From: peterz@infradead.org (Peter Zijlstra)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 0/2] cpustat: use atomic operations to read/update stats
Date: Fri, 22 Feb 2013 15:05:39 +0100 [thread overview]
Message-ID: <1361541939.26780.63.camel@laptop> (raw)
In-Reply-To: <20130222125019.GC17948@somewhere.redhat.com>
On Fri, 2013-02-22 at 13:50 +0100, Frederic Weisbecker wrote:
> > Which is a problem how?
>
> So here is a possible scenario, CPU 0 reads a kcpustat value, and CPU
> 1 writes
> it at the same time:
>
> //Initial value of "cpustat" is 0xffffffff
> == CPU 0 == == CPU 1 ==
>
> //load low part
> mov %eax, [cpustat]
> inc [cpustat]
> //Update the high part if necessary
> jnc 1f
> inc [cpustat + 4]
> 1:
> //load high part
> mov %edx, [cpustat + 4]
>
>
> Afterward, CPU 0 will think the value is 0x1ffffffff while it's
> actually
> 0x100000000.
>
> atomic64_read() and atomic64_set() are supposed to take care of that,
> without
> even the need for _inc() or _add() parts that use LOCK.
Sure I get that, but again, why is that a problem,.. who relies on
these statistics that makes it a problem?
WARNING: multiple messages have this Message-ID (diff)
From: Peter Zijlstra <peterz@infradead.org>
To: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Kevin Hilman <khilman@linaro.org>,
Russell King <rmk+kernel@arm.linux.org.uk>,
Thomas Gleixner <tglx@linutronix.de>,
Steven Rostedt <rostedt@goodmis.org>,
Ingo Molnar <mingo@kernel.org>,
linux-kernel@vger.kernel.org,
linux-arm-kernel@lists.infradead.org,
linaro-kernel@lists.linaro.org
Subject: Re: [PATCH 0/2] cpustat: use atomic operations to read/update stats
Date: Fri, 22 Feb 2013 15:05:39 +0100 [thread overview]
Message-ID: <1361541939.26780.63.camel@laptop> (raw)
In-Reply-To: <20130222125019.GC17948@somewhere.redhat.com>
On Fri, 2013-02-22 at 13:50 +0100, Frederic Weisbecker wrote:
> > Which is a problem how?
>
> So here is a possible scenario, CPU 0 reads a kcpustat value, and CPU
> 1 writes
> it at the same time:
>
> //Initial value of "cpustat" is 0xffffffff
> == CPU 0 == == CPU 1 ==
>
> //load low part
> mov %eax, [cpustat]
> inc [cpustat]
> //Update the high part if necessary
> jnc 1f
> inc [cpustat + 4]
> 1:
> //load high part
> mov %edx, [cpustat + 4]
>
>
> Afterward, CPU 0 will think the value is 0x1ffffffff while it's
> actually
> 0x100000000.
>
> atomic64_read() and atomic64_set() are supposed to take care of that,
> without
> even the need for _inc() or _add() parts that use LOCK.
Sure I get that, but again, why is that a problem,.. who relies on
these statistics that makes it a problem?
next prev parent reply other threads:[~2013-02-22 14:05 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-02-22 5:56 [PATCH 0/2] cpustat: use atomic operations to read/update stats Kevin Hilman
2013-02-22 5:56 ` Kevin Hilman
2013-02-22 5:56 ` [PATCH 1/2] cpustat: use accessor functions for get/set/add Kevin Hilman
2013-02-22 5:56 ` Kevin Hilman
2013-02-22 6:21 ` Viresh Kumar
2013-02-22 6:21 ` Viresh Kumar
2013-02-22 7:17 ` Amit Kucheria
2013-02-22 7:17 ` Amit Kucheria
2013-02-22 7:50 ` Viresh Kumar
2013-02-22 7:50 ` Viresh Kumar
2013-02-22 12:34 ` Frederic Weisbecker
2013-02-22 12:34 ` Frederic Weisbecker
2013-02-22 15:14 ` Kevin Hilman
2013-02-22 15:14 ` Kevin Hilman
2013-02-22 13:38 ` Frederic Weisbecker
2013-02-22 13:38 ` Frederic Weisbecker
2013-02-22 13:58 ` Peter Zijlstra
2013-02-22 13:58 ` Peter Zijlstra
2013-02-22 14:03 ` Frederic Weisbecker
2013-02-22 14:03 ` Frederic Weisbecker
2013-02-22 5:56 ` [PATCH 2/2] cpustat: convert to atomic operations Kevin Hilman
2013-02-22 5:56 ` Kevin Hilman
2013-02-22 8:46 ` [PATCH 0/2] cpustat: use atomic operations to read/update stats Peter Zijlstra
2013-02-22 8:46 ` Peter Zijlstra
2013-02-22 12:50 ` Frederic Weisbecker
2013-02-22 12:50 ` Frederic Weisbecker
2013-02-22 13:48 ` Peter Zijlstra
2013-02-22 13:48 ` Peter Zijlstra
2013-02-22 13:54 ` Ingo Molnar
2013-02-22 13:54 ` Ingo Molnar
2013-02-22 14:04 ` Peter Zijlstra
2013-02-22 14:04 ` Peter Zijlstra
2013-02-22 14:16 ` Ingo Molnar
2013-02-22 14:16 ` Ingo Molnar
2013-02-22 14:25 ` Peter Zijlstra
2013-02-22 14:25 ` Peter Zijlstra
2013-02-22 14:05 ` Peter Zijlstra [this message]
2013-02-22 14:05 ` Peter Zijlstra
2013-02-22 14:15 ` Frederic Weisbecker
2013-02-22 14:15 ` Frederic Weisbecker
2013-02-22 14:09 ` Peter Zijlstra
2013-02-22 14:09 ` Peter Zijlstra
2013-02-22 14:18 ` Ingo Molnar
2013-02-22 14:18 ` Ingo Molnar
2013-02-22 14:21 ` Eric Dumazet
2013-02-22 14:21 ` Eric Dumazet
2013-02-22 14:23 ` Frederic Weisbecker
2013-02-22 14:23 ` Frederic Weisbecker
2013-02-22 19:01 ` Kevin Hilman
2013-02-22 19:01 ` Kevin Hilman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1361541939.26780.63.camel@laptop \
--to=peterz@infradead.org \
--cc=linux-arm-kernel@lists.infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.