From: David Matlack <dmatlack@google.com> To: Paolo Bonzini <pbonzini@redhat.com> Cc: Marc Zyngier <maz@kernel.org>, Huacai Chen <chenhuacai@kernel.org>, Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>, Anup Patel <anup@brainfault.org>, Paul Walmsley <paul.walmsley@sifive.com>, Palmer Dabbelt <palmer@dabbelt.com>, Albert Ou <aou@eecs.berkeley.edu>, Sean Christopherson <seanjc@google.com>, Andrew Jones <drjones@redhat.com>, Ben Gardon <bgardon@google.com>, Peter Xu <peterx@redhat.com>, maciej.szmigiero@oracle.com, "moderated list:KERNEL VIRTUAL MACHINE FOR ARM64 (KVM/arm64)" <kvmarm@lists.cs.columbia.edu>, "open list:KERNEL VIRTUAL MACHINE FOR MIPS (KVM/mips)" <linux-mips@vger.kernel.org>, "open list:KERNEL VIRTUAL MACHINE FOR MIPS (KVM/mips)" <kvm@vger.kernel.org>, "open list:KERNEL VIRTUAL MACHINE FOR RISC-V (KVM/riscv)" <kvm-riscv@lists.infradead.org>, Peter Feiner <pfeiner@google.com>, David Matlack <dmatlack@google.com> Subject: [PATCH v2 01/26] KVM: x86/mmu: Optimize MMU page cache lookup for all direct SPs Date: Fri, 11 Mar 2022 00:25:03 +0000 [thread overview] Message-ID: <20220311002528.2230172-2-dmatlack@google.com> (raw) In-Reply-To: <20220311002528.2230172-1-dmatlack@google.com> Commit fb58a9c345f6 ("KVM: x86/mmu: Optimize MMU page cache lookup for fully direct MMUs") skipped the unsync checks and write flood clearing for full direct MMUs. We can extend this further and skip the checks for all direct shadow pages. Direct shadow pages are never marked unsynced or have a non-zero write-flooding count. Checking sp->role.direct alos generates better code than checking direct_map because, due to register pressure, direct_map has to get shoved onto the stack and then pulled back off. No functional change intended. Reviewed-by: Sean Christopherson <seanjc@google.com> Signed-off-by: David Matlack <dmatlack@google.com> --- arch/x86/kvm/mmu/mmu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 3b8da8b0745e..3ad67f70e51c 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -2034,7 +2034,6 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, int direct, unsigned int access) { - bool direct_mmu = vcpu->arch.mmu->direct_map; union kvm_mmu_page_role role; struct hlist_head *sp_list; unsigned quadrant; @@ -2075,7 +2074,8 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, continue; } - if (direct_mmu) + /* unsync and write-flooding only apply to indirect SPs. */ + if (sp->role.direct) goto trace_get_page; if (sp->unsync) { base-commit: ce41d078aaa9cf15cbbb4a42878cc6160d76525e -- 2.35.1.723.g4982287a31-goog
WARNING: multiple messages have this Message-ID (diff)
From: David Matlack <dmatlack@google.com> To: Paolo Bonzini <pbonzini@redhat.com> Cc: Albert Ou <aou@eecs.berkeley.edu>, "open list:KERNEL VIRTUAL MACHINE FOR MIPS \(KVM/mips\)" <kvm@vger.kernel.org>, Marc Zyngier <maz@kernel.org>, Huacai Chen <chenhuacai@kernel.org>, "open list:KERNEL VIRTUAL MACHINE FOR MIPS \(KVM/mips\)" <linux-mips@vger.kernel.org>, David Matlack <dmatlack@google.com>, Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>, Palmer Dabbelt <palmer@dabbelt.com>, "open list:KERNEL VIRTUAL MACHINE FOR RISC-V \(KVM/riscv\)" <kvm-riscv@lists.infradead.org>, Paul Walmsley <paul.walmsley@sifive.com>, Ben Gardon <bgardon@google.com>, maciej.szmigiero@oracle.com, "moderated list:KERNEL VIRTUAL MACHINE FOR ARM64 \(KVM/arm64\)" <kvmarm@lists.cs.columbia.edu>, Peter Feiner <pfeiner@google.com> Subject: [PATCH v2 01/26] KVM: x86/mmu: Optimize MMU page cache lookup for all direct SPs Date: Fri, 11 Mar 2022 00:25:03 +0000 [thread overview] Message-ID: <20220311002528.2230172-2-dmatlack@google.com> (raw) In-Reply-To: <20220311002528.2230172-1-dmatlack@google.com> Commit fb58a9c345f6 ("KVM: x86/mmu: Optimize MMU page cache lookup for fully direct MMUs") skipped the unsync checks and write flood clearing for full direct MMUs. We can extend this further and skip the checks for all direct shadow pages. Direct shadow pages are never marked unsynced or have a non-zero write-flooding count. Checking sp->role.direct alos generates better code than checking direct_map because, due to register pressure, direct_map has to get shoved onto the stack and then pulled back off. No functional change intended. Reviewed-by: Sean Christopherson <seanjc@google.com> Signed-off-by: David Matlack <dmatlack@google.com> --- arch/x86/kvm/mmu/mmu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 3b8da8b0745e..3ad67f70e51c 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -2034,7 +2034,6 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, int direct, unsigned int access) { - bool direct_mmu = vcpu->arch.mmu->direct_map; union kvm_mmu_page_role role; struct hlist_head *sp_list; unsigned quadrant; @@ -2075,7 +2074,8 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, continue; } - if (direct_mmu) + /* unsync and write-flooding only apply to indirect SPs. */ + if (sp->role.direct) goto trace_get_page; if (sp->unsync) { base-commit: ce41d078aaa9cf15cbbb4a42878cc6160d76525e -- 2.35.1.723.g4982287a31-goog _______________________________________________ kvmarm mailing list kvmarm@lists.cs.columbia.edu https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
next prev parent reply other threads:[~2022-03-11 0:25 UTC|newest] Thread overview: 134+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-03-11 0:25 [PATCH v2 00/26] Extend Eager Page Splitting to the shadow MMU David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-11 0:25 ` David Matlack [this message] 2022-03-11 0:25 ` [PATCH v2 01/26] KVM: x86/mmu: Optimize MMU page cache lookup for all direct SPs David Matlack 2022-03-15 7:40 ` Peter Xu 2022-03-15 7:40 ` Peter Xu 2022-03-22 18:16 ` David Matlack 2022-03-22 18:16 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 02/26] KVM: x86/mmu: Use a bool for direct David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 7:46 ` Peter Xu 2022-03-15 7:46 ` Peter Xu 2022-03-22 18:21 ` David Matlack 2022-03-22 18:21 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 03/26] KVM: x86/mmu: Derive shadow MMU page role from parent David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 8:15 ` Peter Xu 2022-03-15 8:15 ` Peter Xu 2022-03-22 18:30 ` David Matlack 2022-03-22 18:30 ` David Matlack 2022-03-30 14:25 ` Peter Xu 2022-03-30 14:25 ` Peter Xu 2022-03-11 0:25 ` [PATCH v2 04/26] KVM: x86/mmu: Decompose kvm_mmu_get_page() into separate functions David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 8:50 ` Peter Xu 2022-03-15 8:50 ` Peter Xu 2022-03-22 22:09 ` David Matlack 2022-03-22 22:09 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 05/26] KVM: x86/mmu: Rename shadow MMU functions that deal with shadow pages David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 8:52 ` Peter Xu 2022-03-15 8:52 ` Peter Xu 2022-03-22 21:35 ` David Matlack 2022-03-22 21:35 ` David Matlack 2022-03-30 14:28 ` Peter Xu 2022-03-30 14:28 ` Peter Xu 2022-03-11 0:25 ` [PATCH v2 06/26] KVM: x86/mmu: Pass memslot to kvm_mmu_new_shadow_page() David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 9:03 ` Peter Xu 2022-03-15 9:03 ` Peter Xu 2022-03-22 22:05 ` David Matlack 2022-03-22 22:05 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 07/26] KVM: x86/mmu: Separate shadow MMU sp allocation from initialization David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 9:54 ` Peter Xu 2022-03-15 9:54 ` Peter Xu 2022-03-11 0:25 ` [PATCH v2 08/26] KVM: x86/mmu: Link spt to sp during allocation David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 10:04 ` Peter Xu 2022-03-15 10:04 ` Peter Xu 2022-03-22 22:30 ` David Matlack 2022-03-22 22:30 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 09/26] KVM: x86/mmu: Move huge page split sp allocation code to mmu.c David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 10:17 ` Peter Xu 2022-03-15 10:17 ` Peter Xu 2022-03-11 0:25 ` [PATCH v2 10/26] KVM: x86/mmu: Use common code to free kvm_mmu_page structs David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 10:22 ` Peter Xu 2022-03-15 10:22 ` Peter Xu 2022-03-22 22:33 ` David Matlack 2022-03-22 22:33 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 11/26] KVM: x86/mmu: Use common code to allocate kvm_mmu_page structs from vCPU caches David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 10:27 ` Peter Xu 2022-03-15 10:27 ` Peter Xu 2022-03-22 22:35 ` David Matlack 2022-03-22 22:35 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 12/26] KVM: x86/mmu: Pass const memslot to rmap_add() David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 13/26] KVM: x86/mmu: Pass const memslot to init_shadow_page() and descendants David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 14/26] KVM: x86/mmu: Decouple rmap_add() and link_shadow_page() from kvm_vcpu David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 10:37 ` Peter Xu 2022-03-15 10:37 ` Peter Xu 2022-03-11 0:25 ` [PATCH v2 15/26] KVM: x86/mmu: Update page stats in __rmap_add() David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-15 10:39 ` Peter Xu 2022-03-15 10:39 ` Peter Xu 2022-03-11 0:25 ` [PATCH v2 16/26] KVM: x86/mmu: Cache the access bits of shadowed translations David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-16 8:32 ` Peter Xu 2022-03-16 8:32 ` Peter Xu 2022-03-22 22:51 ` David Matlack 2022-03-22 22:51 ` David Matlack 2022-03-30 18:30 ` Peter Xu 2022-03-30 18:30 ` Peter Xu 2022-03-31 21:40 ` David Matlack 2022-03-31 21:40 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 17/26] KVM: x86/mmu: Pass access information to make_huge_page_split_spte() David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-16 8:44 ` Peter Xu 2022-03-16 8:44 ` Peter Xu 2022-03-22 23:08 ` David Matlack 2022-03-22 23:08 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 18/26] KVM: x86/mmu: Zap collapsible SPTEs at all levels in the shadow MMU David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-16 8:49 ` Peter Xu 2022-03-16 8:49 ` Peter Xu 2022-03-22 23:11 ` David Matlack 2022-03-22 23:11 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 19/26] KVM: x86/mmu: Refactor drop_large_spte() David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-16 8:53 ` Peter Xu 2022-03-16 8:53 ` Peter Xu 2022-03-11 0:25 ` [PATCH v2 20/26] KVM: x86/mmu: Extend Eager Page Splitting to the shadow MMU David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-16 10:26 ` Peter Xu 2022-03-16 10:26 ` Peter Xu 2022-03-22 0:07 ` David Matlack 2022-03-22 0:07 ` David Matlack 2022-03-22 23:58 ` David Matlack 2022-03-22 23:58 ` David Matlack 2022-03-30 18:34 ` Peter Xu 2022-03-30 18:34 ` Peter Xu 2022-03-31 19:57 ` David Matlack 2022-03-31 19:57 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 21/26] KVM: Allow for different capacities in kvm_mmu_memory_cache structs David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-19 5:27 ` Anup Patel 2022-03-19 5:27 ` Anup Patel 2022-03-22 23:13 ` David Matlack 2022-03-22 23:13 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 22/26] KVM: Allow GFP flags to be passed when topping up MMU caches David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 23/26] KVM: x86/mmu: Fully split huge pages that require extra pte_list_desc structs David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 24/26] KVM: x86/mmu: Split huge pages aliased by multiple SPTEs David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 25/26] KVM: x86/mmu: Drop NULL pte_list_desc_cache fallback David Matlack 2022-03-11 0:25 ` David Matlack 2022-03-11 0:25 ` [PATCH v2 26/26] KVM: selftests: Map x86_64 guest virtual memory with huge pages David Matlack 2022-03-11 0:25 ` 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=20220311002528.2230172-2-dmatlack@google.com \ --to=dmatlack@google.com \ --cc=aleksandar.qemu.devel@gmail.com \ --cc=anup@brainfault.org \ --cc=aou@eecs.berkeley.edu \ --cc=bgardon@google.com \ --cc=chenhuacai@kernel.org \ --cc=drjones@redhat.com \ --cc=kvm-riscv@lists.infradead.org \ --cc=kvm@vger.kernel.org \ --cc=kvmarm@lists.cs.columbia.edu \ --cc=linux-mips@vger.kernel.org \ --cc=maciej.szmigiero@oracle.com \ --cc=maz@kernel.org \ --cc=palmer@dabbelt.com \ --cc=paul.walmsley@sifive.com \ --cc=pbonzini@redhat.com \ --cc=peterx@redhat.com \ --cc=pfeiner@google.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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.