Post Reply

[Bug report] Music player stops playing music from sd after some time

OP pilgrim011

26th January 2015, 04:34 PM   |  #921  
Senior Recognized Developer
Flag Owego, NY
Thanks Meter: 24,980
 
13,621 posts
Join Date:Joined: Aug 2007
Donate to Me
More
Quote:
Originally Posted by Lanchon

i think he wants those as separate commits. and i think you can abandon your own change yourself in gerrit.

Yes, although in this case, the changeID for the "squashed" commit could have been preserved and used for the final commit in the chain. Yes, you can alter dependency chains for existing commits.
Quote:
Originally Posted by notabenem

Gerrit is quite new to me, apologies for the newbie mistakes. I have submitted 13 separate changes (one for each upstream change, and one for zeitferne's changes).
These are from https://gerrit.omnirom.org/#/c/11621/ to https://gerrit.omnirom.org/#/c/11633/

Yup. Only issue I have is that for the upstream patches, you added "smdk4412" to the commits even though they're just plain old upstream cherry-picks (unless zeitferne did this too???)

It's easy enough to fix this, I can edit the commit messages tonight.
Quote:
Originally Posted by GidiK

I have made a build (omni-4.4, i9100) with all 13 patches applied. I reconfirmed that I could reproduce the fuse bug with a recent nightly and ran the exact same test with the new build. And good news: the bug is gone! So as far as I am concerned, they are good to go.

The test I did:

Code:
adb shell
cd /storage/sdcard1
tar c Movies > /dev/null
while in another terminal window monitor progress with:

Code:
adb shell
iostat -k 5
On the recent nightly it would get stuck the moment the screen went off.

Thanks to all involved in finding this solution!

Based on your input, I'll be ninjamerging these without personally testing them tonight, can you go into Gerrit and +1 them as confirmation?
The Following 5 Users Say Thank You to Entropy512 For This Useful Post: [ View ]
26th January 2015, 04:59 PM   |  #922  
Member
Flag Best
Thanks Meter: 86
 
51 posts
Join Date:Joined: Jan 2013
More
Quote:
Originally Posted by Entropy512

Based on your input, I'll be ninjamerging these without personally testing them tonight, can you go into Gerrit and +1 them as confirmation?

Done. Tried to preserve the order, which I think is important.

Thanks in advance!
The Following 3 Users Say Thank You to GidiK For This Useful Post: [ View ]
26th January 2015, 05:19 PM   |  #923  
Senior Member
Thanks Meter: 73
 
103 posts
Join Date:Joined: Jul 2009
Quote:
Originally Posted by GidiK

Done. Tried to preserve the order, which I think is important.

Thanks in advance!

I changed the commit messages and removed the "smdk4412" prefix as Entropy requested. It seems this removed the +1's from GidiK?
The Following 2 Users Say Thank You to notabenem For This Useful Post: [ View ]
27th January 2015, 08:00 AM   |  #924  
Senior Member
Thanks Meter: 73
 
103 posts
Join Date:Joined: Jul 2009
I am happy to report, that thanks to GidiK and Entropy, the changes are now merged to android-4.4 branch of Omni. There also exists an android-5.0 branch. How do we keep that also up-to-date (knowing, that before this change, it was identical to 4.4)?
Last edited by notabenem; 27th January 2015 at 09:52 AM.
The Following 2 Users Say Thank You to notabenem For This Useful Post: [ View ]
27th January 2015, 09:21 AM   |  #925  
Member
Flag Best
Thanks Meter: 86
 
51 posts
Join Date:Joined: Jan 2013
More
Quote:
Originally Posted by notabenem

I am happy to report, that thanks to GidiK and Entropy, the changes are now merged to android-4.4 branch of Omni. There also exists an android-5.0 branch. How to we keep that also up-to-date (knowing, that before this change, it was identical to 4.4)?

Just tested the 27 Jan 2015 nightly and it is indeed bug-free! Well, free of this bug...

Thanks everybody for all the hard work.

The 5.0 is a whole different issue; for i9100 5.0 doesn't really exist yet and it is not sure it ever will. I assume that anybody who wants to try a private build for 5.0 will have to do his own repo/git-trickery for now (e.g. put a 4.4 reference in your local manifest instead).
The Following User Says Thank You to GidiK For This Useful Post: [ View ]
27th January 2015, 09:51 AM   |  #926  
Senior Member
Thanks Meter: 73
 
103 posts
Join Date:Joined: Jul 2009
Quote:
Originally Posted by GidiK

Just tested the 27 Jan 2015 nightly and it is indeed bug-free! Well, free of this bug...

Thanks everybody for all the hard work.

The 5.0 is a whole different issue; for i9100 5.0 doesn't really exist yet and it is not sure it ever will. I assume that anybody who wants to try a private build for 5.0 will have to do his own repo/git-trickery for now (e.g. put a 4.4 reference in your local manifest instead).

True. I don't expect Omni to miraculously start working on the 5.0 branch of the whole ROM for i9100. However, given the importance of this fix, I suggest to update the android-5.0 branch of the smdk4421 kernel repo. If someone in the future decides to start working on Omni-lollipop for I9100, it would be good to give him a somewhat more stable starting point. This merge is far too important just to have it lost somewhere in 'translation' between 4.4 and 5.0.
The Following 2 Users Say Thank You to notabenem For This Useful Post: [ View ]
27th January 2015, 05:29 PM   |  #927  
Senior Recognized Developer
Flag Owego, NY
Thanks Meter: 24,980
 
13,621 posts
Join Date:Joined: Aug 2007
Donate to Me
More
Quote:
Originally Posted by notabenem

I am happy to report, that thanks to GidiK and Entropy, the changes are now merged to android-4.4 branch of Omni. There also exists an android-5.0 branch. How do we keep that also up-to-date (knowing, that before this change, it was identical to 4.4)?

I can look at merging that. If it's the ONLY delta I might be able to have gerrit do the work for me. Won't be able to look until the weekend though.
The Following 4 Users Say Thank You to Entropy512 For This Useful Post: [ View ]
14th February 2015, 10:03 PM   |  #928  
Senior Member
Thanks Meter: 73
 
103 posts
Join Date:Joined: Jul 2009
Christian Balster has merged the patchset also to Omni's android-5.0 branch: https://gerrit.omnirom.org/11791
It seems there's nothing else to be done for this bug and this thread is ripe to be closed by a mod.

Thanks for everybody who contributed to have this bug discovered, identified, analyzed, tested, fixed and merged, but most notably to: pilgrim, Entropy, cgx, Lanchon and zeitferne!

The Following 10 Users Say Thank You to notabenem For This Useful Post: [ View ]
Post Reply Subscribe to Thread
Previous Thread Next Thread
Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes