> On 26 Jul 2026, at 8:59 PM, Eric Auger <[email protected]> wrote:
> 
> !-------------------------------------------------------------------|
>  CAUTION: External Email
> 
> |-------------------------------------------------------------------!
> 
> This series enhances the current host KVM model with capability to
> set writable ID reg fields. 
> 
> Since v6.7 kernel, KVM/arm allows the userspace to overwrite the values
> of a subset of ID regs. The list of writable fields continues to grow.
> The feature ID range is defined as the AArch64 System register space
> with op0==3, op1=={0, 1, 3}, CRn==0, CRm=={0-7}, op2=={0-7}.
> 
> The end goal is to get more flexibility when migrating guests
> between different host hardware.
> 
> QEMU retrieves the writable ID fields from KVM UAPI [1] and
> match them against a generated description of ID regs and their
> named fields that stem from AARCHMRS Registers.json file.
> Current description is based on latest 2026-03 edition.
> The content of the generated files was compared against kernel
> linux/arch/arm64/tools/sysreg file. It is not straightforward
> to have unit tests for python scripts as there are many cases for
> field extraction. 
> 
> For each writable named field a uint64 property is created
> following the "SYSREG_<REG>_<FIELD>" naming convention. REG and
> FIELD names are those described in ARM ARM Reference manual.
> 
> The list of SYSREG_ID properties can be retrieved through the qmp
> monitor using query-cpu-model-expansion [2].
> 
> For the record, Jinqian was able to migrate between Hisilicon KunPeng
> HIP09 and HIP12 chips with this series using this kind of command:
> 
> -cpu host,pauth=off,pmu=off,sve=off,\
> SYSREG_ID_AA64PFR1_EL1_NMI=0x0,\
> SYSREG_ID_AA64ISAR1_EL1_LS64=0x0,\
> SYSREG_ID_AA64ISAR1_EL1_XS=0x0,\
> SYSREG_ID_AA64ISAR1_EL1_LRCPC=0x2,\
> SYSREG_ID_AA64ISAR2_EL1_RPRFM=0x0,\
> SYSREG_ID_AA64ISAR2_EL1_CLRBHB=0x0,\
> SYSREG_ID_AA64ISAR2_EL1_PAC_frac=0x0,\
> SYSREG_ID_AA64ISAR2_EL1_BC=0x0,\
> SYSREG_ID_AA64ISAR2_EL1_RPRES=0x0,\
> SYSREG_ID_AA64ISAR2_EL1_WFxT=0x0,\
> SYSREG_ID_AA64MMFR0_EL1_FGT=0x0,\
> SYSREG_ID_AA64MMFR0_EL1_BigEnd=0x0,\
> SYSREG_ID_AA64MMFR1_EL1_ECBHB=0x0,\
> SYSREG_ID_AA64MMFR1_EL1_CMOW=0x0,\
> SYSREG_ID_AA64MMFR1_EL1_TIDCP1=0x0,\
> SYSREG_ID_AA64MMFR1_EL1_nTLBPA=0x0,\
> SYSREG_ID_AA64MMFR1_EL1_AFP=0x0,\
> SYSREG_ID_AA64MMFR1_EL1_HCX=0x0,\
> SYSREG_ID_AA64MMFR1_EL1_ETS=0x0,\
> SYSREG_ID_AA64MMFR1_EL1_PAN=0x2,\
> SYSREG_ID_AA64MMFR1_EL1_HAFDBS=0x2,\
> SYSREG_ID_AA64MMFR2_EL1_CnP=0x0,\
> SYSREG_ID_AA64MMFR3_EL1_TCRX=0x0,\
> SYSREG_ID_AA64DFR0_EL1_PMUVer=0x6,\
> SYSREG_ID_AA64DFR0_EL1_DebugVer=0x9,\
> SYSREG_ID_AA64DFR0_EL1_DoubleLock=0xf,\
> SYSREG_ID_AA64ZFR0_EL1_F64MM=0x0,\
> SYSREG_ID_AA64ZFR0_EL1_F32MM=0x0,\
> SYSREG_ID_AA64ZFR0_EL1_SM4=0x0,\
> SYSREG_ID_AA64ZFR0_EL1_SHA3=0x0,\
> SYSREG_ID_AA64ZFR0_EL1_BitPerm=0x0,\
> SYSREG_ID_AA64ZFR0_EL1_AES=0x0,\
> SYSREG_ID_AA64ZFR0_EL1_SVEver=0x0,\
> SYSREG_CTR_EL0_L1Ip=0x2 \
> 
> Connie & Eric
> 
> This series can be found at:
> https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_eauger_qemu_tree_arm-2Dcpu-2Dmodel-2Dv7&d=DwIDAg&c=s883GpUCOChKOHiocYtGcg&r=PGWMyignA0NiDmTlyP7vOTHozBws_VN86yrVmSMkBp0&m=xh5u7-r6auXiWQScafdj-0wW70mXB4FhClh58GryHmExkx8uZN-h9j4B2c90Lowd&s=Ro4v_mjoNaLsznyxm781lzXBcRtXTudjibmgIhoJsNM&e=
>  
> 
> History:
> --------
> v6 -> v7:
> - query-cpu-model-expansion now applies the sysreg props
>  onto a scratch vcpu (POC) allowing a finer validation of settings
>  (not perfect though as init may be revisited depending on
>  settings)
> - Properly handle ValueRange (Khushit)
> - Remove cpu parameter in get_host_cpu_idregs call chain
>  (Khushit)
> 
> v5 -> v6:
> v6 is mostly for Khushit to see how he may work on top
> of this series. It fulfills some of his requirements:
> - Upon Khushit request I added description for enum values
>  and reserved fields
> - set_sysreg_prop checks the input value against enum values
>  if any
> - writable_bitmap is now dispatched into the sysreg descripton
>  and not stored in vcpu state anymore (besides it is VM wide)
> - implemented write for qmp inspection. However this is not
>  yet checked against KVM. That's the next step I think, ie.
>  instantiate a scratch vcpu to test if the field value can be
>  applied?
> - if some inconsistencies between writable mask and idreg
>  field, reserved fields only produce traces
> - this version does not yet fix some reported issues about host
>  supporting nested virt. I will further sync with Khushit.
>  so this is definitively not candidate to be applied, hence the
>  RFC tag.
> 
> Cornelia Huck (3):
>  target/arm/kvm: Retrieve writable ID reg map
>  arm/cpu-features: document ID reg properties
>  arm-qmp-cmds: introspection for ID register props
> 
> Eric Auger (14):
>  scripts: introduce scripts/update-aarch64-cpu-sysregs-header.py
>  target/arm/cpu-sysregs.h.inc: Sort by name alphabetical order
>  target/arm/cpu-sysregs.h.inc: Update with automatic generation
>  arm/cpu: Add infra to handle generated ID register definitions
>  scripts: Introduce scripts/aarch64_sysreg_helpers module
>  scripts: Introduce scripts/update-aarch64-cpu-sysreg-properties.py
>  target/arm/cpu-idregs.h.inc: generate with script
>  target/arm/cpu-idregs.h.inc: Generate enum values
>  arm/kvm: Initialize all writable ID registers from host
>  target/arm/kvm: Introduce kvm_arm_expose_idreg_properties
>  target/arm/cpu: Expose writable ID reg field properties on the kvm
>    host vcpu model
>  target/arm/cpu-idregs.h.inc: Generate reserved fields
>  target/arm/kvm: Ignore and trace unexpected writable reserved fields
>  target/arm/kvm: add utility to write idregs in scratch vcpu
> 
> Shaju Abraham (1):
>  target/arm/cpu_idregs: generate tables for Arm64 ID registers and
>    fields
> 
> docs/system/arm/cpu-features.rst              |  106 +-
> target/arm/cpu-idregs.h                       |   41 +
> target/arm/kvm_arm.h                          |   13 +
> target/arm/cpu-idregs.h.inc                   | 2164 +++++++++++++++++
> target/arm/cpu-sysregs.h.inc                  |   57 +-
> target/arm/arm-qmp-cmds.c                     |   98 +
> target/arm/cpu-idregs.c                       |   96 +
> target/arm/cpu64.c                            |    5 +
> target/arm/kvm-stub.c                         |    6 +
> target/arm/kvm.c                              |  360 ++-
> scripts/aarch64_sysreg_helpers.py             |  109 +
> .../update-aarch64-cpu-sysreg-properties.py   |  290 +++
> scripts/update-aarch64-cpu-sysregs-header.py  |   51 +
> target/arm/meson.build                        |    1 +
> target/arm/trace-events                       |    8 +
> 15 files changed, 3373 insertions(+), 32 deletions(-)
> create mode 100644 target/arm/cpu-idregs.h
> create mode 100644 target/arm/cpu-idregs.h.inc
> create mode 100644 target/arm/cpu-idregs.c
> create mode 100644 scripts/aarch64_sysreg_helpers.py
> create mode 100644 scripts/update-aarch64-cpu-sysreg-properties.py
> create mode 100755 scripts/update-aarch64-cpu-sysregs-header.py
> 
> -- 
> 2.53.0
> 

Hi,

On my box the build is failing with:
[97/2192] Linking target qemu-system-arm
FAILED: qemu-system-arm
cc @qemu-system-arm.rsp
/usr/bin/ld: libsystem_arm.a.p/target_arm_arm-qmp-cmds.c.o: in function 
`qmp_query_cpu_model_expansion':
/home/ahv/qemu/build/../target/arm/arm-qmp-cmds.c:144:(.text+0x4d4): undefined 
reference to `kvm_check_extension'
/usr/bin/ld: 
/home/ahv/qemu/build/../target/arm/arm-qmp-cmds.c:145:(.text+0x4e4): undefined 
reference to `kvm_check_extension'
/usr/bin/ld: 
/home/ahv/qemu/build/../target/arm/arm-qmp-cmds.c:151:(.text+0x514): undefined 
reference to `kvm_check_extension'
/usr/bin/ld: 
/home/ahv/qemu/build/../target/arm/arm-qmp-cmds.c:168:(.text+0x554): undefined 
reference to `kvm_arm_create_scratch_host_vcpu'
/usr/bin/ld: 
/home/ahv/qemu/build/../target/arm/arm-qmp-cmds.c:152:(.text+0x7d4): undefined 
reference to `kvm_check_extension'
/usr/bin/ld: 
/home/ahv/qemu/build/../target/arm/arm-qmp-cmds.c:247:(.text+0x8f4): undefined 
reference to `kvm_arm_destroy_scratch_host_vcpu'
collect2: error: ld returned 1 exit status


Reply via email to