[PATCH net v2] net: ifb: report ethtool stats over num_tx_queues

[PATCH net v2] net: ifb: report ethtool stats over num_tx_queues

[Michael Bommarito]

ifb_dev_init() allocates dp->tx_private to dev->num_tx_queues
entries via kzalloc_objs(*txp, dev->num_tx_queues). Both IFB
per-queue RX and TX stats live in those entries: ifb_xmit() updates
txp->rx_stats using the skb queue mapping, ifb_ri_tasklet() updates
txp->tx_stats, and ifb_stats64() aggregates both over
dev->num_tx_queues.

The ethtool stats callbacks instead size and walk the per-queue
stats with dev->real_num_rx_queues and dev->real_num_tx_queues. With
an asymmetric device where the RX queue count exceeds the TX queue
count, for example:

    ip link add name ifb10 numtxqueues 1 numrxqueues 8 type ifb
    ethtool -S ifb10

ifb_get_ethtool_stats() indexes past the tx_private allocation and
copies adjacent slab data through ETHTOOL_GSTATS.

Use dev->num_tx_queues consistently for the stats strings, the
stats count, and the stats data walks. This reports one RX stats
group and one TX stats group for each backing ifb_q_private entry,
which is the queue set IFB can actually populate.

Reproduced under UML+KASAN at v7.1-rc2:

  BUG: KASAN: slab-out-of-bounds in ifb_fill_stats_data+0x3c/0xae
  Read of size 8 at addr 0000000062dbd228 by task ethtool/36
  ifb_fill_stats_data+0x3c/0xae
  ifb_get_ethtool_stats+0xc0/0x129
  __dev_ethtool+0x1ca5/0x363c
  dev_ethtool+0x123/0x1b3
  dev_ioctl+0x56c/0x744
  sock_do_ioctl+0x15f/0x1b2
  sock_ioctl+0x4d5/0x50a
  sys_ioctl+0xd8b/0xde9

With the patch applied, the same UML+KASAN repro is silent and
ethtool -S ifb10 reports only the stats backed by the single
allocated tx_private entry.

Fixes: a21ee5b2fcb8 ("net: ifb: support ethtools stats")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-4-7
Assisted-by: Codex:gpt-5-5-xhigh
Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com>
---
v2:
- Follow Jakub's review: IFB has no RX-only queue stats independent
  of tx_private, so dump both RX and TX ethtool stats over
  dev->num_tx_queues.
- Update get_strings() and get_sset_count() to use the same queue
  bound as get_ethtool_stats().
- Drop the v1 zero-padding helper and report only the stats slots that
  have a backing ifb_q_private entry.

v1: https://lore.kernel.org/netdev/20260511122835.441911-1-michael.bommarito@gmail.com/

 drivers/net/ifb.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index 5407d2ed71b3..43aa1bfd41cf 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -211,12 +211,12 @@ static void ifb_get_strings(struct net_device *dev, u32 stringset, u8 *buf)
 
 	switch (stringset) {
 	case ETH_SS_STATS:
-		for (i = 0; i < dev->real_num_rx_queues; i++)
+		for (i = 0; i < dev->num_tx_queues; i++)
 			for (j = 0; j < IFB_Q_STATS_LEN; j++)
 				ethtool_sprintf(&p, "rx_queue_%u_%.18s",
 						i, ifb_q_stats_desc[j].desc);
 
-		for (i = 0; i < dev->real_num_tx_queues; i++)
+		for (i = 0; i < dev->num_tx_queues; i++)
 			for (j = 0; j < IFB_Q_STATS_LEN; j++)
 				ethtool_sprintf(&p, "tx_queue_%u_%.18s",
 						i, ifb_q_stats_desc[j].desc);
@@ -229,8 +229,7 @@ static int ifb_get_sset_count(struct net_device *dev, int sset)
 {
 	switch (sset) {
 	case ETH_SS_STATS:
-		return IFB_Q_STATS_LEN * (dev->real_num_rx_queues +
-					  dev->real_num_tx_queues);
+		return IFB_Q_STATS_LEN * dev->num_tx_queues * 2;
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -262,12 +261,12 @@ static void ifb_get_ethtool_stats(struct net_device *dev,
 	struct ifb_q_private *txp;
 	int i;
 
-	for (i = 0; i < dev->real_num_rx_queues; i++) {
+	for (i = 0; i < dev->num_tx_queues; i++) {
 		txp = dp->tx_private + i;
 		ifb_fill_stats_data(&data, &txp->rx_stats);
 	}
 
-	for (i = 0; i < dev->real_num_tx_queues; i++) {
+	for (i = 0; i < dev->num_tx_queues; i++) {
 		txp = dp->tx_private + i;
 		ifb_fill_stats_data(&data, &txp->tx_stats);
 	}
-- 
2.53.0