@@ -154,6 +154,13 @@ size of the address translated by the stage2 level (guest physical to
host physical address translations).
+Also on arm64, if capability KVM_CAP_ARM_SVE is present then the
+KVM_VM_TYPE_ARM_SVE flag may be set in the machine type identifier to
+enable the KVM API extensions for the Arm Scalable Vector Extension
+(SVE) for the created VM. This is required in order to create vcpus
+that support SVE. See section 4.117 (KVM_ARM_SVE_CONFIG) for details.
+
+
4.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATURE_INDEX_LIST
Capability: basic, KVM_CAP_GET_MSR_FEATURES for KVM_GET_MSR_FEATURE_INDEX_LIST
@@ -2099,13 +2106,21 @@ Specifically:
0x6030 0000 0010 004c SPSR_UND 64 spsr[KVM_SPSR_UND]
0x6030 0000 0010 004e SPSR_IRQ 64 spsr[KVM_SPSR_IRQ]
0x6060 0000 0010 0050 SPSR_FIQ 64 spsr[KVM_SPSR_FIQ]
- 0x6040 0000 0010 0054 V0 128 fp_regs.vregs[0]
- 0x6040 0000 0010 0058 V1 128 fp_regs.vregs[1]
+ 0x6040 0000 0010 0054 V0 128 fp_regs.vregs[0] (*)
+ 0x6040 0000 0010 0058 V1 128 fp_regs.vregs[1] (*)
...
- 0x6040 0000 0010 00d0 V31 128 fp_regs.vregs[31]
+ 0x6040 0000 0010 00d0 V31 128 fp_regs.vregs[31] (*)
0x6020 0000 0010 00d4 FPSR 32 fp_regs.fpsr
0x6020 0000 0010 00d5 FPCR 32 fp_regs.fpcr
+(*) These encodings are not accepted for SVE-enabled vcpus. See
+ KVM_ARM_SVE_CONFIG for details of how SVE support is configured for
+ a vcpu.
+
+ The equivalent register content can be accessed via bits [127:0] of
+ the corresponding SVE Zn registers instead for vcpus that have SVE
+ enabled (see below).
+
arm64 CCSIDR registers are demultiplexed by CSSELR value:
0x6020 0000 0011 00 <csselr:8>
@@ -2115,6 +2130,14 @@ arm64 system registers have the following id bit patterns:
arm64 firmware pseudo-registers have the following bit pattern:
0x6030 0000 0014 <regno:16>
+arm64 SVE registers have the following bit patterns:
+ 0x6080 0000 0015 00 <n:5> <slice:5> Zn bits[2048*slice + 2047 : 2048*slice]
+ 0x6050 0000 0015 04 <n:4> <slice:5> Pn bits[256*slice + 255 : 256*slice]
+ 0x6050 0000 0015 060 <slice:5> FFR bits[256*slice + 255 : 256*slice]
+
+ These registers are only accessible on SVE-enabled vcpus. See
+ KVM_ARM_SVE_CONFIG for details.
+
MIPS registers are mapped using the lower 32 bits. The upper 16 of that is
the register group type:
@@ -2632,6 +2655,8 @@ Parameters: struct kvm_vcpu_init (in)
Returns: 0 on success; -1 on error
Errors:
EINVAL: the target is unknown, or the combination of features is invalid.
+ EBADFD: further configuration required before this ioctl (see sections
+ 4.2 KVM_CREATE_VM, 4.117 KVM_ARM_SVE_CONFIG)
ENOENT: a features bit specified is unknown.
This tells KVM what type of CPU to present to the guest, and what
@@ -2694,6 +2719,8 @@ Returns: 0 on success; -1 on error
Errors:
E2BIG: the reg index list is too big to fit in the array specified by
the user (the number required will be written into n).
+ EBADFD: (arm64) further configuration required before this ioctl (see
+ sections 4.2 KVM_CREATE_VM, 4.117 KVM_ARM_SVE_CONFIG)
struct kvm_reg_list {
__u64 n; /* number of registers in reg[] */
@@ -3777,6 +3804,149 @@ Coalesced pio is based on coalesced mmio. There is little difference
between coalesced mmio and pio except that coalesced pio records accesses
to I/O ports.
+4.117 KVM_ARM_SVE_CONFIG
+
+Capability: KVM_CAP_ARM_SVE
+Architectures: arm64
+Type: vm and vcpu ioctl
+Parameters: struct kvm_sve_vls (in/out)
+Returns: 0 on success
+Errors:
+ EINVAL: Unrecognised subcommand or bad arguments, or SVE API not enabled
+ (see section 2.4, KVM_CREATE_VM)
+ EBADFD: vcpu in wrong state for request
+ (KVM_ARM_SVE_CONFIG_SET, KVM_ARM_SVE_CONFIG_SET)
+ ENOMEM: Out of memory
+ EFAULT: Bad user address
+
+struct kvm_sve_vls {
+ __u16 cmd;
+ __u16 max_vq;
+ __u16 _reserved[2];
+ __u64 required_vqs[8];
+};
+
+General:
+
+In addition to requiring KVM_CAP_ARM_SVE, this ioctl is only available
+when the SVE API extensions have been enabled by creating the
+corresponding VM with the KVM_VM_TYPE_ARM_SVE flag. See section 4.2
+(KVM_CREATE_VM) for details.
+
+This should be the first vcpu ioctl issued after creating the vcpu via
+KVM_CREATE_VCPU: until SVE configuration for the vcpu is completed via a
+successful KVM_ARM_SVE_CONFIG_SET subcommand (see below), non-trivial
+vcpu ioctls will be rejected with EBADFD or another appropriate error.
+
+Parameters:
+
+cmd: This ioctl supports a few different subcommands, selected by the
+value of cmd (described in detail in the following sections).
+
+_reserved[]: these fields may be meaningful to later kernels. For
+forward compatibility, they must be zeroed before invoking this ioctl
+for the first time on a given struct kvm_sve_vls object. (So, memset()
+it to zero before first use, or allocate with calloc() for example.)
+
+max_vq, required_vqs[]:
+
+If max_vq == 0, SVE is disabled for this vcpu.
+
+Otherwise, max_vq and required_vqs[] encode a set of SVE vector
+lengths to attempt to configure for this vcpu. The set is encoded as
+follows:
+
+If (a * 64 + b + 1) <= max_vq, then the bit represented by
+
+ required_vqs[a] & ((__u64)1 << b)
+
+(where a is in the range 0..7 and b is in the range 0..63)
+indicates that the vector length (a * 64 + b + 1) * 128 bits is
+supported (KVM_ARM_SVE_CONFIG_QUERY, KVM_ARM_SVE_CONFIG_GET) or required
+(KVM_ARM_SVE_CONFIG_SET).
+
+If (a * 64 + b + 1) > max_vq, then the vector length
+(a * 64 + b + 1) * 128 bits is unsupported or prohibited respectively.
+In other words, only the first max_vq bits in required_vqs[] are
+significant; remaining bits are implicitly treated as if they were zero.
+
+max_vq must be in the range SVE_VQ_MIN (1) to SVE_VQ_MAX (512).
+
+See Documentation/arm64/sve.txt for an explanation of vector lengths and
+the meaning associated with "VQ".
+
+Subcommands:
+
+/* values for cmd: */
+#define KVM_ARM_SVE_CONFIG_QUERY 0 /* query what the host can support */
+#define KVM_ARM_SVE_CONFIG_SET 1 /* enable SVE for vcpu and set VLs */
+#define KVM_ARM_SVE_CONFIG_GET 2 /* read the set of VLs for a vcpu */
+
+Subcommand details:
+
+4.117.1 KVM_ARM_SVE_CONFIG_QUERY
+Type: vm and vcpu
+
+Retrieve the full set of SVE vector lengths available for use by KVM
+guests on this host. The result is independent of which vcpu this
+command is invoked on. As a convenience, it may also be invoked on a
+vm file descriptor, eliminating the need to create a vcpu first.
+
+4.117.2 KVM_ARM_SVE_CONFIG_SET
+Type: vcpu only
+
+Sets whether SVE is enabled for the vcpu, and if so sets the set of
+SVE vector lengths that will be visible to the guest.
+
+This is the only way to enable SVE for a vcpu: if this command is not
+invoked for a vcpu then SVE will not be available to the guest on this
+vcpu.
+
+This subcommand is only permitted once per vcpu, before KVM_RUN has been
+invoked for the vcpu for the first time. Otherwise, the command fails
+with -EBADFD and the state of the vcpu is not modified.
+
+In typical use, the user should call KVM_ARM_SVE_CONFIG_QUERY first to
+populate a struct kvm_sve_vls with the full set of vector lengths
+available on the host, then set cmd = KVM_ARM_SVE_CONFIG_SET and
+re-issue the KVM_ARM_SVE_CONFIG ioctl on the desired vcpu. This will
+configure the best set of vector lengths available. When following this
+approach, the maximum available vector length can also be restricted by
+reducing the value of max_vq before invoking KVM_ARM_SVE_CONFIG_SET.
+
+Every requested vector length in the struct kvm_sve_vls argument must be
+supported by the hardware. In addition, except for vector lengths
+greater than the maximum requested vector length, every vector length
+not requested must *not* be supported by the hardware. (The latter
+restriction may be relaxed in the future.) If the requested set of
+vector lengths is not supportable, the command fails with -EINVAL and
+the state of the vcpu is not modified.
+
+Different vcpus of a vm may be configured with different sets of vector
+lengths. Equally, some vcpus may have SVE enabled and some not.
+However, such configurations are not recommended except for testing and
+experimentation purposes. Architecturally compliant guest OSes will
+work, but may or may not make effective use of the resulting
+configuration.
+
+After a successful KVM_ARM_SVE_CONFIG_SET, KVM_ARM_SVE_CONFIG_GET can be
+used to retrieve the configured set of vector lengths.
+
+4.117.3 KVM_ARM_SVE_CONFIG_GET
+Type: vcpu only
+
+This subcommand returns the set of vector lengths enabled for the vcpu.
+SVE must have been disabled or enabled and configured for this vcpu by a
+successful prior KVM_ARM_SVE_CONFIG_SET call. Otherwise, -EBADFD is
+returned.
+
+If SVE is disabled for this vcpu, this subcommand will yield
+max_vq == 0; otherwise max_vq and required_vqs[] indicate the
+(non-empty) set of configured vector lengths.
+
+The state of the vcpu is unchanged.
+
+
5. The kvm_run structure
------------------------
This patch adds sections to the KVM API documentation describing the extensions for supporting the Scalable Vector Extension (SVE) in guests. Signed-off-by: Dave Martin <Dave.Martin@arm.com> --- Changes since RFC v2: * Fix documentation regarding which SVE Zn register bits must be accessed in order to get at Vn on an SVE-enabled vcpu. * Update documentation to describe KVM_VM_TYPE_ARM_SVE and related API changes. * Move comment about max_vq==0 semantics for KVM_ARM_SVE_CONFIG_GET to the correct place. * Miscellaneous wording updates to describe the new initialisation semantics. The documentation remains inaccurate / misleading in places. Since the current API design is still needs discussion I expect another respin with API changes. I didn't want to waste effort on a time-consuming documentation rewrite in the meantime. --- Documentation/virtual/kvm/api.txt | 176 +++++++++++++++++++++++++++++++++++++- 1 file changed, 173 insertions(+), 3 deletions(-)