[PATCH bpf-next 1/2] bpf: Account for preempt and IRQ state in RCU protection

[PATCH bpf-next 1/2] bpf: Account for preempt and IRQ state in RCU protection

[Ning Ding]

Disabling preemption or local IRQs keeps the current CPU in an RCU
read-side critical section, but in_rcu_cs() does not account for either
state. The verifier therefore rejects safe kptr accesses and invalidates
pointers when another RCU source ends.

Include preemption-disabled and IRQ-disabled state in in_rcu_cs().
Invalidate RCU-protected pointers on RCU unlock, preempt enable, or IRQ
restore only after the final protection ends.

Signed-off-by: Ning Ding <dingning04@gmail.com>
---
 kernel/bpf/verifier.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index d925197c2e5f7..a1005783db300 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4452,8 +4452,9 @@ static bool in_sleepable(struct bpf_verifier_env *env)
 static bool in_rcu_cs(struct bpf_verifier_env *env)
 {
 	return env->cur_state->active_rcu_locks ||
-	       env->cur_state->active_locks ||
-	       !in_sleepable(env);
+	       env->cur_state->active_irq_id ||
+	       env->cur_state->active_preempt_locks ||
+	       env->cur_state->active_locks || !in_sleepable(env);
 }
 
 /* Once GCC supports btf_type_tag the following mechanism will be replaced with tag check */
@@ -11663,6 +11664,9 @@ static int process_irq_flag(struct bpf_verifier_env *env, struct bpf_reg_state *
 		err = unmark_stack_slot_irq_flag(env, reg, kfunc_class);
 		if (err)
 			return err;
+
+		if (!in_rcu_cs(env))
+			invalidate_rcu_protected_refs(env);
 	}
 	return 0;
 }
@@ -13181,7 +13185,8 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 			verbose(env, "unmatched rcu read unlock (kernel function %s)\n", func_name);
 			return -EINVAL;
 		}
-		if (--env->cur_state->active_rcu_locks == 0)
+		env->cur_state->active_rcu_locks--;
+		if (!in_rcu_cs(env))
 			invalidate_rcu_protected_refs(env);
 	} else if (preempt_disable) {
 		env->cur_state->active_preempt_locks++;
@@ -13191,6 +13196,8 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 			return -EINVAL;
 		}
 		env->cur_state->active_preempt_locks--;
+		if (!in_rcu_cs(env))
+			invalidate_rcu_protected_refs(env);
 	}
 
 	if (sleepable && !in_sleepable_context(env)) {
-- 
2.43.0

[PATCH bpf-next 2/2] selftests/bpf: Test overlapping RCU protection

[Ning Ding]

Add task kptr tests that keep RCU protection active after a spin or RCU
unlock when preemption or IRQs remain disabled.

Also test the reverse order with explicit RCU. Verify that task kptrs are
rejected after leaving the final preemption-disabled or IRQ-disabled
region.

Signed-off-by: Ning Ding <dingning04@gmail.com>
---
 .../selftests/bpf/prog_tests/task_kfunc.c     |   6 +
 .../selftests/bpf/progs/task_kfunc_common.h   |   2 +
 .../selftests/bpf/progs/task_kfunc_failure.c  |  49 ++++++
 .../selftests/bpf/progs/task_kfunc_success.c  | 147 ++++++++++++++++++
 4 files changed, 204 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/task_kfunc.c b/tools/testing/selftests/bpf/prog_tests/task_kfunc.c
index fbd7855712c1a..30d403028f984 100644
--- a/tools/testing/selftests/bpf/prog_tests/task_kfunc.c
+++ b/tools/testing/selftests/bpf/prog_tests/task_kfunc.c
@@ -178,6 +178,12 @@ static const char * const success_tests[] = {
 	"task_kfunc_acquire_trusted_walked",
 	"task_kfunc_acquire_after_spin_unlock_non_sleepable",
 	"task_kfunc_acquire_after_spin_unlock_explicit_rcu",
+	"task_kfunc_acquire_after_spin_unlock_preempt_disabled",
+	"task_kfunc_acquire_after_spin_unlock_irq_disabled",
+	"task_kfunc_acquire_after_rcu_unlock_preempt_disabled",
+	"task_kfunc_acquire_after_rcu_unlock_irq_disabled",
+	"task_kfunc_acquire_after_preempt_enable_explicit_rcu",
+	"task_kfunc_acquire_after_irq_restore_explicit_rcu",
 	"test_task_kfunc_flavor_relo",
 	"test_task_kfunc_flavor_relo_not_found",
 };
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_common.h b/tools/testing/selftests/bpf/progs/task_kfunc_common.h
index 052c9d0e3e2a8..a0c599b58c290 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_common.h
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_common.h
@@ -38,6 +38,8 @@ struct task_struct *bpf_task_from_pid(s32 pid) __ksym;
 struct task_struct *bpf_task_from_vpid(s32 vpid) __ksym;
 void bpf_rcu_read_lock(void) __ksym;
 void bpf_rcu_read_unlock(void) __ksym;
+void bpf_local_irq_save(unsigned long *flags) __weak __ksym;
+void bpf_local_irq_restore(unsigned long *flags) __weak __ksym;
 
 static inline struct __tasks_kfunc_map_value *tasks_kfunc_map_value_lookup(struct task_struct *p)
 {
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
index c0e7216b34193..f96b0c13ed1a5 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
@@ -402,3 +402,52 @@ int BPF_PROG(task_kfunc_acquire_after_final_spin_unlock)
 		bpf_task_release(acquired);
 	return 0;
 }
+
+SEC("?fentry.s/" SYS_PREFIX "sys_getpgid")
+__failure __msg("R1 must be a rcu pointer")
+int BPF_PROG(task_kfunc_acquire_after_preempt_enable)
+{
+	struct task_kptr_lock_value *v;
+	struct task_struct *task, *acquired;
+	int key = 0;
+
+	v = bpf_map_lookup_elem(&task_kptr_lock_map, &key);
+	if (!v)
+		return 0;
+
+	bpf_preempt_disable();
+	task = v->task;
+	bpf_preempt_enable();
+	if (!task)
+		return 0;
+
+	acquired = bpf_task_acquire(task);
+	if (acquired)
+		bpf_task_release(acquired);
+	return 0;
+}
+
+SEC("?fentry.s/" SYS_PREFIX "sys_getpgid")
+__failure __msg("R1 must be a rcu pointer")
+int BPF_PROG(task_kfunc_acquire_after_irq_restore)
+{
+	struct task_kptr_lock_value *v;
+	struct task_struct *task, *acquired;
+	unsigned long flags;
+	int key = 0;
+
+	v = bpf_map_lookup_elem(&task_kptr_lock_map, &key);
+	if (!v)
+		return 0;
+
+	bpf_local_irq_save(&flags);
+	task = v->task;
+	bpf_local_irq_restore(&flags);
+	if (!task)
+		return 0;
+
+	acquired = bpf_task_acquire(task);
+	if (acquired)
+		bpf_task_release(acquired);
+	return 0;
+}
diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_success.c b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
index 2bab7634c9dfd..6545b124dee14 100644
--- a/tools/testing/selftests/bpf/progs/task_kfunc_success.c
+++ b/tools/testing/selftests/bpf/progs/task_kfunc_success.c
@@ -414,6 +414,153 @@ int BPF_PROG(task_kfunc_acquire_after_spin_unlock_explicit_rcu)
 	return 0;
 }
 
+SEC("fentry.s/" SYS_PREFIX "sys_getpgid")
+int BPF_PROG(task_kfunc_acquire_after_spin_unlock_preempt_disabled)
+{
+	struct task_kptr_lock_value *v;
+	struct task_struct *task, *acquired;
+	int key = 0;
+
+	v = bpf_map_lookup_elem(&task_kptr_lock_map, &key);
+	if (!v)
+		return 0;
+
+	bpf_preempt_disable();
+	bpf_spin_lock(&v->lock);
+	task = v->task;
+	bpf_spin_unlock(&v->lock);
+	if (task) {
+		acquired = bpf_task_acquire(task);
+		if (acquired)
+			bpf_task_release(acquired);
+	}
+	bpf_preempt_enable();
+	return 0;
+}
+
+SEC("fentry.s/" SYS_PREFIX "sys_getpgid")
+int BPF_PROG(task_kfunc_acquire_after_spin_unlock_irq_disabled)
+{
+	struct task_kptr_lock_value *v;
+	struct task_struct *task, *acquired;
+	unsigned long flags;
+	int key = 0;
+
+	v = bpf_map_lookup_elem(&task_kptr_lock_map, &key);
+	if (!v)
+		return 0;
+
+	bpf_local_irq_save(&flags);
+	bpf_spin_lock(&v->lock);
+	task = v->task;
+	bpf_spin_unlock(&v->lock);
+	if (task) {
+		acquired = bpf_task_acquire(task);
+		if (acquired)
+			bpf_task_release(acquired);
+	}
+	bpf_local_irq_restore(&flags);
+	return 0;
+}
+
+SEC("fentry.s/" SYS_PREFIX "sys_getpgid")
+int BPF_PROG(task_kfunc_acquire_after_rcu_unlock_preempt_disabled)
+{
+	struct task_kptr_lock_value *v;
+	struct task_struct *task, *acquired;
+	int key = 0;
+
+	v = bpf_map_lookup_elem(&task_kptr_lock_map, &key);
+	if (!v)
+		return 0;
+
+	bpf_preempt_disable();
+	bpf_rcu_read_lock();
+	task = v->task;
+	bpf_rcu_read_unlock();
+	if (task) {
+		acquired = bpf_task_acquire(task);
+		if (acquired)
+			bpf_task_release(acquired);
+	}
+	bpf_preempt_enable();
+	return 0;
+}
+
+SEC("fentry.s/" SYS_PREFIX "sys_getpgid")
+int BPF_PROG(task_kfunc_acquire_after_rcu_unlock_irq_disabled)
+{
+	struct task_kptr_lock_value *v;
+	struct task_struct *task, *acquired;
+	unsigned long flags;
+	int key = 0;
+
+	v = bpf_map_lookup_elem(&task_kptr_lock_map, &key);
+	if (!v)
+		return 0;
+
+	bpf_local_irq_save(&flags);
+	bpf_rcu_read_lock();
+	task = v->task;
+	bpf_rcu_read_unlock();
+	if (task) {
+		acquired = bpf_task_acquire(task);
+		if (acquired)
+			bpf_task_release(acquired);
+	}
+	bpf_local_irq_restore(&flags);
+	return 0;
+}
+
+SEC("fentry.s/" SYS_PREFIX "sys_getpgid")
+int BPF_PROG(task_kfunc_acquire_after_preempt_enable_explicit_rcu)
+{
+	struct task_kptr_lock_value *v;
+	struct task_struct *task, *acquired;
+	int key = 0;
+
+	v = bpf_map_lookup_elem(&task_kptr_lock_map, &key);
+	if (!v)
+		return 0;
+
+	bpf_preempt_disable();
+	task = v->task;
+	bpf_rcu_read_lock();
+	bpf_preempt_enable();
+	if (task) {
+		acquired = bpf_task_acquire(task);
+		if (acquired)
+			bpf_task_release(acquired);
+	}
+	bpf_rcu_read_unlock();
+	return 0;
+}
+
+SEC("fentry.s/" SYS_PREFIX "sys_getpgid")
+int BPF_PROG(task_kfunc_acquire_after_irq_restore_explicit_rcu)
+{
+	struct task_kptr_lock_value *v;
+	struct task_struct *task, *acquired;
+	unsigned long flags;
+	int key = 0;
+
+	v = bpf_map_lookup_elem(&task_kptr_lock_map, &key);
+	if (!v)
+		return 0;
+
+	bpf_local_irq_save(&flags);
+	task = v->task;
+	bpf_rcu_read_lock();
+	bpf_local_irq_restore(&flags);
+	if (task) {
+		acquired = bpf_task_acquire(task);
+		if (acquired)
+			bpf_task_release(acquired);
+	}
+	bpf_rcu_read_unlock();
+	return 0;
+}
+
 SEC("syscall")
 int test_task_from_vpid_current(const void *ctx)
 {
-- 
2.43.0