Is CPUTYPE=cortex-A7 supposed to work?

[Andrew Gierth]

Here's another issue that seems to go away as a result of removing the
cortex-a7 option:

On the cortex-a7 build (world and all packages built with CPUTYPE set):

emacs-25.1,3 built with x11 support and running in graphical mode is
unstable; crashes randomly either with a mutex assertion (enqueue_mutex
complaining that the mutex is already owned) or with SEGV. Running in
text mode, it doesn't crash, but hitting pagedown in some buffers (where
there are many similar lines) _reliably_ produces incorrect display
output in which some characters are displaced by 4:

  drwxr-xr-x    2 root  wheel   512 Feb 18 09:35 isp
  drwxr-xr-x    2 root  wheel   512 Feb 18 09:40 ispfw
  drwxr-xr-x    2 root  wheel   512 20 10:31 iwi et
  drwxr-xr-x    2 root  wheel  1024 Feb 18 09:38 iwm

in place of

  drwxr-xr-x    2 root  wheel   512 Feb 18 09:35 isp
  drwxr-xr-x    2 root  wheel   512 Feb 18 09:40 ispfw
  drwxr-xr-x    2 root  wheel   512 Feb 20 10:31 iwi
  drwxr-xr-x    2 root  wheel  1024 Feb 18 09:38 iwm

(the "et" is what was displayed at that point on the previous page)

(by "reliably" I mean that for given buffer contents and sequence of
scroll movements that exhibit the problem, it always happens at the same
place in the same way)

On the non-CPUTYPE build, I can't reproduce either of the issues (though
it's a bit early yet to tell regarding the random-crashes one).
Investigation continues...

-- 
Andrew.

(Clarke's second law: to discover the limits of the possible, one must
venture past them into the realm of the impossible)