7890/aic7xxx 5.1.0pre5 part. success

[Robert G. Brown]

On Fri, 31 Jul 1998, felix wrote:

> Hi all,
> 
> I've tested your patch, but it didn't work for me on an
> ASUS P2B Board. But another patch posted by Wallace E. Owen
> I found on Dejanews worked fine.
> He posted it in c.o.l.h on the 30.Jul.1998.
> He suggested following two things
> 1.
> add |FAILDIS in restart_sequencer() 
> aic_outb(p, FASTMODE | FAILDIS, SEQCTL);
> 2.
> change PCI_COMMAND to DEVCONFIG at line 8531
> #ifdef AIC7XXX_STRICT_PCI_SETUP
> 	pci_read_config_dword(pdev,DEVCONFIG,&devconfig);
> 
> Hope this helps
> and thanks for your work so far.
> 
> Bye
>  Felix and Sven
> 
> 
> To Unsubscribe: send mail to majordomo at FreeBSD.org
> with "unsubscribe aic7xxx" in the body of the message
> 

Yo, dudes.  I tested BOTH of these patches and NEITHER one worked
here.  The patch above had no effect at all.  The previous (lengthy)
patch actually broke my system earlier -- it failed when probing the
first device on the 7860 and never even got to the 7890.

I have to say that I haven't given up on the ideas in the long patch
yet.  The notion that there is a critical delay in the system that
isn't quite long enough is consistent with my observations of marginal
failure and loads that work sometimes and fail sometimes.  And the
register dumping routine will be very helpful as I continue to explore.

Doug, I got the aicasm code and have more or less followed
instructions in the README to build aic7xxx.list.  Pardon my
ignorance, but what am I to do with it now?  I've compared it line by
line to my last copy of Justin's aic7xxx.seq (mentally translating the
conditionals) and there are obvious differences.  I'm comparing it to
the aic7xxx/aic7xxx.seq in 5.1.0pre5, but the aic7xxx.list I've got
seems to have two tabs preceding each line and garbage in front of a
lot of the assembler -- makes it hard to just do a diff.

Also, I don't quite know what to do about the questions concerning
flags in the make list process.  I have both a 7860 and a 7890
onboard, and I would have expected the aic7xxx.seq to not depend on
the particular hardware in place.  Any direction you can give me will
be very helpful.

Thanks

    rgb

Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu




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