On Fri, Mar 03 2023, Peter Maydell <peter.maydell@xxxxxxxxxx> wrote:
> On Tue, 28 Feb 2023 at 15:02, Cornelia Huck <cohuck@xxxxxxxxxx> wrote:
>>
>> Introduce a new cpu feature flag to control MTE support. To preserve
>> backwards compatibility for tcg, MTE will continue to be enabled as
>> long as tag memory has been provided.
>>
>> If MTE has been enabled, we need to disable migration, as we do not
>> yet have a way to migrate the tags as well. Therefore, MTE will stay
>> off with KVM unless requested explicitly.
>>
>> Signed-off-by: Cornelia Huck <cohuck@xxxxxxxxxx>
>> ---
>> docs/system/arm/cpu-features.rst | 21 ++++++
>> hw/arm/virt.c | 2 +-
>> target/arm/cpu.c | 18 ++---
>> target/arm/cpu.h | 1 +
>> target/arm/cpu64.c | 110 +++++++++++++++++++++++++++++++
>> target/arm/internals.h | 1 +
>> target/arm/kvm.c | 29 ++++++++
>> target/arm/kvm64.c | 5 ++
>> target/arm/kvm_arm.h | 19 ++++++
>> target/arm/monitor.c | 1 +
>> 10 files changed, 194 insertions(+), 13 deletions(-)
>
>
>
>> +static inline bool arm_machine_has_tag_memory(void)
>> +{
>> +#ifndef CONFIG_USER_ONLY
>> + Object *obj = object_dynamic_cast(qdev_get_machine(), TYPE_VIRT_MACHINE);
>> +
>> + /* so far, only the virt machine has support for tag memory */
>> + if (obj) {
>> + VirtMachineState *vms = VIRT_MACHINE(obj);
>> +
>> + return vms->mte;
>> + }
>
> Code inside target/arm shouldn't be fishing around inside
> the details of the board model like this. For TCG I think that
> at this point (i.e. at realize) you should be able to tell if
> the board has set up tag memory, because it will have set
> cpu->tag_memory to non-NULL.
I agree that we shouldn't need to poke into the machine innards here,
but I found that it was actually too early to check for cpu->tag_memory --
details have unfortunately been flushed out of my cache already, can try
to repopulate.
