Skip to contentSkip to conversation

AHCI: disabled FBS prior to issuing software reset

PATCH In review Xiangliang Yu 3 participants 10 messages Sep 28, 2013 – Oct 15, 2013 libata

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 AHCI: disabled FBS prior to issuing software reset

If device is attached to port multiplier, the detection process look like this:

	ahci_hardreset(link, class, deadline)
	if (class == ATA_DEV_PMP) {
		sata_pmp_attach(dev)    /* will enable FBS */
		sata_pmp_init_links(ap, nr_ports);
		ata_for_each_link(link, ap, EDGE) {
			sata_std_hardreset(link, class, deadline);
			if (link_is_online)
				ahci_softreset(link, class, deadline);
		}
	}

But, according to chapter 9.3.9 in AHCI spec: Prior to issuing software reset, software shall clear PxCMD.ST to '0' and then clear PxFBS.EN to '0'.

---
 drivers/ata/libahci.c |   10 ++++++++++
 1 files changed, 10 insertions(+), 0 deletions(-)
drivers/ata/libahci.c +10 -0
index acfd0f7..8d024a4 100644
@@ -1267,9 +1267,11 @@ int ahci_do_softreset(struct ata_link *link, unsigned int *class,
 {
 	struct ata_port *ap = link->ap;
 	struct ahci_host_priv *hpriv = ap->host->private_data;
+	struct ahci_port_priv *pp = ap->private_data;
 	const char *reason = NULL;
 	unsigned long now, msecs;
 	struct ata_taskfile tf;
+	bool flag = FALSE;
 	int rc;
 
 	DPRINTK("ENTER\n");
@@ -1279,6 +1281,11 @@ int ahci_do_softreset(struct ata_link *link, unsigned int *class,
 	if (rc && rc != -EOPNOTSUPP)
 		ata_link_warn(link, "failed to reset engine (errno=%d)\n", rc);
 
+	if (!ata_is_host_link(link) && pp->fbs_enabled) {
+		ahci_disable_fbs(ap);
+		flag = TRUE;
+	}
+
 	ata_tf_init(link->device, &tf);
 
 	/* issue the first D2H Register FIS */
@@ -1319,6 +1326,9 @@ int ahci_do_softreset(struct ata_link *link, unsigned int *class,
 	} else
 		*class = ahci_dev_classify(ap);
 
+	if (flag)
+		ahci_enable_fbs(ap);
+
 	DPRINTK("EXIT, class=%u\n", *class);
 	return 0;
 
Signature · 1 line
1.7.1
#1 lore raw

Hello,

Quoted 12 lines · Xiangliang Yu

If device is attached to port multiplier, the detection process look like this:

	ahci_hardreset(link, class, deadline)
	if (class == ATA_DEV_PMP) {
		sata_pmp_attach(dev)    /* will enable FBS */
		sata_pmp_init_links(ap, nr_ports);
		ata_for_each_link(link, ap, EDGE) {
			sata_std_hardreset(link, class, deadline);
			if (link_is_online)
				ahci_softreset(link, class, deadline);
		}
	}
Quoted 2 lines

But, according to chapter 9.3.9 in AHCI spec: Prior to issuing software reset, software shall clear PxCMD.ST to '0' and then clear PxFBS.EN to '0'.

How was this tested? Do you observe any behavior difference? At this point, we are not using SRST on PMP ports anyway, so I can't see how this would make any difference.

Thanks.

Signature · 1 line
tejun
#2 lore raw

hi, Tejun
i had tested the patch with Marvell 88se9235. And driver can find disk if FBS disabled, or can't find disk.
Please see chapter 9.3.9 and 9.3.8 of AHCI spec. Thanks.

2013/9/28 Tejun Heo <[email protected]>:

Quoted 26 lines

Hello,

Quoted 12 lines · Xiangliang Yu

If device is attached to port multiplier, the detection process look like this:

      ahci_hardreset(link, class, deadline)
      if (class == ATA_DEV_PMP) {
              sata_pmp_attach(dev)    /* will enable FBS */
              sata_pmp_init_links(ap, nr_ports);
              ata_for_each_link(link, ap, EDGE) {
                      sata_std_hardreset(link, class, deadline);
                      if (link_is_online)
                              ahci_softreset(link, class, deadline);
              }
      }
Quoted 2 lines

But, according to chapter 9.3.9 in AHCI spec: Prior to issuing software reset, software shall clear PxCMD.ST to '0' and then clear PxFBS.EN to '0'.

How was this tested? Do you observe any behavior difference? At this point, we are not using SRST on PMP ports anyway, so I can't see how this would make any difference.

Thanks.

--
tejun

#3 lore raw

Hello,

Quoted 3 lines · xiangliang yu

hi, Tejun
i had tested the patch with Marvell 88se9235. And driver can find disk if FBS disabled, or can't find disk.

So it can't find the disk if FBS stays enabled? Can you please attach the boot log before & after?

Quoted 1 line

Please see chapter 9.3.9 and 9.3.8 of AHCI spec. Thanks.

Given how crappy and outdated PMP hardware are in general, I'm very unwilling to make changes just because the spec says so, so please stop repeating the spec and provide data.

Thanks.

Signature · 1 line
tejun
#4 lore raw

Hi,

Quoted 7 lines · xiangliang yu
Quoted 3 lines

hi, Tejun
i had tested the patch with Marvell 88se9235. And driver can find disk if FBS disabled, or can't find disk.

So it can't find the disk if FBS stays enabled? Can you please attach the boot log before & after?

Below is boot log:

ata8: SATA link up 6.0 Gbps (SStatus 133 SControl 330) ata_eh_revalidate_and_attach: dev->class = 5.
ata8.15: Port Multiplier 1.2, 0x1b4b:0x9715 r160, 5 ports, feat 0x1/0x1f ahci 0000:03:00.0: FBS is enabled
ata8.00: hard resetting link
ata8.00: SATA link down (SStatus 0 SControl 330)
ata8.01: hard resetting link
ata8.01: SATA link down (SStatus 0 SControl 330)
ata8.02: hard resetting link
ata8.02: SATA link down (SStatus 0 SControl 330)
ata8.03: hard resetting link
ata8.03: SATA link up 6.0 Gbps (SStatus 133 SControl 133) ata8.04: hard resetting link
ata8.04: failed to resume link (SControl 133)
ata8.04: failed to read SCR 0 (Emask=0x40)
ata8.04: failed to read SCR 0 (Emask=0x40)
ata8.04: failed to read SCR 1 (Emask=0x40)
ata8.04: failed to read SCR 0 (Emask=0x40) ata_eh_revalidate_and_attach: dev->class = 1.
ata8.03: native sectors (2) is smaller than sectors (976773168) ata8.03: ATA-8: ST3500413AS, JC4B, max UDMA/133
ata8.03: 976773168 sectors, multi 0: LBA48 NCQ (depth 31/32) ata8.03: configured for UDMA/133 ata_eh_revalidate_and_attach: dev->class = 1.
ata8.04: failed to IDENTIFY (I/O error, err_mask=0x100) ata8.15: hard resetting link
ata8.15: SATA link up 6.0 Gbps (SStatus 133 SControl 330) ata8.15: Port Multiplier vendor mismatch '0x1b4b' != '0x133' ata8.15: PMP revalidation failed (errno=-19)
ata8.15: hard resetting link
ata8.15: SATA link up 6.0 Gbps (SStatus 133 SControl 330) ata8.15: Port Multiplier vendor mismatch '0x1b4b' != '0x133' ata8.15: PMP revalidation failed (errno=-19)
ata8.15: limiting SATA link speed to 3.0 Gbps
ata8.15: hard resetting link
ata8.15: SATA link up 3.0 Gbps (SStatus 123 SControl 320) ata8.15: Port Multiplier vendor mismatch '0x1b4b' != '0x133' ata8.15: PMP revalidation failed (errno=-19)
ata8.15: failed to recover PMP after 5 tries, giving up ata8.15: Port Multiplier detaching
ata8.03: disabled
ata8.00: disabled
ata8: EH complete

#5 lore raw

Hello,

Quoted 4 lines · xiangliang yu
Quoted 2 lines

So it can't find the disk if FBS stays enabled? Can you please attach the boot log before & after?

Below is boot log:

And it works after the patch, right? Can you please update the patch description so that it includes what was failing before and how it was tested?

Thanks.

Signature · 1 line
tejun
#6 lore raw

Hi,

Quoted 10 lines · xiangliang yu
Quoted 4 lines
Quoted 2 lines

So it can't find the disk if FBS stays enabled? Can you please attach the boot log before & after?

Below is boot log:

And it works after the patch, right? Can you please update the patch description so that it includes what was failing before and how it was tested?

Yes. I'll update the patch later.

Thanks.

#7 lore raw
In reply to Tejun Heo

Hi,

Quoted 9 lines · xiangliang yu
Quoted 4 lines
Quoted 2 lines

So it can't find the disk if FBS stays enabled? Can you please attach the boot log before & after?

Below is boot log:

And it works after the patch, right? Can you please update the patch description so that it includes what was failing before and how it was tested?

How about this update:

Subject: [PATCH 1/1] AHCI: disabled FBS prior to issuing software reset

Tested with Marvell 88se9125, attached with one port mulitplier(5 ports) and one disk, we will get following boot log messages if using current code:

ata8: SATA link up 6.0 Gbps (SStatus 133 SControl 330) ata8.15: Port Multiplier 1.2, 0x1b4b:0x9715 r160, 5 ports, feat 0x1/0x1f ahci 0000:03:00.0: FBS is enabled
ata8.00: hard resetting link
ata8.00: SATA link down (SStatus 0 SControl 330)
ata8.01: hard resetting link
ata8.01: SATA link down (SStatus 0 SControl 330)
ata8.02: hard resetting link
ata8.02: SATA link down (SStatus 0 SControl 330)
ata8.03: hard resetting link
ata8.03: SATA link up 6.0 Gbps (SStatus 133 SControl 133) ata8.04: hard resetting link
ata8.04: failed to resume link (SControl 133)
ata8.04: failed to read SCR 0 (Emask=0x40)
ata8.04: failed to read SCR 0 (Emask=0x40)
ata8.04: failed to read SCR 1 (Emask=0x40)
ata8.04: failed to read SCR 0 (Emask=0x40)
ata8.03: native sectors (2) is smaller than sectors (976773168) ata8.03: ATA-8: ST3500413AS, JC4B, max UDMA/133
ata8.03: 976773168 sectors, multi 0: LBA48 NCQ (depth 31/32) ata8.03: configured for UDMA/133
ata8.04: failed to IDENTIFY (I/O error, err_mask=0x100) ata8.15: hard resetting link
ata8.15: SATA link up 6.0 Gbps (SStatus 133 SControl 330) ata8.15: Port Multiplier vendor mismatch '0x1b4b' != '0x133' ata8.15: PMP revalidation failed (errno=-19)
ata8.15: hard resetting link
ata8.15: SATA link up 6.0 Gbps (SStatus 133 SControl 330) ata8.15: Port Multiplier vendor mismatch '0x1b4b' != '0x133' ata8.15: PMP revalidation failed (errno=-19)
ata8.15: limiting SATA link speed to 3.0 Gbps
ata8.15: hard resetting link
ata8.15: SATA link up 3.0 Gbps (SStatus 123 SControl 320) ata8.15: Port Multiplier vendor mismatch '0x1b4b' != '0x133' ata8.15: PMP revalidation failed (errno=-19)
ata8.15: failed to recover PMP after 5 tries, giving up ata8.15: Port Multiplier detaching
ata8.03: disabled
ata8.00: disabled
ata8: EH complete

The reason is that current detection code doesn't follow AHCI spec:

First,the port multiplier detection process look like this:

ahci_hardreset(link, class, deadline)
if (class == ATA_DEV_PMP) {
sata_pmp_attach(dev) /* will enable FBS */ sata_pmp_init_links(ap, nr_ports);
ata_for_each_link(link, ap, EDGE) {
sata_std_hardreset(link, class, deadline);
if (link_is_online) /* do soft reset */
ahci_softreset(link, class, deadline);
}
}
But, according to chapter 9.3.9 in AHCI spec: Prior to issuing software reset, software shall clear PxCMD.ST to '0' and then clear PxFBS.EN to '0'.

The patch test ok with kernel 3.11.1.

---
 drivers/ata/libahci.c |   10 ++++++++++
 1 files changed, 10 insertions(+), 0 deletions(-)
drivers/ata/libahci.c +0 -0
index 34c8216..c2a29bf 100644
@@ -1266,9 +1266,11 @@ int ahci_do_softreset(struct ata_link *link,

unsigned int *class,
{
struct ata_port *ap = link->ap;
struct ahci_host_priv *hpriv = ap->host->private_data;
+ struct ahci_port_priv *pp = ap->private_data; const char *reason = NULL;
unsigned long now, msecs;
struct ata_taskfile tf;
+ bool fbs_flag = false;
int rc;

DPRINTK("ENTER\n");
@@ -1278,6 +1280,11 @@ int ahci_do_softreset(struct ata_link *link, unsigned int *class,
if (rc && rc != -EOPNOTSUPP)
ata_link_warn(link, "failed to reset engine (errno=%d)\n", rc);

+ if (!ata_is_host_link(link) && pp->fbs_enabled) {
+ ahci_disable_fbs(ap);
+ fbs_flag = true;
+ }
+
ata_tf_init(link->device, &tf);

/* issue the first D2H Register FIS */
@@ -1318,6 +1325,9 @@ int ahci_do_softreset(struct ata_link *link, unsigned int *class,
} else
*class = ahci_dev_classify(ap);

+ if (fbs_flag)
+ ahci_enable_fbs(ap);
+
DPRINTK("EXIT, class=%u\n", *class);
return 0;

Signature · 4 lines
1.7.1


Thanks!
#8 lore raw

Hello,

Review int ahci_do_softreset(struct ata_link *link, · line 1283

Quoted 5 lines · xiangliang yu
(quoted diff) +0 -0
@@ -1278,6 +1280,11 @@ int ahci_do_softreset(struct ata_link *link,
 unsigned int *class,
  if (rc && rc != -EOPNOTSUPP)
  ata_link_warn(link, "failed to reset engine (errno=%d)\n", rc);
 

Please add a comment here.

Review int ahci_do_softreset(struct ata_link *link, · line 1287

Quoted 4 lines
(quoted diff) +4 -0
+ if (!ata_is_host_link(link) && pp->fbs_enabled) {
+ ahci_disable_fbs(ap);
+ fbs_flag = true;
+ }

White space damaged?

Review int ahci_do_softreset(struct ata_link *link, · line 1331

Quoted 12 lines
(quoted diff) +1 -0
+
  ata_tf_init(link->device, &tf);
 

/* issue the first D2H Register FIS */

(quoted diff) +3 -0
@@ -1318,6 +1325,9 @@ int ahci_do_softreset(struct ata_link *link,
 unsigned int *class,
  } else
  *class = ahci_dev_classify(ap);
 
+ if (fbs_flag)
+ ahci_enable_fbs(ap);
+

Thanks.

Signature · 1 line
tejun
#9 lore raw
In reply to Xiangliang Yu

Hello.

Quoted 25 lines · Xiangliang Yu

If device is attached to port multiplier, the detection process look like this:

	ahci_hardreset(link, class, deadline)
	if (class == ATA_DEV_PMP) {
		sata_pmp_attach(dev)    /* will enable FBS */
		sata_pmp_init_links(ap, nr_ports);
		ata_for_each_link(link, ap, EDGE) {
			sata_std_hardreset(link, class, deadline);
			if (link_is_online)
				ahci_softreset(link, class, deadline);
		}
	}

But, according to chapter 9.3.9 in AHCI spec: Prior to issuing software reset, software shall clear PxCMD.ST to '0' and then clear PxFBS.EN to '0'.

---
  drivers/ata/libahci.c |   10 ++++++++++
  1 files changed, 10 insertions(+), 0 deletions(-)
drivers/ata/libahci.c +0 -0
index acfd0f7..8d024a4 100644

drivers/ata/libahci.c int ahci_do_softreset(struct ata_link *link, unsigned int *class, · line 1274

Quoted 8 lines
drivers/ata/libahci.c +2 -0
@@ -1267,9 +1267,11 @@ int ahci_do_softreset(struct ata_link *link, unsigned int *class,
  {
  	struct ata_port *ap = link->ap;
  	struct ahci_host_priv *hpriv = ap->host->private_data;
+	struct ahci_port_priv *pp = ap->private_data;
  	const char *reason = NULL;
  	unsigned long now, msecs;
  	struct ata_taskfile tf;
+	bool flag = FALSE;

Why not lowercase 'false'? And this is not a good name for a variable (too
generic), it should better be called 'fbs' or something...

drivers/ata/libahci.c int ahci_do_softreset(struct ata_link *link, unsigned int *class, · line 1286

Quoted 7 lines
drivers/ata/libahci.c +3 -0
@@ -1279,6 +1281,11 @@ int ahci_do_softreset(struct ata_link *link, unsigned int *class,
  	if (rc && rc != -EOPNOTSUPP)
  		ata_link_warn(link, "failed to reset engine (errno=%d)\n", rc);
 
+	if (!ata_is_host_link(link) && pp->fbs_enabled) {
+		ahci_disable_fbs(ap);
+		flag = TRUE;

Why not lowercase 'true'?

WBR, Sergei