From: Valentin Schneider <valentin.schneider@arm.com>
To: Ingo Molnar <mingo@kernel.org>
Cc: linux-kernel@vger.kernel.org, peterz@infradead.org,
vincent.guittot@linaro.org, dietmar.eggemann@arm.com,
morten.rasmussen@arm.com, Quentin Perret <qperret@google.com>
Subject: Re: [PATCH v4 05/10] sched/topology: Define and assign sched_domain flag metadata
Date: Sat, 08 Aug 2020 01:19:24 +0100 [thread overview]
Message-ID: <jhj1rkihtv7.mognet@arm.com> (raw)
In-Reply-To: <jhj8serixd4.mognet@arm.com>
On 06/08/20 17:18, Valentin Schneider wrote:
> In the grand scheme of things I'd actually like to have this file output
> the names of the flags rather than their values (since I now save them when
> SCHED_DEBUG=y), but I didn't find a simple way to hack the existing SD ctl
> table (sd_alloc_ctl_domain_table() and co) into doing this.
>
I "just" had to spend some more time grokking how the whole ctl
proc_handler thing is supposed to work; I now have a mostly working
solution, i.e. on my Juno:
$ cat /proc/sys/kernel/sched_domain/cpu0/domain*/flags
SD_BALANCE_NEWIDLE SD_BALANCE_EXEC SD_BALANCE_FORK SD_WAKE_AFFINE SD_SHARE_PKG_RESOURCES
SD_BALANCE_NEWIDLE SD_BALANCE_EXEC SD_BALANCE_FORK SD_WAKE_AFFINE SD_ASYM_CPUCAPACITY SD_PREFER_SIBLING
I'll clean that up and go for the automagic ordering I previously
described.
>
> Now as to making this fully automagic, I *think* I could do something like
> having a first enum to set up an ordering:
>
> #define SD_FLAG(name, ...) __##name,
> enum {
> #include <linux/sched/sd_flags.h>
> };
>
> A second one to have powers of 2:
>
> #define SD_FLAG(name, ...) name = 1 << __##name,
> enum {
> #include <linux/sched/sd_flags.h>
> };
>
> And finally the metadata array assignment might be doable with:
>
> #define SD_FLAG(_name, mflags) [__##_name] = { .meta_flags = mflags, .name = #_name },
>
> Or, if there is a way somehow to directly get powers of 2 out of an enum:
>
> #define SD_FLAG(_name, mflags) [_ffs(_name)] = { .meta_flags = mflags, .name = #_name },
>
>>> +#ifdef CONFIG_SCHED_DEBUG
>>> +#define SD_FLAG(_name, idx, mflags) [idx] = {.meta_flags = mflags, .name = #_name},
>>
>> s/{./{ .
>> s/e}/e }
>>
>> Thanks,
>>
>> Ingo
next prev parent reply other threads:[~2020-08-08 0:19 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-31 11:54 [PATCH v4 00/10] sched: Instrument sched domain flags Valentin Schneider
2020-07-31 11:54 ` [PATCH v4 01/10] ARM, sched/topology: Remove SD_SHARE_POWERDOMAIN Valentin Schneider
2020-07-31 11:54 ` [PATCH v4 02/10] ARM: Revert back to default scheduler topology Valentin Schneider
2020-07-31 11:54 ` [PATCH v4 03/10] sched/topology: Propagate SD_ASYM_CPUCAPACITY upwards Valentin Schneider
2020-08-06 14:20 ` Ingo Molnar
2020-08-06 16:19 ` Valentin Schneider
2020-08-06 16:46 ` Ingo Molnar
2020-07-31 11:54 ` [PATCH v4 04/10] sched/topology: Split out SD_* flags declaration to its own file Valentin Schneider
2020-07-31 11:54 ` [PATCH v4 05/10] sched/topology: Define and assign sched_domain flag metadata Valentin Schneider
2020-08-04 11:08 ` peterz
2020-08-04 11:12 ` Valentin Schneider
2020-08-06 14:07 ` Ingo Molnar
2020-08-06 16:18 ` Valentin Schneider
2020-08-08 0:19 ` Valentin Schneider [this message]
2020-07-31 11:54 ` [PATCH v4 06/10] sched/topology: Verify SD_* flags setup when sched_debug is on Valentin Schneider
2020-07-31 11:54 ` [PATCH v4 07/10] sched/topology: Add more flags to the SD degeneration mask Valentin Schneider
2020-08-06 13:57 ` Ingo Molnar
2020-07-31 11:55 ` [PATCH v4 08/10] sched/topology: Remove SD_SERIALIZE degeneration special case Valentin Schneider
2020-07-31 11:55 ` [PATCH v4 09/10] sched/topology: Introduce SD metaflag for flags needing > 1 groups Valentin Schneider
2020-07-31 11:55 ` [PATCH v4 10/10] sched/topology: Use prebuilt SD flag degeneration mask Valentin Schneider
2020-08-06 11:25 ` [PATCH v4 00/10] sched: Instrument sched domain flags Dietmar Eggemann
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=jhj1rkihtv7.mognet@arm.com \
--to=valentin.schneider@arm.com \
--cc=dietmar.eggemann@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=morten.rasmussen@arm.com \
--cc=peterz@infradead.org \
--cc=qperret@google.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
Powered by JetHome