Thank you @Peter2121.. I had to modify one header that was linux specific in order to make VNC server work:

@cederom,
It seems that has_header meson function is not recursive.
I could reproduce your problem creating symlinks of epoll.h in /usr/local/include/sys or in /usr/include/sys
I don't know why do you have such links (or strange CFLAGS=I/usr/local/include/libepoll-shim) but anyway it is not a standard situation. So I think that libepoll-shim package installed would not prevent correct building of EFL without epoll.

@cederom,
I cannot reproduce your problem on my test PC under FreeBSD 12.1.
I have libepoll-shim package installed, but during the configuration of EFL meson does not detect sys/epoll.h header. I have it installed in /usr/local/include/libepoll-shim/sys/epoll.h and I have CFLAGS=-I/usr/local/include exported, but meson does not detect it. Don't you have a symlink of this file in /usr/include for any reason?

It's just

I can confirm that this is the EFL build time configuration issue. Older EFL build works fine with Terminology 1.7.0. I will looks into that epoll stuff in a free moment. Thank you!! :-)

In T8716#152454, @cederom wrote:

epoll-shim is required by other packages and cannot be removed from the system. If presence of that package impacts Terminology then terminology build scripts need fix.

Thank you for your feedback that it works for you, this is valuable input.

@cederom,
Try to build EFL and Terminology in jail with the strict minimum of other software installed. Terminology and Enlightenment from Git work fine here with EFL 1.24 on FreeBSD 11.3.

Your terminology 1.6.0 is probably linked with an other version of efl, one that was compiled when epoll-shim wasn't installed.

You cannot build a port package with all this other software installed.

Did you build EFL with those flags?

The problem with Terminology 1.7.0 is not fixed, exactly the same setup and 1.6.0 works fine, I cannot remove libepoll-shim from system because other applications depend on it. Will report a working solution if I find one. Thanks for your quick response :-)

EFL

meson build -Dsystemd=false -Deeze=false

Well, it looks like I am updating the port, because no one did that so far ;-)

Please advise the FreeBSD port maintainers not to build with wayland support . It does not work, it merely builds.

In T8716#152440, @netstar wrote:

@cederom

https://www.enlightenment.org/docs/distros/freebsd-start.md

Follow that guide - remove the libepoll-shim package as suggested by @arrowdodger

These goes the port updates tickets and whole ports sources / patches are there too:

• Terminology: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=245598
• EFL: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=245594
• Enlightenment: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=245596

@cederom

That sounds about right @arrowdodger

@cederom Are you building in unclean environment? It might be that build system picks stuff from libepoll-shim FreeBSD package.

I'm runnning E 24.0 and EFL 1.24 and Terminology 1.7.0 on FreeBSD 12.1 and OpenBSD 6.7-beta

Hey @netstar :-) Which version of Terminology on which E and EFL?

I'm running terminology fine here on OpenBSD and FreeBSD.

Can you show the build options for EFL? There should be no reference to epoll

There are some changes in src/bin/termpty.c but I don't see why it should fail.

Sure thing, I am porting EFL+E to FreeBSD so no problem with building and patching :-)

Would it be possible for you to compile terminology 1.6.0 with the same efl version?

Hello @billiob and thank you for quick response :-)

If you create a new terminal from the menu, is it better?
What if you focus/unfocus the window?

No update in more than a year

Should be fixed by 38e4935427939d21f9c545935e3a0e2ffa674a0b

For now I patched the last release version so it is not the clone from git, but thanks for the tip. 🙇

Are you using terminology from git? Because there was a bug and pasting stuff might not have worked as expected. It was fixed in
https://phab.enlightenment.org/rTRMb76bbbe45507e2c1c62e38b60f12c09eef176e15 ?

that's not transparency. first it's an implied protocol where whoever sets a pixmap; on root window sets a property that gives that's pixmap id. it's a hack. it's also got wonderful side effects like the client has to track it's x,y relative to root and redraw every time it's moved - this doesn't sync. it'll always be a bit out-of sync as yoiu get the events that the window moved after it happened. in the age of real compositing this hack is no longer valid or desirable. as the person who came up with the hack in the first place over 20 years ago... i think i get to call it dead when it is dead. :) move on and use a compositor. :)

I was expecting something like using the picture from the root window of the wm and just use that as background.
I'm using feh to set the background picture.

how is this broken? to be transparent you create a window with an argb visual and fill it with alpha channel pixels... efl actually doesnt have any special different code-path per wm. there is a code path to detect a if the composite extension exists (and xrender and xfixes) etc. and falls back to shaped if it doesn't... it also checks to see if the parent window is argb or not on creation to ensure it matches a parent if done that way - for root window it's special and doesn't enforce it but if it's requested it'll work fine..