Comment 3 for bug 56153

> On Tuesday 23 October 2007 04:28:15 am you wrote:
>> About that special patch, did you have asked upstream for their opinion
>> on this patch?

Jonathan Nelson <email address hidden> writes:
> I posted a patch on the xine bug tracking system over a year ago to
> fix this problem [1]. ... I'm assuming that the xine authors consider
> it either a trivial issue, or just haven't noticed the bug reports
> yet.

I've just talked on IRC with one xine developer (#xine on Freenode):

14:55:50 < siretart> _ds_: how do you think about the patch in https://bugs.edge.launchpad.net/ubuntu/+source/xine-lib/+bug/56153. any reason not to apply it upstream?
14:56:28 < siretart> _ds_: I noticed that a xine user has tried to upload a xine package with that patch to ubuntu (but obviously failed because of his gpg key not in the keyring)
14:56:55 < siretart> I asked him whats up and he said that he wasn't
able to gain any feedback from 'xine upstream'
15:28:49 < _ds_> I don't like that patch - it affects all audio.
15:41:25 < siretart> but it seems to fix problems. hmm
16:44:03 < siretart> what bad could happen with that patch?
16:45:58 < _ds_> It's basically excessive memory allocation (consider also the number of buffers). The proper fix for this should be in the Vorbis decoder.
16:48:02 < siretart> ok. thanks for explanation

I have to admit that I agree to upstream here, and I don't see why we
should carry around a bad patch in ubuntu and/or debian.

> I've been using a patched version of xine-lib since noticing the second bug
> report, and haven't had any issues with xine.

Did you measure the increased memory overhead? Please show some numbers.

> I'm afraid I don't really know the correct procedure for doing this

You did okay: Talking to the relevant parties.

--
Gruesse/greetings,
Reinhard Tartler, KeyID 945348A4