[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sat, 9 Feb 2019 13:09:12 +0100
From: Greg KH <gregkh@...uxfoundation.org>
To: "Reshetova, Elena" <elena.reshetova@...el.com>
Cc: Peter Zijlstra <peterz@...radead.org>,
"kernel-hardening@...ts.openwall.com" <kernel-hardening@...ts.openwall.com>,
"luto@...nel.org" <luto@...nel.org>,
"tglx@...utronix.de" <tglx@...utronix.de>,
"mingo@...hat.com" <mingo@...hat.com>,
"bp@...en8.de" <bp@...en8.de>,
"keescook@...omium.org" <keescook@...omium.org>,
"tytso@....edu" <tytso@....edu>
Subject: Re: [RFC PATCH] x86/entry/64: randomize kernel stack offset upon
system call
On Sat, Feb 09, 2019 at 11:38:00AM +0000, Reshetova, Elena wrote:
>
> On Fri, Feb 08, 2019 at 01:20:09PM +0000, Reshetova, Elena wrote:
> > > > On Fri, Feb 08, 2019 at 02:15:49PM +0200, Elena Reshetova wrote:
> >
> > > I can measure the impact with rdtsc(), I think it is *very* small.
> > > Btw, what should be the good measurement test?
> > > I am not that happy with just looping on fopen-fclose, too
> > > much noise.
> >
> > IIRC amluto had a nice micro-bench for 'pure' syscall overhead
> > somewhere.
> >
> > Something like getpid/gettid in a loop should work fairly well, that's a
> > really light syscall to make.
>
> Thanks! Let me try this for measurements next. I was surprise that
> there nothing ready-made in perf or rt-tools. You would think
> this must be regularly in a need to measure.
lmbench has traditionally been used for something like this in the past,
you might want to look at that too.
greg k-h
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
