Re: BFS for the power kernel
 

So... I'd like to recompile kpv48 with bfs and fcam driver but... I don't know how do that...

Re: BFS for the power kernel
 

already using multiboot with kp48/nitroid and now added latest bfs6.debs. used swappolube to edit available values as when using echo commands from this thread and direct edit on phone or winscp the values reset to default after reboot. unfortunately i found that my phone ran really badly when set up that way with loads of hangs when multitasking and occaisional reboots too. using kp48 again trouble free . unlucky ? or my bad set up ? really not sure...really appreciate this thread and tigrettes etc work to give me the chance to try though even if so far unsuccessful.

Re: BFS for the power kernel
 

Could anyone else with the BFS kernel and the CSSU update 16 or 16.5 check if they still have the call distortions.

It didn't happen to me once since upgrading a few days ago. Since I realized it I also tried to force it by doing something on the phone that utilizes a lot of CPU time and then calling myself. This was something that normally always caused the distortion. But until now it didn't happen even once.

I know the new CSSU update fixes to some memory leaks and probably more stuff of hildon-desktop... so after all it might not have been a directly BFS related thing.

Re: BFS for the power kernel
 

Curiously, which release of BFS kernel are you using?

Re: BFS for the power kernel
 

this one:
http://talk.maemo.org/showpost.php?p...&postcount=254

Btw. as a small update: there was a short call distortion for a few seconds today but it went away and I could talk and hear clearly afterwards.

I hope that this wasn't just a lucky streak but something that happens a lot less now.

Re: BFS for the power kernel
 

In my experience, some short distortion means that the problem is still there, and that you'll eventually get a more severe distorted call later. Please do report back though :).

I've been talking with Tigerite about releasing kernel-bfs to Maemo's repositories with only the BFS patches enabled up to v0.330. This because of the following reasons:
a) I don't see the distortion being fixed anytime soon
b) kernel-bfs has some advantages over kernel-power. For example the -ck patches. (Subjective point)
c) BFS 330 is still BFS. Whilst it isn't the latest version, it still offers BFS' (design) advantages over CFS.

I've been running this configuration as my stable configuration for several weeks now, without any issues. Expect this release soon.

Re: BFS for the power kernel
 

If I understand correctly, it will contain all kp48 advantages + bfs, yes?

Maybe You could include also freemangordon's and colin.stephane's patches/fixes, that are going to made it into kp49? Especially, that it fixes quite severe bugs, causing device reboot in certain applications (so called "save as" bug, but it's not limited to leafpad "save as").

Re: BFS for the power kernel
 

The changes from kernel-power v48 are in kernel-bfs' git repository since 2011-08-23. So yes, it will. Do note that kernel-bfs also got some additional patches over BFS.

Sure, they fit in the project's scope (which includes being a bit more bleeding-edge). We can't keep track of every patch posted on the forums, but I assume you mean the ones posted over here?

--

On some other news: we're 'professionalizing' everything a bit to make kernel-bfs ready for some wider-spread usage. For example, there's a new (small) wiki page* and we're doing little (cosmetic) cleanups where needed. One of the last points on the TO-DO list is this TMO thread. We're thinking about starting a new, proper [ANNOUNCE] thread for when kernel-bfs hits the repositories. We tried contacting coreyoconnor to update the first post of this thread a while ago, but we didn't got a reply :(. What does the general public think about this? Is it justified? Of course, there'll be a reference to this thread if we decide to go with a new thread.

*Feel free to extend it, if you want.

Re: BFS for the power kernel
 

Thanks for explaining it. BTW, it seems that their (freemangordon and colin.stephane) patches to kernel-power are always here:
http://bigbob.fun.free.fr/maemo-kernel-power/

I by no means expert on this, but it seems that changes fit project purposes, they're definitely more on bleeding-edge (especially compcache).

I also totally agree with your idea of starting proper announcement thread, if title of this one can't be changed. Especially, that You're going to revert back to version without call distortions, so, probably, thread won't be bombed with report about this for n pages.

Re: BFS for the power kernel
 

Thanks for the pointer. Their work got merged: https://garage.maemo.org/plugins/ggi...7530190218d138

A little status update: I'm currently waiting for Tigerite to merge his UBIFS patches (as I prefer not to merge other people's work ;)). After his push, I'll push the updated debian/changelog, and send kernel-bfs to Maemo's autobuilder.

To bridge the gap in time, I've compiled kernel-bfs as how it would be released to the repositories. You can find the debs, source, and uImage on kernel-bfs' garage page. The changelog can be found here.
I've been running this configuration for about two weeks now, and a slightly changed configurations for quite some time before that too. Everything should go smoothly :).

Re: BFS for the power kernel
 

FWIW,

fcam-drivers and bleeding-edge-wl1251 will need to be rebuilt against the headers for the most recent bfs6 builds. I'm looking at you, Tigerite :P

Re: BFS for the power kernel
 

IDont, according to last irc meeting @ mamemo-kernel-power, Pali will be releasing kp49 quickly (mostly, to fix bug known as "save as"), so he won't be able to introduce *every* path created by colin.stephane and freemangordon up to date (for example, further tweaking of FM transmitter API's).

That mean bfs kernel will be "little" ahead - which I think is all right due to spirit of project :) Just posting this, to avoid confusion, when You'll find out that some updates you've included didn't made it to kp49. "they" will, just later.

It's also possible (but I think unlikely) that patches to be found on kp49 will be tweaked a little further before releasing (compared to ones from link I've provided), so it may be good to keep an eye on code and update things that seems sane.

I know it sounds little messy, I hope freemangordon and colin.stephane will finally request maintainership and we're gonna get more predictable update stream ;)

Re: BFS for the power kernel
 

this is the most amazing mod ive ever done. without fail. i would suggest it has made my device about 35 percent faster. overclocked to 1100, the same as i had with power48. and limiting the scaling to 4 frequencies. the speed is phenominal. fantastic work everyone! im chuffed!

*edit. only thing is tactile feedback seems to have been enabled and i dunno how to turn it off. seems to vibrate slightly when pressing screen. any ideas?

Re: BFS for the power kernel
 

just uninstall it.

apt-get remove tactile

Re: BFS for the power kernel
 

i get 'tactile is not installed so not removed'.

also when opening xterm get....


BusyBox v1.19.2 (Debian 1.19.2power1) built-in shell (ash)
Enter 'help' for a list of built-in commands.

-sh: /home/user/.bashrc: line 5: can't create /dev/cgroup/cpu/user/2600/tasks: nonexistent directory
-sh: /home/user/.bashrc: line 5: can't create /dev/cgroup/cpu/user/2600/notify_on_release: nonexistent directory

*edit.... now im really confused about the tactile vibration. i dont have it installed, its set to 0 in the cssu transitions.ini. it seems to only vibrate sometimes and vary in vibration. what on earth?!

*edit 2.... fixed it! was just the touch vibration option under display. durrr. somehow reenabled on install

Re: BFS for the power kernel
 

I haven't got any experience with building the bleeding-edge wl1251 driver, but here are the fcam-drivers: https://garage.maemo.org/frs/downloa....7-2_armel.deb.
I've modified the source to include kernel-bfs' fcam drivers next to the ones for kernel-power and kernel-omap, instead of having one of the last two replaced. This new fcam package should be pushable to Maemo's repositories too.

By the way, about the wlan1251 driver, pali noted it's on his TODO list to look into integrating it into kernel-power, which would IMO be a really nice and clean solution.

We've been there before; It's just that we hadn't made any releases back then ;).

I'll keep an eye on the source, thanks for mentioning.

Whoa, I'm glad you're content with kernel-bfs :)

Re: BFS for the power kernel
 

yeah you were right... the call distortion came back now to an extent where I had them even a few times a day. strange thing though that I haven't experienced it even once in some days after the cssu update.

anyway I'm just giving the new kernel with bfs 330 a try and it's working pretty smooth, thanks!

Re: BFS for the power kernel
 

Hey iDont, I have idea, that may make BFS kernel even more bleeding-edge ;)

As You may remember, there was a path in power-kernel tree, that allowed bq27x00_battery module and BME to live together (happily). Path was turned off, due to it being violating standards (theoretically, bq_27x00_battery and BME should not access same i2c). Ho ever, there was *never* true report of it causing any problems (every one I've heard of, resulted in finding other source of problems), so decision was rather "ideological" than practical. AFAIK, no one know why it worked perfectly (when theoretically, it should not), and we won't know, cause BME is closed source - there is possibility, that something there is also "out of standards"l, allowing combo to work ;)

If You could enable it in BFS kernel, many, many people would be happy penguins (including me ;) ). I'm sure it would bring quite wider audience for project, but, more importantly, it would be in spirit of project - incorporating features that are working great and bringing joy, but denied mainline ;)

/Estel

// Edit

If someone wonder, why even bother - let me name first thing from top of my head. Using it with advanced power monitor (and after calibrating fuel gauge chip), we can make our battery status applet to show *real* amount of battery left, with mistake lower than +-0,1% (absolutely true mAh readings + battery percentage readings).

Re: BFS for the power kernel
 

Well, being bleeding-edge is not a goal in itself; it just happens to come with many goodies ;)

Yes, I've noticed the whole i2c-dev patch discussion on IRC. Personally, I think the patch should be left disabled. I know there aren't any confirmed reports on it causing trouble, but this doesn't alleviate the fact that you're removing a very important safety mechanism.

However, to give everyone some freedom of choice, I've created a patch that allows you to toggle the safety mechanism in question on or off at runtime. The patch consists of Matan's original patch + some sysfs code wrapped around it (from the kernel documentation, with love). I've used a slightly provocative name for the sysfs entry, but that seems only fair considering what you're actually doing ;).

If you understand the risks of the (original) patch, then here goes the tutorial:
After enabling the hack, you can stop/start bme even though bq27x00_battery is loaded, as desired:
The above scenario will result in an instant reboot when i2c-dev_break_ioctl is 0. You can also verify that the pre-KP48 path is active by checking your kernel log:
after starting BME ;)

Precompiled debs plus the patch itself can be found on kernel-bfs' garage page. Please test it and report back. I haven't got any previous experience with kobjects/sysfs, so some elements of the patch could probably be implemented in a cleaner manner. This doesn't affect its workings however. Feedback is appreciated / patches are welcome :).

p.s. I'll send the patch to pali later as well.
edit 2: patch is sent. Pali replied that he's already working on a different solution which works by redirecting ioctls. That solution allows us to have both bq27x00_battery and bme working together without removing the safety mechanism in i2c-dev. There's no ETA on that yet though.

Re: BFS for the power kernel
 

hello iDont just wanna ask is the call distortion problem still exist on the current builds?

Re: BFS for the power kernel
 

Nope, we've reverted back to BFS v0.330, which doesn't have the issue.

--

A little update: I'll most likely push kernel-bfs either this evening or tomorrow evening to Maemo's repositories. I haven't heard anything back from Tigerite yet regarding pushing the UBIFS patches to our git tree, but I suppose it isn't too much of a problem that the git tree will lack one patch which the official release does include. Those interested in the patch can find it in the source package anyway. If desired, I can always push the patch myself (with proper credits of course) later.

Re: BFS for the power kernel
 

Thanks very much, iDont! I think your approach is best compromise. Will test it and report back.


// Edit - scratch the following, iDont beat me up to it ;)

Re: BFS for the power kernel
 

Thank you, much appreciated :)

The debs on kernel-bfs's garage page are as how kernel-bfs will be released to the repos.
However, if the KP49 pre-release / the site I should keep an eye on has introduced new changes, then those will be merged into kernel-bfs' release as well. I'll check for those changes this evening too.

Edit: just noticed your edit ;). I'll leave my reply in here for others though.

Re: BFS for the power kernel
 

thanks. Will this auto-upgrade/overwrite older manual installs of kernel-bfs (bfs6) people (including me) have ? And modify the multiboot files ?

Or will we have to remove things manually first ?

Thanks again for all the effort !

Re: BFS for the power kernel
 

It doesn't matter whether you've installed the debs manually or via a package manager, upgrading the package (again either manually or via a package manager) will go fine.

Regarding multiboot: in older kernel-bfs releases you had to create the multiboot entry yourself. If you have done so, then you'll need to remove it manually too.
By the way, not removing the manually created multiboot entry won't even cause any trouble; AFAIK, you'll only end up with kernel-bfs twice in your multiboot list ;).

Re: BFS for the power kernel
 

Sorry for noob question, but when I try to install fcam-drivers-bfs_1.0.7-2_armel.deb :

dpkg: considering removing fcam-drivers in favour of fcam-drivers-bfs ...
dpkg: no, cannot proceed with removal of fcam-drivers (--auto-deconfigure will help):
blessn900 depends on fcam-drivers (>= 1.0.6-1)
fcam-drivers is to be removed.
dpkg: regarding .../fcam-drivers-bfs_1.0.7-2_armel.deb containing fcam-drivers-bfs:
fcam-drivers-bfs conflicts with fcam-drivers
fcam-drivers (version 1.0.7-2) is present and installed.
dpkg: error processing /home/user/MyDocs/bfs/fcam-drivers-bfs_1.0.7-2_armel.deb (--install):
conflicting packages - not installing fcam-drivers-bfs
Errors were encountered while processing:
/home/user/MyDocs/bfs/fcam-drivers-bfs_1.0.7-2_armel.deb


What is correct way of installing fcam drivers for bfs?

Re: BFS for the power kernel
 

You did try to install it the correct way ;). There's a little catch however.

Fcam-drivers-bfs conflicts with, replaces, and provides fcam-drivers. You appear to have an application installed which depends on the fcam-drivers (namely blessn900). Prior to installing fcam-drivers-bfs, fcam-drivers needs to be removed (since we try to replace it), Now this application that depends on fcam-drivers starts to complain, even though we're about to install a package that provides the same functionality!

You can make dpkg automatically deconfigure and reconfigure the applications that depend on fcam-drivers, so that the those applications won't complain in the little time span the fcam-drivers package is removed. To do this, you'll have to run:

--
edit: the most ideal situation is that the official fcam-drivers will get support for kernel-bfs. But this can't be done before kernel-bfs is pushed to the repositories since fcam-drivers will depend on kernel-bfs-headers then. Until that time, we'll have to do with our own package ;). I'll try to contact the maintainer of the fcam-drivers package when kernel-bfs hits the repositories.


--

Update: Kernel-bfs has just hit the repos! The new kernel-bfs [Announce] thread can be found here:
http://talk.maemo.org/showthread.php?p=1099211

Re: BFS for the power kernel
 

What about status of this thread? We should continue to use it as technical-background, or should we move every discussion to new announce one?