netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Kuniyuki Iwashima <kuniyu@amazon.com>
To: <i.abramov@mt-integration.ru>
Cc: <davem@davemloft.net>, <kuba@kernel.org>,
	<linux-kernel@vger.kernel.org>, <lvc-project@linuxtesting.org>,
	<netdev@vger.kernel.org>, <kuniyu@amazon.com>
Subject: Re: [PATCH net 1/4] ieee802154: Restore initial state on failed device_rename() in cfg802154_switch_netns()
Date: Tue, 25 Mar 2025 15:00:47 -0700	[thread overview]
Message-ID: <20250325220115.67524-1-kuniyu@amazon.com> (raw)
In-Reply-To: <20250325141723.499850-2-i.abramov@mt-integration.ru>

From: Ivan Abramov <i.abramov@mt-integration.ru>
Date: Tue, 25 Mar 2025 17:17:20 +0300
> Currently, the return value of device_rename() is not checked or acted
> upon. There is also a pointless WARN_ON() call in case of an allocation
> failure, since it only leads to useless splats caused by deliberate fault
> injections.
> 
> Since it's possible to roll back the changes made before the
> device_rename() call in case of failure, do it.
> 
> Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
> 
> Fixes: 66e5c2672cd1 ("ieee802154: add netns support")
> Signed-off-by: Ivan Abramov <i.abramov@mt-integration.ru>
> ---
>  net/ieee802154/core.c | 44 +++++++++++++++++++++++--------------------
>  1 file changed, 24 insertions(+), 20 deletions(-)
> 
> diff --git a/net/ieee802154/core.c b/net/ieee802154/core.c
> index 88adb04e4072..f9865eb2c7cf 100644
> --- a/net/ieee802154/core.c
> +++ b/net/ieee802154/core.c
> @@ -233,31 +233,35 @@ int cfg802154_switch_netns(struct cfg802154_registered_device *rdev,
>  		wpan_dev->netdev->netns_local = true;
>  	}
>  
> -	if (err) {
> -		/* failed -- clean up to old netns */
> -		net = wpan_phy_net(&rdev->wpan_phy);
> -
> -		list_for_each_entry_continue_reverse(wpan_dev,
> -						     &rdev->wpan_dev_list,
> -						     list) {
> -			if (!wpan_dev->netdev)
> -				continue;
> -			wpan_dev->netdev->netns_local = false;
> -			err = dev_change_net_namespace(wpan_dev->netdev, net,
> -						       "wpan%d");
> -			WARN_ON(err);
> -			wpan_dev->netdev->netns_local = true;
> -		}
> +	if (err)
> +		goto errout;
>  
> -		return err;
> -	}
> +	err = device_rename(&rdev->wpan_phy.dev, dev_name(&rdev->wpan_phy.dev));
>  
> -	wpan_phy_net_set(&rdev->wpan_phy, net);
> +	if (err)
> +		goto errout;
>  
> -	err = device_rename(&rdev->wpan_phy.dev, dev_name(&rdev->wpan_phy.dev));
> -	WARN_ON(err);
> +	wpan_phy_net_set(&rdev->wpan_phy, net);
>  
>  	return 0;
> +
> +errout:
> +	/* failed -- clean up to old netns */
> +	net = wpan_phy_net(&rdev->wpan_phy);
> +
> +	list_for_each_entry_continue_reverse(wpan_dev,
> +					     &rdev->wpan_dev_list,
> +					     list) {
> +		if (!wpan_dev->netdev)
> +			continue;
> +		wpan_dev->netdev->netns_local = false;
> +		err = dev_change_net_namespace(wpan_dev->netdev, net,
> +					       "wpan%d");
> +		WARN_ON(err);

It's still possible to trigger this with -ENOMEM.

For example, see bitmap_zalloc() in __dev_alloc_name().

Perhaps simply use pr_warn() or net_warn_ratelimited() as do_setlink().

I guess the stack trace from here is not so interesting as it doens't
show where it actually failed.


> +		wpan_dev->netdev->netns_local = true;
> +	}
> +
> +	return err;
>  }
>  
>  void cfg802154_dev_free(struct cfg802154_registered_device *rdev)
> -- 
> 2.39.5

  reply	other threads:[~2025-03-25 22:01 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-03-25 14:17 [PATCH net 0/4] Avoid using WARN_ON() on allocation failure in device_rename() Ivan Abramov
2025-03-25 14:17 ` [PATCH net 1/4] ieee802154: Restore initial state on failed device_rename() in cfg802154_switch_netns() Ivan Abramov
2025-03-25 22:00   ` Kuniyuki Iwashima [this message]
2025-03-25 14:17 ` [PATCH net 2/4] ieee802154: Avoid calling WARN_ON() on -ENOMEM in cfg802154_pernet_exit() Ivan Abramov
2025-03-25 14:17 ` [PATCH net 3/4] cfg80211: Avoid calling WARN_ON() on -ENOMEM in cfg80211_switch_netns() Ivan Abramov
2025-03-25 21:39 ` [PATCH net 0/4] Avoid using WARN_ON() on allocation failure in device_rename() Jakub Kicinski

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=20250325220115.67524-1-kuniyu@amazon.com \
    --to=kuniyu@amazon.com \
    --cc=davem@davemloft.net \
    --cc=i.abramov@mt-integration.ru \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lvc-project@linuxtesting.org \
    --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).