bug#27816: 26.0.50; X protocol error: BadPixmap (invalid Pixmap parameter) on protocol request 55

[martin rudalics]

> Could any of this be related to the change whereby we no longer wait
> for the frame to become visible?

You mean that Emacs now doesn't consume CPU cycles any more?  In any
case, I think that the problem has existed ever since.  It just
manifests itself in different ways.  And it would be nice if someone
else who builds Lucid tried the scenario.  It's even simpler than Jean
Louis describes it:

(1) In one terminal do emacs -Q -nw

(2) M-x server-start

(3) In another terminal do emacsclient -c

(4) In the GUI frame do C-x 5 0 and repeat (3) and (4) at least five
    times.

> If you run Emacs in X synchronized mode and put a breakpoint on
> x_error_quitter, can you see why the problem happens?  Which picmap
> parameter is invalid?

I'll try to do that as soon as I understand how it works and how to
debug it.  Also IIRC synchronized mode didn't always work 100% reliably
here.  From the pngs it seems obvious that the arrows above and below
the slider didn't get drawn.

>> /usr/lib/gcc/x86_64-linux-gnu/4.7/include/stddef.h:426:3: note: previous 
declaration of ‘max_align_t’ was here
>
> Sounds like some Gnulib-related issue.  Perhaps Paul could help figure
> this out.

I suppose some leftovers in the build directory.  I have to build from
scratch.

>> BTW, I never use emacsclient and have no idea how to use it.  Is there a
>> way to stop the server from the starting emacs?
>
> I don't understand the question.  When Emacs starts, the server is not
> started automatically, so I'm unsure what is it that you want to do.
> Please tell me what I missed.

The doc-string of ‘server-start’ says that this would "Allow this Emacs
process to be a server for client processes".  How can I "Stop this
Emacs process to be a server for client processes"?  Frankly I have no
idea what emacsclient is, how it works and how it interacts with the
server.

martin