From: Pavel Machek <pavel@ucw.cz>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: "Joerg Roedel" <joro@8bytes.org>,
"Thomas Gleixner" <tglx@linutronix.de>,
"Ingo Molnar" <mingo@kernel.org>, "Peter Anvin" <hpa@zytor.com>,
"the arch/x86 maintainers" <x86@kernel.org>,
"Linux Kernel Mailing List" <linux-kernel@vger.kernel.org>,
linux-mm <linux-mm@kvack.org>,
"Andrew Lutomirski" <luto@kernel.org>,
"Dave Hansen" <dave.hansen@intel.com>,
"Josh Poimboeuf" <jpoimboe@redhat.com>,
"Jürgen Groß" <jgross@suse.com>,
"Peter Zijlstra" <peterz@infradead.org>,
"Borislav Petkov" <bp@alien8.de>, "Jiri Kosina" <jkosina@suse.cz>,
"Boris Ostrovsky" <boris.ostrovsky@oracle.com>,
"Brian Gerst" <brgerst@gmail.com>,
"David Laight" <David.Laight@aculab.com>,
"Denys Vlasenko" <dvlasenk@redhat.com>,
"Eduardo Valentin" <eduval@amazon.com>,
"Greg Kroah-Hartman" <gregkh@linuxfoundation.org>,
"Will Deacon" <will.deacon@arm.com>,
"Liguori, Anthony" <aliguori@amazon.com>,
"Daniel Gruss" <daniel.gruss@iaik.tugraz.at>,
"Hugh Dickins" <hughd@google.com>,
"Kees Cook" <keescook@google.com>,
"Andrea Arcangeli" <aarcange@redhat.com>,
"Waiman Long" <llong@redhat.com>,
"David H . Gutteridge" <dhgutteridge@sympatico.ca>,
"Joerg Roedel" <jroedel@suse.de>,
"Arnaldo Carvalho de Melo" <acme@kernel.org>,
"Alexander Shishkin" <alexander.shishkin@linux.intel.com>,
"Jiri Olsa" <jolsa@redhat.com>,
"Namhyung Kim" <namhyung@kernel.org>
Subject: Re: [PATCH 0/3] PTI for x86-32 Fixes and Updates
Date: Tue, 24 Jul 2018 15:39:35 +0200 [thread overview]
Message-ID: <20180724133935.GA30797@amd> (raw)
In-Reply-To: <CA+55aFynT9Sp7CbnB=GqLbns7GFZbv3pDSQm_h0jFvJpz3ES+g@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 3456 bytes --]
On Mon 2018-07-23 12:00:08, Linus Torvalds wrote:
> On Mon, Jul 23, 2018 at 7:09 AM Pavel Machek <pavel@ucw.cz> wrote:
> >
> > Meanwhile... it looks like gcc is not slowed down significantly, but
> > other stuff sees 30% .. 40% slowdowns... which is rather
> > significant.
>
> That is more or less expected.
Ok, so I was wrong. bzip2 showed 30% slowdown, but running test in a
loop, I get (on v4.18) that, too.
That tells me that something is wrong with machine I'm using for
benchmarking. Whether KPTI is enabled can still be measured with the
bzip2 pipeline, but the effect is far more subtle.
Pavel
pavel@amd:~$ while true; do time cat /dev/urandom | head -c 10000000 |
bzip2 -9 - | wc -c ; done10044031
3.87user 0.91system 4.62 (0m4.622s) elapsed 103.48%CPU
10044234
4.03user 0.82system 4.68 (0m4.688s) elapsed 103.67%CPU
10043664
4.28user 0.85system 4.99 (0m4.994s) elapsed 102.90%CPU
10045959
4.43user 0.85system 5.12 (0m5.121s) elapsed 103.44%CPU
10043829
4.50user 0.89system 5.22 (0m5.228s) elapsed 103.22%CPU
10044296
4.65user 0.93system 5.39 (0m5.398s) elapsed 103.61%CPU
10045311
4.76user 0.93system 5.47 (0m5.479s) elapsed 103.98%CPU
10043819
4.81user 0.93system 5.55 (0m5.556s) elapsed 103.37%CPU
10045097
4.72user 1.04system 5.59 (0m5.597s) elapsed 103.01%CPU
10044012
4.86user 0.97system 5.68 (0m5.684s) elapsed 102.79%CPU
10044569
4.93user 0.96system 5.72 (0m5.728s) elapsed 102.92%CPU
10044141
4.94user 0.98system 5.75 (0m5.752s) elapsed 102.97%CPU
10043695
4.97user 0.95system 5.76 (0m5.768s) elapsed 102.87%CPU
10045690
5.12user 0.94system 5.90 (0m5.901s) elapsed 102.79%CPU
10045153
5.06user 1.00system 5.88 (0m5.883s) elapsed 103.21%CPU
10044560
5.10user 1.01system 5.92 (0m5.927s) elapsed 103.31%CPU
10044845
5.17user 0.99system 5.96 (0m5.960s) elapsed 103.44%CPU
10043884
5.15user 1.03system 6.00 (0m6.004s) elapsed 103.14%CPU
10044286
5.18user 1.01system 6.00 (0m6.002s) elapsed 103.40%CPU
10045749
5.00user 1.22system 6.04 (0m6.044s) elapsed 102.98%CPU
10044098
5.22user 1.02system 6.05 (0m6.053s) elapsed 103.21%CPU
10045326
5.20user 1.01system 6.04 (0m6.048s) elapsed 102.72%CPU
10042365
5.22user 1.03system 6.06 (0m6.061s) elapsed 103.30%CPU
10043952
5.24user 1.00system 6.06 (0m6.069s) elapsed 102.97%CPU
10044569
5.30user 1.00system 6.09 (0m6.099s) elapsed 103.46%CPU
10043241
5.26user 1.00system 6.09 (0m6.097s) elapsed 102.79%CPU
10044797
5.30user 1.01system 6.11 (0m6.114s) elapsed 103.46%CPU
10043711
5.25user 1.02system 6.09 (0m6.093s) elapsed 103.03%CPU
10043882
5.31user 1.01system 6.13 (0m6.131s) elapsed 103.28%CPU
10043571
5.26user 1.05system 6.13 (0m6.133s) elapsed 103.06%CPU
10044742
5.29user 1.03system 6.12 (0m6.122s) elapsed 103.25%CPU
10044170
5.35user 1.04system 6.18 (0m6.183s) elapsed 103.60%CPU
10043542
5.22user 1.12system 6.17 (0m6.172s) elapsed 102.89%CPU
10042985
5.25user 1.13system 6.19 (0m6.193s) elapsed 103.09%CPU
10044102
5.36user 1.01system 6.17 (0m6.177s) elapsed 103.17%CPU
10044609
5.48user 0.99system 6.28 (0m6.284s) elapsed 103.11%CPU
10045185
5.40user 1.03system 6.23 (0m6.236s) elapsed 103.29%CPU
10044444
5.41user 1.06system 6.25 (0m6.255s) elapsed 103.55%CPU
10044859
5.35user 1.04system 6.20 (0m6.201s) elapsed 103.17%CPU
10045613
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]
next prev parent reply other threads:[~2018-07-24 13:39 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-07-20 16:22 Joerg Roedel
2018-07-20 16:22 ` [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables Joerg Roedel
2018-07-20 17:06 ` Andy Lutomirski
2018-07-20 17:48 ` Joerg Roedel
2018-07-20 19:32 ` Andy Lutomirski
2018-07-20 21:37 ` Joerg Roedel
2018-07-20 22:20 ` Andy Lutomirski
2018-07-21 21:06 ` Linus Torvalds
2018-07-20 19:27 ` Thomas Gleixner
2018-07-20 19:33 ` Andy Lutomirski
2018-07-20 19:43 ` Thomas Gleixner
2018-07-20 19:53 ` Thomas Gleixner
2018-07-20 16:22 ` [PATCH 2/3] x86/entry/32: Check for VM86 mode in slow-path check Joerg Roedel
2018-07-21 16:06 ` Pavel Machek
2018-07-20 16:22 ` [PATCH 3/3] x86/entry/32: Copy only ptregs on paranoid entry/exit path Joerg Roedel
2018-07-20 17:09 ` Andy Lutomirski
2018-07-20 21:42 ` Joerg Roedel
2018-07-23 3:49 ` [PATCH 0/3] PTI for x86-32 Fixes and Updates David H. Gutteridge
2018-07-23 7:29 ` Joerg Roedel
2018-07-26 3:47 ` David H. Gutteridge
2018-07-23 14:09 ` Pavel Machek
2018-07-23 19:00 ` Linus Torvalds
2018-07-23 21:38 ` Pavel Machek
2018-07-23 21:50 ` Andy Lutomirski
2018-07-23 21:55 ` Pavel Machek
2018-07-24 21:18 ` Pavel Machek
2018-07-23 21:59 ` Josh Poimboeuf
2018-07-23 22:07 ` Dave Hansen
2018-07-24 13:39 ` Pavel Machek [this message]
2018-07-24 14:39 ` Andy Lutomirski
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180724133935.GA30797@amd \
--to=pavel@ucw.cz \
--cc=David.Laight@aculab.com \
--cc=aarcange@redhat.com \
--cc=acme@kernel.org \
--cc=alexander.shishkin@linux.intel.com \
--cc=aliguori@amazon.com \
--cc=boris.ostrovsky@oracle.com \
--cc=bp@alien8.de \
--cc=brgerst@gmail.com \
--cc=daniel.gruss@iaik.tugraz.at \
--cc=dave.hansen@intel.com \
--cc=dhgutteridge@sympatico.ca \
--cc=dvlasenk@redhat.com \
--cc=eduval@amazon.com \
--cc=gregkh@linuxfoundation.org \
--cc=hpa@zytor.com \
--cc=hughd@google.com \
--cc=jgross@suse.com \
--cc=jkosina@suse.cz \
--cc=jolsa@redhat.com \
--cc=joro@8bytes.org \
--cc=jpoimboe@redhat.com \
--cc=jroedel@suse.de \
--cc=keescook@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=llong@redhat.com \
--cc=luto@kernel.org \
--cc=mingo@kernel.org \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=will.deacon@arm.com \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox