Re: [GENETLINK]: Question: global lock (genl_mutex) possible refinement?

* Richard MUSIL [off-list ref] 2007-07-20 18:15

Patrick McHardy wrote:

quoted

Export the lock/unlock/.. functions. You'll also need a new version 
similar to __rtnl_unlock.

Patrick, you might feel, I am not reading your lines, but in fact I do.
The problem is that I do not feel competent to follow/propose such
changes. So what I propose here (in included patch) is the least change
scenario, which I can think of and on which I feel safe.

If there are some other changes required, as you suggested for example
exporting lock from genetlink module, I hope authors of genetlink will
comment on that. Currently, I do not see any reason for that, but this
could be due to my limited knowledge.

Actually there is no reason to not use separate locks for the
message serialization and the protection of the list of registered
families. There is only one lock simply for the reason that I've
never thought of anybody could think of registering a new genetlink
family while processing a message.

Alternatively you could also postpone the registration of the new
genetlink family to a workqueue.