Re: [PATCH v2] sched: Micro optimization in pick_next_task() and in check_preempt_curr()

[Steven Rostedt]

On Thu, 19 Dec 2019 18:20:58 +0300
Kirill Tkhai <ktkhai@xxxxxxxxxxxxx> wrote:

> diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
> index e00f41aa8ec4..ff12a422ff19 100644
> --- a/include/asm-generic/vmlinux.lds.h
> +++ b/include/asm-generic/vmlinux.lds.h
> @@ -108,6 +108,13 @@
>  #define SBSS_MAIN .sbss
>  #endif

I would also add a comment here:

/*
 * The order of the sched class addresses are important, as they are
 * used to determine the order of the priority of each sched class in
 * relation to each other.
 */

>  
> +#define SCHED_DATA				\
> +	*(__idle_sched_class)			\
> +	*(__fair_sched_class)			\
> +	*(__rt_sched_class)			\
> +	*(__dl_sched_class)			\
> +	*(__stop_sched_class)
> +
>  /*
>   * Align to a 32 byte boundary equal to the
>   * alignment gcc 4.5 uses for a struct
> @@ -308,6 +315,7 @@
>  #define DATA_DATA							\
>  	*(.xiptext)							\
>  	*(DATA_MAIN)							\
> +	SCHED_DATA							\
>  	*(.ref.data)							\
>  	*(.data..shared_aligned) /* percpu related */			\
>  	MEM_KEEP(init.data*)						\