Re: Log for attempted build of webkit2gtk_2.6.2+dfsg1-4 on m68k (dist=unstable)

[Thorsten Glaser <tg@mirbsd.de>]

Alberto Garcia dixit:

>> CMake Error at CMakeLists.txt:109 (message):
>>   Unknown CPU 'm68k'
>
>That architecture is not listed in CMakeLists.txt, that's why it
>fails.
>
>I guess we could try to put it there and see if it builds, is there a

Indeed, but Webkit usually needs quite some amount of patching,
see the bugreports to both qtwebkit and webkit-gtk for those.
Some of it is atomics-related (we do have GCC atomics nowadays,
but only up to 32 bit values), most is making bogus implicit
alignment assumptions explicit by introducing explicit padding
members into various structures (which do not change the struct
anywhere else, but m68k needs less padding than other arches).

>machine where I (or someone else) can test it?

You can get yourself a VM: https://wiki.debian.org/Aranym/Quick

Thanks,
//mirabilos
-- 
<igli> exceptions: a truly awful implementation of quite a nice idea.
<igli> just about the worst way you could do something like that, afaic.
<igli> it's like anti-design.  <mirabilos> that too… may I quote you on that?
<igli> sure, tho i doubt anyone will listen ;)