Re: [PATCH v8 34/43] arm64: RME: Propagate number of breakpoints and watchpoints to userspace

[Gavin Shan <gshan@redhat.com>]

On 4/16/25 11:41 PM, Steven Price wrote:
> From: Jean-Philippe Brucker <jean-philippe@linaro.org>
> 
> The RMM describes the maximum number of BPs/WPs available to the guest
> in the Feature Register 0. Propagate those numbers into ID_AA64DFR0_EL1,
> which is visible to userspace. A VMM needs this information in order to
> set up realm parameters.
> 
> Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
> Signed-off-by: Steven Price <steven.price@arm.com>
> ---
>   arch/arm64/include/asm/kvm_rme.h |  2 ++
>   arch/arm64/kvm/rme.c             | 22 ++++++++++++++++++++++
>   arch/arm64/kvm/sys_regs.c        |  2 +-
>   3 files changed, 25 insertions(+), 1 deletion(-)
> 

If I don't miss anything, it's not enough to apply the filter on reading and
resetting path where sanitise_id_aa64dfr0_el1() is called. id_aa64dfr0_el1
is writable and it's possible that QEMU modifies its value. Afterwards, the
register is read from guest kernel, which will be trapped to host and the
modified value is returned, without this filter applied. So I think the same
filter need to be applied to the write path originated from QEMU.

Thanks,
Gavin

> diff --git a/arch/arm64/include/asm/kvm_rme.h b/arch/arm64/include/asm/kvm_rme.h
> index f786fd978cf6..09cbb61816f3 100644
> --- a/arch/arm64/include/asm/kvm_rme.h
> +++ b/arch/arm64/include/asm/kvm_rme.h
> @@ -94,6 +94,8 @@ void kvm_init_rme(void);
>   u32 kvm_realm_ipa_limit(void);
>   u32 kvm_realm_vgic_nr_lr(void);
>   
> +u64 kvm_realm_reset_id_aa64dfr0_el1(const struct kvm_vcpu *vcpu, u64 val);
> +
>   bool kvm_rme_supports_sve(void);
>   
>   int kvm_realm_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap);
> diff --git a/arch/arm64/kvm/rme.c b/arch/arm64/kvm/rme.c
> index 297b13ef1729..0c358ce0a7a1 100644
> --- a/arch/arm64/kvm/rme.c
> +++ b/arch/arm64/kvm/rme.c
> @@ -87,6 +87,28 @@ u32 kvm_realm_vgic_nr_lr(void)
>   	return u64_get_bits(rmm_feat_reg0, RMI_FEATURE_REGISTER_0_GICV3_NUM_LRS);
>   }
>   
> +u64 kvm_realm_reset_id_aa64dfr0_el1(const struct kvm_vcpu *vcpu, u64 val)
> +{
> +	u32 bps = u64_get_bits(rmm_feat_reg0, RMI_FEATURE_REGISTER_0_NUM_BPS);
> +	u32 wps = u64_get_bits(rmm_feat_reg0, RMI_FEATURE_REGISTER_0_NUM_WPS);
> +	u32 ctx_cmps;
> +
> +	if (!kvm_is_realm(vcpu->kvm))
> +		return val;
> +
> +	/* Ensure CTX_CMPs is still valid */
> +	ctx_cmps = FIELD_GET(ID_AA64DFR0_EL1_CTX_CMPs, val);
> +	ctx_cmps = min(bps, ctx_cmps);
> +
> +	val &= ~(ID_AA64DFR0_EL1_BRPs_MASK | ID_AA64DFR0_EL1_WRPs_MASK |
> +		 ID_AA64DFR0_EL1_CTX_CMPs);
> +	val |= FIELD_PREP(ID_AA64DFR0_EL1_BRPs_MASK, bps) |
> +	       FIELD_PREP(ID_AA64DFR0_EL1_WRPs_MASK, wps) |
> +	       FIELD_PREP(ID_AA64DFR0_EL1_CTX_CMPs, ctx_cmps);
> +
> +	return val;
> +}
> +
>   static int get_start_level(struct realm *realm)
>   {
>   	return 4 - ((realm->ia_bits - 8) / (RMM_PAGE_SHIFT - 3));
> diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
> index de7fe024dbff..36e22ed84e7e 100644
> --- a/arch/arm64/kvm/sys_regs.c
> +++ b/arch/arm64/kvm/sys_regs.c
> @@ -1844,7 +1844,7 @@ static u64 sanitise_id_aa64dfr0_el1(const struct kvm_vcpu *vcpu, u64 val)
>   	/* Hide BRBE from guests */
>   	val &= ~ID_AA64DFR0_EL1_BRBE_MASK;
>   
> -	return val;
> +	return kvm_realm_reset_id_aa64dfr0_el1(vcpu, val);
>   }
>   
>   static int set_id_aa64dfr0_el1(struct kvm_vcpu *vcpu,