[MPlayer-users] Re: Bug report: segmentation fault while playing/encoding a dvd with the -ss option

[Aym/Wedge]

> > I use the -ss option in addition of the -endpos option. With mostly
> > 75% of the dvds, there is no problem, it works fine. But for the 25%
> > left, it ends brutally in a segfault :(.
>
> Yes, PTS resetting to 0 is a known issue.  You can try to use -sb
> instead, or -edl (and probably -hr-edl-seek) with an edl file.

OK, I'll check this, thanx :)

> > [WORKS] mencoder dvd://1 -dvd-device /dev/sr0 -aid 128 -oac mp3lame
> > -lameopts abr:br=128 -ovc xvid -xvidencopts bitrate=850 -vf
> > pp=de,scale=720:-2 -o test.avi -ss 3180 -endpos 20
>
> You really shouldn't postprocess before encoding, it will just waste
> bits for little or nothing.
>
> And scale "-2" won't ensure dimentions that are multiples of 16.

As I don't pretend to be an encoding guru, I did it the simpliest way
to me and to my end-users. My script is designed to users who actually
don't know how to rip a dvd, and want to do it by just answering
*simple* questions.
For the scale option, I've added a -aspect option in my script wich is
designed to replace a failed scale :)

> > This is also a bit annoying because I use the -ss option when the
> > movie is too long to fit in a single 700 Mo cd. I need to split the
> > encoding into parts, and it generally fails at the beginning of the
> > second part of the encoding.
>
> Better to encode to one big file first, and split afterwards
> (preferably with avisplit rather than mencoder)

This was the way I did in previous releases, but for I got audio
issues, space waste (needs twice the size of the final movie ... bad)
and dependancies issues (avisplit is a part of transcode, which is not
very easy to install), I decided to do it only with mencoder and
lsdvd.

> > MPEG Stream reached EOF
> > ds_fill_buffer: EOF reached (stream: video)
> >
> > Flushing video frames
> > Writing index...
> > Writing header...
> > ODML: Aspect information not (yet?) available or unspecified, not
> > writing vprp header.
> >
> > Video stream:      nan kbit/s  (-2147483648 B/s)  size: 0 bytes  0.000
> > secs  0 frames
> >
> > Audio stream:      nan kbit/s  (-2147483648 B/s)  size: 0 bytes  0.000
> > secs Uninit audio filters...
> > [libaf] Removing filter dummy
> > Uninit audio: liba52
> > Uninit video: libmpeg2
>
> That obviously didn't work as intended, but I don't see a segfault.

Well, I though obvious that the segfault was at the end, and it's
true, I didn't put it in the log ...

> >
> >SSE2 supported but disabled
> >libdvdread: Using libdvdcss version 1.2.9 for DVD access
> >
> >libdvdread: Attempting to retrieve all CSS keys
> >libdvdread: This can take a _long_ time, please be patient
> >
> >
> >
> >
> you removed part of the log.

I don't think so ... I didn't copy/paste the log, I've redirected it
into a file using the ">" command, how could some part of the log have
disappear ?

> >Flushing video frames
> >Writing index...
> >Writing header...
> >ODML: Aspect information not (yet?) available or unspecified, not writing vprp header.
> >
> >Video stream:      nan kbit/s  (-2147483648 B/s)  size: 0 bytes  0.000 secs  0 frames
> >
> >Audio stream:      nan kbit/s  (-2147483648 B/s)  size: 0 bytes  0.000 secs
> >Uninit audio filters...
> >[libaf] Removing filter dummy
> >Uninit audio: liba52
> >Uninit video: libmpeg2
> >
> >
> what segfault??
> In any case, use a fresh svn checkout, with which -ss to work reliably
> in dvd streams

As I said higher, the segfault appears just after.
I can't use svn for the simpliest reason that I don't have access to
the net at home, and a restricted access at work.