[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d7371d20-e2d8-c5d1-b59b-850d86bf7a83@infradead.org>
Date: Mon, 19 Apr 2021 11:52:33 -0700
From: Randy Dunlap <rdunlap@...radead.org>
To: Luigi Rizzo <lrizzo@...gle.com>, linux-kernel@...r.kernel.org,
peterz@...radead.org, axboe@...nel.dk, paulmck@...nel.org
Subject: Re: [PATCH] smp: add a best_effort version of
smp_call_function_many()
Hi--
On 4/19/21 11:44 AM, Luigi Rizzo wrote:
>
> Signed-off-by: Luigi Rizzo <lrizzo@...gle.com>
> ---
> include/linux/smp.h | 10 ++++++
> kernel/smp.c | 75 +++++++++++++++++++++++++++++++++++++--------
> 2 files changed, 72 insertions(+), 13 deletions(-)
>
> diff --git a/kernel/smp.c b/kernel/smp.c
> index aeb0adfa0606..75155875fadc 100644
> --- a/kernel/smp.c
> +++ b/kernel/smp.c
> @@ -693,8 +715,32 @@ static void smp_call_function_many_cond(const struct cpumask *mask,
> csd_lock_wait(csd);
> }
> }
> + return busy ? cfd->cpumask : NULL;
> }
>
> +/**
> + * Extended version of smp_call_function_many(). Same constraints.
> + * @mode == 0 same as wait = false, returns 0;
> + * @mode == 1 same as wait = true, returns 0;
> + * @mode = SMP_CFM_BEST_EFFORT: skips CPUs with previous pending requests,
> + * returns 0 and *mask unmodified if no CPUs are skipped,
> + * -EBUSY if CPUs are skipped, and *mask is the set of skipped CPUs
Please convert those comments to real kernel-doc format.
(Documentation/doc-guide/kernel-doc.rst)
> + */
> +int __smp_call_function_many(struct cpumask *mask, smp_call_func_t func,
> + void *info, int mode)
> +{
thanks.
--
~Randy
Powered by blists - more mailing lists