^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0-only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * Copyright (c) 2012, Microsoft Corporation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * Author:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) * K. Y. Srinivasan <kys@microsoft.com>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/kernel.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/jiffies.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/mman.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <linux/delay.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/init.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/module.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/kthread.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <linux/completion.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <linux/memory_hotplug.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <linux/memory.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include <linux/notifier.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <linux/percpu_counter.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #include <linux/hyperv.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) #include <asm/hyperv-tlfs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) #include <asm/mshyperv.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) #define CREATE_TRACE_POINTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) #include "hv_trace_balloon.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) * We begin with definitions supporting the Dynamic Memory protocol
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) * with the host.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) * Begin protocol definitions.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) * Protocol versions. The low word is the minor version, the high word the major
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) * version.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) * History:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) * Initial version 1.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) * Changed to 0.1 on 2009/03/25
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) * Changes to 0.2 on 2009/05/14
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) * Changes to 0.3 on 2009/12/03
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) * Changed to 1.0 on 2011/04/05
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) #define DYNMEM_MAKE_VERSION(Major, Minor) ((__u32)(((Major) << 16) | (Minor)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) #define DYNMEM_MAJOR_VERSION(Version) ((__u32)(Version) >> 16)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) #define DYNMEM_MINOR_VERSION(Version) ((__u32)(Version) & 0xff)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) enum {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) DYNMEM_PROTOCOL_VERSION_1 = DYNMEM_MAKE_VERSION(0, 3),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) DYNMEM_PROTOCOL_VERSION_2 = DYNMEM_MAKE_VERSION(1, 0),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) DYNMEM_PROTOCOL_VERSION_3 = DYNMEM_MAKE_VERSION(2, 0),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) DYNMEM_PROTOCOL_VERSION_WIN7 = DYNMEM_PROTOCOL_VERSION_1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) DYNMEM_PROTOCOL_VERSION_WIN8 = DYNMEM_PROTOCOL_VERSION_2,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) DYNMEM_PROTOCOL_VERSION_WIN10 = DYNMEM_PROTOCOL_VERSION_3,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) DYNMEM_PROTOCOL_VERSION_CURRENT = DYNMEM_PROTOCOL_VERSION_WIN10
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) * Message Types
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) enum dm_message_type {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) * Version 0.3
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) DM_ERROR = 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) DM_VERSION_REQUEST = 1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) DM_VERSION_RESPONSE = 2,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) DM_CAPABILITIES_REPORT = 3,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) DM_CAPABILITIES_RESPONSE = 4,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) DM_STATUS_REPORT = 5,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) DM_BALLOON_REQUEST = 6,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) DM_BALLOON_RESPONSE = 7,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) DM_UNBALLOON_REQUEST = 8,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) DM_UNBALLOON_RESPONSE = 9,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) DM_MEM_HOT_ADD_REQUEST = 10,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) DM_MEM_HOT_ADD_RESPONSE = 11,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) DM_VERSION_03_MAX = 11,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) * Version 1.0.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) DM_INFO_MESSAGE = 12,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) DM_VERSION_1_MAX = 12
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) * Structures defining the dynamic memory management
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) * protocol.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) union dm_version {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) struct {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) __u16 minor_version;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) __u16 major_version;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) __u32 version;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) union dm_caps {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) struct {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) __u64 balloon:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) __u64 hot_add:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) * To support guests that may have alignment
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) * limitations on hot-add, the guest can specify
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) * its alignment requirements; a value of n
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) * represents an alignment of 2^n in mega bytes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) __u64 hot_add_alignment:4;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) __u64 reservedz:58;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) } cap_bits;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) __u64 caps;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) union dm_mem_page_range {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) struct {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) * The PFN number of the first page in the range.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) * 40 bits is the architectural limit of a PFN
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) * number for AMD64.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) __u64 start_page:40;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) * The number of pages in the range.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) __u64 page_cnt:24;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) } finfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) __u64 page_range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) * The header for all dynamic memory messages:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) * type: Type of the message.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) * size: Size of the message in bytes; including the header.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) * trans_id: The guest is responsible for manufacturing this ID.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) struct dm_header {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) __u16 type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) __u16 size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) __u32 trans_id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) * A generic message format for dynamic memory.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) * Specific message formats are defined later in the file.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) struct dm_message {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) __u8 data[]; /* enclosed message */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) * Specific message types supporting the dynamic memory protocol.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) * Version negotiation message. Sent from the guest to the host.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) * The guest is free to try different versions until the host
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) * accepts the version.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) * dm_version: The protocol version requested.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) * is_last_attempt: If TRUE, this is the last version guest will request.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) * reservedz: Reserved field, set to zero.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) struct dm_version_request {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) union dm_version version;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) __u32 is_last_attempt:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) __u32 reservedz:31;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) * Version response message; Host to Guest and indicates
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) * if the host has accepted the version sent by the guest.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) * is_accepted: If TRUE, host has accepted the version and the guest
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) * should proceed to the next stage of the protocol. FALSE indicates that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) * guest should re-try with a different version.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) * reservedz: Reserved field, set to zero.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) struct dm_version_response {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) __u64 is_accepted:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) __u64 reservedz:63;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) * Message reporting capabilities. This is sent from the guest to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) * host.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) struct dm_capabilities {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) union dm_caps caps;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) __u64 min_page_cnt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) __u64 max_page_number;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) * Response to the capabilities message. This is sent from the host to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) * guest. This message notifies if the host has accepted the guest's
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) * capabilities. If the host has not accepted, the guest must shutdown
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) * the service.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) * is_accepted: Indicates if the host has accepted guest's capabilities.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) * reservedz: Must be 0.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) struct dm_capabilities_resp_msg {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) __u64 is_accepted:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) __u64 reservedz:63;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) * This message is used to report memory pressure from the guest.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) * This message is not part of any transaction and there is no
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) * response to this message.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) * num_avail: Available memory in pages.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) * num_committed: Committed memory in pages.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) * page_file_size: The accumulated size of all page files
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) * in the system in pages.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) * zero_free: The nunber of zero and free pages.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) * page_file_writes: The writes to the page file in pages.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) * io_diff: An indicator of file cache efficiency or page file activity,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) * calculated as File Cache Page Fault Count - Page Read Count.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) * This value is in pages.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) * Some of these metrics are Windows specific and fortunately
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) * the algorithm on the host side that computes the guest memory
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) * pressure only uses num_committed value.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) struct dm_status {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) __u64 num_avail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) __u64 num_committed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) __u64 page_file_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) __u64 zero_free;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) __u32 page_file_writes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) __u32 io_diff;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) * Message to ask the guest to allocate memory - balloon up message.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) * This message is sent from the host to the guest. The guest may not be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) * able to allocate as much memory as requested.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) * num_pages: number of pages to allocate.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) struct dm_balloon {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) __u32 num_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) __u32 reservedz;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) * Balloon response message; this message is sent from the guest
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) * to the host in response to the balloon message.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) * reservedz: Reserved; must be set to zero.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) * more_pages: If FALSE, this is the last message of the transaction.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) * if TRUE there will atleast one more message from the guest.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) * range_count: The number of ranges in the range array.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) * range_array: An array of page ranges returned to the host.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) struct dm_balloon_response {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) __u32 reservedz;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) __u32 more_pages:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) __u32 range_count:31;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) union dm_mem_page_range range_array[];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) * Un-balloon message; this message is sent from the host
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) * to the guest to give guest more memory.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) * more_pages: If FALSE, this is the last message of the transaction.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) * if TRUE there will atleast one more message from the guest.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) * reservedz: Reserved; must be set to zero.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) * range_count: The number of ranges in the range array.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) * range_array: An array of page ranges returned to the host.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) struct dm_unballoon_request {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) __u32 more_pages:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) __u32 reservedz:31;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) __u32 range_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) union dm_mem_page_range range_array[];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) * Un-balloon response message; this message is sent from the guest
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) * to the host in response to an unballoon request.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) struct dm_unballoon_response {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) * Hot add request message. Message sent from the host to the guest.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) * mem_range: Memory range to hot add.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) struct dm_hot_add {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) union dm_mem_page_range range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) * Hot add response message.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) * This message is sent by the guest to report the status of a hot add request.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) * If page_count is less than the requested page count, then the host should
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) * assume all further hot add requests will fail, since this indicates that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) * the guest has hit an upper physical memory barrier.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) * Hot adds may also fail due to low resources; in this case, the guest must
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) * not complete this message until the hot add can succeed, and the host must
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) * not send a new hot add request until the response is sent.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) * If VSC fails to hot add memory DYNMEM_NUMBER_OF_UNSUCCESSFUL_HOTADD_ATTEMPTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) * times it fails the request.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) * page_count: number of pages that were successfully hot added.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) * result: result of the operation 1: success, 0: failure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) struct dm_hot_add_response {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) __u32 page_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) __u32 result;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) * Types of information sent from host to the guest.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) enum dm_info_type {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) INFO_TYPE_MAX_PAGE_CNT = 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) MAX_INFO_TYPE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) * Header for the information message.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) struct dm_info_header {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) enum dm_info_type type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) __u32 data_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) * This message is sent from the host to the guest to pass
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) * some relevant information (win8 addition).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) * reserved: no used.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) * info_size: size of the information blob.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) * info: information blob.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) struct dm_info_msg {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) struct dm_header hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) __u32 reserved;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) __u32 info_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) __u8 info[];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) * End protocol definitions.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) * State to manage hot adding memory into the guest.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) * The range start_pfn : end_pfn specifies the range
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) * that the host has asked us to hot add. The range
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) * start_pfn : ha_end_pfn specifies the range that we have
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) * currently hot added. We hot add in multiples of 128M
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) * chunks; it is possible that we may not be able to bring
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) * online all the pages in the region. The range
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) * covered_start_pfn:covered_end_pfn defines the pages that can
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) * be brough online.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) struct hv_hotadd_state {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) struct list_head list;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) unsigned long start_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) unsigned long covered_start_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) unsigned long covered_end_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) unsigned long ha_end_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) unsigned long end_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) * A list of gaps.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) struct list_head gap_list;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) struct hv_hotadd_gap {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) struct list_head list;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) unsigned long start_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) unsigned long end_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) struct balloon_state {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) __u32 num_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) struct work_struct wrk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) struct hot_add_wrk {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) union dm_mem_page_range ha_page_range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) union dm_mem_page_range ha_region_range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) struct work_struct wrk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) static bool allow_hibernation;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) static bool hot_add = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) static bool do_hot_add;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) * Delay reporting memory pressure by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) * the specified number of seconds.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) static uint pressure_report_delay = 45;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) * The last time we posted a pressure report to host.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) static unsigned long last_post_time;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) module_param(hot_add, bool, (S_IRUGO | S_IWUSR));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) MODULE_PARM_DESC(hot_add, "If set attempt memory hot_add");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) module_param(pressure_report_delay, uint, (S_IRUGO | S_IWUSR));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) MODULE_PARM_DESC(pressure_report_delay, "Delay in secs in reporting pressure");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) static atomic_t trans_id = ATOMIC_INIT(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) static int dm_ring_size = 20 * 1024;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) * Driver specific state.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) enum hv_dm_state {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) DM_INITIALIZING = 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) DM_INITIALIZED,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) DM_BALLOON_UP,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) DM_BALLOON_DOWN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) DM_HOT_ADD,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) DM_INIT_ERROR
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) static __u8 recv_buffer[HV_HYP_PAGE_SIZE];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) static __u8 balloon_up_send_buffer[HV_HYP_PAGE_SIZE];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) #define PAGES_IN_2M (2 * 1024 * 1024 / PAGE_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) #define HA_CHUNK (128 * 1024 * 1024 / PAGE_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) struct hv_dynmem_device {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) struct hv_device *dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) enum hv_dm_state state;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) struct completion host_event;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) struct completion config_event;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) * Number of pages we have currently ballooned out.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) unsigned int num_pages_ballooned;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) unsigned int num_pages_onlined;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) unsigned int num_pages_added;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) * State to manage the ballooning (up) operation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) struct balloon_state balloon_wrk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) * State to execute the "hot-add" operation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) struct hot_add_wrk ha_wrk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) * This state tracks if the host has specified a hot-add
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) * region.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) bool host_specified_ha_region;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) * State to synchronize hot-add.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) struct completion ol_waitevent;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) * This thread handles hot-add
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) * requests from the host as well as notifying
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) * the host with regards to memory pressure in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) * the guest.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) struct task_struct *thread;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) * Protects ha_region_list, num_pages_onlined counter and individual
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) * regions from ha_region_list.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) spinlock_t ha_lock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) * A list of hot-add regions.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) struct list_head ha_region_list;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) * We start with the highest version we can support
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) * and downgrade based on the host; we save here the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) * next version to try.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) __u32 next_version;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) * The negotiated version agreed by host.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) __u32 version;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) static struct hv_dynmem_device dm_device;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) static void post_status(struct hv_dynmem_device *dm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) static inline bool has_pfn_is_backed(struct hv_hotadd_state *has,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) struct hv_hotadd_gap *gap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) /* The page is not backed. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) if ((pfn < has->covered_start_pfn) || (pfn >= has->covered_end_pfn))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) /* Check for gaps. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) list_for_each_entry(gap, &has->gap_list, list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) if ((pfn >= gap->start_pfn) && (pfn < gap->end_pfn))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) static unsigned long hv_page_offline_check(unsigned long start_pfn,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) unsigned long nr_pages)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) unsigned long pfn = start_pfn, count = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) struct hv_hotadd_state *has;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) bool found;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) while (pfn < start_pfn + nr_pages) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) * Search for HAS which covers the pfn and when we find one
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) * count how many consequitive PFNs are covered.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) found = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) list_for_each_entry(has, &dm_device.ha_region_list, list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) while ((pfn >= has->start_pfn) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) (pfn < has->end_pfn) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) (pfn < start_pfn + nr_pages)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) found = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) if (has_pfn_is_backed(has, pfn))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) count++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) pfn++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) * This PFN is not in any HAS (e.g. we're offlining a region
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) * which was present at boot), no need to account for it. Go
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) * to the next one.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) if (!found)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) pfn++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) return count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) static int hv_memory_notifier(struct notifier_block *nb, unsigned long val,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) void *v)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) struct memory_notify *mem = (struct memory_notify *)v;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) unsigned long flags, pfn_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) switch (val) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) case MEM_ONLINE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) case MEM_CANCEL_ONLINE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) complete(&dm_device.ol_waitevent);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) case MEM_OFFLINE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) spin_lock_irqsave(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) pfn_count = hv_page_offline_check(mem->start_pfn,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) mem->nr_pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) if (pfn_count <= dm_device.num_pages_onlined) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) dm_device.num_pages_onlined -= pfn_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) * We're offlining more pages than we managed to online.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) * This is unexpected. In any case don't let
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649) * num_pages_onlined wrap around zero.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) WARN_ON_ONCE(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) dm_device.num_pages_onlined = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) spin_unlock_irqrestore(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) case MEM_GOING_ONLINE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) case MEM_GOING_OFFLINE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) case MEM_CANCEL_OFFLINE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) return NOTIFY_OK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) static struct notifier_block hv_memory_nb = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) .notifier_call = hv_memory_notifier,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666) .priority = 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669) /* Check if the particular page is backed and can be onlined and online it. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) static void hv_page_online_one(struct hv_hotadd_state *has, struct page *pg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) if (!has_pfn_is_backed(has, page_to_pfn(pg))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) if (!PageOffline(pg))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674) __SetPageOffline(pg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677) if (PageOffline(pg))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) __ClearPageOffline(pg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) /* This frame is currently backed; online the page. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) generic_online_page(pg, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683) lockdep_assert_held(&dm_device.ha_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) dm_device.num_pages_onlined++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) static void hv_bring_pgs_online(struct hv_hotadd_state *has,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) unsigned long start_pfn, unsigned long size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) pr_debug("Online %lu pages starting at pfn 0x%lx\n", size, start_pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) for (i = 0; i < size; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) hv_page_online_one(has, pfn_to_page(start_pfn + i));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697) static void hv_mem_hot_add(unsigned long start, unsigned long size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) unsigned long pfn_count,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) struct hv_hotadd_state *has)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) int i, nid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) unsigned long start_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) unsigned long processed_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) unsigned long total_pfn = pfn_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) for (i = 0; i < (size/HA_CHUNK); i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) start_pfn = start + (i * HA_CHUNK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) spin_lock_irqsave(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) has->ha_end_pfn += HA_CHUNK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) if (total_pfn > HA_CHUNK) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) processed_pfn = HA_CHUNK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) total_pfn -= HA_CHUNK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) processed_pfn = total_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719) total_pfn = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) has->covered_end_pfn += processed_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) spin_unlock_irqrestore(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) reinit_completion(&dm_device.ol_waitevent);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) nid = memory_add_physaddr_to_nid(PFN_PHYS(start_pfn));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) ret = add_memory(nid, PFN_PHYS((start_pfn)),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) (HA_CHUNK << PAGE_SHIFT), MEMHP_MERGE_RESOURCE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) pr_err("hot_add memory failed error is %d\n", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) if (ret == -EEXIST) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) * This error indicates that the error
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) * is not a transient failure. This is the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) * case where the guest's physical address map
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) * precludes hot adding memory. Stop all further
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) * memory hot-add.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741) do_hot_add = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743) spin_lock_irqsave(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) has->ha_end_pfn -= HA_CHUNK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) has->covered_end_pfn -= processed_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) spin_unlock_irqrestore(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) * Wait for memory to get onlined. If the kernel onlined the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752) * memory when adding it, this will return directly. Otherwise,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) * it will wait for user space to online the memory. This helps
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) * to avoid adding memory faster than it is getting onlined. As
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) * adding succeeded, it is ok to proceed even if the memory was
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) * not onlined in time.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) wait_for_completion_timeout(&dm_device.ol_waitevent, 5 * HZ);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759) post_status(&dm_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763) static void hv_online_page(struct page *pg, unsigned int order)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) struct hv_hotadd_state *has;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767) unsigned long pfn = page_to_pfn(pg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) spin_lock_irqsave(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770) list_for_each_entry(has, &dm_device.ha_region_list, list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) /* The page belongs to a different HAS. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) if ((pfn < has->start_pfn) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) (pfn + (1UL << order) > has->end_pfn))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) hv_bring_pgs_online(has, pfn, 1UL << order);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779) spin_unlock_irqrestore(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782) static int pfn_covered(unsigned long start_pfn, unsigned long pfn_cnt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) struct hv_hotadd_state *has;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) struct hv_hotadd_gap *gap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) unsigned long residual, new_inc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) spin_lock_irqsave(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) list_for_each_entry(has, &dm_device.ha_region_list, list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) * If the pfn range we are dealing with is not in the current
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) * "hot add block", move on.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) if (start_pfn < has->start_pfn || start_pfn >= has->end_pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) * If the current start pfn is not where the covered_end
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) * is, create a gap and update covered_end_pfn.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) if (has->covered_end_pfn != start_pfn) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) gap = kzalloc(sizeof(struct hv_hotadd_gap), GFP_ATOMIC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805) if (!gap) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806) ret = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) INIT_LIST_HEAD(&gap->list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) gap->start_pfn = has->covered_end_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812) gap->end_pfn = start_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) list_add_tail(&gap->list, &has->gap_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815) has->covered_end_pfn = start_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) * If the current hot add-request extends beyond
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) * our current limit; extend it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) if ((start_pfn + pfn_cnt) > has->end_pfn) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) residual = (start_pfn + pfn_cnt - has->end_pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) * Extend the region by multiples of HA_CHUNK.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827) new_inc = (residual / HA_CHUNK) * HA_CHUNK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) if (residual % HA_CHUNK)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) new_inc += HA_CHUNK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) has->end_pfn += new_inc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) ret = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) spin_unlock_irqrestore(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842) static unsigned long handle_pg_range(unsigned long pg_start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) unsigned long pg_count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) unsigned long start_pfn = pg_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846) unsigned long pfn_cnt = pg_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) unsigned long size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) struct hv_hotadd_state *has;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) unsigned long pgs_ol = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850) unsigned long old_covered_state;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) unsigned long res = 0, flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) pr_debug("Hot adding %lu pages starting at pfn 0x%lx.\n", pg_count,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) pg_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) spin_lock_irqsave(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) list_for_each_entry(has, &dm_device.ha_region_list, list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859) * If the pfn range we are dealing with is not in the current
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) * "hot add block", move on.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) if (start_pfn < has->start_pfn || start_pfn >= has->end_pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) old_covered_state = has->covered_end_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) if (start_pfn < has->ha_end_pfn) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) * This is the case where we are backing pages
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) * in an already hot added region. Bring
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) * these pages online first.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) pgs_ol = has->ha_end_pfn - start_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) if (pgs_ol > pfn_cnt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) pgs_ol = pfn_cnt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) has->covered_end_pfn += pgs_ol;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) pfn_cnt -= pgs_ol;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) * Check if the corresponding memory block is already
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881) * online. It is possible to observe struct pages still
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) * being uninitialized here so check section instead.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) * In case the section is online we need to bring the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) * rest of pfns (which were not backed previously)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) * online too.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) if (start_pfn > has->start_pfn &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) online_section_nr(pfn_to_section_nr(start_pfn)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) hv_bring_pgs_online(has, start_pfn, pgs_ol);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) if ((has->ha_end_pfn < has->end_pfn) && (pfn_cnt > 0)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895) * We have some residual hot add range
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) * that needs to be hot added; hot add
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) * it now. Hot add a multiple of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) * of HA_CHUNK that fully covers the pages
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) * we have.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901) size = (has->end_pfn - has->ha_end_pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902) if (pfn_cnt <= size) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) size = ((pfn_cnt / HA_CHUNK) * HA_CHUNK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) if (pfn_cnt % HA_CHUNK)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) size += HA_CHUNK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907) pfn_cnt = size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909) spin_unlock_irqrestore(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910) hv_mem_hot_add(has->ha_end_pfn, size, pfn_cnt, has);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) spin_lock_irqsave(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) * If we managed to online any pages that were given to us,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915) * we declare success.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) res = has->covered_end_pfn - old_covered_state;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920) spin_unlock_irqrestore(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922) return res;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) static unsigned long process_hot_add(unsigned long pg_start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926) unsigned long pfn_cnt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927) unsigned long rg_start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) unsigned long rg_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) struct hv_hotadd_state *ha_region = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) int covered;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) if (pfn_cnt == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) if (!dm_device.host_specified_ha_region) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938) covered = pfn_covered(pg_start, pfn_cnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939) if (covered < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) if (covered)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) goto do_pg_range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947) * If the host has specified a hot-add range; deal with it first.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950) if (rg_size != 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951) ha_region = kzalloc(sizeof(struct hv_hotadd_state), GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952) if (!ha_region)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955) INIT_LIST_HEAD(&ha_region->list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) INIT_LIST_HEAD(&ha_region->gap_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958) ha_region->start_pfn = rg_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) ha_region->ha_end_pfn = rg_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) ha_region->covered_start_pfn = pg_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) ha_region->covered_end_pfn = pg_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) ha_region->end_pfn = rg_start + rg_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964) spin_lock_irqsave(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965) list_add_tail(&ha_region->list, &dm_device.ha_region_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966) spin_unlock_irqrestore(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) do_pg_range:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971) * Process the page range specified; bringing them
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972) * online if possible.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974) return handle_pg_range(pg_start, pfn_cnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) static void hot_add_req(struct work_struct *dummy)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981) struct dm_hot_add_response resp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) unsigned long pg_start, pfn_cnt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984) unsigned long rg_start, rg_sz;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986) struct hv_dynmem_device *dm = &dm_device;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) memset(&resp, 0, sizeof(struct dm_hot_add_response));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989) resp.hdr.type = DM_MEM_HOT_ADD_RESPONSE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990) resp.hdr.size = sizeof(struct dm_hot_add_response);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993) pg_start = dm->ha_wrk.ha_page_range.finfo.start_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994) pfn_cnt = dm->ha_wrk.ha_page_range.finfo.page_cnt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996) rg_start = dm->ha_wrk.ha_region_range.finfo.start_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) rg_sz = dm->ha_wrk.ha_region_range.finfo.page_cnt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) if ((rg_start == 0) && (!dm->host_specified_ha_region)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) unsigned long region_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001) unsigned long region_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004) * The host has not specified the hot-add region.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005) * Based on the hot-add page range being specified,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006) * compute a hot-add region that can cover the pages
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) * that need to be hot-added while ensuring the alignment
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008) * and size requirements of Linux as it relates to hot-add.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010) region_start = pg_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) region_size = (pfn_cnt / HA_CHUNK) * HA_CHUNK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012) if (pfn_cnt % HA_CHUNK)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013) region_size += HA_CHUNK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015) region_start = (pg_start / HA_CHUNK) * HA_CHUNK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) rg_start = region_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018) rg_sz = region_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) if (do_hot_add)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) resp.page_count = process_hot_add(pg_start, pfn_cnt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023) rg_start, rg_sz);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) dm->num_pages_added += resp.page_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) * The result field of the response structure has the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) * following semantics:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) * 1. If all or some pages hot-added: Guest should return success.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033) * 2. If no pages could be hot-added:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) * If the guest returns success, then the host
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) * will not attempt any further hot-add operations. This
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) * signifies a permanent failure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) * If the guest returns failure, then this failure will be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) * treated as a transient failure and the host may retry the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041) * hot-add operation after some delay.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043) if (resp.page_count > 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044) resp.result = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) else if (!do_hot_add)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046) resp.result = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) resp.result = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) if (!do_hot_add || resp.page_count == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) if (!allow_hibernation)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052) pr_err("Memory hot add failed\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) pr_info("Ignore hot-add request!\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) dm->state = DM_INITIALIZED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) resp.hdr.trans_id = atomic_inc_return(&trans_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) vmbus_sendpacket(dm->dev->channel, &resp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060) sizeof(struct dm_hot_add_response),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) (unsigned long)NULL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) VM_PKT_DATA_INBAND, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) static void process_info(struct hv_dynmem_device *dm, struct dm_info_msg *msg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067) struct dm_info_header *info_hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069) info_hdr = (struct dm_info_header *)msg->info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) switch (info_hdr->type) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) case INFO_TYPE_MAX_PAGE_CNT:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) if (info_hdr->data_size == sizeof(__u64)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) __u64 *max_page_count = (__u64 *)&info_hdr[1];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) pr_info("Max. dynamic memory size: %llu MB\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) (*max_page_count) >> (20 - HV_HYP_PAGE_SHIFT));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) pr_warn("Received Unknown type: %d\n", info_hdr->type);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086) static unsigned long compute_balloon_floor(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088) unsigned long min_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) unsigned long nr_pages = totalram_pages();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) #define MB2PAGES(mb) ((mb) << (20 - PAGE_SHIFT))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) /* Simple continuous piecewiese linear function:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) * max MiB -> min MiB gradient
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093) * 0 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094) * 16 16
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) * 32 24
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096) * 128 72 (1/2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) * 512 168 (1/4)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098) * 2048 360 (1/8)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099) * 8192 744 (1/16)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) * 32768 1512 (1/32)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102) if (nr_pages < MB2PAGES(128))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) min_pages = MB2PAGES(8) + (nr_pages >> 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) else if (nr_pages < MB2PAGES(512))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) min_pages = MB2PAGES(40) + (nr_pages >> 2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) else if (nr_pages < MB2PAGES(2048))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107) min_pages = MB2PAGES(104) + (nr_pages >> 3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) else if (nr_pages < MB2PAGES(8192))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109) min_pages = MB2PAGES(232) + (nr_pages >> 4);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) min_pages = MB2PAGES(488) + (nr_pages >> 5);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) #undef MB2PAGES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113) return min_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) * Post our status as it relates memory pressure to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) * host. Host expects the guests to post this status
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) * periodically at 1 second intervals.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) * The metrics specified in this protocol are very Windows
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122) * specific and so we cook up numbers here to convey our memory
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) * pressure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) static void post_status(struct hv_dynmem_device *dm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128) struct dm_status status;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) unsigned long now = jiffies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130) unsigned long last_post = last_post_time;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132) if (pressure_report_delay > 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) --pressure_report_delay;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) if (!time_after(now, (last_post_time + HZ)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) memset(&status, 0, sizeof(struct dm_status));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) status.hdr.type = DM_STATUS_REPORT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142) status.hdr.size = sizeof(struct dm_status);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) status.hdr.trans_id = atomic_inc_return(&trans_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) * The host expects the guest to report free and committed memory.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147) * Furthermore, the host expects the pressure information to include
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148) * the ballooned out pages. For a given amount of memory that we are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149) * managing we need to compute a floor below which we should not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) * balloon. Compute this and add it to the pressure report.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151) * We also need to report all offline pages (num_pages_added -
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152) * num_pages_onlined) as committed to the host, otherwise it can try
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153) * asking us to balloon them out.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155) status.num_avail = si_mem_available();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156) status.num_committed = vm_memory_committed() +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157) dm->num_pages_ballooned +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158) (dm->num_pages_added > dm->num_pages_onlined ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159) dm->num_pages_added - dm->num_pages_onlined : 0) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160) compute_balloon_floor();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162) trace_balloon_status(status.num_avail, status.num_committed,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163) vm_memory_committed(), dm->num_pages_ballooned,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164) dm->num_pages_added, dm->num_pages_onlined);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1165) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1166) * If our transaction ID is no longer current, just don't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1167) * send the status. This can happen if we were interrupted
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168) * after we picked our transaction ID.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170) if (status.hdr.trans_id != atomic_read(&trans_id))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) * If the last post time that we sampled has changed,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175) * we have raced, don't post the status.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) if (last_post != last_post_time)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180) last_post_time = jiffies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181) vmbus_sendpacket(dm->dev->channel, &status,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182) sizeof(struct dm_status),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183) (unsigned long)NULL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184) VM_PKT_DATA_INBAND, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188) static void free_balloon_pages(struct hv_dynmem_device *dm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189) union dm_mem_page_range *range_array)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191) int num_pages = range_array->finfo.page_cnt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192) __u64 start_frame = range_array->finfo.start_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193) struct page *pg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196) for (i = 0; i < num_pages; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197) pg = pfn_to_page(i + start_frame);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198) __ClearPageOffline(pg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199) __free_page(pg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200) dm->num_pages_ballooned--;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1204)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1205)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1206) static unsigned int alloc_balloon_pages(struct hv_dynmem_device *dm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) unsigned int num_pages,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208) struct dm_balloon_response *bl_resp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) int alloc_unit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211) unsigned int i, j;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) struct page *pg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214) for (i = 0; i < num_pages / alloc_unit; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) if (bl_resp->hdr.size + sizeof(union dm_mem_page_range) >
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216) HV_HYP_PAGE_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217) return i * alloc_unit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220) * We execute this code in a thread context. Furthermore,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221) * we don't want the kernel to try too hard.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223) pg = alloc_pages(GFP_HIGHUSER | __GFP_NORETRY |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224) __GFP_NOMEMALLOC | __GFP_NOWARN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225) get_order(alloc_unit << PAGE_SHIFT));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) if (!pg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228) return i * alloc_unit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230) dm->num_pages_ballooned += alloc_unit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233) * If we allocatted 2M pages; split them so we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234) * can free them in any order we get.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) if (alloc_unit != 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238) split_page(pg, get_order(alloc_unit << PAGE_SHIFT));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240) /* mark all pages offline */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241) for (j = 0; j < (1 << get_order(alloc_unit << PAGE_SHIFT)); j++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242) __SetPageOffline(pg + j);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244) bl_resp->range_count++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245) bl_resp->range_array[i].finfo.start_page =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246) page_to_pfn(pg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) bl_resp->range_array[i].finfo.page_cnt = alloc_unit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248) bl_resp->hdr.size += sizeof(union dm_mem_page_range);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) return i * alloc_unit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255) static void balloon_up(struct work_struct *dummy)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257) unsigned int num_pages = dm_device.balloon_wrk.num_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258) unsigned int num_ballooned = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259) struct dm_balloon_response *bl_resp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260) int alloc_unit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262) bool done = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264) long avail_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265) unsigned long floor;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268) * We will attempt 2M allocations. However, if we fail to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269) * allocate 2M chunks, we will go back to PAGE_SIZE allocations.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1271) alloc_unit = PAGES_IN_2M;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1272)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1273) avail_pages = si_mem_available();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1274) floor = compute_balloon_floor();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1275)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1276) /* Refuse to balloon below the floor. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1277) if (avail_pages < num_pages || avail_pages - num_pages < floor) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1278) pr_info("Balloon request will be partially fulfilled. %s\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1279) avail_pages < num_pages ? "Not enough memory." :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1280) "Balloon floor reached.");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1281)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1282) num_pages = avail_pages > floor ? (avail_pages - floor) : 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1283) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1284)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1285) while (!done) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1286) memset(balloon_up_send_buffer, 0, HV_HYP_PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1287) bl_resp = (struct dm_balloon_response *)balloon_up_send_buffer;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1288) bl_resp->hdr.type = DM_BALLOON_RESPONSE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1289) bl_resp->hdr.size = sizeof(struct dm_balloon_response);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1290) bl_resp->more_pages = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1291)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1292) num_pages -= num_ballooned;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1293) num_ballooned = alloc_balloon_pages(&dm_device, num_pages,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1294) bl_resp, alloc_unit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1295)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1296) if (alloc_unit != 1 && num_ballooned == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1297) alloc_unit = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1298) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1299) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1300)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1301) if (num_ballooned == 0 || num_ballooned == num_pages) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1302) pr_debug("Ballooned %u out of %u requested pages.\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1303) num_pages, dm_device.balloon_wrk.num_pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1304)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1305) bl_resp->more_pages = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1306) done = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1307) dm_device.state = DM_INITIALIZED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1308) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1309)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1310) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1311) * We are pushing a lot of data through the channel;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1312) * deal with transient failures caused because of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1313) * lack of space in the ring buffer.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1314) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1315)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1316) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1317) bl_resp->hdr.trans_id = atomic_inc_return(&trans_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1318) ret = vmbus_sendpacket(dm_device.dev->channel,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1319) bl_resp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1320) bl_resp->hdr.size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1321) (unsigned long)NULL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1322) VM_PKT_DATA_INBAND, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1323)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1324) if (ret == -EAGAIN)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1325) msleep(20);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1326) post_status(&dm_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1327) } while (ret == -EAGAIN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1328)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1329) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1330) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1331) * Free up the memory we allocatted.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1332) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1333) pr_err("Balloon response failed\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1334)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1335) for (i = 0; i < bl_resp->range_count; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1336) free_balloon_pages(&dm_device,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1337) &bl_resp->range_array[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1338)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1339) done = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1340) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1341) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1342)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1343) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1345) static void balloon_down(struct hv_dynmem_device *dm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1346) struct dm_unballoon_request *req)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1347) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1348) union dm_mem_page_range *range_array = req->range_array;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1349) int range_count = req->range_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1350) struct dm_unballoon_response resp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1351) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1352) unsigned int prev_pages_ballooned = dm->num_pages_ballooned;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1353)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1354) for (i = 0; i < range_count; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1355) free_balloon_pages(dm, &range_array[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1356) complete(&dm_device.config_event);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1357) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1358)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1359) pr_debug("Freed %u ballooned pages.\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1360) prev_pages_ballooned - dm->num_pages_ballooned);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1361)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1362) if (req->more_pages == 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1363) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1364)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1365) memset(&resp, 0, sizeof(struct dm_unballoon_response));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1366) resp.hdr.type = DM_UNBALLOON_RESPONSE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1367) resp.hdr.trans_id = atomic_inc_return(&trans_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1368) resp.hdr.size = sizeof(struct dm_unballoon_response);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1369)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1370) vmbus_sendpacket(dm_device.dev->channel, &resp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1371) sizeof(struct dm_unballoon_response),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1372) (unsigned long)NULL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1373) VM_PKT_DATA_INBAND, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1374)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1375) dm->state = DM_INITIALIZED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1376) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1377)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1378) static void balloon_onchannelcallback(void *context);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1379)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1380) static int dm_thread_func(void *dm_dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1381) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1382) struct hv_dynmem_device *dm = dm_dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1383)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1384) while (!kthread_should_stop()) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1385) wait_for_completion_interruptible_timeout(
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1386) &dm_device.config_event, 1*HZ);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1387) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1388) * The host expects us to post information on the memory
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1389) * pressure every second.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1390) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1391) reinit_completion(&dm_device.config_event);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1392) post_status(dm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1393) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1394)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1395) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1396) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1397)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1398)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1399) static void version_resp(struct hv_dynmem_device *dm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1400) struct dm_version_response *vresp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1401) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1402) struct dm_version_request version_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1403) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1404)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1405) if (vresp->is_accepted) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1406) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1407) * We are done; wakeup the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1408) * context waiting for version
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1409) * negotiation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1410) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1411) complete(&dm->host_event);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1412) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1413) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1414) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1415) * If there are more versions to try, continue
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1416) * with negotiations; if not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1417) * shutdown the service since we are not able
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1418) * to negotiate a suitable version number
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1419) * with the host.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1420) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1421) if (dm->next_version == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1422) goto version_error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1423)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1424) memset(&version_req, 0, sizeof(struct dm_version_request));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1425) version_req.hdr.type = DM_VERSION_REQUEST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1426) version_req.hdr.size = sizeof(struct dm_version_request);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1427) version_req.hdr.trans_id = atomic_inc_return(&trans_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1428) version_req.version.version = dm->next_version;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1429) dm->version = version_req.version.version;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1430)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1431) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1432) * Set the next version to try in case current version fails.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1433) * Win7 protocol ought to be the last one to try.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1434) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1435) switch (version_req.version.version) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1436) case DYNMEM_PROTOCOL_VERSION_WIN8:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1437) dm->next_version = DYNMEM_PROTOCOL_VERSION_WIN7;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1438) version_req.is_last_attempt = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1439) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1440) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1441) dm->next_version = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1442) version_req.is_last_attempt = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1443) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1444)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1445) ret = vmbus_sendpacket(dm->dev->channel, &version_req,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1446) sizeof(struct dm_version_request),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1447) (unsigned long)NULL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1448) VM_PKT_DATA_INBAND, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1449)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1450) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1451) goto version_error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1452)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1453) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1454)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1455) version_error:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1456) dm->state = DM_INIT_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1457) complete(&dm->host_event);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1458) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1459)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1460) static void cap_resp(struct hv_dynmem_device *dm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1461) struct dm_capabilities_resp_msg *cap_resp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1462) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1463) if (!cap_resp->is_accepted) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1464) pr_err("Capabilities not accepted by host\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1465) dm->state = DM_INIT_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1466) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1467) complete(&dm->host_event);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1468) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1469)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1470) static void balloon_onchannelcallback(void *context)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1471) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1472) struct hv_device *dev = context;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1473) u32 recvlen;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1474) u64 requestid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1475) struct dm_message *dm_msg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1476) struct dm_header *dm_hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1477) struct hv_dynmem_device *dm = hv_get_drvdata(dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1478) struct dm_balloon *bal_msg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1479) struct dm_hot_add *ha_msg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1480) union dm_mem_page_range *ha_pg_range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1481) union dm_mem_page_range *ha_region;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1482)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1483) memset(recv_buffer, 0, sizeof(recv_buffer));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1484) vmbus_recvpacket(dev->channel, recv_buffer,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1485) HV_HYP_PAGE_SIZE, &recvlen, &requestid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1486)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1487) if (recvlen > 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1488) dm_msg = (struct dm_message *)recv_buffer;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1489) dm_hdr = &dm_msg->hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1491) switch (dm_hdr->type) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1492) case DM_VERSION_RESPONSE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1493) version_resp(dm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1494) (struct dm_version_response *)dm_msg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1495) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1496)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1497) case DM_CAPABILITIES_RESPONSE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1498) cap_resp(dm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1499) (struct dm_capabilities_resp_msg *)dm_msg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1500) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1501)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1502) case DM_BALLOON_REQUEST:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1503) if (allow_hibernation) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1504) pr_info("Ignore balloon-up request!\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1505) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1506) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1507)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1508) if (dm->state == DM_BALLOON_UP)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1509) pr_warn("Currently ballooning\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1510) bal_msg = (struct dm_balloon *)recv_buffer;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1511) dm->state = DM_BALLOON_UP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1512) dm_device.balloon_wrk.num_pages = bal_msg->num_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1513) schedule_work(&dm_device.balloon_wrk.wrk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1514) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1515)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1516) case DM_UNBALLOON_REQUEST:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1517) if (allow_hibernation) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1518) pr_info("Ignore balloon-down request!\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1519) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1520) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1521)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1522) dm->state = DM_BALLOON_DOWN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1523) balloon_down(dm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1524) (struct dm_unballoon_request *)recv_buffer);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1525) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1526)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1527) case DM_MEM_HOT_ADD_REQUEST:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1528) if (dm->state == DM_HOT_ADD)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1529) pr_warn("Currently hot-adding\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1530) dm->state = DM_HOT_ADD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1531) ha_msg = (struct dm_hot_add *)recv_buffer;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1532) if (ha_msg->hdr.size == sizeof(struct dm_hot_add)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1533) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1534) * This is a normal hot-add request specifying
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1535) * hot-add memory.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1536) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1537) dm->host_specified_ha_region = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1538) ha_pg_range = &ha_msg->range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1539) dm->ha_wrk.ha_page_range = *ha_pg_range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1540) dm->ha_wrk.ha_region_range.page_range = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1541) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1542) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1543) * Host is specifying that we first hot-add
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1544) * a region and then partially populate this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1545) * region.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1546) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1547) dm->host_specified_ha_region = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1548) ha_pg_range = &ha_msg->range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1549) ha_region = &ha_pg_range[1];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1550) dm->ha_wrk.ha_page_range = *ha_pg_range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1551) dm->ha_wrk.ha_region_range = *ha_region;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1552) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1553) schedule_work(&dm_device.ha_wrk.wrk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1554) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1555)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1556) case DM_INFO_MESSAGE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1557) process_info(dm, (struct dm_info_msg *)dm_msg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1558) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1559)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1560) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1561) pr_warn_ratelimited("Unhandled message: type: %d\n", dm_hdr->type);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1562)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1563) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1564) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1565)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1566) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1567)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1568) static int balloon_connect_vsp(struct hv_device *dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1569) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1570) struct dm_version_request version_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1571) struct dm_capabilities cap_msg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1572) unsigned long t;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1573) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1574)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1575) ret = vmbus_open(dev->channel, dm_ring_size, dm_ring_size, NULL, 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1576) balloon_onchannelcallback, dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1577) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1578) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1579)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1580) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1581) * Initiate the hand shake with the host and negotiate
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1582) * a version that the host can support. We start with the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1583) * highest version number and go down if the host cannot
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1584) * support it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1585) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1586) memset(&version_req, 0, sizeof(struct dm_version_request));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1587) version_req.hdr.type = DM_VERSION_REQUEST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1588) version_req.hdr.size = sizeof(struct dm_version_request);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1589) version_req.hdr.trans_id = atomic_inc_return(&trans_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1590) version_req.version.version = DYNMEM_PROTOCOL_VERSION_WIN10;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1591) version_req.is_last_attempt = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1592) dm_device.version = version_req.version.version;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1593)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1594) ret = vmbus_sendpacket(dev->channel, &version_req,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1595) sizeof(struct dm_version_request),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1596) (unsigned long)NULL, VM_PKT_DATA_INBAND, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1597) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1598) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1599)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1600) t = wait_for_completion_timeout(&dm_device.host_event, 5*HZ);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1601) if (t == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1602) ret = -ETIMEDOUT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1603) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1604) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1605)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1606) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1607) * If we could not negotiate a compatible version with the host
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1608) * fail the probe function.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1609) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1610) if (dm_device.state == DM_INIT_ERROR) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1611) ret = -EPROTO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1612) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1613) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1614)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1615) pr_info("Using Dynamic Memory protocol version %u.%u\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1616) DYNMEM_MAJOR_VERSION(dm_device.version),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1617) DYNMEM_MINOR_VERSION(dm_device.version));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1618)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1619) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1620) * Now submit our capabilities to the host.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1621) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1622) memset(&cap_msg, 0, sizeof(struct dm_capabilities));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1623) cap_msg.hdr.type = DM_CAPABILITIES_REPORT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1624) cap_msg.hdr.size = sizeof(struct dm_capabilities);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1625) cap_msg.hdr.trans_id = atomic_inc_return(&trans_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1626)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1627) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1628) * When hibernation (i.e. virtual ACPI S4 state) is enabled, the host
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1629) * currently still requires the bits to be set, so we have to add code
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1630) * to fail the host's hot-add and balloon up/down requests, if any.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1631) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1632) cap_msg.caps.cap_bits.balloon = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1633) cap_msg.caps.cap_bits.hot_add = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1634)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1635) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1636) * Specify our alignment requirements as it relates
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1637) * memory hot-add. Specify 128MB alignment.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1638) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1639) cap_msg.caps.cap_bits.hot_add_alignment = 7;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1640)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1641) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1642) * Currently the host does not use these
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1643) * values and we set them to what is done in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1644) * Windows driver.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1645) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1646) cap_msg.min_page_cnt = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1647) cap_msg.max_page_number = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1648)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1649) ret = vmbus_sendpacket(dev->channel, &cap_msg,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1650) sizeof(struct dm_capabilities),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1651) (unsigned long)NULL, VM_PKT_DATA_INBAND, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1652) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1653) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1654)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1655) t = wait_for_completion_timeout(&dm_device.host_event, 5*HZ);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1656) if (t == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1657) ret = -ETIMEDOUT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1658) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1659) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1660)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1661) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1662) * If the host does not like our capabilities,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1663) * fail the probe function.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1664) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1665) if (dm_device.state == DM_INIT_ERROR) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1666) ret = -EPROTO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1667) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1668) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1669)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1670) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1671) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1672) vmbus_close(dev->channel);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1673) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1674) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1675)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1676) static int balloon_probe(struct hv_device *dev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1677) const struct hv_vmbus_device_id *dev_id)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1678) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1679) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1680)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1681) allow_hibernation = hv_is_hibernation_supported();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1682) if (allow_hibernation)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1683) hot_add = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1684)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1685) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1686) do_hot_add = hot_add;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1687) #else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1688) do_hot_add = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1689) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1690) dm_device.dev = dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1691) dm_device.state = DM_INITIALIZING;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1692) dm_device.next_version = DYNMEM_PROTOCOL_VERSION_WIN8;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1693) init_completion(&dm_device.host_event);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1694) init_completion(&dm_device.config_event);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1695) INIT_LIST_HEAD(&dm_device.ha_region_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1696) spin_lock_init(&dm_device.ha_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1697) INIT_WORK(&dm_device.balloon_wrk.wrk, balloon_up);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1698) INIT_WORK(&dm_device.ha_wrk.wrk, hot_add_req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1699) dm_device.host_specified_ha_region = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1700)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1701) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1702) set_online_page_callback(&hv_online_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1703) init_completion(&dm_device.ol_waitevent);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1704) register_memory_notifier(&hv_memory_nb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1705) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1706)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1707) hv_set_drvdata(dev, &dm_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1708)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1709) ret = balloon_connect_vsp(dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1710) if (ret != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1711) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1712)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1713) dm_device.state = DM_INITIALIZED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1714)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1715) dm_device.thread =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1716) kthread_run(dm_thread_func, &dm_device, "hv_balloon");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1717) if (IS_ERR(dm_device.thread)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1718) ret = PTR_ERR(dm_device.thread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1719) goto probe_error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1720) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1721)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1722) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1723)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1724) probe_error:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1725) dm_device.state = DM_INIT_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1726) dm_device.thread = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1727) vmbus_close(dev->channel);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1728) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1729) unregister_memory_notifier(&hv_memory_nb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1730) restore_online_page_callback(&hv_online_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1731) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1732) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1733) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1734)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1735) static int balloon_remove(struct hv_device *dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1736) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1737) struct hv_dynmem_device *dm = hv_get_drvdata(dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1738) struct hv_hotadd_state *has, *tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1739) struct hv_hotadd_gap *gap, *tmp_gap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1740) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1741)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1742) if (dm->num_pages_ballooned != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1743) pr_warn("Ballooned pages: %d\n", dm->num_pages_ballooned);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1744)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1745) cancel_work_sync(&dm->balloon_wrk.wrk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1746) cancel_work_sync(&dm->ha_wrk.wrk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1747)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1748) kthread_stop(dm->thread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1749) vmbus_close(dev->channel);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1750) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1751) unregister_memory_notifier(&hv_memory_nb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1752) restore_online_page_callback(&hv_online_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1753) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1754) spin_lock_irqsave(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1755) list_for_each_entry_safe(has, tmp, &dm->ha_region_list, list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1756) list_for_each_entry_safe(gap, tmp_gap, &has->gap_list, list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1757) list_del(&gap->list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1758) kfree(gap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1759) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1760) list_del(&has->list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1761) kfree(has);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1762) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1763) spin_unlock_irqrestore(&dm_device.ha_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1764)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1765) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1766) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1767)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1768) static int balloon_suspend(struct hv_device *hv_dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1769) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1770) struct hv_dynmem_device *dm = hv_get_drvdata(hv_dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1771)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1772) tasklet_disable(&hv_dev->channel->callback_event);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1773)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1774) cancel_work_sync(&dm->balloon_wrk.wrk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1775) cancel_work_sync(&dm->ha_wrk.wrk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1776)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1777) if (dm->thread) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1778) kthread_stop(dm->thread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1779) dm->thread = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1780) vmbus_close(hv_dev->channel);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1781) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1782)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1783) tasklet_enable(&hv_dev->channel->callback_event);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1784)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1785) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1786)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1787) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1788)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1789) static int balloon_resume(struct hv_device *dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1790) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1791) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1792)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1793) dm_device.state = DM_INITIALIZING;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1794)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1795) ret = balloon_connect_vsp(dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1796)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1797) if (ret != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1798) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1799)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1800) dm_device.thread =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1801) kthread_run(dm_thread_func, &dm_device, "hv_balloon");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1802) if (IS_ERR(dm_device.thread)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1803) ret = PTR_ERR(dm_device.thread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1804) dm_device.thread = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1805) goto close_channel;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1806) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1807)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1808) dm_device.state = DM_INITIALIZED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1809) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1810) close_channel:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1811) vmbus_close(dev->channel);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1812) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1813) dm_device.state = DM_INIT_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1814) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1815) unregister_memory_notifier(&hv_memory_nb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1816) restore_online_page_callback(&hv_online_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1817) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1818) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1819) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1820)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1821) static const struct hv_vmbus_device_id id_table[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1822) /* Dynamic Memory Class ID */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1823) /* 525074DC-8985-46e2-8057-A307DC18A502 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1824) { HV_DM_GUID, },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1825) { },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1826) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1827)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1828) MODULE_DEVICE_TABLE(vmbus, id_table);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1829)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1830) static struct hv_driver balloon_drv = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1831) .name = "hv_balloon",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1832) .id_table = id_table,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1833) .probe = balloon_probe,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1834) .remove = balloon_remove,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1835) .suspend = balloon_suspend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1836) .resume = balloon_resume,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1837) .driver = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1838) .probe_type = PROBE_PREFER_ASYNCHRONOUS,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1839) },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1840) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1841)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1842) static int __init init_balloon_drv(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1843) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1844)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1845) return vmbus_driver_register(&balloon_drv);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1846) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1847)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1848) module_init(init_balloon_drv);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1849)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1850) MODULE_DESCRIPTION("Hyper-V Balloon");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1851) MODULE_LICENSE("GPL");
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags