OK ... here goes...
Chris Roy
cmr at uniserve.com
Mon Dec 6 00:42:15 PST 1999
First, some background :)
Dual PII 300 in a gigabyte ga-6bxds board with latest bios upgrade
256m ram, aic7895 ultra wide scsi - dual channel, scsi: sr0, st0, sda
mga g200 w/16mb, modem etc
COL 2.3
Linux test.dot.box 2.2.13 #2 SMP Sun Dec 5 23:14:21 PST 1999 i686
unknown
When the the system boots, during the pci phase, it reports pci 12/0
and 12/1 as being my aic7895 - both on irq 11. Whem linux boots,
/proc/scsi/aic7xxx/0 & 1 shows:
Adapter Configuration:
SCSI Adapter: Adaptec AIC-7895 Ultra SCSI host adapter
Ultra Wide Controller Channel A
PCI MMAPed I/O Base: 0xeb000000
Adapter SEEPROM Config: SEEPROM found and used.
Adaptec SCSI BIOS: Enabled
IRQ: 16
Adapter Configuration:
SCSI Adapter: Adaptec AIC-7895 Ultra SCSI host adapter
Ultra Wide Controller Channel B
PCI MMAPed I/O Base: 0xeb002000
Adapter SEEPROM Config: SEEPROM not found, using leftover BIOS values.
Adaptec SCSI BIOS: Enabled
IRQ: 16
Is this something I should be concerned about? Or is this a result of
the apic translation?
Also, what led me to the above discrepencies was the following error
(which I have as yet not resolved - still searching for answers READ:
HELP):
Dec 5 23:52:45 test kernel: (scsi1:0:0:0) Synchronous at 40.0
Mbyte/sec, offset 8.
Dec 5 23:52:45 test kernel: (scsi1:0:0:0) Data overrun detected in
Data-In phase, tag 1;
Dec 5 23:52:45 test kernel: Have seen Data Phase. Length=255,
NumSGs=1.
Dec 5 23:52:45 test kernel: sg[0] - Addr 0xff9a6c0 : Length 255
Dec 5 23:52:45 test kernel: Vendor: WDIGTL Model:
ENTERPRISE Rev: 1.91
Dec 5 23:52:45 test kernel: Type:
Direct-Access ANSI SCSI revision: 02
I've searched everthing I can but don't really comprehend the error; as
close as I can come was a couple of comments in the kernel config
regarding WD enterpise drives.
Being new to this list, Doug's comment (in an earlier thread) on the
"noapic" parameter were interesting in that that set both channels to
irq 11.
As an aside, with an smp kernel, does the number of downloaded sequencer
instructions decrease? ie from 404 to 374? I ask because, when I first
installed RH5.1 it was 404; after an smp kernel recompile it dropped to
374. This was the same for: Suse 5.3 & 6.0, RH5.2, COL 2.2. The above
data overrun error was not present when the sequencer code was 404
instructions!
...
Dec 5 23:52:45 test kernel: (scsi0) Downloading sequencer code... 374
instructions downloaded
...
Dec 5 23:52:45 test kernel: (scsi1) Downloading sequencer code... 374
instructions downloaded
As always, any help, pointers, etc greatly appreciated.
Chris Roy
To Unsubscribe: send mail to majordomo at FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message
More information about the aic7xxx
mailing list