From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Jann Horn <jannh@google.com>,
Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 4.4 18/73] bpf: fix check_map_func_compatibility logic
Date: Mon, 16 May 2016 18:14:48 -0700 [thread overview]
Message-ID: <20160517011452.765162134@linuxfoundation.org> (raw)
In-Reply-To: <20160517011451.827433776@linuxfoundation.org>
4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alexei Starovoitov <ast@fb.com>
[ Upstream commit 6aff67c85c9e5a4bc99e5211c1bac547936626ca ]
The commit 35578d798400 ("bpf: Implement function bpf_perf_event_read() that get the selected hardware PMU conuter")
introduced clever way to check bpf_helper<->map_type compatibility.
Later on commit a43eec304259 ("bpf: introduce bpf_perf_event_output() helper") adjusted
the logic and inadvertently broke it.
Get rid of the clever bool compare and go back to two-way check
from map and from helper perspective.
Fixes: a43eec304259 ("bpf: introduce bpf_perf_event_output() helper")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
kernel/bpf/verifier.c | 53 ++++++++++++++++++++++++++++++--------------------
1 file changed, 32 insertions(+), 21 deletions(-)
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -239,15 +239,6 @@ static const char * const reg_type_str[]
[CONST_IMM] = "imm",
};
-static const struct {
- int map_type;
- int func_id;
-} func_limit[] = {
- {BPF_MAP_TYPE_PROG_ARRAY, BPF_FUNC_tail_call},
- {BPF_MAP_TYPE_PERF_EVENT_ARRAY, BPF_FUNC_perf_event_read},
- {BPF_MAP_TYPE_PERF_EVENT_ARRAY, BPF_FUNC_perf_event_output},
-};
-
static void print_verifier_state(struct verifier_env *env)
{
enum bpf_reg_type t;
@@ -898,24 +889,44 @@ static int check_func_arg(struct verifie
static int check_map_func_compatibility(struct bpf_map *map, int func_id)
{
- bool bool_map, bool_func;
- int i;
-
if (!map)
return 0;
- for (i = 0; i < ARRAY_SIZE(func_limit); i++) {
- bool_map = (map->map_type == func_limit[i].map_type);
- bool_func = (func_id == func_limit[i].func_id);
- /* only when map & func pair match it can continue.
- * don't allow any other map type to be passed into
- * the special func;
- */
- if (bool_func && bool_map != bool_func)
- return -EINVAL;
+ /* We need a two way check, first is from map perspective ... */
+ switch (map->map_type) {
+ case BPF_MAP_TYPE_PROG_ARRAY:
+ if (func_id != BPF_FUNC_tail_call)
+ goto error;
+ break;
+ case BPF_MAP_TYPE_PERF_EVENT_ARRAY:
+ if (func_id != BPF_FUNC_perf_event_read &&
+ func_id != BPF_FUNC_perf_event_output)
+ goto error;
+ break;
+ default:
+ break;
+ }
+
+ /* ... and second from the function itself. */
+ switch (func_id) {
+ case BPF_FUNC_tail_call:
+ if (map->map_type != BPF_MAP_TYPE_PROG_ARRAY)
+ goto error;
+ break;
+ case BPF_FUNC_perf_event_read:
+ case BPF_FUNC_perf_event_output:
+ if (map->map_type != BPF_MAP_TYPE_PERF_EVENT_ARRAY)
+ goto error;
+ break;
+ default:
+ break;
}
return 0;
+error:
+ verbose("cannot pass map_type %d into func %d\n",
+ map->map_type, func_id);
+ return -EINVAL;
}
static int check_call(struct verifier_env *env, int func_id)
next prev parent reply other threads:[~2016-05-17 1:58 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-17 1:14 [PATCH 4.4 00/73] 4.4.11-stable review Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 01/73] decnet: Do not build routes to devices without decnet private data Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 02/73] route: do not cache fib route info on local routes with oif Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 03/73] packet: fix heap info leak in PACKET_DIAG_MCLIST sock_diag interface Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 04/73] net: sched: do not requeue a NULL skb Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 05/73] bpf/verifier: reject invalid LD_ABS | BPF_DW instruction Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 08/73] net: use skb_postpush_rcsum instead of own implementations Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 09/73] vlan: pull on __vlan_insert_tag error path and fix csum correction Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 10/73] atl2: Disable unimplemented scatter/gather feature Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 11/73] openvswitch: use flow protocol when recalculating ipv6 checksums Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 12/73] net/mlx5e: Devices mtu field is u16 and not int Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 13/73] net/mlx5e: Fix minimum MTU Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 14/73] ipv4/fib: dont warn when primary address is missing if in_dev is dead Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 15/73] net/mlx4_en: fix spurious timestamping callbacks Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 16/73] bpf: fix double-fdput in replace_map_fd_with_map_ptr() Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 17/73] bpf: fix refcnt overflow Greg Kroah-Hartman
2016-05-17 1:14 ` Greg Kroah-Hartman [this message]
2016-05-17 1:14 ` [PATCH 4.4 19/73] samples/bpf: fix trace_output example Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 20/73] net: Implement net_dbg_ratelimited() for CONFIG_DYNAMIC_DEBUG case Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 21/73] gre: do not pull header in ICMP error processing Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 22/73] net_sched: introduce qdisc_replace() helper Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 23/73] net_sched: update hierarchical backlog too Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 24/73] sch_htb: update backlog as well Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 25/73] sch_dsmark: " Greg Kroah-Hartman
2016-05-17 1:14 ` [PATCH 4.4 26/73] netem: Segment GSO packets on enqueue Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 30/73] net/mlx4_en: Fix endianness bug in IPV6 csum calculation Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 31/73] VSOCK: do not disconnect socket when peer has shutdown SEND only Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 32/73] net: bridge: fix old ioctl unlocked net device walk Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 35/73] net: fix a kernel infoleak in x25 module Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 36/73] net: thunderx: avoid exposing kernel stack Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 37/73] tcp: refresh skb timestamp at retransmit time Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 38/73] net/route: enforce hoplimit max value Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 39/73] ocfs2: revert using ocfs2_acl_chmod to avoid inode cluster lock hang Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 40/73] ocfs2: fix posix_acl_create deadlock Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 41/73] zsmalloc: fix zs_can_compact() integer overflow Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 42/73] s390/mm: fix asce_bits handling with dynamic pagetable levels Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 43/73] crypto: qat - fix invalid pf2vf_resp_wq logic Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 44/73] crypto: hash - Fix page length clamping in hash walk Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 45/73] crypto: testmgr - Use kmalloc memory for RSA input Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 46/73] ALSA: usb-audio: Quirk for yet another Phoenix Audio devices (v2) Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 47/73] ALSA: usb-audio: Yet another Phoneix Audio device quirk Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 48/73] ALSA: hda - Fix subwoofer pin on ASUS N751 and N551 Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 49/73] ALSA: hda - Fix white noise on Asus UX501VW headset Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 50/73] ALSA: hda - Fix broken reconfig Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 51/73] spi: pxa2xx: Do not detect number of enabled chip selects on Intel SPT Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 52/73] spi: spi-ti-qspi: Fix FLEN and WLEN settings if bits_per_word is overridden Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 53/73] spi: spi-ti-qspi: Handle truncated frames properly Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 54/73] pinctrl: at91-pio4: fix pull-up/down logic Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 55/73] regmap: spmi: Fix regmap_spmi_ext_read in multi-byte case Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 56/73] perf/core: Disable the event on a truncated AUX record Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 57/73] vfs: add vfs_select_inode() helper Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 58/73] vfs: rename: check backing inode being equal Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 59/73] ARM: dts: at91: sam9x5: Fix the memory range assigned to the PMC Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 61/73] regulator: s2mps11: Fix invalid selector mask and voltages for buck9 Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 62/73] regulator: axp20x: Fix axp22x ldo_io voltage ranges Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 63/73] atomic_open(): fix the handling of create_error Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 64/73] qla1280: Dont allocate 512kb of host tags Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 65/73] tools lib traceevent: Do not reassign parg after collapse_tree() Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 66/73] get_rock_ridge_filename(): handle malformed NM entries Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 67/73] Input: max8997-haptic - fix NULL pointer dereference Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 68/73] Revert "[media] videobuf2-v4l2: Verify planes array in buffer dequeueing" Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 69/73] drm/radeon: fix PLL sharing on DCE6.1 (v2) Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 70/73] drm/i915: Bail out of pipe config compute loop on LPT Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 72/73] drm/radeon: fix DP link training issue with second 4K monitor Greg Kroah-Hartman
2016-05-17 1:15 ` [PATCH 4.4 73/73] nf_conntrack: avoid kernel pointer value leak in slab name Greg Kroah-Hartman
2016-05-17 17:27 ` [PATCH 4.4 00/73] 4.4.11-stable review Guenter Roeck
2016-05-17 17:28 ` 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=20160517011452.765162134@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=ast@kernel.org \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=jannh@google.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