Re: [PATCH V5 01/16] rv: Add Runtime Verification (RV) interface

[Tao Zhou <tao.zhou@linux.dev>]

On Fri, Jul 15, 2022 at 01:47:28PM +0200,
Daniel Bristot de Oliveira wrote:

> On 7/14/22 09:46, Tao Zhou wrote:
> > On Wed, Jul 13, 2022 at 11:17:17PM +0200,
> > Daniel Bristot de Oliveira <bristot@kernel.org> wrote:
> > 
> > [...]
> > 
> >> +void put_task_monitor_slot(int slot)
> >> +{
> >> +	lockdep_assert_held(&rv_interface_lock);
> >> +
> >> +	if (slot < 0 || slot > RV_PER_TASK_MONITORS) {
> > 
> > slot is the array index that should be 0 here. The up bound is not bigger
> > than 0 because the element of array now is RV_PER_TASK_MONITORS. 
> > 
> > So up bound check is 'slot > RV_PER_TASK_MONITORS-1'.
> 
> fixed! (slot >= RV...)
> 
> > [...]
> > 
> >> +/*
> >> + * interface for enabling/disabling a monitor.
> >> + */
> >> +static ssize_t monitor_enable_write_data(struct file *filp, const char __user *user_buf,
> >> +					 size_t count, loff_t *ppos)
> >> +{
> >> +	struct rv_monitor_def *mdef = filp->private_data;
> >> +	int retval;
> >> +	bool val;
> >> +
> >> +	retval = kstrtobool_from_user(user_buf, count, &val);
> >> +	if (retval)
> >> +		return retval;
> >> +
> >> +	retval = count;
> >> +
> >> +	mutex_lock(&rv_interface_lock);
> >> +
> >> +	if (val)
> >> +		retval = enable_monitor(mdef);
> >> +	else
> >> +		retval = disable_monitor(mdef);
> >> +
> >> +	mutex_unlock(&rv_interface_lock);
> >> +
> >> +	return retval ? retval : count;
> > 
> > Feel that this can be written `return retval ? : count;`
> 
> 
> why not...
> 
> > [...]
> > 
> >> +static void *enabled_monitors_start(struct seq_file *m, loff_t *pos)
> >> +{
> >> +	struct rv_monitor_def *m_def;
> >> +	loff_t l;
> >> +
> >> +	mutex_lock(&rv_interface_lock);
> >> +
> >> +	if (list_empty(&rv_monitors_list))
> >> +		return NULL;
> >> +
> >> +	m_def = list_entry(&rv_monitors_list, struct rv_monitor_def, list);
> >> +
> >> +	for (l = 0; l <= *pos; ) {
> >> +		m_def = enabled_monitors_next(m, m_def, &l);
> >> +		if (!m_def)
> >> +			break;
> > 
> > Is this check is inversed. enabled_monitors_start() will stop at first
> > enabled monitor, then enabled_monitors_next() do loop to next. Check
> > like the above, enabled_monitors_start() will loop to the last monitor.
> > But I doubt myself I do not mention/see it. Sorry for these.
> > 
> > the check is:
> > 
> >   if (m_def)
> >      break;
> > 
> > [...]
> 
> 
> see kernel/trace/trace_events.c:s_start...

I presumed @l changed in function enabled_monitors_next() will
impack on the @*pos of enabled_monitors_start(). But it's not.
@l is increased by 1 in enabled_monitors_next() and is used to
check with @*pos passed as parameter argument of enabled_monitors_start().
Absolutely I lost here.. Thanks.

> >> +static ssize_t
> >> +enabled_monitors_write(struct file *filp, const char __user *user_buf,
> >> +		      size_t count, loff_t *ppos)
> >> +{
> >> +	char buff[MAX_RV_MONITOR_NAME_SIZE + 2];
> >> +	struct rv_monitor_def *mdef;
> >> +	int retval = -EINVAL;
> >> +	bool enable = true;
> >> +	char *ptr = buff;
> >> +	int len;
> >> +
> >> +	if (count < 1 || count > MAX_RV_MONITOR_NAME_SIZE + 2)
> > 
> > @count would not include '\0'. That the max val of @count is
> > MAX_RV_MONITOR_NAME_SIZE+1. So the up bound check of @count is
> > `count > MAX_RV_MONITOR_NAME_SIZE + 1`.
> 
> Fixed for v6...
> 
> -- Daniel