Official thread: "zen-sources" - Part III

Message

hajit · Post by **hajit** » Sun Mar 09, 2008 12:35 pm

Hi.

Can you include this patch. It's for Acorp DS-110 dvb-card.
http://liplianin.at.tut.by/dm1105-liplianin.diff

Mod edit by NeddySeagoon. This thread is a continuation of Official thread: "zen-sources" - Part II]

kernelOfTruth · Post by **kernelOfTruth** » Sun Mar 09, 2008 1:27 pm

@NeddySeagoon:

could you please put the following at the beginning of this thread?
people keep on asking on how to check this kernel out etc etc,
thanks

checking the sources out

First Time
Code:
cd /usr/src
git clone git://zen-sources.org/zen/kernel.git
cd zen-sources
make menuconfig

creating a patch
example:

Code: Select all

git diff v2.6.24 v2.6.24-zen1 > 2.6.24-zen1.patch

update sources (inside the directory)

Code: Select all

Update
Code:

cd /usr/src/zen-sources
git pull

you are in the middle of a conflicted merge:

Code: Select all

git checkout -f && git pull

you need to change to another branch ?
example:

Code: Select all

# git checkout master-devel

if the branch doesn't exist in your directory (

Code: Select all

git branch

)
e.g. master-light, then enter

Code: Select all

git branch master-light origin/master-light && git checkout master-light

you want to check out a specific version?

Code: Select all

git checkout v2.6.25-rc5-zen0

speeding up merge (download) of zen-sources (during first time)
1) checkout the linux-2.6 branch of linus (git.eu.kernel.org or git.kernel.org)
2) then merge both (e.g. it's in /usr/src/sources/linux-2.6)
3)

Code: Select all

 git clone --reference /usr/src/sources/linux-2.6/ git://zen-sources.org/zen/kernel.git

what branches are there?
currently (may 2008), there are 5 (?) branches ( heads ):

- master (git checkout master) (e.g. v2.6.24-zen3) [very stable]
- master-new (git checkout master-new) (e.g. v2.6.25-zen1, afaik this serves as test-branch or if master isn't updated for some time as secondary master)
- master-light (git checkout master-light) (e.g. 2.6.25-rc5-ki0) [even more stable than master]
- master-devel (git checkout master-devel) (e.g. 2.6.25-rc5-zen0) [here goes the development]
- master-devel-rt (git checkout master-devel-rt) (e.g. v2.6.24-zen-rt3) [~ stable, for those needing realtime-stuff]

the other "head"s are branches which are added to the main tree

the project's site can be reached over at:
www.zen-sources.org

the repo is reachable via:
http://zen-sources.org/cgi-bin/gitweb.c ... ;a=summary

the SourceForge.net site at:
http://sourceforge.net/projects/zen-sources/ (thanks guys, for hosting

)

edit (June 13th 2008) :

updated to newer zen-sources server + webpage-address

dodo1122 · Post by **dodo1122** » Sun Mar 09, 2008 3:04 pm

2.6.24-zen4 tagged. Loads of cleanups, tuxonice updated, btrfs updated to 0.13 and some fixes.

enjoy ;P

dodo

Cobra78 · Post by **Cobra78** » Sun Mar 09, 2008 6:13 pm

The zen4 kernel had iwlwifi 1.2.23 witch don't work well with my ipw3945, there is a way to have 1.2.25 in zen kernel, the driver downloaded from official iwlwifi site dosen't compile -.-

EnigmaticSeraph · Post by **EnigmaticSeraph** » Sun Mar 09, 2008 9:03 pm

2.6.24-zen4 fails for me with:

Code: Select all

make[3]: *** No rule to make target `/usr/src/zen-sources-stable/include/linux/if_addrlabel.h', needed by `/usr/src/zen-sources-stable/usr/include/linux/if_addrlabel.h'.  Stop.

Greg KH accidentally pushed a patch intended for 2.6.25 (David's fault). This is the patch that needs to be reverted: http://lkml.org/lkml/2008/2/22/520
Simply take out the

Code: Select all

unifdef-y += if_addrlabel.h

line in include/linux/Kbuild

PS@kernelOfTruth: STILL haven't had the free time to go fix the config problem. :/ Hopefully will after this week.

R-Type · Post by **R-Type** » Mon Mar 10, 2008 1:41 am

Code: Select all

  CHK     include/linux/version.h
  CHK     include/linux/utsrelease.h
  CALL    scripts/checksyscalls.sh
  Building modules, stage 2.
  MODPOST 226 modules
ERROR: "sqlzma_fin" [fs/squashfs/squashfs.ko] undefined!
ERROR: "sqlzma_un" [fs/squashfs/squashfs.ko] undefined!
ERROR: "sqlzma_init" [fs/squashfs/squashfs.ko] undefined!
make[1]: *** [__modpost] Error 1
make: *** [modules] Error 2

zen4 seems to break squashfs. zen3 builds fine. Any suggestions?

ilikenwf · Post by **ilikenwf** » Mon Mar 10, 2008 7:58 pm

By chance, could you include the UDF patch in the 2.6.24 main branch? Lots of people want it, and kamikaze had it too...the guys at doom9 are having fits!

http://forum.doom9.org/showthread.php?s ... 727&page=2

deno · Post by **deno** » Mon Mar 10, 2008 9:20 pm

Is it posible to include vesa-tng patch? It would help some of us with 945 gm to have gensplash and be able to resume after hibernate. atm screen becomes black and it remains so, only thing one can do is to shutdown by holding the button.

gimpel · Post by **gimpel** » Mon Mar 10, 2008 9:30 pm

deno wrote:Is it posible to include vesa-tng patch? It would help some of us with 945 gm to have gensplash and be able to resume after hibernate. atm screen becomes black and it remains so, only thing one can do is to shutdown by holding the button.

What's wrong with uvesafb?

unK · Post by **unK** » Mon Mar 10, 2008 9:32 pm

gimpel wrote:What's wrong with uvesafb?

That's another point, but it also doesn't work with gcc-4.3.

gimpel · Post by **gimpel** » Mon Mar 10, 2008 9:35 pm

unK wrote:
gimpel wrote:What's wrong with uvesafb?
That's another point, but it also doesn't work with gcc-4.3.

uvesafb is in mainline, so if it doesn't work, post to lkml o.O

deno · Post by **deno** » Mon Mar 10, 2008 9:36 pm

@gimpel? I have mentioned that in my previous post. people who use uvesafb and v86d with intel cards cannot resume after hibernation. Wth vesa-tng some of them (i.e me) could do it, some again had problems with vesa-tng also. But for me vesa-tng worked excellent.

Jupiter1TX · Post by **Jupiter1TX** » Mon Mar 10, 2008 9:38 pm

deno wrote:Is it posible to include vesa-tng patch? It would help some of us with 945 gm to have gensplash and be able to resume after hibernate. atm screen becomes black and it remains so, only thing one can do is to shutdown by holding the button.

Why not use uvesafb. I switched from vesa-tng with no problems as per Spocks
instructions.
http://dev.gentoo.org/~spock/projects/uvesafb/

IvanMajhen · Post by **IvanMajhen** » Mon Mar 10, 2008 9:39 pm

I have the same problem with uvesafb. After hibernation, screen is blank. In vesa-tng era i had little initramfs image with 915resolution in it and nice widescreen resolution. And it worked great. Now if i done the same with uvesafb (insmoding uvesafb in initramfs after 915resolution) it works... but hibernation (tuxonice) fails every time.

Don't know why vesafb-tng isn't maintained anymore... sad

unK · Post by **unK** » Mon Mar 10, 2008 9:58 pm

gimpel wrote:uvesafb is in mainline, so if it doesn't work, post to lkml o.O

Well, actually it is not that uvesafb doesn't work, it's v86d. see [bug=212531]bug 212531[/bug].

deno · Post by **deno** » Mon Mar 10, 2008 9:59 pm

Jupiter1TX, that's fine with me if you use uvesafb. I was using it for a while but as I mentioned I cannot resume after suspending to ram or to disk.

@Ivan what chip do you have? I never needed i915 res for 1280x800 and did't have to use initrd.

deno · Post by **deno** » Mon Mar 10, 2008 10:03 pm

True it is because of v86d but that doesn't change anything here. It is about getting gensplash and nice looking tty's. With vesa-tng we don't need v86d and that's the point.

IvanMajhen · Post by **IvanMajhen** » Mon Mar 10, 2008 11:00 pm

deno wrote:Jupiter1TX, that's fine with me if you use uvesafb. I was using it for a while but as I mentioned I cannot resume after suspending to ram or to disk.

@Ivan what chip do you have? I never needed i915 res for 1280x800 and did't have to use initrd.

Code: Select all

00:02.1 Display controller: Intel Corporation Mobile 945GM/GMS, 943/940GML Express Integrated Graphics Controller (rev 03)

This is because you have 1280x800 resolution in video bios. I don't.
Uvesafb is good for other arches, but it is just stupid to use it under 32bit x86.
I switched to intelfb, everything works, i have 100% working s3 and s4. But can't stand ugly 1024x768 console on 1280x800 screen.
Vesafb-tng was perfect for me.

Waninkoko · Post by **Waninkoko** » Mon Mar 10, 2008 11:01 pm

ilikenwf wrote:By chance, could you include the UDF patch in the 2.6.24 main branch? Lots of people want it, and kamikaze had it too...the guys at doom9 are having fits!

http://forum.doom9.org/showthread.php?s ... 727&page=2

I'll include it in master-devel and master-light.

ilikenwf · Post by **ilikenwf** » Tue Mar 11, 2008 12:40 am

Waninkoko wrote:
ilikenwf wrote:By chance, could you include the UDF patch in the 2.6.24 main branch? Lots of people want it, and kamikaze had it too...the guys at doom9 are having fits!

http://forum.doom9.org/showthread.php?s ... 727&page=2
I'll include it in master-devel and master-light.

Thanks! The doom9 guys will jump for joy!

Out of curiosity, what is the difference between master-light and master-devel?

Oh..and the zen homepage is down.

dodo1122 · Post by **dodo1122** » Tue Mar 11, 2008 4:47 pm

master-light is the continuation of klight-sources. Basically less patches than master-devel;P

dodo

gimpel · Post by **gimpel** » Tue Mar 11, 2008 8:49 pm

Speaking of uvesafb...
Anyone else getting a kernel panic with current master? Something about swapper tainted vgacon and stuff. It was fine until -rc4-zen0.

I start to wish back vesafb-tng too, no -rt, no 1680x1050, and now it panics anyway, hmm...

Jupiter1TX · Post by **Jupiter1TX** » Wed Mar 12, 2008 1:27 am

IvanMajhen wrote: Uvesafb is good for other arches, but it is just stupid to use it under 32bit x86.

Am just a little curious why you think this is stupid. I have been using this since
the change over with no problems. I use 32bit x86 on a 22" 1680x1050 lcd. Is
there a good reason why this is stupid or did you just make a stupid remark.

IvanMajhen · Post by **IvanMajhen** » Wed Mar 12, 2008 12:25 pm

It is stupid because:

i need to compile klibc for it
needs userspace daemon
needs initramfs if we early wanna see resolution
it simulates x86 calls to query video bios--> why should we need that if we have native calls on 32bit
it doesn't work with tuxonice for me ,and for some other users
it lacks any type of acceleration

The only good thing is that it supports non vesa modes, but vesafb-tng already did that.

ufoman · Post by **ufoman** » Wed Mar 12, 2008 6:56 pm

Any fix for that:

Code: Select all

Initialized empty Git repository in /usr/portage/distfiles/git-src/zen-sources/
remote: Counting objects: 849556, done.
remote: fatal: packfile ./objects/pack/pack-de772f05b2998c8653e84e335eb4fde2dc68a15b.pack cannot be accessed
remote: aborting due to possible repository corruption on the remote side.
fatal: early EOF
fatal: index-pack failed
fetch-pack from 'git://repo.or.cz/linux-2.6/zen-sources.git' failed.

Official thread: "zen-sources" - Part III

Official thread: "zen-sources" - Part III

2.6.24-zen4 failure