From: Andrew Morton <akpm-3NddpPZAyC0@public.gmane.org>
To: Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
Cc: kvm-devel
<kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org>,
linux-kernel
<linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
Subject: Re: [PATCH 0/33] KVM: MMU: Cache shadow page tables
Date: Thu, 4 Jan 2007 09:22:26 -0800 [thread overview]
Message-ID: <20070104092226.91fa2dfe.akpm@osdl.org> (raw)
In-Reply-To: <459D21DD.5090506-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
On Thu, 04 Jan 2007 17:48:45 +0200
Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org> wrote:
> The current kvm shadow page table implementation does not cache shadow
> page tables (except for global translations, used for kernel addresses)
> across context switches. This means that after a context switch, every
> memory access will trap into the host. After a while, the shadow page
> tables will be rebuild, and the guest can proceed at native speed until
> the next context switch.
>
> The natural solution, then, is to cache shadow page tables across
> context switches. Unfortunately, this introduces a bucketload of problems:
>
> - the guest does not notify the processor (and hence kvm) that it
> modifies a page table entry if it has reason to believe that the
> modification will be followed by a tlb flush. It becomes necessary to
> write-protect guest page tables so that we can use the page fault when
> the access occurs as a notification.
> - write protecting the guest page tables means we need to keep track of
> which ptes map those guest page table. We need to add reverse mapping
> for all mapped writable guest pages.
> - when the guest does access the write-protected page, we need to allow
> it to perform the write in some way. We do that either by emulating the
> write, or removing all shadow page tables for that page and allowing the
> write to proceed, depending on circumstances.
>
> This patchset implements the ideas above. While a lot of tuning remains
> to be done (for example, a sane page replacement algorithm), a guest
> running with this patchset applied is much faster and more responsive
> than with 2.6.20-rc3. Some preliminary benchmarks are available in
> http://article.gmane.org/gmane.comp.emulators.kvm.devel/661.
>
> The patchset is bisectable compile-wise.
Is this intended for 2.6.20, or would you prefer that we release what we
have now and hold this off for 2.6.21?
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
next prev parent reply other threads:[~2007-01-04 17:22 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-01-04 15:48 [PATCH 0/33] KVM: MMU: Cache shadow page tables Avi Kivity
2007-01-04 15:50 ` [PATCH 1/33] KVM: MMU: Implement simple reverse mapping Avi Kivity
2007-01-04 16:05 ` [PATCH 16/33] KVM: MMU: kvm_mmu_put_page() only removes one link to the page Avi Kivity
2007-01-04 16:09 ` [PATCH 20/33] KVM: MMU: Handle misaligned accesses to write protected guest page tables Avi Kivity
2007-01-04 16:12 ` [PATCH 23/33] KVM: MMU: If an empty shadow page is not empty, report more info Avi Kivity
[not found] ` <459D21DD.5090506-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
2007-01-04 15:51 ` [PATCH 2/33] KVM: MMU: Teach the page table walker to track guest page table gfns Avi Kivity
2007-01-04 15:52 ` [PATCH 3/33] KVM: MMU: Load the pae pdptrs on cr3 change like the processor does Avi Kivity
2007-01-04 15:53 ` [PATCH 4/33] KVM: MMU: Fold fetch_guest() into init_walker() Avi Kivity
2007-01-04 15:54 ` [PATCH 5/33] KVM: MU: Special treatment for shadow pae root pages Avi Kivity
2007-01-04 15:55 ` [PATCH 6/33] KVM: MMU: Use the guest pdptrs instead of mapping cr3 in pae mode Avi Kivity
2007-01-04 15:56 ` [PATCH 7/33] KVM: MMU: Make the shadow page tables also special-case pae Avi Kivity
2007-01-04 15:57 ` [PATCH 8/33] KVM: MMU: Make kvm_mmu_alloc_page() return a kvm_mmu_page pointer Avi Kivity
2007-01-04 15:58 ` [PATCH 9/33] KVM: MMU: Shadow page table caching Avi Kivity
2007-01-04 15:59 ` [PATCH 10/33] KVM: MMU: Write protect guest pages when a shadow is created for them Avi Kivity
2007-01-04 16:00 ` [PATCH 11/33] KVM: MMU: Let the walker extract the target page gfn from the pte Avi Kivity
2007-01-04 16:01 ` [PATCH 12/33] KVM: MMU: Support emulated writes into RAM Avi Kivity
2007-01-04 16:02 ` [PATCH 13/33] KVM: MMU: Zap shadow page table entries on writes to guest page tables Avi Kivity
2007-01-04 16:03 ` [PATCH 14/33] KVM: MMU: If emulating an instruction fails, try unprotecting the page Avi Kivity
2007-01-04 16:04 ` [PATCH 15/33] KVM: MMU: Implement child shadow unlinking Avi Kivity
2007-01-04 16:06 ` [PATCH 17/33] KVM: MMU: oom handling Avi Kivity
2007-01-04 16:07 ` [PATCH 18/33] KVM: MMU: Remove invlpg interception Avi Kivity
2007-01-04 16:08 ` [PATCH 19/33] KVM: MMU: Remove release_pt_page_64() Avi Kivity
2007-01-04 16:10 ` [PATCH 21/33] KVM: MMU: <ove is_empty_shadow_page() above kvm_mmu_free_page() Avi Kivity
2007-01-04 16:11 ` [PATCH 22/33] KVM: MMU: Ensure freed shadow pages are clean Avi Kivity
2007-01-04 16:13 ` [PATCH 24/33] KVM: MMU: Page table write flood protection Avi Kivity
2007-01-04 16:14 ` [PATCH 25/33] KVM: MMU: Never free a shadow page actively serving as a root Avi Kivity
2007-01-04 16:15 ` [PATCH 26/33] KVM: MMU: Fix cmpxchg8b emulation Avi Kivity
2007-01-04 16:16 ` [PATCH 27/33] KVM: MMU: Treat user-mode faults as a hint that a page is no longer a page table Avi Kivity
2007-01-04 16:17 ` [PATCH 28/33] KVM: MMU: Free pages on kvm destruction Avi Kivity
2007-01-04 16:18 ` [PATCH 29/33] KVM: MMU: Replace atomic allocations by preallocated objects Avi Kivity
2007-01-04 16:19 ` [PATCH 30/33] KVM: MMU: Detect oom conditions and propagate error to userspace Avi Kivity
2007-01-04 16:20 ` [PATCH 31/33] KVM: MMU: Flush guest tlb when reducing permissions on a pte Avi Kivity
2007-01-04 16:21 ` [PATCH 32/33] KVM: MMU: Destroy mmu while we still have a vcpu left Avi Kivity
2007-01-04 16:22 ` [PATCH 33/33] KVM: MMU: add audit code to check mappings, etc are correct Avi Kivity
2007-01-04 17:22 ` Andrew Morton [this message]
[not found] ` <20070104092226.91fa2dfe.akpm-3NddpPZAyC0@public.gmane.org>
2007-01-04 17:41 ` [PATCH 0/33] KVM: MMU: Cache shadow page tables Avi Kivity
[not found] ` <459D3C65.2090703-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
2007-01-04 18:02 ` Ingo Molnar
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=20070104092226.91fa2dfe.akpm@osdl.org \
--to=akpm-3nddppzayc0@public.gmane.org \
--cc=avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org \
--cc=kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.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