diff mbox series

[bpf-next,v2,3/4] inet: null check for close in inet_release

Message ID ad1ecbd205b357f1f73500522a2d495cb6c0cbe1.1719302367.git.tanggeliang@kylinos.cn (mailing list archive)
State New
Headers show
Series Fixes for BPF selftests on Loongarch | expand

Commit Message

Geliang Tang June 25, 2024, 8:24 a.m. UTC
From: Geliang Tang <tanggeliang@kylinos.cn>

Run the following BPF selftests on Loongarch:

./test_progs -t sockmap_listen

A Kernel panic occurs:

'''
 Oops[#1]:
 CPU: 49 PID: 233429 Comm: new_name Tainted: G           OE      6.10.0-rc2+ #20
 Hardware name: LOONGSON Dabieshan/Loongson-TC542F0, BIOS Loongson-UDK2018-V4.0.11
 pc 0000000000000000 ra 90000000051ea4a0 tp 900030008549c000 sp 900030008549fe00
 a0 9000300152524a00 a1 0000000000000000 a2 900030008549fe38 a3 900030008549fe30
 a4 900030008549fe30 a5 90003000c58c8d80 a6 0000000000000000 a7 0000000000000039
 t0 0000000000000000 t1 90003000c58c8d80 t2 0000000000000001 t3 0000000000000000
 t4 0000000000000001 t5 900000011a1bf580 t6 900000011a3aff60 t7 000000000000006b
 t8 00000fffffffffff u0 0000000000000000 s9 00007fffbbe9e930 s0 9000300152524a00
 s1 90003000c58c8d00 s2 9000000006c81568 s3 0000000000000000 s4 90003000c58c8d80
 s5 00007ffff236a000 s6 00007ffffbc292b0 s7 00007ffffbc29998 s8 00007fffbbe9f180
    ra: 90000000051ea4a0 inet_release+0x60/0xc0
   ERA: 0000000000000000 0x0
  CRMD: 000000b0 (PLV0 -IE -DA +PG DACF=CC DACM=CC -WE)
  PRMD: 0000000c (PPLV0 +PIE +PWE)
  EUEN: 00000000 (-FPE -SXE -ASXE -BTE)
  ECFG: 00071c1d (LIE=0,2-4,10-12 VS=7)
 ESTAT: 00030000 [PIF] (IS= ECode=3 EsubCode=0)
  BADV: 0000000000000000
  PRID: 0014c011 (Loongson-64bit, Loongson-3C5000)
 Modules linked in: xt_CHECKSUM xt_MASQUERADE xt_conntrack ipt_REJECT nf_nat_tftp
 Process new_name (pid: 233429, threadinfo=00000000b9196405, task=00000000c01df45b)
 Stack : 0000000000000000 90003000c58c8e20 90003000c58c8d00 900000000505960c
         0000000000000000 9000000101c6ad20 9000300086524540 00000000082e0003
         900030008bf57400 90000000050596bc 900030008bf57400 900000000434acac
         0000000000000016 00007ffff224e060 00007fffbbe9f180 900030008bf57400
         0000000000000000 9000000004341ce0 00007fffbbe9f180 00007ffff2369000
         900030008549fec0 90000000054476ec 000000000000006b 9000000003f71da4
         000000000000003a 00007ffff22b8a44 00007fffbbe9f8e0 00007fffbbe9e680
         ffffffffffffffda 0000000000000000 0000000000000000 0000000000000000
         00007fffbbe9f288 0000000000000000 0000000000000000 0000000000000039
         84c2431493ceab6e 84c23ceb2827425e 0000000000000007 00007ffff2271600
         ...
 Call Trace:
 [<900000000505960c>] __sock_release+0x4c/0xe0
 [<90000000050596bc>] sock_close+0x1c/0x40
 [<900000000434acac>] __fput+0xec/0x2e0
 [<9000000004341ce0>] sys_close+0x40/0xa0
 [<90000000054476ec>] do_syscall+0x8c/0xc0
 [<9000000003f71da4>] handle_syscall+0xc4/0x160

 Code: (Bad address in era)

 ---[ end trace 0000000000000000 ]---
 Kernel panic - not syncing: Fatal exception
 Kernel relocated by 0x3d50000
  .text @ 0x9000000003f50000
  .data @ 0x90000000055b0000
  .bss  @ 0x9000000006ca9400
 ---[ end Kernel panic - not syncing: Fatal exception ]---
'''

This is because "sk->sk_prot->close" pointer is NULL in that case. This
patch adds null check for it in inet_release() to fix this error.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/ipv4/af_inet.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Comments

Eric Dumazet June 25, 2024, 8:31 a.m. UTC | #1
On Tue, Jun 25, 2024 at 10:25 AM Geliang Tang <geliang@kernel.org> wrote:
>
> From: Geliang Tang <tanggeliang@kylinos.cn>
>
> Run the following BPF selftests on Loongarch:
>
> ./test_progs -t sockmap_listen
>
> A Kernel panic occurs:
>
> '''
>  Oops[#1]:
>  CPU: 49 PID: 233429 Comm: new_name Tainted: G           OE      6.10.0-rc2+ #20
>  Hardware name: LOONGSON Dabieshan/Loongson-TC542F0, BIOS Loongson-UDK2018-V4.0.11
>  pc 0000000000000000 ra 90000000051ea4a0 tp 900030008549c000 sp 900030008549fe00
>  a0 9000300152524a00 a1 0000000000000000 a2 900030008549fe38 a3 900030008549fe30
>  a4 900030008549fe30 a5 90003000c58c8d80 a6 0000000000000000 a7 0000000000000039
>  t0 0000000000000000 t1 90003000c58c8d80 t2 0000000000000001 t3 0000000000000000
>  t4 0000000000000001 t5 900000011a1bf580 t6 900000011a3aff60 t7 000000000000006b
>  t8 00000fffffffffff u0 0000000000000000 s9 00007fffbbe9e930 s0 9000300152524a00
>  s1 90003000c58c8d00 s2 9000000006c81568 s3 0000000000000000 s4 90003000c58c8d80
>  s5 00007ffff236a000 s6 00007ffffbc292b0 s7 00007ffffbc29998 s8 00007fffbbe9f180
>     ra: 90000000051ea4a0 inet_release+0x60/0xc0
>    ERA: 0000000000000000 0x0
>   CRMD: 000000b0 (PLV0 -IE -DA +PG DACF=CC DACM=CC -WE)
>   PRMD: 0000000c (PPLV0 +PIE +PWE)
>   EUEN: 00000000 (-FPE -SXE -ASXE -BTE)
>   ECFG: 00071c1d (LIE=0,2-4,10-12 VS=7)
>  ESTAT: 00030000 [PIF] (IS= ECode=3 EsubCode=0)
>   BADV: 0000000000000000
>   PRID: 0014c011 (Loongson-64bit, Loongson-3C5000)
>  Modules linked in: xt_CHECKSUM xt_MASQUERADE xt_conntrack ipt_REJECT nf_nat_tftp
>  Process new_name (pid: 233429, threadinfo=00000000b9196405, task=00000000c01df45b)
>  Stack : 0000000000000000 90003000c58c8e20 90003000c58c8d00 900000000505960c
>          0000000000000000 9000000101c6ad20 9000300086524540 00000000082e0003
>          900030008bf57400 90000000050596bc 900030008bf57400 900000000434acac
>          0000000000000016 00007ffff224e060 00007fffbbe9f180 900030008bf57400
>          0000000000000000 9000000004341ce0 00007fffbbe9f180 00007ffff2369000
>          900030008549fec0 90000000054476ec 000000000000006b 9000000003f71da4
>          000000000000003a 00007ffff22b8a44 00007fffbbe9f8e0 00007fffbbe9e680
>          ffffffffffffffda 0000000000000000 0000000000000000 0000000000000000
>          00007fffbbe9f288 0000000000000000 0000000000000000 0000000000000039
>          84c2431493ceab6e 84c23ceb2827425e 0000000000000007 00007ffff2271600
>          ...
>  Call Trace:
>  [<900000000505960c>] __sock_release+0x4c/0xe0
>  [<90000000050596bc>] sock_close+0x1c/0x40
>  [<900000000434acac>] __fput+0xec/0x2e0
>  [<9000000004341ce0>] sys_close+0x40/0xa0
>  [<90000000054476ec>] do_syscall+0x8c/0xc0
>  [<9000000003f71da4>] handle_syscall+0xc4/0x160
>
>  Code: (Bad address in era)
>
>  ---[ end trace 0000000000000000 ]---
>  Kernel panic - not syncing: Fatal exception
>  Kernel relocated by 0x3d50000
>   .text @ 0x9000000003f50000
>   .data @ 0x90000000055b0000
>   .bss  @ 0x9000000006ca9400
>  ---[ end Kernel panic - not syncing: Fatal exception ]---
> '''
>
> This is because "sk->sk_prot->close" pointer is NULL in that case. This
> patch adds null check for it in inet_release() to fix this error.
>
> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
> Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
> ---
>  net/ipv4/af_inet.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
> index b24d74616637..34a719e98c69 100644
> --- a/net/ipv4/af_inet.c
> +++ b/net/ipv4/af_inet.c
> @@ -434,7 +434,8 @@ int inet_release(struct socket *sock)
>                 if (sock_flag(sk, SOCK_LINGER) &&
>                     !(current->flags & PF_EXITING))
>                         timeout = sk->sk_lingertime;
> -               sk->sk_prot->close(sk, timeout);
> +               if (sk->sk_prot->close)
> +                       sk->sk_prot->close(sk, timeout);

Can you tell us which inet protocol does not have a ->close pointer ?

I find it hard to believe a day-0 bug only hit Loongarch arch in 2024.
diff mbox series

Patch

diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index b24d74616637..34a719e98c69 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -434,7 +434,8 @@  int inet_release(struct socket *sock)
 		if (sock_flag(sk, SOCK_LINGER) &&
 		    !(current->flags & PF_EXITING))
 			timeout = sk->sk_lingertime;
-		sk->sk_prot->close(sk, timeout);
+		if (sk->sk_prot->close)
+			sk->sk_prot->close(sk, timeout);
 		sock->sk = NULL;
 	}
 	return 0;