dm: Add a new uclass driver for VirtIO transport devices

This adds a new virtio uclass driver for “virtio” [1] family of
devices that are are found in virtual environments like QEMU,
yet by design they look like physical devices to the guest.

The uclass driver provides child_pre_probe() and child_post_probe()
methods to do some common operations for virtio device drivers like
device and driver supported feature negotiation, etc.

[1] http://docs.oasis-open.org/virtio/virtio/v1.0/virtio-v1.0.pdf

Signed-off-by: Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>

1 files changed, 707 insertions, 0 deletions

diff --git a/include/virtio.h b/include/virtio.h
new file mode 100644
index 0000000000..654fdf154b
--- /dev/null
+++ b/include/virtio.h
@@ -0,0 +1,707 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/*
+ * Copyright (C) 2018, Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
+ * Copyright (C) 2018, Bin Meng <bmeng.cn@gmail.com>
+ *
+ * VirtIO is a virtualization standard for network and disk device drivers
+ * where just the guest's device driver "knows" it is running in a virtual
+ * environment, and cooperates with the hypervisor. This enables guests to
+ * get high performance network and disk operations, and gives most of the
+ * performance benefits of paravirtualization. In the U-Boot case, the guest
+ * is U-Boot itself, while the virtual environment are normally QEMU targets
+ * like ARM, RISC-V and x86.
+ *
+ * See http://docs.oasis-open.org/virtio/virtio/v1.0/virtio-v1.0.pdf for
+ * the VirtIO specification v1.0.
+ *
+ * This file is largely based on Linux kernel virtio_*.h files
+ */
+
+#ifndef __VIRTIO_H__
+#define __VIRTIO_H__
+
+#define VIRTIO_ID_NET		1 /* virtio net */
+#define VIRTIO_ID_BLOCK		2 /* virtio block */
+#define VIRTIO_ID_MAX_NUM	3
+
+#define VIRTIO_NET_DRV_NAME	"virtio-net"
+#define VIRTIO_BLK_DRV_NAME	"virtio-blk"
+
+/* Status byte for guest to report progress, and synchronize features */
+
+/* We have seen device and processed generic fields (VIRTIO_CONFIG_F_VIRTIO) */
+#define VIRTIO_CONFIG_S_ACKNOWLEDGE	1
+/* We have found a driver for the device */
+#define VIRTIO_CONFIG_S_DRIVER		2
+/* Driver has used its parts of the config, and is happy */
+#define VIRTIO_CONFIG_S_DRIVER_OK	4
+/* Driver has finished configuring features */
+#define VIRTIO_CONFIG_S_FEATURES_OK	8
+/* Device entered invalid state, driver must reset it */
+#define VIRTIO_CONFIG_S_NEEDS_RESET	0x40
+/* We've given up on this device */
+#define VIRTIO_CONFIG_S_FAILED		0x80
+
+/*
+ * Virtio feature bits VIRTIO_TRANSPORT_F_START through VIRTIO_TRANSPORT_F_END
+ * are reserved for the transport being used (eg: virtio_ring, virtio_pci etc.),
+ * the rest are per-device feature bits.
+ */
+#define VIRTIO_TRANSPORT_F_START	28
+#define VIRTIO_TRANSPORT_F_END		38
+
+#ifndef VIRTIO_CONFIG_NO_LEGACY
+/*
+ * Do we get callbacks when the ring is completely used,
+ * even if we've suppressed them?
+ */
+#define VIRTIO_F_NOTIFY_ON_EMPTY	24
+
+/* Can the device handle any descriptor layout? */
+#define VIRTIO_F_ANY_LAYOUT		27
+#endif /* VIRTIO_CONFIG_NO_LEGACY */
+
+/* v1.0 compliant */
+#define VIRTIO_F_VERSION_1		32
+
+/*
+ * If clear - device has the IOMMU bypass quirk feature.
+ * If set - use platform tools to detect the IOMMU.
+ *
+ * Note the reverse polarity (compared to most other features),
+ * this is for compatibility with legacy systems.
+ */
+#define VIRTIO_F_IOMMU_PLATFORM		33
+
+/* Does the device support Single Root I/O Virtualization? */
+#define VIRTIO_F_SR_IOV			37
+
+/**
+ * virtio scatter-gather struct
+ *
+ * @addr:		sg buffer address
+ * @lengh:		sg buffer length
+ */
+struct virtio_sg {
+	void *addr;
+	size_t length;
+};
+
+struct virtqueue;
+
+/* virtio bus operations */
+struct dm_virtio_ops {
+	/**
+	 * get_config() - read the value of a configuration field
+	 *
+	 * @vdev:	the real virtio device
+	 * @offset:	the offset of the configuration field
+	 * @buf:	the buffer to write the field value into
+	 * @len:	the length of the buffer
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*get_config)(struct udevice *vdev, unsigned int offset,
+			  void *buf, unsigned int len);
+	/**
+	 * set_config() - write the value of a configuration field
+	 *
+	 * @vdev:	the real virtio device
+	 * @offset:	the offset of the configuration field
+	 * @buf:	the buffer to read the field value from
+	 * @len:	the length of the buffer
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*set_config)(struct udevice *vdev, unsigned int offset,
+			  const void *buf, unsigned int len);
+	/**
+	 * generation() - config generation counter
+	 *
+	 * @vdev:	the real virtio device
+	 * @counter:	the returned config generation counter
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*generation)(struct udevice *vdev, u32 *counter);
+	/**
+	 * get_status() - read the status byte
+	 *
+	 * @vdev:	the real virtio device
+	 * @status:	the returned status byte
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*get_status)(struct udevice *vdev, u8 *status);
+	/**
+	 * set_status() - write the status byte
+	 *
+	 * @vdev:	the real virtio device
+	 * @status:	the new status byte
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*set_status)(struct udevice *vdev, u8 status);
+	/**
+	 * reset() - reset the device
+	 *
+	 * @vdev:	the real virtio device
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*reset)(struct udevice *vdev);
+	/**
+	 * get_features() - get the array of feature bits for this device
+	 *
+	 * @vdev:	the real virtio device
+	 * @features:	the first 32 feature bits (all we currently need)
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*get_features)(struct udevice *vdev, u64 *features);
+	/**
+	 * set_features() - confirm what device features we'll be using
+	 *
+	 * @vdev:	the real virtio device
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*set_features)(struct udevice *vdev);
+	/**
+	 * find_vqs() - find virtqueues and instantiate them
+	 *
+	 * @vdev:	the real virtio device
+	 * @nvqs:	the number of virtqueues to find
+	 * @vqs:	on success, includes new virtqueues
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*find_vqs)(struct udevice *vdev, unsigned int nvqs,
+			struct virtqueue *vqs[]);
+	/**
+	 * del_vqs() - free virtqueues found by find_vqs()
+	 *
+	 * @vdev:	the real virtio device
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*del_vqs)(struct udevice *vdev);
+	/**
+	 * notify() - notify the device to process the queue
+	 *
+	 * @vdev:	the real virtio device
+	 * @vq:		virtqueue to process
+	 * @return 0 if OK, -ve on error
+	 */
+	int (*notify)(struct udevice *vdev, struct virtqueue *vq);
+};
+
+/* Get access to a virtio bus' operations */
+#define virtio_get_ops(dev)	((struct dm_virtio_ops *)(dev)->driver->ops)
+
+/**
+ * virtio uclass per device private data
+ *
+ * @vqs:			virtualqueue for the virtio device
+ * @vdev:			the real virtio device underneath
+ * @legacy:			is it a legacy device?
+ * @device:			virtio device ID
+ * @vendor:			virtio vendor ID
+ * @features:			negotiated supported features
+ * @feature_table:		an array of feature supported by the driver
+ * @feature_table_size:		number of entries in the feature table array
+ * @feature_table_legacy:	same as feature_table but working in legacy mode
+ * @feature_table_size_legacy:	number of entries in feature table legacy array
+ */
+struct virtio_dev_priv {
+	struct list_head vqs;
+	struct udevice *vdev;
+	bool legacy;
+	u32 device;
+	u32 vendor;
+	u64 features;
+	const u32 *feature_table;
+	u32 feature_table_size;
+	const u32 *feature_table_legacy;
+	u32 feature_table_size_legacy;
+};
+
+/**
+ * virtio_get_config() - read the value of a configuration field
+ *
+ * @vdev:	the real virtio device
+ * @offset:	the offset of the configuration field
+ * @buf:	the buffer to write the field value into
+ * @len:	the length of the buffer
+ * @return 0 if OK, -ve on error
+ */
+int virtio_get_config(struct udevice *vdev, unsigned int offset,
+		      void *buf, unsigned int len);
+
+/**
+ * virtio_set_config() - write the value of a configuration field
+ *
+ * @vdev:	the real virtio device
+ * @offset:	the offset of the configuration field
+ * @buf:	the buffer to read the field value from
+ * @len:	the length of the buffer
+ * @return 0 if OK, -ve on error
+ */
+int virtio_set_config(struct udevice *vdev, unsigned int offset,
+		      void *buf, unsigned int len);
+
+/**
+ * virtio_generation() - config generation counter
+ *
+ * @vdev:	the real virtio device
+ * @counter:	the returned config generation counter
+ * @return 0 if OK, -ve on error
+ */
+int virtio_generation(struct udevice *vdev, u32 *counter);
+
+/**
+ * virtio_get_status() - read the status byte
+ *
+ * @vdev:	the real virtio device
+ * @status:	the returned status byte
+ * @return 0 if OK, -ve on error
+ */
+int virtio_get_status(struct udevice *vdev, u8 *status);
+
+/**
+ * virtio_set_status() - write the status byte
+ *
+ * @vdev:	the real virtio device
+ * @status:	the new status byte
+ * @return 0 if OK, -ve on error
+ */
+int virtio_set_status(struct udevice *vdev, u8 status);
+
+/**
+ * virtio_reset() - reset the device
+ *
+ * @vdev:	the real virtio device
+ * @return 0 if OK, -ve on error
+ */
+int virtio_reset(struct udevice *vdev);
+
+/**
+ * virtio_get_features() - get the array of feature bits for this device
+ *
+ * @vdev:	the real virtio device
+ * @features:	the first 32 feature bits (all we currently need)
+ * @return 0 if OK, -ve on error
+ */
+int virtio_get_features(struct udevice *vdev, u64 *features);
+
+/**
+ * virtio_set_features() - confirm what device features we'll be using
+ *
+ * @vdev:	the real virtio device
+ * @return 0 if OK, -ve on error
+ */
+int virtio_set_features(struct udevice *vdev);
+
+/**
+ * virtio_find_vqs() - find virtqueues and instantiate them
+ *
+ * @vdev:	the real virtio device
+ * @nvqs:	the number of virtqueues to find
+ * @vqs:	on success, includes new virtqueues
+ * @return 0 if OK, -ve on error
+ */
+int virtio_find_vqs(struct udevice *vdev, unsigned int nvqs,
+		    struct virtqueue *vqs[]);
+
+/**
+ * virtio_del_vqs() - free virtqueues found by find_vqs()
+ *
+ * @vdev:	the real virtio device
+ * @return 0 if OK, -ve on error
+ */
+int virtio_del_vqs(struct udevice *vdev);
+
+/**
+ * virtio_notify() - notify the device to process the queue
+ *
+ * @vdev:	the real virtio device
+ * @vq:		virtqueue to process
+ * @return 0 if OK, -ve on error
+ */
+int virtio_notify(struct udevice *vdev, struct virtqueue *vq);
+
+/**
+ * virtio_add_status() - helper to set a new status code to the device
+ *
+ * @vdev:	the real virtio device
+ * @status:	new status code to be added
+ */
+void virtio_add_status(struct udevice *vdev, u8 status);
+
+/**
+ * virtio_finalize_features() - helper to finalize features
+ *
+ * @vdev:	the real virtio device
+ * @return 0 if OK, -ve on error
+ */
+int virtio_finalize_features(struct udevice *vdev);
+
+/**
+ * virtio_driver_features_init() - initialize driver supported features
+ *
+ * This fills in the virtio device parent per child private data with the given
+ * information, which contains driver supported features and legacy features.
+ *
+ * This API should be called in the virtio device driver's bind method, so that
+ * later virtio transport uclass driver can utilize the driver supplied features
+ * to negotiate with the device on the final supported features.
+ *
+ * @priv:		virtio uclass per device private data
+ * @feature:		an array of feature supported by the driver
+ * @feature_size:	number of entries in the feature table array
+ * @feature_legacy:	same as feature_table but working in legacy mode
+ * @feature_legacy_size:number of entries in feature table legacy array
+ */
+void virtio_driver_features_init(struct virtio_dev_priv *priv,
+				 const u32 *feature,
+				 u32 feature_size,
+				 const u32 *feature_legacy,
+				 u32 feature_legacy_size);
+
+/**
+ * virtio_init() - helper to enumerate all known virtio devices
+ *
+ * @return 0 if OK, -ve on error
+ */
+int virtio_init(void);
+
+static inline u16 __virtio16_to_cpu(bool little_endian, __virtio16 val)
+{
+	if (little_endian)
+		return le16_to_cpu((__force __le16)val);
+	else
+		return be16_to_cpu((__force __be16)val);
+}
+
+static inline __virtio16 __cpu_to_virtio16(bool little_endian, u16 val)
+{
+	if (little_endian)
+		return (__force __virtio16)cpu_to_le16(val);
+	else
+		return (__force __virtio16)cpu_to_be16(val);
+}
+
+static inline u32 __virtio32_to_cpu(bool little_endian, __virtio32 val)
+{
+	if (little_endian)
+		return le32_to_cpu((__force __le32)val);
+	else
+		return be32_to_cpu((__force __be32)val);
+}
+
+static inline __virtio32 __cpu_to_virtio32(bool little_endian, u32 val)
+{
+	if (little_endian)
+		return (__force __virtio32)cpu_to_le32(val);
+	else
+		return (__force __virtio32)cpu_to_be32(val);
+}
+
+static inline u64 __virtio64_to_cpu(bool little_endian, __virtio64 val)
+{
+	if (little_endian)
+		return le64_to_cpu((__force __le64)val);
+	else
+		return be64_to_cpu((__force __be64)val);
+}
+
+static inline __virtio64 __cpu_to_virtio64(bool little_endian, u64 val)
+{
+	if (little_endian)
+		return (__force __virtio64)cpu_to_le64(val);
+	else
+		return (__force __virtio64)cpu_to_be64(val);
+}
+
+/**
+ * __virtio_test_bit - helper to test feature bits
+ *
+ * For use by transports. Devices should normally use virtio_has_feature,
+ * which includes more checks.
+ *
+ * @udev: the transport device
+ * @fbit: the feature bit
+ */
+static inline bool __virtio_test_bit(struct udevice *udev, unsigned int fbit)
+{
+	struct virtio_dev_priv *uc_priv = dev_get_uclass_priv(udev);
+
+	/* Did you forget to fix assumptions on max features? */
+	if (__builtin_constant_p(fbit))
+		BUILD_BUG_ON(fbit >= 64);
+	else
+		WARN_ON(fbit >= 64);
+
+	return uc_priv->features & BIT_ULL(fbit);
+}
+
+/**
+ * __virtio_set_bit - helper to set feature bits
+ *
+ * For use by transports.
+ *
+ * @udev: the transport device
+ * @fbit: the feature bit
+ */
+static inline void __virtio_set_bit(struct udevice *udev, unsigned int fbit)
+{
+	struct virtio_dev_priv *uc_priv = dev_get_uclass_priv(udev);
+
+	/* Did you forget to fix assumptions on max features? */
+	if (__builtin_constant_p(fbit))
+		BUILD_BUG_ON(fbit >= 64);
+	else
+		WARN_ON(fbit >= 64);
+
+	uc_priv->features |= BIT_ULL(fbit);
+}
+
+/**
+ * __virtio_clear_bit - helper to clear feature bits
+ *
+ * For use by transports.
+ *
+ * @vdev: the transport device
+ * @fbit: the feature bit
+ */
+static inline void __virtio_clear_bit(struct udevice *udev, unsigned int fbit)
+{
+	struct virtio_dev_priv *uc_priv = dev_get_uclass_priv(udev);
+
+	/* Did you forget to fix assumptions on max features? */
+	if (__builtin_constant_p(fbit))
+		BUILD_BUG_ON(fbit >= 64);
+	else
+		WARN_ON(fbit >= 64);
+
+	uc_priv->features &= ~BIT_ULL(fbit);
+}
+
+/**
+ * virtio_has_feature - helper to determine if this device has this feature
+ *
+ * Note this API is only usable after the virtio device driver's bind phase,
+ * as the feature has been negotiated between the device and the driver.
+ *
+ * @vdev: the virtio device
+ * @fbit: the feature bit
+ */
+static inline bool virtio_has_feature(struct udevice *vdev, unsigned int fbit)
+{
+	if (!(vdev->flags & DM_FLAG_BOUND))
+		WARN_ON(true);
+
+	return __virtio_test_bit(vdev->parent, fbit);
+}
+
+static inline bool virtio_legacy_is_little_endian(void)
+{
+#ifdef __LITTLE_ENDIAN
+	return true;
+#else
+	return false;
+#endif
+}
+
+static inline bool virtio_is_little_endian(struct udevice *vdev)
+{
+	struct virtio_dev_priv *uc_priv = dev_get_uclass_priv(vdev->parent);
+
+	return !uc_priv->legacy || virtio_legacy_is_little_endian();
+}
+
+/* Memory accessors */
+static inline u16 virtio16_to_cpu(struct udevice *vdev, __virtio16 val)
+{
+	return __virtio16_to_cpu(virtio_is_little_endian(vdev), val);
+}
+
+static inline __virtio16 cpu_to_virtio16(struct udevice *vdev, u16 val)
+{
+	return __cpu_to_virtio16(virtio_is_little_endian(vdev), val);
+}
+
+static inline u32 virtio32_to_cpu(struct udevice *vdev, __virtio32 val)
+{
+	return __virtio32_to_cpu(virtio_is_little_endian(vdev), val);
+}
+
+static inline __virtio32 cpu_to_virtio32(struct udevice *vdev, u32 val)
+{
+	return __cpu_to_virtio32(virtio_is_little_endian(vdev), val);
+}
+
+static inline u64 virtio64_to_cpu(struct udevice *vdev, __virtio64 val)
+{
+	return __virtio64_to_cpu(virtio_is_little_endian(vdev), val);
+}
+
+static inline __virtio64 cpu_to_virtio64(struct udevice *vdev, u64 val)
+{
+	return __cpu_to_virtio64(virtio_is_little_endian(vdev), val);
+}
+
+/* Read @count fields, @bytes each */
+static inline void __virtio_cread_many(struct udevice *vdev,
+				       unsigned int offset,
+				       void *buf, size_t count, size_t bytes)
+{
+	u32 old, gen;
+	int i;
+
+	/* no need to check return value as generation can be optional */
+	virtio_generation(vdev, &gen);
+	do {
+		old = gen;
+
+		for (i = 0; i < count; i++)
+			virtio_get_config(vdev, offset + bytes * i,
+					  buf + i * bytes, bytes);
+
+		virtio_generation(vdev, &gen);
+	} while (gen != old);
+}
+
+static inline void virtio_cread_bytes(struct udevice *vdev,
+				      unsigned int offset,
+				      void *buf, size_t len)
+{
+	__virtio_cread_many(vdev, offset, buf, len, 1);
+}
+
+static inline u8 virtio_cread8(struct udevice *vdev, unsigned int offset)
+{
+	u8 ret;
+
+	virtio_get_config(vdev, offset, &ret, sizeof(ret));
+	return ret;
+}
+
+static inline void virtio_cwrite8(struct udevice *vdev,
+				  unsigned int offset, u8 val)
+{
+	virtio_set_config(vdev, offset, &val, sizeof(val));
+}
+
+static inline u16 virtio_cread16(struct udevice *vdev,
+				 unsigned int offset)
+{
+	u16 ret;
+
+	virtio_get_config(vdev, offset, &ret, sizeof(ret));
+	return virtio16_to_cpu(vdev, (__force __virtio16)ret);
+}
+
+static inline void virtio_cwrite16(struct udevice *vdev,
+				   unsigned int offset, u16 val)
+{
+	val = (__force u16)cpu_to_virtio16(vdev, val);
+	virtio_set_config(vdev, offset, &val, sizeof(val));
+}
+
+static inline u32 virtio_cread32(struct udevice *vdev,
+				 unsigned int offset)
+{
+	u32 ret;
+
+	virtio_get_config(vdev, offset, &ret, sizeof(ret));
+	return virtio32_to_cpu(vdev, (__force __virtio32)ret);
+}
+
+static inline void virtio_cwrite32(struct udevice *vdev,
+				   unsigned int offset, u32 val)
+{
+	val = (__force u32)cpu_to_virtio32(vdev, val);
+	virtio_set_config(vdev, offset, &val, sizeof(val));
+}
+
+static inline u64 virtio_cread64(struct udevice *vdev,
+				 unsigned int offset)
+{
+	u64 ret;
+
+	__virtio_cread_many(vdev, offset, &ret, 1, sizeof(ret));
+	return virtio64_to_cpu(vdev, (__force __virtio64)ret);
+}
+
+static inline void virtio_cwrite64(struct udevice *vdev,
+				   unsigned int offset, u64 val)
+{
+	val = (__force u64)cpu_to_virtio64(vdev, val);
+	virtio_set_config(vdev, offset, &val, sizeof(val));
+}
+
+/* Config space read accessor */
+#define virtio_cread(vdev, structname, member, ptr)			\
+	do {								\
+		/* Must match the member's type, and be integer */	\
+		if (!typecheck(typeof((((structname *)0)->member)), *(ptr))) \
+			(*ptr) = 1;					\
+									\
+		switch (sizeof(*ptr)) {					\
+		case 1:							\
+			*(ptr) = virtio_cread8(vdev,			\
+					       offsetof(structname, member)); \
+			break;						\
+		case 2:							\
+			*(ptr) = virtio_cread16(vdev,			\
+						offsetof(structname, member)); \
+			break;						\
+		case 4:							\
+			*(ptr) = virtio_cread32(vdev,			\
+						offsetof(structname, member)); \
+			break;						\
+		case 8:							\
+			*(ptr) = virtio_cread64(vdev,			\
+						offsetof(structname, member)); \
+			break;						\
+		default:						\
+			WARN_ON(true);					\
+		}							\
+	} while (0)
+
+/* Config space write accessor */
+#define virtio_cwrite(vdev, structname, member, ptr)			\
+	do {								\
+		/* Must match the member's type, and be integer */	\
+		if (!typecheck(typeof((((structname *)0)->member)), *(ptr))) \
+			WARN_ON((*ptr) == 1);				\
+									\
+		switch (sizeof(*ptr)) {					\
+		case 1:							\
+			virtio_cwrite8(vdev,				\
+				       offsetof(structname, member),	\
+				       *(ptr));				\
+			break;						\
+		case 2:							\
+			virtio_cwrite16(vdev,				\
+					offsetof(structname, member),	\
+					*(ptr));			\
+			break;						\
+		case 4:							\
+			virtio_cwrite32(vdev,				\
+					offsetof(structname, member),	\
+					*(ptr));			\
+			break;						\
+		case 8:							\
+			virtio_cwrite64(vdev,				\
+					offsetof(structname, member),	\
+					*(ptr));			\
+			break;						\
+		default:						\
+			WARN_ON(true);					\
+		}							\
+	} while (0)
+
+/* Conditional config space accessors */
+#define virtio_cread_feature(vdev, fbit, structname, member, ptr)	\
+	({								\
+		int _r = 0;						\
+		if (!virtio_has_feature(vdev, fbit))			\
+			_r = -ENOENT;					\
+		else							\
+			virtio_cread(vdev, structname, member, ptr);	\
+		_r;							\
+	})
+
+#endif /* __VIRTIO_H__ */