From: Ingo Molnar <mingo@elte.hu>
To: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: tglx@linutronix.de, Dan Hecht <dhecht@vmware.com>,
Zachary Amsden <zach@vmware.com>,
akpm@linux-foundation.org, ak@suse.de,
Virtualization Mailing List <virtualization@lists.osdl.org>,
Rusty Russell <rusty@rustcorp.com.au>,
LKML <linux-kernel@vger.kernel.org>,
john stultz <johnstul@us.ibm.com>
Subject: Re: + stupid-hack-to-make-mainline-build.patch added to -mm tree
Date: Thu, 8 Mar 2007 09:01:31 +0100 [thread overview]
Message-ID: <20070308080131.GB9295@elte.hu> (raw)
In-Reply-To: <45EF2991.1020204@goop.org>
* Jeremy Fitzhardinge <jeremy@goop.org> wrote:
> > Your implementation is almost the perfect prototype, if you move the
> > 128 bit hackery into the hypervisor and hide it away from the kernel
> > :)
>
> The point is to use the tsc to avoid making any hypercalls, so dealing
> with the tsc->ns conversion has to happen on the guest side somehow.
you are obsessed with avoiding a hypercall, but why? Granted it's slow
especially on things like SVN/VMX, but it's not fundamentally slow. We
definitely do not want to design our whole APIs and abstractions around
the temporary notion that 'hypercalls are slow'. I'd expect hypercalls
to be put into silicon just as much as SYSENTER was put into silicon.
Anyway, in terms of guest time code, a /big/ amount of design junk can
be avoided by not trying to do sillynesses like 'virtual time'. The TSC
is awfully unreliable.
really, it's a bit as if Linus looked at his 386DX CPU when he bought it
16 years ago and decided that: "this CPU executes 16-bit code much
faster than 32-bit code, so lets base this new toy OS on 16-bit code.
Sure, it's a bit of a pain to use, compared to 32-bit code, but users
demand performance!".
/THIS/ is the kind of junk we are trying to protect Linux against.
Basically hypervisors are a way to prolong hardware legacies, and
because unlike real hardware software ABIs dont actually burn out with
time, and people are stubborn about using them, their effects are alot
worse and alot longer than that of legacy hardware.
Ingo
next prev parent reply other threads:[~2007-03-08 8:02 UTC|newest]
Thread overview: 114+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <200703060654.l266sVxr014860@shell0.pdx.osdl.net>
[not found] ` <45ED16D2.3000202@vmware.com>
[not found] ` <20070306084258.GA15745@elte.hu>
[not found] ` <20070306084647.GA16280@elte.hu>
[not found] ` <45ED2C82.3080008@vmware.com>
2007-03-06 10:59 ` Thomas Gleixner
2007-03-06 21:07 ` Dan Hecht
2007-03-06 22:21 ` Andi Kleen
2007-03-06 21:32 ` Dan Hecht
2007-03-06 23:53 ` Thomas Gleixner
2007-03-07 0:24 ` Jeremy Fitzhardinge
2007-03-07 0:35 ` Dan Hecht
2007-03-07 0:49 ` Thomas Gleixner
2007-03-07 0:53 ` Dan Hecht
2007-03-07 1:18 ` Thomas Gleixner
2007-03-07 2:08 ` Dan Hecht
2007-03-07 8:37 ` Thomas Gleixner
2007-03-07 17:41 ` Jeremy Fitzhardinge
2007-03-07 17:49 ` Ingo Molnar
2007-03-07 18:03 ` James Morris
2007-03-07 18:35 ` Jeremy Fitzhardinge
2007-03-08 0:45 ` Alan Cox
2007-03-07 17:52 ` Ingo Molnar
2007-03-07 18:28 ` Jeremy Fitzhardinge
2007-03-07 18:53 ` Thomas Gleixner
2007-03-07 18:11 ` James Morris
2007-03-07 18:56 ` Thomas Gleixner
2007-03-07 19:05 ` Jeremy Fitzhardinge
2007-03-07 19:49 ` Dan Hecht
2007-03-07 20:11 ` Jeremy Fitzhardinge
2007-03-07 20:49 ` Dan Hecht
2007-03-07 21:14 ` Thomas Gleixner
2007-03-07 20:57 ` Thomas Gleixner
2007-03-07 21:02 ` Dan Hecht
2007-03-07 21:08 ` Jeremy Fitzhardinge
2007-03-07 21:19 ` Thomas Gleixner
2007-03-07 21:14 ` Dan Hecht
2007-03-07 21:21 ` Thomas Gleixner
2007-03-07 21:33 ` Dan Hecht
2007-03-07 22:05 ` Jeremy Fitzhardinge
2007-03-07 23:05 ` Thomas Gleixner
2007-03-07 23:25 ` Zachary Amsden
2007-03-07 23:36 ` Jeremy Fitzhardinge
2007-03-07 23:40 ` Zachary Amsden
2007-03-08 18:30 ` Chris Wright
2007-03-08 0:22 ` Thomas Gleixner
2007-03-08 1:01 ` Daniel Arai
2007-03-08 1:23 ` Jeremy Fitzhardinge
2007-03-08 7:02 ` Thomas Gleixner
2007-03-08 7:28 ` Thomas Gleixner
2007-03-08 8:01 ` Zachary Amsden
2007-03-08 18:24 ` Chris Wright
2007-03-08 18:44 ` Daniel Arai
2007-03-08 19:14 ` Chris Wright
2007-03-08 19:17 ` Ingo Molnar
2007-03-08 19:42 ` Jeremy Fitzhardinge
2007-03-08 19:47 ` Chris Wright
2007-03-08 19:52 ` Jeremy Fitzhardinge
2007-03-08 20:10 ` Chris Wright
2007-03-08 20:18 ` Jeremy Fitzhardinge
2007-03-08 20:23 ` Chris Wright
2007-03-08 20:33 ` Jeremy Fitzhardinge
2007-03-08 20:42 ` Chris Wright
2007-03-08 20:42 ` Jeremy Fitzhardinge
2007-03-08 21:45 ` Andi Kleen
2007-03-08 19:54 ` Ingo Molnar
2007-03-08 9:10 ` hardwired VMI crap Ingo Molnar
2007-03-08 10:06 ` Zachary Amsden
2007-03-08 11:09 ` Thomas Gleixner
2007-03-08 20:46 ` Zachary Amsden
2007-03-08 21:13 ` Ingo Molnar
2007-03-08 22:17 ` Zachary Amsden
2007-03-08 22:33 ` Ingo Molnar
2007-03-08 22:39 ` Zachary Amsden
2007-03-16 10:12 ` Pavel Machek
2007-03-08 21:15 ` Jeremy Fitzhardinge
2007-03-08 21:34 ` Ingo Molnar
2007-03-08 21:43 ` Andi Kleen
2007-03-08 22:30 ` Ingo Molnar
2007-03-08 22:36 ` Zachary Amsden
2007-03-08 23:39 ` Jeremy Fitzhardinge
2007-03-08 23:55 ` Zachary Amsden
2007-03-09 0:10 ` Jeremy Fitzhardinge
2007-03-09 0:29 ` Linus Torvalds
2007-03-09 0:22 ` Daniel Walker
2007-03-09 0:28 ` Thomas Gleixner
2007-03-09 0:04 ` Thomas Gleixner
2007-03-09 0:44 ` Jeremy Fitzhardinge
2007-03-08 22:31 ` Zachary Amsden
2007-03-08 21:39 ` Andi Kleen
2007-03-08 22:58 ` Zachary Amsden
2007-03-08 22:42 ` Ingo Molnar
2007-03-08 23:39 ` Zachary Amsden
2007-03-08 18:35 ` Chris Wright
2007-03-07 23:33 ` + stupid-hack-to-make-mainline-build.patch added to -mm tree Jeremy Fitzhardinge
2007-03-07 23:52 ` Dan Hecht
2007-03-08 0:19 ` Jeremy Fitzhardinge
2007-03-08 0:35 ` Thomas Gleixner
2007-03-08 0:38 ` Jeremy Fitzhardinge
2007-03-07 20:40 ` Thomas Gleixner
2007-03-07 21:07 ` Jeremy Fitzhardinge
2007-03-07 21:40 ` Thomas Gleixner
2007-03-07 21:34 ` Dan Hecht
2007-03-07 22:14 ` Thomas Gleixner
2007-03-07 22:17 ` Zachary Amsden
2007-03-07 22:31 ` Thomas Gleixner
2007-03-07 22:28 ` Dan Hecht
2007-03-08 8:01 ` Ingo Molnar [this message]
2007-03-08 8:15 ` Keir Fraser
2007-03-08 8:41 ` Jeremy Fitzhardinge
2007-03-08 10:26 ` Rusty Russell
2007-03-07 21:42 ` Dan Hecht
2007-03-07 22:07 ` Thomas Gleixner
2007-03-07 5:10 ` Jeremy Fitzhardinge
2007-03-07 0:40 ` Thomas Gleixner
2007-03-07 0:42 ` Dan Hecht
2007-03-07 1:22 ` Thomas Gleixner
2007-03-07 1:44 ` Dan Hecht
2007-03-07 7:48 ` Thomas Gleixner
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=20070308080131.GB9295@elte.hu \
--to=mingo@elte.hu \
--cc=ak@suse.de \
--cc=akpm@linux-foundation.org \
--cc=dhecht@vmware.com \
--cc=jeremy@goop.org \
--cc=johnstul@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=rusty@rustcorp.com.au \
--cc=tglx@linutronix.de \
--cc=virtualization@lists.osdl.org \
--cc=zach@vmware.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
all inboxes | Powered by JetHome®