From: Paolo Bonzini <pbonzini@redhat.com>
To: Ben Gardon <bgardon@google.com>,
linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Cc: Cannon Matthews <cannonmatthews@google.com>,
Peter Xu <peterx@redhat.com>,
Sean Christopherson <sean.j.christopherson@intel.com>,
Peter Shier <pshier@google.com>,
Peter Feiner <pfeiner@google.com>,
Junaid Shahid <junaids@google.com>,
Jim Mattson <jmattson@google.com>,
Yulei Zhang <yulei.kernel@gmail.com>,
Wanpeng Li <kernellwp@gmail.com>,
Vitaly Kuznetsov <vkuznets@redhat.com>,
Xiao Guangrong <xiaoguangrong.eric@gmail.com>
Subject: Re: [PATCH 00/22] Introduce the TDP MMU
Date: Sat, 26 Sep 2020 03:14:01 +0200 [thread overview]
Message-ID: <34785fca-4d6d-88d7-f90a-2e17815d02e6@redhat.com> (raw)
In-Reply-To: <20200925212302.3979661-1-bgardon@google.com>
On 25/09/20 23:22, Ben Gardon wrote:
> Over the years, the needs for KVM's x86 MMU have grown from running small
> guests to live migrating multi-terabyte VMs with hundreds of vCPUs. Where
> we previously depended on shadow paging to run all guests, we now have
> two dimensional paging (TDP). This patch set introduces a new
> implementation of much of the KVM MMU, optimized for running guests with
> TDP. We have re-implemented many of the MMU functions to take advantage of
> the relative simplicity of TDP and eliminate the need for an rmap.
> Building on this simplified implementation, a future patch set will change
> the synchronization model for this "TDP MMU" to enable more parallelism
> than the monolithic MMU lock. A TDP MMU is currently in use at Google
> and has given us the performance necessary to live migrate our 416 vCPU,
> 12TiB m2-ultramem-416 VMs.
>
> This work was motivated by the need to handle page faults in parallel for
> very large VMs. When VMs have hundreds of vCPUs and terabytes of memory,
> KVM's MMU lock suffers extreme contention, resulting in soft-lockups and
> long latency on guest page faults. This contention can be easily seen
> running the KVM selftests demand_paging_test with a couple hundred vCPUs.
> Over a 1 second profile of the demand_paging_test, with 416 vCPUs and 4G
> per vCPU, 98% of the time was spent waiting for the MMU lock. At Google,
> the TDP MMU reduced the test duration by 89% and the execution was
> dominated by get_user_pages and the user fault FD ioctl instead of the
> MMU lock.
>
> This series is the first of two. In this series we add a basic
> implementation of the TDP MMU. In the next series we will improve the
> performance of the TDP MMU and allow it to execute MMU operations
> in parallel.
>
> The overall purpose of the KVM MMU is to program paging structures
> (CR3/EPT/NPT) to encode the mapping of guest addresses to host physical
> addresses (HPA), and to provide utilities for other KVM features, for
> example dirty logging. The definition of the L1 guest physical address
> (GPA) to HPA mapping comes in two parts: KVM's memslots map GPA to HVA,
> and the kernel MM/x86 host page tables map HVA -> HPA. Without TDP, the
> MMU must program the x86 page tables to encode the full translation of
> guest virtual addresses (GVA) to HPA. This requires "shadowing" the
> guest's page tables to create a composite x86 paging structure. This
> solution is complicated, requires separate paging structures for each
> guest CR3, and requires emulating guest page table changes. The TDP case
> is much simpler. In this case, KVM lets the guest control CR3 and programs
> the EPT/NPT paging structures with the GPA -> HPA mapping. The guest has
> no way to change this mapping and only one version of the paging structure
> is needed per L1 paging mode. In this case the paging mode is some
> combination of the number of levels in the paging structure, the address
> space (normal execution or system management mode, on x86), and other
> attributes. Most VMs only ever use 1 paging mode and so only ever need one
> TDP structure.
>
> This series implements a "TDP MMU" through alternative implementations of
> MMU functions for running L1 guests with TDP. The TDP MMU falls back to
> the existing shadow paging implementation when TDP is not available, and
> interoperates with the existing shadow paging implementation for nesting.
> The use of the TDP MMU can be controlled by a module parameter which is
> snapshot on VM creation and follows the life of the VM. This snapshot
> is used in many functions to decide whether or not to use TDP MMU handlers
> for a given operation.
>
> This series can also be viewed in Gerrit here:
> https://linux-review.googlesource.com/c/virt/kvm/kvm/+/2538
> (Thanks to Dmitry Vyukov <dvyukov@google.com> for setting up the
> Gerrit instance)
>
> Ben Gardon (22):
> kvm: mmu: Separate making SPTEs from set_spte
> kvm: mmu: Introduce tdp_iter
> kvm: mmu: Init / Uninit the TDP MMU
> kvm: mmu: Allocate and free TDP MMU roots
> kvm: mmu: Add functions to handle changed TDP SPTEs
> kvm: mmu: Make address space ID a property of memslots
> kvm: mmu: Support zapping SPTEs in the TDP MMU
> kvm: mmu: Separate making non-leaf sptes from link_shadow_page
> kvm: mmu: Remove disallowed_hugepage_adjust shadow_walk_iterator arg
> kvm: mmu: Add TDP MMU PF handler
> kvm: mmu: Factor out allocating a new tdp_mmu_page
> kvm: mmu: Allocate struct kvm_mmu_pages for all pages in TDP MMU
> kvm: mmu: Support invalidate range MMU notifier for TDP MMU
> kvm: mmu: Add access tracking for tdp_mmu
> kvm: mmu: Support changed pte notifier in tdp MMU
> kvm: mmu: Add dirty logging handler for changed sptes
> kvm: mmu: Support dirty logging for the TDP MMU
> kvm: mmu: Support disabling dirty logging for the tdp MMU
> kvm: mmu: Support write protection for nesting in tdp MMU
> kvm: mmu: NX largepage recovery for TDP MMU
> kvm: mmu: Support MMIO in the TDP MMU
> kvm: mmu: Don't clear write flooding count for direct roots
>
> arch/x86/include/asm/kvm_host.h | 17 +
> arch/x86/kvm/Makefile | 3 +-
> arch/x86/kvm/mmu/mmu.c | 437 ++++++----
> arch/x86/kvm/mmu/mmu_internal.h | 98 +++
> arch/x86/kvm/mmu/paging_tmpl.h | 3 +-
> arch/x86/kvm/mmu/tdp_iter.c | 198 +++++
> arch/x86/kvm/mmu/tdp_iter.h | 55 ++
> arch/x86/kvm/mmu/tdp_mmu.c | 1315 +++++++++++++++++++++++++++++++
> arch/x86/kvm/mmu/tdp_mmu.h | 52 ++
> include/linux/kvm_host.h | 2 +
> virt/kvm/kvm_main.c | 7 +-
> 11 files changed, 2022 insertions(+), 165 deletions(-)
> create mode 100644 arch/x86/kvm/mmu/tdp_iter.c
> create mode 100644 arch/x86/kvm/mmu/tdp_iter.h
> create mode 100644 arch/x86/kvm/mmu/tdp_mmu.c
> create mode 100644 arch/x86/kvm/mmu/tdp_mmu.h
>
Ok, I've not finished reading the code but I have already an idea of
what it's like. I really think we should fast track this as the basis
for more 5.11 work. I'll finish reviewing it and, if you don't mind, I
might make some of the changes myself so I have the occasion to play and
get accustomed to the code; speak up if you disagree with them though!
Another thing I'd like to add is a few tracepoints.
Paolo
next prev parent reply other threads:[~2020-09-26 1:14 UTC|newest]
Thread overview: 106+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-25 21:22 Ben Gardon
2020-09-25 21:22 ` [PATCH 01/22] kvm: mmu: Separate making SPTEs from set_spte Ben Gardon
2020-09-30 4:55 ` Sean Christopherson
2020-09-30 23:03 ` Ben Gardon
2020-09-25 21:22 ` [PATCH 02/22] kvm: mmu: Introduce tdp_iter Ben Gardon
2020-09-26 0:04 ` Paolo Bonzini
2020-09-30 5:06 ` Sean Christopherson
2020-09-26 0:54 ` Paolo Bonzini
2020-09-30 5:08 ` Sean Christopherson
2020-09-30 5:24 ` Sean Christopherson
2020-09-30 6:24 ` Paolo Bonzini
2020-09-30 23:20 ` Eric van Tassell
2020-09-30 23:34 ` Paolo Bonzini
2020-10-01 0:07 ` Sean Christopherson
2020-09-25 21:22 ` [PATCH 03/22] kvm: mmu: Init / Uninit the TDP MMU Ben Gardon
2020-09-26 0:06 ` Paolo Bonzini
2020-09-30 5:34 ` Sean Christopherson
2020-09-30 18:36 ` Ben Gardon
2020-09-30 16:57 ` Sean Christopherson
2020-09-30 17:39 ` Paolo Bonzini
2020-09-30 18:42 ` Ben Gardon
2020-09-25 21:22 ` [PATCH 04/22] kvm: mmu: Allocate and free TDP MMU roots Ben Gardon
2020-09-30 6:06 ` Sean Christopherson
2020-09-30 6:26 ` Paolo Bonzini
2020-09-30 15:38 ` Sean Christopherson
2020-10-12 22:59 ` Ben Gardon
2020-10-12 23:59 ` Sean Christopherson
2020-09-25 21:22 ` [PATCH 05/22] kvm: mmu: Add functions to handle changed TDP SPTEs Ben Gardon
2020-09-26 0:39 ` Paolo Bonzini
2020-09-28 17:23 ` Paolo Bonzini
2020-09-25 21:22 ` [PATCH 06/22] kvm: mmu: Make address space ID a property of memslots Ben Gardon
2020-09-30 6:10 ` Sean Christopherson
2020-09-30 23:11 ` Ben Gardon
2020-09-25 21:22 ` [PATCH 07/22] kvm: mmu: Support zapping SPTEs in the TDP MMU Ben Gardon
2020-09-26 0:14 ` Paolo Bonzini
2020-09-30 6:15 ` Sean Christopherson
2020-09-30 6:28 ` Paolo Bonzini
2020-09-25 21:22 ` [PATCH 08/22] kvm: mmu: Separate making non-leaf sptes from link_shadow_page Ben Gardon
2020-09-25 21:22 ` [PATCH 09/22] kvm: mmu: Remove disallowed_hugepage_adjust shadow_walk_iterator arg Ben Gardon
2020-09-30 16:19 ` Sean Christopherson
2020-09-25 21:22 ` [PATCH 10/22] kvm: mmu: Add TDP MMU PF handler Ben Gardon
2020-09-26 0:24 ` Paolo Bonzini
2020-09-30 16:37 ` Sean Christopherson
2020-09-30 16:55 ` Paolo Bonzini
2020-09-30 17:37 ` Paolo Bonzini
2020-10-06 22:35 ` Ben Gardon
2020-10-06 22:33 ` Ben Gardon
2020-10-07 20:55 ` Sean Christopherson
2020-09-25 21:22 ` [PATCH 11/22] kvm: mmu: Factor out allocating a new tdp_mmu_page Ben Gardon
2020-09-26 0:22 ` Paolo Bonzini
2020-09-30 18:53 ` Ben Gardon
2020-09-25 21:22 ` [PATCH 12/22] kvm: mmu: Allocate struct kvm_mmu_pages for all pages in TDP MMU Ben Gardon
2020-09-25 21:22 ` [PATCH 13/22] kvm: mmu: Support invalidate range MMU notifier for " Ben Gardon
2020-09-30 17:03 ` Sean Christopherson
2020-09-30 23:15 ` Ben Gardon
2020-09-30 23:24 ` Sean Christopherson
2020-09-30 23:27 ` Ben Gardon
2020-09-25 21:22 ` [PATCH 14/22] kvm: mmu: Add access tracking for tdp_mmu Ben Gardon
2020-09-26 0:32 ` Paolo Bonzini
2020-09-30 17:48 ` Sean Christopherson
2020-10-06 23:38 ` Ben Gardon
2020-09-25 21:22 ` [PATCH 15/22] kvm: mmu: Support changed pte notifier in tdp MMU Ben Gardon
2020-09-26 0:33 ` Paolo Bonzini
2020-09-28 15:11 ` Paolo Bonzini
2020-10-07 16:53 ` Ben Gardon
2020-10-07 17:18 ` Paolo Bonzini
2020-10-07 17:30 ` Ben Gardon
2020-10-07 17:54 ` Paolo Bonzini
2020-09-25 21:22 ` [PATCH 16/22] kvm: mmu: Add dirty logging handler for changed sptes Ben Gardon
2020-09-26 0:45 ` Paolo Bonzini
2020-09-25 21:22 ` [PATCH 17/22] kvm: mmu: Support dirty logging for the TDP MMU Ben Gardon
2020-09-26 1:04 ` Paolo Bonzini
2020-10-08 18:27 ` Ben Gardon
2020-09-29 15:07 ` Paolo Bonzini
2020-09-30 18:04 ` Sean Christopherson
2020-09-30 18:08 ` Paolo Bonzini
2020-09-25 21:22 ` [PATCH 18/22] kvm: mmu: Support disabling dirty logging for the tdp MMU Ben Gardon
2020-09-26 1:09 ` Paolo Bonzini
2020-10-07 16:30 ` Ben Gardon
2020-10-07 17:21 ` Paolo Bonzini
2020-10-07 17:28 ` Ben Gardon
2020-10-07 17:53 ` Paolo Bonzini
2020-09-25 21:22 ` [PATCH 19/22] kvm: mmu: Support write protection for nesting in " Ben Gardon
2020-09-30 18:06 ` Sean Christopherson
2020-09-25 21:23 ` [PATCH 20/22] kvm: mmu: NX largepage recovery for TDP MMU Ben Gardon
2020-09-26 1:14 ` Paolo Bonzini
2020-09-30 22:23 ` Ben Gardon
2020-09-29 18:24 ` Paolo Bonzini
2020-09-30 18:15 ` Sean Christopherson
2020-09-30 19:56 ` Paolo Bonzini
2020-09-30 22:33 ` Ben Gardon
2020-09-30 22:27 ` Ben Gardon
2020-09-25 21:23 ` [PATCH 21/22] kvm: mmu: Support MMIO in the " Ben Gardon
2020-09-30 18:19 ` Sean Christopherson
2020-09-25 21:23 ` [PATCH 22/22] kvm: mmu: Don't clear write flooding count for direct roots Ben Gardon
2020-09-26 1:25 ` Paolo Bonzini
2020-10-05 22:48 ` Ben Gardon
2020-10-05 23:44 ` Sean Christopherson
2020-10-06 16:19 ` Ben Gardon
2020-09-26 1:14 ` Paolo Bonzini [this message]
2020-09-28 17:31 ` [PATCH 00/22] Introduce the TDP MMU Paolo Bonzini
2020-09-29 17:40 ` Ben Gardon
2020-09-29 18:10 ` Paolo Bonzini
2020-09-30 6:19 ` Sean Christopherson
2020-09-30 6:30 ` Paolo Bonzini
2020-10-23 16:30 Paolo Bonzini
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=34785fca-4d6d-88d7-f90a-2e17815d02e6@redhat.com \
--to=pbonzini@redhat.com \
--cc=bgardon@google.com \
--cc=cannonmatthews@google.com \
--cc=jmattson@google.com \
--cc=junaids@google.com \
--cc=kernellwp@gmail.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=peterx@redhat.com \
--cc=pfeiner@google.com \
--cc=pshier@google.com \
--cc=sean.j.christopherson@intel.com \
--cc=vkuznets@redhat.com \
--cc=xiaoguangrong.eric@gmail.com \
--cc=yulei.kernel@gmail.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®