[PATCH net v2] net: xps: reject an out of range traffic class

[PATCH net v2] net: xps: reject an out of range traffic class

[Norbert Szetei]

Only the entries below dev->num_tc are valid in dev->tc_to_txq[], and
dev->prio_tc_map[] may only name classes below it. netdev_set_num_tc()
lowers dev->num_tc without touching either array.

netdev_txq_to_tc() walks all TC_MAX_QUEUE slots and
netdev_get_prio_tc_map() returns the entry as it stands, so a leftover
entry is handed out as a traffic class >= dev->num_tc. Taking that
class from netdev_txq_to_tc(), __netif_set_xps_queue() rejects only a
negative one and indexes an XPS map sized for dev->num_tc classes:

	tci = j * num_tc + tc;
	RCU_INIT_POINTER(new_dev_maps->attr_map[tci], map);

Any caller that lowers num_tc leaves such entries behind, and
mqprio_destroy() tears down with netdev_set_num_tc(dev, 0) rather than
netdev_reset_tc(). After mqprio with 8 classes then 1, tc_to_txq[1..7]
still describe txq 1..7. The splat is from an XPS write to txq 2:

  BUG: KASAN: slab-out-of-bounds in __netif_set_xps_queue+0x1eb9/0x2440
  Write of size 8 at addr ffff888110e978d8 by task xps_oob/573
   __netif_set_xps_queue+0x1eb9/0x2440
   xps_rxqs_store+0x24d/0x360
   netdev_queue_attr_store+0x61/0x90
  Allocated by task 573:
   __kmalloc_noprof+0x246/0x6c0
   __netif_set_xps_queue+0x8ca/0x2440
  The buggy address is located 0 bytes to the right of
   allocated 88-byte region [ffff888110e97880, ffff888110e978d8)

Reject a class the map has no room for.

Fixes: 184c449f91fe ("net: Add support for XPS with QoS via traffic classes")
Assisted-by: LLM
Signed-off-by: Norbert Szetei <norbert@doyensec.com>
---
v2:
  - bound the class in __netif_set_xps_queue() instead of clearing
    dev->tc_to_txq[]/dev->prio_tc_map[] in netdev_set_num_tc(), per the
    Sashiko review of v1
  - dropped the memory-ordering claim from the changelog
  - retitled
v1: https://lore.kernel.org/netdev/16E3A318-5532-4B5E-8D03-86D21B463A2D@doyensec.com/

 net/core/dev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index ecfbd72d5d1a..7e76860602e2 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2901,7 +2901,7 @@ int __netif_set_xps_queue(struct net_device *dev, const unsigned long *mask,
 		dev = netdev_get_tx_queue(dev, index)->sb_dev ? : dev;
 
 		tc = netdev_txq_to_tc(dev, index);
-		if (tc < 0)
+		if (tc < 0 || tc >= num_tc)
 			return -EINVAL;
 	}
 
-- 
2.55.0

Re: [PATCH net v2] net: xps: reject an out of range traffic class

[Simon Horman]

On Sat, Sep 12, 2026 at 08:01:06AM +0200, Norbert Szetei wrote:
> Only the entries below dev->num_tc are valid in dev->tc_to_txq[], and
> dev->prio_tc_map[] may only name classes below it. netdev_set_num_tc()
> lowers dev->num_tc without touching either array.
> 
> netdev_txq_to_tc() walks all TC_MAX_QUEUE slots and
> netdev_get_prio_tc_map() returns the entry as it stands, so a leftover
> entry is handed out as a traffic class >= dev->num_tc. Taking that
> class from netdev_txq_to_tc(), __netif_set_xps_queue() rejects only a
> negative one and indexes an XPS map sized for dev->num_tc classes:
> 
> 	tci = j * num_tc + tc;
> 	RCU_INIT_POINTER(new_dev_maps->attr_map[tci], map);
> 
> Any caller that lowers num_tc leaves such entries behind, and
> mqprio_destroy() tears down with netdev_set_num_tc(dev, 0) rather than
> netdev_reset_tc(). After mqprio with 8 classes then 1, tc_to_txq[1..7]
> still describe txq 1..7. The splat is from an XPS write to txq 2:
> 
>   BUG: KASAN: slab-out-of-bounds in __netif_set_xps_queue+0x1eb9/0x2440
>   Write of size 8 at addr ffff888110e978d8 by task xps_oob/573
>    __netif_set_xps_queue+0x1eb9/0x2440
>    xps_rxqs_store+0x24d/0x360
>    netdev_queue_attr_store+0x61/0x90
>   Allocated by task 573:
>    __kmalloc_noprof+0x246/0x6c0
>    __netif_set_xps_queue+0x8ca/0x2440
>   The buggy address is located 0 bytes to the right of
>    allocated 88-byte region [ffff888110e97880, ffff888110e978d8)

I think it would be useful to provide the stack trace with line numbers
decoded. This can be done using scripts/decode_stacktrace.sh

And perhaps be a bit more explicit in the description about how
an overrun of attr_map occurs; I believe that is what
the splat is telling us has happened.

> 
> Reject a class the map has no room for.
> 
> Fixes: 184c449f91fe ("net: Add support for XPS with QoS via traffic classes")
> Assisted-by: LLM
> Signed-off-by: Norbert Szetei <norbert@doyensec.com>
> ---
> v2:
>   - bound the class in __netif_set_xps_queue() instead of clearing
>     dev->tc_to_txq[]/dev->prio_tc_map[] in netdev_set_num_tc(), per the
>     Sashiko review of v1
>   - dropped the memory-ordering claim from the changelog
>   - retitled
> v1: https://lore.kernel.org/netdev/16E3A318-5532-4B5E-8D03-86D21B463A2D@doyensec.com/

...