aic7xxx timeout errors on boot

[Jim Shepherd]

Is there any work being done to resolve these timeout errors.  It seems
that at least a few people are hampered by them.  As mentioned before, I
am willing to help in any way. I do not have any background in
programming drivers, but I can apply patches and will provide any
information about the system or errors even if it requires re-entering
by hand the boot-up messages.

-Jim Shepherd

For reference, attached is the original message and one of my
follow-ups:


I am receiving timeout errors on boot preventing the OS to load when
using an old CDROM drive and a Quantum Atlas IV drive:

SCSI: aborting command due to timeout: pid 0, scsi0, channel 0, id 4,
lun 0, Test Unit Ready 00 00 00 00 00
SCSI host 0 abort (pid 0) timed out - resetting
SCSI bus is being reset for host 0 channel 0
(last 2 lines repeat indefinitely)

With the same hardware configuration, I did not receive these errors
with the Linux 2.2.3 kernel, but they I first came across them with the
2.2.10 kernel.  Also using the Adaptec SCSI software under DOS does not
result in any obvious errors (no errors are displayed during loading). 
I have looked through the archives and looked at termination issues as
well as trying the "no_reset", "panic_on_abort", and "override_term"
parameters at bootup.  Following is my configuration.  If I provide
anything else that would be usefull, please let me know.  Thanks in
advance.

System Configuration:
Redhat Linux 6.1 with 2.3.49 kernel (aic7xxx support is compiled into
the kernel)
AIR 54TPI motherboard with on board AIC-7880P SCSI chip
  68 pin connector
     Quantum Atlas IV 9GB harddrive
     End of ribbon cable is terminated
  50 pin connector
     Chinon CDS-535 Internal CDROM Drive (SCSI-2 drive)
     End of ribbon cable is terminated
     (Have also tried using middle connector on ribbon attached to
      motherboard connector and terminating both ends)
No IDE devices

Information from /proc:
Attached devices: 
Host: scsi0 Channel: 00 Id: 00 Lun: 00
  Vendor: QUANTUM  Model: ATLAS IV 9 WLS   Rev: 0808
  Type:   Direct-Access                    ANSI SCSI revision: 03

Adaptec AIC7xxx driver version: 5.2.0/3.2.4
Compile Options:
  TCQ Enabled By Default : Disabled
  AIC7XXX_PROC_STATS     : Disabled
  AIC7XXX_RESET_DELAY    : 5

Adapter Configuration:
           SCSI Adapter: Adaptec AIC-7880 Ultra SCSI host adapter
                           Ultra Wide Controller at PCI 0/8/0
    PCI MMAPed I/O Base: 0xfebfb000
 Adapter SEEPROM Config: SEEPROM found and used.
      Adaptec SCSI BIOS: Enabled
                    IRQ: 10
                   SCBs: Active 0, Max Active 1,
                         Allocated 31, HW 16, Page 255
             Interrupts: 3679
      BIOS Control Word: 0x18b6
   Adapter Control Word: 0x005c
   Extended Translation: Enabled
Disconnect Enable Flags: 0xffff
     Ultra Enable Flags: 0x0000
 Tag Queue Enable Flags: 0x0000
Ordered Queue Tag Flags: 0x0000
Default Tag Queue Depth: 8
    Tagged Queue By Device array for aic7xxx host instance 0:
      {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255}
    Actual queue depth per device for aic7xxx host instance 0:
      {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}

Statistics:

(scsi0:0:0:0)
  Device using Wide/Sync transfers at 20.0 MByte/sec, offset 8
  Transinfo settings: current(25/8/1/0), goal(25/8/1/0), user(25/15/1/0)
  Total transfers 3612 (2650 reads and 962 writes)

Ariel wrote:
<snip> 
> Doug Ledford <dledford at redhat.com> knows about it but he said he needs
> more info to try to solve it.

I am more than willing to help provide as much info as needed and/or try
out new patches.

> I noticed the various boot options to the driver, panic on abort for
> instance, but they all give a very large number of HEX values - with no
> way of sending the data other then perhaps a digital camera - or error
> prone write it down on paper and copy it.
> 
> Is there any way to redirect these numbers to a file?

I believe that you can setup the kernel so that the console messages are
redirected to a printer (even parallel port printers).  This method
would be a little bit better than writing the info and typing it back
into an e-mail.  If beneficial, I could also fax the output to
interested parties.  I vaguely remember reading somewhere that the
console could be redirected to a floppy (This could have just been a
suggestion though or a non-mainstream patch).

 
> I don't think I can load the driver as module since I need SCSI to boot my
> system - am I wrong? Can I perhaps cause it to skip the ID with the chinon
> CD, and try it again after the system booted?
> 
>         -Ariel


This capability would even be helpful in the meantime.  My CDROM drive
is internal, so I have to physically disconnect the SCSI cable before I
boot into Linux and reattach when I boot into DOS to access a CDROM
(Currently, the only reason I ever boot into DOS is to copy files off of
the CDROM so I can access them from Linux)

-Jim Shepherd


To Unsubscribe: send mail to majordomo at FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message