Release 4.12 include/linux/hyperv.h
/*
*
* 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 <haiyangz@microsoft.com>
* Hank Janssen <hjanssen@microsoft.com>
* K. Y. Srinivasan <kys@microsoft.com>
*
*/
#ifndef _HYPERV_H
#define _HYPERV_H
#include <uapi/linux/hyperv.h>
#include <uapi/asm/hyperv.h>
#include <linux/types.h>
#include <linux/scatterlist.h>
#include <linux/list.h>
#include <linux/timer.h>
#include <linux/completion.h>
#include <linux/device.h>
#include <linux/mod_devicetable.h>
#include <linux/interrupt.h>
#define MAX_PAGE_BUFFER_COUNT 32
#define MAX_MULTIPAGE_BUFFER_COUNT 32
/* 128K */
#pragma pack(push, 1)
/* Single-page buffer */
struct hv_page_buffer {
u32 len;
u32 offset;
u64 pfn;
};
/* Multiple-page buffer */
struct hv_multipage_buffer {
/* Length and Offset determines the # of pfns in the array */
u32 len;
u32 offset;
u64 pfn_array[MAX_MULTIPAGE_BUFFER_COUNT];
};
/*
* Multiple-page buffer array; the pfn array is variable size:
* The number of entries in the PFN array is determined by
* "len" and "offset".
*/
struct hv_mpb_array {
/* Length and Offset determines the # of pfns in the array */
u32 len;
u32 offset;
u64 pfn_array[];
};
/* 0x18 includes the proprietary packet header */
#define MAX_PAGE_BUFFER_PACKET (0x18 + \
(sizeof(struct hv_page_buffer) * \
MAX_PAGE_BUFFER_COUNT))
#define MAX_MULTIPAGE_BUFFER_PACKET (0x18 + \
sizeof(struct hv_multipage_buffer))
#pragma pack(pop)
struct hv_ring_buffer {
/* Offset in bytes from the start of ring data below */
u32 write_index;
/* Offset in bytes from the start of ring data below */
u32 read_index;
u32 interrupt_mask;
/*
* Win8 uses some of the reserved bits to implement
* interrupt driven flow management. On the send side
* we can request that the receiver interrupt the sender
* when the ring transitions from being full to being able
* to handle a message of size "pending_send_sz".
*
* Add necessary state for this enhancement.
*/
u32 pending_send_sz;
u32 reserved1[12];
union {
struct {
u32 feat_pending_send_sz:1;
};
u32 value;
} feature_bits;
/* Pad it to PAGE_SIZE so that data starts on page boundary */
u8 reserved2[4028];
/*
* Ring data starts here + RingDataStartOffset
* !!! DO NOT place any fields below this !!!
*/
u8 buffer[0];
} __packed;
struct hv_ring_buffer_info {
struct hv_ring_buffer *ring_buffer;
u32 ring_size; /* Include the shared header */
spinlock_t ring_lock;
u32 ring_datasize; /* < ring_size */
u32 ring_data_startoffset;
u32 priv_write_index;
u32 priv_read_index;
u32 cached_read_index;
};
/*
*
* hv_get_ringbuffer_availbytes()
*
* Get number of bytes available to read and to write to
* for the specified ring buffer
*/
static inline void
hv_get_ringbuffer_availbytes(const struct hv_ring_buffer_info *rbi,
u32 *read, u32 *write)
{
u32 read_loc, write_loc, dsize;
/* Capture the read/write indices before they changed */
read_loc = rbi->ring_buffer->read_index;
write_loc = rbi->ring_buffer->write_index;
dsize = rbi->ring_datasize;
*write = write_loc >= read_loc ? dsize - (write_loc - read_loc) :
read_loc - write_loc;
*read = dsize - *write;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| Haiyang Zhang | 77 | 98.72% | 1 | 50.00% |
| Stephen Hemminger | 1 | 1.28% | 1 | 50.00% |
| Total | 78 | 100.00% | 2 | 100.00% |
static inline u32 hv_get_bytes_to_read(const struct hv_ring_buffer_info *rbi)
{
u32 read_loc, write_loc, dsize, read;
dsize = rbi->ring_datasize;
read_loc = rbi->ring_buffer->read_index;
write_loc = READ_ONCE(rbi->ring_buffer->write_index);
read = write_loc >= read_loc ? (write_loc - read_loc) :
(dsize - read_loc) + write_loc;
return read;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| K. Y. Srinivasan | 69 | 98.57% | 1 | 50.00% |
| Stephen Hemminger | 1 | 1.43% | 1 | 50.00% |
| Total | 70 | 100.00% | 2 | 100.00% |
static inline u32 hv_get_bytes_to_write(const struct hv_ring_buffer_info *rbi)
{
u32 read_loc, write_loc, dsize, write;
dsize = rbi->ring_datasize;
read_loc = READ_ONCE(rbi->ring_buffer->read_index);
write_loc = rbi->ring_buffer->write_index;
write = write_loc >= read_loc ? dsize - (write_loc - read_loc) :
read_loc - write_loc;
return write;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| K. Y. Srinivasan | 67 | 98.53% | 1 | 50.00% |
| Stephen Hemminger | 1 | 1.47% | 1 | 50.00% |
| Total | 68 | 100.00% | 2 | 100.00% |
static inline u32 hv_get_cached_bytes_to_write(
const struct hv_ring_buffer_info *rbi)
{
u32 read_loc, write_loc, dsize, write;
dsize = rbi->ring_datasize;
read_loc = rbi->cached_read_index;
write_loc = rbi->ring_buffer->write_index;
write = write_loc >= read_loc ? dsize - (write_loc - read_loc) :
read_loc - write_loc;
return write;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| Dexuan Cui | 63 | 100.00% | 1 | 100.00% |
| Total | 63 | 100.00% | 1 | 100.00% |
/*
* VMBUS version is 32 bit entity broken up into
* two 16 bit quantities: major_number. minor_number.
*
* 0 . 13 (Windows Server 2008)
* 1 . 1 (Windows 7)
* 2 . 4 (Windows 8)
* 3 . 0 (Windows 8 R2)
* 4 . 0 (Windows 10)
*/
#define VERSION_WS2008 ((0 << 16) | (13))
#define VERSION_WIN7 ((1 << 16) | (1))
#define VERSION_WIN8 ((2 << 16) | (4))
#define VERSION_WIN8_1 ((3 << 16) | (0))
#define VERSION_WIN10 ((4 << 16) | (0))
#define VERSION_INVAL -1
#define VERSION_CURRENT VERSION_WIN10
/* Make maximum size of pipe payload of 16K */
#define MAX_PIPE_DATA_PAYLOAD (sizeof(u8) * 16384)
/* Define PipeMode values. */
#define VMBUS_PIPE_TYPE_BYTE 0x00000000
#define VMBUS_PIPE_TYPE_MESSAGE 0x00000004
/* The size of the user defined data buffer for non-pipe offers. */
#define MAX_USER_DEFINED_BYTES 120
/* The size of the user defined data buffer for pipe offers. */
#define MAX_PIPE_USER_DEFINED_BYTES 116
/*
* At the center of the Channel Management library is the Channel Offer. This
* struct contains the fundamental information about an offer.
*/
struct vmbus_channel_offer {
uuid_le if_type;
uuid_le if_instance;
/*
* These two fields are not currently used.
*/
u64 reserved1;
u64 reserved2;
u16 chn_flags;
u16 mmio_megabytes; /* in bytes * 1024 * 1024 */
union {
/* Non-pipes: The user has MAX_USER_DEFINED_BYTES bytes. */
struct {
unsigned char user_def[MAX_USER_DEFINED_BYTES];
} std;
/*
* Pipes:
* The following sructure is an integrated pipe protocol, which
* is implemented on top of standard user-defined data. Pipe
* clients have MAX_PIPE_USER_DEFINED_BYTES left for their own
* use.
*/
struct {
u32 pipe_mode;
unsigned char user_def[MAX_PIPE_USER_DEFINED_BYTES];
} pipe;
} u;
/*
* The sub_channel_index is defined in win8.
*/
u16 sub_channel_index;
u16 reserved3;
} __packed;
/* Server Flags */
#define VMBUS_CHANNEL_ENUMERATE_DEVICE_INTERFACE 1
#define VMBUS_CHANNEL_SERVER_SUPPORTS_TRANSFER_PAGES 2
#define VMBUS_CHANNEL_SERVER_SUPPORTS_GPADLS 4
#define VMBUS_CHANNEL_NAMED_PIPE_MODE 0x10
#define VMBUS_CHANNEL_LOOPBACK_OFFER 0x100
#define VMBUS_CHANNEL_PARENT_OFFER 0x200
#define VMBUS_CHANNEL_REQUEST_MONITORED_NOTIFICATION 0x400
#define VMBUS_CHANNEL_TLNPI_PROVIDER_OFFER 0x2000
struct vmpacket_descriptor {
u16 type;
u16 offset8;
u16 len8;
u16 flags;
u64 trans_id;
} __packed;
struct vmpacket_header {
u32 prev_pkt_start_offset;
struct vmpacket_descriptor descriptor;
} __packed;
struct vmtransfer_page_range {
u32 byte_count;
u32 byte_offset;
} __packed;
struct vmtransfer_page_packet_header {
struct vmpacket_descriptor d;
u16 xfer_pageset_id;
u8 sender_owns_set;
u8 reserved;
u32 range_cnt;
struct vmtransfer_page_range ranges[1];
} __packed;
struct vmgpadl_packet_header {
struct vmpacket_descriptor d;
u32 gpadl;
u32 reserved;
} __packed;
struct vmadd_remove_transfer_page_set {
struct vmpacket_descriptor d;
u32 gpadl;
u16 xfer_pageset_id;
u16 reserved;
} __packed;
/*
* This structure defines a range in guest physical space that can be made to
* look virtually contiguous.
*/
struct gpa_range {
u32 byte_count;
u32 byte_offset;
u64 pfn_array[0];
};
/*
* This is the format for an Establish Gpadl packet, which contains a handle by
* which this GPADL will be known and a set of GPA ranges associated with it.
* This can be converted to a MDL by the guest OS. If there are multiple GPA
* ranges, then the resulting MDL will be "chained," representing multiple VA
* ranges.
*/
struct vmestablish_gpadl {
struct vmpacket_descriptor d;
u32 gpadl;
u32 range_cnt;
struct gpa_range range[1];
} __packed;
/*
* This is the format for a Teardown Gpadl packet, which indicates that the
* GPADL handle in the Establish Gpadl packet will never be referenced again.
*/
struct vmteardown_gpadl {
struct vmpacket_descriptor d;
u32 gpadl;
u32 reserved; /* for alignment to a 8-byte boundary */
} __packed;
/*
* This is the format for a GPA-Direct packet, which contains a set of GPA
* ranges, in addition to commands and/or data.
*/
struct vmdata_gpa_direct {
struct vmpacket_descriptor d;
u32 reserved;
u32 range_cnt;
struct gpa_range range[1];
} __packed;
/* This is the format for a Additional Data Packet. */
struct vmadditional_data {
struct vmpacket_descriptor d;
u64 total_bytes;
u32 offset;
u32 byte_cnt;
unsigned char data[1];
} __packed;
union vmpacket_largest_possible_header {
struct vmpacket_descriptor simple_hdr;
struct vmtransfer_page_packet_header xfer_page_hdr;
struct vmgpadl_packet_header gpadl_hdr;
struct vmadd_remove_transfer_page_set add_rm_xfer_page_hdr;
struct vmestablish_gpadl establish_gpadl_hdr;
struct vmteardown_gpadl teardown_gpadl_hdr;
struct vmdata_gpa_direct data_gpa_direct_hdr;
};
#define VMPACKET_DATA_START_ADDRESS(__packet) \
(void *)(((unsigned char *)__packet) + \
((struct vmpacket_descriptor)__packet)->offset8 * 8)
#define VMPACKET_DATA_LENGTH(__packet) \
((((struct vmpacket_descriptor)__packet)->len8 - \
((struct vmpacket_descriptor)__packet)->offset8) * 8)
#define VMPACKET_TRANSFER_MODE(__packet) \
(((struct IMPACT)__packet)->type)
enum vmbus_packet_type {
VM_PKT_INVALID = 0x0,
VM_PKT_SYNCH = 0x1,
VM_PKT_ADD_XFER_PAGESET = 0x2,
VM_PKT_RM_XFER_PAGESET = 0x3,
VM_PKT_ESTABLISH_GPADL = 0x4,
VM_PKT_TEARDOWN_GPADL = 0x5,
VM_PKT_DATA_INBAND = 0x6,
VM_PKT_DATA_USING_XFER_PAGES = 0x7,
VM_PKT_DATA_USING_GPADL = 0x8,
VM_PKT_DATA_USING_GPA_DIRECT = 0x9,
VM_PKT_CANCEL_REQUEST = 0xa,
VM_PKT_COMP = 0xb,
VM_PKT_DATA_USING_ADDITIONAL_PKT = 0xc,
VM_PKT_ADDITIONAL_DATA = 0xd
};
#define VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED 1
/* Version 1 messages */
enum vmbus_channel_message_type {
CHANNELMSG_INVALID = 0,
CHANNELMSG_OFFERCHANNEL = 1,
CHANNELMSG_RESCIND_CHANNELOFFER = 2,
CHANNELMSG_REQUESTOFFERS = 3,
CHANNELMSG_ALLOFFERS_DELIVERED = 4,
CHANNELMSG_OPENCHANNEL = 5,
CHANNELMSG_OPENCHANNEL_RESULT = 6,
CHANNELMSG_CLOSECHANNEL = 7,
CHANNELMSG_GPADL_HEADER = 8,
CHANNELMSG_GPADL_BODY = 9,
CHANNELMSG_GPADL_CREATED = 10,
CHANNELMSG_GPADL_TEARDOWN = 11,
CHANNELMSG_GPADL_TORNDOWN = 12,
CHANNELMSG_RELID_RELEASED = 13,
CHANNELMSG_INITIATE_CONTACT = 14,
CHANNELMSG_VERSION_RESPONSE = 15,
CHANNELMSG_UNLOAD = 16,
CHANNELMSG_UNLOAD_RESPONSE = 17,
CHANNELMSG_18 = 18,
CHANNELMSG_19 = 19,
CHANNELMSG_20 = 20,
CHANNELMSG_TL_CONNECT_REQUEST = 21,
CHANNELMSG_COUNT
};
struct vmbus_channel_message_header {
enum vmbus_channel_message_type msgtype;
u32 padding;
} __packed;
/* Query VMBus Version parameters */
struct vmbus_channel_query_vmbus_version {
struct vmbus_channel_message_header header;
u32 version;
} __packed;
/* VMBus Version Supported parameters */
struct vmbus_channel_version_supported {
struct vmbus_channel_message_header header;
u8 version_supported;
} __packed;
/* Offer Channel parameters */
struct vmbus_channel_offer_channel {
struct vmbus_channel_message_header header;
struct vmbus_channel_offer offer;
u32 child_relid;
u8 monitorid;
/*
* win7 and beyond splits this field into a bit field.
*/
u8 monitor_allocated:1;
u8 reserved:7;
/*
* These are new fields added in win7 and later.
* Do not access these fields without checking the
* negotiated protocol.
*
* If "is_dedicated_interrupt" is set, we must not set the
* associated bit in the channel bitmap while sending the
* interrupt to the host.
*
* connection_id is to be used in signaling the host.
*/
u16 is_dedicated_interrupt:1;
u16 reserved1:15;
u32 connection_id;
} __packed;
/* Rescind Offer parameters */
struct vmbus_channel_rescind_offer {
struct vmbus_channel_message_header header;
u32 child_relid;
} __packed;
static inline u32
hv_ringbuffer_pending_size(const struct hv_ring_buffer_info *rbi)
{
return rbi->ring_buffer->pending_send_sz;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| Stephen Hemminger | 20 | 100.00% | 1 | 100.00% |
| Total | 20 | 100.00% | 1 | 100.00% |
/*
* Request Offer -- no parameters, SynIC message contains the partition ID
* Set Snoop -- no parameters, SynIC message contains the partition ID
* Clear Snoop -- no parameters, SynIC message contains the partition ID
* All Offers Delivered -- no parameters, SynIC message contains the partition
* ID
* Flush Client -- no parameters, SynIC message contains the partition ID
*/
/* Open Channel parameters */
struct vmbus_channel_open_channel {
struct vmbus_channel_message_header header;
/* Identifies the specific VMBus channel that is being opened. */
u32 child_relid;
/* ID making a particular open request at a channel offer unique. */
u32 openid;
/* GPADL for the channel's ring buffer. */
u32 ringbuffer_gpadlhandle;
/*
* Starting with win8, this field will be used to specify
* the target virtual processor on which to deliver the interrupt for
* the host to guest communication.
* Prior to win8, incoming channel interrupts would only
* be delivered on cpu 0. Setting this value to 0 would
* preserve the earlier behavior.
*/
u32 target_vp;
/*
* The upstream ring buffer begins at offset zero in the memory
* described by RingBufferGpadlHandle. The downstream ring buffer
* follows it at this offset (in pages).
*/
u32 downstream_ringbuffer_pageoffset;
/* User-specific data to be passed along to the server endpoint. */
unsigned char userdata[MAX_USER_DEFINED_BYTES];
} __packed;
/* Open Channel Result parameters */
struct vmbus_channel_open_result {
struct vmbus_channel_message_header header;
u32 child_relid;
u32 openid;
u32 status;
} __packed;
/* Close channel parameters; */
struct vmbus_channel_close_channel {
struct vmbus_channel_message_header header;
u32 child_relid;
} __packed;
/* Channel Message GPADL */
#define GPADL_TYPE_RING_BUFFER 1
#define GPADL_TYPE_SERVER_SAVE_AREA 2
#define GPADL_TYPE_TRANSACTION 8
/*
* The number of PFNs in a GPADL message is defined by the number of
* pages that would be spanned by ByteCount and ByteOffset. If the
* implied number of PFNs won't fit in this packet, there will be a
* follow-up packet that contains more.
*/
struct vmbus_channel_gpadl_header {
struct vmbus_channel_message_header header;
u32 child_relid;
u32 gpadl;
u16 range_buflen;
u16 rangecount;
struct gpa_range range[0];
} __packed;
/* This is the followup packet that contains more PFNs. */
struct vmbus_channel_gpadl_body {
struct vmbus_channel_message_header header;
u32 msgnumber;
u32 gpadl;
u64 pfn[0];
} __packed;
struct vmbus_channel_gpadl_created {
struct vmbus_channel_message_header header;
u32 child_relid;
u32 gpadl;
u32 creation_status;
} __packed;
struct vmbus_channel_gpadl_teardown {
struct vmbus_channel_message_header header;
u32 child_relid;
u32 gpadl;
} __packed;
struct vmbus_channel_gpadl_torndown {
struct vmbus_channel_message_header header;
u32 gpadl;
} __packed;
struct vmbus_channel_relid_released {
struct vmbus_channel_message_header header;
u32 child_relid;
} __packed;
struct vmbus_channel_initiate_contact {
struct vmbus_channel_message_header header;
u32 vmbus_version_requested;
u32 target_vcpu; /* The VCPU the host should respond to */
u64 interrupt_page;
u64 monitor_page1;
u64 monitor_page2;
} __packed;
/* Hyper-V socket: guest's connect()-ing to host */
struct vmbus_channel_tl_connect_request {
struct vmbus_channel_message_header header;
uuid_le guest_endpoint_id;
uuid_le host_service_id;
} __packed;
struct vmbus_channel_version_response {
struct vmbus_channel_message_header header;
u8 version_supported;
} __packed;
enum vmbus_channel_state {
CHANNEL_OFFER_STATE,
CHANNEL_OPENING_STATE,
CHANNEL_OPEN_STATE,
CHANNEL_OPENED_STATE,
};
/*
* Represents each channel msg on the vmbus connection This is a
* variable-size data structure depending on the msg type itself
*/
struct vmbus_channel_msginfo {
/* Bookkeeping stuff */
struct list_head msglistentry;
/* So far, this is only used to handle gpadl body message */
struct list_head submsglist;
/* Synchronize the request/response if needed */
struct completion waitevent;
struct vmbus_channel *waiting_channel;
union {
struct vmbus_channel_version_supported version_supported;
struct vmbus_channel_open_result open_result;
struct vmbus_channel_gpadl_torndown gpadl_torndown;
struct vmbus_channel_gpadl_created gpadl_created;
struct vmbus_channel_version_response version_response;
} response;
u32 msgsize;
/*
* The channel message that goes out on the "wire".
* It will contain at minimum the VMBUS_CHANNEL_MESSAGE_HEADER header
*/
unsigned char msg[0];
};
struct vmbus_close_msg {
struct vmbus_channel_msginfo info;
struct vmbus_channel_close_channel msg;
};
/* Define connection identifier type. */
union hv_connection_id {
u32 asu32;
struct {
u32 id:24;
u32 reserved:8;
} u;
};
/* Definition of the hv_signal_event hypercall input structure. */
struct hv_input_signal_event {
union hv_connection_id connectionid;
u16 flag_number;
u16 rsvdz;
};
struct hv_input_signal_event_buffer {
u64 align8;
struct hv_input_signal_event event;
};
enum hv_numa_policy {
HV_BALANCED = 0,
HV_LOCALIZED,
};
enum vmbus_device_type {
HV_IDE = 0,
HV_SCSI,
HV_FC,
HV_NIC,
HV_ND,
HV_PCIE,
HV_FB,
HV_KBD,
HV_MOUSE,
HV_KVP,
HV_TS,
HV_HB,
HV_SHUTDOWN,
HV_FCOPY,
HV_BACKUP,
HV_DM,
HV_UNKNOWN,
};
struct vmbus_device {
u16 dev_type;
uuid_le guid;
bool perf_device;
};
struct vmbus_channel {
struct list_head listentry;
struct hv_device *device_obj;
enum vmbus_channel_state state;
struct vmbus_channel_offer_channel offermsg;
/*
* These are based on the OfferMsg.MonitorId.
* Save it here for easy access.
*/
u8 monitor_grp;
u8 monitor_bit;
bool rescind; /* got rescind msg */
u32 ringbuffer_gpadlhandle;
/* Allocated memory for ring buffer */
void *ringbuffer_pages;
u32 ringbuffer_pagecount;
struct hv_ring_buffer_info outbound; /* send to parent */
struct hv_ring_buffer_info inbound; /* receive from parent */
spinlock_t inbound_lock;
struct vmbus_close_msg close_msg;
/* Channel callback's invoked in softirq context */
struct tasklet_struct callback_event;
void (*onchannel_callback)(void *context);
void *channel_callback_context;
/*
* A channel can be marked for one of three modes of reading:
* BATCHED - callback called from taslket and should read
* channel until empty. Interrupts from the host
* are masked while read is in process (default).
* DIRECT - callback called from tasklet (softirq).
* ISR - callback called in interrupt context and must
* invoke its own deferred processing.
* Host interrupts are disabled and must be re-enabled
* when ring is empty.
*/
enum hv_callback_mode {
HV_CALL_BATCHED,
HV_CALL_DIRECT,
HV_CALL_ISR
} callback_mode;
bool is_dedicated_interrupt;
struct hv_input_signal_event_buffer sig_buf;
struct hv_input_signal_event *sig_event;
/*
* Starting with win8, this field will be used to specify
* the target virtual processor on which to deliver the interrupt for
* the host to guest communication.
* Prior to win8, incoming channel interrupts would only
* be delivered on cpu 0. Setting this value to 0 would
* preserve the earlier behavior.
*/
u32 target_vp;
/* The corresponding CPUID in the guest */
u32 target_cpu;
/*
* State to manage the CPU affiliation of channels.
*/
struct cpumask alloced_cpus_in_node;
int numa_node;
/*
* Support for sub-channels. For high performance devices,
* it will be useful to have multiple sub-channels to support
* a scalable communication infrastructure with the host.
* The support for sub-channels is implemented as an extention
* to the current infrastructure.
* The initial offer is considered the primary channel and this
* offer message will indicate if the host supports sub-channels.
* The guest is free to ask for sub-channels to be offerred and can
* open these sub-channels as a normal "primary" channel. However,
* all sub-channels will have the same type and instance guids as the
* primary channel. Requests sent on a given channel will result in a
* response on the same channel.
*/
/*
* Sub-channel creation callback. This callback will be called in
* process context when a sub-channel offer is received from the host.
* The guest can open the sub-channel in the context of this callback.
*/
void (*sc_creation_callback)(struct vmbus_channel *new_sc);
/*
* Channel rescind callback. Some channels (the hvsock ones), need to
* register a callback which is invoked in vmbus_onoffer_rescind().
*/
void (*chn_rescind_callback)(struct vmbus_channel *channel);
/*
* The spinlock to protect the structure. It is being used to protect
* test-and-set access to various attributes of the structure as well
* as all sc_list operations.
*/
spinlock_t lock;
/*
* All Sub-channels of a primary channel are linked here.
*/
struct list_head sc_list;
/*
* Current number of sub-channels.
*/
int num_sc;
/*
* Number of a sub-channel (position within sc_list) which is supposed
* to be used as the next outgoing channel.
*/
int next_oc;
/*
* The primary channel this sub-channel belongs to.
* This will be NULL for the primary channel.
*/
struct vmbus_channel *primary_channel;
/*
* Support per-channel state for use by vmbus drivers.
*/
void *per_channel_state;
/*
* To support per-cpu lookup mapping of relid to channel,
* link up channels based on their CPU affinity.
*/
struct list_head percpu_list;
/*
* Defer freeing channel until after all cpu's have
* gone through grace period.
*/
struct rcu_head rcu;
/*
* For performance critical channels (storage, networking
* etc,), Hyper-V has a mechanism to enhance the throughput
* at the expense of latency:
* When the host is to be signaled, we just set a bit in a shared page
* and this bit will be inspected by the hypervisor within a certain
* window and if the bit is set, the host will be signaled. The window
* of time is the monitor latency - currently around 100 usecs. This
* mechanism improves throughput by:
*
* A) Making the host more efficient - each time it wakes up,
* potentially it will process morev number of packets. The
* monitor latency allows a batch to build up.
* B) By deferring the hypercall to signal, we will also minimize
* the interrupts.
*
* Clearly, these optimizations improve throughput at the expense of
* latency. Furthermore, since the channel is shared for both
* control and data messages, control messages currently suffer
* unnecessary latency adversley impacting performance and boot
* time. To fix this issue, permit tagging the channel as being
* in "low latency" mode. In this mode, we will bypass the monitor
* mechanism.
*/
bool low_latency;
/*
* NUMA distribution policy:
* We support teo policies:
* 1) Balanced: Here all performance critical channels are
* distributed evenly amongst all the NUMA nodes.
* This policy will be the default policy.
* 2) Localized: All channels of a given instance of a
* performance critical service will be assigned CPUs
* within a selected NUMA node.
*/
enum hv_numa_policy affinity_policy;
};
static inline bool is_hvsock_channel(const struct vmbus_channel *c)
{
return !!(c->offermsg.offer.chn_flags &
VMBUS_CHANNEL_TLNPI_PROVIDER_OFFER);
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| Dexuan Cui | 28 | 100.00% | 1 | 100.00% |
| Total | 28 | 100.00% | 1 | 100.00% |
static inline void set_channel_affinity_state(struct vmbus_channel *c,
enum hv_numa_policy policy)
{
c->affinity_policy = policy;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| K. Y. Srinivasan | 22 | 100.00% | 1 | 100.00% |
| Total | 22 | 100.00% | 1 | 100.00% |
static inline void set_channel_read_mode(struct vmbus_channel *c,
enum hv_callback_mode mode)
{
c->callback_mode = mode;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| K. Y. Srinivasan | 16 | 72.73% | 1 | 50.00% |
| Stephen Hemminger | 6 | 27.27% | 1 | 50.00% |
| Total | 22 | 100.00% | 2 | 100.00% |
static inline void set_per_channel_state(struct vmbus_channel *c, void *s)
{
c->per_channel_state = s;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| K. Y. Srinivasan | 22 | 100.00% | 1 | 100.00% |
| Total | 22 | 100.00% | 1 | 100.00% |
static inline void *get_per_channel_state(struct vmbus_channel *c)
{
return c->per_channel_state;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| K. Y. Srinivasan | 18 | 100.00% | 1 | 100.00% |
| Total | 18 | 100.00% | 1 | 100.00% |
static inline void set_channel_pending_send_size(struct vmbus_channel *c,
u32 size)
{
c->outbound.ring_buffer->pending_send_sz = size;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| Dexuan Cui | 25 | 100.00% | 1 | 100.00% |
| Total | 25 | 100.00% | 1 | 100.00% |
static inline void set_low_latency_mode(struct vmbus_channel *c)
{
c->low_latency = true;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| K. Y. Srinivasan | 18 | 100.00% | 1 | 100.00% |
| Total | 18 | 100.00% | 1 | 100.00% |
static inline void clear_low_latency_mode(struct vmbus_channel *c)
{
c->low_latency = false;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
| K. Y. Srinivasan | 18 | 100.00% | 1 | 100.00% |
| Total | 18 | 100.00% | 1 | 100.00% |
void vmbus_onmessage(void *context);
int vmbus_request_offers(void);
/*
* APIs for managing sub-channels.
*/
void vmbus_set_sc_create_callback(struct vmbus_channel *primary_channel,
void (*sc_cr_cb)(struct vmbus_channel *new_sc));
void vmbus_set_chn_rescind_callback(struct vmbus_channel *channel,
void (*chn_rescind_cb)(struct vmbus_channel *));
/*
* Retrieve the (sub) channel on which to send an outgoing request.
* When a primary channel has multiple sub-channels, we choose a
* channel whose VCPU binding is closest to the VCPU on which
* this call is being made.
*/
struct vmbus_channel *vmbus_get_outgoing_channel(struct vmbus_channel *primary);
/*
* Check if sub-channels have already been offerred. This API will be useful
* when the driver is unloaded after establishing sub-channels. In this case,
* when the driver is re-loaded, the driver would have to check if the
* subchannels have already been established before attempting to request
* the creation of sub-channels.
* This function returns TRUE to indicate that subchannels have already been
* created.
* This function should be invoked after setting the callback function for
* sub-channel creation.
*/
bool vmbus_are_subchannels_present(struct vmbus_channel *primary);
/* The format must be the same as struct vmdata_gpa_direct */
struct vmbus_channel_packet_page_buffer {
u16 type;
u16 dataoffset8;
u16 length8;
u16 flags;
u64 transactionid;
u32 reserved;
u32 rangecount;
struct hv_page_buffer range[MAX_PAGE_BUFFER_COUNT];
} __packed;
/* The format must be the same as struct vmdata_gpa_direct */
struct vmbus_channel_packet_multipage_buffer {
u16 type;
u16 dataoffset8;
u16 length8;
u16 flags;
u64 transactionid;
u32 reserved;
u32 rangecount; /* Always 1 in this case */
struct hv_multipage_buffer range;
} __packed;
/* The format must be the same as struct vmdata_gpa_direct */
struct vmbus_packet_mpb_array {
u16 type;
u16 dataoffset8;
u16 length8;
u16 flags;
u64 transactionid;
u32 reserved;
u32 rangecount; /* Always 1 in this case */
struct hv_mpb_array range;
} __packed;
extern int vmbus_open(struct vmbus_channel *channel,
u32 send_ringbuffersize,
u32 recv_ringbuffersize,
void *userdata,
u32 userdatalen,
void (*onchannel_callback)(void *context),
void *context);
extern void vmbus_close(struct vmbus_channel *channel);
extern int vmbus_sendpacket(struct vmbus_channel *channel,
void *buffer,
u32 bufferLen,
u64 requestid,
enum vmbus_packet_type type,
u32 flags);
extern int vmbus_sendpacket_ctl(struct vmbus_channel *channel,
void *buffer,
u32 bufferLen,
u64 requestid,
enum vmbus_packet_type type,
u32 flags);
extern int vmbus_sendpacket_pagebuffer(struct vmbus_channel *channel,
struct hv_page_buffer pagebuffers[],
u32 pagecount,
void *buffer,
u32 bufferlen,
u64 requestid);
extern int vmbus_sendpacket_pagebuffer_ctl(struct vmbus_channel *