Kodi Community Forum
OpenELEC Testbuilds for RaspberryPi Part 2 - Printable Version

+- Kodi Community Forum (https://forum.kodi.tv)
+-- Forum: Support (https://forum.kodi.tv/forumdisplay.php?fid=33)
+--- Forum: General Support (https://forum.kodi.tv/forumdisplay.php?fid=111)
+---- Forum: Raspberry Pi (https://forum.kodi.tv/forumdisplay.php?fid=166)
+---- Thread: OpenELEC Testbuilds for RaspberryPi Part 2 (/showthread.php?tid=184866)



RE: OpenELEC Testbuilds for RaspberryPi Part 2 - popcornmix - 2014-03-11

(2014-03-10, 23:18)ijsbeer79 Wrote: Running overclocked mode. Can test more tomorrow: what do you suggest: test with 'older' build or remove overclock?

Jonathan says the mask=2 contains most of the complicated code, so removing that bit is most likely to avoid the problem (but also avoid most of the benefit).
The mask=0 failure is more surprising.

Perhaps remove overclock, and check mask=0 case a few times. Try to confirm if it is consistently worse than mask=1 or mask=4 (I'd predict it to be the same).


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - allan87 - 2014-03-11

Is this new kernel also in the Openelec 4.0 beta?


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - tinnyskillz - 2014-03-11

Is it me, or is wireless and wired connection at startup beginning real slow (connects like two minutes after boot/ sometimes not at all until a reboot or two)?


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - popcornmix - 2014-03-11

(2014-03-09, 23:25)RichG Wrote: Testing with two files:
If I play file one then stop and restart it - no noise burst
Play file two, stop, restart - sometimes get noise
Play file one then start file two - loud burst
Play file two then start file one - quieter burst most times, occasionally louder.

Thanks for info. A fix has been pushed to newclock3.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - popcornmix - 2014-03-11

(2014-03-11, 02:05)allan87 Wrote: Is this new kernel also in the Openelec 4.0 beta?
I believe so.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - Milhouse - 2014-03-11

(2014-03-11, 02:05)allan87 Wrote: Is this new kernel also in the Openelec 4.0 beta?

Pretty much - the only kernel patch not present in OE 4.0 Beta is the accelerated support of high-speed isochronous transactions that appeared in #0307. Everything else is the same, kernel-wise, including the support for FIQ FSM which is enabled by default in OE 4.0 Beta.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - Milhouse - 2014-03-11

New OpenELEC Gotham build: #0311
(Supercedes previous build)

Code:
# uname -a
Linux rpi512 3.13.6 #1 PREEMPT Tue Mar 11 00:19:44 GMT 2014 armv6l GNU/Linux

# vcgencmd version
Mar 10 2014 21:47:27
Copyright (c) 2012 Broadcom
version 2375441df366a702b88684435fa0df35497342eb (clean) (release)

# lsb_release
OpenELEC (Milhouse) - Version: devel-20140311001839-r17885-ge97abed

Based on tip of XBMC master (d49c4fe, changelog) and tip of OpenELEC master (e97abed, changelog) with the following modifications:
  • Includes newclock3 commits (except for 3653a4a which I've replaced with a static spinner to avoid hammering the GUI)
  • Includes kernel 3.13.y-next commits (see below for details on enabling FIQ_FSM patch)
  • Excludes the OpenELEC fernetmenta patches (due to conflict with newclock3)
  • Default setting for "Show RSS Feed" changed to disabled
  • Disabled "Total Duration" in Confluence (see build #0221 for details)
  • ALSA is enabled and any suitable hardware should be available in XBMC Audio Settings
Build Highlights:
  1. New Firmware Mar 10:
    • firmware: audio_render: add support for float samples and non-power-of-two channels
    • firmware: hdmi: Allow hdmi channel map to be overridden with a gencmd
    • firmware: audioplus: limit sample rates to ones supported by hardware
    • firmware: mailbox: Add property to get memory handle from dispmanx resource
    • firmware: Allow interrupts to be masked from GPU (e.g. when arm is handling them)
    • firmware: memory reduction of cutdown firmware (saves about 1M)
  2. OpenELEC:
    • Default to paplayer for music playback
    • PR3677 has been reverted in OpenELEC master due to crash problems with buggy skins, but remains present in this build as the risk is low
    • lirc updates to fix issue with Pi GPIO support
  3. XBMC:
    • Several music, video and MySQL database fixes, plus fixes here and there to other areas. See changelog for details.
  4. newclock3:
    • Merged into XBMC master: "[PiSink] Report all hdmi samples rate as supported. Required for passthrough"
    • Add AE_FMT_FLOAT (and AE_FMT_S32) to Pi sink, which hopefully reduces CPU when audio is already in that format. Also avoids expanding audio from 6->8 channels (GPU previously only supported powers of 2 channels). Requires updated firmware. The channel ordering is still not right for multichannel PCM - waiting for more information on that, but the plumbing is in place so should be a small change when I [popcornmix] get the answer.
    • Fix for white noise at start of files (omxplayer)

Additional Testing Notes:
  1. Testers should try adding the following entry to their advancedsettings.xml:
    Code:
    <advancedsettings>
      <video>
        <defaultplayer>dvdplayer</defaultplayer>
        <defaultdvdplayer>dvdplayer</defaultdvdplayer>
      </video>
    </advancedsettings>
    and report if it is better/worse than omxplayer. You can still play files with omxplayer using the context-menu "Play using... OMXPlayer".

  2. The following settings are no longer required in config.txt and should be removed:
    Code:
    no_hdmi_resample=1
    hdmi_stream_channels=1
    no_resample_audio is now a default, and hdmi_stream_channels is switched based on audio content. For the time being when using passthrough, 2.0 speaker layout should continue to be used (you will still get 5.1 with AC3/DTS).

  3. The FIQ_FSM patch is now enabled by default in OpenELEC master. See the FIQ_FSM announce thread for details.

    One new feature, currently disabled by default, is accelerated support of high-speed isochronous transactions (webcams, real time devices, etc.). There's a possibility this could be used by ALSA or DVB modules (although not always) which may affect some users (hopefully for the better). If you might benefit from this, enable by adding the following option to the end of the line in your /flash/cmdline.txt file:
    Code:
    dwc_otg.fiq_fsm_mask=0x7



RE: OpenELEC Testbuilds for RaspberryPi Part 2 - RichG - 2014-03-11

(2014-03-11, 02:27)popcornmix Wrote:
(2014-03-09, 23:25)RichG Wrote: Testing with two files:
If I play file one then stop and restart it - no noise burst
Play file two, stop, restart - sometimes get noise
Play file one then start file two - loud burst
Play file two then start file one - quieter burst most times, occasionally louder.

Thanks for info. A fix has been pushed to newclock3.

Thanks for the fix. Will try the new build later.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - robe1 - 2014-03-11

Hi popcornmix

As suggested I tried the latest Milhouse build to stop the film pauses with DTS passthrough enabled and it did not work, still had long pauses every 30 seconds or so. I also disabled the fanart add-on and increased the gpu memory from 128 to 256 but with no success.

I have uploaded a copy of the logs here - http://pastebin.com/aG4jEJq8

A couple of things I have noticed, not sure if they will help:-

I have recently upgraded my amplifier. On my old amp is only supported 1080i and I didn't have the pauses. My mew amp supports 1080p and the pauses are very bad. Is there a way of forcing XBMC to 1080i so I can test this out on the new amp?

I also tried experimenting with different frame rates - set at 24, 30 and 50 the pauses happened but it seemed to get further into the film before they started. Withe the frame rate at 60 they started almost immediately.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - RichG - 2014-03-11

(2014-03-11, 02:27)popcornmix Wrote:
(2014-03-09, 23:25)RichG Wrote: Testing with two files:
If I play file one then stop and restart it - no noise burst
Play file two, stop, restart - sometimes get noise
Play file one then start file two - loud burst
Play file two then start file one - quieter burst most times, occasionally louder.

Thanks for info. A fix has been pushed to newclock3.

Sorry to say, but the fix hasn't worked. Upgraded to the new MilhouseVH build and I'm now getting a loud burst every time I start file one, but nothing going from one to two.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - fightnight - 2014-03-11

#0311 fixed loud burst.

Thanks!

Edit: at 3rd time I get loud burst again :S


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - ken_s - 2014-03-11

(2014-03-10, 20:41)popcornmix Wrote:
(2014-03-10, 20:19)Shanyel Wrote: I can confirm that since I've added this setting, no lockups

This setting has 3 enable bits in. It would be interested to determine which bit causes the lockup. Can you (anyone who sees the lockup) try:
Code:
dwc_otg.fiq_fsm_enable=1
Code:
dwc_otg.fiq_fsm_enable=2
Code:
dwc_otg.fiq_fsm_enable=4

Hopefully just one of those will lock up.

All 3 settings led to freezes on my pi, so leaving it on 0 for now which seems to work.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - User 203163 - 2014-03-11

(2014-03-10, 20:41)popcornmix Wrote: This setting has 3 enable bits in. It would be interested to determine which bit causes the lockup. Can you (anyone who sees the lockup) try:
Code:
dwc_otg.fiq_fsm_enable=1
Code:
dwc_otg.fiq_fsm_enable=2
Code:
dwc_otg.fiq_fsm_enable=4

Hopefully just one of those will lock up.

Copied two ~4GB .mkv-files from PC to USB-HDD connected to my RPi.

Average speed 3MB/sec (over wired nic)

Nothing in cmdline.txt other than boot=LABEL=SYSTEM disk=LABEL=STORAGE ssh quiet Lockup after 14-20 minutes of copying. I could still use ssh but it was sloooow, bcmstat showed over 9,000 IRQ/s


boot=LABEL=SYSTEM disk=LABEL=STORAGE dwc_otg.fiq_fsm_enable=x ssh quiet
x=0 No lockups
x=1 Lockup
x=2 Lockup
x=4 Lockup


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - Silo100 - 2014-03-11

(2014-03-11, 12:47)IntelSwe Wrote:
(2014-03-10, 20:41)popcornmix Wrote: This setting has 3 enable bits in. It would be interested to determine which bit causes the lockup. Can you (anyone who sees the lockup) try:
Code:
dwc_otg.fiq_fsm_enable=1
Code:
dwc_otg.fiq_fsm_enable=2
Code:
dwc_otg.fiq_fsm_enable=4

Hopefully just one of those will lock up.

boot=LABEL=SYSTEM disk=LABEL=STORAGE dwc_otg.fiq_fsm_enable=x ssh quiet
x=0 No lockups
x=1 Lockup
x=2 Lockup
x=4 Lockup

Exactly the same results here.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - ramekers - 2014-03-11

(2014-03-09, 14:53)ramekers Wrote: ever since the latest build, but probably a few builds earlier also,
I can't seem to get dual audio output working;
I can enable this in the settings menu but the audio output stays at the currently selected device,
being either Analog or HDMI. Both of them work properly, but not at the same time.

René

Nobody else having this problem ??
To make things worse, toggling the audio channel from my Phone doesn't work either,
it just seems to toggle the "enable passthrough" option instead of toggling the audio channel,
even though the Phone does send the command "Action(audiotoggledigital)" to XBMC.
(I use Yatse's remote app)