Re: [patch] sony-laptop: fix a couple signedness bugs

[Mattia Dongili <malattia@xxxxxxxx>]

On Thu, Jun 07, 2012 at 12:17:37PM +0200, walter harms wrote:
> 
> 
> Am 07.06.2012 11:59, schrieb Dan Carpenter:
> > On Thu, Jun 07, 2012 at 11:47:30AM +0200, walter harms wrote:
> >>
> >>
> >> Am 07.06.2012 10:22, schrieb Dan Carpenter:
> >>> These need to be signed to handle negative error codes.
> >>>
> >>> Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>

Thanks Dan, I'll take this and the other patch you sent earlier and
submit to Matthew together with other patches I have pending here.
See below for a quick comment.

...
> >> sony_find_snc_handle() should return an int, did something change ?
> > 
> > offset is a u64.  It has to be 64 bit.  I thought about changing it
> > to s64, but decided to just cast it here instead.
> > 
> > regards,
> > dan carpenter
> > 
> 
> If it this function
>  http://lxr.free-electrons.com/source/drivers/platform/x86/sony-laptop.c?a=powerpc#L817
> it just returns an index expected to be <0x10.
> also offset was an int
>  http://lxr.free-electrons.com/source/drivers/platform/x86/sony-laptop.c?a=powerpc#L1563
> 
> it make we wonder why somebody made the change ...

it was changed to call into sony_nc_buffer_call which is a new function
(ebcef1b0e41f2ff972e5c5487a30e8f4ee2b6f13).

...
> >>> -	if (offset < 0)
> >>> +	if ((int)offset < 0)
> >>>  		return;

This check is actually redundant in a way as
sony_nc_backlight_ng_read_limits is only called with a valid handle.
Might as well remove the if statement altogether. sony_nc_rfkill_setup
has similar code without the check for a valid offset.

Thanks
-- 
mattia
:wq!
--
To unsubscribe from this list: send the line "unsubscribe platform-driver-x86" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html