Re: Fast Userspace Mutexes (futex) vs. msem_*

[Rusty Russell <rusty@rustcorp.com.au>]

On Fri, 8 Mar 2002 00:15:52 -0600 (CST)
Doug Siebert <dsiebert@divms.uiowa.edu> wrote:
 
> The direction that the futex implementation is going is looking a lot like
> how they are implemented on HP-UX (as well as Tru64 and AIX)  I am curious
> though why the case of "what happens if the process holding the lock dies"
> is considered unimportant by some people.  It wouldn't be all that much
> more work to "do it right" (IMHO) and handle this case.  AFAIK, on HP-UX
> the implementation kept a "locker id" and a linked list of waiters' lock
> ids (to allow first come first served as well as handling the case of a
> lock holder dying)

Fundamentally, these locks are fast because they *don't* tell the kernel.
Also, if you die with a lock, "autocleaning" it is often far worse then
just locking up, since the data it was protecting may be inconcsistant.

At the end of the day, this is the place for an application-specific
answer.  And this is nothing that can't be done in userspace (ie. if
you can't get the lock for some length of time, notify the lock master
and it will do all the cleanups and release the lock).

Hope that clarifies,
Rusty.
-- 
  Anyone who quotes me in their sig is an idiot. -- Rusty Russell.