From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
To: bpf@vger.kernel.org
Cc: Alexei Starovoitov <ast@kernel.org>,
Andrii Nakryiko <andrii@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Eduard Zingerman <eddyz87@gmail.com>,
Emil Tsalapatis <emil@etsalapatis.com>,
kkd@meta.com, kernel-team@meta.com
Subject: [PATCH bpf-next v1 1/2] bpf: Fix sleepable context checks and remove in_sleepable helper
Date: Thu, 6 Aug 2026 18:40:48 +0200 [thread overview]
Message-ID: <20260806164049.3158887-2-memxor@gmail.com> (raw)
In-Reply-To: <20260806164049.3158887-1-memxor@gmail.com>
The verifier has two helpers to determine if the current program can
sleep:
- in_sleepable() checks only env->cur_state->in_sleepable
- in_sleepable_context() checks active RCU, preempt, lock and IRQ
state in addition to in_sleepable
Several call sites incorrectly used in_sleepable() where
in_sleepable_context() is required. This allowed sleepable programs
in non-sleepable contexts (e.g., inside bpf_rcu_read_lock()) to
incorrectly use sleepable iterators and kfuncs:
- check_css_task_iter_allowlist() returned in_sleepable() and
therefore allowed css_task iterator in any sleepable program,
even inside RCU/preempt/lock/IRQ-disabled regions. Fix it to
use in_sleepable_context().
- The first sleepable check in check_kfunc_call() used
in_sleepable() and only verified the program flag, not the
current context. Replace it with in_sleepable_context() so
sleepable kfuncs are rejected when called from any
non-sleepable context, consistent with the second check in the
same function.
To prevent future misuse, remove the in_sleepable() helper
completely and replace its legitimate uses with a direct check of
env->cur_state->in_sleepable. We do not want anyone to test this
bit directly and in isolation; all sleepable-context decisions must
go through in_sleepable_context(), which accounts for locks and
preemption. The remaining direct checks are in in_rcu_cs() and
in_sleepable_context() itself, where testing the bit in isolation
is intentional.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
---
kernel/bpf/verifier.c | 13 ++++---------
1 file changed, 4 insertions(+), 9 deletions(-)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index e6233c0081d1..3dadf2b33a36 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4441,11 +4441,6 @@ static int map_kptr_match_type(struct bpf_verifier_env *env,
return -EINVAL;
}
-static bool in_sleepable(struct bpf_verifier_env *env)
-{
- return env->cur_state->in_sleepable;
-}
-
/* The non-sleepable programs and sleepable programs with explicit bpf_rcu_read_lock()
* can dereference RCU protected pointers and result is PTR_TRUSTED.
*/
@@ -4455,7 +4450,7 @@ static bool in_rcu_cs(struct bpf_verifier_env *env)
env->cur_state->active_preempt_locks ||
env->cur_state->active_locks ||
env->cur_state->active_irq_id ||
- !in_sleepable(env);
+ !env->cur_state->in_sleepable;
}
/* Once GCC supports btf_type_tag the following mechanism will be replaced with tag check */
@@ -10258,7 +10253,7 @@ static inline bool in_sleepable_context(struct bpf_verifier_env *env)
!env->cur_state->active_preempt_locks &&
!env->cur_state->active_locks &&
!env->cur_state->active_irq_id &&
- in_sleepable(env);
+ env->cur_state->in_sleepable;
}
static const char *non_sleepable_context_description(struct bpf_verifier_env *env)
@@ -12083,7 +12078,7 @@ static bool check_css_task_iter_allowlist(struct bpf_verifier_env *env)
return true;
fallthrough;
default:
- return in_sleepable(env);
+ return in_sleepable_context(env);
}
}
@@ -13090,7 +13085,7 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
}
sleepable = bpf_is_kfunc_sleepable(&meta);
- if (sleepable && !in_sleepable(env)) {
+ if (sleepable && !in_sleepable_context(env)) {
verbose(env, "program must be sleepable to call sleepable kfunc %s\n", func_name);
return -EACCES;
}
--
2.53.0-Meta
next prev parent reply other threads:[~2026-08-06 16:40 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-06 16:40 [PATCH bpf-next v1 0/2] Fix sleepable context checks in verifier Kumar Kartikeya Dwivedi
2026-08-06 16:40 ` Kumar Kartikeya Dwivedi [this message]
2026-08-06 17:14 ` [PATCH bpf-next v1 1/2] bpf: Fix sleepable context checks and remove in_sleepable helper sashiko-bot
2026-08-06 20:31 ` Eduard Zingerman
2026-08-06 21:22 ` Kumar Kartikeya Dwivedi
2026-08-06 22:17 ` Eduard Zingerman
2026-08-06 21:10 ` Eduard Zingerman
2026-08-06 21:20 ` Kumar Kartikeya Dwivedi
2026-08-06 21:46 ` Kumar Kartikeya Dwivedi
2026-08-06 16:40 ` [PATCH bpf-next v1 2/2] selftests/bpf: Test sleepable context distinctions Kumar Kartikeya Dwivedi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260806164049.3158887-2-memxor@gmail.com \
--to=memxor@gmail.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=eddyz87@gmail.com \
--cc=emil@etsalapatis.com \
--cc=kernel-team@meta.com \
--cc=kkd@meta.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.