diff options
author | Shmulik Ladkani <shmulik@metanetworks.com> | 2020-12-14 14:38:32 +0100 |
---|---|---|
committer | Steffen Klassert <steffen.klassert@secunet.com> | 2020-12-19 08:12:17 +0100 |
commit | 56ce7c25ae1525d83cf80a880cf506ead1914250 (patch) | |
tree | bf05f75d9ba2d4e7a81853207ceb918662977ddb /net/xfrm/xfrm_input.c | |
parent | Merge tag 'mtd/fixes-for-5.10-rc8' of git://git.kernel.org/pub/scm/linux/kern... (diff) | |
download | linux-56ce7c25ae1525d83cf80a880cf506ead1914250.tar.xz linux-56ce7c25ae1525d83cf80a880cf506ead1914250.zip |
xfrm: Fix oops in xfrm_replay_advance_bmp
When setting xfrm replay_window to values higher than 32, a rare
page-fault occurs in xfrm_replay_advance_bmp:
BUG: unable to handle page fault for address: ffff8af350ad7920
#PF: supervisor write access in kernel mode
#PF: error_code(0x0002) - not-present page
PGD ad001067 P4D ad001067 PUD 0
Oops: 0002 [#1] SMP PTI
CPU: 3 PID: 30 Comm: ksoftirqd/3 Kdump: loaded Not tainted 5.4.52-050452-generic #202007160732
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014
RIP: 0010:xfrm_replay_advance_bmp+0xbb/0x130
RSP: 0018:ffffa1304013ba40 EFLAGS: 00010206
RAX: 000000000000010d RBX: 0000000000000002 RCX: 00000000ffffff4b
RDX: 0000000000000018 RSI: 00000000004c234c RDI: 00000000ffb3dbff
RBP: ffffa1304013ba50 R08: ffff8af330ad7920 R09: 0000000007fffffa
R10: 0000000000000800 R11: 0000000000000010 R12: ffff8af29d6258c0
R13: ffff8af28b95c700 R14: 0000000000000000 R15: ffff8af29d6258fc
FS: 0000000000000000(0000) GS:ffff8af339ac0000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffff8af350ad7920 CR3: 0000000015ee4000 CR4: 00000000001406e0
Call Trace:
xfrm_input+0x4e5/0xa10
xfrm4_rcv_encap+0xb5/0xe0
xfrm4_udp_encap_rcv+0x140/0x1c0
Analysis revealed offending code is when accessing:
replay_esn->bmp[nr] |= (1U << bitnr);
with 'nr' being 0x07fffffa.
This happened in an SMP system when reordering of packets was present;
A packet arrived with a "too old" sequence number (outside the window,
i.e 'diff > replay_window'), and therefore the following calculation:
bitnr = replay_esn->replay_window - (diff - pos);
yields a negative result, but since bitnr is u32 we get a large unsigned
quantity (in crash dump above: 0xffffff4b seen in ecx).
This was supposed to be protected by xfrm_input()'s former call to:
if (x->repl->check(x, skb, seq)) {
However, the state's spinlock x->lock is *released* after '->check()'
is performed, and gets re-acquired before '->advance()' - which gives a
chance for a different core to update the xfrm state, e.g. by advancing
'replay_esn->seq' when it encounters more packets - leading to a
'diff > replay_window' situation when original core continues to
xfrm_replay_advance_bmp().
An attempt to fix this issue was suggested in commit bcf66bf54aab
("xfrm: Perform a replay check after return from async codepaths"),
by calling 'x->repl->recheck()' after lock is re-acquired, but fix
applied only to asyncronous crypto algorithms.
Augment the fix, by *always* calling 'recheck()' - irrespective if we're
using async crypto.
Fixes: 0ebea8ef3559 ("[IPSEC]: Move state lock into x->type->input")
Signed-off-by: Shmulik Ladkani <shmulik.ladkani@gmail.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Diffstat (limited to 'net/xfrm/xfrm_input.c')
-rw-r--r-- | net/xfrm/xfrm_input.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index 37456d022cfa..61e6220ddd5a 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -660,7 +660,7 @@ resume: /* only the first xfrm gets the encap type */ encap_type = 0; - if (async && x->repl->recheck(x, skb, seq)) { + if (x->repl->recheck(x, skb, seq)) { XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATESEQERROR); goto drop_unlock; } |