Re: CMD controller not found on 4000/500A

[George Harvey <fr30%dial.pipex.com@localhost>]

On Sun, 18 Feb 2024 10:17:58 -0500
Dave McGuire <mcguire%neurotica.com@localhost> wrote:

> 
>    I'm not sure I'd call it a smoking gun, but my 705A boots reliably 
> now from a CMD CQD-223/TM.
> 
>    In sys/arch/vax/vax/ka680.c, in struct ka680_calls, there's a 
> hard-coded timing value in member cpu_vups.  Change that to 45 or so. 
> That made the difference here.
> 

I have built a kernel (nearly 17 hours to do a native build on the 500A) with vups increased to 45 and it still fails to find the CMD controller. There is now a noticeable pause after it detects mscpbus0, presumably due to the increased delay, but it fails to detect mscpbus1.

Since network booting always works, and disk booting doesn't, could the bootloader be doing something that is interfering with the kernel's hardware detection? My bootloader is [1.12 (Tue Jan 16 08:28:51 UTC 2024)], which version do you have on your 705A?

George