Re: [Gnu-arch-users] hook questions

[Dustin Sallings]

On Jan 14, 2004, at 17:43, Robert Collins wrote:

> Npo, because if you don't lock where I'm suggesting, the race condition
> can allow two competing commits to both succeed:
>
> commit A sees that X-1 is the current
> commit B sees that X-1 is the current
> commit A & B both set X in their files.
> now, A is a slower machine, and B
> locks, commits, unlocks
> before A trys to lock.

	In tla, in particular, you lock the patch you have, though, right?  
The lock no longer exists in that patch when you go to aquire it.  At 
this point, you abort the commit because it's out of date, anyway.

        In your scenario (with my understanding of arch locking):

commit A sees that X-1 is the current (local)
commit B sees that X-1 is the current (local)
commit A & B both set X in their files.
now, A is a slower machine, and B
locks X-1, commits, unlocks
A attempts to aquire lock X-1 and fails, update/replay required

--
Dustin Sallings