Re: [PATCH] Process Aggregates (PAGG) for 2.6.7

[Peter Williams <pwil3058@bigpond.net.au>]

Erik Jacobson wrote:
> Attached is a PAGG patch to kernel 2.6.7.
> 
> The maintainers of two patches that make use of PAGG will post their patches
> in to this discussion thread shortly.
> 
> The biggest change in this patch from the last one I posted is that
> Peter Williams supplied an implementation for the init function pointer
> in the pagg hook.  We kicked this around a few times to flush out
> locking issues.


I wish that you had included me in this discussion.  Can you explain 
exactly what the locking issues with my code were?  We might have been 
able to come up with a better solution than the one you have used which 
places unnecessary restrictions (i.e. no blocking) on the init() 
callbacks which weren't applicable in the code that I provided.  Since 
these callbacks are highly likely to want to allocate dynamic memory 
there is always a chance that they will block and the no blocking 
restriction becomes an unnecessary burden.

Peter
-- 
Peter Williams                                   pwil3058@bigpond.net.au

"Learning, n. The kind of ignorance distinguishing the studious."
  -- Ambrose Bierce