From: Ben Gardon <bgardon@google.com>
To: Sean Christopherson <seanjc@google.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
Vitaly Kuznetsov <vkuznets@redhat.com>,
Wanpeng Li <wanpengli@tencent.com>,
Jim Mattson <jmattson@google.com>, Joerg Roedel <joro@8bytes.org>,
kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
Hou Wenlong <houwenlong93@linux.alibaba.com>
Subject: Re: [PATCH 25/28] KVM: x86/mmu: Require mmu_lock be held for write to zap TDP MMU range
Date: Tue, 23 Nov 2021 11:58:30 -0800 [thread overview]
Message-ID: <CANgfPd8EmhE3wWCp9cYat-GQ_uB83TTyLXyMj0tBXhnZ1yVwig@mail.gmail.com> (raw)
In-Reply-To: <20211120045046.3940942-26-seanjc@google.com>
On Fri, Nov 19, 2021 at 8:51 PM Sean Christopherson <seanjc@google.com> wrote:
>
> Now that all callers of zap_gfn_range() hold mmu_lock for write, drop
> support for zapping with mmu_lock held for read. That all callers hold
> mmu_lock for write isn't a random coincedence; now that the paths that
> need to zap _everything_ have their own path, the only callers left are
> those that need to zap for functional correctness. And when zapping is
> required for functional correctness, mmu_lock must be held for write,
> otherwise the caller has no guarantees about the state of the TDP MMU
> page tables after it has run, e.g. the SPTE(s) it zapped can be
> immediately replaced by a vCPU faulting in a page.
>
> Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
> ---
> arch/x86/kvm/mmu/tdp_mmu.c | 29 ++++++-----------------------
> 1 file changed, 6 insertions(+), 23 deletions(-)
>
> diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
> index 0e5a0d40e54a..926e92473e92 100644
> --- a/arch/x86/kvm/mmu/tdp_mmu.c
> +++ b/arch/x86/kvm/mmu/tdp_mmu.c
> @@ -844,15 +844,9 @@ bool kvm_tdp_mmu_zap_sp(struct kvm *kvm, struct kvm_mmu_page *sp)
> * function cannot yield, it will not release the MMU lock or reschedule and
> * the caller must ensure it does not supply too large a GFN range, or the
> * operation can cause a soft lockup.
> - *
> - * If shared is true, this thread holds the MMU lock in read mode and must
> - * account for the possibility that other threads are modifying the paging
> - * structures concurrently. If shared is false, this thread should hold the
> - * MMU lock in write mode.
> */
> static bool zap_gfn_range(struct kvm *kvm, struct kvm_mmu_page *root,
> - gfn_t start, gfn_t end, bool can_yield, bool flush,
> - bool shared)
> + gfn_t start, gfn_t end, bool can_yield, bool flush)
> {
> bool zap_all = (start == 0 && end >= tdp_mmu_max_gfn_host());
> struct tdp_iter iter;
> @@ -865,15 +859,14 @@ static bool zap_gfn_range(struct kvm *kvm, struct kvm_mmu_page *root,
>
> end = min(end, tdp_mmu_max_gfn_host());
>
> - kvm_lockdep_assert_mmu_lock_held(kvm, shared);
> + lockdep_assert_held_write(&kvm->mmu_lock);
>
> rcu_read_lock();
>
> for_each_tdp_pte_min_level(iter, root->spt, root->role.level,
> min_level, start, end) {
> -retry:
> if (can_yield &&
> - tdp_mmu_iter_cond_resched(kvm, &iter, flush, shared)) {
> + tdp_mmu_iter_cond_resched(kvm, &iter, flush, false)) {
> flush = false;
> continue;
> }
> @@ -892,17 +885,8 @@ static bool zap_gfn_range(struct kvm *kvm, struct kvm_mmu_page *root,
> !is_last_spte(iter.old_spte, iter.level))
> continue;
>
> - if (!shared) {
> - tdp_mmu_set_spte(kvm, &iter, 0);
> - flush = true;
> - } else if (!tdp_mmu_zap_spte_atomic(kvm, &iter)) {
> - /*
> - * The iter must explicitly re-read the SPTE because
> - * the atomic cmpxchg failed.
> - */
> - iter.old_spte = kvm_tdp_mmu_read_spte(iter.sptep);
> - goto retry;
> - }
> + tdp_mmu_set_spte(kvm, &iter, 0);
> + flush = true;
> }
>
> rcu_read_unlock();
> @@ -921,8 +905,7 @@ bool __kvm_tdp_mmu_zap_gfn_range(struct kvm *kvm, int as_id, gfn_t start,
> struct kvm_mmu_page *root;
>
> for_each_tdp_mmu_root_yield_safe(kvm, root, as_id, false)
> - flush = zap_gfn_range(kvm, root, start, end, can_yield, flush,
> - false);
> + flush = zap_gfn_range(kvm, root, start, end, can_yield, flush);
>
> return flush;
> }
> --
> 2.34.0.rc2.393.gf8c9666880-goog
>
next prev parent reply other threads:[~2021-11-23 19:58 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-20 4:50 [PATCH 00/28] KVM: x86/mmu: Overhaul TDP MMU zapping and flushing Sean Christopherson
2021-11-20 4:50 ` [PATCH 01/28] KVM: x86/mmu: Use yield-safe TDP MMU root iter in MMU notifier unmapping Sean Christopherson
2021-11-22 19:48 ` Ben Gardon
2021-11-30 8:03 ` Paolo Bonzini
2021-11-20 4:50 ` [PATCH 02/28] KVM: x86/mmu: Skip tlb flush if it has been done in zap_gfn_range() Sean Christopherson
2021-11-20 4:50 ` [PATCH 03/28] KVM: x86/mmu: Remove spurious TLB flushes in TDP MMU zap collapsible path Sean Christopherson
2021-11-20 4:50 ` [PATCH 04/28] KVM: x86/mmu: Retry page fault if root is invalidated by memslot update Sean Christopherson
2021-11-22 19:54 ` Ben Gardon
2021-12-01 20:49 ` Paolo Bonzini
2021-12-08 19:17 ` Sean Christopherson
2021-11-20 4:50 ` [PATCH 05/28] KVM: x86/mmu: Check for present SPTE when clearing dirty bit in TDP MMU Sean Christopherson
2021-11-22 19:57 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 06/28] KVM: x86/mmu: Formalize TDP MMU's (unintended?) deferred TLB flush logic Sean Christopherson
2021-11-20 4:50 ` [PATCH 07/28] KVM: x86/mmu: Document that zapping invalidated roots doesn't need to flush Sean Christopherson
2021-11-20 4:50 ` [PATCH 08/28] KVM: x86/mmu: Drop unused @kvm param from kvm_tdp_mmu_get_root() Sean Christopherson
2021-11-22 20:02 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 09/28] KVM: x86/mmu: Require mmu_lock be held for write in unyielding root iter Sean Christopherson
2021-11-22 20:10 ` Ben Gardon
2021-11-22 20:19 ` Sean Christopherson
2021-11-20 4:50 ` [PATCH 10/28] KVM: x86/mmu: Allow yielding when zapping GFNs for defunct TDP MMU root Sean Christopherson
2021-11-22 21:30 ` Ben Gardon
2021-11-22 22:40 ` Sean Christopherson
2021-11-22 23:03 ` Ben Gardon
2021-12-14 23:45 ` Sean Christopherson
2021-12-14 23:52 ` Sean Christopherson
2021-11-20 4:50 ` [PATCH 11/28] KVM: x86/mmu: Check for !leaf=>leaf, not PFN change, in TDP MMU SP removal Sean Christopherson
2021-11-20 4:50 ` [PATCH 12/28] KVM: x86/mmu: Batch TLB flushes from TDP MMU for MMU notifier change_spte Sean Christopherson
2021-11-22 21:45 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 13/28] KVM: x86/mmu: Drop RCU after processing each root in MMU notifier hooks Sean Christopherson
2021-11-22 21:47 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 14/28] KVM: x86/mmu: Add helpers to read/write TDP MMU SPTEs and document RCU Sean Christopherson
2021-11-22 21:55 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 15/28] KVM: x86/mmu: Take TDP MMU roots off list when invalidating all roots Sean Christopherson
2021-11-22 22:20 ` Ben Gardon
2021-11-22 23:08 ` Sean Christopherson
2021-11-23 0:03 ` Ben Gardon
2021-12-14 23:34 ` Sean Christopherson
2021-11-20 4:50 ` [PATCH 16/28] KVM: x86/mmu: WARN if old _or_ new SPTE is REMOVED in non-atomic path Sean Christopherson
2021-11-22 21:57 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 17/28] KVM: x86/mmu: Terminate yield-friendly walk if invalid root observed Sean Christopherson
2021-11-22 22:25 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 18/28] KVM: x86/mmu: Refactor low-level TDP MMU set SPTE helper to take raw vals Sean Christopherson
2021-11-22 22:29 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 19/28] KVM: x86/mmu: Zap only the target TDP MMU shadow page in NX recovery Sean Christopherson
2021-11-22 22:43 ` Ben Gardon
2021-11-23 1:16 ` Sean Christopherson
2021-11-23 19:35 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 20/28] KVM: x86/mmu: Use common TDP MMU zap helper for MMU notifier unmap hook Sean Christopherson
2021-11-22 22:49 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 21/28] KVM: x86/mmu: Add TDP MMU helper to zap a root Sean Christopherson
2021-11-22 22:54 ` Ben Gardon
2021-11-22 23:15 ` Sean Christopherson
2021-11-22 23:38 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 22/28] KVM: x86/mmu: Skip remote TLB flush when zapping all of TDP MMU Sean Christopherson
2021-11-22 23:00 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 23/28] KVM: x86/mmu: Use "zap root" path for "slow" zap of all TDP MMU SPTEs Sean Christopherson
2021-11-20 4:50 ` [PATCH 24/28] KVM: x86/mmu: Add dedicated helper to zap TDP MMU root shadow page Sean Christopherson
2021-11-23 1:04 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 25/28] KVM: x86/mmu: Require mmu_lock be held for write to zap TDP MMU range Sean Christopherson
2021-11-23 19:58 ` Ben Gardon [this message]
2021-11-20 4:50 ` [PATCH 26/28] KVM: x86/mmu: Zap only TDP MMU leafs in kvm_zap_gfn_range() Sean Christopherson
2021-11-23 19:58 ` Ben Gardon
2021-11-20 4:50 ` [PATCH 27/28] KVM: x86/mmu: Do remote TLB flush before dropping RCU in TDP MMU resched Sean Christopherson
2021-11-23 19:58 ` Ben Gardon
2021-11-24 18:42 ` Sean Christopherson
2021-11-30 11:29 ` Paolo Bonzini
2021-11-30 15:45 ` Sean Christopherson
2021-11-30 16:16 ` Paolo Bonzini
2021-11-20 4:50 ` [PATCH 28/28] KVM: x86/mmu: Defer TLB flush to caller when freeing TDP MMU shadow pages Sean Christopherson
2021-11-23 20:12 ` Ben Gardon
2021-12-01 17:53 ` [PATCH 00/28] KVM: x86/mmu: Overhaul TDP MMU zapping and flushing David Matlack
2021-12-02 2:03 ` Sean Christopherson
2021-12-03 0:16 ` David Matlack
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=CANgfPd8EmhE3wWCp9cYat-GQ_uB83TTyLXyMj0tBXhnZ1yVwig@mail.gmail.com \
--to=bgardon@google.com \
--cc=houwenlong93@linux.alibaba.com \
--cc=jmattson@google.com \
--cc=joro@8bytes.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=seanjc@google.com \
--cc=vkuznets@redhat.com \
--cc=wanpengli@tencent.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;
as well as URLs for NNTP newsgroup(s).