Re: [PATCH] lglock: make lg_lock_global() actually lock globally

[Nick Piggin]

On Thu, Aug 26, 2010 at 01:45:39PM +0200, Peter Zijlstra wrote:
> On Thu, 2010-08-26 at 21:38 +1000, Nick Piggin wrote:
> > I think for CPU plug, stop_machine is reasonable (especially
> > considering it is required in unload, which means any frequent
> > amount of cpu plug activity already will require stop_machine to
> > run anyway). 
> 
> How is it required? 

Well, as is implemented.

 
> Its currently implemented as such, and its sure a lot easier to do that
> way, but I could imagine that unplugging a CPU could be done without it.
 
I would much prefer the rules to be simpler and easier for all
other kernel code, and keep complexity and overheads in cpu
plug/unplug.

I don't see what is so nice about stop_machine()less cpu plug/unplug
or module unload. Module load definitely is nice because you can
have a lot of modules and on demand loading from non-privileged
operations.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/