libata-eh.c

来自「linux 内核源代码」· C语言 代码 · 共 2,561 行 · 第 1/5 页

	arg.since = j64 - min(j64, j5mins);
	ata_ering_map(&dev->ering, speed_down_verdict_cb, &arg);

	if (arg.nr_errors[1] + arg.nr_errors[2] + arg.nr_errors[3] > 10)
		verdict |= ATA_EH_SPDN_FALLBACK_TO_PIO;

	return verdict;
}

/**
 *	ata_eh_speed_down - record error and speed down if necessary
 *	@dev: Failed device
 *	@is_io: Did the device fail during normal IO?
 *	@err_mask: err_mask of the error
 *
 *	Record error and examine error history to determine whether
 *	adjusting transmission speed is necessary.  It also sets
 *	transmission limits appropriately if such adjustment is
 *	necessary.
 *
 *	LOCKING:
 *	Kernel thread context (may sleep).
 *
 *	RETURNS:
 *	Determined recovery action.
 */
static unsigned int ata_eh_speed_down(struct ata_device *dev, int is_io,
				      unsigned int err_mask)
{
	unsigned int verdict;
	unsigned int action = 0;

	/* don't bother if Cat-0 error */
	if (ata_eh_categorize_error(is_io, err_mask) == 0)
		return 0;

	/* record error and determine whether speed down is necessary */
	ata_ering_record(&dev->ering, is_io, err_mask);
	verdict = ata_eh_speed_down_verdict(dev);

	/* turn off NCQ? */
	if ((verdict & ATA_EH_SPDN_NCQ_OFF) &&
	    (dev->flags & (ATA_DFLAG_PIO | ATA_DFLAG_NCQ |
			   ATA_DFLAG_NCQ_OFF)) == ATA_DFLAG_NCQ) {
		dev->flags |= ATA_DFLAG_NCQ_OFF;
		ata_dev_printk(dev, KERN_WARNING,
			       "NCQ disabled due to excessive errors\n");
		goto done;
	}

	/* speed down? */
	if (verdict & ATA_EH_SPDN_SPEED_DOWN) {
		/* speed down SATA link speed if possible */
		if (sata_down_spd_limit(dev->link) == 0) {
			action |= ATA_EH_HARDRESET;
			goto done;
		}

		/* lower transfer mode */
		if (dev->spdn_cnt < 2) {
			static const int dma_dnxfer_sel[] =
				{ ATA_DNXFER_DMA, ATA_DNXFER_40C };
			static const int pio_dnxfer_sel[] =
				{ ATA_DNXFER_PIO, ATA_DNXFER_FORCE_PIO0 };
			int sel;

			if (dev->xfer_shift != ATA_SHIFT_PIO)
				sel = dma_dnxfer_sel[dev->spdn_cnt];
			else
				sel = pio_dnxfer_sel[dev->spdn_cnt];

			dev->spdn_cnt++;

			if (ata_down_xfermask_limit(dev, sel) == 0) {
				action |= ATA_EH_SOFTRESET;
				goto done;
			}
		}
	}

	/* Fall back to PIO?  Slowing down to PIO is meaningless for
	 * SATA.  Consider it only for PATA.
	 */
	if ((verdict & ATA_EH_SPDN_FALLBACK_TO_PIO) && (dev->spdn_cnt >= 2) &&
	    (dev->link->ap->cbl != ATA_CBL_SATA) &&
	    (dev->xfer_shift != ATA_SHIFT_PIO)) {
		if (ata_down_xfermask_limit(dev, ATA_DNXFER_FORCE_PIO) == 0) {
			dev->spdn_cnt = 0;
			action |= ATA_EH_SOFTRESET;
			goto done;
		}
	}

	return 0;
 done:
	/* device has been slowed down, blow error history */
	ata_ering_clear(&dev->ering);
	return action;
}

/**
 *	ata_eh_link_autopsy - analyze error and determine recovery action
 *	@link: host link to perform autopsy on
 *
 *	Analyze why @link failed and determine which recovery actions
 *	are needed.  This function also sets more detailed AC_ERR_*
 *	values and fills sense data for ATAPI CHECK SENSE.
 *
 *	LOCKING:
 *	Kernel thread context (may sleep).
 */
static void ata_eh_link_autopsy(struct ata_link *link)
{
	struct ata_port *ap = link->ap;
	struct ata_eh_context *ehc = &link->eh_context;
	struct ata_device *dev;
	unsigned int all_err_mask = 0;
	int tag, is_io = 0;
	u32 serror;
	int rc;

	DPRINTK("ENTER\n");

	if (ehc->i.flags & ATA_EHI_NO_AUTOPSY)
		return;

	/* obtain and analyze SError */
	rc = sata_scr_read(link, SCR_ERROR, &serror);
	if (rc == 0) {
		ehc->i.serror |= serror;
		ata_eh_analyze_serror(link);
	} else if (rc != -EOPNOTSUPP) {
		/* SError read failed, force hardreset and probing */
		ata_ehi_schedule_probe(&ehc->i);
		ehc->i.action |= ATA_EH_HARDRESET;
		ehc->i.err_mask |= AC_ERR_OTHER;
	}

	/* analyze NCQ failure */
	ata_eh_analyze_ncq_error(link);

	/* any real error trumps AC_ERR_OTHER */
	if (ehc->i.err_mask & ~AC_ERR_OTHER)
		ehc->i.err_mask &= ~AC_ERR_OTHER;

	all_err_mask |= ehc->i.err_mask;

	for (tag = 0; tag < ATA_MAX_QUEUE; tag++) {
		struct ata_queued_cmd *qc = __ata_qc_from_tag(ap, tag);

		if (!(qc->flags & ATA_QCFLAG_FAILED) || qc->dev->link != link)
			continue;

		/* inherit upper level err_mask */
		qc->err_mask |= ehc->i.err_mask;

		/* analyze TF */
		ehc->i.action |= ata_eh_analyze_tf(qc, &qc->result_tf);

		/* DEV errors are probably spurious in case of ATA_BUS error */
		if (qc->err_mask & AC_ERR_ATA_BUS)
			qc->err_mask &= ~(AC_ERR_DEV | AC_ERR_MEDIA |
					  AC_ERR_INVALID);

		/* any real error trumps unknown error */
		if (qc->err_mask & ~AC_ERR_OTHER)
			qc->err_mask &= ~AC_ERR_OTHER;

		/* SENSE_VALID trumps dev/unknown error and revalidation */
		if (qc->flags & ATA_QCFLAG_SENSE_VALID)
			qc->err_mask &= ~(AC_ERR_DEV | AC_ERR_OTHER);

		/* accumulate error info */
		ehc->i.dev = qc->dev;
		all_err_mask |= qc->err_mask;
		if (qc->flags & ATA_QCFLAG_IO)
			is_io = 1;
	}

	/* enforce default EH actions */
	if (ap->pflags & ATA_PFLAG_FROZEN ||
	    all_err_mask & (AC_ERR_HSM | AC_ERR_TIMEOUT))
		ehc->i.action |= ATA_EH_SOFTRESET;
	else if ((is_io && all_err_mask) ||
		 (!is_io && (all_err_mask & ~AC_ERR_DEV)))
		ehc->i.action |= ATA_EH_REVALIDATE;

	/* If we have offending qcs and the associated failed device,
	 * perform per-dev EH action only on the offending device.
	 */
	if (ehc->i.dev) {
		ehc->i.dev_action[ehc->i.dev->devno] |=
			ehc->i.action & ATA_EH_PERDEV_MASK;
		ehc->i.action &= ~ATA_EH_PERDEV_MASK;
	}

	/* propagate timeout to host link */
	if ((all_err_mask & AC_ERR_TIMEOUT) && !ata_is_host_link(link))
		ap->link.eh_context.i.err_mask |= AC_ERR_TIMEOUT;

	/* record error and consider speeding down */
	dev = ehc->i.dev;
	if (!dev && ((ata_link_max_devices(link) == 1 &&
		      ata_dev_enabled(link->device))))
	    dev = link->device;

	if (dev)
		ehc->i.action |= ata_eh_speed_down(dev, is_io, all_err_mask);

	DPRINTK("EXIT\n");
}

/**
 *	ata_eh_autopsy - analyze error and determine recovery action
 *	@ap: host port to perform autopsy on
 *
 *	Analyze all links of @ap and determine why they failed and
 *	which recovery actions are needed.
 *
 *	LOCKING:
 *	Kernel thread context (may sleep).
 */
void ata_eh_autopsy(struct ata_port *ap)
{
	struct ata_link *link;

	ata_port_for_each_link(link, ap)
		ata_eh_link_autopsy(link);

	/* Autopsy of fanout ports can affect host link autopsy.
	 * Perform host link autopsy last.
	 */
	if (ap->nr_pmp_links)
		ata_eh_link_autopsy(&ap->link);
}

/**
 *	ata_eh_link_report - report error handling to user
 *	@link: ATA link EH is going on
 *
 *	Report EH to user.
 *
 *	LOCKING:
 *	None.
 */
static void ata_eh_link_report(struct ata_link *link)
{
	struct ata_port *ap = link->ap;
	struct ata_eh_context *ehc = &link->eh_context;
	const char *frozen, *desc;
	char tries_buf[6];
	int tag, nr_failed = 0;

	if (ehc->i.flags & ATA_EHI_QUIET)
		return;

	desc = NULL;
	if (ehc->i.desc[0] != '\0')
		desc = ehc->i.desc;

	for (tag = 0; tag < ATA_MAX_QUEUE; tag++) {
		struct ata_queued_cmd *qc = __ata_qc_from_tag(ap, tag);

		if (!(qc->flags & ATA_QCFLAG_FAILED) || qc->dev->link != link ||
		    ((qc->flags & ATA_QCFLAG_QUIET) &&
		     qc->err_mask == AC_ERR_DEV))
			continue;
		if (qc->flags & ATA_QCFLAG_SENSE_VALID && !qc->err_mask)
			continue;

		nr_failed++;
	}

	if (!nr_failed && !ehc->i.err_mask)
		return;

	frozen = "";
	if (ap->pflags & ATA_PFLAG_FROZEN)
		frozen = " frozen";

	memset(tries_buf, 0, sizeof(tries_buf));
	if (ap->eh_tries < ATA_EH_MAX_TRIES)
		snprintf(tries_buf, sizeof(tries_buf) - 1, " t%d",
			 ap->eh_tries);

	if (ehc->i.dev) {
		ata_dev_printk(ehc->i.dev, KERN_ERR, "exception Emask 0x%x "
			       "SAct 0x%x SErr 0x%x action 0x%x%s%s\n",
			       ehc->i.err_mask, link->sactive, ehc->i.serror,
			       ehc->i.action, frozen, tries_buf);
		if (desc)
			ata_dev_printk(ehc->i.dev, KERN_ERR, "%s\n", desc);
	} else {
		ata_link_printk(link, KERN_ERR, "exception Emask 0x%x "
				"SAct 0x%x SErr 0x%x action 0x%x%s%s\n",
				ehc->i.err_mask, link->sactive, ehc->i.serror,
				ehc->i.action, frozen, tries_buf);
		if (desc)
			ata_link_printk(link, KERN_ERR, "%s\n", desc);
	}

	if (ehc->i.serror)
		ata_port_printk(ap, KERN_ERR,
		  "SError: { %s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s}\n",
		  ehc->i.serror & SERR_DATA_RECOVERED ? "RecovData " : "",
		  ehc->i.serror & SERR_COMM_RECOVERED ? "RecovComm " : "",
		  ehc->i.serror & SERR_DATA ? "UnrecovData " : "",
		  ehc->i.serror & SERR_PERSISTENT ? "Persist " : "",
		  ehc->i.serror & SERR_PROTOCOL ? "Proto " : "",
		  ehc->i.serror & SERR_INTERNAL ? "HostInt " : "",
		  ehc->i.serror & SERR_PHYRDY_CHG ? "PHYRdyChg " : "",
		  ehc->i.serror & SERR_PHY_INT_ERR ? "PHYInt " : "",
		  ehc->i.serror & SERR_COMM_WAKE ? "CommWake " : "",
		  ehc->i.serror & SERR_10B_8B_ERR ? "10B8B " : "",
		  ehc->i.serror & SERR_DISPARITY ? "Dispar " : "",
		  ehc->i.serror & SERR_CRC ? "BadCRC " : "",
		  ehc->i.serror & SERR_HANDSHAKE ? "Handshk " : "",
		  ehc->i.serror & SERR_LINK_SEQ_ERR ? "LinkSeq " : "",
		  ehc->i.serror & SERR_TRANS_ST_ERROR ? "TrStaTrns " : "",
		  ehc->i.serror & SERR_UNRECOG_FIS ? "UnrecFIS " : "",
		  ehc->i.serror & SERR_DEV_XCHG ? "DevExch " : "");

	for (tag = 0; tag < ATA_MAX_QUEUE; tag++) {
		struct ata_queued_cmd *qc = __ata_qc_from_tag(ap, tag);
		struct ata_taskfile *cmd = &qc->tf, *res = &qc->result_tf;
		const u8 *cdb = qc->cdb;
		char data_buf[20] = "";
		char cdb_buf[70] = "";

		if (!(qc->flags & ATA_QCFLAG_FAILED) ||
		    qc->dev->link != link || !qc->err_mask)
			continue;

		if (qc->dma_dir != DMA_NONE) {
			static const char *dma_str[] = {
				[DMA_BIDIRECTIONAL]	= "bidi",
				[DMA_TO_DEVICE]		= "out",
				[DMA_FROM_DEVICE]	= "in",
			};
			static const char *prot_str[] = {
				[ATA_PROT_PIO]		= "pio",
				[ATA_PROT_DMA]		= "dma",
				[ATA_PROT_NCQ]		= "ncq",
				[ATA_PROT_ATAPI]	= "pio",
				[ATA_PROT_ATAPI_DMA]	= "dma",
			};

			snprintf(data_buf, sizeof(data_buf), " %s %u %s",
				 prot_str[qc->tf.protocol], qc->nbytes,
				 dma_str[qc->dma_dir]);
		}

		if (is_atapi_taskfile(&qc->tf))
			snprintf(cdb_buf, sizeof(cdb_buf),
				 "cdb %02x %02x %02x %02x %02x %02x %02x %02x  "
				 "%02x %02x %02x %02x %02x %02x %02x %02x\n         ",
				 cdb[0], cdb[1], cdb[2], cdb[3],
				 cdb[4], cdb[5], cdb[6], cdb[7],
				 cdb[8], cdb[9], cdb[10], cdb[11],
				 cdb[12], cdb[13], cdb[14], cdb[15]);

		ata_dev_printk(qc->dev, KERN_ERR,
			"cmd %02x/%02x:%02x:%02x:%02x:%02x/%02x:%02x:%02x:%02x:%02x/%02x "
			"tag %d%s\n         %s"
			"res %02x/%02x:%02x:%02x:%02x:%02x/%02x:%02x:%02x:%02x:%02x/%02x "
			"Emask 0x%x (%s)%s\n",
			cmd->command, cmd->feature, cmd->nsect,
			cmd->lbal, cmd->lbam, cmd->lbah,
			cmd->hob_feature, cmd->hob_nsect,
			cmd->hob_lbal, cmd->hob_lbam, cmd->hob_lbah,
			cmd->device, qc->tag, data_buf, cdb_buf,
			res->command, res->feature, res->nsect,
			res->lbal, res->lbam, res->lbah,
			res->hob_feature, res->hob_nsect,
			res->hob_lbal, res->hob_lbam, res->hob_lbah,
			res->device, qc->err_mask, ata_err_string(qc->err_mask),
			qc->err_mask & AC_ERR_NCQ ? " <F>" : "");

		if (res->command & (ATA_BUSY | ATA_DRDY | ATA_DF | ATA_DRQ |
				    ATA_ERR)) {
			if (res->command & ATA_BUSY)
				ata_dev_printk(qc->dev, KERN_ERR,
				  "status: { Busy }\n");
			else
				ata_dev_printk(qc->dev, KERN_ERR,
				  "status: { %s%s%s%s}\n",
				  res->command & ATA_DRDY ? "DRDY " : "",
				  res->command & ATA_DF ? "DF " : "",
				  res->command & ATA_DRQ ? "DRQ " : "",
				  res->command & ATA_ERR ? "ERR " : "");
		}

		if (cmd->command != ATA_CMD_PACKET &&
		    (res->feature & (ATA_ICRC | ATA_UNC | ATA_IDNF |
				     ATA_ABORTED)))
			ata_dev_printk(qc->dev, KERN_ERR,
			  "error: { %s%s%s%s}\n",
			  res->feature & ATA_ICRC ? "ICRC " : "",
			  res->feature & ATA_UNC ? "UNC " : "",
			  res->feature & ATA_IDNF ? "IDNF " : "",
			  res->feature & ATA_ABORTED ? "ABRT " : "");
	}
}

/**
 *	ata_eh_report - report error handling to user
 *	@ap: ATA port to report EH about
 *
 *	Report EH to user.
 *
 *	LOCKING:
 *	None.
 */
void ata_eh_report(struct ata_port *ap)
{
	struct ata_link *link;

	__ata_port_for_each_link(link, ap)
		ata_eh_link_report(link);
}

static int ata_do_reset(struct ata_link *link, ata_reset_fn_t reset,
			unsigned int *classes, unsigned long deadline)
{
	struct ata_device *dev;
	int rc;

	ata_link_for_each_dev(dev, link)
		classes[dev->devno] = ATA_DEV_UNKNOWN;

	rc = reset(link, classes, deadline);
	if (rc)
		return rc;

	/* If any class isn't ATA_DEV_UNKNOWN, consider classification
	 * is complete and convert all ATA_DEV_UNKNOWN to
	 * ATA_DEV_NONE.
	 */
	ata_link_for_each_dev(dev, link)
		if (classes[dev->devno] != ATA_DEV_UNKNOWN)
			break;

	if (dev) {
		ata_link_for_each_dev(dev, link) {
			if (classes[dev->devno] == ATA_DEV_UNKNOWN)
				classes[dev->devno] = ATA_DEV_NONE;
		}
	}

	return 0;
}

static int ata_eh_followup_srst_needed(struct ata_link *link,
				       int rc, int classify,
				       const unsigned int *classes)
{
	if (link->flags & ATA_LFLAG_NO_SRST)
		return 0;
	if (rc == -EAGAIN)
		return 1;
	if (rc != 0)
		return 0;
	if ((link->ap->flags & ATA_FLAG_PMP) && ata_is_host_link(link))
		return 1;
	if (classify && !(link->flags & ATA_LFLAG_ASSUME_CLASS) &&
	    classes[0] == ATA_DEV_UNKNOWN)
		return 1;
	return 0;
}

int ata_eh_reset(struct ata_link *link, int classify,
		 ata_prereset_fn_t prereset, ata_reset_fn_t softreset,
		 ata_reset_fn_t hardreset, ata_postreset_fn_t postreset)
{
	const int max_tries = ARRAY_SIZE(ata_eh_reset_timeouts);
	struct ata_port *ap = link->ap;
	struct ata_eh_context *ehc = &link->eh_context;
	unsigned int *classes = ehc->classes;
	unsigned int lflags = link->flags;
	int verbose = !(ehc->i.flags & ATA_EHI_QUIET);
	int try = 0;
	struct ata_device *dev;
	unsigned long deadline, now;
	unsigned int tmp_action;
	ata_reset_fn_t reset;
	unsigned long flags;
	u32 sstatus;
	int rc;

	/* about to reset */
	spin_lock_irqsave(ap->lock, flags);
	ap->pflags |= ATA_PFLAG_RESETTING;
	spin_unlock_irqrestore(ap->lock, flags);

	ata_eh_about_to_do(link, NULL, ehc->i.action & ATA_EH_RESET_MASK);

	ata_link_for_each_dev(dev, link) {
		/* If we issue an SRST then an ATA drive (not ATAPI)
		 * may change configuration and be in PIO0 timing. If
		 * we do a hard reset (or are coming from power on)
		 * this is true for ATA or ATAPI. Until we've set a
		 * suitable controller mode we should not touch the
		 * bus as we may be talking too fast.
		 */
		dev->pio_mode = XFER_PIO_0;

		/* If the controller has a pio mode setup function
		 * then use it to set the chipset to rights. Don't
		 * touch the DMA setup as that will be dealt with when
		 * configuring devices.
		 */
		if (ap->ops->set_piomode)
			ap->ops->set_piomode(ap, dev);