mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Vipin Sharma <vipinsh@google.com>
To: seanjc@google.com, pbonzini@redhat.com, bgardon@google.com,
	dmatlack@google.com
Cc: jmattson@google.com, mizhang@google.com, kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org, Vipin Sharma <vipinsh@google.com>
Subject: [Patch v4 11/18] KVM: x86/mmu: Add documentation of NUMA aware page table capability
Date: Mon,  6 Mar 2023 14:41:20 -0800	[thread overview]
Message-ID: <20230306224127.1689967-12-vipinsh@google.com> (raw)
In-Reply-To: <20230306224127.1689967-1-vipinsh@google.com>

Add documentation for KVM_CAP_NUMA_AWARE_PAGE_TABLE capability and
explain why it is needed.

Signed-off-by: Vipin Sharma <vipinsh@google.com>
---
 Documentation/virt/kvm/api.rst | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst
index 62de0768d6aa..7e3a1299ca8e 100644
--- a/Documentation/virt/kvm/api.rst
+++ b/Documentation/virt/kvm/api.rst
@@ -7669,6 +7669,35 @@ This capability is aimed to mitigate the threat that malicious VMs can
 cause CPU stuck (due to event windows don't open up) and make the CPU
 unavailable to host or other VMs.
 
+7.34 KVM_CAP_NUMA_AWARE_PAGE_TABLE
+------------------------------
+
+:Architectures: x86
+:Target: VM
+:Returns: 0 on success, -EINVAL if vCPUs are already created.
+
+This capability allows userspace to enable NUMA aware page tables allocations.
+NUMA aware page tables are disabled by default. Once enabled, prior to vCPU
+creation, any page table allocated during the life of a VM will be allocated
+preferably from the NUMA node of the leaf page.
+
+Without this capability, default feature is to use current thread mempolicy and
+allocate page table based on that.
+
+This capability is useful to improve page accesses by a guest. For example, an
+initialization thread which access lots of remote memory and ends up creating
+page tables on local NUMA node, or some service thread allocates memory on
+remote NUMA nodes and later worker/background threads accessing that memory
+will end up accessing remote NUMA node page tables. So, a multi NUMA node
+guest, can with high confidence access local memory faster instead of going
+through remote page tables first.
+
+This capability is also helpful for host to reduce live migration impact when
+splitting huge pages during dirty log operations. If the thread splitting huge
+page is on remote NUMA node it will create page tables on remote node. Even if
+guest is careful in making sure that it only access local memory they will end
+up accessing remote page tables.
+
 8. Other capabilities.
 ======================
 
-- 
2.40.0.rc0.216.gc4246ad0f0-goog


  parent reply	other threads:[~2023-03-06 22:42 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-06 22:41 [Patch v4 00/18] NUMA aware page table allocation Vipin Sharma
2023-03-06 22:41 ` [Patch v4 01/18] KVM: x86/mmu: Change KVM mmu shrinker to no-op Vipin Sharma
2023-03-06 22:41 ` [Patch v4 02/18] KVM: x86/mmu: Remove zapped_obsolete_pages from struct kvm_arch{} Vipin Sharma
2023-03-06 22:41 ` [Patch v4 03/18] KVM: x86/mmu: Track count of pages in KVM MMU page caches globally Vipin Sharma
2023-03-07 11:32   ` kernel test robot
2023-03-07 19:13     ` Vipin Sharma
2023-03-07 20:18       ` Sean Christopherson
2023-03-07 12:13   ` kernel test robot
2023-03-08 20:33   ` Zhi Wang
2023-03-08 22:16     ` Vipin Sharma
2023-03-09  5:18       ` Mingwei Zhang
2023-03-09 12:52         ` Zhi Wang
2023-03-09 19:52           ` Vipin Sharma
2023-03-09 15:37   ` Zhi Wang
2023-03-09 18:19     ` Vipin Sharma
2023-03-09 23:53   ` David Matlack
2023-03-10  0:28     ` Vipin Sharma
2023-03-10  0:55       ` David Matlack
2023-03-10  1:09         ` Vipin Sharma
2023-03-10  0:22   ` David Matlack
2023-03-10  0:36     ` Vipin Sharma
2023-03-06 22:41 ` [Patch v4 04/18] KVM: x86/mmu: Shrink shadow page caches via MMU shrinker Vipin Sharma
2023-03-06 22:41 ` [Patch v4 05/18] KVM: x86/mmu: Add split_shadow_page_cache pages to global count of MMU cache pages Vipin Sharma
2023-03-09 15:58   ` Zhi Wang
2023-03-09 19:59     ` Vipin Sharma
2023-03-10  0:05       ` David Matlack
2023-03-10  0:06         ` David Matlack
2023-03-06 22:41 ` [Patch v4 06/18] KVM: x86/mmu: Shrink split_shadow_page_cache via MMU shrinker Vipin Sharma
2023-03-09 16:01   ` Zhi Wang
2023-03-09 19:59     ` Vipin Sharma
2023-03-06 22:41 ` [Patch v4 07/18] KVM: x86/mmu: Unconditionally count allocations from MMU page caches Vipin Sharma
2023-03-09 16:03   ` Zhi Wang
2023-03-06 22:41 ` [Patch v4 08/18] KVM: x86/mmu: Track unused mmu_shadowed_info_cache pages count via global counter Vipin Sharma
2023-03-30  4:53   ` Yang, Weijiang
2023-04-03 23:02     ` Vipin Sharma
2023-03-06 22:41 ` [Patch v4 09/18] KVM: x86/mmu: Shrink mmu_shadowed_info_cache via MMU shrinker Vipin Sharma
2023-03-06 22:41 ` [Patch v4 10/18] KVM: x86/mmu: Add per VM NUMA aware page table capability Vipin Sharma
2023-03-06 22:41 ` Vipin Sharma [this message]
2023-03-23 21:59   ` [Patch v4 11/18] KVM: x86/mmu: Add documentation of " David Matlack
2023-03-28 16:47     ` Vipin Sharma
2023-03-06 22:41 ` [Patch v4 12/18] KVM: x86/mmu: Allocate NUMA aware page tables on TDP huge page splits Vipin Sharma
2023-03-23 22:15   ` David Matlack
2023-03-28 17:12     ` Vipin Sharma
2023-03-06 22:41 ` [Patch v4 13/18] KVM: mmu: Add common initialization logic for struct kvm_mmu_memory_cache{} Vipin Sharma
2023-03-23 22:23   ` David Matlack
2023-03-28 17:16     ` Vipin Sharma
2023-03-06 22:41 ` [Patch v4 14/18] KVM: mmu: Initialize kvm_mmu_memory_cache.gfp_zero to __GFP_ZERO by default Vipin Sharma
2023-03-23 22:28   ` David Matlack
2023-03-28 17:31     ` Vipin Sharma
2023-03-28 23:13       ` David Matlack
2023-03-06 22:41 ` [Patch v4 15/18] KVM: mmu: Add NUMA node support in struct kvm_mmu_memory_cache{} Vipin Sharma
2023-03-23 22:30   ` David Matlack
2023-03-28 17:50     ` Vipin Sharma
2023-03-28 23:24       ` David Matlack
2023-04-03 22:57         ` Vipin Sharma
2023-03-06 22:41 ` [Patch v4 16/18] KVM: x86/mmu: Allocate numa aware page tables during page fault Vipin Sharma
2023-03-29  0:21   ` David Matlack
2023-03-29  0:28     ` David Matlack
2023-03-29 19:03     ` David Matlack
2023-04-03 22:54       ` Vipin Sharma
2023-04-03 22:50     ` Vipin Sharma
2023-03-06 22:41 ` [Patch v4 17/18] KVM: x86/mmu: Allocate shadow mmu page table on huge page split on the same NUMA node Vipin Sharma
2023-03-06 22:41 ` [Patch v4 18/18] KVM: x86/mmu: Reduce default mmu memory cache size Vipin Sharma
2023-03-07 18:19 ` [Patch v4 00/18] NUMA aware page table allocation Mingwei Zhang
2023-03-07 18:33   ` Vipin Sharma

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=20230306224127.1689967-12-vipinsh@google.com \
    --to=vipinsh@google.com \
    --cc=bgardon@google.com \
    --cc=dmatlack@google.com \
    --cc=jmattson@google.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mizhang@google.com \
    --cc=pbonzini@redhat.com \
    --cc=seanjc@google.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®