Prev: x86: Use EOI register in io-apic on intel platforms
Next: sysfs: Update s_iattr on link and unlink.
From: Andre Tomt on 3 Nov 2009 20:10
Andre Tomt wrote:
> Jeff Garzik wrote:
>> On 09/09/2009 12:30 PM, Thomas Fjellstrom wrote:
>>> No errors on that disk. Other than the one above, and its more of a
>>> warning.
>>> However, I just rebooted to add some extra drives, thinking
>>> everything was
>>> working a little better now that I've updated to 2.6.31-rc9, I'm
>>> treated to
>>> the following two messages right after boot (and a system lockup to
>>> boot):
>
> Jeff & Co,
>
> Just chiming in with a "me too" comment.
>
> Identical issues with the AOC-SASLP-MV8 card with 8 Seagate Barracuda
> 7200.11 1.5TB drives using the mvsas driver. Creating a md raid array
> seems to trigger it nearly instantly, however other loads do not trigger
> it as fast (if at all, only tested briefly with paralell dd).
>
> Not sure if its related in any way, but I also noticed that with
> parallell dd read on all ports, the card/driver tops out at 600MB/s even
> though the card is in a northbridge connected slot running in x4 mode.
> Only the first 4-5 ports run at nearly full speed, while the rest barely
> gets to read any data at all. To illustrate, it looks sort of like this
> in iostat -m:
> sda 122
> sdb 122
> sbc 120
> sdd 121
> sde 100
> sdf 10
> sdg 5
> sdh 5
>
> Starting to think this driver has some issues with concurrency.. The
> other controllers have no issues saturating all their ports, even if
> they are on a south bridge connected slot.

Err, saturating all the disks. I'm not expecting 300MB/s*8, even though
it would be very cool 8)

>
> Same disks/setup works just fine on AHCI and sata_mv controllers on the
> same computer, easily pushes 8-900MB/s with md software raid5/6 (Yipes!)
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Andre Tomt on 3 Nov 2009 20:30
Jeff Garzik wrote:
> On 09/09/2009 12:30 PM, Thomas Fjellstrom wrote:
>> No errors on that disk. Other than the one above, and its more of a
>> warning.
>> However, I just rebooted to add some extra drives, thinking everything
>> was
>> working a little better now that I've updated to 2.6.31-rc9, I'm
>> treated to
>> the following two messages right after boot (and a system lockup to
>> boot):

Jeff & Co,

Just chiming in with a "me too" comment.

Identical issues with the AOC-SASLP-MV8 card with 8 Seagate Barracuda
7200.11 1.5TB drives using the mvsas driver. Creating a md raid array
seems to trigger it nearly instantly, however other loads do not trigger
it as fast (if at all, only tested briefly with paralell dd).

Not sure if its related in any way, but I also noticed that with
parallell dd read on all ports, the card/driver tops out at 600MB/s even
though the card is in a northbridge connected slot running in x4 mode.
Only the first 4-5 ports run at nearly full speed, while the rest barely
gets to read any data at all. To illustrate, it looks sort of like this
in iostat -m:
sda 122
sdb 122
sbc 120
sdd 121
sde 100
sdf 10
sdg 5
sdh 5

Starting to think this driver has some issues with concurrency.. The
other controllers have no issues saturating all their ports, even if
they are on a south bridge connected slot.

Same disks/setup works just fine on AHCI and sata_mv controllers on the
same computer, easily pushes 8-900MB/s with md software raid5/6 (Yipes!)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Mikael Abrahamsson on 3 Dec 2009 16:20
On Wed, 4 Nov 2009, Andre Tomt wrote:

> Identical issues with the AOC-SASLP-MV8 card with 8 Seagate Barracuda
> 7200.11 1.5TB drives using the mvsas driver. Creating a md raid array
> seems to trigger it nearly instantly, however other loads do not trigger
> it as fast (if at all, only tested briefly with paralell dd).

Would just like to chime in with a "me too" on this. I've tried both
ubuntu 9.10 kernel (2.6.31) and their 10.04 pre-alpha (guess it's
2.6.32-rc8) and mvsas + AOC-SASLP-MV8 is pretty much unusable. If I start
md creation it stalls immediately and gets nowhere, last time I tried it
md claimed it had written 160 blocks then it stalled (all drive LEDs on
the controller is constantly lit).

If I boot without drives and then hot-plug the drives, it immediately
oops:es.

Any code/fixes I can try, this is a lab machine so I can easily do
whatever needed with it...

--
Mikael Abrahamsson email: swmike(a)swm.pp.se
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Kristleifur Daðason on 3 Dec 2009 16:40
On Thu, Dec 3, 2009 at 9:17 PM, Mikael Abrahamsson <swmike(a)swm.pp.se> wrote:
>
> On Wed, 4 Nov 2009, Andre Tomt wrote:
>
>> Identical issues with the AOC-SASLP-MV8 card with 8 Seagate Barracuda
>> 7200.11 1.5TB drives using the mvsas driver. Creating a md raid array seems
>> to trigger it nearly instantly, however other loads do not trigger it
>> as fast (if at all, only tested briefly with paralell dd).
>
> Would just like to chime in with a "me too" on this. I've tried both ubuntu
> 9.10 kernel (2.6.31) and their 10.04 pre-alpha (guess it's 2.6.32-rc8) and
> mvsas + AOC-SASLP-MV8 is pretty much unusable. If I start md creation it
> stalls immediately and gets nowhere, last time I tried it md claimed it had
> written 160 blocks then it stalled (all drive LEDs on the controller is
> constantly lit).
>
> If I boot without drives and then hot-plug the drives, it immediately
> oops:es.
>
> Any code/fixes I can try, this is a lab machine so I can easily do whatever
> needed with it...
>

Hi,

I had a bit of trouble finding some patches that were released by Andy Yan at
Marvell on November 9th, so here is a link:

1/7: http://kerneltrap.org/mailarchive/linux-scsi/2009/11/9/6558753
2/7: http://kerneltrap.org/mailarchive/linux-scsi/2009/11/9/6558763
3/7: http://kerneltrap.org/mailarchive/linux-scsi/2009/11/9/6558773
4/7: http://kerneltrap.org/mailarchive/linux-scsi/2009/11/9/6558783
5/7: http://kerneltrap.org/mailarchive/linux-scsi/2009/11/9/6558793
6/7: http://kerneltrap.org/mailarchive/linux-scsi/2009/11/9/6558803
7/7: http://kerneltrap.org/mailarchive/linux-scsi/2009/11/9/6558813

I haven't tried these myself, but I believe that mr. Fjellstrom tried, with
success.

Note to the lists - the patches weren't properly construction for inclusion in
the kernel. If the outstanding issues could be fixed, this would be a boon to
the open-soure community as these cards are very good value for money. They
have been recommended for use on OpenSolaris by one of Stanford's ZFS
admins, which indicates their quality to be very much acceptable.

-- Kristleifur
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Mikael Abrahamsson on 4 Dec 2009 14:10
On Thu, 3 Dec 2009, Kristleifur Da�ason wrote:

> On Thu, Dec 3, 2009 at 9:17 PM, Mikael Abrahamsson <swmike(a)swm.pp.se> wrote:
>>
>> On Wed, 4 Nov 2009, Andre Tomt wrote:
>>
>>> Identical issues with the AOC-SASLP-MV8 card with 8 Seagate Barracuda
>>> 7200.11 1.5TB drives using the mvsas driver. Creating a md raid array seems
>>> to trigger it nearly instantly, however other loads do not trigger it
>>> as fast (if at all, only tested briefly with paralell dd).
>>
>> Would just like to chime in with a "me too" on this. I've tried both ubuntu
>> 9.10 kernel (2.6.31) and their 10.04 pre-alpha (guess it's 2.6.32-rc8) and
>> mvsas + AOC-SASLP-MV8 is pretty much unusable. If I start md creation it
>> stalls immediately and gets nowhere, last time I tried it md claimed it had
>> written 160 blocks then it stalled (all drive LEDs on the controller is
>> constantly lit).
>>
>> If I boot without drives and then hot-plug the drives, it immediately
>> oops:es.
>>
>> Any code/fixes I can try, this is a lab machine so I can easily do whatever
>> needed with it...
>>
>
> Hi,
>
> I had a bit of trouble finding some patches that were released by Andy Yan at
> Marvell on November 9th, so here is a link:
>
> 1/7: http://kerneltrap.org/mailarchive/linux-scsi/2009/11/9/6558753

They applied cleanly to the ubuntu 10.04 2.6.32 kernel source package but
when compiling I received an error about "sas_change_queue_depth" having
wrong amount of arguments:

- sas_change_queue_depth(sdev, MVS_QUEUE_SIZE);
+ sas_change_queue_depth(sdev, MVS_QUEUE_SIZE, SCSI_QDEPTH_DEFAULT);

So I removed that change and then things compiled, booted properly, but
just froze when I plugged in the disks (including some garbled graphics on
my X screen). That patch set doesn't seem to be ok.

--
Mikael Abrahamsson email: swmike(a)swm.pp.se
 |  Next  |  Last
Pages: 1 2
Prev: x86: Use EOI register in io-apic on intel platforms
Next: sysfs: Update s_iattr on link and unlink.