linux/drivers/hv/hv_snapshot.c

459 lines
12 KiB
C
Raw Permalink Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* An implementation of host initiated guest snapshot.
*
* Copyright (C) 2013, Microsoft, Inc.
* Author : K. Y. Srinivasan <kys@microsoft.com>
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/net.h>
#include <linux/nls.h>
#include <linux/connector.h>
#include <linux/workqueue.h>
#include <linux/hyperv.h>
#include <asm/hyperv-tlfs.h>
#include "hyperv_vmbus.h"
#include "hv_utils_transport.h"
#define VSS_MAJOR 5
#define VSS_MINOR 0
#define VSS_VERSION (VSS_MAJOR << 16 | VSS_MINOR)
#define VSS_VER_COUNT 1
static const int vss_versions[] = {
VSS_VERSION
};
#define FW_VER_COUNT 1
static const int fw_versions[] = {
UTIL_FW_VERSION
};
/* See comment with struct hv_vss_msg regarding the max VMbus packet size */
#define VSS_MAX_PKT_SIZE (HV_HYP_PAGE_SIZE * 2)
/*
* Timeout values are based on expecations from host
*/
#define VSS_FREEZE_TIMEOUT (15 * 60)
/*
* Global state maintained for transaction that is being processed. For a class
* of integration services, including the "VSS service", the specified protocol
* is a "request/response" protocol which means that there can only be single
* outstanding transaction from the host at any given point in time. We use
* this to simplify memory management in this driver - we cache and process
* only one message at a time.
*
* While the request/response protocol is guaranteed by the host, we further
* ensure this by serializing packet processing in this driver - we do not
* read additional packets from the VMBUs until the current packet is fully
* handled.
*/
static struct {
int state; /* hvutil_device_state */
int recv_len; /* number of bytes received. */
struct vmbus_channel *recv_channel; /* chn we got the request */
u64 recv_req_id; /* request ID. */
struct hv_vss_msg *msg; /* current message */
} vss_transaction;
static void vss_respond_to_host(int error);
/*
* This state maintains the version number registered by the daemon.
*/
static int dm_reg_value;
static const char vss_devname[] = "vmbus/hv_vss";
static __u8 *recv_buffer;
static struct hvutil_transport *hvt;
static void vss_timeout_func(struct work_struct *dummy);
static void vss_handle_request(struct work_struct *dummy);
static DECLARE_DELAYED_WORK(vss_timeout_work, vss_timeout_func);
static DECLARE_WORK(vss_handle_request_work, vss_handle_request);
static void vss_poll_wrapper(void *channel)
{
/* Transaction is finished, reset the state here to avoid races. */
vss_transaction.state = HVUTIL_READY;
tasklet_schedule(&((struct vmbus_channel *)channel)->callback_event);
}
/*
* Callback when data is received from user mode.
*/
static void vss_timeout_func(struct work_struct *dummy)
{
/*
* Timeout waiting for userspace component to reply happened.
*/
pr_warn("VSS: timeout waiting for daemon to reply\n");
vss_respond_to_host(HV_E_FAIL);
hv_poll_channel(vss_transaction.recv_channel, vss_poll_wrapper);
}
Drivers: hv: utils: fix a race on userspace daemons registration Background: userspace daemons registration protocol for Hyper-V utilities drivers has two steps: 1) daemon writes its own version to kernel 2) kernel reads it and replies with module version at this point we consider the handshake procedure being completed and we do hv_poll_channel() transitioning the utility device to HVUTIL_READY state. At this point we're ready to handle messages from kernel. When hvutil_transport is in HVUTIL_TRANSPORT_CHARDEV mode we have a single buffer for outgoing message. hvutil_transport_send() puts to this buffer and till the buffer is cleared with hvt_op_read() returns -EFAULT to all consequent calls. Host<->guest protocol guarantees there is no more than one request at a time and we will not get new requests till we reply to the previous one so this single message buffer is enough. Now to the race. When we finish negotiation procedure and send kernel module version to userspace with hvutil_transport_send() it goes into the above mentioned buffer and if the daemon is slow enough to read it from there we can get a collision when a request from the host comes, we won't be able to put anything to the buffer so the request will be lost. To solve the issue we need to know when the negotiation is really done (when the version message is read by the daemon) and transition to HVUTIL_READY state after this happens. Implement a callback on read to support this. Old style netlink communication is not affected by the change, we don't really know when these messages are delivered but we don't have a single message buffer there. Reported-by: Barry Davis <barry_davis@stormagic.com> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2016-06-10 08:08:57 +08:00
static void vss_register_done(void)
{
hv_poll_channel(vss_transaction.recv_channel, vss_poll_wrapper);
pr_debug("VSS: userspace daemon registered\n");
}
static int vss_handle_handshake(struct hv_vss_msg *vss_msg)
{
u32 our_ver = VSS_OP_REGISTER1;
switch (vss_msg->vss_hdr.operation) {
case VSS_OP_REGISTER:
/* Daemon doesn't expect us to reply */
dm_reg_value = VSS_OP_REGISTER;
break;
case VSS_OP_REGISTER1:
Drivers: hv: utils: fix a race on userspace daemons registration Background: userspace daemons registration protocol for Hyper-V utilities drivers has two steps: 1) daemon writes its own version to kernel 2) kernel reads it and replies with module version at this point we consider the handshake procedure being completed and we do hv_poll_channel() transitioning the utility device to HVUTIL_READY state. At this point we're ready to handle messages from kernel. When hvutil_transport is in HVUTIL_TRANSPORT_CHARDEV mode we have a single buffer for outgoing message. hvutil_transport_send() puts to this buffer and till the buffer is cleared with hvt_op_read() returns -EFAULT to all consequent calls. Host<->guest protocol guarantees there is no more than one request at a time and we will not get new requests till we reply to the previous one so this single message buffer is enough. Now to the race. When we finish negotiation procedure and send kernel module version to userspace with hvutil_transport_send() it goes into the above mentioned buffer and if the daemon is slow enough to read it from there we can get a collision when a request from the host comes, we won't be able to put anything to the buffer so the request will be lost. To solve the issue we need to know when the negotiation is really done (when the version message is read by the daemon) and transition to HVUTIL_READY state after this happens. Implement a callback on read to support this. Old style netlink communication is not affected by the change, we don't really know when these messages are delivered but we don't have a single message buffer there. Reported-by: Barry Davis <barry_davis@stormagic.com> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2016-06-10 08:08:57 +08:00
/* Daemon expects us to reply with our own version */
if (hvutil_transport_send(hvt, &our_ver, sizeof(our_ver),
vss_register_done))
return -EFAULT;
dm_reg_value = VSS_OP_REGISTER1;
break;
default:
return -EINVAL;
}
pr_info("VSS: userspace daemon ver. %d connected\n", dm_reg_value);
return 0;
}
static int vss_on_msg(void *msg, int len)
{
struct hv_vss_msg *vss_msg = (struct hv_vss_msg *)msg;
if (len != sizeof(*vss_msg)) {
pr_debug("VSS: Message size does not match length\n");
return -EINVAL;
}
if (vss_msg->vss_hdr.operation == VSS_OP_REGISTER ||
vss_msg->vss_hdr.operation == VSS_OP_REGISTER1) {
/*
* Don't process registration messages if we're in the middle
* of a transaction processing.
*/
if (vss_transaction.state > HVUTIL_READY) {
pr_debug("VSS: Got unexpected registration request\n");
return -EINVAL;
}
return vss_handle_handshake(vss_msg);
} else if (vss_transaction.state == HVUTIL_USERSPACE_REQ) {
vss_transaction.state = HVUTIL_USERSPACE_RECV;
if (vss_msg->vss_hdr.operation == VSS_OP_HOT_BACKUP)
vss_transaction.msg->vss_cf.flags =
VSS_HBU_NO_AUTO_RECOVERY;
if (cancel_delayed_work_sync(&vss_timeout_work)) {
vss_respond_to_host(vss_msg->error);
/* Transaction is finished, reset the state. */
hv_poll_channel(vss_transaction.recv_channel,
vss_poll_wrapper);
}
} else {
/* This is a spurious call! */
pr_debug("VSS: Transaction not active\n");
return -EINVAL;
}
return 0;
}
static void vss_send_op(void)
{
int op = vss_transaction.msg->vss_hdr.operation;
int rc;
struct hv_vss_msg *vss_msg;
/* The transaction state is wrong. */
if (vss_transaction.state != HVUTIL_HOSTMSG_RECEIVED) {
pr_debug("VSS: Unexpected attempt to send to daemon\n");
return;
}
vss_msg = kzalloc(sizeof(*vss_msg), GFP_KERNEL);
if (!vss_msg)
return;
vss_msg->vss_hdr.operation = op;
vss_transaction.state = HVUTIL_USERSPACE_REQ;
schedule_delayed_work(&vss_timeout_work, op == VSS_OP_FREEZE ?
VSS_FREEZE_TIMEOUT * HZ : HV_UTIL_TIMEOUT * HZ);
Drivers: hv: utils: fix a race on userspace daemons registration Background: userspace daemons registration protocol for Hyper-V utilities drivers has two steps: 1) daemon writes its own version to kernel 2) kernel reads it and replies with module version at this point we consider the handshake procedure being completed and we do hv_poll_channel() transitioning the utility device to HVUTIL_READY state. At this point we're ready to handle messages from kernel. When hvutil_transport is in HVUTIL_TRANSPORT_CHARDEV mode we have a single buffer for outgoing message. hvutil_transport_send() puts to this buffer and till the buffer is cleared with hvt_op_read() returns -EFAULT to all consequent calls. Host<->guest protocol guarantees there is no more than one request at a time and we will not get new requests till we reply to the previous one so this single message buffer is enough. Now to the race. When we finish negotiation procedure and send kernel module version to userspace with hvutil_transport_send() it goes into the above mentioned buffer and if the daemon is slow enough to read it from there we can get a collision when a request from the host comes, we won't be able to put anything to the buffer so the request will be lost. To solve the issue we need to know when the negotiation is really done (when the version message is read by the daemon) and transition to HVUTIL_READY state after this happens. Implement a callback on read to support this. Old style netlink communication is not affected by the change, we don't really know when these messages are delivered but we don't have a single message buffer there. Reported-by: Barry Davis <barry_davis@stormagic.com> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2016-06-10 08:08:57 +08:00
rc = hvutil_transport_send(hvt, vss_msg, sizeof(*vss_msg), NULL);
if (rc) {
pr_warn("VSS: failed to communicate to the daemon: %d\n", rc);
if (cancel_delayed_work_sync(&vss_timeout_work)) {
vss_respond_to_host(HV_E_FAIL);
vss_transaction.state = HVUTIL_READY;
}
}
kfree(vss_msg);
}
static void vss_handle_request(struct work_struct *dummy)
{
switch (vss_transaction.msg->vss_hdr.operation) {
/*
* Initiate a "freeze/thaw" operation in the guest.
* We respond to the host once the operation is complete.
*
* We send the message to the user space daemon and the operation is
* performed in the daemon.
*/
case VSS_OP_THAW:
case VSS_OP_FREEZE:
case VSS_OP_HOT_BACKUP:
if (vss_transaction.state < HVUTIL_READY) {
/* Userspace is not registered yet */
pr_debug("VSS: Not ready for request.\n");
vss_respond_to_host(HV_E_FAIL);
return;
}
pr_debug("VSS: Received request for op code: %d\n",
vss_transaction.msg->vss_hdr.operation);
vss_transaction.state = HVUTIL_HOSTMSG_RECEIVED;
vss_send_op();
return;
case VSS_OP_GET_DM_INFO:
vss_transaction.msg->dm_info.flags = 0;
break;
default:
break;
}
vss_respond_to_host(0);
hv_poll_channel(vss_transaction.recv_channel, vss_poll_wrapper);
}
/*
* Send a response back to the host.
*/
static void
vss_respond_to_host(int error)
{
struct icmsg_hdr *icmsghdrp;
u32 buf_len;
struct vmbus_channel *channel;
u64 req_id;
/*
* Copy the global state for completing the transaction. Note that
* only one transaction can be active at a time.
*/
buf_len = vss_transaction.recv_len;
channel = vss_transaction.recv_channel;
req_id = vss_transaction.recv_req_id;
icmsghdrp = (struct icmsg_hdr *)
&recv_buffer[sizeof(struct vmbuspipe_hdr)];
if (channel->onchannel_callback == NULL)
/*
* We have raced with util driver being unloaded;
* silently return.
*/
return;
icmsghdrp->status = error;
icmsghdrp->icflags = ICMSGHDRFLAG_TRANSACTION | ICMSGHDRFLAG_RESPONSE;
vmbus_sendpacket(channel, recv_buffer, buf_len, req_id,
VM_PKT_DATA_INBAND, 0);
}
/*
* This callback is invoked when we get a VSS message from the host.
* The host ensures that only one VSS transaction can be active at a time.
*/
void hv_vss_onchannelcallback(void *context)
{
struct vmbus_channel *channel = context;
u32 recvlen;
u64 requestid;
struct hv_vss_msg *vss_msg;
int vss_srv_version;
struct icmsg_hdr *icmsghdrp;
if (vss_transaction.state > HVUTIL_READY)
return;
if (vmbus_recvpacket(channel, recv_buffer, VSS_MAX_PKT_SIZE, &recvlen, &requestid)) {
pr_err_ratelimited("VSS request received. Could not read into recv buf\n");
return;
}
if (!recvlen)
return;
/* Ensure recvlen is big enough to read header data */
if (recvlen < ICMSG_HDR) {
pr_err_ratelimited("VSS request received. Packet length too small: %d\n",
recvlen);
return;
}
icmsghdrp = (struct icmsg_hdr *)&recv_buffer[sizeof(struct vmbuspipe_hdr)];
if (icmsghdrp->icmsgtype == ICMSGTYPE_NEGOTIATE) {
if (vmbus_prep_negotiate_resp(icmsghdrp,
recv_buffer, recvlen,
fw_versions, FW_VER_COUNT,
vss_versions, VSS_VER_COUNT,
NULL, &vss_srv_version)) {
pr_info("VSS IC version %d.%d\n",
vss_srv_version >> 16,
vss_srv_version & 0xFFFF);
}
} else if (icmsghdrp->icmsgtype == ICMSGTYPE_VSS) {
/* Ensure recvlen is big enough to contain hv_vss_msg */
if (recvlen < ICMSG_HDR + sizeof(struct hv_vss_msg)) {
pr_err_ratelimited("Invalid VSS msg. Packet length too small: %u\n",
recvlen);
return;
}
vss_msg = (struct hv_vss_msg *)&recv_buffer[ICMSG_HDR];
/*
* Stash away this global state for completing the
* transaction; note transactions are serialized.
*/
vss_transaction.recv_len = recvlen;
vss_transaction.recv_req_id = requestid;
vss_transaction.msg = (struct hv_vss_msg *)vss_msg;
schedule_work(&vss_handle_request_work);
return;
} else {
pr_err_ratelimited("VSS request received. Invalid msg type: %d\n",
icmsghdrp->icmsgtype);
return;
}
icmsghdrp->icflags = ICMSGHDRFLAG_TRANSACTION |
ICMSGHDRFLAG_RESPONSE;
vmbus_sendpacket(channel, recv_buffer, recvlen, requestid,
VM_PKT_DATA_INBAND, 0);
}
static void vss_on_reset(void)
{
if (cancel_delayed_work_sync(&vss_timeout_work))
vss_respond_to_host(HV_E_FAIL);
vss_transaction.state = HVUTIL_DEVICE_INIT;
}
int
hv_vss_init(struct hv_util_service *srv)
{
if (vmbus_proto_version < VERSION_WIN8_1) {
pr_warn("Integration service 'Backup (volume snapshot)'"
" not supported on this host version.\n");
return -ENOTSUPP;
}
recv_buffer = srv->recv_buffer;
vss_transaction.recv_channel = srv->channel;
vss_transaction.recv_channel->max_pkt_size = VSS_MAX_PKT_SIZE;
/*
* When this driver loads, the user level daemon that
* processes the host requests may not yet be running.
* Defer processing channel callbacks until the daemon
* has registered.
*/
vss_transaction.state = HVUTIL_DEVICE_INIT;
hvt = hvutil_transport_init(vss_devname, CN_VSS_IDX, CN_VSS_VAL,
vss_on_msg, vss_on_reset);
if (!hvt) {
pr_warn("VSS: Failed to initialize transport\n");
return -EFAULT;
}
return 0;
}
static void hv_vss_cancel_work(void)
{
cancel_delayed_work_sync(&vss_timeout_work);
cancel_work_sync(&vss_handle_request_work);
}
int hv_vss_pre_suspend(void)
{
struct vmbus_channel *channel = vss_transaction.recv_channel;
struct hv_vss_msg *vss_msg;
/*
* Fake a THAW message for the user space daemon in case the daemon
* has frozen the file systems. It doesn't matter if there is already
* a message pending to be delivered to the user space since we force
* vss_transaction.state to be HVUTIL_READY, so the user space daemon's
* write() will fail with EINVAL (see vss_on_msg()), and the daemon
* will reset the device by closing and re-opening it.
*/
vss_msg = kzalloc(sizeof(*vss_msg), GFP_KERNEL);
if (!vss_msg)
return -ENOMEM;
tasklet_disable(&channel->callback_event);
vss_msg->vss_hdr.operation = VSS_OP_THAW;
/* Cancel any possible pending work. */
hv_vss_cancel_work();
/* We don't care about the return value. */
hvutil_transport_send(hvt, vss_msg, sizeof(*vss_msg), NULL);
kfree(vss_msg);
vss_transaction.state = HVUTIL_READY;
/* tasklet_enable() will be called in hv_vss_pre_resume(). */
return 0;
}
int hv_vss_pre_resume(void)
{
struct vmbus_channel *channel = vss_transaction.recv_channel;
tasklet_enable(&channel->callback_event);
return 0;
}
void hv_vss_deinit(void)
{
vss_transaction.state = HVUTIL_DEVICE_DYING;
hv_vss_cancel_work();
hvutil_transport_destroy(hvt);
}