Re: [uclinux-dist-devel] [linux-pm] freezer: should barriers be smp ?

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Saturday, April 16, 2011, Mike Frysinger wrote:
> On Fri, Apr 15, 2011 at 19:11, Rafael J. Wysocki wrote:
> > On Friday, April 15, 2011, Mike Frysinger wrote:
> >> On Fri, Apr 15, 2011 at 12:29, Pavel Machek wrote:
> >> >> > > I believe the code is correct as is.
> >> >> >
> >> >> > that isnt what the code / documentation says.  unless i'm reading them
> >> >> > wrong, both seem to indicate that the proposed patch is what we
> >> >> > actually want.
> >> >>
> >> >> The existing code is correct but it isn't optimal.
> >> >>
> >> >> wmb() and rmb() are heavy-duty operations, and you don't want to call
> >> >> them when they aren't needed.  That's exactly what smp_wmb() and
> >> >> smp_rmb() are for -- they call wmb() and rmb(), but only in SMP
> >> >> kernels.
> >> >>
> >> >> Unless you need to synchronize with another processor (not necessarily
> >> >> a CPU, it could be something embedded within a device), you should
> >> >> always use smp_wmb() and smp_rmb() rather than wmb() and rmb().
> >> >
> >> > Maybe; but this code is not performance critical and I believe being
> >> > obvious here is better...
> >>
> >> isnt it though ?  especially when we talk about suspending/resuming on
> >> embedded systems to get more savings over just cpu idle ?  we want
> >> that latency to be as low as possible.
> >
> > I agree, we can switch the freezer to smp_ barriers, but not for the reason
> > you gave before. :-)
> >
> > Care to repost the patch with a suitable changelog?
> 
> np
> 
> to be clear, what you said wrt the Blackfin smp barriers still holds
> true right ?

I think so.

> so this changset i merged doesnt need any tweaking ...
> http://git.kernel.org/?p=linux/kernel/git/vapier/blackfin.git;a=commitdiff;h=943aee0c685d0563228d5a2ad9c8394ad0300fb5

Rafael