About Work-Around Difficulty

1. If the expiration of a record is connected with the execution of a critical section in the kernel, it can be of paramount importance that the terminating use of the record be accomplished as quickly as possible, running in a real-memory task. No page faults, no screwing around, do the real work and scram, as when servicing any kind of real-time interrupt.

The work-around in such a case is non-trivial and it may have footprint and other consequences.

2. These actions were brought in 2006. There have been at least 5 years to figure out how to avoid the essential claims of the patent for which infringement has been found. What's up with that? I would assume the kernel team (and Linus) know what they need to do already.