On Thu, 18 Oct 2012 14:51:03 -0500 Shirish Pargaonkar <shirishpargaonkar@xxxxxxxxx> wrote: > On Wed, Oct 17, 2012 at 1:09 PM, Jeff Layton <jlayton@xxxxxxxxxx> wrote: > [...] > > All the patches look correct except this question, would not this key_put > call end up purging the mapping from the cache. In the current code > this is OK since cifsacl has its own cache but once it gone, we are > relying on keys API cache to provide mapping. > No. The key (aka cache entry) will not garbage collected from the cache (aka the keyring) until it has expired and all references are put. Currently, I don't think we're setting any expiry on the key, which I think means that it'll live in the cache indefinitely. That does bring up a good point though. We should probably allow these to eventually expire. That's fairly simple to implement though -- any time we do a request key, we can just reset the timeout to some point in the future. I'll plan to add that in. Thanks for the review. -- Jeff Layton <jlayton@xxxxxxxxxx> -- To unsubscribe from this list: send the line "unsubscribe linux-cifs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
