Re: [PATCH v3 02/34] KVM: x86: hyper-v: Introduce TLB flush ring

From: Maxim Levitsky
Date: Wed May 11 2022 - 07:20:11 EST

Next message: Maxim Levitsky: "Re: [PATCH v3 03/34] KVM: x86: hyper-v: Add helper to read hypercall data for array"
Previous message: Daniil Lunev: "Re: [PATCH 0/2] Prevent re-use of FUSE superblock after force unmount"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Thu, 2022-04-14 at 15:19 +0200, Vitaly Kuznetsov wrote:
> To allow flushing individual GVAs instead of always flushing the whole
> VPID a per-vCPU structure to pass the requests is needed. Introduce a
> simple ring write-locked structure to hold two types of entries:
> individual GVA (GFN + up to 4095 following GFNs in the lower 12 bits)
> and 'flush all'.
>
> The queuing rule is: if there's not enough space on the ring to put
> the request and leave at least 1 entry for 'flush all' - put 'flush
> all' entry.
>
> The size of the ring is arbitrary set to '16'.
>
> Note, kvm_hv_flush_tlb() only queues 'flush all' entries for now so
> there's very small functional change but the infrastructure is
> prepared to handle individual GVA flush requests.

As I see from this patch, also the code doesn't process the requests
from the ring buffer yet, but rather just ignores it completely,
and resets the whole ring buffer (kvm_hv_vcpu_empty_flush_tlb)
Maybe you should mention it here.

>
> Signed-off-by: Vitaly Kuznetsov <vkuznets@xxxxxxxxxx>
> ---
> arch/x86/include/asm/kvm_host.h | 16 +++++++
> arch/x86/kvm/hyperv.c | 83 +++++++++++++++++++++++++++++++++
> arch/x86/kvm/hyperv.h | 13 ++++++
> arch/x86/kvm/x86.c | 5 +-
> arch/x86/kvm/x86.h | 1 +
> 5 files changed, 116 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
> index 1de3ad9308d8..b4dd2ff61658 100644
> --- a/arch/x86/include/asm/kvm_host.h
> +++ b/arch/x86/include/asm/kvm_host.h
> @@ -578,6 +578,20 @@ struct kvm_vcpu_hv_synic {
> bool dont_zero_synic_pages;
> };
>
> +#define KVM_HV_TLB_FLUSH_RING_SIZE (16)
> +
> +struct kvm_vcpu_hv_tlb_flush_entry {
> + u64 addr;
> + u64 flush_all:1;
> + u64 pad:63;
> +};

Have you considered using kfifo.h library instead?

> +
> +struct kvm_vcpu_hv_tlb_flush_ring {
> + int read_idx, write_idx;
> + spinlock_t write_lock;
> + struct kvm_vcpu_hv_tlb_flush_entry entries[KVM_HV_TLB_FLUSH_RING_SIZE];
> +};
> +
> /* Hyper-V per vcpu emulation context */
> struct kvm_vcpu_hv {
> struct kvm_vcpu *vcpu;
> @@ -597,6 +611,8 @@ struct kvm_vcpu_hv {
> u32 enlightenments_ebx; /* HYPERV_CPUID_ENLIGHTMENT_INFO.EBX */
> u32 syndbg_cap_eax; /* HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES.EAX */
> } cpuid_cache;
> +
> + struct kvm_vcpu_hv_tlb_flush_ring tlb_flush_ring;
> };
>
> /* Xen HVM per vcpu emulation context */
> diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
> index b402ad059eb9..fb716cf919ed 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 +955,8 @@ static int kvm_hv_vcpu_init(struct kvm_vcpu *vcpu)
>
> hv_vcpu->vp_index = vcpu->vcpu_idx;
>
> + spin_lock_init(&hv_vcpu->tlb_flush_ring.write_lock);
> +
> return 0;
> }
>
> @@ -1789,6 +1792,74 @@ static u64 kvm_get_sparse_vp_set(struct kvm *kvm, struct kvm_hv_hcall *hc,
> var_cnt * sizeof(*sparse_banks));
> }
>
> +static inline int hv_tlb_flush_ring_free(struct kvm_vcpu_hv *hv_vcpu,
> + int read_idx, int write_idx)
> +{
> + if (write_idx >= read_idx)
> + return KVM_HV_TLB_FLUSH_RING_SIZE - (write_idx - read_idx) - 1;
> +
> + return read_idx - write_idx - 1;
> +}
> +
> +static void hv_tlb_flush_ring_enqueue(struct kvm_vcpu *vcpu)
> +{
> + struct kvm_vcpu_hv_tlb_flush_ring *tlb_flush_ring;
> + struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
> + int ring_free, write_idx, read_idx;
> + unsigned long flags;
> +
> + if (!hv_vcpu)
> + return;
> +
> + tlb_flush_ring = &hv_vcpu->tlb_flush_ring;
> +
> + spin_lock_irqsave(&tlb_flush_ring->write_lock, flags);
> +
> + /*
> + * 'read_idx' is updated by the vCPU which does the flush, this
> + * happens without 'tlb_flush_ring->write_lock' being held; make
> + * sure we read it once.
> + */
> + read_idx = READ_ONCE(tlb_flush_ring->read_idx);
> + /*
> + * 'write_idx' is only updated here, under 'tlb_flush_ring->write_lock'.
> + * allow the compiler to re-read it, it can't change.
> + */
> + write_idx = tlb_flush_ring->write_idx;
> +
> + ring_free = hv_tlb_flush_ring_free(hv_vcpu, read_idx, write_idx);
> + /* Full ring always contains 'flush all' entry */
> + if (!ring_free)
> + goto out_unlock;
> +
> + tlb_flush_ring->entries[write_idx].addr = 0;
> + tlb_flush_ring->entries[write_idx].flush_all = 1;
> + /*
> + * Advance write index only after filling in the entry to
> + * synchronize with lockless reader.
> + */
> + smp_wmb();
> + tlb_flush_ring->write_idx = (write_idx + 1) % KVM_HV_TLB_FLUSH_RING_SIZE;
> +
> +out_unlock:
> + spin_unlock_irqrestore(&tlb_flush_ring->write_lock, flags);
> +}
> +
> +void kvm_hv_vcpu_flush_tlb(struct kvm_vcpu *vcpu)
> +{
> + struct kvm_vcpu_hv_tlb_flush_ring *tlb_flush_ring;
> + struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
> +
> + kvm_vcpu_flush_tlb_guest(vcpu);
> +
> + if (!hv_vcpu)
> + return;
> +
> + tlb_flush_ring = &hv_vcpu->tlb_flush_ring;
> +
> + tlb_flush_ring->read_idx = tlb_flush_ring->write_idx;
> +}
> +
> static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
> {
> struct kvm *kvm = vcpu->kvm;
> @@ -1797,6 +1868,8 @@ 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;
>
> /*
> @@ -1876,10 +1949,20 @@ 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_ring_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_ring_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 da2737f2a956..6847caeaaf84 100644
> --- a/arch/x86/kvm/hyperv.h
> +++ b/arch/x86/kvm/hyperv.h
> @@ -147,4 +147,17 @@ 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_empty_flush_tlb(struct kvm_vcpu *vcpu)
> +{
> + struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
> +
> + if (!hv_vcpu)
> + return;
> +
> + hv_vcpu->tlb_flush_ring.read_idx = hv_vcpu->tlb_flush_ring.write_idx;
> +}
> +void kvm_hv_vcpu_flush_tlb(struct kvm_vcpu *vcpu);
> +
> +
> #endif
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index f633cff8cd7f..e5aec386d299 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -3324,7 +3324,7 @@ static void kvm_vcpu_flush_tlb_all(struct kvm_vcpu *vcpu)
> static_call(kvm_x86_flush_tlb_all)(vcpu);
> }
>
> -static void kvm_vcpu_flush_tlb_guest(struct kvm_vcpu *vcpu)
> +void kvm_vcpu_flush_tlb_guest(struct kvm_vcpu *vcpu)
> {
> ++vcpu->stat.tlb_flush;
>
> @@ -3362,7 +3362,8 @@ void kvm_service_local_tlb_flush_requests(struct kvm_vcpu *vcpu)
>
> if (kvm_check_request(KVM_REQ_TLB_FLUSH_GUEST, vcpu)) {
> kvm_vcpu_flush_tlb_guest(vcpu);
> - kvm_clear_request(KVM_REQ_HV_TLB_FLUSH, vcpu);
> + if (kvm_check_request(KVM_REQ_HV_TLB_FLUSH, vcpu))
> + kvm_hv_vcpu_empty_flush_tlb(vcpu);
> } else if (kvm_check_request(KVM_REQ_HV_TLB_FLUSH, vcpu)) {
> kvm_vcpu_flush_tlb_guest(vcpu);
> }
> diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h
> index 588792f00334..2324f496c500 100644
> --- a/arch/x86/kvm/x86.h
> +++ b/arch/x86/kvm/x86.h
> @@ -58,6 +58,7 @@ static inline unsigned int __shrink_ple_window(unsigned int val,
>
> #define MSR_IA32_CR_PAT_DEFAULT 0x0007040600070406ULL
>
> +void kvm_vcpu_flush_tlb_guest(struct kvm_vcpu *vcpu);
> void kvm_service_local_tlb_flush_requests(struct kvm_vcpu *vcpu);
> int kvm_check_nested_events(struct kvm_vcpu *vcpu);
>

Overall looks good to me. I might have missed something though.

Best regards,
Maxim Levitsky

Next message: Maxim Levitsky: "Re: [PATCH v3 03/34] KVM: x86: hyper-v: Add helper to read hypercall data for array"
Previous message: Daniil Lunev: "Re: [PATCH 0/2] Prevent re-use of FUSE superblock after force unmount"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]