Re: [PATCH v16 03/20] unwind_user/sframe: Store .sframe section data in per-mm maple tree

[Steven Rostedt]

On Thu, 21 May 2026 16:25:29 +0200
Jens Remus <jremus@xxxxxxxxxxxxx> wrote:

>  int sframe_remove_section(unsigned long sframe_start)
>  {
> -	return -ENOSYS;
> +	struct mm_struct *mm = current->mm;
> +	struct sframe_section *sec;
> +	unsigned long index = 0;
> +	bool found = false;
> +	int ret = 0;
> +
> +	guard(srcu)(&sframe_srcu);
> +
> +	mt_for_each(&mm->sframe_mt, sec, index, ULONG_MAX) {
> +		if (sec->sframe_start == sframe_start) {
> +			found = true;
> +			ret |= __sframe_remove_section(mm, sec);

Because this is all internal data, the __sframe_remove_section() should
never fail. Perhaps we should add a WARN_ON() if it does?

		if (sec->sframe_start == sframe_start) {
			ret |= __sframe_remove_section(mm, sec);
			WARN_ON(!found && ret);
			found = true;

-- Steve



> +		}
> +	}
> +
> +	if (!found || ret)
> +		return -EINVAL;
> +
> +	return 0;
> +}
> +