mbox series

[v6,0/2] KVM: arm64: mixed-width check should be skipped for uninitialized vCPUs

Message ID 20220329031924.619453-1-reijiw@google.com (mailing list archive)
Headers show
Series KVM: arm64: mixed-width check should be skipped for uninitialized vCPUs | expand

Message

Reiji Watanabe March 29, 2022, 3:19 a.m. UTC
KVM allows userspace to configure either all EL1 32bit or 64bit vCPUs
for a guest.  At vCPU reset, vcpu_allowed_register_width() checks
if the vcpu's register width is consistent with all other vCPUs'.
Since the checking is done even against vCPUs that are not initialized
(KVM_ARM_VCPU_INIT has not been done) yet, the uninitialized vCPUs
are erroneously treated as 64bit vCPU, which causes the function to
incorrectly detect a mixed-width VM.

This series will fix this problem by introducing a new VM flag that
indicates the guest needs to be configured with all 32bit or 64bit
vCPUs and checking vcpu's register width against the new flag at
the vcpu's KVM_ARM_VCPU_INIT (instead of against other vCPUs'
register width).

Patch-1 introduces KVM_ARCH_FLAG_EL1_32BIT and
KVM_ARCH_FLAG_REG_WIDTH_CONFIGURED bits for kvm->arch.flags and
uses them to check vcpu's register width to fix the problem.

Patch-2 introduces a selftest that can test non-mixed-width vCPUs (all
64bit vCPUs or all 32bit vcPUs) can be configured, and mixed-width
vCPUs cannot be configured.

The series is based on kvmarm/fixes at:

  commit 8872d9b3e35a ("KVM: arm64: Drop unneeded minor version check from PSCI v1.x handler")

v6:
  - Fix typo and minor nits in the selftests [Oliver]

v5: https://lore.kernel.org/all/20220321050804.2701035-1-reijiw@google.com/
  - Rebase to kvmarm/next (and drop the patch-1 "KVM: arm64: Generalise
    VM features into a set of flags")
  - Use kernel-doc style comments for kvm_set_vm_width() [Oliver]
  - Change kvm_set_vm_width() to use if/else instead of a ternary
    operator for KVM_ARCH_FLAG_EL1_32BIT check [Oliver]

v4: https://lore.kernel.org/all/20220314061959.3349716-1-reijiw@google.com/
  - Use different implementation of vcpu_el1_is_32bit() depending on
    the context. [Marc]
  - Rename kvm_register_width_check_or_init() to kvm_set_vm_width(), and
    call it from kvm_rest_vcpu() instead of from kvm_vcpu_set_target()
  - Remove vcpu_allowed_register_width(), and does the same checking
    in kvm_set_vm_width() instead.

v3: https://lore.kernel.org/all/20220303035408.3708241-1-reijiw@google.com/
  - Introduced 'flags' to kvm_arch, and use bits of the flags for
    a set of booleans for VM feature.
  - Changed 'el1_reg_width' to two bits of 'flags' of kvm_arch.

v2: https://lore.kernel.org/all/20220118041923.3384602-1-reijiw@google.com/
  - Introduced 'el1_reg_width' for kvm_arch and use it to check vcpu's
    register width against the flag at the vcpu's KVM_ARM_VCPU_INIT.

v1: https://lore.kernel.org/all/20220110054042.1079932-1-reijiw@google.com/

[1] https://lore.kernel.org/all/20210715163159.1480168-2-maz@kernel.org/

Reiji Watanabe (2):
  KVM: arm64: mixed-width check should be skipped for uninitialized
    vCPUs
  KVM: arm64: selftests: Introduce vcpu_width_config

 arch/arm64/include/asm/kvm_emulate.h          |  27 ++--
 arch/arm64/include/asm/kvm_host.h             |  10 ++
 arch/arm64/kvm/reset.c                        |  65 ++++++----
 tools/testing/selftests/kvm/.gitignore        |   1 +
 tools/testing/selftests/kvm/Makefile          |   1 +
 .../selftests/kvm/aarch64/vcpu_width_config.c | 122 ++++++++++++++++++
 6 files changed, 196 insertions(+), 30 deletions(-)
 create mode 100644 tools/testing/selftests/kvm/aarch64/vcpu_width_config.c

Comments

Marc Zyngier April 6, 2022, 1:21 p.m. UTC | #1
On Mon, 28 Mar 2022 20:19:22 -0700, Reiji Watanabe wrote:
> KVM allows userspace to configure either all EL1 32bit or 64bit vCPUs
> for a guest.  At vCPU reset, vcpu_allowed_register_width() checks
> if the vcpu's register width is consistent with all other vCPUs'.
> Since the checking is done even against vCPUs that are not initialized
> (KVM_ARM_VCPU_INIT has not been done) yet, the uninitialized vCPUs
> are erroneously treated as 64bit vCPU, which causes the function to
> incorrectly detect a mixed-width VM.
> 
> [...]

Applied to fixes, thanks!

[1/2] KVM: arm64: mixed-width check should be skipped for uninitialized vCPUs
      commit: 26bf74bd9f6ff0f1545b4f0c92a37c232d076014
[2/2] KVM: arm64: selftests: Introduce vcpu_width_config
      commit: 2f5d27e6cf14efe652748bad89ee529ed5a5d577

Note that I have somewhat tweaked the first patch to my own liking.

Cheers,

	M.