Re: [PATCH v4 6/7] xen: introduce Kconfig ARCH_PAGING_MEMPOOL

[Jan Beulich <jbeulich@xxxxxxxx>]

On 01.04.2025 10:58, Luca Fancellu wrote:
> From: Penny Zheng <Penny.Zheng@xxxxxxx>
> 
> ARM MPU system doesn't need to use paging memory pool, as MPU memory
> mapping table at most takes only one 4KB page, which is enough to
> manage the maximum 255 MPU memory regions, for all EL2 stage 1
> translation and EL1 stage 2 translation.
> 
> Introduce ARCH_PAGING_MEMPOOL Kconfig common symbol, selected for Arm
> MMU systems and x86. Removed stubs from RISC-V now that the common code
> provide them and the functions are not gonna be used.
> 
> Wrap the code inside 'construct_domU' that deal with p2m paging
> allocation in a new function 'domain_p2m_set_allocation', protected
> by ARCH_PAGING_MEMPOOL, this is done in this way to prevent polluting
> the former function with #ifdefs and improve readability
> 
> Introduce arch_{get,set}_paging_mempool_size stubs for architecture
> with !ARCH_PAGING_MEMPOOL.
> 
> Remove 'struct paging_domain' from Arm 'struct arch_domain' when the
> field is not required.
> 
> Signed-off-by: Penny Zheng <penny.zheng@xxxxxxx>
> Signed-off-by: Wei Chen <wei.chen@xxxxxxx>
> Signed-off-by: Luca Fancellu <luca.fancellu@xxxxxxx>
> Reviewed-by: Michal Orzel <michal.orzel@xxxxxxx> # arm
> Reviewed-by: Oleksii Kurochko <oleksii.kurochko@xxxxxxxxx> # riscv

Acked-by: Jan Beulich <jbeulich@xxxxxxxx>

> @@ -114,9 +115,25 @@ void arch_get_info_guest(struct vcpu *v, 
> vcpu_guest_context_u c);
>  int arch_initialise_vcpu(struct vcpu *v, XEN_GUEST_HANDLE_PARAM(void) arg);
>  int default_initialise_vcpu(struct vcpu *v, XEN_GUEST_HANDLE_PARAM(void) 
> arg);
>  
> +#ifdef CONFIG_ARCH_PAGING_MEMPOOL
> +
>  int arch_get_paging_mempool_size(struct domain *d, uint64_t *size /* bytes 
> */);
>  int arch_set_paging_mempool_size(struct domain *d, uint64_t size /* bytes 
> */);
>  
> +#else /* !CONFIG_ARCH_PAGING_MEMPOOL */
> +
> +static inline int arch_get_paging_mempool_size(struct domain *d, uint64_t 
> *size)
> +{
> +    return -EOPNOTSUPP;
> +}
> +
> +static inline int arch_set_paging_mempool_size(struct domain *d, uint64_t 
> size)
> +{
> +    return -EOPNOTSUPP;
> +}

Arguably while "set" will of course need to fail (perhaps unless size was zero),
"get" may be fine to uniformly succeed, reporting back a size of 0. But we can
switch to that alternative model whenever a need arises, I guess.

Jan