fwlog.state != ACTIVE) {
+		spin_unlock_irq(&phba->hbalock);
+		return -EINVAL;
+	}
+	spin_unlock_irq(&phba->hbalock);
+
+	list_for_each_entry_safe(dmabuf, next,
+				 &phba->ras_fwlog.fwlog_buff_list, list) {
+		memcpy(buffer + copied, dmabuf->virt, LPFC_RAS_MAX_ENTRY_SIZE);
+		copied += LPFC_RAS_MAX_ENTRY_SIZE;
+		if (size > copied)
+			break;
+	}
+	return copied;
+}
+
+static int
+lpfc_debugfs_ras_log_release(struct inode *inode, struct file *file)
+{
+	struct lpfc_debug *debug = file->private_data;
+
+	vfree(debug->buffer);
+	kfree(debug);
+
+	return 0;
+}
+
+/**
+ * lpfc_debugfs_ras_log_open - Open the RAS log debugfs buffer
+ * @inode: The inode pointer that contains a vport pointer.
+ * @file: The file pointer to attach the log output.
+ *
+ * Description:
+ * This routine is the entry point for the debugfs open file operation. It gets
+ * the vport from the i_private field in @inode, allocates the necessary buffer
+ * for the log, fills the buffer from the in-memory log for this vport, and then
+ * returns a pointer to that log in the private_data field in @file.
+ *
+ * Returns:
+ * This function returns zero if successful. On error it will return a negative
+ * error value.
+ **/
+static int
+lpfc_debugfs_ras_log_open(struct inode *inode, struct file *file)
+{
+	struct lpfc_hba *phba = inode->i_private;
+	struct lpfc_debug *debug;
+	int size;
+	int rc = -ENOMEM;
+
+	spin_lock_irq(&phba->hbalock);
+	if (phba->ras_fwlog.state != ACTIVE) {
+		spin_unlock_irq(&phba->hbalock);
+		rc = -EINVAL;
+		goto out;
+	}
+	spin_unlock_irq(&phba->hbalock);
+	debug = kmalloc(sizeof(*debug), GFP_KERNEL);
+	if (!debug)
+		goto out;
+
+	size = LPFC_RAS_MIN_BUFF_POST_SIZE * phba->cfg_ras_fwlog_buffsize;
+	debug->buffer = vmalloc(size);
+	if (!debug->buffer)
+		goto free_debug;
+
+	debug->len = lpfc_debugfs_ras_log_data(phba, debug->buffer, size);
+	if (debug->len < 0) {
+		rc = -EINVAL;
+		goto free_buffer;
+	}
+	file->private_data = debug;
+
+	return 0;
+
+free_buffer:
+	vfree(debug->buffer);
+free_debug:
+	kfree(debug);
+out:
+	return rc;
+}
+
 /**
  * lpfc_debugfs_dumpHBASlim_open - Open the Dump HBA SLIM debugfs buffer
  * @inode: The inode pointer that contains a vport pointer.
@@ -5286,6 +5376,16 @@ static const struct file_operations lpfc_debugfs_op_lockstat = {
 };
 #endif
 
+#undef lpfc_debugfs_ras_log
+static const struct file_operations lpfc_debugfs_ras_log = {
+	.owner =        THIS_MODULE,
+	.open =         lpfc_debugfs_ras_log_open,
+	.llseek =       lpfc_debugfs_lseek,
+	.read =         lpfc_debugfs_read,
+	.release =      lpfc_debugfs_ras_log_release,
+};
+#endif
+
 #undef lpfc_debugfs_op_dumpHBASlim
 static const struct file_operations lpfc_debugfs_op_dumpHBASlim = {
 	.owner =        THIS_MODULE,
@@ -5457,7 +5557,6 @@ static const struct file_operations lpfc_idiag_op_extAcc = {
 	.release =      lpfc_idiag_cmd_release,
 };
 
-#endif
 
 /* lpfc_idiag_mbxacc_dump_bsg_mbox - idiag debugfs dump bsg mailbox command
  * @phba: Pointer to HBA context object.
@@ -5707,6 +5806,19 @@ lpfc_debugfs_initialize(struct lpfc_vport *vport)
 			goto debug_failed;
 		}
 
+		/* RAS log */
+		snprintf(name, sizeof(name), "ras_log");
+		phba->debug_ras_log =
+			debugfs_create_file(name, 0644,
+					    phba->hba_debugfs_root,
+					    phba, &lpfc_debugfs_ras_log);
+		if (!phba->debug_ras_log) {
+			lpfc_printf_vlog(vport, KERN_ERR, LOG_INIT,
+					 "6148 Cannot create debugfs"
+					 " ras_log\n");
+			goto debug_failed;
+		}
+
 		/* Setup hbqinfo */
 		snprintf(name, sizeof(name), "hbqinfo");
 		phba->debug_hbqinfo =
@@ -6117,6 +6229,9 @@ lpfc_debugfs_terminate(struct lpfc_vport *vport)
 		debugfs_remove(phba->debug_hbqinfo); /* hbqinfo */
 		phba->debug_hbqinfo = NULL;
 
+		debugfs_remove(phba->debug_ras_log);
+		phba->debug_ras_log = NULL;
+
 #ifdef LPFC_HDWQ_LOCK_STAT
 		debugfs_remove(phba->debug_lockstat); /* lockstat */
 		phba->debug_lockstat = NULL;
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 0e6674bd15c6..294f041961a8 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -6219,11 +6219,16 @@ lpfc_ras_stop_fwlog(struct lpfc_hba *phba)
 {
 	struct lpfc_ras_fwlog *ras_fwlog = &phba->ras_fwlog;
 
-	ras_fwlog->ras_active = false;
+	spin_lock_irq(&phba->hbalock);
+	ras_fwlog->state = INACTIVE;
+	spin_unlock_irq(&phba->hbalock);
 
 	/* Disable FW logging to host memory */
 	writel(LPFC_CTL_PDEV_CTL_DDL_RAS,
 	       phba->sli4_hba.conf_regs_memmap_p + LPFC_CTL_PDEV_CTL_OFFSET);
+
+	/* Wait 10ms for firmware to stop using DMA buffer */
+	usleep_range(10 * 1000, 20 * 1000);
 }
 
 /**
@@ -6259,7 +6264,9 @@ lpfc_sli4_ras_dma_free(struct lpfc_hba *phba)
 		ras_fwlog->lwpd.virt = NULL;
 	}
 
-	ras_fwlog->ras_active = false;
+	spin_lock_irq(&phba->hbalock);
+	ras_fwlog->state = INACTIVE;
+	spin_unlock_irq(&phba->hbalock);
 }
 
 /**
@@ -6361,7 +6368,9 @@ lpfc_sli4_ras_mbox_cmpl(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 		goto disable_ras;
 	}
 
-	ras_fwlog->ras_active = true;
+	spin_lock_irq(&phba->hbalock);
+	ras_fwlog->state = ACTIVE;
+	spin_unlock_irq(&phba->hbalock);
 	mempool_free(pmb, phba->mbox_mem_pool);
 
 	return;
@@ -6393,6 +6402,10 @@ lpfc_sli4_ras_fwlog_init(struct lpfc_hba *phba,
 	uint32_t len = 0, fwlog_buffsize, fwlog_entry_count;
 	int rc = 0;
 
+	spin_lock_irq(&phba->hbalock);
+	ras_fwlog->state = INACTIVE;
+	spin_unlock_irq(&phba->hbalock);
+
 	fwlog_buffsize = (LPFC_RAS_MIN_BUFF_POST_SIZE *
 			  phba->cfg_ras_fwlog_buffsize);
 	fwlog_entry_count = (fwlog_buffsize/LPFC_RAS_MAX_ENTRY_SIZE);
@@ -6452,6 +6465,9 @@ lpfc_sli4_ras_fwlog_init(struct lpfc_hba *phba,
 	mbx_fwlog->u.request.lwpd.addr_lo = putPaddrLow(ras_fwlog->lwpd.phys);
 	mbx_fwlog->u.request.lwpd.addr_hi = putPaddrHigh(ras_fwlog->lwpd.phys);
 
+	spin_lock_irq(&phba->hbalock);
+	ras_fwlog->state = REG_INPROGRESS;
+	spin_unlock_irq(&phba->hbalock);
 	mbox->vport = phba->pport;
 	mbox->mbox_cmpl = lpfc_sli4_ras_mbox_cmpl;
 
-- 
cgit v1.2.3-59-g8ed1b


From 7cfd5639d99bec0d27af089d0c8c114330e43a72 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Mon, 4 Nov 2019 16:56:58 -0800
Subject: scsi: lpfc: Fix duplicate unreg_rpi error in port offline flow

If the driver receives a login that is later then LOGO'd by the remote port
(aka ndlp), the driver, upon the completion of the LOGO ACC transmission,
will logout the node and unregister the rpi that is being used for the
node.  As part of the unreg, the node's rpi value is replaced by the
LPFC_RPI_ALLOC_ERROR value.  If the port is subsequently offlined, the
offline walks the nodes and ensures they are logged out, which possibly
entails unreg'ing their rpi values.  This path does not validate the node's
rpi value, thus doesn't detect that it has been unreg'd already.  The
replaced rpi value is then used when accessing the rpi bitmask array which
tracks active rpi values.  As the LPFC_RPI_ALLOC_ERROR value is not a valid
index for the bitmask, it may fault the system.

Revise the rpi release code to detect when the rpi value is the replaced
RPI_ALLOC_ERROR value and ignore further release steps.

Link: https://lore.kernel.org/r/20191105005708.7399-2-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 294f041961a8..660f96218b25 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -18247,6 +18247,13 @@ lpfc_sli4_alloc_rpi(struct lpfc_hba *phba)
 static void
 __lpfc_sli4_free_rpi(struct lpfc_hba *phba, int rpi)
 {
+	/*
+	 * if the rpi value indicates a prior unreg has already
+	 * been done, skip the unreg.
+	 */
+	if (rpi == LPFC_RPI_ALLOC_ERROR)
+		return;
+
 	if (test_and_clear_bit(rpi, phba->sli4_hba.rpi_bmask)) {
 		phba->sli4_hba.rpi_count--;
 		phba->sli4_hba.max_cfg_param.rpi_used--;
-- 
cgit v1.2.3-59-g8ed1b


From 93a4d6f40198dffcca35d9a928c409f9290f1fe0 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Mon, 4 Nov 2019 16:57:05 -0800
Subject: scsi: lpfc: Add registration for CPU Offline/Online events

The recent affinitization didn't address cpu offlining/onlining.  If an
interrupt vector is shared and the low order cpu owning the vector is
offlined, as interrupts are managed, the vector is taken offline. This
causes the other CPUs sharing the vector will hang as they can't get io
completions.

Correct by registering callbacks with the system for Offline/Online
events. When a cpu is taken offline, its eq, which is tied to an interrupt
vector is found. If the cpu is the "owner" of the vector and if the
eq/vector is shared by other CPUs, the eq is placed into a polled mode.
Additionally, code paths that perform io submission on the "sharing CPUs"
will check the eq state and poll for completion after submission of new io
to a wq that uses the eq.

Similarly, when a cpu comes back online and owns an offlined vector, the eq
is taken out of polled mode and rearmed to start driving interrupts for eq.

Link: https://lore.kernel.org/r/20191105005708.7399-9-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc.h      |   7 ++
 drivers/scsi/lpfc/lpfc_crtn.h |   6 ++
 drivers/scsi/lpfc/lpfc_init.c | 202 +++++++++++++++++++++++++++++++++++++++++-
 drivers/scsi/lpfc/lpfc_sli.c  | 164 ++++++++++++++++++++++++++++++++--
 drivers/scsi/lpfc/lpfc_sli4.h |  21 ++++-
 5 files changed, 388 insertions(+), 12 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc.h b/drivers/scsi/lpfc/lpfc.h
index 4559f1700c6d..88d5fd99f5ec 100644
--- a/drivers/scsi/lpfc/lpfc.h
+++ b/drivers/scsi/lpfc/lpfc.h
@@ -1215,6 +1215,13 @@ struct lpfc_hba {
 	uint64_t ktime_seg10_min;
 	uint64_t ktime_seg10_max;
 #endif
+
+	struct hlist_node cpuhp;	/* used for cpuhp per hba callback */
+	struct timer_list cpuhp_poll_timer;
+	struct list_head poll_list;	/* slowpath eq polling list */
+#define LPFC_POLL_HB	1		/* slowpath heartbeat */
+#define LPFC_POLL_FASTPATH	0	/* called from fastpath */
+#define LPFC_POLL_SLOWPATH	1	/* called from slowpath */
 };
 
 static inline struct Scsi_Host *
diff --git a/drivers/scsi/lpfc/lpfc_crtn.h b/drivers/scsi/lpfc/lpfc_crtn.h
index 6e09fd98a922..d91aa5330306 100644
--- a/drivers/scsi/lpfc/lpfc_crtn.h
+++ b/drivers/scsi/lpfc/lpfc_crtn.h
@@ -215,6 +215,12 @@ irqreturn_t lpfc_sli_fp_intr_handler(int, void *);
 irqreturn_t lpfc_sli4_intr_handler(int, void *);
 irqreturn_t lpfc_sli4_hba_intr_handler(int, void *);
 
+inline void lpfc_sli4_cleanup_poll_list(struct lpfc_hba *phba);
+int lpfc_sli4_poll_eq(struct lpfc_queue *q, uint8_t path);
+void lpfc_sli4_poll_hbtimer(struct timer_list *t);
+void lpfc_sli4_start_polling(struct lpfc_queue *q);
+void lpfc_sli4_stop_polling(struct lpfc_queue *q);
+
 void lpfc_read_rev(struct lpfc_hba *, LPFC_MBOXQ_t *);
 void lpfc_sli4_swap_str(struct lpfc_hba *, LPFC_MBOXQ_t *);
 void lpfc_config_ring(struct lpfc_hba *, int, LPFC_MBOXQ_t *);
diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c
index ec9dbc042a41..888ad32d5267 100644
--- a/drivers/scsi/lpfc/lpfc_init.c
+++ b/drivers/scsi/lpfc/lpfc_init.c
@@ -40,6 +40,7 @@
 #include <linux/irq.h>
 #include <linux/bitops.h>
 #include <linux/crash_dump.h>
+#include <linux/cpuhotplug.h>
 
 #include <scsi/scsi.h>
 #include <scsi/scsi_device.h>
@@ -66,9 +67,13 @@
 #include "lpfc_version.h"
 #include "lpfc_ids.h"
 
+static enum cpuhp_state lpfc_cpuhp_state;
 /* Used when mapping IRQ vectors in a driver centric manner */
 static uint32_t lpfc_present_cpu;
 
+static void __lpfc_cpuhp_remove(struct lpfc_hba *phba);
+static void lpfc_cpuhp_remove(struct lpfc_hba *phba);
+static void lpfc_cpuhp_add(struct lpfc_hba *phba);
 static void lpfc_get_hba_model_desc(struct lpfc_hba *, uint8_t *, uint8_t *);
 static int lpfc_post_rcv_buf(struct lpfc_hba *);
 static int lpfc_sli4_queue_verify(struct lpfc_hba *);
@@ -3379,6 +3384,8 @@ lpfc_online(struct lpfc_hba *phba)
 	if (phba->cfg_xri_rebalancing)
 		lpfc_create_multixri_pools(phba);
 
+	lpfc_cpuhp_add(phba);
+
 	lpfc_unblock_mgmt_io(phba);
 	return 0;
 }
@@ -3542,6 +3549,7 @@ lpfc_offline(struct lpfc_hba *phba)
 			spin_unlock_irq(shost->host_lock);
 		}
 	lpfc_destroy_vport_work_array(phba, vports);
+	__lpfc_cpuhp_remove(phba);
 
 	if (phba->cfg_xri_rebalancing)
 		lpfc_destroy_multixri_pools(phba);
@@ -9255,6 +9263,8 @@ lpfc_sli4_queue_destroy(struct lpfc_hba *phba)
 	}
 	spin_unlock_irq(&phba->hbalock);
 
+	lpfc_sli4_cleanup_poll_list(phba);
+
 	/* Release HBA eqs */
 	if (phba->sli4_hba.hdwq)
 		lpfc_sli4_release_hdwq(phba);
@@ -11057,6 +11067,170 @@ found_any:
 	return;
 }
 
+/**
+ * lpfc_cpuhp_get_eq
+ *
+ * @phba:   pointer to lpfc hba data structure.
+ * @cpu:    cpu going offline
+ * @eqlist:
+ */
+static void
+lpfc_cpuhp_get_eq(struct lpfc_hba *phba, unsigned int cpu,
+		  struct list_head *eqlist)
+{
+	struct lpfc_vector_map_info *map;
+	const struct cpumask *maskp;
+	struct lpfc_queue *eq;
+	unsigned int i;
+	cpumask_t tmp;
+	u16 idx;
+
+	for (idx = 0; idx < phba->cfg_irq_chann; idx++) {
+		maskp = pci_irq_get_affinity(phba->pcidev, idx);
+		if (!maskp)
+			continue;
+		/*
+		 * if irq is not affinitized to the cpu going
+		 * then we don't need to poll the eq attached
+		 * to it.
+		 */
+		if (!cpumask_and(&tmp, maskp, cpumask_of(cpu)))
+			continue;
+		/* get the cpus that are online and are affini-
+		 * tized to this irq vector.  If the count is
+		 * more than 1 then cpuhp is not going to shut-
+		 * down this vector.  Since this cpu has not
+		 * gone offline yet, we need >1.
+		 */
+		cpumask_and(&tmp, maskp, cpu_online_mask);
+		if (cpumask_weight(&tmp) > 1)
+			continue;
+
+		/* Now that we have an irq to shutdown, get the eq
+		 * mapped to this irq.  Note: multiple hdwq's in
+		 * the software can share an eq, but eventually
+		 * only eq will be mapped to this vector
+		 */
+		for_each_possible_cpu(i) {
+			map = &phba->sli4_hba.cpu_map[i];
+			if (!(map->irq == pci_irq_vector(phba->pcidev, idx)))
+				continue;
+			eq = phba->sli4_hba.hdwq[map->hdwq].hba_eq;
+			list_add(&eq->_poll_list, eqlist);
+			/* 1 is good enough. others will be a copy of this */
+			break;
+		}
+	}
+}
+
+static void __lpfc_cpuhp_remove(struct lpfc_hba *phba)
+{
+	if (phba->sli_rev != LPFC_SLI_REV4)
+		return;
+
+	cpuhp_state_remove_instance_nocalls(lpfc_cpuhp_state,
+					    &phba->cpuhp);
+	/*
+	 * unregistering the instance doesn't stop the polling
+	 * timer. Wait for the poll timer to retire.
+	 */
+	synchronize_rcu();
+	del_timer_sync(&phba->cpuhp_poll_timer);
+}
+
+static void lpfc_cpuhp_remove(struct lpfc_hba *phba)
+{
+	if (phba->pport->fc_flag & FC_OFFLINE_MODE)
+		return;
+
+	__lpfc_cpuhp_remove(phba);
+}
+
+static void lpfc_cpuhp_add(struct lpfc_hba *phba)
+{
+	if (phba->sli_rev != LPFC_SLI_REV4)
+		return;
+
+	rcu_read_lock();
+
+	if (!list_empty(&phba->poll_list)) {
+		timer_setup(&phba->cpuhp_poll_timer, lpfc_sli4_poll_hbtimer, 0);
+		mod_timer(&phba->cpuhp_poll_timer,
+			  jiffies + msecs_to_jiffies(LPFC_POLL_HB));
+	}
+
+	rcu_read_unlock();
+
+	cpuhp_state_add_instance_nocalls(lpfc_cpuhp_state,
+					 &phba->cpuhp);
+}
+
+static int __lpfc_cpuhp_checks(struct lpfc_hba *phba, int *retval)
+{
+	if (phba->pport->load_flag & FC_UNLOADING) {
+		*retval = -EAGAIN;
+		return true;
+	}
+
+	if (phba->sli_rev != LPFC_SLI_REV4) {
+		*retval = 0;
+		return true;
+	}
+
+	/* proceed with the hotplug */
+	return false;
+}
+
+static int lpfc_cpu_offline(unsigned int cpu, struct hlist_node *node)
+{
+	struct lpfc_hba *phba = hlist_entry_safe(node, struct lpfc_hba, cpuhp);
+	struct lpfc_queue *eq, *next;
+	LIST_HEAD(eqlist);
+	int retval;
+
+	if (!phba) {
+		WARN_ONCE(!phba, "cpu: %u. phba:NULL", raw_smp_processor_id());
+		return 0;
+	}
+
+	if (__lpfc_cpuhp_checks(phba, &retval))
+		return retval;
+
+	lpfc_cpuhp_get_eq(phba, cpu, &eqlist);
+
+	/* start polling on these eq's */
+	list_for_each_entry_safe(eq, next, &eqlist, _poll_list) {
+		list_del_init(&eq->_poll_list);
+		lpfc_sli4_start_polling(eq);
+	}
+
+	return 0;
+}
+
+static int lpfc_cpu_online(unsigned int cpu, struct hlist_node *node)
+{
+	struct lpfc_hba *phba = hlist_entry_safe(node, struct lpfc_hba, cpuhp);
+	struct lpfc_queue *eq, *next;
+	unsigned int n;
+	int retval;
+
+	if (!phba) {
+		WARN_ONCE(!phba, "cpu: %u. phba:NULL", raw_smp_processor_id());
+		return 0;
+	}
+
+	if (__lpfc_cpuhp_checks(phba, &retval))
+		return retval;
+
+	list_for_each_entry_safe(eq, next, &phba->poll_list, _poll_list) {
+		n = lpfc_find_cpu_handle(phba, eq->hdwq, LPFC_FIND_BY_HDWQ);
+		if (n == cpu)
+			lpfc_sli4_stop_polling(eq);
+	}
+
+	return 0;
+}
+
 /**
  * lpfc_sli4_enable_msix - Enable MSI-X interrupt mode to SLI-4 device
  * @phba: pointer to lpfc hba data structure.
@@ -11460,6 +11634,9 @@ lpfc_sli4_hba_unset(struct lpfc_hba *phba)
 	/* Wait for completion of device XRI exchange busy */
 	lpfc_sli4_xri_exchange_busy_wait(phba);
 
+	/* per-phba callback de-registration for hotplug event */
+	lpfc_cpuhp_remove(phba);
+
 	/* Disable PCI subsystem interrupt */
 	lpfc_sli4_disable_intr(phba);
 
@@ -12752,6 +12929,9 @@ lpfc_pci_probe_one_s4(struct pci_dev *pdev, const struct pci_device_id *pid)
 	/* Enable RAS FW log support */
 	lpfc_sli4_ras_setup(phba);
 
+	INIT_LIST_HEAD(&phba->poll_list);
+	cpuhp_state_add_instance_nocalls(lpfc_cpuhp_state, &phba->cpuhp);
+
 	return 0;
 
 out_free_sysfs_attr:
@@ -13569,11 +13749,24 @@ lpfc_init(void)
 	/* Initialize in case vector mapping is needed */
 	lpfc_present_cpu = num_present_cpus();
 
+	error = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN,
+					"lpfc/sli4:online",
+					lpfc_cpu_online, lpfc_cpu_offline);
+	if (error < 0)
+		goto cpuhp_failure;
+	lpfc_cpuhp_state = error;
+
 	error = pci_register_driver(&lpfc_driver);
-	if (error) {
-		fc_release_transport(lpfc_transport_template);
-		fc_release_transport(lpfc_vport_transport_template);
-	}
+	if (error)
+		goto unwind;
+
+	return error;
+
+unwind:
+	cpuhp_remove_multi_state(lpfc_cpuhp_state);
+cpuhp_failure:
+	fc_release_transport(lpfc_transport_template);
+	fc_release_transport(lpfc_vport_transport_template);
 
 	return error;
 }
@@ -13590,6 +13783,7 @@ lpfc_exit(void)
 {
 	misc_deregister(&lpfc_mgmt_dev);
 	pci_unregister_driver(&lpfc_driver);
+	cpuhp_remove_multi_state(lpfc_cpuhp_state);
 	fc_release_transport(lpfc_transport_template);
 	fc_release_transport(lpfc_vport_transport_template);
 	idr_destroy(&lpfc_hba_index);
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 660f96218b25..f8de313d592c 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -515,7 +515,8 @@ lpfc_sli4_eqcq_flush(struct lpfc_hba *phba, struct lpfc_queue *eq)
 }
 
 static int
-lpfc_sli4_process_eq(struct lpfc_hba *phba, struct lpfc_queue *eq)
+lpfc_sli4_process_eq(struct lpfc_hba *phba, struct lpfc_queue *eq,
+		     uint8_t rearm)
 {
 	struct lpfc_eqe *eqe;
 	int count = 0, consumed = 0;
@@ -549,8 +550,8 @@ lpfc_sli4_process_eq(struct lpfc_hba *phba, struct lpfc_queue *eq)
 	eq->queue_claimed = 0;
 
 rearm_and_exit:
-	/* Always clear and re-arm the EQ */
-	phba->sli4_hba.sli4_write_eq_db(phba, eq, consumed, LPFC_QUEUE_REARM);
+	/* Always clear the EQ. */
+	phba->sli4_hba.sli4_write_eq_db(phba, eq, consumed, rearm);
 
 	return count;
 }
@@ -7948,7 +7949,7 @@ lpfc_sli4_process_missed_mbox_completions(struct lpfc_hba *phba)
 
 	if (mbox_pending)
 		/* process and rearm the EQ */
-		lpfc_sli4_process_eq(phba, fpeq);
+		lpfc_sli4_process_eq(phba, fpeq, LPFC_QUEUE_REARM);
 	else
 		/* Always clear and re-arm the EQ */
 		sli4_hba->sli4_write_eq_db(phba, fpeq, 0, LPFC_QUEUE_REARM);
@@ -10113,10 +10114,13 @@ lpfc_sli_issue_iocb(struct lpfc_hba *phba, uint32_t ring_number,
 		    struct lpfc_iocbq *piocb, uint32_t flag)
 {
 	struct lpfc_sli_ring *pring;
+	struct lpfc_queue *eq;
 	unsigned long iflags;
 	int rc;
 
 	if (phba->sli_rev == LPFC_SLI_REV4) {
+		eq = phba->sli4_hba.hdwq[piocb->hba_wqidx].hba_eq;
+
 		pring = lpfc_sli4_calc_ring(phba, piocb);
 		if (unlikely(pring == NULL))
 			return IOCB_ERROR;
@@ -10124,6 +10128,8 @@ lpfc_sli_issue_iocb(struct lpfc_hba *phba, uint32_t ring_number,
 		spin_lock_irqsave(&pring->ring_lock, iflags);
 		rc = __lpfc_sli_issue_iocb(phba, ring_number, piocb, flag);
 		spin_unlock_irqrestore(&pring->ring_lock, iflags);
+
+		lpfc_sli4_poll_eq(eq, LPFC_POLL_FASTPATH);
 	} else {
 		/* For now, SLI2/3 will still use hbalock */
 		spin_lock_irqsave(&phba->hbalock, iflags);
@@ -14302,7 +14308,7 @@ lpfc_sli4_hba_intr_handler(int irq, void *dev_id)
 		lpfc_sli4_mod_hba_eq_delay(phba, fpeq, LPFC_MAX_AUTO_EQ_DELAY);
 
 	/* process and rearm the EQ */
-	ecount = lpfc_sli4_process_eq(phba, fpeq);
+	ecount = lpfc_sli4_process_eq(phba, fpeq, LPFC_QUEUE_REARM);
 
 	if (unlikely(ecount == 0)) {
 		fpeq->EQ_no_entry++;
@@ -14362,6 +14368,147 @@ lpfc_sli4_intr_handler(int irq, void *dev_id)
 	return (hba_handled == true) ? IRQ_HANDLED : IRQ_NONE;
 } /* lpfc_sli4_intr_handler */
 
+void lpfc_sli4_poll_hbtimer(struct timer_list *t)
+{
+	struct lpfc_hba *phba = from_timer(phba, t, cpuhp_poll_timer);
+	struct lpfc_queue *eq;
+	int i = 0;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(eq, &phba->poll_list, _poll_list)
+		i += lpfc_sli4_poll_eq(eq, LPFC_POLL_SLOWPATH);
+	if (!list_empty(&phba->poll_list))
+		mod_timer(&phba->cpuhp_poll_timer,
+			  jiffies + msecs_to_jiffies(LPFC_POLL_HB));
+
+	rcu_read_unlock();
+}
+
+inline int lpfc_sli4_poll_eq(struct lpfc_queue *eq, uint8_t path)
+{
+	struct lpfc_hba *phba = eq->phba;
+	int i = 0;
+
+	/*
+	 * Unlocking an irq is one of the entry point to check
+	 * for re-schedule, but we are good for io submission
+	 * path as midlayer does a get_cpu to glue us in. Flush
+	 * out the invalidate queue so we can see the updated
+	 * value for flag.
+	 */
+	smp_rmb();
+
+	if (READ_ONCE(eq->mode) == LPFC_EQ_POLL)
+		/* We will not likely get the completion for the caller
+		 * during this iteration but i guess that's fine.
+		 * Future io's coming on this eq should be able to
+		 * pick it up.  As for the case of single io's, they
+		 * will be handled through a sched from polling timer
+		 * function which is currently triggered every 1msec.
+		 */
+		i = lpfc_sli4_process_eq(phba, eq, LPFC_QUEUE_NOARM);
+
+	return i;
+}
+
+static inline void lpfc_sli4_add_to_poll_list(struct lpfc_queue *eq)
+{
+	struct lpfc_hba *phba = eq->phba;
+
+	if (list_empty(&phba->poll_list)) {
+		timer_setup(&phba->cpuhp_poll_timer, lpfc_sli4_poll_hbtimer, 0);
+		/* kickstart slowpath processing for this eq */
+		mod_timer(&phba->cpuhp_poll_timer,
+			  jiffies + msecs_to_jiffies(LPFC_POLL_HB));
+	}
+
+	list_add_rcu(&eq->_poll_list, &phba->poll_list);
+	synchronize_rcu();
+}
+
+static inline void lpfc_sli4_remove_from_poll_list(struct lpfc_queue *eq)
+{
+	struct lpfc_hba *phba = eq->phba;
+
+	/* Disable slowpath processing for this eq.  Kick start the eq
+	 * by RE-ARMING the eq's ASAP
+	 */
+	list_del_rcu(&eq->_poll_list);
+	synchronize_rcu();
+
+	if (list_empty(&phba->poll_list))
+		del_timer_sync(&phba->cpuhp_poll_timer);
+}
+
+inline void lpfc_sli4_cleanup_poll_list(struct lpfc_hba *phba)
+{
+	struct lpfc_queue *eq, *next;
+
+	list_for_each_entry_safe(eq, next, &phba->poll_list, _poll_list)
+		list_del(&eq->_poll_list);
+
+	INIT_LIST_HEAD(&phba->poll_list);
+	synchronize_rcu();
+}
+
+static inline void
+__lpfc_sli4_switch_eqmode(struct lpfc_queue *eq, uint8_t mode)
+{
+	if (mode == eq->mode)
+		return;
+	/*
+	 * currently this function is only called during a hotplug
+	 * event and the cpu on which this function is executing
+	 * is going offline.  By now the hotplug has instructed
+	 * the scheduler to remove this cpu from cpu active mask.
+	 * So we don't need to work about being put aside by the
+	 * scheduler for a high priority process.  Yes, the inte-
+	 * rrupts could come but they are known to retire ASAP.
+	 */
+
+	/* Disable polling in the fastpath */
+	WRITE_ONCE(eq->mode, mode);
+	/* flush out the store buffer */
+	smp_wmb();
+
+	/*
+	 * Add this eq to the polling list and start polling. For
+	 * a grace period both interrupt handler and poller will
+	 * try to process the eq _but_ that's fine.  We have a
+	 * synchronization mechanism in place (queue_claimed) to
+	 * deal with it.  This is just a draining phase for int-
+	 * errupt handler (not eq's) as we have guranteed through
+	 * barrier that all the CPUs have seen the new CQ_POLLED
+	 * state. which will effectively disable the REARMING of
+	 * the EQ.  The whole idea is eq's die off eventually as
+	 * we are not rearming EQ's anymore.
+	 */
+	mode ? lpfc_sli4_add_to_poll_list(eq) :
+	       lpfc_sli4_remove_from_poll_list(eq);
+}
+
+void lpfc_sli4_start_polling(struct lpfc_queue *eq)
+{
+	__lpfc_sli4_switch_eqmode(eq, LPFC_EQ_POLL);
+}
+
+void lpfc_sli4_stop_polling(struct lpfc_queue *eq)
+{
+	struct lpfc_hba *phba = eq->phba;
+
+	__lpfc_sli4_switch_eqmode(eq, LPFC_EQ_INTERRUPT);
+
+	/* Kick start for the pending io's in h/w.
+	 * Once we switch back to interrupt processing on a eq
+	 * the io path completion will only arm eq's when it
+	 * receives a completion.  But since eq's are in disa-
+	 * rmed state it doesn't receive a completion.  This
+	 * creates a deadlock scenaro.
+	 */
+	phba->sli4_hba.sli4_write_eq_db(phba, eq, 0, LPFC_QUEUE_REARM);
+}
+
 /**
  * lpfc_sli4_queue_free - free a queue structure and associated memory
  * @queue: The queue structure to free.
@@ -14436,6 +14583,7 @@ lpfc_sli4_queue_alloc(struct lpfc_hba *phba, uint32_t page_size,
 		return NULL;
 
 	INIT_LIST_HEAD(&queue->list);
+	INIT_LIST_HEAD(&queue->_poll_list);
 	INIT_LIST_HEAD(&queue->wq_list);
 	INIT_LIST_HEAD(&queue->wqfull_list);
 	INIT_LIST_HEAD(&queue->page_list);
@@ -19757,6 +19905,8 @@ lpfc_sli4_issue_wqe(struct lpfc_hba *phba, struct lpfc_sli4_hdw_queue *qp,
 
 		lpfc_sli_ringtxcmpl_put(phba, pring, pwqe);
 		spin_unlock_irqrestore(&pring->ring_lock, iflags);
+
+		lpfc_sli4_poll_eq(qp->hba_eq, LPFC_POLL_FASTPATH);
 		return 0;
 	}
 
@@ -19777,6 +19927,8 @@ lpfc_sli4_issue_wqe(struct lpfc_hba *phba, struct lpfc_sli4_hdw_queue *qp,
 		}
 		lpfc_sli_ringtxcmpl_put(phba, pring, pwqe);
 		spin_unlock_irqrestore(&pring->ring_lock, iflags);
+
+		lpfc_sli4_poll_eq(qp->hba_eq, LPFC_POLL_FASTPATH);
 		return 0;
 	}
 
@@ -19805,6 +19957,8 @@ lpfc_sli4_issue_wqe(struct lpfc_hba *phba, struct lpfc_sli4_hdw_queue *qp,
 		}
 		lpfc_sli_ringtxcmpl_put(phba, pring, pwqe);
 		spin_unlock_irqrestore(&pring->ring_lock, iflags);
+
+		lpfc_sli4_poll_eq(qp->hba_eq, LPFC_POLL_FASTPATH);
 		return 0;
 	}
 	return WQE_ERROR;
diff --git a/drivers/scsi/lpfc/lpfc_sli4.h b/drivers/scsi/lpfc/lpfc_sli4.h
index bbe24c19b1d9..ef32159248d7 100644
--- a/drivers/scsi/lpfc/lpfc_sli4.h
+++ b/drivers/scsi/lpfc/lpfc_sli4.h
@@ -133,6 +133,23 @@ struct lpfc_rqb {
 struct lpfc_queue {
 	struct list_head list;
 	struct list_head wq_list;
+
+	/*
+	 * If interrupts are in effect on _all_ the eq's the footprint
+	 * of polling code is zero (except mode). This memory is chec-
+	 * ked for every io to see if the io needs to be polled and
+	 * while completion to check if the eq's needs to be rearmed.
+	 * Keep in same cacheline as the queue ptr to avoid cpu fetch
+	 * stalls. Using 1B memory will leave us with 7B hole. Fill
+	 * it with other frequently used members.
+	 */
+	uint16_t last_cpu;	/* most recent cpu */
+	uint16_t hdwq;
+	uint8_t	 qe_valid;
+	uint8_t  mode;	/* interrupt or polling */
+#define LPFC_EQ_INTERRUPT	0
+#define LPFC_EQ_POLL		1
+
 	struct list_head wqfull_list;
 	enum lpfc_sli4_queue_type type;
 	enum lpfc_sli4_queue_subtype subtype;
@@ -240,10 +257,8 @@ struct lpfc_queue {
 	struct delayed_work	sched_spwork;
 
 	uint64_t isr_timestamp;
-	uint16_t hdwq;
-	uint16_t last_cpu;	/* most recent cpu */
-	uint8_t	qe_valid;
 	struct lpfc_queue *assoc_qp;
+	struct list_head _poll_list;
 	void **q_pgs;	/* array to index entries per page */
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 171f6c41949f6e9d5e09dcac842a10bf8dda8dcc Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Mon, 4 Nov 2019 16:57:07 -0800
Subject: scsi: lpfc: Add enablement of multiple adapter dumps

Some adapters support the ability to hold multiple adapter dumps on the
adapter flash. Some adapters default to enabling this feature while others
default to single-dump.

Make support uniform by enabling dual dump by default.

Link: https://lore.kernel.org/r/20191105005708.7399-11-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_hw4.h | 10 ++++++++++
 drivers/scsi/lpfc/lpfc_sli.c | 27 ++++++++++++++++++++++++++-
 2 files changed, 36 insertions(+), 1 deletion(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_hw4.h b/drivers/scsi/lpfc/lpfc_hw4.h
index 9daa2b494b5c..25cdcbc2b02f 100644
--- a/drivers/scsi/lpfc/lpfc_hw4.h
+++ b/drivers/scsi/lpfc/lpfc_hw4.h
@@ -3530,6 +3530,7 @@ struct lpfc_sli4_parameters {
 
 #define LPFC_SET_UE_RECOVERY		0x10
 #define LPFC_SET_MDS_DIAGS		0x11
+#define LPFC_SET_DUAL_DUMP		0x1e
 struct lpfc_mbx_set_feature {
 	struct mbox_header header;
 	uint32_t feature;
@@ -3544,6 +3545,15 @@ struct lpfc_mbx_set_feature {
 #define lpfc_mbx_set_feature_mds_deep_loopbk_SHIFT  1
 #define lpfc_mbx_set_feature_mds_deep_loopbk_MASK   0x00000001
 #define lpfc_mbx_set_feature_mds_deep_loopbk_WORD   word6
+#define lpfc_mbx_set_feature_dd_SHIFT		0
+#define lpfc_mbx_set_feature_dd_MASK		0x00000001
+#define lpfc_mbx_set_feature_dd_WORD		word6
+#define lpfc_mbx_set_feature_ddquery_SHIFT	1
+#define lpfc_mbx_set_feature_ddquery_MASK	0x00000001
+#define lpfc_mbx_set_feature_ddquery_WORD	word6
+#define LPFC_DISABLE_DUAL_DUMP		0
+#define LPFC_ENABLE_DUAL_DUMP		1
+#define LPFC_QUERY_OP_DUAL_DUMP		2
 	uint32_t word7;
 #define lpfc_mbx_set_feature_UERP_SHIFT 0
 #define lpfc_mbx_set_feature_UERP_MASK  0x0000ffff
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index f8de313d592c..fad890cea21a 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -6203,6 +6203,14 @@ lpfc_set_features(struct lpfc_hba *phba, LPFC_MBOXQ_t *mbox,
 		mbox->u.mqe.un.set_feature.feature = LPFC_SET_MDS_DIAGS;
 		mbox->u.mqe.un.set_feature.param_len = 8;
 		break;
+	case LPFC_SET_DUAL_DUMP:
+		bf_set(lpfc_mbx_set_feature_dd,
+		       &mbox->u.mqe.un.set_feature, LPFC_ENABLE_DUAL_DUMP);
+		bf_set(lpfc_mbx_set_feature_ddquery,
+		       &mbox->u.mqe.un.set_feature, 0);
+		mbox->u.mqe.un.set_feature.feature = LPFC_SET_DUAL_DUMP;
+		mbox->u.mqe.un.set_feature.param_len = 4;
+		break;
 	}
 
 	return;
@@ -7200,7 +7208,7 @@ lpfc_post_rq_buffer(struct lpfc_hba *phba, struct lpfc_queue *hrq,
 int
 lpfc_sli4_hba_setup(struct lpfc_hba *phba)
 {
-	int rc, i, cnt, len;
+	int rc, i, cnt, len, dd;
 	LPFC_MBOXQ_t *mboxq;
 	struct lpfc_mqe *mqe;
 	uint8_t *vpd;
@@ -7451,6 +7459,23 @@ lpfc_sli4_hba_setup(struct lpfc_hba *phba)
 	phba->sli3_options |= (LPFC_SLI3_NPIV_ENABLED | LPFC_SLI3_HBQ_ENABLED);
 	spin_unlock_irq(&phba->hbalock);
 
+	/* Always try to enable dual dump feature if we can */
+	lpfc_set_features(phba, mboxq, LPFC_SET_DUAL_DUMP);
+	rc = lpfc_sli_issue_mbox(phba, mboxq, MBX_POLL);
+	dd = bf_get(lpfc_mbx_set_feature_dd, &mboxq->u.mqe.un.set_feature);
+	if ((rc == MBX_SUCCESS) && (dd == LPFC_ENABLE_DUAL_DUMP))
+		lpfc_printf_log(phba, KERN_ERR, LOG_SLI | LOG_INIT,
+				"6448 Dual Dump is enabled\n");
+	else
+		lpfc_printf_log(phba, KERN_INFO, LOG_SLI | LOG_INIT,
+				"6447 Dual Dump Mailbox x%x (x%x/x%x) failed, "
+				"rc:x%x dd:x%x\n",
+				bf_get(lpfc_mqe_command, &mboxq->u.mqe),
+				lpfc_sli_config_mbox_subsys_get(
+					phba, mboxq),
+				lpfc_sli_config_mbox_opcode_get(
+					phba, mboxq),
+				rc, dd);
 	/*
 	 * Allocate all resources (xri,rpi,vpi,vfi) now.  Subsequent
 	 * calls depends on these resources to complete port setup.
-- 
cgit v1.2.3-59-g8ed1b


From 765ab6cdac3b681952da0e22184bf6cf1ae41cf8 Mon Sep 17 00:00:00 2001
From: Bart Van Assche <bvanassche@acm.org>
Date: Wed, 6 Nov 2019 21:21:54 -0800
Subject: scsi: lpfc: Fix a kernel warning triggered by lpfc_get_sgl_per_hdwq()

Fix the following kernel bug report:

BUG: using smp_processor_id() in preemptible [00000000] code: systemd-udevd/954

Fixes: d79c9e9d4b3d ("scsi: lpfc: Support dynamic unbounded SGL lists on G7 hardware.")
Link: https://lore.kernel.org/r/20191107052158.25788-2-bvanassche@acm.org
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index fad890cea21a..613fbf4a7da9 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -20668,7 +20668,7 @@ lpfc_get_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 		/* allocate more */
 		spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 		tmp = kmalloc_node(sizeof(*tmp), GFP_ATOMIC,
-				   cpu_to_node(smp_processor_id()));
+				   cpu_to_node(raw_smp_processor_id()));
 		if (!tmp) {
 			lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
 					"8353 error kmalloc memory for HDWQ "
@@ -20811,7 +20811,7 @@ lpfc_get_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 		/* allocate more */
 		spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 		tmp = kmalloc_node(sizeof(*tmp), GFP_ATOMIC,
-				   cpu_to_node(smp_processor_id()));
+				   cpu_to_node(raw_smp_processor_id()));
 		if (!tmp) {
 			lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
 					"8355 error kmalloc memory for HDWQ "
-- 
cgit v1.2.3-59-g8ed1b


From d480e57809a043333a3b9e755c0bdd43e10a9f12 Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Mon, 11 Nov 2019 15:03:58 -0800
Subject: scsi: lpfc: fix inlining of lpfc_sli4_cleanup_poll_list()

Compilation can fail due to having an inline function reference where the
function body is not present.

Fix by removing the inline tag.

Fixes: 93a4d6f40198 ("scsi: lpfc: Add registration for CPU Offline/Online events")

Link: https://lore.kernel.org/r/20191111230401.12958-4-jsmart2021@gmail.com
Reviewed-by: Ewan D. Milne <emilne@redhat.com>
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_crtn.h | 2 +-
 drivers/scsi/lpfc/lpfc_sli.c  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_crtn.h b/drivers/scsi/lpfc/lpfc_crtn.h
index d91aa5330306..ee353c84a097 100644
--- a/drivers/scsi/lpfc/lpfc_crtn.h
+++ b/drivers/scsi/lpfc/lpfc_crtn.h
@@ -215,7 +215,7 @@ irqreturn_t lpfc_sli_fp_intr_handler(int, void *);
 irqreturn_t lpfc_sli4_intr_handler(int, void *);
 irqreturn_t lpfc_sli4_hba_intr_handler(int, void *);
 
-inline void lpfc_sli4_cleanup_poll_list(struct lpfc_hba *phba);
+void lpfc_sli4_cleanup_poll_list(struct lpfc_hba *phba);
 int lpfc_sli4_poll_eq(struct lpfc_queue *q, uint8_t path);
 void lpfc_sli4_poll_hbtimer(struct timer_list *t);
 void lpfc_sli4_start_polling(struct lpfc_queue *q);
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 613fbf4a7da9..5286c78645ac 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -14466,7 +14466,7 @@ static inline void lpfc_sli4_remove_from_poll_list(struct lpfc_queue *eq)
 		del_timer_sync(&phba->cpuhp_poll_timer);
 }
 
-inline void lpfc_sli4_cleanup_poll_list(struct lpfc_hba *phba)
+void lpfc_sli4_cleanup_poll_list(struct lpfc_hba *phba)
 {
 	struct lpfc_queue *eq, *next;
 
-- 
cgit v1.2.3-59-g8ed1b


From 4583a4f66b323c6e4d774be2649e83a4e7c7b78c Mon Sep 17 00:00:00 2001
From: James Smart <jsmart2021@gmail.com>
Date: Fri, 15 Nov 2019 16:38:47 -0800
Subject: scsi: lpfc: use hdwq assigned cpu for allocation

Looking at the recent conversion from smp_processor_id() to
raw_smp_processor_id(), realized that the allocation should be based on the
cpu the hdwq is bound to, not the executing cpu.

Revise to pull cpu number from the hdwq

Fixes: 765ab6cdac3b ("scsi: lpfc: Fix a kernel warning triggered by lpfc_get_sgl_per_hdwq()")
Link: https://lore.kernel.org/r/20191116003847.6141-1-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'drivers/scsi/lpfc/lpfc_sli.c')

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 5286c78645ac..7c8527bd1677 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -20668,7 +20668,7 @@ lpfc_get_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 		/* allocate more */
 		spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 		tmp = kmalloc_node(sizeof(*tmp), GFP_ATOMIC,
-				   cpu_to_node(raw_smp_processor_id()));
+				   cpu_to_node(hdwq->io_wq->chann));
 		if (!tmp) {
 			lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
 					"8353 error kmalloc memory for HDWQ "
@@ -20811,7 +20811,7 @@ lpfc_get_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 		/* allocate more */
 		spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 		tmp = kmalloc_node(sizeof(*tmp), GFP_ATOMIC,
-				   cpu_to_node(raw_smp_processor_id()));
+				   cpu_to_node(hdwq->io_wq->chann));
 		if (!tmp) {
 			lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
 					"8355 error kmalloc memory for HDWQ "
-- 
cgit v1.2.3-59-g8ed1b