Messages in this thread | | | Date | Wed, 29 Jan 2014 00:30:19 -0800 (PST) | From | David Rientjes <> | Subject | Re: [PATCH] mm: slub: fix page->_count corruption (again) |
| |
On Tue, 28 Jan 2014, Dave Hansen wrote:
> It has measurable performance benefits, and the benefits go up as the > cost of en/disabling interrupts goes up (like if it takes you a hypercall). > > Fengguang, could you run a set of tests for the top patch in this branch > to see if we'd be giving much up by axing the code? > > https://github.com/hansendc/linux/tree/slub-nocmpxchg-for-Fengguang-20140128 > > I was talking with one of the distros about turning it off as well. > They mentioned that they saw a few performance regressions when it was > turned off. I'll share details when I get them. >
FWIW, I've compared netperf TCP_RR on all machine types I have available with and without cmpxchg_double and I've never measured a regression.
| |