dasd_3990_erp.c

linux-2.6.15.6

/* 
 * File...........: linux/drivers/s390/block/dasd_3990_erp.c
 * Author(s)......: Horst  Hummel    <[e-mail]> 
 *		    Holger Smolinski <[e-mail]>
 * Bugreports.to..: <[e-mail]>
 * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 2000, 2001
 *
 * $Revision: 1.36 $
 */

#include <linux/timer.h>
#include <linux/slab.h>
#include <asm/idals.h>
#include <asm/todclk.h>

#define PRINTK_HEADER "dasd_erp(3990): "

#include "dasd_int.h"
#include "dasd_eckd.h"


struct DCTL_data {
	unsigned char subcommand;  /* e.g Inhibit Write, Enable Write,... */
	unsigned char modifier;	   /* Subcommand modifier */
	unsigned short res;	   /* reserved */
} __attribute__ ((packed));

/*
 ***************************************************************************** 
 * SECTION ERP EXAMINATION
 ***************************************************************************** 
 */

/*
 * DASD_3990_ERP_EXAMINE_24 
 *
 * DESCRIPTION
 *   Checks only for fatal (unrecoverable) error. 
 *   A detailed examination of the sense data is done later outside
 *   the interrupt handler.
 *
 *   Each bit configuration leading to an action code 2 (Exit with
 *   programming error or unusual condition indication)
 *   are handled as fatal error磗.
 * 
 *   All other configurations are handled as recoverable errors.
 *
 * RETURN VALUES
 *   dasd_era_fatal	for all fatal (unrecoverable errors)
 *   dasd_era_recover	for all others.
 */
static dasd_era_t
dasd_3990_erp_examine_24(struct dasd_ccw_req * cqr, char *sense)
{

	struct dasd_device *device = cqr->device;

	/* check for 'Command Reject' */
	if ((sense[0] & SNS0_CMD_REJECT) &&
	    (!(sense[2] & SNS2_ENV_DATA_PRESENT))) {

		DEV_MESSAGE(KERN_ERR, device, "%s",
			    "EXAMINE 24: Command Reject detected - "
			    "fatal error");

		return dasd_era_fatal;
	}

	/* check for 'Invalid Track Format' */
	if ((sense[1] & SNS1_INV_TRACK_FORMAT) &&
	    (!(sense[2] & SNS2_ENV_DATA_PRESENT))) {

		DEV_MESSAGE(KERN_ERR, device, "%s",
			    "EXAMINE 24: Invalid Track Format detected "
			    "- fatal error");

		return dasd_era_fatal;
	}

	/* check for 'No Record Found' */
	if (sense[1] & SNS1_NO_REC_FOUND) {

                /* FIXME: fatal error ?!? */
		DEV_MESSAGE(KERN_ERR, device,
			    "EXAMINE 24: No Record Found detected %s",
                            device->state <= DASD_STATE_BASIC ?
			    " " : "- fatal error");

		return dasd_era_fatal;
	}

	/* return recoverable for all others */
	return dasd_era_recover;
}				/* END dasd_3990_erp_examine_24 */

/*
 * DASD_3990_ERP_EXAMINE_32 
 *
 * DESCRIPTION
 *   Checks only for fatal/no/recoverable error. 
 *   A detailed examination of the sense data is done later outside
 *   the interrupt handler.
 *
 * RETURN VALUES
 *   dasd_era_none	no error 
 *   dasd_era_fatal	for all fatal (unrecoverable errors)
 *   dasd_era_recover	for recoverable others.
 */
static dasd_era_t
dasd_3990_erp_examine_32(struct dasd_ccw_req * cqr, char *sense)
{

	struct dasd_device *device = cqr->device;

	switch (sense[25]) {
	case 0x00:
		return dasd_era_none;

	case 0x01:
		DEV_MESSAGE(KERN_ERR, device, "%s", "EXAMINE 32: fatal error");

		return dasd_era_fatal;

	default:

		return dasd_era_recover;
	}

}				/* end dasd_3990_erp_examine_32 */

/*
 * DASD_3990_ERP_EXAMINE 
 *
 * DESCRIPTION
 *   Checks only for fatal/no/recover error. 
 *   A detailed examination of the sense data is done later outside
 *   the interrupt handler.
 *
 *   The logic is based on the 'IBM 3990 Storage Control  Reference' manual
 *   'Chapter 7. Error Recovery Procedures'.
 *
 * RETURN VALUES
 *   dasd_era_none	no error 
 *   dasd_era_fatal	for all fatal (unrecoverable errors)
 *   dasd_era_recover	for all others.
 */
dasd_era_t
dasd_3990_erp_examine(struct dasd_ccw_req * cqr, struct irb * irb)
{

	char *sense = irb->ecw;
	dasd_era_t era = dasd_era_recover;
	struct dasd_device *device = cqr->device;

	/* check for successful execution first */
	if (irb->scsw.cstat == 0x00 &&
	    irb->scsw.dstat == (DEV_STAT_CHN_END | DEV_STAT_DEV_END))
		return dasd_era_none;

	/* distinguish between 24 and 32 byte sense data */
	if (sense[27] & DASD_SENSE_BIT_0) {

		era = dasd_3990_erp_examine_24(cqr, sense);

	} else {

		era = dasd_3990_erp_examine_32(cqr, sense);

	}

	/* log the erp chain if fatal error occurred */
	if ((era == dasd_era_fatal) && (device->state >= DASD_STATE_READY)) {
		dasd_log_sense(cqr, irb);
		dasd_log_ccw(cqr, 0, irb->scsw.cpa);
	}

	return era;

}				/* END dasd_3990_erp_examine */

/*
 ***************************************************************************** 
 * SECTION ERP HANDLING
 ***************************************************************************** 
 */
/*
 ***************************************************************************** 
 * 24 and 32 byte sense ERP functions
 ***************************************************************************** 
 */

/*
 * DASD_3990_ERP_CLEANUP 
 *
 * DESCRIPTION
 *   Removes the already build but not necessary ERP request and sets
 *   the status of the original cqr / erp to the given (final) status
 *
 *  PARAMETER
 *   erp		request to be blocked
 *   final_status	either DASD_CQR_DONE or DASD_CQR_FAILED 
 *
 * RETURN VALUES
 *   cqr		original cqr		   
 */
static struct dasd_ccw_req *
dasd_3990_erp_cleanup(struct dasd_ccw_req * erp, char final_status)
{
	struct dasd_ccw_req *cqr = erp->refers;

	dasd_free_erp_request(erp, erp->device);
	cqr->status = final_status;
	return cqr;

}				/* end dasd_3990_erp_cleanup */

/*
 * DASD_3990_ERP_BLOCK_QUEUE 
 *
 * DESCRIPTION
 *   Block the given device request queue to prevent from further
 *   processing until the started timer has expired or an related
 *   interrupt was received.
 */
static void
dasd_3990_erp_block_queue(struct dasd_ccw_req * erp, int expires)
{

	struct dasd_device *device = erp->device;

	DEV_MESSAGE(KERN_INFO, device,
		    "blocking request queue for %is", expires/HZ);

	device->stopped |= DASD_STOPPED_PENDING;
	erp->status = DASD_CQR_QUEUED;

	dasd_set_timer(device, expires);
}

/*
 * DASD_3990_ERP_INT_REQ 
 *
 * DESCRIPTION
 *   Handles 'Intervention Required' error.
 *   This means either device offline or not installed.
 *
 * PARAMETER
 *   erp		current erp
 * RETURN VALUES
 *   erp		modified erp
 */
static struct dasd_ccw_req *
dasd_3990_erp_int_req(struct dasd_ccw_req * erp)
{

	struct dasd_device *device = erp->device;

	/* first time set initial retry counter and erp_function */
	/* and retry once without blocking queue		 */
	/* (this enables easier enqueing of the cqr)		 */
	if (erp->function != dasd_3990_erp_int_req) {

		erp->retries = 256;
		erp->function = dasd_3990_erp_int_req;

	} else {

		/* issue a message and wait for 'device ready' interrupt */
		DEV_MESSAGE(KERN_ERR, device, "%s",
			    "is offline or not installed - "
			    "INTERVENTION REQUIRED!!");

		dasd_3990_erp_block_queue(erp, 60*HZ);
	}

	return erp;

}				/* end dasd_3990_erp_int_req */

/*
 * DASD_3990_ERP_ALTERNATE_PATH 
 *
 * DESCRIPTION
 *   Repeat the operation on a different channel path.
 *   If all alternate paths have been tried, the request is posted with a
 *   permanent error.
 *
 *  PARAMETER
 *   erp		pointer to the current ERP
 *
 * RETURN VALUES
 *   erp		modified pointer to the ERP
 */
static void
dasd_3990_erp_alternate_path(struct dasd_ccw_req * erp)
{
	struct dasd_device *device = erp->device;
	__u8 opm;

	/* try alternate valid path */
	opm = ccw_device_get_path_mask(device->cdev);
	//FIXME: start with get_opm ?
	if (erp->lpm == 0)
		erp->lpm = LPM_ANYPATH & ~(erp->irb.esw.esw0.sublog.lpum);
	else
		erp->lpm &= ~(erp->irb.esw.esw0.sublog.lpum);

	if ((erp->lpm & opm) != 0x00) {

		DEV_MESSAGE(KERN_DEBUG, device,
			    "try alternate lpm=%x (lpum=%x / opm=%x)",
			    erp->lpm, erp->irb.esw.esw0.sublog.lpum, opm);

		/* reset status to queued to handle the request again... */
		if (erp->status > DASD_CQR_QUEUED)
			erp->status = DASD_CQR_QUEUED;
		erp->retries = 1;
	} else {
		DEV_MESSAGE(KERN_ERR, device,
			    "No alternate channel path left (lpum=%x / "
			    "opm=%x) -> permanent error",
			    erp->irb.esw.esw0.sublog.lpum, opm);

		/* post request with permanent error */
		if (erp->status > DASD_CQR_QUEUED)
			erp->status = DASD_CQR_FAILED;
	}
}				/* end dasd_3990_erp_alternate_path */

/*
 * DASD_3990_ERP_DCTL
 *
 * DESCRIPTION
 *   Setup cqr to do the Diagnostic Control (DCTL) command with an 
 *   Inhibit Write subcommand (0x20) and the given modifier.
 *
 *  PARAMETER
 *   erp		pointer to the current (failed) ERP
 *   modifier		subcommand modifier
 *   
 * RETURN VALUES
 *   dctl_cqr		pointer to NEW dctl_cqr 
 *
 */
static struct dasd_ccw_req *
dasd_3990_erp_DCTL(struct dasd_ccw_req * erp, char modifier)
{

	struct dasd_device *device = erp->device;
	struct DCTL_data *DCTL_data;
	struct ccw1 *ccw;
	struct dasd_ccw_req *dctl_cqr;

	dctl_cqr = dasd_alloc_erp_request((char *) &erp->magic, 1,
					  sizeof (struct DCTL_data),
					  erp->device);
	if (IS_ERR(dctl_cqr)) {
		DEV_MESSAGE(KERN_ERR, device, "%s",
			    "Unable to allocate DCTL-CQR");
		erp->status = DASD_CQR_FAILED;
		return erp;
	}

	DCTL_data = dctl_cqr->data;

	DCTL_data->subcommand = 0x02;	/* Inhibit Write */
	DCTL_data->modifier = modifier;

	ccw = dctl_cqr->cpaddr;
	memset(ccw, 0, sizeof (struct ccw1));
	ccw->cmd_code = CCW_CMD_DCTL;
	ccw->count = 4;
	ccw->cda = (__u32)(addr_t) DCTL_data;
	dctl_cqr->function = dasd_3990_erp_DCTL;
	dctl_cqr->refers = erp;
	dctl_cqr->device = erp->device;
	dctl_cqr->magic = erp->magic;
	dctl_cqr->expires = 5 * 60 * HZ;
	dctl_cqr->retries = 2;

	dctl_cqr->buildclk = get_clock();

	dctl_cqr->status = DASD_CQR_FILLED;

	return dctl_cqr;

}				/* end dasd_3990_erp_DCTL */

/*
 * DASD_3990_ERP_ACTION_1 
 *
 * DESCRIPTION
 *   Setup ERP to do the ERP action 1 (see Reference manual).
 *   Repeat the operation on a different channel path.
 *   If all alternate paths have been tried, the request is posted with a
 *   permanent error.
 *   Note: duplex handling is not implemented (yet).
 *
 *  PARAMETER
 *   erp		pointer to the current ERP
 *
 * RETURN VALUES
 *   erp		pointer to the ERP
 *
 */
static struct dasd_ccw_req *
dasd_3990_erp_action_1(struct dasd_ccw_req * erp)
{

	erp->function = dasd_3990_erp_action_1;

	dasd_3990_erp_alternate_path(erp);

	return erp;

}				/* end dasd_3990_erp_action_1 */

/*
 * DASD_3990_ERP_ACTION_4 
 *
 * DESCRIPTION
 *   Setup ERP to do the ERP action 4 (see Reference manual).
 *   Set the current request to PENDING to block the CQR queue for that device
 *   until the state change interrupt appears.
 *   Use a timer (20 seconds) to retry the cqr if the interrupt is still
 *   missing.
 *
 *  PARAMETER
 *   sense		sense data of the actual error
 *   erp		pointer to the current ERP
 *
 * RETURN VALUES
 *   erp		pointer to the ERP
 *
 */
static struct dasd_ccw_req *
dasd_3990_erp_action_4(struct dasd_ccw_req * erp, char *sense)
{

	struct dasd_device *device = erp->device;

	/* first time set initial retry counter and erp_function    */
	/* and retry once without waiting for state change pending  */
	/* interrupt (this enables easier enqueing of the cqr)	    */
	if (erp->function != dasd_3990_erp_action_4) {

		DEV_MESSAGE(KERN_INFO, device, "%s",
			    "dasd_3990_erp_action_4: first time retry");

		erp->retries = 256;
		erp->function = dasd_3990_erp_action_4;

	} else {

		if (sense[25] == 0x1D) {	/* state change pending */

			DEV_MESSAGE(KERN_INFO, device, 
				    "waiting for state change pending "
				    "interrupt, %d retries left",
				    erp->retries);
			
			dasd_3990_erp_block_queue(erp, 30*HZ);

                } else if (sense[25] == 0x1E) {	/* busy */
			DEV_MESSAGE(KERN_INFO, device,
				    "busy - redriving request later, "
				    "%d retries left",
				    erp->retries);
                        dasd_3990_erp_block_queue(erp, HZ);
		} else {

			/* no state change pending - retry */
			DEV_MESSAGE (KERN_INFO, device, 
				     "redriving request immediately, "
				     "%d retries left", 
				     erp->retries);
			erp->status = DASD_CQR_QUEUED;
		}
	}

	return erp;

}				/* end dasd_3990_erp_action_4 */

/*
 ***************************************************************************** 
 * 24 byte sense ERP functions (only)
 ***************************************************************************** 
 */

/*
 * DASD_3990_ERP_ACTION_5 
 *
 * DESCRIPTION
 *   Setup ERP to do the ERP action 5 (see Reference manual).
 *   NOTE: Further handling is done in xxx_further_erp after the retries.
 *
 *  PARAMETER
 *   erp		pointer to the current ERP
 *
 * RETURN VALUES
 *   erp		pointer to the ERP
 *
 */
static struct dasd_ccw_req *
dasd_3990_erp_action_5(struct dasd_ccw_req * erp)
{

	/* first of all retry */
	erp->retries = 10;
	erp->function = dasd_3990_erp_action_5;

	return erp;

}				/* end dasd_3990_erp_action_5 */

/*
 * DASD_3990_HANDLE_ENV_DATA
 *
 * DESCRIPTION
 *   Handles 24 byte 'Environmental data present'.