Re: [PATCH] libata: add Samsung HD401LJ to the NCQ blacklist

Re: [PATCH] libata: add Samsung HD401LJ to the NCQ blacklist

[Robert Hancock]

Max Kellermann wrote:
> Yet another hard drive which doesn't seem to get NCQ right.
> 
>  ata1: EH in ADMA mode, notifier 0x0 notifier_error 0x0 gen_ctl
>  0x1501000 status 0x400 next cpb count 0xB next cpb idx 0x0
>  [...]
>  ata1: timeout waiting for ADMA IDLE, stat=0x400
>  ata1: timeout waiting for ADMA LEGACY, stat=0x400
>  ata1.00: exception Emask 0x0 SAct 0x3fff SErr 0x200000 action 0x2
>  frozen
>  [...]
>  ata1: soft resetting port

Don't know if I noticed this before, but the controller seems to be 
reporting an SError, looks like a link layer CRC error. This could be a 
potential symptom of the drive and controller just not getting along, 
but you might want to try a different SATA cable as well.

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

[PATCH] libata: add Samsung HD401LJ to the NCQ blacklist

[Max Kellermann]

Yet another hard drive which doesn't seem to get NCQ right.

 ata1: EH in ADMA mode, notifier 0x0 notifier_error 0x0 gen_ctl
 0x1501000 status 0x400 next cpb count 0xB next cpb idx 0x0
 [...]
 ata1: timeout waiting for ADMA IDLE, stat=0x400
 ata1: timeout waiting for ADMA LEGACY, stat=0x400
 ata1.00: exception Emask 0x0 SAct 0x3fff SErr 0x200000 action 0x2
 frozen
 [...]
 ata1: soft resetting port

http://lkml.org/lkml/2007/3/16/184

Signed-off-by: Max Kellermann <max@duempel.org>

 drivers/ata/libata-core.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 0abd72d..39a4c05 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -3374,6 +3374,7 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
 	{ "Maxtor 6L250S0",     "BANC1G10",     ATA_HORKAGE_NONCQ },
 	/* NCQ hard hangs device under heavier load, needs hard power cycle */
 	{ "Maxtor 6B250S0",	"BANC1B70",	ATA_HORKAGE_NONCQ },
+        { "SAMSUNG HD401LJ",	NULL,		ATA_HORKAGE_NONCQ, },
 	/* Blacklist entries taken from Silicon Image 3124/3132
 	   Windows driver .inf file - also several Linux problem reports */
 	{ "HTS541060G9SA00",    "MB3OC60D",     ATA_HORKAGE_NONCQ, },

Re: [PATCH] libata: add Samsung HD401LJ to the NCQ blacklist

[Tejun Heo]

Max Kellermann wrote:
> Yet another hard drive which doesn't seem to get NCQ right.
> 
>  ata1: EH in ADMA mode, notifier 0x0 notifier_error 0x0 gen_ctl
>  0x1501000 status 0x400 next cpb count 0xB next cpb idx 0x0
>  [...]
>  ata1: timeout waiting for ADMA IDLE, stat=0x400
>  ata1: timeout waiting for ADMA LEGACY, stat=0x400
>  ata1.00: exception Emask 0x0 SAct 0x3fff SErr 0x200000 action 0x2
>  frozen
>  [...]
>  ata1: soft resetting port
> 
> http://lkml.org/lkml/2007/3/16/184
> 
> Signed-off-by: Max Kellermann <max@duempel.org>
> 
>  drivers/ata/libata-core.c |    1 +
>  1 files changed, 1 insertions(+), 0 deletions(-)
> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
> index 0abd72d..39a4c05 100644
> --- a/drivers/ata/libata-core.c
> +++ b/drivers/ata/libata-core.c
> @@ -3374,6 +3374,7 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
>  	{ "Maxtor 6L250S0",     "BANC1G10",     ATA_HORKAGE_NONCQ },
>  	/* NCQ hard hangs device under heavier load, needs hard power cycle */
>  	{ "Maxtor 6B250S0",	"BANC1B70",	ATA_HORKAGE_NONCQ },
> +        { "SAMSUNG HD401LJ",	NULL,		ATA_HORKAGE_NONCQ, },
>  	/* Blacklist entries taken from Silicon Image 3124/3132
>  	   Windows driver .inf file - also several Linux problem reports */
>  	{ "HTS541060G9SA00",    "MB3OC60D",     ATA_HORKAGE_NONCQ, },

Can you try it on another controller?  Say, a ahci or sil24?

-- 
tejun

Re: [PATCH] libata: add Samsung HD401LJ to the NCQ blacklist

[Max Kellermann]

On 2007/04/18 08:26, Tejun Heo <htejun@gmail.com> wrote:
> Can you try it on another controller?  Say, a ahci or sil24?

Seems to work ok on a Via VT8251 controller (AHCI) on kernel
2.6.21-rc6-git4 without my NCQ disabling patch.

By the way, on this machine, I had a lot of SATA timeouts during
kernel boot, which were solved by appending "nomsi".  Do you believe
there might be a relation to the NCQ problems on my first (Nforce)
machine?

Max

Re: [PATCH] libata: add Samsung HD401LJ to the NCQ blacklist

[Tejun Heo]

Max Kellermann wrote:
> On 2007/04/18 08:26, Tejun Heo <htejun@gmail.com> wrote:
>> Can you try it on another controller?  Say, a ahci or sil24?
> 
> Seems to work ok on a Via VT8251 controller (AHCI) on kernel
> 2.6.21-rc6-git4 without my NCQ disabling patch.

Unfortunately, NCQ is not supported on VT8251, so it doesn't prove
anything.  :-(

> By the way, on this machine, I had a lot of SATA timeouts during
> kernel boot, which were solved by appending "nomsi".  Do you believe
> there might be a relation to the NCQ problems on my first (Nforce)
> machine?

It's more likely your chipset just has busted MSI support.  Please post
the result of 'lspci -tv' and 'lspci -nn'.

Thanks.

-- 
tejun