Skip to content

Commit

Permalink
bpf: Send signals asynchronously if !preemptible
Browse files Browse the repository at this point in the history
[ Upstream commit 87c5441 ]

BPF programs can execute in all kinds of contexts and when a program
running in a non-preemptible context uses the bpf_send_signal() kfunc,
it will cause issues because this kfunc can sleep.
Change `irqs_disabled()` to `!preemptible()`.

Reported-by: syzbot+97da3d7e0112d59971de@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/all/67486b09.050a0220.253251.0084.GAE@google.com/
Fixes: 1bc7896 ("bpf: Fix deadlock with rq_lock in bpf_send_signal()")
Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20250115103647.38487-1-puranjay@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
  • Loading branch information
puranjaymohan authored and gregkh committed Feb 8, 2025
1 parent 904c93c commit 092fc76
Showing 1 changed file with 1 addition and 1 deletion.
2 changes: 1 addition & 1 deletion kernel/trace/bpf_trace.c
Original file line number Diff line number Diff line change
Expand Up @@ -854,7 +854,7 @@ static int bpf_send_signal_common(u32 sig, enum pid_type type, struct task_struc
if (unlikely(is_global_init(task)))
return -EPERM;

if (irqs_disabled()) {
if (!preemptible()) {
/* Do an early check on signal validity. Otherwise,
* the error is lost in deferred irq_work.
*/
Expand Down

0 comments on commit 092fc76

Please sign in to comment.