[rescue] SUN2 behavior

[Dan Moisa]

This is great, thanks for putting it together!!
I'm definitely not even close. Trying to take everything off the address
bus since I'm seeing some weird half-level driving behavior.

At least it's clearly not worth writing a stub boot program since
diagnostic LEDs are so early in the sequence.

Dan.

On Sat, Sep 5, 2026 at 11:32 PM Romain Dolbeau via rescue <
rescue at sunhelp.org> wrote:

> Le dim. 6 sept. 2026 à 06:02, Dan Moisa via rescue
> <rescue at sunhelp.org> a écrit :
> > 1. What LED diagnostic sequence do you get, if any, if you run the board
> just by itself - no RAM, nothing else on multibus,
>
> Out of a cold reset. The CPU first fetches two words, the initial SP
> and PC (<
> https://github.com/calmsacibis995/sunos-34-src/blob/7fcdbbecbbde0ccf28a923c533a5f20ed90d9ccb/sun/prom_monitor/rsun/mon/kernel/romvec.s#L15-L16
> >).
>
> It will then jump to the address stored in the SP
> (<
> https://github.com/calmsacibis995/sunos-34-src/blob/7fcdbbecbbde0ccf28a923c533a5f20ed90d9ccb/sun/prom_monitor/rsun/mon/kernel/trap.s#L71
> >)
>
> At that point it sets a few systems registers and then set the LEds to
> L_INITIAL (<
> https://github.com/calmsacibis995/sunos-34-src/blob/7fcdbbecbbde0ccf28a923c533a5f20ed90d9ccb/sun/prom_monitor/rsun/mon/kernel/trap.s#L109
> >
> ,  <
> https://github.com/calmsacibis995/sunos-34-src/blob/7fcdbbecbbde0ccf28a923c533a5f20ed90d9ccb/sun/prom_monitor/rsun/mon/h/s2led.h#L22
> >).
> That is the first deliberate change to the Leds.
>
> After more system setup, it jumps to PowerUpReset that immediately go
> to _qpdiag (<
> https://github.com/calmsacibis995/sunos-34-src/blob/7fcdbbecbbde0ccf28a923c533a5f20ed90d9ccb/sun/prom_monitor/rsun/mon/diag/diag.s#L35
> >).
> The first thing _qpdiag will do is the "marching one" so the user can
> check all the Leds are working.
>
> After that, it goes for hardware testing starting with the MMU, and the
> Leds do:
>
>  => L_SM_CONST
>  => L_SM_DATA
>  => L_SM_ADDR
>  => L_PM_CONST
>  => L_PM_DATA
>  => L_PM_ADDR
>  => L_PROM
>  => L_M_MAP
>  => L_SETUP_MEM [that one fails without memory]
>  => L_SETUP_MAP
>  => L_SETUP_FB
>  => L_SETUP_KEYB
>  => L_RUNNING
>  => L_HEARTBEAT
>  => L_RUNNING
>  => L_HEARTBEAT
> (repeat the blink)
>
> So if you never reach the Leds as 0x1 (L_INITIAL), the CPU can't even
> fetch the initial words or can't set the Leds at all.
> If you see the "marching one", CPU is good, and so is the /DTACK logic.
> After that, you need the Led pattern to tell where things went wrong.
>
> Cordially,
>
> --
> Romain Dolbeau
>
> _______________________________________________
> rescue list - http://sunhelp.org/mailman/listinfo/rescue_sunhelp.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sunhelp.org/pipermail/rescue_sunhelp.org/attachments/20260906/b22c1b6d/attachment.html>