public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
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,
	David Matlack <dmatlack@google.com>,
	Isaku Yamahata <isaku.yamahata@intel.com>
Subject: [PATCH v4 10/11] KVM: x86/mmu: Use static key/branches for checking if TDP MMU is enabled
Date: Wed, 12 Oct 2022 18:17:01 +0000	[thread overview]
Message-ID: <20221012181702.3663607-11-seanjc@google.com> (raw)
In-Reply-To: <20221012181702.3663607-1-seanjc@google.com>

Now that the TDP MMU being enabled is read-only after the vendor module
is loaded, use a static key to track whether or not the TDP MMU is
enabled to avoid conditional branches in hot paths, e.g. in
direct_page_fault() and fast_page_fault().

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 arch/x86/kvm/mmu.h     |  5 +++--
 arch/x86/kvm/mmu/mmu.c | 14 ++++++++++----
 2 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h
index 1ad6d02e103f..bc0d8a5c09f9 100644
--- a/arch/x86/kvm/mmu.h
+++ b/arch/x86/kvm/mmu.h
@@ -2,6 +2,7 @@
 #ifndef __KVM_X86_MMU_H
 #define __KVM_X86_MMU_H
 
+#include <linux/jump_label.h>
 #include <linux/kvm_host.h>
 #include "kvm_cache_regs.h"
 #include "cpuid.h"
@@ -230,13 +231,13 @@ static inline bool kvm_shadow_root_allocated(struct kvm *kvm)
 }
 
 #ifdef CONFIG_X86_64
-extern bool tdp_mmu_enabled;
+DECLARE_STATIC_KEY_TRUE(tdp_mmu_enabled);
 #endif
 
 static inline bool is_tdp_mmu_enabled(void)
 {
 #ifdef CONFIG_X86_64
-	return tdp_mmu_enabled;
+	return static_branch_likely(&tdp_mmu_enabled);
 #else
 	return false;
 #endif
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index 4792d76edd6d..a5ba7b41263d 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -101,8 +101,10 @@ bool tdp_enabled = false;
 #ifdef CONFIG_X86_64
 static bool __ro_after_init tdp_mmu_allowed;
 
-bool __read_mostly tdp_mmu_enabled = true;
-module_param_named(tdp_mmu, tdp_mmu_enabled, bool, 0444);
+static bool __read_mostly __tdp_mmu_enabled = true;
+module_param_named(tdp_mmu, __tdp_mmu_enabled, bool, 0444);
+
+DEFINE_STATIC_KEY_TRUE(tdp_mmu_enabled);
 #endif
 
 static int max_huge_page_level __read_mostly;
@@ -5702,7 +5704,11 @@ void kvm_configure_mmu(bool enable_tdp, int tdp_forced_root_level,
 	max_tdp_level = tdp_max_root_level;
 
 #ifdef CONFIG_X86_64
-	tdp_mmu_enabled = tdp_mmu_allowed && tdp_enabled;
+	__tdp_mmu_enabled = tdp_mmu_allowed && tdp_enabled;
+	if (__tdp_mmu_enabled)
+		static_branch_enable(&tdp_mmu_enabled);
+	else
+		static_branch_disable(&tdp_mmu_enabled);
 #endif
 	/*
 	 * max_huge_page_level reflects KVM's MMU capabilities irrespective
@@ -6712,7 +6718,7 @@ void __init kvm_mmu_x86_module_init(void)
 	 * TDP MMU is actually enabled is determined in kvm_configure_mmu()
 	 * when the vendor module is loaded.
 	 */
-	tdp_mmu_allowed = tdp_mmu_enabled;
+	tdp_mmu_allowed = __tdp_mmu_enabled;
 #endif
 
 	kvm_mmu_spte_module_init();
-- 
2.38.0.rc1.362.ged0d419d3c-goog


  parent reply	other threads:[~2022-10-12 18:17 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-12 18:16 [PATCH v4 00/11] KVM: x86/mmu: Make tdp_mmu a read-only parameter Sean Christopherson
2022-10-12 18:16 ` [PATCH v4 01/11] KVM: x86/mmu: Change tdp_mmu to " Sean Christopherson
2022-10-12 21:37   ` Huang, Kai
2022-10-12 18:16 ` [PATCH v4 02/11] KVM: x86/mmu: Move TDP MMU VM init/uninit behind tdp_mmu_enabled Sean Christopherson
2022-10-12 18:16 ` [PATCH v4 03/11] KVM: x86/mmu: Grab mmu_invalidate_seq in kvm_faultin_pfn() Sean Christopherson
2022-10-12 18:16 ` [PATCH v4 04/11] KVM: x86/mmu: Handle error PFNs " Sean Christopherson
2022-10-12 18:16 ` [PATCH v4 05/11] KVM: x86/mmu: Avoid memslot lookup during KVM_PFN_ERR_HWPOISON handling Sean Christopherson
2022-10-12 18:16 ` [PATCH v4 06/11] KVM: x86/mmu: Handle no-slot faults in kvm_faultin_pfn() Sean Christopherson
2022-10-12 18:16 ` [PATCH v4 07/11] KVM: x86/mmu: Pivot on "TDP MMU enabled" when handling direct page faults Sean Christopherson
2022-10-12 18:16 ` [PATCH v4 08/11] KVM: x86/mmu: Pivot on "TDP MMU enabled" to check if active MMU is TDP MMU Sean Christopherson
2022-10-12 18:17 ` [PATCH v4 09/11] KVM: x86/mmu: Replace open coded usage of tdp_mmu_page with is_tdp_mmu_page() Sean Christopherson
2022-10-12 18:17 ` Sean Christopherson [this message]
2022-10-12 18:17 ` [PATCH v4 11/11] KVM: x86/mmu: Stop needlessly making MMU pages available for TDP MMU Sean Christopherson
2022-10-13 17:21 ` [PATCH v4 00/11] KVM: x86/mmu: Make tdp_mmu a read-only parameter David Matlack
2022-10-13 20:12   ` Sean Christopherson
2022-10-13 22:56     ` 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=20221012181702.3663607-11-seanjc@google.com \
    --to=seanjc@google.com \
    --cc=dmatlack@google.com \
    --cc=isaku.yamahata@intel.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.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