[lkml]   [2008]   [Aug]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH 2/2] smp_call_function: use rwlocks on queues rather than rcu
On Mon, Aug 25, 2008 at 10:44:03AM -0500, Christoph Lameter wrote:
> Paul E. McKenney wrote:
> > But I am not sure that this gets the grace periods to go fast enough to
> > cover Christoph's use case -- he seems to be in a "faster is better"
> > space rather than in an "at least this fast" space. Still, it would
> > likely help in some important cases.
> I think there was an AIM9 regression in the close/open tests when the struct
> file was switched to RCU. That test could be run with various intervals to
> figure out if a shorter RCU period is beneficial and how short an RCU period
> is needed to avoid the regression.

Well, with some luck, I can get an RCU implementation that allows the
grace-period duration to be varied, which would allow someone to check
the AIM9 regression.

Thanx, Paul

 \ /
  Last update: 2008-08-25 22:07    [W:0.082 / U:34.764 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site