I actually didn't know that was allowed.  I'll change it before merging.

Ethan

On Mon, Jul 18, 2011 at 14:38, Ben Pfaff <b...@nicira.com> wrote:
> On Mon, Jul 18, 2011 at 02:35:06PM -0700, Ethan Jackson wrote:
>> Sounds good to me, here is another version.
>
> Thanks, just one comment left:
>
>> +/* Returns a string representation of 'fields'. */
>> +const char *
>> +flow_hash_fields_to_str(enum nx_hash_fields fields)
>> +{
>> +    static char *symmetric_l4 = "symmetric_l4";
>> +    static char *eth_src = "eth_src";
>> +    static char *unknown = "<unknown>";
>> +
>> +    switch (fields) {
>> +    case NX_HASH_FIELDS_ETH_SRC: return eth_src;
>> +    case NX_HASH_FIELDS_SYMMETRIC_L4: return symmetric_l4;
>> +    default: return unknown;
>> +    }
>> +}
>
> What's the benefit of separate declarations for the strings?  Why not
> just write, e.g. return "symmetric_l4"?
>
> Thanks,
>
> Ben.
>
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev

Reply via email to