Re: [PATCH] at24: make module parameters changeable via sysfs

[Jean Delvare <khali-PUYAD+kWke1g9hUCZPvPmw@public.gmane.org>]

Uwe,

On Fri, 14 Sep 2012 10:25:36 +0200, Jean Delvare wrote:
> On Wed, 12 Sep 2012 11:43:32 +0200, Uwe Kleine-König wrote:
> > The respective values are evaluated at each read/write, so no further
> > action is required than to change the perm argument to module_param.
> > 
> > Note there is no sanity check so root can make the driver effectively
> > unusable but that's what root is for :-)
> >
> > Signed-off-by: Uwe Kleine-König <u.kleine-koenig-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
> > ---
> >  drivers/misc/eeprom/at24.c |    4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/misc/eeprom/at24.c b/drivers/misc/eeprom/at24.c
> > index ab1ad41..8a5a192 100644
> > --- a/drivers/misc/eeprom/at24.c
> > +++ b/drivers/misc/eeprom/at24.c
> > @@ -85,7 +85,7 @@ struct at24_data {
> >   * This value is forced to be a power of two so that writes align on pages.
> >   */
> >  static unsigned io_limit = 128;
> > -module_param(io_limit, uint, 0);
> > +module_param(io_limit, uint, S_IRUGO | S_IWUSR);
> 
> This won't work. Not only there is no validation of the value, while
> there is such a validation (and value adjustment!) in at24_init(); you
> seem to not care, but I do. But the more important problem is that
> changing io_limit at run-time will only affect reads, not writes. The
> size limit from writes is computed at device probing time:
> 
> static int at24_probe(struct i2c_client *client, const struct i2c_device_id *id)
> {
> (...)
> 	if (writable) {
> 	(...)
> 			if (write_max > io_limit)
> 				write_max = io_limit;
> 
> So changing the value through sysfs will have no effect. If you want it
> to have an effect, you have to move the check from at24_probe() to
> at24_eeprom_write().
> 
> Back to the validation issue, I think it would be worth looking into
> module_param_cb(). Using it, it may not be that difficult to get
> validation when the value is changed through sysfs. Otherwise I'll ask
> you to check what exactly happens if someone sets io_limit to 0. We
> can't afford infinite loops or EEPROM corruption on root mistyping.
> 
> >  MODULE_PARM_DESC(io_limit, "Maximum bytes per I/O (default 128)");
> >  
> >  /*
> > @@ -93,7 +93,7 @@ MODULE_PARM_DESC(io_limit, "Maximum bytes per I/O (default 128)");
> >   * it's important to recover from write timeouts.
> >   */
> >  static unsigned write_timeout = 25;
> > -module_param(write_timeout, uint, 0);
> > +module_param(write_timeout, uint, S_IRUGO | S_IWUSR);
> 
> This one is OK.
> 
> >  MODULE_PARM_DESC(write_timeout, "Time (in ms) to try writes (default 25)");
> >  
> >  #define AT24_SIZE_BYTELEN 5

I reviewed this patch 3 months ago and did not hear back. Are you going
to update this patch and resubmit, or should I just drop it?

-- 
Jean Delvare