stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Daniel Borkmann <daniel@iogearbox.net>,
	Alexei Starovoitov <ast@kernel.org>,
	Martin KaFai Lau <kafai@fb.com>,
	Edward Cree <ecree@solarflare.com>,
	"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 4.9 11/72] bpf: prevent leaking pointer via xadd on unpriviledged
Date: Wed, 19 Jul 2017 12:24:05 +0200	[thread overview]
Message-ID: <20170719102437.488482453@linuxfoundation.org> (raw)
In-Reply-To: <20170719102435.760649060@linuxfoundation.org>

4.9-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Daniel Borkmann <daniel@iogearbox.net>

commit 6bdf6abc56b53103324dfd270a86580306e1a232 upstream.

Leaking kernel addresses on unpriviledged is generally disallowed,
for example, verifier rejects the following:

  0: (b7) r0 = 0
  1: (18) r2 = 0xffff897e82304400
  3: (7b) *(u64 *)(r1 +48) = r2
  R2 leaks addr into ctx

Doing pointer arithmetic on them is also forbidden, so that they
don't turn into unknown value and then get leaked out. However,
there's xadd as a special case, where we don't check the src reg
for being a pointer register, e.g. the following will pass:

  0: (b7) r0 = 0
  1: (7b) *(u64 *)(r1 +48) = r0
  2: (18) r2 = 0xffff897e82304400 ; map
  4: (db) lock *(u64 *)(r1 +48) += r2
  5: (95) exit

We could store the pointer into skb->cb, loose the type context,
and then read it out from there again to leak it eventually out
of a map value. Or more easily in a different variant, too:

   0: (bf) r6 = r1
   1: (7a) *(u64 *)(r10 -8) = 0
   2: (bf) r2 = r10
   3: (07) r2 += -8
   4: (18) r1 = 0x0
   6: (85) call bpf_map_lookup_elem#1
   7: (15) if r0 == 0x0 goto pc+3
   R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R6=ctx R10=fp
   8: (b7) r3 = 0
   9: (7b) *(u64 *)(r0 +0) = r3
  10: (db) lock *(u64 *)(r0 +0) += r6
  11: (b7) r0 = 0
  12: (95) exit

  from 7 to 11: R0=inv,min_value=0,max_value=0 R6=ctx R10=fp
  11: (b7) r0 = 0
  12: (95) exit

Prevent this by checking xadd src reg for pointer types. Also
add a couple of test cases related to this.

Fixes: 1be7f75d1668 ("bpf: enable non-root eBPF programs")
Fixes: 17a5267067f3 ("bpf: verifier (add verifier core)")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 kernel/bpf/verifier.c |    5 +++++
 1 file changed, 5 insertions(+)

--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -885,6 +885,11 @@ static int check_xadd(struct bpf_verifie
 	if (err)
 		return err;
 
+	if (is_pointer_value(env, insn->src_reg)) {
+		verbose("R%d leaks addr into mem\n", insn->src_reg);
+		return -EACCES;
+	}
+
 	/* check whether atomic_add can read the memory */
 	err = check_mem_access(env, insn->dst_reg, insn->off,
 			       BPF_SIZE(insn->code), BPF_READ, -1);

  parent reply	other threads:[~2017-07-19 10:26 UTC|newest]

Thread overview: 67+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-19 10:23 [PATCH 4.9 00/72] 4.9.39-stable review Greg Kroah-Hartman
2017-07-19 10:23 ` [PATCH 4.9 01/72] xen-netfront: Rework the fix for Rx stall during OOM and network stress Greg Kroah-Hartman
2017-07-19 10:23 ` [PATCH 4.9 02/72] net_sched: fix error recovery at qdisc creation Greg Kroah-Hartman
2017-07-19 10:23 ` [PATCH 4.9 03/72] net: sched: Fix one possible panic when no destroy callback Greg Kroah-Hartman
2017-07-19 10:23 ` [PATCH 4.9 04/72] net/phy: micrel: configure intterupts after autoneg workaround Greg Kroah-Hartman
2017-07-19 10:23 ` [PATCH 4.9 05/72] ipv6: avoid unregistering inet6_dev for loopback Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 06/72] net: dp83640: Avoid NULL pointer dereference Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 07/72] tcp: reset sk_rx_dst in tcp_disconnect() Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 08/72] net: prevent sign extension in dev_get_stats() Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 09/72] bridge: mdb: fix leak on complete_info ptr on fail path Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 10/72] rocker: move dereference before free Greg Kroah-Hartman
2017-07-19 10:24 ` Greg Kroah-Hartman [this message]
2017-07-19 10:24 ` [PATCH 4.9 13/72] net/mlx5: Cancel delayed recovery work when unloading the driver Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 14/72] liquidio: fix bug in soft reset failure detection Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 15/72] net/mlx5e: Fix TX carrier errors report in get stats ndo Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 16/72] ipv6: dad: dont remove dynamic addresses if link is down Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 17/72] vxlan: fix hlist corruption Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 18/72] net: core: Fix slab-out-of-bounds in netdev_stats_to_stats64 Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 20/72] vrf: fix bug_on triggered by rx when destroying a vrf Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 21/72] rds: tcp: use sock_create_lite() to create the accept socket Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 23/72] brcmfmac: Fix a memory leak in error handling path in brcmf_cfg80211_attach Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 24/72] brcmfmac: Fix glom_skb leak in brcmf_sdiod_recv_chain Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 25/72] sfc: dont read beyond unicast address list Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 26/72] cfg80211: Define nla_policy for NL80211_ATTR_LOCAL_MESH_POWER_MODE Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 27/72] cfg80211: Validate frequencies nested in NL80211_ATTR_SCAN_FREQUENCIES Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 28/72] cfg80211: Check if PMKID attribute is of expected size Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 29/72] cfg80211: Check if NAN service ID " Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 30/72] irqchip/gic-v3: Fix out-of-bound access in gic_set_affinity Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 31/72] parisc: Report SIGSEGV instead of SIGBUS when running out of stack Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 32/72] parisc: use compat_sys_keyctl() Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 33/72] parisc: DMA API: return error instead of BUG_ON for dma ops on non dma devs Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 34/72] parisc/mm: Ensure IRQs are off in switch_mm() Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 35/72] tools/lib/lockdep: Reduce MAX_LOCK_DEPTH to avoid overflowing lock_chain/: Depth Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 36/72] thp, mm: fix crash due race in MADV_FREE handling Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 37/72] kernel/extable.c: mark core_kernel_text notrace Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 38/72] mm/list_lru.c: fix list_lru_count_node() to be race free Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 39/72] fs/dcache.c: fix spin lockup issue on nlru->lock Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 40/72] checkpatch: silence perl 5.26.0 unescaped left brace warnings Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 41/72] binfmt_elf: use ELF_ET_DYN_BASE only for PIE Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 42/72] arm: move ELF_ET_DYN_BASE to 4MB Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 43/72] arm64: move ELF_ET_DYN_BASE to 4GB / 4MB Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 44/72] powerpc: " Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 45/72] s390: reduce ELF_ET_DYN_BASE Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 46/72] exec: Limit arg stack to at most 75% of _STK_LIM Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 47/72] ARM64: dts: marvell: armada37xx: Fix timer interrupt specifiers Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 48/72] vt: fix unchecked __put_user() in tioclinux ioctls Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 49/72] rcu: Add memory barriers for NOCB leader wakeup Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 50/72] nvmem: core: fix leaks on registration errors Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 51/72] mnt: In umount propagation reparent in a separate pass Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 52/72] mnt: In propgate_umount handle visiting mounts in any order Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 53/72] mnt: Make propagate_umount less slow for overlapping mount propagation trees Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 54/72] selftests/capabilities: Fix the test_execve test Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 57/72] crypto: atmel - only treat EBUSY as transient if backlog Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 58/72] crypto: sha1-ssse3 - Disable avx2 Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 61/72] Revert "sched/core: Optimize SCHED_SMT" Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 62/72] sched/fair, cpumask: Export for_each_cpu_wrap() Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 63/72] sched/topology: Fix building of overlapping sched-groups Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 64/72] sched/topology: Optimize build_group_mask() Greg Kroah-Hartman
2017-07-19 10:24 ` [PATCH 4.9 65/72] sched/topology: Fix overlapping sched_group_mask Greg Kroah-Hartman
2017-07-19 10:25 ` [PATCH 4.9 66/72] PM / wakeirq: Convert to SRCU Greg Kroah-Hartman
2017-07-19 10:25 ` [PATCH 4.9 67/72] PM / QoS: return -EINVAL for bogus strings Greg Kroah-Hartman
2017-07-19 10:25 ` [PATCH 4.9 68/72] tracing: Use SOFTIRQ_OFFSET for softirq dectection for more accurate results Greg Kroah-Hartman
2017-07-19 10:25 ` [PATCH 4.9 72/72] kvm: vmx: allow host to access guest MSR_IA32_BNDCFGS Greg Kroah-Hartman
2017-07-19 15:58 ` [PATCH 4.9 00/72] 4.9.39-stable review Sumit Semwal
2017-07-20  5:08   ` Greg Kroah-Hartman
2017-07-19 20:34 ` Guenter Roeck
2017-07-19 23:39 ` Shuah Khan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170719102437.488482453@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=ast@kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=davem@davemloft.net \
    --cc=ecree@solarflare.com \
    --cc=kafai@fb.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).