[PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Robert Hancock]

This patch adds in some NCQ blacklist entries taken from the Silicon
Image Windows drivers' .inf files for the 3124 and 3132 controllers.
These entries were marked as ""DisableSataQueueing". Assume these are
in their blacklist for a reason and disable NCQ on these drives.

Signed-off-by: Robert Hancock <hancockr@shaw.ca>

--- linux-2.6.21-rc1edit/drivers/ata/libata-core.c.prev	2007-02-21 22:23:05.000000000 -0600
+++ linux-2.6.21-rc1edit/drivers/ata/libata-core.c	2007-02-21 22:25:44.000000000 -0600
@@ -3269,6 +3269,13 @@ static const struct ata_blacklist_entry 
 
 	/* Devices with NCQ limits */
 
+	/* The following blacklist entries are taken from the Windows
+	   driver .inf files for the Silicon Image 3124 and 3132. */
+	{ "Maxtor 7B250S0",	"BANC1B70",	ATA_HORKAGE_NONCQ, },
+	{ "HTS541060G9SA00",	"MB3OC60D",	ATA_HORKAGE_NONCQ, },
+	{ "HTS541080G9SA00",	"MB4OC60D",	ATA_HORKAGE_NONCQ, },
+	{ "HTS541010G9SA00",	"MBZOC60D",	ATA_HORKAGE_NONCQ, },
+
 	/* End Marker */
 	{ }
 };

Re: [PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Jeff Garzik]

Robert Hancock wrote:
> This patch adds in some NCQ blacklist entries taken from the Silicon
> Image Windows drivers' .inf files for the 3124 and 3132 controllers.
> These entries were marked as ""DisableSataQueueing". Assume these are
> in their blacklist for a reason and disable NCQ on these drives.
> 
> Signed-off-by: Robert Hancock <hancockr@shaw.ca>
> 
> --- linux-2.6.21-rc1edit/drivers/ata/libata-core.c.prev    2007-02-21 
> 22:23:05.000000000 -0600
> +++ linux-2.6.21-rc1edit/drivers/ata/libata-core.c    2007-02-21 
> 22:25:44.000000000 -0600
> @@ -3269,6 +3269,13 @@ static const struct ata_blacklist_entry
>     /* Devices with NCQ limits */
> 
> +    /* The following blacklist entries are taken from the Windows
> +       driver .inf files for the Silicon Image 3124 and 3132. */
> +    { "Maxtor 7B250S0",    "BANC1B70",    ATA_HORKAGE_NONCQ, },
> +    { "HTS541060G9SA00",    "MB3OC60D",    ATA_HORKAGE_NONCQ, },
> +    { "HTS541080G9SA00",    "MB4OC60D",    ATA_HORKAGE_NONCQ, },
> +    { "HTS541010G9SA00",    "MBZOC60D",    ATA_HORKAGE_NONCQ, },

Do we have information that these drives fail on non-SiI controllers?

Sometimes the problem can be related to a single family of controllers.

	Jeff

Re: [PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Jens Axboe]

On Thu, Feb 22 2007, Jeff Garzik wrote:
> Robert Hancock wrote:
> >This patch adds in some NCQ blacklist entries taken from the Silicon
> >Image Windows drivers' .inf files for the 3124 and 3132 controllers.
> >These entries were marked as ""DisableSataQueueing". Assume these are
> >in their blacklist for a reason and disable NCQ on these drives.
> >
> >Signed-off-by: Robert Hancock <hancockr@shaw.ca>
> >
> >--- linux-2.6.21-rc1edit/drivers/ata/libata-core.c.prev    2007-02-21 
> >22:23:05.000000000 -0600
> >+++ linux-2.6.21-rc1edit/drivers/ata/libata-core.c    2007-02-21 
> >22:25:44.000000000 -0600
> >@@ -3269,6 +3269,13 @@ static const struct ata_blacklist_entry
> >    /* Devices with NCQ limits */
> >
> >+    /* The following blacklist entries are taken from the Windows
> >+       driver .inf files for the Silicon Image 3124 and 3132. */
> >+    { "Maxtor 7B250S0",    "BANC1B70",    ATA_HORKAGE_NONCQ, },
> >+    { "HTS541060G9SA00",    "MB3OC60D",    ATA_HORKAGE_NONCQ, },
> >+    { "HTS541080G9SA00",    "MB4OC60D",    ATA_HORKAGE_NONCQ, },
> >+    { "HTS541010G9SA00",    "MBZOC60D",    ATA_HORKAGE_NONCQ, },
> 
> Do we have information that these drives fail on non-SiI controllers?
> 
> Sometimes the problem can be related to a single family of controllers.

I don't know about the Hitachi's, but the Maxtor with that firmware is
definitely broken. It _appeared_ to work if the depth was limited to 4,
but I didn't test it long enough to be absolutely certain. So the safest
is indeed to blacklist it.

I'm pretty sure my initial NCQ patches had a blacklist entry for that
drive as well, but apparently the blacklist got lost somewhere along the
way?

-- 
Jens Axboe

Re: [PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Mark Lord]

Jens Axboe wrote:
>> Robert Hancock wrote:
>..
>>> +    /* The following blacklist entries are taken from the Windows
>>> +       driver .inf files for the Silicon Image 3124 and 3132. */
>>> +    { "Maxtor 7B250S0",    "BANC1B70",    ATA_HORKAGE_NONCQ, },
>>> +    { "HTS541060G9SA00",    "MB3OC60D",    ATA_HORKAGE_NONCQ, },
>>> +    { "HTS541080G9SA00",    "MB4OC60D",    ATA_HORKAGE_NONCQ, },
>>> +    { "HTS541010G9SA00",    "MBZOC60D",    ATA_HORKAGE_NONCQ, },
>> Do we have information that these drives fail on non-SiI controllers?
>>
>> Sometimes the problem can be related to a single family of controllers.
> 
> I don't know about the Hitachi's, but the Maxtor with that firmware is
> definitely broken. It _appeared_ to work if the depth was limited to 4,
> but I didn't test it long enough to be absolutely certain. So the safest
> is indeed to blacklist it.

Yes, broken on the Silicon Image controllers for sure.
But what type of controller did you observe the failures on, Jens ?

Re: [PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Jens Axboe]

On Thu, Feb 22 2007, Mark Lord wrote:
> Jens Axboe wrote:
> >>Robert Hancock wrote:
> >..
> >>>+    /* The following blacklist entries are taken from the Windows
> >>>+       driver .inf files for the Silicon Image 3124 and 3132. */
> >>>+    { "Maxtor 7B250S0",    "BANC1B70",    ATA_HORKAGE_NONCQ, },
> >>>+    { "HTS541060G9SA00",    "MB3OC60D",    ATA_HORKAGE_NONCQ, },
> >>>+    { "HTS541080G9SA00",    "MB4OC60D",    ATA_HORKAGE_NONCQ, },
> >>>+    { "HTS541010G9SA00",    "MBZOC60D",    ATA_HORKAGE_NONCQ, },
> >>Do we have information that these drives fail on non-SiI controllers?
> >>
> >>Sometimes the problem can be related to a single family of controllers.
> >
> >I don't know about the Hitachi's, but the Maxtor with that firmware is
> >definitely broken. It _appeared_ to work if the depth was limited to 4,
> >but I didn't test it long enough to be absolutely certain. So the safest
> >is indeed to blacklist it.
> 
> Yes, broken on the Silicon Image controllers for sure.
> But what type of controller did you observe the failures on, Jens ?

achi

-- 
Jens Axboe

Re: [PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Ask Bjørn Hansen]

On Feb 21, 2007, at 10:57 PM, Jeff Garzik wrote:

>> +    /* The following blacklist entries are taken from the Windows
>> +       driver .inf files for the Silicon Image 3124 and 3132. */
>> +    { "Maxtor 7B250S0",    "BANC1B70",    ATA_HORKAGE_NONCQ, },
[...]
>
> Do we have information that these drives fail on non-SiI controllers?

At least tangentially related:

On one of my boxes (running 2.6.18-1.2869 from Fedora) I have a  
couple of other Maxtor drives that didn't like NCQ.   They are on a  
JMicron 20360/20363 (ahci driver).  (There's also a Promise 300 TX4  
card in the box and an Intel ICH8 that shows up with ata_piix).

model and (partial) firmware revision of the drives:
Maxtor 7V300F0  VA11
Maxtor 7B300S0  BANC

Until I disabled NCQ I got gazillions of messages like the ones below  
and absymal performance.

   - ask


ata5: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive 0xf)
ata6: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive  
0x33)
ata5: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive 0x0)
ata6: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive 0x0)
ata5: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive 0x0)
ata6: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive 0x0)
ata6: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive  
0x38)
ata5: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive  
0x3c)
ata5: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive  
0xf2)
ata5: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive  
0x1c)



-- 
http://develooper.com/ - http://askask.com/

Re: [PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Robert Hancock]

Ask Bjørn Hansen wrote:
> 
> On Feb 21, 2007, at 10:57 PM, Jeff Garzik wrote:
> 
>>> +    /* The following blacklist entries are taken from the Windows
>>> +       driver .inf files for the Silicon Image 3124 and 3132. */
>>> +    { "Maxtor 7B250S0",    "BANC1B70",    ATA_HORKAGE_NONCQ, },
> [...]
>>
>> Do we have information that these drives fail on non-SiI controllers?
> 
> At least tangentially related:
> 
> On one of my boxes (running 2.6.18-1.2869 from Fedora) I have a couple 
> of other Maxtor drives that didn't like NCQ.   They are on a JMicron 
> 20360/20363 (ahci driver).  (There's also a Promise 300 TX4 card in the 
> box and an Intel ICH8 that shows up with ata_piix).
> 
> model and (partial) firmware revision of the drives:
> Maxtor 7V300F0  VA11
> Maxtor 7B300S0  BANC
> 
> Until I disabled NCQ I got gazillions of messages like the ones below 
> and absymal performance.
> 
>   - ask
> 
> 
> ata5: spurious interrupt (irq_stat 0x8 active_tag -84148995 sactive 0xf)

Sounds like those are some that we should be blacklisting as well, 
unless Eric has a good reason why not (CCing). Can you provide the full 
firmware revision strings from those drives, i.e. from "hdparm -I"?

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

Re: [PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Ask Bjørn Hansen]

On Feb 22, 2007, at 4:18 PM, Robert Hancock wrote:

>> model and (partial) firmware revision of the drives:
>> Maxtor 7V300F0  VA11
>> Maxtor 7B300S0  BANC
>> Until I disabled NCQ I got gazillions of messages like the ones  
>> below and absymal performance.
>>   - ask
>> ata5: spurious interrupt (irq_stat 0x8 active_tag -84148995  
>> sactive 0xf)
>
> Sounds like those are some that we should be blacklisting as well,  
> unless Eric has a good reason why not (CCing). Can you provide the  
> full firmware revision strings from those drives, i.e. from "hdparm  
> -I"?

ATA device, with non-removable media
         Model Number:       Maxtor 7B300S0
         Firmware Revision:  BANC1B70

ATA device, with non-removable media
         Model Number:       Maxtor 7V300F0
         Firmware Revision:  VA111630


  - ask

-- 
http://develooper.com/ - http://askask.com/

Re: [PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Jeff Garzik]

Robert Hancock wrote:
> This patch adds in some NCQ blacklist entries taken from the Silicon
> Image Windows drivers' .inf files for the 3124 and 3132 controllers.
> These entries were marked as ""DisableSataQueueing". Assume these are
> in their blacklist for a reason and disable NCQ on these drives.
> 
> Signed-off-by: Robert Hancock <hancockr@shaw.ca>
> 
> --- linux-2.6.21-rc1edit/drivers/ata/libata-core.c.prev    2007-02-21 
> 22:23:05.000000000 -0600
> +++ linux-2.6.21-rc1edit/drivers/ata/libata-core.c    2007-02-21 
> 22:25:44.000000000 -0600
> @@ -3269,6 +3269,13 @@ static const struct ata_blacklist_entry
>     /* Devices with NCQ limits */
> 
> +    /* The following blacklist entries are taken from the Windows
> +       driver .inf files for the Silicon Image 3124 and 3132. */
> +    { "Maxtor 7B250S0",    "BANC1B70",    ATA_HORKAGE_NONCQ, },
> +    { "HTS541060G9SA00",    "MB3OC60D",    ATA_HORKAGE_NONCQ, },
> +    { "HTS541080G9SA00",    "MB4OC60D",    ATA_HORKAGE_NONCQ, },
> +    { "HTS541010G9SA00",    "MBZOC60D",    ATA_HORKAGE_NONCQ, },


The general consensus of private emails seems to be that its ok to 
blacklist the Maxtor one, at least.

So let's go ahead and at least get that entry in.

	Jeff

Re: [PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Tejun Heo]

Jeff Garzik wrote:
> Robert Hancock wrote:
>> This patch adds in some NCQ blacklist entries taken from the Silicon
>> Image Windows drivers' .inf files for the 3124 and 3132 controllers.
>> These entries were marked as ""DisableSataQueueing". Assume these are
>> in their blacklist for a reason and disable NCQ on these drives.
>>
>> Signed-off-by: Robert Hancock <hancockr@shaw.ca>
>>
>> --- linux-2.6.21-rc1edit/drivers/ata/libata-core.c.prev    2007-02-21
>> 22:23:05.000000000 -0600
>> +++ linux-2.6.21-rc1edit/drivers/ata/libata-core.c    2007-02-21
>> 22:25:44.000000000 -0600
>> @@ -3269,6 +3269,13 @@ static const struct ata_blacklist_entry
>>     /* Devices with NCQ limits */
>>
>> +    /* The following blacklist entries are taken from the Windows
>> +       driver .inf files for the Silicon Image 3124 and 3132. */
>> +    { "Maxtor 7B250S0",    "BANC1B70",    ATA_HORKAGE_NONCQ, },
>> +    { "HTS541060G9SA00",    "MB3OC60D",    ATA_HORKAGE_NONCQ, },
>> +    { "HTS541080G9SA00",    "MB4OC60D",    ATA_HORKAGE_NONCQ, },
>> +    { "HTS541010G9SA00",    "MBZOC60D",    ATA_HORKAGE_NONCQ, },

There was a recent regression report on ahci + HTS541010G9SA00 from
Mathieu Berard (cc'd).  NCQ + device initiated link power saving (which,
BTW, got activated by executing _GTF.  surprise!) combination seems
broken.  I know that ahci controllers (at least ich6m, ich7 and ich8)
work with device initiated PS, so it's most likely the disk that's
malfunctioning.

This and the fact sata_sil24 windows driver blacklists them make me
think those HTS5410s should be added to our blacklist too.  What do you
think?

Mathieu, just to be sure, can you post the result of 'hdparm -I
/dev/sdX' and 'lspci -nn'?

Thanks.

-- 
tejun

Re: [PATCH] libata: add NCQ blacklist entries from Silicon Image Windows driver

[Mathieu Bérard]

Tejun Heo a écrit :
> Jeff Garzik wrote:
>
> Mathieu, just to be sure, can you post the result of 'hdparm -I
> /dev/sdX' and 'lspci -nn'?
>
Hi,
no problem :

lspci -nn:
00:00.0 Host bridge [0600]: Intel Corporation Mobile 915GM/PM/GMS/910GML
Express Processor to DRAM Controller [8086:2590] (rev 03)
00:01.0 PCI bridge [0604]: Intel Corporation Mobile 915GM/PM Express PCI
Express Root Port [8086:2591] (rev 03)
00:1c.0 PCI bridge [0604]: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6
Family) PCI Express Port 1 [8086:2660] (rev 03)
00:1c.1 PCI bridge [0604]: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6
Family) PCI Express Port 2 [8086:2662] (rev 03)
00:1d.0 USB Controller [0c03]: Intel Corporation 82801FB/FBM/FR/FW/FRW
(ICH6 Family) USB UHCI #1 [8086:2658] (rev 03)
00:1d.1 USB Controller [0c03]: Intel Corporation 82801FB/FBM/FR/FW/FRW
(ICH6 Family) USB UHCI #2 [8086:2659] (rev 03)
00:1d.2 USB Controller [0c03]: Intel Corporation 82801FB/FBM/FR/FW/FRW
(ICH6 Family) USB UHCI #3 [8086:265a] (rev 03)
00:1d.3 USB Controller [0c03]: Intel Corporation 82801FB/FBM/FR/FW/FRW
(ICH6 Family) USB UHCI #4 [8086:265b] (rev 03)
00:1d.7 USB Controller [0c03]: Intel Corporation 82801FB/FBM/FR/FW/FRW
(ICH6 Family) USB2 EHCI Controller [8086:265c] (rev 03)
00:1e.0 PCI bridge [0604]: Intel Corporation 82801 Mobile PCI Bridge
[8086:2448] (rev d3)
00:1e.2 Multimedia audio controller [0401]: Intel Corporation
82801FB/FBM/FR/FW/FRW (ICH6 Family) AC'97 Audio Controller [8086:266e]
(rev 03)
00:1e.3 Modem [0703]: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6
Family) AC'97 Modem Controller [8086:266d] (rev 03)
00:1f.0 ISA bridge [0601]: Intel Corporation 82801FBM (ICH6M) LPC
Interface Bridge [8086:2641] (rev 03)
00:1f.1 IDE interface [0101]: Intel Corporation 82801FB/FBM/FR/FW/FRW
(ICH6 Family) IDE Controller [8086:266f] (rev 03)
00:1f.2 IDE interface [0101]: Intel Corporation 82801FBM (ICH6M) SATA
Controller [8086:2653] (rev 03)
00:1f.3 SMBus [0c05]: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6
Family) SMBus Controller [8086:266a] (rev 03)
01:00.0 VGA compatible controller [0300]: ATI Technologies Inc M24
[Radeon Mobility X600] [1002:5462]
06:01.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd.
RTL-8139/8139C/8139C+ [10ec:8139] (rev 10)
06:02.0 Network controller [0280]: Intel Corporation PRO/Wireless 2200BG
Network Connection [8086:4220] (rev 05)
06:04.0 CardBus bridge [0607]: Texas Instruments PCIxx21/x515 Cardbus
Controller [104c:8031]
06:04.2 FireWire (IEEE 1394) [0c00]: Texas Instruments OHCI Compliant
IEEE 1394 Host Controller [104c:8032]
06:04.3 Mass storage controller [0180]: Texas Instruments PCIxx21
Integrated FlashMedia Controller [104c:8033]
06:04.4 Generic system peripheral [0805]: Texas Instruments
PCI6411/6421/6611/6621/7411/7421/7611/7621 Secure Digital Controller
[104c:8034]

hdparm -I /dev/sda:

/dev/sda:

ATA device, with non-removable media
        Model Number:       HTS541010G9SA00                        
        Serial Number:      MP2ZM4X0GSWJKR
        Firmware Revision:  MBZOC60D
Standards:
        Used: ATA/ATAPI-7 T13 1532D revision 1
        Supported: 7 6 5 4
Configuration:
        Logical         max     current
        cylinders       16383   16383
        heads           16      16
        sectors/track   63      63
        --
        CHS current addressable sectors:   16514064
        LBA    user addressable sectors:  195371568
        LBA48  user addressable sectors:  195371568
        device size with M = 1024*1024:       95396 MBytes
        device size with M = 1000*1000:      100030 MBytes (100 GB)
Capabilities:
        LBA, IORDY(can be disabled)
        Queue depth: 32
        Standby timer values: spec'd by Vendor, no device specific minimum
        R/W multiple sector transfer: Max = 16  Current = 0
        Advanced power management level: 128 (0x80)
        Recommended acoustic management value: 128, current value: 254
        DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
             Cycle time: min=120ns recommended=120ns
        PIO: pio0 pio1 pio2 pio3 pio4
             Cycle time: no flow control=240ns  IORDY flow control=120ns
Commands/features:
        Enabled Supported:
           *    SMART feature set
                Security Mode feature set
           *    Power Management feature set
           *    Write cache
           *    Look-ahead
           *    Host Protected Area feature set
           *    WRITE_BUFFER command
           *    READ_BUFFER command
           *    NOP cmd
           *    DOWNLOAD_MICROCODE
           *    Advanced Power Management feature set
                Power-Up In Standby feature set
           *    SET_FEATURES required to spinup after power up
                SET_MAX security extension
                Automatic Acoustic Management feature set
           *    48-bit Address feature set
           *    Device Configuration Overlay feature set
           *    Mandatory FLUSH_CACHE
           *    FLUSH_CACHE_EXT
           *    SMART error logging
           *    SMART self-test
           *    General Purpose Logging feature set
           *    WRITE_{DMA|MULTIPLE}_FUA_EXT
           *    IDLE_IMMEDIATE with UNLOAD
           *    SATA-I signaling speed (1.5Gb/s)
           *    Native Command Queueing (NCQ)
           *    Host-initiated interface power management
                Non-Zero buffer offsets in DMA Setup FIS
                DMA Setup Auto-Activate optimization
                Device-initiated interface power management
                In-order data delivery
           *    Software settings preservation
Security:
        Master password revision code = 65534
                supported
        not     enabled
        not     locked
        not     frozen
        not     expired: security count
        not     supported: enhanced erase
        64min for SECURITY ERASE UNIT.
Checksum: correct

-- 
Mathieu