[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20191107124918.GH4131@hirez.programming.kicks-ass.net>
Date: Thu, 7 Nov 2019 13:49:18 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: Jan Stancek <jstancek@...hat.com>,
LKML <linux-kernel@...r.kernel.org>, ltp@...ts.linux.it,
Al Viro <viro@...iv.linux.org.uk>,
Greg KH <gregkh@...uxfoundation.org>,
Ingo Molnar <mingo@...nel.org>
Subject: Re: [PATCH] kernel: use ktime_get_real_ts64() to calculate
acct.ac_btime
On Thu, Nov 07, 2019 at 09:56:12AM +0100, Thomas Gleixner wrote:
> But it would be way simpler just to store the CLOCK_REALTIME start time
> along with BOOT and MONOTONIC and just get rid of all these horrible
> calculations which are bound to be wrong.
>
> Peter?
So I'd really hate to do that, as that gives the impression REALTIME is
a sane clock to use. As I argued in that other email, REALTIME is
horrible crap.
Powered by blists - more mailing lists