From: Xiangliang Yu <yxlraid@gmail.com>
To: tj@kernel.org, JBottomley@parallels.com
Cc: dan.j.williams@intel.com, todd.e.brandt@intel.com,
lukasz.dorau@intel.com, linux-ide@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org,
Xiangliang Yu <yxlraid@gmail.com>
Subject: [PATCH 1/3] libsas: modify SATA error handler
Date: Thu, 24 Apr 2014 21:27:27 +0800 [thread overview]
Message-ID: <1398346047-10276-1-git-send-email-yxlraid@gmail.com> (raw)
Add support for SATA port softreset and port multiplier error
handling.
Signed-off-by: Xiangliang Yu <yxlraid@gmail.com>
---
drivers/scsi/libsas/sas_ata.c | 226 ++++++++++++++++++++++++++++++++++++++++-
include/scsi/libsas.h | 6 +
2 files changed, 231 insertions(+), 1 deletions(-)
diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c
index 766098a..29a19fd 100644
--- a/drivers/scsi/libsas/sas_ata.c
+++ b/drivers/scsi/libsas/sas_ata.c
@@ -442,6 +442,226 @@ static int sas_ata_hard_reset(struct ata_link *link, unsigned int *class,
return ret;
}
+static void sas_ata_freeze(struct ata_port *ap)
+{
+ struct domain_device *dev = ap->private_data;
+ struct sas_ha_struct *sas_ha = dev->port->ha;
+ struct Scsi_Host *host = sas_ha->core.shost;
+ struct sas_internal *i = to_sas_internal(host->transportt);
+
+ if (i->dft->lldd_dev_freeze)
+ i->dft->lldd_dev_freeze(dev);
+}
+
+static void sas_ata_thaw(struct ata_port *ap)
+{
+ struct domain_device *dev = ap->private_data;
+ struct sas_ha_struct *sas_ha = dev->port->ha;
+ struct Scsi_Host *host = sas_ha->core.shost;
+ struct sas_internal *i = to_sas_internal(host->transportt);
+
+ if (i->dft->lldd_dev_thaw)
+ i->dft->lldd_dev_thaw(dev);
+}
+
+static int sas_ata_wait_task_done(struct sas_task *task, unsigned long timeout,
+ int (*check_done)(struct sas_task *task))
+{
+ struct ata_port *ap = task->uldd_task;
+ unsigned long deadline;
+ int done;
+
+ if (!check_done) {
+ SAS_DPRINTK("check function is null.\n");
+ return -1;
+ }
+
+ deadline = ata_deadline(jiffies, timeout);
+ done = check_done(task);
+
+ while (done && time_before(jiffies, deadline)) {
+ ata_msleep(ap, 1);
+
+ done = check_done(task);
+ }
+
+ return done;
+}
+
+static int sas_ata_exec_polled_cmd(struct ata_port *ap, struct ata_taskfile *tf,
+ int pmp, unsigned long timeout)
+{
+ struct domain_device *dev = ap->private_data;
+ struct sas_ha_struct *sas_ha = dev->port->ha;
+ struct Scsi_Host *host = sas_ha->core.shost;
+ struct sas_internal *i = to_sas_internal(host->transportt);
+ struct sas_task *task = NULL;
+ int ret = -1;
+
+ if (!i->dft->lldd_execute_task) {
+ SAS_DPRINTK("execute function is null.\n");
+ return ret;
+ }
+
+ task = sas_alloc_task(GFP_ATOMIC);
+ if (!task) {
+ SAS_DPRINTK("failed to alloc sas task.\n");
+ goto fail;
+ }
+
+ task->dev = dev;
+ task->task_proto = SAS_PROTOCOL_SATA;
+ task->uldd_task = ap;
+
+ ata_tf_to_fis(tf, pmp, 0, (u8 *)&task->ata_task.fis);
+ task->ata_task.retry_count = 1;
+ task->task_state_flags = SAS_TASK_STATE_PENDING;
+ task->task_state_flags |= SAS_TASK_NEED_DEV_RESET;
+
+ ret = i->dft->lldd_execute_task(task, 1, GFP_ATOMIC);
+ if (ret) {
+ SAS_DPRINTK("failed to send internal task.\n");
+ goto fail;
+ }
+
+ if (timeout) {
+ ret = sas_ata_wait_task_done(task, timeout,
+ i->dft->lldd_wait_task_done);
+ if (ret) {
+ SAS_DPRINTK("get wrong status.\n");
+ goto fail;
+ }
+ }
+ list_del_init(&task->list);
+ sas_free_task(task);
+
+ return 0;
+fail:
+ if (task) {
+ list_del_init(&task->list);
+ sas_free_task(task);
+ }
+
+ return ret;
+}
+
+static int sas_ata_soft_reset(struct ata_link *link, unsigned int *class,
+ unsigned long deadline)
+{
+ struct ata_taskfile tf;
+ struct ata_port *ap = link->ap;
+ struct domain_device *dev = ap->private_data;
+ struct sas_ha_struct *sas_ha = dev->port->ha;
+ struct Scsi_Host *host = sas_ha->core.shost;
+ struct sas_internal *i = to_sas_internal(host->transportt);
+ struct sas_phy *phy;
+ unsigned long now, msecs;
+ unsigned int pmp;
+ int ret = -1;
+ int (*check_ready)(struct ata_link *link);
+
+ phy = sas_get_local_phy(dev);
+ if (scsi_is_sas_phy_local(phy))
+ check_ready = local_ata_check_ready;
+ else
+ check_ready = smp_ata_check_ready;
+ sas_put_local_phy(phy);
+
+ pmp = sata_srst_pmp(link);
+
+ msecs = 0;
+ now = jiffies;
+ if (time_after(deadline, now))
+ msecs = jiffies_to_msecs(deadline - now);
+
+ memset(&tf, 0, sizeof(struct ata_taskfile));
+ tf.ctl = ATA_SRST;
+ tf.device = ATA_DEVICE_OBS;
+
+ if (sas_ata_exec_polled_cmd(ap, &tf, pmp, msecs)) {
+ ret = -EIO;
+ goto fail;
+ }
+
+ tf.ctl &= ~ATA_SRST;
+ sas_ata_exec_polled_cmd(ap, &tf, pmp, msecs);
+
+ ret = ata_wait_after_reset(link, deadline, check_ready);
+ if (ret) {
+ SAS_DPRINTK("device is not ready.\n");
+ ret = -EIO;
+ goto fail;
+ } else {
+ if (i->dft->lldd_dev_classify)
+ *class = i->dft->lldd_dev_classify(dev);
+ }
+
+ return 0;
+
+fail:
+ SAS_DPRINTK("failed to reset.\n");
+ return ret;
+}
+
+/* According sata 3.0 spec 13.15.4.2, enable the device port */
+static int sas_ata_pmp_hard_reset(struct ata_link *link, unsigned int *class,
+ unsigned long deadline)
+{
+ struct ata_port *ap = link->ap;
+ struct domain_device *dev = ap->private_data;
+ struct sas_ha_struct *sas_ha = dev->port->ha;
+ struct Scsi_Host *host = sas_ha->core.shost;
+ struct sas_internal *i = to_sas_internal(host->transportt);
+ int ret = -1;
+ u32 scontrol = 0;
+
+ set_bit(SAS_DEV_RESET, &dev->state);
+
+ ret = sata_scr_read(link, SCR_CONTROL, &scontrol);
+ if (ret)
+ goto error;
+
+ /* enable device port */
+ scontrol = 0x1;
+ ret = sata_scr_write(link, SCR_CONTROL, scontrol);
+ if (ret)
+ goto error;
+
+ ata_msleep(ap, 1);
+
+ scontrol = 0x0;
+ ret = sata_scr_write(link, SCR_CONTROL, scontrol);
+ if (ret)
+ goto error;
+
+ ata_msleep(ap, 10);
+
+ /* check device link status */
+ if (ata_link_offline(link)) {
+ SAS_DPRINTK("link is offline.\n");
+ goto error;
+ }
+
+ /* clear X bit */
+ scontrol = 0xFFFFFFFF;
+ ret = sata_scr_write(link, SCR_ERROR, scontrol);
+ if (ret)
+ goto error;
+
+ /* may be need to wait for device sig */
+ ata_msleep(ap, 3);
+
+ /* check device class */
+ if (i->dft->lldd_dev_classify)
+ *class = i->dft->lldd_dev_classify(dev);
+
+ return 0;
+
+error:
+ SAS_DPRINTK("failed to hard reset.\n");
+ return ret;
+}
+
/*
* notify the lldd to forget the sas_task for this internal ata command
* that bypasses scsi-eh
@@ -551,8 +771,12 @@ void sas_ata_end_eh(struct ata_port *ap)
static struct ata_port_operations sas_sata_ops = {
.prereset = ata_std_prereset,
.hardreset = sas_ata_hard_reset,
+ .softreset = sas_ata_soft_reset,
+ .pmp_hardreset = sas_ata_pmp_hard_reset,
+ .freeze = sas_ata_freeze,
+ .thaw = sas_ata_thaw,
.postreset = ata_std_postreset,
- .error_handler = ata_std_error_handler,
+ .error_handler = sata_pmp_error_handler,
.post_internal_cmd = sas_ata_post_internal,
.qc_defer = ata_std_qc_defer,
.qc_prep = ata_noop_qc_prep,
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index ef7872c..a26466a 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -689,6 +689,12 @@ struct sas_domain_function_template {
/* GPIO support */
int (*lldd_write_gpio)(struct sas_ha_struct *, u8 reg_type,
u8 reg_index, u8 reg_count, u8 *write_data);
+
+ /* ATA EH functions */
+ void (*lldd_dev_freeze)(struct domain_device *);
+ void (*lldd_dev_thaw)(struct domain_device *);
+ int (*lldd_wait_task_done)(struct sas_task *);
+ int (*lldd_dev_classify)(struct domain_device *);
};
extern int sas_register_ha(struct sas_ha_struct *);
--
1.7.1
next reply other threads:[~2014-04-24 13:28 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-04-24 13:27 Xiangliang Yu [this message]
[not found] ` <CAP0P+NWxxxnQH1e3tXcqwud7EaV4kRRMW46Ad-PWoMWFzgy1eg@mail.gmail.com>
2014-06-03 7:13 ` Xiangliang Yu
2014-06-03 16:05 ` Dan Williams
2014-08-06 10:50 ` Xiangliang Yu
2014-08-06 17:21 ` Dan Williams
2014-08-07 1:50 ` Xiangliang Yu
2014-06-04 1:41 ` Dan Williams
2014-08-07 23:54 ` Dan Williams
2014-08-08 9:46 ` Xiangliang Yu
2014-08-25 16:02 ` Dan Williams
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1398346047-10276-1-git-send-email-yxlraid@gmail.com \
--to=yxlraid@gmail.com \
--cc=JBottomley@parallels.com \
--cc=dan.j.williams@intel.com \
--cc=linux-ide@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=lukasz.dorau@intel.com \
--cc=tj@kernel.org \
--cc=todd.e.brandt@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®