public inbox for linux-usb@vger.kernel.org
 help / color / mirror / Atom feed
From: Jakub Kicinski <kuba@kernel.org>
To: Oliver Neukum <oneukum@suse.com>
Cc: Paolo Abeni <pabeni@redhat.com>,
	"Peter GJ. Park" <gyujoon.park@samsung.com>,
	Andrew Lunn <andrew+netdev@lunn.ch>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	netdev@vger.kernel.org, linux-usb@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH] net: usb: usbnet: fix use-after-free in race on workqueue
Date: Tue, 1 Jul 2025 18:26:17 -0700	[thread overview]
Message-ID: <20250701182617.07d6e437@kernel.org> (raw)
In-Reply-To: <ebd0bb9b-8e66-4119-b011-c1a737749fb2@suse.com>

On Tue, 1 Jul 2025 15:22:54 +0200 Oliver Neukum wrote:
> On 26.06.25 11:21, Paolo Abeni wrote:
> >>   drivers/net/usb/usbnet.c | 3 +++
> >>   1 file changed, 3 insertions(+)
> >>
> >> diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
> >> index c04e715a4c2ade3bc5587b0df71643a25cf88c55..3c5d9ba7fa6660273137c80106746103f84f5a37 100644
> >> --- a/drivers/net/usb/usbnet.c
> >> +++ b/drivers/net/usb/usbnet.c
> >> @@ -1660,6 +1660,9 @@ void usbnet_disconnect (struct usb_interface *intf)
> >>   	usb_free_urb(dev->interrupt);
> >>   	kfree(dev->padding_pkt);
> >>   
> >> +	timer_delete_sync(&dev->delay);
> >> +	tasklet_kill(&dev->bh);
> >> +	cancel_work_sync(&dev->kevent);
> >>   	free_netdev(net);  
> > This happens after unregister_netdev(), which calls usbnet_stop() that
> > already performs the above cleanup. How the race is supposed to take place?  
> 
> That is indeed a core question, which we really need an answer to.
> Do I interpret dev_close_many() correctly, if I state that the
> ndo_stop() method will _not_ be called if the device has never been
> opened?

Correct, open and close are paired. Most drivers would crash if we
tried to close them before they ever got opened. 

> I am sorry to be a stickler here, but if that turns out to be true,
> usbnet is not the only driver that has this bug.

Shooting from the hip slightly, but its unusual for a driver to start
link monitoring before open. After all there can be no packets on a
device that's closed. Why not something like:

diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index 9564478a79cc..b75b0b5c3abc 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -896,6 +896,9 @@ int usbnet_open (struct net_device *net)
        int                     retval;
        const struct driver_info *info = dev->driver_info;
 
+       if (dev->driver_info->flags & FLAG_LINK_INTR)
+               usbnet_link_change(dev, 0, 0);
+
        if ((retval = usb_autopm_get_interface(dev->intf)) < 0) {
                netif_info(dev, ifup, dev->net,
                           "resumption fail (%d) usbnet usb-%s-%s, %s\n",
@@ -1862,8 +1865,7 @@ usbnet_probe (struct usb_interface *udev, const struct usb_device_id *prod)
 
        netif_device_attach (net);
 
-       if (dev->driver_info->flags & FLAG_LINK_INTR)
-               usbnet_link_change(dev, 0, 0);
+       netif_carrier_off(net);
 
        return 0;
 

  reply	other threads:[~2025-07-02  1:26 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20250625093354epcas1p1c9817df6e1d1599e8b4eb16c5715a6fd@epcas1p1.samsung.com>
2025-06-25  9:33 ` [PATCH] net: usb: usbnet: fix use-after-free in race on workqueue Peter GJ. Park
2025-06-26  9:21   ` Paolo Abeni
2025-06-27 10:47     ` [PATCH v2] " Peter GJ. Park
2025-06-27 10:59     ` [PATCH net " Peter GJ. Park
2025-07-02  1:27       ` Jakub Kicinski
2025-07-01  4:19     ` [PATCH net] " Peter GJ. Park
2025-07-01 13:22     ` [PATCH] " Oliver Neukum
2025-07-02  1:26       ` Jakub Kicinski [this message]
2025-07-02 10:54         ` Oliver Neukum
2025-07-02 18:22           ` 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=20250701182617.07d6e437@kernel.org \
    --to=kuba@kernel.org \
    --cc=andrew+netdev@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=gyujoon.park@samsung.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=oneukum@suse.com \
    --cc=pabeni@redhat.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