FreeBSD 11.3-RELEASE and 11.2-RELEASE images fail to boot on BeagleBone Black

[Dennis Clarke]

On 7/14/19 1:28 PM, Mark Linimon wrote:
> On Sun, Jul 14, 2019 at 04:19:37AM -0400, Dennis Clarke wrote:
>> I have the same issue with ppc64 and most certainly for RISC-V where
>> we are really well and truely in outer darkness.
> 
> I don't understand.  Can you explain this to me more clearly?
>

Essentially one my checkout the sources and do a clean build only to see
the kernel panic on ppc64 hardware.  There are times where things work
and there are mostly times where things simply don't and in very bad
ways. On a PowerMac G5 it is usually the latter.


> fwiw, I have personally invested a lot of time in getting our powerpc64
> documentation up to date on the wiki.  As well:
> 
>    blackbird1% uname -a
>    FreeBSD blackbird1.lonesome.com 13.0-CURRENT FreeBSD 13.0-CURRENT r349478 GENERIC  powerpc
>    blackbird1% uptime
>    5:25PM  up 12:52, 2 users, load averages: 0.01, 0.05, 0.0
> 
> (It would have been up longer but I moved it into my rack.)
> 
> What have I missed?

Well I think I last saw things sort-of working with r345425 but only in
single user mode and only with a single core.  I wiped the machine out
and started over today with 12.0-RELEASE r341666 and again only with a
single core available.

I am watching r350018 compile right now but I am still fuzzy on what the
boot loaded magic words are. Could be usefdt=1 and maybe not. Could be
that I must have kern.smp.disabled=1 and maybe not. Or it could panic so
early that not much of this matters.

As I say, outer darkness.

> As for RISC-V, IIUC, support for the architecture is very much in an
> alpha state.

Strangely I have an instance that boots neatly inside a qemu vm and that
even includes a ZFS zpool.


-- 
Dennis Clarke
RISC-V/SPARC/PPC/ARM/CISC
UNIX and Linux spoken
GreyBeard and suspenders optional