Zen-Kernel Discussion/Support Thread - Part 4

Message

Strowi · Post by **Strowi** » Sat Aug 21, 2010 5:07 pm

mhh with 2.6.35-zen1 i get this when trying to make on amd64:


root@Sleipnir:/home/strowi/src/zen-stable-[git/master]> make
scripts/kconfig/conf -s arch/x86/Kconfig
  CHK     include/linux/version.h
  CHK     include/generated/utsrelease.h
  CC      arch/x86/kernel/asm-offsets.s
In file included from include/linux/module.h:18,
                 from include/linux/crypto.h:21,
                 from arch/x86/kernel/asm-offsets_64.c:8,
                 from arch/x86/kernel/asm-offsets.c:4:
include/linux/tracepoint.h: In Funktion »tracepoint_synchronize_unregister«:
include/linux/tracepoint.h:80: Fehler: Implizite Deklaration der Funktion »synchronize_sched«
make[1]: *** [arch/x86/kernel/asm-offsets.s] Fehler 1
make: *** [prepare0] Fehler 2

Any ideas?

Jinxter2K7 · Post by **Jinxter2K7** » Sat Aug 21, 2010 6:51 pm

My guess is that you have Classic RCU enabled..

Strowi · Post by **Strowi** » Sat Aug 21, 2010 7:02 pm

Thx, problem solved, wonder how it got enabled... always have been using tree...
Anyway it compiled fine and seems to do the job well!

cheater1034 · Post by **cheater1034** » Sun Aug 22, 2010 1:42 am

albright wrote:
A tar snapshot of latest stable *2.6.35-r* for the portage
yeah to that ... we're already up to 35.3 I think

1. I have no control over portagew
2. zen1 release is done and final but I need to get in touch with damentz or dodo (who is unavailable atm) before i can upload the tar

peter4 · Post by **peter4** » Sun Aug 22, 2010 2:24 am

Hi,

do you guys (the zen guys) mess with the intel graphics drivers? I'm having problems with kwin crashing and at kde bugzilla they're saying it's a driver issue: https://bugs.kde.org/show_bug.cgi?id=241263

I'd like to make sure before reporting the bug on gentoo bugzilla.

rahulthewall · Post by **rahulthewall** » Sun Aug 22, 2010 7:29 am

peter4 wrote:Hi,

do you guys (the zen guys) mess with the intel graphics drivers? I'm having problems with kwin crashing and at kde bugzilla they're saying it's a driver issue: https://bugs.kde.org/show_bug.cgi?id=241263

I'd like to make sure before reporting the bug on gentoo bugzilla.

See this bug: https://bugs.kde.org/show_bug.cgi?id=241402

Try to use Xrender as your compositing backed, it will solve the problems.

shaumux · Post by **shaumux** » Sun Aug 22, 2010 7:32 am

Whats the vhba and scholastic blue in 2.6.35-zen1?
its was displayed at the initial screen as included but couldn't find it in the config screen.

peter4 · Post by **peter4** » Sun Aug 22, 2010 7:35 am

rahulthewall wrote: Try to use Xrender as your compositing backed, it will solve the problems.

With xrender backend, kwin says it can't enable half the effects...

No offense, but it's a bit like saying "your 3d driver is bugged and crashes? no biggie, just use the vesa driver!"

rahulthewall · Post by **rahulthewall** » Sun Aug 22, 2010 9:30 am

peter4 wrote:
rahulthewall wrote: Try to use Xrender as your compositing backed, it will solve the problems.
With xrender backend, kwin says it can't enable half the effects...

No offense, but it's a bit like saying "your 3d driver is bugged and crashes? no biggie, just use the vesa driver!"

I know man. I am just pointing out that it is a problem that many users are having and thus not limited to the Zen kernel. OpenGL compositing in KDE 4.5 is buggy, very buggy (for me).

See this as well: https://bugs.kde.org/show_bug.cgi?id=247839

peter4 · Post by **peter4** » Sun Aug 22, 2010 2:54 pm

Hi,

on the just-synced zen-stable I'm getting this:

Code: Select all

[   93.340744] rt73usb 1-6:1.0: no reset_resume for driver rt73usb?
[   93.344016] ------------[ cut here ]------------
[   93.344023] WARNING: at net/mac80211/work.c:884 ieee80211_work_work+0x2ef/0x13b0()
[   93.344025] Hardware name: EP45-UD3R
[   93.344026] work scheduled while going to suspend
[   93.344028] Modules linked in: nvidia(P) snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep rt73usb rt2x00usb rt2x00lib r8169 mii scsi_wait_scan usbhid uhci_hcd usb_storage hid ehci_hcd usbcore ata_piix ahci libahci pata_jmicron libata
[   93.344043] Pid: 4452, comm: async/27 Tainted: P            2.6.35-zen1+ #3
[   93.344045] Call Trace:
[   93.344051]  [<ffffffff8103aabb>] ? warn_slowpath_common+0x7b/0xc0
[   93.344054]  [<ffffffff8103abb5>] ? warn_slowpath_fmt+0x45/0x50
[   93.344057]  [<ffffffff813af93f>] ? ieee80211_work_work+0x2ef/0x13b0
[   93.344061]  [<ffffffff811d362c>] ? string+0x4c/0xe0
[   93.344064]  [<ffffffff81036c60>] ? default_wake_function+0x0/0x20
[   93.344067]  [<ffffffff813b0a74>] ? ieee80211_work_purge+0x74/0xd0
[   93.344070]  [<ffffffff813b1857>] ? ieee80211_stop+0x57/0x4e0
[   93.344073]  [<ffffffff8133c44a>] ? dev_deactivate+0x1aa/0x1d0
[   93.344077]  [<ffffffff81328551>] ? __dev_close+0x41/0x70
[   93.344079]  [<ffffffff81328592>] ? dev_close+0x12/0x40
[   93.344082]  [<ffffffff81328615>] ? rollback_registered_many+0x55/0x2c0
[   93.344085]  [<ffffffff8132888e>] ? unregister_netdevice_many+0xe/0x70
[   93.344088]  [<ffffffff813b0da4>] ? ieee80211_remove_interfaces+0xa4/0xe0
[   93.344092]  [<ffffffff813a2536>] ? ieee80211_unregister_hw+0x46/0x100
[   93.344097]  [<ffffffffa00dea29>] ? rt2x00lib_remove_dev+0xc9/0xd0 [rt2x00lib]
[   93.344101]  [<ffffffffa0054136>] ? rt2x00usb_disconnect+0x36/0x80 [rt2x00usb]
[   93.344111]  [<ffffffffa00a6254>] ? usb_unbind_interface+0xc4/0x120 [usbcore]
[   93.344115]  [<ffffffff81282090>] ? __device_release_driver+0x60/0xd0
[   93.344118]  [<ffffffff812821d5>] ? device_release_driver+0x25/0x40
[   93.344126]  [<ffffffffa00a64dc>] ? usb_rebind_intf+0x1c/0x60 [usbcore]
[   93.344133]  [<ffffffffa00a65bd>] ? do_unbind_rebind+0x7d/0xb0 [usbcore]
[   93.344141]  [<ffffffffa00a66ce>] ? usb_resume+0xde/0x180 [usbcore]
[   93.344144]  [<ffffffff81286ac2>] ? pm_op+0x1a2/0x1c0
[   93.344147]  [<ffffffff81286cc8>] ? device_resume+0x88/0x130
[   93.344151]  [<ffffffff81059740>] ? async_thread+0x0/0x260
[   93.344153]  [<ffffffff81286d84>] ? async_resume+0x14/0x40
[   93.344156]  [<ffffffff81059845>] ? async_thread+0x105/0x260
[   93.344159]  [<ffffffff81036c60>] ? default_wake_function+0x0/0x20
[   93.344162]  [<ffffffff81059740>] ? async_thread+0x0/0x260
[   93.344165]  [<ffffffff81052afe>] ? kthread+0x8e/0xa0
[   93.344168]  [<ffffffff81003b94>] ? kernel_thread_helper+0x4/0x10
[   93.344171]  [<ffffffff81052a70>] ? kthread+0x0/0xa0
[   93.344178]  [<ffffffff81003b90>] ? kernel_thread_helper+0x0/0x10
[   93.344179] ---[ end trace 8223f2954477be04 ]---

every time I resume from hibernate. Any thoughts?

HecHacker1 · Post by **HecHacker1** » Wed Aug 25, 2010 10:10 pm

Any chance you can revert this patch:

http://lkml.org/lkml/2010/6/17/134

I am having the same problem as reported. Basically, every few seconds my mouse freezes and the kslowd000 process uses all of my CPU. I don't know if this has been fixed upstream yet.

kernelOfTruth · Post by **kernelOfTruth** » Thu Aug 26, 2010 7:36 pm

a very interesting low-latency patchset for CFS (!) is coming up:

RFC PATCH 00/11] sched: CFS low-latency features

kudos to Mathieu Desnoyers & Peter Zijlstra and Nokia !

kernelOfTruth · Post by **kernelOfTruth** » Sat Aug 28, 2010 10:38 am

kernelOfTruth wrote:a very interesting low-latency patchset for CFS (!) is coming up:

RFC PATCH 00/11] sched: CFS low-latency features

kudos to Mathieu Desnoyers & Peter Zijlstra and Nokia !

@Steven Barrett:

thanks for adding !

for this to work it needs to be explicitly enabled - only do if you really use your box as a desktop

otherwise throughput / performance might be negatively affected:

for opt in DYN_MIN_VRUNTIME \
NO_FAIR_SLEEPERS FAIR_SLEEPERS_INTERACTIVE FAIR_SLEEPERS_TIMER \
INTERACTIVE TIMER \
INTERACTIVE_FORK_EXPEDITED TIMER_FORK_EXPEDITED;
do echo $opt > /sys/kernel/debug/sched_features;
done

edit:

this might break waking up after suspending (X froze for me)

so watch if it works for you before and after enabling those features

P.Kosunen · Post by **P.Kosunen** » Sat Aug 28, 2010 11:31 am

What are requirements for BFQ?

Just installed sys-kernel/zen-sources-2.6.34_p1-r2, but can't find bfq with menuconfig search or grepping .config.

nico-- · Post by **nico--** » Sun Aug 29, 2010 12:51 am

P.Kosunen wrote:What are requirements for BFQ?

Just installed sys-kernel/zen-sources-2.6.34_p1-r2, but can't find bfq with menuconfig search or grepping .config.

There's a bfq use flag that you have to enable.

Saundersx · Post by **Saundersx** » Tue Aug 31, 2010 1:47 am

Just want to say I compiled the latest kernel and the "CFS low-latency features" seem to improve responsiveness considerably, and I don't believe it's a placebo effect

kernelOfTruth · Post by **kernelOfTruth** » Tue Aug 31, 2010 10:14 am

Saundersx wrote:Just want to say I compiled the latest kernel and the "CFS low-latency features" seem to improve responsiveness considerably, and I don't believe it's a placebo effect

you enabled it explicitely ?

for this you need to compile debugfs in / activate it and activate the low-latency CFS features (like mentioned above):

Code: Select all

for opt in DYN_MIN_VRUNTIME \ 
NO_FAIR_SLEEPERS FAIR_SLEEPERS_INTERACTIVE FAIR_SLEEPERS_TIMER \ 
INTERACTIVE TIMER \ 
INTERACTIVE_FORK_EXPEDITED TIMER_FORK_EXPEDITED; 
do echo $opt > /sys/kernel/debug/sched_features; 
done

anyways: yeah, it's really responsive, but throughput get's really down

Anon-E-moose · Post by **Anon-E-moose** » Tue Aug 31, 2010 11:11 am

Has the guy(s) maintaining the website fallen off the face of the earth?

Edit to add: Not bitching, just curious.

dodo1122 · Post by **dodo1122** » Tue Aug 31, 2010 11:29 am

damentz doesnt have access yet, cheater-conrad is busy and i just came back from holidays and i'm without my linux machine atm. i'll upload the patches as soon as i have access to linux and zen git repo.

cheater1034 · Post by **cheater1034** » Tue Aug 31, 2010 12:56 pm

I gotta get linux reinstalled on my laptop, so... :\

Ant P. · Post by **Ant P.** » Tue Aug 31, 2010 9:32 pm

Can we have compcache upstream's patches instead of the in-kernel staging driver? According to their bugtracker (and my own experience with them) the kernel one doesn't work.

cheater1034 · Post by **cheater1034** » Wed Sep 01, 2010 7:03 am

Ant_P wrote:Can we have compcache upstream's patches instead of the in-kernel staging driver? According to their bugtracker (and my own experience with them) the kernel one doesn't work.

I'll look into it and just fix the staging driver if that's the case

benneque · Post by **benneque** » Wed Sep 01, 2010 1:56 pm

Still no zen 35 in portage

my nouveau driver is waiting for 35

Yuu · Post by **Yuu** » Wed Sep 01, 2010 3:27 pm

Is this project is still active ? There's no more releases, and their website is down from here (and here)

I'm using the 2.6.34-zen1, I'm very happy whith it, and I'm desperately waiting for the 2.6.35 :/

benneque · Post by **benneque** » Wed Sep 01, 2010 3:43 pm

Huh? Homepage is alive: http://zen-kernel.org/

Zen-Kernel Discussion/Support Thread - Part 4

upstream bug