Skip to contentSkip to conversation

Support writable CPU ID registers from userspace

PATCH Superseded Jing Zhang 3 participants 20 messages May 22, 2023 – May 31, 2023

Subsystems (3)

v10 of 11 Superseded Latest is v11 →

lore
Share & tools
lore

Jing Zhang

author
#0 lore raw

v10 cover Support writable CPU ID registers from userspace

This patchset refactors/adds code to support writable per guest CPU ID feature registers. Part of the code/ideas are from
https://lore.kernel.org/all/[email protected] . No functional change is intended in this patchset. With the new CPU ID feature registers infrastructure, only writtings of ID_AA64PFR0_EL1.[CSV2|CSV3], ID_AA64DFR0_EL1.PMUVer and ID_DFR0_ELF.PerfMon are allowed as KVM does before.

Writable (Configurable) per guest CPU ID feature registers are useful for creating/migrating guest on ARM CPUs with different kinds of features.

This patchset uses kvm->arch.config_lock from Oliver's lock inversion fixes at
https://lore.kernel.org/linux-arm-kernel/[email protected]/

---

* v9 - v10
- Rebased to v6.4-rc3
- Addressed some review comments from v8/v9.

* v8 -> v9
- Rebased to v6.4-rc2.
- Don't create new file id_regs.c and don't move out id regs from sys_reg_descs array to reduce the changes.

* v7 -> v8
- Move idregs table sanity check to kvm_sys_reg_table_init.
- Only allow userspace writing before VM running.
- No lock is hold for guest access to idregs.
- Addressed some other comments from Reiji and Oliver.

* v6 -> v7
- Rebased to v6.3-rc7.
- Add helpers for idregs read/write.
- Guard all idregs reads/writes.
- Add code to fix features' safe value type which is different for KVM than for the host.

* v5 -> v6
- Rebased to v6.3-rc5.
- Reuse struct sys_reg_desc's reset() callback and field val for KVM. sanitisation function and writable mask instead of creating a new data structure for idregs.
- Use get_arm64_ftr_reg() instead of exposing idregs ftr_bits array.

* v4 -> v5
- Rebased to 2fad20ae05cb (kvmarm/next)
Merge branch kvm-arm64/selftest/misc-6,4 into kvmarm-master/next
- Use kvm->arch.config_lock to guard update to multiple VM scope idregs to avoid lock inversion
- Add back IDREG() macro for idregs access
- Refactor struct id_reg_desc by using existing infrastructure.
- Addressed many other comments from Marc.

* v3 -> v4
- Remove IDREG() macro for ID reg access, use simple array access instead
- Rename kvm_arm_read_id_reg_with_encoding() to kvm_arm_read_id_reg()
- Save perfmon value in ID_DFR0_EL1 instead of pmuver
- Update perfmon in ID_DFR0_EL1 and pmuver in ID_AA64DFR0_EL1 atomically
- Remove kvm_vcpu_has_pmu() in macro kvm_pmu_is_3p5()
- Improve ID register sanity checking in kvm_arm_check_idreg_table()

* v2 -> v3
- Rebased to 96a4627dbbd4 (kvmarm/next)
Merge tag ' https://github.com/oupton/linux tags/kvmarm-6.3' from into kvmarm-master/next
- Add id registere emulation entry point function emulate_id_reg
- Fix consistency for ID_AA64DFR0_EL1.PMUVer and ID_DFR0_EL1.PerfMon
- Improve the checking for id register table by ensuring that every entry has the correct id register encoding.
- Addressed other comments from Reiji and Marc.

* v1 -> v2
- Rebase to 7121a2e1d107 (kvmarm/next) Merge branch kvm-arm64/nv-prefix into kvmarm/next
- Address writing issue for PMUVer

[1] https://lore.kernel.org/all/[email protected]
[2] https://lore.kernel.org/all/[email protected]
[3] https://lore.kernel.org/all/[email protected]
[4] https://lore.kernel.org/all/[email protected]
[5] https://lore.kernel.org/all/[email protected]
[6] https://lore.kernel.org/all/[email protected]
[7] https://lore.kernel.org/all/[email protected]
[8] https://lore.kernel.org/all/[email protected]
[9] https://lore.kernel.org/all/[email protected]

---

Jing Zhang (5):
KVM: arm64: Save ID registers' sanitized value per guest
KVM: arm64: Use per guest ID register for ID_AA64PFR0_EL1.[CSV2|CSV3]
KVM: arm64: Use per guest ID register for ID_AA64DFR0_EL1.PMUVer
KVM: arm64: Reuse fields of sys_reg_desc for idreg
KVM: arm64: Refactor writings for PMUVer/CSV2/CSV3

 arch/arm64/include/asm/cpufeature.h |   1 +
 arch/arm64/include/asm/kvm_host.h   |  34 +-
 arch/arm64/kernel/cpufeature.c      |   2 +-
 arch/arm64/kvm/arm.c                |  24 +-
 arch/arm64/kvm/sys_regs.c           | 469 +++++++++++++++++++++++-----
 arch/arm64/kvm/sys_regs.h           |  22 +-
 include/kvm/arm_pmu.h               |   5 +-
 7 files changed, 437 insertions(+), 120 deletions(-)

base-commit: 44c026a73be8038f03dbdeef028b642880cf1511

Signature · 1 line
2.40.1.698.g37aff9b760-goog
author
#1 lore raw

PATCH v10 1/5 KVM: arm64: Save ID registers' sanitized value per guest

Introduce id_regs[] in kvm_arch as a storage of guest's ID registers, and save ID registers' sanitized value in the array at KVM_CREATE_VM. Use the saved ones when ID registers are read by the guest or userspace (via KVM_GET_ONE_REG).

No functional change intended.

---
 arch/arm64/include/asm/kvm_host.h | 20 +++++++++
 arch/arm64/kvm/arm.c              |  1 +
 arch/arm64/kvm/sys_regs.c         | 69 +++++++++++++++++++++++++------
 arch/arm64/kvm/sys_regs.h         |  7 ++++
 4 files changed, 85 insertions(+), 12 deletions(-)
arch/arm64/include/asm/kvm_host.h +20 -0
index 7e7e19ef6993..069606170c82 100644
@@ -178,6 +178,21 @@ struct kvm_smccc_features {
 	unsigned long vendor_hyp_bmap;
 };
 
+/*
+ * Emulated CPU ID registers per VM
+ * (Op0, Op1, CRn, CRm, Op2) of the ID registers to be saved in it
+ * is (3, 0, 0, crm, op2), where 1<=crm<8, 0<=op2<8.
+ *
+ * These emulated idregs are VM-wide, but accessed from the context of a vCPU.
+ * Atomic access to multiple idregs are guarded by kvm_arch.config_lock.
+ */
+#define IDREG_IDX(id)		(((sys_reg_CRm(id) - 1) << 3) | sys_reg_Op2(id))
+#define IDREG(kvm, id)		((kvm)->arch.idregs.regs[IDREG_IDX(id)])
+#define KVM_ARM_ID_REG_NUM	(IDREG_IDX(sys_reg(3, 0, 0, 7, 7)) + 1)
+struct kvm_idregs {
+	u64 regs[KVM_ARM_ID_REG_NUM];
+};
+
 typedef unsigned int pkvm_handle_t;
 
 struct kvm_protected_vm {
@@ -253,6 +268,9 @@ struct kvm_arch {
 	struct kvm_smccc_features smccc_feat;
 	struct maple_tree smccc_filter;
 
+	/* Emulated CPU ID registers */
+	struct kvm_idregs idregs;
+
 	/*
 	 * For an untrusted host VM, 'pkvm.handle' is used to lookup
 	 * the associated pKVM instance in the hypervisor.
@@ -1045,6 +1063,8 @@ int kvm_vm_ioctl_mte_copy_tags(struct kvm *kvm,
 int kvm_vm_ioctl_set_counter_offset(struct kvm *kvm,
 				    struct kvm_arm_counter_offset *offset);
 
+void kvm_arm_init_id_regs(struct kvm *kvm);
+
 /* Guest/host FPSIMD coordination helpers */
 int kvm_arch_vcpu_run_map_fp(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_load_fp(struct kvm_vcpu *vcpu);
arch/arm64/kvm/arm.c +1 -0
index 14391826241c..774656a0718d 100644
@@ -163,6 +163,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 
 	set_default_spectre(kvm);
 	kvm_arm_init_hypercalls(kvm);
+	kvm_arm_init_id_regs(kvm);
 
 	/*
 	 * Initialise the default PMUver before there is a chance to
arch/arm64/kvm/sys_regs.c +57 -12

106 lines not on this page · raw message

arch/arm64/kvm/sys_regs.h +7 -0
index 6b11f2cc7146..eba10de2e7ae 100644
@@ -27,6 +27,13 @@ struct sys_reg_params {
 	bool	is_write;
 };
 
+#define encoding_to_params(reg)						\
+	((struct sys_reg_params){ .Op0 = sys_reg_Op0(reg),		\
+				  .Op1 = sys_reg_Op1(reg),		\
+				  .CRn = sys_reg_CRn(reg),		\
+				  .CRm = sys_reg_CRm(reg),		\
+				  .Op2 = sys_reg_Op2(reg) })
+
 #define esr_sys64_to_params(esr)                                               \
 	((struct sys_reg_params){ .Op0 = ((esr) >> 20) & 3,                    \
 				  .Op1 = ((esr) >> 14) & 0x7,                  \
Signature · 1 line
2.40.1.698.g37aff9b760-goog
maintainer
#2 lore raw

Re: [PATCH v10 1/5] KVM: arm64: Save ID registers' sanitized value per guest

Quoted 97 lines · Jing Zhang

Introduce id_regs[] in kvm_arch as a storage of guest's ID registers, and save ID registers' sanitized value in the array at KVM_CREATE_VM. Use the saved ones when ID registers are read by the guest or userspace (via KVM_GET_ONE_REG).

No functional change intended.

---
 arch/arm64/include/asm/kvm_host.h | 20 +++++++++
 arch/arm64/kvm/arm.c              |  1 +
 arch/arm64/kvm/sys_regs.c         | 69 +++++++++++++++++++++++++------
 arch/arm64/kvm/sys_regs.h         |  7 ++++
 4 files changed, 85 insertions(+), 12 deletions(-)
arch/arm64/include/asm/kvm_host.h +20 -0
index 7e7e19ef6993..069606170c82 100644
@@ -178,6 +178,21 @@ struct kvm_smccc_features {
 	unsigned long vendor_hyp_bmap;
 };
 
+/*
+ * Emulated CPU ID registers per VM
+ * (Op0, Op1, CRn, CRm, Op2) of the ID registers to be saved in it
+ * is (3, 0, 0, crm, op2), where 1<=crm<8, 0<=op2<8.
+ *
+ * These emulated idregs are VM-wide, but accessed from the context of a vCPU.
+ * Atomic access to multiple idregs are guarded by kvm_arch.config_lock.
+ */
+#define IDREG_IDX(id)		(((sys_reg_CRm(id) - 1) << 3) | sys_reg_Op2(id))
+#define IDREG(kvm, id)		((kvm)->arch.idregs.regs[IDREG_IDX(id)])
+#define KVM_ARM_ID_REG_NUM	(IDREG_IDX(sys_reg(3, 0, 0, 7, 7)) + 1)
+struct kvm_idregs {
+	u64 regs[KVM_ARM_ID_REG_NUM];
+};
+
 typedef unsigned int pkvm_handle_t;
 
 struct kvm_protected_vm {
@@ -253,6 +268,9 @@ struct kvm_arch {
 	struct kvm_smccc_features smccc_feat;
 	struct maple_tree smccc_filter;
 
+	/* Emulated CPU ID registers */
+	struct kvm_idregs idregs;
+
 	/*
 	 * For an untrusted host VM, 'pkvm.handle' is used to lookup
 	 * the associated pKVM instance in the hypervisor.
@@ -1045,6 +1063,8 @@ int kvm_vm_ioctl_mte_copy_tags(struct kvm *kvm,
 int kvm_vm_ioctl_set_counter_offset(struct kvm *kvm,
 				    struct kvm_arm_counter_offset *offset);
 
+void kvm_arm_init_id_regs(struct kvm *kvm);
+
 /* Guest/host FPSIMD coordination helpers */
 int kvm_arch_vcpu_run_map_fp(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_load_fp(struct kvm_vcpu *vcpu);
arch/arm64/kvm/arm.c +1 -0
index 14391826241c..774656a0718d 100644
@@ -163,6 +163,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 
 	set_default_spectre(kvm);
 	kvm_arm_init_hypercalls(kvm);
+	kvm_arm_init_id_regs(kvm);
 
 	/*
 	 * Initialise the default PMUver before there is a chance to
arch/arm64/kvm/sys_regs.c +2 -1
index 71b12094d613..d2ee3a1c7f03 100644
@@ -41,6 +41,7 @@
  * 64bit interface.
  */
 
+static u64 kvm_arm_read_id_reg(const struct kvm_vcpu *vcpu, u32 id);
 static u64 sys_reg_to_index(const struct sys_reg_desc *reg);
 
 static bool read_from_write_only(struct kvm_vcpu *vcpu,
@@ -364,7 +365,7 @@ static bool trap_loregion(struct kvm_vcpu *vcpu,
 			  struct sys_reg_params *p,
 			  const struct sys_reg_desc *r)
 {
-	u64 val = read_sanitised_ftr_reg(SYS_ID_AA64MMFR1_EL1);
+	u64 val = kvm_arm_read_id_reg(vcpu, SYS_ID_AA64MMFR1_EL1);
 	u32 sr = reg_to_encoding(r);
 
 	if (!(val & (0xfUL << ID_AA64MMFR1_EL1_LO_SHIFT))) {

arch/arm64/kvm/sys_regs.c static u8 pmuver_to_perfmon(u8 pmuver) · line 1212

Quoted 6 lines
arch/arm64/kvm/sys_regs.c +1 -2
@@ -1208,16 +1209,9 @@ static u8 pmuver_to_perfmon(u8 pmuver)
 	}
 }
 
-/* Read a sanitised cpufeature ID register by sys_reg_desc */
-static u64 read_id_reg(const struct kvm_vcpu *vcpu, struct sys_reg_desc const *r)
+static u64 kvm_arm_read_id_reg(const struct kvm_vcpu *vcpu, u32 id)

'id' is misleading here. Is it an encoding? Or is it the index in the array? Reading the code, it is likely to be the former, but maybe calling the parameter 'encoding' would help.

Quoted 59 lines
arch/arm64/kvm/sys_regs.c +28 -9
 {
-	u32 id = reg_to_encoding(r);
-	u64 val;
-
-	if (sysreg_visible_as_raz(vcpu, r))
-		return 0;
-
-	val = read_sanitised_ftr_reg(id);
+	u64 val = IDREG(vcpu->kvm, id);
 
 	switch (id) {
 	case SYS_ID_AA64PFR0_EL1:
@@ -1280,6 +1274,26 @@ static u64 read_id_reg(const struct kvm_vcpu *vcpu, struct sys_reg_desc const *r
 	return val;
 }
 
+/* Read a sanitised cpufeature ID register by sys_reg_desc */
+static u64 read_id_reg(const struct kvm_vcpu *vcpu, struct sys_reg_desc const *r)
+{
+	if (sysreg_visible_as_raz(vcpu, r))
+		return 0;
+
+	return kvm_arm_read_id_reg(vcpu, reg_to_encoding(r));
+}
+
+/*
+ * Return true if the register's (Op0, Op1, CRn, CRm, Op2) is
+ * (3, 0, 0, crm, op2), where 1<=crm<8, 0<=op2<8.
+ */
+static inline bool is_id_reg(u32 id)
+{
+	return (sys_reg_Op0(id) == 3 && sys_reg_Op1(id) == 0 &&
+		sys_reg_CRn(id) == 0 && sys_reg_CRm(id) >= 1 &&
+		sys_reg_CRm(id) < 8);
+}
+
 static unsigned int id_visibility(const struct kvm_vcpu *vcpu,
 				  const struct sys_reg_desc *r)
 {
@@ -2244,8 +2258,8 @@ static bool trap_dbgdidr(struct kvm_vcpu *vcpu,
 	if (p->is_write) {
 		return ignore_write(vcpu, p);
 	} else {
-		u64 dfr = read_sanitised_ftr_reg(SYS_ID_AA64DFR0_EL1);
-		u64 pfr = read_sanitised_ftr_reg(SYS_ID_AA64PFR0_EL1);
+		u64 dfr = kvm_arm_read_id_reg(vcpu, SYS_ID_AA64DFR0_EL1);
+		u64 pfr = kvm_arm_read_id_reg(vcpu, SYS_ID_AA64PFR0_EL1);
 		u32 el3 = !!cpuid_feature_extract_unsigned_field(pfr, ID_AA64PFR0_EL1_EL3_SHIFT);
 
 		p->regval = ((((dfr >> ID_AA64DFR0_EL1_WRPs_SHIFT) & 0xf) << 28) |
@@ -3343,6 +3357,37 @@ int kvm_arm_copy_sys_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
 	return write_demux_regids(uindices);
 }
 
+/*
+ * Set the guest's ID registers with ID_SANITISED() to the host's sanitized value.
+ */
+void kvm_arm_init_id_regs(struct kvm *kvm)
+{

arch/arm64/kvm/sys_regs.c int kvm_arm_copy_sys_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices) · line 3374

Quoted 10 lines
arch/arm64/kvm/sys_regs.c +10 -0
+	const struct sys_reg_desc *idreg;
+	struct sys_reg_params params;
+	u32 id;
+
+	/* Find the first idreg (SYS_ID_PFR0_EL1) in sys_reg_descs. */
+	id = SYS_ID_PFR0_EL1;
+	params = encoding_to_params(id);
+	idreg = find_reg(&params, sys_reg_descs, ARRAY_SIZE(sys_reg_descs));
+	if (WARN_ON(!idreg))
+		return;

What is this trying to guard against? Not finding ID_PFR0_EL1 in the sysreg table? But this says nothing about the following registers (all 55 of them), so why do we need to special-case this one?

Quoted 27 lines
arch/arm64/kvm/sys_regs.c +16 -0
+
+	/* Initialize all idregs */
+	while (is_id_reg(id)) {
+		/*
+		 * Some hidden ID registers which are not in arm64_ftr_regs[]
+		 * would cause warnings from read_sanitised_ftr_reg().
+		 * Skip those ID registers to avoid the warnings.
+		 */
+		if (idreg->visibility != raz_visibility)
+			IDREG(kvm, id) = read_sanitised_ftr_reg(id);
+
+		idreg++;
+		id = reg_to_encoding(idreg);
+	}
+}
+
 int __init kvm_sys_reg_table_init(void)
 {
 	bool valid = true;
arch/arm64/kvm/sys_regs.h +0 -0
index 6b11f2cc7146..eba10de2e7ae 100644
@@ -27,6 +27,13 @@ struct sys_reg_params {
 	bool	is_write;
 };
 

arch/arm64/kvm/sys_regs.h struct sys_reg_params { · line 39

Quoted 10 lines
arch/arm64/kvm/sys_regs.h +7 -0
+#define encoding_to_params(reg)						\
+	((struct sys_reg_params){ .Op0 = sys_reg_Op0(reg),		\
+				  .Op1 = sys_reg_Op1(reg),		\
+				  .CRn = sys_reg_CRn(reg),		\
+				  .CRm = sys_reg_CRm(reg),		\
+				  .Op2 = sys_reg_Op2(reg) })
+
 #define esr_sys64_to_params(esr)                                               \
 	((struct sys_reg_params){ .Op0 = ((esr) >> 20) & 3,                    \
 				  .Op1 = ((esr) >> 14) & 0x7,                  \

Thanks,

	M.
Signature · 1 line
Without deviation from the norm, progress is not possible.
author
#3 lore raw

Re: [PATCH v10 1/5] KVM: arm64: Save ID registers' sanitized value per guest

Quoted 103 lines · Marc Zyngier
Quoted 100 lines · Jing Zhang

Introduce id_regs[] in kvm_arch as a storage of guest's ID registers, and save ID registers' sanitized value in the array at KVM_CREATE_VM. Use the saved ones when ID registers are read by the guest or userspace (via KVM_GET_ONE_REG).

No functional change intended.

---
 arch/arm64/include/asm/kvm_host.h | 20 +++++++++
 arch/arm64/kvm/arm.c              |  1 +
 arch/arm64/kvm/sys_regs.c         | 69 +++++++++++++++++++++++++------
 arch/arm64/kvm/sys_regs.h         |  7 ++++
 4 files changed, 85 insertions(+), 12 deletions(-)
arch/arm64/include/asm/kvm_host.h +20 -0
index 7e7e19ef6993..069606170c82 100644
@@ -178,6 +178,21 @@ struct kvm_smccc_features {
      unsigned long vendor_hyp_bmap;
 };
 
+/*
+ * Emulated CPU ID registers per VM
+ * (Op0, Op1, CRn, CRm, Op2) of the ID registers to be saved in it
+ * is (3, 0, 0, crm, op2), where 1<=crm<8, 0<=op2<8.
+ *
+ * These emulated idregs are VM-wide, but accessed from the context of a vCPU.
+ * Atomic access to multiple idregs are guarded by kvm_arch.config_lock.
+ */
+#define IDREG_IDX(id)                (((sys_reg_CRm(id) - 1) << 3) | sys_reg_Op2(id))
+#define IDREG(kvm, id)               ((kvm)->arch.idregs.regs[IDREG_IDX(id)])
+#define KVM_ARM_ID_REG_NUM   (IDREG_IDX(sys_reg(3, 0, 0, 7, 7)) + 1)
+struct kvm_idregs {
+     u64 regs[KVM_ARM_ID_REG_NUM];
+};
+
 typedef unsigned int pkvm_handle_t;
 
 struct kvm_protected_vm {
@@ -253,6 +268,9 @@ struct kvm_arch {
      struct kvm_smccc_features smccc_feat;
      struct maple_tree smccc_filter;
 
+     /* Emulated CPU ID registers */
+     struct kvm_idregs idregs;
+
      /*
       * For an untrusted host VM, 'pkvm.handle' is used to lookup
       * the associated pKVM instance in the hypervisor.
@@ -1045,6 +1063,8 @@ int kvm_vm_ioctl_mte_copy_tags(struct kvm *kvm,
 int kvm_vm_ioctl_set_counter_offset(struct kvm *kvm,
                                  struct kvm_arm_counter_offset *offset);
 
+void kvm_arm_init_id_regs(struct kvm *kvm);
+
 /* Guest/host FPSIMD coordination helpers */
 int kvm_arch_vcpu_run_map_fp(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_load_fp(struct kvm_vcpu *vcpu);
arch/arm64/kvm/arm.c +1 -0
index 14391826241c..774656a0718d 100644
@@ -163,6 +163,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 
      set_default_spectre(kvm);
      kvm_arm_init_hypercalls(kvm);
+     kvm_arm_init_id_regs(kvm);
 
      /*
       * Initialise the default PMUver before there is a chance to
arch/arm64/kvm/sys_regs.c +2 -1
index 71b12094d613..d2ee3a1c7f03 100644
@@ -41,6 +41,7 @@
  * 64bit interface.
  */
 
+static u64 kvm_arm_read_id_reg(const struct kvm_vcpu *vcpu, u32 id);
 static u64 sys_reg_to_index(const struct sys_reg_desc *reg);
 
 static bool read_from_write_only(struct kvm_vcpu *vcpu,
@@ -364,7 +365,7 @@ static bool trap_loregion(struct kvm_vcpu *vcpu,
                        struct sys_reg_params *p,
                        const struct sys_reg_desc *r)
 {
-     u64 val = read_sanitised_ftr_reg(SYS_ID_AA64MMFR1_EL1);
+     u64 val = kvm_arm_read_id_reg(vcpu, SYS_ID_AA64MMFR1_EL1);
      u32 sr = reg_to_encoding(r);
 
      if (!(val & (0xfUL << ID_AA64MMFR1_EL1_LO_SHIFT))) {
@@ -1208,16 +1209,9 @@ static u8 pmuver_to_perfmon(u8 pmuver)
      }
 }
 
Quoted 7 lines

arch/arm64/kvm/sys_regs.c static u8 pmuver_to_perfmon(u8 pmuver) · line 1212

Quoted 3 lines
arch/arm64/kvm/sys_regs.c +1 -2
-/* Read a sanitised cpufeature ID register by sys_reg_desc */
-static u64 read_id_reg(const struct kvm_vcpu *vcpu, struct sys_reg_desc const *r)
+static u64 kvm_arm_read_id_reg(const struct kvm_vcpu *vcpu, u32 id)

'id' is misleading here. Is it an encoding? Or is it the index in the array? Reading the code, it is likely to be the former, but maybe calling the parameter 'encoding' would help.

Yes, it is an encoding. Will replace it with 'encoding".

Quoted 66 lines
Quoted 66 lines
arch/arm64/kvm/sys_regs.c +34 -9
 {
-     u32 id = reg_to_encoding(r);
-     u64 val;
-
-     if (sysreg_visible_as_raz(vcpu, r))
-             return 0;
-
-     val = read_sanitised_ftr_reg(id);
+     u64 val = IDREG(vcpu->kvm, id);
 
      switch (id) {
      case SYS_ID_AA64PFR0_EL1:
@@ -1280,6 +1274,26 @@ static u64 read_id_reg(const struct kvm_vcpu *vcpu, struct sys_reg_desc const *r
      return val;
 }
 
+/* Read a sanitised cpufeature ID register by sys_reg_desc */
+static u64 read_id_reg(const struct kvm_vcpu *vcpu, struct sys_reg_desc const *r)
+{
+     if (sysreg_visible_as_raz(vcpu, r))
+             return 0;
+
+     return kvm_arm_read_id_reg(vcpu, reg_to_encoding(r));
+}
+
+/*
+ * Return true if the register's (Op0, Op1, CRn, CRm, Op2) is
+ * (3, 0, 0, crm, op2), where 1<=crm<8, 0<=op2<8.
+ */
+static inline bool is_id_reg(u32 id)
+{
+     return (sys_reg_Op0(id) == 3 && sys_reg_Op1(id) == 0 &&
+             sys_reg_CRn(id) == 0 && sys_reg_CRm(id) >= 1 &&
+             sys_reg_CRm(id) < 8);
+}
+
 static unsigned int id_visibility(const struct kvm_vcpu *vcpu,
                                const struct sys_reg_desc *r)
 {
@@ -2244,8 +2258,8 @@ static bool trap_dbgdidr(struct kvm_vcpu *vcpu,
      if (p->is_write) {
              return ignore_write(vcpu, p);
      } else {
-             u64 dfr = read_sanitised_ftr_reg(SYS_ID_AA64DFR0_EL1);
-             u64 pfr = read_sanitised_ftr_reg(SYS_ID_AA64PFR0_EL1);
+             u64 dfr = kvm_arm_read_id_reg(vcpu, SYS_ID_AA64DFR0_EL1);
+             u64 pfr = kvm_arm_read_id_reg(vcpu, SYS_ID_AA64PFR0_EL1);
              u32 el3 = !!cpuid_feature_extract_unsigned_field(pfr, ID_AA64PFR0_EL1_EL3_SHIFT);
 
              p->regval = ((((dfr >> ID_AA64DFR0_EL1_WRPs_SHIFT) & 0xf) << 28) |
@@ -3343,6 +3357,37 @@ int kvm_arm_copy_sys_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
      return write_demux_regids(uindices);
 }
 
+/*
+ * Set the guest's ID registers with ID_SANITISED() to the host's sanitized value.
+ */
+void kvm_arm_init_id_regs(struct kvm *kvm)
+{
+     const struct sys_reg_desc *idreg;
+     struct sys_reg_params params;
+     u32 id;
+
+     /* Find the first idreg (SYS_ID_PFR0_EL1) in sys_reg_descs. */
+     id = SYS_ID_PFR0_EL1;
Quoted 8 lines

arch/arm64/kvm/sys_regs.c int kvm_arm_copy_sys_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices) · line 3374

Quoted 4 lines
arch/arm64/kvm/sys_regs.c +4 -0
+     params = encoding_to_params(id);
+     idreg = find_reg(&params, sys_reg_descs, ARRAY_SIZE(sys_reg_descs));
+     if (WARN_ON(!idreg))
+             return;

What is this trying to guard against? Not finding ID_PFR0_EL1 in the sysreg table? But this says nothing about the following registers (all 55 of them), so why do we need to special-case this one?

Here is to find the first idreg in the array and warn that no idregs found in the array with the assumption that ID_PFR0_EL1 is the first one defined and if it is not found, then no other idregs are defined either.
Another way is to go through all the regs in array sys_reg_descs and do the initialization if it is a idreg.

Quoted 41 lines
Quoted 28 lines
arch/arm64/kvm/sys_regs.c +16 -0
+
+     /* Initialize all idregs */
+     while (is_id_reg(id)) {
+             /*
+              * Some hidden ID registers which are not in arm64_ftr_regs[]
+              * would cause warnings from read_sanitised_ftr_reg().
+              * Skip those ID registers to avoid the warnings.
+              */
+             if (idreg->visibility != raz_visibility)
+                     IDREG(kvm, id) = read_sanitised_ftr_reg(id);
+
+             idreg++;
+             id = reg_to_encoding(idreg);
+     }
+}
+
 int __init kvm_sys_reg_table_init(void)
 {
      bool valid = true;
arch/arm64/kvm/sys_regs.h +0 -0
index 6b11f2cc7146..eba10de2e7ae 100644
@@ -27,6 +27,13 @@ struct sys_reg_params {
      bool    is_write;
 };
 

arch/arm64/kvm/sys_regs.h struct sys_reg_params { · line 39

Quoted 10 lines
arch/arm64/kvm/sys_regs.h +7 -0
+#define encoding_to_params(reg)                                              \
+     ((struct sys_reg_params){ .Op0 = sys_reg_Op0(reg),              \
+                               .Op1 = sys_reg_Op1(reg),              \
+                               .CRn = sys_reg_CRn(reg),              \
+                               .CRm = sys_reg_CRm(reg),              \
+                               .Op2 = sys_reg_Op2(reg) })
+
 #define esr_sys64_to_params(esr)                                               \
      ((struct sys_reg_params){ .Op0 = ((esr) >> 20) & 3,                    \
                                .Op1 = ((esr) >> 14) & 0x7,                  \

Thanks,

Quoted 4 lines
        M.

--
Without deviation from the norm, progress is not possible.

Thanks,
Jing

maintainer
#4 lore raw

Re: [PATCH v10 1/5] KVM: arm64: Save ID registers' sanitized value per guest

On Tue, 30 May 2023 19:02:03 +0100,
Jing Zhang <[email protected]> wrote:

[...]

Quoted 19 lines
Quoted 11 lines
Quoted 11 lines
arch/arm64/kvm/sys_regs.c +11 -0
+/*
+ * Set the guest's ID registers with ID_SANITISED() to the host's sanitized value.
+ */
+void kvm_arm_init_id_regs(struct kvm *kvm)
+{
+     const struct sys_reg_desc *idreg;
+     struct sys_reg_params params;
+     u32 id;
+
+     /* Find the first idreg (SYS_ID_PFR0_EL1) in sys_reg_descs. */
+     id = SYS_ID_PFR0_EL1;
Quoted 8 lines

arch/arm64/kvm/sys_regs.c int kvm_arm_copy_sys_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices) · line 3374

Quoted 4 lines
arch/arm64/kvm/sys_regs.c +4 -0
+     params = encoding_to_params(id);
+     idreg = find_reg(&params, sys_reg_descs, ARRAY_SIZE(sys_reg_descs));
+     if (WARN_ON(!idreg))
+             return;

What is this trying to guard against? Not finding ID_PFR0_EL1 in the sysreg table? But this says nothing about the following registers (all 55 of them), so why do we need to special-case this one?

Quoted 4 lines

Here is to find the first idreg in the array and warn that no idregs found in the array with the assumption that ID_PFR0_EL1 is the first one defined and if it is not found, then no other idregs are defined either.

I didn't make my point clear. What we have is a purely static array. Why should we perform such a test on every single VM creation? Any structural validation should only happen once, at KVM init time.

Quoted 2 lines

Another way is to go through all the regs in array sys_reg_descs and do the initialization if it is a idreg.

That'd be a waste of precious cycles.

This WARN_ON()+early return should go, but the rest is fine.

Thanks,

	M.
Signature · 1 line
Without deviation from the norm, progress is not possible.
author
#5 lore raw

Re: [PATCH v10 1/5] KVM: arm64: Save ID registers' sanitized value per guest

Hi Marc,

Quoted 39 lines · Marc Zyngier

On Tue, 30 May 2023 19:02:03 +0100,
Jing Zhang <[email protected]> wrote:

[...]

Quoted 19 lines
Quoted 11 lines
Quoted 11 lines
arch/arm64/kvm/sys_regs.c +11 -0
+/*
+ * Set the guest's ID registers with ID_SANITISED() to the host's sanitized value.
+ */
+void kvm_arm_init_id_regs(struct kvm *kvm)
+{
+     const struct sys_reg_desc *idreg;
+     struct sys_reg_params params;
+     u32 id;
+
+     /* Find the first idreg (SYS_ID_PFR0_EL1) in sys_reg_descs. */
+     id = SYS_ID_PFR0_EL1;
Quoted 8 lines

arch/arm64/kvm/sys_regs.c int kvm_arm_copy_sys_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices) · line 3374

Quoted 4 lines
arch/arm64/kvm/sys_regs.c +4 -0
+     params = encoding_to_params(id);
+     idreg = find_reg(&params, sys_reg_descs, ARRAY_SIZE(sys_reg_descs));
+     if (WARN_ON(!idreg))
+             return;

What is this trying to guard against? Not finding ID_PFR0_EL1 in the sysreg table? But this says nothing about the following registers (all 55 of them), so why do we need to special-case this one?

Quoted 4 lines

Here is to find the first idreg in the array and warn that no idregs found in the array with the assumption that ID_PFR0_EL1 is the first one defined and if it is not found, then no other idregs are defined either.

I didn't make my point clear. What we have is a purely static array. Why should we perform such a test on every single VM creation? Any structural validation should only happen once, at KVM init time.

Quoted 2 lines

Another way is to go through all the regs in array sys_reg_descs and do the initialization if it is a idreg.

That'd be a waste of precious cycles.

Quoted 8 lines

This WARN_ON()+early return should go, but the rest is fine.

Thanks,

        M.

--
Without deviation from the norm, progress is not possible.

That makes sense. We don't have to do the check on every single VM creation. And the code to find the ID_PFR0_EL1 should not be done for every VM creation either.
I'll use a static variable to save the pointer to ID_PFR0_EL1 and do the check and search in kvm_sys_reg_table_init().

Thanks,
Jing

author
#6 lore raw

PATCH v10 2/5 KVM: arm64: Use per guest ID register for ID_AA64PFR0_EL1.[CSV2|CSV3]

In reply to Jing Zhang

With per guest ID registers, ID_AA64PFR0_EL1.[CSV2|CSV3] settings from userspace can be stored in its corresponding ID register.

The setting of CSV bits for protected VMs are removed according to the discussion from Fuad below:
https://lore.kernel.org/all/CA+EHjTwXA9TprX4jeG+-D+c8v9XG+oFdU1o6TSkvVye145_OvA@mail.gmail.com

Besides the removal of CSV bits setting for protected VMs, No other functional change intended.

---
 arch/arm64/include/asm/kvm_host.h |  2 --
 arch/arm64/kvm/arm.c              | 17 ---------
 arch/arm64/kvm/sys_regs.c         | 58 +++++++++++++++++++++++++------
 3 files changed, 47 insertions(+), 30 deletions(-)
arch/arm64/include/asm/kvm_host.h +0 -2
index 069606170c82..8a2fde6c04c4 100644
@@ -257,8 +257,6 @@ struct kvm_arch {
 
 	cpumask_var_t supported_cpus;
 
-	u8 pfr0_csv2;
-	u8 pfr0_csv3;
 	struct {
 		u8 imp:4;
 		u8 unimp:4;
arch/arm64/kvm/arm.c +0 -17
index 774656a0718d..5114521ace60 100644
@@ -102,22 +102,6 @@ static int kvm_arm_default_max_vcpus(void)
 	return vgic_present ? kvm_vgic_get_max_vcpus() : KVM_MAX_VCPUS;
 }
 
-static void set_default_spectre(struct kvm *kvm)
-{
-	/*
-	 * The default is to expose CSV2 == 1 if the HW isn't affected.
-	 * Although this is a per-CPU feature, we make it global because
-	 * asymmetric systems are just a nuisance.
-	 *
-	 * Userspace can override this as long as it doesn't promise
-	 * the impossible.
-	 */
-	if (arm64_get_spectre_v2_state() == SPECTRE_UNAFFECTED)
-		kvm->arch.pfr0_csv2 = 1;
-	if (arm64_get_meltdown_state() == SPECTRE_UNAFFECTED)
-		kvm->arch.pfr0_csv3 = 1;
-}
-
 /**
  * kvm_arch_init_vm - initializes a VM data structure
  * @kvm:	pointer to the KVM struct
@@ -161,7 +145,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	/* The maximum number of VCPUs is limited by the host's GIC model */
 	kvm->max_vcpus = kvm_arm_default_max_vcpus();
 
-	set_default_spectre(kvm);
 	kvm_arm_init_hypercalls(kvm);
 	kvm_arm_init_id_regs(kvm);
 
arch/arm64/kvm/sys_regs.c +47 -11

100 lines not on this page · raw message

Signature · 1 line
2.40.1.698.g37aff9b760-goog
maintainer
#7 lore raw

Re: [PATCH v10 2/5] KVM: arm64: Use per guest ID register for ID_AA64PFR0_EL1.[CSV2|CSV3]

Quoted 10 lines · Jing Zhang

With per guest ID registers, ID_AA64PFR0_EL1.[CSV2|CSV3] settings from userspace can be stored in its corresponding ID register.

The setting of CSV bits for protected VMs are removed according to the discussion from Fuad below:
https://lore.kernel.org/all/CA+EHjTwXA9TprX4jeG+-D+c8v9XG+oFdU1o6TSkvVye145_OvA@mail.gmail.com

Besides the removal of CSV bits setting for protected VMs, No other functional change intended.

One thing that you don't mention is the addition of some locking, which is a pretty significant change.

Quoted 72 lines
---
 arch/arm64/include/asm/kvm_host.h |  2 --
 arch/arm64/kvm/arm.c              | 17 ---------
 arch/arm64/kvm/sys_regs.c         | 58 +++++++++++++++++++++++++------
 3 files changed, 47 insertions(+), 30 deletions(-)
arch/arm64/include/asm/kvm_host.h +0 -2
index 069606170c82..8a2fde6c04c4 100644
@@ -257,8 +257,6 @@ struct kvm_arch {
 
 	cpumask_var_t supported_cpus;
 
-	u8 pfr0_csv2;
-	u8 pfr0_csv3;
 	struct {
 		u8 imp:4;
 		u8 unimp:4;
arch/arm64/kvm/arm.c +0 -17
index 774656a0718d..5114521ace60 100644
@@ -102,22 +102,6 @@ static int kvm_arm_default_max_vcpus(void)
 	return vgic_present ? kvm_vgic_get_max_vcpus() : KVM_MAX_VCPUS;
 }
 
-static void set_default_spectre(struct kvm *kvm)
-{
-	/*
-	 * The default is to expose CSV2 == 1 if the HW isn't affected.
-	 * Although this is a per-CPU feature, we make it global because
-	 * asymmetric systems are just a nuisance.
-	 *
-	 * Userspace can override this as long as it doesn't promise
-	 * the impossible.
-	 */
-	if (arm64_get_spectre_v2_state() == SPECTRE_UNAFFECTED)
-		kvm->arch.pfr0_csv2 = 1;
-	if (arm64_get_meltdown_state() == SPECTRE_UNAFFECTED)
-		kvm->arch.pfr0_csv3 = 1;
-}
-
 /**
  * kvm_arch_init_vm - initializes a VM data structure
  * @kvm:	pointer to the KVM struct
@@ -161,7 +145,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	/* The maximum number of VCPUs is limited by the host's GIC model */
 	kvm->max_vcpus = kvm_arm_default_max_vcpus();
 
-	set_default_spectre(kvm);
 	kvm_arm_init_hypercalls(kvm);
 	kvm_arm_init_id_regs(kvm);
 
arch/arm64/kvm/sys_regs.c +0 -4
index d2ee3a1c7f03..9fb1c2f8f5a5 100644
@@ -1218,10 +1218,6 @@ static u64 kvm_arm_read_id_reg(const struct kvm_vcpu *vcpu, u32 id)
 		if (!vcpu_has_sve(vcpu))
 			val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_SVE);
 		val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_AMU);
-		val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2);
-		val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2), (u64)vcpu->kvm->arch.pfr0_csv2);
-		val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3);
-		val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3), (u64)vcpu->kvm->arch.pfr0_csv3);
 		if (kvm_vgic_global_state.type == VGIC_V3) {
 			val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_GIC);
 			val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_GIC), 1);

arch/arm64/kvm/sys_regs.c static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu, · line 1358

Quoted 4 lines
arch/arm64/kvm/sys_regs.c +1 -0
@@ -1359,7 +1355,11 @@ static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu,
 			       const struct sys_reg_desc *rd,
 			       u64 val)
 {
+	struct kvm_arch *arch = &vcpu->kvm->arch;

The use of kvm_arch as an anchor is very non-idiomatic. Use the kvm pointer for this if you must, but I'd rather you spell the whole thing out.

Quoted 22 lines
arch/arm64/kvm/sys_regs.c +5 -6
+	u64 old_val = read_id_reg(vcpu, rd);
+	u64 new_val = val;
 	u8 csv2, csv3;
+	int ret = 0;
 
 	/*
 	 * Allow AA64PFR0_EL1.CSV2 to be set from userspace as long as
@@ -1377,17 +1377,26 @@ static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu,
 	    (csv3 && arm64_get_meltdown_state() != SPECTRE_UNAFFECTED))
 		return -EINVAL;
 
+	mutex_lock(&arch->config_lock);
 	/* We can only differ with CSV[23], and anything else is an error */
-	val ^= read_id_reg(vcpu, rd);
+	val ^= old_val;
 	val &= ~(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2) |
 		 ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3));
-	if (val)
-		return -EINVAL;
-
-	vcpu->kvm->arch.pfr0_csv2 = csv2;
-	vcpu->kvm->arch.pfr0_csv3 = csv3;

arch/arm64/kvm/sys_regs.c static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu, · line 1393

Quoted 10 lines
arch/arm64/kvm/sys_regs.c +8 -1
+	if (val) {
+		ret = -EINVAL;
+		goto out;
+	}
 
-	return 0;
+	/* Only allow userspace to change the idregs before VM running */
+	if (kvm_vm_has_ran_once(vcpu->kvm)) {
+		if (new_val != old_val)
+			ret = -EBUSY;

This sort of check should be done exactly once in a central spot. For similar reasons, the config_lock should be take in a unique location so that we can actually reason about this globally rather than at a microscopic level.

Something like this (which applies to the full series):

arch/arm64/kvm/sys_regs.c +24 -36
index b3eacfc592eb..e184b9350166 100644
@@ -1534,7 +1534,6 @@ static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
 			       const struct sys_reg_desc *rd,
 			       u64 val)
 {
-	struct kvm_arch *arch = &vcpu->kvm->arch;
 	u8 pmuver, host_pmuver;
 	bool valid_pmu;
 	int ret = 0;
@@ -1557,14 +1556,6 @@ static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
 	if (kvm_vcpu_has_pmu(vcpu) != valid_pmu)
 		return -EINVAL;
 
-	mutex_lock(&arch->config_lock);
-	/* Only allow userspace to change the idregs before VM running */
-	if (kvm_vm_has_ran_once(vcpu->kvm)) {
-		if (val != read_id_reg(vcpu, rd))
-			ret = -EBUSY;
-		goto out;
-	}
-
 	if (!valid_pmu) {
 		/*
 		 * Ignore the PMUVer field in @val. The PMUVer would be determined
@@ -1592,7 +1583,6 @@ static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
 			   pmuver == ID_AA64DFR0_EL1_PMUVer_IMP_DEF);
 
 out:
-	mutex_unlock(&arch->config_lock);
 	return ret;
 }
 
@@ -1617,7 +1607,6 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
 			   const struct sys_reg_desc *rd,
 			   u64 val)
 {
-	struct kvm_arch *arch = &vcpu->kvm->arch;
 	u8 perfmon, host_perfmon;
 	bool valid_pmu;
 	int ret = 0;
@@ -1641,14 +1630,6 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
 	if (kvm_vcpu_has_pmu(vcpu) != valid_pmu)
 		return -EINVAL;
 
-	mutex_lock(&arch->config_lock);
-	/* Only allow userspace to change the idregs before VM running */
-	if (kvm_vm_has_ran_once(vcpu->kvm)) {
-		if (val != read_id_reg(vcpu, rd))
-			ret = -EBUSY;
-		goto out;
-	}
-
 	if (!valid_pmu) {
 		/*
 		 * Ignore the PerfMon field in @val. The PerfMon would be determined
@@ -1676,7 +1657,6 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
 			   perfmon == ID_DFR0_EL1_PerfMon_IMPDEF);
 
 out:
-	mutex_unlock(&arch->config_lock);
 	return ret;
 }
 
@@ -1690,11 +1670,7 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
 static int get_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
 		      u64 *val)
 {
-	struct kvm_arch *arch = &vcpu->kvm->arch;
-
-	mutex_lock(&arch->config_lock);
 	*val = read_id_reg(vcpu, rd);
-	mutex_unlock(&arch->config_lock);
 
 	return 0;
 }
@@ -1702,21 +1678,12 @@ static int get_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
 static int set_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
 		      u64 val)
 {
-	struct kvm_arch *arch = &vcpu->kvm->arch;
 	u32 id = reg_to_encoding(rd);
 	int ret = 0;
 
-	mutex_lock(&arch->config_lock);
-	/* Only allow userspace to change the idregs before VM running */
-	if (kvm_vm_has_ran_once(vcpu->kvm)) {
-		if (val != read_id_reg(vcpu, rd))
-			ret = -EBUSY;
-	} else {
-		ret = arm64_check_features(vcpu, rd, val);
-		if (!ret)
-			IDREG(vcpu->kvm, id) = val;
-	}
-	mutex_unlock(&arch->config_lock);
+	ret = arm64_check_features(vcpu, rd, val);
+	if (!ret)
+		IDREG(vcpu->kvm, id) = val;
 
 	return ret;
 }
@@ -3438,6 +3405,9 @@ int kvm_sys_reg_get_user(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg,
 	if (!r || sysreg_hidden_user(vcpu, r))
 		return -ENOENT;
 
+	if (is_id_reg(reg_to_encoding(r)))
+		mutex_lock(&vcpu->kvm->arch.config_lock);
+
 	if (r->get_user) {
 		ret = (r->get_user)(vcpu, r, &val);
 	} else {
@@ -3445,6 +3415,9 @@ int kvm_sys_reg_get_user(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg,
 		ret = 0;
 	}
 
+	if (is_id_reg(reg_to_encoding(r)))
+		mutex_unlock(&vcpu->kvm->arch.config_lock);
+
 	if (!ret)
 		ret = put_user(val, uaddr);
 
@@ -3482,9 +3455,21 @@ int kvm_sys_reg_set_user(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg,
 	if (!r || sysreg_hidden_user(vcpu, r))
 		return -ENOENT;
 
+	/* Only allow userspace to change the idregs before VM running */
+	if (is_id_reg(reg_to_encoding(r)) &&
+	    kvm_vm_has_ran_once(vcpu->kvm)) {
+		if (val == read_id_reg(vcpu, r))
+			return 0;
+		return -EBUSY;
+	}
+
 	if (sysreg_user_write_ignore(vcpu, r))
 		return 0;
 
+	/* ID regs are global to the VM and cannot be updated concurrently */
+	if (is_id_reg(reg_to_encoding(r)))
+		mutex_lock(&vcpu->kvm->arch.config_lock);
+
 	if (r->set_user) {
 		ret = (r->set_user)(vcpu, r, val);
 	} else {
@@ -3492,6 +3477,9 @@ int kvm_sys_reg_set_user(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg,
 		ret = 0;
 	}
 
+	if (is_id_reg(reg_to_encoding(r)))
+		mutex_unlock(&vcpu->kvm->arch.config_lock);
+
 	return ret;
 }
 

and you can then restore the code to its original shape, as there is no need to change the control flow anymore.

arch/arm64/kvm/sys_regs.c static int set_id_dfr0_el1(struct kvm_vcpu *vcpu, · line 1490

Quoted 13 lines
arch/arm64/kvm/sys_regs.c +6 -0
+	} else {
+		IDREG(vcpu->kvm, reg_to_encoding(rd)) = new_val;
+	}
+out:
+	mutex_unlock(&arch->config_lock);
+	return ret;
 }
 
 static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
@@ -1479,7 +1488,12 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
 static int get_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
 		      u64 *val)
 {

Right above this function is a comment that says the idreg are immutable. Time to revisit it?

Quoted 35 lines
arch/arm64/kvm/sys_regs.c +20 -0
+	struct kvm_arch *arch = &vcpu->kvm->arch;
+
+	mutex_lock(&arch->config_lock);
 	*val = read_id_reg(vcpu, rd);
+	mutex_unlock(&arch->config_lock);
+
 	return 0;
 }
 
@@ -3364,6 +3378,7 @@ void kvm_arm_init_id_regs(struct kvm *kvm)
 {
 	const struct sys_reg_desc *idreg;
 	struct sys_reg_params params;
+	u64 val;
 	u32 id;
 
 	/* Find the first idreg (SYS_ID_PFR0_EL1) in sys_reg_descs. */
@@ -3386,6 +3401,27 @@ void kvm_arm_init_id_regs(struct kvm *kvm)
 		idreg++;
 		id = reg_to_encoding(idreg);
 	}
+
+	/*
+	 * The default is to expose CSV2 == 1 if the HW isn't affected.
+	 * Although this is a per-CPU feature, we make it global because
+	 * asymmetric systems are just a nuisance.
+	 *
+	 * Userspace can override this as long as it doesn't promise
+	 * the impossible.
+	 */
+	val = IDREG(kvm, SYS_ID_AA64PFR0_EL1);
+
+	if (arm64_get_spectre_v2_state() == SPECTRE_UNAFFECTED) {
+		val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2);
+		val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2), 1);

arch/arm64/kvm/sys_regs.c void kvm_arm_init_id_regs(struct kvm *kvm) · line 3427

Quoted 10 lines
arch/arm64/kvm/sys_regs.c +7 -0
+	}
+	if (arm64_get_meltdown_state() == SPECTRE_UNAFFECTED) {
+		val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3);
+		val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3), 1);
+	}
+
+	IDREG(kvm, SYS_ID_AA64PFR0_EL1) = val;
 }
 
 int __init kvm_sys_reg_table_init(void)

Thanks,

	M.
Signature · 1 line
Without deviation from the norm, progress is not possible.
author
#8 lore raw

Re: [PATCH v10 2/5] KVM: arm64: Use per guest ID register for ID_AA64PFR0_EL1.[CSV2|CSV3]

Quoted 11 lines · Marc Zyngier
Quoted 8 lines · Jing Zhang

With per guest ID registers, ID_AA64PFR0_EL1.[CSV2|CSV3] settings from userspace can be stored in its corresponding ID register.

The setting of CSV bits for protected VMs are removed according to the discussion from Fuad below:
https://lore.kernel.org/all/CA+EHjTwXA9TprX4jeG+-D+c8v9XG+oFdU1o6TSkvVye145_OvA@mail.gmail.com

Quoted 5 lines
Quoted 2 lines

Besides the removal of CSV bits setting for protected VMs, No other functional change intended.

One thing that you don't mention is the addition of some locking, which is a pretty significant change.

Will update the commit message.

Quoted 73 lines
Quoted 73 lines
---
 arch/arm64/include/asm/kvm_host.h |  2 --
 arch/arm64/kvm/arm.c              | 17 ---------
 arch/arm64/kvm/sys_regs.c         | 58 +++++++++++++++++++++++++------
 3 files changed, 47 insertions(+), 30 deletions(-)
arch/arm64/include/asm/kvm_host.h +0 -2
index 069606170c82..8a2fde6c04c4 100644
@@ -257,8 +257,6 @@ struct kvm_arch {
 
      cpumask_var_t supported_cpus;
 
-     u8 pfr0_csv2;
-     u8 pfr0_csv3;
      struct {
              u8 imp:4;
              u8 unimp:4;
arch/arm64/kvm/arm.c +0 -17
index 774656a0718d..5114521ace60 100644
@@ -102,22 +102,6 @@ static int kvm_arm_default_max_vcpus(void)
      return vgic_present ? kvm_vgic_get_max_vcpus() : KVM_MAX_VCPUS;
 }
 
-static void set_default_spectre(struct kvm *kvm)
-{
-     /*
-      * The default is to expose CSV2 == 1 if the HW isn't affected.
-      * Although this is a per-CPU feature, we make it global because
-      * asymmetric systems are just a nuisance.
-      *
-      * Userspace can override this as long as it doesn't promise
-      * the impossible.
-      */
-     if (arm64_get_spectre_v2_state() == SPECTRE_UNAFFECTED)
-             kvm->arch.pfr0_csv2 = 1;
-     if (arm64_get_meltdown_state() == SPECTRE_UNAFFECTED)
-             kvm->arch.pfr0_csv3 = 1;
-}
-
 /**
  * kvm_arch_init_vm - initializes a VM data structure
  * @kvm:     pointer to the KVM struct
@@ -161,7 +145,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
      /* The maximum number of VCPUs is limited by the host's GIC model */
      kvm->max_vcpus = kvm_arm_default_max_vcpus();
 
-     set_default_spectre(kvm);
      kvm_arm_init_hypercalls(kvm);
      kvm_arm_init_id_regs(kvm);
 
arch/arm64/kvm/sys_regs.c +0 -4
index d2ee3a1c7f03..9fb1c2f8f5a5 100644
@@ -1218,10 +1218,6 @@ static u64 kvm_arm_read_id_reg(const struct kvm_vcpu *vcpu, u32 id)
              if (!vcpu_has_sve(vcpu))
                      val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_SVE);
              val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_AMU);
-             val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2);
-             val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2), (u64)vcpu->kvm->arch.pfr0_csv2);
-             val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3);
-             val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3), (u64)vcpu->kvm->arch.pfr0_csv3);
              if (kvm_vgic_global_state.type == VGIC_V3) {
                      val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_GIC);
                      val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_GIC), 1);
@@ -1359,7 +1355,11 @@ static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu,
Quoted 8 lines

arch/arm64/kvm/sys_regs.c static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,

Quoted 4 lines
arch/arm64/kvm/sys_regs.c +1 -0
                             const struct sys_reg_desc *rd,
                             u64 val)
 {
+     struct kvm_arch *arch = &vcpu->kvm->arch;

The use of kvm_arch as an anchor is very non-idiomatic. Use the kvm pointer for this if you must, but I'd rather you spell the whole thing out.

Sure, will spell the whole thing out.

Quoted 196 lines
Quoted 23 lines
arch/arm64/kvm/sys_regs.c +5 -6
+     u64 old_val = read_id_reg(vcpu, rd);
+     u64 new_val = val;
      u8 csv2, csv3;
+     int ret = 0;
 
      /*
       * Allow AA64PFR0_EL1.CSV2 to be set from userspace as long as
@@ -1377,17 +1377,26 @@ static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu,
          (csv3 && arm64_get_meltdown_state() != SPECTRE_UNAFFECTED))
              return -EINVAL;
 
+     mutex_lock(&arch->config_lock);
      /* We can only differ with CSV[23], and anything else is an error */
-     val ^= read_id_reg(vcpu, rd);
+     val ^= old_val;
      val &= ~(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2) |
               ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3));
-     if (val)
-             return -EINVAL;
-
-     vcpu->kvm->arch.pfr0_csv2 = csv2;
-     vcpu->kvm->arch.pfr0_csv3 = csv3;

arch/arm64/kvm/sys_regs.c static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu, · line 1393

Quoted 10 lines
arch/arm64/kvm/sys_regs.c +8 -1
+     if (val) {
+             ret = -EINVAL;
+             goto out;
+     }
 
-     return 0;
+     /* Only allow userspace to change the idregs before VM running */
+     if (kvm_vm_has_ran_once(vcpu->kvm)) {
+             if (new_val != old_val)
+                     ret = -EBUSY;

This sort of check should be done exactly once in a central spot. For similar reasons, the config_lock should be take in a unique location so that we can actually reason about this globally rather than at a microscopic level.

Something like this (which applies to the full series):

arch/arm64/kvm/sys_regs.c +24 -36
index b3eacfc592eb..e184b9350166 100644
@@ -1534,7 +1534,6 @@ static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
                               const struct sys_reg_desc *rd,
                               u64 val)
 {
-       struct kvm_arch *arch = &vcpu->kvm->arch;
        u8 pmuver, host_pmuver;
        bool valid_pmu;
        int ret = 0;
@@ -1557,14 +1556,6 @@ static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
        if (kvm_vcpu_has_pmu(vcpu) != valid_pmu)
                return -EINVAL;
 
-       mutex_lock(&arch->config_lock);
-       /* Only allow userspace to change the idregs before VM running */
-       if (kvm_vm_has_ran_once(vcpu->kvm)) {
-               if (val != read_id_reg(vcpu, rd))
-                       ret = -EBUSY;
-               goto out;
-       }
-
        if (!valid_pmu) {
                /*
                 * Ignore the PMUVer field in @val. The PMUVer would be determined
@@ -1592,7 +1583,6 @@ static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
                           pmuver == ID_AA64DFR0_EL1_PMUVer_IMP_DEF);
 
 out:
-       mutex_unlock(&arch->config_lock);
        return ret;
 }
 
@@ -1617,7 +1607,6 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
                           const struct sys_reg_desc *rd,
                           u64 val)
 {
-       struct kvm_arch *arch = &vcpu->kvm->arch;
        u8 perfmon, host_perfmon;
        bool valid_pmu;
        int ret = 0;
@@ -1641,14 +1630,6 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
        if (kvm_vcpu_has_pmu(vcpu) != valid_pmu)
                return -EINVAL;
 
-       mutex_lock(&arch->config_lock);
-       /* Only allow userspace to change the idregs before VM running */
-       if (kvm_vm_has_ran_once(vcpu->kvm)) {
-               if (val != read_id_reg(vcpu, rd))
-                       ret = -EBUSY;
-               goto out;
-       }
-
        if (!valid_pmu) {
                /*
                 * Ignore the PerfMon field in @val. The PerfMon would be determined
@@ -1676,7 +1657,6 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
                           perfmon == ID_DFR0_EL1_PerfMon_IMPDEF);
 
 out:
-       mutex_unlock(&arch->config_lock);
        return ret;
 }
 
@@ -1690,11 +1670,7 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
 static int get_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
                      u64 *val)
 {
-       struct kvm_arch *arch = &vcpu->kvm->arch;
-
-       mutex_lock(&arch->config_lock);
        *val = read_id_reg(vcpu, rd);
-       mutex_unlock(&arch->config_lock);
 
        return 0;
 }
@@ -1702,21 +1678,12 @@ static int get_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
 static int set_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
                      u64 val)
 {
-       struct kvm_arch *arch = &vcpu->kvm->arch;
        u32 id = reg_to_encoding(rd);
        int ret = 0;
 
-       mutex_lock(&arch->config_lock);
-       /* Only allow userspace to change the idregs before VM running */
-       if (kvm_vm_has_ran_once(vcpu->kvm)) {
-               if (val != read_id_reg(vcpu, rd))
-                       ret = -EBUSY;
-       } else {
-               ret = arm64_check_features(vcpu, rd, val);
-               if (!ret)
-                       IDREG(vcpu->kvm, id) = val;
-       }
-       mutex_unlock(&arch->config_lock);
+       ret = arm64_check_features(vcpu, rd, val);
+       if (!ret)
+               IDREG(vcpu->kvm, id) = val;
 
        return ret;
 }
@@ -3438,6 +3405,9 @@ int kvm_sys_reg_get_user(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg,
        if (!r || sysreg_hidden_user(vcpu, r))
                return -ENOENT;
 
+       if (is_id_reg(reg_to_encoding(r)))
+               mutex_lock(&vcpu->kvm->arch.config_lock);
+
        if (r->get_user) {
                ret = (r->get_user)(vcpu, r, &val);
        } else {
@@ -3445,6 +3415,9 @@ int kvm_sys_reg_get_user(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg,
                ret = 0;
        }
 
+       if (is_id_reg(reg_to_encoding(r)))
+               mutex_unlock(&vcpu->kvm->arch.config_lock);
+
        if (!ret)
                ret = put_user(val, uaddr);
 
@@ -3482,9 +3455,21 @@ int kvm_sys_reg_set_user(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg,
        if (!r || sysreg_hidden_user(vcpu, r))
                return -ENOENT;
 
+       /* Only allow userspace to change the idregs before VM running */
+       if (is_id_reg(reg_to_encoding(r)) &&
+           kvm_vm_has_ran_once(vcpu->kvm)) {
+               if (val == read_id_reg(vcpu, r))
+                       return 0;
+               return -EBUSY;
+       }
+
        if (sysreg_user_write_ignore(vcpu, r))
                return 0;
 
+       /* ID regs are global to the VM and cannot be updated concurrently */
+       if (is_id_reg(reg_to_encoding(r)))
+               mutex_lock(&vcpu->kvm->arch.config_lock);
+
        if (r->set_user) {
                ret = (r->set_user)(vcpu, r, val);
        } else {
@@ -3492,6 +3477,9 @@ int kvm_sys_reg_set_user(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg,
                ret = 0;
        }
 
+       if (is_id_reg(reg_to_encoding(r)))
+               mutex_unlock(&vcpu->kvm->arch.config_lock);
+
        return ret;
 }
 
Quoted 2 lines

and you can then restore the code to its original shape, as there is no need to change the control flow anymore.

Thanks. Will do as you suggested.

Quoted 9 lines
Quoted 9 lines
arch/arm64/kvm/sys_regs.c +6 -0
+     } else {
+             IDREG(vcpu->kvm, reg_to_encoding(rd)) = new_val;
+     }
+out:
+     mutex_unlock(&arch->config_lock);
+     return ret;
 }
 
Quoted 8 lines

arch/arm64/kvm/sys_regs.c static int set_id_dfr0_el1(struct kvm_vcpu *vcpu, · line 1490

Quoted 5 lines
arch/arm64/kvm/sys_regs.c +0 -0
 static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
@@ -1479,7 +1488,12 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
 static int get_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
                    u64 *val)
 {

Right above this function is a comment that says the idreg are immutable. Time to revisit it?

Will update the comment.

Quoted 45 lines
Quoted 45 lines
arch/arm64/kvm/sys_regs.c +27 -0
+     struct kvm_arch *arch = &vcpu->kvm->arch;
+
+     mutex_lock(&arch->config_lock);
      *val = read_id_reg(vcpu, rd);
+     mutex_unlock(&arch->config_lock);
+
      return 0;
 }
 
@@ -3364,6 +3378,7 @@ void kvm_arm_init_id_regs(struct kvm *kvm)
 {
      const struct sys_reg_desc *idreg;
      struct sys_reg_params params;
+     u64 val;
      u32 id;
 
      /* Find the first idreg (SYS_ID_PFR0_EL1) in sys_reg_descs. */
@@ -3386,6 +3401,27 @@ void kvm_arm_init_id_regs(struct kvm *kvm)
              idreg++;
              id = reg_to_encoding(idreg);
      }
+
+     /*
+      * The default is to expose CSV2 == 1 if the HW isn't affected.
+      * Although this is a per-CPU feature, we make it global because
+      * asymmetric systems are just a nuisance.
+      *
+      * Userspace can override this as long as it doesn't promise
+      * the impossible.
+      */
+     val = IDREG(kvm, SYS_ID_AA64PFR0_EL1);
+
+     if (arm64_get_spectre_v2_state() == SPECTRE_UNAFFECTED) {
+             val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2);
+             val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV2), 1);
+     }
+     if (arm64_get_meltdown_state() == SPECTRE_UNAFFECTED) {
+             val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3);
+             val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1_CSV3), 1);
+     }
+
+     IDREG(kvm, SYS_ID_AA64PFR0_EL1) = val;
 }
 
Quoted 8 lines

arch/arm64/kvm/sys_regs.c void kvm_arm_init_id_regs(struct kvm *kvm) · line 3427

Quoted 1 line
arch/arm64/kvm/sys_regs.c +0 -0
 int __init kvm_sys_reg_table_init(void)

Thanks,

        M.

--
Without deviation from the norm, progress is not possible.

Thanks,
Jing

author
#9 lore raw

PATCH v10 3/5 KVM: arm64: Use per guest ID register for ID_AA64DFR0_EL1.PMUVer

In reply to Jing Zhang

With per guest ID registers, PMUver settings from userspace can be stored in its corresponding ID register.

No functional change intended.

---
 arch/arm64/include/asm/kvm_host.h |  12 ++--
 arch/arm64/kvm/arm.c              |   6 --
 arch/arm64/kvm/sys_regs.c         | 100 ++++++++++++++++++++++++------
 include/kvm/arm_pmu.h             |   5 +-
 4 files changed, 92 insertions(+), 31 deletions(-)
arch/arm64/include/asm/kvm_host.h +7 -5
index 8a2fde6c04c4..7b0f43373dbe 100644
@@ -246,6 +246,13 @@ struct kvm_arch {
 #define KVM_ARCH_FLAG_TIMER_PPIS_IMMUTABLE		7
 	/* SMCCC filter initialized for the VM */
 #define KVM_ARCH_FLAG_SMCCC_FILTER_CONFIGURED		8
+	/*
+	 * AA64DFR0_EL1.PMUver was set as ID_AA64DFR0_EL1_PMUVer_IMP_DEF
+	 * or DFR0_EL1.PerfMon was set as ID_DFR0_EL1_PerfMon_IMPDEF from
+	 * userspace for VCPUs without PMU.
+	 */
+#define KVM_ARCH_FLAG_VCPU_HAS_IMP_DEF_PMU		9
+
 	unsigned long flags;
 
 	/*
@@ -257,11 +264,6 @@ struct kvm_arch {
 
 	cpumask_var_t supported_cpus;
 
-	struct {
-		u8 imp:4;
-		u8 unimp:4;
-	} dfr0_pmuver;
-
 	/* Hypercall features firmware registers' descriptor */
 	struct kvm_smccc_features smccc_feat;
 	struct maple_tree smccc_filter;
arch/arm64/kvm/arm.c +0 -6
index 5114521ace60..ca18c09ccf82 100644
@@ -148,12 +148,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	kvm_arm_init_hypercalls(kvm);
 	kvm_arm_init_id_regs(kvm);
 
-	/*
-	 * Initialise the default PMUver before there is a chance to
-	 * create an actual PMU.
-	 */
-	kvm->arch.dfr0_pmuver.imp = kvm_arm_pmu_get_pmuver_limit();
-
 	return 0;
 
 err_free_cpumask:
arch/arm64/kvm/sys_regs.c +82 -18

149 lines not on this page · raw message

include/kvm/arm_pmu.h +3 -2
index 1a6a695ca67a..8d70dbdc1e0a 100644
@@ -92,8 +92,9 @@ void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu);
 /*
  * Evaluates as true when emulating PMUv3p5, and false otherwise.
  */
-#define kvm_pmu_is_3p5(vcpu)						\
-	(vcpu->kvm->arch.dfr0_pmuver.imp >= ID_AA64DFR0_EL1_PMUVer_V3P5)
+#define kvm_pmu_is_3p5(vcpu)									\
+	 (FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer),					\
+		    IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1)) >= ID_AA64DFR0_EL1_PMUVer_V3P5)
 
 u8 kvm_arm_pmu_get_pmuver_limit(void);
 
Signature · 1 line
2.40.1.698.g37aff9b760-goog
maintainer
#10 lore raw

Re: [PATCH v10 3/5] KVM: arm64: Use per guest ID register for ID_AA64DFR0_EL1.PMUVer

Quoted 183 lines · Jing Zhang

With per guest ID registers, PMUver settings from userspace can be stored in its corresponding ID register.

No functional change intended.

---
 arch/arm64/include/asm/kvm_host.h |  12 ++--
 arch/arm64/kvm/arm.c              |   6 --
 arch/arm64/kvm/sys_regs.c         | 100 ++++++++++++++++++++++++------
 include/kvm/arm_pmu.h             |   5 +-
 4 files changed, 92 insertions(+), 31 deletions(-)
arch/arm64/include/asm/kvm_host.h +7 -5
index 8a2fde6c04c4..7b0f43373dbe 100644
@@ -246,6 +246,13 @@ struct kvm_arch {
 #define KVM_ARCH_FLAG_TIMER_PPIS_IMMUTABLE		7
 	/* SMCCC filter initialized for the VM */
 #define KVM_ARCH_FLAG_SMCCC_FILTER_CONFIGURED		8
+	/*
+	 * AA64DFR0_EL1.PMUver was set as ID_AA64DFR0_EL1_PMUVer_IMP_DEF
+	 * or DFR0_EL1.PerfMon was set as ID_DFR0_EL1_PerfMon_IMPDEF from
+	 * userspace for VCPUs without PMU.
+	 */
+#define KVM_ARCH_FLAG_VCPU_HAS_IMP_DEF_PMU		9
+
 	unsigned long flags;
 
 	/*
@@ -257,11 +264,6 @@ struct kvm_arch {
 
 	cpumask_var_t supported_cpus;
 
-	struct {
-		u8 imp:4;
-		u8 unimp:4;
-	} dfr0_pmuver;
-
 	/* Hypercall features firmware registers' descriptor */
 	struct kvm_smccc_features smccc_feat;
 	struct maple_tree smccc_filter;
arch/arm64/kvm/arm.c +0 -6
index 5114521ace60..ca18c09ccf82 100644
@@ -148,12 +148,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	kvm_arm_init_hypercalls(kvm);
 	kvm_arm_init_id_regs(kvm);
 
-	/*
-	 * Initialise the default PMUver before there is a chance to
-	 * create an actual PMU.
-	 */
-	kvm->arch.dfr0_pmuver.imp = kvm_arm_pmu_get_pmuver_limit();
-
 	return 0;
 
 err_free_cpumask:
arch/arm64/kvm/sys_regs.c +58 -17
index 9fb1c2f8f5a5..84d9e4baa4f8 100644
@@ -1178,9 +1178,12 @@ static bool access_arch_timer(struct kvm_vcpu *vcpu,
 static u8 vcpu_pmuver(const struct kvm_vcpu *vcpu)
 {
 	if (kvm_vcpu_has_pmu(vcpu))
-		return vcpu->kvm->arch.dfr0_pmuver.imp;
+		return FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer),
+				 IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1));
+	else if (test_bit(KVM_ARCH_FLAG_VCPU_HAS_IMP_DEF_PMU, &vcpu->kvm->arch.flags))
+		return ID_AA64DFR0_EL1_PMUVer_IMP_DEF;
 
-	return vcpu->kvm->arch.dfr0_pmuver.unimp;
+	return 0;
 }
 
 static u8 perfmon_to_pmuver(u8 perfmon)
@@ -1403,8 +1406,12 @@ static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
 			       const struct sys_reg_desc *rd,
 			       u64 val)
 {
+	struct kvm_arch *arch = &vcpu->kvm->arch;
+	u64 old_val = read_id_reg(vcpu, rd);
 	u8 pmuver, host_pmuver;
+	u64 new_val = val;
 	bool valid_pmu;
+	int ret = 0;
 
 	host_pmuver = kvm_arm_pmu_get_pmuver_limit();
 
@@ -1424,26 +1431,51 @@ static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
 	if (kvm_vcpu_has_pmu(vcpu) != valid_pmu)
 		return -EINVAL;
 
+	mutex_lock(&arch->config_lock);
 	/* We can only differ with PMUver, and anything else is an error */
-	val ^= read_id_reg(vcpu, rd);
+	val ^= old_val;
 	val &= ~ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer);
-	if (val)
-		return -EINVAL;
+	if (val) {
+		ret = -EINVAL;
+		goto out;
+	}
 
-	if (valid_pmu)
-		vcpu->kvm->arch.dfr0_pmuver.imp = pmuver;
-	else
-		vcpu->kvm->arch.dfr0_pmuver.unimp = pmuver;
+	/* Only allow userspace to change the idregs before VM running */
+	if (kvm_vm_has_ran_once(vcpu->kvm)) {
+		if (new_val != old_val)
+			ret = -EBUSY;
+	} else {
+		if (valid_pmu) {
+			val = IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1);
+			val &= ~ID_AA64DFR0_EL1_PMUVer_MASK;
+			val |= FIELD_PREP(ID_AA64DFR0_EL1_PMUVer_MASK, pmuver);
+			IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1) = val;
+
+			val = IDREG(vcpu->kvm, SYS_ID_DFR0_EL1);
+			val &= ~ID_DFR0_EL1_PerfMon_MASK;
+			val |= FIELD_PREP(ID_DFR0_EL1_PerfMon_MASK, pmuver_to_perfmon(pmuver));
+			IDREG(vcpu->kvm, SYS_ID_DFR0_EL1) = val;
+		} else {
+			assign_bit(KVM_ARCH_FLAG_VCPU_HAS_IMP_DEF_PMU, &vcpu->kvm->arch.flags,
+				   pmuver == ID_AA64DFR0_EL1_PMUVer_IMP_DEF);
+		}
+	}
 
-	return 0;
+out:
+	mutex_unlock(&arch->config_lock);
+	return ret;
 }
 
 static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
 			   const struct sys_reg_desc *rd,
 			   u64 val)
 {
+	struct kvm_arch *arch = &vcpu->kvm->arch;
+	u64 old_val = read_id_reg(vcpu, rd);
 	u8 perfmon, host_perfmon;
+	u64 new_val = val;
 	bool valid_pmu;
+	int ret = 0;
 
 	host_perfmon = pmuver_to_perfmon(kvm_arm_pmu_get_pmuver_limit());
 
@@ -1464,18 +1496,39 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
 	if (kvm_vcpu_has_pmu(vcpu) != valid_pmu)
 		return -EINVAL;
 
+	mutex_lock(&arch->config_lock);
 	/* We can only differ with PerfMon, and anything else is an error */
-	val ^= read_id_reg(vcpu, rd);
+	val ^= old_val;
 	val &= ~ARM64_FEATURE_MASK(ID_DFR0_EL1_PerfMon);
-	if (val)
-		return -EINVAL;
+	if (val) {
+		ret = -EINVAL;
+		goto out;
+	}
 
-	if (valid_pmu)
-		vcpu->kvm->arch.dfr0_pmuver.imp = perfmon_to_pmuver(perfmon);
-	else
-		vcpu->kvm->arch.dfr0_pmuver.unimp = perfmon_to_pmuver(perfmon);
+	/* Only allow userspace to change the idregs before VM running */
+	if (kvm_vm_has_ran_once(vcpu->kvm)) {
+		if (new_val != old_val)
+			ret = -EBUSY;
+	} else {
+		if (valid_pmu) {
+			val = IDREG(vcpu->kvm, SYS_ID_DFR0_EL1);
+			val &= ~ID_DFR0_EL1_PerfMon_MASK;
+			val |= FIELD_PREP(ID_DFR0_EL1_PerfMon_MASK, perfmon);
+			IDREG(vcpu->kvm, SYS_ID_DFR0_EL1) = val;

arch/arm64/kvm/sys_regs.c static int set_id_dfr0_el1(struct kvm_vcpu *vcpu, · line 1527

Quoted 10 lines
arch/arm64/kvm/sys_regs.c +10 -0
+
+			val = IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1);
+			val &= ~ID_AA64DFR0_EL1_PMUVer_MASK;
+			val |= FIELD_PREP(ID_AA64DFR0_EL1_PMUVer_MASK, perfmon_to_pmuver(perfmon));
+			IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1) = val;
+		} else {
+			assign_bit(KVM_ARCH_FLAG_VCPU_HAS_IMP_DEF_PMU, &vcpu->kvm->arch.flags,
+				   perfmon == ID_DFR0_EL1_PerfMon_IMPDEF);
+		}
+	}

This is the exact same code as for aa64fdr0. Make it a helper, please.

Quoted 31 lines
arch/arm64/kvm/sys_regs.c +14 -1
-	return 0;
+out:
+	mutex_unlock(&arch->config_lock);
+	return ret;
 }
 
 /*
@@ -3422,6 +3475,17 @@ void kvm_arm_init_id_regs(struct kvm *kvm)
 	}
 
 	IDREG(kvm, SYS_ID_AA64PFR0_EL1) = val;
+	/*
+	 * Initialise the default PMUver before there is a chance to
+	 * create an actual PMU.
+	 */
+	val = IDREG(kvm, SYS_ID_AA64DFR0_EL1);
+
+	val &= ~ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer);
+	val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer),
+			  kvm_arm_pmu_get_pmuver_limit());
+
+	IDREG(kvm, SYS_ID_AA64DFR0_EL1) = val;
 }
 
 int __init kvm_sys_reg_table_init(void)
include/kvm/arm_pmu.h +0 -0
index 1a6a695ca67a..8d70dbdc1e0a 100644

include/kvm/arm_pmu.h void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu); · line 97

Quoted 8 lines
include/kvm/arm_pmu.h +3 -2
@@ -92,8 +92,9 @@ void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu);
 /*
  * Evaluates as true when emulating PMUv3p5, and false otherwise.
  */
-#define kvm_pmu_is_3p5(vcpu)						\
-	(vcpu->kvm->arch.dfr0_pmuver.imp >= ID_AA64DFR0_EL1_PMUVer_V3P5)
+#define kvm_pmu_is_3p5(vcpu)									\
+	 (FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer),					\
+		    IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1)) >= ID_AA64DFR0_EL1_PMUVer_V3P5)

This is getting unreadable. How about something like:

include/kvm/arm_pmu.h +7 -3
index 8d70dbdc1e0a..ecb55d87fa36 100644
@@ -92,9 +92,13 @@ void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu);
 /*
  * Evaluates as true when emulating PMUv3p5, and false otherwise.
  */
-#define kvm_pmu_is_3p5(vcpu)									\
-	 (FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer),					\
-		    IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1)) >= ID_AA64DFR0_EL1_PMUVer_V3P5)
+#define kvm_pmu_is_3p5(vcpu)	({					\
+	u64 val = IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1); 		\
+	u8 v;								\
+									\
+	v = FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer), val);	\
+	v >= ID_AA64DFR0_EL1_PMUVer_V3P5;				\
+})
 
 u8 kvm_arm_pmu_get_pmuver_limit(void);
 

Thanks,

	M.
Signature · 1 line
Without deviation from the norm, progress is not possible.
author
#11 lore raw

Re: [PATCH v10 3/5] KVM: arm64: Use per guest ID register for ID_AA64DFR0_EL1.PMUVer

Quoted 190 lines · Marc Zyngier
Quoted 187 lines · Jing Zhang

With per guest ID registers, PMUver settings from userspace can be stored in its corresponding ID register.

No functional change intended.

---
 arch/arm64/include/asm/kvm_host.h |  12 ++--
 arch/arm64/kvm/arm.c              |   6 --
 arch/arm64/kvm/sys_regs.c         | 100 ++++++++++++++++++++++++------
 include/kvm/arm_pmu.h             |   5 +-
 4 files changed, 92 insertions(+), 31 deletions(-)
arch/arm64/include/asm/kvm_host.h +7 -5
index 8a2fde6c04c4..7b0f43373dbe 100644
@@ -246,6 +246,13 @@ struct kvm_arch {
 #define KVM_ARCH_FLAG_TIMER_PPIS_IMMUTABLE           7
      /* SMCCC filter initialized for the VM */
 #define KVM_ARCH_FLAG_SMCCC_FILTER_CONFIGURED                8
+     /*
+      * AA64DFR0_EL1.PMUver was set as ID_AA64DFR0_EL1_PMUVer_IMP_DEF
+      * or DFR0_EL1.PerfMon was set as ID_DFR0_EL1_PerfMon_IMPDEF from
+      * userspace for VCPUs without PMU.
+      */
+#define KVM_ARCH_FLAG_VCPU_HAS_IMP_DEF_PMU           9
+
      unsigned long flags;
 
      /*
@@ -257,11 +264,6 @@ struct kvm_arch {
 
      cpumask_var_t supported_cpus;
 
-     struct {
-             u8 imp:4;
-             u8 unimp:4;
-     } dfr0_pmuver;
-
      /* Hypercall features firmware registers' descriptor */
      struct kvm_smccc_features smccc_feat;
      struct maple_tree smccc_filter;
arch/arm64/kvm/arm.c +0 -6
index 5114521ace60..ca18c09ccf82 100644
@@ -148,12 +148,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
      kvm_arm_init_hypercalls(kvm);
      kvm_arm_init_id_regs(kvm);
 
-     /*
-      * Initialise the default PMUver before there is a chance to
-      * create an actual PMU.
-      */
-     kvm->arch.dfr0_pmuver.imp = kvm_arm_pmu_get_pmuver_limit();
-
      return 0;
 
 err_free_cpumask:
arch/arm64/kvm/sys_regs.c +62 -17
index 9fb1c2f8f5a5..84d9e4baa4f8 100644
@@ -1178,9 +1178,12 @@ static bool access_arch_timer(struct kvm_vcpu *vcpu,
 static u8 vcpu_pmuver(const struct kvm_vcpu *vcpu)
 {
      if (kvm_vcpu_has_pmu(vcpu))
-             return vcpu->kvm->arch.dfr0_pmuver.imp;
+             return FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer),
+                              IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1));
+     else if (test_bit(KVM_ARCH_FLAG_VCPU_HAS_IMP_DEF_PMU, &vcpu->kvm->arch.flags))
+             return ID_AA64DFR0_EL1_PMUVer_IMP_DEF;
 
-     return vcpu->kvm->arch.dfr0_pmuver.unimp;
+     return 0;
 }
 
 static u8 perfmon_to_pmuver(u8 perfmon)
@@ -1403,8 +1406,12 @@ static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
                             const struct sys_reg_desc *rd,
                             u64 val)
 {
+     struct kvm_arch *arch = &vcpu->kvm->arch;
+     u64 old_val = read_id_reg(vcpu, rd);
      u8 pmuver, host_pmuver;
+     u64 new_val = val;
      bool valid_pmu;
+     int ret = 0;
 
      host_pmuver = kvm_arm_pmu_get_pmuver_limit();
 
@@ -1424,26 +1431,51 @@ static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,
      if (kvm_vcpu_has_pmu(vcpu) != valid_pmu)
              return -EINVAL;
 
+     mutex_lock(&arch->config_lock);
      /* We can only differ with PMUver, and anything else is an error */
-     val ^= read_id_reg(vcpu, rd);
+     val ^= old_val;
      val &= ~ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer);
-     if (val)
-             return -EINVAL;
+     if (val) {
+             ret = -EINVAL;
+             goto out;
+     }
 
-     if (valid_pmu)
-             vcpu->kvm->arch.dfr0_pmuver.imp = pmuver;
-     else
-             vcpu->kvm->arch.dfr0_pmuver.unimp = pmuver;
+     /* Only allow userspace to change the idregs before VM running */
+     if (kvm_vm_has_ran_once(vcpu->kvm)) {
+             if (new_val != old_val)
+                     ret = -EBUSY;
+     } else {
+             if (valid_pmu) {
+                     val = IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1);
+                     val &= ~ID_AA64DFR0_EL1_PMUVer_MASK;
+                     val |= FIELD_PREP(ID_AA64DFR0_EL1_PMUVer_MASK, pmuver);
+                     IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1) = val;
+
+                     val = IDREG(vcpu->kvm, SYS_ID_DFR0_EL1);
+                     val &= ~ID_DFR0_EL1_PerfMon_MASK;
+                     val |= FIELD_PREP(ID_DFR0_EL1_PerfMon_MASK, pmuver_to_perfmon(pmuver));
+                     IDREG(vcpu->kvm, SYS_ID_DFR0_EL1) = val;
+             } else {
+                     assign_bit(KVM_ARCH_FLAG_VCPU_HAS_IMP_DEF_PMU, &vcpu->kvm->arch.flags,
+                                pmuver == ID_AA64DFR0_EL1_PMUVer_IMP_DEF);
+             }
+     }
 
-     return 0;
+out:
+     mutex_unlock(&arch->config_lock);
+     return ret;
 }
 
 static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
                         const struct sys_reg_desc *rd,
                         u64 val)
 {
+     struct kvm_arch *arch = &vcpu->kvm->arch;
+     u64 old_val = read_id_reg(vcpu, rd);
      u8 perfmon, host_perfmon;
+     u64 new_val = val;
      bool valid_pmu;
+     int ret = 0;
 
      host_perfmon = pmuver_to_perfmon(kvm_arm_pmu_get_pmuver_limit());
 
@@ -1464,18 +1496,39 @@ static int set_id_dfr0_el1(struct kvm_vcpu *vcpu,
      if (kvm_vcpu_has_pmu(vcpu) != valid_pmu)
              return -EINVAL;
 
+     mutex_lock(&arch->config_lock);
      /* We can only differ with PerfMon, and anything else is an error */
-     val ^= read_id_reg(vcpu, rd);
+     val ^= old_val;
      val &= ~ARM64_FEATURE_MASK(ID_DFR0_EL1_PerfMon);
-     if (val)
-             return -EINVAL;
+     if (val) {
+             ret = -EINVAL;
+             goto out;
+     }
 
-     if (valid_pmu)
-             vcpu->kvm->arch.dfr0_pmuver.imp = perfmon_to_pmuver(perfmon);
-     else
-             vcpu->kvm->arch.dfr0_pmuver.unimp = perfmon_to_pmuver(perfmon);
+     /* Only allow userspace to change the idregs before VM running */
+     if (kvm_vm_has_ran_once(vcpu->kvm)) {
+             if (new_val != old_val)
+                     ret = -EBUSY;
+     } else {
+             if (valid_pmu) {
+                     val = IDREG(vcpu->kvm, SYS_ID_DFR0_EL1);
+                     val &= ~ID_DFR0_EL1_PerfMon_MASK;
+                     val |= FIELD_PREP(ID_DFR0_EL1_PerfMon_MASK, perfmon);
+                     IDREG(vcpu->kvm, SYS_ID_DFR0_EL1) = val;
+
+                     val = IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1);
+                     val &= ~ID_AA64DFR0_EL1_PMUVer_MASK;
+                     val |= FIELD_PREP(ID_AA64DFR0_EL1_PMUVer_MASK, perfmon_to_pmuver(perfmon));
Quoted 8 lines

arch/arm64/kvm/sys_regs.c static int set_id_dfr0_el1(struct kvm_vcpu *vcpu, · line 1527

Quoted 6 lines
arch/arm64/kvm/sys_regs.c +6 -0
+                     IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1) = val;
+             } else {
+                     assign_bit(KVM_ARCH_FLAG_VCPU_HAS_IMP_DEF_PMU, &vcpu->kvm->arch.flags,
+                                perfmon == ID_DFR0_EL1_PerfMon_IMPDEF);
+             }
+     }

This is the exact same code as for aa64fdr0. Make it a helper, please.

Will do.

Quoted 53 lines
Quoted 32 lines
arch/arm64/kvm/sys_regs.c +14 -1
-     return 0;
+out:
+     mutex_unlock(&arch->config_lock);
+     return ret;
 }
 
 /*
@@ -3422,6 +3475,17 @@ void kvm_arm_init_id_regs(struct kvm *kvm)
      }
 
      IDREG(kvm, SYS_ID_AA64PFR0_EL1) = val;
+     /*
+      * Initialise the default PMUver before there is a chance to
+      * create an actual PMU.
+      */
+     val = IDREG(kvm, SYS_ID_AA64DFR0_EL1);
+
+     val &= ~ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer);
+     val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer),
+                       kvm_arm_pmu_get_pmuver_limit());
+
+     IDREG(kvm, SYS_ID_AA64DFR0_EL1) = val;
 }
 
 int __init kvm_sys_reg_table_init(void)
include/kvm/arm_pmu.h +0 -0
index 1a6a695ca67a..8d70dbdc1e0a 100644

include/kvm/arm_pmu.h void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu); · line 97

Quoted 8 lines
include/kvm/arm_pmu.h +3 -2
@@ -92,8 +92,9 @@ void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu);
 /*
  * Evaluates as true when emulating PMUv3p5, and false otherwise.
  */
-#define kvm_pmu_is_3p5(vcpu)                                         \
-     (vcpu->kvm->arch.dfr0_pmuver.imp >= ID_AA64DFR0_EL1_PMUVer_V3P5)
+#define kvm_pmu_is_3p5(vcpu)                                                                 \
+      (FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer),                                 \
+                 IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1)) >= ID_AA64DFR0_EL1_PMUVer_V3P5)

This is getting unreadable. How about something like:

include/kvm/arm_pmu.h +0 -2
index 8d70dbdc1e0a..ecb55d87fa36 100644
@@ -92,9 +92,13 @@ void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu);
 /*
  * Evaluates as true when emulating PMUv3p5, and false otherwise.
  */
-#define kvm_pmu_is_3p5(vcpu)                                                                   \
-        (FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer),                                 \

include/kvm/arm_pmu.h void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu); · line 103

Quoted 10 lines
include/kvm/arm_pmu.h +7 -1
-                   IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1)) >= ID_AA64DFR0_EL1_PMUVer_V3P5)
+#define kvm_pmu_is_3p5(vcpu)   ({                                      \
+       u64 val = IDREG(vcpu->kvm, SYS_ID_AA64DFR0_EL1);                \
+       u8 v;                                                           \
+                                                                       \
+       v = FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMUVer), val); \
+       v >= ID_AA64DFR0_EL1_PMUVer_V3P5;                               \
+})
 
 u8 kvm_arm_pmu_get_pmuver_limit(void);

Sure, will use your suggestion.

Quoted 7 lines
include/kvm/arm_pmu.h +0 -0
 Thanks,
        M.

--
Without deviation from the norm, progress is not possible.

Thanks,
Jing

author
#12 lore raw

PATCH v10 4/5 KVM: arm64: Reuse fields of sys_reg_desc for idreg

In reply to Jing Zhang

Since reset() and val are not used for idreg in sys_reg_desc, they would be used with other purposes for idregs.
The callback reset() would be used to return KVM sanitised id register values. The u64 val would be used as mask for writable fields in idregs. Only bits with 1 in val are writable from userspace.

---
 arch/arm64/kvm/sys_regs.c | 101 +++++++++++++++++++++++++++-----------
 arch/arm64/kvm/sys_regs.h |  15 ++++--
 2 files changed, 82 insertions(+), 34 deletions(-)
arch/arm64/kvm/sys_regs.c +72 -29
index 84d9e4baa4f8..72255dea8027 100644
@@ -541,10 +541,11 @@ static int get_bvr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
 	return 0;
 }
 
-static void reset_bvr(struct kvm_vcpu *vcpu,
+static u64 reset_bvr(struct kvm_vcpu *vcpu,
 		      const struct sys_reg_desc *rd)
 {
 	vcpu->arch.vcpu_debug_state.dbg_bvr[rd->CRm] = rd->val;
+	return rd->val;
 }
 
 static bool trap_bcr(struct kvm_vcpu *vcpu,
@@ -577,10 +578,11 @@ static int get_bcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
 	return 0;
 }
 
-static void reset_bcr(struct kvm_vcpu *vcpu,
+static u64 reset_bcr(struct kvm_vcpu *vcpu,
 		      const struct sys_reg_desc *rd)
 {
 	vcpu->arch.vcpu_debug_state.dbg_bcr[rd->CRm] = rd->val;
+	return rd->val;
 }
 
 static bool trap_wvr(struct kvm_vcpu *vcpu,
@@ -614,10 +616,11 @@ static int get_wvr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
 	return 0;
 }
 
-static void reset_wvr(struct kvm_vcpu *vcpu,
+static u64 reset_wvr(struct kvm_vcpu *vcpu,
 		      const struct sys_reg_desc *rd)
 {
 	vcpu->arch.vcpu_debug_state.dbg_wvr[rd->CRm] = rd->val;
+	return rd->val;
 }
 
 static bool trap_wcr(struct kvm_vcpu *vcpu,
@@ -650,25 +653,28 @@ static int get_wcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
 	return 0;
 }
 
-static void reset_wcr(struct kvm_vcpu *vcpu,
+static u64 reset_wcr(struct kvm_vcpu *vcpu,
 		      const struct sys_reg_desc *rd)
 {
 	vcpu->arch.vcpu_debug_state.dbg_wcr[rd->CRm] = rd->val;
+	return rd->val;
 }
 
-static void reset_amair_el1(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
+static u64 reset_amair_el1(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
 	u64 amair = read_sysreg(amair_el1);
 	vcpu_write_sys_reg(vcpu, amair, AMAIR_EL1);
+	return amair;
 }
 
-static void reset_actlr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
+static u64 reset_actlr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
 	u64 actlr = read_sysreg(actlr_el1);
 	vcpu_write_sys_reg(vcpu, actlr, ACTLR_EL1);
+	return actlr;
 }
 
-static void reset_mpidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
+static u64 reset_mpidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
 	u64 mpidr;
 
@@ -682,7 +688,10 @@ static void reset_mpidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 	mpidr = (vcpu->vcpu_id & 0x0f) << MPIDR_LEVEL_SHIFT(0);
 	mpidr |= ((vcpu->vcpu_id >> 4) & 0xff) << MPIDR_LEVEL_SHIFT(1);
 	mpidr |= ((vcpu->vcpu_id >> 12) & 0xff) << MPIDR_LEVEL_SHIFT(2);
-	vcpu_write_sys_reg(vcpu, (1ULL << 31) | mpidr, MPIDR_EL1);
+	mpidr |= (1ULL << 31);
+	vcpu_write_sys_reg(vcpu, mpidr, MPIDR_EL1);
+
+	return mpidr;
 }
 
 static unsigned int pmu_visibility(const struct kvm_vcpu *vcpu,
@@ -694,13 +703,13 @@ static unsigned int pmu_visibility(const struct kvm_vcpu *vcpu,
 	return REG_HIDDEN;
 }
 
-static void reset_pmu_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
+static u64 reset_pmu_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
 	u64 n, mask = BIT(ARMV8_PMU_CYCLE_IDX);
 
 	/* No PMU available, any PMU reg may UNDEF... */
 	if (!kvm_arm_support_pmu_v3())
-		return;
+		return 0;
 
 	n = read_sysreg(pmcr_el0) >> ARMV8_PMU_PMCR_N_SHIFT;
 	n &= ARMV8_PMU_PMCR_N_MASK;
@@ -709,33 +718,41 @@ static void reset_pmu_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 
 	reset_unknown(vcpu, r);
 	__vcpu_sys_reg(vcpu, r->reg) &= mask;
+
+	return __vcpu_sys_reg(vcpu, r->reg);
 }
 
-static void reset_pmevcntr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
+static u64 reset_pmevcntr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
 	reset_unknown(vcpu, r);
 	__vcpu_sys_reg(vcpu, r->reg) &= GENMASK(31, 0);
+
+	return __vcpu_sys_reg(vcpu, r->reg);
 }
 
-static void reset_pmevtyper(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
+static u64 reset_pmevtyper(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
 	reset_unknown(vcpu, r);
 	__vcpu_sys_reg(vcpu, r->reg) &= ARMV8_PMU_EVTYPE_MASK;
+
+	return __vcpu_sys_reg(vcpu, r->reg);
 }
 
-static void reset_pmselr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
+static u64 reset_pmselr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
 	reset_unknown(vcpu, r);
 	__vcpu_sys_reg(vcpu, r->reg) &= ARMV8_PMU_COUNTER_MASK;
+
+	return __vcpu_sys_reg(vcpu, r->reg);
 }
 
-static void reset_pmcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
+static u64 reset_pmcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
 	u64 pmcr;
 
 	/* No PMU available, PMCR_EL0 may UNDEF... */
 	if (!kvm_arm_support_pmu_v3())
-		return;
+		return 0;
 
 	/* Only preserve PMCR_EL0.N, and reset the rest to 0 */
 	pmcr = read_sysreg(pmcr_el0) & (ARMV8_PMU_PMCR_N_MASK << ARMV8_PMU_PMCR_N_SHIFT);
@@ -743,6 +760,8 @@ static void reset_pmcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 		pmcr |= ARMV8_PMU_PMCR_LC;
 
 	__vcpu_sys_reg(vcpu, r->reg) = pmcr;
+
+	return __vcpu_sys_reg(vcpu, r->reg);
 }
 
 static bool check_pmu_access_disabled(struct kvm_vcpu *vcpu, u64 flags)
@@ -1212,6 +1231,11 @@ static u8 pmuver_to_perfmon(u8 pmuver)
 	}
 }
 
+static u64 general_read_kvm_sanitised_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd)
+{
+	return read_sanitised_ftr_reg(reg_to_encoding(rd));
+}
+
 static u64 kvm_arm_read_id_reg(const struct kvm_vcpu *vcpu, u32 id)
 {
 	u64 val = IDREG(vcpu->kvm, id);
@@ -1597,7 +1621,7 @@ static bool access_clidr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
  * Fabricate a CLIDR_EL1 value instead of using the real value, which can vary
  * by the physical CPU which the vcpu currently resides in.
  */
-static void reset_clidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
+static u64 reset_clidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
 	u64 ctr_el0 = read_sanitised_ftr_reg(SYS_CTR_EL0);
 	u64 clidr;
@@ -1645,6 +1669,8 @@ static void reset_clidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 		clidr |= 2 << CLIDR_TTYPE_SHIFT(loc);
 
 	__vcpu_sys_reg(vcpu, r->reg) = clidr;
+
+	return __vcpu_sys_reg(vcpu, r->reg);
 }
 
 static int set_clidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
@@ -1744,6 +1770,17 @@ static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu,
 	.visibility = elx2_visibility,		\
 }
 
+/*
+ * Since reset() callback and field val are not used for idregs, they will be
+ * used for specific purposes for idregs.
+ * The reset() would return KVM sanitised register value. The value would be the
+ * same as the host kernel sanitised value if there is no KVM sanitisation.
+ * The val would be used as a mask indicating writable fields for the idreg.
+ * Only bits with 1 are writable from userspace. This mask might not be
+ * necessary in the future whenever all ID registers are enabled as writable
+ * from userspace.
+ */
+
 /* sys_reg_desc initialiser for known cpufeature ID registers */
 #define ID_SANITISED(name) {			\
 	SYS_DESC(SYS_##name),			\
@@ -1751,6 +1788,8 @@ static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu,
 	.get_user = get_id_reg,			\
 	.set_user = set_id_reg,			\
 	.visibility = id_visibility,		\
+	.reset = general_read_kvm_sanitised_reg,\
+	.val = 0,				\
 }
 
 /* sys_reg_desc initialiser for known cpufeature ID registers */
@@ -1760,6 +1799,8 @@ static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu,
 	.get_user = get_id_reg,			\
 	.set_user = set_id_reg,			\
 	.visibility = aa32_id_visibility,	\
+	.reset = general_read_kvm_sanitised_reg,\
+	.val = 0,				\
 }
 
 /*
@@ -1772,7 +1813,9 @@ static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu,
 	.access = access_id_reg,			\
 	.get_user = get_id_reg,				\
 	.set_user = set_id_reg,				\
-	.visibility = raz_visibility			\
+	.visibility = raz_visibility,			\
+	.reset = NULL,					\
+	.val = 0,					\
 }
 
 /*
@@ -1786,6 +1829,8 @@ static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu,
 	.get_user = get_id_reg,			\
 	.set_user = set_id_reg,			\
 	.visibility = raz_visibility,		\
+	.reset = NULL,				\
+	.val = 0,				\
 }
 
 static bool access_sp_el1(struct kvm_vcpu *vcpu,
@@ -3122,19 +3167,21 @@ id_to_sys_reg_desc(struct kvm_vcpu *vcpu, u64 id,
  */
 
 #define FUNCTION_INVARIANT(reg)						\
-	static void get_##reg(struct kvm_vcpu *v,			\
+	static u64 get_##reg(struct kvm_vcpu *v,			\
 			      const struct sys_reg_desc *r)		\
 	{								\
 		((struct sys_reg_desc *)r)->val = read_sysreg(reg);	\
+		return ((struct sys_reg_desc *)r)->val;			\
 	}
 
 FUNCTION_INVARIANT(midr_el1)
 FUNCTION_INVARIANT(revidr_el1)
 FUNCTION_INVARIANT(aidr_el1)
 
-static void get_ctr_el0(struct kvm_vcpu *v, const struct sys_reg_desc *r)
+static u64 get_ctr_el0(struct kvm_vcpu *v, const struct sys_reg_desc *r)
 {
 	((struct sys_reg_desc *)r)->val = read_sanitised_ftr_reg(SYS_CTR_EL0);
+	return ((struct sys_reg_desc *)r)->val;
 }
 
 /* ->val is filled in by kvm_sys_reg_table_init() */
@@ -3424,9 +3471,7 @@ int kvm_arm_copy_sys_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
 	return write_demux_regids(uindices);
 }
 
-/*
- * Set the guest's ID registers with ID_SANITISED() to the host's sanitized value.
- */
+/* Initialize the guest's ID registers with KVM sanitised values. */
 void kvm_arm_init_id_regs(struct kvm *kvm)
 {
 	const struct sys_reg_desc *idreg;
@@ -3443,13 +3488,11 @@ void kvm_arm_init_id_regs(struct kvm *kvm)
 
 	/* Initialize all idregs */
 	while (is_id_reg(id)) {
-		/*
-		 * Some hidden ID registers which are not in arm64_ftr_regs[]
-		 * would cause warnings from read_sanitised_ftr_reg().
-		 * Skip those ID registers to avoid the warnings.
-		 */
-		if (idreg->visibility != raz_visibility)
-			IDREG(kvm, id) = read_sanitised_ftr_reg(id);
+		val = 0;
+		/* Read KVM sanitised register value if available */
+		if (idreg->reset)
+			val = idreg->reset(NULL, idreg);
+		IDREG(kvm, id) = val;
 
 		idreg++;
 		id = reg_to_encoding(idreg);
arch/arm64/kvm/sys_regs.h +10 -5
index eba10de2e7ae..c65c129b3500 100644
@@ -71,13 +71,16 @@ struct sys_reg_desc {
 		       struct sys_reg_params *,
 		       const struct sys_reg_desc *);
 
-	/* Initialization for vcpu. */
-	void (*reset)(struct kvm_vcpu *, const struct sys_reg_desc *);
+	/*
+	 * Initialization for vcpu. Return initialized value, or KVM
+	 * sanitized value for ID registers.
+	 */
+	u64 (*reset)(struct kvm_vcpu *, const struct sys_reg_desc *);
 
 	/* Index into sys_reg[], or 0 if we don't need to save it. */
 	int reg;
 
-	/* Value (usually reset value) */
+	/* Value (usually reset value), or write mask for idregs */
 	u64 val;
 
 	/* Custom get/set_user functions, fallback to generic if NULL */
@@ -130,19 +133,21 @@ static inline bool read_zero(struct kvm_vcpu *vcpu,
 }
 
 /* Reset functions */
-static inline void reset_unknown(struct kvm_vcpu *vcpu,
+static inline u64 reset_unknown(struct kvm_vcpu *vcpu,
 				 const struct sys_reg_desc *r)
 {
 	BUG_ON(!r->reg);
 	BUG_ON(r->reg >= NR_SYS_REGS);
 	__vcpu_sys_reg(vcpu, r->reg) = 0x1de7ec7edbadc0deULL;
+	return __vcpu_sys_reg(vcpu, r->reg);
 }
 
-static inline void reset_val(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
+static inline u64 reset_val(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 {
 	BUG_ON(!r->reg);
 	BUG_ON(r->reg >= NR_SYS_REGS);
 	__vcpu_sys_reg(vcpu, r->reg) = r->val;
+	return __vcpu_sys_reg(vcpu, r->reg);
 }
 
 static inline unsigned int sysreg_visibility(const struct kvm_vcpu *vcpu,
Signature · 1 line
2.40.1.698.g37aff9b760-goog
#13 lore raw

Re: [PATCH v10 4/5] KVM: arm64: Reuse fields of sys_reg_desc for idreg

Hi Jing,

Quoted 5 lines · Jing Zhang

Since reset() and val are not used for idreg in sys_reg_desc, they would be used with other purposes for idregs.
The callback reset() would be used to return KVM sanitised id register values. The u64 val would be used as mask for writable fields in idregs. Only bits with 1 in val are writable from userspace.

The tense of the changelog is wrong (should be in an imperative mood). Maybe something like:

sys_reg_desc::{reset, val} are presently unused for ID register descriptors. Repurpose these fields to support user-configurable ID registers.

Use the ::reset() function pointer to return the sanitised value of a given ID register, optionally with KVM-specific feature sanitisation. Additionally, keep a mask of writable register fields in ::val.

Quoted 6 lines
---
 arch/arm64/kvm/sys_regs.c | 101 +++++++++++++++++++++++++++-----------
 arch/arm64/kvm/sys_regs.h |  15 ++++--
 2 files changed, 82 insertions(+), 34 deletions(-)

[...]

Quoted 15 lines
arch/arm64/kvm/sys_regs.c +11 -0
+/*
+ * Since reset() callback and field val are not used for idregs, they will be
+ * used for specific purposes for idregs.
+ * The reset() would return KVM sanitised register value. The value would be the
+ * same as the host kernel sanitised value if there is no KVM sanitisation.
+ * The val would be used as a mask indicating writable fields for the idreg.
+ * Only bits with 1 are writable from userspace. This mask might not be
+ * necessary in the future whenever all ID registers are enabled as writable
+ * from userspace.
+ */
+
 /* sys_reg_desc initialiser for known cpufeature ID registers */
 #define ID_SANITISED(name) {			\
 	SYS_DESC(SYS_##name),			\

arch/arm64/kvm/sys_regs.c static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu, · line 1792

Quoted 5 lines
arch/arm64/kvm/sys_regs.c +2 -0
@@ -1751,6 +1788,8 @@ static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu,
 	.get_user = get_id_reg,			\
 	.set_user = set_id_reg,			\
 	.visibility = id_visibility,		\
+	.reset = general_read_kvm_sanitised_reg,\
+	.val = 0,				\

I generally think unions are more trouble than they're worth, but it might make sense to throw the fields with dual meaning into one, like

  struct sys_reg_desc {

[...]

	union {
		struct {
			void (*reset)(struct kvm_vcpu *, const struct sys_reg_desc *);
			u64 val;
		};
		struct {
			u64 (*read_sanitised)(struct kvm_vcpu *vcpu, const struct sys_reg_desc *);
			u64 mask;
		};
	};

}

You could then avoid repainting the world to handle ->reset() returning a value and usage of the fields in an id register context become a bit more self-documenting. And you get to play with fire while you do it!

Let's see if the other side of the pond agrees with my bikeshedding...

Signature · 2 lines
Thanks,
Oliver
maintainer
#14 lore raw

Re: [PATCH v10 4/5] KVM: arm64: Reuse fields of sys_reg_desc for idreg

Quoted 67 lines · Oliver Upton

Hi Jing,

Quoted 5 lines · Jing Zhang

Since reset() and val are not used for idreg in sys_reg_desc, they would be used with other purposes for idregs.
The callback reset() would be used to return KVM sanitised id register values. The u64 val would be used as mask for writable fields in idregs. Only bits with 1 in val are writable from userspace.

The tense of the changelog is wrong (should be in an imperative mood). Maybe something like:

sys_reg_desc::{reset, val} are presently unused for ID register descriptors. Repurpose these fields to support user-configurable ID registers.

Use the ::reset() function pointer to return the sanitised value of a given ID register, optionally with KVM-specific feature sanitisation. Additionally, keep a mask of writable register fields in ::val.

Quoted 6 lines
---
 arch/arm64/kvm/sys_regs.c | 101 +++++++++++++++++++++++++++-----------
 arch/arm64/kvm/sys_regs.h |  15 ++++--
 2 files changed, 82 insertions(+), 34 deletions(-)

[...]

Quoted 15 lines
arch/arm64/kvm/sys_regs.c +11 -0
+/*
+ * Since reset() callback and field val are not used for idregs, they will be
+ * used for specific purposes for idregs.
+ * The reset() would return KVM sanitised register value. The value would be the
+ * same as the host kernel sanitised value if there is no KVM sanitisation.
+ * The val would be used as a mask indicating writable fields for the idreg.
+ * Only bits with 1 are writable from userspace. This mask might not be
+ * necessary in the future whenever all ID registers are enabled as writable
+ * from userspace.
+ */
+
 /* sys_reg_desc initialiser for known cpufeature ID registers */
 #define ID_SANITISED(name) {			\
 	SYS_DESC(SYS_##name),			\

arch/arm64/kvm/sys_regs.c static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu, · line 1792

Quoted 5 lines
arch/arm64/kvm/sys_regs.c +2 -0
@@ -1751,6 +1788,8 @@ static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu,
 	.get_user = get_id_reg,			\
 	.set_user = set_id_reg,			\
 	.visibility = id_visibility,		\
+	.reset = general_read_kvm_sanitised_reg,\
+	.val = 0,				\

I generally think unions are more trouble than they're worth, but it might make sense to throw the fields with dual meaning into one, like

  struct sys_reg_desc {

[...]

	union {
		struct {
			void (*reset)(struct kvm_vcpu *, const struct sys_reg_desc *);
			u64 val;
		};
		struct {
			u64 (*read_sanitised)(struct kvm_vcpu *vcpu, const struct sys_reg_desc *);
			u64 mask;
		};
	};
Quoted 7 lines

}

You could then avoid repainting the world to handle ->reset() returning a value and usage of the fields in an id register context become a bit more self-documenting. And you get to play with fire while you do it!

Let's see if the other side of the pond agrees with my bikeshedding...

I don't think this works just as suggested. It completely breaks all the existing macros that use implicit (in order) initialisers. Not hard to fix, but pretty invasive (I count 656 warnings in my current tree).

I agree it is much cleaner though.

	M.
Signature · 1 line
Without deviation from the norm, progress is not possible.
author
#15 lore raw

PATCH v10 5/5 KVM: arm64: Refactor writings for PMUVer/CSV2/CSV3

In reply to Jing Zhang

Refactor writings for ID_AA64PFR0_EL1.[CSV2|CSV3], ID_AA64DFR0_EL1.PMUVer and ID_DFR0_ELF.PerfMon based on utilities specific to ID register.

---
 arch/arm64/include/asm/cpufeature.h |   1 +
 arch/arm64/kernel/cpufeature.c      |   2 +-
 arch/arm64/kvm/sys_regs.c           | 365 ++++++++++++++++++----------
 3 files changed, 243 insertions(+), 125 deletions(-)
arch/arm64/include/asm/cpufeature.h +1 -0
index 6bf013fb110d..dc769c2eb7a4 100644
@@ -915,6 +915,7 @@ static inline unsigned int get_vmid_bits(u64 mmfr1)
 	return 8;
 }
 
+s64 arm64_ftr_safe_value(const struct arm64_ftr_bits *ftrp, s64 new, s64 cur);
 struct arm64_ftr_reg *get_arm64_ftr_reg(u32 sys_id);
 
 extern struct arm64_ftr_override id_aa64mmfr1_override;
arch/arm64/kernel/cpufeature.c +1 -1
index 7d7128c65161..3317a7b6deac 100644
@@ -798,7 +798,7 @@ static u64 arm64_ftr_set_value(const struct arm64_ftr_bits *ftrp, s64 reg,
 	return reg;
 }
 
-static s64 arm64_ftr_safe_value(const struct arm64_ftr_bits *ftrp, s64 new,
+s64 arm64_ftr_safe_value(const struct arm64_ftr_bits *ftrp, s64 new,
 				s64 cur)
 {
 	s64 ret = 0;
arch/arm64/kvm/sys_regs.c +241 -124

482 lines not on this page · raw message

Signature · 1 line
2.40.1.698.g37aff9b760-goog
maintainer
#16 lore raw

Re: [PATCH v10 5/5] KVM: arm64: Refactor writings for PMUVer/CSV2/CSV3

Quoted 11 lines · Jing Zhang

Refactor writings for ID_AA64PFR0_EL1.[CSV2|CSV3], ID_AA64DFR0_EL1.PMUVer and ID_DFR0_ELF.PerfMon based on utilities specific to ID register.

---
 arch/arm64/include/asm/cpufeature.h |   1 +
 arch/arm64/kernel/cpufeature.c      |   2 +-
 arch/arm64/kvm/sys_regs.c           | 365 ++++++++++++++++++----------
 3 files changed, 243 insertions(+), 125 deletions(-)

Reading the result after applying this series, I feel like a stuck record. This final series still contains gems like this:

static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu,

			       const struct sys_reg_desc *rd,
			       u64 val)
{
	u8 csv2, csv3;

	/*
	 * Allow AA64PFR0_EL1.CSV2 to be set from userspace as long as
	 * it doesn't promise more than what is actually provided (the
	 * guest could otherwise be covered in ectoplasmic residue).
	 */
	csv2 = cpuid_feature_extract_unsigned_field(val, ID_AA64PFR0_EL1_CSV2_SHIFT);
	if (csv2 > 1 ||
	    (csv2 && arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED))
		return -EINVAL;

	/* Same thing for CSV3 */
	csv3 = cpuid_feature_extract_unsigned_field(val, ID_AA64PFR0_EL1_CSV3_SHIFT);
	if (csv3 > 1 ||
	    (csv3 && arm64_get_meltdown_state() != SPECTRE_UNAFFECTED))
		return -EINVAL;

	return set_id_reg(vcpu, rd, val);
}

Why do we have this? I've asked the question at least 3 times in the previous versions, and I still see the same code.

If we have sane limits, the call to arm64_check_features() in set_id_reg() will catch the illegal write. So why do we have this at all? The whole point of the exercise was to unify the handling. But you're actually making it worse.

So what's the catch?

	M.
Signature · 1 line
Without deviation from the norm, progress is not possible.
author
#17 lore raw

Re: [PATCH v10 5/5] KVM: arm64: Refactor writings for PMUVer/CSV2/CSV3

Hi Marc,

Quoted 46 lines · Marc Zyngier
Quoted 11 lines · Jing Zhang

Refactor writings for ID_AA64PFR0_EL1.[CSV2|CSV3], ID_AA64DFR0_EL1.PMUVer and ID_DFR0_ELF.PerfMon based on utilities specific to ID register.

---
 arch/arm64/include/asm/cpufeature.h |   1 +
 arch/arm64/kernel/cpufeature.c      |   2 +-
 arch/arm64/kvm/sys_regs.c           | 365 ++++++++++++++++++----------
 3 files changed, 243 insertions(+), 125 deletions(-)

Reading the result after applying this series, I feel like a stuck record. This final series still contains gems like this:

static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu,
const struct sys_reg_desc *rd, u64 val)

{
        u8 csv2, csv3;

        /*
         * Allow AA64PFR0_EL1.CSV2 to be set from userspace as long as
         * it doesn't promise more than what is actually provided (the
         * guest could otherwise be covered in ectoplasmic residue).
         */
        csv2 = cpuid_feature_extract_unsigned_field(val, ID_AA64PFR0_EL1_CSV2_SHIFT);
        if (csv2 > 1 ||
            (csv2 && arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED))
                return -EINVAL;

        /* Same thing for CSV3 */
        csv3 = cpuid_feature_extract_unsigned_field(val, ID_AA64PFR0_EL1_CSV3_SHIFT);
        if (csv3 > 1 ||
            (csv3 && arm64_get_meltdown_state() != SPECTRE_UNAFFECTED))
                return -EINVAL;

        return set_id_reg(vcpu, rd, val);
}

Why do we have this? I've asked the question at least 3 times in the previous versions, and I still see the same code.

Quoted 6 lines

If we have sane limits, the call to arm64_check_features() in set_id_reg() will catch the illegal write. So why do we have this at all? The whole point of the exercise was to unify the handling. But you're actually making it worse.

So what's the catch?

Sorry, I am only aware of one discussion of this code in v8. The reason I still keep the check here is that the arm64_check_features() can not catch all illegal writes as this code does.
For example, for CSV2, one concern is:
When arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED, this code only allows guest CSV2 to be set to 0, any non-zero value would lead to -EINVAL. If we remove the check here, the guest CSV2 can be set to any value lower or equal to host CSV2.
Of course, we can set the sane limit of CSV2 to 0 when arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED in read_sanitised_id_aa64pfr0_el1(). Then we can remove all the checks here and no specific set_id function for AA64PFR0_EL1 is needed.

Quoted 5 lines
        M.

--
Without deviation from the norm, progress is not possible.

Thanks,
Jing

maintainer
#18 lore raw

Re: [PATCH v10 5/5] KVM: arm64: Refactor writings for PMUVer/CSV2/CSV3

Quoted 56 lines · Jing Zhang

Hi Marc,

Quoted 46 lines · Marc Zyngier
Quoted 11 lines · Jing Zhang

Refactor writings for ID_AA64PFR0_EL1.[CSV2|CSV3], ID_AA64DFR0_EL1.PMUVer and ID_DFR0_ELF.PerfMon based on utilities specific to ID register.

---
 arch/arm64/include/asm/cpufeature.h |   1 +
 arch/arm64/kernel/cpufeature.c      |   2 +-
 arch/arm64/kvm/sys_regs.c           | 365 ++++++++++++++++++----------
 3 files changed, 243 insertions(+), 125 deletions(-)

Reading the result after applying this series, I feel like a stuck record. This final series still contains gems like this:

static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu,
const struct sys_reg_desc *rd, u64 val)

{
        u8 csv2, csv3;

        /*
         * Allow AA64PFR0_EL1.CSV2 to be set from userspace as long as
         * it doesn't promise more than what is actually provided (the
         * guest could otherwise be covered in ectoplasmic residue).
         */
        csv2 = cpuid_feature_extract_unsigned_field(val, ID_AA64PFR0_EL1_CSV2_SHIFT);
        if (csv2 > 1 ||
            (csv2 && arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED))
                return -EINVAL;

        /* Same thing for CSV3 */
        csv3 = cpuid_feature_extract_unsigned_field(val, ID_AA64PFR0_EL1_CSV3_SHIFT);
        if (csv3 > 1 ||
            (csv3 && arm64_get_meltdown_state() != SPECTRE_UNAFFECTED))
                return -EINVAL;

        return set_id_reg(vcpu, rd, val);
}

Why do we have this? I've asked the question at least 3 times in the previous versions, and I still see the same code.

Quoted 6 lines

If we have sane limits, the call to arm64_check_features() in set_id_reg() will catch the illegal write. So why do we have this at all? The whole point of the exercise was to unify the handling. But you're actually making it worse.

So what's the catch?

Quoted 8 lines

Sorry, I am only aware of one discussion of this code in v8. The reason I still keep the check here is that the arm64_check_features() can not catch all illegal writes as this code does.
For example, for CSV2, one concern is:
When arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED, this code only allows guest CSV2 to be set to 0, any non-zero value would lead to -EINVAL. If we remove the check here, the guest CSV2 can be set to any value lower or equal to host CSV2.

Sorry, this doesn't make sense. Lower is always fine. If you meant 'higher', then I agree that it would be bad. But that doesn't make keeping this code the right outcome.

Quoted 4 lines

Of course, we can set the sane limit of CSV2 to 0 when arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED in read_sanitised_id_aa64pfr0_el1(). Then we can remove all the checks here and no specific set_id function for AA64PFR0_EL1 is needed.

This is what I have been asking for all along: the "sanitised" view of the register *must* return the absolute limit for the fields that are flagged as writable by "mask".

If we need extra code, then something is really wrong. The core feature code manages that without any special casing, and we should be able to reach the same level.

Thanks,

	M.
Signature · 1 line
Without deviation from the norm, progress is not possible.
author
#19 lore raw

Re: [PATCH v10 5/5] KVM: arm64: Refactor writings for PMUVer/CSV2/CSV3

Hi Marc,

Quoted 68 lines · Marc Zyngier
Quoted 56 lines · Jing Zhang

Hi Marc,

Quoted 46 lines · Marc Zyngier
Quoted 11 lines · Jing Zhang

Refactor writings for ID_AA64PFR0_EL1.[CSV2|CSV3], ID_AA64DFR0_EL1.PMUVer and ID_DFR0_ELF.PerfMon based on utilities specific to ID register.

---
 arch/arm64/include/asm/cpufeature.h |   1 +
 arch/arm64/kernel/cpufeature.c      |   2 +-
 arch/arm64/kvm/sys_regs.c           | 365 ++++++++++++++++++----------
 3 files changed, 243 insertions(+), 125 deletions(-)

Reading the result after applying this series, I feel like a stuck record. This final series still contains gems like this:

static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu,
const struct sys_reg_desc *rd, u64 val)

{
        u8 csv2, csv3;

        /*
         * Allow AA64PFR0_EL1.CSV2 to be set from userspace as long as
         * it doesn't promise more than what is actually provided (the
         * guest could otherwise be covered in ectoplasmic residue).
         */
        csv2 = cpuid_feature_extract_unsigned_field(val, ID_AA64PFR0_EL1_CSV2_SHIFT);
        if (csv2 > 1 ||
            (csv2 && arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED))
                return -EINVAL;

        /* Same thing for CSV3 */
        csv3 = cpuid_feature_extract_unsigned_field(val, ID_AA64PFR0_EL1_CSV3_SHIFT);
        if (csv3 > 1 ||
            (csv3 && arm64_get_meltdown_state() != SPECTRE_UNAFFECTED))
                return -EINVAL;

        return set_id_reg(vcpu, rd, val);
}

Why do we have this? I've asked the question at least 3 times in the previous versions, and I still see the same code.

Quoted 6 lines

If we have sane limits, the call to arm64_check_features() in set_id_reg() will catch the illegal write. So why do we have this at all? The whole point of the exercise was to unify the handling. But you're actually making it worse.

So what's the catch?

Quoted 8 lines

Sorry, I am only aware of one discussion of this code in v8. The reason I still keep the check here is that the arm64_check_features() can not catch all illegal writes as this code does.
For example, for CSV2, one concern is:
When arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED, this code only allows guest CSV2 to be set to 0, any non-zero value would lead to -EINVAL. If we remove the check here, the guest CSV2 can be set to any value lower or equal to host CSV2.

Quoted 3 lines

Sorry, this doesn't make sense. Lower is always fine. If you meant 'higher', then I agree that it would be bad. But that doesn't make keeping this code the right outcome.

Got it. Then it would be good to remove the check here. Will do that.

Quoted 6 lines
Quoted 5 lines

Of course, we can set the sane limit of CSV2 to 0 when arm64_get_spectre_v2_state() != SPECTRE_UNAFFECTED in read_sanitised_id_aa64pfr0_el1(). Then we can remove all the checks here and no specific set_id function for AA64PFR0_EL1 is needed.

Quoted 7 lines

This is what I have been asking for all along: the "sanitised" view of the register *must* return the absolute limit for the fields that are flagged as writable by "mask".

If we need extra code, then something is really wrong. The core feature code manages that without any special casing, and we should be able to reach the same level.

Understood.

Quoted 7 lines

Thanks,

        M.

--
Without deviation from the norm, progress is not possible.

Thanks,
Jing

_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel