linux/kernel/bpf
Martin KaFai Lau 7c4cd051ad bpf: Fix syscall's stackmap lookup potential deadlock
The map_lookup_elem used to not acquiring spinlock
in order to optimize the reader.

It was true until commit 557c0c6e7d ("bpf: convert stackmap to pre-allocation")
The syscall's map_lookup_elem(stackmap) calls bpf_stackmap_copy().
bpf_stackmap_copy() may find the elem no longer needed after the copy is done.
If that is the case, pcpu_freelist_push() saves this elem for reuse later.
This push requires a spinlock.

If a tracing bpf_prog got run in the middle of the syscall's
map_lookup_elem(stackmap) and this tracing bpf_prog is calling
bpf_get_stackid(stackmap) which also requires the same pcpu_freelist's
spinlock, it may end up with a dead lock situation as reported by
Eric Dumazet in https://patchwork.ozlabs.org/patch/1030266/

The situation is the same as the syscall's map_update_elem() which
needs to acquire the pcpu_freelist's spinlock and could race
with tracing bpf_prog.  Hence, this patch fixes it by protecting
bpf_stackmap_copy() with this_cpu_inc(bpf_prog_active)
to prevent tracing bpf_prog from running.

A later syscall's map_lookup_elem commit f1a2e44a3a ("bpf: add queue and stack maps")
also acquires a spinlock and races with tracing bpf_prog similarly.
Hence, this patch is forward looking and protects the majority
of the map lookups.  bpf_map_offload_lookup_elem() is the exception
since it is for network bpf_prog only (i.e. never called by tracing
bpf_prog).

Fixes: 557c0c6e7d ("bpf: convert stackmap to pre-allocation")
Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2019-01-31 23:18:21 +01:00
..
Makefile
arraymap.c bpf: pass struct btf pointer to the map_check_btf() callback 2018-12-12 15:33:33 -08:00
bpf_lru_list.c
bpf_lru_list.h
btf.c bpf: btf: allow typedef func_proto 2019-01-29 19:15:32 -08:00
cgroup.c bpf: run bpf programs with preemption disabled 2019-01-31 23:14:55 +01:00
core.c bpf: enable access to ax register also from verifier rewrite 2019-01-02 16:01:24 -08:00
cpumap.c bpf/cpumap: make sure frame_size for build_skb is aligned if headroom isn't 2018-12-20 23:19:12 +01:00
devmap.c
disasm.c
disasm.h
hashtab.c bpf: fix lockdep false positive in percpu_freelist 2019-01-31 23:18:21 +01:00
helpers.c
inode.c
local_storage.c bpf: enable cgroup local storage map pretty print with kind_flag 2018-12-18 01:11:59 +01:00
lpm_trie.c bpf: pass struct btf pointer to the map_check_btf() callback 2018-12-12 15:33:33 -08:00
map_in_map.c bpf: fix inner map masking to prevent oob under speculation 2019-01-18 15:19:56 -08:00
map_in_map.h
offload.c
percpu_freelist.c bpf: fix lockdep false positive in percpu_freelist 2019-01-31 23:18:21 +01:00
percpu_freelist.h bpf: fix lockdep false positive in percpu_freelist 2019-01-31 23:18:21 +01:00
queue_stack_maps.c
reuseport_array.c
stackmap.c bpf: zero out build_id for BPF_STACK_BUILD_ID_IP 2019-01-17 16:42:35 +01:00
syscall.c bpf: Fix syscall's stackmap lookup potential deadlock 2019-01-31 23:18:21 +01:00
tnum.c
verifier.c bpf: fix sanitation of alu op with pointer / scalar type from different paths 2019-01-05 21:32:38 -08:00
xskmap.c