diff options
| author | Alexei Starovoitov <ast@kernel.org> | 2019-01-28 18:43:34 -0800 |
|---|---|---|
| committer | Daniel Borkmann <daniel@iogearbox.net> | 2019-01-31 23:14:55 +0100 |
| commit | 6cab5e90ab2bd323c9f3811b6c70a4687df51e27 (patch) | |
| tree | 0fc39a3f8010ea54edab9f8d70014b447ccb6bd2 /kernel/bpf | |
| parent | 1bb54c4071f585ebef56ce8fdfe6026fa2cbcddd (diff) | |
bpf: run bpf programs with preemption disabled
Disabled preemption is necessary for proper access to per-cpu maps
from BPF programs.
But the sender side of socket filters didn't have preemption disabled:
unix_dgram_sendmsg->sk_filter->sk_filter_trim_cap->bpf_prog_run_save_cb->BPF_PROG_RUN
and a combination of af_packet with tun device didn't disable either:
tpacket_snd->packet_direct_xmit->packet_pick_tx_queue->ndo_select_queue->
tun_select_queue->tun_ebpf_select_queue->bpf_prog_run_clear_cb->BPF_PROG_RUN
Disable preemption before executing BPF programs (both classic and extended).
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Diffstat (limited to 'kernel/bpf')
| -rw-r--r-- | kernel/bpf/cgroup.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c index ab612fe9862f..d17d05570a3f 100644 --- a/kernel/bpf/cgroup.c +++ b/kernel/bpf/cgroup.c @@ -572,7 +572,7 @@ int __cgroup_bpf_run_filter_skb(struct sock *sk, bpf_compute_and_save_data_end(skb, &saved_data_end); ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[type], skb, - bpf_prog_run_save_cb); + __bpf_prog_run_save_cb); bpf_restore_data_end(skb, saved_data_end); __skb_pull(skb, offset); skb->sk = save_sk; |