diff options
| author | Kashyap Desai <kashyap.desai@broadcom.com> | 2021-05-20 20:55:30 +0530 |
|---|---|---|
| committer | Martin K. Petersen <martin.petersen@oracle.com> | 2021-06-02 00:56:16 -0400 |
| commit | fb9b04574f147831d96b6aead161c8ca26670c97 (patch) | |
| tree | 102fc9acfea63a8b12ba15fa28f306b852f5c9ed /drivers/scsi/mpi3mr/mpi3mr.h | |
| parent | e36710dc06e3994c9abe78b957a07cfd9243a674 (diff) | |
scsi: mpi3mr: Add support for recovering controller
Detection of firmware fault or any kind of unresponsiveness in the
controller (any admin command which times out) results in resetting the
controller. The primary reset mechanisms used are either soft reset or diag
fault reset. A reset is performed if the host sets the ResetAction field in
the HostDiagnostic register to either 001b (soft reset) or 007b (diag fault
reset). After successfully resetting the controller the driver
reinitializes the controller by going through start of the day
initialization procedure. Pending I/Os during the reset are returned back
to the SCSI midlayer for retry.
Link: https://lore.kernel.org/r/20210520152545.2710479-10-kashyap.desai@broadcom.com
Cc: sathya.prakash@broadcom.co
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Tomas Henzl <thenzl@redhat.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Diffstat (limited to 'drivers/scsi/mpi3mr/mpi3mr.h')
| -rw-r--r-- | drivers/scsi/mpi3mr/mpi3mr.h | 15 |
1 files changed, 13 insertions, 2 deletions
diff --git a/drivers/scsi/mpi3mr/mpi3mr.h b/drivers/scsi/mpi3mr/mpi3mr.h index 5c01ae31cbab..31167a912a81 100644 --- a/drivers/scsi/mpi3mr/mpi3mr.h +++ b/drivers/scsi/mpi3mr/mpi3mr.h @@ -103,6 +103,7 @@ extern struct list_head mrioc_list; #define MPI3MR_INTADMCMD_TIMEOUT 10 #define MPI3MR_PORTENABLE_TIMEOUT 300 #define MPI3MR_RESETTM_TIMEOUT 30 +#define MPI3MR_RESET_HOST_IOWAIT_TIMEOUT 5 #define MPI3MR_DEFAULT_SHUTDOWN_TIME 120 #define MPI3MR_WATCHDOG_INTERVAL 1000 /* in milli seconds */ @@ -635,10 +636,14 @@ struct scmd_priv { * @dev_handle_bitmap_sz: Device handle bitmap size * @removepend_bitmap: Remove pending bitmap * @delayed_rmhs_list: Delayed device removal list + * @fault_dbg: Fault debug flag * @reset_in_progress: Reset in progress flag * @unrecoverable: Controller unrecoverable flag + * @reset_mutex: Controller reset mutex + * @reset_waitq: Controller reset wait queue * @diagsave_timeout: Diagnostic information save timeout * @logging_level: Controller debug logging level + * @flush_io_count: I/O count to flush after reset * @current_event: Firmware event currently in process * @driver_info: Driver, Kernel, OS information to firmware * @change_count: Topology change count @@ -753,11 +758,15 @@ struct mpi3mr_ioc { void *removepend_bitmap; struct list_head delayed_rmhs_list; + u8 fault_dbg; u8 reset_in_progress; u8 unrecoverable; + struct mutex reset_mutex; + wait_queue_head_t reset_waitq; u16 diagsave_timeout; int logging_level; + u16 flush_io_count; struct mpi3mr_fwevt *current_event; struct mpi3_driver_info_layout driver_info; @@ -806,8 +815,8 @@ struct delayed_dev_rmhs_node { int mpi3mr_setup_resources(struct mpi3mr_ioc *mrioc); void mpi3mr_cleanup_resources(struct mpi3mr_ioc *mrioc); -int mpi3mr_init_ioc(struct mpi3mr_ioc *mrioc); -void mpi3mr_cleanup_ioc(struct mpi3mr_ioc *mrioc); +int mpi3mr_init_ioc(struct mpi3mr_ioc *mrioc, u8 re_init); +void mpi3mr_cleanup_ioc(struct mpi3mr_ioc *mrioc, u8 re_init); int mpi3mr_issue_port_enable(struct mpi3mr_ioc *mrioc, u8 async); int mpi3mr_admin_request_post(struct mpi3mr_ioc *mrioc, void *admin_req, u16 admin_req_sz, u8 ignore_reset); @@ -833,6 +842,8 @@ void mpi3mr_stop_watchdog(struct mpi3mr_ioc *mrioc); int mpi3mr_soft_reset_handler(struct mpi3mr_ioc *mrioc, u32 reset_reason, u8 snapdump); +int mpi3mr_diagfault_reset_handler(struct mpi3mr_ioc *mrioc, + u32 reset_reason); void mpi3mr_ioc_disable_intr(struct mpi3mr_ioc *mrioc); void mpi3mr_ioc_enable_intr(struct mpi3mr_ioc *mrioc); |