Official thread: "zen-sources" - Part II

Message

Waninkoko · Post by **Waninkoko** » Sat Jan 05, 2008 3:37 pm

Post #1000

Veldrin · Post by **Veldrin** » Sun Jan 06, 2008 10:04 am

Well, actually it was the 1001st post, and the 1000th answer...

but who cares?

awesome topic, and - I often forget to mention it - your sources rock.

cheers
V.

RobertDavid · Post by **RobertDavid** » Sun Jan 06, 2008 11:28 am

BTW dont you want to make some new thread or something like monthly new thread because, reading 1000 posts back is torture:)

It would be superb if there will be some subthread support in forums to devide this long posts and make them more clean.

termite · Post by **termite** » Sun Jan 06, 2008 3:10 pm

Agreed. I think zen-sources should have its own forums, given how popular it has become. There are a myriad of different issues and it would be much easier to find answers if things were better organized.

rmh3093 · Post by **rmh3093** » Sun Jan 06, 2008 4:45 pm

termite wrote:Agreed. I think zen-sources should have its own forums, given how popular it has become. There are a myriad of different issues and it would be much easier to find answers if things were better organized.

I was thinking about that too

creidiki · Post by **creidiki** » Sun Jan 06, 2008 9:15 pm

It will be even more true when stable releases are available - some of us are not so kernel-crazed we like to mess with developing kernels

F.eg, i'm still on 2.6.22-klight3 (now at build 11). 2.6.23 kernels don't count :p

ffwd · Post by **ffwd** » Sun Jan 06, 2008 10:26 pm

2.6.24 zens are nice, but im still sticking to 2.6.23-kamikaze5 until can i have a working madwifi-ng and open-vm-tools without manually hacking the ebuilds/compiles/whatever (not to mention the power saving regression in 24's)...

eatnumber1 · Post by **eatnumber1** » Mon Jan 07, 2008 4:09 am

ffwd wrote:2.6.24 zens are nice, but im still sticking to 2.6.23-kamikaze5 until can i have a working madwifi-ng and open-vm-tools without manually hacking the ebuilds/compiles/whatever (not to mention the power saving regression in 24's)...

I wasn't aware there was a power saving regression... what is it? Can you link to some info?

tranquilcool · Post by **tranquilcool** » Mon Jan 07, 2008 7:13 am

errors with latest git pull;

CC [M] drivers/media/video/bt8xx/bttv-driver.o
drivers/media/video/bt8xx/bttv-driver.c

error: unknown field ‘vidioc_g_register’ specified in initializer
drivers/media/video/bt8xx/bttv-driver.c

error: ‘bttv_g_register’ undeclared here (not in a function)
drivers/media/video/bt8xx/bttv-driver.c

error: unknown field ‘vidioc_s_register’ specified in initializer
drivers/media/video/bt8xx/bttv-driver.c

error: ‘bttv_s_register’ undeclared here (not in a function)
make[4]: *** [drivers/media/video/bt8xx/bttv-driver.o] Error 1
make[3]: *** [drivers/media/video/bt8xx] Error 2
make[2]: *** [drivers/media/video] Error 2
make[1]: *** [drivers/media] Error 2
make: *** [drivers] Error 2

buddabrod · Post by **buddabrod** » Mon Jan 07, 2008 7:55 am

eatnumber1 wrote:
ffwd wrote:2.6.24 zens are nice, but im still sticking to 2.6.23-kamikaze5 until can i have a working madwifi-ng and open-vm-tools without manually hacking the ebuilds/compiles/whatever (not to mention the power saving regression in 24's)...
I wasn't aware there was a power saving regression... what is it? Can you link to some info?

I stated this many times, waninkoko knows it.
Iirc this occured after pulling in some -rt stuff, now my notebook consumes about 2Watt more than vanilla!
For the meantime i'm back to 2.6-wireless-git.

Edit: dunno if there is any information to be found but i'm quite sure it's the -rt stuff. (as i said, vanilla's fine)

oRDeX · Post by **oRDeX** » Mon Jan 07, 2008 9:34 am

buddabrod wrote:
eatnumber1 wrote:
ffwd wrote:2.6.24 zens are nice, but im still sticking to 2.6.23-kamikaze5 until can i have a working madwifi-ng and open-vm-tools without manually hacking the ebuilds/compiles/whatever (not to mention the power saving regression in 24's)...
I wasn't aware there was a power saving regression... what is it? Can you link to some info?
I stated this many times, waninkoko knows it.
Iirc this occured after pulling in some -rt stuff, now my notebook consumes about 2Watt more than vanilla!
For the meantime i'm back to 2.6-wireless-git.

Edit: dunno if there is any information to be found but i'm quite sure it's the -rt stuff. (as i said, vanilla's fine)

How do you know how many watts your nb consumes?

hirakendu · Post by **hirakendu** » Mon Jan 07, 2008 11:19 am

soft way - use this program called 'powertop' and run your notebook on battery. cruder way will be to run on battery for fixed time and calculate using battery percentage, total capacity (e.g., if typical 53 Watt-hr battery goes down to 90 % in 15 minutes, its roughly 20 watts) etc (or cat /proc/acpi/battery/BAT0/info etc). hard way - use wattmeter :p.

kernelOfTruth · Post by **kernelOfTruth** » Mon Jan 07, 2008 12:17 pm

*subscribes*

@buddabrod:

you tried disabling softirq & hardirq and measuring / comparing watt usage ?

buddabrod · Post by **buddabrod** » Mon Jan 07, 2008 1:47 pm

kernelOfTruth wrote:*subscribes*

@buddabrod:

you tried disabling softirq & hardirq and measuring / comparing watt usage ?

Yes and there was no difference..

But i have to admit... vipernicus release consumed the least power of all

So i'd really appreciate another release (for vipernicus

) based on stable .23

rmh3093 · Post by **rmh3093** » Mon Jan 07, 2008 2:02 pm

oRDeX wrote:
buddabrod wrote:
eatnumber1 wrote:
ffwd wrote:2.6.24 zens are nice, but im still sticking to 2.6.23-kamikaze5 until can i have a working madwifi-ng and open-vm-tools without manually hacking the ebuilds/compiles/whatever (not to mention the power saving regression in 24's)...
I wasn't aware there was a power saving regression... what is it? Can you link to some info?
I stated this many times, waninkoko knows it.
Iirc this occured after pulling in some -rt stuff, now my notebook consumes about 2Watt more than vanilla!
For the meantime i'm back to 2.6-wireless-git.

Edit: dunno if there is any information to be found but i'm quite sure it's the -rt stuff. (as i said, vanilla's fine)
How do you know how many watts your nb consumes?

an amp meter

eatnumber1 · Post by **eatnumber1** » Mon Jan 07, 2008 3:01 pm

buddabrod wrote:
eatnumber1 wrote:
ffwd wrote:2.6.24 zens are nice, but im still sticking to 2.6.23-kamikaze5 until can i have a working madwifi-ng and open-vm-tools without manually hacking the ebuilds/compiles/whatever (not to mention the power saving regression in 24's)...
I wasn't aware there was a power saving regression... what is it? Can you link to some info?
I stated this many times, waninkoko knows it.
Iirc this occured after pulling in some -rt stuff, now my notebook consumes about 2Watt more than vanilla!
For the meantime i'm back to 2.6-wireless-git.

Edit: dunno if there is any information to be found but i'm quite sure it's the -rt stuff. (as i said, vanilla's fine)

When you get a chance, can you try to git bisect to find the commit that caused the problem?

eatnumber1 · Post by **eatnumber1** » Mon Jan 07, 2008 3:18 pm

tranquilcool wrote:errors with latest git pull;

CC [M] drivers/media/video/bt8xx/bttv-driver.o
drivers/media/video/bt8xx/bttv-driver.c error: unknown field ‘vidioc_g_register’ specified in initializer
drivers/media/video/bt8xx/bttv-driver.c error: ‘bttv_g_register’ undeclared here (not in a function)
drivers/media/video/bt8xx/bttv-driver.c error: unknown field ‘vidioc_s_register’ specified in initializer
drivers/media/video/bt8xx/bttv-driver.c error: ‘bttv_s_register’ undeclared here (not in a function)
make[4]: *** [drivers/media/video/bt8xx/bttv-driver.o] Error 1
make[3]: *** [drivers/media/video/bt8xx] Error 2
make[2]: *** [drivers/media/video] Error 2
make[1]: *** [drivers/media] Error 2
make: *** [drivers] Error 2

Compiles fine for me... please post your .config

EDIT: I take that back... I'll look at it later.

EDIT: Fixed...

StifflerStealth · Post by **StifflerStealth** » Mon Jan 07, 2008 3:19 pm

buddabrod wrote:So i'd really appreciate another release (for vipernicus ) based on stable .23

He is working on a super kernel that is far superior to Linux, I expect by now. Or he is just resting in peace. That is ... if you believe in heaven. I'd like to think he is working on the universe's super kernel.

Could be wrong though.

Woot! New thread! I love the new thread smell.

hoacker · Post by **hoacker** » Mon Jan 07, 2008 8:31 pm

I was wondering why it was so silent about zen the last few days. Just found part II and subscribed

Have a problem with latest git sources (rc7). Some .config as for rc6 (except for new options set to "n") gives me an error:

LD .tmp_vmlinux1
drivers/built-in.o: In function `v4l2_i2c_attach':
(.text+0x95566): undefined reference to `i2c_attach_client'
make: *** [.tmp_vmlinux1] Error 1

Switching off the v4l stuff certainly removes the error.

x86_64, .config: http://pastebin.com/m589dac09

RobertDavid · Post by **RobertDavid** » Mon Jan 07, 2008 9:02 pm

Code: Select all

[root@robert-server zen-sources]# git-pull
remote: Counting objects: 3914, done.
remote: Compressing objects: 100% (805/805), done.
Indexing 3434 objects...
remote: Total 3434 (delta 2965), reused 3083 (delta 2629)
 100% (3434/3434) done
Resolving 2965 deltas...
 100% (2965/2965) done
288 objects were added to complete this thin pack.
* refs/remotes/origin/aufs: fast forward to branch 'aufs' of git://repo.or.cz/linux-2.6/zen-sources
  old..new: 88eee02..42a35b7
* refs/remotes/origin/madwifi: fast forward to branch 'madwifi' of git://repo.or.cz/linux-2.6/zen-sources
  old..new: e5cff40..01617ea
* refs/remotes/origin/master: forcing update to non-fast forward branch 'master' of git://repo.or.cz/linux-2.6/zen-sources
  old...new: 9d87b69...3ce5445
* refs/remotes/origin/master-devel: storing branch 'master-devel' of git://repo.or.cz/linux-2.6/zen-sources
  commit: 1704788
* refs/remotes/origin/uvc: fast forward to branch 'uvc' of git://repo.or.cz/linux-2.6/zen-sources
  old..new: 80ba9de..66c4402
* refs/remotes/origin/v4l-dvb: storing branch 'v4l-dvb' of git://repo.or.cz/linux-2.6/zen-sources
  commit: 6b15bb0
Auto-following refs/tags/v2.6.24-rc7-zen0
Auto-following refs/tags/v2.6.24-rc7
remote: Counting objects: 1, done.
remote: Total 1 (delta 0), reused 1 (delta 0)
Unpacking 1 objects...
 100% (1/1) done
* refs/tags/v2.6.24-rc7-zen0: storing tag 'v2.6.24-rc7-zen0' of git://repo.or.cz/linux-2.6/zen-sources
  commit: 1704788
* refs/tags/v2.6.24-rc7: storing tag 'v2.6.24-rc7' of git://repo.or.cz/linux-2.6/zen-sources
  tag: fcb31af
Auto-merged Documentation/kernel-parameters.txt
CONFLICT (content): Merge conflict in Documentation/kernel-parameters.txt
Auto-merged Makefile
CONFLICT (content): Merge conflict in Makefile
Auto-merged drivers/scsi/scsi_lib.c
Auto-merged fs/nfs/nfs4proc.c
Auto-merged include/linux/cpu.h
Auto-merged include/linux/cpumask.h
Auto-merged sound/pci/hda/hda_intel.c
Automatic merge failed; fix conflicts and then commit the result.
[root@robert-server zen-sources]# git-pull
You are in the middle of a conflicted merge.
[root@robert-server zen-sources]

Looks like some error in repository. Maybe I will solve that clean git cloning. (but I dont think it is mistake on my side)

tranquilcool · Post by **tranquilcool** » Mon Jan 07, 2008 11:47 pm

@RobertDavid

git reset --hard
git pull

yulo99 · Post by **yulo99** » Tue Jan 08, 2008 9:21 am

hi,
at first thank you for great work with zen! but I have the problem with tuxonice hibernation (to disk)on my IBM/Lenovo T60.
I hangs during suspending on "Going atomic". I was trying to detect what is the reason, so these are some facts:
- hibernation was working fine till 'linux-2.6.22-kamikaze9' no hibernation on 2.6.23 (kamikaze) and 24 (zen)
- I compiled all possible device drivers as modules (only some sata, ahci left), booted in single mode, removed all modules - hibernation failed
- lately I tried tuxonice-sources (2.6.23-r6) and it works (!) with the same kernel .config as tested before with kamikaze 2.6.23. So I started to test patches from kamikaze, result: after applying 'thinkpad-2.6.23-1.patch' hibernation failed on "going atomic"

The last test was done with kamikaze but I think that it is similar problem with zen.

Cheers
yulo.

RobertDavid · Post by **RobertDavid** » Tue Jan 08, 2008 9:31 am

tranquilcool wrote:@RobertDavid

git reset --hard
git pull

Thanks, worked.

omnimuc · Post by **omnimuc** » Tue Jan 08, 2008 2:16 pm

Hi,
just pulled 2.6.24-rc7-zen0
my config: http://pastebin.com/m6d5421c3

I just get something like:
Booting Zen-Neu
boot /vmlinuz-zen root=/dev/hda3
--some activity on the disk without any output--

after that nothing happens

no problem with 2.6.24-rc6-zen0-gb125960d

any Ideas?

Poe · Post by **Poe** » Tue Jan 08, 2008 2:18 pm

could you tell my why there are no ebuild - http://zen.waninkoko.info/ebuilds/ for 2.6.24? yes, file is exist, but it has only few lines of ebuild, I dont want 9999 version.

Official thread: "zen-sources" - Part II

t60 hibernate problem

won`t boot