[linux-2.6] / drivers / s390 / kvm / kvm_virtio.c

/*
 * kvm_virtio.c - virtio for kvm on s390
 *
 * Copyright IBM Corp. 2008
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License (version 2 only)
 * as published by the Free Software Foundation.
 *
 *    Author(s): Christian Borntraeger <borntraeger@de.ibm.com>
 */

#include <linux/init.h>
#include <linux/bootmem.h>
#include <linux/err.h>
#include <linux/virtio.h>
#include <linux/virtio_config.h>
#include <linux/virtio_console.h>
#include <linux/interrupt.h>
#include <linux/virtio_ring.h>
#include <linux/pfn.h>
#include <asm/io.h>
#include <asm/kvm_para.h>
#include <asm/kvm_virtio.h>
#include <asm/setup.h>
#include <asm/s390_ext.h>
#include <asm/s390_rdev.h>

#define VIRTIO_SUBCODE_64 0x0D00

/*
 * The pointer to our (page) of device descriptions.
 */
static void *kvm_devices;

struct kvm_device {
	struct virtio_device vdev;
	struct kvm_device_desc *desc;
};

#define to_kvmdev(vd) container_of(vd, struct kvm_device, vdev)

/*
 * memory layout:
 * - kvm_device_descriptor
 *        struct kvm_device_desc
 * - configuration
 *        struct kvm_vqconfig
 * - feature bits
 * - config space
 */
static struct kvm_vqconfig *kvm_vq_config(const struct kvm_device_desc *desc)
{
	return (struct kvm_vqconfig *)(desc + 1);
}

static u8 *kvm_vq_features(const struct kvm_device_desc *desc)
{
	return (u8 *)(kvm_vq_config(desc) + desc->num_vq);
}

static u8 *kvm_vq_configspace(const struct kvm_device_desc *desc)
{
	return kvm_vq_features(desc) + desc->feature_len * 2;
}

/*
 * The total size of the config page used by this device (incl. desc)
 */
static unsigned desc_size(const struct kvm_device_desc *desc)
{
	return sizeof(*desc)
		+ desc->num_vq * sizeof(struct kvm_vqconfig)
		+ desc->feature_len * 2
		+ desc->config_len;
}

/* This gets the device's feature bits. */
static u32 kvm_get_features(struct virtio_device *vdev)
{
	unsigned int i;
	u32 features = 0;
	struct kvm_device_desc *desc = to_kvmdev(vdev)->desc;
	u8 *in_features = kvm_vq_features(desc);

	for (i = 0; i < min(desc->feature_len * 8, 32); i++)
		if (in_features[i / 8] & (1 << (i % 8)))
			features |= (1 << i);
	return features;
}

static void kvm_finalize_features(struct virtio_device *vdev)
{
	unsigned int i, bits;
	struct kvm_device_desc *desc = to_kvmdev(vdev)->desc;
	/* Second half of bitmap is features we accept. */
	u8 *out_features = kvm_vq_features(desc) + desc->feature_len;

	/* Give virtio_ring a chance to accept features. */
	vring_transport_features(vdev);

	memset(out_features, 0, desc->feature_len);
	bits = min_t(unsigned, desc->feature_len, sizeof(vdev->features)) * 8;
	for (i = 0; i < bits; i++) {
		if (test_bit(i, vdev->features))
			out_features[i / 8] |= (1 << (i % 8));
	}
}

/*
 * Reading and writing elements in config space
 */
static void kvm_get(struct virtio_device *vdev, unsigned int offset,
		   void *buf, unsigned len)
{
	struct kvm_device_desc *desc = to_kvmdev(vdev)->desc;

	BUG_ON(offset + len > desc->config_len);
	memcpy(buf, kvm_vq_configspace(desc) + offset, len);
}

static void kvm_set(struct virtio_device *vdev, unsigned int offset,
		   const void *buf, unsigned len)
{
	struct kvm_device_desc *desc = to_kvmdev(vdev)->desc;

	BUG_ON(offset + len > desc->config_len);
	memcpy(kvm_vq_configspace(desc) + offset, buf, len);
}

/*
 * The operations to get and set the status word just access
 * the status field of the device descriptor. set_status will also
 * make a hypercall to the host, to tell about status changes
 */
static u8 kvm_get_status(struct virtio_device *vdev)
{
	return to_kvmdev(vdev)->desc->status;
}

static void kvm_set_status(struct virtio_device *vdev, u8 status)
{
	BUG_ON(!status);
	to_kvmdev(vdev)->desc->status = status;
	kvm_hypercall1(KVM_S390_VIRTIO_SET_STATUS,
		       (unsigned long) to_kvmdev(vdev)->desc);
}

/*
 * To reset the device, we use the KVM_VIRTIO_RESET hypercall, using the
 * descriptor address. The Host will zero the status and all the
 * features.
 */
static void kvm_reset(struct virtio_device *vdev)
{
	kvm_hypercall1(KVM_S390_VIRTIO_RESET,
		       (unsigned long) to_kvmdev(vdev)->desc);
}

/*
 * When the virtio_ring code wants to notify the Host, it calls us here and we
 * make a hypercall.  We hand the address  of the virtqueue so the Host
 * knows which virtqueue we're talking about.
 */
static void kvm_notify(struct virtqueue *vq)
{
	struct kvm_vqconfig *config = vq->priv;

	kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, config->address);
}

/*
 * This routine finds the first virtqueue described in the configuration of
 * this device and sets it up.
 */
static struct virtqueue *kvm_find_vq(struct virtio_device *vdev,
				    unsigned index,
				    void (*callback)(struct virtqueue *vq))
{
	struct kvm_device *kdev = to_kvmdev(vdev);
	struct kvm_vqconfig *config;
	struct virtqueue *vq;
	int err;

	if (index >= kdev->desc->num_vq)
		return ERR_PTR(-ENOENT);

	config = kvm_vq_config(kdev->desc)+index;

	err = vmem_add_mapping(config->address,
			       vring_size(config->num, PAGE_SIZE));
	if (err)
		goto out;

	vq = vring_new_virtqueue(config->num, vdev, (void *) config->address,
				 kvm_notify, callback);
	if (!vq) {
		err = -ENOMEM;
		goto unmap;
	}

	/*
	 * register a callback token
	 * The host will sent this via the external interrupt parameter
	 */
	config->token = (u64) vq;

	vq->priv = config;
	return vq;
unmap:
	vmem_remove_mapping(config->address,
			    vring_size(config->num, PAGE_SIZE));
out:
	return ERR_PTR(err);
}

static void kvm_del_vq(struct virtqueue *vq)
{
	struct kvm_vqconfig *config = vq->priv;

	vring_del_virtqueue(vq);
	vmem_remove_mapping(config->address,
			    vring_size(config->num, PAGE_SIZE));
}

/*
 * The config ops structure as defined by virtio config
 */
static struct virtio_config_ops kvm_vq_configspace_ops = {
	.get_features = kvm_get_features,
	.finalize_features = kvm_finalize_features,
	.get = kvm_get,
	.set = kvm_set,
	.get_status = kvm_get_status,
	.set_status = kvm_set_status,
	.reset = kvm_reset,
	.find_vq = kvm_find_vq,
	.del_vq = kvm_del_vq,
};

/*
 * The root device for the kvm virtio devices.
 * This makes them appear as /sys/devices/kvm_s390/0,1,2 not /sys/devices/0,1,2.
 */
static struct device *kvm_root;

/*
 * adds a new device and register it with virtio
 * appropriate drivers are loaded by the device model
 */
static void add_kvm_device(struct kvm_device_desc *d, unsigned int offset)
{
	struct kvm_device *kdev;

	kdev = kzalloc(sizeof(*kdev), GFP_KERNEL);
	if (!kdev) {
		printk(KERN_EMERG "Cannot allocate kvm dev %u type %u\n",
		       offset, d->type);
		return;
	}

	kdev->vdev.dev.parent = kvm_root;
	kdev->vdev.id.device = d->type;
	kdev->vdev.config = &kvm_vq_configspace_ops;
	kdev->desc = d;

	if (register_virtio_device(&kdev->vdev) != 0) {
		printk(KERN_ERR "Failed to register kvm device %u type %u\n",
		       offset, d->type);
		kfree(kdev);
	}
}

/*
 * scan_devices() simply iterates through the device page.
 * The type 0 is reserved to mean "end of devices".
 */
static void scan_devices(void)
{
	unsigned int i;
	struct kvm_device_desc *d;

	for (i = 0; i < PAGE_SIZE; i += desc_size(d)) {
		d = kvm_devices + i;

		if (d->type == 0)
			break;

		add_kvm_device(d, i);
	}
}

/*
 * we emulate the request_irq behaviour on top of s390 extints
 */
static void kvm_extint_handler(u16 code)
{
	void *data = (void *) *(long *) __LC_PFAULT_INTPARM;
	u16 subcode = S390_lowcore.cpu_addr;

	if ((subcode & 0xff00) != VIRTIO_SUBCODE_64)
		return;

	vring_interrupt(0, data);
}

/*
 * Init function for virtio
 * devices are in a single page above top of "normal" mem
 */
static int __init kvm_devices_init(void)
{
	int rc;

	if (!MACHINE_IS_KVM)
		return -ENODEV;

	kvm_root = s390_root_dev_register("kvm_s390");
	if (IS_ERR(kvm_root)) {
		rc = PTR_ERR(kvm_root);
		printk(KERN_ERR "Could not register kvm_s390 root device");
		return rc;
	}

	rc = vmem_add_mapping(real_memory_size, PAGE_SIZE);
	if (rc) {
		s390_root_dev_unregister(kvm_root);
		return rc;
	}

	kvm_devices = (void *) real_memory_size;

	ctl_set_bit(0, 9);
	register_external_interrupt(0x2603, kvm_extint_handler);

	scan_devices();
	return 0;
}

/* code for early console output with virtio_console */
static __init int early_put_chars(u32 vtermno, const char *buf, int count)
{
	char scratch[17];
	unsigned int len = count;

	if (len > sizeof(scratch) - 1)
		len = sizeof(scratch) - 1;
	scratch[len] = '\0';
	memcpy(scratch, buf, len);
	kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, __pa(scratch));
	return len;
}

void __init s390_virtio_console_init(void)
{
	virtio_cons_early_init(early_put_chars);
}

/*
 * We do this after core stuff, but before the drivers.
 */
postcore_initcall(kvm_devices_init);
Commit	Line	Data
e976a2b9 CB	1	/*
	2	* kvm_virtio.c - virtio for kvm on s390
	3	*
	4	* Copyright IBM Corp. 2008
	5	*
	6	* This program is free software; you can redistribute it and/or modify
	7	* it under the terms of the GNU General Public License (version 2 only)
	8	* as published by the Free Software Foundation.
	9	*
	10	* Author(s): Christian Borntraeger <borntraeger@de.ibm.com>
	11	*/
	12
	13	#include <linux/init.h>
	14	#include <linux/bootmem.h>
	15	#include <linux/err.h>
	16	#include <linux/virtio.h>
	17	#include <linux/virtio_config.h>
faeba830	18	#include <linux/virtio_console.h>
e976a2b9 CB	19	#include <linux/interrupt.h>
e976a2b9 CB	20	#include <linux/virtio_ring.h>
17f34580	21	#include <linux/pfn.h>
e976a2b9 CB	22	#include <asm/io.h>
	23	#include <asm/kvm_para.h>
	24	#include <asm/kvm_virtio.h>
	25	#include <asm/setup.h>
	26	#include <asm/s390_ext.h>
37f1c012	27	#include <asm/s390_rdev.h>
e976a2b9 CB	28
	29	#define VIRTIO_SUBCODE_64 0x0D00
	30
	31	/*
	32	* The pointer to our (page) of device descriptions.
	33	*/
	34	static void *kvm_devices;
	35
e976a2b9 CB	36	struct kvm_device {
	37	struct virtio_device vdev;
	38	struct kvm_device_desc *desc;
	39	};
	40
	41	#define to_kvmdev(vd) container_of(vd, struct kvm_device, vdev)
	42
	43	/*
	44	* memory layout:
	45	* - kvm_device_descriptor
	46	* struct kvm_device_desc
	47	* - configuration
	48	* struct kvm_vqconfig
	49	* - feature bits
	50	* - config space
	51	*/
	52	static struct kvm_vqconfig kvm_vq_config(const struct kvm_device_desc desc)
	53	{
	54	return (struct kvm_vqconfig *)(desc + 1);
	55	}
	56
	57	static u8 kvm_vq_features(const struct kvm_device_desc desc)
	58	{
	59	return (u8 *)(kvm_vq_config(desc) + desc->num_vq);
	60	}
	61
	62	static u8 kvm_vq_configspace(const struct kvm_device_desc desc)
	63	{
	64	return kvm_vq_features(desc) + desc->feature_len * 2;
	65	}
	66
	67	/*
	68	* The total size of the config page used by this device (incl. desc)
	69	*/
	70	static unsigned desc_size(const struct kvm_device_desc *desc)
	71	{
	72	return sizeof(*desc)
	73	+ desc->num_vq * sizeof(struct kvm_vqconfig)
	74	+ desc->feature_len * 2
	75	+ desc->config_len;
	76	}
	77
5ca9fd54 HC	78	/* This gets the device's feature bits. */
5ca9fd54 HC	79	static u32 kvm_get_features(struct virtio_device *vdev)
e976a2b9	80	{
5ca9fd54 HC	81	unsigned int i;
5ca9fd54 HC	82	u32 features = 0;
e976a2b9	83	struct kvm_device_desc *desc = to_kvmdev(vdev)->desc;
5ca9fd54	84	u8 *in_features = kvm_vq_features(desc);
e976a2b9	85
5ca9fd54 HC	86	for (i = 0; i < min(desc->feature_len * 8, 32); i++)
	87	if (in_features[i / 8] & (1 << (i % 8)))
	88	features \|= (1 << i);
	89	return features;
	90	}
e976a2b9	91
c624896e	92	static void kvm_finalize_features(struct virtio_device *vdev)
5ca9fd54	93	{
c624896e	94	unsigned int i, bits;
5ca9fd54 HC	95	struct kvm_device_desc *desc = to_kvmdev(vdev)->desc;
	96	/* Second half of bitmap is features we accept. */
	97	u8 *out_features = kvm_vq_features(desc) + desc->feature_len;
e976a2b9	98
e34f8725 RR	99	/* Give virtio_ring a chance to accept features. */
	100	vring_transport_features(vdev);
	101
5ca9fd54	102	memset(out_features, 0, desc->feature_len);
c624896e RR	103	bits = min_t(unsigned, desc->feature_len, sizeof(vdev->features)) * 8;
	104	for (i = 0; i < bits; i++) {
	105	if (test_bit(i, vdev->features))
5ca9fd54 HC	106	out_features[i / 8] \|= (1 << (i % 8));
5ca9fd54 HC	107	}
e976a2b9 CB	108	}
	109
	110	/*
	111	* Reading and writing elements in config space
	112	*/
	113	static void kvm_get(struct virtio_device *vdev, unsigned int offset,
	114	void *buf, unsigned len)
	115	{
	116	struct kvm_device_desc *desc = to_kvmdev(vdev)->desc;
	117
	118	BUG_ON(offset + len > desc->config_len);
	119	memcpy(buf, kvm_vq_configspace(desc) + offset, len);
	120	}
	121
	122	static void kvm_set(struct virtio_device *vdev, unsigned int offset,
	123	const void *buf, unsigned len)
	124	{
	125	struct kvm_device_desc *desc = to_kvmdev(vdev)->desc;
	126
	127	BUG_ON(offset + len > desc->config_len);
	128	memcpy(kvm_vq_configspace(desc) + offset, buf, len);
	129	}
	130
	131	/*
	132	* The operations to get and set the status word just access
	133	* the status field of the device descriptor. set_status will also
	134	* make a hypercall to the host, to tell about status changes
	135	*/
	136	static u8 kvm_get_status(struct virtio_device *vdev)
	137	{
	138	return to_kvmdev(vdev)->desc->status;
	139	}
	140
	141	static void kvm_set_status(struct virtio_device *vdev, u8 status)
	142	{
	143	BUG_ON(!status);
	144	to_kvmdev(vdev)->desc->status = status;
	145	kvm_hypercall1(KVM_S390_VIRTIO_SET_STATUS,
	146	(unsigned long) to_kvmdev(vdev)->desc);
	147	}
	148
	149	/*
	150	* To reset the device, we use the KVM_VIRTIO_RESET hypercall, using the
	151	* descriptor address. The Host will zero the status and all the
	152	* features.
	153	*/
	154	static void kvm_reset(struct virtio_device *vdev)
	155	{
	156	kvm_hypercall1(KVM_S390_VIRTIO_RESET,
	157	(unsigned long) to_kvmdev(vdev)->desc);
	158	}
	159
	160	/*
	161	* When the virtio_ring code wants to notify the Host, it calls us here and we
	162	* make a hypercall. We hand the address of the virtqueue so the Host
	163	* knows which virtqueue we're talking about.
	164	*/
	165	static void kvm_notify(struct virtqueue *vq)
	166	{
	167	struct kvm_vqconfig *config = vq->priv;
	168
	169	kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, config->address);
	170	}
	171
172	/*
173	* This routine finds the first virtqueue described in the configuration of
174	* this device and sets it up.
175	*/
176	static struct virtqueue kvm_find_vq(struct virtio_device vdev,
177	unsigned index,
178	void (callback)(struct virtqueue vq))
179	{
180	struct kvm_device *kdev = to_kvmdev(vdev);
181	struct kvm_vqconfig *config;
182	struct virtqueue *vq;
183	int err;
184
185	if (index >= kdev->desc->num_vq)
186	return ERR_PTR(-ENOENT);
187
188	config = kvm_vq_config(kdev->desc)+index;
189
17f34580 HC	190	err = vmem_add_mapping(config->address,
	191	vring_size(config->num, PAGE_SIZE));
	192	if (err)
e976a2b9	193	goto out;
e976a2b9 CB	194
	195	vq = vring_new_virtqueue(config->num, vdev, (void *) config->address,
	196	kvm_notify, callback);
	197	if (!vq) {
	198	err = -ENOMEM;
	199	goto unmap;
	200	}
	201
	202	/*
	203	* register a callback token
	204	* The host will sent this via the external interrupt parameter
	205	*/
	206	config->token = (u64) vq;
	207
	208	vq->priv = config;
	209	return vq;
	210	unmap:
17f34580 HC	211	vmem_remove_mapping(config->address,
17f34580 HC	212	vring_size(config->num, PAGE_SIZE));
e976a2b9 CB	213	out:
	214	return ERR_PTR(err);
	215	}
	216
	217	static void kvm_del_vq(struct virtqueue *vq)
	218	{
	219	struct kvm_vqconfig *config = vq->priv;
	220
	221	vring_del_virtqueue(vq);
17f34580 HC	222	vmem_remove_mapping(config->address,
17f34580 HC	223	vring_size(config->num, PAGE_SIZE));
e976a2b9 CB	224	}
	225
	226	/*
	227	* The config ops structure as defined by virtio config
	228	*/
	229	static struct virtio_config_ops kvm_vq_configspace_ops = {
5ca9fd54	230	.get_features = kvm_get_features,
c624896e	231	.finalize_features = kvm_finalize_features,
e976a2b9 CB	232	.get = kvm_get,
	233	.set = kvm_set,
	234	.get_status = kvm_get_status,
	235	.set_status = kvm_set_status,
	236	.reset = kvm_reset,
	237	.find_vq = kvm_find_vq,
	238	.del_vq = kvm_del_vq,
	239	};
	240
	241	/*
	242	* The root device for the kvm virtio devices.
	243	* This makes them appear as /sys/devices/kvm_s390/0,1,2 not /sys/devices/0,1,2.
	244	*/
37f1c012	245	static struct device *kvm_root;
e976a2b9 CB	246
	247	/*
	248	* adds a new device and register it with virtio
	249	* appropriate drivers are loaded by the device model
	250	*/
b769f579	251	static void add_kvm_device(struct kvm_device_desc *d, unsigned int offset)
e976a2b9 CB	252	{
	253	struct kvm_device *kdev;
	254
	255	kdev = kzalloc(sizeof(*kdev), GFP_KERNEL);
	256	if (!kdev) {
b769f579 RR	257	printk(KERN_EMERG "Cannot allocate kvm dev %u type %u\n",
b769f579 RR	258	offset, d->type);
e976a2b9 CB	259	return;
	260	}
	261
37f1c012	262	kdev->vdev.dev.parent = kvm_root;
e976a2b9 CB	263	kdev->vdev.id.device = d->type;
	264	kdev->vdev.config = &kvm_vq_configspace_ops;
	265	kdev->desc = d;
	266
	267	if (register_virtio_device(&kdev->vdev) != 0) {
b769f579 RR	268	printk(KERN_ERR "Failed to register kvm device %u type %u\n",
b769f579 RR	269	offset, d->type);
e976a2b9 CB	270	kfree(kdev);
	271	}
	272	}
	273
	274	/*
	275	* scan_devices() simply iterates through the device page.
	276	* The type 0 is reserved to mean "end of devices".
	277	*/
	278	static void scan_devices(void)
	279	{
	280	unsigned int i;
	281	struct kvm_device_desc *d;
	282
	283	for (i = 0; i < PAGE_SIZE; i += desc_size(d)) {
	284	d = kvm_devices + i;
	285
	286	if (d->type == 0)
	287	break;
	288
b769f579	289	add_kvm_device(d, i);
e976a2b9 CB	290	}
	291	}
	292
	293	/*
	294	* we emulate the request_irq behaviour on top of s390 extints
	295	*/
	296	static void kvm_extint_handler(u16 code)
	297	{
	298	void data = (void ) (long ) __LC_PFAULT_INTPARM;
	299	u16 subcode = S390_lowcore.cpu_addr;
	300
	301	if ((subcode & 0xff00) != VIRTIO_SUBCODE_64)
	302	return;
	303
	304	vring_interrupt(0, data);
	305	}
	306
	307	/*
	308	* Init function for virtio
	309	* devices are in a single page above top of "normal" mem
	310	*/
	311	static int __init kvm_devices_init(void)
	312	{
	313	int rc;
	314
	315	if (!MACHINE_IS_KVM)
	316	return -ENODEV;
	317
37f1c012 CH	318	kvm_root = s390_root_dev_register("kvm_s390");
	319	if (IS_ERR(kvm_root)) {
	320	rc = PTR_ERR(kvm_root);
e976a2b9 CB	321	printk(KERN_ERR "Could not register kvm_s390 root device");
	322	return rc;
	323	}
	324
cc835f78	325	rc = vmem_add_mapping(real_memory_size, PAGE_SIZE);
17f34580	326	if (rc) {
37f1c012	327	s390_root_dev_unregister(kvm_root);
17f34580	328	return rc;
e976a2b9 CB	329	}
e976a2b9 CB	330
cc835f78	331	kvm_devices = (void *) real_memory_size;
e976a2b9 CB	332
	333	ctl_set_bit(0, 9);
	334	register_external_interrupt(0x2603, kvm_extint_handler);
	335
	336	scan_devices();
	337	return 0;
	338	}
	339
faeba830 CB	340	/* code for early console output with virtio_console */
	341	static __init int early_put_chars(u32 vtermno, const char *buf, int count)
	342	{
	343	char scratch[17];
	344	unsigned int len = count;
	345
	346	if (len > sizeof(scratch) - 1)
	347	len = sizeof(scratch) - 1;
	348	scratch[len] = '\0';
	349	memcpy(scratch, buf, len);
	350	kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, __pa(scratch));
	351	return len;
	352	}
	353
26f746f3	354	void __init s390_virtio_console_init(void)
faeba830 CB	355	{
	356	virtio_cons_early_init(early_put_chars);
	357	}
	358
e976a2b9 CB	359	/*
	360	* We do this after core stuff, but before the drivers.
	361	*/
	362	postcore_initcall(kvm_devices_init);