cyclic: Add ratelimiting schedule() via CYCLIC_RATELIMIT_US
PATCH 1/1 cyclic: Add ratelimiting schedule() via CYCLIC_RATELIMIT_US
This patch adds the new Kconfig option CONFIG_CYCLIC_RATELIMIT_US which defines the min allowed time after with a new call into the cyclic infrastructure is allowed. This results in a ratelimiting of the all functions hooked into the cyclic interface. As it's been noticed that on some platforms, that high frequent calls to schedule() (replaced from WATCHDOG_RESET) may lead to a performance degration.
When a high frequent calling into schedule() is detected, a warning is logged only once to help indentify this frantic caller.
--- common/Kconfig | 11 +++++++++++ common/cyclic.c | 27 +++++++++++++++++++++++++-- include/asm-generic/global_data.h | 1 + 3 files changed, 37 insertions(+), 2 deletions(-)
common/Kconfig +11 -0
@@ -601,6 +601,17 @@ config CYCLIC_MAX_CPU_TIME_US takes longer than this duration this function will get unregistered automatically. +config CYCLIC_RATELIMIT_US + int "Sets the min time/delay in us after a new call into schedule is done" + default 100 + help + The min allowed time after with a new call into the cyclic + infrastructure is allowed. This results in a ratelimiting of the + all functions hooked into the cyclic interface. + As it's been noticed that on some platforms, that high frequent + calls to schedule() (replaced from WATCHDOG_RESET) may lead to a + performance degration. + endif # CYCLIC config EVENT
common/cyclic.c +25 -2
@@ -12,6 +12,7 @@#include <log.h> #include <malloc.h> #include <time.h> +#include <linux/bug.h> #include <linux/errno.h> #include <linux/list.h> #include <asm/global_data.h>
@@ -109,8 +110,30 @@ void schedule(void)* schedule() might get called very early before the cyclic IF is * ready. Make sure to only call cyclic_run() when it's initalized. */ -if (gd) -cyclic_run (); +if (gd) { +uint64_t now; + +/* +* Add some ratelimiting to not call into the cyclic IF too +* ofter. As it's been noticed that on some platforms, that +* high frequent calls to schedule() (replaced from +* WATCHDOG_RESET) may lead to a performance degration. +*/ +/* Do not reset the watchdog too often */ + now =timer_get_us (); +if (time_after_eq64 (now, gd->cyclic_next_call)) { +cyclic_run (); + gd->cyclic_next_call = now +CONFIG_CYCLIC_RATELIMIT_US ; + }else { +/* +* Throw a warning (only once) to help identifying +* frantic callers +*/ +WARN_ONCE (1 ,"schedule() called very often, now = %lld us next call at %lld us, dt = %lld us\n" , + now, gd->cyclic_next_call, + gd->cyclic_next_call - now); + } + } }int cyclic_unregister_all (void )
include/asm-generic/global_data.h +1 -0
@@ -484,6 +484,7 @@ struct global_data {* @cyclic_list: list of registered cyclic functions */ struct hlist_head cyclic_list; +uint64_t cyclic_next_call;#endif /** * @dmtag_list: List of DM tags
Signature · 1 line
2.41.0
Hi Stefan,
Quoted 25 lines · Stefan Roese
This patch adds the new
KconfigoptionCONFIG_CYCLIC_RATELIMIT_USwhich defines the min allowed time after with a new call into the cyclic infrastructure is allowed. This results in a ratelimiting of the all functions hooked into the cyclic interface. As it's been noticed that on some platforms, that high frequent calls toschedule()(replaced fromWATCHDOG_RESET) may lead to a performance degration.When a high frequent calling into
schedule()is detected, a warning is logged only once to help indentify this frantic caller.Signed-off-by Stefan Roese <[email protected]>Cc Simon Glass <[email protected]>Cc Christophe Leroy <[email protected]>--- common/Kconfig | 11 +++++++++++ common/cyclic.c | 27 +++++++++++++++++++++++++-- include/asm-generic/global_data.h | 1 + 3 files changed, 37 insertions(+), 2 deletions(-)common/Kconfig +0
-0
common/Kconfig config CYCLIC_MAX_CPU_TIME_US · line 606
Quoted 6 lines
common/Kconfig +3
-0@@ -601,6 +601,17 @@ config CYCLIC_MAX_CPU_TIME_US takes longer than this duration this function will get unregistered automatically. +config CYCLIC_RATELIMIT_US + int "Sets the min time/delay in us after a new call into schedule is done" + default 100
1000 ?
common/Kconfig config CYCLIC_MAX_CPU_TIME_US · line 609
Quoted 3 lines
common/Kconfig +3
-0+ help + The min allowed time after with a new call into the cyclic + infrastructure is allowed. This results in a ratelimiting of the
rate limiting
common/Kconfig config CYCLIC_MAX_CPU_TIME_US · line 613
Quoted 4 lines
common/Kconfig +4
-0+ all functions hooked into the cyclic interface. + As it's been noticed that on some platforms, that high frequent + calls to schedule() (replaced from WATCHDOG_RESET) may lead to a + performance degration.
degradation
also below
Quoted 17 lines
common/Kconfig +1
-0+ endif # CYCLIC config EVENTcommon/cyclic.c +1
-0@@ -12,6 +12,7 @@#include <log.h> #include <malloc.h> #include <time.h> +#include <linux/bug.h> #include <linux/errno.h> #include <linux/list.h> #include <asm/global_data.h>
common/cyclic.c void schedule(void) · line 114
Quoted 7 lines
common/cyclic.c +2
-2@@ -109,8 +110,30 @@ void schedule(void)* schedule() might get called very early before the cyclic IF is * ready. Make sure to only call cyclic_run() when it's initalized. */ -if (gd) -cyclic_run (); +if (gd) { +uint64_t now;
Should be u64 in here I think.
common/cyclic.c void schedule(void) · line 122
Quoted 8 lines
common/cyclic.c +8
-0+ +/* +* Add some ratelimiting to not call into the cyclic IF too +* ofter. As it's been noticed that on some platforms, that +* high frequent calls to schedule() (replaced from +* WATCHDOG_RESET) may lead to a performance degration. +*/ +/* Do not reset the watchdog too often */
join with above comment
Quoted 30 lines
common/cyclic.c +14
-0+ now =timer_get_us (); +if (time_after_eq64 (now, gd->cyclic_next_call)) { +cyclic_run (); + gd->cyclic_next_call = now +CONFIG_CYCLIC_RATELIMIT_US ; + }else { +/* +* Throw a warning (only once) to help identifying +* frantic callers +*/ +WARN_ONCE (1 ,"schedule() called very often, now = %lld us next call at %lld us, dt = %lld us\n" , + now, gd->cyclic_next_call, + gd->cyclic_next_call - now); + } + } }int cyclic_unregister_all (void )include/asm-generic/global_data.h +1
-1@@ -484,6 +484,7 @@ struct global_data {* @cyclic_list: list of registered cyclic functions */ struct hlist_head cyclic_list; +uint64_t cyclic_next_call;#endif /** * @dmtag_list: List of DM tags --
Quoted 2 lines
2.41.0
Regards,
Simon
Hi Simon,
Quoted 31 lines · Simon Glass
Hi Stefan,
Quoted 28 lines · Stefan Roese
This patch adds the new
KconfigoptionCONFIG_CYCLIC_RATELIMIT_USwhich defines the min allowed time after with a new call into the cyclic infrastructure is allowed. This results in a ratelimiting of the all functions hooked into the cyclic interface. As it's been noticed that on some platforms, that high frequent calls toschedule()(replaced fromWATCHDOG_RESET) may lead to a performance degration.When a high frequent calling into
schedule()is detected, a warning is logged only once to help indentify this frantic caller.Signed-off-by Stefan Roese <[email protected]>Cc Simon Glass <[email protected]>Cc Christophe Leroy <[email protected]>--- common/Kconfig | 11 +++++++++++ common/cyclic.c | 27 +++++++++++++++++++++++++-- include/asm-generic/global_data.h | 1 + 3 files changed, 37 insertions(+), 2 deletions(-)common/Kconfig +0
-0@@ -601,6 +601,17 @@ config CYCLIC_MAX_CPU_TIME_US takes longer than this duration this function will get unregistered automatically.
Quoted 5 lines
common/Kconfig config CYCLIC_MAX_CPU_TIME_US · line 606
Quoted 3 lines
common/Kconfig +3
-0+config CYCLIC_RATELIMIT_US + int "Sets the min time/delay in us after a new call into schedule is done" + default 1001000 ?
1000 was my first pick as well. Leading to this WARN_ONE warning:
schedule() called very often, now = 1085562us next call at 1086380us
This is why I switched to a less strict rate limitation as default. Other platforms may choose to override this value if this matches their use cases.
Quoted 5 lines
common/Kconfig config CYCLIC_MAX_CPU_TIME_US · line 609
Quoted 3 lines
common/Kconfig +3
-0+ help + The min allowed time after with a new call into the cyclic + infrastructure is allowed. This results in a ratelimiting of therate limiting
Thanks.
Quoted 8 lines
common/Kconfig config CYCLIC_MAX_CPU_TIME_US · line 613
Quoted 4 lines
common/Kconfig +4
-0+ all functions hooked into the cyclic interface. + As it's been noticed that on some platforms, that high frequent + calls to schedule() (replaced from WATCHDOG_RESET) may lead to a + performance degration.degradation
also below
Thanks, will update in v2.
Quoted 18 lines
Quoted 18 lines
common/Kconfig +1
-0+ endif # CYCLIC config EVENTcommon/cyclic.c +1
-0@@ -12,6 +12,7 @@#include <log.h> #include <malloc.h> #include <time.h> +#include <linux/bug.h> #include <linux/errno.h> #include <linux/list.h> #include <asm/global_data.h> @@ -109,8 +110,30 @@ void schedule(void)* schedule() might get called very early before the cyclic IF is
Quoted 8 lines
common/cyclic.c void schedule(void) · line 114
Quoted 6 lines
common/cyclic.c +2
-2* ready. Make sure to only call cyclic_run() when it's initalized. */ -if (gd) -cyclic_run (); +if (gd) { +uint64_t now;Should be u64 in here I think.
This file is already using uint64_t instead of u64. This is why I used this type here as well.
Quoted 10 lines
common/cyclic.c void schedule(void) · line 122
Quoted 8 lines
common/cyclic.c +8
-0+ +/* +* Add some ratelimiting to not call into the cyclic IF too +* ofter. As it's been noticed that on some platforms, that +* high frequent calls to schedule() (replaced from +* WATCHDOG_RESET) may lead to a performance degration. +*/ +/* Do not reset the watchdog too often */ join with above comment
Ok, will do.
Thanks,
Stefan
Quoted 27 lines
Quoted 27 lines
common/cyclic.c +14
-0+ now =timer_get_us (); +if (time_after_eq64 (now, gd->cyclic_next_call)) { +cyclic_run (); + gd->cyclic_next_call = now +CONFIG_CYCLIC_RATELIMIT_US ; + }else { +/* +* Throw a warning (only once) to help identifying +* frantic callers +*/ +WARN_ONCE (1 ,"schedule() called very often, now = %lld us next call at %lld us, dt = %lld us\n" , + now, gd->cyclic_next_call, + gd->cyclic_next_call - now); + } + } }int cyclic_unregister_all (void )include/asm-generic/global_data.h +1
-0@@ -484,6 +484,7 @@ struct global_data {* @cyclic_list: list of registered cyclic functions */ struct hlist_head cyclic_list; +uint64_t cyclic_next_call;#endif
Quoted 8 lines
Quoted 5 lines
include/asm-generic/global_data.h +0
-1/** * @dmtag_list: List of DM tags -- 2.41.0
Regards,
Simon
Viele Grüße,
Stefan Roese
Signature · 3 lines
DENX Software Engineering GmbH, Managing Director: Erika Unter HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-51 Fax: (+49)-8142-66989-80 Email: [email protected]