Netdev List
 help / color / mirror / Atom feed
From: Kory Maincent <kory.maincent@bootlin.com>
To: Paul Barker <paul.barker.ct@bp.renesas.com>
Cc: "Niklas Söderlund" <niklas.soderlund@ragnatech.se>,
	"Andrew Lunn" <andrew+netdev@lunn.ch>,
	"David S. Miller" <davem@davemloft.net>,
	"Eric Dumazet" <edumazet@google.com>,
	"Jakub Kicinski" <kuba@kernel.org>,
	"Paolo Abeni" <pabeni@redhat.com>,
	"Mikhail Ulyanov" <mikhail.ulyanov@cogentembedded.com>,
	"Sergei Shtylyov" <sergei.shtylyov@cogentembedded.com>,
	"Thomas Petazzoni" <thomas.petazzoni@bootlin.com>,
	"Niklas Söderlund" <niklas.soderlund+renesas@ragnatech.se>,
	"Claudiu Beznea" <claudiu.beznea.uj@bp.renesas.com>,
	netdev@vger.kernel.org, linux-renesas-soc@vger.kernel.org,
	linux-kernel@vger.kernel.org,
	"Sergey Shtylyov" <s.shtylyov@omp.ru>
Subject: Re: [PATCH net v2 1/2] net: ravb: Fix missing rtnl lock in suspend path
Date: Mon, 27 Jan 2025 11:28:50 +0100	[thread overview]
Message-ID: <20250127112850.05d7769b@kmaincent-XPS-13-7390> (raw)
In-Reply-To: <20250123183358.502e8032@kmaincent-XPS-13-7390>

On Thu, 23 Jan 2025 18:33:58 +0100
Kory Maincent <kory.maincent@bootlin.com> wrote:
 
> > >  
> > > @@ -3247,7 +3253,9 @@ static int ravb_resume(struct device *dev)
> > >  
> > >  	/* If WoL is enabled restore the interface. */
> > >  	if (priv->wol_enabled) {
> > > +		rtnl_lock();
> > >  		ret = ravb_wol_restore(ndev);
> > > +		rtnl_unlock();
> > >  		if (ret)
> > >  			return ret;
> > >  	} else {
> > > @@ -3257,7 +3265,9 @@ static int ravb_resume(struct device *dev)
> > >  	}
> > >  
> > >  	/* Reopening the interface will restore the device to the working
> > > state. */
> > > +	rtnl_lock();
> > >  	ret = ravb_open(ndev);
> > > +	rtnl_unlock();
> > >  	if (ret < 0)
> > >  		goto out_rpm_put;
>
> > I don't like the multiple lock/unlock calls in each function. I think v1
> > was better, where we take the lock once in each function and then unlock
> > when it is no longer needed or when we're about to return.
> 
> You will need to achieve a consensus on it with Claudiu. His point of view has
> that the locking scheme looks complicated.
> 
> On my side I don't have really an opinion, maybe a small preference for v1
> which is protecting wol_enabled flag even if it is not needed.

Claudiu any remarks?

If not I will come back to the first version as asked by Paul who is the
Maintainer of the ravb driver.

Sergey have asked to remove the duplicate of the if condition.
Paul is this ok for you?

@@ -3245,19 +3250,21 @@ static int ravb_resume(struct device *dev)
        if (!netif_running(ndev))
                return 0;
 
+       rtnl_lock();
        /* If WoL is enabled restore the interface. */
-       if (priv->wol_enabled) {
+       if (priv->wol_enabled)
                ret = ravb_wol_restore(ndev);
-               if (ret)
-                       return ret;
-       } else {
+       else
                ret = pm_runtime_force_resume(dev);
-               if (ret)
-                       return ret;
+
+       if (ret) {
+               rtnl_unlock();
+               return ret;
        }
 
        /* Reopening the interface will restore the device to the working
state. */
        ret = ravb_open(ndev);
+       rtnl_unlock();
        if (ret < 0)
                goto out_rpm_put;


Note: Sergey, I have received your mail as spam. 

Regards,
-- 
Köry Maincent, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com

  reply	other threads:[~2025-01-27 10:29 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-23 16:57 [PATCH net v2 0/2] Fix missing rtnl lock in suspend path Kory Maincent
2025-01-23 16:58 ` [PATCH net v2 1/2] net: ravb: " Kory Maincent
2025-01-23 17:23   ` Paul Barker
2025-01-23 17:33     ` Kory Maincent
2025-01-27 10:28       ` Kory Maincent [this message]
2025-01-29 14:36         ` Claudiu Beznea
2025-01-23 16:58 ` [PATCH net v2 2/2] net: sh_eth: " Kory Maincent

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=20250127112850.05d7769b@kmaincent-XPS-13-7390 \
    --to=kory.maincent@bootlin.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=claudiu.beznea.uj@bp.renesas.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-renesas-soc@vger.kernel.org \
    --cc=mikhail.ulyanov@cogentembedded.com \
    --cc=netdev@vger.kernel.org \
    --cc=niklas.soderlund+renesas@ragnatech.se \
    --cc=niklas.soderlund@ragnatech.se \
    --cc=pabeni@redhat.com \
    --cc=paul.barker.ct@bp.renesas.com \
    --cc=s.shtylyov@omp.ru \
    --cc=sergei.shtylyov@cogentembedded.com \
    --cc=thomas.petazzoni@bootlin.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox