[concurrency-interest] Soliciting input about removeAllThreadLocals
Sun, 08 Dec 2002 22:49:41 -0800
>Sorry but thats a vacuous example. What does the thread local do if it
>is totally independent of any task the thread may execute.
I already gave the logging example. Or how about a call counting
facility? I agree that these aren't great, detailed examples, but I
*strongly* believe that the burden of proof is on those who want to add
the new facility.
>>>I can easily imagine usage of thread-locals in a way that
>>>task-local and thus it would be wrong not to reset them when a new
>>In this sort of use, the ThreadLocal is generally updated
>>manually in a finally block so there's no need ot reset.
>You're assuming that the task knows ahead of time that it is using
Not necessarily. The use of thread locals could be hidden behind a
library that (say) begins and ends transactions. The transaction data
block would be automagically carried along by the thread. (For what
it's worth, I did this at CMU and Transarc, and it works fine.)
Sorry everybody. I realize that David and I are beating this one to
death. I promise to shut up.