Home » Mailing list archives » linux-kernel » 2007 » October » 8

Re: [PATCH] [3/6] scheduler: Do devirtualization for sched_fair

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
[view in full thread]
From: Andi Kleen <ak@...>
To: Ingo Molnar <mingo@...>
Cc: <linux-kernel@...>
Subject: Re: [PATCH] [3/6] scheduler: Do devirtualization for sched_fair
Date: Monday, October 8, 2007 - 8:32 am

> hm, i'm not convinced about this one. It increases the code size a bit 

Tiny bit (<200 bytes) and the wait_for/sleep_on refactor patch in the series
saves over 1K so I should have some room for code size increase. Overall
it will be still considerable smaller.


Unfortunately not -- for this to work (especially for inlining) requires to 
#include files implementing the sub calls. Except for the scheduler that 
is pretty uncommon unfortunately. Also the situation regarding which
call target is the common one is typically much less clear than with 
sched_fair / other scheduling classes.

I considered making it generic, but I don't think it would make sense
at the current time.

Also most paths are not as time critical as the scheduler.


It might have been measurable if the context switch was measurable at all.
Unfortunately the lmbench3 lat_ctx test I tired fluctuated by itself
over 50%. Ok I suppose it would be possible to instrument the kernel itself
to measure cycles. Would that convince you?

-Andi
-
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
Messages in current thread:
[PATCH] [0/6] Some scheduler changes for sched-devel, Andi Kleen, (Sun Oct 7, 4:59 pm)
[PATCH] [5/6] scheduler: Protect important kernel threads ag..., Andi Kleen, (Sun Oct 7, 4:59 pm)
Re: [PATCH] [5/6] scheduler: Protect important kernel thread..., Ingo Molnar, (Mon Oct 8, 7:51 am)
Re: [PATCH] [5/6] scheduler: Protect important kernel thread..., Andi Kleen, (Mon Oct 8, 8:33 am)
Re: [PATCH] [5/6] scheduler: Protect important kernel thread..., Ingo Molnar, (Mon Oct 8, 8:43 am)
Re: [PATCH] [5/6] scheduler: Protect important kernel thread..., Andi Kleen, (Mon Oct 8, 9:08 am)
[PATCH] [4/6] scheduler: Refactor normalize_rt_tasks, Andi Kleen, (Sun Oct 7, 4:59 pm)
Re: [PATCH] [4/6] scheduler: Refactor normalize_rt_tasks, Ingo Molnar, (Mon Oct 8, 7:44 am)
[PATCH] [6/6] scheduler: Remove bogus comment in sched_group..., Andi Kleen, (Sun Oct 7, 4:59 pm)
Re: [PATCH] [6/6] scheduler: Remove bogus comment in sched_g..., Ingo Molnar, (Mon Oct 8, 7:52 am)
[PATCH] [2/6] scheduler: Refactor common code of sleep_on / ..., Andi Kleen, (Sun Oct 7, 4:59 pm)
Re: [PATCH] [2/6] scheduler: Refactor common code of sleep_o..., Andi Kleen, (Sun Oct 7, 6:22 pm)
Re: [PATCH] [2/6] scheduler: Refactor common code of sleep_o..., Ingo Molnar, (Mon Oct 8, 7:39 am)
Re: [PATCH] [2/6] scheduler: Refactor common code of sleep_o..., Ingo Molnar, (Mon Oct 8, 8:03 am)
[PATCH] [3/6] scheduler: Do devirtualization for sched_fair, Andi Kleen, (Sun Oct 7, 4:59 pm)
Re: [PATCH] [3/6] scheduler: Do devirtualization for sched_f..., Ingo Molnar, (Mon Oct 8, 7:42 am)
Re: [PATCH] [3/6] scheduler: Do devirtualization for sched_f..., Andi Kleen, (Mon Oct 8, 8:32 am)
Re: [PATCH] [3/6] scheduler: Do devirtualization for sched_f..., Ingo Molnar, (Mon Oct 8, 8:39 am)
Re: [PATCH] [3/6] scheduler: Do devirtualization for sched_f..., Andi Kleen, (Mon Oct 8, 10:33 am)
[PATCH] [1/6] scheduler: Remove some unnecessary gotos in sc..., Andi Kleen, (Sun Oct 7, 4:59 pm)
Re: [PATCH] [1/6] scheduler: Remove some unnecessary gotos i..., Ingo Molnar, (Tue Oct 9, 3:17 pm)
Re: [PATCH] [1/6] scheduler: Remove some unnecessary gotos i..., Andi Kleen, (Tue Oct 9, 8:55 pm)
Re: [PATCH] [1/6] scheduler: Remove some unnecessary gotos i..., Ingo Molnar, (Wed Oct 10, 7:25 am)
Re: [PATCH] [1/6] scheduler: Remove some unnecessary gotos i..., Ingo Molnar, (Wed Oct 10, 7:26 am)
Re: [PATCH] [1/6] scheduler: Remove some unnecessary gotos i..., Ingo Molnar, (Mon Oct 8, 7:36 am)