SiI3112 lost interrupts

SiI3112 lost interrupts

[Hugo Mills]

[-- Attachment #1: Type: text/plain, Size: 2827 bytes --]

   I bought a SiI3112-based SATA adapter (Adaptec 1210SA) and a 120Gb
drive for my computer a while ago. Unfortunately, I can't get them to
work in any sensible manner at the moment.

   I'm running 2.4.22-pre6-ac1, so by all accounts the chip should be
recognised (it is), and put into some form of DMA mode (it isn't) on
boot up. Any attempts to access the disk appear to work, but take a
_very_ long time, and cause lots of lost interrupts. 

   The relevant sections from the logs when the machine starts up look
like this:

Adaptec AAR-1210SA: IDE controller at PCI slot 00:0b.0
PCI: Found IRQ 10 for device 00:0b.0
Adaptec AAR-1210SA: chipset revision 2
Adaptec AAR-1210SA: not 100% native mode: will probe irqs later
ide0: MMIO-DMA , BIOS settings: hda:pio, hdb:pio
ide1: MMIO-DMA , BIOS settings: hdc:pio, hdd:pio
ALI15X3: IDE controller at PCI slot 00:0f.0
ALI15X3: chipset revision 193
ALI15X3: not 100% native mode: will probe irqs later
ide2: BM-DMA at 0x9000-0x9007, BIOS settings: hde:DMA, hdf:pio
ide3: BM-DMA at 0x9008-0x900f, BIOS settings: hdg:pio, hdh:pio
hda: ST3120026AS, ATA DISK drive
hdc: no response (status = 0xfe)
hde: Maxtor 90648D3, ATA DISK drive
blk: queue c0340238, I/O limit 4095Mb (mask 0xffffffff)
hdg: Pioneer DVD-ROM ATAPIModel DVD-120S, ATAPI CD/DVD-ROM drive
ide0 at 0xd4816080-0xd4816087,0xd481608a on irq 10
ide2 at 0x1f0-0x1f7,0x3f6 on irq 14
ide3 at 0x170-0x177,0x376 on irq 15
hda: attached ide-disk driver.
hda: lost interrupt
hda: lost interrupt
hda: lost interrupt
hda: host protected area => 1
hda: lost interrupt
hda: 234441648 sectors (120034 MB) w/8192KiB Cache, CHS=14593/255/63
hda: lost interrupt
hda: lost interrupt
hde: attached ide-disk driver.
hde: host protected area => 1
hde: 12656448 sectors (6480 MB) w/512KiB Cache, CHS=787/255/63, UDMA(33)
Partition check:
/dev/ide/host0/bus0/target0/lun0:<3>hda: lost interrupt
p1 p2 p3
/dev/ide/host2/bus0/target0/lun0: p1

   I've tried most, if not all, of the "fixes" that people have
suggested on LKML for this chipset (all the hdparm invocations, for
example), and haven't managed to change any of the symptoms I
mentioned above. Can anyone suggest any way that I can get this [for
me] rather expensive and currently useless piece of hardware working
under Linux?

   I know that Andre Hedrick has fixed some things in an unreleased
patch, but it's unclear exactly which problems he's fixed, or when the
patch is going to be released. Is anyone in a position to say whether
my problem is going to be fixed by his changes, or when that patch is
going to be released, if ever?

   Thanks,
   Hugo.

-- 
=== Hugo Mills: hugo@... carfax.org.uk | darksatanic.net | lug.org.uk ===
  PGP key: 1C335860 from wwwkeys.eu.pgp.net or http://www.carfax.org.uk
           --- There are three mistaiks in this sentance. ---            

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: SiI3112 lost interrupts

[Andre Hedrick]

Well I just now got it refixed (for two environments) and discussing with
SiI how to address this in a final solution.

Working the driver not to yeild CRC reports from the PATA core when they
need to be address in the SATA layer from PHY level reporting.

Cheers,

Andre Hedrick
LAD Storage Consulting Group

On Sun, 20 Jul 2003, Hugo Mills wrote:

>    I bought a SiI3112-based SATA adapter (Adaptec 1210SA) and a 120Gb
> drive for my computer a while ago. Unfortunately, I can't get them to
> work in any sensible manner at the moment.
> 
>    I'm running 2.4.22-pre6-ac1, so by all accounts the chip should be
> recognised (it is), and put into some form of DMA mode (it isn't) on
> boot up. Any attempts to access the disk appear to work, but take a
> _very_ long time, and cause lots of lost interrupts. 
> 
>    The relevant sections from the logs when the machine starts up look
> like this:
> 
> Adaptec AAR-1210SA: IDE controller at PCI slot 00:0b.0
> PCI: Found IRQ 10 for device 00:0b.0
> Adaptec AAR-1210SA: chipset revision 2
> Adaptec AAR-1210SA: not 100% native mode: will probe irqs later
> ide0: MMIO-DMA , BIOS settings: hda:pio, hdb:pio
> ide1: MMIO-DMA , BIOS settings: hdc:pio, hdd:pio
> ALI15X3: IDE controller at PCI slot 00:0f.0
> ALI15X3: chipset revision 193
> ALI15X3: not 100% native mode: will probe irqs later
> ide2: BM-DMA at 0x9000-0x9007, BIOS settings: hde:DMA, hdf:pio
> ide3: BM-DMA at 0x9008-0x900f, BIOS settings: hdg:pio, hdh:pio
> hda: ST3120026AS, ATA DISK drive
> hdc: no response (status = 0xfe)
> hde: Maxtor 90648D3, ATA DISK drive
> blk: queue c0340238, I/O limit 4095Mb (mask 0xffffffff)
> hdg: Pioneer DVD-ROM ATAPIModel DVD-120S, ATAPI CD/DVD-ROM drive
> ide0 at 0xd4816080-0xd4816087,0xd481608a on irq 10
> ide2 at 0x1f0-0x1f7,0x3f6 on irq 14
> ide3 at 0x170-0x177,0x376 on irq 15
> hda: attached ide-disk driver.
> hda: lost interrupt
> hda: lost interrupt
> hda: lost interrupt
> hda: host protected area => 1
> hda: lost interrupt
> hda: 234441648 sectors (120034 MB) w/8192KiB Cache, CHS=14593/255/63
> hda: lost interrupt
> hda: lost interrupt
> hde: attached ide-disk driver.
> hde: host protected area => 1
> hde: 12656448 sectors (6480 MB) w/512KiB Cache, CHS=787/255/63, UDMA(33)
> Partition check:
> /dev/ide/host0/bus0/target0/lun0:<3>hda: lost interrupt
> p1 p2 p3
> /dev/ide/host2/bus0/target0/lun0: p1
> 
>    I've tried most, if not all, of the "fixes" that people have
> suggested on LKML for this chipset (all the hdparm invocations, for
> example), and haven't managed to change any of the symptoms I
> mentioned above. Can anyone suggest any way that I can get this [for
> me] rather expensive and currently useless piece of hardware working
> under Linux?
> 
>    I know that Andre Hedrick has fixed some things in an unreleased
> patch, but it's unclear exactly which problems he's fixed, or when the
> patch is going to be released. Is anyone in a position to say whether
> my problem is going to be fixed by his changes, or when that patch is
> going to be released, if ever?
> 
>    Thanks,
>    Hugo.
> 
> -- 
> === Hugo Mills: hugo@... carfax.org.uk | darksatanic.net | lug.org.uk ===
>   PGP key: 1C335860 from wwwkeys.eu.pgp.net or http://www.carfax.org.uk
>            --- There are three mistaiks in this sentance. ---            
> 

Re: SiI3112 lost interrupts

[Hugo Mills]

[-- Attachment #1: Type: text/plain, Size: 640 bytes --]

On Sun, Jul 20, 2003 at 01:51:36PM -0700, Andre Hedrick wrote:
> 
> Well I just now got it refixed (for two environments) and discussing with
> SiI how to address this in a final solution.

   That's good to know what's happening with these drivers, and that
there's a fix on the way. Thank you.

   Are you able & willing to give an estimated timescale for the
public release of the solution?

   Thanks,
   Hugo.

-- 
=== Hugo Mills: hugo@... carfax.org.uk | darksatanic.net | lug.org.uk ===
  PGP key: 1C335860 from wwwkeys.eu.pgp.net or http://www.carfax.org.uk
           --- There are three mistaiks in this sentance. ---            

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]