Re: [PATCH] firmware: Google VPD: fix error handling on allocation failures

[walter harms <wharms@bfs.de>]

Am 03.05.2017 17:49, schrieb Colin King:
> From: Colin Ian King <colin.king@canonical.com>
> 
> Fix two allocation failure checks. Firstly, ensure info is checked
> for a failed allocation; this fixes a potential null pointer
> dereference issue on info->key.  Secondly, free info is info->key
> fails to allocate to fix a memory leak.
> 
> Detected by CoverityScan, CID#1430064 ("Resource Leak")
> 
> Signed-off-by: Colin Ian King <colin.king@canonical.com>
> ---
>  drivers/firmware/google/vpd.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/firmware/google/vpd.c b/drivers/firmware/google/vpd.c
> index 3ce813110d5e..2eb15b1dabcc 100644
> --- a/drivers/firmware/google/vpd.c
> +++ b/drivers/firmware/google/vpd.c
> @@ -116,9 +116,13 @@ static int vpd_section_attrib_add(const u8 *key, s32 key_len,
>  		return VPD_OK;
>  
>  	info = kzalloc(sizeof(*info), GFP_KERNEL);
> +	if (!info)
> +		return -ENOMEM;
>  	info->key = kzalloc(key_len + 1, GFP_KERNEL);
> -	if (!info->key)
> +	if (!info->key) {
> +		kfree(info);
>  		return -ENOMEM;
> +	}
>  
>  	memcpy(info->key, key, key_len);
>  


the use of key_len+1 looks like preparing space for a string.
so kstrdup() seems more fitting. If this is not a string
there is also a kmemdup().


re,
 wh