SB600 in AHCI mode will not run SATA optical drive

SB600 in AHCI mode will not run SATA optical drive

[Darrick Burch]

Hey, all:

I'm having a bit of an issue with running my Pioneer BDC-202BK BD-ROM
drive on my motherboard's (MSI K9A2-CF) AMD SB600-based SATA controller
and I was wondering if anyone has seen anything similar.

The problem is that the drive does not initialize at all under Linux.  I
get errors from the ahci driver which, from what I've read from the wiki,
is the only driver under which the SATA ports on the SB600 are supported.

The errors I'm getting loop infinitely until I remove the drive data cable:

Nov  9 21:51:51 linux kernel: ata3: soft resetting link
Nov  9 21:51:51 linux kernel: ata3: SATA link down (SStatus 1 SControl 300)
Nov  9 21:51:51 linux kernel: ata3: EH complete
Nov  9 21:51:51 linux kernel: ata3: exception Emask 0x10 SAct 0x0 SErr
0x40c0000 action 0xa frozen
Nov  9 21:51:51 linux kernel: ata3: irq_stat 0x00000040, connection status
changed
Nov  9 21:51:51 linux kernel: ata3: SError: { CommWake 10B8B DevExch }
Nov  9 21:51:51 linux kernel: ata3: soft resetting link
Nov  9 21:51:51 linux kernel: ata3: SATA link down (SStatus 1 SControl 300)
Nov  9 21:51:51 linux kernel: ata3: EH complete
Nov  9 21:51:51 linux kernel: ata3: exception Emask 0x10 SAct 0x0 SErr
0x40c0000 action 0xa frozen
Nov  9 21:51:51 linux kernel: ata3: irq_stat 0x00000040, connection status
changed
Nov  9 21:51:51 linux kernel: ata3: SError: { CommWake 10B8B DevExch }

I've been able to get around this problem (which prevents me from booting
completely) by commenting the PCI_VDEVICE definition for the SB600.  Of
course, this means that the controller is effectively disabled.  I'm not
in desperate need for my optical drive in Linux at the moment, but it
would be nice to have access to it someday.

I suppose my questions are the following:

1. Is it possible that the drive simply does not want to run under AHCI
mode?  I should note that the drive works correctly under Windows XP with
the controller set to IDE emulation mode.

2. I'm thinking that the best thing for me to do is to get another PCIe
SATA controller and dedicate it to this one drive.  Does anyone want to
give their opinion on this?

Thanks in advance,
-Darrick

PS - I have no other devices on the SATA controller; my two hard drives
are PATA and they're connected to the single PATA header available on the
motherboard.  These, of course, are working fine.

Re: SB600 in AHCI mode will not run SATA optical drive

[Robert Hancock]

Darrick Burch wrote:
> Hey, all:
> 
> I'm having a bit of an issue with running my Pioneer BDC-202BK BD-ROM
> drive on my motherboard's (MSI K9A2-CF) AMD SB600-based SATA controller
> and I was wondering if anyone has seen anything similar.
> 
> The problem is that the drive does not initialize at all under Linux.  I
> get errors from the ahci driver which, from what I've read from the wiki,
> is the only driver under which the SATA ports on the SB600 are supported.
> 
> The errors I'm getting loop infinitely until I remove the drive data cable:
> 
> Nov  9 21:51:51 linux kernel: ata3: soft resetting link
> Nov  9 21:51:51 linux kernel: ata3: SATA link down (SStatus 1 SControl 300)
> Nov  9 21:51:51 linux kernel: ata3: EH complete
> Nov  9 21:51:51 linux kernel: ata3: exception Emask 0x10 SAct 0x0 SErr
> 0x40c0000 action 0xa frozen
> Nov  9 21:51:51 linux kernel: ata3: irq_stat 0x00000040, connection status
> changed
> Nov  9 21:51:51 linux kernel: ata3: SError: { CommWake 10B8B DevExch }
> Nov  9 21:51:51 linux kernel: ata3: soft resetting link
> Nov  9 21:51:51 linux kernel: ata3: SATA link down (SStatus 1 SControl 300)
> Nov  9 21:51:51 linux kernel: ata3: EH complete
> Nov  9 21:51:51 linux kernel: ata3: exception Emask 0x10 SAct 0x0 SErr
> 0x40c0000 action 0xa frozen
> Nov  9 21:51:51 linux kernel: ata3: irq_stat 0x00000040, connection status
> changed
> Nov  9 21:51:51 linux kernel: ata3: SError: { CommWake 10B8B DevExch }
> 
> I've been able to get around this problem (which prevents me from booting
> completely) by commenting the PCI_VDEVICE definition for the SB600.  Of
> course, this means that the controller is effectively disabled.  I'm not
> in desperate need for my optical drive in Linux at the moment, but it
> would be nice to have access to it someday.
> 
> I suppose my questions are the following:
> 
> 1. Is it possible that the drive simply does not want to run under AHCI
> mode?  I should note that the drive works correctly under Windows XP with
> the controller set to IDE emulation mode.

Could be some kind of SATA PHY quirk or incompatibility that only shows 
up in AHCI mode or something. It seems from the output that the SATA 
link may be going up and down. You could try forcing 1.5Gbps 
(libata.force=1.5Gbps if libata is built in, or force=1.5Gbps when the 
libata module gets loaded) and see if that helps.

Maybe Tejun has some more ideas..

> 2. I'm thinking that the best thing for me to do is to get another PCIe
> SATA controller and dedicate it to this one drive.  Does anyone want to
> give their opinion on this?
> 
> Thanks in advance,
> -Darrick
> 
> PS - I have no other devices on the SATA controller; my two hard drives
> are PATA and they're connected to the single PATA header available on the
> motherboard.  These, of course, are working fine.
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ide" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: SB600 in AHCI mode will not run SATA optical drive

[Darrick Burch]

Robert Hancock wrote:

> Could be some kind of SATA PHY quirk or incompatibility that only shows
> up in AHCI mode or something. It seems from the output that the SATA
> link may be going up and down. You could try forcing 1.5Gbps
> (libata.force=1.5Gbps if libata is built in, or force=1.5Gbps when the
> libata module gets loaded) and see if that helps.

That appears to have done it--thanks!  I feel kinda bad now because I was
suspecting a speed issue, but couldn't figure out how to force it.  If I
had read the kernel documentation a little more, I would have found the
kernel option you mentioned.  Duh.  At any rate, thanks again.

Re: SB600 in AHCI mode will not run SATA optical drive

[Robert Hancock]

Darrick Burch wrote:
> Robert Hancock wrote:
> 
>> Could be some kind of SATA PHY quirk or incompatibility that only shows
>> up in AHCI mode or something. It seems from the output that the SATA
>> link may be going up and down. You could try forcing 1.5Gbps
>> (libata.force=1.5Gbps if libata is built in, or force=1.5Gbps when the
>> libata module gets loaded) and see if that helps.
> 
> That appears to have done it--thanks!  I feel kinda bad now because I was
> suspecting a speed issue, but couldn't figure out how to force it.  If I
> had read the kernel documentation a little more, I would have found the
> kernel option you mentioned.  Duh.  At any rate, thanks again.

Could be we need a quirk in libata for that drive to do that 
automatically, or something.. Can you post the dmesg output from bootup 
with the force 1.5Gbps option in place?

Re: SB600 in AHCI mode will not run SATA optical drive

[Darrick Burch]

>> Robert Hancock wrote:
>>
>>> Could be some kind of SATA PHY quirk or incompatibility that only shows
>>> up in AHCI mode or something. It seems from the output that the SATA
>>> link may be going up and down. You could try forcing 1.5Gbps
>>> (libata.force=1.5Gbps if libata is built in, or force=1.5Gbps when the
>>> libata module gets loaded) and see if that helps.
>>
>> That appears to have done it--thanks!  I feel kinda bad now because I
>> was
>> suspecting a speed issue, but couldn't figure out how to force it.  If I
>> had read the kernel documentation a little more, I would have found the
>> kernel option you mentioned.  Duh.  At any rate, thanks again.
>

> Could be we need a quirk in libata for that drive to do that
> automatically, or something.. Can you post the dmesg output from bootup
> with the force 1.5Gbps option in place?

Robert:

I hope I got the part of the log that you're looking for.  If not, let me
know because there's more where that came from.  Anyway:

libata version 3.00 loaded.
ACPI: PCI Interrupt 0000:00:14.1[A] -> GSI 16 (level, low) -> IRQ 16
PCI: Setting latency timer of device 0000:00:14.1 to 64
scsi0 : pata_atiixp
scsi1 : pata_atiixp
ata1: PATA max UDMA/100 cmd 0x1f0 ctl 0x3f6 bmdma 0xff00 irq 14
ata2: PATA max UDMA/100 cmd 0x170 ctl 0x376 bmdma 0xff08 irq 15
ata1.00: ATA-6: WDC WD1600JB-00GVC0, 08.02D08, max UDMA/100
ata1.00: 312581808 sectors, multi 16: LBA48
ata1.01: ATA-6: WDC WD1600JB-00GVC0, 08.02D08, max UDMA/100
ata1.01: 312581808 sectors, multi 16: LBA48
ata1.00: configured for UDMA/100
ata1.01: configured for UDMA/100
scsi 0:0:0:0: Direct-Access     ATA      WDC WD1600JB-00G 08.0 PQ: 0 ANSI: 5
scsi 0:0:1:0: Direct-Access     ATA      WDC WD1600JB-00G 08.0 PQ: 0 ANSI: 5
ahci 0000:00:12.0: version 3.0
ACPI: PCI Interrupt 0000:00:12.0[A] -> GSI 22 (level, low) -> IRQ 22
ahci 0000:00:12.0: controller can't do 64bit DMA, forcing 32bit
ahci 0000:00:12.0: controller can't do PMP, turning off CAP_PMP
ahci 0000:00:12.0: AHCI 0001.0100 32 slots 4 ports 3 Gbps 0xf impl IDE mode
ahci 0000:00:12.0: flags: ncq sntf ilck pm led clo pio slum part
scsi2 : ahci
scsi3 : ahci
scsi4 : ahci
scsi5 : ahci
ata3: FORCE: PHY spd limit set to 1.5Gbps
ata3: SATA max UDMA/133 irq_stat 0x00000040 irq 22, connection status changed
ata4: FORCE: PHY spd limit set to 1.5Gbps
ata4: SATA max UDMA/133 abar m1024@0xfe8ff800 port 0xfe8ff980 irq 22
ata5: FORCE: PHY spd limit set to 1.5Gbps
ata5: SATA max UDMA/133 abar m1024@0xfe8ff800 port 0xfe8ffa00 irq 22
ata6: FORCE: PHY spd limit set to 1.5Gbps
ata6: SATA max UDMA/133 abar m1024@0xfe8ff800 port 0xfe8ffa80 irq 22
ata3: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
ata3.00: ATAPI: PIONEER BD-ROM  BDC-202, 1.07, max UDMA/66
ata3.00: SB600 AHCI: limiting to 255 sectors per cmd
ata3.00: SB600 AHCI: limiting to 255 sectors per cmd
ata3.00: configured for UDMA/66
ata4: SATA link down (SStatus 0 SControl 310)
ata5: SATA link down (SStatus 0 SControl 310)
ata6: SATA link down (SStatus 0 SControl 310)
scsi 2:0:0:0: CD-ROM            PIONEER  BD-ROM  BDC-202  1.07 PQ: 0 ANSI: 5

Oh, yeah, in case it matters: I'm running SuSE 11.0 with x86_64 kernel
2.6.25.18-0.2-default.

Thanks!
-Darrick

Re: SB600 in AHCI mode will not run SATA optical drive

[Tejun Heo]

Robert Hancock wrote:
> Darrick Burch wrote:
>> Robert Hancock wrote:
>>
>>> Could be some kind of SATA PHY quirk or incompatibility that only shows
>>> up in AHCI mode or something. It seems from the output that the SATA
>>> link may be going up and down. You could try forcing 1.5Gbps
>>> (libata.force=1.5Gbps if libata is built in, or force=1.5Gbps when the
>>> libata module gets loaded) and see if that helps.
>>
>> That appears to have done it--thanks!  I feel kinda bad now because I was
>> suspecting a speed issue, but couldn't figure out how to force it.  If I
>> had read the kernel documentation a little more, I would have found the
>> kernel option you mentioned.  Duh.  At any rate, thanks again.
> 
> Could be we need a quirk in libata for that drive to do that
> automatically, or something.. Can you post the dmesg output from bootup
> with the force 1.5Gbps option in place?

I don't think we can do a quirk here as it happens before the driver
has any way to identify the device.  libata EH usually can react to
PHY problems like this correctly and lowers speed automatically but in
this case it doesn't work because the PHY stays online too briefly for
libata EH to consider it to be the one device failing.  It just resets
error history and restarts probing sequence on each PHY event.  :-(

Maybe libata EH needs to keep PHY errors over failing probings and act
on it.  Hmm...

-- 
tejun