747 replies to this topic

[littlesat]

As should recommend pieterg and or milo will look into this first... I have no idea what this might break... But at least thanks for the recommendation...

[mik9]

That is probably something I do not use as I only use Movie Selection and play just one file. Anyway, thanks for this solution, it was easy to adjust and works for me.

[littlesat]

If you know how to patch that row... Then please investigate further an come with a better patch.... I just try to motivate making is by yourself better...

[mik9]

Did you try to destroy it on close?

[pieterg]

I have a "simple" solution for this. The user requested a skip, so why sending EOF event in this case? I see no reason for this.
So change filepush.cpp to:

143						 if (m_stop)
144								 break;
141						 sendEvent(evtEOF);
145						 if (m_stream_mode)

By the looks of it, that would break end-of-file detection (playback should stop when the file ends)

Or only sending EOF if eofcount=10.

that won't fix things on low-bandwidth channels, you'll easily get more than 10 EOF iterations before the buffer has grown to 188*1024 again, so you'd still get the EOF event.

[littlesat]

Ik had bij het bekijken ook al het idee dat het de gehele eof detectie stuk zou maken...

As I looked at the patch I got the impression that the eof detection was damaged by this.

Edited by Frenske, 30 January 2013 - 03:53.
English please. ;)

[Apple740]

Dreambox8000 user here. Problems with 1/3/4/6/7/9 keys not always doing what they should do within a timeshift exists for years, and i always thought they were DreamMultimedia related.
However, to my big surprise, last week i tried an OE2.0 based image and found out that the timeshift works flawless here!

Sorry but this is the end of Openpli for me (unless issue's are fully resolved ).

Edited by Apple740, 30 January 2013 - 18:29.

[betacentauri]

Hi Pieterg,

thanks for looking into it.
I made several tests:
1. Playing recording with skipping before start of file or after file end. And watching until normal end of file -> no problems
2. Normal live tv -> as expected no problems
3. Normal timeshift -> the problem is fixed

My fix "only" sends no EOF, if there was a stop/pause request. So end of file recognition works as before.
If there is a bigger timeshift buffer, then also no EOF event is sent while skipping.

You and everyone else with an ET9x00 can test it. I have attached the enigma2 binary.

Attached Files

•  enigma_bin.tar.gz   824.42KB   10 downloads

[ajeeb]

Thanks for fix this issue .
but there is a bug runs all the way with pli even pli2
this happen when you start timeshift a channel that included a dvb subs . I mean when resume play TShift the last sub choosed disappear and to get it back need press yellow button and choose it AGAIN.

hope you understand me.

BR

[greatred]

@betacentauri

I would like to use your fix but I`m having a VU+Uno.
My build environement is setup and I was able to successfuly build my first OpenPLi image.
Could you please tell me how I can build only the enigma2 binary with the problematic commit reverted?
Many thanks in advance.

[betacentauri]

@all: Already seven downloads. Did somebody test it and can confirm, that it's working??

@greatred: Here's the howto for my et9200. Your directories might be named in a different way.
1. Go to your build directory: <path_to_build_dir>/openpli-oe-core
2. Edit site.conf and put a # before last line:

#INHERIT += "rm_work"

3. Change to <path_to_build_dir>/openpli-oe-core/build-et9x00 directory
4. Execute this

source env.source; bitbake -f -c compile -b ../meta-openpli/recipes-openpli/enigma2/enigma2.bb

5. Enigma2 unpatched binary is now located in this directory:
<path_to_build_dir>/openpli-oe-core/build-et9x00/tmp/work/et9x00-oe-linux/enigma2-2.7+......./git/main
6. Edit this file: <path_to_build_dir>/openpli-oe-core/build-et9x00/tmp/work/et9x00-oe-linux/enigma2-2.7+......./git/lib/base/filepush.cpp
Goto line 141 and change

sendEvent(evtEOF);
if (m_stop)
  break;

into this

if (m_stop)
  break;
sendEvent(evtEOF);

And save file.
7. Change to <path_to_build_dir>/openpli-oe-core/build-et9x00 directory
8. Execute this

source env.source; bitbake -f -c compile -b ../meta-openpli/recipes-openpli/enigma2/enigma2.bb

9. If there were no error messages, binary is located here: <path_to_build_dir>/openpli-oe-core/build-et9x00/tmp/work/et9x00-oe-linux/enigma2-2.7+......./git/main
Please check file date of the binary. It must have the date of the second compile!!
10. You can strip binary to make it smaller (I should be around 40MB. After stripping it's only 2,8MB big). You don't need to do this:
To strip execute this:

<path_to_build_dir>/openpli-oe-core/build-et9x00/tmp/sysroots/i686-linux/usr/bin/mips32el-oe-linux/mipsel-oe-linux-strip<space><path_to_enigma2_binary>

11. Put the binary on your stb. E.g. in /tmp and do a chmod 755 enigma2
12. Do a init 4
13. Start enigma2 from /tmp with ./enigma2

Edited by betacentauri, 31 January 2013 - 23:17.

[blzr]

@vu-boxes users -
seems that thanks to this @pieterg's commit:
http://openpli.git.s...5dab92ecab10ef7
vu last timeshit issues (an incorrect timeshif behavior on un-pausing /ie jump to 'live'/) should be fixed from the next build
//at least on my duo it's ok now ...

[greatred]

@betacentauri

many thanks for this great guide
Instead of patching the filepush.cpp I reverted tstools.h and tstools.cpp to versions prior to 31.12.2012
And just few minutes later I see that your patch has been commited.

BIG THANKS to betacentauri

[blzr]

btw. speaking of 'how to compile own modified e2' I recommend slightly different method (which does not require keeping entire 'work' just to build one binary; efficient use of resources, you know )
assuming you have openpli oe core build enviroment set up:
1. clone (openpli) enigma2 repository
2. make your changes in there
3. bitbake enigma2 pointing to the recipe included in your local e2 repo, something like this:

brt@brt:~$ cd openpli-oe-core/build-vuduo/
brt@brt:~/openpli-oe-core/build-vuduo$ source env.source
brt@brt:~/openpli-oe-core/build-vuduo$ bitbake -b ~/enigma2/enigma2.bb

it will fail on do_package, but you'll find package content in ~/enigma2/build/package
(and of course on subsequent tests/builds you'll have to remember all these pull / reset / cleansstate etc...)
//and sorry for OT

Edited by blzr, 2 February 2013 - 22:58.

[betacentauri]

@greatred: The todays commit fixes only the black screen problem. See here http://openpli.org/f...-two-questions/
I don't believe(but cannot exclude) that it fixes the problem with seeking back when you are almost at the end of the timeshift buffer (as described in this thread).
For this you should change the filepush.cpp
It would be great if you can test it.

@blzr: Thanks! That seems easier. My current problem was that everytime someone commited a patch a new folder was checked out and I had to move my changes to the new folder.

[greatred]

It looks like it fixed more than black screen issue from mentioned thread
Just did some random timeshifting and I can confirm that unpausing is working again as expected.

[Dimitrij]

I just tested it.
Problems with a black screen during rewinding remained.

[betacentauri]

@greatred: thanks for testing and great to hear that it works for you!

@dima: black screen while seeking? I don't have that. I only had black screens when I started normal time shift. What Stb do you use? What do we have to do to reproduce your problem?

[Dimitrij]

ЕТ9000

action ->  InfobarTimeshiftActivateActions timeshiftActivateEnd
activateTimeshiftEnd
play, ...
decoder state: play, vpid=-1, apid=-1
DEMUX_STOP - pcr - ok
DEMUX_STOP - video - ok
VIDEO_STOP - ok
AUDIO_STOP - ok
DEMUX_STOP - audio - ok
AUDIO_CONTINUE - ok
cleaning up
TuxTxt cache cleared
alloc PVR
allocate demux
loading streaminfo for /media/hdd/timeshift/timeshift.aPMI3H
setIoPrio best-effort level 0 ok
FILEPUSH THREAD START
ok ... now we start!!
eventNewProgramInfo TS
have 1 video stream(s) (0647), and 1 audio stream(s) (0a2f), and the pcr pid is 0647, and the text pid is ffffffff
seeking relative to end. len=5527800, seek = -1000
samples step 41942988, pts begin 6850798943, pts end 6856326743, offs begin 51724, offs end 109346816:
eDVBTSTools::getPTS got it from sc file offset=51724 pts=6850798943
adding sample 51724: pts 0 -> pos 51724 (diff 0 bytes)
eDVBTSTools::getPTS got it from sc file offset=41987004 pts=6852865343
adding sample 41994712: pts 2066400 -> pos 41987004 (diff -7708 bytes)
eDVBTSTools::getPTS got it from sc file offset=72085240 pts=6854483543
adding sample 83937700: pts 3684600 -> pos 72085240 (diff -11852460 bytes)
using: 3684600:5527800 -> 72085240:109346816
eDVBTSTools::getPTS got it from sc file offset=72085240 pts=6854483543
adding sample 109326512: pts 3684600 -> pos 72085240 (diff -37241272 bytes)
calculated diff -20468 ms
diff to big, refining
using: 3684600:5527800 -> 72085240:109346816
eDVBTSTools::getPTS got it from sc file offset=72085240 pts=6854483543
adding sample 109326512: pts 3684600 -> pos 72085240 (diff -37241272 bytes)
calculated diff -20468 ms
diff to big, refining
using: 3684600:5527800 -> 72085240:109346816
eDVBTSTools::getPTS got it from sc file offset=72085240 pts=6854483543
adding sample 109326512: pts 3684600 -> pos 72085240 (diff -37241272 bytes)
calculated diff -20468 ms
diff to big, refining
using: 3684600:5527800 -> 72085240:109346816
eDVBTSTools::getPTS got it from sc file offset=72085240 pts=6854483543
adding sample 109326512: pts 3684600 -> pos 72085240 (diff -37241272 bytes)
calculated diff -20468 ms
diff to big, refining
using: 3684600:5527800 -> 72085240:109346816
eDVBTSTools::getPTS got it from sc file offset=72085240 pts=6854483543
adding sample 109326512: pts 3684600 -> pos 72085240 (diff -37241272 bytes)
calculated diff -20468 ms
TuxTxt cache cleared
diff to big, refining
aborting. Taking 109326512 as offset for 3684600
ok, resolved skip (rel: 0, diff 3684600), now at 068430b0
decoder state: play, vpid=1607, apid=2607
DMX_SET_PES_FILTER(0xa2f) - audio - ok
DEMUX_START - audio - ok
AUDIO_SET_BYPASS(1) - wait for driver eof ok
reached EOF, but the file may grow. delaying 1 second.
ok
AUDIO_PAUSE - ok
AUDIO_PLAY - ok
Video Device: /dev/dvb/adapter0/video0
demux device: /dev/dvb/adapter0/demux6
VIDEO_SET_STREAMTYPE 1 - ok
DMX_SET_PES_FILTER(0x647) - video - ok
DEMUX_START - video - ok
VIDEO_FREEZE - ok
VIDEO_PLAY - ok
VIDEO_SLOWMOTION(0) - ok
VIDEO_FAST_FORWARD(0) - ok
VIDEO_CONTINUE - ok
AUDIO_CONTINUE - ok
disable teletext subtitles
eDVBServicePlay::switchToTimeshift, in pause mode now.
eDVBServicePlay::pause
decoder state: pause, vpid=1607, apid=2607
VIDEO_SLOWMOTION(0) - ok
VIDEO_FAST_FORWARD(0) - ok
VIDEO_FREEZE - ok
AUDIO_PAUSE - ok
have 1 video stream(s) (0647), and 1 audio stream(s) (0a2f), and the pcr pid is 0647, and the text pid is ffffffff
decoder state: pause, vpid=1607, apid=2607
VIDEO_SLOWMOTION(0) - ok
VIDEO_FAST_FORWARD(0) - ok
VIDEO_FREEZE - ok
AUDIO_PAUSE - ok
resolved to PAUSE
eDVBServicePlay::pause
eDVBServicePlay::seekTo: jump -90000
seek.
eFilePushThread stopping thread
FILEPUSH THREAD STOP
thread joined 0
AUDIO_CLEAR_BUFFER - ok
VIDEO_CLEAR_BUFFER - ok
setIoPrio best-effort level 0 ok
FILEPUSH THREAD START
seeking relative to end. len=5544000, seek = -90000
using: 3684600:5527800 -> 72085240:109346816
eDVBTSTools::getPTS got it from sc file offset=107839080 pts=6856256543
adding sample 107854848: pts 5457600 -> pos 107839080 (diff -15768 bytes)
calculated diff 40 ms
aborting. Taking 107854848 as offset for 5457600
ok, resolved skip (rel: 0, diff 5457600), now at 066dbc00
eDVBChannel: End of file!
PATready
use pmtpid 025f for service_id 025f
eDVBChannel: pvrEvent evtStopped
eventNewProgramInfo TS
have 1 video stream(s) (0647), and 2 audio stream(s) (0a2f, 0e17), and the pcr pid is 0647, and the text pid is ffffffff
decoder state: pause, vpid=1607, apid=2607
VIDEO_SLOWMOTION(0) - ok
VIDEO_FAST_FORWARD(0) - ok
VIDEO_FREEZE - ok
AUDIO_PAUSE - ok
resolved to FAST FORWARD
eDVBServicePlay::setFastForward(-2)
setting cue skipmode to -2
skipmode ratio is -180000:90000, bitrate is 14272223 bit/s
resolved to: -951788 524288
flush pvr
eFilePushThread stopping thread
wait for driver eof aborted by signal
FILEPUSH THREAD STOP
thread joined 0
AUDIO_CLEAR_BUFFER - ok
VIDEO_CLEAR_BUFFER - ok
setIoPrio best-effort level 0 ok
FILEPUSH THREAD START
done
decoder state: trickmode, vpid=1607, apid=2607
AUDIO_STOP - using: 5457600:5527800 -> 107839080:109346816
eDVBTSTools::getPTS got it from sc file offset=109319204 pts=6856326743
adding sample 37343289760: pts 5527800 -> pos 109319204 (diff -37233970556 bytes)
calculated diff -19262309 ms
diff to big, refining
using: 5457600:5527800 -> 107839080:109319204
eDVBTSTools::getPTS got it from sc file offset=109319204 pts=6856326743
adding sample 36661376348: pts 5527800 -> pos 109319204 (diff -36552057144 bytes)
calculated diff -19262309 ms
diff to big, refining
using: 5457600:5527800 -> 107839080:109319204
eDVBTSTools::getPTS got it from sc file offset=109319204 pts=6856326743
adding sample 36661376348: pts 5527800 -> pos 109319204 (diff -36552057144 bytes)
calculated diff -19262309 ms
diff to big, refining
using: 5457600:5527800 -> 107839080:109319204
ok
DEMUX_STOP - audio - eDVBTSTools::getPTS got it from sc file offset=109319204 pts=6856326743
ok
AUDIO_CONTINUE - ok
VIDEO_SLOWMOTION(0) - ok
VIDEO_FAST_FORWARD(1) - ok
VIDEO_CONTINUE - adding sample 36661376348: pts 5527800 -> pos 109319204 (diff -36552057144 bytes)
calculated diff -19262309 ms
diff to big, refining
using: 5457600:5527800 -> 107839080:109319204
eDVBTSTools::getPTS got it from sc file offset=109319204 pts=6856326743
adding sample 36661376348: pts 5527800 -> pos 109319204 (diff -36552057144 bytes)
calculated diff -19262309 ms
diff to big, refining
aborting. Taking 36661376348 as offset for 5527800
ok, resolved skip (rel: 1, diff 5527800), now at 88930355c
ok
eDVBChannel: End of file!
eDVBChannel: pvrEvent evtStopped
wait for driver eof ok
reached EOF, but the file may grow. delaying 1 second.
eDVBChannel: End of file!
wait for driver eof ok
reached EOF, but the file may grow. delaying 1 second.
eDVBChannel: End of file!
wait for driver eof ok
reached EOF, but the file may grow. delaying 1 second.
eDVBChannel: End of file!
wait for driver eof ok
reached EOF, but the file may grow. delaying 1 second.
eDVBChannel: End of file!
wait for driver eof ok
reached EOF, but the file may grow. delaying 1 second.
eDVBChannel: End of file!
wait for driver eof ok
reached EOF, but the file may grow. delaying 1 second.
eDVBChannel: End of file!
wait for driver eof ok
reached EOF, but the file may grow. delaying 1 second.
eDVBChannel: End of file!
wait for driver eof ok
reached EOF, but the file may grow. delaying 1 second.
eDVBChannel: End of file!
wait for driver eof ok
reached EOF, but the file may grow. delaying 1 second.
eDVBChannel: End of file!
wait for driver eof ok
FILEPUSH THREAD STOP
eDVBChannel: End of file!
eDVBChannel: pvrEvent evtStopped
4416

[pieterg]

I made several tests:
1. Playing recording with skipping before start of file or after file end. And watching until normal end of file -> no problems
2. Normal live tv -> as expected no problems
3. Normal timeshift -> the problem is fixed

My fix "only" sends no EOF, if there was a stop/pause request. So end of file recognition works as before.
If there is a bigger timeshift buffer, then also no EOF event is sent while skipping.

I've taken a closer look at the evtEOF handling in filepush, and indeed when we avoid emitting evtEOF when the thread is stopping (thus ignoring the decoder being empty, which it is because of the skip attempt), avoids that the timeshift switches to the live point.
And as far as I can see there are no side effects.

Only one thing is remaining, as Dima73 points out, sometimes you cannot rewind from the live point, you'll get a black screen.
In that case, the only way to get away from the live point, is press pause, and wait a few seconds.