From: Andrew Morton <akpm@linux-foundation.org>
To: daumas <daumas77@gmail.com>
Cc: linux-kernel@vger.kernel.org, linux-ide@vger.kernel.org
Subject: Re: Problem with SI3124-based controller and SI 4726-based PMP in RAID10 mode
Date: Wed, 10 Dec 2008 13:49:38 -0800 [thread overview]
Message-ID: <20081210134938.2dbe6d21.akpm@linux-foundation.org> (raw)
In-Reply-To: <49401F04.90408@gmail.com>
(cc linux-ide)
On Wed, 10 Dec 2008 13:56:52 -0600
daumas <daumas77@gmail.com> wrote:
>
> I have a sata_sil24 problem involving a 4-port SI3124-based controller
> connected to a 3-port eSATA enclosure. Each eSATA port is an SI
> 4726-based H/W PMP. The PMPs are connected to (13) 1 TB hard drives, as
> indicated below.
>
> PCI3124 Port 1
> 4726 (Configured in RAID10 mode)
> 4726 Port 1 => 1 TB drive
> 4726 Port 2 => 1 TB drive
> 4726 Port 3 => 1 TB drive
> 4726 Port 4 => 1 TB drive
> 4726 Port 5 => 1 TB drive
> PCI3124 Port 2
> 4726 (Configured in RAID10 mode)
> 4726 Port 1 => 1 TB drive
> 4726 Port 2 => 1 TB drive
> 4726 Port 3 => 1 TB drive
> 4726 Port 4 => 1 TB drive
> 4726 Port 5 => 1 TB drive
> PCI3124 Port 3
> 4726 (Configured in JBOD/Passthru mode)
> 4726 Port 1 => 1 TB drive
> 4726 Port 2 => 1 TB drive
> 4726 Port 3 => 1 TB drive
> 4726 Port 4 => no connection
> 4726 Port 5 => no connection
> PCI3124 Port 4 =>no connection
>
> I am running openSuse 11.0 with a 2.6.27.7-4.1 kernel
>
> For the 4726 PMP configured in JBOD mode, all is well, and I can see all
> 3 disks.
>
> However, for the 4726 PMPs configured in RAID10 mode, I can see NONE of
> the disks.
>
> As I understand it, the 4726 is a hardware-based PMP which can handle
> RAID10 by itself and thereby offload the RAID processing. I would expect
> the 3124 controller to see the drives hidden behind the 4726 as a single
> unit, but this does not happen. Instead I get the errors listed at end,
> and the RAID10 device is not 'seen' at all. (Key error of note is "both
> IDENTIFYs aborted, assuming NODEV")
>
> JBOD is fine for host-only based RAID functionality, but I want the
> higher performance the 4726 offers for RAID10.
> Is is not possible to off-load the RAID10 function to the 4726 under linux?
>
> Any help, workarounds (beyond strict JBOD mode) are appreciated.
>
> Thanks!
>
> Dec 10 11:02:23 buildhost36 sudo: e-build : TTY=pts/1 ;
> PWD=/home/e-build ; USER=root ; COMMAND=/sbin/modprobe sata_sil24
> Dec 10 11:02:23 buildhost36 kernel: sata_sil24 0000:03:03.0: version 1.1
> Dec 10 11:02:23 buildhost36 kernel: vendor=8086 device=0329
> Dec 10 11:02:23 buildhost36 kernel: vendor=8086 device=3595
> Dec 10 11:02:23 buildhost36 kernel: sata_sil24 0000:03:03.0: PCI INT A
> -> GSI 72 (level, low) -> IRQ 64
> Dec 10 11:02:23 buildhost36 kernel: sata_sil24 0000:03:03.0: Applying
> completion IRQ loss on PCI-X errata fix
> Dec 10 11:02:23 buildhost36 kernel: scsi8 : sata_sil24
> Dec 10 11:02:23 buildhost36 kernel: scsi9 : sata_sil24
> Dec 10 11:02:23 buildhost36 kernel: scsi10 : sata_sil24
> Dec 10 11:02:23 buildhost36 kernel: scsi11 : sata_sil24
> Dec 10 11:02:23 buildhost36 kernel: ata7: SATA max UDMA/100 host
> m128@0xdefffc00 port 0xdeff0000 irq 64
> Dec 10 11:02:23 buildhost36 kernel: ata8: SATA max UDMA/100 host
> m128@0xdefffc00 port 0xdeff2000 irq 64
> Dec 10 11:02:23 buildhost36 kernel: ata9: SATA max UDMA/100 host
> m128@0xdefffc00 port 0xdeff4000 irq 64
> Dec 10 11:02:23 buildhost36 kernel: ata10: SATA max UDMA/100 host
> m128@0xdefffc00 port 0xdeff6000 irq 64
> Dec 10 11:02:25 buildhost36 kernel: ata7: SATA link up 3.0 Gbps (SStatus
> 123 SControl 0)
> Dec 10 11:02:25 buildhost36 kernel: ata7.15: Port Multiplier 1.1,
> 0x1095:0x4726 r31, 3 ports, feat 0x1/0x9
> Dec 10 11:02:25 buildhost36 kernel: ata7.00: hard resetting link
> Dec 10 11:02:25 buildhost36 kernel: ata7.00: SATA link down (SStatus 4
> SControl 0)
> Dec 10 11:02:25 buildhost36 kernel: ata7.01: hard resetting link
> Dec 10 11:02:25 buildhost36 kernel: ata7.01: SATA link up 3.0 Gbps
> (SStatus 123 SControl 0)
> Dec 10 11:02:25 buildhost36 kernel: ata7.02: hard resetting link
> Dec 10 11:02:26 buildhost36 kernel: ata7.02: SATA link up 1.5 Gbps
> (SStatus 113 SControl 320)
> Dec 10 11:02:26 buildhost36 kernel: ata7.01: unsupported device, disabling
> Dec 10 11:02:26 buildhost36 kernel: ata7.01: disabled
> Dec 10 11:02:26 buildhost36 kernel: ata7.02: both IDENTIFYs aborted,
> assuming NODEV
> Dec 10 11:02:26 buildhost36 kernel: ata7: EH complete
> Dec 10 11:02:28 buildhost36 kernel: ata8: SATA link up 3.0 Gbps (SStatus
> 123 SControl 0)
> Dec 10 11:02:28 buildhost36 kernel: ata8.15: Port Multiplier 1.1,
> 0x1095:0x4726 r31, 3 ports, feat 0x1/0x9
> Dec 10 11:02:28 buildhost36 kernel: ata8.00: hard resetting link
> Dec 10 11:02:28 buildhost36 kernel: ata8.00: SATA link down (SStatus 4
> SControl 0)
> Dec 10 11:02:28 buildhost36 kernel: ata8.01: hard resetting link
> Dec 10 11:02:29 buildhost36 kernel: ata8.01: SATA link up 3.0 Gbps
> (SStatus 123 SControl 0)
> Dec 10 11:02:29 buildhost36 kernel: ata8.02: hard resetting link
> Dec 10 11:02:29 buildhost36 kernel: ata8.02: SATA link up 1.5 Gbps
> (SStatus 113 SControl 320)
> Dec 10 11:02:29 buildhost36 kernel: ata8.01: unsupported device, disabling
> Dec 10 11:02:29 buildhost36 kernel: ata8.01: disabled
> Dec 10 11:02:29 buildhost36 kernel: ata8.02: both IDENTIFYs aborted,
> assuming NODEV
> Dec 10 11:02:29 buildhost36 kernel: ata8: EH complete
> Dec 10 11:02:31 buildhost36 kernel: ata9: SATA link up 3.0 Gbps (SStatus
> 123 SControl 0)
> Dec 10 11:02:31 buildhost36 kernel: ata9.15: Port Multiplier 1.1,
> 0x1095:0x4726 r31, 7 ports, feat 0x1/0x9
> Dec 10 11:02:31 buildhost36 kernel: ata9.00: hard resetting link
> Dec 10 11:02:32 buildhost36 kernel: ata9.00: SATA link up 1.5 Gbps
> (SStatus 113 SControl 320)
> Dec 10 11:02:32 buildhost36 kernel: ata9.01: hard resetting link
> Dec 10 11:02:32 buildhost36 kernel: ata9.01: SATA link up 1.5 Gbps
> (SStatus 113 SControl 300)
> Dec 10 11:02:32 buildhost36 kernel: ata9.02: hard resetting link
> Dec 10 11:02:32 buildhost36 kernel: ata9.02: SATA link up 1.5 Gbps
> (SStatus 113 SControl 300)
> Dec 10 11:02:32 buildhost36 kernel: ata9.03: hard resetting link
> Dec 10 11:02:33 buildhost36 kernel: ata9.03: SATA link down (SStatus 0
> SControl 320)
> Dec 10 11:02:33 buildhost36 kernel: ata9.04: hard resetting link
> Dec 10 11:02:33 buildhost36 kernel: ata9.04: SATA link down (SStatus 0
> SControl 320)
> Dec 10 11:02:33 buildhost36 kernel: ata9.05: hard resetting link
> Dec 10 11:02:33 buildhost36 kernel: ata9.05: SATA link up 3.0 Gbps
> (SStatus 123 SControl 0)
> Dec 10 11:02:33 buildhost36 kernel: ata9.06: hard resetting link
> Dec 10 11:02:34 buildhost36 kernel: ata9.06: SATA link up 1.5 Gbps
> (SStatus 113 SControl 320)
> Dec 10 11:02:34 buildhost36 kernel: ata9.00: ATA-8: ST31000340AS, SD15,
> max UDMA/133
> Dec 10 11:02:34 buildhost36 kernel: ata9.00: 1953525168 sectors, multi
> 0: LBA48 NCQ (depth 31/32)
> Dec 10 11:02:34 buildhost36 kernel: ata9.00: configured for UDMA/100
> Dec 10 11:02:34 buildhost36 kernel: ata9.01: ATA-8: ST31000340AS, SD15,
> max UDMA/133
> Dec 10 11:02:34 buildhost36 kernel: ata9.01: 1953525168 sectors, multi
> 0: LBA48 NCQ (depth 31/32)
> Dec 10 11:02:34 buildhost36 kernel: ata9.01: configured for UDMA/100
> Dec 10 11:02:34 buildhost36 kernel: ata9.02: ATA-8: ST31000340AS, SD15,
> max UDMA/133
> Dec 10 11:02:34 buildhost36 kernel: ata9.02: 1953525168 sectors, multi
> 0: LBA48 NCQ (depth 31/32)
> Dec 10 11:02:34 buildhost36 kernel: ata9.02: configured for UDMA/100
> Dec 10 11:02:34 buildhost36 kernel: ata9.05: unsupported device, disabling
> Dec 10 11:02:34 buildhost36 kernel: ata9.05: disabled
> Dec 10 11:02:34 buildhost36 kernel: ata9: EH complete
> Dec 10 11:02:36 buildhost36 kernel: ata10: SATA link down (SStatus 0
> SControl 0)
> Dec 10 11:02:36 buildhost36 kernel: scsi 10:0:0:0: Direct-Access ATA
> ST31000340AS SD15 PQ: 0 ANSI: 5
> Dec 10 11:02:36 buildhost36 kernel: sd 10:0:0:0: [sdb] 1953525168
> 512-byte hardware sectors: (1000GB/931GiB)
> Dec 10 11:02:36 buildhost36 kernel: sd 10:0:0:0: [sdb] Write Protect is off
> Dec 10 11:02:36 buildhost36 kernel: sd 10:0:0:0: [sdb] Mode Sense: 00 3a
> 00 00
> Dec 10 11:02:36 buildhost36 kernel: sd 10:0:0:0: [sdb] Write cache:
> enabled, read cache: enabled, doesn't support DPO or FUA
> Dec 10 11:02:36 buildhost36 kernel: sd 10:0:0:0: [sdb] 1953525168
> 512-byte hardware sectors: (1000GB/931GiB)
> Dec 10 11:02:36 buildhost36 kernel: sd 10:0:0:0: [sdb] Write Protect is off
> Dec 10 11:02:36 buildhost36 kernel: sd 10:0:0:0: [sdb] Mode Sense: 00 3a
> 00 00
> Dec 10 11:02:36 buildhost36 kernel: sd 10:0:0:0: [sdb] Write cache:
> enabled, read cache: enabled, doesn't support DPO or FUA
> Dec 10 11:02:36 buildhost36 kernel: sdb:
> Dec 10 11:02:36 buildhost36 kernel: sd 10:0:0:0: [sdb] Attached SCSI disk
> Dec 10 11:02:36 buildhost36 kernel: sd 10:0:0:0: Attached scsi generic
> sg2 type 0
> Dec 10 11:02:36 buildhost36 kernel: scsi 10:1:0:0: Direct-Access ATA
> ST31000340AS SD15 PQ: 0 ANSI: 5
> Dec 10 11:02:36 buildhost36 kernel: sd 10:1:0:0: [sdc] 1953525168
> 512-byte hardware sectors: (1000GB/931GiB)
> Dec 10 11:02:36 buildhost36 kernel: sd 10:1:0:0: [sdc] Write Protect is off
> Dec 10 11:02:36 buildhost36 kernel: sd 10:1:0:0: [sdc] Mode Sense: 00 3a
> 00 00
> Dec 10 11:02:36 buildhost36 kernel: sd 10:1:0:0: [sdc] Write cache:
> enabled, read cache: enabled, doesn't support DPO or FUA
> Dec 10 11:02:36 buildhost36 kernel: sd 10:1:0:0: [sdc] 1953525168
> 512-byte hardware sectors: (1000GB/931GiB)
> Dec 10 11:02:36 buildhost36 kernel: sd 10:1:0:0: [sdc] Write Protect is off
> Dec 10 11:02:36 buildhost36 kernel: sd 10:1:0:0: [sdc] Mode Sense: 00 3a
> 00 00
> Dec 10 11:02:36 buildhost36 kernel: sd 10:1:0:0: [sdc] Write cache:
> enabled, read cache: enabled, doesn't support DPO or FUA
> Dec 10 11:02:36 buildhost36 kernel: sdc:
> Dec 10 11:02:36 buildhost36 kernel: sd 10:1:0:0: [sdc] Attached SCSI disk
> Dec 10 11:02:36 buildhost36 kernel: sd 10:1:0:0: Attached scsi generic
> sg3 type 0
> Dec 10 11:02:36 buildhost36 kernel: scsi 10:2:0:0: Direct-Access ATA
> ST31000340AS SD15 PQ: 0 ANSI: 5
> Dec 10 11:02:36 buildhost36 kernel: sd 10:2:0:0: [sdd] 1953525168
> 512-byte hardware sectors: (1000GB/931GiB)
> Dec 10 11:02:36 buildhost36 kernel: sd 10:2:0:0: [sdd] Write Protect is off
> Dec 10 11:02:36 buildhost36 kernel: sd 10:2:0:0: [sdd] Mode Sense: 00 3a
> 00 00
> Dec 10 11:02:36 buildhost36 kernel: sd 10:2:0:0: [sdd] Write cache:
> enabled, read cache: enabled, doesn't support DPO or FUA
> Dec 10 11:02:36 buildhost36 kernel: sd 10:2:0:0: [sdd] 1953525168
> 512-byte hardware sectors: (1000GB/931GiB)
> Dec 10 11:02:36 buildhost36 kernel: sd 10:2:0:0: [sdd] Write Protect is off
> Dec 10 11:02:36 buildhost36 kernel: sd 10:2:0:0: [sdd] Mode Sense: 00 3a
> 00 00
> Dec 10 11:02:36 buildhost36 kernel: sd 10:2:0:0: [sdd] Write cache:
> enabled, read cache: enabled, doesn't support DPO or FUA
> Dec 10 11:02:36 buildhost36 kernel: sdd: sdd1
> Dec 10 11:02:36 buildhost36 kernel: sd 10:2:0:0: [sdd] Attached SCSI disk
> Dec 10 11:02:36 buildhost36 kernel: sd 10:2:0:0: Attached scsi generic
> sg4 type 0
> Dec 10 11:02:55 buildhost36 sshd[6534]: Accepted
> keyboard-interactive/pam for e-build from 9.65.98.119 port 34222 ssh2
> Dec 10 11:03:57 buildhost36 sshd[6566]: Accepted
> keyboard-interactive/pam for root from 9.65.98.119 port 44494 ssh2
>
> Henri Daumas
> Global TIM Competency - Integrated Operations
> Configuration Management/Build
>
>
prev parent reply other threads:[~2008-12-10 21:49 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-12-10 19:56 Problem with SI3124-based controller and SI 4726-based PMP in RAID10 mode daumas
2008-12-10 21:49 ` Andrew Morton [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20081210134938.2dbe6d21.akpm@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=daumas77@gmail.com \
--cc=linux-ide@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.