[PATCH v2] sched_ext: initialize built-in idle state before ops.init()

[PATCH v2] sched_ext: initialize built-in idle state before ops.init()

[Andrea Righi]

A BPF scheduler may want to use the built-in idle cpumasks in ops.init()
before the scheduler is fully initialized, either directly or through a
BPF timer for example.

However, this would result in an error, since the idle state has not
been properly initialized yet.

This can be easily verified by modifying scx_simple to call
scx_bpf_get_idle_cpumask() in ops.init():

$ sudo scx_simple

DEBUG DUMP
===========================================================================

scx_simple[121] triggered exit kind 1024:
  runtime error (built-in idle tracking is disabled)
...

Fix this by properly initializing the idle state before ops.init() is
called. With this change applied:

$ sudo scx_simple
local=2 global=0
local=19 global=11
local=23 global=11
...

Fixes: d73249f88743d ("sched_ext: idle: Make idle static keys private")
Signed-off-by: Andrea Righi <arighi@nvidia.com>
---
 kernel/sched/ext.c      | 4 ++--
 kernel/sched/ext_idle.c | 8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

ChangeLog v1 -> v2:
 - fix potential deadlock with cpu_hotplug_lock

diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
index d82e9d3cbc0dc..3d20228052217 100644
--- a/kernel/sched/ext.c
+++ b/kernel/sched/ext.c
@@ -5361,6 +5361,8 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link)
 	 */
 	cpus_read_lock();
 
+	scx_idle_enable(ops);
+
 	if (scx_ops.init) {
 		ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, init);
 		if (ret) {
@@ -5427,8 +5429,6 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link)
 	if (scx_ops.cpu_acquire || scx_ops.cpu_release)
 		static_branch_enable(&scx_ops_cpu_preempt);
 
-	scx_idle_enable(ops);
-
 	/*
 	 * Lock out forks, cgroup on/offlining and moves before opening the
 	 * floodgate so that they don't wander into the operations prematurely.
diff --git a/kernel/sched/ext_idle.c b/kernel/sched/ext_idle.c
index 746fd36050045..811f8bff2ff02 100644
--- a/kernel/sched/ext_idle.c
+++ b/kernel/sched/ext_idle.c
@@ -794,14 +794,14 @@ static void reset_idle_masks(struct sched_ext_ops *ops)
 void scx_idle_enable(struct sched_ext_ops *ops)
 {
 	if (!ops->update_idle || (ops->flags & SCX_OPS_KEEP_BUILTIN_IDLE))
-		static_branch_enable(&scx_builtin_idle_enabled);
+		static_branch_enable_cpuslocked(&scx_builtin_idle_enabled);
 	else
-		static_branch_disable(&scx_builtin_idle_enabled);
+		static_branch_disable_cpuslocked(&scx_builtin_idle_enabled);
 
 	if (ops->flags & SCX_OPS_BUILTIN_IDLE_PER_NODE)
-		static_branch_enable(&scx_builtin_idle_per_node);
+		static_branch_enable_cpuslocked(&scx_builtin_idle_per_node);
 	else
-		static_branch_disable(&scx_builtin_idle_per_node);
+		static_branch_disable_cpuslocked(&scx_builtin_idle_per_node);
 
 #ifdef CONFIG_SMP
 	reset_idle_masks(ops);
-- 
2.49.0

Re: [PATCH v2] sched_ext: initialize built-in idle state before ops.init()

[Joel Fernandes]

On Tue, Mar 25, 2025 at 10:32:12AM +0100, Andrea Righi wrote:
> A BPF scheduler may want to use the built-in idle cpumasks in ops.init()
> before the scheduler is fully initialized, either directly or through a
> BPF timer for example.
> 
> However, this would result in an error, since the idle state has not
> been properly initialized yet.
> 
> This can be easily verified by modifying scx_simple to call
> scx_bpf_get_idle_cpumask() in ops.init():
> 
> $ sudo scx_simple
> 
> DEBUG DUMP
> ===========================================================================
> 
> scx_simple[121] triggered exit kind 1024:
>   runtime error (built-in idle tracking is disabled)
> ...
> 
> Fix this by properly initializing the idle state before ops.init() is
> called. With this change applied:
> 
> $ sudo scx_simple
> local=2 global=0
> local=19 global=11
> local=23 global=11
> ...
> 
> Fixes: d73249f88743d ("sched_ext: idle: Make idle static keys private")
> Signed-off-by: Andrea Righi <arighi@nvidia.com>

Reviewed-by: Joel Fernandes <joelagnelf@nvidia.com>

Makes sense, I think this is restriction is because static branch will use
IPIs in some architectures which requires to exclude hotplug operations.

thanks,

 - Joel


> ---
>  kernel/sched/ext.c      | 4 ++--
>  kernel/sched/ext_idle.c | 8 ++++----
>  2 files changed, 6 insertions(+), 6 deletions(-)
> 
> ChangeLog v1 -> v2:
>  - fix potential deadlock with cpu_hotplug_lock
> 
> diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
> index d82e9d3cbc0dc..3d20228052217 100644
> --- a/kernel/sched/ext.c
> +++ b/kernel/sched/ext.c
> @@ -5361,6 +5361,8 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link)
>  	 */
>  	cpus_read_lock();
>  
> +	scx_idle_enable(ops);
> +
>  	if (scx_ops.init) {
>  		ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, init);
>  		if (ret) {
> @@ -5427,8 +5429,6 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link)
>  	if (scx_ops.cpu_acquire || scx_ops.cpu_release)
>  		static_branch_enable(&scx_ops_cpu_preempt);
>  
> -	scx_idle_enable(ops);
> -
>  	/*
>  	 * Lock out forks, cgroup on/offlining and moves before opening the
>  	 * floodgate so that they don't wander into the operations prematurely.
> diff --git a/kernel/sched/ext_idle.c b/kernel/sched/ext_idle.c
> index 746fd36050045..811f8bff2ff02 100644
> --- a/kernel/sched/ext_idle.c
> +++ b/kernel/sched/ext_idle.c
> @@ -794,14 +794,14 @@ static void reset_idle_masks(struct sched_ext_ops *ops)
>  void scx_idle_enable(struct sched_ext_ops *ops)
>  {
>  	if (!ops->update_idle || (ops->flags & SCX_OPS_KEEP_BUILTIN_IDLE))
> -		static_branch_enable(&scx_builtin_idle_enabled);
> +		static_branch_enable_cpuslocked(&scx_builtin_idle_enabled);
>  	else
> -		static_branch_disable(&scx_builtin_idle_enabled);
> +		static_branch_disable_cpuslocked(&scx_builtin_idle_enabled);
>  
>  	if (ops->flags & SCX_OPS_BUILTIN_IDLE_PER_NODE)
> -		static_branch_enable(&scx_builtin_idle_per_node);
> +		static_branch_enable_cpuslocked(&scx_builtin_idle_per_node);
>  	else
> -		static_branch_disable(&scx_builtin_idle_per_node);
> +		static_branch_disable_cpuslocked(&scx_builtin_idle_per_node);
>  
>  #ifdef CONFIG_SMP
>  	reset_idle_masks(ops);
> -- 
> 2.49.0
> 

Re: [PATCH v2] sched_ext: initialize built-in idle state before ops.init()

[Tejun Heo]

On Tue, Mar 25, 2025 at 10:32:12AM +0100, Andrea Righi wrote:
> A BPF scheduler may want to use the built-in idle cpumasks in ops.init()
> before the scheduler is fully initialized, either directly or through a
> BPF timer for example.
> 
> However, this would result in an error, since the idle state has not
> been properly initialized yet.
> 
> This can be easily verified by modifying scx_simple to call
> scx_bpf_get_idle_cpumask() in ops.init():
> 
> $ sudo scx_simple
> 
> DEBUG DUMP
> ===========================================================================
> 
> scx_simple[121] triggered exit kind 1024:
>   runtime error (built-in idle tracking is disabled)
> ...
> 
> Fix this by properly initializing the idle state before ops.init() is
> called. With this change applied:
> 
> $ sudo scx_simple
> local=2 global=0
> local=19 global=11
> local=23 global=11
> ...
> 
> Fixes: d73249f88743d ("sched_ext: idle: Make idle static keys private")
> Signed-off-by: Andrea Righi <arighi@nvidia.com>

Applied to sched_ext/for-6.15-fixes.

Thanks.

-- 
tejun