All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nick Piggin <npiggin@suse.de>
To: Miao Xie <miaox@cn.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>,
	Lee Schermerhorn <lee.schermerhorn@hp.com>,
	Paul Menage <menage@google.com>,
	Linux-Kernel <linux-kernel@vger.kernel.org>,
	Linux-MM <linux-mm@kvack.org>
Subject: Re: [PATCH 4/4] cpuset,mm: use rwlock to protect task->mempolicy and mems_allowed
Date: Thu, 4 Mar 2010 14:30:17 +1100	[thread overview]
Message-ID: <20100304033017.GN8653@laptop> (raw)
In-Reply-To: <4B8E3F77.6070201@cn.fujitsu.com>

On Wed, Mar 03, 2010 at 06:52:39PM +0800, Miao Xie wrote:
> if MAX_NUMNODES > BITS_PER_LONG, loading/storing task->mems_allowed or mems_allowed in
> task->mempolicy are not atomic operations, and the kernel page allocator gets an empty
> mems_allowed when updating task->mems_allowed or mems_allowed in task->mempolicy. So we
> use a rwlock to protect them to fix this probelm.

Thanks for working on this. However, rwlocks are pretty nasty to use
when you have short critical sections and hot read-side (they're twice
as heavy as even spinlocks in that case).

It's being used in the page allocator path, so I would say rwlocks are
almost a showstopper. Wouldn't it be possible to use a seqlock for this?

WARNING: multiple messages have this Message-ID (diff)
From: Nick Piggin <npiggin@suse.de>
To: Miao Xie <miaox@cn.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>,
	Lee Schermerhorn <lee.schermerhorn@hp.com>,
	Paul Menage <menage@google.com>,
	Linux-Kernel <linux-kernel@vger.kernel.org>,
	Linux-MM <linux-mm@kvack.org>
Subject: Re: [PATCH 4/4] cpuset,mm: use rwlock to protect task->mempolicy and mems_allowed
Date: Thu, 4 Mar 2010 14:30:17 +1100	[thread overview]
Message-ID: <20100304033017.GN8653@laptop> (raw)
In-Reply-To: <4B8E3F77.6070201@cn.fujitsu.com>

On Wed, Mar 03, 2010 at 06:52:39PM +0800, Miao Xie wrote:
> if MAX_NUMNODES > BITS_PER_LONG, loading/storing task->mems_allowed or mems_allowed in
> task->mempolicy are not atomic operations, and the kernel page allocator gets an empty
> mems_allowed when updating task->mems_allowed or mems_allowed in task->mempolicy. So we
> use a rwlock to protect them to fix this probelm.

Thanks for working on this. However, rwlocks are pretty nasty to use
when you have short critical sections and hot read-side (they're twice
as heavy as even spinlocks in that case).

It's being used in the page allocator path, so I would say rwlocks are
almost a showstopper. Wouldn't it be possible to use a seqlock for this?

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2010-03-04  3:30 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-03-03 10:52 [PATCH 4/4] cpuset,mm: use rwlock to protect task->mempolicy and mems_allowed Miao Xie
2010-03-03 10:52 ` Miao Xie
2010-03-03 23:50 ` Andrew Morton
2010-03-03 23:50   ` Andrew Morton
2010-03-04  9:03   ` Miao Xie
2010-03-04  9:03     ` Miao Xie
2010-03-04  3:30 ` Nick Piggin [this message]
2010-03-04  3:30   ` Nick Piggin
2010-03-04  9:36   ` Miao Xie
2010-03-04  9:36     ` Miao Xie
2010-03-04 14:58   ` Peter Zijlstra
2010-03-04 14:58     ` Peter Zijlstra
2010-03-04 16:34     ` Nick Piggin
2010-03-04 16:34       ` Nick Piggin
2010-03-04  4:53 ` Nick Piggin
2010-03-04  4:53   ` Nick Piggin
2010-03-04 14:31 ` Lee Schermerhorn
2010-03-04 14:31   ` Lee Schermerhorn
2010-03-05 13:05   ` Lee Schermerhorn
2010-03-05 13:05     ` Lee Schermerhorn
2010-03-05 12:03 ` Paul Menage
2010-03-05 12:03   ` Paul Menage
2010-03-07  2:33   ` Miao Xie
2010-03-07  2:33     ` Miao Xie
2010-03-09 19:42     ` Paul Menage
2010-03-09 19:42       ` Paul Menage
2010-03-11  5:04       ` Miao Xie
2010-03-11  5:04         ` Miao Xie
2010-03-11  5:30         ` Nick Piggin
2010-03-11  5:30           ` Nick Piggin
2010-03-11  7:57           ` Miao Xie
2010-03-11  7:57             ` Miao Xie

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=20100304033017.GN8653@laptop \
    --to=npiggin@suse.de \
    --cc=lee.schermerhorn@hp.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=menage@google.com \
    --cc=miaox@cn.fujitsu.com \
    --cc=rientjes@google.com \
    /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.