[PATCH v2 04/16] KVM: arm64: Optimize host memory aborts
Marc Zyngier
maz at kernel.org
Mon Jul 26 06:24:43 PDT 2021
On Mon, 26 Jul 2021 14:13:06 +0100,
Quentin Perret <qperret at google.com> wrote:
>
> On Monday 26 Jul 2021 at 11:35:10 (+0100), Marc Zyngier wrote:
[...]
> > You could also use a kvm_mem_range for the iteration, and add a helper
> > that checks for the inclusion.
>
> Something like this (untested)?
>
> diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
> index 75273166d2c5..07d228163090 100644
> --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c
> +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
> @@ -234,9 +234,15 @@ static inline int __host_stage2_idmap(u64 start, u64 end,
> __ret; \
> })
>
> +static inline bool range_included(struct kvm_mem_range *child,
> + struct kvm_mem_range *parent)
> +{
> + return parent->start <= child->start && child->end <= parent->end;
> +}
> +
> static int host_stage2_find_range(u64 addr, struct kvm_mem_range *range)
> {
> - u64 granule, start, end;
> + struct kvm_mem_range cur;
> kvm_pte_t pte;
> u32 level;
> int ret;
> @@ -252,16 +258,15 @@ static int host_stage2_find_range(u64 addr, struct kvm_mem_range *range)
> return -EPERM;
>
> do {
> - granule = kvm_granule_size(level);
> - start = ALIGN_DOWN(addr, granule);
> - end = start + granule;
> + u64 granule = kvm_granule_size(level);
> + cur.start = ALIGN_DOWN(addr, granule);
> + cur.end = cur.start + granule;
> level++;
> } while ((level < KVM_PGTABLE_MAX_LEVELS) &&
> - (!kvm_level_supports_block_mapping(level) ||
> - start < range->start || range->end < end));
> + !(kvm_level_supports_block_mapping(level) &&
> + range_included(&cur, parent)));
>
> - range->start = start;
> - range->end = end;
> + *range = cur;
>
> return 0;
> }
>
Beautiful.
M.
--
Without deviation from the norm, progress is not possible.
More information about the linux-arm-kernel
mailing list