mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Christoph Lameter <cl@linux.com>, Tejun Heo <tj@kernel.org>,
	akpm@linuxfoundation.org, rostedt@goodmis.org,
	linux-kernel@vger.kernel.org, Ingo Molnar <mingo@kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Eric Dumazet <edumazet@google.com>
Subject: Re: [PATCH 5/6] net: __this_cpu_inc in route.c
Date: Wed, 16 Oct 2013 12:25:57 +0200	[thread overview]
Message-ID: <20131016102557.GV10651@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <1381915369.2045.118.camel@edumazet-glaptop.roam.corp.google.com>

On Wed, Oct 16, 2013 at 02:22:49AM -0700, Eric Dumazet wrote:
> On Wed, 2013-10-16 at 10:46 +0200, Peter Zijlstra wrote:
> 
> > Are we sure all !x86 implementations will DTRT in that it will increment
> > some CPU and not get horribly confused? I suppose it would; but is
> > that a guarantee given someplace?
> 
> I think we should be fine, these are only stats exposed
> on /proc/net/stat/rt_cache

My concern was really that an unprotected __this_cpu op would complete
coherent and not corrupt state. If we cannot guarantee this it should
always be a full BUG to use it without proper protection.

For x86 its fairly easy to see its correct this way; but for load-store
archs this isn't immediately obvious to me.

Suppose; r1 contains our per-cpu pointer:

	LOAD  r2, per-cpu-base
	ADD   r1, r2
	LOAD  r2, $(r1) # load value
	INC   r2
	STORE $(r1), r2 # store value

If such a thing is done without preempt disable; we could be
preempted/migrated at any place along that chain. In that case the STORE
could be to another CPUs memory (we get migrated near the INC) and could
conflict with a per-cpu operation on that CPU corrupting state.

If I look at percpu.h; the generic __this_cpu versions look like they
generate the above for such archs.

In that case; I don't see how even for statistics (where we really don't
care what cpu the op happens on, as long as it happens to a cpu,
coherently) it is correct to use the raw_this_cpu stuff without
preemption protection.

In fact; I think the comment near __this_cpu_read actually alludes to
this.

  reply	other threads:[~2013-10-16 10:26 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-15 17:47 [PATCH 0/6] percpu: Implement Preemption checks for __this_cpu operations V4b Christoph Lameter
2013-10-15 17:47 ` [PATCH 1/6] net: ip4_datagram_connect: Use correct form of statistics update Christoph Lameter
2013-10-15 18:36   ` Eric Dumazet
2013-10-16  6:09     ` Ingo Molnar
2013-10-16  8:35   ` Peter Zijlstra
2013-10-16  9:14     ` Eric Dumazet
2013-10-16  9:26       ` Ingo Molnar
2013-10-16 14:27         ` Christoph Lameter
2013-10-16 14:37           ` Eric Dumazet
2013-10-15 17:47 ` [PATCH 2/6] percpu: Add raw_cpu_ops Christoph Lameter
2013-10-15 17:47 ` [PATCH 3/6] mm: Use raw_cpu ops for determining current NUMA node Christoph Lameter
2013-10-16  8:38   ` Peter Zijlstra
2013-10-16 14:22     ` Christoph Lameter
2013-10-15 17:47 ` [PATCH 4/6] Use raw_cpu_write for initialization of per cpu refcount Christoph Lameter
2013-10-16  8:43   ` Peter Zijlstra
2013-10-15 17:47 ` [PATCH 5/6] net: __this_cpu_inc in route.c Christoph Lameter
2013-10-16  8:46   ` Peter Zijlstra
2013-10-16  9:22     ` Eric Dumazet
2013-10-16 10:25       ` Peter Zijlstra [this message]
2013-10-16 15:07         ` Christoph Lameter
2013-10-15 17:47 ` [PATCH 6/6] percpu: Add preemption checks to __this_cpu ops Christoph Lameter
2013-10-16  8:49   ` Peter Zijlstra
2013-10-16 15:09     ` Christoph Lameter
2013-10-16 15:36       ` Peter Zijlstra
2013-10-16 15:55         ` Christoph Lameter
2013-10-16 16:25           ` Peter Zijlstra
2013-10-16 16:52             ` Steven Rostedt
2013-10-16 17:11               ` Peter Zijlstra
2013-10-16 17:39                 ` Steven Rostedt
2013-10-16 18:38             ` Peter Zijlstra
2013-10-17 19:22               ` Christoph Lameter
2013-10-17 21:13                 ` Peter Zijlstra
     [not found] <20131011175518.634285474@linux.com>
2013-10-11 17:54 ` [PATCH 5/6] net: __this_cpu_inc in route.c Christoph Lameter

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=20131016102557.GV10651@twins.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=akpm@linuxfoundation.org \
    --cc=cl@linux.com \
    --cc=edumazet@google.com \
    --cc=eric.dumazet@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=rostedt@goodmis.org \
    --cc=tglx@linutronix.de \
    --cc=tj@kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®