Re: Performance Benchmark for PTI (aka Meltdown mitigation)

From: Slawa Olhovchenkov <slw_at_zxy.spb.ru>
Date: Fri, 9 Mar 2018 18:16:06 +0300
On Fri, Mar 09, 2018 at 09:58:55AM -0500, Ed Maste wrote:

> On 9 March 2018 at 07:01, Slawa Olhovchenkov <slw_at_zxy.spb.ru> wrote:
> > On Thu, Mar 08, 2018 at 05:04:11PM -0500, Arshan Khanifar wrote:
> >
> >> Executive Summary:
> >> - The PTI feature increases the system call times by more than 100%.
> >> - As a macrobenchmark, buildworld was used. Wall clock and user time
> >> showed no statistically-significant changes, while system time
> >> increased by less than 5%.
> >>
> >> This email contains the results for benchmarking the performance of the
> >> PTI patch on FreeBSD 12-current. As a microbenchmark, timing of
> >> getppid(2) system call was used, and as a macrobenchmark, a number of
> >> buildworld tasks were timed.
> >>
> >
> > Can you also run pre-patched kernel?
> 
> It's not easy to do an apples-to-apples comparison as there were a few
> followup changes to the PTI work, interspersed with unrelated changes.
> That said, I think Arshan has some benchmarks obtained during the
> development of the PTI changes that may be illustrative.
> 
> The best approach is probably to compare stable/11 at r329450 (last
> stable/11 revision before the merge) with r329462 with PTI and IBRS
> disabled.

Stable/11 more interesting to me, nice to see.
Received on Fri Mar 09 2018 - 14:16:10 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:41:15 UTC