Re: hsm violation

Re: hsm violation

[Robert Hancock]

Andrew Morton wrote:
> On Sun, 24 Jun 2007 14:32:22 +0200 Enrico Sardi <enricoss@tiscali.it> wrote:
>> [   61.176000] ata1.00: exception Emask 0x2 SAct 0x2 SErr 0x0 action 0x2 
>> frozen
>> [   61.176000] ata1.00: (spurious completions during NCQ issue=0x0 
>> SAct=0x2 FIS=005040a1:00000004)

..

> 
> It's not obvious (to me) whether this is a driver bug, a hardware bug,
> expected-normal-behaviour or what - those diagnostics (which we get to
> see distressingly frequently) are pretty obscure.

The spurious completions during NCQ error is indicating that the drive 
has indicated it's completed NCQ command tags which weren't outstanding. 
  It's normally a result of a bad NCQ implementation on the drive. 
Technically we can live with it, but it's rather dangerous (if it 
indicates completions for non-outstanding commands, how do we know it 
doesn't indicate completions for actually outstanding commands that 
aren't actually completed yet..)

-- 
Robert Hancock      Saskatoon, SK, Canada
To email, remove "nospam" from hancockr@nospamshaw.ca
Home Page: http://www.roberthancock.com/

Re: hsm violation

[Tejun Heo]

Robert Hancock wrote:
> Andrew Morton wrote:
>> On Sun, 24 Jun 2007 14:32:22 +0200 Enrico Sardi <enricoss@tiscali.it>
>> wrote:
>>> [   61.176000] ata1.00: exception Emask 0x2 SAct 0x2 SErr 0x0 action
>>> 0x2 frozen
>>> [   61.176000] ata1.00: (spurious completions during NCQ issue=0x0
>>> SAct=0x2 FIS=005040a1:00000004)
>>
>> It's not obvious (to me) whether this is a driver bug, a hardware bug,
>> expected-normal-behaviour or what - those diagnostics (which we get to
>> see distressingly frequently) are pretty obscure.
> 
> The spurious completions during NCQ error is indicating that the drive
> has indicated it's completed NCQ command tags which weren't outstanding.
>  It's normally a result of a bad NCQ implementation on the drive.
> Technically we can live with it, but it's rather dangerous (if it
> indicates completions for non-outstanding commands, how do we know it
> doesn't indicate completions for actually outstanding commands that
> aren't actually completed yet..)

There is a small race window there.  Please consider the following sequence.

1. drive sends SDB FIS with spurious completion in it.
2. block layer issues new r/w command to the drive.  SDB FIS is still in
flight.
3. ata driver issues the command (the pending bit is set prior to
transmitting command FIS).
4. controller completes receiving FIS from #1.  Driver reads the mask
and completes all indicated commands.  If spurious completion in #1
happens to match the slot allocated in #3, the driver just completed a
command which hasn't been issued to the drive yet.

So, it actually is dangerous.  We might even be seeing the real
completion as spurious one (as the command is completed prematurely).

It seems all those HTS541* drives share this problem.  Four of them are
already on the blacklist and the other OS reportedly blacklists three of
them too.  I'll submit a patch to add HTS541616J9SA00.

Thanks.

-- 
tejun

[PATCH 2.6.22-rc5] libata: add HTS541616J9SA00 to NCQ blacklist

[Tejun Heo]

Another member of HTS5416* family doing spurious NCQ completion.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Enrico Sardi <enricoss@tiscali.it>
---
 drivers/ata/libata-core.c |    1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index adfae9d..fbca8d8 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -3803,6 +3803,7 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
 	/* Drives which do spurious command completion */
 	{ "HTS541680J9SA00",	"SB2IC7EP",	ATA_HORKAGE_NONCQ, },
 	{ "HTS541612J9SA00",	"SBDIC7JP",	ATA_HORKAGE_NONCQ, },
+	{ "Hitachi HTS541616J9SA00", "SB4OC70P", ATA_HORKAGE_NONCQ, },
 	{ "WDC WD740ADFD-00NLR1", NULL,		ATA_HORKAGE_NONCQ, },
 
 	/* Devices with NCQ limits */

Re: [PATCH 2.6.22-rc5] libata: add HTS541616J9SA00 to NCQ blacklist

[Robert Hancock]

Tejun Heo wrote:
> Another member of HTS5416* family doing spurious NCQ completion.
> 
> Signed-off-by: Tejun Heo <htejun@gmail.com>
> Cc: Enrico Sardi <enricoss@tiscali.it>
> ---
>  drivers/ata/libata-core.c |    1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
> index adfae9d..fbca8d8 100644
> --- a/drivers/ata/libata-core.c
> +++ b/drivers/ata/libata-core.c
> @@ -3803,6 +3803,7 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
>  	/* Drives which do spurious command completion */
>  	{ "HTS541680J9SA00",	"SB2IC7EP",	ATA_HORKAGE_NONCQ, },
>  	{ "HTS541612J9SA00",	"SBDIC7JP",	ATA_HORKAGE_NONCQ, },
> +	{ "Hitachi HTS541616J9SA00", "SB4OC70P", ATA_HORKAGE_NONCQ, },
>  	{ "WDC WD740ADFD-00NLR1", NULL,		ATA_HORKAGE_NONCQ, },
>  
>  	/* Devices with NCQ limits */
> 

Is that the right ID string? Strange that that one has Hitachi at the 
front and the others don't..

-- 
Robert Hancock      Saskatoon, SK, Canada
To email, remove "nospam" from hancockr@nospamshaw.ca
Home Page: http://www.roberthancock.com/

Re: [PATCH 2.6.22-rc5] libata: add HTS541616J9SA00 to NCQ blacklist

[Tejun Heo]

Robert Hancock wrote:
> Tejun Heo wrote:
>> Another member of HTS5416* family doing spurious NCQ completion.
>>
>> Signed-off-by: Tejun Heo <htejun@gmail.com>
>> Cc: Enrico Sardi <enricoss@tiscali.it>
>> ---
>>  drivers/ata/libata-core.c |    1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
>> index adfae9d..fbca8d8 100644
>> --- a/drivers/ata/libata-core.c
>> +++ b/drivers/ata/libata-core.c
>> @@ -3803,6 +3803,7 @@ static const struct ata_blacklist_entry
>> ata_device_blacklist [] = {
>>      /* Drives which do spurious command completion */
>>      { "HTS541680J9SA00",    "SB2IC7EP",    ATA_HORKAGE_NONCQ, },
>>      { "HTS541612J9SA00",    "SBDIC7JP",    ATA_HORKAGE_NONCQ, },
>> +    { "Hitachi HTS541616J9SA00", "SB4OC70P", ATA_HORKAGE_NONCQ, },
>>      { "WDC WD740ADFD-00NLR1", NULL,        ATA_HORKAGE_NONCQ, },
>>  
>>      /* Devices with NCQ limits */
>>
> 
> Is that the right ID string? Strange that that one has Hitachi at the
> front and the others don't..

Yeah, I realized that and asked Enrico about it.  :-)

-- 
tejun

Re: [PATCH 2.6.22-rc5] libata: add HTS541616J9SA00 to NCQ blacklist

[Petr Vandrovec]

Tejun Heo wrote:
> Robert Hancock wrote:
>> Tejun Heo wrote:
>>> Another member of HTS5416* family doing spurious NCQ completion.
>>>
>>> Signed-off-by: Tejun Heo <htejun@gmail.com>
>>> Cc: Enrico Sardi <enricoss@tiscali.it>
>>> ---
>>>  drivers/ata/libata-core.c |    1 +
>>>  1 file changed, 1 insertion(+)
>>>
>>> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
>>> index adfae9d..fbca8d8 100644
>>> --- a/drivers/ata/libata-core.c
>>> +++ b/drivers/ata/libata-core.c
>>> @@ -3803,6 +3803,7 @@ static const struct ata_blacklist_entry
>>> ata_device_blacklist [] = {
>>>      /* Drives which do spurious command completion */
>>>      { "HTS541680J9SA00",    "SB2IC7EP",    ATA_HORKAGE_NONCQ, },
>>>      { "HTS541612J9SA00",    "SBDIC7JP",    ATA_HORKAGE_NONCQ, },
>>> +    { "Hitachi HTS541616J9SA00", "SB4OC70P", ATA_HORKAGE_NONCQ, },
>>>      { "WDC WD740ADFD-00NLR1", NULL,        ATA_HORKAGE_NONCQ, },
>>>  
>>>      /* Devices with NCQ limits */
>>>
>> Is that the right ID string? Strange that that one has Hitachi at the
>> front and the others don't..
> 
> Yeah, I realized that and asked Enrico about it.  :-)

I think that "new" one is correct, while old ones are incorrect (unless 
it uses strstr()) - all my Hitachis claim to be Hitachis - like this one 
(which seems to work fine with NCQ):

gwy:~# hdparm -i /dev/sda

/dev/sda:

  Model=Hitachi HDT725032VLA380                 , FwRev=V54OA52A, 
SerialNo=      VFA200R208LH5J
  Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }

								Petr

Re: [PATCH 2.6.22-rc5] libata: add HTS541616J9SA00 to NCQ blacklist

[Tejun Heo]

Petr Vandrovec wrote:
>>>> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
>>>> index adfae9d..fbca8d8 100644
>>>> --- a/drivers/ata/libata-core.c
>>>> +++ b/drivers/ata/libata-core.c
>>>> @@ -3803,6 +3803,7 @@ static const struct ata_blacklist_entry
>>>> ata_device_blacklist [] = {
>>>>      /* Drives which do spurious command completion */
>>>>      { "HTS541680J9SA00",    "SB2IC7EP",    ATA_HORKAGE_NONCQ, },
>>>>      { "HTS541612J9SA00",    "SBDIC7JP",    ATA_HORKAGE_NONCQ, },
>>>> +    { "Hitachi HTS541616J9SA00", "SB4OC70P", ATA_HORKAGE_NONCQ, },
>>>>      { "WDC WD740ADFD-00NLR1", NULL,        ATA_HORKAGE_NONCQ, },
>>>>  
>>>>      /* Devices with NCQ limits */
>>>>
>>> Is that the right ID string? Strange that that one has Hitachi at the
>>> front and the others don't..
>>
>> Yeah, I realized that and asked Enrico about it.  :-)
> 
> I think that "new" one is correct, while old ones are incorrect (unless
> it uses strstr()) - all my Hitachis claim to be Hitachis - like this one
> (which seems to work fine with NCQ):
> 
> gwy:~# hdparm -i /dev/sda
> 
> /dev/sda:
> 
>  Model=Hitachi HDT725032VLA380                 , FwRev=V54OA52A,
> SerialNo=      VFA200R208LH5J
>  Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }

Hmmm... The last one (HTS541612J9SA00) is taken directly from hdparm
output, and I think I verified the patch with the reporter.  Hmm... Can
anyone verify these module strings?

-- 
tejun

Re: [PATCH 2.6.22-rc5] libata: add HTS541616J9SA00 to NCQ blacklist

[Robert Hancock]

Tejun Heo wrote:
> Petr Vandrovec wrote:
>>>>> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
>>>>> index adfae9d..fbca8d8 100644
>>>>> --- a/drivers/ata/libata-core.c
>>>>> +++ b/drivers/ata/libata-core.c
>>>>> @@ -3803,6 +3803,7 @@ static const struct ata_blacklist_entry
>>>>> ata_device_blacklist [] = {
>>>>>      /* Drives which do spurious command completion */
>>>>>      { "HTS541680J9SA00",    "SB2IC7EP",    ATA_HORKAGE_NONCQ, },
>>>>>      { "HTS541612J9SA00",    "SBDIC7JP",    ATA_HORKAGE_NONCQ, },
>>>>> +    { "Hitachi HTS541616J9SA00", "SB4OC70P", ATA_HORKAGE_NONCQ, },
>>>>>      { "WDC WD740ADFD-00NLR1", NULL,        ATA_HORKAGE_NONCQ, },
>>>>>  
>>>>>      /* Devices with NCQ limits */
>>>>>
>>>> Is that the right ID string? Strange that that one has Hitachi at the
>>>> front and the others don't..
>>> Yeah, I realized that and asked Enrico about it.  :-)
>> I think that "new" one is correct, while old ones are incorrect (unless
>> it uses strstr()) - all my Hitachis claim to be Hitachis - like this one
>> (which seems to work fine with NCQ):
>>
>> gwy:~# hdparm -i /dev/sda
>>
>> /dev/sda:
>>
>>  Model=Hitachi HDT725032VLA380                 , FwRev=V54OA52A,
>> SerialNo=      VFA200R208LH5J
>>  Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
> 
> Hmmm... The last one (HTS541612J9SA00) is taken directly from hdparm
> output, and I think I verified the patch with the reporter.  Hmm... Can
> anyone verify these module strings?

Could well be that they've started attaching Hitachi to the ID strings 
now.. In the past it hasn't seemed to have been Hitachi's (and IBM's 
before that) practice to have it there, but maybe they see the advantage 
of being able to figure out who made the drive now :-)

-- 
Robert Hancock      Saskatoon, SK, Canada
To email, remove "nospam" from hancockr@nospamshaw.ca
Home Page: http://www.roberthancock.com/

Re: [PATCH 2.6.22-rc5] libata: add HTS541616J9SA00 to NCQ blacklist

[Petr Vandrovec]

Robert Hancock wrote:
> Tejun Heo wrote:
>> Petr Vandrovec wrote:
>>>>>> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
>>>>>> index adfae9d..fbca8d8 100644
>>>>>> --- a/drivers/ata/libata-core.c
>>>>>> +++ b/drivers/ata/libata-core.c
>>>>>> @@ -3803,6 +3803,7 @@ static const struct ata_blacklist_entry
>>>>>> ata_device_blacklist [] = {
>>>>>>      /* Drives which do spurious command completion */
>>>>>>      { "HTS541680J9SA00",    "SB2IC7EP",    ATA_HORKAGE_NONCQ, },
>>>>>>      { "HTS541612J9SA00",    "SBDIC7JP",    ATA_HORKAGE_NONCQ, },
>>>>>> +    { "Hitachi HTS541616J9SA00", "SB4OC70P", ATA_HORKAGE_NONCQ, },
>>>>>>      { "WDC WD740ADFD-00NLR1", NULL,        ATA_HORKAGE_NONCQ, },
>>>>>>  
>>>>>>      /* Devices with NCQ limits */
>>>>>>
>>>>> Is that the right ID string? Strange that that one has Hitachi at the
>>>>> front and the others don't..
>>>> Yeah, I realized that and asked Enrico about it.  :-)
>>> I think that "new" one is correct, while old ones are incorrect (unless
>>> it uses strstr()) - all my Hitachis claim to be Hitachis - like this one
>>> (which seems to work fine with NCQ):
>>>
>>> gwy:~# hdparm -i /dev/sda
>>>
>>> /dev/sda:
>>>
>>>  Model=Hitachi HDT725032VLA380                 , FwRev=V54OA52A,
>>> SerialNo=      VFA200R208LH5J
>>>  Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
>>
>> Hmmm... The last one (HTS541612J9SA00) is taken directly from hdparm
>> output, and I think I verified the patch with the reporter.  Hmm... Can
>> anyone verify these module strings?
> 
> Could well be that they've started attaching Hitachi to the ID strings 
> now.. In the past it hasn't seemed to have been Hitachi's (and IBM's 
> before that) practice to have it there, but maybe they see the advantage 
> of being able to figure out who made the drive now :-)

Perhaps ones sold directly by Hitachi are Hitachi, while ones sold 
through OEMs are no-name?
							Petr

Re: [PATCH 2.6.22-rc5] libata: add HTS541616J9SA00 to NCQ blacklist

[Tejun Heo]

Petr Vandrovec wrote:
>>> Hmmm... The last one (HTS541612J9SA00) is taken directly from hdparm
>>> output, and I think I verified the patch with the reporter.  Hmm... Can
>>> anyone verify these module strings?
>>
>> Could well be that they've started attaching Hitachi to the ID strings
>> now.. In the past it hasn't seemed to have been Hitachi's (and IBM's
>> before that) practice to have it there, but maybe they see the
>> advantage of being able to figure out who made the drive now :-)
> 
> Perhaps ones sold directly by Hitachi are Hitachi, while ones sold
> through OEMs are no-name?

I dunno but I doubt they care that much about the model string but one
way or the other we might have to do pattern matching on HTS541* anyway.

-- 
tejun

Re: [PATCH 2.6.22-rc5] libata: add HTS541616J9SA00 to NCQ blacklist

[Tejun Heo]

Tejun Heo wrote:
> Another member of HTS5416* family doing spurious NCQ completion.
> 
> Signed-off-by: Tejun Heo <htejun@gmail.com>
> Cc: Enrico Sardi <enricoss@tiscali.it>

Jeff, ping.

-- 
tejun

Re: [PATCH 2.6.22-rc5] libata: add HTS541616J9SA00 to NCQ blacklist

[Jeff Garzik]

Tejun Heo wrote:
> Another member of HTS5416* family doing spurious NCQ completion.
> 
> Signed-off-by: Tejun Heo <htejun@gmail.com>
> Cc: Enrico Sardi <enricoss@tiscali.it>
> ---
>  drivers/ata/libata-core.c |    1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
> index adfae9d..fbca8d8 100644
> --- a/drivers/ata/libata-core.c
> +++ b/drivers/ata/libata-core.c
> @@ -3803,6 +3803,7 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
>  	/* Drives which do spurious command completion */
>  	{ "HTS541680J9SA00",	"SB2IC7EP",	ATA_HORKAGE_NONCQ, },
>  	{ "HTS541612J9SA00",	"SBDIC7JP",	ATA_HORKAGE_NONCQ, },
> +	{ "Hitachi HTS541616J9SA00", "SB4OC70P", ATA_HORKAGE_NONCQ, },
>  	{ "WDC WD740ADFD-00NLR1", NULL,		ATA_HORKAGE_NONCQ, },

applied