net: sched: fix 32-bit backlog wrap in gred, bfifo and plug enqueue
Subsystems (4)
Jamal Hadi Salim
maintainerPATCH v2 1/1 net: sched: fix 32-bit backlog wrap in gred, bfifo and plug enqueue
gred_enqueue(), bfifo_enqueue() and plug_enqueue() admit a packet when the current backlog plus the packet length fits within the queue limit:
sch->qstats.backlog + qdisc_pkt_len(skb) <= sch->limit (gred default VQ) gred_backlog+qdisc_pkt_len(skb) <= q->limit (gred configured VQ) sch->qstats.backlog + qdisc_pkt_len(skb) <= sch->limit (bfifo)
sch->qstats.backlog + skb->len <= q->limit (plug)
sch->qstats.backlog and q->backlog are u32, and qdisc_pkt_len()/skb->len are unsigned int, so all sums are computed in 32 bits and wrap at 2^32. Once the true backlog exceeds 4 GiB the wrapped sum becomes small and admission keeps succeeding, so the queue grows without bound and the kernel can be driven to OOM.
Promote the sums to u64 so admission stops once the true backlog exceeds the limit. The limit is u32, so the bounded queue stays below 2^32 and the stored u32 backlog never wraps.
The bug can only be reproduced as root (albeit with ridiculous setup):
attach a gred (or bfifo/plug) qdisc with a limit near 4 GiB, leaving the default VQ unconfigured (for gred), and drive >4 GiB of queued traffic (e.g. via a size table / stab to inflate qdisc_pkt_len, or sustained high-rate traffic). The u32 backlog+len sum wraps at 2^32, admission keeps succeeding, and the queue grows unboundedly to OOM.
---v1->v2:
1. Added bfifo+plug into the same patch as gred since it is the same pattern. Flagged by Zhan Xusheng and Sashiko[1][2].
2. Starting this patch, and for the rest of AI found bugs, i will start adding the conditions required to reproduce the patch (see the above commentary "The bug can only be reproduced as root...").
[1] https://sashiko.dev/#/patchset/[email protected]
[2] https://netdev-ai.bots.linux.dev/sashiko/#/patchset/[email protected]
--- net/sched/sch_fifo.c | 2 +- net/sched/sch_gred.c | 4 ++-- net/sched/sch_plug.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-)
net/sched/sch_fifo.c +1 -1
@@ -19,7 +19,7 @@static int bfifo_enqueue (struct sk_buff *skb,struct Qdisc *sch,struct sk_buff **to_free) { -if (likely (sch->qstats.backlog +qdisc_pkt_len (skb) <= +if (likely ((u64 )sch->qstats.backlog +qdisc_pkt_len (skb) <=READ_ONCE (sch->limit)))return qdisc_enqueue_tail (skb, sch);
net/sched/sch_gred.c +2 -2
@@ -179,7 +179,7 @@ static int gred_enqueue(struct sk_buff *skb, struct Qdisc *sch,* if no default DP has been configured. This * allows for DP flows to be left untouched. */ -if (likely (sch->qstats.backlog +qdisc_pkt_len (skb) <= +if (likely ((u64 )sch->qstats.backlog +qdisc_pkt_len (skb) <= sch->limit))return qdisc_enqueue_tail (skb, sch);else
@@ -244,7 +244,7 @@ static int gred_enqueue(struct sk_buff *skb, struct Qdisc *sch,break ; } -if (gred_backlog (t, q, sch) +qdisc_pkt_len (skb) <= q->limit) { +if ((u64 )gred_backlog (t, q, sch) +qdisc_pkt_len (skb) <= q->limit) { q->backlog +=qdisc_pkt_len (skb);return qdisc_enqueue_tail (skb, sch); }
net/sched/sch_plug.c +1 -1
@@ -89,7 +89,7 @@ static int plug_enqueue(struct sk_buff *skb, struct Qdisc *sch, {struct plug_sched_data *q =qdisc_priv (sch); -if (likely (sch->qstats.backlog + skb->len <= q->limit)) { +if (likely ((u64 )sch->qstats.backlog + skb->len <= q->limit)) {if (!q->unplug_indefinite) q->pkts_current_epoch++;return qdisc_enqueue_tail (skb, sch);
Signature · 1 line
2.43.0
Quoted 32 lines · Jamal Hadi Salim
gred_enqueue(),bfifo_enqueue()andplug_enqueue()admit a packet when the current backlog plus the packet length fits within the queue limit:
sch->qstats.backlog+qdisc_pkt_len(skb)<=sch->limit(gred default VQ) gred_backlog+qdisc_pkt_len(skb) <=q->limit(gred configured VQ)sch->qstats.backlog+qdisc_pkt_len(skb)<=sch->limit(bfifo)
sch->qstats.backlog+skb->len<=q->limit(plug)
sch->qstats.backlogandq->backlogare u32, and qdisc_pkt_len()/skb->len are unsigned int, so all sums are computed in 32 bits and wrap at 2^32. Once the true backlog exceeds 4 GiB the wrapped sum becomes small and admission keeps succeeding, so the queue grows without bound and the kernel can be driven to OOM.Promote the sums to u64 so admission stops once the true backlog exceeds the limit. The limit is u32, so the bounded queue stays below 2^32 and the stored u32 backlog never wraps.
The bug can only be reproduced as root (albeit with ridiculous setup):
attach a gred (or bfifo/plug) qdisc with a limit near 4 GiB, leaving the default VQ unconfigured (for gred), and drive >4 GiB of queued traffic (e.g. via a size table / stab to inflateqdisc_pkt_len, or sustained high-rate traffic). The u32 backlog+len sum wraps at 2^32, admission keeps succeeding, and the queue grows unboundedly to OOM.Fixes a3eb95f891d6 ("net_sched: gred: add TCA_GRED_LIMIT attribute")Reported-by [email protected]Tested-by Victor Nogueira <[email protected]>Signed-off-by Jamal Hadi Salim <[email protected]>---
v1->v2:
Quoted 8 lines
1. Added bfifo+plug into the same patch as gred since it is the same pattern. Flagged by Zhan Xusheng and Sashiko[1][2].
2. Starting this patch, and for the rest of AI found bugs, i will start adding the conditions required to reproduce the patch (see the above commentary "The bug can only be reproduced as root...").[1] https://sashiko.dev/#/patchset/[email protected]
[2] https://netdev-ai.bots.linux.dev/sashiko/#/patchset/[email protected]
This patch was applied to netdev/net.git (main) by Jakub Kicinski <[email protected]>:
Hello:
This patch was applied to netdev/net.git (main)
by Jakub Kicinski <[email protected]>:
Quoted 9 lines · On Tue, 18 Aug 2026 05:59:27 -0400 you wrote:
gred_enqueue(),bfifo_enqueue()andplug_enqueue()admit a packet when the current backlog plus the packet length fits within the queue limit:
sch->qstats.backlog+qdisc_pkt_len(skb)<=sch->limit(gred default VQ) gred_backlog+qdisc_pkt_len(skb) <=q->limit(gred configured VQ)sch->qstats.backlog+qdisc_pkt_len(skb)<=sch->limit(bfifo)
sch->qstats.backlog+skb->len<=q->limit(plug)[...]
Here is the summary with links:
- [net,v2] net: sched: fix 32-bit backlog wrap in gred, bfifo and plug enqueue
https://git.kernel.org/netdev/net/c/4c660ee8c809
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html