Merge patch series "mpi3mr: Support PCI Error Recovery"

Sumit Saxena <sumit.saxena@broadcom.com> says:

This patch series contains the changes done in the driver to support
PCI error recovery. It is rework of older patch series from Ranjan
Kumar, see [1].

[1] https://lore.kernel.org/all/20231214205900.270488-1-ranjan.kumar@broadcom.com/

Link: https://lore.kernel.org/r/20240627101735.18286-1-sumit.saxena@broadcom.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

Author: martinkpetersen

drivers/scsi/mpi3mr/mpi3mr.h

@@ -23,6 +23,7 @@
 #include <linux/miscdevice.h>
 #include <linux/module.h>
 #include <linux/pci.h>
+#include <linux/aer.h>
 #include <linux/poll.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
@@ -56,8 +57,8 @@ extern struct list_head mrioc_list;
 extern int prot_mask;
 extern atomic64_t event_counter;
 
-#define MPI3MR_DRIVER_VERSION	"8.9.1.0.50"
-#define MPI3MR_DRIVER_RELDATE	"14-May-2024"
+#define MPI3MR_DRIVER_VERSION	"8.9.1.0.51"
+#define MPI3MR_DRIVER_RELDATE	"29-May-2024"
 
 #define MPI3MR_DRIVER_NAME	"mpi3mr"
 #define MPI3MR_DRIVER_LICENSE	"GPL"
@@ -129,6 +130,7 @@ extern atomic64_t event_counter;
 #define MPI3MR_PREPARE_FOR_RESET_TIMEOUT	180
 #define MPI3MR_RESET_ACK_TIMEOUT		30
 #define MPI3MR_MUR_TIMEOUT			120
+#define MPI3MR_RESET_TIMEOUT			510
 
 #define MPI3MR_WATCHDOG_INTERVAL		1000 /* in milli seconds */
 
@@ -517,6 +519,7 @@ struct mpi3mr_throttle_group_info {
 
 /* HBA port flags */
 #define MPI3MR_HBA_PORT_FLAG_DIRTY	0x01
+#define MPI3MR_HBA_PORT_FLAG_NEW       0x02
 
 /* IOCTL data transfer sge*/
 #define MPI3MR_NUM_IOCTL_SGE		256
@@ -1153,6 +1156,8 @@ struct scmd_priv {
  * @trace_release_trigger_active: Trace trigger active flag
  * @fw_release_trigger_active: Fw release trigger active flag
  * @snapdump_trigger_active: Snapdump trigger active flag
+ * @pci_err_recovery: PCI error recovery in progress
+ * @block_on_pci_err: Block IO during PCI error recovery
  */
 struct mpi3mr_ioc {
 	struct list_head list;
@@ -1353,6 +1358,8 @@ struct mpi3mr_ioc {
 	bool snapdump_trigger_active;
 	bool trace_release_trigger_active;
 	bool fw_release_trigger_active;
+	bool pci_err_recovery;
+	bool block_on_pci_err;
 };
 
 /**

drivers/scsi/mpi3mr/mpi3mr_app.c

@@ -846,7 +846,7 @@ static int mpi3mr_bsg_pel_abort(struct mpi3mr_ioc *mrioc)
 		dprint_bsg_err(mrioc, "%s: reset in progress\n", __func__);
 		return -1;
 	}
-	if (mrioc->stop_bsgs) {
+	if (mrioc->stop_bsgs || mrioc->block_on_pci_err) {
 		dprint_bsg_err(mrioc, "%s: bsgs are blocked\n", __func__);
 		return -1;
 	}
@@ -1492,6 +1492,9 @@ static long mpi3mr_bsg_adp_reset(struct mpi3mr_ioc *mrioc,
 		goto out;
 	}
 
+	if (mrioc->unrecoverable || mrioc->block_on_pci_err)
+		return -EINVAL;
+
 	sg_copy_to_buffer(job->request_payload.sg_list,
 			  job->request_payload.sg_cnt,
 			  &adpreset, sizeof(adpreset));
@@ -2575,7 +2578,7 @@ static long mpi3mr_bsg_process_mpt_cmds(struct bsg_job *job)
 		mutex_unlock(&mrioc->bsg_cmds.mutex);
 		goto out;
 	}
-	if (mrioc->stop_bsgs) {
+	if (mrioc->stop_bsgs || mrioc->block_on_pci_err) {
 		dprint_bsg_err(mrioc, "%s: bsgs are blocked\n", __func__);
 		rval = -EAGAIN;
 		mutex_unlock(&mrioc->bsg_cmds.mutex);
@@ -3108,7 +3111,8 @@ adp_state_show(struct device *dev, struct device_attribute *attr,
 	ioc_state = mpi3mr_get_iocstate(mrioc);
 	if (ioc_state == MRIOC_STATE_UNRECOVERABLE)
 		adp_state = MPI3MR_BSG_ADPSTATE_UNRECOVERABLE;
-	else if ((mrioc->reset_in_progress) || (mrioc->stop_bsgs))
+	else if (mrioc->reset_in_progress || mrioc->stop_bsgs ||
+		 mrioc->block_on_pci_err)
 		adp_state = MPI3MR_BSG_ADPSTATE_IN_RESET;
 	else if (ioc_state == MRIOC_STATE_FAULT)
 		adp_state = MPI3MR_BSG_ADPSTATE_FAULT;

drivers/scsi/mpi3mr/mpi3mr_fw.c

@@ -608,7 +608,7 @@ int mpi3mr_blk_mq_poll(struct Scsi_Host *shost, unsigned int queue_num)
 	mrioc = (struct mpi3mr_ioc *)shost->hostdata;
 
 	if ((mrioc->reset_in_progress || mrioc->prepare_for_reset ||
-	    mrioc->unrecoverable))
+	    mrioc->unrecoverable || mrioc->pci_err_recovery))
 		return 0;
 
 	num_entries = mpi3mr_process_op_reply_q(mrioc,
@@ -1693,6 +1693,12 @@ int mpi3mr_admin_request_post(struct mpi3mr_ioc *mrioc, void *admin_req,
 		retval = -EAGAIN;
 		goto out;
 	}
+	if (mrioc->pci_err_recovery) {
+		ioc_err(mrioc, "admin request queue submission failed due to pci error recovery in progress\n");
+		retval = -EAGAIN;
+		goto out;
+	}
+
 	areq_entry = (u8 *)mrioc->admin_req_base +
 	    (areq_pi * MPI3MR_ADMIN_REQ_FRAME_SZ);
 	memset(areq_entry, 0, MPI3MR_ADMIN_REQ_FRAME_SZ);
@@ -2363,6 +2369,11 @@ int mpi3mr_op_request_post(struct mpi3mr_ioc *mrioc,
 		retval = -EAGAIN;
 		goto out;
 	}
+	if (mrioc->pci_err_recovery) {
+		ioc_err(mrioc, "operational request queue submission failed due to pci error recovery in progress\n");
+		retval = -EAGAIN;
+		goto out;
+	}
 
 	segment_base_addr = segments[pi / op_req_q->segment_qd].segment;
 	req_entry = (u8 *)segment_base_addr +
@@ -2627,7 +2638,7 @@ static void mpi3mr_watchdog_work(struct work_struct *work)
 	union mpi3mr_trigger_data trigger_data;
 	u16 reset_reason = MPI3MR_RESET_FROM_FAULT_WATCH;
 
-	if (mrioc->reset_in_progress)
+	if (mrioc->reset_in_progress || mrioc->pci_err_recovery)
 		return;
 
 	if (!mrioc->unrecoverable && !pci_device_is_present(mrioc->pdev)) {
@@ -4268,7 +4279,7 @@ int mpi3mr_reinit_ioc(struct mpi3mr_ioc *mrioc, u8 is_resume)
 		goto out_failed_noretry;
 	}
 
-	if (is_resume) {
+	if (is_resume || mrioc->block_on_pci_err) {
 		dprint_reset(mrioc, "setting up single ISR\n");
 		retval = mpi3mr_setup_isr(mrioc, 1);
 		if (retval) {
@@ -4319,7 +4330,7 @@ int mpi3mr_reinit_ioc(struct mpi3mr_ioc *mrioc, u8 is_resume)
 		goto out_failed;
 	}
 
-	if (is_resume) {
+	if (is_resume || mrioc->block_on_pci_err) {
 		dprint_reset(mrioc, "setting up multiple ISR\n");
 		retval = mpi3mr_setup_isr(mrioc, 0);
 		if (retval) {
@@ -4807,7 +4818,8 @@ void mpi3mr_cleanup_ioc(struct mpi3mr_ioc *mrioc)
 
 	ioc_state = mpi3mr_get_iocstate(mrioc);
 
-	if ((!mrioc->unrecoverable) && (!mrioc->reset_in_progress) &&
+	if (!mrioc->unrecoverable && !mrioc->reset_in_progress &&
+	    !mrioc->pci_err_recovery &&
 	    (ioc_state == MRIOC_STATE_READY)) {
 		if (mpi3mr_issue_and_process_mur(mrioc,
 		    MPI3MR_RESET_FROM_CTLR_CLEANUP))