linux-kernel - Re: [PATCH 1/4] KVM: arm64: Allow userspace to change MIDR

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <99f6301e-69dd-47e6-2cc6-10f44f6760cd@redhat.com>
Date: Tue, 11 Feb 2025 13:43:18 +0100 (CET)
From: Sebastian Ott <sebott@...hat.com>
To: Oliver Upton <oliver.upton@...ux.dev>
cc: Marc Zyngier <maz@...nel.org>, Joey Gouly <joey.gouly@....com>, 
    Suzuki K Poulose <suzuki.poulose@....com>, 
    Zenghui Yu <yuzenghui@...wei.com>, 
    Catalin Marinas <catalin.marinas@....com>, Will Deacon <will@...nel.org>, 
    Shameer Kolothum <shameerali.kolothum.thodi@...wei.com>, 
    Cornelia Huck <cohuck@...hat.com>, Eric Auger <eric.auger@...hat.com>, 
    linux-arm-kernel@...ts.infradead.org, kvmarm@...ts.linux.dev, 
    linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/4] KVM: arm64: Allow userspace to change MIDR_EL1

Hi Oliver,

On Mon, 10 Feb 2025, Oliver Upton wrote:
> On Mon, Feb 10, 2025 at 04:49:50PM +0100, Sebastian Ott wrote:
>> Enable VMMs to write MIDR_EL1 by treating it as a VM ID register.
>> Since MIDR_EL1 is not handled as a proper arm64_ftr_reg apply only
>> a sanity check against the writable mask to ensure the reserved
>> bits are 0.
>
> How exactly does the VMM's MIDR_EL1 find its way to the guest? VPIDR_EL2
> is still set to the hardware value.

Ouch. Completely missed that part, sry.

>
>> @@ -2021,7 +2037,11 @@ static int set_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
>>  		return ret;
>>  	}
>>
>> -	ret = arm64_check_features(vcpu, rd, val);
>> +	if (skip_feature_check(id))
>> +		ret = arm64_check_mask(rd, val);
>> +	else
>> +		ret = arm64_check_features(vcpu, rd, val);
>> +
>
> Can you add a new implementation of ->set_user() for MIDR/REVIDR/AIDR
> instead?

Yes, sure.

>> @@ -2542,6 +2571,8 @@ static const struct sys_reg_desc sys_reg_descs[] = {
>>
>>  	{ SYS_DESC(SYS_DBGVCR32_EL2), undef_access, reset_val, DBGVCR32_EL2, 0 },
>>
>> +	{ ID_DESC(MIDR_EL1), .set_user = set_id_reg, .visibility = id_visibility,
>> +	  .reset = reset_midr_el1, .val = (u32)-1 },
>
> nit: GENMASK() instead of truncation by casting.

All done. I add a test and send out V2.
Thanks a lot!
Sebastian