From: Raghavendra Rao Ananta <rananta@google.com>
To: Marc Zyngier <maz@kernel.org>, Andrew Jones <drjones@redhat.com>,
James Morse <james.morse@arm.com>,
Alexandru Elisei <alexandru.elisei@arm.com>,
Suzuki K Poulose <suzuki.poulose@arm.com>
Cc: kvm@vger.kernel.org, Catalin Marinas <catalin.marinas@arm.com>,
Peter Shier <pshier@google.com>,
linux-kernel@vger.kernel.org, Will Deacon <will@kernel.org>,
kvmarm@lists.cs.columbia.edu,
linux-arm-kernel@lists.infradead.org
Subject: [RFC PATCH 5/8] KVM: arm64: Add vendor hypervisor service calls firmware register
Date: Tue, 2 Nov 2021 00:22:00 +0000 [thread overview]
Message-ID: <20211102002203.1046069-6-rananta@google.com> (raw)
In-Reply-To: <20211102002203.1046069-1-rananta@google.com>
Introduce the firmware register to hold the vendor specific
hypervisor service calls (owner value 6) as a bitmap. The
bitmap represents the features that'll be enabled for the
guest, as configured by the user-space. Currently, this
includes support only for Precision Time Protocol (PTP),
represented by bit-0.
Signed-off-by: Raghavendra Rao Ananta <rananta@google.com>
---
Documentation/virt/kvm/arm/hypercalls.rst | 23 +++++++++++-----
arch/arm64/include/asm/kvm_host.h | 1 +
arch/arm64/include/uapi/asm/kvm.h | 6 +++++
arch/arm64/kvm/arm.c | 1 +
arch/arm64/kvm/hypercalls.c | 33 ++++++++++++++++++++++-
include/kvm/arm_hypercalls.h | 3 +++
6 files changed, 59 insertions(+), 8 deletions(-)
diff --git a/Documentation/virt/kvm/arm/hypercalls.rst b/Documentation/virt/kvm/arm/hypercalls.rst
index 2cb82c694868..61d95f4f1ddf 100644
--- a/Documentation/virt/kvm/arm/hypercalls.rst
+++ b/Documentation/virt/kvm/arm/hypercalls.rst
@@ -20,13 +20,14 @@ pseudo-registers" that can be manipulated using the GET/SET_ONE_REG
interface. These registers can be saved/restored by userspace, and set
to a convenient value if required.
-The firmware registers, KVM_REG_ARM_STD and KVM_REG_ARM_STD_HYP exposes
-the hypercall services in the form of a feature bitmap. Upon VM creation,
-by default, KVM exposes all the features to the guest, which can be learnt
-using GET_ONE_REG interface. Conversely, the features can be enabled or
-disabled via the SET_ONE_REG interface. These registers allow the user-space
-modification only until the VM has started running, after which they turn to
-read-only registers. SET_ONE_REG in this scenario will return -EBUSY.
+The firmware registers, KVM_REG_ARM_STD, KVM_REG_ARM_STD_HYP and
+KVM_REG_ARM_VENDOR_HYP exposes the hypercall services in the form of a
+feature bitmap. Upon VM creation, by default, KVM exposes all the features
+to the guest, which can be learnt using GET_ONE_REG interface. Conversely,
+the features can be enabled or disabled via the SET_ONE_REG interface.
+These registers allow the user-space modification only until the VM has
+started running, after which they turn to read-only registers.
+SET_ONE_REG in this scenario will return -EBUSY.
The following register is defined:
@@ -100,4 +101,12 @@ The following register is defined:
The bit represents the Paravirtualized Time service (also known as stolen time) as
represented by ARM DEN0057A.
+* KVM_REG_ARM_VENDOR_HYP
+ Controls the bitmap of the Vendor specific Hypervisor Service Calls.
+
+ The following bits are accepted:
+
+ KVM_REG_ARM_STD_HYP_PTP:
+ The bit represents the Precision Time Protocol KVM service.
+
.. [1] https://developer.arm.com/-/media/developer/pdf/ARM_DEN_0070A_Firmware_interfaces_for_mitigating_CVE-2017-5715.pdf
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index cee4f4b8a756..27861b3bd25f 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -108,6 +108,7 @@ struct hvc_reg_desc {
u64 kvm_std_bmap;
u64 kvm_std_hyp_bmap;
+ u64 kvm_vendor_hyp_bmap;
};
struct kvm_arch {
diff --git a/arch/arm64/include/uapi/asm/kvm.h b/arch/arm64/include/uapi/asm/kvm.h
index 46701da1a27d..a1d0e8e69eed 100644
--- a/arch/arm64/include/uapi/asm/kvm.h
+++ b/arch/arm64/include/uapi/asm/kvm.h
@@ -293,6 +293,12 @@ enum kvm_reg_arm_std_hyp_bmap {
KVM_REG_ARM_STD_HYP_BMAP_MAX,
};
+#define KVM_REG_ARM_VENDOR_HYP KVM_REG_ARM_FW_REG(5)
+enum kvm_reg_arm_vendor_hyp_bmap {
+ KVM_REG_ARM_VENDOR_HYP_PTP,
+ KVM_REG_ARM_VENDOR_HYP_BMAP_MAX,
+};
+
/* SVE registers */
#define KVM_REG_ARM64_SVE (0x15 << KVM_REG_ARM_COPROC_SHIFT)
diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index 1c69d2a71b86..5c89db8336eb 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -136,6 +136,7 @@ static void set_default_hypercalls(struct kvm *kvm)
hvc_desc->kvm_std_bmap = ARM_SMCCC_STD_FEATURES;
hvc_desc->kvm_std_hyp_bmap = ARM_SMCCC_STD_HYP_FEATURES;
+ hvc_desc->kvm_vendor_hyp_bmap = ARM_SMCCC_VENDOR_HYP_FEATURES;
}
/**
diff --git a/arch/arm64/kvm/hypercalls.c b/arch/arm64/kvm/hypercalls.c
index 46064c515058..74ebe5355dc0 100644
--- a/arch/arm64/kvm/hypercalls.c
+++ b/arch/arm64/kvm/hypercalls.c
@@ -67,6 +67,8 @@ static u64 *kvm_fw_reg_to_bmap(struct kvm *kvm, u64 fw_reg)
return &hvc_desc->kvm_std_bmap;
case KVM_REG_ARM_STD_HYP:
return &hvc_desc->kvm_std_hyp_bmap;
+ case KVM_REG_ARM_VENDOR_HYP:
+ return &hvc_desc->kvm_vendor_hyp_bmap;
default:
return NULL;
}
@@ -93,6 +95,10 @@ static const struct kvm_hvc_func_map hvc_std_hyp_map[] = {
HVC_FUNC_MAP_DESC(ARM_SMCCC_HV_PV_TIME_ST, KVM_REG_ARM_STD_HYP_PV_TIME_ST),
};
+static const struct kvm_hvc_func_map hvc_vendor_hyp_map[] = {
+ HVC_FUNC_MAP_DESC(ARM_SMCCC_VENDOR_HYP_KVM_PTP_FUNC_ID, KVM_REG_ARM_VENDOR_HYP_PTP),
+};
+
bool kvm_hvc_call_supported(struct kvm_vcpu *vcpu, u32 func_id)
{
struct kvm *kvm = vcpu->kvm;
@@ -113,6 +119,11 @@ bool kvm_hvc_call_supported(struct kvm_vcpu *vcpu, u32 func_id)
hvc_func_map = hvc_std_hyp_map;
map_sz = ARRAY_SIZE(hvc_std_hyp_map);
break;
+ case ARM_SMCCC_OWNER_VENDOR_HYP:
+ fw_reg = KVM_REG_ARM_VENDOR_HYP;
+ hvc_func_map = hvc_vendor_hyp_map;
+ map_sz = ARRAY_SIZE(hvc_vendor_hyp_map);
+ break;
default:
/* Allow all the owners that aren't mapped */
return true;
@@ -133,6 +144,7 @@ bool kvm_hvc_call_supported(struct kvm_vcpu *vcpu, u32 func_id)
int kvm_hvc_call_handler(struct kvm_vcpu *vcpu)
{
+ struct hvc_reg_desc *hvc_desc = &vcpu->kvm->arch.hvc_desc;
u32 func_id = smccc_get_function(vcpu);
u64 val[4] = {SMCCC_RET_NOT_SUPPORTED};
u32 feature;
@@ -204,7 +216,14 @@ int kvm_hvc_call_handler(struct kvm_vcpu *vcpu)
break;
case ARM_SMCCC_VENDOR_HYP_KVM_FEATURES_FUNC_ID:
val[0] = BIT(ARM_SMCCC_KVM_FUNC_FEATURES);
- val[0] |= BIT(ARM_SMCCC_KVM_FUNC_PTP);
+
+ /*
+ * The feature bits exposed to user-space doesn't include
+ * ARM_SMCCC_KVM_FUNC_FEATURES. However, we expose this to
+ * the guest as bit-0. Hence, left-shift the user-space
+ * exposed bitmap by 1 to accommodate this.
+ */
+ val[0] |= (hvc_desc->kvm_vendor_hyp_bmap << 1);
break;
case ARM_SMCCC_VENDOR_HYP_KVM_PTP_FUNC_ID:
kvm_ptp_get_time(vcpu, val);
@@ -230,6 +249,7 @@ static const u64 fw_reg_ids[] = {
KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2,
KVM_REG_ARM_STD,
KVM_REG_ARM_STD_HYP,
+ KVM_REG_ARM_VENDOR_HYP,
};
int kvm_arm_get_fw_num_regs(struct kvm_vcpu *vcpu)
@@ -310,6 +330,9 @@ int kvm_arm_get_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
case KVM_REG_ARM_STD_HYP:
val = hvc_desc->kvm_std_hyp_bmap;
break;
+ case KVM_REG_ARM_VENDOR_HYP:
+ val = hvc_desc->kvm_vendor_hyp_bmap;
+ break;
default:
return -ENOENT;
}
@@ -446,6 +469,14 @@ int kvm_arm_set_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
hvc_desc->kvm_std_hyp_bmap = val;
return 0;
+
+ case KVM_REG_ARM_VENDOR_HYP:
+ if (val & ~ARM_SMCCC_VENDOR_HYP_FEATURES)
+ return -EINVAL;
+
+ hvc_desc->kvm_vendor_hyp_bmap = val;
+ return 0;
+
default:
return -ENOENT;
}
diff --git a/include/kvm/arm_hypercalls.h b/include/kvm/arm_hypercalls.h
index bbb3b12b10e3..d8c17d161ee5 100644
--- a/include/kvm/arm_hypercalls.h
+++ b/include/kvm/arm_hypercalls.h
@@ -12,6 +12,9 @@
#define ARM_SMCCC_STD_HYP_FEATURES \
GENMASK_ULL(KVM_REG_ARM_STD_HYP_BMAP_MAX - 1, 0)
+#define ARM_SMCCC_VENDOR_HYP_FEATURES \
+ GENMASK_ULL(KVM_REG_ARM_VENDOR_HYP_BMAP_MAX - 1, 0)
+
int kvm_hvc_call_handler(struct kvm_vcpu *vcpu);
static inline u32 smccc_get_function(struct kvm_vcpu *vcpu)
--
2.33.1.1089.g2158813163f-goog
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
next prev parent reply other threads:[~2021-11-02 0:22 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-02 0:21 [RFC PATCH 0/8] KVM: arm64: Add support for hypercall services selection Raghavendra Rao Ananta
2021-11-02 0:21 ` [RFC PATCH 1/8] KVM: arm64: Factor out firmware register handling from psci.c Raghavendra Rao Ananta
2021-11-03 21:43 ` Oliver Upton
2021-11-04 17:16 ` Raghavendra Rao Ananta
2021-11-08 21:33 ` Oliver Upton
2021-11-02 0:21 ` [RFC PATCH 2/8] KVM: arm64: Setup base for hypercall firmware registers Raghavendra Rao Ananta
2021-11-03 22:18 ` Oliver Upton
2021-11-04 19:04 ` Raghavendra Rao Ananta
2021-11-02 0:21 ` [RFC PATCH 3/8] KVM: arm64: Add standard secure service calls firmware register Raghavendra Rao Ananta
2021-11-04 0:15 ` Oliver Upton
2021-11-04 18:00 ` Raghavendra Rao Ananta
2021-11-02 0:21 ` [RFC PATCH 4/8] KVM: arm64: Add standard hypervisor " Raghavendra Rao Ananta
2021-11-02 0:22 ` Raghavendra Rao Ananta [this message]
2021-11-02 0:22 ` [RFC PATCH 6/8] tools: Import the firmware registers Raghavendra Rao Ananta
2021-11-04 0:23 ` Oliver Upton
2021-11-04 18:58 ` Raghavendra Rao Ananta
2021-11-02 0:22 ` [RFC PATCH 7/8] tools: Import ARM SMCCC definitions Raghavendra Rao Ananta
2021-11-02 0:22 ` [RFC PATCH 8/8] selftests: KVM: aarch64: Introduce hypercall ABI test Raghavendra Rao Ananta
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20211102002203.1046069-6-rananta@google.com \
--to=rananta@google.com \
--cc=alexandru.elisei@arm.com \
--cc=catalin.marinas@arm.com \
--cc=drjones@redhat.com \
--cc=james.morse@arm.com \
--cc=kvm@vger.kernel.org \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=maz@kernel.org \
--cc=pshier@google.com \
--cc=suzuki.poulose@arm.com \
--cc=will@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).