Re: [PATCH 15/30] staging: wilc1000: use kmemdup instead of kmalloc in add_network_to_shadow()

[Ajay Singh <ajay.kathat@microchip.com>]

On Wed, 9 May 2018 16:42:59 +0300
Claudiu Beznea <Claudiu.Beznea@microchip.com> wrote:

> On 07.05.2018 11:43, Ajay Singh wrote:
> > Use kmemdup instead of kmalloc & memcpy in add_network_to_shadow().
> > 
> > Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
> > ---
> >  drivers/staging/wilc1000/wilc_wfi_cfgoperations.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c
> > b/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c index
> > 0ae2065..ca221f1 100644 ---
> > a/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c +++
> > b/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c @@ -331,8
> > +331,8 @@ static void add_network_to_shadow(struct network_info
> > *nw_info, shadow_nw_info->tsf_hi = nw_info->tsf_hi; if (ap_found !=
> > -1) kfree(shadow_nw_info->ies);
> > -	shadow_nw_info->ies = kmalloc(nw_info->ies_len,
> > GFP_KERNEL);
> > -	memcpy(shadow_nw_info->ies, nw_info->ies,
> > nw_info->ies_len);
> > +	shadow_nw_info->ies = kmemdup(nw_info->ies,
> > nw_info->ies_len,
> > +				      GFP_KERNEL);  
> 
> Maybe, in case of NULL, you will want to set ies_len = 0 ?


I couldn't find code where 'ies_len' is check to validity of data.
Mostly we use NULL check for "ies" pointer for data
validity.So in my opinion setting it to zero would be
irrelevant.


> 
> >  	shadow_nw_info->time_scan = jiffies;
> >  	shadow_nw_info->time_scan_cached = jiffies;
> >  	shadow_nw_info->found = 1;
> >   


Regards,
Ajay