From: Ben Gardon <bgardon@google.com>
To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Cc: Paolo Bonzini <pbonzini@redhat.com>, Peter Xu <peterx@redhat.com>,
Sean Christopherson <seanjc@google.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>,
Ben Gardon <bgardon@google.com>
Subject: [PATCH 13/13] KVM: x86/mmu: Tear down roots in fast invalidation thread
Date: Wed, 31 Mar 2021 14:08:41 -0700 [thread overview]
Message-ID: <20210331210841.3996155-14-bgardon@google.com> (raw)
In-Reply-To: <20210331210841.3996155-1-bgardon@google.com>
To avoid saddling a vCPU thread with the work of tearing down an entire
paging structure, take a reference on each root before they become
obsolete, so that the thread initiating the fast invalidation can tear
down the paging structure and (most likely) release the last reference.
As a bonus, this teardown can happen under the MMU lock in read mode so
as not to block the progress of vCPU threads.
Signed-off-by: Ben Gardon <bgardon@google.com>
---
arch/x86/kvm/mmu/mmu.c | 6 ++++
arch/x86/kvm/mmu/tdp_mmu.c | 74 +++++++++++++++++++++++++++++++++++++-
arch/x86/kvm/mmu/tdp_mmu.h | 1 +
3 files changed, 80 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index 49b7097fb55b..22742619698d 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -5455,6 +5455,12 @@ static void kvm_mmu_zap_all_fast(struct kvm *kvm)
kvm_zap_obsolete_pages(kvm);
write_unlock(&kvm->mmu_lock);
+
+ if (is_tdp_mmu_enabled(kvm)) {
+ read_lock(&kvm->mmu_lock);
+ kvm_tdp_mmu_zap_all_fast(kvm);
+ read_unlock(&kvm->mmu_lock);
+ }
}
static bool kvm_has_zapped_obsolete_pages(struct kvm *kvm)
diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
index 428ff6778426..5498df7e2e1f 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.c
+++ b/arch/x86/kvm/mmu/tdp_mmu.c
@@ -794,13 +794,85 @@ void kvm_tdp_mmu_zap_all(struct kvm *kvm)
* kvm_reload_remote_mmus. Since this is in the same critical section, no new
* roots will be created between this function and the MMU reload signals
* being sent.
+ * Take a reference on all roots so that this thread can do the bulk of
+ * the work required to free the roots once they are invalidated. Without
+ * this reference, a vCPU thread might drop the last reference to a root
+ * and get stuck with tearing down the entire paging structure.
*/
void kvm_tdp_mmu_invalidate_roots(struct kvm *kvm)
{
struct kvm_mmu_page *root;
for_each_tdp_mmu_root(kvm, root)
- root->role.invalid = true;
+ if (refcount_inc_not_zero(&root->tdp_mmu_root_count))
+ root->role.invalid = true;
+}
+
+static struct kvm_mmu_page *next_invalidated_root(struct kvm *kvm,
+ struct kvm_mmu_page *prev_root)
+{
+ struct kvm_mmu_page *next_root;
+
+ if (prev_root)
+ next_root = list_next_or_null_rcu(&kvm->arch.tdp_mmu_roots,
+ &prev_root->link,
+ typeof(*prev_root), link);
+ else
+ next_root = list_first_or_null_rcu(&kvm->arch.tdp_mmu_roots,
+ typeof(*next_root), link);
+
+ while (next_root && !(next_root->role.invalid &&
+ refcount_read(&next_root->tdp_mmu_root_count)))
+ next_root = list_next_or_null_rcu(&kvm->arch.tdp_mmu_roots,
+ &next_root->link,
+ typeof(*next_root), link);
+
+ return next_root;
+}
+
+/*
+ * Since kvm_tdp_mmu_invalidate_roots has acquired a reference to each
+ * invalidated root, they will not be freed until this function drops the
+ * reference. Before dropping that reference, tear down the paging
+ * structure so that whichever thread does drop the last reference
+ * only has to do a trivial ammount of work. Since the roots are invalid,
+ * no new SPTEs should be created under them.
+ */
+void kvm_tdp_mmu_zap_all_fast(struct kvm *kvm)
+{
+ gfn_t max_gfn = 1ULL << (shadow_phys_bits - PAGE_SHIFT);
+ struct kvm_mmu_page *next_root;
+ struct kvm_mmu_page *root;
+ bool flush = false;
+
+ lockdep_assert_held_read(&kvm->mmu_lock);
+
+ rcu_read_lock();
+
+ root = next_invalidated_root(kvm, NULL);
+
+ while (root) {
+ next_root = next_invalidated_root(kvm, root);
+
+ rcu_read_unlock();
+
+ flush |= zap_gfn_range(kvm, root, 0, max_gfn, true, true);
+
+ /*
+ * Put the reference acquired in
+ * kvm_tdp_mmu_invalidate_roots
+ */
+ kvm_tdp_mmu_put_root(kvm, root, true);
+
+ root = next_root;
+
+ rcu_read_lock();
+ }
+
+ rcu_read_unlock();
+
+ if (flush)
+ kvm_flush_remote_tlbs(kvm);
}
/*
diff --git a/arch/x86/kvm/mmu/tdp_mmu.h b/arch/x86/kvm/mmu/tdp_mmu.h
index ff4978817fb8..d6d98f9047cd 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.h
+++ b/arch/x86/kvm/mmu/tdp_mmu.h
@@ -24,6 +24,7 @@ bool kvm_tdp_mmu_zap_gfn_range(struct kvm *kvm, gfn_t start, gfn_t end,
void kvm_tdp_mmu_zap_all(struct kvm *kvm);
void kvm_tdp_mmu_invalidate_roots(struct kvm *kvm);
+void kvm_tdp_mmu_zap_all_fast(struct kvm *kvm);
int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code,
int map_writable, int max_level, kvm_pfn_t pfn,
--
2.31.0.291.g576ba9dcdaf-goog
next prev parent reply other threads:[~2021-03-31 21:10 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-31 21:08 [PATCH 00/13] More parallel operations for the TDP MMU Ben Gardon
2021-03-31 21:08 ` [PATCH 01/13] KVM: x86/mmu: Re-add const qualifier in kvm_tdp_mmu_zap_collapsible_sptes Ben Gardon
2021-03-31 21:08 ` [PATCH 02/13] KVM: x86/mmu: Move kvm_mmu_(get|put)_root to TDP MMU Ben Gardon
2021-03-31 21:08 ` [PATCH 03/13] KVM: x86/mmu: use tdp_mmu_free_sp to free roots Ben Gardon
2021-03-31 21:08 ` [PATCH 04/13] KVM: x86/mmu: Merge TDP MMU put and free root Ben Gardon
2021-03-31 21:08 ` [PATCH 05/13] KVM: x86/mmu: comment for_each_tdp_mmu_root requires MMU write lock Ben Gardon
2021-03-31 21:08 ` [PATCH 06/13] KVM: x86/mmu: Refactor yield safe root iterator Ben Gardon
2021-03-31 21:08 ` [PATCH 07/13] KVM: x86/mmu: Make TDP MMU root refcount atomic Ben Gardon
2021-03-31 22:21 ` Sean Christopherson
2021-04-01 16:50 ` Ben Gardon
2021-03-31 21:08 ` [PATCH 08/13] KVM: x86/mmu: Protect the tdp_mmu_roots list with RCU Ben Gardon
2021-04-01 9:37 ` Paolo Bonzini
2021-04-01 16:48 ` Ben Gardon
2021-04-01 13:16 ` kernel test robot
2021-04-01 16:50 ` Ben Gardon
2021-03-31 21:08 ` [PATCH 09/13] KVM: x86/mmu: Allow zap gfn range to operate under the mmu read lock Ben Gardon
2021-04-01 9:58 ` Paolo Bonzini
2021-04-01 16:50 ` Ben Gardon
2021-03-31 21:08 ` [PATCH 10/13] KVM: x86/mmu: Allow zapping collapsible SPTEs to use MMU " Ben Gardon
2021-04-01 10:12 ` Paolo Bonzini
2021-03-31 21:08 ` [PATCH 11/13] KVM: x86/mmu: Allow enabling / disabling dirty logging under " Ben Gardon
2021-03-31 21:08 ` [PATCH 12/13] KVM: x86/mmu: Fast invalidation for TDP MMU Ben Gardon
2021-03-31 22:27 ` Sean Christopherson
2021-04-01 16:50 ` Ben Gardon
2021-04-01 10:36 ` Paolo Bonzini
2021-04-01 16:50 ` Ben Gardon
2021-04-01 17:02 ` Paolo Bonzini
2021-03-31 21:08 ` Ben Gardon [this message]
2021-03-31 22:29 ` [PATCH 13/13] KVM: x86/mmu: Tear down roots in fast invalidation thread Sean Christopherson
2021-04-01 10:41 ` Paolo Bonzini
2021-04-01 10:37 ` Paolo Bonzini
2021-04-01 10:43 ` [PATCH 00/13] More parallel operations for the TDP MMU 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=20210331210841.3996155-14-bgardon@google.com \
--to=bgardon@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=pbonzini@redhat.com \
--cc=peterx@redhat.com \
--cc=pfeiner@google.com \
--cc=pshier@google.com \
--cc=seanjc@google.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;
as well as URLs for NNTP newsgroup(s).