GSoC project ideas for improving LOs core-code / performance

[Michael Stahl]

On 24.02.2018 00:57, Tomaž Vajngerl wrote:
> On Fri, Feb 23, 2018 at 3:48 PM, Clemens Eisserer <linuxhippy at gmail.com> wrote:

>> The X11 backend has certain issues (e.g. reading back huge images via
>> XGetImage), the GTK3 backend performs all rendering via software and
>> the OpenGL backend on Linux lacks certain features which are crucial
>> to achieve acceptable performance (e.g. accelerated text rendering).
> 
> I have added the idea to improve OpenGL backend, which I'm also
> willing to mentor. For GTK3 - Caolan is the right guy who knows what
> could be improved there performance wise. But OTOH any X11 are
> probably pointless as the idea AFAIK is to get rid of those backends
> in the long run, so I'm not sure who would even be interested in
> mentoring that. :)

the gen X11 backend is useful as a fail-safe fallback when the other
ones don't work; i've got bad experiences running some of the bibisect
repos on Fedora 27 where gen is the only one that doesn't crash on
startup, and i guess one of the reasons for that is that ~nobody does
changes in gen or the system's X11 libs nowadays.

so i'm mostly interested in preserving gen as-is and not adding fancy
new stuff with unknown compatibility issues to it...