[PATCH 5.15 0/1] Request to cherry-pick 74e7e1efdad4 to 5.15.y

[PATCH 5.15 0/1] Request to cherry-pick 74e7e1efdad4 to 5.15.y

[Meena Shanmugam]

The commit 74e7e1efdad4 (xen/netback: don't call kfree_skb() with
interrupts disabled) fixes deadlock in Linux netback driver. This seems
to be a good candidate for the stable trees. This patch didn't apply
cleanly in 5.15 kernel due to difference in function prototypes in
drivers/net/xen-netback/common.h.

Juergen Gross (1):
  xen/netback: don't call kfree_skb() with interrupts disabled

 drivers/net/xen-netback/common.h    | 2 +-
 drivers/net/xen-netback/interface.c | 6 ++++--
 drivers/net/xen-netback/rx.c        | 8 +++++---
 3 files changed, 10 insertions(+), 6 deletions(-)

-- 
2.39.0.rc1.256.g54fd8350bd-goog

[PATCH 5.15 1/1] xen/netback: don't call kfree_skb() with interrupts disabled

[Meena Shanmugam]

From: Juergen Gross <jgross@suse.com>

commit 74e7e1efdad45580cc3839f2a155174cf158f9b5 upstream.

It is not allowed to call kfree_skb() from hardware interrupt
context or with interrupts being disabled. So remove kfree_skb()
from the spin_lock_irqsave() section and use the already existing
"drop" label in xenvif_start_xmit() for dropping the SKB. At the
same time replace the dev_kfree_skb() call there with a call of
dev_kfree_skb_any(), as xenvif_start_xmit() can be called with
disabled interrupts.

This is XSA-424 / CVE-2022-42328 / CVE-2022-42329.

Fixes: be81992f9086 ("xen/netback: don't queue unlimited number of packages")
Reported-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Meena Shanmugam <meenashanmugam@google.com>
---
 drivers/net/xen-netback/common.h    | 2 +-
 drivers/net/xen-netback/interface.c | 6 ++++--
 drivers/net/xen-netback/rx.c        | 8 +++++---
 3 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index d9dea4829c86..bdb3139c7162 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -395,7 +395,7 @@ irqreturn_t xenvif_ctrl_irq_fn(int irq, void *data);
 
 bool xenvif_have_rx_work(struct xenvif_queue *queue, bool test_kthread);
 void xenvif_rx_action(struct xenvif_queue *queue);
-void xenvif_rx_queue_tail(struct xenvif_queue *queue, struct sk_buff *skb);
+bool xenvif_rx_queue_tail(struct xenvif_queue *queue, struct sk_buff *skb);
 
 void xenvif_carrier_on(struct xenvif *vif);
 
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index c58996c1e230..6a35772fde7a 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -269,14 +269,16 @@ xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (vif->hash.alg == XEN_NETIF_CTRL_HASH_ALGORITHM_NONE)
 		skb_clear_hash(skb);
 
-	xenvif_rx_queue_tail(queue, skb);
+	if (!xenvif_rx_queue_tail(queue, skb))
+		goto drop;
+
 	xenvif_kick_thread(queue);
 
 	return NETDEV_TX_OK;
 
  drop:
 	vif->dev->stats.tx_dropped++;
-	dev_kfree_skb(skb);
+	dev_kfree_skb_any(skb);
 	return NETDEV_TX_OK;
 }
 
diff --git a/drivers/net/xen-netback/rx.c b/drivers/net/xen-netback/rx.c
index a0335407be42..c2671eb6ad93 100644
--- a/drivers/net/xen-netback/rx.c
+++ b/drivers/net/xen-netback/rx.c
@@ -82,9 +82,10 @@ static bool xenvif_rx_ring_slots_available(struct xenvif_queue *queue)
 	return false;
 }
 
-void xenvif_rx_queue_tail(struct xenvif_queue *queue, struct sk_buff *skb)
+bool xenvif_rx_queue_tail(struct xenvif_queue *queue, struct sk_buff *skb)
 {
 	unsigned long flags;
+	bool ret = true;
 
 	spin_lock_irqsave(&queue->rx_queue.lock, flags);
 
@@ -92,8 +93,7 @@ void xenvif_rx_queue_tail(struct xenvif_queue *queue, struct sk_buff *skb)
 		struct net_device *dev = queue->vif->dev;
 
 		netif_tx_stop_queue(netdev_get_tx_queue(dev, queue->id));
-		kfree_skb(skb);
-		queue->vif->dev->stats.rx_dropped++;
+		ret = false;
 	} else {
 		if (skb_queue_empty(&queue->rx_queue))
 			xenvif_update_needed_slots(queue, skb);
@@ -104,6 +104,8 @@ void xenvif_rx_queue_tail(struct xenvif_queue *queue, struct sk_buff *skb)
 	}
 
 	spin_unlock_irqrestore(&queue->rx_queue.lock, flags);
+
+	return ret;
 }
 
 static struct sk_buff *xenvif_rx_dequeue(struct xenvif_queue *queue)
-- 
2.39.0.rc1.256.g54fd8350bd-goog

Re: [PATCH 5.15 0/1] Request to cherry-pick 74e7e1efdad4 to 5.15.y

[Juergen Gross]

[-- Attachment #1.1.1: Type: text/plain, Size: 720 bytes --]

On 13.12.22 22:53, Meena Shanmugam wrote:
> The commit 74e7e1efdad4 (xen/netback: don't call kfree_skb() with
> interrupts disabled) fixes deadlock in Linux netback driver. This seems
> to be a good candidate for the stable trees. This patch didn't apply
> cleanly in 5.15 kernel due to difference in function prototypes in
> drivers/net/xen-netback/common.h.
> 
> Juergen Gross (1):
>    xen/netback: don't call kfree_skb() with interrupts disabled
> 
>   drivers/net/xen-netback/common.h    | 2 +-
>   drivers/net/xen-netback/interface.c | 6 ++++--
>   drivers/net/xen-netback/rx.c        | 8 +++++---
>   3 files changed, 10 insertions(+), 6 deletions(-)
> 

It has already been picked.


Juergen

[-- Attachment #1.1.2: OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 3149 bytes --]

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 495 bytes --]

Re: [PATCH 5.15 0/1] Request to cherry-pick 74e7e1efdad4 to 5.15.y

[Greg KH]

On Tue, Dec 13, 2022 at 09:53:38PM +0000, Meena Shanmugam wrote:
> The commit 74e7e1efdad4 (xen/netback: don't call kfree_skb() with
> interrupts disabled) fixes deadlock in Linux netback driver. This seems
> to be a good candidate for the stable trees. This patch didn't apply
> cleanly in 5.15 kernel due to difference in function prototypes in
> drivers/net/xen-netback/common.h.
> 
> Juergen Gross (1):
>   xen/netback: don't call kfree_skb() with interrupts disabled
> 
>  drivers/net/xen-netback/common.h    | 2 +-
>  drivers/net/xen-netback/interface.c | 6 ++++--
>  drivers/net/xen-netback/rx.c        | 8 +++++---
>  3 files changed, 10 insertions(+), 6 deletions(-)
> 
> -- 
> 2.39.0.rc1.256.g54fd8350bd-goog
> 

Can you just test the latest stable -rc releases that were announced a
few days ago instead?  It has this commit in it, right?

thanks,

greg k-h

Re: [PATCH 5.15 0/1] Request to cherry-pick 74e7e1efdad4 to 5.15.y

[Meena Shanmugam]

On Wed, Dec 14, 2022 at 12:53 AM Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Tue, Dec 13, 2022 at 09:53:38PM +0000, Meena Shanmugam wrote:
> > The commit 74e7e1efdad4 (xen/netback: don't call kfree_skb() with
> > interrupts disabled) fixes deadlock in Linux netback driver. This seems
> > to be a good candidate for the stable trees. This patch didn't apply
> > cleanly in 5.15 kernel due to difference in function prototypes in
> > drivers/net/xen-netback/common.h.
> >
> > Juergen Gross (1):
> >   xen/netback: don't call kfree_skb() with interrupts disabled
> >
> >  drivers/net/xen-netback/common.h    | 2 +-
> >  drivers/net/xen-netback/interface.c | 6 ++++--
> >  drivers/net/xen-netback/rx.c        | 8 +++++---
> >  3 files changed, 10 insertions(+), 6 deletions(-)
> >
> > --
> > 2.39.0.rc1.256.g54fd8350bd-goog
> >
>
> Can you just test the latest stable -rc releases that were announced a
> few days ago instead?  It has this commit in it, right?
>
> thanks,
>
> greg k-h

Sorry, I was testing using 5.15.82 and I didn't realize that it was
already queued for 5.15.83.

Thanks,
Meena