Re: server.el test failures (was: Re: bug#9800: Incomplete truncated file buffers from the /proc filesystem)

[Jim Porter]

On 2/23/2023 11:52 PM, Eli Zaretskii wrote:
> > Date: Thu, 23 Feb 2023 18:20:39 -0800
> > From: Jim Porter <jporterbugs@gmail.com>
> > Cc: emacs-devel@gnu.org
> >
> > The segfault is in FACE_FROM_ID_OR_NULL, called from
> > Finternal_merge_in_global_face. It happens because the face_cache is
> > null during these tests (since Emacs is noninteractive).
> >
> > The attached patch fixes the issue for me, though I'm not totally sure
> > it's the *right* fix. Any thoughts? (I'm also not 100% sure this is the
> > same issue you're seeing...)
>
> Please show the C backtrace from the crash, and include the Lisp
> backtrace (the "xbacktrace" command in src/.gdbinit).

Ok, I ran Emacs (master branch) under GDB with the following arguments:

  -Q -L ":../test" -l ert -l lisp/server-tests.el --batch --eval 
'(ert-run-tests-batch-and-exit (quote (not (or (tag :unstable) (tag 
:nativecomp)))))'

Attached is the backtrace from 'xbacktrace'.

I think this only occurs when starting the Emacs server in a batch-mode 
Emacs process and then starting a client via "emacsclient -c". I see the 
segfault when running any of the tests in test/lisp/server-tests.el that 
create a frame (e.g. 'server-tests/server-start/stop-prompt-with-client').

This also only seems to occur if the first face that 
'face-set-after-frame-default' passes to 'internal-merge-in-global-face' 
is the default face.

gdb.txt
Description: Text document