From mboxrd@z Thu Jan  1 00:00:00 1970
From: Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [PATCH v6 07/17] add minimal virtio support for devtree virtio-mmio
Date: Fri, 11 Jul 2014 10:32:51 +0200
Message-ID: <53BFA133.7080608@redhat.com>
References: <1405066787-5793-1-git-send-email-drjones@redhat.com> <1405066787-5793-8-git-send-email-drjones@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 7bit
Cc: christoffer.dall@linaro.org
To: Andrew Jones <drjones@redhat.com>, kvmarm@lists.cs.columbia.edu,
	kvm@vger.kernel.org
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:52396 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751119AbaGKIc6 (ORCPT <rfc822;kvm@vger.kernel.org>);
	Fri, 11 Jul 2014 04:32:58 -0400
In-Reply-To: <1405066787-5793-8-git-send-email-drjones@redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

Il 11/07/2014 10:19, Andrew Jones ha scritto:
> +enum virtio_hwdesc_type {
> +	VIRTIO_HWDESC_TYPE_DT = 0,	/* device tree */
> +	NR_VIRTIO_HWDESC_TYPES,
> +};
> +
> +enum virtio_bus_type {
> +	VIRTIO_BUS_TYPE_MMIO = 0,	/* virtio-mmio */
> +	NR_VIRTIO_BUS_TYPES,
> +};
> +
> +struct virtio_bind_bus {
> +	bool (*hwdesc_probe)(void);
> +	struct virtio_device *(*device_bind)(u32 devid);
> +};
> +
> +static struct virtio_device *vm_dt_device_bind(u32 devid);
> +
> +static struct virtio_bind_bus
> +virtio_bind_busses[NR_VIRTIO_HWDESC_TYPES][NR_VIRTIO_BUS_TYPES] = {
> +
> +[VIRTIO_HWDESC_TYPE_DT] = {
> +
> +	[VIRTIO_BUS_TYPE_MMIO] = {
> +		.hwdesc_probe = dt_available,
> +		.device_bind = vm_dt_device_bind,
> +	},
> +},
> +};
> +

If you put this in lib/virtio.c, it is overengineered.  It would make 
sense if something else provided virtio_bind_busses[][].

I suggest that you drop it and split this file in four:

lib/virtio.c
lib/virtio-mmio.c
lib/virtio-mmio-dt.c
lib/arm/virtio.c

where virtio_bind is in lib/arm/virtio.c.

Paolo