Bug 77533 – xine-lib does not compile in support for xxmc with vld extensions ( for cle266 )

• Product: Gentoo Linux
• Status: RESOLVED TEST-REQUEST

[Description Jörg Sommersberg 2005-01-11 08:38:22 UTC]

when emerging xine-lib with USE="xvmc" it does not compile in the support for the cle266 graphics on epia boards.
although this is in principle possible with the new xine-lib-1.0.
has to be a problem of the ebuild

Reproducible: Always
Steps to Reproduce:
1.USE="xvmc"
2.emerge xine-lib
3.

Actual Results:  
xine-lib compiles but without xvmc ( respective xxmc ) support.

Expected Results:  
compile with xvmc support ( resp. xxmc ).


See also this bug
http://bugs.gentoo.org/show_bug.cgi?id=71099


Tried to fix it according this attchment
http://bugs.gentoo.org/attachment.cgi?id=43889

But the code in ebuild for xine-lib-1 to set tha vars for the xvmc libraries is
too complex for me.

[Comment 1 Jörg Sommersberg 2005-01-12 05:14:49 UTC]

Created attachment 48288 [details]
Suggestion for a patch of the ebuild

OK it does compile in the support for xxmc.
But the ebuild does complain if "nvidia" and "i8x0" are both not set.
In case someone is emerging xine-lib who has the XvMC-libary from the unichrome
project installed ( and of course the XvMC-wrapper ) this is unnecessary.
In that case it compiles and links properly against the installed libraries on
the system.

My suggestion :
Change the ebuild to look for the "cle266" USE-Flag and not complain if it is
set.

[Comment 2 Jörg Sommersberg 2005-01-12 05:18:36 UTC]

Ebuild warning is confusing for users that have the xvmc form the unichrome project and dont need "i8x0" or "nvidia".

For these poeple the warning from the ebuild makes no sense.

[Comment 3 Jörg Sommersberg 2005-01-14 05:26:58 UTC]

Created attachment 48449 [details]
Proposed Patch as diff output

Reopened that bug because i am not sure if my input will be recognized.
If that action is not adequate please tell me, so i can act accordingly next
time

Also attached the proposed patch as diff file for convenience.

Regards

J

[Comment 4 Jörg Sommersberg 2005-01-14 05:26:58 UTC]

Created attachment 48449 [details]
Proposed Patch as diff output

Reopened that bug because i am not sure if my input will be recognized.
If that action is not adequate please tell me, so i can act accordingly next
time

Also attached the proposed patch as diff file for convenience.

Regards

Jörg