Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Reply-To: Sean Christopherson <seanjc@google.com>
Date:   Tue, 21 Mar 2023 15:00:08 -0700
Mime-Version: 1.0
Message-ID: <20230321220021.2119033-1-seanjc@google.com>
Subject: [PATCH v4 00/13] KVM: x86/mmu: Optimize clear dirty log
From:   Sean Christopherson <seanjc@google.com>
To:     Sean Christopherson <seanjc@google.com>,
        Paolo Bonzini <pbonzini@redhat.com>
Cc:     kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
        Vipin Sharma <vipinsh@google.com>,
        David Matlack <dmatlack@google.com>,
        Ben Gardon <bgardon@google.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

This is a massaged version of Vipin's series to optimize clearing dirty
state in the TDP MMU.  It's basically the same as v3, just spread out over
more patches.  The only meaningful difference in the end is that
clear_dirty_gfn_range() also gets similar treatment in handling Dirty vs.
Writable logic.

Vipin, I'm still planning on applying this for 6.4, but the changes ended
up being a wee bit bigger than I'm comfortable making on the fly, thus the
formal posting.

v4:
- Split patches into more fine-grained chunks.
- Massage changelogs as needed.
- Collect reviews. [David]

v3:
- https://lore.kernel.org/all/20230211014626.3659152-1-vipinsh@google.com
- Tried to do better job at writing commit messages.
- Made kvm_tdp_mmu_clear_spte_bits() similar to the kvm_tdp_mmu_write_spte().
- clear_dirty_pt_masked() evaluates mask for the bit to be cleared outside the
  loop and use that for all of the SPTEs instead of calculating for each SPTE.
- Some naming changes based on the feedbacks.
- Split out the dead code clean from the optimization code.


v2: https://lore.kernel.org/lkml/20230203192822.106773-1-vipinsh@google.com/
- Clear dirty log and age gfn range does not go through
  handle_changed_spte, they handle their SPTE changes locally to improve
  their speed.
- Clear only specific bits atomically when updating SPTEs in clearing
  dirty log and aging gfn range functions.
- Removed tdp_mmu_set_spte_no_[acc_track|dirty_log] APIs.
- Converged all handle_changed_spte related functions to one place.

v1: https://lore.kernel.org/lkml/20230125213857.824959-1-vipinsh@google.com

Vipin Sharma (13):
  KVM: x86/mmu: Add a helper function to check if an SPTE needs atomic
    write
  KVM: x86/mmu: Use kvm_ad_enabled() to determine if TDP MMU SPTEs need
    wrprot
  KVM: x86/mmu: Consolidate Dirty vs. Writable clearing logic in TDP MMU
  KVM: x86/mmu: Atomically clear SPTE dirty state in the clear-dirty-log
    flow
  KVM: x86/mmu: Drop access tracking checks when clearing TDP MMU dirty
    bits
  KVM: x86/mmu: Bypass __handle_changed_spte() when clearing TDP MMU
    dirty bits
  KVM: x86/mmu: Remove "record_dirty_log" in __tdp_mmu_set_spte()
  KVM: x86/mmu: Clear only A-bit (if enabled) when aging TDP MMU SPTEs
  KVM: x86/mmu: Drop unnecessary dirty log checks when aging TDP MMU
    SPTEs
  KVM: x86/mmu: Bypass __handle_changed_spte() when aging TDP MMU SPTEs
  KVM: x86/mmu: Remove "record_acc_track" in __tdp_mmu_set_spte()
  KVM: x86/mmu: Remove handle_changed_spte_dirty_log()
  KVM: x86/mmu: Merge all handle_changed_pte*() functions

 arch/x86/kvm/mmu/tdp_iter.h |  48 +++++---
 arch/x86/kvm/mmu/tdp_mmu.c  | 215 ++++++++++++------------------------
 2 files changed, 106 insertions(+), 157 deletions(-)


base-commit: f3d90f901d18749dca096719540a075f59240051
-- 
2.40.0.rc2.332.ga46443480c-goog