Re: [PATCH 1/4] VM/RMAP: Add infrastructure for batching the rmap chain locking

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, 09 May 2011 21:27:28 -0400 Rik van Riel <riel@xxxxxxxxxx> wrote:

> On 05/09/2011 09:26 PM, Andrew Morton wrote:
> > On Tue, 10 May 2011 02:05:17 +0200 Andi Kleen<andi@xxxxxxxxxxxxxx>  wrote:
> >
> >>> hm, which atomic ops are those?  We shouldn't need buslocked operations
> >>> on UP.
> >>
> >> Good point. I forgot they're patched out.
> >>
> >> Then it's likely somewhat slower, but I doubt it's a significant
> >> slowdown. Essentially it's just a few more if ()s
> >
> > And presumably some more instruction cachelines here and there.  It'll
> > be small, but 100*small == large.  We can reduce the overhead to zero,
> > it's a question of how ugly the end result is.
> 
> More instruction cachelines, but fewer pointer dereferences
> due to not looking up the lock all the time (if the root
> anon_vma is the same).
> 

Why would we be looking up a lock which doesn't exist?

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxxx  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>


[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux]     [Linux OMAP]     [Linux MIPS]     [ECOS]     [Asterisk Internet PBX]     [Linux API]