From: Morten Rasmussen <morten.rasmussen@arm.com>
To: Valentin Schneider <valentin.schneider@arm.com>
Cc: linux-kernel@vger.kernel.org, mingo@kernel.org,
peterz@infradead.org, vincent.guittot@linaro.org,
dietmar.eggemann@arm.com
Subject: Re: [RFC PATCH 1/3] sched/topology: Split out SD_* flags declaration to its own file
Date: Mon, 23 Mar 2020 14:42:34 +0100 [thread overview]
Message-ID: <20200323134234.GD6103@e123083-lin> (raw)
In-Reply-To: <20200311183320.19186-2-valentin.schneider@arm.com>
On Wed, Mar 11, 2020 at 06:33:18PM +0000, Valentin Schneider wrote:
> diff --git a/include/linux/sched/sd_flags.h b/include/linux/sched/sd_flags.h
> new file mode 100644
> index 000000000000..685bbe736945
> --- /dev/null
> +++ b/include/linux/sched/sd_flags.h
> @@ -0,0 +1,33 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * sched-domains (multiprocessor balancing) flag declarations.
> + */
> +
> +/* Balance when about to become idle */
> +SD_FLAG(SD_BALANCE_NEWIDLE, 0)
> +/* Balance on exec */
> +SD_FLAG(SD_BALANCE_EXEC, 1)
> +/* Balance on fork, clone */
> +SD_FLAG(SD_BALANCE_FORK, 2)
> +/* Balance on wakeup */
> +SD_FLAG(SD_BALANCE_WAKE, 3)
> +/* Wake task to waking CPU */
> +SD_FLAG(SD_WAKE_AFFINE, 4)
Isn't it more like: "Consider waking task on waking CPU"?
IIRC, with this flag set the wake-up can happen either near prev_cpu or
this_cpu.
> +/* Domain members have different CPU capacities */
> +SD_FLAG(SD_ASYM_CPUCAPACITY, 5)
> +/* Domain members share CPU capacity */
> +SD_FLAG(SD_SHARE_CPUCAPACITY, 6)
Perhaps add +" (SMT)" to the comment to help the uninitiated
understanding it a bit easier?
> +/* Domain members share power domain */
> +SD_FLAG(SD_SHARE_POWERDOMAIN, 7)
This flag is set only by 32-bit arm and has never had any effect. I
think it was the beginning of something years ago that hasn't
progressed. Perhaps we can remove it now?
> +/* Domain members share CPU pkg resources */
> +SD_FLAG(SD_SHARE_PKG_RESOURCES, 8)
+" (e.g. caches)" ?
> +/* Only a single load balancing instance */
> +SD_FLAG(SD_SERIALIZE, 9)
> +/* Place busy groups earlier in the domain */
> +SD_FLAG(SD_ASYM_PACKING, 10)
Place busy _tasks_ earlier in the domain?
It is a bit unclear what 'earlier' means here but since the packing
ordering can actually be defined by the architecture, we can't be much
more specific I guess.
Morten
next prev parent reply other threads:[~2020-03-23 13:42 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-11 18:33 [RFC PATCH 0/3] sched: Instrument sched domain flags Valentin Schneider
2020-03-11 18:33 ` [RFC PATCH 1/3] sched/topology: Split out SD_* flags declaration to its own file Valentin Schneider
2020-03-23 13:42 ` Morten Rasmussen [this message]
2020-03-23 17:10 ` Valentin Schneider
2020-03-24 8:35 ` Morten Rasmussen
2020-03-11 18:33 ` [RFC PATCH 2/3] sched/topology: Define and assign sched_domain flag metadata Valentin Schneider
2020-03-11 18:33 ` [RFC PATCH 3/3] sched/topology: Verify SD_* flags setup when sched_debug is on 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=20200323134234.GD6103@e123083-lin \
--to=morten.rasmussen@arm.com \
--cc=dietmar.eggemann@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.