[Milhouse]

New OpenELEC Jarvis build #0807: RPi / RPi2
(Supercedes previous build)

Code:

# uname -a
Linux rpi512 4.1.4 #1 Fri Aug 7 21:02:59 BST 2015 armv6l GNU/Linux

# vcgencmd version
Aug  3 2015 14:56:05
Copyright (c) 2012 Broadcom
version 4b51d81eb0068a875b336f4cc2c468cbdd06d0c5 (clean) (release)

# lsb_release
OpenELEC (Milhouse) - Version: devel-20150807210212-#0807-g15ce945 [Build #0807]

# vcdbg log msg 2>&1 | grep DTOK
001580.260: Kernel trailer DTOK property says yes

# Kernel device tree status: Enabled

Based on tip of OpenELEC master (15ce9459, changelog) and tip of XBMC master (c09c73f8, changelog) with the following modifications:

• Includes newclock4 patches
  
• Excludes the OpenELEC fernetmenta patches due to conflicts with newclock4
  
• Excludes the OpenELEC linux-01-RPi_support patch in favour of sourcing these and possibly more recent patches directly from kernel branch rpi-4.1.y
  
• Default setting for "Show RSS Feed" changed to disabled (new installs only) [patch details]
  
• Disabled "Total Duration" in Confluence (see build #0221 for details)
  
• Includes latest dcadec master (2a9186e3, ahead +34)
  
• Includes latest kodi-platform master (15edaf78, ahead +2)
  
• Includes latest libcec master (e8506b3d, ahead +29)
  
• Includes latest libnfs master (cc69b9c1, ahead +29)
  
• Includes latest platform master (081032fd, ahead +3)
  
• Includes latest addons: adsp.basic (da6ee5dd), adsp.biquad.filters (b6a31a90), adsp.freesurround (4dd3202f), asplib (490de460), pvr.argustv (b5fef151, +22), pvr.demo (bb2c46a6, +13), pvr.dvblink (65073157, +15), pvr.dvbviewer (d1052eb5, +14), pvr.filmon (5d1a5c13, +12), pvr.hts (676ea53e, +36), pvr.iptvsimple (353cb28d, +12), pvr.mediaportal.tvserver (0de32b2b, +26), pvr.mythtv (75aa8eee, +59), pvr.nextpvr (a731f3c4, +15), pvr.njoy (43b4698f, +12), pvr.pctv (ff5d5360, +12), pvr.stalker (b8a61cb8, +12), pvr.vbox (48ebfdae, +24), pvr.vdr.vnsi (8dedf209, +11), pvr.vuplus (0702fc50, +14), pvr.wmc (e87f1328, +14)
  
• Exclude kodi-100.10-handle-SIGTERM.patch: Use updated Kodi patch from stefansaraev repo
  
• Include b24d9e06: handle SIGTERM
  
• Include patch: Add ffmpeg dependency and includes for HEVC optimisations
  
• Include patch: Fix devinput 18b8ba44
  
• Include patch: Add experimental splash video
  
• Include patch: Enable pvr addons, disable PVR updates
  
• Include PR:4228: ffmpeg: update to ffmpeg-2.7.1
  
• Include PR:4260: sshd: restart on failure. dont wait for network
  
• Include PR:4262: buildsys improvements
  
• Include PR:7462: added: hide thumbs for unwatched episodes option to show plot is off
  
• Revert 25a51dad: Not yet ready for prime time
  

Build Highlights:

1. Kodi cosmetics
   

Build Details:

1. XBMC:
   • [PVR] 'Confim channel switch' only for up/down, not ch+/- (PR:7733, 1 commit, 5 files changed)
     
   • changed: Rename cache: to forward: as that's what it really is (PR:7704, 1 commit, 1 file changed)
     
   • [rsxs] Fix compilation error on GCC 5 (PR:7672, 1 commit, 1 file changed)
     
   • [dx11] Don't cull a texture which is shown with reverse side. (PR:7740, 1 commit, 1 file changed)
     
   • [PVR] Conflence: Add EpisodeName to MyPVRChannels (PR:7731, 1 commit, 1 file changed)
     
   • [pydocs] update sortMethods to only suggest using string (PR:7745, 2 commits, 7 files changed)
     
2. pvr.dvbviewer:
   • Automatically fill in platform and library name (PR:17, 2 commits, 2 files changed)
     
3. kernel 4.1.y:
   • New commits in this build:
     • bcm2835 camera planar/packed stride length (869c3bc3)
       
4. Additional commits/pull requests/changes not yet merged upstream:
   • Updated: patch: Fix devinput 18b8ba44
     

[zaphod24]

Latest build, 0807, is working great for me on my rpi2 watching 1080i60 live TV (which is mpeg2 for my provider) with advanced deinterlacing. Seems to be fast-forwarding better/quicker than the last few builds. I am using the following memory and overclock settings that I got from https://www.raspberrypi.org/forums/viewt...6&t=100994 which might be helpful to others:

dtoverlay=sdhost,overclock_50=84
gpu_mem_1024=320
arm_freq=1050
core_freq=525
gpu_freq=350
over_voltage=4

and also (instead of force_turbo=1) in autostart.sh:

echo performance > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor

These seem quite stable for me. I am using 2 heat sinks, one each on the main chips.

[Milhouse]

Yes this particular crash has only happened once. The assert that fails is this one https://github.com/sahlberg/libnfs/blame.../pdu.c#L92

Are you able to re-test #0802 and confirm if this also crashes? If it has the same problem as #0803 (to be honest, I'd expect it to, assuming the problem is due to libnfs changes) then it will result in a segfault as non-debug enabled builds don't include the assert() statements, so you're most likely barrelling into whatever issue the assert is trying to detect.

Also, if you're able to reproduce the crash in the latest #0807 build, it would be worth trying RPi2 debug build #0807x which has the recent libnfs changes reverted (ie. libnfs HEAD is 68941809) and see if there is still a problem - this would definitely rule libnfs in or out. And since #0807x is debug-enabled, the asserts will be enabled.

[doveman2]

Latest build, 0807, is working great for me on my rpi2 watching 1080i60 live TV (which is mpeg2 for my provider) with advanced deinterlacing. Seems to be fast-forwarding better/quicker than the last few builds. I am using the following memory and overclock settings that I got from https://www.raspberrypi.org/forums/viewt...6&t=100994 which might be helpful to others:

dtoverlay=sdhost,overclock_50=84
gpu_mem_1024=320
arm_freq=1050
core_freq=525
gpu_freq=350
over_voltage=4

and also (instead of force_turbo=1) in autostart.sh:

echo performance > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor

These seem quite stable for me. I am using 2 heat sinks, one each on the main chips.

I'm not familiar with
dtoverlay=sdhost,overclock_50=84
what does that do?

Likewise for
echo performance > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
instead of force_turbo=1, is that safer? I seem to recall force_turbo=1 voided the warranty at one point but I think that's no longer the case but I still don't use it as it seems like it might be asking for trouble!

[illiac4]

Why are in latest builds debug logs so huge?
Anyway here is the debug log using mmal, livetv, no overclock. The picture freezes, sound continues, when i want to stop it has started to switch resolutions and eventually i had to reset it because i was not able to control it anymore with remote.
And the network is perfectly ok because i was testing lately also with 1037U and livetv there works perfectly.

Debug log (i had to compress it because it is so big):
http://bite-in.com/debug/debug.tgz

[illiac4]

One question. I have disabled mmal acceleration and am using now only omxplayer and livetv works. ASAIK the mmal should be used for video files. But is it possible to use both? I'd like to use omxplayer for livetv only and not mmal since it causes problems.
So would be ideal to be able to enable both under acceleration but when using livetv only omx will be used for it. Is this possible to achive with advancedsettings.xml?

And another question:
What is used when using omxplayer as default with disabled mmal and then you select mmal as deinterlacing method?
P.S.:
I see if i enable both omx and mmal thne for livetv by default is used omx, so no problem.

[ivota]

Code:

ERROR: CDVDPlayerAudio::DecodeFrame - Decode Error. Skipping audio packet (-1094995529)

http://xbmclogs.com/psn9rfbaw

[Milhouse]

Code:

ERROR: CDVDPlayerAudio::DecodeFrame - Decode Error. Skipping audio packet (-1094995529)

http://xbmclogs.com/psn9rfbaw

Can you upload a debug log so that the media info for the stream is visible. I can't access the stream you're playing ("Wrong IP"). If you can download the stream and make a sample available that might also help.

Edit: Actually I guess the following might be enough to describe the audio:

Code:

08:23:20 144.408569 T:1448076352  NOTICE: Creating audio stream (codec id: 86018, channels: 2, sample rate: 48000, no pass-through)

[ivota]

debug log ??!! ( I add enabled audio and video debug in kodi log)

http://xbmclogs.com/pddr3s0z8

[Milhouse]

I'm not familiar with
dtoverlay=sdhost,overclock_50=84
what does that do?

http://forum.kodi.tv/showthread.php?tid=...pid2005396

[Heiko123]

Hello,

I've explored that after one of the last updates, the TV-epg and TV-recordings won't be updated from time to time.
The only way is a reboot, within a new scan and reading.

Is it necessary that anything is buggy or is it a add-on problem ?

http://pastebin.com/LpZw7tqB

[evanspae]

Hello,

I've explored that after one of the last updates, the TV-epg and TV-recordings won't be updated from time to time.
The only way is a reboot, within a new scan and reading.

Is it necessary that anything is buggy or is it a add-on problem ?

http://pastebin.com/LpZw7tqB

I get same no guide details with ArgusTV plugin and with build #0806

[Heiko123]

Hello,

I've explored that after one of the last updates, the TV-epg and TV-recordings won't be updated from time to time.
The only way is a reboot, within a new scan and reading.

Is it necessary that anything is buggy or is it a add-on problem ?

http://pastebin.com/LpZw7tqB

I get same no guide details with ArgusTV plugin and with build #0806

Did you have after reboot the guide or not ?
I've guide and records and I see it , but it wont updated.

[Milhouse]

@Heiko12, @evanspae: @zaphod24 thinks the problem started "between 0802 and 0803":

Not RPI specific I'm sure, but between 0802 and 0803 the EPG no longer seems to be updating for live TV. There are no strange errors in the logs either.

Edit: Deleted Epg10.db and rebooted once or twice and now it's working...

Can either of you narrow down exactly which build introduced this problem, presumably either #0802 or #0803?

If I were to hazard a guess, PR6937 (in #0803) is most probably the culprit, and PR7734 is being worked on which may be trying to fix this issue (but doesn't seem to be working just yet). Or the problem might have been caused by some other PR...

[ksooo]

Just a quick question. Does "EPG does not get updated" mean that there do no new EPG data arrive (e.g EPG window runs dry and no longer shows data for "now") or are you talking about the EPG window not automatically following the timeline, but having data?