Re: [PATCH] arm/kvm: add support for MTE

[Ganapatrao Kulkarni]

Hi Peter,

On 16-07-2024 09:15 pm, Peter Maydell wrote:
> In target/arm/cpu.c:arm_cpu_realizefn() there is this code:
>
>      if (cpu_isar_feature(aa64_mte, cpu)) {
>          /*
>           * The architectural range of GM blocksize is 2-6, however qemu
>           * doesn't support blocksize of 2 (see HELPER(ldgm)).
>           */
>          if (tcg_enabled()) {
>              assert(cpu->gm_blocksize >= 3 && cpu->gm_blocksize <= 6);
>          }
>
> #ifndef CONFIG_USER_ONLY
>          /*
>           * If we do not have tag-memory provided by the machine,
>           * reduce MTE support to instructions enabled at EL0.
>           * This matches Cortex-A710 BROADCASTMTE input being LOW.
>           */
>          if (cpu->tag_memory == NULL) {
>              cpu->isar.id_aa64pfr1 =
>                  FIELD_DP64(cpu->isar.id_aa64pfr1, ID_AA64PFR1, MTE, 1);
>          }
> #endif
>      }
>
> With this patch, for KVM we will end up going through the
> "squash ID_AA64PFR1_EL1.MTE to 1" codepath, because KVM doesn't
> set cpu->tag_memory and this is still using that as its check.
>
> More generally, how does the enabling of the MTE KVM cap
> interact with the ID_AA64PFR1_EL1 value that we read from
> the host in kvm_arm_get_host_cpu_features() ? We care that we

Linux kernel masks the MTE bits of register id_aa64pfr1 until unless the 
MTE is enabled for that VM. I have modified to enable 
MTE(KVM_CAP_ARM_MTE) before we read the register id_aa64pfr1 in 
kvm_arm_get_host_cpu_features to make sure we get the unmasked/actual 
MTE bits. I will post this change in the V2 patchset.

> have the right ID register values because we use ID field
> checks to determine whether the vcpu has a feature or not,
> even in the KVM case.
>
> Since Cornelia first wrote the patch this is based on, we've
> landed gdbstub support for MTE (so gdb can find out which
> addresses in the memory map have tags and read and write
> those tags). So I think the KVM MTE support now also needs to
> handle that. (See aarch64_cpu_register_gdb_commands() in
> target/arm/gdbstub64.c.)

I looked at this code and it looks like, complete code is under
ifdef CONFIG_USER_ONLY and for kvm(target aarch64-softmmu) this is not 
getting enabled. Are you asking to remove these ifdef and make
mte-gdbstub commands available for the KVM mode as well?


--
Thanks,
Ganapat/GK