[cairo] cairo-glesv2 and pango: CAIRO_STATUS_DEVICE_ERROR
psychon at znc.in
Fri Jul 5 15:08:17 UTC 2019
Just a quick reply for the mailing list archives: This was now reported
(who has no clue about GL and has his local cairo built without GL support)
On 01.07.19 14:01, Christopher Pahl wrote:
> Hello dear cairo devs,
> I need your help troubleshooting what possibly could be a bug in cairo.
> I'm developing an embedded UI that uses cairo for drawing its contents and
> pango for layouting text.
> For hardware acceleration I'm planning to use the cairo-gl surface backend
> (--enable-glesv2 to be exact), since that
> is supported well on the target hardware (RaPi 3b+) and showed promising
> results overall on first tests (big kudos for that!)
> The UI will run inside a weston compositor, as native wayland client, using
> EGL to get a OpenGL ES 2.0 context.
> Sadly, I get pretty weird behavior related to font rendering. There are
> three possible states:
> - The UI totally works. This was mostly the case only when development
> - The UI refuses to start, every call to cairo_show_text() or
> - The UI starts, but everything drawn before the text calls is blacked out,
> only the elements drawn after are visible.
> I tried to put a minimal example of this here:
> The behavior is the same on my development machine and on the target
> hardware. Now, here's the catch:
> If I compile cairo with --enable-glesv3, everything works perfectly fine on
> my machine (except that glesv3 does not make sense on the Pi...)
> I'm happy to provide more data if required, but I'm a bit stuck here on
> what to try. It would be very nice if you could push me in the right
> For reference:
> - both cairo 1.16 and latest cairo master from git show the same issue
> - I'm using weston 6.0.0
> - I'm using pango 1.43.0 and freetype 2.10.0
> Best regards,
"In the beginning the Universe was created. This has made a lot of
people very angry and has been widely regarded as a bad move."
More information about the cairo