[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <3fb752bb-793d-4c0b-bf7f-12d04a4e7cc0@amd.com>
Date: Fri, 25 Apr 2025 09:18:14 +0530
From: K Prateek Nayak <kprateek.nayak@....com>
To: Prakash Sangappa <prakash.sangappa@...cle.com>, Sebastian Andrzej Siewior
<bigeasy@...utronix.de>
CC: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"peterz@...radead.org" <peterz@...radead.org>, "rostedt@...dmis.org"
<rostedt@...dmis.org>, "mathieu.desnoyers@...icios.com"
<mathieu.desnoyers@...icios.com>, "tglx@...utronix.de" <tglx@...utronix.de>
Subject: Re: [PATCH V2 1/3] Sched: Scheduler time slice extension
Hello Prakash,
On 4/25/2025 6:49 AM, Prakash Sangappa wrote:
>>> static void hrtick_clear(struct rq *rq)
>>> {
>>> + rseq_delay_resched_tick();
>> This is called from __schedule(). If you set the need-resched flag here,
>> it gets removed shortly after. Do I miss something?
> hrtick_clear() is also called when the cpu is being removed in sched_cpu_dying().
> We need to set resched there?
sched_cpu_dying() is called from the cpuhp thread which will go away
once the hotplug is done and shouldn't need this. Furthermore, cpuhp
thread will not use the "rseq_sched_delay" API so removing this should
be fine.
--
Thanks and Regards,
Prateek
Powered by blists - more mailing lists