net: fix frag_list leak in skb_segment_list() error path
Subsystems (2)
Eric Dumazet
maintainerPATCH 1/1 net: fix frag_list leak in skb_segment_list() error path
skb_segment_list() detaches the frag_list from the head skb and moves its members one by one to the skb->next chain. On error, err_linearize frees that chain with kfree_skb_list(skb->next), relying on the members not yet processed to still be linked after the current one.
This does not hold when the current member was shared: it is replaced by skb_clone(), and __skb_clone() clears ->next. If __skb_linearize() then fails on that clone, the remaining members are no longer reachable from skb->next nor from skb_shinfo(skb)->frag_list, and they leak.
The skb_shared() error branch already relinks nskb->next to list_skb before jumping to err_linearize. Do it unconditionally right after nskb is added to the chain, so every error exit sees the remaining members. On the last member list_skb is NULL, which still terminates the chain.
---
Bug discovered while reviewing at Simon Polack upcoming patch in skb_segment_list()
net/core/skbuff.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
net/core/skbuff.c +6 -3
@@ -4734,10 +4734,13 @@ struct sk_buff *skb_segment_list(struct sk_buff *skb,else tail->next = nskb; -if (unlikely (err)) { - nskb->next = list_skb; +/* nskb may be a clone with a NULL ->next: keep the remaining +* members linked so that err_linearize frees them. +*/ + nskb->next = list_skb; + +if (unlikely (err)) goto err_linearize; - } tail = nskb;
Signature · 1 line
2.53.0
Quoted 10 lines · Eric Dumazet
skb_segment_list()detaches thefrag_listfrom the head skb and moves its members one by one to theskb->nextchain. On error,err_linearizefrees that chain withkfree_skb_list(skb->next), relying on the members not yet processed to still be linked after the current one.This does not hold when the current member was shared: it is replaced by
skb_clone(), and__skb_clone()clears ->next. If__skb_linearize()then fails on that clone, the remaining members are no longer reachable fromskb->nextnor from skb_shinfo(skb)->frag_list, and they leak.
Quoted 8 lines
The
skb_shared()error branch already relinksnskb->nexttolist_skbbefore jumping toerr_linearize. Do it unconditionally right after nskb is added to the chain, so every error exit sees the remaining members. On the last memberlist_skbis NULL, which still terminates the chain.Fixes 53475c5dd856 ("net: fix use-after-free when UDP GRO with shared fraglist")Signed-off-by Eric Dumazet <[email protected]>Cc Simon Polack <[email protected]>
Quoted 10 lines · On Wed, 21 Oct 2026 18:18:58 +0200 Eric Dumazet wrote:
skb_segment_list()detaches thefrag_listfrom the head skb and moves its members one by one to theskb->nextchain. On error,err_linearizefrees that chain withkfree_skb_list(skb->next), relying on the members not yet processed to still be linked after the current one.This does not hold when the current member was shared: it is replaced by
skb_clone(), and__skb_clone()clears ->next. If__skb_linearize()then fails on that clone, the remaining members are no longer reachable fromskb->nextnor from skb_shinfo(skb)->frag_list, and they leak.
Quoted 8 lines
The
skb_shared()error branch already relinksnskb->nexttolist_skbbefore jumping toerr_linearize. Do it unconditionally right after nskb is added to the chain, so every error exit sees the remaining members. On the last memberlist_skbis NULL, which still terminates the chain.Fixes 53475c5dd856 ("net: fix use-after-free when UDP GRO with shared fraglist")Signed-off-by Eric Dumazet <[email protected]>Cc Simon Polack <[email protected]>
Quoted 10 lines · Eric Dumazet
skb_segment_list()detaches thefrag_listfrom the head skb and moves its members one by one to theskb->nextchain. On error,err_linearizefrees that chain withkfree_skb_list(skb->next), relying on the members not yet processed to still be linked after the current one.This does not hold when the current member was shared: it is replaced by
skb_clone(), and__skb_clone()clears ->next. If__skb_linearize()then fails on that clone, the remaining members are no longer reachable fromskb->nextnor from skb_shinfo(skb)->frag_list, and they leak.
Quoted 8 lines
The
skb_shared()error branch already relinksnskb->nexttolist_skbbefore jumping toerr_linearize. Do it unconditionally right after nskb is added to the chain, so every error exit sees the remaining members. On the last memberlist_skbis NULL, which still terminates the chain.Fixes 53475c5dd856 ("net: fix use-after-free when UDP GRO with shared fraglist")Signed-off-by Eric Dumazet <[email protected]>Cc Simon Polack <[email protected]>
Quoted 28 lines
---
Bug discovered while reviewing at Simon Polack upcoming patch inskb_segment_list()net/core/skbuff.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)net/core/skbuff.c +6
-3@@ -4734,10 +4734,13 @@ struct sk_buff *skb_segment_list(struct sk_buff *skb,else tail->next = nskb; -if (unlikely (err)) { - nskb->next = list_skb; +/* nskb may be a clone with a NULL ->next: keep the remaining +* members linked so that err_linearize frees them. +*/ + nskb->next = list_skb; + +if (unlikely (err)) goto err_linearize; - } tail = nskb;
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 11 lines · On Wed, 21 Oct 2026 18:18:58 +0200 you wrote:
skb_segment_list()detaches thefrag_listfrom the head skb and moves its members one by one to theskb->nextchain. On error,err_linearizefrees that chain withkfree_skb_list(skb->next), relying on the members not yet processed to still be linked after the current one.This does not hold when the current member was shared: it is replaced by
skb_clone(), and__skb_clone()clears ->next. If__skb_linearize()then fails on that clone, the remaining members are no longer reachable fromskb->nextnor from skb_shinfo(skb)->frag_list, and they leak.[...]
Here is the summary with links:
- [net] net: fix frag_list leak in skb_segment_list() error path
https://git.kernel.org/netdev/net/c/546daf344420
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html