public inbox for linux-arch@vger.kernel.org
 help / color / mirror / Atom feed
From: Benjamin Herrenschmidt <benh-XVmvHMARGAS8U2dJNN8I7kB+6BGkLq7r@public.gmane.org>
To: Hugh Dickins <hugh-DTz5qymZ9yRBDgjK7y7TUQ@public.gmane.org>
Cc: Nick Piggin <npiggin-l3A5Bk7waGM@public.gmane.org>,
	Linus Torvalds
	<torvalds-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>,
	linux-arch-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Linux Memory Management List
	<linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org>
Subject: Re: [rfc] data race in page table setup/walking?
Date: Wed, 30 Apr 2008 07:37:39 +1000	[thread overview]
Message-ID: <1209505059.18023.193.camel@pasglop> (raw)
In-Reply-To: <Pine.LNX.4.64.0804291333540.22025-popGQ1T0qN76K7/ahGyk6A@public.gmane.org>


On Tue, 2008-04-29 at 13:36 +0100, Hugh Dickins wrote:
> 
> Ugh.  It's just so irritating to introduce these blockages against
> such a remote possibility (but there again, that's what so much of
> kernel code has to be about).  Is there any other way of handling it?

Not that much overhead... I think smp_read_barrier_depends() is a nop on
most archs no ? The data dependency between all the pointers takes care
of ordering in many cases. So it boils down to smp_wmb's when setting
which is not that expensive.

Cheers,
Ben.

WARNING: multiple messages have this Message-ID (diff)
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <npiggin@suse.de>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	linux-arch@vger.kernel.org,
	Linux Memory Management List <linux-mm@kvack.org>
Subject: Re: [rfc] data race in page table setup/walking?
Date: Wed, 30 Apr 2008 07:37:39 +1000	[thread overview]
Message-ID: <1209505059.18023.193.camel@pasglop> (raw)
Message-ID: <20080429213739.kRUOkpL3C3ASbJr2zcujJGmwsMq3x4PxckQcSsIx0_0@z> (raw)
In-Reply-To: <Pine.LNX.4.64.0804291333540.22025@blonde.site>


On Tue, 2008-04-29 at 13:36 +0100, Hugh Dickins wrote:
> 
> Ugh.  It's just so irritating to introduce these blockages against
> such a remote possibility (but there again, that's what so much of
> kernel code has to be about).  Is there any other way of handling it?

Not that much overhead... I think smp_read_barrier_depends() is a nop on
most archs no ? The data dependency between all the pointers takes care
of ordering in many cases. So it boils down to smp_wmb's when setting
which is not that expensive.

Cheers,
Ben.



  parent reply	other threads:[~2008-04-29 21:37 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-04-29  5:00 [rfc] data race in page table setup/walking? Nick Piggin
2008-04-29  5:00 ` Nick Piggin
     [not found] ` <20080429050054.GC21795-B4tOwbsTzaBolqkO4TVVkw@public.gmane.org>
2008-04-29  5:08   ` Benjamin Herrenschmidt
2008-04-29  5:08     ` Benjamin Herrenschmidt
2008-04-29  5:41     ` Nick Piggin
2008-04-29  5:41       ` Nick Piggin
2008-04-29 10:56   ` David Miller
2008-04-29 10:56     ` David Miller
2008-04-29 12:36   ` Hugh Dickins
2008-04-29 12:36     ` Hugh Dickins
     [not found]     ` <Pine.LNX.4.64.0804291333540.22025-popGQ1T0qN76K7/ahGyk6A@public.gmane.org>
2008-04-29 21:37       ` Benjamin Herrenschmidt [this message]
2008-04-29 21:37         ` Benjamin Herrenschmidt
2008-04-29 22:47         ` Hugh Dickins
2008-04-29 22:47           ` Hugh Dickins
     [not found]           ` <Pine.LNX.4.64.0804292328110.23470-popGQ1T0qN76K7/ahGyk6A@public.gmane.org>
2008-04-30  0:09             ` Benjamin Herrenschmidt
2008-04-30  0:09               ` Benjamin Herrenschmidt
2008-04-30  6:03       ` Nick Piggin
2008-04-30  6:03         ` Nick Piggin
     [not found]         ` <20080430060340.GE27652-B4tOwbsTzaBolqkO4TVVkw@public.gmane.org>
2008-04-30  6:05           ` David Miller
2008-04-30  6:05             ` David Miller
     [not found]             ` <20080429.230543.98200575.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2008-04-30  6:17               ` Nick Piggin
2008-04-30  6:17                 ` Nick Piggin
2008-04-30 11:14           ` Hugh Dickins
2008-04-30 11:14             ` Hugh Dickins
     [not found]             ` <Pine.LNX.4.64.0804301140490.4651-popGQ1T0qN76K7/ahGyk6A@public.gmane.org>
2008-05-01  0:35               ` Nick Piggin
2008-05-01  0:35                 ` Nick Piggin
     [not found]                 ` <20080501003542.GB11312-B4tOwbsTzaBolqkO4TVVkw@public.gmane.org>
2008-05-01 12:45                   ` Hugh Dickins
2008-05-01 12:45                     ` Hugh Dickins
2008-04-30 15:53           ` Linus Torvalds
2008-04-30 15:53             ` Linus Torvalds
     [not found]             ` <alpine.LFD.1.10.0804300848390.2997-5CScLwifNT1QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org>
2008-05-01  0:29               ` Nick Piggin
2008-05-01  0:29                 ` Nick Piggin
     [not found]                 ` <20080501002955.GA11312-B4tOwbsTzaBolqkO4TVVkw@public.gmane.org>
2008-05-01  3:24                   ` Linus Torvalds
2008-05-01  3:24                     ` Linus Torvalds
2008-05-02  1:20                     ` Nick Piggin
2008-05-02  1:33                       ` Linus Torvalds
2008-05-02  1:43                         ` Nick Piggin

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=1209505059.18023.193.camel@pasglop \
    --to=benh-xvmvhmargas8u2djnn8i7kb+6bgklq7r@public.gmane.org \
    --cc=hugh-DTz5qymZ9yRBDgjK7y7TUQ@public.gmane.org \
    --cc=linux-arch-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
    --cc=npiggin-l3A5Bk7waGM@public.gmane.org \
    --cc=torvalds-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.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