[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20191219092026.GW2844@hirez.programming.kicks-ass.net>
Date: Thu, 19 Dec 2019 10:20:26 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Ming Lei <ming.lei@...hat.com>
Cc: Thomas Gleixner <tglx@...utronix.de>, Jens Axboe <axboe@...nel.dk>,
linux-kernel@...r.kernel.org, linux-block@...r.kernel.org,
Long Li <longli@...rosoft.com>, Ingo Molnar <mingo@...hat.com>,
Keith Busch <keith.busch@...el.com>,
Christoph Hellwig <hch@....de>,
Sagi Grimberg <sagi@...mberg.me>,
John Garry <john.garry@...wei.com>,
Hannes Reinecke <hare@...e.com>
Subject: Re: [RFC PATCH 1/3] sched/core: add API for exporting runqueue clock
On Thu, Dec 19, 2019 at 09:29:14AM +0800, Ming Lei wrote:
> On Wed, Dec 18, 2019 at 10:51:01AM +0100, Peter Zijlstra wrote:
> > On Wed, Dec 18, 2019 at 03:19:40PM +0800, Ming Lei wrote:
> > > Scheduler runqueue maintains its own software clock that is periodically
> > > synchronised with hardware. Export this clock so that it can be used
> > > by interrupt flood detection for saving the cost of reading from hardware.
> >
> > But you don't have much, if any, guarantees the thing gets updated.
>
> Any software clock won't be guaranteed to be updated in time, however,
> they are still useful.
It still is broken, and I really don't want to expose this for whatever
reason.
Powered by blists - more mailing lists