Re: [PATCH bpf-next v2 1/4] skmsg: null check for sg_page in sk_msg_recvmsg

[Geliang Tang <geliang@kernel.org>]

On Tue, 2024-06-25 at 12:37 -0700, John Fastabend wrote:
> Eric Dumazet wrote:
> > 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_basic
> > > 
> > > A Kernel panic occurs:
> > > 
> > > '''
> > >  Oops[#1]:
> > >  CPU: 22 PID: 2824 Comm: test_progs Tainted: G           OE     
> > > 6.10.0-rc2+ #18
> > >  Hardware name: LOONGSON Dabieshan/Loongson-TC542F0, BIOS
> > > Loongson-UDK2018-V4.0.11
> > >  pc 9000000004162774 ra 90000000048bf6c0 tp 90001000aa16c000 sp
> > > 90001000aa16fb90
> > >  a0 0000000000000000 a1 0000000000000000 a2 0000000000000000 a3
> > > 90001000aa16fd70
> > >  a4 0000000000000800 a5 0000000000000000 a6 000055557b63aae8 a7
> > > 00000000000000cf
> > >  t0 0000000000000000 t1 0000000000004000 t2 0000000000000048 t3
> > > 0000000000000000
> > >  t4 0000000000000001 t5 0000000000000002 t6 0000000000000001 t7
> > > 0000000000000002
> > >  t8 0000000000000018 u0 9000000004856150 s9 0000000000000000 s0
> > > 0000000000000000
> > >  s1 0000000000000000 s2 90001000aa16fd70 s3 0000000000000000 s4
> > > 0000000000000000
> > >  s5 0000000000004000 s6 900010009284dc00 s7 0000000000000001 s8
> > > 900010009284dc00
> > >     ra: 90000000048bf6c0 sk_msg_recvmsg+0x120/0x560
> > >    ERA: 9000000004162774 copy_page_to_iter+0x74/0x1c0
> > >   CRMD: 000000b0 (PLV0 -IE -DA +PG DACF=CC DACM=CC -WE)
> > >   PRMD: 0000000c (PPLV0 +PIE +PWE)
> > >   EUEN: 00000007 (+FPE +SXE +ASXE -BTE)
> > >   ECFG: 00071c1d (LIE=0,2-4,10-12 VS=7)
> > >  ESTAT: 00010000 [PIL] (IS= ECode=1 EsubCode=0)
> > >   BADV: 0000000000000040
> > >   PRID: 0014c011 (Loongson-64bit, Loongson-3C5000)
> > >  Modules linked in: bpf_testmod(OE) xt_CHECKSUM xt_MASQUERADE
> > > xt_conntrack
> > >  Process test_progs (pid: 2824, threadinfo=0000000000863a31,
> > > task=000000001cba0874)
> > >  Stack : 0000000000000001 fffffffffffffffc 0000000000000000
> > > 0000000000000000
> > >          0000000000000018 0000000000000000 0000000000000000
> > > 90000000048bf6c0
> > >          90000000052cd638 90001000aa16fd70 900010008bf51580
> > > 900010009284f000
> > >          90000000049f2b90 900010009284f188 900010009284f178
> > > 90001000861d4780
> > >          9000100084dccd00 0000000000000800 0000000000000007
> > > fffffffffffffff2
> > >          000000000453e92f 90000000049aae34 90001000aa16fd60
> > > 900010009284f000
> > >          0000000000000000 0000000000000000 900010008bf51580
> > > 90000000049f2b90
> > >          0000000000000001 0000000000000000 9000100084dc3a10
> > > 900010009284f1ac
> > >          90001000aa16fd40 0000555559953278 0000000000000001
> > > 0000000000000000
> > >          90001000aa16fdc8 9000000005a5a000 90001000861d4780
> > > 0000000000000800
> > >          ...
> > >  Call Trace:
> > >  [<9000000004162774>] copy_page_to_iter+0x74/0x1c0
> > >  [<90000000048bf6c0>] sk_msg_recvmsg+0x120/0x560
> > >  [<90000000049f2b90>] tcp_bpf_recvmsg_parser+0x170/0x4e0
> > >  [<90000000049aae34>] inet_recvmsg+0x54/0x100
> > >  [<900000000481ad5c>] sock_recvmsg+0x7c/0xe0
> > >  [<900000000481e1a8>] __sys_recvfrom+0x108/0x1c0
> > >  [<900000000481e27c>] sys_recvfrom+0x1c/0x40
> > >  [<9000000004c076ec>] do_syscall+0x8c/0xc0
> > >  [<9000000003731da4>] handle_syscall+0xc4/0x160
> > > 
> > >  Code: 0010b09b  440125a0  0011df8d <28c10364> 0012b70c 
> > > 00133305  0013b1ac  0010dc84  00151585
> > > 
> > >  ---[ end trace 0000000000000000 ]---
> > >  Kernel panic - not syncing: Fatal exception
> > >  Kernel relocated by 0x3510000
> > >   .text @ 0x9000000003710000
> > >   .data @ 0x9000000004d70000
> > >   .bss  @ 0x9000000006469400
> > >  ---[ end Kernel panic - not syncing: Fatal exception ]---
> > > '''
> > > 
> > > This is because "sg_page(sge)" is NULL in that case. This patch
> > > adds null
> > > check for it in sk_msg_recvmsg() to fix this error.
> > > 
> > > Fixes: 604326b41a6f ("bpf, sockmap: convert to generic sk_msg
> > > interface")
> > > Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
> > > ---
> > >  net/core/skmsg.c | 2 ++
> > >  1 file changed, 2 insertions(+)
> > > 
> > > diff --git a/net/core/skmsg.c b/net/core/skmsg.c
> > > index fd20aae30be2..bafcc1e2eadf 100644
> > > --- a/net/core/skmsg.c
> > > +++ b/net/core/skmsg.c
> > > @@ -432,6 +432,8 @@ int sk_msg_recvmsg(struct sock *sk, struct
> > > sk_psock *psock, struct msghdr *msg,
> > >                         sge = sk_msg_elem(msg_rx, i);
> > >                         copy = sge->length;
> > >                         page = sg_page(sge);
> > > +                       if (!page)
> > > +                               goto out;
> > >                         if (copied + copy > len)
> > >                                 copy = len - copied;
> > >                         copy = copy_page_to_iter(page, sge-
> > > >offset, copy, iter);
> > > --
> > > 2.43.0
> > > 
> > 
> > This looks pretty much random to me.
> > 
> > Please find the root cause, instead of desperately trying to fix
> > 'tests'.
> 
> If this happens then either we put a bad msg_rx on the queue see a
> few lines
> up and we need to sort out why that msg_rx was built. Or we walked
> off the
> end of a scatter gather list and need to see why this test isn't
> sufficient?
> 
>   } while ((i != msg_rx->sg.end) && !sg_is_last(sge))
> 
> is this happening every time you run the command or did you run this
> for
> a long iteration and eventually hit this? I don't see why this would
> be

This happens every time when run test_sockmap_skb_verdict_shutdown test
in sockmap_basic. It hits this null page case on X86_64 platform too.

> specific to your arch though.

Kernel panics when a null page is passed to kmap_local_page() on
Loongarch only, and this function is an arch specific one. I think this
issue is somehow related to Loongarch's memory management.

Thanks,
-Geliang