Posts by popcornmix

You can always switch to 720p in system/video settings.

That will prove if the issue is purely resolution dependent, or if something else in the update has caused it.

Do you have "adjust refresh rate to match video" enabled in system/video settings?

That is recommended for smoothest video.

I notice you are using omxplayer. That isn't the default (or recommended) on PI2/Pi3.

Do you have the problem with omxplayer disabled (leaving MMAL enabled)?

I see you have omxplayer enabled, which isn't the default (or recommended) on Pi2/Pi3.

Do you still have the issue if you disable this?

You can customise the filename tags used for 3d: Advancedsettings.xml#video

See:

  <stereoscopicregex3d>[-. _]3d[-. _]</stereoscopicregex3d>
  <stereoscopicregexsbs>[-. _]h?sbs[-. _]</stereoscopicregexsbs>
  <stereoscopicregextab>[-. _]h?tab[-. _]</stereoscopicregextab>

Quote from metaron

Is there a log somewhere I might be able to resurrect? (git bisect log obviously only shows the most recent bisect)

I think only the current git bisect log is saved.

You can run "git bisect log" to see what commits you reported as good/bad.

You might want to double check if you reported bad to 54a2ec67f1db62a763f57b7f8f2e82874f5f358b or an older commit.

54a2ec67f1db62a763f57b7f8f2e82874f5f358b is not going to be the commit that caused the problem (it only affects the

SOC_OMAP5 platform which isn't the Pi or PC).

If you ever report the wrong answer to git bisect good/bad then it will identify the wrong commit.

The occasional quick compiles are expected. Especially after a few iterations the number of commits jumped will be small and if these don't affect header files the amount of rebuilding may be small.

How are you naming the file?

You should add tags like: 3D#Video_filenames_flags

Quote from metaron

Only shows 31 individual commits. Might it be because I cloned with --depth=1 then ran --deepen=100?

Yes, I don't think a shallow clone will have the required history.

I think you should clone the complete repo for this to work.

metaron good work!

That provides evidence that it is an upstream linux kernel bug that affects multiple platforms.

Reporting upstream may be useful: Reporting bugs for the Linux kernel

You have identified a merge commit that seems responsible which is helpful, but it is a huge merge of 343 commits.

You can view these commits with:

git log  e6445f52d9c8b0e6557a45fa7d0e8e088d430a8c^..e6445f52d9c8b0e6557a45fa7d0e8e088d430a8c

I think you can narrow it down to a single commit with:

dc4@dc4-XPS13-9333:~/projects/linux$ git bisect start
dc4@dc4-XPS13-9333:~/projects/linux$ git bisect good e6dce825fba05f447bd22c865e27233182ab3d79
dc4@dc4-XPS13-9333:~/projects/linux$ git bisect bad e6445f52d9c8b0e6557a45fa7d0e8e088d430a8c
Bisecting: 184 revisions left to test after this (roughly 8 steps)

Build this kernel and check if it is good/bad. Then run either "git bisect good" or "git bisect bad" and it will provide you with a new commit to build from. After 8 builds you should have the answer what the specific commit was.

Quote from gert

I am fairly certain I just did apt-get update && apt-get upgrade (or maybe: apt-get dist-upgrade)

That gets you the 4.9 kernel. The next raspbian sdcard image will use this kernel.
(rpi-update currently gets a slightly newer 4.9 kernel and is the easiest way to downgrade to a previous kernel).

Quote from metaron

OK, so now I'm really confused. I've blown a jesse-lite image to a spare SD card, logged in and run:

sudo apt-get install --reinstall raspberrypi-bootloader raspberrypi-kernel

but ls /lib/modules shows me 4.4.50+ 4.4.50-v7+ (which isn't the 4.9.24 version maroi says he has on his box)

Typing sudo apt-get dist-upgrade tells me 0 upgraded, 0 newly installed, o to remove and 0 not upgraded. - so that doesn't work either.

Try running "apt-get update" beforehand to get the indexes up to date.

Quote from metaron

Edit: Maybe kernel: Bump to 4.3.24 · Hexxeh/rpi-firmware@bbd611a · GitHub is it?

(It's a bit late now - will try again another day!)

Yes - 4.3.24 was a typo (for 4.9.24) but there is no safe way of fixing a git commit message.

Quote from gert

The RPi2 has worked flawlessly for about 2 years, until it suddenly began to produce Continuity Counter Errors.

If you want to narrow this down, can you back up your sdcard and try reverting the kernel.

sudo rpi-update 7a47836821b92efa569500b3382b1812082e42d3

will get you the last 4.8 kernel (4.8.13) which from my understanding should be good.

sudo rpi-update bcc6146e102d85b1aa214855ad7aae278d3bd269

will get you the first 4.9 kernel (4.9.0) which should have the problems.

It would be useful if you (or others) can prove or disprove this is correct.

Quote from metaron

Anyone know how to track down #993?

sudo rpi-update e253b1dcaa3aa774444f6c0224f04237f60ba2d1

Might be interesting to make a list of USB DVB devices that have this problem (artefacts with 4.9 kernel but fine with 4.4 kernel).

If you have this issue report make/model of device (a link to where it was obtained from may be useful).

Output of "lsusb -v"

Quote from metaron

after running rpi-update on my box

$ uname -a

Linux pi2 4.9.30-v7+ #1001 SMP Fri May 26 16:09:18 BST 2017 armv7l ARMv7 Processor rev 5 (v7l) BCM2835 GNU/Linux

Watching BBC2 HD the green artifacts are back (intermittently as they were before).

Returning to the 4.8.16 build.

sudo apt-get install --reinstall raspberrypi-bootloader raspberrypi-kernel

should get you back to latest stable 4.9 kernel/firmware (i.e. matching marol). Could you double check if that is good/bad?

Also maril/metaron are you using a Pi 3? Any non-default config.txt options?