Re: [PATCH v5] bpf: test_run: Fix the null pointer dereference issue in bpf_lwt_xmit_push_encap
From: Feng Yang <hidden>
Date: 2026-02-11 07:54:07
Also in:
bpf, lkml
On Tue, 10 Feb 2026 11:10:03 -0800 Martin KaFai Lau [off-list ref] wrote:
On 2/10/26 1:06 AM, Feng Yang wrote:quoted
From: Feng Yang <redacted> The bpf_lwt_xmit_push_encap helper needs to access skb_dst(skb)->dev to calculate the needed headroom: err = skb_cow_head(skb, len + LL_RESERVED_SPACE(skb_dst(skb)->dev)); But skb->_skb_refdst may not be initialized when the skb is set up by bpf_prog_test_run_skb function. Executing bpf_lwt_push_ip_encap function in this scenario will trigger null pointer dereference, causing a kernel crash as Yinhao reported: [ 105.186365] BUG: kernel NULL pointer dereference, address: 0000000000000000 [ 105.186382] #PF: supervisor read access in kernel mode [ 105.186388] #PF: error_code(0x0000) - not-present page [ 105.186393] PGD 121d3d067 P4D 121d3d067 PUD 106c83067 PMD 0 [ 105.186404] Oops: 0000 [#1] PREEMPT SMP NOPTI [ 105.186412] CPU: 3 PID: 3250 Comm: poc Kdump: loaded Not tainted 6.19.0-rc5 #1 [ 105.186423] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 [ 105.186427] RIP: 0010:bpf_lwt_push_ip_encap+0x1eb/0x520 [ 105.186443] Code: 0f 84 de 01 00 00 0f b7 4a 04 66 85 c9 0f 85 47 01 00 00 31 c0 5b 5d 41 5c 41 5d 41 5e c3 cc cc cc cc 48 8b 73 58 48 83 e6 fe <48> 8b 36 0f > b7 be ec 00 00 00 0f b7 b6 e6 00 00 00 01 fe 83 e6 f0 [ 105.186449] RSP: 0018:ffffbb0e0387bc50 EFLAGS: 00010246 [ 105.186455] RAX: 000000000000004e RBX: ffff94c74e036500 RCX: ffff94c74874da00 [ 105.186460] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff94c74e036500 [ 105.186463] RBP: 0000000000000001 R08: 0000000000000002 R09: 0000000000000000 [ 105.186467] R10: ffffbb0e0387bd50 R11: 0000000000000000 R12: ffffbb0e0387bc98 [ 105.186471] R13: 0000000000000014 R14: 0000000000000000 R15: 0000000000000002 [ 105.186484] FS: 00007f166aa4d680(0000) GS:ffff94c8b7780000(0000) knlGS:0000000000000000 [ 105.186490] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 105.186494] CR2: 0000000000000000 CR3: 000000015eade001 CR4: 0000000000770ee0 [ 105.186499] PKRU: 55555554 [ 105.186502] Call Trace: [ 105.186507] <TASK> [ 105.186513] bpf_lwt_xmit_push_encap+0x2b/0x40 [ 105.186522] bpf_prog_a75eaad51e517912+0x41/0x49 [ 105.186536] ? kvm_clock_get_cycles+0x18/0x30 [ 105.186547] ? ktime_get+0x3c/0xa0 [ 105.186554] bpf_test_run+0x195/0x320 [ 105.186563] ? bpf_test_run+0x10f/0x320 [ 105.186579] bpf_prog_test_run_skb+0x2f5/0x4f0 [ 105.186590] __sys_bpf+0x69c/0xa40 [ 105.186603] __x64_sys_bpf+0x1e/0x30 [ 105.186611] do_syscall_64+0x59/0x110 [ 105.186620] entry_SYSCALL_64_after_hwframe+0x76/0xe0 [ 105.186649] RIP: 0033:0x7f166a97455d Temporarily add the setting of skb->_skb_refdst before bpf_test_run to resolve the issue. Fixes: 52f278774e79 ("bpf: implement BPF_LWT_ENCAP_IP mode in bpf_lwt_push_encap") Reported-by: Yinhao Hu <redacted> Reported-by: Kaiyan Mei <redacted> Closes: https://groups.google.com/g/hust-os-kernel-patches/c/8-a0kPpBW2s Signed-off-by: Yun Lu <redacted> Signed-off-by: Feng Yang <redacted> Tested-by: syzbot@syzkaller.appspotmail.com --- Changes in v5: - Refer to the bpf_lwt_xmit_reroute function to configure the dst parameter. - Link to v4: https://lore.kernel.org/all/20260209015111.28144-1-yangfeng59949@163.com/ (local) Changes in v4: - add rcu lock - Link to v3: https://lore.kernel.org/all/20260206055113.63476-1-yangfeng59949@163.com/ (local) Changes in v3: - use dst_init - Link to v2: https://lore.kernel.org/all/20260205092227.126665-1-yangfeng59949@163.com/ (local) Changes in v2: - Link to v1: https://lore.kernel.org/all/20260127084520.13890-1-luyun_611@163.com/ (local)The earlier syzbot reports are crying for a selftest which is still missing in v5. The CI has also reported errors in the test_progs. Did you run any test before posting ?
My apologies. I only tested whether a crash would occur when using `bpf_prog_test_run_skb` to execute `bpf_lwt_push_ip_encap` without a dst entry. I will include a selftest in my next submission.
pw-bot: crquoted
--- net/bpf/test_run.c | 47 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 47 insertions(+)diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 178c4738e63b..dbd2c44da7e5 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c@@ -24,6 +24,7 @@ #include <net/netdev_rx_queue.h> #include <net/xdp.h> #include <net/netfilter/nf_bpf_link.h> +#include <net/ip6_route.h> #define CREATE_TRACE_POINTS #include <trace/events/bpf_test_run.h>@@ -992,6 +993,7 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr, u32 headroom = NET_SKB_PAD + NET_IP_ALIGN; u32 linear_sz = kattr->test.data_size_in; u32 repeat = kattr->test.repeat; + struct dst_entry *dst = NULL; struct __sk_buff *ctx = NULL; struct sk_buff *skb = NULL; struct sock *sk = NULL;@@ -1156,6 +1158,51 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr, skb->ip_summed = CHECKSUM_COMPLETE; } + if (prog->type == BPF_PROG_TYPE_LWT_XMIT && !skb_dst(skb)) { + struct flowi4 fl4 = {}; + struct flowi6 fl6 = {}; + struct rtable *rt; + + switch (skb->protocol) { + case htons(ETH_P_IP): + if (sizeof(struct iphdr) <= skb_headlen(skb)) { + fl4.saddr = ip_hdr(skb)->saddr; + fl4.daddr = ip_hdr(skb)->daddr; + } + + rt = ip_route_output_key(net, &fl4);What can be expected from the return value rt if fl4 is 0.
If it is empty, `ip_route_output_key_hash_rcu` will assign the loopback address, and the returned `rt` is valid, not an error value.
quoted
+ if (IS_ERR(rt)) { + ret = PTR_ERR(rt);I suspect this is probably what failed in CI. Add a NULL check in bpf_lwt_push_ip_encap instead. No change is needed in the bpf_prog_test_run_skb.
The root cause of the CI failure is that the default path was taken, resulting in a return value of -EINVAL. So, is your suggestion to modify the bpf_lwt_push_ip_encap function as in the v1(https://lore.kernel.org/all/20260127084520.13890-1-luyun_611@163.com/ (local)) version? However, the previous suggestion was to make changes to the bpf_prog_test_run_skb function instead. Thank you very much for your reply.
quoted
+ goto out; + } + dst = &rt->dst; + break; +#if IS_ENABLED(CONFIG_IPV6) + case htons(ETH_P_IPV6): + if (sizeof(struct ipv6hdr) <= skb_headlen(skb)) { + fl6.saddr = ipv6_hdr(skb)->saddr; + fl6.daddr = ipv6_hdr(skb)->daddr; + } + + dst = ip6_route_output(net, NULL, &fl6); + if (IS_ERR(dst)) { + ret = PTR_ERR(dst); + goto out; + } + break; +#endif + default: + ret = -EINVAL;
The reason for the CI test failure.
quoted
+ goto out; + } + + if (unlikely(dst->error)) { + ret = dst->error; + dst_release(dst); + goto out; + } + skb_dst_set(skb, dst); + } ret = bpf_test_run(prog, skb, repeat, &retval, &duration, false); if (ret) goto out;