Re: [PATCH v5 1/3] sched_ext: Introduce scx_bpf_cpu_rq_locked()

[Tejun Heo <tj@kernel.org>]

On Mon, Sep 01, 2025 at 02:26:02PM +0100, Christian Loehle wrote:
> Most fields in scx_bpf_cpu_rq() assume that its rq_lock is held.
> Furthermore they become meaningless without rq lock, too.
> Make a safer version of scx_bpf_cpu_rq() that only returns a rq
> if we hold rq lock of that rq.
> 
> Also mark the new scx_bpf_cpu_rq_locked() as returning NULL.
> 
> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
> ---
>  kernel/sched/ext.c                       | 23 +++++++++++++++++++++++
>  tools/sched_ext/include/scx/common.bpf.h |  1 +
>  2 files changed, 24 insertions(+)
> 
> diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
> index 4ae32ef179dd..9fcc310d85d5 100644
> --- a/kernel/sched/ext.c
> +++ b/kernel/sched/ext.c
> @@ -7430,6 +7430,28 @@ __bpf_kfunc struct rq *scx_bpf_cpu_rq(s32 cpu)
>  	return cpu_rq(cpu);
>  }
>  
> +/**
> + * scx_bpf_cpu_rq_locked - Return the rq currently locked by SCX
> + *
> + * Returns the rq if a rq lock is currently held by SCX.
> + * Otherwise emits an error and returns NULL.
> + */
> +__bpf_kfunc struct rq *scx_bpf_cpu_rq_locked(void)

How about naming it scx_bpf_locked_rq()? That reads a lot easier to me and
given that it doesn't take @cpu anymore, the _cpu_ part of the name isn't
necessary.

Thanks.

-- 
tejun