Re: [PATCH bpf 1/2] bpf: Forbid bpf_ktime_get_coarse_ns and bpf_timer_* in tracing progs
From: Alexei Starovoitov <hidden>
Date: 2021-11-09 21:52:17
Also in:
bpf
On Mon, Nov 8, 2021 at 8:46 AM Dmitrii Banshchikov [off-list ref] wrote:
bpf_ktime_get_coarse_ns() helper uses ktime_get_coarse_ns() time accessor that isn't safe for any context. This results in locking issues:
Please trim the cc. vger thinks it's a spam. The patch didn't reach the mailing list or patchwork.
const struct bpf_func_proto bpf_ktime_get_coarse_ns_proto = {
.func = bpf_ktime_get_coarse_ns,
.gpl_only = false,
.ret_type = RET_INTEGER,
+ .allowed = bpf_ktime_get_coarse_ns_allowed,I think it's easier to move to networking prog types instead of going through a callback.
quoted hunk ↗ jump to hunk
static const struct bpf_func_proto bpf_timer_init_proto = { .func = bpf_timer_init, .gpl_only = true,@@ -1147,6 +1173,7 @@ static const struct bpf_func_proto bpf_timer_init_proto = { .arg1_type = ARG_PTR_TO_TIMER, .arg2_type = ARG_CONST_MAP_PTR, .arg3_type = ARG_ANYTHING, + .allowed = bpf_timer_allowed,
I think disabling timers in check_map_prog_compatibility (similar to how bpf_spin_lock is restricted) would be cleaner.