[PATCH v5 net 0/2] udp: Fix two integer overflows when sk->sk_rcvbuf is close to INT_MAX.

[PATCH v5 net 0/2] udp: Fix two integer overflows when sk->sk_rcvbuf is close to INT_MAX.

[Kuniyuki Iwashima]

I got a report that UDP mem usage in /proc/net/sockstat did not
drop even after an application was terminated.

The issue could happen if sk->sk_rmem_alloc wraps around due
to a large sk->sk_rcvbuf, which was INT_MAX in our case.

The patch 2 fixes the issue, and the patch 1 fixes yet another
overflow I found while investigating the issue.


v5:
  * Drop flaky test
  * Patch 1
    * Update size after skb_condense()

v4:
  * Patch 3
    * Wait RCU for at most 30 sec

v3: https://lore.kernel.org/netdev/20250327202722.63756-1-kuniyu@amazon.com/
  * Rebase
  * Add Willem's tags

v2: https://lore.kernel.org/netdev/20250325195826.52385-1-kuniyu@amazon.com/
  * Patch 1
    * Define rmem and rcvbuf as unsigned int (Eric)
    * Take skb->truesize into account for sk with large rcvbuf (Willem)

  * Patch 3
    * Add a comment

v1: https://lore.kernel.org/netdev/20250323231016.74813-1-kuniyu@amazon.com/


Kuniyuki Iwashima (2):
  udp: Fix multiple wraparounds of sk->sk_rmem_alloc.
  udp: Fix memory accounting leak.

 net/ipv4/udp.c | 42 ++++++++++++++++++++++++------------------
 1 file changed, 24 insertions(+), 18 deletions(-)

-- 
2.48.1

[PATCH v5 net 1/2] udp: Fix multiple wraparounds of sk->sk_rmem_alloc.

[Kuniyuki Iwashima]

__udp_enqueue_schedule_skb() has the following condition:

  if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf)
          goto drop;

sk->sk_rcvbuf is initialised by net.core.rmem_default and later can
be configured by SO_RCVBUF, which is limited by net.core.rmem_max,
or SO_RCVBUFFORCE.

If we set INT_MAX to sk->sk_rcvbuf, the condition is always false
as sk->sk_rmem_alloc is also signed int.

Then, the size of the incoming skb is added to sk->sk_rmem_alloc
unconditionally.

This results in integer overflow (possibly multiple times) on
sk->sk_rmem_alloc and allows a single socket to have skb up to
net.core.udp_mem[1].

For example, if we set a large value to udp_mem[1] and INT_MAX to
sk->sk_rcvbuf and flood packets to the socket, we can see multiple
overflows:

  # cat /proc/net/sockstat | grep UDP:
  UDP: inuse 3 mem 7956736  <-- (7956736 << 12) bytes > INT_MAX * 15
                                             ^- PAGE_SHIFT
  # ss -uam
  State  Recv-Q      ...
  UNCONN -1757018048 ...    <-- flipping the sign repeatedly
         skmem:(r2537949248,rb2147483646,t0,tb212992,f1984,w0,o0,bl0,d0)

Previously, we had a boundary check for INT_MAX, which was removed by
commit 6a1f12dd85a8 ("udp: relax atomic operation on sk->sk_rmem_alloc").

A complete fix would be to revert it and cap the right operand by
INT_MAX:

  rmem = atomic_add_return(size, &sk->sk_rmem_alloc);
  if (rmem > min(size + (unsigned int)sk->sk_rcvbuf, INT_MAX))
          goto uncharge_drop;

but we do not want to add the expensive atomic_add_return() back just
for the corner case.

Casting rmem to unsigned int prevents multiple wraparounds, but we still
allow a single wraparound.

  # cat /proc/net/sockstat | grep UDP:
  UDP: inuse 3 mem 524288  <-- (INT_MAX + 1) >> 12

  # ss -uam
  State  Recv-Q      ...
  UNCONN -2147482816 ...   <-- INT_MAX + 831 bytes
         skmem:(r2147484480,rb2147483646,t0,tb212992,f3264,w0,o0,bl0,d14468947)

So, let's define rmem and rcvbuf as unsigned int and check skb->truesize
only when rcvbuf is large enough to lower the overflow possibility.

Note that we still have a small chance to see overflow if multiple skbs
to the same socket are processed on different core at the same time and
each size does not exceed the limit but the total size does.

Note also that we must ignore skb->truesize for a small buffer as
explained in commit 363dc73acacb ("udp: be less conservative with
sock rmem accounting").

Fixes: 6a1f12dd85a8 ("udp: relax atomic operation on sk->sk_rmem_alloc")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
v5:
  * Update size after skb_condense()

v2:
  * Define rmem and rcvbuf as unsigned int
  * Take skb->truesize into account for the large rcvbuf case
---
 net/ipv4/udp.c | 26 +++++++++++++++++---------
 1 file changed, 17 insertions(+), 9 deletions(-)

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index d0bffcfa56d8..354779b22faf 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1725,17 +1725,25 @@ static int udp_rmem_schedule(struct sock *sk, int size)
 int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
 {
 	struct sk_buff_head *list = &sk->sk_receive_queue;
-	int rmem, err = -ENOMEM;
+	unsigned int rmem, rcvbuf;
 	spinlock_t *busy = NULL;
-	int size, rcvbuf;
+	int size, err = -ENOMEM;
 
-	/* Immediately drop when the receive queue is full.
-	 * Always allow at least one packet.
-	 */
 	rmem = atomic_read(&sk->sk_rmem_alloc);
 	rcvbuf = READ_ONCE(sk->sk_rcvbuf);
-	if (rmem > rcvbuf)
-		goto drop;
+	size = skb->truesize;
+
+	/* Immediately drop when the receive queue is full.
+	 * Cast to unsigned int performs the boundary check for INT_MAX.
+	 */
+	if (rmem + size > rcvbuf) {
+		if (rcvbuf > INT_MAX >> 1)
+			goto drop;
+
+		/* Always allow at least one packet for small buffer. */
+		if (rmem > rcvbuf)
+			goto drop;
+	}
 
 	/* Under mem pressure, it might be helpful to help udp_recvmsg()
 	 * having linear skbs :
@@ -1745,10 +1753,10 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
 	 */
 	if (rmem > (rcvbuf >> 1)) {
 		skb_condense(skb);
-
+		size = skb->truesize;
 		busy = busylock_acquire(sk);
 	}
-	size = skb->truesize;
+
 	udp_set_dev_scratch(skb);
 
 	atomic_add(size, &sk->sk_rmem_alloc);
-- 
2.48.1

[PATCH v5 net 2/2] udp: Fix memory accounting leak.

[Kuniyuki Iwashima]

Matt Dowling reported a weird UDP memory usage issue.

Under normal operation, the UDP memory usage reported in /proc/net/sockstat
remains close to zero.  However, it occasionally spiked to 524,288 pages
and never dropped.  Moreover, the value doubled when the application was
terminated.  Finally, it caused intermittent packet drops.

We can reproduce the issue with the script below [0]:

  1. /proc/net/sockstat reports 0 pages

    # cat /proc/net/sockstat | grep UDP:
    UDP: inuse 1 mem 0

  2. Run the script till the report reaches 524,288

    # python3 test.py & sleep 5
    # cat /proc/net/sockstat | grep UDP:
    UDP: inuse 3 mem 524288  <-- (INT_MAX + 1) >> PAGE_SHIFT

  3. Kill the socket and confirm the number never drops

    # pkill python3 && sleep 5
    # cat /proc/net/sockstat | grep UDP:
    UDP: inuse 1 mem 524288

  4. (necessary since v6.0) Trigger proto_memory_pcpu_drain()

    # python3 test.py & sleep 1 && pkill python3

  5. The number doubles

    # cat /proc/net/sockstat | grep UDP:
    UDP: inuse 1 mem 1048577

The application set INT_MAX to SO_RCVBUF, which triggered an integer
overflow in udp_rmem_release().

When a socket is close()d, udp_destruct_common() purges its receive
queue and sums up skb->truesize in the queue.  This total is calculated
and stored in a local unsigned integer variable.

The total size is then passed to udp_rmem_release() to adjust memory
accounting.  However, because the function takes a signed integer
argument, the total size can wrap around, causing an overflow.

Then, the released amount is calculated as follows:

  1) Add size to sk->sk_forward_alloc.
  2) Round down sk->sk_forward_alloc to the nearest lower multiple of
      PAGE_SIZE and assign it to amount.
  3) Subtract amount from sk->sk_forward_alloc.
  4) Pass amount >> PAGE_SHIFT to __sk_mem_reduce_allocated().

When the issue occurred, the total in udp_destruct_common() was 2147484480
(INT_MAX + 833), which was cast to -2147482816 in udp_rmem_release().

At 1) sk->sk_forward_alloc is changed from 3264 to -2147479552, and
2) sets -2147479552 to amount.  3) reverts the wraparound, so we don't
see a warning in inet_sock_destruct().  However, udp_memory_allocated
ends up doubling at 4).

Since commit 3cd3399dd7a8 ("net: implement per-cpu reserves for
memory_allocated"), memory usage no longer doubles immediately after
a socket is close()d because __sk_mem_reduce_allocated() caches the
amount in udp_memory_per_cpu_fw_alloc.  However, the next time a UDP
socket receives a packet, the subtraction takes effect, causing UDP
memory usage to double.

This issue makes further memory allocation fail once the socket's
sk->sk_rmem_alloc exceeds net.ipv4.udp_rmem_min, resulting in packet
drops.

To prevent this issue, let's use unsigned int for the calculation and
call sk_forward_alloc_add() only once for the small delta.

Note that first_packet_length() also potentially has the same problem.

[0]:
from socket import *

SO_RCVBUFFORCE = 33
INT_MAX = (2 ** 31) - 1

s = socket(AF_INET, SOCK_DGRAM)
s.bind(('', 0))
s.setsockopt(SOL_SOCKET, SO_RCVBUFFORCE, INT_MAX)

c = socket(AF_INET, SOCK_DGRAM)
c.connect(s.getsockname())

data = b'a' * 100

while True:
    c.send(data)

Fixes: f970bd9e3a06 ("udp: implement memory accounting helpers")
Reported-by: Matt Dowling <madowlin@amazon.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
 net/ipv4/udp.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 354779b22faf..2742cc7602bb 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1625,12 +1625,12 @@ static bool udp_skb_has_head_state(struct sk_buff *skb)
 }
 
 /* fully reclaim rmem/fwd memory allocated for skb */
-static void udp_rmem_release(struct sock *sk, int size, int partial,
-			     bool rx_queue_lock_held)
+static void udp_rmem_release(struct sock *sk, unsigned int size,
+			     int partial, bool rx_queue_lock_held)
 {
 	struct udp_sock *up = udp_sk(sk);
 	struct sk_buff_head *sk_queue;
-	int amt;
+	unsigned int amt;
 
 	if (likely(partial)) {
 		up->forward_deficit += size;
@@ -1650,10 +1650,8 @@ static void udp_rmem_release(struct sock *sk, int size, int partial,
 	if (!rx_queue_lock_held)
 		spin_lock(&sk_queue->lock);
 
-
-	sk_forward_alloc_add(sk, size);
-	amt = (sk->sk_forward_alloc - partial) & ~(PAGE_SIZE - 1);
-	sk_forward_alloc_add(sk, -amt);
+	amt = (size + sk->sk_forward_alloc - partial) & ~(PAGE_SIZE - 1);
+	sk_forward_alloc_add(sk, size - amt);
 
 	if (amt)
 		__sk_mem_reduce_allocated(sk, amt >> PAGE_SHIFT);
@@ -1843,7 +1841,7 @@ EXPORT_IPV6_MOD_GPL(skb_consume_udp);
 
 static struct sk_buff *__first_packet_length(struct sock *sk,
 					     struct sk_buff_head *rcvq,
-					     int *total)
+					     unsigned int *total)
 {
 	struct sk_buff *skb;
 
@@ -1876,8 +1874,8 @@ static int first_packet_length(struct sock *sk)
 {
 	struct sk_buff_head *rcvq = &udp_sk(sk)->reader_queue;
 	struct sk_buff_head *sk_queue = &sk->sk_receive_queue;
+	unsigned int total = 0;
 	struct sk_buff *skb;
-	int total = 0;
 	int res;
 
 	spin_lock_bh(&rcvq->lock);
-- 
2.48.1

Re: [PATCH v5 net 0/2] udp: Fix two integer overflows when sk->sk_rcvbuf is close to INT_MAX.

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Tue, 1 Apr 2025 11:44:41 -0700 you wrote:
> I got a report that UDP mem usage in /proc/net/sockstat did not
> drop even after an application was terminated.
> 
> The issue could happen if sk->sk_rmem_alloc wraps around due
> to a large sk->sk_rcvbuf, which was INT_MAX in our case.
> 
> The patch 2 fixes the issue, and the patch 1 fixes yet another
> overflow I found while investigating the issue.
> 
> [...]

Here is the summary with links:
  - [v5,net,1/2] udp: Fix multiple wraparounds of sk->sk_rmem_alloc.
    https://git.kernel.org/netdev/net/c/5a465a0da13e
  - [v5,net,2/2] udp: Fix memory accounting leak.
    https://git.kernel.org/netdev/net/c/df207de9d9e7

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html