Re: Adaptec AHD U320 operating as only U160

On Feb 17, 2008 8:33 AM, Justin T. Gibbs <gibbs_at_scsiguy.com> wrote:
> Niki Denev wrote:
>  > I was playing around with DTrace, tracing cam/xpt and the ahd driver and
>  > found out that if i comment the following code :
>  >
>  >         if ((spi3caps & SID_SPI_IUS) == 0)
>  >                 spi->ppr_options &= ~MSG_EXT_PPR_IU_REQ;
>  >
>  > at line 6655 in sys/cam/cam_xpt.c my disks again negotiate as U320 :
>  >
>  >   da0 at ahd0 bus 0 target 0 lun 0
>  >   da0: <SEAGATE ST336807LW 0C01> Fixed Direct Access SCSI-3 device
>  >   da0: 320.000MB/s transfers (160.000MHz DT, offset 63, 16bit)
>  >   da0: Command Queueing Enabled
>  >   da0: 35003MB (71687372 512 byte sectors: 255H 63S/T 4462C)
>
> The aic79xx driver was not properly exporting its capabilities to
> CAM.  This has been addressed as of version 1.30 of aic79xx_osm.c.
> Please let me know if you still have problems.
>

With aic79xx_osm.c 1.30 all of my U320 drives are detected as such. Thanks!
I'm only wondering why with the latest version
"cpi->transport_version" is set to 2, and
then set to 4 on the next line? Probably you left it there for readability?

>  >
>  > Unfortunately I began seeing again the "Invalid sequencer interrupt"
>  > messages that i was seeing before(with fbsd 6.2) with Seagate drives
>  > on Adaptec at U320 speeds, and I prey that they are harmless (as they
>  > used to be?)
>
> While I do not know their root cause, they do appear to be harmless.
> Do you happen to have your drives in a SES enclosure (on a backplane
> with a SES chip)?  One user claimed this was only reproducible when
> a GEM318 SES chip was on the bus.
>
> --
> Justin

Nope, no SES enclosure, just plain 68pin scsi drives.
I have now put version 1.30 of aic79xx_osm.c on two machines,
One of them is running the Dtrace snapshot of 8.0-current
with dual channel PCI-X Adaptec U320 controller (on a plain
32bit/33mhz pci slot) :

ahd0_at_pci0:4:1:0:        class=0x010000 card=0x00429005 chip=0x80129005
rev=0x03 hdr=0x00
    vendor     = 'Adaptec Inc'
    device     = 'ASC-29320 Ultra320 SCSI Controller'
    class      = mass storage
    subclass   = SCSI

ahd0: <Adaptec 29320 Ultra320 SCSI adapter> port
0xe000-0xe0ff,0xd800-0xd8ff mem 0xfeb9c000-0xfeb9dfff irq 17 at device
1.0 on pci4
ahd0: [ITHREAD]

and has four Seagate 73G drives attached to it :
da0 at ahd1 bus 0 target 0 lun 0
da0: <SEAGATE ST373307LW 0005> Fixed Direct Access SCSI-3 device
da0: 320.000MB/s transfers (160.000MHz DT, offset 63, 16bit)
da0: Command Queueing Enabled
da0: 70007MB (143374744 512 byte sectors: 255H 63S/T 8924C)

on this machine i haven't seen "Invalid sequencer interrupt" messages yet.

The other machine is running 7.0-PRERELEASE
with Adaptec U320 on PCIe (actually it's still PCI/X with a PCIe bridge chip) :

pcib4_at_pci0:3:0:0:       class=0x060400 card=0x00000000 chip=0x811410b5
rev=0xbc hdr=0x01
    vendor     = 'PLX Technology Inc.'
    class      = bridge
    subclass   = PCI-PCI
ahd0_at_pci0:4:4:0:        class=0x010000 card=0x00459005 chip=0x80179005
rev=0x10 hdr=0x00
    vendor     = 'Adaptec Inc'
    device     = 'ASC-29320ALP Ultra320 SCSI Controller'
    class      = mass storage
    subclass   = SCSI

and with four 36G Seagate drives (they are smaller but are newer
revision than the ones in the first machine) :

da0 at ahd0 bus 0 target 0 lun 0
da0: <SEAGATE ST336807LW 0C01> Fixed Direct Access SCSI-3 device
da0: 320.000MB/s transfers (160.000MHz DT, offset 63, 16bit)
da0: Command Queueing Enabled
da0: 35003MB (71687372 512 byte sectors: 255H 63S/T 4462C)

On this machine, i'm seeing the sequencer messages during boot,
which are somewhat lenghty and i'm putting them here :
  http://www.bg.freebsd.org/~ndenev/ahdinvseqintr.txt

Then there are several messages like this one too :
Copied 18 bytes of sense data offset 12: 0x70 0x0 0x5 0x0 0x0 0x0 0x0
0xa 0x0 0x0 0x0 0x0 0x20 0x0 0x2 0xcf 0x0 0x0

  --Niki