Re: [PATCH bpf-next v2] bpf: Use an on-stack pt_regs in the raw tracepoint helpers

From: Alexei Starovoitov

Date: Thu Nov 19 2026 - 20:11:10 EST


On Thu, Nov 19, 2026 at 02:37 AM Mike Bazov via B4 Relay <devnull+mike.bazov.gmail.com@xxxxxxxxxx> wrote:
> Each caller then has its own registers, however it nests or gets
> preempted, and the helpers no longer fail with -EBUSY when nested more
> than three deep.

bpf-ci is correct.
let's remove that warn_on as part of the series.

we have another hole:

prog 1 on mmap_lock_acquire_returned
bpf_get_stack(BPF_F_USER_STACK | BPF_F_USER_BUILD_ID)
mmap_read_trylock()
prog 2 on mmap_lock_acquire_returned

stack colection with build_id's will disappear.

Let's fix all together with somethign like:

diff --git a/kernel/bpf/mmap_unlock_work.h b/kernel/bpf/mmap_unlock_work.h
index 1834db2..65673d1 100644
--- a/kernel/bpf/mmap_unlock_work.h
+++ b/kernel/bpf/mmap_unlock_work.h
@@ -13,6 +13,7 @@ struct mmap_unlock_irq_work {
struct irq_work irq_work;
struct mm_struct *mm;
atomic_t active;
+ int nest_level;
};

DECLARE_PER_CPU(struct mmap_unlock_irq_work, mmap_unlock_work);
@@ -28,8 +29,13 @@ static inline struct mmap_unlock_irq_work *bpf_mmap_unlock_guard_get(void)
{
struct mmap_unlock_irq_work *work;

- if (!irqs_disabled())
- return NULL;
+ if (!irqs_disabled()) {
+ /* progs on mmap_lock tracepoints can get here again */
+ if (this_cpu_inc_return(mmap_unlock_work.nest_level) <= 3)
+ return NULL;
+ this_cpu_dec(mmap_unlock_work.nest_level);
+ return ERR_PTR(-EBUSY);
+ }

/*
* PREEMPT_RT does not allow to trylock mmap sem in interrupt
@@ -51,12 +57,15 @@ bpf_mmap_unlock_guard_put(struct mmap_unlock_irq_work *work)
{
if (work)
atomic_set_release(&work->active, 0);
+ else
+ this_cpu_dec(mmap_unlock_work.nest_level);
}

static inline void bpf_mmap_unlock_mm(struct mmap_unlock_irq_work *work, struct mm_struct *mm)
{
if (!work) {
mmap_read_unlock(mm);
+ bpf_mmap_unlock_guard_put(NULL);
} else {
work->mm = mm;

diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 3f1901a..5c49c13 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -702,7 +702,7 @@ BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map,
sds = this_cpu_ptr(&bpf_trace_sds);
nest_level = this_cpu_inc_return(bpf_trace_nest_level);

- if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(sds->sds))) {
+ if (nest_level > ARRAY_SIZE(sds->sds)) {
err = -EBUSY;
goto out;
}

pw-bot: cr