Hi Yunshui,

On 05/01/26 13:30, Yunshui Jiang wrote:

From: Eric Dumazet edumazet@google.com

[ Upstream commit d8b57135fd9ffe9a5b445350a686442a531c5339 ]

Changes from upstream: Modify the logic of frame_get_stripped_skb() instead of hsr_get_untagged_frame because the 5.4 kernel API has not been updated yet. Keep the core logic identical to upstream commit d8b57135fd9f.

syzbot got a crash [1] in skb_clone(), caused by a bug in hsr_get_untagged_frame().

When/if create_stripped_skb_hsr() returns NULL, we must not attempt to call skb_clone().

While we are at it, replace a WARN_ONCE() by netdev_warn_once().

[1] general protection fault, probably for non-canonical address 0xdffffc000000000f: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x0000000000000078-0x000000000000007f] CPU: 1 PID: 754 Comm: syz-executor.0 Not tainted 6.0.0-syzkaller-02734-g0326074ff465 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/22/2022 RIP: 0010:skb_clone+0x108/0x3c0 net/core/skbuff.c:1641 Code: 93 02 00 00 49 83 7c 24 28 00 0f 85 e9 00 00 00 e8 5d 4a 29 fa 4c 8d 75 7e 48 b8 00 00 00 00 00 fc ff df 4c 89 f2 48 c1 ea 03 <0f> b6 04 02 4c 89 f2 83 e2 07 38 d0 7f 08 84 c0 0f 85 9e 01 00 00 RSP: 0018:ffffc90003ccf4e0 EFLAGS: 00010207

RAX: dffffc0000000000 RBX: ffffc90003ccf5f8 RCX: ffffc9000c24b000 RDX: 000000000000000f RSI: ffffffff8751cb13 RDI: 0000000000000000 RBP: 0000000000000000 R08: 00000000000000f0 R09: 0000000000000140 R10: fffffbfff181d972 R11: 0000000000000000 R12: ffff888161fc3640 R13: 0000000000000a20 R14: 000000000000007e R15: ffffffff8dc5f620 FS: 00007feb621e4700(0000) GS:ffff8880b9b00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007feb621e3ff8 CR3: 00000001643a9000 CR4: 00000000003506e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace:

<TASK> hsr_get_untagged_frame+0x4e/0x610 net/hsr/hsr_forward.c:164 hsr_forward_do net/hsr/hsr_forward.c:461 [inline] hsr_forward_skb+0xcca/0x1d50 net/hsr/hsr_forward.c:623 hsr_handle_frame+0x588/0x7c0 net/hsr/hsr_slave.c:69 __netif_receive_skb_core+0x9fe/0x38f0 net/core/dev.c:5379 __netif_receive_skb_one_core+0xae/0x180 net/core/dev.c:5483 __netif_receive_skb+0x1f/0x1c0 net/core/dev.c:5599 netif_receive_skb_internal net/core/dev.c:5685 [inline] netif_receive_skb+0x12f/0x8d0 net/core/dev.c:5744 tun_rx_batched+0x4ab/0x7a0 drivers/net/tun.c:1544 tun_get_user+0x2686/0x3a00 drivers/net/tun.c:1995 tun_chr_write_iter+0xdb/0x200 drivers/net/tun.c:2025 call_write_iter include/linux/fs.h:2187 [inline] new_sync_write fs/read_write.c:491 [inline] vfs_write+0x9e9/0xdd0 fs/read_write.c:584 ksys_write+0x127/0x250 fs/read_write.c:637 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd

Cc: stable@vger.kernel.org Fixes: f266a683a480 ("net/hsr: Better frame dispatch") Reported-by: syzbot syzkaller@googlegroups.com Signed-off-by: Eric Dumazet edumazet@google.com Link: https://lore.kernel.org/r/20221017165928.2150130-1-edumazet@google.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Yunshui Jiang jiangyunshui@kylinos.cn

Note: 5.4.y stable branch is End-of-life as of Dec 3.

Please find the actively maintained LTS branches here: https://www.kernel.org/category/releases.html

Greg's email has more details: https://lore.kernel.org/all/2025120319-blip-grime-93e8@gregkh/

Thanks, Harshit

---

net/hsr/hsr_forward.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/net/hsr/hsr_forward.c b/net/hsr/hsr_forward.c index 7073724fdfa6..5c1b9cd6dd52 100644 --- a/net/hsr/hsr_forward.c +++ b/net/hsr/hsr_forward.c @@ -115,8 +115,17 @@ static struct sk_buff *create_stripped_skb(struct sk_buff *skb_in, static struct sk_buff *frame_get_stripped_skb(struct hsr_frame_info *frame, struct hsr_port *port) {

• if (!frame->skb_std)

• frame->skb_std = create_stripped_skb(frame->skb_hsr, frame);
  

• if (!frame->skb_std) {

• if (frame->skb_hsr)
  

• 	frame->skb_std =
  

• 		create_stripped_skb(frame->skb_hsr, frame);
  

• else
  

• 	netdev_warn_once(port->dev,
  

• 		"Unexpected frame received in hsr_get_untagged_frame()\n");
  

• if (!frame->skb_std)
  

• 	return NULL;
  

• } return skb_clone(frame->skb_std, GFP_ATOMIC); }