netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: Jay Vosburgh <fubar@us.ibm.com>
Cc: netdev@vger.kernel.org, jgarzik@pobox.com, davem@davemloft.net,
	andy@greyhouse.net, fubar@us.ibm.com
Subject: Re: [PATCH 6/7] bonding: fix lock ordering for rtnl and bonding_rwsem
Date: Thu, 17 Jan 2008 16:38:01 -0800	[thread overview]
Message-ID: <20080117163801.77c62170.akpm@linux-foundation.org> (raw)
In-Reply-To: <12006159123054-git-send-email-fubar@us.ibm.com>

On Thu, 17 Jan 2008 16:25:02 -0800
Jay Vosburgh <fubar@us.ibm.com> wrote:

> Fix the handling of rtnl and the bonding_rwsem to always be acquired
> in a consistent order (rtnl, then bonding_rwsem).
> 
> The existing code sometimes acquired them in this order, and sometimes
> in the opposite order, which opens a window for deadlock between ifenslave
> and sysfs.
> 
> ...
>
>  int bond_create(char *name, struct bond_params *params, struct bonding **newbond)
>  {
>  	struct net_device *bond_dev;
> +	struct bonding *bond, *nxt;
>  	int res;
>  
>  	rtnl_lock();
> +	down_write(&bonding_rwsem);
> +
> +	/* Check to see if the bond already exists. */
> +	list_for_each_entry_safe(bond, nxt, &bond_dev_list, bond_list)

this could (should) use list_for_each_entry().

> +		if (strnicmp(bond->dev->name, name, IFNAMSIZ) == 0) {
> +			printk(KERN_ERR DRV_NAME
> +			       ": cannot add bond %s; it already exists\n",
> +			       name);
> +			res = -EPERM;
> +			goto out_rtnl;
> +		}
> +
>  	bond_dev = alloc_netdev(sizeof(struct bonding), name ? name : "",
>  				ether_setup);
>  	if (!bond_dev) {
> @@ -4915,10 +4928,12 @@ int bond_create(char *name, struct bond_params *params, struct bonding **newbond
>  
>  	netif_carrier_off(bond_dev);
>  
> +	up_write(&bonding_rwsem);
>  	rtnl_unlock(); /* allows sysfs registration of net device */
>  	res = bond_create_sysfs_entry(bond_dev->priv);
>  	if (res < 0) {
>  		rtnl_lock();
> +		down_write(&bonding_rwsem);
>  		goto out_bond;
>  	}
>  
> @@ -4929,6 +4944,7 @@ out_bond:
>  out_netdev:
>  	free_netdev(bond_dev);
>  out_rtnl:
> +	up_write(&bonding_rwsem);
>  	rtnl_unlock();
>  	return res;
>  }
> @@ -4949,6 +4965,9 @@ static int __init bonding_init(void)
>  #ifdef CONFIG_PROC_FS
>  	bond_create_proc_dir();
>  #endif
> +
> +	init_rwsem(&bonding_rwsem);

It would be better to initialise this at compile time with DECLARE_RWSEM().


But neither of those things need to be done for 2.6.24.


  parent reply	other threads:[~2008-01-18  0:38 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-01-18  0:24 [PATCH 0/7] bonding: 7 fixes for 2.6.24 Jay Vosburgh
2008-01-18  0:24 ` [PATCH 1/7] bonding: fix locking in sysfs primary/active selection Jay Vosburgh
2008-01-18  0:24   ` [PATCH 2/7] bonding: fix ASSERT_RTNL that produces spurious warnings Jay Vosburgh
2008-01-18  0:24     ` [PATCH 3/7] bonding: fix locking during alb failover and slave removal Jay Vosburgh
2008-01-18  0:25       ` [PATCH 4/7] bonding: release slaves when master removed via sysfs Jay Vosburgh
2008-01-18  0:25         ` [PATCH 5/7] bonding: Fix up parameter parsing Jay Vosburgh
2008-01-18  0:25           ` [PATCH 6/7] bonding: fix lock ordering for rtnl and bonding_rwsem Jay Vosburgh
2008-01-18  0:25             ` [PATCH 7/7] bonding: Don't hold lock when calling rtnl_unlock Jay Vosburgh
2008-01-18  0:38             ` Andrew Morton [this message]
2008-01-18 19:39   ` [PATCH 1/7] bonding: fix locking in sysfs primary/active selection Jeff Garzik
2008-01-21  1:15 ` [PATCH 0/7] bonding: 7 fixes for 2.6.24 David Miller
2008-01-21  1:36   ` Andrew Morton
2008-01-21  3:58     ` David Miller
2008-01-21  1:57   ` Jeff Garzik

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=20080117163801.77c62170.akpm@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=andy@greyhouse.net \
    --cc=davem@davemloft.net \
    --cc=fubar@us.ibm.com \
    --cc=jgarzik@pobox.com \
    --cc=netdev@vger.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;
as well as URLs for NNTP newsgroup(s).