2.6.31 goes stable

combuster · 2009-09-10 12:13:31

http://wiki.archlinux.org/index.php/Ker … rom_Source

Follow the instructions in the wiki, no big deal...

nicosaturno · 2009-09-10 12:17:09

thanks combuster and flamelab that's a great community ;D

Acecero · 2009-09-10 17:22:16

combuster wrote:

Looks like they mess(a)ed things up this time Gonna try with git snapshots and see what happenes...
No changes in performance between 2.6.31-rc9 and 2.6.31 stable...

I guess you can't expect much difference between the last RC and the first stable. Thanks for trying.

nicosaturno · 2009-09-10 17:48:15

compiled the kernel and KMS is enabled (thanks guys ) KMS is not so fast, but I just wanted to give a try.
If needed I can publish the kernel .tar.gz somewhere.

bye,
Nico

combuster · 2009-09-10 18:13:31

@Acecero

Well there was some last minute commit's, hope dies last

mente · 2009-09-10 21:48:53

Took kernel26 PKGBUILD and slightly modified it for 2.6.31 compilation. Compiled, installed, good. But loading system takes several minutes instead 20 seconds previous!
My changes: removed archlinux patches. Most long loader is udev. It takes more than 1.30 minutes. Any ideas?
P.S. System: x86 laptop built on Intel i945.
updated: Fixed the issue with long time loading. I haven't updated kernel26-firmware

Last edited by mente (2009-09-11 08:43:16)

Pierre · 2009-09-10 22:13:12

Why don't you try the package from [testing]?

foutrelis · 2009-09-10 22:14:59

@mente: Give the kernel that is in [testing] right now a try.

Anonymo · 2009-09-10 22:20:13

nicosaturno wrote:

compiled the kernel and KMS is enabled (thanks guys ) KMS is not so fast, but I just wanted to give a try.
If needed I can publish the kernel .tar.gz somewhere.
bye,
Nico

Is this i686 with bfs?
I needed this file since it takes over 4 hours on this old computer.

flamelab · 2009-09-10 22:57:10

Pierre wrote:

Why don't you try the package from [testing]?

Will you enable Radeon kms support before it passes to core ? KMS will be enabled on boot when xf86-video-ati 6.13 and mesa 7.6 (and ati-dri, libgl) pass to extra. Till then, I don't think there is any problem to enable it.

mente · 2009-09-11 07:21:43

foutrelis wrote:

@mente: Give the kernel that is in [testing] right now a try.

Ok, I'll try. I have forgotten about [testing] repo

mente · 2009-09-11 07:41:21

Logging in as anonymous ... Logged in!
==> SYST ... done.    ==> PWD ... done.
==> TYPE I ... done.  ==> CWD /other/kernel26 ... done.
==> SIZE patch-2.6.31-1-ARCH.bz2 ... done.
==> PASV ... done.    ==> RETR patch-2.6.31-1-ARCH.bz2 ...
No such file `patch-2.6.31-1-ARCH.bz2'.

==> ERROR: Failure while downloading patch-2.6.31-1-ARCH.bz2
    Aborting...

Ahaha. Tried with ABS. Giving a try with removed Arch patches.

pedepy · 2009-09-11 07:55:08

emaemmmm .. im trying to compile xf86-video-ati-git from AUR to try kms (using the 2.6.31-1 kernel from [testing], mind you) and i keep getting this:

In file included from /usr/include/xorg/randrstr.h:47,
                 from radeon_probe.h:42,
                 from radeon.h:69,
                 from radeon_accel.c:79:
/usr/include/xorg/rrtransform.h:33: error: field 'f_transform' has incomplete type
/usr/include/xorg/rrtransform.h:34: error: field 'f_inverse' has incomplete type
In file included from radeon_probe.h:42,
                 from radeon.h:69,
                 from radeon_accel.c:79:
/usr/include/xorg/randrstr.h:128: error: field 'f_transform' has incomplete type
/usr/include/xorg/randrstr.h:129: error: field 'f_inverse' has incomplete type
In file included from radeon_probe.h:44,
                 from radeon.h:69,
                 from radeon_accel.c:79:
/usr/include/xorg/xf86Crtc.h:315: error: field 'f_crtc_to_framebuffer' has incomplete type
/usr/include/xorg/xf86Crtc.h:316: error: field 'f_framebuffer_to_crtc' has incomplete type
make[2]: *** [radeon_accel.lo] Error 1
make[1]: *** [all-recursive] Error 1
make: *** [all] Error 2

... i had xorg-server from testing as well, but i also tried with the 'real' one and i get the same errors ...

also, is kms even enabled in the testing kernel ? radeon.modeset=1 option in grub returns an unknown error, however when i use it everything is SLOW .. i mean i get 19 fps in glxgears ..

help.

Perry3D · 2009-09-11 07:55:56

In the testing repo config file: # CONFIG_DRM_RADEON_KMS is not set.
Is it enabled or not?

/edit: ok, i have to wait until madwifi is updated.

Last edited by Perry3D (2009-09-11 08:01:01)

mente · 2009-09-11 08:42:41

Fixed the issue with long time loading. I haven't updated kernel26-firmware

nicosaturno · 2009-09-11 09:26:10

Anonymo wrote:

nicosaturno wrote:
compiled the kernel and KMS is enabled (thanks guys ) KMS is not so fast, but I just wanted to give a try.
If needed I can publish the kernel .tar.gz somewhere.
bye,
Nico
Is this i686 with bfs?
I needed this file since it takes over 4 hours on this old computer.

yes i've compiled the kernel26-bfs with KMS but i've disabled some new addiction.
If needed I can upload it on rapidshare and you can give a try.

Beini · 2009-09-11 10:53:19

mente wrote:

Logging in as anonymous ... Logged in!
==> SYST ... done.    ==> PWD ... done.
==> TYPE I ... done.  ==> CWD /other/kernel26 ... done.
==> SIZE patch-2.6.31-1-ARCH.bz2 ... done.
==> PASV ... done.    ==> RETR patch-2.6.31-1-ARCH.bz2 ...
No such file `patch-2.6.31-1-ARCH.bz2'.

==> ERROR: Failure while downloading patch-2.6.31-1-ARCH.bz2
    Aborting...

Ahaha. Tried with ABS. Giving a try with removed Arch patches.

You can get the ARCH patches from git:

$ git clone git://projects.archlinux.org/linux-2.6-ARCH.git
$ cd linux-2.6-ARCH
$ ./gen_kernel_patch

mente · 2009-09-11 14:25:22

Git... Should try it in PKGBUILD

bangkok_manouel · 2009-09-11 14:31:05

patches are mirrored too ftp://ftp.archlinux.org/other/kernel26/

Beini · 2009-09-11 14:40:23

bangkok_manouel wrote:

patches are mirrored too ftp://ftp.archlinux.org/other/kernel26/

They are now

So no need for the git, but when I was compiling the kernel I couldn't find the 2.6.31 patch from the mirrors

anrxc · 2009-09-11 14:45:28

Nothing new in the Intel land. ET Radar still drops to 5 FPS and crashes on my GM965 as soon as there is some rain and thunder (amazing effects) and all that with the minimal setup possible, borderline with cheating.

combuster · 2009-09-11 16:04:33

Mesa's fault, you should see their commit's, popping up all the time They are really busy these days...

anrxc · 2009-09-11 17:02:16

I wrote about SwapbuffersWait several times before, and since you guys are seeing some performance drop I will mention it again. Here glxgears went from 300 to 500 and game play improved just a bit with:

# Improve games that render frames at less than refresh rate
Option "SwapbuffersWait" "false"

But as a side effect you might get tearing... search for the initial commit for more information.

dhave · 2009-09-12 18:58:37

I assume from reading the kernelnewbies page on 2.6.31 that online defrag for ext4 is still not there yet. Is this correct?

No big deal, really, but I thought I'd like to try this as soon as it's roadworthy.

lucke · 2009-09-12 19:14:32

Defragmentation is there - EXT4_IOC_MOVE_EXT ioctl. Search the ext4 mailing list for the userspace component. I managed to defragment a newly created partition, for some reason couldn't defragment my old partition. I'm not sure how final all this is; seems more final than not.

Arch Linux

#26 2009-09-10 12:13:31

Re: 2.6.31 goes stable

#27 2009-09-10 12:17:09

Re: 2.6.31 goes stable

#28 2009-09-10 17:22:16

Re: 2.6.31 goes stable

#29 2009-09-10 17:48:15

Re: 2.6.31 goes stable

#30 2009-09-10 18:13:31

Re: 2.6.31 goes stable

#31 2009-09-10 21:48:53

Re: 2.6.31 goes stable

#32 2009-09-10 22:13:12

Re: 2.6.31 goes stable

#33 2009-09-10 22:14:59

Re: 2.6.31 goes stable

#34 2009-09-10 22:20:13

Re: 2.6.31 goes stable

#35 2009-09-10 22:57:10

Re: 2.6.31 goes stable

#36 2009-09-11 07:21:43

Re: 2.6.31 goes stable

#37 2009-09-11 07:41:21

Re: 2.6.31 goes stable

#38 2009-09-11 07:55:08

Re: 2.6.31 goes stable

#39 2009-09-11 07:55:56

Re: 2.6.31 goes stable

#40 2009-09-11 08:42:41

Re: 2.6.31 goes stable

#41 2009-09-11 09:26:10

Re: 2.6.31 goes stable

#42 2009-09-11 10:53:19

Re: 2.6.31 goes stable

#43 2009-09-11 14:25:22

Re: 2.6.31 goes stable

#44 2009-09-11 14:31:05

Re: 2.6.31 goes stable

#45 2009-09-11 14:40:23

Re: 2.6.31 goes stable

#46 2009-09-11 14:45:28

Re: 2.6.31 goes stable

#47 2009-09-11 16:04:33

Re: 2.6.31 goes stable

#48 2009-09-11 17:02:16

Re: 2.6.31 goes stable

#49 2009-09-12 18:58:37

Re: 2.6.31 goes stable

#50 2009-09-12 19:14:32

Re: 2.6.31 goes stable

Board footer