// SPDX-License-Identifier: GPL-2.0-only /* * scsi_error.c Copyright (C) 1997 Eric Youngdale * * SCSI error/timeout handling * Initial versions: Eric Youngdale. Based upon conversations with * Leonard Zubkoff and David Miller at Linux Expo, * ideas originating from all over the place. * * Restructured scsi_unjam_host and associated functions. * September 04, 2002 Mike Anderson ([email protected]) * * Forward port of Russell King's ([email protected]) changes and * minor cleanups. * September 30, 2002 Mike Anderson ([email protected]) */ #include <linux/module.h> #include <linux/sched.h> #include <linux/gfp.h> #include <linux/timer.h> #include <linux/string.h> #include <linux/kernel.h> #include <linux/freezer.h> #include <linux/kthread.h> #include <linux/interrupt.h> #include <linux/blkdev.h> #include <linux/delay.h> #include <linux/jiffies.h> #include <scsi/scsi.h> #include <scsi/scsi_cmnd.h> #include <scsi/scsi_dbg.h> #include <scsi/scsi_device.h> #include <scsi/scsi_driver.h> #include <scsi/scsi_eh.h> #include <scsi/scsi_common.h> #include <scsi/scsi_transport.h> #include <scsi/scsi_host.h> #include <scsi/scsi_ioctl.h> #include <scsi/scsi_dh.h> #include <scsi/scsi_devinfo.h> #include <scsi/sg.h> #include "scsi_priv.h" #include "scsi_logging.h" #include "scsi_transport_api.h" #include <trace/events/scsi.h> #include <asm/unaligned.h> /* * These should *probably* be handled by the host itself. * Since it is allowed to sleep, it probably should. */ #define BUS_RESET_SETTLE_TIME … #define HOST_RESET_SETTLE_TIME … static int scsi_eh_try_stu(struct scsi_cmnd *scmd); static enum scsi_disposition scsi_try_to_abort_cmd(const struct scsi_host_template *, struct scsi_cmnd *); void scsi_eh_wakeup(struct Scsi_Host *shost, unsigned int busy) { … } /** * scsi_schedule_eh - schedule EH for SCSI host * @shost: SCSI host to invoke error handling on. * * Schedule SCSI EH without scmd. */ void scsi_schedule_eh(struct Scsi_Host *shost) { … } EXPORT_SYMBOL_GPL(…); static int scsi_host_eh_past_deadline(struct Scsi_Host *shost) { … } static bool scsi_cmd_retry_allowed(struct scsi_cmnd *cmd) { … } static bool scsi_eh_should_retry_cmd(struct scsi_cmnd *cmd) { … } /** * scmd_eh_abort_handler - Handle command aborts * @work: command to be aborted. * * Note: this function must be called only for a command that has timed out. * Because the block layer marks a request as complete before it calls * scsi_timeout(), a .scsi_done() call from the LLD for a command that has * timed out do not have any effect. Hence it is safe to call * scsi_finish_command() from this function. */ void scmd_eh_abort_handler(struct work_struct *work) { … } /** * scsi_abort_command - schedule a command abort * @scmd: scmd to abort. * * We only need to abort commands after a command timeout */ static int scsi_abort_command(struct scsi_cmnd *scmd) { … } /** * scsi_eh_reset - call into ->eh_action to reset internal counters * @scmd: scmd to run eh on. * * The scsi driver might be carrying internal state about the * devices, so we need to call into the driver to reset the * internal state once the error handler is started. */ static void scsi_eh_reset(struct scsi_cmnd *scmd) { … } static void scsi_eh_inc_host_failed(struct rcu_head *head) { … } /** * scsi_eh_scmd_add - add scsi cmd to error handling. * @scmd: scmd to run eh on. */ void scsi_eh_scmd_add(struct scsi_cmnd *scmd) { … } /** * scsi_timeout - Timeout function for normal scsi commands. * @req: request that is timing out. * * Notes: * We do not need to lock this. There is the potential for a race * only in that the normal completion handling might run, but if the * normal completion function determines that the timer has already * fired, then it mustn't do anything. */ enum blk_eh_timer_return scsi_timeout(struct request *req) { … } /** * scsi_block_when_processing_errors - Prevent cmds from being queued. * @sdev: Device on which we are performing recovery. * * Description: * We block until the host is out of error recovery, and then check to * see whether the host or the device is offline. * * Return value: * 0 when dev was taken offline by error recovery. 1 OK to proceed. */ int scsi_block_when_processing_errors(struct scsi_device *sdev) { … } EXPORT_SYMBOL(…); #ifdef CONFIG_SCSI_LOGGING /** * scsi_eh_prt_fail_stats - Log info on failures. * @shost: scsi host being recovered. * @work_q: Queue of scsi cmds to process. */ static inline void scsi_eh_prt_fail_stats(struct Scsi_Host *shost, struct list_head *work_q) { … } #endif /** * scsi_report_lun_change - Set flag on all *other* devices on the same target * to indicate that a UNIT ATTENTION is expected. * @sdev: Device reporting the UNIT ATTENTION */ static void scsi_report_lun_change(struct scsi_device *sdev) { … } /** * scsi_report_sense - Examine scsi sense information and log messages for * certain conditions, also issue uevents for some of them. * @sdev: Device reporting the sense code * @sshdr: sshdr to be examined */ static void scsi_report_sense(struct scsi_device *sdev, struct scsi_sense_hdr *sshdr) { … } static inline void set_scsi_ml_byte(struct scsi_cmnd *cmd, u8 status) { … } /** * scsi_check_sense - Examine scsi cmd sense * @scmd: Cmd to have sense checked. * * Return value: * SUCCESS or FAILED or NEEDS_RETRY or ADD_TO_MLQUEUE * * Notes: * When a deferred error is detected the current command has * not been executed and needs retrying. */ enum scsi_disposition scsi_check_sense(struct scsi_cmnd *scmd) { … } EXPORT_SYMBOL_GPL(…); static void scsi_handle_queue_ramp_up(struct scsi_device *sdev) { … } static void scsi_handle_queue_full(struct scsi_device *sdev) { … } /** * scsi_eh_completed_normally - Disposition a eh cmd on return from LLD. * @scmd: SCSI cmd to examine. * * Notes: * This is *only* called when we are examining the status of commands * queued during error recovery. the main difference here is that we * don't allow for the possibility of retries here, and we are a lot * more restrictive about what we consider acceptable. */ static enum scsi_disposition scsi_eh_completed_normally(struct scsi_cmnd *scmd) { … } /** * scsi_eh_done - Completion function for error handling. * @scmd: Cmd that is done. */ void scsi_eh_done(struct scsi_cmnd *scmd) { … } /** * scsi_try_host_reset - ask host adapter to reset itself * @scmd: SCSI cmd to send host reset. */ static enum scsi_disposition scsi_try_host_reset(struct scsi_cmnd *scmd) { … } /** * scsi_try_bus_reset - ask host to perform a bus reset * @scmd: SCSI cmd to send bus reset. */ static enum scsi_disposition scsi_try_bus_reset(struct scsi_cmnd *scmd) { … } static void __scsi_report_device_reset(struct scsi_device *sdev, void *data) { … } /** * scsi_try_target_reset - Ask host to perform a target reset * @scmd: SCSI cmd used to send a target reset * * Notes: * There is no timeout for this operation. if this operation is * unreliable for a given host, then the host itself needs to put a * timer on it, and set the host back to a consistent state prior to * returning. */ static enum scsi_disposition scsi_try_target_reset(struct scsi_cmnd *scmd) { … } /** * scsi_try_bus_device_reset - Ask host to perform a BDR on a dev * @scmd: SCSI cmd used to send BDR * * Notes: * There is no timeout for this operation. if this operation is * unreliable for a given host, then the host itself needs to put a * timer on it, and set the host back to a consistent state prior to * returning. */ static enum scsi_disposition scsi_try_bus_device_reset(struct scsi_cmnd *scmd) { … } /** * scsi_try_to_abort_cmd - Ask host to abort a SCSI command * @hostt: SCSI driver host template * @scmd: SCSI cmd used to send a target reset * * Return value: * SUCCESS, FAILED, or FAST_IO_FAIL * * Notes: * SUCCESS does not necessarily indicate that the command * has been aborted; it only indicates that the LLDDs * has cleared all references to that command. * LLDDs should return FAILED only if an abort was required * but could not be executed. LLDDs should return FAST_IO_FAIL * if the device is temporarily unavailable (eg due to a * link down on FibreChannel) */ static enum scsi_disposition scsi_try_to_abort_cmd(const struct scsi_host_template *hostt, struct scsi_cmnd *scmd) { … } static void scsi_abort_eh_cmnd(struct scsi_cmnd *scmd) { … } /** * scsi_eh_prep_cmnd - Save a scsi command info as part of error recovery * @scmd: SCSI command structure to hijack * @ses: structure to save restore information * @cmnd: CDB to send. Can be NULL if no new cmnd is needed * @cmnd_size: size in bytes of @cmnd (must be <= MAX_COMMAND_SIZE) * @sense_bytes: size of sense data to copy. or 0 (if != 0 @cmnd is ignored) * * This function is used to save a scsi command information before re-execution * as part of the error recovery process. If @sense_bytes is 0 the command * sent must be one that does not transfer any data. If @sense_bytes != 0 * @cmnd is ignored and this functions sets up a REQUEST_SENSE command * and cmnd buffers to read @sense_bytes into @scmd->sense_buffer. */ void scsi_eh_prep_cmnd(struct scsi_cmnd *scmd, struct scsi_eh_save *ses, unsigned char *cmnd, int cmnd_size, unsigned sense_bytes) { … } EXPORT_SYMBOL(…); /** * scsi_eh_restore_cmnd - Restore a scsi command info as part of error recovery * @scmd: SCSI command structure to restore * @ses: saved information from a coresponding call to scsi_eh_prep_cmnd * * Undo any damage done by above scsi_eh_prep_cmnd(). */ void scsi_eh_restore_cmnd(struct scsi_cmnd* scmd, struct scsi_eh_save *ses) { … } EXPORT_SYMBOL(…); /** * scsi_send_eh_cmnd - submit a scsi command as part of error recovery * @scmd: SCSI command structure to hijack * @cmnd: CDB to send * @cmnd_size: size in bytes of @cmnd * @timeout: timeout for this request * @sense_bytes: size of sense data to copy or 0 * * This function is used to send a scsi command down to a target device * as part of the error recovery process. See also scsi_eh_prep_cmnd() above. * * Return value: * SUCCESS or FAILED or NEEDS_RETRY */ static enum scsi_disposition scsi_send_eh_cmnd(struct scsi_cmnd *scmd, unsigned char *cmnd, int cmnd_size, int timeout, unsigned sense_bytes) { … } /** * scsi_request_sense - Request sense data from a particular target. * @scmd: SCSI cmd for request sense. * * Notes: * Some hosts automatically obtain this information, others require * that we obtain it on our own. This function will *not* return until * the command either times out, or it completes. */ static enum scsi_disposition scsi_request_sense(struct scsi_cmnd *scmd) { … } static enum scsi_disposition scsi_eh_action(struct scsi_cmnd *scmd, enum scsi_disposition rtn) { … } /** * scsi_eh_finish_cmd - Handle a cmd that eh is finished with. * @scmd: Original SCSI cmd that eh has finished. * @done_q: Queue for processed commands. * * Notes: * We don't want to use the normal command completion while we are are * still handling errors - it may cause other commands to be queued, * and that would disturb what we are doing. Thus we really want to * keep a list of pending commands for final completion, and once we * are ready to leave error handling we handle completion for real. */ void scsi_eh_finish_cmd(struct scsi_cmnd *scmd, struct list_head *done_q) { … } EXPORT_SYMBOL(…); /** * scsi_eh_get_sense - Get device sense data. * @work_q: Queue of commands to process. * @done_q: Queue of processed commands. * * Description: * See if we need to request sense information. if so, then get it * now, so we have a better idea of what to do. * * Notes: * This has the unfortunate side effect that if a shost adapter does * not automatically request sense information, we end up shutting * it down before we request it. * * All drivers should request sense information internally these days, * so for now all I have to say is tough noogies if you end up in here. * * XXX: Long term this code should go away, but that needs an audit of * all LLDDs first. */ int scsi_eh_get_sense(struct list_head *work_q, struct list_head *done_q) { … } EXPORT_SYMBOL_GPL(…); /** * scsi_eh_tur - Send TUR to device. * @scmd: &scsi_cmnd to send TUR * * Return value: * 0 - Device is ready. 1 - Device NOT ready. */ static int scsi_eh_tur(struct scsi_cmnd *scmd) { … } /** * scsi_eh_test_devices - check if devices are responding from error recovery. * @cmd_list: scsi commands in error recovery. * @work_q: queue for commands which still need more error recovery * @done_q: queue for commands which are finished * @try_stu: boolean on if a STU command should be tried in addition to TUR. * * Decription: * Tests if devices are in a working state. Commands to devices now in * a working state are sent to the done_q while commands to devices which * are still failing to respond are returned to the work_q for more * processing. **/ static int scsi_eh_test_devices(struct list_head *cmd_list, struct list_head *work_q, struct list_head *done_q, int try_stu) { … } /** * scsi_eh_try_stu - Send START_UNIT to device. * @scmd: &scsi_cmnd to send START_UNIT * * Return value: * 0 - Device is ready. 1 - Device NOT ready. */ static int scsi_eh_try_stu(struct scsi_cmnd *scmd) { … } /** * scsi_eh_stu - send START_UNIT if needed * @shost: &scsi host being recovered. * @work_q: &list_head for pending commands. * @done_q: &list_head for processed commands. * * Notes: * If commands are failing due to not ready, initializing command required, * try revalidating the device, which will end up sending a start unit. */ static int scsi_eh_stu(struct Scsi_Host *shost, struct list_head *work_q, struct list_head *done_q) { … } /** * scsi_eh_bus_device_reset - send bdr if needed * @shost: scsi host being recovered. * @work_q: &list_head for pending commands. * @done_q: &list_head for processed commands. * * Notes: * Try a bus device reset. Still, look to see whether we have multiple * devices that are jammed or not - if we have multiple devices, it * makes no sense to try bus_device_reset - we really would need to try * a bus_reset instead. */ static int scsi_eh_bus_device_reset(struct Scsi_Host *shost, struct list_head *work_q, struct list_head *done_q) { … } /** * scsi_eh_target_reset - send target reset if needed * @shost: scsi host being recovered. * @work_q: &list_head for pending commands. * @done_q: &list_head for processed commands. * * Notes: * Try a target reset. */ static int scsi_eh_target_reset(struct Scsi_Host *shost, struct list_head *work_q, struct list_head *done_q) { … } /** * scsi_eh_bus_reset - send a bus reset * @shost: &scsi host being recovered. * @work_q: &list_head for pending commands. * @done_q: &list_head for processed commands. */ static int scsi_eh_bus_reset(struct Scsi_Host *shost, struct list_head *work_q, struct list_head *done_q) { … } /** * scsi_eh_host_reset - send a host reset * @shost: host to be reset. * @work_q: &list_head for pending commands. * @done_q: &list_head for processed commands. */ static int scsi_eh_host_reset(struct Scsi_Host *shost, struct list_head *work_q, struct list_head *done_q) { … } /** * scsi_eh_offline_sdevs - offline scsi devices that fail to recover * @work_q: &list_head for pending commands. * @done_q: &list_head for processed commands. */ static void scsi_eh_offline_sdevs(struct list_head *work_q, struct list_head *done_q) { … } /** * scsi_noretry_cmd - determine if command should be failed fast * @scmd: SCSI cmd to examine. */ bool scsi_noretry_cmd(struct scsi_cmnd *scmd) { … } /** * scsi_decide_disposition - Disposition a cmd on return from LLD. * @scmd: SCSI cmd to examine. * * Notes: * This is *only* called when we are examining the status after sending * out the actual data command. any commands that are queued for error * recovery (e.g. test_unit_ready) do *not* come through here. * * When this routine returns failed, it means the error handler thread * is woken. In cases where the error code indicates an error that * doesn't require the error handler read (i.e. we don't need to * abort/reset), this function should return SUCCESS. */ enum scsi_disposition scsi_decide_disposition(struct scsi_cmnd *scmd) { … } static enum rq_end_io_ret eh_lock_door_done(struct request *req, blk_status_t status) { … } /** * scsi_eh_lock_door - Prevent medium removal for the specified device * @sdev: SCSI device to prevent medium removal * * Locking: * We must be called from process context. * * Notes: * We queue up an asynchronous "ALLOW MEDIUM REMOVAL" request on the * head of the devices request queue, and continue. */ static void scsi_eh_lock_door(struct scsi_device *sdev) { … } /** * scsi_restart_operations - restart io operations to the specified host. * @shost: Host we are restarting. * * Notes: * When we entered the error handler, we blocked all further i/o to * this device. we need to 'reverse' this process. */ static void scsi_restart_operations(struct Scsi_Host *shost) { … } /** * scsi_eh_ready_devs - check device ready state and recover if not. * @shost: host to be recovered. * @work_q: &list_head for pending commands. * @done_q: &list_head for processed commands. */ void scsi_eh_ready_devs(struct Scsi_Host *shost, struct list_head *work_q, struct list_head *done_q) { … } EXPORT_SYMBOL_GPL(…); /** * scsi_eh_flush_done_q - finish processed commands or retry them. * @done_q: list_head of processed commands. */ void scsi_eh_flush_done_q(struct list_head *done_q) { … } EXPORT_SYMBOL(…); /** * scsi_unjam_host - Attempt to fix a host which has a cmd that failed. * @shost: Host to unjam. * * Notes: * When we come in here, we *know* that all commands on the bus have * either completed, failed or timed out. we also know that no further * commands are being sent to the host, so things are relatively quiet * and we have freedom to fiddle with things as we wish. * * This is only the *default* implementation. it is possible for * individual drivers to supply their own version of this function, and * if the maintainer wishes to do this, it is strongly suggested that * this function be taken as a template and modified. this function * was designed to correctly handle problems for about 95% of the * different cases out there, and it should always provide at least a * reasonable amount of error recovery. * * Any command marked 'failed' or 'timeout' must eventually have * scsi_finish_cmd() called for it. we do all of the retry stuff * here, so when we restart the host after we return it should have an * empty queue. */ static void scsi_unjam_host(struct Scsi_Host *shost) { … } /** * scsi_error_handler - SCSI error handler thread * @data: Host for which we are running. * * Notes: * This is the main error handling loop. This is run as a kernel thread * for every SCSI host and handles all error handling activity. */ int scsi_error_handler(void *data) { … } /* * Function: scsi_report_bus_reset() * * Purpose: Utility function used by low-level drivers to report that * they have observed a bus reset on the bus being handled. * * Arguments: shost - Host in question * channel - channel on which reset was observed. * * Returns: Nothing * * Lock status: Host lock must be held. * * Notes: This only needs to be called if the reset is one which * originates from an unknown location. Resets originated * by the mid-level itself don't need to call this, but there * should be no harm. * * The main purpose of this is to make sure that a CHECK_CONDITION * is properly treated. */ void scsi_report_bus_reset(struct Scsi_Host *shost, int channel) { … } EXPORT_SYMBOL(…); /* * Function: scsi_report_device_reset() * * Purpose: Utility function used by low-level drivers to report that * they have observed a device reset on the device being handled. * * Arguments: shost - Host in question * channel - channel on which reset was observed * target - target on which reset was observed * * Returns: Nothing * * Lock status: Host lock must be held * * Notes: This only needs to be called if the reset is one which * originates from an unknown location. Resets originated * by the mid-level itself don't need to call this, but there * should be no harm. * * The main purpose of this is to make sure that a CHECK_CONDITION * is properly treated. */ void scsi_report_device_reset(struct Scsi_Host *shost, int channel, int target) { … } EXPORT_SYMBOL(…); /** * scsi_ioctl_reset: explicitly reset a host/bus/target/device * @dev: scsi_device to operate on * @arg: reset type (see sg.h) */ int scsi_ioctl_reset(struct scsi_device *dev, int __user *arg) { … } bool scsi_command_normalize_sense(const struct scsi_cmnd *cmd, struct scsi_sense_hdr *sshdr) { … } EXPORT_SYMBOL(…); /** * scsi_get_sense_info_fld - get information field from sense data (either fixed or descriptor format) * @sense_buffer: byte array of sense data * @sb_len: number of valid bytes in sense_buffer * @info_out: pointer to 64 integer where 8 or 4 byte information * field will be placed if found. * * Return value: * true if information field found, false if not found. */ bool scsi_get_sense_info_fld(const u8 *sense_buffer, int sb_len, u64 *info_out) { … } EXPORT_SYMBOL(…);
Codebase Browser
linux