Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: MythTV: Dev

mythtv-vid branch needs testing

 

 

First page Previous page 1 2 3 Next page Last page  View All MythTV dev RSS feed   Index | Next | Previous | View Threaded


danielk at cuymedia

Aug 1, 2007, 12:28 PM

Post #1 of 69 (3971 views)
Permalink
mythtv-vid branch needs testing

Aside from a memory leak it looks like most of the OSX issues have
been addressed. A few of us have been using this branch for a while,
but it would be good to get some people to test this before we merge
it into the trunk.

Instead of only being able to pick one video decoder, one video
rendering method, one OSD rendering method and one deinterlacing
method at a time, the mythtv-vid branch allows you to configure
different methods depending on the resolution of the video.
You can also chose the fallback deinterlacer to use if the first
once can't be used; useful if you want to fallback to kernel
deinterlacing or linear blend when bob deinterlacing fails.

There are also some additional rendering methods: For example, OpenGL
rendering, which uses OpenGL 2.0 instead of XVideo for color conversion,
scaling and compositing. Another one is XvMC-OpenGL rendering which
uses an nVidia extension to allow XvMC decoded frames to be rendered
using OpenGL with OpenGL composited OSD. Both methods require greater
CPU and GPU resources, but can result in a better looking video. The
XvMC-OpenGL rendering method will not be compiled in unless you link
directly to the XvMCNVIDIA library using the --xvmc-lib ./configure
parameter.

Three video profiles groups are defined by default "CPU++", "CPU+",
"CPU--". The first "CPU++" is the default for all frontends and
just uses ffmpeg and the old default video rendering method. The
second "CPU+" uses ffmpeg and XVideo for SDTV resolutions and then
degrades to the lower output quality XvMC and libmpeg2 decoders for
higher resolutions. Finally, "CPU--" tries to use hardware decoding
whenever possible and falls back to libmpeg2 with XVideo. The profile
groups are global for all your for all your frontends, but you can
configure which one to use on any particular frontend and create new
ones if you wish. The first matching rule which is usable with your
hardware is used, so you can have ivtv configured even if you don't
have the hardware. If no matching rule is found MythTV will pick
a decoder and rendering methods which will work, but of course these
may not be ideal for your particular hardware.

-- Daniel

_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


freedenizen at gmail

Aug 1, 2007, 12:52 PM

Post #2 of 69 (3869 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

The main thing I have is the lack of overscan with the opengl
renderer. Some of the videos have the vbi info at the top and it is
pretty distracting to have it there.

I also noticed a lot of prebuffer pauses when playing back HD content.
I set it to just use regular XvMC with (I believe) the same settings
that I use for trunk without problem.

I don't have my logs accessible right now but there were messages in
them stating something along the lines of "deinterlacer 'xxxxx' not
active' It didn't matter which deinterlacer I chose, even none it
would say that.

All that said, the video quality does look better for SD shows, and I
love the full sized OSD displays
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


xanium4332 at googlemail

Aug 2, 2007, 8:04 AM

Post #3 of 69 (3861 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On 01/08/07, Asher <freedenizen [at] gmail> wrote:
>
> The main thing I have is the lack of overscan with the opengl
> renderer. Some of the videos have the vbi info at the top and it is
> pretty distracting to have it there.
>
> I also noticed a lot of prebuffer pauses when playing back HD content.
> I set it to just use regular XvMC with (I believe) the same settings
> that I use for trunk without problem.
>
> I don't have my logs accessible right now but there were messages in
> them stating something along the lines of "deinterlacer 'xxxxx' not
> active' It didn't matter which deinterlacer I chose, even none it
> would say that.
>
> All that said, the video quality does look better for SD shows, and I
> love the full sized OSD displays


I've been running the branch for a while, and am currently in the process of
creating some ubuntu debs (for easy install). I'll setup a repository
hopefully, which would let people easily install + test them. The biggest
problem at the moment is the plugins, as they've changed quite a bit since
the 0.20 release (which the current debs are based on).

As a side note, using some of the more exotic deinterlacers, such as linear
blend/kernel (2x, hardware), when the video is panning, or an object moves
quickly, 'shimmering' (best way to describe it) occurs between colour
differences. I guess it occurs across the whole image, but is only
noticeable between 2 different colours (e.g. white and red). Just want to
check that it is a known attribute of the deinterlacer, and not a
configuration problem. Even still, it's nice to have no 'flicker' issues
created by BOB, yet still keep the framerate.

Is there any timescale for the merge, within 6 months, a year?

Regards, James


hendrixski at storsint

Aug 2, 2007, 9:36 AM

Post #4 of 69 (3855 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

James Buckley wrote:
> On 01/08/07, *Asher* <freedenizen [at] gmail
> <mailto:freedenizen [at] gmail>> wrote:
> I've been running the branch for a while, and am currently in the
> process of creating some ubuntu debs (for easy install). I'll setup a
> repository hopefully, which would let people easily install + test
> them. The biggest problem at the moment is the plugins, as they've
> changed quite a bit since the 0.20 release (which the current debs are
> based on).
>

Hey, that's a pretty good idea. Make .debs for people to be able to
easily install the test version. that lowers the barrier to entry for
people to become testers, and they can get the latest trunk through
automatic updates every week or so without having to think about it.

I've been meaning to contribute to mythtv but got sidetracked with
debian packaging for a little while, so now I know the bare basics of
how to make an Ubuntu .deb... coincidence? If you need a partner in
crime, shoot me an email.
- hendrixski
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


danielk at cuymedia

Aug 2, 2007, 2:29 PM

Post #5 of 69 (3844 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On Wed, 2007-08-01 at 12:52 -0700, Asher wrote:
> The main thing I have is the lack of overscan with the opengl
> renderer. Some of the videos have the vbi info at the top and it is
> pretty distracting to have it there.

Mark Kendal is working on a patch for this.

> I also noticed a lot of prebuffer pauses when playing back HD content.
> I set it to just use regular XvMC with (I believe) the same settings
> that I use for trunk without problem.

Which settings are those? chromakey or ia44 OSD? Which deinterlacer?
What XvMC hardware are you using? VIA, nVidia, other?

If you use the OpenGL OSD with XvMC then you will of course have
much worse performance, but you would know if you were using this
OSD renderer because the OSD would look much better.

> I don't have my logs accessible right now but there were messages in
> them stating something along the lines of "deinterlacer 'xxxxx' not
> active' It didn't matter which deinterlacer I chose, even none it
> would say that.

I would need to see the logs. Bob deint will only work if the refresh
rate is high enough, and the 'none' deinterlacer will never load. I'll
make sure we don't print anything confusing to the log in this case.

> All that said, the video quality does look better for SD shows, and I
> love the full sized OSD displays
:)

-- Daniel

_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


freedenizen at gmail

Aug 2, 2007, 2:36 PM

Post #6 of 69 (3838 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On 8/2/07, Daniel Kristjansson <danielk [at] cuymedia> wrote:
> On Wed, 2007-08-01 at 12:52 -0700, Asher wrote:
> > I also noticed a lot of prebuffer pauses when playing back HD content.
> > I set it to just use regular XvMC with (I believe) the same settings
> > that I use for trunk without problem.
>
> Which settings are those? chromakey or ia44 OSD? Which deinterlacer?
> What XvMC hardware are you using? VIA, nVidia, other?

I will try to test tonight and get you all the info, different setting
etc. I did notice that it wasn't a cpu limitation, usage was around
40% so I'm thinking it is gpu limited, only running a 6600GT

> > I don't have my logs accessible right now but there were messages in
> > them stating something along the lines of "deinterlacer 'xxxxx' not
> > active' It didn't matter which deinterlacer I chose, even none it
> > would say that.
>
> I would need to see the logs. Bob deint will only work if the refresh
> rate is high enough, and the 'none' deinterlacer will never load. I'll
> make sure we don't print anything confusing to the log in this case.

ok, so I'm guessing 60Hz isn't high enough? Unfortunately that is all
my display will do. I think I tried linearblend and kernel deint and
was getting the same message but once again I will test later and get
logs, do you want them with -v playback?
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


danielk at cuymedia

Aug 2, 2007, 2:47 PM

Post #7 of 69 (3847 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On Thu, 2007-08-02 at 14:36 -0700, Asher wrote:
> On 8/2/07, Daniel Kristjansson <danielk [at] cuymedia> wrote:
> > On Wed, 2007-08-01 at 12:52 -0700, Asher wrote:
> > I would need to see the logs. Bob deint will only work if the refresh
> > rate is high enough, and the 'none' deinterlacer will never load. I'll
> > make sure we don't print anything confusing to the log in this case.
>
> ok, so I'm guessing 60Hz isn't high enough? Unfortunately that is all
> my display will do. I think I tried linearblend and kernel deint and
> was getting the same message but once again I will test later and get
> logs, do you want them with -v playback?

60Hz would be enough for NTSC playback without time-stretch, but
it needs to be >= twice the framerate. Sometimes its just a few
milliseconds under that and fiddling with the modelines can make
it work (there is lots of info on this in the mythtv users mailing
list archive).

Linear blend and kernel deinterlacing should work for any of the
software decoders, while XvMC only accepts bob and one-field
deinterlacing. The video profiles editor should only allow you to
select deinterlacers which at least can be valid when the moons
are aligned. If it is allowing you to configure kernel deint for
XvMC that is a bug.

-- Daniel

_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


danielk at cuymedia

Aug 3, 2007, 12:24 PM

Post #8 of 69 (3803 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On Thu, 2007-08-02 at 16:04 +0100, James Buckley wrote:
> On 01/08/07, Asher <freedenizen [at] gmail> wrote:


> I've been running the branch for a while, and am currently in the
> process of creating some ubuntu debs (for easy install). I'll setup a
> repository hopefully, which would let people easily install + test
> them. The biggest problem at the moment is the plugins, as they've
> changed quite a bit since the 0.20 release (which the current debs are
> based on).

Please don't make any debs public. I don't want bug reports from
people can't compile from the source. I need testers to be able
to test patches and have the most up to date version of the branch.

> Is there any timescale for the merge, within 6 months, a year?

More like a week or two, depending on bug reports.

-- Daniel


_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


msc at antzsystem

Aug 3, 2007, 12:42 PM

Post #9 of 69 (3811 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

Am Mittwoch, 1. August 2007 schrieb Daniel Kristjansson:
> Aside from a memory leak it looks like most of the OSX issues have
> been addressed. A few of us have been using this branch for a while,
> but it would be good to get some people to test this before we merge
> it into the trunk.

sounds nice, i will try it again.

Is there something new regarding scaling (high-quality software scaler?)
or full-framerate rendering?

Is there a list with all new/improved features besides your posting?


--
Markus Schulz
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


hendrixski at storsint

Aug 3, 2007, 1:16 PM

Post #10 of 69 (3810 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

:-) There would be two ways to ensure testers get the most up-to-date
patches via debian packages.

1) Make a script that compiles and debianizes the latest svn every night
then posts it on the repository (not advisable).

2) Make a script that converts the patches to dpatches and inserts them
into the deb, then post that deb on the repository every night. (this
way if someone suspects that a patch caused more bugs, they can just
remove one line of code from debian/patches/00patch, run debuild -S, run
dpkg -i mythtv-latest.deb, and they've uninstalled the patch, and can
test run without it).

Would that increase the number of people who can test the latest, most
up-to-date patches? After all, with enough eyeballs all bugs are shallow.

- hendrixski

Daniel Kristjansson wrote:
>> I've been running the branch for a while, and am currently in the
>> process of creating some ubuntu debs (for easy install). I'll setup a
>> repository hopefully, which would let people easily install + test
>> them. The biggest problem at the moment is the plugins, as they've
>> changed quite a bit since the 0.20 release (which the current debs are
>> based on).
>>
>
> Please don't make any debs public. I don't want bug reports from
> people can't compile from the source. I need testers to be able
> to test patches and have the most up to date version of the branch.
>

_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


danielk at cuymedia

Aug 3, 2007, 1:37 PM

Post #11 of 69 (3823 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On Fri, 2007-08-03 at 16:16 -0400, hendrixski [at] storsint wrote:
> :-) There would be two ways to ensure testers get the most up-to-date
> patches via debian packages.
>
> 1) Make a script that compiles and debianizes the latest svn every night
> then posts it on the repository (not advisable).
>
> 2) Make a script that converts the patches to dpatches and inserts them
> into the deb, then post that deb on the repository every night. (this
> way if someone suspects that a patch caused more bugs, they can just
> remove one line of code from debian/patches/00patch, run debuild -S, run
> dpkg -i mythtv-latest.deb, and they've uninstalled the patch, and can
> test run without it).
>
> Would that increase the number of people who can test the latest, most
> up-to-date patches? After all, with enough eyeballs all bugs are shallow.

I don't want everyone to test a patch for tester X, just tester X.
When I write a debug patch it may very well break important things,
such as playback of video, but it will help reveal the problem that
the tester experienced. I may at the same time have a patch for
tester Y and and another for tester Z, none of which are compatible.
If someone can't be bothered to compile MythTV then I'm afraid they
are unlikely to put in the time required to figure out the cause of
a bug so it can be fixed, nor are they likely to even know how to
apply a patch. When someone tells me they applied a patch and got
a certain result, when in fact they failed to apply the patch or
failed to compile or failed to install the updated software, then
this sends me looking for the problem elsewhere, wasting time that
could be spent solving problems.

-- Daniel

_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


jppoet at gmail

Aug 3, 2007, 9:03 PM

Post #12 of 69 (3825 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On 8/1/07, Daniel Kristjansson <danielk [at] cuymedia> wrote:
>
> Aside from a memory leak it looks like most of the OSX issues have
> been addressed. A few of us have been using this branch for a while,
> but it would be good to get some people to test this before we merge
> it into the trunk.

I have been using this for the last few of days. In general it looks
very good. In that time I have had mythfrontend crash twice when
starting to play a show, but I cannot force it to happen, and it may
not be related.

Using OpenGL when playing a standard-def show, right at the beginning
it plays fine for a couple of seconds, then has a quick pause, then
continues just fine. It does not really bother me, since it only
happens right at the beginning.

I can even use OpenGL for 720p shows, as long as I turn of OSD fade.
However, using "time stretch" results in pretty jerky playback.

OpenGL with 1080i is jerky.

My frontend has a E6600 and a 7300GT.

John
--
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


freedenizen at gmail

Aug 4, 2007, 9:57 AM

Post #13 of 69 (3788 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

Of course now when I try to do testing everything I was mentioning
seems to be working.

when moving to the -vid branch from trunk if I don't recompile the
plugins, myth fails to load, but going the other way (from -vid to
trunk) the plugins are just disabled if they aren't recompiled. Is it
possible for the -vid branch to have the same behavior?

I did see it crashing too, I need to run it through gdb and see if I
can get something more meaningful.
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


freedenizen at gmail

Aug 4, 2007, 11:06 AM

Post #14 of 69 (3784 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

I attached a log and backtrace from a crash. I started playing the
video and a second into it I hit fast forward, crash was pretty
immediate after that.
Attachments: gdb.txt (6.69 KB)
  vid_ff_crash.log (14.1 KB)


xanium4332 at googlemail

Aug 4, 2007, 12:28 PM

Post #15 of 69 (3800 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On 03/08/07, Daniel Kristjansson <danielk [at] cuymedia> wrote:
>
> On Fri, 2007-08-03 at 16:16 -0400, hendrixski [at] storsint wrote:
> > :-) There would be two ways to ensure testers get the most up-to-date
> > patches via debian packages.
> >
> > 1) Make a script that compiles and debianizes the latest svn every night
> > then posts it on the repository (not advisable).
> >
> > 2) Make a script that converts the patches to dpatches and inserts them
> > into the deb, then post that deb on the repository every night. (this
> > way if someone suspects that a patch caused more bugs, they can just
> > remove one line of code from debian/patches/00patch, run debuild -S, run
> > dpkg -i mythtv-latest.deb, and they've uninstalled the patch, and can
> > test run without it).
> >
> > Would that increase the number of people who can test the latest, most
> > up-to-date patches? After all, with enough eyeballs all bugs are
> shallow.
>
> I don't want everyone to test a patch for tester X, just tester X.
> When I write a debug patch it may very well break important things,
> such as playback of video, but it will help reveal the problem that
> the tester experienced. I may at the same time have a patch for
> tester Y and and another for tester Z, none of which are compatible.
> If someone can't be bothered to compile MythTV then I'm afraid they
> are unlikely to put in the time required to figure out the cause of
> a bug so it can be fixed, nor are they likely to even know how to
> apply a patch. When someone tells me they applied a patch and got
> a certain result, when in fact they failed to apply the patch or
> failed to compile or failed to install the updated software, then
> this sends me looking for the problem elsewhere, wasting time that
> could be spent solving problems.
>
> -- Daniel


I see your point on the debs, however I personally like the helper scripts
and etc.. that come with the ubuntu ones (a year on Gentoo has worn away at
my 'configure-everything-yourself' section of my brain, hopefully it should
regenerate though :P). There's also no point if it's going to get merged so
soon, I thought it the -vid branch was going to be around for a while longer
(like Compiz Fusion, where at the moment you can get daily builds for the
masses).


doug at parkercat

Aug 4, 2007, 12:53 PM

Post #16 of 69 (3796 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

First, a minor compilation issue:

GL_FRAMEBUFFER_INCOMPLETE_DUPLICATE_ATTACHMENT_EXT is no longer
defined in Debian's /usr/include/GL/glext.h version 7.0.1~rc2-1 (from
experimental archive)... see
http://www.nabble.com/Missing-entry-in-GL_EXT_framebuffer_object-t2029545.html
for the likely reason.

Commenting this out (in openglcontext.cpp) allows build to complete, as
it's just used in outputting possible error status.

A few brief testing results...

1. My Intel-G965-video system (desktop/dev system) doesn't have recent
enough GLX to run OpenGL video out -- has 1.2, needs 1.3. Too bad -- I
was hoping to test out OpenGL video rendering on this modern graphics
hardware w/ good open-source drivers. OTOH from what I've read GLX 1.3
is slated for xorg 1.5 (~November 2007). Normal Xv output is working
fine on this system (except for tearing at HD resolutions, but that's a
driver issue). This somewhat beefy system (Core 2 Duo 2.4 GHz) lacks
the CPU horsepower to do kernel deint at 1080i resolution -- pegs one
core; mythfrontend must be killed.

2. My main Myth system (nVidia FX 5200) lacks the GPU horsepower to do
OpenGL output of even standard-def material with any deinterlacing
turned on -- though it's close. Switching into Timestretch (to slow
things down) hard-hung the system requiring power cycle. This may also
have been trying to fall back to kernel deint, though the hang killed my
database and I had to restore it, so I can't check. I will test with an
nVidia 6200 card as well once it's not 90 degrees in Boston and I can
bear to do work on that system.

-Doug
Attachments: signature.asc (0.25 KB)


doug at parkercat

Aug 5, 2007, 6:12 AM

Post #17 of 69 (3790 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

Doug Larrick wrote:
> 2. My main Myth system (nVidia FX 5200) lacks the GPU horsepower to do
> OpenGL output of even standard-def material with any deinterlacing
> turned on -- though it's close. Switching into Timestretch (to slow
> things down) hard-hung the system requiring power cycle. This may also
> have been trying to fall back to kernel deint, though the hang killed my
> database and I had to restore it, so I can't check. I will test with an
> nVidia 6200 card as well once it's not 90 degrees in Boston and I can
> bear to do work on that system.

The 6200 has the horsepower to do standard decode & OpenGL video out for
standard-def material. Looks good.

I am seeing a semi-reproducible system hang (no response to keypresses
or ssh) when switching between timestretch value of 1 and 1.05 -- which
presumably is switching OpenGL deinterlace methods. I was using Bob
(2x,HW) as primary and Onefield (HW) as fallback.

-Doug
Attachments: signature.asc (0.25 KB)


mark.kendall at gmail

Aug 6, 2007, 3:59 AM

Post #18 of 69 (3768 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On 8/5/07, Doug Larrick <doug [at] parkercat> wrote:
> I am seeing a semi-reproducible system hang (no response to keypresses
> or ssh) when switching between timestretch value of 1 and 1.05 -- which
> presumably is switching OpenGL deinterlace methods. I was using Bob
> (2x,HW) as primary and Onefield (HW) as fallback.

Doug

I can't reproduce this on my system. Can you catch some logs so I can
see what's going on?

Thanks and regards

Mark
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


mark.kendall at gmail

Aug 6, 2007, 5:49 AM

Post #19 of 69 (3752 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On 8/6/07, Mark Kendall <mark.kendall [at] gmail> wrote:
> I can't reproduce this on my system. Can you catch some logs so I can
> see what's going on?

Actually I can :)

Looks like the same issue for Asher and Doug. When FallbackDeint is
called, for some reason we can't make the ogl context current. For
Asher and Doug this is fatal but not on my system...

I'll have another look later but can't see any obvious causes.

Regards

Mark
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


doug at parkercat

Aug 6, 2007, 4:47 PM

Post #20 of 69 (3739 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

Mark Kendall wrote:
> On 8/6/07, Mark Kendall <mark.kendall [at] gmail> wrote:
>> I can't reproduce this on my system. Can you catch some logs so I can
>> see what's going on?
>
> Actually I can :)
>
> Looks like the same issue for Asher and Doug. When FallbackDeint is
> called, for some reason we can't make the ogl context current. For
> Asher and Doug this is fatal but not on my system...
>
> I'll have another look later but can't see any obvious causes.

If it's any help I have OpenGL vsync and OpenGL menu painter turned on.
Though I expect that's a pretty common configuration.

nVidia version is Debian's 100.14.11-1.

I built Myth on a system with Intel graphics (and thus Xorg OpenGL
headers, not nVidia). Do you think it's worthwhile trying to build
directly on the target system?

-Doug
Attachments: signature.asc (0.25 KB)


Dibblahmythml0015 at pendor

Aug 8, 2007, 1:56 AM

Post #21 of 69 (3700 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

Daniel Kristjansson wrote:
> it would be good to get some people to test this before we merge
> it into the trunk.

Tested on a Athlon X2 4000, NVidia Geforce 6150 embedded
video. Works OK for CPU+. Appears to be too slow for the
opengl renderer.

There's a few issues with tabbing order on the Settings / TV
/ Playback UI - I may have time to look at these later.

On another system (effectively a Mac Mini - Core 2 Duo,
i845GM integrated video), I get a segfault which I'm not
managing to track down. This system definitely works with trunk.




Starting program: /usr/local/bin/mythfrontend
Failed to read a valid object file image from memory.
[Thread debugging using libthread_db enabled]
[New Thread 47035281451680 (LWP 1419)]
[New Thread 1082132800 (LWP 1424)]
[New Thread 1090525504 (LWP 1425)]
[Thread 1090525504 (LWP 1425) exited]
[New Thread 1090525504 (LWP 1426)]
[New Thread 1098918208 (LWP 1427)]
[New Thread 1107310912 (LWP 1428)]
[New Thread 1115703616 (LWP 1429)]

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1115703616 (LWP 1429)]
0x00002aaaaac65d20 in ?? ()

Thread 7 (Thread 1115703616 (LWP 1429)):
#0 0x00002aaaaac65d20 in ?? ()
No symbol table info available.
#1 0x00002ac73f1b960d in __glXInitialize ()
from //usr/lib64/opengl/xorg-x11/lib/libGL.so.1
No symbol table info available.
#2 0x00002ac73f1b4e59 in glXQueryVersion ()
from //usr/lib64/opengl/xorg-x11/lib/libGL.so.1
No symbol table info available.
#3 0x00002ac73c49902b in get_glx_version (XJ_disp=0x83fb90,
major=@0x4280258c, minor=@0x42802588) at util-opengl.cpp:83
ret = 1
errbase = 155
eventbase = 77
gl_major = 1061374025
gl_minor = 10951
tmp_disp = (Display *) 0x7ee7e0
#4 0x00002ac73c499a07 in OpenGLContext::IsGLXSupported
(display=0x83fb90,
min_major=1, min_minor=3) at openglcontext.cpp:567
major = 0
minor = 0
#5 0x00002ac73c4768df in SetFromHW (d=0x83fb90,
useXvMC=@0x4280288d,
useXVideo=@0x4280288c, useShm=@0x4280288b,
useXvMCOpenGL=@0x4280288a,
useOpenGL=@0x42802889) at videoout_xv.cpp:4114
glx_1_3 = 66
#6 0x00002ac73c476a32 in allowed_video_renderers
(myth_codec_id=kCodec_NONE,
XJ_disp=0x83fb90) at videoout_xv.cpp:4145
list = (QStringList &) @0x42802b90:
{<QValueList<QString>> = {
sh = 0x0}, <No data fields>}
vld = false
idct = false
mc = false
xv = true
shm = true
xvmc_opengl = false
opengl = true
list = {<QValueList<QString>> = {sh = 0x400}, <No
data fields>}
#7 0x00002ac73c485fed in VideoOutputXv::InitSetupBuffers
(this=0x7f2680)
at videoout_xv.cpp:1395
renderers = {<QValueList<QString>> = {sh = 0x0},
<No data fields>}
renderer = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x7f2b38,
static shared_null = 0x665da0}
tmp = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d =
0x3faaaa8f007f26e0,
static shared_null = 0x665da0}
use_xv = false
use_shm = false
use_opengl = false
ok = false
osdrenderer = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d =
0x44000000780,
static shared_null = 0x665da0}
#8 0x00002ac73c4893f6 in VideoOutputXv::Init
(this=0x7f2680, width=720,
height=576, aspect=1.33333004, winid=8389078, winx=0,
winy=0, winw=1280,
winh=1024, embedid=0) at videoout_xv.cpp:1511
No locals.
#9 0x00002ac73c436034 in VideoOutput::Create
(decoder=@0x42803a30,
codec_id=kCodec_NONE, codec_priv=0x0,
video_dim=@0x42803a00,
video_aspect=1.33333004, win_id=8389078,
display_rect=@0x428039f0,
embed_id=0) at videooutbase.cpp:149
it = {node = 0x7e6970}
vo = (VideoOutput *) 0x7f2680
renderers = {<QValueList<QString>> = {sh =
0x84a130}, <No data fields>}
xvlist = {<QValueList<QString>> = {sh = 0xdccf40},
<No data fields>}
renderer = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x82ae90,
static shared_null = 0x665da0}
#10 0x00002ac73c385382 in NuppelVideoPlayer::InitVideo
(this=0x2aaab25aedb0)
at NuppelVideoPlayer.cpp:578
widget = (struct QWidget *) 0x2aaab00738b0
video_dim = {wd = 720, ht = 576}
display_rect = {x1 = 0, y1 = 0, x2 = 1279, y2 = 1023}
db_scale = false
__PRETTY_FUNCTION__ = "bool
NuppelVideoPlayer::InitVideo()"
#11 0x00002ac73c3882a3 in NuppelVideoPlayer::StartPlaying
(this=0x2aaab25aedb0)
at NuppelVideoPlayer.cpp:3137
output_video = 50
decoder_thread = 46912625110448
#12 0x00002ac73c34bb88 in SpawnDecode (param=0x2aaab25aedb0)
at tv_play.cpp:460
decoder_thread_pool = (void *) 0x0
nvp = (class NuppelVideoPlayer *) 0x2aaab25aedb0
#13 0x00002ac73f747135 in start_thread () from
/lib/libpthread.so.0
No symbol table info available.
#14 0x00002ac73fd7e2ce in clone () from /lib/libc.so.6
No symbol table info available.
#15 0x0000000000000000 in ?? ()
No symbol table info available.

Thread 6 (Thread 1107310912 (LWP 1428)):
#0 0x00002ac73fd78322 in select () from /lib/libc.so.6
No symbol table info available.
#1 0x00002ac73d92ab33 in MythSocket::readyReadThread () at
mythsocket.cpp:743
it = {<QGListIterator> = {list = 0x2ac73db26540,
curNode = 0x0}, <No data fields>}
rval = 0
rfds = {fds_bits = {81920, 0 <repeats 15 times>}}
timeout = {tv_sec = 0, tv_usec = 4000}
sock = (MythSocket *) 0x0
maxfd = 16
found = false
#2 0x00002ac73f747135 in start_thread () from
/lib/libpthread.so.0
No symbol table info available.
#3 0x00002ac73fd7e2ce in clone () from /lib/libc.so.6
No symbol table info available.
#4 0x0000000000000000 in ?? ()
No symbol table info available.

Thread 5 (Thread 1098918208 (LWP 1427)):
#0 0x00002ac73fd511b1 in nanosleep () from /lib/libc.so.6
No symbol table info available.
#1 0x00002ac73fd78a94 in usleep () from /lib/libc.so.6
No symbol table info available.
#2 0x00002ac73c32cbad in TV::StartPlayer (this=0x2aaab001a650,
isWatchingRecording=false, maxWait=20000) at
tv_play.cpp:1396
t = {m_timer = {ds = 33419572}}
#3 0x00002ac73c33e787 in TV::HandleStateChange
(this=0x2aaab001a650)
at tv_play.cpp:1158
name = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x7c5730,
static shared_null = 0x665da0}
ok = false
changed = true
nextState = kState_WatchingLiveTV
desiredNextState = kState_WatchingLiveTV
lastState = -1321545224
#4 0x00002ac73c340280 in TV::RunTV (this=0x2aaab001a650) at
tv_play.cpp:1749
doHandle = true
keypressed = (QKeyEvent *) 0x2ac73bcabbc2
netCmd = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x7689d0,
static shared_null = 0x665da0}
updatecheck = 3
#5 0x00002ac73c3422a9 in TV::EventThread (param=0x2aaab001a650)
at tv_play.cpp:1710
thetv = (class TV *) 0x2aaab001a650
#6 0x00002ac73f747135 in start_thread () from
/lib/libpthread.so.0
No symbol table info available.
#7 0x00002ac73fd7e2ce in clone () from /lib/libc.so.6
No symbol table info available.
#8 0x0000000000000000 in ?? ()
No symbol table info available.

Thread 4 (Thread 1090525504 (LWP 1426)):
#0 0x00002ac73f74b917 in pthread_cond_timedwait@@GLIBC_2.3.2 ()
from /lib/libpthread.so.0
No symbol table info available.
#1 0x00002ac73e9d0c25 in ?? () from /usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#2 0x00002ac73e9d0d7e in QThread::msleep () from
/usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#3 0x00002ac73d8ea1da in MonitorThread::run (this=0x805650)
at mythmediamonitor.cpp:47
No locals.
#4 0x00002ac73e9d0f9b in QThreadInstance::start ()
from /usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#5 0x00002ac73f747135 in start_thread () from
/lib/libpthread.so.0
No symbol table info available.
#6 0x00002ac73fd7e2ce in clone () from /lib/libc.so.6
No symbol table info available.
#7 0x0000000000000000 in ?? ()
No symbol table info available.

Thread 2 (Thread 1082132800 (LWP 1424)):
#0 0x00002ac73f74b726 in pthread_cond_wait@@GLIBC_2.3.2 ()
from /lib/libpthread.so.0
No symbol table info available.
#1 0x00002ac73ec7f8b2 in QWaitCondition::wait ()
from /usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#2 0x00002ac73d80977f in MythContext::waitPrivRequest
(this=0x75c130)
at mythcontext.cpp:2884
No locals.
#3 0x000000000042be10 in run_priv_thread (data=0x0) at
main.cpp:719
No locals.
#4 0x00002ac73f747135 in start_thread () from
/lib/libpthread.so.0
No symbol table info available.
#5 0x00002ac73fd7e2ce in clone () from /lib/libc.so.6
No symbol table info available.
#6 0x0000000000000000 in ?? ()
No symbol table info available.

Thread 1 (Thread 47035281451680 (LWP 1419)):
#0 0x00002ac73fd511b1 in nanosleep () from /lib/libc.so.6
No symbol table info available.
#1 0x00002ac73fd78a94 in usleep () from /lib/libc.so.6
No symbol table info available.
#2 0x00002ac73c356ece in TV::StartTV (tvrec=0x0,
startInGuide=false,
inPlaylist=false, initByNetworkCommand=false) at
tv_play.cpp:185
freeRecorders = 3
tv = (class TV *) 0x2aaab001a650
quitAll = false
showDialogs = true
playCompleted = false
curProgram = (ProgramInfo *) 0x0
allowrerecord = false
deleterecording = false
#3 0x000000000042d547 in startTVNormal () at main.cpp:274
No locals.
#4 0x0000000000434f5a in TVMenuCallback (data=0x75c130,
selection=@0x2aaab0018c58) at main.cpp:308
sel = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d =
0x2aaab0022460,
static shared_null = 0x665da0}
#5 0x00002ac73dbb7d80 in MythThemedMenuPrivate::handleAction (
this=0x2aaab0018c00, action=@0x2aaab0029810) at
myththemedmenu.cpp:2212
No locals.
#6 0x00002ac73dbb833f in MythThemedMenuPrivate::keyHandler (
this=0x2aaab0018c00, actions=@0x7fff6ee0b950,
fullexit=true)
at myththemedmenu.cpp:1960
it = {node = 0x2aaab0029800}
action = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x850480,
static shared_null = 0x665da0}
i = 0
lastbutton = (ThemedButton *) 0x2aaab13a3d30
oldrow = 0
oldcolumn = 0
handled = true
#7 0x00002ac73dbb8b2e in
MythThemedMenuPrivate::keyPressHandler (
this=0x2aaab0018c00, e=0x7fff6ee0c070) at
myththemedmenu.cpp:1869
actions = {<QValueList<QString>> = {
sh = 0x2aaab1399900}, <No data fields>}
#8 0x00002ac73dbb8bbb in MythThemedMenu::keyPressEvent
(this=0x2aaab0007780,
e=0x7fff6ee0c070) at myththemedmenu.cpp:2385
ret = true
#9 0x00002ac73db6d68e in MythMainWindow::eventFilter
(this=0x851510,
e=0x7fff6ee0c070) at mythmainwindow.cpp:1114
top = (class MythScreenType *) 0x2aaab0007780
it = (class MythScreenStack **) 0xdcbd88
ge = (MythGestureEvent *) 0x2aaab139e4b0
#10 0x00002ac73ea2b852 in QObject::activate_filters ()
from /usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#11 0x00002ac73ea2b8a7 in QObject::event () from
/usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#12 0x00002ac73ea5db58 in QWidget::event () from
/usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#13 0x00002ac73e9d7485 in QApplication::internalNotify ()
from /usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#14 0x00002ac73e9d870e in QApplication::notify ()
from /usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#15 0x00002ac73e97e5c9 in QETWidget::translateKeyEvent ()
from /usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#16 0x00002ac73e97ff13 in QApplication::x11ProcessEvent ()
from /usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#17 0x00002ac73e98e3cf in QEventLoop::processEvents ()
from /usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#18 0x00002ac73e9e9f52 in QEventLoop::enterLoop ()
from /usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#19 0x00002ac73e9e9e02 in QEventLoop::exec () from
/usr/qt/3/lib/libqt-mt.so.3
No symbol table info available.
#20 0x00000000004335e2 in main (argc=1, argv=0x7fff6ee0d2c8)
at main.cpp:1255
geometry = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x665da0,
static shared_null = 0x665da0}
display = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x665da0,
static shared_null = 0x665da0}
a = <incomplete type>
logfile = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x75bd70,
static shared_null = 0x665da0}
pluginname = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x75bca0,
static shared_null = 0x665da0}
settingsOverride = {sh = 0x75bff0}
finfo = {fn = {static null = {
static null = <same as static member of an already
seen type>,
d = 0x665da0, static shared_null = 0x665da0}, d =
0x75c050,
static shared_null = 0x665da0}, fic = 0x0, cache =
true, symLink = false}
binname = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x75c0a0,
static shared_null = 0x665da0}
ResetSettings = false
logfd = -1
fileprefix = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d = 0x7995a0,
static shared_null = 0x665da0}
dir = <incomplete type>
priv_thread = 1082132800
status = 0
themename = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d =
0x2aaab0001000,
static shared_null = 0x665da0}
randomtheme = false
themedir = {static null = {
static null = <same as static member of an already seen
type>,
d = 0x665da0, static shared_null = 0x665da0}, d =
0x2aaab0001380,
static shared_null = 0x665da0}
mainWindow = (class MythMainWindow *) 0x851510
pmanager = (MythPluginManager *) 0x8461e0
mon = (class MediaMonitor *) 0x7d3eb0
networkControl = (class NetworkControl *) 0x0
exitstatus = 0
#0 0x00002aaaaac65d20 in ?? ()


_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


danielk at cuymedia

Aug 8, 2007, 9:51 AM

Post #22 of 69 (3697 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

On Wed, 2007-08-08 at 09:56 +0100, Allan Stirling wrote:
> Daniel Kristjansson wrote:
> > it would be good to get some people to test this before we merge
> > it into the trunk.
>
> Tested on a Athlon X2 4000, NVidia Geforce 6150 embedded
> video. Works OK for CPU+. Appears to be too slow for the
> opengl renderer.

Sounds about right, I can't imagine that any integrated
GeForce could handle the OpenGL renderer. You need a fairly
beefy card even for NTSC/PAL, especially if you use any
of the OpenGL filters.

> There's a few issues with tabbing order on the Settings / TV
> / Playback UI - I may have time to look at these later.

Ok, just point them out to me.

> On another system (effectively a Mac Mini - Core 2 Duo,
> i845GM integrated video), I get a segfault which I'm not
> managing to track down. This system definitely works with trunk.
<snip>
> Thread 7 (Thread 1115703616 (LWP 1429)):
> #0 0x00002aaaaac65d20 in ?? ()
> No symbol table info available.
> #1 0x00002ac73f1b960d in __glXInitialize ()
> from //usr/lib64/opengl/xorg-x11/lib/libGL.so.1
> No symbol table info available.
> #2 0x00002ac73f1b4e59 in glXQueryVersion ()
> from //usr/lib64/opengl/xorg-x11/lib/libGL.so.1
> No symbol table info available.
> #3 0x00002ac73c49902b in get_glx_version (XJ_disp=0x83fb90,
> major=@0x4280258c, minor=@0x42802588) at util-opengl.cpp:83
> ret = 1
> errbase = 155
> eventbase = 77
> gl_major = 1061374025
> gl_minor = 10951
> tmp_disp = (Display *) 0x7ee7e0
<snip>

It looks like you are linked to the Xorg libGL but I presume are using
the nVidia proprietary driver. You must use libraries that match your
drivers, otherwise segfaults and kernel lockups are sure to follow.

-- Daniel

_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


Dibblahmythml0015 at pendor

Aug 8, 2007, 10:04 AM

Post #23 of 69 (3698 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

Daniel Kristjansson wrote:
> On Wed, 2007-08-08 at 09:56 +0100, Allan Stirling wrote:
>> Daniel Kristjansson wrote:
>>> it would be good to get some people to test this before we merge
>>> it into the trunk.
>
>> There's a few issues with tabbing order on the Settings / TV
>> / Playback UI - I may have time to look at these later.
>
> Ok, just point them out to me.
Haven't got the box in front of me at the moment. Maybe
later tonight.

>
>> On another system (effectively a Mac Mini - Core 2 Duo,
>> i845GM integrated video), I get a segfault which I'm not
>> managing to track down. This system definitely works with trunk.
> <snip>
>> Thread 7 (Thread 1115703616 (LWP 1429)):
>> #0 0x00002aaaaac65d20 in ?? ()
>> No symbol table info available.
>> #1 0x00002ac73f1b960d in __glXInitialize ()
>> from //usr/lib64/opengl/xorg-x11/lib/libGL.so.1
>> No symbol table info available.
>> #2 0x00002ac73f1b4e59 in glXQueryVersion ()
>> from //usr/lib64/opengl/xorg-x11/lib/libGL.so.1
>> No symbol table info available.
>> #3 0x00002ac73c49902b in get_glx_version (XJ_disp=0x83fb90,
>> major=@0x4280258c, minor=@0x42802588) at util-opengl.cpp:83
>> ret = 1
>> errbase = 155
>> eventbase = 77
>> gl_major = 1061374025
>> gl_minor = 10951
>> tmp_disp = (Display *) 0x7ee7e0
> <snip>
>
> It looks like you are linked to the Xorg libGL but I presume are using
> the nVidia proprietary driver. You must use libraries that match your
> drivers, otherwise segfaults and kernel lockups are sure to follow.

Apologies for not being clear. These are two separate
systems - The one with the segfault is an i845 with the
intel driver - Therefore mesa provides the opengl.

Cheers,

Allan.
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


adeffs.mythtv at gmail

Aug 8, 2007, 6:15 PM

Post #24 of 69 (3670 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

Daniel, I compiled mythtv-vid [at] 1414 and it crashes with the following backtrace:
[Thread debugging using libthread_db enabled]
[New Thread -1255340320 (LWP 26001)]
[New Thread -1258964048 (LWP 26011)]
[New Thread -1267352656 (LWP 26016)]
[New Thread -1288139856 (LWP 26018)]

Program received signal SIGINT, Interrupt.
[Switching to Thread -1255340320 (LWP 26001)]
0xb5e610c1 in open () from /lib/tls/libc.so.6
The program is running. Exit anyway? (y or n)
Thread 2 (process 564):
#0 0xb5e40327 in ?? ()
No symbol table info available.
#1 0xb6f00fdc in ?? ()
No symbol table info available.
#2 0x00000000 in ?? ()
No symbol table info available.

Thread 1 (process 563):
#0 0xb5da3b66 in ?? ()
No symbol table info available.
#1 0xb7fc54f8 in ?? ()
No symbol table info available.
#2 0x00000001 in ?? ()
No symbol table info available.
#3 0xb67ea6a8 in ?? ()
No symbol table info available.
#4 0xb5ea9ff4 in ?? ()
No symbol table info available.
#5 0xb5eab880 in ?? ()
No symbol table info available.
#6 0x08127aa8 in CustomPriority::installClicked (this=0x0) at
custompriority.cpp:380
query = <incomplete type>
#7 0x00000001 in ?? ()
No symbol table info available.
#8 0xb7fba6c8 in ?? ()
No symbol table info available.

Thread 4 (process 7587):
#0 0xb5d47b81 in pthread_cond_wait@@GLIBC_2.3.2 () from
/lib/tls/libpthread.so.0
No symbol table info available.
#1 0xb652b238 in QWaitCondition::wait () from /usr/lib/libqt-mt.so.3
No symbol table info available.
#2 0xb6b63e35 in MythContext::waitPrivRequest () from
/usr/local/lib/libmyth-0.20.so.0
No symbol table info available.
#3 0x08061c78 in ?? ()
No symbol table info available.
#4 0xb5d45ced in start_thread () from /lib/tls/libpthread.so.0
No symbol table info available.
#5 0xb5bc3dee in clone () from /lib/tls/libc.so.6
No symbol table info available.

Thread 3 (process 7596):
#0 0xb5bbc327 in select () from /lib/tls/libc.so.6
No symbol table info available.
#1 0xb6cbb7fd in MythSocket::readyReadThread () from
/usr/local/lib/libmyth-0.20.so.0
No symbol table info available.
#2 0xb5d45ced in start_thread () from /lib/tls/libpthread.so.0
No symbol table info available.
#3 0xb5bc3dee in clone () from /lib/tls/libc.so.6
No symbol table info available.

Thread 2 (process 7632):
#0 0xb5d4a411 in __read_nocancel () from /lib/tls/libpthread.so.0
No symbol table info available.
#1 0xb5aedf2c in lirc_nextcode () from /usr/lib/liblirc_client.so.0
No symbol table info available.
#2 0xb6ce4446 in LircClient::Process () from /usr/local/lib/libmyth-0.20.so.0
No symbol table info available.
#3 0xb6a6d944 in SpawnLirc (param=0xb3b8b508) at mythmainwindow.cpp:65
config_file = {static null = {static null = <same as static
member of an already seen
type>, d = 0x819d418, static shared_null = 0x819d418}, d = 0x81a4ce0,
static shared_null = 0
x819d418}
cl = (struct LircClient *) 0x8274c28
#4 0xb5d45ced in start_thread () from /lib/tls/libpthread.so.0
No symbol table info available.
#5 0xb5bc3dee in clone () from /lib/tls/libc.so.6
No symbol table info available.

Thread 1 (process 7583):
#0 0xb5b1b7c7 in raise () from /lib/tls/libc.so.6
No symbol table info available.
#1 0xb5b1d06b in abort () from /lib/tls/libc.so.6
No symbol table info available.
#2 0xb5b52545 in __fsetlocking () from /lib/tls/libc.so.6
No symbol table info available.
#3 0xb5b59aaf in free () from /lib/tls/libc.so.6
No symbol table info available.
#4 0xb5b5adfc in malloc () from /lib/tls/libc.so.6
No symbol table info available.
#5 0xb5d0b8b8 in operator new () from /usr/lib/libstdc++.so.6
No symbol table info available.
#6 0xb3a55306 in MythArchiveTempDir () at archivesettings.cpp:16
gc = (HostLineEdit *) 0x14
#7 0xb3a58f3e in ArchiveSettings (this=0xbfd50798) at archivesettings.cpp:275
vcg1 = (struct VerticalConfigurationGroup *) 0xb3b234d8
vcg2 = (struct VerticalConfigurationGroup *) 0xb7fa2470
vcg3 = (struct VerticalConfigurationGroup *) 0xb5c1bee8
vcg4 = (struct VerticalConfigurationGroup *) 0xb3b1af38
#8 0xb3a54973 in mythplugin_init (libversion=0xb6d1bac4
"0.20.20070717-1") at main.cpp:540
mpSettings = {<> = {<No data fields>}, <No data fields>}
#9 0xb6c38528 in MythPlugin::init () from /usr/local/lib/libmyth-0.20.so.0
No symbol table info available.
#10 0xb6c38b97 in MythPluginManager::init_plugin () from
/usr/local/lib/libmyth-0.20.so.0
No symbol table info available.
#11 0xb6c39b66 in MythPluginManager::MythPluginManager () from
/usr/local/lib/libmyth-0.20.so
.0
No symbol table info available.
#12 0x08067eb2 in ?? ()
No symbol table info available.
#13 0xb5b06eb0 in __libc_start_main () from /lib/tls/libc.so.6
No symbol table info available.
#14 0x0805e791 in ?? ()
No symbol table info available.


--
Steve
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev


adeffs.mythtv at gmail

Aug 8, 2007, 6:54 PM

Post #25 of 69 (3672 views)
Permalink
Re: mythtv-vid branch needs testing [In reply to]

sorry for that last email, i needed to recompile mythplugins.

Now I'm trying to watch an SD show to see if it uses the OpenGL
renderer and see how the new scaling works out. I'm noticing this in
the log: which shows its turning on and off the deinterlacing,
something it used to do before. These are from captures of analog
stations over firewire. Is there a way to force deinterlacing on
automatically?

Also, is there more than one option for scaling in the OpenGL renderer
or is the default the "best quality" its going to be able to provide?

the log:
2007-08-08 21:51:33.443 NVP: progressive frame seen after 3 interlaced frames
2007-08-08 21:51:33.510 NVP: interlaced frame seen after 2 progressive frames
2007-08-08 21:51:33.543 NVP: progressive frame seen after 1 interlaced frames
2007-08-08 21:51:33.610 GLVid: Turning off deinterlacing.
2007-08-08 21:51:33.610 Disabled deinterlacing
2007-08-08 21:51:33.643 NVP: interlaced frame seen after 3 progressive frames
2007-08-08 21:51:33.710 GLVid: Turning on deinterlacing.
2007-08-08 21:51:33.710 Enabled deinterlacing
2007-08-08 21:51:33.777 NVP: progressive frame seen after 4 interlaced frames
2007-08-08 21:51:33.810 NVP: interlaced frame seen after 1 progressive frames
2007-08-08 21:51:34.177 NVP: progressive frame seen after 11 interlaced frames
2007-08-08 21:51:34.243 NVP: interlaced frame seen after 2 progressive frames
'video_output' mean = '33330.69', std. dev. = '3387.24', fps = '30.00'
2007-08-08 21:51:34.543 NVP: progressive frame seen after 9 interlaced frames
2007-08-08 21:51:34.610 NVP: interlaced frame seen after 2 progressive frames
2007-08-08 21:51:34.677 NVP: progressive frame seen after 2 interlaced frames
2007-08-08 21:51:34.743 GLVid: Turning off deinterlacing.
2007-08-08 21:51:34.744 Disabled deinterlacing
2007-08-08 21:51:34.777 NVP: interlaced frame seen after 3 progressive frames
2007-08-08 21:51:34.843 NVP: progressive frame seen after 2 interlaced frames
2007-08-08 21:51:34.877 NVP: interlaced frame seen after 1 progressive frames
2007-08-08 21:51:34.943 GLVid: Turning on deinterlacing.
2007-08-08 21:51:34.944 Enabled deinterlacing
2007-08-08 21:51:35.080 NVP: progressive frame seen after 6 interlaced frames
2007-08-08 21:51:35.127 NVP: interlaced frame seen after 1 progressive frames
2007-08-08 21:51:35.677 NVP: progressive frame seen after 17 interlaced frames
2007-08-08 21:51:35.710 NVP: interlaced frame seen after 1 progressive frames
2007-08-08 21:51:35.977 NVP: progressive frame seen after 8 interlaced frames
2007-08-08 21:51:36.043 GLVid: Turning off deinterlacing.
2007-08-08 21:51:36.044 Disabled deinterlacing
2007-08-08 21:51:36.410 NVP: interlaced frame seen after 13 progressive frames
2007-08-08 21:51:36.477 GLVid: Turning on deinterlacing.
2007-08-08 21:51:36.477 Enabled deinterlacing
2007-08-08 21:51:36.777 NVP: progressive frame seen after 11 interlaced frames
2007-08-08 21:51:36.844 GLVid: Turning off deinterlacing.
2007-08-08 21:51:36.844 Disabled deinterlacing
2007-08-08 21:51:36.877 NVP: interlaced frame seen after 3 progressive frames
2007-08-08 21:51:36.944 NVP: progressive frame seen after 2 interlaced frames

On 8/8/07, Steven Adeff <adeffs.mythtv [at] gmail> wrote:
> Daniel, I compiled mythtv-vid [at] 1414 and it crashes with the following backtrace:


--
Steve
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev

First page Previous page 1 2 3 Next page Last page  View All MythTV dev RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.