Re: [PATCH] schedule: use unlikely()

[Greg KH <gregkh@linuxfoundation.org>]

On Fri, Dec 08, 2017 at 03:30:18PM +0100, Pavel Machek wrote:
> On Thu 2017-11-30 08:07:44, Greg KH wrote:
> > On Thu, Nov 30, 2017 at 02:04:01AM -0500, Mikulas Patocka wrote:
> > > 
> > > 
> > > On Tue, 28 Nov 2017, Greg KH wrote:
> > > 
> > > > On Mon, Nov 27, 2017 at 07:05:22PM -0500, Mikulas Patocka wrote:
> > > > > 
> > > > > 
> > > > > On Sat, 25 Nov 2017, Greg KH wrote:
> > > > > 
> > > > > > On Mon, Nov 13, 2017 at 02:00:45PM -0500, Mikulas Patocka wrote:
> > > > > > > A small patch for schedule(), so that the code goes straght in the common
> > > > > > > case.
> > > > > > > 
> > > > > > > Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
> > > > > > 
> > > > > > Was this a measurable difference?  If so, great, please provide the
> > > > > > numbers and how you tested in the changelog.  If it can't be measured,
> > > > > > then it is not worth it to add these markings
> > > > > 
> > > > > It is much easier to make microoptimizations (such as using likely() and 
> > > > > unlikely()) than to measure their effect.
> > > > > 
> > > > > If a programmer were required to measure performance every time he uses 
> > > > > likely() or unlikely() in his code, he wouldn't use them at all.
> > > > 
> > > > If you can not measure it, you should not use it.  You are forgetting
> > > > about the testing that was done a few years ago that found that some
> > > > huge percentage (80? 75? 90?) of all of these markings were wrong and
> > > > harmful or did absolutely nothing.
> > > 
> > > The whole kernel has 19878 likely/unlikely tags.
> > 
> > And most of them are wrong.  Don't add new ones unless you can prove it
> > is correct.
> 
> _Most_ of them wrong? Really? Where is your data for _that_?

Andi Kleen ran tests about 5 years ago or so which showed this.  It's in
the archives somewhere...

greg k-h