Re: Arm v7 RPi2 -current unresponsive to debugger escape during buildworld

On Nov 7, 2025, at 08:16, bob prohaska <fbsd@www.zefox.net> wrote:

> On Thu, Nov 06, 2025 at 08:15:49PM -0800, Mark Millard wrote:
>> 
>> The enter-tilda-control-B sequence is via the tty driver and
>> kernel. It is not tied to a specific process.
>> 
> 
> Would debugging via JTAG offer any better recourse to a
> silent hang?

I have no experience with JTAG or analogous. My logic
analyzer background does not really apply.

Even if I managed to replicate the behavior, I've not
come up with any way of being able to gather useful
evidence based on what you have reported.

One avoid-the-problem technique is to build on a more
capable system that supports armv7 chroot use --by
mounting the drive and chrooting to it, doing the
build and install in the chroot, and then exit,
dismount, and move the media back to the RPI2.

This does have the advantage that the live kernel is
not being changed and the system processes are not
being updated, so reboots during the install are not
required: There are no runtime conflicts with the live
system.

But it also assumes that your context supports also
having the disk media mounted on the more capable
system for a while.

===
Mark Millard
marklmi at yahoo.com