linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Vitaly Kuznetsov <vkuznets@redhat.com>
To: kvm@vger.kernel.org, Paolo Bonzini <pbonzini@redhat.com>,
	Sean Christopherson <seanjc@google.com>
Cc: Wanpeng Li <wanpengli@tencent.com>,
	Jim Mattson <jmattson@google.com>,
	Michael Kelley <mikelley@microsoft.com>,
	Siddharth Chandrasekaran <sidcha@amazon.de>,
	Yuan Yao <yuan.yao@linux.intel.com>,
	Maxim Levitsky <mlevitsk@redhat.com>,
	linux-hyperv@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH v13 16/48] KVM: x86: hyper-v: Don't use sparse_set_to_vcpu_mask() in kvm_hv_send_ipi()
Date: Tue,  1 Nov 2022 15:53:54 +0100	[thread overview]
Message-ID: <20221101145426.251680-17-vkuznets@redhat.com> (raw)
In-Reply-To: <20221101145426.251680-1-vkuznets@redhat.com>

Get rid of on-stack allocation of vcpu_mask and optimize kvm_hv_send_ipi()
for a smaller number of vCPUs in the request. When Hyper-V TLB flush
is in  use, HvSendSyntheticClusterIpi{,Ex} calls are not commonly used to
send IPIs to a large number of vCPUs (and are rarely used in general).

Introduce hv_is_vp_in_sparse_set() to directly check if the specified
VP_ID is present in sparse vCPU set.

Reviewed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
---
 arch/x86/kvm/hyperv.c | 42 +++++++++++++++++++++++++++++++-----------
 1 file changed, 31 insertions(+), 11 deletions(-)

diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
index 2fceb85687c1..0bfa59838e0a 100644
--- a/arch/x86/kvm/hyperv.c
+++ b/arch/x86/kvm/hyperv.c
@@ -1741,6 +1741,28 @@ static void sparse_set_to_vcpu_mask(struct kvm *kvm, u64 *sparse_banks,
 	}
 }
 
+static bool hv_is_vp_in_sparse_set(u32 vp_id, u64 valid_bank_mask, u64 sparse_banks[])
+{
+	int valid_bit_nr = vp_id / HV_VCPUS_PER_SPARSE_BANK;
+	unsigned long sbank;
+
+	if (!test_bit(valid_bit_nr, (unsigned long *)&valid_bank_mask))
+		return false;
+
+	/*
+	 * The index into the sparse bank is the number of preceding bits in
+	 * the valid mask.  Optimize for VMs with <64 vCPUs by skipping the
+	 * fancy math if there can't possibly be preceding bits.
+	 */
+	if (valid_bit_nr)
+		sbank = hweight64(valid_bank_mask & GENMASK_ULL(valid_bit_nr - 1, 0));
+	else
+		sbank = 0;
+
+	return test_bit(vp_id % HV_VCPUS_PER_SPARSE_BANK,
+			(unsigned long *)&sparse_banks[sbank]);
+}
+
 struct kvm_hv_hcall {
 	u64 param;
 	u64 ingpa;
@@ -2035,8 +2057,8 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
 		((u64)hc->rep_cnt << HV_HYPERCALL_REP_COMP_OFFSET);
 }
 
-static void kvm_send_ipi_to_many(struct kvm *kvm, u32 vector,
-				 unsigned long *vcpu_bitmap)
+static void kvm_hv_send_ipi_to_many(struct kvm *kvm, u32 vector,
+				    u64 *sparse_banks, u64 valid_bank_mask)
 {
 	struct kvm_lapic_irq irq = {
 		.delivery_mode = APIC_DM_FIXED,
@@ -2046,7 +2068,9 @@ static void kvm_send_ipi_to_many(struct kvm *kvm, u32 vector,
 	unsigned long i;
 
 	kvm_for_each_vcpu(i, vcpu, kvm) {
-		if (vcpu_bitmap && !test_bit(i, vcpu_bitmap))
+		if (sparse_banks &&
+		    !hv_is_vp_in_sparse_set(kvm_hv_get_vpindex(vcpu),
+					    valid_bank_mask, sparse_banks))
 			continue;
 
 		/* We fail only when APIC is disabled */
@@ -2059,7 +2083,6 @@ static u64 kvm_hv_send_ipi(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
 	struct kvm *kvm = vcpu->kvm;
 	struct hv_send_ipi_ex send_ipi_ex;
 	struct hv_send_ipi send_ipi;
-	DECLARE_BITMAP(vcpu_mask, KVM_MAX_VCPUS);
 	u64 valid_bank_mask;
 	u64 sparse_banks[KVM_HV_MAX_SPARSE_VCPU_SET_BITS];
 	u32 vector;
@@ -2121,13 +2144,10 @@ static u64 kvm_hv_send_ipi(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
 	if ((vector < HV_IPI_LOW_VECTOR) || (vector > HV_IPI_HIGH_VECTOR))
 		return HV_STATUS_INVALID_HYPERCALL_INPUT;
 
-	if (all_cpus) {
-		kvm_send_ipi_to_many(kvm, vector, NULL);
-	} else {
-		sparse_set_to_vcpu_mask(kvm, sparse_banks, valid_bank_mask, vcpu_mask);
-
-		kvm_send_ipi_to_many(kvm, vector, vcpu_mask);
-	}
+	if (all_cpus)
+		kvm_hv_send_ipi_to_many(kvm, vector, NULL, 0);
+	else
+		kvm_hv_send_ipi_to_many(kvm, vector, sparse_banks, valid_bank_mask);
 
 ret_success:
 	return HV_STATUS_SUCCESS;
-- 
2.37.3


  parent reply	other threads:[~2022-11-01 14:57 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-01 14:53 [PATCH v13 00/48] KVM: x86: hyper-v: Fine-grained TLB flush + L2 TLB flush features Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 01/48] x86/hyperv: Move VMCB enlightenment definitions to hyperv-tlfs.h Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 02/48] KVM: selftests: Move "struct hv_enlightenments" to x86_64/svm.h Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 03/48] KVM: SVM: Add a proper field for Hyper-V VMCB enlightenments Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 04/48] x86/hyperv: KVM: Rename "hv_enlightenments" to "hv_vmcb_enlightenments" Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 05/48] KVM: x86: Rename 'enable_direct_tlbflush' to 'enable_l2_tlb_flush' Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 06/48] KVM: VMX: Rename "vmx/evmcs.{ch}" to "vmx/hyperv.{ch}" Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 07/48] KVM: x86: Move clearing of TLB_FLUSH_CURRENT to kvm_vcpu_flush_tlb_all() Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 08/48] KVM: x86: hyper-v: Resurrect dedicated KVM_REQ_HV_TLB_FLUSH flag Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 09/48] KVM: x86: hyper-v: Introduce TLB flush fifo Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 10/48] KVM: x86: hyper-v: Add helper to read hypercall data for array Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 11/48] KVM: x86: hyper-v: Handle HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST{,EX} calls gently Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 12/48] KVM: x86: hyper-v: Expose support for extended gva ranges for flush hypercalls Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 13/48] KVM: x86: Prepare kvm_hv_flush_tlb() to handle L2's GPAs Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 14/48] x86/hyperv: Introduce HV_MAX_SPARSE_VCPU_BANKS/HV_VCPUS_PER_SPARSE_BANK constants Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 15/48] KVM: x86: hyper-v: Use HV_MAX_SPARSE_VCPU_BANKS/HV_VCPUS_PER_SPARSE_BANK instead of raw '64' Vitaly Kuznetsov
2022-11-01 14:53 ` Vitaly Kuznetsov [this message]
2022-11-01 14:53 ` [PATCH v13 17/48] KVM: x86: hyper-v: Create a separate fifo for L2 TLB flush Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 18/48] KVM: x86: hyper-v: Use preallocated buffer in 'struct kvm_vcpu_hv' instead of on-stack 'sparse_banks' Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 19/48] KVM: nVMX: Keep track of hv_vm_id/hv_vp_id when eVMCS is in use Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 20/48] KVM: nSVM: Keep track of Hyper-V hv_vm_id/hv_vp_id Vitaly Kuznetsov
2022-11-01 14:53 ` [PATCH v13 21/48] KVM: x86: Introduce .hv_inject_synthetic_vmexit_post_tlb_flush() nested hook Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 22/48] KVM: x86: hyper-v: Introduce kvm_hv_is_tlb_flush_hcall() Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 23/48] KVM: x86: hyper-v: L2 TLB flush Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 24/48] KVM: x86: hyper-v: Introduce fast guest_hv_cpuid_has_l2_tlb_flush() check Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 25/48] KVM: nVMX: hyper-v: Cache VP assist page in 'struct kvm_vcpu_hv' Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 26/48] KVM: nVMX: hyper-v: Enable L2 TLB flush Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 27/48] KVM: x86: Make kvm_hv_get_assist_page() return 0/-errno Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 28/48] KVM: nSVM: hyper-v: Enable L2 TLB flush Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 29/48] KVM: x86: Expose Hyper-V L2 TLB flush feature Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 30/48] KVM: selftests: Better XMM read/write helpers Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 31/48] KVM: selftests: Move HYPERV_LINUX_OS_ID definition to a common header Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 32/48] KVM: selftests: Move the function doing Hyper-V hypercall " Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 33/48] KVM: selftests: Hyper-V PV IPI selftest Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 34/48] KVM: selftests: Fill in vm->vpages_mapped bitmap in virt_map() too Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 35/48] KVM: selftests: Export vm_vaddr_unused_gap() to make it possible to request unmapped ranges Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 36/48] KVM: selftests: Drop helpers to read/write page table entries Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 37/48] KVM: selftests: Hyper-V PV TLB flush selftest Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 38/48] KVM: selftests: Sync 'struct hv_enlightened_vmcs' definition with hyperv-tlfs.h Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 39/48] KVM: selftests: Sync 'struct hv_vp_assist_page' " Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 40/48] KVM: selftests: Move Hyper-V VP assist page enablement out of evmcs.h Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 41/48] KVM: selftests: Split off load_evmcs() from load_vmcs() Vitaly Kuznetsov
2022-11-01 16:13   ` Sean Christopherson
2022-11-01 14:54 ` [PATCH v13 42/48] KVM: selftests: Create a vendor independent helper to allocate Hyper-V specific test pages Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 43/48] KVM: selftests: Allocate Hyper-V partition assist page Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 44/48] KVM: selftests: Stuff RAX/RCX with 'safe' values in vmmcall()/vmcall() Vitaly Kuznetsov
2022-11-01 15:21   ` Sean Christopherson
2022-11-01 14:54 ` [PATCH v13 45/48] KVM: selftests: Introduce rdmsr_from_l2() and use it for MSR-Bitmap tests Vitaly Kuznetsov
2022-11-01 16:11   ` Sean Christopherson
2022-11-01 16:27     ` Vitaly Kuznetsov
2022-11-01 17:26       ` Sean Christopherson
2022-11-01 14:54 ` [PATCH v13 46/48] KVM: selftests: evmcs_test: Introduce L2 TLB flush test Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 47/48] KVM: selftests: hyperv_svm_test: " Vitaly Kuznetsov
2022-11-01 14:54 ` [PATCH v13 48/48] KVM: selftests: Rename 'evmcs_test' to 'hyperv_evmcs' Vitaly Kuznetsov
2022-11-01 15:21 ` [PATCH v13 00/48] KVM: x86: hyper-v: Fine-grained TLB flush + L2 TLB flush features Sean Christopherson
2022-11-01 16:29   ` Vitaly Kuznetsov
2022-11-18 18:16     ` Paolo Bonzini
2022-11-21  9:20       ` Vitaly Kuznetsov

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=20221101145426.251680-17-vkuznets@redhat.com \
    --to=vkuznets@redhat.com \
    --cc=jmattson@google.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-hyperv@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mikelley@microsoft.com \
    --cc=mlevitsk@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=seanjc@google.com \
    --cc=sidcha@amazon.de \
    --cc=wanpengli@tencent.com \
    --cc=yuan.yao@linux.intel.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).