lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <da036c786700032b32e68ebece06fd1a6b6bf344.1618244920.git.sidcha@amazon.de>
Date:   Mon, 12 Apr 2021 19:00:16 +0200
From:   Siddharth Chandrasekaran <sidcha@...zon.de>
To:     "K. Y. Srinivasan" <kys@...rosoft.com>,
        Haiyang Zhang <haiyangz@...rosoft.com>,
        Stephen Hemminger <sthemmin@...rosoft.com>,
        Wei Liu <wei.liu@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        <x86@...nel.org>, "H. Peter Anvin" <hpa@...or.com>,
        Paolo Bonzini <pbonzini@...hat.com>,
        "Sean Christopherson" <seanjc@...gle.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Wanpeng Li <wanpengli@...cent.com>,
        Jim Mattson <jmattson@...gle.com>,
        "Joerg Roedel" <joro@...tes.org>
CC:     Siddharth Chandrasekaran <sidcha@...zon.de>,
        Alexander Graf <graf@...zon.com>,
        Evgeny Iakovlev <eyakovl@...zon.de>,
        Liran Alon <liran@...zon.com>,
        Ioannis Aslanidis <iaslan@...zon.de>,
        <linux-hyperv@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <kvm@...r.kernel.org>
Subject: [PATCH v2 3/4] KVM: x86: kvm_hv_flush_tlb use inputs from XMM registers

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 <graf@...zon.com>
Co-developed-by: Evgeny Iakovlev <eyakovl@...zon.de>
Signed-off-by: Evgeny Iakovlev <eyakovl@...zon.de>
Signed-off-by: Siddharth Chandrasekaran <sidcha@...zon.de>
---
 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..1f9959aba70d 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 (hc.fast && 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



Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ