[PATCH] KVM: arm64: selftests: Sync ID_AA64PFR1, MPIDR, CLIDR in guest

Zenghui Yu yuzenghui at huawei.com
Mon Oct 13 23:59:21 PDT 2025


On 2025/10/14 0:14, Ben Horgan wrote:
> 
> 
> On 10/13/25 13:20, Ben Horgan wrote:
> > Hi Zenghui,
> >
> > On 10/12/25 16:43, Zenghui Yu wrote:
> > > We forgot to sync several registers (ID_AA64PFR1, MPIDR, CLIDR) in guest to
> > > make sure that the guest had seen the written value.
> > >
> > > Add them to the list.
> > >
> > > Signed-off-by: Zenghui Yu <zenghui.yu at linux.dev>
> > > ---
> > >  tools/testing/selftests/kvm/arm64/set_id_regs.c | 3 +++
> > >  1 file changed, 3 insertions(+)
> > >
> > > diff --git a/tools/testing/selftests/kvm/arm64/set_id_regs.c b/tools/testing/selftests/kvm/arm64/set_id_regs.c
> > > index 8ff1e853f7f8..5e24f77868b5 100644
> > > --- a/tools/testing/selftests/kvm/arm64/set_id_regs.c
> > > +++ b/tools/testing/selftests/kvm/arm64/set_id_regs.c
> > > @@ -249,11 +249,14 @@ static void guest_code(void)
> > >  	GUEST_REG_SYNC(SYS_ID_AA64ISAR2_EL1);
> > >  	GUEST_REG_SYNC(SYS_ID_AA64ISAR3_EL1);
> > >  	GUEST_REG_SYNC(SYS_ID_AA64PFR0_EL1);
> > > +	GUEST_REG_SYNC(SYS_ID_AA64PFR1_EL1);
> > >  	GUEST_REG_SYNC(SYS_ID_AA64MMFR0_EL1);
> > >  	GUEST_REG_SYNC(SYS_ID_AA64MMFR1_EL1);
> > >  	GUEST_REG_SYNC(SYS_ID_AA64MMFR2_EL1);
> > >  	GUEST_REG_SYNC(SYS_ID_AA64MMFR3_EL1);
> > >  	GUEST_REG_SYNC(SYS_ID_AA64ZFR0_EL1);
> > > +	GUEST_REG_SYNC(SYS_MPIDR_EL1);
> > > +	GUEST_REG_SYNC(SYS_CLIDR_EL1);
> >
> > Why did you choose this position in the list for these 2?

Technically there are no restrictions on the order. I just want it to
look consistent with test_reset_preserves_id_regs().

> >
> >>  	GUEST_REG_SYNC(SYS_CTR_EL0);
> >>  	GUEST_REG_SYNC(SYS_MIDR_EL1);
> >>  	GUEST_REG_SYNC(SYS_REVIDR_EL1);
> >
> > Looks good to me.
> >
> > Reviewed-By: Ben Horgan <ben.horgan at arm.com>

Thanks!

> On running this, I note test_guest_reg_read() doesn't announce its
> presence. How about adding a ksft_test_result_pass("%s\n", __func__) and
> bumping the number of tests?

Yup, I agreed.

Thanks,
Zenghui



More information about the linux-arm-kernel mailing list