Hyper-V supports the use of XMM registers to perform fast hypercalls.
This allows guests to take advantage of the improved performance of the
fast hypercall interface even though a hypercall may require more than
(the current maximum of) two input registers.
The XMM fast hypercall interface uses six additional XMM registers (XMM0
to XMM5) to allow the guest to pass an input parameter block of up to
112 bytes. Hyper-V can also return data back to the guest in the
remaining XMM registers that are not used by the current hypercall.
Add framework to read/write to XMM registers in kvm_hv_hypercall() and
use the additional hypercall inputs from XMM registers in
kvm_hv_flush_tlb() when possible.
Cc: Alexander Graf <[email protected]>
Co-developed-by: Evgeny Iakovlev <[email protected]>
Signed-off-by: Evgeny Iakovlev <[email protected]>
Signed-off-by: Siddharth Chandrasekaran <[email protected]>
---
arch/x86/kvm/hyperv.c | 109 ++++++++++++++++++++++++++++++++++--------
1 file changed, 90 insertions(+), 19 deletions(-)
diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
index 8f6babd1ea0d..bf2f86f263f1 100644
--- a/arch/x86/kvm/hyperv.c
+++ b/arch/x86/kvm/hyperv.c
@@ -36,6 +36,7 @@
#include "trace.h"
#include "irq.h"
+#include "fpu.h"
/* "Hv#1" signature */
#define HYPERV_CPUID_SIGNATURE_EAX 0x31237648
@@ -1623,6 +1624,8 @@ static __always_inline unsigned long *sparse_set_to_vcpu_mask(
return vcpu_bitmap;
}
+#define KVM_HV_HYPERCALL_MAX_XMM_REGISTERS 6
+
struct kvm_hv_hcall {
u64 param;
u64 ingpa;
@@ -1632,10 +1635,14 @@ struct kvm_hv_hcall {
u16 rep_idx;
bool fast;
bool rep;
+ sse128_t xmm[KVM_HV_HYPERCALL_MAX_XMM_REGISTERS];
+ bool xmm_dirty;
};
static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc, bool ex)
{
+ int i, j;
+ gpa_t gpa;
struct kvm *kvm = vcpu->kvm;
struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
struct hv_tlb_flush_ex flush_ex;
@@ -1649,8 +1656,15 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc, bool
bool all_cpus;
if (!ex) {
- if (unlikely(kvm_read_guest(kvm, hc->ingpa, &flush, sizeof(flush))))
- return HV_STATUS_INVALID_HYPERCALL_INPUT;
+ if (hc->fast) {
+ flush.address_space = hc->ingpa;
+ flush.flags = hc->outgpa;
+ flush.processor_mask = sse128_lo(hc->xmm[0]);
+ } else {
+ if (unlikely(kvm_read_guest(kvm, hc->ingpa,
+ &flush, sizeof(flush))))
+ return HV_STATUS_INVALID_HYPERCALL_INPUT;
+ }
trace_kvm_hv_flush_tlb(flush.processor_mask,
flush.address_space, flush.flags);
@@ -1668,9 +1682,16 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc, bool
all_cpus = (flush.flags & HV_FLUSH_ALL_PROCESSORS) ||
flush.processor_mask == 0;
} else {
- if (unlikely(kvm_read_guest(kvm, hc->ingpa, &flush_ex,
- sizeof(flush_ex))))
- return HV_STATUS_INVALID_HYPERCALL_INPUT;
+ if (hc->fast) {
+ flush_ex.address_space = hc->ingpa;
+ flush_ex.flags = hc->outgpa;
+ memcpy(&flush_ex.hv_vp_set,
+ &hc->xmm[0], sizeof(hc->xmm[0]));
+ } else {
+ if (unlikely(kvm_read_guest(kvm, hc->ingpa, &flush_ex,
+ sizeof(flush_ex))))
+ return HV_STATUS_INVALID_HYPERCALL_INPUT;
+ }
trace_kvm_hv_flush_tlb_ex(flush_ex.hv_vp_set.valid_bank_mask,
flush_ex.hv_vp_set.format,
@@ -1681,20 +1702,29 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc, bool
all_cpus = flush_ex.hv_vp_set.format !=
HV_GENERIC_SET_SPARSE_4K;
- sparse_banks_len =
- bitmap_weight((unsigned long *)&valid_bank_mask, 64) *
- sizeof(sparse_banks[0]);
+ sparse_banks_len = bitmap_weight((unsigned long *)&valid_bank_mask, 64);
if (!sparse_banks_len && !all_cpus)
goto ret_success;
- if (!all_cpus &&
- kvm_read_guest(kvm,
- hc->ingpa + offsetof(struct hv_tlb_flush_ex,
- hv_vp_set.bank_contents),
- sparse_banks,
- sparse_banks_len))
- return HV_STATUS_INVALID_HYPERCALL_INPUT;
+ if (!all_cpus) {
+ if (hc->fast) {
+ if (sparse_banks_len > KVM_HV_HYPERCALL_MAX_XMM_REGISTERS - 1)
+ return HV_STATUS_INVALID_HYPERCALL_INPUT;
+ for (i = 0, j = 1; i < sparse_banks_len; i += 2, j++) {
+ sparse_banks[i + 0] = sse128_lo(hc->xmm[j]);
+ sparse_banks[i + 1] = sse128_hi(hc->xmm[j]);
+ }
+ } else {
+ gpa = hc->ingpa;
+ gpa += offsetof(struct hv_tlb_flush_ex,
+ hv_vp_set.bank_contents);
+ if (unlikely(kvm_read_guest(kvm, gpa, sparse_banks,
+ sparse_banks_len *
+ sizeof(sparse_banks[0]))))
+ return HV_STATUS_INVALID_HYPERCALL_INPUT;
+ }
+ }
}
cpumask_clear(&hv_vcpu->tlb_flush);
@@ -1890,6 +1920,41 @@ static u16 kvm_hvcall_signal_event(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *h
return HV_STATUS_SUCCESS;
}
+static bool is_xmm_fast_hypercall(struct kvm_hv_hcall *hc)
+{
+ switch (hc->code) {
+ case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST:
+ case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE:
+ case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST_EX:
+ case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX:
+ return true;
+ }
+
+ return false;
+}
+
+static inline void kvm_hv_hypercall_read_xmm(struct kvm_hv_hcall *hc)
+{
+ int reg;
+
+ kvm_fpu_get();
+ for (reg = 0; reg < KVM_HV_HYPERCALL_MAX_XMM_REGISTERS; reg++)
+ _kvm_read_sse_reg(reg, &hc->xmm[reg]);
+ kvm_fpu_put();
+ hc->xmm_dirty = false;
+}
+
+static inline void kvm_hv_hypercall_write_xmm(struct kvm_hv_hcall *hc)
+{
+ int reg;
+
+ kvm_fpu_get();
+ for (reg = 0; reg < KVM_HV_HYPERCALL_MAX_XMM_REGISTERS; reg++)
+ _kvm_write_sse_reg(reg, &hc->xmm[reg]);
+ kvm_fpu_put();
+ hc->xmm_dirty = false;
+}
+
int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
{
struct kvm_hv_hcall hc;
@@ -1926,6 +1991,9 @@ int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
hc.rep_idx = (hc.param >> HV_HYPERCALL_REP_START_OFFSET) & 0xfff;
hc.rep = !!(hc.rep_cnt || hc.rep_idx);
+ if (is_xmm_fast_hypercall(&hc))
+ kvm_hv_hypercall_read_xmm(&hc);
+
trace_kvm_hv_hypercall(hc.code, hc.fast, hc.rep_cnt, hc.rep_idx,
hc.ingpa, hc.outgpa);
@@ -1961,28 +2029,28 @@ int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
kvm_hv_hypercall_complete_userspace;
return 0;
case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST:
- if (unlikely(hc.fast || !hc.rep_cnt || hc.rep_idx)) {
+ if (unlikely(!hc.rep_cnt || hc.rep_idx)) {
ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
break;
}
ret = kvm_hv_flush_tlb(vcpu, &hc, false);
break;
case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE:
- if (unlikely(hc.fast || hc.rep)) {
+ if (unlikely(hc.rep)) {
ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
break;
}
ret = kvm_hv_flush_tlb(vcpu, &hc, false);
break;
case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST_EX:
- if (unlikely(hc.fast || !hc.rep_cnt || hc.rep_idx)) {
+ if (unlikely(!hc.rep_cnt || hc.rep_idx)) {
ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
break;
}
ret = kvm_hv_flush_tlb(vcpu, &hc, true);
break;
case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX:
- if (unlikely(hc.fast || hc.rep)) {
+ if (unlikely(hc.rep)) {
ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
break;
}
@@ -2035,6 +2103,9 @@ int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
break;
}
+ if (hc.xmm_dirty)
+ kvm_hv_hypercall_write_xmm(&hc);
+
return kvm_hv_hypercall_complete(vcpu, ret);
}
--
2.17.1
Amazon Development Center Germany GmbH
Krausenstr. 38
10117 Berlin
Geschaeftsfuehrung: Christian Schlaeger, Jonathan Weiss
Eingetragen am Amtsgericht Charlottenburg unter HRB 149173 B
Sitz: Berlin
Ust-ID: DE 289 237 879
Siddharth Chandrasekaran <[email protected]> writes:
> Hyper-V supports the use of XMM registers to perform fast hypercalls.
> This allows guests to take advantage of the improved performance of the
> fast hypercall interface even though a hypercall may require more than
> (the current maximum of) two input registers.
>
> The XMM fast hypercall interface uses six additional XMM registers (XMM0
> to XMM5) to allow the guest to pass an input parameter block of up to
> 112 bytes. Hyper-V can also return data back to the guest in the
> remaining XMM registers that are not used by the current hypercall.
>
> Add framework to read/write to XMM registers in kvm_hv_hypercall() and
> use the additional hypercall inputs from XMM registers in
> kvm_hv_flush_tlb() when possible.
>
> Cc: Alexander Graf <[email protected]>
> Co-developed-by: Evgeny Iakovlev <[email protected]>
> Signed-off-by: Evgeny Iakovlev <[email protected]>
> Signed-off-by: Siddharth Chandrasekaran <[email protected]>
> ---
> arch/x86/kvm/hyperv.c | 109 ++++++++++++++++++++++++++++++++++--------
> 1 file changed, 90 insertions(+), 19 deletions(-)
>
> diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
> index 8f6babd1ea0d..bf2f86f263f1 100644
> --- a/arch/x86/kvm/hyperv.c
> +++ b/arch/x86/kvm/hyperv.c
> @@ -36,6 +36,7 @@
>
> #include "trace.h"
> #include "irq.h"
> +#include "fpu.h"
>
> /* "Hv#1" signature */
> #define HYPERV_CPUID_SIGNATURE_EAX 0x31237648
> @@ -1623,6 +1624,8 @@ static __always_inline unsigned long *sparse_set_to_vcpu_mask(
> return vcpu_bitmap;
> }
>
> +#define KVM_HV_HYPERCALL_MAX_XMM_REGISTERS 6
> +
> struct kvm_hv_hcall {
> u64 param;
> u64 ingpa;
> @@ -1632,10 +1635,14 @@ struct kvm_hv_hcall {
> u16 rep_idx;
> bool fast;
> bool rep;
> + sse128_t xmm[KVM_HV_HYPERCALL_MAX_XMM_REGISTERS];
> + bool xmm_dirty;
> };
>
> static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc, bool ex)
> {
> + int i, j;
> + gpa_t gpa;
> struct kvm *kvm = vcpu->kvm;
> struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
> struct hv_tlb_flush_ex flush_ex;
> @@ -1649,8 +1656,15 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc, bool
> bool all_cpus;
>
> if (!ex) {
> - if (unlikely(kvm_read_guest(kvm, hc->ingpa, &flush, sizeof(flush))))
> - return HV_STATUS_INVALID_HYPERCALL_INPUT;
> + if (hc->fast) {
> + flush.address_space = hc->ingpa;
> + flush.flags = hc->outgpa;
> + flush.processor_mask = sse128_lo(hc->xmm[0]);
> + } else {
> + if (unlikely(kvm_read_guest(kvm, hc->ingpa,
> + &flush, sizeof(flush))))
> + return HV_STATUS_INVALID_HYPERCALL_INPUT;
> + }
>
> trace_kvm_hv_flush_tlb(flush.processor_mask,
> flush.address_space, flush.flags);
> @@ -1668,9 +1682,16 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc, bool
> all_cpus = (flush.flags & HV_FLUSH_ALL_PROCESSORS) ||
> flush.processor_mask == 0;
> } else {
> - if (unlikely(kvm_read_guest(kvm, hc->ingpa, &flush_ex,
> - sizeof(flush_ex))))
> - return HV_STATUS_INVALID_HYPERCALL_INPUT;
> + if (hc->fast) {
> + flush_ex.address_space = hc->ingpa;
> + flush_ex.flags = hc->outgpa;
> + memcpy(&flush_ex.hv_vp_set,
> + &hc->xmm[0], sizeof(hc->xmm[0]));
> + } else {
> + if (unlikely(kvm_read_guest(kvm, hc->ingpa, &flush_ex,
> + sizeof(flush_ex))))
> + return HV_STATUS_INVALID_HYPERCALL_INPUT;
> + }
>
> trace_kvm_hv_flush_tlb_ex(flush_ex.hv_vp_set.valid_bank_mask,
> flush_ex.hv_vp_set.format,
> @@ -1681,20 +1702,29 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc, bool
> all_cpus = flush_ex.hv_vp_set.format !=
> HV_GENERIC_SET_SPARSE_4K;
>
> - sparse_banks_len =
> - bitmap_weight((unsigned long *)&valid_bank_mask, 64) *
> - sizeof(sparse_banks[0]);
> + sparse_banks_len = bitmap_weight((unsigned long *)&valid_bank_mask, 64);
>
> if (!sparse_banks_len && !all_cpus)
> goto ret_success;
>
> - if (!all_cpus &&
> - kvm_read_guest(kvm,
> - hc->ingpa + offsetof(struct hv_tlb_flush_ex,
> - hv_vp_set.bank_contents),
> - sparse_banks,
> - sparse_banks_len))
> - return HV_STATUS_INVALID_HYPERCALL_INPUT;
> + if (!all_cpus) {
> + if (hc->fast) {
> + if (sparse_banks_len > KVM_HV_HYPERCALL_MAX_XMM_REGISTERS - 1)
> + return HV_STATUS_INVALID_HYPERCALL_INPUT;
> + for (i = 0, j = 1; i < sparse_banks_len; i += 2, j++) {
> + sparse_banks[i + 0] = sse128_lo(hc->xmm[j]);
> + sparse_banks[i + 1] = sse128_hi(hc->xmm[j]);
> + }
> + } else {
> + gpa = hc->ingpa;
> + gpa += offsetof(struct hv_tlb_flush_ex,
> + hv_vp_set.bank_contents);
> + if (unlikely(kvm_read_guest(kvm, gpa, sparse_banks,
> + sparse_banks_len *
> + sizeof(sparse_banks[0]))))
> + return HV_STATUS_INVALID_HYPERCALL_INPUT;
> + }
> + }
> }
>
> cpumask_clear(&hv_vcpu->tlb_flush);
> @@ -1890,6 +1920,41 @@ static u16 kvm_hvcall_signal_event(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *h
> return HV_STATUS_SUCCESS;
> }
>
> +static bool is_xmm_fast_hypercall(struct kvm_hv_hcall *hc)
> +{
> + switch (hc->code) {
> + case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST:
> + case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE:
> + case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST_EX:
> + case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX:
> + return true;
> + }
> +
> + return false;
> +}
> +
> +static inline void kvm_hv_hypercall_read_xmm(struct kvm_hv_hcall *hc)
> +{
> + int reg;
> +
> + kvm_fpu_get();
> + for (reg = 0; reg < KVM_HV_HYPERCALL_MAX_XMM_REGISTERS; reg++)
> + _kvm_read_sse_reg(reg, &hc->xmm[reg]);
> + kvm_fpu_put();
> + hc->xmm_dirty = false;
> +}
> +
> +static inline void kvm_hv_hypercall_write_xmm(struct kvm_hv_hcall *hc)
> +{
> + int reg;
> +
> + kvm_fpu_get();
> + for (reg = 0; reg < KVM_HV_HYPERCALL_MAX_XMM_REGISTERS; reg++)
> + _kvm_write_sse_reg(reg, &hc->xmm[reg]);
> + kvm_fpu_put();
> + hc->xmm_dirty = false;
> +}
> +
> int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
> {
> struct kvm_hv_hcall hc;
> @@ -1926,6 +1991,9 @@ int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
> hc.rep_idx = (hc.param >> HV_HYPERCALL_REP_START_OFFSET) & 0xfff;
> hc.rep = !!(hc.rep_cnt || hc.rep_idx);
>
> + if (is_xmm_fast_hypercall(&hc))
> + kvm_hv_hypercall_read_xmm(&hc);
is_xmm_fast_hypercall() check should probably be complemented with " &&
hc.fast" as there's no point in reading this regs when the hypercall is
not 'fast'.
Also, we can probably defer kvm_hv_hypercall_read_xmm() until we know
how many regs we actually need to not read them all (we will always
need xmm[0] I guess so we can as well read it here).
> +
> trace_kvm_hv_hypercall(hc.code, hc.fast, hc.rep_cnt, hc.rep_idx,
> hc.ingpa, hc.outgpa);
>
> @@ -1961,28 +2029,28 @@ int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
> kvm_hv_hypercall_complete_userspace;
> return 0;
> case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST:
> - if (unlikely(hc.fast || !hc.rep_cnt || hc.rep_idx)) {
> + if (unlikely(!hc.rep_cnt || hc.rep_idx)) {
> ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
> break;
> }
> ret = kvm_hv_flush_tlb(vcpu, &hc, false);
> break;
> case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE:
> - if (unlikely(hc.fast || hc.rep)) {
> + if (unlikely(hc.rep)) {
> ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
> break;
> }
> ret = kvm_hv_flush_tlb(vcpu, &hc, false);
> break;
> case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST_EX:
> - if (unlikely(hc.fast || !hc.rep_cnt || hc.rep_idx)) {
> + if (unlikely(!hc.rep_cnt || hc.rep_idx)) {
> ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
> break;
> }
> ret = kvm_hv_flush_tlb(vcpu, &hc, true);
> break;
> case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX:
> - if (unlikely(hc.fast || hc.rep)) {
> + if (unlikely(hc.rep)) {
> ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
> break;
> }
> @@ -2035,6 +2103,9 @@ int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
> break;
> }
>
> + if (hc.xmm_dirty)
> + kvm_hv_hypercall_write_xmm(&hc);
> +
> return kvm_hv_hypercall_complete(vcpu, ret);
> }
--
Vitaly
On 08/04/21 14:01, Vitaly Kuznetsov wrote:
>
> Also, we can probably defer kvm_hv_hypercall_read_xmm() until we know
> how many regs we actually need to not read them all (we will always
> need xmm[0] I guess so we can as well read it here).
The cost is get/put FPU, so I think there's not much to gain from that.
Paolo
Paolo Bonzini <[email protected]> writes:
> On 08/04/21 14:01, Vitaly Kuznetsov wrote:
>>
>> Also, we can probably defer kvm_hv_hypercall_read_xmm() until we know
>> how many regs we actually need to not read them all (we will always
>> need xmm[0] I guess so we can as well read it here).
>
> The cost is get/put FPU, so I think there's not much to gain from that.
>
Maybe, I just think that in most cases we will only need xmm0. To make
the optimization work we can probably do kvm_get_fpu() once we figured
out that we're dealing with XMM hypercall and do kvm_put_fpu() when
we're done processing hypercall parameters. This way we don't need to do
get/put twice. We can certainly leave this idea to the (possible) future
optimizations.
--
Vitaly