linux_dsm_epyc7002/kernel/bpf
Jakub Kicinski 7640ead939 bpf: verifier: make sure callees don't prune with caller differences
Currently for liveness and state pruning the register parentage
chains don't include states of the callee.  This makes some sense
as the callee can't access those registers.  However, this means
that READs done after the callee returns will not propagate into
the states of the callee.  Callee will then perform pruning
disregarding differences in caller state.

Example:

   0: (85) call bpf_user_rnd_u32
   1: (b7) r8 = 0
   2: (55) if r0 != 0x0 goto pc+1
   3: (b7) r8 = 1
   4: (bf) r1 = r8
   5: (85) call pc+4
   6: (15) if r8 == 0x1 goto pc+1
   7: (05) *(u64 *)(r9 - 8) = r3
   8: (b7) r0 = 0
   9: (95) exit

   10: (15) if r1 == 0x0 goto pc+0
   11: (95) exit

Here we acquire unknown state with call to get_random() [1].  Then
we store this random state in r8 (either 0 or 1) [1 - 3], and make
a call on line 5.  Callee does nothing but a trivial conditional
jump (to create a pruning point).  Upon return caller checks the
state of r8 and either performs an unsafe read or not.

Verifier will first explore the path with r8 == 1, creating a pruning
point at [11].  The parentage chain for r8 will include only callers
states so once verifier reaches [6] it will mark liveness only on states
in the caller, and not [11].  Now when verifier walks the paths with
r8 == 0 it will reach [11] and since REG_LIVE_READ on r8 was not
propagated there it will prune the walk entirely (stop walking
the entire program, not just the callee).  Since [6] was never walked
with r8 == 0, [7] will be considered dead and replaced with "goto -1"
causing hang at runtime.

This patch weaves the callee's explored states onto the callers
parentage chain.  Rough parentage for r8 would have looked like this
before:

[0] [1] [2] [3] [4] [5]   [10]      [11]      [6]      [7]
     |           |      ,---|----.    |        |        |
  sl0:         sl0:    / sl0:     \ sl0:      sl0:     sl0:
  fr0: r8 <-- fr0: r8<+--fr0: r8   `fr0: r8  ,fr0: r8<-fr0: r8
                       \ fr1: r8 <- fr1: r8 /
                        \__________________/

after:

[0] [1] [2] [3] [4] [5]   [10]      [11]      [6]      [7]
     |           |          |         |        |        |
   sl0:         sl0:      sl0:       sl0:      sl0:     sl0:
   fr0: r8 <-- fr0: r8 <- fr0: r8 <- fr0: r8 <-fr0: r8<-fr0: r8
                          fr1: r8 <- fr1: r8

Now the mark from instruction 6 will travel through callees states.

Note that we don't have to connect r0 because its overwritten by
callees state on return and r1 - r5 because those are not alive
any more once a call is made.

v2:
 - don't connect the callees registers twice (Alexei: suggestion & code)
 - add more details to the comment (Ed & Alexei)
v1: don't unnecessarily link caller saved regs (Jiong)

Fixes: f4d7e40a5b ("bpf: introduce function calls (verification)")
Reported-by: David Beckett <david.beckett@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-12-13 10:35:40 -08:00
..
arraymap.c bpf: return EOPNOTSUPP when map lookup isn't supported 2018-10-09 21:52:20 -07:00
bpf_lru_list.c
bpf_lru_list.h bpf: Only set node->ref = 1 if it has not been set 2017-09-01 09:57:39 -07:00
btf.c bpf: btf: check name validity for various types 2018-11-28 16:03:04 -08:00
cgroup.c bpf: add cg_skb_is_valid_access for BPF_PROG_TYPE_CGROUP_SKB 2018-10-19 13:49:34 -07:00
core.c bpf: fix bpf_jit_limit knob for PAGE_SIZE >= 64K 2018-12-11 19:12:21 -08:00
cpumap.c bpf: fix redirect to map under tail calls 2018-08-17 15:56:23 -07:00
devmap.c bpf: devmap: fix wrong interface selection in notifier_call 2018-10-26 00:32:21 +02:00
disasm.c bpf: Remove struct bpf_verifier_env argument from print_bpf_insn 2018-03-23 17:38:57 +01:00
disasm.h bpf: Remove struct bpf_verifier_env argument from print_bpf_insn 2018-03-23 17:38:57 +01:00
hashtab.c bpf: add bpffs pretty print for percpu arraymap/hash/lru_hash 2018-08-30 14:03:53 +02:00
helpers.c bpf: fix direct packet write into pop/peek helpers 2018-10-25 17:02:06 -07:00
inode.c bpf: decouple btf from seq bpf fs dump and enable more maps 2018-08-13 00:52:45 +02:00
local_storage.c bpf: allocate local storage buffers using GFP_ATOMIC 2018-11-16 21:17:49 -08:00
lpm_trie.c bpf: decouple btf from seq bpf fs dump and enable more maps 2018-08-13 00:52:45 +02:00
Makefile bpf: add queue and stack maps 2018-10-19 13:24:31 -07:00
map_in_map.c bpf: don't allow create maps of per-cpu cgroup local storages 2018-10-01 16:18:33 +02:00
map_in_map.h
offload.c bpf: add verifier callback to get stack usage info for offloaded progs 2018-10-08 10:24:12 +02:00
percpu_freelist.c bpf: fix lockdep splat 2017-11-15 19:46:32 +09:00
percpu_freelist.h
queue_stack_maps.c bpf: fix integer overflow in queue_stack_map 2018-11-22 21:29:40 +01:00
reuseport_array.c bpf: Introduce BPF_MAP_TYPE_REUSEPORT_SOCKARRAY 2018-08-11 01:58:46 +02:00
stackmap.c bpf: rename stack trace map operations 2018-10-19 13:24:30 -07:00
syscall.c bpf: fix bpf_prog_get_info_by_fd to return 0 func_lens for unpriv 2018-11-02 13:51:15 -07:00
tnum.c bpf/verifier: improve register value range tracking with ARSH 2018-04-29 08:45:53 -07:00
verifier.c bpf: verifier: make sure callees don't prune with caller differences 2018-12-13 10:35:40 -08:00
xskmap.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2018-10-19 11:03:06 -07:00