Re: [RFC/PATCH] printk: detect incomplete lines

[Joe Perches]

On Mon, 2008-05-12 at 13:49 +0200, Vegard Nossum wrote:
> So we are now using __builtin_frame_address(). This should work everywhere
> and should not cause any harmful side effects.

Perhaps print an identifier using __builtin_frame_address(0)
to allow an external tool to reassemble complete messages?

> +	if (prev_caller != __builtin_frame_address(0) && !log_level_unknown) {
> +		emit_log_char('.');
> +		emit_log_char('.');
> +		emit_log_char('.');
> +		emit_log_char('\n');
> +		log_level_unknown = 1;
> +	}
> +
> +	prev_caller = __builtin_frame_address(0);
> +

maybe something like:

static void emit_log_id(void *addr)
{
	unsigned int i = 0;
	char id[sizeof(void *) * 2 + 6];
	sprintf(id, "{@%p}", addr);
	while (*id[i])
		emit_log_char(id[i++]);
}
...
	caller = __builtin_frame_address(0);
	if (prev_caller != caller && !log_level_unknown) {
		emit_log_id(caller);
		log_level_unknown = 1;
		prev_caller = caller;
	}


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/