• 1
  • 38
  • 39
  • 40(current)
  • 41
  • 42
  • 111
OpenELEC Testbuilds for RaspberryPi (Kodi 15.0) Part 1
Is this not the 5VDC hardwired "Status" led?

On all my 4 Rpi's since 2014 have only one Option - switch off the Powersupply an nothing red will burn :-) sorry, ther's in germany "fastnacht" -> Hellau ;-)


ok,ok, i have build for some times ago a little hardwareSwitch, out of the Box for starting the Pi and also to turn off with a little sript (over gpio Pins) - but with testbuilds it's now no Option for me - to much work...not stable ;-)

btw: http://www.gtkdb.de/index_36_2238.html

http://i53.photobucket.com/albums/g79/re...pi_psu.jpg
Kinda don't want to make a new thread, but I just got a Pi2 and tried using the 5.03 build for the Armv7 builds, and when I power on I get no image. It's a completely new install. Any suggestions?
(2015-02-15, 22:08)doldi Wrote: Is this not the 5VDC hardwired "Status" led?

On all my 4 Rpi's since 2014 have only one Option - switch off the Powersupply an nothing red will burn :-) sorry, ther's in germany "fastnacht" -> Hellau ;-)
Helau and alaaf!
Good ideaBig Grin Will check this after Tuesday (end of Fasching in Bavaria).
(2015-02-15, 22:39)Abshole Wrote: Kinda don't want to make a new thread, but I just got a Pi2 and tried using the 5.03 build for the Armv7 builds, and when I power on I get no image. It's a completely new install. Any suggestions?

I have having a similar problem too. I have a RPi2 with the newest OpenElec installed using the instructions from Kodi's homepage (using Win32DiskImager and the image file)

In general, everything works great, but like reported here, it hangs upon a reboot or power on. It hangs on the OpenElec screen. After rebooting once or twice, it works and boots like normal and everything works. Are there any changes I could make to the config.txt or anything to fix this?

I am new to the RPi / OpenElec world, so if there is further information needed to help diagnose this, please let me know.

Thanks.
(2015-02-15, 23:41)kcsims Wrote:
(2015-02-15, 22:39)Abshole Wrote: Kinda don't want to make a new thread, but I just got a Pi2 and tried using the 5.03 build for the Armv7 builds, and when I power on I get no image. It's a completely new install. Any suggestions?

I have having a similar problem too. I have a RPi2 with the newest OpenElec installed using the instructions from Kodi's homepage (using Win32DiskImager and the image file)

In general, everything works great, but like reported here, it hangs upon a reboot or power on. It hangs on the OpenElec screen. After rebooting once or twice, it works and boots like normal and everything works. Are there any changes I could make to the config.txt or anything to fix this?

I am new to the RPi / OpenElec world, so if there is further information needed to help diagnose this, please let me know.

Thanks.

I tried a different SD card and got everything to work properly.. Kinda sucks because the first card I tried was a brand new one. Hopefully it'll work with Raspbian.
Any chance that your sd card is a current Samsung one?
I've had huge problems with a brand new 16 GB Samsung Evo on my RPi2 - even without overclock and using 4 different chargers (10A multi-charger with high current port, 2A samsung charger, 1.2A charger from amazon, 1.5A charger from samsung) I encounter system corruption every 2 to 3 boots.
After switching to a 64GB sandisk ultra micro sd I've used in my phone previously all my problems are gone.
I think there has been some acknowledgement on the official boards that current Samsung cards might cause problems, although the can't reliably reproduce it yet.
Not sure if that's related but I'd try different (old) cards if you have problems booting your RPi using new cards.
Yes, I'm using a Samsung 16gb class 6 card (http://www.amazon.com/gp/product/B00IVPU...UTF8&psc=1)

Don't think I need a 64gb card, but would any sandisk be fine? For just running kodi, would anything larger than 8gb be recommended?

Thanks.
(2015-02-16, 01:18)Cy4n1d3 Wrote: Any chance that your sd card is a current Samsung one?
I've had huge problems with a brand new 16 GB Samsung Evo on my RPi2 - even without overclock and using 4 different chargers (10A multi-charger with high current port, 2A samsung charger, 1.2A charger from amazon, 1.5A charger from samsung) I encounter system corruption every 2 to 3 boots.
After switching to a 64GB sandisk ultra micro sd I've used in my phone previously all my problems are gone.
I think there has been some acknowledgement on the official boards that current Samsung cards might cause problems, although the can't reliably reproduce it yet.
Not sure if that's related but I'd try different (old) cards if you have problems booting your RPi using new cards.

This is the one I first tried. I got OpenELEC to work, but I have this small colored square in the upper right hand corner. Not sure what it's from.
Please, start a separate thread for 5.0.3 problems. Or post on the OpenELEC forum. If it's not related to a test build, it's OT discussion and just spamming everyone subscribed to this thread. Thank you.
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
New OpenELEC I****** build #0215: RPi / RPi2
(Supercedes previous build)

Code:
# uname -a
Linux rpi512 3.19.0 #1 Sun Feb 15 21:02:00 GMT 2015 armv6l GNU/Linux

# vcgencmd version
Feb 14 2015 22:20:17
Copyright (c) 2012 Broadcom
version 7789db485409720b0e523a3d6b86b12ed56fd152 (clean) (release)

# lsb_release
OpenELEC (Milhouse) - Version: devel-20150215210110-r20281-g2b2d44a [Build #0215]

# vcdbg log msg 2>&1 | grep DTOK
001531.587: Kernel trailer DTOK property says yes

# Kernel device tree status: Enabled

Based on tip of OpenELEC master (2b2d44aa, changelog) and tip of XBMC master (64621811, changelog) with the following modifications: Build Highlights:
  1. New firmware
  2. Build with gcc armv7ve support
  3. Update ffmpeg to 2.5.4
  4. Updated kodi-pvr-addons
  5. RPi2: Disable omxplayer by default
Build Details:
  1. Firmware (Feb 14):
    • firmware: arm_loader: Avoid race condition that can cause an over temp warning on boot
    • firmware: arm_loader: Remove hack for sdram and turbo mode
  2. OpenELEC:
    • WeTek update (PR:3916, 3 commits, 22 files changed)
    • GCC 4.9 has support for armv7ve (PR:3894, 1 commit, 1 file changed)
    • projects/*/linux: enable CONFIG_UHID, this fixes #3514 (3ee63df0)
    • ffmpeg: update to ffmpeg-2.5.4 (2fb0892a)
  3. XBMC:
    • Cleanup of legacy upgrade code pre-Frodo (PR:6423, 2 commits, 1 file changed)
    • FFmpeg: Bump version to 2.5.4-Isengard-alpha (PR:6430, 1 commit, 1 file changed)
    • [AE] clean up - drop enumeration for vanished engines (PR:6431, 1 commit, 2 files changed)
    • Bump droid api to 17 (PR:6425, 1 commit, 7 files changed)
    • FIX: [gles;amc;deint] invert fields because pic is inverted (PR:6363, 1 commit, 1 file changed)
    • FIX: [amc] proper fix to overbuffer (PR:6109, 1 commit, 2 files changed)
    • [omxplayer] Remove test of removed setting, avoids a log message (PR:6443, 1 commit, 1 file changed)
    • [Confluence] cosmetics (3a2a417d)
    • [skin] update re-touched (3a1fa229)
    • [osx/ios/atv2] - kill 300+ warnings about unused variables by disabling the build of all the neptune tls keys we don't make use of (c4a440d8)
    • [osx/ios/atv2] - sync xcode project after libsquish was moved to depends - fixes linker warning (f22685a5)
    • [ios] - fix 2 more warnings about unhandled enum value in switch case (b52084b1)
  4. kodi-pvr-addons:
    • set socket to invalid if open fails, fixes false positive IsOpen respons... (PR:429, 1 commit, 1 file changed)
    • [pvr.mythtv] fix compile error wir c++11 (PR:436, 1 commit, 1 file changed)
    • platform posix: fix memory leak in pthread (PR:435, 1 commit, 1 file changed)
    • [pvr.hts] Fixed typo: load defaults in ADDON_ReadSettings(void) (PR:428, 1 commit, 1 file changed)
    • [pvr.hts] Fixed CTvheadend::GetRecordings(...) return value. (PR:424, 1 commit, 1 file changed)
    • [pvr.hts] fix extern "C" scope in client.cpp (PR:437, 1 commit, 1 file changed)
    • [pvr.hts] Fix: mark recordings, schedules and events as "del" on (re)conect (PR:423, 1 commit, 1 file changed)
  5. newclock4:
    • New commits in this build:
      • [mmalcodec] Reduce number of extra video buffers (e8b1e0ce)
      • [mmalrenderer] Ensure we have requested number of buffers in non-accelerated case (9f861b6d)
      • [mmalrenderer] Render new from frame from RenderUpdate rather than FlipPage (487c6dc1)
      • [mmalrenderer] Avoid grabbing the g_graphicsContext lock - it can deadlock (85ad72a0)
      • [rbp] Add Pi 2 specific settings defaults (5795c59f)
    • Commits no longer in build:
      • [mmalcodec] Try reducing number of video frames (d4ec2a5a)
      • [mmalcodec] Ensure we have requested number of buffers in non-accelerated case (7787d4ba)
      • [MMALCoded] Call port settings changed from callback to avoid thread unsafe code (c180884c)
      • [mmalcodec] Avoid double framerate deinterlace for dvd menus for now (b33567c5)
      • [mmalcodec] squash: Avoid deinterlace for dvd menus for now (7ccadf42)
      • Revert "[MMALCoded] Call port settings changed from callback to avoid thread unsafe code" (c1b5b75a)
      • [omxplayer] Remove test of removed setting, avoids a log message (e1cb964e)
      • [mmalcodec] Restrict number of frames pushed to GPU to 16 (7116642a)
      • Revert "[mmalcodec] Restrict number of frames pushed to GPU to 16" (08dd97aa)
      • test: Render new from frame from RenderUpdate rather than FlipPage (809caebe)
      • MMALRenderer: Avoid grabbing the g_graphicsContext lock - it can deadlock (a65f95d5)
      • [dvdcodec] Use libmpeg2 when mpg2 licence not available to improve software decoded dvd menus (fb0308b8)
  6. kernel 3.19.y:
    • New commits in this build:
      • platform: Remove bcmrpi_small_defconfig (9883079d)
      • platform: Remove bcmrpi_sdcard_defconfig (d5a3db1a)
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
I already mentioned it in my PR6235 which is part of newclock4 branch - it appears that monoscopic 3D playback is broken atm for both, PI1 and PI2. SBS/OU is working fine.
Quote:New commits in this build:
  • [rbp] Add Pi 2 specific settings defaults (5795c59f)
Interesting - I have tested this.

I'm using the Amber skin, it shows the plot in an autoscrolling textbox when pressing "i" while playing.
With default rbp2 settings (limit GUI fps to 10, OMXPlayer disabled) that textbox is scrolling so fast that you can't read it. Disabling the GUI fps limiter (setting it to "Off") "slows" it down to normal speed.
OMXPlayer doesn't show this issue (in fact I had it set to 10 already).
I guess it's the issue you were discussing in PR 6090 and you haven't found a proper solution yet.
(2015-02-16, 02:30)Abshole Wrote:
(2015-02-16, 01:18)Cy4n1d3 Wrote: Any chance that your sd card is a current Samsung one?
I've had huge problems with a brand new 16 GB Samsung Evo on my RPi2 - even without overclock and using 4 different chargers (10A multi-charger with high current port, 2A samsung charger, 1.2A charger from amazon, 1.5A charger from samsung) I encounter system corruption every 2 to 3 boots.
After switching to a 64GB sandisk ultra micro sd I've used in my phone previously all my problems are gone.
I think there has been some acknowledgement on the official boards that current Samsung cards might cause problems, although the can't reliably reproduce it yet.
Not sure if that's related but I'd try different (old) cards if you have problems booting your RPi using new cards.

This is the one I first tried. I got OpenELEC to work, but I have this small colored square in the upper right hand corner. Not sure what it's from.

The small rainbow square is an indicator to say that your pi is not getting enough current. Try with a better charger.
I have a b+ and a pi2 both using 16gb Samsung evo cards running at 1ghz without issues. (They are cron jobbed to reboot every night)
Maybe I'm just lucky but I've seen no such corruption issues using the Samsung cards.
New OpenELEC I****** build #0216: RPi / RPi2
(Supercedes previous build)

Code:
# uname -a
Linux rpi512 3.19.0 #1 Mon Feb 16 21:03:26 GMT 2015 armv6l GNU/Linux

# vcgencmd version
Feb 14 2015 22:20:17
Copyright (c) 2012 Broadcom
version 7789db485409720b0e523a3d6b86b12ed56fd152 (clean) (release)

# lsb_release
OpenELEC (Milhouse) - Version: devel-20150216210235-r20283-g5c70a3c [Build #0216]

# vcdbg log msg 2>&1 | grep DTOK
001549.639: Kernel trailer DTOK property says yes

# Kernel device tree status: Enabled

Based on tip of OpenELEC master (5c70a3c0, changelog) and tip of XBMC master (11ca9830, changelog) with the following modifications: Build Highlights:
  1. Potential fix for libcdio when ripping CDs (see discussion)
  2. Added: PR:6453: webserver: add support for WSGI
Build Details:
  1. OpenELEC:
    • Update 98-eventlircd.rules (PS3 BD remote hardware Version #2) (PR:3918, 1 commit, 1 file changed)
  2. XBMC:
    • [omxcore] Skip out of flush on error (PR:6446, 1 commit, 1 file changed)
    • [linux] fix building with gcc 4.9 on i386 (PR:6440, 1 commit, 1 file changed)
    • [skin.re-touched] fix win32 build (PR:6458, 1 commit, 1 file changed)
    • [resamplepi] Try to report the same numbers as ffmpeg (PR:6449, 1 commit, 1 file changed)
    • [omxplayer] Add some info about hdmi sync to codec overlay (PR:6445, 1 commit, 3 files changed)
    • [omxplayer] Avoid extra frame allocation when deinterlace might be enabled (PR:6444, 1 commit, 1 file changed)
    • [TexturePacker] - fix compilation with c++11 (0c2e2418)
  3. kodi-pvr-addons:
    • Increase API version to 1.9.4 and add related changes (PR:427, 9 commits, 55 files changed)
    • [pvr.mythtv] Release v2.0.0 (PR:432, 3 commits, 380 files changed)
    • [pvr.mythtv] fix addon.map since 1.9.4 (PR:439, 2 commits, 3 files changed)
  4. newclock4:
    • New commits in this build:
      • - allow reads < CDIO_CD_FRAMESIZE_RAW by using a buffer - fixes #15794 (ea2e2ff5)
      • Revert "don't back out from 3D handling on resolution changes if we're switching to a different 3D mode" (77048e0e)
      • [graphics] Don't set stereo mode based on resolution (b58b4b74)
    • Commits no longer in build:
      • [omxcore] Skip out of flush on error (6e034150)
      • [resamplepi] Try to report the same numbers as ffmpeg (d3a1fde1)
      • [omxplayer] Add some info about hdmi sync to codec overlay (c87c06b0)
      • [omxplayer] Avoid extra frame allocation when deinterlace might be enabled (8dbbaaeb)
  5. kernel 3.19.y:
    • New commits in this build:
      • Create bmp085_i2c-sensor-overlay.dts and update Makefile (6e8273a7)
      • Merge pull request #831 from h616/rpi-3.19.y (33ae3fbd)
      • lirc-rpi: use getnstimeofday instead of read_current_timer (a8f400dc)
  6. Additional commits/pull requests/changes not yet merged upstream:
    • Added: PR:6453: webserver: add support for WSGI
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
Hi!

From version 20272 to build 20281 something must be changed since LiveTV when not in full screen starts to jitter (runs not fluent, it runs but stops every few seconds and then runs
and again stops and so on) every few seconds. I have also noted that the sound can get corrupted from this build on LiveTV. At first i thought it is network but then i have started to find the builds where this has started and the result is that it was not present
in build 20272 but has started with build 20281 and the same behavior is still present in latest build.

Debug log from 20281: http://pastebin.com/DBqWL9bT
Another debug from 2081: with error present: http://pastebin.com/rRpS8H6x
Debug log from 20272 with perfect working LiveTV: http://pastebin.com/z3PuXmTJ
  • 1
  • 38
  • 39
  • 40(current)
  • 41
  • 42
  • 111

Logout Mark Read Team Forum Stats Members Help
OpenELEC Testbuilds for RaspberryPi (Kodi 15.0) Part 112