[GBouerat]

Hello,

with build #0215 on an Intel NUC NUC7I3BNH https://www.intel.com/content/www/us/en/...i3bnh.html
a SONY OLED A1 TV
and a Sony STR-DN1080 receiver

I have some issues :
- long audio cuts/drops (I don't know the right word) when playing videos with Dolby Atmos track with passthrough enable
- white screen when playing HDR 10 bits videos

[Milhouse]

Hello,

with build #0215 on an Intel NUC NUC7I3BNH https://www.intel.com/content/www/us/en/...i3bnh.html
a SONY OLED A1 TV
and a Sony STR-DN1080 receiver

I have some issues :
- long audio cuts/drops (I don't know the right word) when playing videos with Dolby Atmos track with passthrough enable
- white screen when playing HDR 10 bits videos

And #0214 doesn't have this issue?

[GBouerat]

I don't know, #0215 is the first testbuild I installed.

[HeresJohnny]

Returning to my MySQL problem.. I did a bad thing. Wanted to update from 5.7.19 to 5.7.21 and had forgotten my root password. For kodi, there was a special user with admin priviliges so I tried to reset the root passwort and shot the database. Deleted everything and started with 5.7.21 from scratch. Unfortunately that means I can do no more tests with that database.

I have saved the complete old config, except the value for query_cache_size which I don't remember exactly but which was something between 80M and 100M (because I had read this article https://haydenjames.io/mysql-query-cache...rformance/). In a comment on the user manual for query_cache, Hayden James states: With the popularity of InnoDB as of MySQL 5.6.8 query cache is now disabled by default. Previously I recommended a small query_cache_size here: http://haydenjames.io/mysql-query-cache-...rformance/ However, with InnoDB it's almost always better to disable query cache now. Which is probably part of the reason for the new default.

Here's the config of my.ini
https://pastebin.com/G86KesjM

There are actually two ways to disable the query cache which are
query_cache_size = 0 
  or
query_cache_type = 0

The manual recommends:
If query_cache_size is 0, you should also set query_cache_type variable to 0. In this case, the server does not acquire the query cache mutex at all, which means that the query cache cannot be enabled at runtime and there is reduced overhead in query execution.

When I reported that I could now boot into KODI I had this config
query_cache_size = 0
query_cache_type = 1

Just for completeness sake, I run MySQL on Windows 10, 8GB, Intel Celeron 2930. There's also a mail server running on MySQL.

[Milhouse]

I don't know, #0215 is the first testbuild I installed.

That's why there's daily builds, so you can go back and identify when the problem started.

[Milhouse]

New LibreELEC.tv Leia build #0217: Generic
(Supercedes previous build)

SHA256 Checksum: 66bc106d6145151c77968f1f5b8bfe068e3ec3a4e0baffd2b422328ada07a507 (Generic)

text:
# uname -a

Linux NUC 4.14.18 #1 SMP Sun Feb 18 01:23:46 GMT 2018 x86_64 GNU/Linux



# lsb_release

LibreELEC (Milhouse): devel-20180218010737-#0217-gc2fd843 [Build #0217]



# Kodi version

(18.0-ALPHA1 Git:f5871d9). Platform: Linux x86 64-bit

Based on tip of LibreELEC.tv master (c2fd843, changelog) and tip of XBMC master (866f392, changelog) with the following modifications:

• Includes latest kodi-platform master (36fb493)
  
• Includes latest libcec master (8adc786, ahead +26)
  
• Includes latest libnfs master (54405d9, ahead +120)
  
• Includes latest p8-platform master (a822e19)
  
• Includes latest addons: inputstream.adaptive (b28d3f0), inputstream.rtmp (b3ffc49), peripheral.joystick (12e7d52, +1), pvr.argustv (a4d3ec7), pvr.demo (8f62053), pvr.dvblink (89d1687), pvr.dvbviewer (64c4356), pvr.filmon (2721611), pvr.hdhomerun (7379fbc), pvr.hts (095784b), pvr.iptvsimple (933a5c4), pvr.mediaportal.tvserver (dec444f), pvr.mythtv (3200b09), pvr.nextpvr (aa1c8d2), pvr.njoy (67b837d), pvr.octonet (689afbf), pvr.pctv (5b47d9f), pvr.stalker (77eec69), pvr.teleboy (3ae0297), pvr.vbox (57f3198), pvr.vdr.vnsi (4db947c), pvr.vuplus (f05f158), pvr.wmc (38e670a), pvr.zattoo (3c46173), vfs.libarchive (128394c)
  
• Include [env] compare (perma): kodi: reduce arm memory fragmentation
  
• Include [env] compare (perma): busybox: 2nd attempt at disabling busybox parallel build
  
• Include [env] compare (perma): dmidecode: initial package
  
• Include [env] patch: RPi/RPi2: enable Broadcom WiFi debugging (see details)
  
• Include [env] patch: libcec: fix PR390
  
• Include [env] patch: libcec: don't link non-existant libtinfo
  
• Include [env] patch: rev hack for kodi
  
• Include [env] patch: HACK: Disable multiple PVR addons during migration. Always enable inputstream.* and os.*
  
• Include [env] patch: Add experimental splash video for RPi
  
• Include [env] patch: Bump included addon versions to prevent online updates
  
• Include [env] patch: Add kodi binary addons (pvr, adsp, inputstream, vfs, other)
  
• Include [env] PR:2466 (perma): linux: update to linux-4.14.19 (-ish)
  
• Include [env] PR:2501 (perma): buildsystem: selectively build packages with debug
  
• Include [env] PR:2504 (perma): busybox: bump package to v1.28.1
  
• Include [pkg] patch: skin.estuary: reduce system info font size (kodi)
  
• Include [pkg] PR:13274 (perma): Mode whitelist
  

Build Highlights:

1. busybox: bump package to v1.28.1
   
2. [Estuary] - fix info display for sets
   
3. Python: Normalize Python exception before processing
   
4. RetroPlayer GL(ES)/GUI fixes
   

Build Details:

1. XBMC:
   • [Estuary] - fix info display for sets (PR:13530, 1 commit, 4 files changed)
     
   • RetroPlayer: GL(ES) don't hardcode pixelformats (PR:13504, 3 commits, 23 files changed)
     
   • Python: Normalize Python exception before processing (PR:13554, 1 commit, 1 file changed)
     
   • RetroPlayer: Move and document GUI code (PR:13553, 2 commits, 32 files changed)
     
   • [cleanup] remove unnecessary PlatformInclude.h & system.h includes (PR:13550, 4 commits, 295 files changed)
     
   • amlcodec: reduce video decoder pre-buffering time, pre-buffer more only when decoder is busy (PR:13541, 1 commit, 1 file changed)
     
2. peripheral.joystick:
   • [android] Add buttonmap for Xbox One S wireless controller (PR:139, 1 commit, 1 file changed)
     
3. Additional commits/pull requests/changes not yet merged upstream:
   • Added: [env] PR:2504 (perma): busybox: bump package to v1.28.1
     

[toasterboy1]

Anyone else having problems with DVD playback? Tried searching the thread but did not come up with anything. Currently on build 215 but updating to 217 as I type. Have tried with internal and USB drive. I do not currently have a debug log but can get one if it will help the project. Just do not want to take up space for something if everyone is aware it does not work.

Update: Gets hung up trying to load main menu. Turning off VDPAU seems to fix the problem.

[Milhouse]

Anyone else having problems with DVD playback? Tried searching the thread but did not come up with anything. Currently on build 215 but updating to 217 as I type. Have tried with internal and USB drive. I do not currently have a debug log but can get one if it will help the project. Just do not want to take up space for something if everyone is aware it does not work.

Update: Gets hung up trying to load main menu. Turning off VDPAU seems to fix the problem.

I can reproduce an issue with a DVD ISO test disc I use (streamed over smb:// or from local HDD, it makes no difference), which seems to be VDPAU HW decode related.

It's the Hitchhikers Guide to the Galaxy "extras" disc which consists of a two page menu, from which you can play various short videos.

With VDPAU HW decode enabled, videos played from the first menu page will end normally, returning to the menu, but extras played from the second page remain permanently on a black screen at end of playback and never return to the second menu page (GUI is still working, "stop" from remote will end playback). Disabling VDPAU HW decode fixes the issue. This issue is present in LE 8.2.3 so it's also in Kodi 17, is reproducible in Ubuntu with latest Kodi master and Radeon R7 250E with VDPAU HW enabled, so is not specific to these builds (or even LibreELEC). You might want to see when your VDPAU issue started.

Intel NUC with VAAPI HW encode enabled and lastest test build has no issue with either menu, everything is normal, so this seems specific to VDPAU HW decode.

While testing the same ISO on RPi2 with latest test build the second menu page is not working at all when either OMX or MMAL is enabled. I can navigate to it from the first menu page but cannot navigate within the second menu page - the menu selection cannot be changed ("STOP" on the remote will end "playback" of the menu). This problem starts with #0613 so probably due to the VideoPlayer changes as ffmpeg in LE didn't bump in #0613 (still ffmpeg-3.3.1).

If both OMX and MMAL are disabled then the second menu can be navigated successfully. The second menu is working normally with LibreELEC 8.2.3/Kodi 17.6, with OMX disabled/MMAL enabled, or both disabled, but not OMX enabled/MMAL disabled (second menu page appears, but menu item selector doesn't appear although menu can still be navigated blind).

Edit: On RPi2 with #0217, videos from the first menu page appear as a black screen (while audio can be heard) with MMAL enabled or both OMX and MMAL disabled. With OMX enabled and MMAL disabled, the video appears but is very stuttery. Odd.

I may try and upload this 7.5GB DVD ISO somewhere if anyone wants to test it/investigate (PM me, it will take a couple of days to upload unfortunately, and disk space is also at a premium).

[Martijn]

Milhouse
It should be possible to rip out the video parts perhaps and you only keep the troubled menu's.
Else you can probably use team FTP so our Devs can look where and why it fails

[HeresJohnny]

This doesn't really belong here, but since the team advocates the switch from default MySQL to MariaDB I tried it out for myself. What can I say, MariaDB is much more responsive. When a longer query is active MariaDB does a much better job of committing the results from a queue. With MySQL the whole server got swamped more often than not, resulting in high CPU and unresponsiveness. When exiting Kodi, most of the times it just crashed. This might have been due to tinkering with the my.ini options, but anyway, the default settings of MariaDB are already really good. When the query queue gets too large it looks like MariaDB slows down everything so no locks occur. Quitting Kodi really works now, so I know the database has received all updates.

[fab73]

Hello,

with build #0215 on an Intel NUC NUC7I3BNH https://www.intel.com/content/www/us/en/...i3bnh.html
a SONY OLED A1 TV
and a Sony STR-DN1080 receiver

I have some issues :
- long audio cuts/drops (I don't know the right word) when playing videos with Dolby Atmos track with passthrough enable
- white screen when playing HDR 10 bits videos

And #0214 doesn't have this issue?  

 Hello,

I have the same issue of white screen with some HDR 10bits videos on Asrock J4205 since build #210.
No worries with build #209.

[Milhouse]

I've uploaded a reduced version of my test DVD ISO:

http://milhouse.libreelec.tv/other/HITCH..._REDUX.ISO (1.1GB)

This includes the two menus, plus 3 small videos that are playable (tested with VLC on Windows, and LE9 on Skylake NUC with VAAPI):

Video #1: (Menu 1, The Inner Planets) -> Behind the Scenes
Video #2: (Menu 2, The Outer Planets) -> Pebble Mill at One
Video #3: (Menu 2, The Outer Planets) -> Deleted scene from Episode 2

All other items will be unplayable.

Kodi problems with this ISO:

x86_64 with VDPAU HW decode (confirmed with LibreELEC 8.2.3/Kodi 17.6, LibreELEC 9.0/Kodi 18, Ubuntu/Kodi 18 with Radeon R7 250E):

1. Playback video #2 and video #3 (from menu 2):
VDPAU HW decode enabled: at the end of playback menu 2 will not reappear - just a persistent black screen
VDPAU HW decode disabled: menu 2 appears as expected

I'll post about the RPi issues in the RPi test thread.

[futuretristan]

First, thanks for this amazing software package! I want to use LibreELEC to watch TV, to stream Amazon prime contents and as a Spotify client and all of this works pretty well. However, I installed #0216 this weekend and encounter some problems.

Hardware facts: Self-build HTPC based on the J4205-ITX platform with a TBS 6281 PCIe-card to watch DVB-C.

Starting with a fresh LibreELEC 8 installation, I upgraded to #0216 by placing the corresponding archive file into the update folder and rebooted. I then installed the crazy cat driver for the TBS card and Tvheadend server as well as client. Live TV works very well, however I have three questions/problems.

1)
When I suspend LibreELEC to standby instead of powering down completely, then after wake up the Tvheadend server is not running/working any more.

2)
When I set-up Kodi such that it automatically starts live TV when powering up, then live TV works (I can also change channels), but no EPG information is shown and neither samba nor ssh access is possible. Only when I stop live TV by hitting the stop button, EPG information becomes available, however samba and ssh do not start working. I have to manually set-up Kodi to not start live TV when booting and do a reboot to make samba and ssh work again.

3)
To save some time when booting LibreELEC, I activated the ultra fast boot mode in UEFI settings. Booting indeed is some seconds quicker now. However, I cannot access UEFI setting any more (as expected). In Windows there is an option to reboot into UEFI and in Ubuntu you can reboot into UEFI with the command "systemctl reboot --firmware-setup". Is there also some solution available in LibreELEC (the latter command does not work)?

[gwilly7]

Hi Milhouse,

Thank you for all of your hard work. I built a new system built on AMD's new Raven Ridge this weekend and I was wondering if any of your builds might have drivers for these chips yet?

[Milhouse]

Hi Milhouse,

Thank you for all of your hard work. I built a new system built on AMD's new Raven Ridge this weekend and I was wondering if any of your builds might have drivers for these chips yet?

The drivers are all present, it just depends if the drivers have been updated to support the various PCI IDs of the new hardware - it might work, so worth a try.