> On 16 Sep 2026, at 8:15 PM, Eric Auger <[email protected]> wrote:
>
> !-------------------------------------------------------------------|
> CAUTION: External Email
>
> |-------------------------------------------------------------------!
>
> From: Cornelia Huck <[email protected]>
>
> Add an helper to retrieve the writable id reg bitmask.
s/an/a
> then retrieve the writable_map array and dispatch it in the writable_map
s/writable_map/writable_mask
> fields of arm64_id_regs[] elements.
s/fields/field
> Signed-off-by: Eric Auger <[email protected]>
> Signed-off-by: Cornelia Huck <[email protected]>
>
> ---
>
> v7 -> v8
> - added kvm_enabled()
>
> v5 -> v6
> - use arm64_id_regs[].writable_map
> - squashed 2 commits (helper + user)
> - removed stub
> ---
> target/arm/kvm.c | 61 ++++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 61 insertions(+)
>
> diff --git a/target/arm/kvm.c b/target/arm/kvm.c
> index ed99be7fd8..c6eac41991 100644
> --- a/target/arm/kvm.c
> +++ b/target/arm/kvm.c
> @@ -28,6 +28,7 @@
> #include "kvm_arm.h"
> #include "cpu.h"
> #include "cpu-sysregs.h"
> +#include "cpu-idregs.h"
> #include "trace.h"
> #include "internals.h"
> #include "hw/pci/pci.h"
> @@ -51,6 +52,7 @@ const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
> static bool cap_has_mp_state;
> static bool cap_has_inject_serror_esr;
> static bool cap_has_inject_ext_dabt;
> +static bool writable_map_dispatched;
>
> /**
> * ARMHostCPUFeatures: information about the host CPU (identified
> @@ -273,6 +275,21 @@ static uint32_t kvm_arm_sve_get_vls(int fd)
> return vls[0] & MAKE_64BIT_MASK(0, ARM_MAX_VQ);
> }
>
> +static int kvm_feature_idx_to_idregs_idx(int kidx)
> +{
> + int op1, crm, op2;
> + ARMSysRegs sysreg;
> +
> + op1 = kidx / 64;
> + if (op1 == 2) {
> + op1 = 3;
> + }
> + crm = (kidx % 64) / 8;
> + op2 = kidx % 8;
> + sysreg = ENCODE_ID_REG(3, op1, 0, crm, op2);
> + return get_sysreg_idx(sysreg);
> +}
> +
> static void kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf)
> {
> /* Identify the feature bits corresponding to the host CPU, and
> @@ -480,9 +497,53 @@ static void
> kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf)
> ahcf->features = features;
> }
>
> +static int kvm_arm_get_writable_id_regs(uint64_t *idregmap)
> +{
> + int cap_writable_id_regs;
> + struct reg_mask_range range = {
> + .range = KVM_ARM_FEATURE_ID_RANGE,
> + .reserved = {0},
> + .addr = (uint64_t)idregmap,
> + };
> +
> + cap_writable_id_regs =
> + kvm_check_extension(kvm_state,
> KVM_CAP_ARM_SUPPORTED_REG_MASK_RANGES);
> +
> + if (!cap_writable_id_regs ||
> + !(cap_writable_id_regs & (1 << KVM_ARM_FEATURE_ID_RANGE))) {
> + return -ENOSYS;
> + }
> +
> + if (kvm_vm_ioctl(kvm_state, KVM_ARM_GET_REG_WRITABLE_MASKS, &range)) {
> + return -errno;
> + }
I think just doing return kvm_vm_ioctl(kvm_state,
KVM_ARM_GET_REG_WRITABLE_MASKS, &range) is enough.
> + return 0;
> +}
> +
> void kvm_arm_set_cpu_features_from_host(ARMCPU *cpu)
> {
> CPUARMState *env = &cpu->env;
> + uint64_t *writable_map;
> +
> + writable_map = g_new(uint64_t, KVM_ARM_FEATURE_ID_RANGE_SIZE);
> +
> + if (kvm_enabled() && !writable_map_dispatched &&
> + !kvm_arm_get_writable_id_regs(writable_map)) {
> + for (int i = 0; i < KVM_ARM_FEATURE_ID_RANGE_SIZE; i++) {
> + uint64_t mask = writable_map[i];
> +
> + if (mask) {
> + int idx = kvm_feature_idx_to_idregs_idx(i);
> +
> + if (idx < 0 || idx > ARRAY_SIZE(arm64_id_regs)) {
Bug: >=
Also, this condition will never be true, as get_sysreg_idx() asserts out on an
unknown ID register. Which probably
should not happen.
> + continue;
> + }
> + arm64_id_regs[idx].writable_mask = mask;
> + }
> + }
> + writable_map_dispatched = true;
What is the need of writable_map_dispatched, why can’t we use
arm_host_cpu_features.dtb_compatible as below?
Warm Regards,
Khushit
> + }
> + g_free(writable_map);
>
> if (!arm_host_cpu_features.dtb_compatible) {
> kvm_arm_get_host_cpu_features(&arm_host_cpu_features);
> --
> 2.53.0
>
