Merge branch 'bpf-track-overlapping-rcu-protection'

Ning Ding says:

====================
bpf: Track overlapping RCU protection

Preemption-disabled and IRQ-disabled regions provide RCU protection, but
the verifier does not account for them. Current implementation can invalidate
a task kptr while another RCU source remains active, or keep it valid
after the final source ends.

Track these regions and invalidate RCU-protected pointers only after the
last protection ends. Add task kptr tests for overlapping protection and
final-exit rejection.

This follows review of the applied spin-unlock fix series [1].

Tested in QEMU/KVM:
  ./test_progs -t task_kfunc
  ./test_progs -t preempt_lock
  ./test_progs -t irq

[1] https://lore.kernel.org/r/20260803112615.3362122-1-dingning04@gmail.com
====================

Link: https://patch.msgid.link/20260805233940.3966981-1-dingning04@gmail.com
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
This commit is contained in:
Kumar Kartikeya Dwivedi 2026-08-06 15:13:40 +02:00
commit 8c7f55d60a
6 changed files with 220 additions and 6 deletions

View File

@ -4452,7 +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_preempt_locks ||
env->cur_state->active_locks ||
env->cur_state->active_irq_id ||
!in_sleepable(env);
}
@ -7166,7 +7168,6 @@ static int process_spin_lock(struct bpf_verifier_env *env, struct bpf_reg_state
return err;
}
} else {
bool was_in_rcu_cs;
void *ptr;
int type;
@ -7194,12 +7195,11 @@ static int process_spin_lock(struct bpf_verifier_env *env, struct bpf_reg_state
verbose(env, "%s_unlock cannot be out of order\n", lock_str);
return -EINVAL;
}
was_in_rcu_cs = in_rcu_cs(env);
if (release_lock_state(cur, type, reg->id, ptr)) {
verbose(env, "%s_unlock of different lock\n", lock_str);
return -EINVAL;
}
if (was_in_rcu_cs && !in_rcu_cs(env))
if (!in_rcu_cs(env))
invalidate_rcu_protected_refs(env);
invalidate_non_owning_refs(env);
@ -11663,6 +11663,9 @@ 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;
}
@ -13159,7 +13162,8 @@ 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++;
@ -13169,6 +13173,8 @@ 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)) {

View File

@ -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",
};

View File

@ -116,9 +116,9 @@ int BPF_PROG(test_cpumask_null, struct task_struct *task, u64 clone_flags)
return 0;
}
SEC("tp_btf/task_newtask")
SEC("?fentry.s/" SYS_PREFIX "sys_getpgid")
__failure __msg("R2 must be a rcu pointer")
int BPF_PROG(test_global_mask_out_of_rcu, struct task_struct *task, u64 clone_flags)
int BPF_PROG(test_global_mask_out_of_rcu)
{
struct bpf_cpumask *local, *prev;
@ -133,6 +133,10 @@ int BPF_PROG(test_global_mask_out_of_rcu, struct task_struct *task, u64 clone_fl
return 0;
}
/*
* Use a sleepable program so explicit RCU is the only source of RCU
* protection.
*/
bpf_rcu_read_lock();
local = global_mask;
if (!local) {

View File

@ -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)
{

View File

@ -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;
}

View File

@ -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)
{