From: Khalid Aziz <khalid.aziz@oracle.com>
To: Julian Stecklina <jsteckli@amazon.de>,
Linus Torvalds <torvalds@linux-foundation.org>
Cc: David Woodhouse <dwmw@amazon.co.uk>,
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
juerg.haefliger@hpe.com, deepa.srinivasan@oracle.com,
Jim Mattson <jmattson@google.com>,
Andrew Cooper <andrew.cooper3@citrix.com>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Boris Ostrovsky <boris.ostrovsky@oracle.com>,
linux-mm <linux-mm@kvack.org>,
Thomas Gleixner <tglx@linutronix.de>,
joao.m.martins@oracle.com, pradeep.vincent@oracle.com,
Andi Kleen <ak@linux.intel.com>,
kanth.ghatraju@oracle.com, Liran Alon <liran.alon@oracle.com>,
Kees Cook <keescook@google.com>,
Kernel Hardening <kernel-hardening@lists.openwall.com>,
chris.hyser@oracle.com, Tyler Hicks <tyhicks@canonical.com>,
John Haxby <john.haxby@oracle.com>, Jon Masters <jcm@redhat.com>
Subject: Re: Redoing eXclusive Page Frame Ownership (XPFO) with isolated CPUs in mind (for KVM to isolate its guests per CPU)
Date: Fri, 14 Sep 2018 11:06:53 -0600 [thread overview]
Message-ID: <5efc291c-b0ed-577e-02d1-285d080c293d@oracle.com> (raw)
In-Reply-To: <ciirm8efdy916l.fsf@u54ee758033e858cfa736.ant.amazon.com>
On 09/12/2018 09:37 AM, Julian Stecklina wrote:
> Julian Stecklina <jsteckli@amazon.de> writes:
>
>> Linus Torvalds <torvalds@linux-foundation.org> writes:
>>
>>> On Fri, Aug 31, 2018 at 12:45 AM Julian Stecklina <jsteckli@amazon.de> wrote:
>>>>
>>>> I've been spending some cycles on the XPFO patch set this week. For the
>>>> patch set as it was posted for v4.13, the performance overhead of
>>>> compiling a Linux kernel is ~40% on x86_64[1]. The overhead comes almost
>>>> completely from TLB flushing. If we can live with stale TLB entries
>>>> allowing temporary access (which I think is reasonable), we can remove
>>>> all TLB flushing (on x86). This reduces the overhead to 2-3% for
>>>> kernel compile.
>>>
>>> I have to say, even 2-3% for a kernel compile sounds absolutely horrendous.
>>
>> Well, it's at least in a range where it doesn't look hopeless.
>>
>>> Kernel bullds are 90% user space at least for me, so a 2-3% slowdown
>>> from a kernel is not some small unnoticeable thing.
>>
>> The overhead seems to come from the hooks that XPFO adds to
>> alloc/free_pages. These hooks add a couple of atomic operations per
>> allocated (4K) page for book keeping. Some of these atomic ops are only
>> for debugging and could be removed. There is also some opportunity to
>> streamline the per-page space overhead of XPFO.
>
> I've updated my XPFO branch[1] to make some of the debugging optional
> and also integrated the XPFO bookkeeping with struct page, instead of
> requiring CONFIG_PAGE_EXTENSION, which removes some checks in the hot
> path. These changes push the overhead down to somewhere between 1.5 and
> 2% for my quad core box in kernel compile. This is close to the
> measurement noise, so I take suggestions for a better benchmark here.
>
> Of course, if you hit contention on the xpfo spinlock then performance
> will suffer. I guess this is what happened on Khalid's large box.
>
> I'll try to remove the spinlocks and add fixup code to the pagefault
> handler to see whether this improves the situation on large boxes. This
> might turn out to be ugly, though.
>
Hi Julian,
I ran tests with your updated code and gathered lock statistics. Change in system time for "make -j60" was in the noise margin (It actually went up by about 2%). There is some contention on xpfo_lock. Average wait time does not look high compared to other locks. Max hold time looks a little long. From /proc/lock_stat:
&(&page->xpfo_lock)->rlock: 29698 29897 0.06 134.39 15345.58 0.51 422474670 960222532 0.05 30362.05 195807002.62 0.20
Nevertheless even a smaller average wait time can add up.
--
Khalid
next prev parent reply other threads:[~2018-09-14 17:07 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-08-20 21:25 Redoing eXclusive Page Frame Ownership (XPFO) with isolated CPUs in mind (for KVM to isolate its guests per CPU) Konrad Rzeszutek Wilk
2018-08-20 21:48 ` Linus Torvalds
2018-08-20 21:52 ` Woodhouse, David
2018-08-20 22:18 ` Kees Cook
2018-08-20 22:27 ` Linus Torvalds
2018-08-20 22:35 ` Tycho Andersen
2018-08-20 22:59 ` Dave Hansen
2018-08-20 23:14 ` David Woodhouse
2018-08-20 23:26 ` Dave Hansen
2018-08-20 23:38 ` Linus Torvalds
2018-08-21 9:57 ` David Woodhouse
2018-08-21 14:01 ` Liran Alon
2018-08-21 14:22 ` David Woodhouse
2018-08-21 23:04 ` Liran Alon
2018-08-30 16:00 ` Julian Stecklina
2018-08-31 15:26 ` Tycho Andersen
2018-09-01 21:38 ` Linus Torvalds
2018-09-01 22:33 ` Wes Turner
2018-09-03 15:36 ` Andi Kleen
2018-09-03 14:51 ` Julian Stecklina
2018-09-12 15:37 ` Julian Stecklina
2018-09-13 6:11 ` Juerg Haefliger
2018-09-17 10:01 ` Julian Stecklina
2018-09-17 10:19 ` Tycho Andersen
2018-09-17 13:27 ` Christoph Hellwig
2018-09-14 17:06 ` Khalid Aziz [this message]
2018-09-17 9:51 ` Julian Stecklina
2018-09-18 23:00 ` Khalid Aziz
2018-09-24 14:45 ` Stecklina, Julian
2018-10-15 8:07 ` Khalid Aziz
2018-10-24 11:00 ` Khalid Aziz
2018-10-24 15:00 ` Tycho Andersen
2018-09-03 15:26 ` Andi Kleen
2018-09-04 9:37 ` Julian Stecklina
2018-09-07 21:30 ` Khalid Aziz
2018-08-31 8:43 ` James Bottomley
2018-09-19 1:03 ` Balbir Singh
2018-09-19 15:34 ` Jonathan Adams
2018-09-19 15:38 ` Jonathan Adams
2018-09-19 15:43 ` Jonathan Adams
2018-09-23 2:33 ` Balbir Singh
2018-09-25 14:12 ` Stecklina, Julian
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=5efc291c-b0ed-577e-02d1-285d080c293d@oracle.com \
--to=khalid.aziz@oracle.com \
--cc=ak@linux.intel.com \
--cc=andrew.cooper3@citrix.com \
--cc=boris.ostrovsky@oracle.com \
--cc=chris.hyser@oracle.com \
--cc=deepa.srinivasan@oracle.com \
--cc=dwmw@amazon.co.uk \
--cc=jcm@redhat.com \
--cc=jmattson@google.com \
--cc=joao.m.martins@oracle.com \
--cc=john.haxby@oracle.com \
--cc=jsteckli@amazon.de \
--cc=juerg.haefliger@hpe.com \
--cc=kanth.ghatraju@oracle.com \
--cc=keescook@google.com \
--cc=kernel-hardening@lists.openwall.com \
--cc=konrad.wilk@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=liran.alon@oracle.com \
--cc=pradeep.vincent@oracle.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=tyhicks@canonical.com \
/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;
as well as URLs for NNTP newsgroup(s).