On 03/14/2013 03:59 AM, Paul Mackerras wrote:
> On Fri, Mar 08, 2013 at 10:02:31PM -0600, Nathan Fontenot wrote:
>> This patch adds a platform_has_feature() function to check features selected
>> by firmware and reported via the device tree 'ibm,architecture-vec5'
>> property. As part of this the #defines used for the architecture vector are
>> moved to prom.h and re-defined such that the vector 5 options have the vector
>> index and the feature bits encoded into them. This allows for callers of
>> platform_has_feature() to pass in a single pre-defined value.
>
> One other comment...
>
>> +#if defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_POWERNV)
>> +bool platform_has_feature(unsigned int feature)
>> +{
>> + struct device_node *chosen;
>> + const char *vec5;
>> + bool has_option;
>> +
>> + chosen = of_find_node_by_path("/chosen");
>> + if (!chosen)
>> + return false;
>> +
>> + vec5 = of_get_property(chosen, "ibm,architecture-vec-5", NULL);
>> + has_option = vec5 && (vec5[OV5_INDX(feature)] & OV5_FEAT(feature));
>
> You access vec5[index] without checking that the vector is at least
> index+1 bytes long, according to either the length byte at the
> beginning of the vector, or the total length of the property.
> Checking both would be a good idea.
>
Yep. Thanks for letting me know.
--
-Nathan
_______________________________________________
Linuxppc-dev mailing list
[email protected]
https://lists.ozlabs.org/listinfo/linuxppc-dev
