From: Nishanth Aravamudan <nacc@us.ibm.com>
To: Tony Lindgren <tony@atomide.com>
Cc: Srivatsa Vaddagiri <vatsa@in.ibm.com>,
Con Kolivas <kernel@kolivas.org>,
linux-kernel@vger.kernel.org, akpm@osdl.org,
ck list <ck@vds.kolivas.org>,
rmk+lkml@arm.linux.org.uk
Subject: Re: [PATCH 1/3] dynticks - implement no idle hz for x86
Date: Wed, 7 Sep 2005 08:53:52 -0700 [thread overview]
Message-ID: <20050907155352.GD4590@us.ibm.com> (raw)
In-Reply-To: <20050907081303.GC5804@atomide.com>
On 07.09.2005 [11:13:04 +0300], Tony Lindgren wrote:
> * Nishanth Aravamudan <nacc@us.ibm.com> [050906 23:55]:
>
> ...
>
> > Sigh, later than I had hoped, but here is what I have hashed out so far.
> > Does it seem like a step in the right direction? Rather hand-wavy, but I
> > think it's mostly correct ;)
>
> Some comments below.
Updated document below.
Thanks,
Nish
- include/linux/timer.h
with definitions in kernel/timer.c
OR better in
- include/linux/ticksource.h
with definitions in kernel/ticksource.c?
#define DYN_TICK_ENABLED (1 << 1)
#define DYN_TICK_SUITABLE (1 << 0)
#define DYN_TICK_MIN_SKIP 2
/* Abstraction of a tick source
* @state: current state
* @max_skip: current maximum number of ticks to skip
* @init: initialization routine
* @enable_dyn_tick: called via sysfs to enable interrupt skipping
* @disable_dyn_tick: called via sysfs to disable interrupt
* skipping
* @set_all_cpus_idle: last cpu to go idle calls this, which should
* disable any timesource (e.g. PIT on x86)
* @recover_time: handler for returning from skipped ticks and keeping
* time consistent
*/
struct tick_source {
unsigned int state;
unsigned long max_skip;
int (*init) (void);
void (*enable_dyn_tick) (void);
void (*disable_dyn_tick) (void);
unsigned long (*reprogram) (unsigned long); /* return number of ticks skipped */
unsigned long (*recover_time) (int, void *, struct pt_regs *); /* handler in arm */
/* following empty in UP */
void (*set_all_cpus_idle) (int);
spinlock_t lock;
};
extern void tick_source_register(struct tick_source *new_tick_source);
extern struct tick_source *current_ticksource;
#ifdef CONFIG_NO_IDLE_HZ /* which means CONFIG_DYNTICK is also on */
extern void set_tick_max_skip(unsigned long max_skip);
/* idle_reprogram_tick calls reprogram_tick calls current_ticksource->reprogram()
* do we really need the first step? */
extern void idle_reprogram_tick(void);
/* return number of ticks skipped, potentially for accounting purposes? */
extern unsigned long reprogram_tick(void);
extern struct tick_source * __init arch_select_tick_source(void);
extern void __init dyn_tick_init(void); /* calls select_tick_source(), verifies source is usable, then calls tick_source_register() */
static inline int dyn_tick_enabled(void)
{
return (current_ticksource->state & DYN_TICK_ENABLED);
}
#else /* CONFIG_NO_IDLE_HZ */
static inline void set_tick_max_skip(unsigned long max_skip)
{
}
static inline void idle_reprogram_tick(void)
{
}
static inline unsigned long reprogram_tick(void)
{
return 0;
}
static inline void dyn_tick_init(void)
{
}
static inline int dyn_tick_enabled(void)
{
return 0;
}
#endif /* CONFIG_NO_IDLE_HZ */
/* Pick up arch specific header */
#include <asm/timer.h> /* or <asm/ticksource.h>, depending */
- sched.c / sched.h
/* do we want these elsewhere? */
cpumask_t no_idle_hz_cpumask;
- each arch-specific file pair needs to provide:
arch_select_tick_source();
appropriate struct tick_source definitions, functions, etc.
- include/asm-i386/timer.h /* or ticksource.h */
with defines in arch/i386/timer.c /* or ticksource.c */
- include/asm-arm/arch-omap/timer.h /* or ticksource.h */
with definitions in arch/arm/mach-omap/timer.c /* or ticksource.c */
- include/asm-s390/timer.h /* or ticksource.h */
with definitions in arch/s390/timer.c /* or ticksource.c */
next prev parent reply other threads:[~2005-09-07 15:54 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-08-31 16:58 Updated dynamic tick patches Srivatsa Vaddagiri
2005-08-31 17:12 ` [PATCH 1/3] Updated dynamic tick patches - Fix lost tick calculation in timer_pm.c Srivatsa Vaddagiri
2005-08-31 22:36 ` Zachary Amsden
2005-08-31 22:47 ` john stultz
2005-09-02 15:43 ` [PATCH 1/3] dynticks - implement no idle hz for x86 Con Kolivas
2005-09-02 15:45 ` [PATCH 2/3] dyntick - Fix lost tick calculation in timer pm.c Con Kolivas
2005-09-02 15:46 ` [PATCH 3/3] dyntick - Recover walltime upon wakeup Con Kolivas
2005-09-02 17:25 ` [PATCH 2/3] dyntick - Fix lost tick calculation in timer pm.c Srivatsa Vaddagiri
2005-09-02 20:18 ` Thomas Schlichter
2005-09-02 21:21 ` john stultz
2005-09-02 16:56 ` [PATCH 1/3] dynticks - implement no idle hz for x86 Russell King
2005-09-02 17:12 ` Srivatsa Vaddagiri
2005-09-03 6:13 ` Con Kolivas
2005-09-03 7:58 ` Russell King
2005-09-03 8:01 ` Con Kolivas
2005-09-03 8:06 ` Russell King
2005-09-03 8:14 ` Con Kolivas
2005-09-04 20:10 ` Nishanth Aravamudan
2005-09-04 20:26 ` Russell King
2005-09-04 20:37 ` Nishanth Aravamudan
2005-09-04 21:17 ` Russell King
2005-09-05 3:08 ` Con Kolivas
2005-09-05 16:28 ` Nishanth Aravamudan
2005-09-05 6:58 ` Tony Lindgren
2005-09-05 16:30 ` Nishanth Aravamudan
2005-09-04 20:41 ` Nishanth Aravamudan
2005-09-05 5:32 ` Srivatsa Vaddagiri
2005-09-05 5:48 ` Nishanth Aravamudan
2005-09-05 6:32 ` Srivatsa Vaddagiri
2005-09-05 6:44 ` Nishanth Aravamudan
2005-09-06 20:51 ` Nishanth Aravamudan
2005-09-07 8:13 ` Tony Lindgren
2005-09-07 15:00 ` Nishanth Aravamudan
2005-09-07 15:53 ` Nishanth Aravamudan [this message]
2005-09-07 17:07 ` Srivatsa Vaddagiri
2005-09-07 17:23 ` Nishanth Aravamudan
2005-09-07 18:14 ` Srivatsa Vaddagiri
2005-09-07 18:22 ` Nishanth Aravamudan
2005-09-07 16:14 ` Bill Davidsen
2005-09-07 16:42 ` Nish Aravamudan
2005-09-07 17:17 ` Srivatsa Vaddagiri
2005-09-07 17:27 ` Nish Aravamudan
2005-09-07 18:18 ` Srivatsa Vaddagiri
2005-09-07 18:33 ` Nish Aravamudan
2005-09-09 16:27 ` Bill Davidsen
2005-09-05 7:37 ` Russell King
2005-09-05 7:49 ` Srivatsa Vaddagiri
2005-09-05 8:00 ` Russell King
2005-09-05 16:33 ` Nishanth Aravamudan
2005-09-05 7:00 ` Srivatsa Vaddagiri
2005-09-05 7:27 ` Tony Lindgren
2005-09-05 17:02 ` Nishanth Aravamudan
2005-09-07 7:37 ` Tony Lindgren
2005-09-07 15:05 ` Nishanth Aravamudan
2005-09-08 10:00 ` Tony Lindgren
2005-09-08 21:22 ` Nishanth Aravamudan
2005-09-08 22:08 ` Nishanth Aravamudan
2005-09-09 22:30 ` Nishanth Aravamudan
2005-09-20 11:06 ` Srivatsa Vaddagiri
2005-09-20 14:58 ` Nishanth Aravamudan
2005-09-22 13:38 ` Martin Schwidefsky
2005-09-22 14:52 ` Nishanth Aravamudan
2005-09-22 18:32 ` Srivatsa Vaddagiri
2005-09-26 15:08 ` Srivatsa Vaddagiri
2005-09-23 6:55 ` Srivatsa Vaddagiri
2005-09-05 7:44 ` Russell King
2005-09-05 8:19 ` Srivatsa Vaddagiri
2005-09-05 8:32 ` Russell King
2005-09-05 9:24 ` Srivatsa Vaddagiri
2005-09-05 17:06 ` Nishanth Aravamudan
2005-09-05 17:04 ` Nishanth Aravamudan
2005-09-05 17:27 ` Srivatsa Vaddagiri
2005-09-05 18:06 ` Nishanth Aravamudan
2005-09-05 13:19 ` Srivatsa Vaddagiri
2005-09-05 16:57 ` Nishanth Aravamudan
2005-09-05 17:25 ` Srivatsa Vaddagiri
2005-09-05 18:11 ` Nishanth Aravamudan
2005-09-03 4:05 ` [PATCH 1/3] Updated dynamic tick patches - Fix lost tick calculation in timer_pm.c Lee Revell
2005-09-03 4:18 ` Peter Williams
2005-09-03 4:34 ` Lee Revell
2005-09-03 4:48 ` Peter Williams
2005-09-03 5:15 ` Parag Warudkar
2005-09-03 5:30 ` Lee Revell
2005-09-03 5:20 ` Srivatsa Vaddagiri
2005-09-06 10:32 ` Pavel Machek
2005-09-06 10:46 ` Srivatsa Vaddagiri
2005-09-06 18:04 ` john stultz
2005-08-31 17:26 ` [PATCH 2/3] Updated dynamic tick patches - Cleanup Srivatsa Vaddagiri
2005-08-31 17:27 ` [PATCH 3/3] Updated dynamic tick patches - Recover walltime upon wakeup Srivatsa Vaddagiri
2005-09-01 5:23 ` Updated dynamic tick patches Con Kolivas
2005-09-01 13:07 ` Tony Lindgren
2005-09-01 13:19 ` David Weinehall
2005-09-01 13:46 ` Tony Lindgren
2005-09-01 14:11 ` Srivatsa Vaddagiri
2005-09-02 17:34 ` Srivatsa Vaddagiri
2005-09-03 10:16 ` Tony Lindgren
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20050907155352.GD4590@us.ibm.com \
--to=nacc@us.ibm.com \
--cc=akpm@osdl.org \
--cc=ck@vds.kolivas.org \
--cc=kernel@kolivas.org \
--cc=linux-kernel@vger.kernel.org \
--cc=rmk+lkml@arm.linux.org.uk \
--cc=tony@atomide.com \
--cc=vatsa@in.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.