sch_sfb: Don't assume the skb is still around after enqueueing to child
authorToke Høiland-Jørgensen <toke@toke.dk>
Wed, 31 Aug 2022 21:52:18 +0000 (23:52 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Thu, 15 Sep 2022 10:39:46 +0000 (12:39 +0200)
commitb5aa83141aa97f81c8e06051e4bd925bfb5474fb
treed3d2d4fcd4620423cd75626b1e7975456b581370
parenteb4d8d6b44a23ff2b6e2af06c8240de73dff8a7d
sch_sfb: Don't assume the skb is still around after enqueueing to child

[ Upstream commit 9efd23297cca530bb35e1848665805d3fcdd7889 ]

The sch_sfb enqueue() routine assumes the skb is still alive after it has
been enqueued into a child qdisc, using the data in the skb cb field in the
increment_qlen() routine after enqueue. However, the skb may in fact have
been freed, causing a use-after-free in this case. In particular, this
happens if sch_cake is used as a child of sfb, and the GSO splitting mode
of CAKE is enabled (in which case the skb will be split into segments and
the original skb freed).

Fix this by copying the sfb cb data to the stack before enqueueing the skb,
and using this stack copy in increment_qlen() instead of the skb pointer
itself.

Reported-by: zdi-disclosures@trendmicro.com # ZDI-CAN-18231
Fixes: e13e02a3c68d ("net_sched: SFB flow scheduler")
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
net/sched/sch_sfb.c