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 09/48] KVM: x86: hyper-v: Introduce TLB flush fifo
Date: Tue, 1 Nov 2022 15:53:47 +0100 [thread overview]
Message-ID: <20221101145426.251680-10-vkuznets@redhat.com> (raw)
In-Reply-To: <20221101145426.251680-1-vkuznets@redhat.com>
To allow flushing individual GVAs instead of always flushing the whole
VPID a per-vCPU structure to pass the requests is needed. Use standard
'kfifo' to queue two types of entries: individual GVA (GFN + up to 4095
following GFNs in the lower 12 bits) and 'flush all'.
The size of the fifo is arbitrarily set to '16'.
Note, kvm_hv_flush_tlb() only queues 'flush all' entries for now and
kvm_hv_vcpu_flush_tlb() doesn't actually read the fifo just resets the
queue before returning -EOPNOTSUPP (which triggers full TLB flush) so
the functional change is very small but the infrastructure is prepared
to handle individual GVA flush requests.
Reviewed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
---
arch/x86/include/asm/kvm_host.h | 20 ++++++++++++++
arch/x86/kvm/hyperv.c | 47 +++++++++++++++++++++++++++++++++
arch/x86/kvm/hyperv.h | 15 +++++++++++
arch/x86/kvm/svm/svm.c | 2 +-
arch/x86/kvm/x86.c | 11 ++++++--
5 files changed, 92 insertions(+), 3 deletions(-)
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 05d90e11e535..f8ee72dd8660 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -25,6 +25,7 @@
#include <linux/clocksource.h>
#include <linux/irqbypass.h>
#include <linux/hyperv.h>
+#include <linux/kfifo.h>
#include <asm/apic.h>
#include <asm/pvclock-abi.h>
@@ -599,6 +600,23 @@ struct kvm_vcpu_hv_synic {
bool dont_zero_synic_pages;
};
+/* The maximum number of entries on the TLB flush fifo. */
+#define KVM_HV_TLB_FLUSH_FIFO_SIZE (16)
+/*
+ * Note: the following 'magic' entry is made up by KVM to avoid putting
+ * anything besides GVA on the TLB flush fifo. It is theoretically possible
+ * to observe a request to flush 4095 PFNs starting from 0xfffffffffffff000
+ * which will look identical. KVM's action to 'flush everything' instead of
+ * flushing these particular addresses is, however, fully legitimate as
+ * flushing more than requested is always OK.
+ */
+#define KVM_HV_TLB_FLUSHALL_ENTRY ((u64)-1)
+
+struct kvm_vcpu_hv_tlb_flush_fifo {
+ spinlock_t write_lock;
+ DECLARE_KFIFO(entries, u64, KVM_HV_TLB_FLUSH_FIFO_SIZE);
+};
+
/* Hyper-V per vcpu emulation context */
struct kvm_vcpu_hv {
struct kvm_vcpu *vcpu;
@@ -620,6 +638,8 @@ struct kvm_vcpu_hv {
u32 nested_eax; /* HYPERV_CPUID_NESTED_FEATURES.EAX */
u32 nested_ebx; /* HYPERV_CPUID_NESTED_FEATURES.EBX */
} cpuid_cache;
+
+ struct kvm_vcpu_hv_tlb_flush_fifo tlb_flush_fifo;
};
/* Xen HVM per vcpu emulation context */
diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
index 3c0f639f6a05..9d9a5ff2d54b 100644
--- a/arch/x86/kvm/hyperv.c
+++ b/arch/x86/kvm/hyperv.c
@@ -29,6 +29,7 @@
#include <linux/kvm_host.h>
#include <linux/highmem.h>
#include <linux/sched/cputime.h>
+#include <linux/spinlock.h>
#include <linux/eventfd.h>
#include <asm/apicdef.h>
@@ -954,6 +955,9 @@ int kvm_hv_vcpu_init(struct kvm_vcpu *vcpu)
hv_vcpu->vp_index = vcpu->vcpu_idx;
+ INIT_KFIFO(hv_vcpu->tlb_flush_fifo.entries);
+ spin_lock_init(&hv_vcpu->tlb_flush_fifo.write_lock);
+
return 0;
}
@@ -1783,6 +1787,37 @@ static u64 kvm_get_sparse_vp_set(struct kvm *kvm, struct kvm_hv_hcall *hc,
var_cnt * sizeof(*sparse_banks));
}
+static void hv_tlb_flush_enqueue(struct kvm_vcpu *vcpu)
+{
+ struct kvm_vcpu_hv_tlb_flush_fifo *tlb_flush_fifo;
+ struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
+ u64 flush_all_entry = KVM_HV_TLB_FLUSHALL_ENTRY;
+
+ if (!hv_vcpu)
+ return;
+
+ tlb_flush_fifo = &hv_vcpu->tlb_flush_fifo;
+
+ kfifo_in_spinlocked_noirqsave(&tlb_flush_fifo->entries, &flush_all_entry,
+ 1, &tlb_flush_fifo->write_lock);
+}
+
+int kvm_hv_vcpu_flush_tlb(struct kvm_vcpu *vcpu)
+{
+ struct kvm_vcpu_hv_tlb_flush_fifo *tlb_flush_fifo;
+ struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
+
+ if (!hv_vcpu)
+ return -EINVAL;
+
+ tlb_flush_fifo = &hv_vcpu->tlb_flush_fifo;
+
+ kfifo_reset_out(&tlb_flush_fifo->entries);
+
+ /* Precise flushing isn't implemented yet. */
+ return -EOPNOTSUPP;
+}
+
static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
{
struct kvm *kvm = vcpu->kvm;
@@ -1791,6 +1826,8 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
DECLARE_BITMAP(vcpu_mask, KVM_MAX_VCPUS);
u64 valid_bank_mask;
u64 sparse_banks[KVM_HV_MAX_SPARSE_VCPU_SET_BITS];
+ struct kvm_vcpu *v;
+ unsigned long i;
bool all_cpus;
/*
@@ -1870,10 +1907,20 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
* analyze it here, flush TLB regardless of the specified address space.
*/
if (all_cpus) {
+ kvm_for_each_vcpu(i, v, kvm)
+ hv_tlb_flush_enqueue(v);
+
kvm_make_all_cpus_request(kvm, KVM_REQ_HV_TLB_FLUSH);
} else {
sparse_set_to_vcpu_mask(kvm, sparse_banks, valid_bank_mask, vcpu_mask);
+ for_each_set_bit(i, vcpu_mask, KVM_MAX_VCPUS) {
+ v = kvm_get_vcpu(kvm, i);
+ if (!v)
+ continue;
+ hv_tlb_flush_enqueue(v);
+ }
+
kvm_make_vcpus_request_mask(kvm, KVM_REQ_HV_TLB_FLUSH, vcpu_mask);
}
diff --git a/arch/x86/kvm/hyperv.h b/arch/x86/kvm/hyperv.h
index 1030b1b50552..f79edf9234cd 100644
--- a/arch/x86/kvm/hyperv.h
+++ b/arch/x86/kvm/hyperv.h
@@ -151,4 +151,19 @@ int kvm_vm_ioctl_hv_eventfd(struct kvm *kvm, struct kvm_hyperv_eventfd *args);
int kvm_get_hv_cpuid(struct kvm_vcpu *vcpu, struct kvm_cpuid2 *cpuid,
struct kvm_cpuid_entry2 __user *entries);
+static inline void kvm_hv_vcpu_purge_flush_tlb(struct kvm_vcpu *vcpu)
+{
+ struct kvm_vcpu_hv_tlb_flush_fifo *tlb_flush_fifo;
+ struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
+
+ if (!hv_vcpu || !kvm_check_request(KVM_REQ_HV_TLB_FLUSH, vcpu))
+ return;
+
+ tlb_flush_fifo = &hv_vcpu->tlb_flush_fifo;
+
+ kfifo_reset_out(&tlb_flush_fifo->entries);
+}
+
+int kvm_hv_vcpu_flush_tlb(struct kvm_vcpu *vcpu);
+
#endif
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index 9210a8840224..25dc38a94fa0 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -3743,7 +3743,7 @@ static void svm_flush_tlb_current(struct kvm_vcpu *vcpu)
* A TLB flush for the current ASID flushes both "host" and "guest" TLB
* entries, and thus is a superset of Hyper-V's fine grained flushing.
*/
- kvm_clear_request(KVM_REQ_HV_TLB_FLUSH, vcpu);
+ kvm_hv_vcpu_purge_flush_tlb(vcpu);
/*
* Flush only the current ASID even if the TLB flush was invoked via
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index bf2aa77b7a2f..ed8b7802ba88 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3422,7 +3422,7 @@ static void kvm_vcpu_flush_tlb_guest(struct kvm_vcpu *vcpu)
* Flushing all "guest" TLB is always a superset of Hyper-V's fine
* grained flushing.
*/
- kvm_clear_request(KVM_REQ_HV_TLB_FLUSH, vcpu);
+ kvm_hv_vcpu_purge_flush_tlb(vcpu);
}
@@ -10497,7 +10497,14 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
kvm_service_local_tlb_flush_requests(vcpu);
- if (kvm_check_request(KVM_REQ_HV_TLB_FLUSH, vcpu))
+ /*
+ * Fall back to a "full" guest flush if Hyper-V's precise
+ * flushing fails. Note, Hyper-V's flushing is per-vCPU, but
+ * the flushes are considered "remote" and not "local" because
+ * the requests can be initiated from other vCPUs.
+ */
+ if (kvm_check_request(KVM_REQ_HV_TLB_FLUSH, vcpu) &&
+ kvm_hv_vcpu_flush_tlb(vcpu))
kvm_vcpu_flush_tlb_guest(vcpu);
if (kvm_check_request(KVM_REQ_REPORT_TPR_ACCESS, vcpu)) {
--
2.37.3
next prev parent reply other threads:[~2022-11-01 14:56 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 ` Vitaly Kuznetsov [this message]
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 ` [PATCH v13 16/48] KVM: x86: hyper-v: Don't use sparse_set_to_vcpu_mask() in kvm_hv_send_ipi() Vitaly Kuznetsov
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=20221101145426.251680-10-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).