From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-fsdevel-owner@vger.kernel.org>
Received: from mga02.intel.com ([134.134.136.20]:18445 "EHLO mga02.intel.com"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1752983AbeFTQTT (ORCPT <rfc822;linux-fsdevel@vger.kernel.org>);
        Wed, 20 Jun 2018 12:19:19 -0400
From: Andi Kleen <ak@linux.intel.com>
To: Arnd Bergmann <arnd@arndb.de>
Cc: Jens Axboe <axboe@kernel.dk>, Jan Kara <jack@suse.cz>,
        Jeff Layton <jlayton@redhat.com>,
        "Darrick J. Wong" <darrick.wong@oracle.com>,
        y2038 Mailman List <y2038@lists.linaro.org>,
        Brian Foster <bfoster@redhat.com>,
        Miklos Szeredi <miklos@szeredi.hu>,
        Pavel Tatashin <pasha.tatashin@oracle.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Linux FS-devel Mailing List <linux-fsdevel@vger.kernel.org>,
        Alexander Viro <viro@zeniv.linux.org.uk>,
        Andi Kleen <andi.kleen@intel.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Deepa Dinamani <deepa.kernel@gmail.com>
Subject: Re: [PATCH] vfs: replace current_kernel_time64 with ktime equivalent
References: <20180620150138.49380-1-arnd@arndb.de>
        <87r2l1sby0.fsf@linux.intel.com>
        <CAK8P3a1J9j7pz-d+otbJweWgOZZePdaMJEb_sU2GJKi3oYq7Xg@mail.gmail.com>
Date: Wed, 20 Jun 2018 09:19:18 -0700
In-Reply-To: <CAK8P3a1J9j7pz-d+otbJweWgOZZePdaMJEb_sU2GJKi3oYq7Xg@mail.gmail.com>
        (Arnd Bergmann's message of "Wed, 20 Jun 2018 18:14:50 +0200")
Message-ID: <87muvpsa61.fsf@linux.intel.com>
MIME-Version: 1.0
Content-Type: text/plain
Sender: linux-fsdevel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

Arnd Bergmann <arnd@arndb.de> writes:
>
> To clarify: current_kernel_time() uses at most millisecond resolution rather
> than microsecond, as tkr_mono.xtime_nsec only gets updated during the
> timer tick.

Ah you're right. I remember now: the motivation was to make sure there
is basically no overhead. In some setups the full gtod can be rather
slow, particularly if it falls back to some crappy timer.

I think it would be ok if it falls back to jiffies if TSC or a similar
fast timer doesn't work. But the function you're using likely
doesn't do that?

> Has that time scale changed over the past 16 years as CPUs got faster
> (and system call entry times slower down again with recent changes)?

Maybe a bit, but not substantially.



-Andi