Re: [PATCH v2] arm/kvm: Enable support for KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE

[Gavin Shan]

Hi Shameer,

On 8/15/23 19:27, Shameer Kolothum wrote:
> Now that we have Eager Page Split support added for ARM in the kernel,
> enable it in Qemu. This adds,
>   -eager-split-size to -accel sub-options to set the eager page split chunk 
> size.
>   -enable KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE.
>
> The chunk size specifies how many pages to break at a time, using a
> single allocation. Bigger the chunk size, more pages need to be
> allocated ahead of time.
>
> Signed-off-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
> ---
> RFC v1: 
> https://lore.kernel.org/qemu-devel/20230725150002.621-1-shameerali.kolothum.thodi@huawei.com/
>    -Updated qemu-options.hx with description
>    -Addressed review comments from Peter and Gavin(Thanks).
> ---
>   include/sysemu/kvm_int.h |  1 +
>   qemu-options.hx          | 14 +++++++++
>   target/arm/kvm.c         | 62 ++++++++++++++++++++++++++++++++++++++++
>   3 files changed, 77 insertions(+)

[...]

>   
> +static void kvm_arch_get_eager_split_size(Object *obj, Visitor *v,
> +                                          const char *name, void *opaque,
> +                                          Error **errp)
> +{
> +    KVMState *s = KVM_STATE(obj);
> +    uint64_t value = s->kvm_eager_split_size;
> +
> +    visit_type_size(v, name, &value, errp);
> +}
> +
> +static void kvm_arch_set_eager_split_size(Object *obj, Visitor *v,
> +                                          const char *name, void *opaque,
> +                                          Error **errp)
> +{
> +    KVMState *s = KVM_STATE(obj);
> +    uint64_t value;
> +
> +    if (s->fd != -1) {
> +        error_setg(errp, "Cannot set properties after the accelerator has been 
> initialized");
> +        return;
> +    }
> +

Errors spotted by './scripts/checkpatch.pl', as below:

ERROR: line over 90 characters
#139: FILE: target/arm/kvm.c:1112:
+        error_setg(errp, "Cannot set properties after the accelerator has been 
initialized");

Thanks,
Gavin