On Fri, Jul 21, 2023 at 08:16:14PM +0800, Hou Tao wrote: > > > On 7/20/2023 4:57 PM, Jiri Olsa wrote: > > We received report [1] of kernel crash, which is caused by > > using nesting protection without disabled preemption. > > > > The bpf_event_output can be called by programs executed by > > bpf_prog_run_array_cg function that disabled migration but > > keeps preemption enabled. > > > > This can cause task to be preempted by another one inside the > > nesting protection and lead eventually to two tasks using same > > perf_sample_data buffer and cause crashes like: > > > > BUG: kernel NULL pointer dereference, address: 0000000000000001 > > #PF: supervisor instruction fetch in kernel mode > > #PF: error_code(0x0010) - not-present page > > ... > > ? perf_output_sample+0x12a/0x9a0 > > ? finish_task_switch.isra.0+0x81/0x280 > > ? perf_event_output+0x66/0xa0 > > ? bpf_event_output+0x13a/0x190 > > ? bpf_event_output_data+0x22/0x40 > > ? bpf_prog_dfc84bbde731b257_cil_sock4_connect+0x40a/0xacb > > ? xa_load+0x87/0xe0 > > ? __cgroup_bpf_run_filter_sock_addr+0xc1/0x1a0 > > ? release_sock+0x3e/0x90 > > ? sk_setsockopt+0x1a1/0x12f0 > > ? udp_pre_connect+0x36/0x50 > > ? inet_dgram_connect+0x93/0xa0 > > ? __sys_connect+0xb4/0xe0 > > ? udp_setsockopt+0x27/0x40 > > ? __pfx_udp_push_pending_frames+0x10/0x10 > > ? __sys_setsockopt+0xdf/0x1a0 > > ? __x64_sys_connect+0xf/0x20 > > ? do_syscall_64+0x3a/0x90 > > ? entry_SYSCALL_64_after_hwframe+0x72/0xdc > > > > Fixing this by disabling preemption in bpf_event_output. > > > > [1] https://github.com/cilium/cilium/issues/26756 > > Cc: stable@xxxxxxxxxxxxxxx > > Reported-by: Oleg "livelace" Popov <o.popov@xxxxxxxxxxx> > > Fixes: 2a916f2f546c bpf: Use migrate_disable/enable in array macros and cgroup/lirc code. > > Signed-off-by: Jiri Olsa <jolsa@xxxxxxxxxx> > > Acked-by: Hou Tao <houtao1@xxxxxxxxxx> > > With one nit above. The format of the Fixes tags should be 2a916f2f546c > ("bpf: Use migrate_disable/enable in array macros and cgroup/lirc code.") > right, sorry about that.. should I resend? thanks, jirka
