Things that killed my motivation to play with wayland

[darxus at chaosreigns.com]

On 06/12, sardemff7+wayland at sardemff7.net wrote:
> On 12/06/2013 02:49, darxus at chaosreigns.com wrote:
> >1) This mesa bug:  https://bugs.freedesktop.org/show_bug.cgi?id=61919

> I provided a patch on this one, and even if I didn’t test it, it

Thank you.  My first step in testing it was attempting to reproduce the
error, which failed, so I closed the bug.  I'd guess due to upgrading
my linux distro.

I now have my automated build testing running again.  Only needed to
remove one repo (libXi - probably just needs more updated deps) from it
to get it working, which is kind of surprising.

> >2) The XWayland segfault bug:
> >https://bugs.freedesktop.org/show_bug.cgi?id=59983

> I just rebased the full xwayland branch to xorg-server master, and
> also the intel and wlshm DDX.
> 
> http://git.sardemff7.net/wayland/xorg-server/log/?h=xwayland-1.15
> http://git.sardemff7.net/wayland/xf86-video-intel/log/?h=xwayland
> http://git.sardemff7.net/wayland/xf86-video-wlshm/

Cool.  

> If Tiago can point me to the fix commit, I can include it in there,
> so we can start using a new base to hack on.

I got the impression that it would only work with some substantial changes
he made, which krh decided not to use (splitting xwm out to a separate
process).

> Tests are good, but if all of them fail badly, you *can* ignore
> them. In this case, they should be fixed, but that shouldn’t prevent
> you from using cairo. IMO, you should keep trying to get the tests
> fixed, without forcing them in your test script.

That's exactly what I did.  I just have running tests disabled for things
where it doesn't pass.  It's just another thing that really frustrated me.

> I’ll try to support you and provide patches on cairo too, please
> point me to the relevant bugs.

Thanks.


On 06/12, Tiago Vignatti wrote:
> I just sent the fix to the mailing list. I don't consider that patch
> enough though cause it exposes now another bug in XWayland
> infrastructure.

Thanks.


On 06/14, Uli Schlachter wrote:
> I don't think that it will ever be possible to have the test suite succeed for
> all tests on e.g. cairo-gl or cairo-xlib, because the results depend too much
> on the graphics driver and are too different across different drivers. Also,
> AFAIK cairo-pdf/cairo-ps often hit ghostscript bugs. :-(

Understandable.  But with this one bug fixed, I can at least get all of,
what was it, the image tests to pass, on the software X server recommended
in the instructions.  

> At the very least, could you write up the approach you came up with?

Sure.  

> I remember that I helped with some magic which didn't compare the test results
> against the expected results, but instead compared them against the results of
> the previous commit. No idea about the details, so it would be nice to be able
> to say "here, this mail describes one possible approach".

Yup.

> I just sent said patch to the cairo mailing list. If no one objects, I will
> just push it to fdo next week. Hopefully, Chris will instead show up and
> figure out what the right thing is (I guess he has a reason for dumping the
> patch into Bugzilla and not into Git, but I don't know that reason).

Nice, thanks.  (Saw some of that.)

-- 
"Hello, babies. Welcome to Earth. It's hot in the summer and cold in
the winter. It's round and wet and crowded. At the outside, babies,
you've got about a hundred years here. There's only one rule that I
know of, babies—God damn it, you've got to be kind." - Kurt Vonnegut
http://www.ChaosReigns.com