This patch introduces CONFIG_CPUSETS_V1 and guard cpuset-v1 code under
CONFIG_CPUSETS_V1. The default value of CONFIG_CPUSETS_V1 is N, so that
user who adopted v2 don't have 'pay' for cpuset v1.
Besides, to make code succinct, rename '__cpuset_memory_pressure_bump()'
to 'cpuset_memory_pressure_bump()', and expose it to the world, which
takes place of the old mocro 'cpuset_memory_pressure_bump'.
Signed-off-by: Chen Ridong <chenridong@huawei.com>
---
include/linux/cpuset.h | 8 +-------
init/Kconfig | 13 +++++++++++++
kernel/cgroup/Makefile | 3 ++-
kernel/cgroup/cpuset-internal.h | 15 +++++++++++++++
kernel/cgroup/cpuset-v1.c | 10 ++++++----
kernel/cgroup/cpuset.c | 2 ++
6 files changed, 39 insertions(+), 12 deletions(-)
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index 2a6981eeebf8..f91f1f61f482 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -99,13 +99,7 @@ static inline bool cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
extern int cpuset_mems_allowed_intersects(const struct task_struct *tsk1,
const struct task_struct *tsk2);
-#define cpuset_memory_pressure_bump() \
- do { \
- if (cpuset_memory_pressure_enabled) \
- __cpuset_memory_pressure_bump(); \
- } while (0)
-extern int cpuset_memory_pressure_enabled;
-extern void __cpuset_memory_pressure_bump(void);
+extern void cpuset_memory_pressure_bump(void);
extern void cpuset_task_status_allowed(struct seq_file *m,
struct task_struct *task);
diff --git a/init/Kconfig b/init/Kconfig
index b9ec6b6ef3a5..369941de9c2c 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -1144,6 +1144,19 @@ config CPUSETS
Say N if unsure.
+config CPUSETS_V1
+ bool "Legacy cgroup v1 cpusets controller"
+ depends on CPUSETS
+ default n
+ help
+ Legacy cgroup v1 cpusets controller which has been deprecated by
+ cgroup v2 implementation. The v1 is there for legacy applications
+ which haven't migrated to the new cgroup v2 interface yet. If you
+ do not have any such application then you are completely fine leaving
+ this option disabled.
+
+ Say N if unsure.
+
config PROC_PID_CPUSET
bool "Include legacy /proc/<pid>/cpuset file"
depends on CPUSETS
diff --git a/kernel/cgroup/Makefile b/kernel/cgroup/Makefile
index 005ac4c675cb..a5c9359d516f 100644
--- a/kernel/cgroup/Makefile
+++ b/kernel/cgroup/Makefile
@@ -4,6 +4,7 @@ obj-y := cgroup.o rstat.o namespace.o cgroup-v1.o freezer.o
obj-$(CONFIG_CGROUP_FREEZER) += legacy_freezer.o
obj-$(CONFIG_CGROUP_PIDS) += pids.o
obj-$(CONFIG_CGROUP_RDMA) += rdma.o
-obj-$(CONFIG_CPUSETS) += cpuset.o cpuset-v1.o
+obj-$(CONFIG_CPUSETS) += cpuset.o
+obj-$(CONFIG_CPUSETS_V1) += cpuset-v1.o
obj-$(CONFIG_CGROUP_MISC) += misc.o
obj-$(CONFIG_CGROUP_DEBUG) += debug.o
diff --git a/kernel/cgroup/cpuset-internal.h b/kernel/cgroup/cpuset-internal.h
index 36e9686c7bba..9ccc4f54eefd 100644
--- a/kernel/cgroup/cpuset-internal.h
+++ b/kernel/cgroup/cpuset-internal.h
@@ -279,6 +279,7 @@ int cpuset_common_seq_show(struct seq_file *sf, void *v);
/*
* cpuset-v1.c
*/
+#ifdef CONFIG_CPUSETS_V1
extern struct cftype legacy_files[];
void fmeter_init(struct fmeter *fmp);
void cpuset_update_task_spread_flags(struct cpuset *cs,
@@ -288,5 +289,19 @@ void hotplug_update_tasks_legacy(struct cpuset *cs,
struct cpumask *new_cpus, nodemask_t *new_mems,
bool cpus_updated, bool mems_updated);
int validate_change_legacy(struct cpuset *cur, struct cpuset *trial);
+void cpuset_memory_pressure_bump(void);
+
+#else
+static inline void fmeter_init(struct fmeter *fmp) {}
+static inline void cpuset_update_task_spread_flags(struct cpuset *cs,
+ struct task_struct *tsk) {}
+static inline void update_tasks_flags(struct cpuset *cs) {}
+static inline void hotplug_update_tasks_legacy(struct cpuset *cs,
+ struct cpumask *new_cpus, nodemask_t *new_mems,
+ bool cpus_updated, bool mems_updated) {}
+static inline int validate_change_legacy(struct cpuset *cur,
+ struct cpuset *trial) { return 0; }
+void cpuset_memory_pressure_bump(void) {}
+#endif /* CONFIG_CPUSETS_V1 */
#endif /* __CPUSET_INTERNAL_H */
diff --git a/kernel/cgroup/cpuset-v1.c b/kernel/cgroup/cpuset-v1.c
index ffb8711cc8fa..efa5ab0167b6 100644
--- a/kernel/cgroup/cpuset-v1.c
+++ b/kernel/cgroup/cpuset-v1.c
@@ -137,11 +137,13 @@ int cpuset_memory_pressure_enabled __read_mostly;
* (direct) page reclaim by any task attached to the cpuset.
*/
-void __cpuset_memory_pressure_bump(void)
+void cpuset_memory_pressure_bump(void)
{
- rcu_read_lock();
- fmeter_markevent(&task_cs(current)->fmeter);
- rcu_read_unlock();
+ if (cpuset_memory_pressure_enabled) {
+ rcu_read_lock();
+ fmeter_markevent(&task_cs(current)->fmeter);
+ rcu_read_unlock();
+ }
}
static int update_relax_domain_level(struct cpuset *cs, s64 val)
diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index 903e7bab399a..658a9e00671b 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -3617,7 +3617,9 @@ struct cgroup_subsys cpuset_cgrp_subsys = {
.can_fork = cpuset_can_fork,
.cancel_fork = cpuset_cancel_fork,
.fork = cpuset_fork,
+#ifdef CONFIG_CPUSETS_V1
.legacy_cftypes = legacy_files,
+#endif
.dfl_cftypes = dfl_files,
.early_init = true,
.threaded = true,
--
2.34.1On 8/23/24 06:01, Chen Ridong wrote:
> This patch introduces CONFIG_CPUSETS_V1 and guard cpuset-v1 code under
> CONFIG_CPUSETS_V1. The default value of CONFIG_CPUSETS_V1 is N, so that
> user who adopted v2 don't have 'pay' for cpuset v1.
> Besides, to make code succinct, rename '__cpuset_memory_pressure_bump()'
> to 'cpuset_memory_pressure_bump()', and expose it to the world, which
> takes place of the old mocro 'cpuset_memory_pressure_bump'.
>
> Signed-off-by: Chen Ridong <chenridong@huawei.com>
> ---
> include/linux/cpuset.h | 8 +-------
> init/Kconfig | 13 +++++++++++++
> kernel/cgroup/Makefile | 3 ++-
> kernel/cgroup/cpuset-internal.h | 15 +++++++++++++++
> kernel/cgroup/cpuset-v1.c | 10 ++++++----
> kernel/cgroup/cpuset.c | 2 ++
> 6 files changed, 39 insertions(+), 12 deletions(-)
>
> diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
> index 2a6981eeebf8..f91f1f61f482 100644
> --- a/include/linux/cpuset.h
> +++ b/include/linux/cpuset.h
> @@ -99,13 +99,7 @@ static inline bool cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
> extern int cpuset_mems_allowed_intersects(const struct task_struct *tsk1,
> const struct task_struct *tsk2);
>
> -#define cpuset_memory_pressure_bump() \
> - do { \
> - if (cpuset_memory_pressure_enabled) \
> - __cpuset_memory_pressure_bump(); \
> - } while (0)
> -extern int cpuset_memory_pressure_enabled;
> -extern void __cpuset_memory_pressure_bump(void);
> +extern void cpuset_memory_pressure_bump(void);
>
> extern void cpuset_task_status_allowed(struct seq_file *m,
> struct task_struct *task);
As you are introducing a new CONFIG_CPUSET_V1 kconfig option, you can
use it to eliminate a useless function call if cgroup v1 isn't enabled.
Not just this function, all the v1 specific functions should have a
!CONFIG_CPUSET_V1 version that can be optimized out.
Cheers,
Longman
On 2024/8/26 1:25, Waiman Long wrote:
> On 8/23/24 06:01, Chen Ridong wrote:
>> This patch introduces CONFIG_CPUSETS_V1 and guard cpuset-v1 code under
>> CONFIG_CPUSETS_V1. The default value of CONFIG_CPUSETS_V1 is N, so that
>> user who adopted v2 don't have 'pay' for cpuset v1.
>> Besides, to make code succinct, rename '__cpuset_memory_pressure_bump()'
>> to 'cpuset_memory_pressure_bump()', and expose it to the world, which
>> takes place of the old mocro 'cpuset_memory_pressure_bump'.
>>
>> Signed-off-by: Chen Ridong <chenridong@huawei.com>
>> ---
>> include/linux/cpuset.h | 8 +-------
>> init/Kconfig | 13 +++++++++++++
>> kernel/cgroup/Makefile | 3 ++-
>> kernel/cgroup/cpuset-internal.h | 15 +++++++++++++++
>> kernel/cgroup/cpuset-v1.c | 10 ++++++----
>> kernel/cgroup/cpuset.c | 2 ++
>> 6 files changed, 39 insertions(+), 12 deletions(-)
>>
>> diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
>> index 2a6981eeebf8..f91f1f61f482 100644
>> --- a/include/linux/cpuset.h
>> +++ b/include/linux/cpuset.h
>> @@ -99,13 +99,7 @@ static inline bool cpuset_zone_allowed(struct zone
>> *z, gfp_t gfp_mask)
>> extern int cpuset_mems_allowed_intersects(const struct task_struct
>> *tsk1,
>> const struct task_struct *tsk2);
>> -#define cpuset_memory_pressure_bump() \
>> - do { \
>> - if (cpuset_memory_pressure_enabled) \
>> - __cpuset_memory_pressure_bump(); \
>> - } while (0)
>> -extern int cpuset_memory_pressure_enabled;
>> -extern void __cpuset_memory_pressure_bump(void);
>> +extern void cpuset_memory_pressure_bump(void);
>> extern void cpuset_task_status_allowed(struct seq_file *m,
>> struct task_struct *task);
>
> As you are introducing a new CONFIG_CPUSET_V1 kconfig option, you can
> use it to eliminate a useless function call if cgroup v1 isn't enabled.
> Not just this function, all the v1 specific functions should have a
> !CONFIG_CPUSET_V1 version that can be optimized out.
>
> Cheers,
> Longman
>
>
Hi, Longman, currently, we have only added one place to manage all the
functions whose are empty if !CONFIG_CPUSET_V1 is not defined , and the
caller does not need to care about using cpuset v1 or v2, which makes it
succinct.
However, we have to add !CONFIG_CPUSET_V1 to many places to avoid
useless function calls. I am not opposed to do this if you thick it 'a
better implementation.
Thanks,
Ridong
On 8/25/24 21:34, Chen Ridong wrote:
>
>
> On 2024/8/26 1:25, Waiman Long wrote:
>> On 8/23/24 06:01, Chen Ridong wrote:
>>> This patch introduces CONFIG_CPUSETS_V1 and guard cpuset-v1 code under
>>> CONFIG_CPUSETS_V1. The default value of CONFIG_CPUSETS_V1 is N, so that
>>> user who adopted v2 don't have 'pay' for cpuset v1.
>>> Besides, to make code succinct, rename
>>> '__cpuset_memory_pressure_bump()'
>>> to 'cpuset_memory_pressure_bump()', and expose it to the world, which
>>> takes place of the old mocro 'cpuset_memory_pressure_bump'.
>>>
>>> Signed-off-by: Chen Ridong <chenridong@huawei.com>
>>> ---
>>> include/linux/cpuset.h | 8 +-------
>>> init/Kconfig | 13 +++++++++++++
>>> kernel/cgroup/Makefile | 3 ++-
>>> kernel/cgroup/cpuset-internal.h | 15 +++++++++++++++
>>> kernel/cgroup/cpuset-v1.c | 10 ++++++----
>>> kernel/cgroup/cpuset.c | 2 ++
>>> 6 files changed, 39 insertions(+), 12 deletions(-)
>>>
>>> diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
>>> index 2a6981eeebf8..f91f1f61f482 100644
>>> --- a/include/linux/cpuset.h
>>> +++ b/include/linux/cpuset.h
>>> @@ -99,13 +99,7 @@ static inline bool cpuset_zone_allowed(struct
>>> zone *z, gfp_t gfp_mask)
>>> extern int cpuset_mems_allowed_intersects(const struct task_struct
>>> *tsk1,
>>> const struct task_struct *tsk2);
>>> -#define cpuset_memory_pressure_bump() \
>>> - do { \
>>> - if (cpuset_memory_pressure_enabled) \
>>> - __cpuset_memory_pressure_bump(); \
>>> - } while (0)
>>> -extern int cpuset_memory_pressure_enabled;
>>> -extern void __cpuset_memory_pressure_bump(void);
>>> +extern void cpuset_memory_pressure_bump(void);
>>> extern void cpuset_task_status_allowed(struct seq_file *m,
>>> struct task_struct *task);
>>
>> As you are introducing a new CONFIG_CPUSET_V1 kconfig option, you can
>> use it to eliminate a useless function call if cgroup v1 isn't
>> enabled. Not just this function, all the v1 specific functions should
>> have a !CONFIG_CPUSET_V1 version that can be optimized out.
>>
>> Cheers,
>> Longman
>>
>>
> Hi, Longman, currently, we have only added one place to manage all the
> functions whose are empty if !CONFIG_CPUSET_V1 is not defined , and
> the caller does not need to care about using cpuset v1 or v2, which
> makes it succinct.
>
> However, we have to add !CONFIG_CPUSET_V1 to many places to avoid
> useless function calls. I am not opposed to do this if you thick it 'a
> better implementation.
You don't need to touch any files that use these cgroup v1 only
functions. You only need to add some #ifdef code in the cpuset.h header
file like
#ifdef CONFIG_CPUSET_V1
extern void cpuset_memory_pressure_bump(void);
#else
static inline void cpuset_memory_pressure_bump(void) { }
#endif
BTW, try not to introduce unnecessary performance regression. The reason
cpuset_memory_pressure_bump() is a macro is to avoid a function call if
cpuset_memory_pressure_enabled isn't set which is most likely case. It
is cheaper to read a variable than do a function call.
Cheers,
Longman
On 2024/8/26 10:18, Waiman Long wrote:
> On 8/25/24 21:34, Chen Ridong wrote:
>>
>>
>> On 2024/8/26 1:25, Waiman Long wrote:
>>> On 8/23/24 06:01, Chen Ridong wrote:
>>>> This patch introduces CONFIG_CPUSETS_V1 and guard cpuset-v1 code under
>>>> CONFIG_CPUSETS_V1. The default value of CONFIG_CPUSETS_V1 is N, so that
>>>> user who adopted v2 don't have 'pay' for cpuset v1.
>>>> Besides, to make code succinct, rename
>>>> '__cpuset_memory_pressure_bump()'
>>>> to 'cpuset_memory_pressure_bump()', and expose it to the world, which
>>>> takes place of the old mocro 'cpuset_memory_pressure_bump'.
>>>>
>>>> Signed-off-by: Chen Ridong <chenridong@huawei.com>
>>>> ---
>>>> include/linux/cpuset.h | 8 +-------
>>>> init/Kconfig | 13 +++++++++++++
>>>> kernel/cgroup/Makefile | 3 ++-
>>>> kernel/cgroup/cpuset-internal.h | 15 +++++++++++++++
>>>> kernel/cgroup/cpuset-v1.c | 10 ++++++----
>>>> kernel/cgroup/cpuset.c | 2 ++
>>>> 6 files changed, 39 insertions(+), 12 deletions(-)
>>>>
>>>> diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
>>>> index 2a6981eeebf8..f91f1f61f482 100644
>>>> --- a/include/linux/cpuset.h
>>>> +++ b/include/linux/cpuset.h
>>>> @@ -99,13 +99,7 @@ static inline bool cpuset_zone_allowed(struct
>>>> zone *z, gfp_t gfp_mask)
>>>> extern int cpuset_mems_allowed_intersects(const struct task_struct
>>>> *tsk1,
>>>> const struct task_struct *tsk2);
>>>> -#define cpuset_memory_pressure_bump() \
>>>> - do { \
>>>> - if (cpuset_memory_pressure_enabled) \
>>>> - __cpuset_memory_pressure_bump(); \
>>>> - } while (0)
>>>> -extern int cpuset_memory_pressure_enabled;
>>>> -extern void __cpuset_memory_pressure_bump(void);
>>>> +extern void cpuset_memory_pressure_bump(void);
>>>> extern void cpuset_task_status_allowed(struct seq_file *m,
>>>> struct task_struct *task);
>>>
>>> As you are introducing a new CONFIG_CPUSET_V1 kconfig option, you can
>>> use it to eliminate a useless function call if cgroup v1 isn't
>>> enabled. Not just this function, all the v1 specific functions should
>>> have a !CONFIG_CPUSET_V1 version that can be optimized out.
>>>
>>> Cheers,
>>> Longman
>>>
>>>
>> Hi, Longman, currently, we have only added one place to manage all the
>> functions whose are empty if !CONFIG_CPUSET_V1 is not defined , and
>> the caller does not need to care about using cpuset v1 or v2, which
>> makes it succinct.
>>
>> However, we have to add !CONFIG_CPUSET_V1 to many places to avoid
>> useless function calls. I am not opposed to do this if you thick it 'a
>> better implementation.
>
> You don't need to touch any files that use these cgroup v1 only
> functions. You only need to add some #ifdef code in the cpuset.h header
> file like
>
> #ifdef CONFIG_CPUSET_V1
> extern void cpuset_memory_pressure_bump(void);
> #else
> static inline void cpuset_memory_pressure_bump(void) { }
> #endif
>
> BTW, try not to introduce unnecessary performance regression. The reason
> cpuset_memory_pressure_bump() is a macro is to avoid a function call if
> cpuset_memory_pressure_enabled isn't set which is most likely case. It
> is cheaper to read a variable than do a function call.
>
> Cheers,
> Longman
>
>
>
>
Thank you for your patience, I just misunderstood.
I will do it in v2.
Best regards,
Ridong
© 2016 - 2026 Red Hat, Inc.