Re: Panic while making buildlernel on RPi3

From: Mark Millard <marklmi_at_yahoo.com>
Date: Sat, 05 Mar 2022 06:51:00 UTC
On 2022-Mar-4, at 20:30, bob prohaska <fbsd@www.zefox.net> wrote:

> On Fri, Mar 04, 2022 at 05:36:20PM -0800, Mark Millard wrote:
>> 
>> 
>>> 
>>> Unfortunately, the crash means that the buffers were not
>>> flushed to media when the problem occurred. The build log
>>> is incomplete compared to what actually built and where
>>> it actually stopped.
>>> 
>>> So I'm not sure how you got:
>>> 
>>> --- ng_pppoe.kld ---
>>> ld: error: no input files
>>> *** [ng_pppoe.kld] Error code 1
>> 
>> Poorly worded on my part. I mean that I'm not sure how
>> you would know it was "in the same place" when you
>> re-ran.
>> 
>>> When I looked at the buildkernel.log file provided,
>>> it stopped without such:
>>> 
>>> --- all_subdir_netgraph ---
>>> Building /usr/obj/usr/src/arm64.aarch64/sys/GENERIC/modules/usr/src/sys/modules/netgraph/pptpgre/ng_pptpgre.o
>>> 
> 
> Apologies for the ambiguity!
> 
> After the panic I simply re-ran the original buildkernel command, which furnished the
> added output and then stopped in _roughly_ the same place. It didn't panic, and maybe
> the panic was unrelated to the error. I've tried hard to _make_ the machine panic,
> using stress2, and failed. 
> 
> Perhaps more curious, I subsequently ran a make -j4 buildkernel with no
> extra options and it completed successfully. It's rebooted now and seems
> to work no worse than lately.

[Split to show temporary context switch.]

> It still displays the strange ping behavior,
> answering a few percent of incoming pings  unless an outbound ping is running. 
> Then it answers maybe half of pings. That seems to allow enough network access
> to permit ssh and git to function, with a sort of stutter. 

Are you going to do the official-builds-on-microsd-card
sorts of boot and try tests that I've suggested, such as a
13.1-PRERELEASE (snapshot) test? (This avoids your having
built anything tested and all but whatever minimal
configuration that you need to do to allow the test.)

Getting a failure from such an installation of official
installation materials might be more likely to lead to
getting help isolating the issue.

[Back to the Subject line's issue . . .]

> I've been using -DWITH_META_MODE as a default setting for buildworld and 
> buildkernel. Might this be part of my  problems with the Pi3's ? 

NO_CLEAN is more likely to have the result messed up: it
does less dependency checking and can miss more that should
be rebuilt/relinked.

WITH_META_MODE is likely to rebuild more than NO_CLEAN, and
so, less likely to include stale material.

Without special knowledge of the details of what all needs to
be rebuilt at the time, WITH_META_MODE is normally safer.

Rebuilding from scratch each time takes a lot of time
for each rebuild.


===
Mark Millard
marklmi at yahoo.com