SuperZaxxon v1.73 - maintenance release


#1138 also works like #1137. No squashfs errors. PaleMoon recovers every time. Thanks!
Thanks again, so it's the DMA parameters. The bad thing is that switching to those old parameters hurts wifi for me, the transfer becomes less stable, I get ~200kB/s fluctuations (to the slow side of course), while with problematic parameters it's much more stable.

The obvious solution is to only set the new params for wifi and revert them for SD cards (they don't help SD card rates anyway), so could you do one more (and hopefully last) test? I'm almost sure it will work, but it would be great to have your confirmation.

The file is the same as always: http://notaz.gp2x.de/tmp/uImage-3
 
I hate to ask this after you put in the time to update it, @notaz, but is there an easy way I can downgrade bluez4? The new version doesn't seem to play nice with bluetooth audio for me.

Thanks again for all your work. This is a pretty minor thing in the long run.

EDIT: Or perhaps I simply need to also upgrade bluez-alsa? Not sure but I will look into that.
 
I hate to ask this after you put in the time to update it, @notaz, but is there an easy way I can downgrade bluez4?
Not any I can think of.
Somebody needs to debug what is going wrong with the updated bluez version. I don't have any BT audio hardware and can't do that.

EDIT: Or perhaps I simply need to also upgrade bluez-alsa?
IIRC it was already updated.
 
Slackware just got pulse audio for bluetooth audio, otherwise, no bluetooth audio...
 
Well I will attempt to do some investigating into the problem. I may load an SD card with SZ 1.72 and boot that for when I want to use BT audio until I can help figure it out. I have further related investigation to do in order to determine why auto PIN generation does not work when paring with the Pebble watch.

EDIT: I spliced out the pebble specific stuff, as that fits better in the BT issues thread in the support section.

@Linux-SWAT: What is the best way to try pulse on Slackware? I just installed it again recently and was unable to get any audio working. Also, it had similar BT PIN issues to SZ with regards to pairing with my Pebble.

LAST EDIT: Suggestion from notaz in another thread fixes bluetooth audio in 1.73. Awesome!

To fix, simply make sure /etc/bluetooth/audio.conf has this socket line in the general section:

Code:
[General]
Enable=Socket
 
Last edited:
Ok, so this has nothing to do with this specific release, I'm just not going to make a thread for one post and forget thing.

Code:
https://gist.github.com/urjaman/38c43d90e5615fccc817

EDIT: I Apparently cannot make that a link, it just becomes a MEDIA=gist thing that doesnt seem to work atleast for me locally... just a big area of white. So there you go, link in a code block. FFS.

There's a patch for making the BT work after suspend (apply order is that first and the auto-hiber later...)
That should actually be better than current in any case, but yeah.

The second thing (which is what i wanted to drop a reference to somewhere) is a script that allows the pandora to intelligently (when nothing prevents this) suspend after the lid is closed with a timeout so it doesnt interrupt active use etc.
One cool thing with this is that plugging the pandora to charger and closing lid causes it to charge to 100% and suspend when the charge is stopped :)

Currently it is a bit hacky and has hardcoded stuff (like to consider ssh connections active use, and the 7.5 minute timeout etc), but if somebody wants to work on integrating this to the OS I'd be quite happy :).
 
Did something change with zram in the last updates? I just noticed I don't have any zram swap anymore, but it's not impossible I somehow broke it myself - does it still work for others?

Code:
~ # mkswap /dev/zram0
mkswap: error: swap area needs to be at least 40 KiB
Usage: mkswap [-c] [-pPAGESZ] [-L label] [-U UUID] /dev/name [blocks]
~ # ls -al /dev/zram0
brw-rw---- 1 root disk 254, 0 Feb 24 10:02 /dev/zram0
~ # uname -a
Linux box 3.2.74 #1134 Sun Dec 13 20:44:55 EET 2015 armv7l GNU/Linux
~ # dmesg | grep zram
[  8.708984] zram: module is from the staging directory, the quality is unknown, you have been warned.
[  8.711364] zram: num_devices not specified. Using default: 1
[  8.711364] zram: Creating 1 devices ...
 
It's now disabled because it really does more harm than good.

But it should still work, not sure what's going on for you. Maybe it can't allocate memory because the memory got fragmented and you just need to reboot?

edit: hard -> harm
 
Last edited:
Hrm, I guess I'll just disable it then...

The zram module is loaded at boot time, so memory fragmentation should not be a problem. I used to have the impression it takes some disk swap load off my Rebirth especially when running a web browser (not much else does get it into swap, anyway), but on the other hand it generates a lot of ugly kernel messages when zram swap is full, even when another swap device is available.
[doublepost=1456314787,1456314654][/doublepost]Thanks again for still working on this stuff, by the way.
 
Ok, so this has nothing to do with this specific release, I'm just not going to make a thread for one post and forget thing.

Code:
https://gist.github.com/urjaman/38c43d90e5615fccc817
I took the BT part but leaving autohiber out until someone makes it optional. I need my pandora to be accessible over usbnet at all times and would not like it to autosuspend, I'm guessing it would also confuse some users.
 
Last edited:
Back
Top