2018-05-30, 22:47
2018-05-31, 00:51
New LibreELEC.tv Leia build #0530: RPi / RPi2
(Supercedes previous build)
SHA256 Checksum:
SHA256 Checksum:
Based on tip of LibreELEC.tv master (e1f38c5, changelog) and tip of XBMC master (d0cc291, changelog) with the following modifications:
(Supercedes previous build)
SHA256 Checksum:
9af8db423c21125caad5551cd2e89ded6a1b0ad578244f105bad2b10a5f4a0aa
(RPi)SHA256 Checksum:
5067d2daea221a0a1999a1a3898fa4fbf66afe43d5df9904c2d63180dceae844
(RPi2)text:# uname -a
Linux rpi512 4.14.44 #1 Wed May 30 21:10:55 BST 2018 armv6l GNU/Linux
# vcgencmd version
May 25 2018 14:36:49
Copyright © 2012 Broadcom
version 1d00025257e46ea069b2a0f59f0f443b2be2124e (clean) (release)
# lsb_release
LibreELEC (Milhouse): devel-20180530210545-#0530-ge1f38c5 [Build #0530]
# Kodi version
(18.0-ALPHA2 Git:795b10a). Platform: Linux ARM 32-bit
Based on tip of LibreELEC.tv master (e1f38c5, changelog) and tip of XBMC master (d0cc291, changelog) with the following modifications:
- Includes newclock5 patches
- Includes latest bcm2835-driver master (784fe6c)
- Includes latest kodi-platform master (e8574b8)
- Includes latest libcec master (8adc786, ahead +26)
- Includes latest libnfs master (6a33413, ahead +132)
- Includes latest p8-platform master (a822e19)
- Includes latest addons: inputstream.adaptive (d4a9954), inputstream.rtmp (b8e3f39), peripheral.joystick (80224e1), pvr.argustv (2bce465), pvr.demo (20d81d8), pvr.dvblink (a87258b), pvr.dvbviewer (884b732), pvr.filmon (470ca1c), pvr.hdhomerun (4639740), pvr.hts (4f7196d), pvr.iptvsimple (e220777), pvr.mediaportal.tvserver (c4e32b0), pvr.mythtv (9d15025), pvr.nextpvr (78a80de), pvr.njoy (5a2c2d3), pvr.octonet (54680a3), pvr.pctv (17c1897), pvr.stalker (0700069), pvr.teleboy (3e9e537), pvr.vbox (48ffcba), pvr.vdr.vnsi (a2880c7), pvr.vuplus (6c94eec), pvr.wmc (a7ec576), pvr.zattoo (f04367b), vfs.libarchive (84a4876)
- Exclude [env] kodi-0100-temp-add-pr13594.patch: merged upstream
- 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: kodi: remove annoying, excessively repetitive addon logging
- 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:2449 (perma): Adding support for AMD GPUs
- Include [env] PR:2680 (perma): kodi: mid-May 2018
- Include [env] PR:2696 (perma): linux: update ouya patch
- Include [env] PR:2722 (perma): fix "code may be misoptimized" warnings
- Include [env] PR:2725 (perma): Python2: update to Python2-2.7.15, bump Python-related packages
- Include [env] PR:2729 (perma): linux: update to linux-4.14.44
- Include [pkg] patch: skin.estuary: reduce system info font size (kodi)
- Include [pkg] PR:13904 (perma): Fix whitelistable resolutions on some platforms
- Fix OMX stutter
- ffmpeg: hevc: Fix performance regression + latest ben optimisations
- LibreELEC.tv:
- scripts/image: fix noobs tarball (PR:2732, 1 commit, 1 file changed)
- scripts/image: fix noobs tarball (PR:2732, 1 commit, 1 file changed)
- XBMC:
- Fixed: use fadelabel instead of textbox for Estuary notications so th (PR:13929, 1 commit, 1 file changed)
- Fixed: use fadelabel instead of textbox for Estuary notications so th (PR:13929, 1 commit, 1 file changed)
- newclock5:
- Additional commits/pull requests/changes not yet merged upstream:
2018-05-31, 05:15
I've noticed an issue with the list of recorded TV shows. If I watch a few minutes of a show and press stop, the item disappears from the list. Group Items has to be disabled for the item to disappear. The recorded show is visible if Group Items is enabled.
I'll have to jump back through the builds to find when this first occurred and grab a log file too when it happens.
I'll have to jump back through the builds to find when this first occurred and grab a log file too when it happens.
2018-05-31, 06:17
(2018-05-06, 00:47)Milhouse Wrote:Sorry, I missed your reply. I wouldn't have been able to "journalctl -a | pastebinit" anyway as I had no remote access when it was in this state. I believe I can access the SD card in a Linux Virtualbox to run fsck.ext4 but it's been a while since I last did that so I'll have to refresh my memory for next time. This time I ended up just formatting the 2nd partition using a Windows partitioning program and rebuilding the databases but I've made a Kodi backup now (and copied it to my PC) so it should be less painful next time. I've bought a new SD card anyway as this one does seem to corrupt a bit too easily and I've also bought a RPI 3 B+ so I'll be transferring from my RPI 2 to that soon.(2018-05-05, 20:52)doveman2 Wrote: What's going on when I boot and it says "A start job is running to create volatiles files and directories" (I might not have got that exactly right as it's scrolled off the screen now)?
Last time I went to use my Pi it was frozen and non-responsive, so I rebooted it using the cmdline on my PC (to avoid corrupting the SD by doing a hard reset) but I got this message and it wouldn't boot.
I've just overwritten the files on the PC-accessible partition with those from #504 but I'm still getting this message. When I've had this problem in the past I've had to start from scratch and write a new image to the SD to clear the linux partition.
It seems to have done a bit more than it has in the past as it's thrown up a load of messages, the last of which is "Starting Avahi defaults" but that's been there for several minutes now so I don't think it's going to boot, so will I just have to wipe the card again?
Not really sure, that's the log message from a systemd unit that you wouldn't normally see unless the unit has failed or a prior unit has failed. Can you provide a link to [inline]journalctl -a | pastebinit[/url]. It could be you have some corrupt files or a corrupted file system in the second partition - if you are able to run fsck.ext4 on the second partition it should tell you if the file system is clean or not, and hopefully also fix any issues.
2018-05-31, 10:47
When I turn off the Raspberry PI 3 model B+, the USB ports are not disabled (0530 and earlier). My external USB DAC stopped shutting down automatically. Although with 8.2.5 everything works - DAC will turn off automatically after switching off in the 8.2.5.
But when measuring the voltage on the USB with a multimeter in the off state "0", in the switched on "+5V". Something does not allow the ports to completely shut down. On Raspberry PI 3 also works as it should.
But when measuring the voltage on the USB with a multimeter in the off state "0", in the switched on "+5V". Something does not allow the ports to completely shut down. On Raspberry PI 3 also works as it should.
2018-05-31, 13:24
(2018-05-30, 20:24)popcornmix Wrote:Hi @popcornmix & Ben,(2018-05-30, 16:42)bleep42 Wrote: To reiterate, I am quite frankly stunned at how well HEVC plays on a Pi3, even a Pi2, when I think of how it was a year or so ago, and how everyone said it would never work without pure hardware decoding, all due to your and your teams fantastic work. Thank you.
If you do, or don't manage to claw back the extra few percent, is academic, as I would suspect, that the majority of people, using sensible bitrate files, it works admirably. :-)
Thank you. Kevin.
Looks like it was a bug in ffmpeg 4 that stopped inline assembly from being used: http://ffmpeg.org/pipermail/ffmpeg-devel...30732.html
Hopefully will be fixed in tonight's build.
Wow... I now see a 20 to 25% improvement in decoding efficiency for HEVC, my test file that was maxing out yesterday is now peaking at just a tad over 300% (out of 400%) or 80% per CPU (Pi3B+).
That is amazing :-) where yesterday my test file would overheat the CPU in a few minutes, today it's not quite managing to get it to 70C and even when it does, by me insulating the CPU a bit, to force it to throttle to 1200MHz, the file still plays perfectly. :-) I'm impressed. That's the highest bitrate test file that I have, I'll have to generate something even more tricky now. ;-)
Thanks to everyone for all their hard work.
Kevin.
2018-05-31, 13:37
PS. I just tried the 10bit Jelly fish files and none of them will play past the first few frames, the play bar along the bottom continues but the picture freezes, #0530.
Personally I don't use, or see any dramatic improvement for 10bit, so I wouldn't normally test it.
Regards,
Kevin.
Personally I don't use, or see any dramatic improvement for 10bit, so I wouldn't normally test it.
Regards,
Kevin.
2018-05-31, 14:05
@Milhouse
Due to LE crashes with the addon zattoo PVR: How do I enable stack trace on a RasPi2?
Thx for your support,
oliaros
Due to LE crashes with the addon zattoo PVR: How do I enable stack trace on a RasPi2?
Thx for your support,
oliaros
2018-05-31, 15:00
I think I've fixed this issue. I went into the System settings > Audio and changed Resample Quality from 'GPU accelerated' to 'Low (fast).' This seems to have cured the problem, although I'm not sure why as I thought this setting only really applied to video.
2018-05-31, 15:02
Quote:BUG?: Tiny dropout at beginning of music track
On the latest nightly, I'm playing an album of FLAC-encoded music files, streamed from a NAS via ethernet. At the beginning of most (but not all) songs, there's a tiny dropout just after the track starts.
I think I've fixed this issue. I went into the System settings > Audio and changed Resample Quality from 'GPU accelerated' to 'Low (fast).' This seems to have cured the problem, although I'm not sure why as I thought this setting only really applied to video.
2018-05-31, 15:52
(2018-05-30, 06:39)garbear Wrote:I am not using IARL. I just have a folder of ROMS on my NAS.Quote:Retroplayer is still crashing (since #402). #401 is the last working version Crash Log@quig This should be fixed by https://github.com/zach-morris/plugin.pr...rl/pull/52 . Can you update to the latest IARL?
21:55:29.897 T:1940258832 ERROR: Failed to open nfs://192.168.1.2/Multimedia/ROMs/SNES/Chrono Trigger (U) [!].xml: Failed to open file
It looks like it is changing the file extension from .smc to .xml
ping @garbear
2018-05-31, 18:37
(2018-05-31, 14:05)oliaros Wrote: @Milhouse
Due to LE crashes with the addon zattoo PVR: How do I enable stack trace on a RasPi2?
Thx for your support,
oliaros
See the "Bug reports" section in the first post, or testing note #4 also in the first post. A crash log with stacktrace is (in most cases) generated automatically following a crash.
2018-05-31, 18:40
(2018-05-31, 15:42)tjost Wrote: log
listen to music
Hmmm... another "Busy dialog" related crash.
There are more fixes under development that should address at least some of the remaining issues.
2018-05-31, 21:17