From: Artem Bityutskiy <[email protected]>
This patch makes the cache_cleaner workqueue deferrable, to prevent
unnecessary system wake-ups, which is very important for embedded
battery-powered devices.
do_cache_clean() is called every 30 seconds at the moment, and often
makes the system wake up from its power-save sleep state. With this
change, when the workqueue uses a deferrable timer, the
do_cache_clean() invocation will be delayed and combined with the
closest "real" wake-up. This improves the power consumption situation.
Note, I tried to create a DECLARE_DELAYED_WORK_DEFERRABLE() helper
macro, similar to DECLARE_DELAYED_WORK(), but failed because of the
way the timer wheel core stores the deferrable flag (it is the
LSBit in the time->base pointer). My attempt to define a static
variable with this bit set ended up with the "initializer element is
not constant" error.
Thus, I have to use run-time initialization, so I created a new
cache_initialize() function which is called once when sunrpc is
being initialized.
Signed-off-by: Artem Bityutskiy <[email protected]>
---
include/linux/sunrpc/cache.h | 1 +
net/sunrpc/cache.c | 7 ++++++-
net/sunrpc/sunrpc_syms.c | 1 +
3 files changed, 8 insertions(+), 1 deletions(-)
diff --git a/include/linux/sunrpc/cache.h b/include/linux/sunrpc/cache.h
index 6f52b4d..7bf3e84 100644
--- a/include/linux/sunrpc/cache.h
+++ b/include/linux/sunrpc/cache.h
@@ -192,6 +192,7 @@ extern int cache_check(struct cache_detail *detail,
extern void cache_flush(void);
extern void cache_purge(struct cache_detail *detail);
#define NEVER (0x7FFFFFFF)
+extern void __init cache_initialize(void);
extern int cache_register(struct cache_detail *cd);
extern void cache_unregister(struct cache_detail *cd);
diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c
index 58de76c..939d048 100644
--- a/net/sunrpc/cache.c
+++ b/net/sunrpc/cache.c
@@ -320,7 +320,7 @@ static struct cache_detail *current_detail;
static int current_index;
static void do_cache_clean(struct work_struct *work);
-static DECLARE_DELAYED_WORK(cache_cleaner, do_cache_clean);
+static struct delayed_work cache_cleaner;
static void sunrpc_init_cache_detail(struct cache_detail *cd)
{
@@ -1504,6 +1504,11 @@ static int create_cache_proc_entries(struct cache_detail *cd)
}
#endif
+void __init cache_initialize(void)
+{
+ INIT_DELAYED_WORK_DEFERRABLE(&cache_cleaner, do_cache_clean);
+}
+
int cache_register(struct cache_detail *cd)
{
int ret;
diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index f438347..c52b184 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -43,6 +43,7 @@ init_sunrpc(void)
#ifdef CONFIG_PROC_FS
rpc_proc_init();
#endif
+ cache_initialize();
cache_register(&ip_map_cache);
cache_register(&unix_gid_cache);
svc_init_xprt_sock(); /* svc sock transport */
--
1.7.0.1
From: J. Bruce Fields <[email protected]>
Steal some text from 6e453a67510 "Add support for deferrable timers".
A reader shouldn't have to dig through the git logs for the basic
description of a deferrable timer.
Signed-off-by: J. Bruce Fields <[email protected]>
---
kernel/timer.c | 9 +++++++--
1 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/kernel/timer.c b/kernel/timer.c
index ee305c8..947bcf6 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -90,8 +90,13 @@ static DEFINE_PER_CPU(struct tvec_base *, tvec_bases) = &boot_tvec_bases;
/*
* Note that all tvec_bases are 2 byte aligned and lower bit of
- * base in timer_list is guaranteed to be zero. Use the LSB for
- * the new flag to indicate whether the timer is deferrable
+ * base in timer_list is guaranteed to be zero. Use the LSB to
+ * indicate whether the timer is deferrable.
+ *
+ * A deferrable timer will work normally when the system is busy, but
+ * will not cause a CPU to come out of idle just to service it; instead,
+ * the timer will be serviced when the CPU eventually wakes up with a
+ * subsequent non-deferrable timer.
*/
#define TBASE_DEFERRABLE_FLAG (0x1)
--
1.7.0.4
On Thu, 2010-07-01 at 18:05 +0300, Artem Bityutskiy wrote:
> From: Artem Bityutskiy <[email protected]>
>
> This patch makes the cache_cleaner workqueue deferrable, to prevent
> unnecessary system wake-ups, which is very important for embedded
> battery-powered devices.
>
> do_cache_clean() is called every 30 seconds at the moment, and often
> makes the system wake up from its power-save sleep state. With this
> change, when the workqueue uses a deferrable timer, the
> do_cache_clean() invocation will be delayed and combined with the
> closest "real" wake-up. This improves the power consumption situation.
>
> Note, I tried to create a DECLARE_DELAYED_WORK_DEFERRABLE() helper
> macro, similar to DECLARE_DELAYED_WORK(), but failed because of the
> way the timer wheel core stores the deferrable flag (it is the
> LSBit in the time->base pointer). My attempt to define a static
> variable with this bit set ended up with the "initializer element is
> not constant" error.
>
> Thus, I have to use run-time initialization, so I created a new
> cache_initialize() function which is called once when sunrpc is
> being initialized.
>
> Signed-off-by: Artem Bityutskiy <[email protected]>
Bruce,
I did some git-logging, and found out that sunrpc stuff goes from you,
from Trond, and from David, and MAINTAINERS also mentions Neil. Whom
should I ping WRT to merging this patch?
--
Best Regards,
Artem Bityutskiy (Артём Битюцкий)
On Tue, 2010-07-06 at 13:19 -0400, J. Bruce Fields wrote:
> From: J. Bruce Fields <[email protected]>
>=20
> Steal some text from 6e453a67510 "Add support for deferrable timers".
> A reader shouldn't have to dig through the git logs for the basic
> description of a deferrable timer.
>=20
> Signed-off-by: J. Bruce Fields <[email protected]>
Yes, this is a good idea. Someone cares to pick this?
--=20
Best Regards,
Artem Bityutskiy (=D0=90=D1=80=D1=82=D1=91=D0=BC =D0=91=D0=B8=D1=82=D1=8E=
=D1=86=D0=BA=D0=B8=D0=B9)
On Thu, Jul 01, 2010 at 06:05:56PM +0300, Artem Bityutskiy wrote:
> From: Artem Bityutskiy <[email protected]>
>
> This patch makes the cache_cleaner workqueue deferrable, to prevent
> unnecessary system wake-ups, which is very important for embedded
> battery-powered devices.
>
> do_cache_clean() is called every 30 seconds at the moment, and often
> makes the system wake up from its power-save sleep state. With this
> change, when the workqueue uses a deferrable timer, the
> do_cache_clean() invocation will be delayed and combined with the
> closest "real" wake-up. This improves the power consumption situation.
>
> Note, I tried to create a DECLARE_DELAYED_WORK_DEFERRABLE() helper
> macro, similar to DECLARE_DELAYED_WORK(), but failed because of the
> way the timer wheel core stores the deferrable flag (it is the
> LSBit in the time->base pointer). My attempt to define a static
> variable with this bit set ended up with the "initializer element is
> not constant" error.
>
> Thus, I have to use run-time initialization, so I created a new
> cache_initialize() function which is called once when sunrpc is
> being initialized.
Thanks, applying for 2.6.36.
--b.
>
> Signed-off-by: Artem Bityutskiy <[email protected]>
> ---
> include/linux/sunrpc/cache.h | 1 +
> net/sunrpc/cache.c | 7 ++++++-
> net/sunrpc/sunrpc_syms.c | 1 +
> 3 files changed, 8 insertions(+), 1 deletions(-)
>
> diff --git a/include/linux/sunrpc/cache.h b/include/linux/sunrpc/cache.h
> index 6f52b4d..7bf3e84 100644
> --- a/include/linux/sunrpc/cache.h
> +++ b/include/linux/sunrpc/cache.h
> @@ -192,6 +192,7 @@ extern int cache_check(struct cache_detail *detail,
> extern void cache_flush(void);
> extern void cache_purge(struct cache_detail *detail);
> #define NEVER (0x7FFFFFFF)
> +extern void __init cache_initialize(void);
> extern int cache_register(struct cache_detail *cd);
> extern void cache_unregister(struct cache_detail *cd);
>
> diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c
> index 58de76c..939d048 100644
> --- a/net/sunrpc/cache.c
> +++ b/net/sunrpc/cache.c
> @@ -320,7 +320,7 @@ static struct cache_detail *current_detail;
> static int current_index;
>
> static void do_cache_clean(struct work_struct *work);
> -static DECLARE_DELAYED_WORK(cache_cleaner, do_cache_clean);
> +static struct delayed_work cache_cleaner;
>
> static void sunrpc_init_cache_detail(struct cache_detail *cd)
> {
> @@ -1504,6 +1504,11 @@ static int create_cache_proc_entries(struct cache_detail *cd)
> }
> #endif
>
> +void __init cache_initialize(void)
> +{
> + INIT_DELAYED_WORK_DEFERRABLE(&cache_cleaner, do_cache_clean);
> +}
> +
> int cache_register(struct cache_detail *cd)
> {
> int ret;
> diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
> index f438347..c52b184 100644
> --- a/net/sunrpc/sunrpc_syms.c
> +++ b/net/sunrpc/sunrpc_syms.c
> @@ -43,6 +43,7 @@ init_sunrpc(void)
> #ifdef CONFIG_PROC_FS
> rpc_proc_init();
> #endif
> + cache_initialize();
> cache_register(&ip_map_cache);
> cache_register(&unix_gid_cache);
> svc_init_xprt_sock(); /* svc sock transport */
> --
> 1.7.0.1
>
On Mon, Jul 05, 2010 at 03:31:30PM +0300, Artem Bityutskiy wrote:
> On Thu, 2010-07-01 at 18:05 +0300, Artem Bityutskiy wrote:
> > From: Artem Bityutskiy <[email protected]>
> >
> > This patch makes the cache_cleaner workqueue deferrable, to prevent
> > unnecessary system wake-ups, which is very important for embedded
> > battery-powered devices.
> >
> > do_cache_clean() is called every 30 seconds at the moment, and often
> > makes the system wake up from its power-save sleep state. With this
> > change, when the workqueue uses a deferrable timer, the
> > do_cache_clean() invocation will be delayed and combined with the
> > closest "real" wake-up. This improves the power consumption situation.
> >
> > Note, I tried to create a DECLARE_DELAYED_WORK_DEFERRABLE() helper
> > macro, similar to DECLARE_DELAYED_WORK(), but failed because of the
> > way the timer wheel core stores the deferrable flag (it is the
> > LSBit in the time->base pointer). My attempt to define a static
> > variable with this bit set ended up with the "initializer element is
> > not constant" error.
> >
> > Thus, I have to use run-time initialization, so I created a new
> > cache_initialize() function which is called once when sunrpc is
> > being initialized.
> >
> > Signed-off-by: Artem Bityutskiy <[email protected]>
>
> Bruce,
>
> I did some git-logging, and found out that sunrpc stuff goes from you,
> from Trond, and from David, and MAINTAINERS also mentions Neil. Whom
> should I ping WRT to merging this patch?
I got it. (I'm just getting back from vacation, and very backlogged.)
--b.