On Fri, 2007-29-06 at 14:57 +0200, Johannes Berg wrote: > On Wed, 2007-06-27 at 19:24 -0400, jamal wrote: > Hm. I'm starting to dislike the dynamic registration the more I think > about it. Now when a group is unregistered I'd have to unbind everybody > who's currently using it... I understood you earlier to say only one family (in kernel) can own a group else it is an error, no? So there is "somebody" but not "everybody" BTW, refer to my earlier email in response to Patrick. I think preserving a group per family is also an interesting idea. i.e the first 2^16 are reserved/static and the others are in all for grabs area. > At least when I want to enforce > root/non-root binds which is a further goal. Permissions would be group attributes - maybe something in the group struct to say who is allowed to bind? > > c) Use a global hash table to store all the genl_multicast_groups; > > I think this (handwaving) should be searchable by i) name ii)ID and iii) > > family. > > Also, it doesn't look like I'll ever be searching for a group unless we > want to allow userspace to explicitly ask for a group ID, but I'd think > it should just get the list of all IDs when asking for family > information and then cache that. I think its useful to have the filters in the kernel when possible/sensible (one of my beefs against netlink in general is it lacks that). cheers, jamal > johannes - To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
