From: Quinn Tran <qutran@marvell.com>

Date: Wed, 15 Jun 2022 22:35:00 -0700

Subject: scsi: qla2xxx: Wind down adapter after PCIe error

Patch-mainline: v5.20-rc1

Git-commit: d3117c83ba316b3200d9f2fe900f2b9a5525a25c

References: bsc#1201958

Put adapter into a wind down state if OS does not make any attempt to

recover the adapter after PCIe error.

Link: https://lore.kernel.org/r/20220616053508.27186-4-njavali@marvell.com

Cc: stable@vger.kernel.org

Signed-off-by: Quinn Tran <qutran@marvell.com>

Signed-off-by: Nilesh Javali <njavali@marvell.com>

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

Acked-by: Daniel Wagner <dwagner@suse.de>

---

 drivers/scsi/qla2xxx/qla_bsg.c  |   10 +++++++-

 drivers/scsi/qla2xxx/qla_def.h  |    4 +++

 drivers/scsi/qla2xxx/qla_init.c |   20 ++++++++++++++++

 drivers/scsi/qla2xxx/qla_os.c   |   48 ++++++++++++++++++++++++++++++++++++++++

 4 files changed, 81 insertions(+), 1 deletion(-)

--- a/drivers/scsi/qla2xxx/qla_bsg.c

+++ b/drivers/scsi/qla2xxx/qla_bsg.c

@@ -3062,6 +3062,13 @@ qla24xx_bsg_timeout(struct bsg_job *bsg_

 	ql_log(ql_log_info, vha, 0x708b, "%s CMD timeout. bsg ptr %p.\n",

 	    __func__, bsg_job);

+

+	if (qla2x00_isp_reg_stat(ha)) {

+		ql_log(ql_log_info, vha, 0x9007,

+		    "PCI/Register disconnect.\n");

+		qla_pci_set_eeh_busy(vha);

+	}

+

 	/* find the bsg job from the active list of commands */

 	spin_lock_irqsave(&ha->hardware_lock, flags);

 	for (que = 0; que < ha->max_req_queues; que++) {

@@ -3079,7 +3086,8 @@ qla24xx_bsg_timeout(struct bsg_job *bsg_

 			    sp->u.bsg_job == bsg_job) {

 				req->outstanding_cmds[cnt] = NULL;

 				spin_unlock_irqrestore(&ha->hardware_lock, flags);

-				if (ha->isp_ops->abort_command(sp)) {

+

+				if (!ha->flags.eeh_busy && ha->isp_ops->abort_command(sp)) {

 					ql_log(ql_log_warn, vha, 0x7089,

 					    "mbx abort_command failed.\n");

 					bsg_reply->result = -EIO;

--- a/drivers/scsi/qla2xxx/qla_def.h

+++ b/drivers/scsi/qla2xxx/qla_def.h

@@ -4054,6 +4054,9 @@ struct qla_hw_data {

 		uint32_t	n2n_fw_acc_sec:1;

 		uint32_t	plogi_template_valid:1;

 		uint32_t	port_isolated:1;

+		uint32_t	eeh_flush:2;

+#define EEH_FLUSH_RDY  1

+#define EEH_FLUSH_DONE 2

 	} flags;

 	uint16_t max_exchg;

@@ -4088,6 +4091,7 @@ struct qla_hw_data {

 	uint32_t		rsp_que_len;

 	uint32_t		req_que_off;

 	uint32_t		rsp_que_off;

+	unsigned long		eeh_jif;

 	/* Multi queue data structs */

 	device_reg_t *mqiobase;

--- a/drivers/scsi/qla2xxx/qla_init.c

+++ b/drivers/scsi/qla2xxx/qla_init.c

@@ -48,6 +48,7 @@ qla2x00_sp_timeout(unsigned long __data)

 {

 	srb_t *sp = (srb_t *)__data;

 	struct srb_iocb *iocb;

+	scsi_qla_host_t *vha = sp->vha;

 	WARN_ON(irqs_disabled());

 	iocb = &sp->u.iocb_cmd;

@@ -55,6 +56,12 @@ qla2x00_sp_timeout(unsigned long __data)

 	/* ref: TMR */

 	kref_put(&sp->cmd_kref, qla2x00_sp_release);

+

+	if (vha && qla2x00_isp_reg_stat(vha->hw)) {

+		ql_log(ql_log_info, vha, 0x9008,

+		    "PCI/Register disconnect.\n");

+		qla_pci_set_eeh_busy(vha);

+	}

 }

 void qla2x00_sp_free(srb_t *sp)

@@ -9671,6 +9678,12 @@ int qla2xxx_disable_port(struct Scsi_Hos

 	vha->hw->flags.port_isolated = 1;

+	if (qla2x00_isp_reg_stat(vha->hw)) {

+		ql_log(ql_log_info, vha, 0x9006,

+		    "PCI/Register disconnect, exiting.\n");

+		qla_pci_set_eeh_busy(vha);

+		return FAILED;

+	}

 	if (qla2x00_chip_is_down(vha))

 		return 0;

@@ -9686,6 +9699,13 @@ int qla2xxx_enable_port(struct Scsi_Host

 {

 	scsi_qla_host_t *vha = shost_priv(host);

+	if (qla2x00_isp_reg_stat(vha->hw)) {

+		ql_log(ql_log_info, vha, 0x9001,

+		    "PCI/Register disconnect, exiting.\n");

+		qla_pci_set_eeh_busy(vha);

+		return FAILED;

+	}

+

 	vha->hw->flags.port_isolated = 0;

 	/* Set the flag to 1, so that isp_abort can proceed */

 	vha->flags.online = 1;

--- a/drivers/scsi/qla2xxx/qla_os.c

+++ b/drivers/scsi/qla2xxx/qla_os.c

@@ -340,6 +340,11 @@ MODULE_PARM_DESC(ql2xabts_wait_nvme,

 		 "To wait for ABTS response on I/O timeouts for NVMe. (default: 1)");

+u32 ql2xdelay_before_pci_error_handling = 5;

+module_param(ql2xdelay_before_pci_error_handling, uint, 0644);

+MODULE_PARM_DESC(ql2xdelay_before_pci_error_handling,

+	"Number of seconds delayed before qla begin PCI error self-handling (default: 5).\n");

+

 static void qla2x00_clear_drv_active(struct qla_hw_data *);

 static void qla2x00_free_device(scsi_qla_host_t *);

 static int qla2xxx_map_queues(struct Scsi_Host *shost);

@@ -7275,6 +7280,44 @@ static void qla_heart_beat(struct scsi_q

 	}

 }

+static void qla_wind_down_chip(scsi_qla_host_t *vha)

+{

+	struct qla_hw_data *ha = vha->hw;

+

+	if (!ha->flags.eeh_busy)

+		return;

+	if (ha->pci_error_state)

+		/* system is trying to recover */

+		return;

+

+	/*

+	 * Current system is not handling PCIE error.  At this point, this is

+	 * best effort to wind down the adapter.

+	 */

+	if (time_after_eq(jiffies, ha->eeh_jif + ql2xdelay_before_pci_error_handling * HZ) &&

+	    !ha->flags.eeh_flush) {

+		ql_log(ql_log_info, vha, 0x9009,

+		    "PCI Error detected, attempting to reset hardware.\n");

+

+		ha->isp_ops->reset_chip(vha);

+		ha->isp_ops->disable_intrs(ha);

+

+		ha->flags.eeh_flush = EEH_FLUSH_RDY;

+		ha->eeh_jif = jiffies;

+

+	} else if (ha->flags.eeh_flush == EEH_FLUSH_RDY &&

+	    time_after_eq(jiffies, ha->eeh_jif +  5 * HZ)) {

+		pci_clear_master(ha->pdev);

+

+		/* flush all command */

+		qla2x00_abort_isp_cleanup(vha);

+		ha->flags.eeh_flush = EEH_FLUSH_DONE;

+

+		ql_log(ql_log_info, vha, 0x900a,

+		    "PCI Error handling complete, all IOs aborted.\n");

+	}

+}

+

 /**************************************************************************

 *   qla2x00_timer

 *

@@ -7297,6 +7340,8 @@ qla2x00_timer(scsi_qla_host_t *vha)

 	fc_port_t *fcport = NULL;

 	if (ha->flags.eeh_busy) {

+		qla_wind_down_chip(vha);

+

 		ql_dbg(ql_dbg_timer, vha, 0x6000,

 		    "EEH = %d, restarting timer.\n",

 		    ha->flags.eeh_busy);

@@ -7877,6 +7922,9 @@ void qla_pci_set_eeh_busy(struct scsi_ql

 	spin_lock_irqsave(&base_vha->work_lock, flags);

 	if (!ha->flags.eeh_busy) {

+		ha->eeh_jif = jiffies;

+		ha->flags.eeh_flush = 0;

+

 		ha->flags.eeh_busy = 1;

 		do_cleanup = true;

 	}