mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Dietmar Eggemann <dietmar.eggemann@arm.com>
To: Valentin Schneider <valentin.schneider@arm.com>,
	linux-kernel@vger.kernel.org
Cc: mingo@kernel.org, peterz@infradead.org,
	vincent.guittot@linaro.org, morten.rasmussen@arm.com
Subject: Re: [PATCH v3 5/7] sched/topology: Add more flags to the SD degeneration mask
Date: Thu, 2 Jul 2020 20:28:38 +0200	[thread overview]
Message-ID: <3406fe6b-405b-d542-6459-732f4ae890f6@arm.com> (raw)
In-Reply-To: <20200701190656.10126-6-valentin.schneider@arm.com>

On 01/07/2020 21:06, Valentin Schneider wrote:
> I don't think it is going to change much in practice, but we were missing
> those:
> 
> o SD_BALANCE_WAKE: Used just like the other SD_BALANCE_* flags, so also
>   needs > 1 group.
> o SD_ASYM_PACKING: Hinges on load balancing (periodic / wakeup), thus needs
>   > 1 group to happen
> o SD_OVERLAP: Describes domains with overlapping groups; can't have
>   overlaps with a single group.
> 
> SD_PREFER_SIBLING is as always the odd one out: we currently consider it
> in sd_parent_degenerate() but not in sd_degenerate(). It too hinges on load
> balancing, and thus won't have any effect when set on a domain with a
> single group. Add it to the sd_degenerate() groups mask.
> 
> Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
> ---
>  kernel/sched/topology.c | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
> index 6047d491abe9..fe393b295444 100644
> --- a/kernel/sched/topology.c
> +++ b/kernel/sched/topology.c
> @@ -161,11 +161,15 @@ static int sd_degenerate(struct sched_domain *sd)
>  
>  	/* Following flags need at least 2 groups */
>  	if (sd->flags & (SD_BALANCE_NEWIDLE |
> +			 SD_BALANCE_WAKE |
>  			 SD_BALANCE_FORK |
>  			 SD_BALANCE_EXEC |
> +			 SD_ASYM_PACKING |
>  			 SD_SHARE_CPUCAPACITY |
>  			 SD_ASYM_CPUCAPACITY |
> -			 SD_SHARE_PKG_RESOURCES)) {
> +			 SD_SHARE_PKG_RESOURCES |
> +			 SD_OVERLAP |
> +			 SD_PREFER_SIBLING)) {

Why do you add SD_PREFER_SIBLING here? It doesn't have SDF_NEEDS_GROUPS
set in [PATCH v3 6/7].

$ cat ./include/linux/sched/sd_flags.h | grep ^SD_FLAG
SD_FLAG(SD_BALANCE_NEWIDLE,     0, SDF_SHARED_CHILD | SDF_NEEDS_GROUPS)
SD_FLAG(SD_BALANCE_EXEC,        1, SDF_SHARED_CHILD | SDF_NEEDS_GROUPS)
SD_FLAG(SD_BALANCE_FORK,        2, SDF_SHARED_CHILD | SDF_NEEDS_GROUPS)
SD_FLAG(SD_BALANCE_WAKE,        3, SDF_SHARED_CHILD | SDF_NEEDS_GROUPS)
SD_FLAG(SD_WAKE_AFFINE,         4, SDF_SHARED_CHILD)
SD_FLAG(SD_ASYM_CPUCAPACITY,    5, SDF_SHARED_PARENT | SDF_NEEDS_GROUPS)
SD_FLAG(SD_SHARE_CPUCAPACITY,   6, SDF_SHARED_CHILD | SDF_NEEDS_GROUPS)
SD_FLAG(SD_SHARE_PKG_RESOURCES, 7, SDF_SHARED_CHILD | SDF_NEEDS_GROUPS)
SD_FLAG(SD_SERIALIZE,           8, SDF_SHARED_PARENT)
SD_FLAG(SD_ASYM_PACKING,        9, SDF_SHARED_CHILD | SDF_NEEDS_GROUPS)
SD_FLAG(SD_PREFER_SIBLING,      10, 0)                           <-- !!!
SD_FLAG(SD_OVERLAP,             11, SDF_SHARED_PARENT |SDF_NEEDS_GROUPS)
SD_FLAG(SD_NUMA,                12, SDF_SHARED_PARENT)

  reply	other threads:[~2020-07-02 18:29 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-01 19:06 [PATCH v3 0/7] sched: Instrument sched domain flags Valentin Schneider
2020-07-01 19:06 ` [PATCH v3 1/7] sched/topology: Split out SD_* flags declaration to its own file Valentin Schneider
2020-07-01 19:06 ` [PATCH v3 2/7] sched/topology: Define and assign sched_domain flag metadata Valentin Schneider
2020-07-02 12:15   ` Quentin Perret
2020-07-02 14:31     ` Valentin Schneider
2020-07-02 15:45       ` Quentin Perret
2020-07-02 16:25         ` Valentin Schneider
2020-07-02 16:37           ` Quentin Perret
2020-07-02 16:49             ` Valentin Schneider
2020-07-01 19:06 ` [PATCH v3 3/7] sched/topology: Verify SD_* flags setup when sched_debug is on Valentin Schneider
2020-07-02 14:20   ` Peter Zijlstra
2020-07-02 14:32     ` Valentin Schneider
2020-07-01 19:06 ` [PATCH v3 4/7] arm, sched/topology: Remove SD_SHARE_POWERDOMAIN Valentin Schneider
2020-07-02 16:44   ` Dietmar Eggemann
2020-07-02 18:46     ` Valentin Schneider
2020-07-01 19:06 ` [PATCH v3 5/7] sched/topology: Add more flags to the SD degeneration mask Valentin Schneider
2020-07-02 18:28   ` Dietmar Eggemann [this message]
2020-07-01 19:06 ` [PATCH v3 6/7] sched/topology: Introduce SD metaflag for flags needing > 1 groups Valentin Schneider
2020-07-02 18:29   ` Dietmar Eggemann
2020-07-02 18:46     ` Valentin Schneider
2020-07-13 12:39   ` Peter Zijlstra
2020-07-13 13:25     ` Valentin Schneider
2020-07-01 19:06 ` [PATCH v3 7/7] sched/topology: Use prebuilt SD flag degeneration mask Valentin Schneider
2020-07-13 12:55   ` Peter Zijlstra
2020-07-13 13:28     ` Valentin Schneider
2020-07-13 13:43       ` Peter Zijlstra
2020-07-13 13:52         ` Valentin Schneider

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3406fe6b-405b-d542-6459-732f4ae890f6@arm.com \
    --to=dietmar.eggemann@arm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=morten.rasmussen@arm.com \
    --cc=peterz@infradead.org \
    --cc=valentin.schneider@arm.com \
    --cc=vincent.guittot@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®