Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
References: <20200311181601.18314-1-valentin.schneider@arm.com> <20200311181601.18314-3-valentin.schneider@arm.com> <4a7fe6ae-3587-4a55-1cf2-c4fe568a5ffa@arm.com>
User-agent: mu4e 0.9.17; emacs 26.3
From:   Valentin Schneider <valentin.schneider@arm.com>
To:     Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc:     linux-kernel@vger.kernel.org, mingo@kernel.org,
        peterz@infradead.org, vincent.guittot@linaro.org
Subject: Re: [PATCH v2 2/9] sched/debug: Make sd->flags sysctl read-only
In-reply-to: <4a7fe6ae-3587-4a55-1cf2-c4fe568a5ffa@arm.com>
Date:   Thu, 19 Mar 2020 12:04:56 +0000
Message-ID: <jhjwo7go8if.mognet@arm.com>
MIME-Version: 1.0
Content-Type: text/plain
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On Thu, Mar 19 2020, Dietmar Eggemann wrote:
> On 11.03.20 19:15, Valentin Schneider wrote:
>> Writing to the sysctl of a sched_domain->flags directly updates the value of
>> the field, and goes nowhere near update_top_cache_domain(). This means that
>> the cached domain pointers can end up containing stale data (e.g. the
>> domain pointed to doesn't have the relevant flag set anymore).
>>
>> Explicit domain walks that check for flags will be affected by
>> the write, but this won't be in sync with the cached pointers which will
>> still point to the domains that were cached at the last sched_domain
>> build.
>>
>> In other words, writing to this interface is playing a dangerous game. It
>> could be made to trigger an update of the cached sched_domain pointers when
>> written to, but this does not seem to be worth the trouble. Make it
>> read-only.
>
> As long as I don't change SD flags for which cached SD pointers exist
> (SD_SHARE_PKG_RESOURCES, SD_NUMA, SD_ASYM_PACKING or
> SD_ASYM_CPUCAPACITY) the write-able interface still could make some sense.
>
> E.g. by enabling SD_BALANCE_WAKE on the fly, I can force !want_affine
> wakees into slow path.
>

True, although since there is no explicit differenciation between the
cached and !cached flags, this is still a landmined interface.

> The question is, do people use the writable flags interface to tweak
> select_task_rq_fair() behavior in this regard?

I did try asking around on IRC (which admittedly is just a small subset
of people) but didn't find anyone.