[PATCH 2/3] arm64/kvm: Fine grain _EL2 system registers list that affect nested virtualization

[Miguel Luis]

Hi Marc,

> On 18 Sep 2023, at 09:40, Marc Zyngier <maz at kernel.org> wrote:
> 
> Hi Miguel,
> 
> On Wed, 13 Sep 2023 19:52:07 +0100,
> Miguel Luis <miguel.luis at oracle.com> wrote:
>> 
>> Some _EL1 registers got included in the _EL2 ranges, which are not
>> affected by NV. Remove them and fine grain the ranges to exclusively
>> include the _EL2 ones.
>> 
>> Signed-off-by: Miguel Luis <miguel.luis at oracle.com>
>> ---
>> arch/arm64/kvm/emulate-nested.c | 44 ++++++++++++++++++++++++++++-----
>> 1 file changed, 38 insertions(+), 6 deletions(-)
>> 
>> diff --git a/arch/arm64/kvm/emulate-nested.c b/arch/arm64/kvm/emulate-nested.c
>> index 9ced1bf0c2b7..9aa1c06abdb7 100644
>> --- a/arch/arm64/kvm/emulate-nested.c
>> +++ b/arch/arm64/kvm/emulate-nested.c
>> @@ -649,14 +649,46 @@ static const struct encoding_to_trap_config encoding_to_cgt[] __initconst = {
>> SR_TRAP(SYS_APGAKEYHI_EL1, CGT_HCR_APK),
>> /* All _EL2 registers */
>> SR_RANGE_TRAP(sys_reg(3, 4, 0, 0, 0),
>> -       sys_reg(3, 4, 3, 15, 7), CGT_HCR_NV),
>> + sys_reg(3, 4, 4, 0, 1), CGT_HCR_NV),
> 
> It would be good if the commit message explained that you are folding
> SPSR/ELR into the existing range. Also, please keep the two ends of
> the ranges vertically aligned.
> 

OK.

>> /* Skip the SP_EL1 encoding... */
>> - SR_TRAP(SYS_SPSR_EL2, CGT_HCR_NV),
>> - SR_TRAP(SYS_ELR_EL2, CGT_HCR_NV),
>> - SR_RANGE_TRAP(sys_reg(3, 4, 4, 1, 1),
>> -       sys_reg(3, 4, 10, 15, 7), CGT_HCR_NV),
>> + SR_RANGE_TRAP(sys_reg(3, 4, 4, 3, 0),
>> + sys_reg(3, 4, 10, 6, 7), CGT_HCR_NV),
>> + /* skip MECID_A0_EL2, MECID_A1_EL2, MECID_P0_EL2,
>> +  *      MECID_P1_EL2, MECIDR_EL2, VMECID_A_EL2,
>> +  *      VMECID_P_EL2.
>> +  */
> 
> Please follow the kernel comment format. Also, why are you skipping
> the MEC registers, but not the MPAM ones? At least indicate a
> rationale for this.
> 

I’m not aware of any exceptions for MPAM registers, although there are for MEC
when HCR_EL2.NV2 is 0.

>> SR_RANGE_TRAP(sys_reg(3, 4, 12, 0, 0),
>> -       sys_reg(3, 4, 14, 15, 7), CGT_HCR_NV),
>> + sys_reg(3, 4, 12, 1, 1), CGT_HCR_NV),
>> + /* ICH_AP0R<m>_EL2 */
>> + SR_RANGE_TRAP(SYS_ICH_AP0R0_EL2,
>> + SYS_ICH_AP0R3_EL2, CGT_HCR_NV),
>> + /* ICH_AP1R<m>_EL2 */
>> + SR_RANGE_TRAP(SYS_ICH_AP1R0_EL2,
>> + SYS_ICH_AP1R3_EL2, CGT_HCR_NV),
>> + SR_RANGE_TRAP(sys_reg(3, 4, 12, 9, 5),
>> + sys_reg(3, 4, 12, 11, 7), CGT_HCR_NV),
>> + /* ICH_LR<m>_EL2 */
>> + SR_TRAP(SYS_ICH_LR0_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR1_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR2_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR3_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR4_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR5_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR6_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR7_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR8_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR9_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR10_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR11_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR12_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR13_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR14_EL2, CGT_HCR_NV),
>> + SR_TRAP(SYS_ICH_LR15_EL2, CGT_HCR_NV),
> 
> You could describe all the LRs a single range.
> 

Should we skip the gap between LR7 - LR8 ?

>> + SR_RANGE_TRAP(sys_reg(3, 4, 13, 0, 1),
>> + sys_reg(3, 4, 13, 0, 7), CGT_HCR_NV),
>> + /* skip AMEVCNTVOFF0<n>_EL2 and AMEVCNTVOFF1<n>_EL2 */
> 
> Why?

I didn’t find its definition TBH although these could use a single range.

Thanks,

Miguel

> 
>> + SR_RANGE_TRAP(sys_reg(3, 4, 14, 0, 3),
>> + sys_reg(3, 4, 14, 5, 2), CGT_HCR_NV),
>> /* All _EL02, _EL12 registers */
>> SR_RANGE_TRAP(sys_reg(3, 5, 0, 0, 0),
>>       sys_reg(3, 5, 10, 15, 7), CGT_HCR_NV),
> 
> Thanks,
> 
> M.
> 
> -- 
> Without deviation from the norm, progress is not possible.