From 95fa0405c5991726e06c08ffcd8ff872f7fb4f2d Mon Sep 17 00:00:00 2001 From: Haiyang Zhang Date: Mon, 28 Nov 2011 13:35:35 -0800 Subject: staging: hv: move hv_netvsc out of staging area hv_netvsc has been reviewed on netdev mailing list on 6/09/2011. All recommended changes have been made. We are requesting to move it out of staging area. Signed-off-by: Haiyang Zhang Signed-off-by: KY Srinivasan Signed-off-by: Mike Sterling Acked-by: Stephen Hemminger Acked-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- drivers/net/Kconfig | 2 + drivers/net/Makefile | 2 + drivers/net/hyperv/Kconfig | 5 + drivers/net/hyperv/Makefile | 3 + drivers/net/hyperv/hyperv_net.h | 1058 +++++++++++++++++++++++++++++++++++++ drivers/net/hyperv/netvsc.c | 939 ++++++++++++++++++++++++++++++++ drivers/net/hyperv/netvsc_drv.c | 456 ++++++++++++++++ drivers/net/hyperv/rndis_filter.c | 855 ++++++++++++++++++++++++++++++ 8 files changed, 3320 insertions(+) create mode 100644 drivers/net/hyperv/Kconfig create mode 100644 drivers/net/hyperv/Makefile create mode 100644 drivers/net/hyperv/hyperv_net.h create mode 100644 drivers/net/hyperv/netvsc.c create mode 100644 drivers/net/hyperv/netvsc_drv.c create mode 100644 drivers/net/hyperv/rndis_filter.c (limited to 'drivers/net') diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 654a5e9..99aa7fa 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -338,4 +338,6 @@ config VMXNET3 To compile this driver as a module, choose M here: the module will be called vmxnet3. +source "drivers/net/hyperv/Kconfig" + endif # NETDEVICES diff --git a/drivers/net/Makefile b/drivers/net/Makefile index fa877cd..a81192b 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -66,3 +66,5 @@ obj-$(CONFIG_USB_USBNET) += usb/ obj-$(CONFIG_USB_ZD1201) += usb/ obj-$(CONFIG_USB_IPHETH) += usb/ obj-$(CONFIG_USB_CDC_PHONET) += usb/ + +obj-$(CONFIG_HYPERV_NET) += hyperv/ diff --git a/drivers/net/hyperv/Kconfig b/drivers/net/hyperv/Kconfig new file mode 100644 index 0000000..936968d --- /dev/null +++ b/drivers/net/hyperv/Kconfig @@ -0,0 +1,5 @@ +config HYPERV_NET + tristate "Microsoft Hyper-V virtual network driver" + depends on HYPERV + help + Select this option to enable the Hyper-V virtual network driver. diff --git a/drivers/net/hyperv/Makefile b/drivers/net/hyperv/Makefile new file mode 100644 index 0000000..c8a6682 --- /dev/null +++ b/drivers/net/hyperv/Makefile @@ -0,0 +1,3 @@ +obj-$(CONFIG_HYPERV_NET) += hv_netvsc.o + +hv_netvsc-y := netvsc_drv.o netvsc.o rndis_filter.o diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h new file mode 100644 index 0000000..ac1ec84 --- /dev/null +++ b/drivers/net/hyperv/hyperv_net.h @@ -0,0 +1,1058 @@ +/* + * + * Copyright (c) 2011, Microsoft Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple + * Place - Suite 330, Boston, MA 02111-1307 USA. + * + * Authors: + * Haiyang Zhang + * Hank Janssen + * K. Y. Srinivasan + * + */ + +#ifndef _HYPERV_NET_H +#define _HYPERV_NET_H + +#include +#include + +/* Fwd declaration */ +struct hv_netvsc_packet; + +/* Represent the xfer page packet which contains 1 or more netvsc packet */ +struct xferpage_packet { + struct list_head list_ent; + + /* # of netvsc packets this xfer packet contains */ + u32 count; +}; + +/* The number of pages which are enough to cover jumbo frame buffer. */ +#define NETVSC_PACKET_MAXPAGE 4 + +/* + * Represent netvsc packet which contains 1 RNDIS and 1 ethernet frame + * within the RNDIS + */ +struct hv_netvsc_packet { + /* Bookkeeping stuff */ + struct list_head list_ent; + + struct hv_device *device; + bool is_data_pkt; + + /* + * Valid only for receives when we break a xfer page packet + * into multiple netvsc packets + */ + struct xferpage_packet *xfer_page_pkt; + + union { + struct { + u64 recv_completion_tid; + void *recv_completion_ctx; + void (*recv_completion)(void *context); + } recv; + struct { + u64 send_completion_tid; + void *send_completion_ctx; + void (*send_completion)(void *context); + } send; + } completion; + + /* This points to the memory after page_buf */ + void *extension; + + u32 total_data_buflen; + /* Points to the send/receive buffer where the ethernet frame is */ + u32 page_buf_cnt; + struct hv_page_buffer page_buf[NETVSC_PACKET_MAXPAGE]; +}; + +struct netvsc_device_info { + unsigned char mac_adr[6]; + bool link_state; /* 0 - link up, 1 - link down */ + int ring_size; +}; + +/* Interface */ +int netvsc_device_add(struct hv_device *device, void *additional_info); +int netvsc_device_remove(struct hv_device *device); +int netvsc_send(struct hv_device *device, + struct hv_netvsc_packet *packet); +void netvsc_linkstatus_callback(struct hv_device *device_obj, + unsigned int status); +int netvsc_recv_callback(struct hv_device *device_obj, + struct hv_netvsc_packet *packet); +int rndis_filter_open(struct hv_device *dev); +int rndis_filter_close(struct hv_device *dev); +int rndis_filter_device_add(struct hv_device *dev, + void *additional_info); +void rndis_filter_device_remove(struct hv_device *dev); +int rndis_filter_receive(struct hv_device *dev, + struct hv_netvsc_packet *pkt); + + + +int rndis_filter_send(struct hv_device *dev, + struct hv_netvsc_packet *pkt); + +#define NVSP_INVALID_PROTOCOL_VERSION ((u32)0xFFFFFFFF) + +#define NVSP_PROTOCOL_VERSION_1 2 +#define NVSP_MIN_PROTOCOL_VERSION NVSP_PROTOCOL_VERSION_1 +#define NVSP_MAX_PROTOCOL_VERSION NVSP_PROTOCOL_VERSION_1 + +enum { + NVSP_MSG_TYPE_NONE = 0, + + /* Init Messages */ + NVSP_MSG_TYPE_INIT = 1, + NVSP_MSG_TYPE_INIT_COMPLETE = 2, + + NVSP_VERSION_MSG_START = 100, + + /* Version 1 Messages */ + NVSP_MSG1_TYPE_SEND_NDIS_VER = NVSP_VERSION_MSG_START, + + NVSP_MSG1_TYPE_SEND_RECV_BUF, + NVSP_MSG1_TYPE_SEND_RECV_BUF_COMPLETE, + NVSP_MSG1_TYPE_REVOKE_RECV_BUF, + + NVSP_MSG1_TYPE_SEND_SEND_BUF, + NVSP_MSG1_TYPE_SEND_SEND_BUF_COMPLETE, + NVSP_MSG1_TYPE_REVOKE_SEND_BUF, + + NVSP_MSG1_TYPE_SEND_RNDIS_PKT, + NVSP_MSG1_TYPE_SEND_RNDIS_PKT_COMPLETE, + + /* + * This should be set to the number of messages for the version with + * the maximum number of messages. + */ + NVSP_NUM_MSG_PER_VERSION = 9, +}; + +enum { + NVSP_STAT_NONE = 0, + NVSP_STAT_SUCCESS, + NVSP_STAT_FAIL, + NVSP_STAT_PROTOCOL_TOO_NEW, + NVSP_STAT_PROTOCOL_TOO_OLD, + NVSP_STAT_INVALID_RNDIS_PKT, + NVSP_STAT_BUSY, + NVSP_STAT_MAX, +}; + +struct nvsp_message_header { + u32 msg_type; +}; + +/* Init Messages */ + +/* + * This message is used by the VSC to initialize the channel after the channels + * has been opened. This message should never include anything other then + * versioning (i.e. this message will be the same for ever). + */ +struct nvsp_message_init { + u32 min_protocol_ver; + u32 max_protocol_ver; +} __packed; + +/* + * This message is used by the VSP to complete the initialization of the + * channel. This message should never include anything other then versioning + * (i.e. this message will be the same for ever). + */ +struct nvsp_message_init_complete { + u32 negotiated_protocol_ver; + u32 max_mdl_chain_len; + u32 status; +} __packed; + +union nvsp_message_init_uber { + struct nvsp_message_init init; + struct nvsp_message_init_complete init_complete; +} __packed; + +/* Version 1 Messages */ + +/* + * This message is used by the VSC to send the NDIS version to the VSP. The VSP + * can use this information when handling OIDs sent by the VSC. + */ +struct nvsp_1_message_send_ndis_version { + u32 ndis_major_ver; + u32 ndis_minor_ver; +} __packed; + +/* + * This message is used by the VSC to send a receive buffer to the VSP. The VSP + * can then use the receive buffer to send data to the VSC. + */ +struct nvsp_1_message_send_receive_buffer { + u32 gpadl_handle; + u16 id; +} __packed; + +struct nvsp_1_receive_buffer_section { + u32 offset; + u32 sub_alloc_size; + u32 num_sub_allocs; + u32 end_offset; +} __packed; + +/* + * This message is used by the VSP to acknowledge a receive buffer send by the + * VSC. This message must be sent by the VSP before the VSP uses the receive + * buffer. + */ +struct nvsp_1_message_send_receive_buffer_complete { + u32 status; + u32 num_sections; + + /* + * The receive buffer is split into two parts, a large suballocation + * section and a small suballocation section. These sections are then + * suballocated by a certain size. + */ + + /* + * For example, the following break up of the receive buffer has 6 + * large suballocations and 10 small suballocations. + */ + + /* + * | Large Section | | Small Section | + * ------------------------------------------------------------ + * | | | | | | | | | | | | | | | | | | + * | | + * LargeOffset SmallOffset + */ + + struct nvsp_1_receive_buffer_section sections[1]; +} __packed; + +/* + * This message is sent by the VSC to revoke the receive buffer. After the VSP + * completes this transaction, the vsp should never use the receive buffer + * again. + */ +struct nvsp_1_message_revoke_receive_buffer { + u16 id; +}; + +/* + * This message is used by the VSC to send a send buffer to the VSP. The VSC + * can then use the send buffer to send data to the VSP. + */ +struct nvsp_1_message_send_send_buffer { + u32 gpadl_handle; + u16 id; +} __packed; + +/* + * This message is used by the VSP to acknowledge a send buffer sent by the + * VSC. This message must be sent by the VSP before the VSP uses the sent + * buffer. + */ +struct nvsp_1_message_send_send_buffer_complete { + u32 status; + + /* + * The VSC gets to choose the size of the send buffer and the VSP gets + * to choose the sections size of the buffer. This was done to enable + * dynamic reconfigurations when the cost of GPA-direct buffers + * decreases. + */ + u32 section_size; +} __packed; + +/* + * This message is sent by the VSC to revoke the send buffer. After the VSP + * completes this transaction, the vsp should never use the send buffer again. + */ +struct nvsp_1_message_revoke_send_buffer { + u16 id; +}; + +/* + * This message is used by both the VSP and the VSC to send a RNDIS message to + * the opposite channel endpoint. + */ +struct nvsp_1_message_send_rndis_packet { + /* + * This field is specified by RNIDS. They assume there's two different + * channels of communication. However, the Network VSP only has one. + * Therefore, the channel travels with the RNDIS packet. + */ + u32 channel_type; + + /* + * This field is used to send part or all of the data through a send + * buffer. This values specifies an index into the send buffer. If the + * index is 0xFFFFFFFF, then the send buffer is not being used and all + * of the data was sent through other VMBus mechanisms. + */ + u32 send_buf_section_index; + u32 send_buf_section_size; +} __packed; + +/* + * This message is used by both the VSP and the VSC to complete a RNDIS message + * to the opposite channel endpoint. At this point, the initiator of this + * message cannot use any resources associated with the original RNDIS packet. + */ +struct nvsp_1_message_send_rndis_packet_complete { + u32 status; +}; + +union nvsp_1_message_uber { + struct nvsp_1_message_send_ndis_version send_ndis_ver; + + struct nvsp_1_message_send_receive_buffer send_recv_buf; + struct nvsp_1_message_send_receive_buffer_complete + send_recv_buf_complete; + struct nvsp_1_message_revoke_receive_buffer revoke_recv_buf; + + struct nvsp_1_message_send_send_buffer send_send_buf; + struct nvsp_1_message_send_send_buffer_complete send_send_buf_complete; + struct nvsp_1_message_revoke_send_buffer revoke_send_buf; + + struct nvsp_1_message_send_rndis_packet send_rndis_pkt; + struct nvsp_1_message_send_rndis_packet_complete + send_rndis_pkt_complete; +} __packed; + +union nvsp_all_messages { + union nvsp_message_init_uber init_msg; + union nvsp_1_message_uber v1_msg; +} __packed; + +/* ALL Messages */ +struct nvsp_message { + struct nvsp_message_header hdr; + union nvsp_all_messages msg; +} __packed; + + + + +/* #define NVSC_MIN_PROTOCOL_VERSION 1 */ +/* #define NVSC_MAX_PROTOCOL_VERSION 1 */ + +#define NETVSC_RECEIVE_BUFFER_SIZE (1024*1024) /* 1MB */ + +#define NETVSC_RECEIVE_BUFFER_ID 0xcafe + +#define NETVSC_RECEIVE_SG_COUNT 1 + +/* Preallocated receive packets */ +#define NETVSC_RECEIVE_PACKETLIST_COUNT 256 + +#define NETVSC_PACKET_SIZE 2048 + +/* Per netvsc channel-specific */ +struct netvsc_device { + struct hv_device *dev; + + atomic_t num_outstanding_sends; + bool destroy; + /* + * List of free preallocated hv_netvsc_packet to represent receive + * packet + */ + struct list_head recv_pkt_list; + spinlock_t recv_pkt_list_lock; + + /* Receive buffer allocated by us but manages by NetVSP */ + void *recv_buf; + u32 recv_buf_size; + u32 recv_buf_gpadl_handle; + u32 recv_section_cnt; + struct nvsp_1_receive_buffer_section *recv_section; + + /* Used for NetVSP initialization protocol */ + struct completion channel_init_wait; + struct nvsp_message channel_init_pkt; + + struct nvsp_message revoke_packet; + /* unsigned char HwMacAddr[HW_MACADDR_LEN]; */ + + struct net_device *ndev; + + /* Holds rndis device info */ + void *extension; +}; + + +/* Status codes */ + + +#ifndef STATUS_SUCCESS +#define STATUS_SUCCESS (0x00000000L) +#endif + +#ifndef STATUS_UNSUCCESSFUL +#define STATUS_UNSUCCESSFUL (0xC0000001L) +#endif + +#ifndef STATUS_PENDING +#define STATUS_PENDING (0x00000103L) +#endif + +#ifndef STATUS_INSUFFICIENT_RESOURCES +#define STATUS_INSUFFICIENT_RESOURCES (0xC000009AL) +#endif + +#ifndef STATUS_BUFFER_OVERFLOW +#define STATUS_BUFFER_OVERFLOW (0x80000005L) +#endif + +#ifndef STATUS_NOT_SUPPORTED +#define STATUS_NOT_SUPPORTED (0xC00000BBL) +#endif + +#define RNDIS_STATUS_SUCCESS (STATUS_SUCCESS) +#define RNDIS_STATUS_PENDING (STATUS_PENDING) +#define RNDIS_STATUS_NOT_RECOGNIZED (0x00010001L) +#define RNDIS_STATUS_NOT_COPIED (0x00010002L) +#define RNDIS_STATUS_NOT_ACCEPTED (0x00010003L) +#define RNDIS_STATUS_CALL_ACTIVE (0x00010007L) + +#define RNDIS_STATUS_ONLINE (0x40010003L) +#define RNDIS_STATUS_RESET_START (0x40010004L) +#define RNDIS_STATUS_RESET_END (0x40010005L) +#define RNDIS_STATUS_RING_STATUS (0x40010006L) +#define RNDIS_STATUS_CLOSED (0x40010007L) +#define RNDIS_STATUS_WAN_LINE_UP (0x40010008L) +#define RNDIS_STATUS_WAN_LINE_DOWN (0x40010009L) +#define RNDIS_STATUS_WAN_FRAGMENT (0x4001000AL) +#define RNDIS_STATUS_MEDIA_CONNECT (0x4001000BL) +#define RNDIS_STATUS_MEDIA_DISCONNECT (0x4001000CL) +#define RNDIS_STATUS_HARDWARE_LINE_UP (0x4001000DL) +#define RNDIS_STATUS_HARDWARE_LINE_DOWN (0x4001000EL) +#define RNDIS_STATUS_INTERFACE_UP (0x4001000FL) +#define RNDIS_STATUS_INTERFACE_DOWN (0x40010010L) +#define RNDIS_STATUS_MEDIA_BUSY (0x40010011L) +#define RNDIS_STATUS_MEDIA_SPECIFIC_INDICATION (0x40010012L) +#define RNDIS_STATUS_WW_INDICATION RDIA_SPECIFIC_INDICATION +#define RNDIS_STATUS_LINK_SPEED_CHANGE (0x40010013L) + +#define RNDIS_STATUS_NOT_RESETTABLE (0x80010001L) +#define RNDIS_STATUS_SOFT_ERRORS (0x80010003L) +#define RNDIS_STATUS_HARD_ERRORS (0x80010004L) +#define RNDIS_STATUS_BUFFER_OVERFLOW (STATUS_BUFFER_OVERFLOW) + +#define RNDIS_STATUS_FAILURE (STATUS_UNSUCCESSFUL) +#define RNDIS_STATUS_RESOURCES (STATUS_INSUFFICIENT_RESOURCES) +#define RNDIS_STATUS_CLOSING (0xC0010002L) +#define RNDIS_STATUS_BAD_VERSION (0xC0010004L) +#define RNDIS_STATUS_BAD_CHARACTERISTICS (0xC0010005L) +#define RNDIS_STATUS_ADAPTER_NOT_FOUND (0xC0010006L) +#define RNDIS_STATUS_OPEN_FAILED (0xC0010007L) +#define RNDIS_STATUS_DEVICE_FAILED (0xC0010008L) +#define RNDIS_STATUS_MULTICAST_FULL (0xC0010009L) +#define RNDIS_STATUS_MULTICAST_EXISTS (0xC001000AL) +#define RNDIS_STATUS_MULTICAST_NOT_FOUND (0xC001000BL) +#define RNDIS_STATUS_REQUEST_ABORTED (0xC001000CL) +#define RNDIS_STATUS_RESET_IN_PROGRESS (0xC001000DL) +#define RNDIS_STATUS_CLOSING_INDICATING (0xC001000EL) +#define RNDIS_STATUS_NOT_SUPPORTED (STATUS_NOT_SUPPORTED) +#define RNDIS_STATUS_INVALID_PACKET (0xC001000FL) +#define RNDIS_STATUS_OPEN_LIST_FULL (0xC0010010L) +#define RNDIS_STATUS_ADAPTER_NOT_READY (0xC0010011L) +#define RNDIS_STATUS_ADAPTER_NOT_OPEN (0xC0010012L) +#define RNDIS_STATUS_NOT_INDICATING (0xC0010013L) +#define RNDIS_STATUS_INVALID_LENGTH (0xC0010014L) +#define RNDIS_STATUS_INVALID_DATA (0xC0010015L) +#define RNDIS_STATUS_BUFFER_TOO_SHORT (0xC0010016L) +#define RNDIS_STATUS_INVALID_OID (0xC0010017L) +#define RNDIS_STATUS_ADAPTER_REMOVED (0xC0010018L) +#define RNDIS_STATUS_UNSUPPORTED_MEDIA (0xC0010019L) +#define RNDIS_STATUS_GROUP_ADDRESS_IN_USE (0xC001001AL) +#define RNDIS_STATUS_FILE_NOT_FOUND (0xC001001BL) +#define RNDIS_STATUS_ERROR_READING_FILE (0xC001001CL) +#define RNDIS_STATUS_ALREADY_MAPPED (0xC001001DL) +#define RNDIS_STATUS_RESOURCE_CONFLICT (0xC001001EL) +#define RNDIS_STATUS_NO_CABLE (0xC001001FL) + +#define RNDIS_STATUS_INVALID_SAP (0xC0010020L) +#define RNDIS_STATUS_SAP_IN_USE (0xC0010021L) +#define RNDIS_STATUS_INVALID_ADDRESS (0xC0010022L) +#define RNDIS_STATUS_VC_NOT_ACTIVATED (0xC0010023L) +#define RNDIS_STATUS_DEST_OUT_OF_ORDER (0xC0010024L) +#define RNDIS_STATUS_VC_NOT_AVAILABLE (0xC0010025L) +#define RNDIS_STATUS_CELLRATE_NOT_AVAILABLE (0xC0010026L) +#define RNDIS_STATUS_INCOMPATABLE_QOS (0xC0010027L) +#define RNDIS_STATUS_AAL_PARAMS_UNSUPPORTED (0xC0010028L) +#define RNDIS_STATUS_NO_ROUTE_TO_DESTINATION (0xC0010029L) + +#define RNDIS_STATUS_TOKEN_RING_OPEN_ERROR (0xC0011000L) + +/* Object Identifiers used by NdisRequest Query/Set Information */ +/* General Objects */ +#define RNDIS_OID_GEN_SUPPORTED_LIST 0x00010101 +#define RNDIS_OID_GEN_HARDWARE_STATUS 0x00010102 +#define RNDIS_OID_GEN_MEDIA_SUPPORTED 0x00010103 +#define RNDIS_OID_GEN_MEDIA_IN_USE 0x00010104 +#define RNDIS_OID_GEN_MAXIMUM_LOOKAHEAD 0x00010105 +#define RNDIS_OID_GEN_MAXIMUM_FRAME_SIZE 0x00010106 +#define RNDIS_OID_GEN_LINK_SPEED 0x00010107 +#define RNDIS_OID_GEN_TRANSMIT_BUFFER_SPACE 0x00010108 +#define RNDIS_OID_GEN_RECEIVE_BUFFER_SPACE 0x00010109 +#define RNDIS_OID_GEN_TRANSMIT_BLOCK_SIZE 0x0001010A +#define RNDIS_OID_GEN_RECEIVE_BLOCK_SIZE 0x0001010B +#define RNDIS_OID_GEN_VENDOR_ID 0x0001010C +#define RNDIS_OID_GEN_VENDOR_DESCRIPTION 0x0001010D +#define RNDIS_OID_GEN_CURRENT_PACKET_FILTER 0x0001010E +#define RNDIS_OID_GEN_CURRENT_LOOKAHEAD 0x0001010F +#define RNDIS_OID_GEN_DRIVER_VERSION 0x00010110 +#define RNDIS_OID_GEN_MAXIMUM_TOTAL_SIZE 0x00010111 +#define RNDIS_OID_GEN_PROTOCOL_OPTIONS 0x00010112 +#define RNDIS_OID_GEN_MAC_OPTIONS 0x00010113 +#define RNDIS_OID_GEN_MEDIA_CONNECT_STATUS 0x00010114 +#define RNDIS_OID_GEN_MAXIMUM_SEND_PACKETS 0x00010115 +#define RNDIS_OID_GEN_VENDOR_DRIVER_VERSION 0x00010116 +#define RNDIS_OID_GEN_NETWORK_LAYER_ADDRESSES 0x00010118 +#define RNDIS_OID_GEN_TRANSPORT_HEADER_OFFSET 0x00010119 +#define RNDIS_OID_GEN_MACHINE_NAME 0x0001021A +#define RNDIS_OID_GEN_RNDIS_CONFIG_PARAMETER 0x0001021B + +#define RNDIS_OID_GEN_XMIT_OK 0x00020101 +#define RNDIS_OID_GEN_RCV_OK 0x00020102 +#define RNDIS_OID_GEN_XMIT_ERROR 0x00020103 +#define RNDIS_OID_GEN_RCV_ERROR 0x00020104 +#define RNDIS_OID_GEN_RCV_NO_BUFFER 0x00020105 + +#define RNDIS_OID_GEN_DIRECTED_BYTES_XMIT 0x00020201 +#define RNDIS_OID_GEN_DIRECTED_FRAMES_XMIT 0x00020202 +#define RNDIS_OID_GEN_MULTICAST_BYTES_XMIT 0x00020203 +#define RNDIS_OID_GEN_MULTICAST_FRAMES_XMIT 0x00020204 +#define RNDIS_OID_GEN_BROADCAST_BYTES_XMIT 0x00020205 +#define RNDIS_OID_GEN_BROADCAST_FRAMES_XMIT 0x00020206 +#define RNDIS_OID_GEN_DIRECTED_BYTES_RCV 0x00020207 +#define RNDIS_OID_GEN_DIRECTED_FRAMES_RCV 0x00020208 +#define RNDIS_OID_GEN_MULTICAST_BYTES_RCV 0x00020209 +#define RNDIS_OID_GEN_MULTICAST_FRAMES_RCV 0x0002020A +#define RNDIS_OID_GEN_BROADCAST_BYTES_RCV 0x0002020B +#define RNDIS_OID_GEN_BROADCAST_FRAMES_RCV 0x0002020C + +#define RNDIS_OID_GEN_RCV_CRC_ERROR 0x0002020D +#define RNDIS_OID_GEN_TRANSMIT_QUEUE_LENGTH 0x0002020E + +#define RNDIS_OID_GEN_GET_TIME_CAPS 0x0002020F +#define RNDIS_OID_GEN_GET_NETCARD_TIME 0x00020210 + +/* These are connection-oriented general OIDs. */ +/* These replace the above OIDs for connection-oriented media. */ +#define RNDIS_OID_GEN_CO_SUPPORTED_LIST 0x00010101 +#define RNDIS_OID_GEN_CO_HARDWARE_STATUS 0x00010102 +#define RNDIS_OID_GEN_CO_MEDIA_SUPPORTED 0x00010103 +#define RNDIS_OID_GEN_CO_MEDIA_IN_USE 0x00010104 +#define RNDIS_OID_GEN_CO_LINK_SPEED 0x00010105 +#define RNDIS_OID_GEN_CO_VENDOR_ID 0x00010106 +#define RNDIS_OID_GEN_CO_VENDOR_DESCRIPTION 0x00010107 +#define RNDIS_OID_GEN_CO_DRIVER_VERSION 0x00010108 +#define RNDIS_OID_GEN_CO_PROTOCOL_OPTIONS 0x00010109 +#define RNDIS_OID_GEN_CO_MAC_OPTIONS 0x0001010A +#define RNDIS_OID_GEN_CO_MEDIA_CONNECT_STATUS 0x0001010B +#define RNDIS_OID_GEN_CO_VENDOR_DRIVER_VERSION 0x0001010C +#define RNDIS_OID_GEN_CO_MINIMUM_LINK_SPEED 0x0001010D + +#define RNDIS_OID_GEN_CO_GET_TIME_CAPS 0x00010201 +#define RNDIS_OID_GEN_CO_GET_NETCARD_TIME 0x00010202 + +/* These are connection-oriented statistics OIDs. */ +#define RNDIS_OID_GEN_CO_XMIT_PDUS_OK 0x00020101 +#define RNDIS_OID_GEN_CO_RCV_PDUS_OK 0x00020102 +#define RNDIS_OID_GEN_CO_XMIT_PDUS_ERROR 0x00020103 +#define RNDIS_OID_GEN_CO_RCV_PDUS_ERROR 0x00020104 +#define RNDIS_OID_GEN_CO_RCV_PDUS_NO_BUFFER 0x00020105 + + +#define RNDIS_OID_GEN_CO_RCV_CRC_ERROR 0x00020201 +#define RNDIS_OID_GEN_CO_TRANSMIT_QUEUE_LENGTH 0x00020202 +#define RNDIS_OID_GEN_CO_BYTES_XMIT 0x00020203 +#define RNDIS_OID_GEN_CO_BYTES_RCV 0x00020204 +#define RNDIS_OID_GEN_CO_BYTES_XMIT_OUTSTANDING 0x00020205 +#define RNDIS_OID_GEN_CO_NETCARD_LOAD 0x00020206 + +/* These are objects for Connection-oriented media call-managers. */ +#define RNDIS_OID_CO_ADD_PVC 0xFF000001 +#define RNDIS_OID_CO_DELETE_PVC 0xFF000002 +#define RNDIS_OID_CO_GET_CALL_INFORMATION 0xFF000003 +#define RNDIS_OID_CO_ADD_ADDRESS 0xFF000004 +#define RNDIS_OID_CO_DELETE_ADDRESS 0xFF000005 +#define RNDIS_OID_CO_GET_ADDRESSES 0xFF000006 +#define RNDIS_OID_CO_ADDRESS_CHANGE 0xFF000007 +#define RNDIS_OID_CO_SIGNALING_ENABLED 0xFF000008 +#define RNDIS_OID_CO_SIGNALING_DISABLED 0xFF000009 + +/* 802.3 Objects (Ethernet) */ +#define RNDIS_OID_802_3_PERMANENT_ADDRESS 0x01010101 +#define RNDIS_OID_802_3_CURRENT_ADDRESS 0x01010102 +#define RNDIS_OID_802_3_MULTICAST_LIST 0x01010103 +#define RNDIS_OID_802_3_MAXIMUM_LIST_SIZE 0x01010104 +#define RNDIS_OID_802_3_MAC_OPTIONS 0x01010105 + +#define NDIS_802_3_MAC_OPTION_PRIORITY 0x00000001 + +#define RNDIS_OID_802_3_RCV_ERROR_ALIGNMENT 0x01020101 +#define RNDIS_OID_802_3_XMIT_ONE_COLLISION 0x01020102 +#define RNDIS_OID_802_3_XMIT_MORE_COLLISIONS 0x01020103 + +#define RNDIS_OID_802_3_XMIT_DEFERRED 0x01020201 +#define RNDIS_OID_802_3_XMIT_MAX_COLLISIONS 0x01020202 +#define RNDIS_OID_802_3_RCV_OVERRUN 0x01020203 +#define RNDIS_OID_802_3_XMIT_UNDERRUN 0x01020204 +#define RNDIS_OID_802_3_XMIT_HEARTBEAT_FAILURE 0x01020205 +#define RNDIS_OID_802_3_XMIT_TIMES_CRS_LOST 0x01020206 +#define RNDIS_OID_802_3_XMIT_LATE_COLLISIONS 0x01020207 + +/* Remote NDIS message types */ +#define REMOTE_NDIS_PACKET_MSG 0x00000001 +#define REMOTE_NDIS_INITIALIZE_MSG 0x00000002 +#define REMOTE_NDIS_HALT_MSG 0x00000003 +#define REMOTE_NDIS_QUERY_MSG 0x00000004 +#define REMOTE_NDIS_SET_MSG 0x00000005 +#define REMOTE_NDIS_RESET_MSG 0x00000006 +#define REMOTE_NDIS_INDICATE_STATUS_MSG 0x00000007 +#define REMOTE_NDIS_KEEPALIVE_MSG 0x00000008 + +#define REMOTE_CONDIS_MP_CREATE_VC_MSG 0x00008001 +#define REMOTE_CONDIS_MP_DELETE_VC_MSG 0x00008002 +#define REMOTE_CONDIS_MP_ACTIVATE_VC_MSG 0x00008005 +#define REMOTE_CONDIS_MP_DEACTIVATE_VC_MSG 0x00008006 +#define REMOTE_CONDIS_INDICATE_STATUS_MSG 0x00008007 + +/* Remote NDIS message completion types */ +#define REMOTE_NDIS_INITIALIZE_CMPLT 0x80000002 +#define REMOTE_NDIS_QUERY_CMPLT 0x80000004 +#define REMOTE_NDIS_SET_CMPLT 0x80000005 +#define REMOTE_NDIS_RESET_CMPLT 0x80000006 +#define REMOTE_NDIS_KEEPALIVE_CMPLT 0x80000008 + +#define REMOTE_CONDIS_MP_CREATE_VC_CMPLT 0x80008001 +#define REMOTE_CONDIS_MP_DELETE_VC_CMPLT 0x80008002 +#define REMOTE_CONDIS_MP_ACTIVATE_VC_CMPLT 0x80008005 +#define REMOTE_CONDIS_MP_DEACTIVATE_VC_CMPLT 0x80008006 + +/* + * Reserved message type for private communication between lower-layer host + * driver and remote device, if necessary. + */ +#define REMOTE_NDIS_BUS_MSG 0xff000001 + +/* Defines for DeviceFlags in struct rndis_initialize_complete */ +#define RNDIS_DF_CONNECTIONLESS 0x00000001 +#define RNDIS_DF_CONNECTION_ORIENTED 0x00000002 +#define RNDIS_DF_RAW_DATA 0x00000004 + +/* Remote NDIS medium types. */ +#define RNDIS_MEDIUM_802_3 0x00000000 +#define RNDIS_MEDIUM_802_5 0x00000001 +#define RNDIS_MEDIUM_FDDI 0x00000002 +#define RNDIS_MEDIUM_WAN 0x00000003 +#define RNDIS_MEDIUM_LOCAL_TALK 0x00000004 +#define RNDIS_MEDIUM_ARCNET_RAW 0x00000006 +#define RNDIS_MEDIUM_ARCNET_878_2 0x00000007 +#define RNDIS_MEDIUM_ATM 0x00000008 +#define RNDIS_MEDIUM_WIRELESS_WAN 0x00000009 +#define RNDIS_MEDIUM_IRDA 0x0000000a +#define RNDIS_MEDIUM_CO_WAN 0x0000000b +/* Not a real medium, defined as an upper-bound */ +#define RNDIS_MEDIUM_MAX 0x0000000d + + +/* Remote NDIS medium connection states. */ +#define RNDIS_MEDIA_STATE_CONNECTED 0x00000000 +#define RNDIS_MEDIA_STATE_DISCONNECTED 0x00000001 + +/* Remote NDIS version numbers */ +#define RNDIS_MAJOR_VERSION 0x00000001 +#define RNDIS_MINOR_VERSION 0x00000000 + + +/* NdisInitialize message */ +struct rndis_initialize_request { + u32 req_id; + u32 major_ver; + u32 minor_ver; + u32 max_xfer_size; +}; + +/* Response to NdisInitialize */ +struct rndis_initialize_complete { + u32 req_id; + u32 status; + u32 major_ver; + u32 minor_ver; + u32 dev_flags; + u32 medium; + u32 max_pkt_per_msg; + u32 max_xfer_size; + u32 pkt_alignment_factor; + u32 af_list_offset; + u32 af_list_size; +}; + +/* Call manager devices only: Information about an address family */ +/* supported by the device is appended to the response to NdisInitialize. */ +struct rndis_co_address_family { + u32 address_family; + u32 major_ver; + u32 minor_ver; +}; + +/* NdisHalt message */ +struct rndis_halt_request { + u32 req_id; +}; + +/* NdisQueryRequest message */ +struct rndis_query_request { + u32 req_id; + u32 oid; + u32 info_buflen; + u32 info_buf_offset; + u32 dev_vc_handle; +}; + +/* Response to NdisQueryRequest */ +struct rndis_query_complete { + u32 req_id; + u32 status; + u32 info_buflen; + u32 info_buf_offset; +}; + +/* NdisSetRequest message */ +struct rndis_set_request { + u32 req_id; + u32 oid; + u32 info_buflen; + u32 info_buf_offset; + u32 dev_vc_handle; +}; + +/* Response to NdisSetRequest */ +struct rndis_set_complete { + u32 req_id; + u32 status; +}; + +/* NdisReset message */ +struct rndis_reset_request { + u32 reserved; +}; + +/* Response to NdisReset */ +struct rndis_reset_complete { + u32 status; + u32 addressing_reset; +}; + +/* NdisMIndicateStatus message */ +struct rndis_indicate_status { + u32 status; + u32 status_buflen; + u32 status_buf_offset; +}; + +/* Diagnostic information passed as the status buffer in */ +/* struct rndis_indicate_status messages signifying error conditions. */ +struct rndis_diagnostic_info { + u32 diag_status; + u32 error_offset; +}; + +/* NdisKeepAlive message */ +struct rndis_keepalive_request { + u32 req_id; +}; + +/* Response to NdisKeepAlive */ +struct rndis_keepalive_complete { + u32 req_id; + u32 status; +}; + +/* + * Data message. All Offset fields contain byte offsets from the beginning of + * struct rndis_packet. All Length fields are in bytes. VcHandle is set + * to 0 for connectionless data, otherwise it contains the VC handle. + */ +struct rndis_packet { + u32 data_offset; + u32 data_len; + u32 oob_data_offset; + u32 oob_data_len; + u32 num_oob_data_elements; + u32 per_pkt_info_offset; + u32 per_pkt_info_len; + u32 vc_handle; + u32 reserved; +}; + +/* Optional Out of Band data associated with a Data message. */ +struct rndis_oobd { + u32 size; + u32 type; + u32 class_info_offset; +}; + +/* Packet extension field contents associated with a Data message. */ +struct rndis_per_packet_info { + u32 size; + u32 type; + u32 per_pkt_info_offset; +}; + +/* Format of Information buffer passed in a SetRequest for the OID */ +/* OID_GEN_RNDIS_CONFIG_PARAMETER. */ +struct rndis_config_parameter_info { + u32 parameter_name_offset; + u32 parameter_name_length; + u32 parameter_type; + u32 parameter_value_offset; + u32 parameter_value_length; +}; + +/* Values for ParameterType in struct rndis_config_parameter_info */ +#define RNDIS_CONFIG_PARAM_TYPE_INTEGER 0 +#define RNDIS_CONFIG_PARAM_TYPE_STRING 2 + +/* CONDIS Miniport messages for connection oriented devices */ +/* that do not implement a call manager. */ + +/* CoNdisMiniportCreateVc message */ +struct rcondis_mp_create_vc { + u32 req_id; + u32 ndis_vc_handle; +}; + +/* Response to CoNdisMiniportCreateVc */ +struct rcondis_mp_create_vc_complete { + u32 req_id; + u32 dev_vc_handle; + u32 status; +}; + +/* CoNdisMiniportDeleteVc message */ +struct rcondis_mp_delete_vc { + u32 req_id; + u32 dev_vc_handle; +}; + +/* Response to CoNdisMiniportDeleteVc */ +struct rcondis_mp_delete_vc_complete { + u32 req_id; + u32 status; +}; + +/* CoNdisMiniportQueryRequest message */ +struct rcondis_mp_query_request { + u32 req_id; + u32 request_type; + u32 oid; + u32 dev_vc_handle; + u32 info_buflen; + u32 info_buf_offset; +}; + +/* CoNdisMiniportSetRequest message */ +struct rcondis_mp_set_request { + u32 req_id; + u32 request_type; + u32 oid; + u32 dev_vc_handle; + u32 info_buflen; + u32 info_buf_offset; +}; + +/* CoNdisIndicateStatus message */ +struct rcondis_indicate_status { + u32 ndis_vc_handle; + u32 status; + u32 status_buflen; + u32 status_buf_offset; +}; + +/* CONDIS Call/VC parameters */ +struct rcondis_specific_parameters { + u32 parameter_type; + u32 parameter_length; + u32 parameter_lffset; +}; + +struct rcondis_media_parameters { + u32 flags; + u32 reserved1; + u32 reserved2; + struct rcondis_specific_parameters media_specific; +}; + +struct rndis_flowspec { + u32 token_rate; + u32 token_bucket_size; + u32 peak_bandwidth; + u32 latency; + u32 delay_variation; + u32 service_type; + u32 max_sdu_size; + u32 minimum_policed_size; +}; + +struct rcondis_call_manager_parameters { + struct rndis_flowspec transmit; + struct rndis_flowspec receive; + struct rcondis_specific_parameters call_mgr_specific; +}; + +/* CoNdisMiniportActivateVc message */ +struct rcondis_mp_activate_vc_request { + u32 req_id; + u32 flags; + u32 dev_vc_handle; + u32 media_params_offset; + u32 media_params_length; + u32 call_mgr_params_offset; + u32 call_mgr_params_length; +}; + +/* Response to CoNdisMiniportActivateVc */ +struct rcondis_mp_activate_vc_complete { + u32 req_id; + u32 status; +}; + +/* CoNdisMiniportDeactivateVc message */ +struct rcondis_mp_deactivate_vc_request { + u32 req_id; + u32 flags; + u32 dev_vc_handle; +}; + +/* Response to CoNdisMiniportDeactivateVc */ +struct rcondis_mp_deactivate_vc_complete { + u32 req_id; + u32 status; +}; + + +/* union with all of the RNDIS messages */ +union rndis_message_container { + struct rndis_packet pkt; + struct rndis_initialize_request init_req; + struct rndis_halt_request halt_req; + struct rndis_query_request query_req; + struct rndis_set_request set_req; + struct rndis_reset_request reset_req; + struct rndis_keepalive_request keep_alive_req; + struct rndis_indicate_status indicate_status; + struct rndis_initialize_complete init_complete; + struct rndis_query_complete query_complete; + struct rndis_set_complete set_complete; + struct rndis_reset_complete reset_complete; + struct rndis_keepalive_complete keep_alive_complete; + struct rcondis_mp_create_vc co_miniport_create_vc; + struct rcondis_mp_delete_vc co_miniport_delete_vc; + struct rcondis_indicate_status co_indicate_status; + struct rcondis_mp_activate_vc_request co_miniport_activate_vc; + struct rcondis_mp_deactivate_vc_request co_miniport_deactivate_vc; + struct rcondis_mp_create_vc_complete co_miniport_create_vc_complete; + struct rcondis_mp_delete_vc_complete co_miniport_delete_vc_complete; + struct rcondis_mp_activate_vc_complete co_miniport_activate_vc_complete; + struct rcondis_mp_deactivate_vc_complete + co_miniport_deactivate_vc_complete; +}; + +/* Remote NDIS message format */ +struct rndis_message { + u32 ndis_msg_type; + + /* Total length of this message, from the beginning */ + /* of the sruct rndis_message, in bytes. */ + u32 msg_len; + + /* Actual message */ + union rndis_message_container msg; +}; + + +struct rndis_filter_packet { + void *completion_ctx; + void (*completion)(void *context); + struct rndis_message msg; +}; + +/* Handy macros */ + +/* get the size of an RNDIS message. Pass in the message type, */ +/* struct rndis_set_request, struct rndis_packet for example */ +#define RNDIS_MESSAGE_SIZE(msg) \ + (sizeof(msg) + (sizeof(struct rndis_message) - \ + sizeof(union rndis_message_container))) + +/* get pointer to info buffer with message pointer */ +#define MESSAGE_TO_INFO_BUFFER(msg) \ + (((unsigned char *)(msg)) + msg->info_buf_offset) + +/* get pointer to status buffer with message pointer */ +#define MESSAGE_TO_STATUS_BUFFER(msg) \ + (((unsigned char *)(msg)) + msg->status_buf_offset) + +/* get pointer to OOBD buffer with message pointer */ +#define MESSAGE_TO_OOBD_BUFFER(msg) \ + (((unsigned char *)(msg)) + msg->oob_data_offset) + +/* get pointer to data buffer with message pointer */ +#define MESSAGE_TO_DATA_BUFFER(msg) \ + (((unsigned char *)(msg)) + msg->per_pkt_info_offset) + +/* get pointer to contained message from NDIS_MESSAGE pointer */ +#define RNDIS_MESSAGE_PTR_TO_MESSAGE_PTR(rndis_msg) \ + ((void *) &rndis_msg->msg) + +/* get pointer to contained message from NDIS_MESSAGE pointer */ +#define RNDIS_MESSAGE_RAW_PTR_TO_MESSAGE_PTR(rndis_msg) \ + ((void *) rndis_msg) + + +#define __struct_bcount(x) + + + +#define RNDIS_HEADER_SIZE (sizeof(struct rndis_message) - \ + sizeof(union rndis_message_container)) + +#define NDIS_PACKET_TYPE_DIRECTED 0x00000001 +#define NDIS_PACKET_TYPE_MULTICAST 0x00000002 +#define NDIS_PACKET_TYPE_ALL_MULTICAST 0x00000004 +#define NDIS_PACKET_TYPE_BROADCAST 0x00000008 +#define NDIS_PACKET_TYPE_SOURCE_ROUTING 0x00000010 +#define NDIS_PACKET_TYPE_PROMISCUOUS 0x00000020 +#define NDIS_PACKET_TYPE_SMT 0x00000040 +#define NDIS_PACKET_TYPE_ALL_LOCAL 0x00000080 +#define NDIS_PACKET_TYPE_GROUP 0x00000100 +#define NDIS_PACKET_TYPE_ALL_FUNCTIONAL 0x00000200 +#define NDIS_PACKET_TYPE_FUNCTIONAL 0x00000400 +#define NDIS_PACKET_TYPE_MAC_FRAME 0x00000800 + + + +#endif /* _HYPERV_NET_H */ diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c new file mode 100644 index 0000000..28e69a6 --- /dev/null +++ b/drivers/net/hyperv/netvsc.c @@ -0,0 +1,939 @@ +/* + * Copyright (c) 2009, Microsoft Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple + * Place - Suite 330, Boston, MA 02111-1307 USA. + * + * Authors: + * Haiyang Zhang + * Hank Janssen + */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "hyperv_net.h" + + +static struct netvsc_device *alloc_net_device(struct hv_device *device) +{ + struct netvsc_device *net_device; + struct net_device *ndev = hv_get_drvdata(device); + + net_device = kzalloc(sizeof(struct netvsc_device), GFP_KERNEL); + if (!net_device) + return NULL; + + + net_device->destroy = false; + net_device->dev = device; + net_device->ndev = ndev; + + hv_set_drvdata(device, net_device); + return net_device; +} + +static struct netvsc_device *get_outbound_net_device(struct hv_device *device) +{ + struct netvsc_device *net_device; + + net_device = hv_get_drvdata(device); + if (net_device && net_device->destroy) + net_device = NULL; + + return net_device; +} + +static struct netvsc_device *get_inbound_net_device(struct hv_device *device) +{ + struct netvsc_device *net_device; + + net_device = hv_get_drvdata(device); + + if (!net_device) + goto get_in_err; + + if (net_device->destroy && + atomic_read(&net_device->num_outstanding_sends) == 0) + net_device = NULL; + +get_in_err: + return net_device; +} + + +static int netvsc_destroy_recv_buf(struct netvsc_device *net_device) +{ + struct nvsp_message *revoke_packet; + int ret = 0; + struct net_device *ndev = net_device->ndev; + + /* + * If we got a section count, it means we received a + * SendReceiveBufferComplete msg (ie sent + * NvspMessage1TypeSendReceiveBuffer msg) therefore, we need + * to send a revoke msg here + */ + if (net_device->recv_section_cnt) { + /* Send the revoke receive buffer */ + revoke_packet = &net_device->revoke_packet; + memset(revoke_packet, 0, sizeof(struct nvsp_message)); + + revoke_packet->hdr.msg_type = + NVSP_MSG1_TYPE_REVOKE_RECV_BUF; + revoke_packet->msg.v1_msg. + revoke_recv_buf.id = NETVSC_RECEIVE_BUFFER_ID; + + ret = vmbus_sendpacket(net_device->dev->channel, + revoke_packet, + sizeof(struct nvsp_message), + (unsigned long)revoke_packet, + VM_PKT_DATA_INBAND, 0); + /* + * If we failed here, we might as well return and + * have a leak rather than continue and a bugchk + */ + if (ret != 0) { + netdev_err(ndev, "unable to send " + "revoke receive buffer to netvsp\n"); + return ret; + } + } + + /* Teardown the gpadl on the vsp end */ + if (net_device->recv_buf_gpadl_handle) { + ret = vmbus_teardown_gpadl(net_device->dev->channel, + net_device->recv_buf_gpadl_handle); + + /* If we failed here, we might as well return and have a leak + * rather than continue and a bugchk + */ + if (ret != 0) { + netdev_err(ndev, + "unable to teardown receive buffer's gpadl\n"); + return ret; + } + net_device->recv_buf_gpadl_handle = 0; + } + + if (net_device->recv_buf) { + /* Free up the receive buffer */ + free_pages((unsigned long)net_device->recv_buf, + get_order(net_device->recv_buf_size)); + net_device->recv_buf = NULL; + } + + if (net_device->recv_section) { + net_device->recv_section_cnt = 0; + kfree(net_device->recv_section); + net_device->recv_section = NULL; + } + + return ret; +} + +static int netvsc_init_recv_buf(struct hv_device *device) +{ + int ret = 0; + int t; + struct netvsc_device *net_device; + struct nvsp_message *init_packet; + struct net_device *ndev; + + net_device = get_outbound_net_device(device); + if (!net_device) + return -ENODEV; + ndev = net_device->ndev; + + net_device->recv_buf = + (void *)__get_free_pages(GFP_KERNEL|__GFP_ZERO, + get_order(net_device->recv_buf_size)); + if (!net_device->recv_buf) { + netdev_err(ndev, "unable to allocate receive " + "buffer of size %d\n", net_device->recv_buf_size); + ret = -ENOMEM; + goto cleanup; + } + + /* + * Establish the gpadl handle for this buffer on this + * channel. Note: This call uses the vmbus connection rather + * than the channel to establish the gpadl handle. + */ + ret = vmbus_establish_gpadl(device->channel, net_device->recv_buf, + net_device->recv_buf_size, + &net_device->recv_buf_gpadl_handle); + if (ret != 0) { + netdev_err(ndev, + "unable to establish receive buffer's gpadl\n"); + goto cleanup; + } + + + /* Notify the NetVsp of the gpadl handle */ + init_packet = &net_device->channel_init_pkt; + + memset(init_packet, 0, sizeof(struct nvsp_message)); + + init_packet->hdr.msg_type = NVSP_MSG1_TYPE_SEND_RECV_BUF; + init_packet->msg.v1_msg.send_recv_buf. + gpadl_handle = net_device->recv_buf_gpadl_handle; + init_packet->msg.v1_msg. + send_recv_buf.id = NETVSC_RECEIVE_BUFFER_ID; + + /* Send the gpadl notification request */ + ret = vmbus_sendpacket(device->channel, init_packet, + sizeof(struct nvsp_message), + (unsigned long)init_packet, + VM_PKT_DATA_INBAND, + VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED); + if (ret != 0) { + netdev_err(ndev, + "unable to send receive buffer's gpadl to netvsp\n"); + goto cleanup; + } + + t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ); + BUG_ON(t == 0); + + + /* Check the response */ + if (init_packet->msg.v1_msg. + send_recv_buf_complete.status != NVSP_STAT_SUCCESS) { + netdev_err(ndev, "Unable to complete receive buffer " + "initialization with NetVsp - status %d\n", + init_packet->msg.v1_msg. + send_recv_buf_complete.status); + ret = -EINVAL; + goto cleanup; + } + + /* Parse the response */ + + net_device->recv_section_cnt = init_packet->msg. + v1_msg.send_recv_buf_complete.num_sections; + + net_device->recv_section = kmemdup(init_packet->msg.v1_msg.send_recv_buf_complete.sections, + net_device->recv_section_cnt * sizeof(struct nvsp_1_receive_buffer_section), + GFP_KERNEL); + if (net_device->recv_section == NULL) { + ret = -EINVAL; + goto cleanup; + } + + /* + * For 1st release, there should only be 1 section that represents the + * entire receive buffer + */ + if (net_device->recv_section_cnt != 1 || + net_device->recv_section->offset != 0) { + ret = -EINVAL; + goto cleanup; + } + + goto exit; + +cleanup: + netvsc_destroy_recv_buf(net_device); + +exit: + return ret; +} + + +static int netvsc_connect_vsp(struct hv_device *device) +{ + int ret, t; + struct netvsc_device *net_device; + struct nvsp_message *init_packet; + int ndis_version; + struct net_device *ndev; + + net_device = get_outbound_net_device(device); + if (!net_device) + return -ENODEV; + ndev = net_device->ndev; + + init_packet = &net_device->channel_init_pkt; + + memset(init_packet, 0, sizeof(struct nvsp_message)); + init_packet->hdr.msg_type = NVSP_MSG_TYPE_INIT; + init_packet->msg.init_msg.init.min_protocol_ver = + NVSP_MIN_PROTOCOL_VERSION; + init_packet->msg.init_msg.init.max_protocol_ver = + NVSP_MAX_PROTOCOL_VERSION; + + /* Send the init request */ + ret = vmbus_sendpacket(device->channel, init_packet, + sizeof(struct nvsp_message), + (unsigned long)init_packet, + VM_PKT_DATA_INBAND, + VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED); + + if (ret != 0) + goto cleanup; + + t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ); + + if (t == 0) { + ret = -ETIMEDOUT; + goto cleanup; + } + + if (init_packet->msg.init_msg.init_complete.status != + NVSP_STAT_SUCCESS) { + ret = -EINVAL; + goto cleanup; + } + + if (init_packet->msg.init_msg.init_complete. + negotiated_protocol_ver != NVSP_PROTOCOL_VERSION_1) { + ret = -EPROTO; + goto cleanup; + } + /* Send the ndis version */ + memset(init_packet, 0, sizeof(struct nvsp_message)); + + ndis_version = 0x00050000; + + init_packet->hdr.msg_type = NVSP_MSG1_TYPE_SEND_NDIS_VER; + init_packet->msg.v1_msg. + send_ndis_ver.ndis_major_ver = + (ndis_version & 0xFFFF0000) >> 16; + init_packet->msg.v1_msg. + send_ndis_ver.ndis_minor_ver = + ndis_version & 0xFFFF; + + /* Send the init request */ + ret = vmbus_sendpacket(device->channel, init_packet, + sizeof(struct nvsp_message), + (unsigned long)init_packet, + VM_PKT_DATA_INBAND, 0); + if (ret != 0) + goto cleanup; + + /* Post the big receive buffer to NetVSP */ + ret = netvsc_init_recv_buf(device); + +cleanup: + return ret; +} + +static void netvsc_disconnect_vsp(struct netvsc_device *net_device) +{ + netvsc_destroy_recv_buf(net_device); +} + +/* + * netvsc_device_remove - Callback when the root bus device is removed + */ +int netvsc_device_remove(struct hv_device *device) +{ + struct netvsc_device *net_device; + struct hv_netvsc_packet *netvsc_packet, *pos; + unsigned long flags; + + net_device = hv_get_drvdata(device); + spin_lock_irqsave(&device->channel->inbound_lock, flags); + net_device->destroy = true; + spin_unlock_irqrestore(&device->channel->inbound_lock, flags); + + /* Wait for all send completions */ + while (atomic_read(&net_device->num_outstanding_sends)) { + dev_info(&device->device, + "waiting for %d requests to complete...\n", + atomic_read(&net_device->num_outstanding_sends)); + udelay(100); + } + + netvsc_disconnect_vsp(net_device); + + /* + * Since we have already drained, we don't need to busy wait + * as was done in final_release_stor_device() + * Note that we cannot set the ext pointer to NULL until + * we have drained - to drain the outgoing packets, we need to + * allow incoming packets. + */ + + spin_lock_irqsave(&device->channel->inbound_lock, flags); + hv_set_drvdata(device, NULL); + spin_unlock_irqrestore(&device->channel->inbound_lock, flags); + + /* + * At this point, no one should be accessing net_device + * except in here + */ + dev_notice(&device->device, "net device safe to remove\n"); + + /* Now, we can close the channel safely */ + vmbus_close(device->channel); + + /* Release all resources */ + list_for_each_entry_safe(netvsc_packet, pos, + &net_device->recv_pkt_list, list_ent) { + list_del(&netvsc_packet->list_ent); + kfree(netvsc_packet); + } + + kfree(net_device); + return 0; +} + +static void netvsc_send_completion(struct hv_device *device, + struct vmpacket_descriptor *packet) +{ + struct netvsc_device *net_device; + struct nvsp_message *nvsp_packet; + struct hv_netvsc_packet *nvsc_packet; + struct net_device *ndev; + + net_device = get_inbound_net_device(device); + if (!net_device) + return; + ndev = net_device->ndev; + + nvsp_packet = (struct nvsp_message *)((unsigned long)packet + + (packet->offset8 << 3)); + + if ((nvsp_packet->hdr.msg_type == NVSP_MSG_TYPE_INIT_COMPLETE) || + (nvsp_packet->hdr.msg_type == + NVSP_MSG1_TYPE_SEND_RECV_BUF_COMPLETE) || + (nvsp_packet->hdr.msg_type == + NVSP_MSG1_TYPE_SEND_SEND_BUF_COMPLETE)) { + /* Copy the response back */ + memcpy(&net_device->channel_init_pkt, nvsp_packet, + sizeof(struct nvsp_message)); + complete(&net_device->channel_init_wait); + } else if (nvsp_packet->hdr.msg_type == + NVSP_MSG1_TYPE_SEND_RNDIS_PKT_COMPLETE) { + /* Get the send context */ + nvsc_packet = (struct hv_netvsc_packet *)(unsigned long) + packet->trans_id; + + /* Notify the layer above us */ + nvsc_packet->completion.send.send_completion( + nvsc_packet->completion.send.send_completion_ctx); + + atomic_dec(&net_device->num_outstanding_sends); + } else { + netdev_err(ndev, "Unknown send completion packet type- " + "%d received!!\n", nvsp_packet->hdr.msg_type); + } + +} + +int netvsc_send(struct hv_device *device, + struct hv_netvsc_packet *packet) +{ + struct netvsc_device *net_device; + int ret = 0; + struct nvsp_message sendMessage; + struct net_device *ndev; + + net_device = get_outbound_net_device(device); + if (!net_device) + return -ENODEV; + ndev = net_device->ndev; + + sendMessage.hdr.msg_type = NVSP_MSG1_TYPE_SEND_RNDIS_PKT; + if (packet->is_data_pkt) { + /* 0 is RMC_DATA; */ + sendMessage.msg.v1_msg.send_rndis_pkt.channel_type = 0; + } else { + /* 1 is RMC_CONTROL; */ + sendMessage.msg.v1_msg.send_rndis_pkt.channel_type = 1; + } + + /* Not using send buffer section */ + sendMessage.msg.v1_msg.send_rndis_pkt.send_buf_section_index = + 0xFFFFFFFF; + sendMessage.msg.v1_msg.send_rndis_pkt.send_buf_section_size = 0; + + if (packet->page_buf_cnt) { + ret = vmbus_sendpacket_pagebuffer(device->channel, + packet->page_buf, + packet->page_buf_cnt, + &sendMessage, + sizeof(struct nvsp_message), + (unsigned long)packet); + } else { + ret = vmbus_sendpacket(device->channel, &sendMessage, + sizeof(struct nvsp_message), + (unsigned long)packet, + VM_PKT_DATA_INBAND, + VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED); + + } + + if (ret != 0) + netdev_err(ndev, "Unable to send packet %p ret %d\n", + packet, ret); + else + atomic_inc(&net_device->num_outstanding_sends); + + return ret; +} + +static void netvsc_send_recv_completion(struct hv_device *device, + u64 transaction_id) +{ + struct nvsp_message recvcompMessage; + int retries = 0; + int ret; + struct net_device *ndev; + struct netvsc_device *net_device = hv_get_drvdata(device); + + ndev = net_device->ndev; + + recvcompMessage.hdr.msg_type = + NVSP_MSG1_TYPE_SEND_RNDIS_PKT_COMPLETE; + + /* FIXME: Pass in the status */ + recvcompMessage.msg.v1_msg.send_rndis_pkt_complete.status = + NVSP_STAT_SUCCESS; + +retry_send_cmplt: + /* Send the completion */ + ret = vmbus_sendpacket(device->channel, &recvcompMessage, + sizeof(struct nvsp_message), transaction_id, + VM_PKT_COMP, 0); + if (ret == 0) { + /* success */ + /* no-op */ + } else if (ret == -EAGAIN) { + /* no more room...wait a bit and attempt to retry 3 times */ + retries++; + netdev_err(ndev, "unable to send receive completion pkt" + " (tid %llx)...retrying %d\n", transaction_id, retries); + + if (retries < 4) { + udelay(100); + goto retry_send_cmplt; + } else { + netdev_err(ndev, "unable to send receive " + "completion pkt (tid %llx)...give up retrying\n", + transaction_id); + } + } else { + netdev_err(ndev, "unable to send receive " + "completion pkt - %llx\n", transaction_id); + } +} + +/* Send a receive completion packet to RNDIS device (ie NetVsp) */ +static void netvsc_receive_completion(void *context) +{ + struct hv_netvsc_packet *packet = context; + struct hv_device *device = (struct hv_device *)packet->device; + struct netvsc_device *net_device; + u64 transaction_id = 0; + bool fsend_receive_comp = false; + unsigned long flags; + struct net_device *ndev; + + /* + * Even though it seems logical to do a GetOutboundNetDevice() here to + * send out receive completion, we are using GetInboundNetDevice() + * since we may have disable outbound traffic already. + */ + net_device = get_inbound_net_device(device); + if (!net_device) + return; + ndev = net_device->ndev; + + /* Overloading use of the lock. */ + spin_lock_irqsave(&net_device->recv_pkt_list_lock, flags); + + packet->xfer_page_pkt->count--; + + /* + * Last one in the line that represent 1 xfer page packet. + * Return the xfer page packet itself to the freelist + */ + if (packet->xfer_page_pkt->count == 0) { + fsend_receive_comp = true; + transaction_id = packet->completion.recv.recv_completion_tid; + list_add_tail(&packet->xfer_page_pkt->list_ent, + &net_device->recv_pkt_list); + + } + + /* Put the packet back */ + list_add_tail(&packet->list_ent, &net_device->recv_pkt_list); + spin_unlock_irqrestore(&net_device->recv_pkt_list_lock, flags); + + /* Send a receive completion for the xfer page packet */ + if (fsend_receive_comp) + netvsc_send_recv_completion(device, transaction_id); + +} + +static void netvsc_receive(struct hv_device *device, + struct vmpacket_descriptor *packet) +{ + struct netvsc_device *net_device; + struct vmtransfer_page_packet_header *vmxferpage_packet; + struct nvsp_message *nvsp_packet; + struct hv_netvsc_packet *netvsc_packet = NULL; + unsigned long start; + unsigned long end, end_virtual; + /* struct netvsc_driver *netvscDriver; */ + struct xferpage_packet *xferpage_packet = NULL; + int i, j; + int count = 0, bytes_remain = 0; + unsigned long flags; + struct net_device *ndev; + + LIST_HEAD(listHead); + + net_device = get_inbound_net_device(device); + if (!net_device) + return; + ndev = net_device->ndev; + + /* + * All inbound packets other than send completion should be xfer page + * packet + */ + if (packet->type != VM_PKT_DATA_USING_XFER_PAGES) { + netdev_err(ndev, "Unknown packet type received - %d\n", + packet->type); + return; + } + + nvsp_packet = (struct nvsp_message *)((unsigned long)packet + + (packet->offset8 << 3)); + + /* Make sure this is a valid nvsp packet */ + if (nvsp_packet->hdr.msg_type != + NVSP_MSG1_TYPE_SEND_RNDIS_PKT) { + netdev_err(ndev, "Unknown nvsp packet type received-" + " %d\n", nvsp_packet->hdr.msg_type); + return; + } + + vmxferpage_packet = (struct vmtransfer_page_packet_header *)packet; + + if (vmxferpage_packet->xfer_pageset_id != NETVSC_RECEIVE_BUFFER_ID) { + netdev_err(ndev, "Invalid xfer page set id - " + "expecting %x got %x\n", NETVSC_RECEIVE_BUFFER_ID, + vmxferpage_packet->xfer_pageset_id); + return; + } + + /* + * Grab free packets (range count + 1) to represent this xfer + * page packet. +1 to represent the xfer page packet itself. + * We grab it here so that we know exactly how many we can + * fulfil + */ + spin_lock_irqsave(&net_device->recv_pkt_list_lock, flags); + while (!list_empty(&net_device->recv_pkt_list)) { + list_move_tail(net_device->recv_pkt_list.next, &listHead); + if (++count == vmxferpage_packet->range_cnt + 1) + break; + } + spin_unlock_irqrestore(&net_device->recv_pkt_list_lock, flags); + + /* + * We need at least 2 netvsc pkts (1 to represent the xfer + * page and at least 1 for the range) i.e. we can handled + * some of the xfer page packet ranges... + */ + if (count < 2) { + netdev_err(ndev, "Got only %d netvsc pkt...needed " + "%d pkts. Dropping this xfer page packet completely!\n", + count, vmxferpage_packet->range_cnt + 1); + + /* Return it to the freelist */ + spin_lock_irqsave(&net_device->recv_pkt_list_lock, flags); + for (i = count; i != 0; i--) { + list_move_tail(listHead.next, + &net_device->recv_pkt_list); + } + spin_unlock_irqrestore(&net_device->recv_pkt_list_lock, + flags); + + netvsc_send_recv_completion(device, + vmxferpage_packet->d.trans_id); + + return; + } + + /* Remove the 1st packet to represent the xfer page packet itself */ + xferpage_packet = (struct xferpage_packet *)listHead.next; + list_del(&xferpage_packet->list_ent); + + /* This is how much we can satisfy */ + xferpage_packet->count = count - 1; + + if (xferpage_packet->count != vmxferpage_packet->range_cnt) { + netdev_err(ndev, "Needed %d netvsc pkts to satisfy " + "this xfer page...got %d\n", + vmxferpage_packet->range_cnt, xferpage_packet->count); + } + + /* Each range represents 1 RNDIS pkt that contains 1 ethernet frame */ + for (i = 0; i < (count - 1); i++) { + netvsc_packet = (struct hv_netvsc_packet *)listHead.next; + list_del(&netvsc_packet->list_ent); + + /* Initialize the netvsc packet */ + netvsc_packet->xfer_page_pkt = xferpage_packet; + netvsc_packet->completion.recv.recv_completion = + netvsc_receive_completion; + netvsc_packet->completion.recv.recv_completion_ctx = + netvsc_packet; + netvsc_packet->device = device; + /* Save this so that we can send it back */ + netvsc_packet->completion.recv.recv_completion_tid = + vmxferpage_packet->d.trans_id; + + netvsc_packet->total_data_buflen = + vmxferpage_packet->ranges[i].byte_count; + netvsc_packet->page_buf_cnt = 1; + + netvsc_packet->page_buf[0].len = + vmxferpage_packet->ranges[i].byte_count; + + start = virt_to_phys((void *)((unsigned long)net_device-> + recv_buf + vmxferpage_packet->ranges[i].byte_offset)); + + netvsc_packet->page_buf[0].pfn = start >> PAGE_SHIFT; + end_virtual = (unsigned long)net_device->recv_buf + + vmxferpage_packet->ranges[i].byte_offset + + vmxferpage_packet->ranges[i].byte_count - 1; + end = virt_to_phys((void *)end_virtual); + + /* Calculate the page relative offset */ + netvsc_packet->page_buf[0].offset = + vmxferpage_packet->ranges[i].byte_offset & + (PAGE_SIZE - 1); + if ((end >> PAGE_SHIFT) != (start >> PAGE_SHIFT)) { + /* Handle frame across multiple pages: */ + netvsc_packet->page_buf[0].len = + (netvsc_packet->page_buf[0].pfn << + PAGE_SHIFT) + + PAGE_SIZE - start; + bytes_remain = netvsc_packet->total_data_buflen - + netvsc_packet->page_buf[0].len; + for (j = 1; j < NETVSC_PACKET_MAXPAGE; j++) { + netvsc_packet->page_buf[j].offset = 0; + if (bytes_remain <= PAGE_SIZE) { + netvsc_packet->page_buf[j].len = + bytes_remain; + bytes_remain = 0; + } else { + netvsc_packet->page_buf[j].len = + PAGE_SIZE; + bytes_remain -= PAGE_SIZE; + } + netvsc_packet->page_buf[j].pfn = + virt_to_phys((void *)(end_virtual - + bytes_remain)) >> PAGE_SHIFT; + netvsc_packet->page_buf_cnt++; + if (bytes_remain == 0) + break; + } + } + + /* Pass it to the upper layer */ + rndis_filter_receive(device, netvsc_packet); + + netvsc_receive_completion(netvsc_packet-> + completion.recv.recv_completion_ctx); + } + +} + +static void netvsc_channel_cb(void *context) +{ + int ret; + struct hv_device *device = context; + struct netvsc_device *net_device; + u32 bytes_recvd; + u64 request_id; + unsigned char *packet; + struct vmpacket_descriptor *desc; + unsigned char *buffer; + int bufferlen = NETVSC_PACKET_SIZE; + struct net_device *ndev; + + packet = kzalloc(NETVSC_PACKET_SIZE * sizeof(unsigned char), + GFP_ATOMIC); + if (!packet) + return; + buffer = packet; + + net_device = get_inbound_net_device(device); + if (!net_device) + goto out; + ndev = net_device->ndev; + + do { + ret = vmbus_recvpacket_raw(device->channel, buffer, bufferlen, + &bytes_recvd, &request_id); + if (ret == 0) { + if (bytes_recvd > 0) { + desc = (struct vmpacket_descriptor *)buffer; + switch (desc->type) { + case VM_PKT_COMP: + netvsc_send_completion(device, desc); + break; + + case VM_PKT_DATA_USING_XFER_PAGES: + netvsc_receive(device, desc); + break; + + default: + netdev_err(ndev, + "unhandled packet type %d, " + "tid %llx len %d\n", + desc->type, request_id, + bytes_recvd); + break; + } + + /* reset */ + if (bufferlen > NETVSC_PACKET_SIZE) { + kfree(buffer); + buffer = packet; + bufferlen = NETVSC_PACKET_SIZE; + } + } else { + /* reset */ + if (bufferlen > NETVSC_PACKET_SIZE) { + kfree(buffer); + buffer = packet; + bufferlen = NETVSC_PACKET_SIZE; + } + + break; + } + } else if (ret == -ENOBUFS) { + /* Handle large packet */ + buffer = kmalloc(bytes_recvd, GFP_ATOMIC); + if (buffer == NULL) { + /* Try again next time around */ + netdev_err(ndev, + "unable to allocate buffer of size " + "(%d)!!\n", bytes_recvd); + break; + } + + bufferlen = bytes_recvd; + } + } while (1); + +out: + kfree(buffer); + return; +} + +/* + * netvsc_device_add - Callback when the device belonging to this + * driver is added + */ +int netvsc_device_add(struct hv_device *device, void *additional_info) +{ + int ret = 0; + int i; + int ring_size = + ((struct netvsc_device_info *)additional_info)->ring_size; + struct netvsc_device *net_device; + struct hv_netvsc_packet *packet, *pos; + struct net_device *ndev; + + net_device = alloc_net_device(device); + if (!net_device) { + ret = -ENOMEM; + goto cleanup; + } + + /* + * Coming into this function, struct net_device * is + * registered as the driver private data. + * In alloc_net_device(), we register struct netvsc_device * + * as the driver private data and stash away struct net_device * + * in struct netvsc_device *. + */ + ndev = net_device->ndev; + + /* Initialize the NetVSC channel extension */ + net_device->recv_buf_size = NETVSC_RECEIVE_BUFFER_SIZE; + spin_lock_init(&net_device->recv_pkt_list_lock); + + INIT_LIST_HEAD(&net_device->recv_pkt_list); + + for (i = 0; i < NETVSC_RECEIVE_PACKETLIST_COUNT; i++) { + packet = kzalloc(sizeof(struct hv_netvsc_packet) + + (NETVSC_RECEIVE_SG_COUNT * + sizeof(struct hv_page_buffer)), GFP_KERNEL); + if (!packet) + break; + + list_add_tail(&packet->list_ent, + &net_device->recv_pkt_list); + } + init_completion(&net_device->channel_init_wait); + + /* Open the channel */ + ret = vmbus_open(device->channel, ring_size * PAGE_SIZE, + ring_size * PAGE_SIZE, NULL, 0, + netvsc_channel_cb, device); + + if (ret != 0) { + netdev_err(ndev, "unable to open channel: %d\n", ret); + goto cleanup; + } + + /* Channel is opened */ + pr_info("hv_netvsc channel opened successfully\n"); + + /* Connect with the NetVsp */ + ret = netvsc_connect_vsp(device); + if (ret != 0) { + netdev_err(ndev, + "unable to connect to NetVSP - %d\n", ret); + goto close; + } + + return ret; + +close: + /* Now, we can close the channel safely */ + vmbus_close(device->channel); + +cleanup: + + if (net_device) { + list_for_each_entry_safe(packet, pos, + &net_device->recv_pkt_list, + list_ent) { + list_del(&packet->list_ent); + kfree(packet); + } + + kfree(net_device); + } + + return ret; +} diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c new file mode 100644 index 0000000..93b0e91 --- /dev/null +++ b/drivers/net/hyperv/netvsc_drv.c @@ -0,0 +1,456 @@ +/* + * Copyright (c) 2009, Microsoft Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple + * Place - Suite 330, Boston, MA 02111-1307 USA. + * + * Authors: + * Haiyang Zhang + * Hank Janssen + */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "hyperv_net.h" + +struct net_device_context { + /* point back to our device context */ + struct hv_device *device_ctx; + atomic_t avail; + struct delayed_work dwork; +}; + + +#define PACKET_PAGES_LOWATER 8 +/* Need this many pages to handle worst case fragmented packet */ +#define PACKET_PAGES_HIWATER (MAX_SKB_FRAGS + 2) + +static int ring_size = 128; +module_param(ring_size, int, S_IRUGO); +MODULE_PARM_DESC(ring_size, "Ring buffer size (# of pages)"); + +/* no-op so the netdev core doesn't return -EINVAL when modifying the the + * multicast address list in SIOCADDMULTI. hv is setup to get all multicast + * when it calls RndisFilterOnOpen() */ +static void netvsc_set_multicast_list(struct net_device *net) +{ +} + +static int netvsc_open(struct net_device *net) +{ + struct net_device_context *net_device_ctx = netdev_priv(net); + struct hv_device *device_obj = net_device_ctx->device_ctx; + int ret = 0; + + /* Open up the device */ + ret = rndis_filter_open(device_obj); + if (ret != 0) { + netdev_err(net, "unable to open device (ret %d).\n", ret); + return ret; + } + + netif_start_queue(net); + + return ret; +} + +static int netvsc_close(struct net_device *net) +{ + struct net_device_context *net_device_ctx = netdev_priv(net); + struct hv_device *device_obj = net_device_ctx->device_ctx; + int ret; + + netif_stop_queue(net); + + ret = rndis_filter_close(device_obj); + if (ret != 0) + netdev_err(net, "unable to close device (ret %d).\n", ret); + + return ret; +} + +static void netvsc_xmit_completion(void *context) +{ + struct hv_netvsc_packet *packet = (struct hv_netvsc_packet *)context; + struct sk_buff *skb = (struct sk_buff *) + (unsigned long)packet->completion.send.send_completion_tid; + + kfree(packet); + + if (skb) { + struct net_device *net = skb->dev; + struct net_device_context *net_device_ctx = netdev_priv(net); + unsigned int num_pages = skb_shinfo(skb)->nr_frags + 2; + + dev_kfree_skb_any(skb); + + atomic_add(num_pages, &net_device_ctx->avail); + if (atomic_read(&net_device_ctx->avail) >= + PACKET_PAGES_HIWATER) + netif_wake_queue(net); + } +} + +static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net) +{ + struct net_device_context *net_device_ctx = netdev_priv(net); + struct hv_netvsc_packet *packet; + int ret; + unsigned int i, num_pages; + + /* Add 1 for skb->data and additional one for RNDIS */ + num_pages = skb_shinfo(skb)->nr_frags + 1 + 1; + if (num_pages > atomic_read(&net_device_ctx->avail)) + return NETDEV_TX_BUSY; + + /* Allocate a netvsc packet based on # of frags. */ + packet = kzalloc(sizeof(struct hv_netvsc_packet) + + (num_pages * sizeof(struct hv_page_buffer)) + + sizeof(struct rndis_filter_packet), GFP_ATOMIC); + if (!packet) { + /* out of memory, drop packet */ + netdev_err(net, "unable to allocate hv_netvsc_packet\n"); + + dev_kfree_skb(skb); + net->stats.tx_dropped++; + return NETDEV_TX_BUSY; + } + + packet->extension = (void *)(unsigned long)packet + + sizeof(struct hv_netvsc_packet) + + (num_pages * sizeof(struct hv_page_buffer)); + + /* Setup the rndis header */ + packet->page_buf_cnt = num_pages; + + /* Initialize it from the skb */ + packet->total_data_buflen = skb->len; + + /* Start filling in the page buffers starting after RNDIS buffer. */ + packet->page_buf[1].pfn = virt_to_phys(skb->data) >> PAGE_SHIFT; + packet->page_buf[1].offset + = (unsigned long)skb->data & (PAGE_SIZE - 1); + packet->page_buf[1].len = skb_headlen(skb); + + /* Additional fragments are after SKB data */ + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + const skb_frag_t *f = &skb_shinfo(skb)->frags[i]; + + packet->page_buf[i+2].pfn = page_to_pfn(skb_frag_page(f)); + packet->page_buf[i+2].offset = f->page_offset; + packet->page_buf[i+2].len = skb_frag_size(f); + } + + /* Set the completion routine */ + packet->completion.send.send_completion = netvsc_xmit_completion; + packet->completion.send.send_completion_ctx = packet; + packet->completion.send.send_completion_tid = (unsigned long)skb; + + ret = rndis_filter_send(net_device_ctx->device_ctx, + packet); + if (ret == 0) { + net->stats.tx_bytes += skb->len; + net->stats.tx_packets++; + + atomic_sub(num_pages, &net_device_ctx->avail); + if (atomic_read(&net_device_ctx->avail) < PACKET_PAGES_LOWATER) + netif_stop_queue(net); + } else { + /* we are shutting down or bus overloaded, just drop packet */ + net->stats.tx_dropped++; + kfree(packet); + dev_kfree_skb_any(skb); + } + + return ret ? NETDEV_TX_BUSY : NETDEV_TX_OK; +} + +/* + * netvsc_linkstatus_callback - Link up/down notification + */ +void netvsc_linkstatus_callback(struct hv_device *device_obj, + unsigned int status) +{ + struct net_device *net; + struct net_device_context *ndev_ctx; + struct netvsc_device *net_device; + + net_device = hv_get_drvdata(device_obj); + net = net_device->ndev; + + if (!net) { + netdev_err(net, "got link status but net device " + "not initialized yet\n"); + return; + } + + if (status == 1) { + netif_carrier_on(net); + netif_wake_queue(net); + ndev_ctx = netdev_priv(net); + schedule_delayed_work(&ndev_ctx->dwork, 0); + schedule_delayed_work(&ndev_ctx->dwork, msecs_to_jiffies(20)); + } else { + netif_carrier_off(net); + netif_stop_queue(net); + } +} + +/* + * netvsc_recv_callback - Callback when we receive a packet from the + * "wire" on the specified device. + */ +int netvsc_recv_callback(struct hv_device *device_obj, + struct hv_netvsc_packet *packet) +{ + struct net_device *net = dev_get_drvdata(&device_obj->device); + struct sk_buff *skb; + void *data; + int i; + unsigned long flags; + struct netvsc_device *net_device; + + net_device = hv_get_drvdata(device_obj); + net = net_device->ndev; + + if (!net) { + netdev_err(net, "got receive callback but net device" + " not initialized yet\n"); + return 0; + } + + /* Allocate a skb - TODO direct I/O to pages? */ + skb = netdev_alloc_skb_ip_align(net, packet->total_data_buflen); + if (unlikely(!skb)) { + ++net->stats.rx_dropped; + return 0; + } + + /* for kmap_atomic */ + local_irq_save(flags); + + /* + * Copy to skb. This copy is needed here since the memory pointed by + * hv_netvsc_packet cannot be deallocated + */ + for (i = 0; i < packet->page_buf_cnt; i++) { + data = kmap_atomic(pfn_to_page(packet->page_buf[i].pfn), + KM_IRQ1); + data = (void *)(unsigned long)data + + packet->page_buf[i].offset; + + memcpy(skb_put(skb, packet->page_buf[i].len), data, + packet->page_buf[i].len); + + kunmap_atomic((void *)((unsigned long)data - + packet->page_buf[i].offset), KM_IRQ1); + } + + local_irq_restore(flags); + + skb->protocol = eth_type_trans(skb, net); + skb->ip_summed = CHECKSUM_NONE; + + net->stats.rx_packets++; + net->stats.rx_bytes += skb->len; + + /* + * Pass the skb back up. Network stack will deallocate the skb when it + * is done. + * TODO - use NAPI? + */ + netif_rx(skb); + + return 0; +} + +static void netvsc_get_drvinfo(struct net_device *net, + struct ethtool_drvinfo *info) +{ + strcpy(info->driver, "hv_netvsc"); + strcpy(info->version, HV_DRV_VERSION); + strcpy(info->fw_version, "N/A"); +} + +static const struct ethtool_ops ethtool_ops = { + .get_drvinfo = netvsc_get_drvinfo, + .get_link = ethtool_op_get_link, +}; + +static const struct net_device_ops device_ops = { + .ndo_open = netvsc_open, + .ndo_stop = netvsc_close, + .ndo_start_xmit = netvsc_start_xmit, + .ndo_set_rx_mode = netvsc_set_multicast_list, + .ndo_change_mtu = eth_change_mtu, + .ndo_validate_addr = eth_validate_addr, + .ndo_set_mac_address = eth_mac_addr, +}; + +/* + * Send GARP packet to network peers after migrations. + * After Quick Migration, the network is not immediately operational in the + * current context when receiving RNDIS_STATUS_MEDIA_CONNECT event. So, add + * another netif_notify_peers() into a delayed work, otherwise GARP packet + * will not be sent after quick migration, and cause network disconnection. + */ +static void netvsc_send_garp(struct work_struct *w) +{ + struct net_device_context *ndev_ctx; + struct net_device *net; + struct netvsc_device *net_device; + + ndev_ctx = container_of(w, struct net_device_context, dwork.work); + net_device = hv_get_drvdata(ndev_ctx->device_ctx); + net = net_device->ndev; + netif_notify_peers(net); +} + + +static int netvsc_probe(struct hv_device *dev, + const struct hv_vmbus_device_id *dev_id) +{ + struct net_device *net = NULL; + struct net_device_context *net_device_ctx; + struct netvsc_device_info device_info; + int ret; + + net = alloc_etherdev(sizeof(struct net_device_context)); + if (!net) + return -ENOMEM; + + /* Set initial state */ + netif_carrier_off(net); + + net_device_ctx = netdev_priv(net); + net_device_ctx->device_ctx = dev; + atomic_set(&net_device_ctx->avail, ring_size); + hv_set_drvdata(dev, net); + INIT_DELAYED_WORK(&net_device_ctx->dwork, netvsc_send_garp); + + net->netdev_ops = &device_ops; + + /* TODO: Add GSO and Checksum offload */ + net->hw_features = NETIF_F_SG; + net->features = NETIF_F_SG; + + SET_ETHTOOL_OPS(net, ðtool_ops); + SET_NETDEV_DEV(net, &dev->device); + + ret = register_netdev(net); + if (ret != 0) { + pr_err("Unable to register netdev.\n"); + free_netdev(net); + goto out; + } + + /* Notify the netvsc driver of the new device */ + device_info.ring_size = ring_size; + ret = rndis_filter_device_add(dev, &device_info); + if (ret != 0) { + netdev_err(net, "unable to add netvsc device (ret %d)\n", ret); + unregister_netdev(net); + free_netdev(net); + hv_set_drvdata(dev, NULL); + return ret; + } + memcpy(net->dev_addr, device_info.mac_adr, ETH_ALEN); + + netif_carrier_on(net); + +out: + return ret; +} + +static int netvsc_remove(struct hv_device *dev) +{ + struct net_device *net; + struct net_device_context *ndev_ctx; + struct netvsc_device *net_device; + + net_device = hv_get_drvdata(dev); + net = net_device->ndev; + + if (net == NULL) { + dev_err(&dev->device, "No net device to remove\n"); + return 0; + } + + ndev_ctx = netdev_priv(net); + cancel_delayed_work_sync(&ndev_ctx->dwork); + + /* Stop outbound asap */ + netif_stop_queue(net); + + unregister_netdev(net); + + /* + * Call to the vsc driver to let it know that the device is being + * removed + */ + rndis_filter_device_remove(dev); + + free_netdev(net); + return 0; +} + +static const struct hv_vmbus_device_id id_table[] = { + /* Network guid */ + { VMBUS_DEVICE(0x63, 0x51, 0x61, 0xF8, 0x3E, 0xDF, 0xc5, 0x46, + 0x91, 0x3F, 0xF2, 0xD2, 0xF9, 0x65, 0xED, 0x0E) }, + { }, +}; + +MODULE_DEVICE_TABLE(vmbus, id_table); + +/* The one and only one */ +static struct hv_driver netvsc_drv = { + .name = "netvsc", + .id_table = id_table, + .probe = netvsc_probe, + .remove = netvsc_remove, +}; + +static void __exit netvsc_drv_exit(void) +{ + vmbus_driver_unregister(&netvsc_drv); +} + +static int __init netvsc_drv_init(void) +{ + return vmbus_driver_register(&netvsc_drv); +} + +MODULE_LICENSE("GPL"); +MODULE_VERSION(HV_DRV_VERSION); +MODULE_DESCRIPTION("Microsoft Hyper-V network driver"); + +module_init(netvsc_drv_init); +module_exit(netvsc_drv_exit); diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c new file mode 100644 index 0000000..bafccb3 --- /dev/null +++ b/drivers/net/hyperv/rndis_filter.c @@ -0,0 +1,855 @@ +/* + * Copyright (c) 2009, Microsoft Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple + * Place - Suite 330, Boston, MA 02111-1307 USA. + * + * Authors: + * Haiyang Zhang + * Hank Janssen + */ +#include +#include +#include +#include +#include +#include +#include +#include + +#include "hyperv_net.h" + + +enum rndis_device_state { + RNDIS_DEV_UNINITIALIZED = 0, + RNDIS_DEV_INITIALIZING, + RNDIS_DEV_INITIALIZED, + RNDIS_DEV_DATAINITIALIZED, +}; + +struct rndis_device { + struct netvsc_device *net_dev; + + enum rndis_device_state state; + bool link_state; + atomic_t new_req_id; + + spinlock_t request_lock; + struct list_head req_list; + + unsigned char hw_mac_adr[ETH_ALEN]; +}; + +struct rndis_request { + struct list_head list_ent; + struct completion wait_event; + + /* + * FIXME: We assumed a fixed size response here. If we do ever need to + * handle a bigger response, we can either define a max response + * message or add a response buffer variable above this field + */ + struct rndis_message response_msg; + + /* Simplify allocation by having a netvsc packet inline */ + struct hv_netvsc_packet pkt; + struct hv_page_buffer buf; + /* FIXME: We assumed a fixed size request here. */ + struct rndis_message request_msg; +}; + +static void rndis_filter_send_completion(void *ctx); + +static void rndis_filter_send_request_completion(void *ctx); + + + +static struct rndis_device *get_rndis_device(void) +{ + struct rndis_device *device; + + device = kzalloc(sizeof(struct rndis_device), GFP_KERNEL); + if (!device) + return NULL; + + spin_lock_init(&device->request_lock); + + INIT_LIST_HEAD(&device->req_list); + + device->state = RNDIS_DEV_UNINITIALIZED; + + return device; +} + +static struct rndis_request *get_rndis_request(struct rndis_device *dev, + u32 msg_type, + u32 msg_len) +{ + struct rndis_request *request; + struct rndis_message *rndis_msg; + struct rndis_set_request *set; + unsigned long flags; + + request = kzalloc(sizeof(struct rndis_request), GFP_KERNEL); + if (!request) + return NULL; + + init_completion(&request->wait_event); + + rndis_msg = &request->request_msg; + rndis_msg->ndis_msg_type = msg_type; + rndis_msg->msg_len = msg_len; + + /* + * Set the request id. This field is always after the rndis header for + * request/response packet types so we just used the SetRequest as a + * template + */ + set = &rndis_msg->msg.set_req; + set->req_id = atomic_inc_return(&dev->new_req_id); + + /* Add to the request list */ + spin_lock_irqsave(&dev->request_lock, flags); + list_add_tail(&request->list_ent, &dev->req_list); + spin_unlock_irqrestore(&dev->request_lock, flags); + + return request; +} + +static void put_rndis_request(struct rndis_device *dev, + struct rndis_request *req) +{ + unsigned long flags; + + spin_lock_irqsave(&dev->request_lock, flags); + list_del(&req->list_ent); + spin_unlock_irqrestore(&dev->request_lock, flags); + + kfree(req); +} + +static void dump_rndis_message(struct hv_device *hv_dev, + struct rndis_message *rndis_msg) +{ + struct net_device *netdev; + struct netvsc_device *net_device; + + net_device = hv_get_drvdata(hv_dev); + netdev = net_device->ndev; + + switch (rndis_msg->ndis_msg_type) { + case REMOTE_NDIS_PACKET_MSG: + netdev_dbg(netdev, "REMOTE_NDIS_PACKET_MSG (len %u, " + "data offset %u data len %u, # oob %u, " + "oob offset %u, oob len %u, pkt offset %u, " + "pkt len %u\n", + rndis_msg->msg_len, + rndis_msg->msg.pkt.data_offset, + rndis_msg->msg.pkt.data_len, + rndis_msg->msg.pkt.num_oob_data_elements, + rndis_msg->msg.pkt.oob_data_offset, + rndis_msg->msg.pkt.oob_data_len, + rndis_msg->msg.pkt.per_pkt_info_offset, + rndis_msg->msg.pkt.per_pkt_info_len); + break; + + case REMOTE_NDIS_INITIALIZE_CMPLT: + netdev_dbg(netdev, "REMOTE_NDIS_INITIALIZE_CMPLT " + "(len %u, id 0x%x, status 0x%x, major %d, minor %d, " + "device flags %d, max xfer size 0x%x, max pkts %u, " + "pkt aligned %u)\n", + rndis_msg->msg_len, + rndis_msg->msg.init_complete.req_id, + rndis_msg->msg.init_complete.status, + rndis_msg->msg.init_complete.major_ver, + rndis_msg->msg.init_complete.minor_ver, + rndis_msg->msg.init_complete.dev_flags, + rndis_msg->msg.init_complete.max_xfer_size, + rndis_msg->msg.init_complete. + max_pkt_per_msg, + rndis_msg->msg.init_complete. + pkt_alignment_factor); + break; + + case REMOTE_NDIS_QUERY_CMPLT: + netdev_dbg(netdev, "REMOTE_NDIS_QUERY_CMPLT " + "(len %u, id 0x%x, status 0x%x, buf len %u, " + "buf offset %u)\n", + rndis_msg->msg_len, + rndis_msg->msg.query_complete.req_id, + rndis_msg->msg.query_complete.status, + rndis_msg->msg.query_complete. + info_buflen, + rndis_msg->msg.query_complete. + info_buf_offset); + break; + + case REMOTE_NDIS_SET_CMPLT: + netdev_dbg(netdev, + "REMOTE_NDIS_SET_CMPLT (len %u, id 0x%x, status 0x%x)\n", + rndis_msg->msg_len, + rndis_msg->msg.set_complete.req_id, + rndis_msg->msg.set_complete.status); + break; + + case REMOTE_NDIS_INDICATE_STATUS_MSG: + netdev_dbg(netdev, "REMOTE_NDIS_INDICATE_STATUS_MSG " + "(len %u, status 0x%x, buf len %u, buf offset %u)\n", + rndis_msg->msg_len, + rndis_msg->msg.indicate_status.status, + rndis_msg->msg.indicate_status.status_buflen, + rndis_msg->msg.indicate_status.status_buf_offset); + break; + + default: + netdev_dbg(netdev, "0x%x (len %u)\n", + rndis_msg->ndis_msg_type, + rndis_msg->msg_len); + break; + } +} + +static int rndis_filter_send_request(struct rndis_device *dev, + struct rndis_request *req) +{ + int ret; + struct hv_netvsc_packet *packet; + + /* Setup the packet to send it */ + packet = &req->pkt; + + packet->is_data_pkt = false; + packet->total_data_buflen = req->request_msg.msg_len; + packet->page_buf_cnt = 1; + + packet->page_buf[0].pfn = virt_to_phys(&req->request_msg) >> + PAGE_SHIFT; + packet->page_buf[0].len = req->request_msg.msg_len; + packet->page_buf[0].offset = + (unsigned long)&req->request_msg & (PAGE_SIZE - 1); + + packet->completion.send.send_completion_ctx = req;/* packet; */ + packet->completion.send.send_completion = + rndis_filter_send_request_completion; + packet->completion.send.send_completion_tid = (unsigned long)dev; + + ret = netvsc_send(dev->net_dev->dev, packet); + return ret; +} + +static void rndis_filter_receive_response(struct rndis_device *dev, + struct rndis_message *resp) +{ + struct rndis_request *request = NULL; + bool found = false; + unsigned long flags; + struct net_device *ndev; + + ndev = dev->net_dev->ndev; + + spin_lock_irqsave(&dev->request_lock, flags); + list_for_each_entry(request, &dev->req_list, list_ent) { + /* + * All request/response message contains RequestId as the 1st + * field + */ + if (request->request_msg.msg.init_req.req_id + == resp->msg.init_complete.req_id) { + found = true; + break; + } + } + spin_unlock_irqrestore(&dev->request_lock, flags); + + if (found) { + if (resp->msg_len <= sizeof(struct rndis_message)) { + memcpy(&request->response_msg, resp, + resp->msg_len); + } else { + netdev_err(ndev, + "rndis response buffer overflow " + "detected (size %u max %zu)\n", + resp->msg_len, + sizeof(struct rndis_filter_packet)); + + if (resp->ndis_msg_type == + REMOTE_NDIS_RESET_CMPLT) { + /* does not have a request id field */ + request->response_msg.msg.reset_complete. + status = STATUS_BUFFER_OVERFLOW; + } else { + request->response_msg.msg. + init_complete.status = + STATUS_BUFFER_OVERFLOW; + } + } + + complete(&request->wait_event); + } else { + netdev_err(ndev, + "no rndis request found for this response " + "(id 0x%x res type 0x%x)\n", + resp->msg.init_complete.req_id, + resp->ndis_msg_type); + } +} + +static void rndis_filter_receive_indicate_status(struct rndis_device *dev, + struct rndis_message *resp) +{ + struct rndis_indicate_status *indicate = + &resp->msg.indicate_status; + + if (indicate->status == RNDIS_STATUS_MEDIA_CONNECT) { + netvsc_linkstatus_callback( + dev->net_dev->dev, 1); + } else if (indicate->status == RNDIS_STATUS_MEDIA_DISCONNECT) { + netvsc_linkstatus_callback( + dev->net_dev->dev, 0); + } else { + /* + * TODO: + */ + } +} + +static void rndis_filter_receive_data(struct rndis_device *dev, + struct rndis_message *msg, + struct hv_netvsc_packet *pkt) +{ + struct rndis_packet *rndis_pkt; + u32 data_offset; + int i; + + rndis_pkt = &msg->msg.pkt; + + /* + * FIXME: Handle multiple rndis pkt msgs that maybe enclosed in this + * netvsc packet (ie TotalDataBufferLength != MessageLength) + */ + + /* Remove the rndis header and pass it back up the stack */ + data_offset = RNDIS_HEADER_SIZE + rndis_pkt->data_offset; + + pkt->total_data_buflen -= data_offset; + pkt->page_buf[0].offset += data_offset; + pkt->page_buf[0].len -= data_offset; + + /* Drop the 0th page, if rndis data go beyond page boundary */ + if (pkt->page_buf[0].offset >= PAGE_SIZE) { + pkt->page_buf[1].offset = pkt->page_buf[0].offset - PAGE_SIZE; + pkt->page_buf[1].len -= pkt->page_buf[1].offset; + pkt->page_buf_cnt--; + for (i = 0; i < pkt->page_buf_cnt; i++) + pkt->page_buf[i] = pkt->page_buf[i+1]; + } + + pkt->is_data_pkt = true; + + netvsc_recv_callback(dev->net_dev->dev, pkt); +} + +int rndis_filter_receive(struct hv_device *dev, + struct hv_netvsc_packet *pkt) +{ + struct netvsc_device *net_dev = hv_get_drvdata(dev); + struct rndis_device *rndis_dev; + struct rndis_message rndis_msg; + struct rndis_message *rndis_hdr; + struct net_device *ndev; + + if (!net_dev) + return -EINVAL; + + ndev = net_dev->ndev; + + /* Make sure the rndis device state is initialized */ + if (!net_dev->extension) { + netdev_err(ndev, "got rndis message but no rndis device - " + "dropping this message!\n"); + return -ENODEV; + } + + rndis_dev = (struct rndis_device *)net_dev->extension; + if (rndis_dev->state == RNDIS_DEV_UNINITIALIZED) { + netdev_err(ndev, "got rndis message but rndis device " + "uninitialized...dropping this message!\n"); + return -ENODEV; + } + + rndis_hdr = (struct rndis_message *)kmap_atomic( + pfn_to_page(pkt->page_buf[0].pfn), KM_IRQ0); + + rndis_hdr = (void *)((unsigned long)rndis_hdr + + pkt->page_buf[0].offset); + + /* Make sure we got a valid rndis message */ + if ((rndis_hdr->ndis_msg_type != REMOTE_NDIS_PACKET_MSG) && + (rndis_hdr->msg_len > sizeof(struct rndis_message))) { + netdev_err(ndev, "incoming rndis message buffer overflow " + "detected (got %u, max %zu)..marking it an error!\n", + rndis_hdr->msg_len, + sizeof(struct rndis_message)); + } + + memcpy(&rndis_msg, rndis_hdr, + (rndis_hdr->msg_len > sizeof(struct rndis_message)) ? + sizeof(struct rndis_message) : + rndis_hdr->msg_len); + + kunmap_atomic(rndis_hdr - pkt->page_buf[0].offset, KM_IRQ0); + + dump_rndis_message(dev, &rndis_msg); + + switch (rndis_msg.ndis_msg_type) { + case REMOTE_NDIS_PACKET_MSG: + /* data msg */ + rndis_filter_receive_data(rndis_dev, &rndis_msg, pkt); + break; + + case REMOTE_NDIS_INITIALIZE_CMPLT: + case REMOTE_NDIS_QUERY_CMPLT: + case REMOTE_NDIS_SET_CMPLT: + /* completion msgs */ + rndis_filter_receive_response(rndis_dev, &rndis_msg); + break; + + case REMOTE_NDIS_INDICATE_STATUS_MSG: + /* notification msgs */ + rndis_filter_receive_indicate_status(rndis_dev, &rndis_msg); + break; + default: + netdev_err(ndev, + "unhandled rndis message (type %u len %u)\n", + rndis_msg.ndis_msg_type, + rndis_msg.msg_len); + break; + } + + return 0; +} + +static int rndis_filter_query_device(struct rndis_device *dev, u32 oid, + void *result, u32 *result_size) +{ + struct rndis_request *request; + u32 inresult_size = *result_size; + struct rndis_query_request *query; + struct rndis_query_complete *query_complete; + int ret = 0; + int t; + + if (!result) + return -EINVAL; + + *result_size = 0; + request = get_rndis_request(dev, REMOTE_NDIS_QUERY_MSG, + RNDIS_MESSAGE_SIZE(struct rndis_query_request)); + if (!request) { + ret = -ENOMEM; + goto cleanup; + } + + /* Setup the rndis query */ + query = &request->request_msg.msg.query_req; + query->oid = oid; + query->info_buf_offset = sizeof(struct rndis_query_request); + query->info_buflen = 0; + query->dev_vc_handle = 0; + + ret = rndis_filter_send_request(dev, request); + if (ret != 0) + goto cleanup; + + t = wait_for_completion_timeout(&request->wait_event, 5*HZ); + if (t == 0) { + ret = -ETIMEDOUT; + goto cleanup; + } + + /* Copy the response back */ + query_complete = &request->response_msg.msg.query_complete; + + if (query_complete->info_buflen > inresult_size) { + ret = -1; + goto cleanup; + } + + memcpy(result, + (void *)((unsigned long)query_complete + + query_complete->info_buf_offset), + query_complete->info_buflen); + + *result_size = query_complete->info_buflen; + +cleanup: + if (request) + put_rndis_request(dev, request); + + return ret; +} + +static int rndis_filter_query_device_mac(struct rndis_device *dev) +{ + u32 size = ETH_ALEN; + + return rndis_filter_query_device(dev, + RNDIS_OID_802_3_PERMANENT_ADDRESS, + dev->hw_mac_adr, &size); +} + +static int rndis_filter_query_device_link_status(struct rndis_device *dev) +{ + u32 size = sizeof(u32); + u32 link_status; + int ret; + + ret = rndis_filter_query_device(dev, + RNDIS_OID_GEN_MEDIA_CONNECT_STATUS, + &link_status, &size); + dev->link_state = (link_status != 0) ? true : false; + + return ret; +} + +static int rndis_filter_set_packet_filter(struct rndis_device *dev, + u32 new_filter) +{ + struct rndis_request *request; + struct rndis_set_request *set; + struct rndis_set_complete *set_complete; + u32 status; + int ret, t; + struct net_device *ndev; + + ndev = dev->net_dev->ndev; + + request = get_rndis_request(dev, REMOTE_NDIS_SET_MSG, + RNDIS_MESSAGE_SIZE(struct rndis_set_request) + + sizeof(u32)); + if (!request) { + ret = -ENOMEM; + goto cleanup; + } + + /* Setup the rndis set */ + set = &request->request_msg.msg.set_req; + set->oid = RNDIS_OID_GEN_CURRENT_PACKET_FILTER; + set->info_buflen = sizeof(u32); + set->info_buf_offset = sizeof(struct rndis_set_request); + + memcpy((void *)(unsigned long)set + sizeof(struct rndis_set_request), + &new_filter, sizeof(u32)); + + ret = rndis_filter_send_request(dev, request); + if (ret != 0) + goto cleanup; + + t = wait_for_completion_timeout(&request->wait_event, 5*HZ); + + if (t == 0) { + netdev_err(ndev, + "timeout before we got a set response...\n"); + /* + * We can't deallocate the request since we may still receive a + * send completion for it. + */ + goto exit; + } else { + set_complete = &request->response_msg.msg.set_complete; + status = set_complete->status; + } + +cleanup: + if (request) + put_rndis_request(dev, request); +exit: + return ret; +} + + +static int rndis_filter_init_device(struct rndis_device *dev) +{ + struct rndis_request *request; + struct rndis_initialize_request *init; + struct rndis_initialize_complete *init_complete; + u32 status; + int ret, t; + + request = get_rndis_request(dev, REMOTE_NDIS_INITIALIZE_MSG, + RNDIS_MESSAGE_SIZE(struct rndis_initialize_request)); + if (!request) { + ret = -ENOMEM; + goto cleanup; + } + + /* Setup the rndis set */ + init = &request->request_msg.msg.init_req; + init->major_ver = RNDIS_MAJOR_VERSION; + init->minor_ver = RNDIS_MINOR_VERSION; + /* FIXME: Use 1536 - rounded ethernet frame size */ + init->max_xfer_size = 2048; + + dev->state = RNDIS_DEV_INITIALIZING; + + ret = rndis_filter_send_request(dev, request); + if (ret != 0) { + dev->state = RNDIS_DEV_UNINITIALIZED; + goto cleanup; + } + + + t = wait_for_completion_timeout(&request->wait_event, 5*HZ); + + if (t == 0) { + ret = -ETIMEDOUT; + goto cleanup; + } + + init_complete = &request->response_msg.msg.init_complete; + status = init_complete->status; + if (status == RNDIS_STATUS_SUCCESS) { + dev->state = RNDIS_DEV_INITIALIZED; + ret = 0; + } else { + dev->state = RNDIS_DEV_UNINITIALIZED; + ret = -EINVAL; + } + +cleanup: + if (request) + put_rndis_request(dev, request); + + return ret; +} + +static void rndis_filter_halt_device(struct rndis_device *dev) +{ + struct rndis_request *request; + struct rndis_halt_request *halt; + + /* Attempt to do a rndis device halt */ + request = get_rndis_request(dev, REMOTE_NDIS_HALT_MSG, + RNDIS_MESSAGE_SIZE(struct rndis_halt_request)); + if (!request) + goto cleanup; + + /* Setup the rndis set */ + halt = &request->request_msg.msg.halt_req; + halt->req_id = atomic_inc_return(&dev->new_req_id); + + /* Ignore return since this msg is optional. */ + rndis_filter_send_request(dev, request); + + dev->state = RNDIS_DEV_UNINITIALIZED; + +cleanup: + if (request) + put_rndis_request(dev, request); + return; +} + +static int rndis_filter_open_device(struct rndis_device *dev) +{ + int ret; + + if (dev->state != RNDIS_DEV_INITIALIZED) + return 0; + + ret = rndis_filter_set_packet_filter(dev, + NDIS_PACKET_TYPE_BROADCAST | + NDIS_PACKET_TYPE_ALL_MULTICAST | + NDIS_PACKET_TYPE_DIRECTED); + if (ret == 0) + dev->state = RNDIS_DEV_DATAINITIALIZED; + + return ret; +} + +static int rndis_filter_close_device(struct rndis_device *dev) +{ + int ret; + + if (dev->state != RNDIS_DEV_DATAINITIALIZED) + return 0; + + ret = rndis_filter_set_packet_filter(dev, 0); + if (ret == 0) + dev->state = RNDIS_DEV_INITIALIZED; + + return ret; +} + +int rndis_filter_device_add(struct hv_device *dev, + void *additional_info) +{ + int ret; + struct netvsc_device *net_device; + struct rndis_device *rndis_device; + struct netvsc_device_info *device_info = additional_info; + + rndis_device = get_rndis_device(); + if (!rndis_device) + return -ENODEV; + + /* + * Let the inner driver handle this first to create the netvsc channel + * NOTE! Once the channel is created, we may get a receive callback + * (RndisFilterOnReceive()) before this call is completed + */ + ret = netvsc_device_add(dev, additional_info); + if (ret != 0) { + kfree(rndis_device); + return ret; + } + + + /* Initialize the rndis device */ + net_device = hv_get_drvdata(dev); + + net_device->extension = rndis_device; + rndis_device->net_dev = net_device; + + /* Send the rndis initialization message */ + ret = rndis_filter_init_device(rndis_device); + if (ret != 0) { + /* + * TODO: If rndis init failed, we will need to shut down the + * channel + */ + } + + /* Get the mac address */ + ret = rndis_filter_query_device_mac(rndis_device); + if (ret != 0) { + /* + * TODO: shutdown rndis device and the channel + */ + } + + memcpy(device_info->mac_adr, rndis_device->hw_mac_adr, ETH_ALEN); + + rndis_filter_query_device_link_status(rndis_device); + + device_info->link_state = rndis_device->link_state; + + dev_info(&dev->device, "Device MAC %pM link state %s\n", + rndis_device->hw_mac_adr, + device_info->link_state ? "down" : "up"); + + return ret; +} + +void rndis_filter_device_remove(struct hv_device *dev) +{ + struct netvsc_device *net_dev = hv_get_drvdata(dev); + struct rndis_device *rndis_dev = net_dev->extension; + + /* Halt and release the rndis device */ + rndis_filter_halt_device(rndis_dev); + + kfree(rndis_dev); + net_dev->extension = NULL; + + netvsc_device_remove(dev); +} + + +int rndis_filter_open(struct hv_device *dev) +{ + struct netvsc_device *net_device = hv_get_drvdata(dev); + + if (!net_device) + return -EINVAL; + + return rndis_filter_open_device(net_device->extension); +} + +int rndis_filter_close(struct hv_device *dev) +{ + struct netvsc_device *netDevice = hv_get_drvdata(dev); + + if (!netDevice) + return -EINVAL; + + return rndis_filter_close_device(netDevice->extension); +} + +int rndis_filter_send(struct hv_device *dev, + struct hv_netvsc_packet *pkt) +{ + int ret; + struct rndis_filter_packet *filterPacket; + struct rndis_message *rndisMessage; + struct rndis_packet *rndisPacket; + u32 rndisMessageSize; + + /* Add the rndis header */ + filterPacket = (struct rndis_filter_packet *)pkt->extension; + + memset(filterPacket, 0, sizeof(struct rndis_filter_packet)); + + rndisMessage = &filterPacket->msg; + rndisMessageSize = RNDIS_MESSAGE_SIZE(struct rndis_packet); + + rndisMessage->ndis_msg_type = REMOTE_NDIS_PACKET_MSG; + rndisMessage->msg_len = pkt->total_data_buflen + + rndisMessageSize; + + rndisPacket = &rndisMessage->msg.pkt; + rndisPacket->data_offset = sizeof(struct rndis_packet); + rndisPacket->data_len = pkt->total_data_buflen; + + pkt->is_data_pkt = true; + pkt->page_buf[0].pfn = virt_to_phys(rndisMessage) >> PAGE_SHIFT; + pkt->page_buf[0].offset = + (unsigned long)rndisMessage & (PAGE_SIZE-1); + pkt->page_buf[0].len = rndisMessageSize; + + /* Save the packet send completion and context */ + filterPacket->completion = pkt->completion.send.send_completion; + filterPacket->completion_ctx = + pkt->completion.send.send_completion_ctx; + + /* Use ours */ + pkt->completion.send.send_completion = rndis_filter_send_completion; + pkt->completion.send.send_completion_ctx = filterPacket; + + ret = netvsc_send(dev, pkt); + if (ret != 0) { + /* + * Reset the completion to originals to allow retries from + * above + */ + pkt->completion.send.send_completion = + filterPacket->completion; + pkt->completion.send.send_completion_ctx = + filterPacket->completion_ctx; + } + + return ret; +} + +static void rndis_filter_send_completion(void *ctx) +{ + struct rndis_filter_packet *filterPacket = ctx; + + /* Pass it back to the original handler */ + filterPacket->completion(filterPacket->completion_ctx); +} + + +static void rndis_filter_send_request_completion(void *ctx) +{ + /* Noop */ +} -- cgit v1.1