RIFS-ES Linux Kernel Scheduler Released

[3766691]

Chen,
could you look into the sleepy thingy once more?
I recently had a little annoyance with my wlan, this happened two times :
- doing something not internet related (watching soccer dvb-t using kaffeine) longer than an hour
- lost my wlan internet connection
The same happened directly after this a second time the same conditions.

Why I think it is RIFS related: On Gentoo, normaly loosing connection it is sufficient to
rc-service dhcpcd restart

But I had to do:
rc-service dhcpcd stop
rc-service wpa_supplicant restart
rc-service dhcpcd start

Which indicates a more deeply my broadcom-sta-wl module involved ...

I think this is not related to the scheduler.(Maybe I have to do a test to prove)
Actually Linux kernel sucks with wireless

[ulenrich]

Chen,
really? Please look closely into wakeup features before publishing!

more generally I meant:
If a connection I have to reassure in certain time frames,
could this be a side effect of the RIFS:
Missing an answer in a timely manner?
Which disconnected my wlan because of a timeout?

[3766691]

Chen,
really? Please look closely into wakeup features before publishing!

more generally I meant:
If a connection I have to reassure in certain time frames,
could this be a side effect of the RIFS:
Missing an answer in a timely manner?
Which disconnected my wlan because of a timeout?

I would try to made the scheduler in modular form and find whether it can be solved.

[ulenrich]

Chen,
I just wanted to say, could you test more closely if
RIFS suits needs and requirements of wakeup threads in general terms.

[3766691]

Chen,
I just wanted to say, could you test more closely if
RIFS suits needs and requirements of wakeup threads in general terms.

If you have time please contact with me by mail and it could be better for me to solve and test.
Chen

[3766691]

++

I second that


will try it next week - by then additional fixes for btrfs probably will get in and the kernel should be pretty usable/stable

Would you please share the patch? Would love to test it.

Here is RIFS.ES-V1 patch For kernel 3.5.x:
http://rifs-scheduler.googlecode.com...v1-kernel3.5.x
I am now using this

[TAXI]

Here is RIFS.ES-V1 patch For kernel 3.5.x

There seems to be an error with NO_HZ enabled:

kernel/time/tick-sched.c: In Funktion »tick_nohz_idle_exit«:
kernel/time/tick-sched.c:584:2: Fehler: Implizite Deklaration der Funktion »update_cpu_load_nohz« [-Werror=implicit-function-declaration]

Also with CGROUPS it won't compile:

LD vmlinux
kernel/built-in.o.data+0x43e0): undefined reference to `cpu_cgroup_subsys'
make: *** [vmlinux] Fehler 1

[kernelOfTruth]

There seems to be an error with NO_HZ enabled:

kernel/time/tick-sched.c: In Funktion »tick_nohz_idle_exit«:
kernel/time/tick-sched.c:584:2: Fehler: Implizite Deklaration der Funktion »update_cpu_load_nohz« [-Werror=implicit-function-declaration]

Also with CGROUPS it won't compile:

LD vmlinux
kernel/built-in.o.data+0x43e0): undefined reference to `cpu_cgroup_subsys'
make: *** [vmlinux] Fehler 1

the problem with cgroups is a known issue - afaik RIFS doesn't support cgroups (yet)

[3766691]

There seems to be an error with NO_HZ enabled:

kernel/time/tick-sched.c: In Funktion »tick_nohz_idle_exit«:
kernel/time/tick-sched.c:584:2: Fehler: Implizite Deklaration der Funktion »update_cpu_load_nohz« [-Werror=implicit-function-declaration]

Also with CGROUPS it won't compile:

LD vmlinux
kernel/built-in.o.data+0x43e0): undefined reference to `cpu_cgroup_subsys'
make: *** [vmlinux] Fehler 1

Please disable them. Yes desktop doesn't need these

EDIT 2:
Now I post a new one.

[3766691]

the problem with cgroups is a known issue - afaik RIFS doesn't support cgroups (yet)

OK Now I totally disabled cgroups and non-preemptible and voluntary-sleep preemption model, only preemptible option is left.Tickless is also disabled. These are the feature that desktop users won't care about