From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: zhanghailiang <zhang.zhanghailiang@huawei.com>
Cc: lizhijian@cn.fujitsu.com, quintela@redhat.com,
yunhong.jiang@intel.com, eddie.dong@intel.com,
peter.huangpeng@huawei.com, qemu-devel@nongnu.org,
arei.gonglei@huawei.com, amit.shah@redhat.com,
Yang Hongyang <yanghy@cn.fujitsu.com>
Subject: Re: [Qemu-devel] [PATCH COLO-Frame v8 08/34] COLO: Implement colo checkpoint protocol
Date: Thu, 27 Aug 2015 11:40:52 +0100 [thread overview]
Message-ID: <20150827104051.GC2247@work-vm> (raw)
In-Reply-To: <1438159544-6224-9-git-send-email-zhang.zhanghailiang@huawei.com>
* zhanghailiang (zhang.zhanghailiang@huawei.com) wrote:
> We need communications protocol of user-defined to control the checkpoint
> process.
>
> The new checkpoint request is started by Primary VM, and the interactive process
> like below:
> Checkpoint synchronizing points,
>
> Primary Secondary
> NEW @
> Suspend
> SUSPENDED @
> Suspend&Save state
> SEND @
> Send state Receive state
> RECEIVED @
> Flush network Load state
> LOADED @
> Resume Resume
>
> Start Comparing
> NOTE:
> 1) '@' who sends the message
> 2) Every sync-point is synchronized by two sides with only
> one handshake(single direction) for low-latency.
> If more strict synchronization is required, a opposite direction
> sync-point should be added.
> 3) Since sync-points are single direction, the remote side may
> go forward a lot when this side just receives the sync-point.
>
> Signed-off-by: Yang Hongyang <yanghy@cn.fujitsu.com>
> Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
> Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
> Signed-off-by: Gonglei <arei.gonglei@huawei.com>
> ---
> migration/colo.c | 248 ++++++++++++++++++++++++++++++++++++++++++++++++++++++-
> trace-events | 3 +-
> 2 files changed, 248 insertions(+), 3 deletions(-)
>
> diff --git a/migration/colo.c b/migration/colo.c
> index 364e0dd..4ba6f65 100644
> --- a/migration/colo.c
> +++ b/migration/colo.c
> @@ -14,6 +14,55 @@
> #include "migration/colo.h"
> #include "trace.h"
> #include "qemu/error-report.h"
> +#include "qemu/sockets.h"
> +
> +/* Fix me: Convert to use QAPI */
> +typedef enum COLOCommand {
> + COLO_CHECPOINT_READY = 0x46,
Typo: CHEC*K*POINT
> + /*
> + * Checkpoint synchronizing points.
> + *
> + * Primary Secondary
> + * NEW @
> + * Suspend
> + * SUSPENDED @
> + * Suspend&Save state
> + * SEND @
> + * Send state Receive state
> + * RECEIVED @
> + * Flush network Load state
> + * LOADED @
> + * Resume Resume
> + *
> + * Start Comparing
> + * NOTE:
> + * 1) '@' who sends the message
> + * 2) Every sync-point is synchronized by two sides with only
> + * one handshake(single direction) for low-latency.
> + * If more strict synchronization is required, a opposite direction
> + * sync-point should be added.
> + * 3) Since sync-points are single direction, the remote side may
> + * go forward a lot when this side just receives the sync-point.
> + */
> + COLO_CHECKPOINT_NEW,
> + COLO_CHECKPOINT_SUSPENDED,
> + COLO_CHECKPOINT_SEND,
> + COLO_CHECKPOINT_RECEIVED,
> + COLO_CHECKPOINT_LOADED,
> +
> + COLO_CHECKPOINT_MAX
> +} COLOCommand;
> +
> +const char * const COLOCommand_lookup[] = {
Unneeded ' ' after the *.
> + [COLO_CHECPOINT_READY] = "checkpoint-ready",
> + [COLO_CHECKPOINT_NEW] = "checkpoint-new",
> + [COLO_CHECKPOINT_SUSPENDED] = "checkpoint-suspend",
> + [COLO_CHECKPOINT_SEND] = "checheckpoint-send",
> + [COLO_CHECKPOINT_RECEIVED] = "checkpoint-received",
> + [COLO_CHECKPOINT_LOADED] = "checkpoint-loaded",
> + [COLO_CHECKPOINT_MAX] = NULL,
> +};
>
> static QEMUBH *colo_bh;
>
> @@ -36,20 +85,137 @@ bool migration_incoming_in_colo_state(void)
> return (mis && (mis->state == MIGRATION_STATUS_COLO));
> }
>
> +/* colo checkpoint control helper */
> +static int colo_ctl_put(QEMUFile *f, uint64_t request)
> +{
> + int ret = 0;
> +
> + qemu_put_be64(f, request);
> + qemu_fflush(f);
> +
> + ret = qemu_file_get_error(f);
> + if (request < COLO_CHECKPOINT_MAX) {
> + trace_colo_ctl_put(COLOCommand_lookup[request]);
> + }
> + return ret;
> +}
> +
> +static int colo_ctl_get_value(QEMUFile *f, uint64_t *value)
> +{
> + int ret = 0;
> + uint64_t temp;
> +
> + temp = qemu_get_be64(f);
> +
> + ret = qemu_file_get_error(f);
> + if (ret < 0) {
> + return -1;
Why not just return ret rather than -1?
> + }
> +
> + *value = temp;
> + return 0;
> +}
> +
> +static int colo_ctl_get(QEMUFile *f, uint64_t require)
> +{
> + int ret;
> + uint64_t value;
> +
> + ret = colo_ctl_get_value(f, &value);
> + if (ret < 0) {
> + return ret;
> + }
> +
> + if (value != require) {
> + error_report("unexpected state! expected: %"PRIu64
> + ", received: %"PRIu64, require, value);
> + exit(1);
> + }
Do you really want to exit? If you change that to something like
return -EINVAL;
then if it happens on the primary side, the primary side would
survive.
> +
> + trace_colo_ctl_get(COLOCommand_lookup[require]);
> + return ret;
That's always 0 ?
> +}
> +
> +static int colo_do_checkpoint_transaction(MigrationState *s, QEMUFile *control)
> +{
> + int ret;
> +
> + ret = colo_ctl_put(s->file, COLO_CHECKPOINT_NEW);
> + if (ret < 0) {
> + goto out;
> + }
> +
> + ret = colo_ctl_get(control, COLO_CHECKPOINT_SUSPENDED);
> + if (ret < 0) {
> + goto out;
> + }
> +
> + /* TODO: suspend and save vm state to colo buffer */
> +
> + ret = colo_ctl_put(s->file, COLO_CHECKPOINT_SEND);
> + if (ret < 0) {
> + goto out;
> + }
> +
> + /* TODO: send vmstate to Secondary */
> +
> + ret = colo_ctl_get(control, COLO_CHECKPOINT_RECEIVED);
> + if (ret < 0) {
> + goto out;
> + }
> +
> + ret = colo_ctl_get(control, COLO_CHECKPOINT_LOADED);
> + if (ret < 0) {
> + goto out;
> + }
> +
> + /* TODO: resume Primary */
> +
> +out:
> + return ret;
> +}
> +
> static void *colo_thread(void *opaque)
> {
> MigrationState *s = opaque;
> + QEMUFile *colo_control = NULL;
> + int ret;
> +
> + colo_control = qemu_fopen_socket(qemu_get_fd(s->file), "rb");
> + if (!colo_control) {
> + error_report("Open colo_control failed!");
> + goto out;
> + }
> +
> + /*
> + * Wait for Secondary finish loading vm states and enter COLO
> + * restore.
> + */
> + ret = colo_ctl_get(colo_control, COLO_CHECPOINT_READY);
> + if (ret < 0) {
> + goto out;
> + }
>
> qemu_mutex_lock_iothread();
> vm_start();
> qemu_mutex_unlock_iothread();
> trace_colo_vm_state_change("stop", "run");
>
> - /*TODO: COLO checkpoint savevm loop*/
> + while (s->state == MIGRATION_STATUS_COLO) {
> + /* start a colo checkpoint */
> + if (colo_do_checkpoint_transaction(s, colo_control)) {
> + goto out;
> + }
> + }
>
> +out:
> migrate_set_state(&s->state, MIGRATION_STATUS_COLO,
> MIGRATION_STATUS_COMPLETED);
>
> + if (colo_control) {
> + qemu_fclose(colo_control);
> + }
> +
> qemu_mutex_lock_iothread();
> qemu_bh_schedule(s->cleanup_bh);
> qemu_mutex_unlock_iothread();
> @@ -83,15 +249,93 @@ void colo_init_checkpointer(MigrationState *s)
> qemu_bh_schedule(colo_bh);
> }
>
> +/*
> + * return:
> + * 0: start a checkpoint
> + * -1: some error happened, exit colo restore
> + */
> +static int colo_wait_handle_cmd(QEMUFile *f, int *checkpoint_request)
> +{
> + int ret;
> + uint64_t cmd;
> +
> + ret = colo_ctl_get_value(f, &cmd);
> + if (ret < 0) {
> + return -1;
return ret ?
> + }
> +
> + switch (cmd) {
> + case COLO_CHECKPOINT_NEW:
> + *checkpoint_request = 1;
> + return 0;
> + default:
> + return -1;
> + }
> +}
> +
> void *colo_process_incoming_checkpoints(void *opaque)
> {
> MigrationIncomingState *mis = opaque;
> + QEMUFile *f = mis->file;
> + int fd = qemu_get_fd(f);
> + QEMUFile *ctl = NULL;
> + int ret;
>
> migrate_set_state(&mis->state, MIGRATION_STATUS_ACTIVE,
> MIGRATION_STATUS_COLO);
>
> - /* TODO: COLO checkpoint restore loop */
> + ctl = qemu_fopen_socket(fd, "wb");
> + if (!ctl) {
> + error_report("Can't open incoming channel!");
> + goto out;
> + }
> + ret = colo_ctl_put(ctl, COLO_CHECPOINT_READY);
> + if (ret < 0) {
> + goto out;
> + }
> + /* TODO: in COLO mode, Secondary is runing, so start the vm */
> + while (mis->state == MIGRATION_STATUS_COLO) {
> + int request = 0;
> + int ret = colo_wait_handle_cmd(f, &request);
> +
> + if (ret < 0) {
> + break;
> + } else {
> + if (!request) {
> + continue;
> + }
> + }
>
> + /* TODO: suspend guest */
> + ret = colo_ctl_put(ctl, COLO_CHECKPOINT_SUSPENDED);
> + if (ret < 0) {
> + goto out;
> + }
> +
> + ret = colo_ctl_get(f, COLO_CHECKPOINT_SEND);
> + if (ret < 0) {
> + goto out;
> + }
> +
> + /* TODO: read migration data into colo buffer */
> +
> + ret = colo_ctl_put(ctl, COLO_CHECKPOINT_RECEIVED);
> + if (ret < 0) {
> + goto out;
> + }
> +
> + /* TODO: load vm state */
> +
> + ret = colo_ctl_put(ctl, COLO_CHECKPOINT_LOADED);
> + if (ret < 0) {
> + goto out;
> + }
> +}
> +
> +out:
> + if (ctl) {
> + qemu_fclose(ctl);
> + }
> migration_incoming_exit_colo();
>
> return NULL;
> diff --git a/trace-events b/trace-events
> index 025d71c..4487633 100644
> --- a/trace-events
> +++ b/trace-events
> @@ -1473,7 +1473,8 @@ rdma_start_outgoing_migration_after_rdma_source_init(void) ""
>
> # migration/colo.c
> colo_vm_state_change(const char *old, const char *new) "Change '%s' => '%s'"
> -colo_receive_message(const char *msg) "Receive '%s'"
> +colo_ctl_put(const char *msg) "Send '%s'"
> +colo_ctl_get(const char *msg) "Receive '%s'"
>
> # kvm-all.c
> kvm_ioctl(int type, void *arg) "type 0x%x, arg %p"
> --
> 1.8.3.1
>
>
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
next prev parent reply other threads:[~2015-08-27 10:41 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-29 8:45 [Qemu-devel] [PATCH COLO-Frame v8 00/34] COarse-grain LOck-stepping(COLO) Virtual Machines for Non-stop Service (FT) zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 01/34] configure: Add parameter for configure to enable/disable COLO support zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 02/34] migration: Introduce capability 'colo' to migration zhanghailiang
2015-08-28 21:54 ` Eric Blake
2015-08-31 2:18 ` zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 03/34] COLO: migrate colo related info to slave zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 04/34] colo-comm/migration: skip colo info section for special cases zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 05/34] migration: Add state records for migration incoming zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 06/34] migration: Integrate COLO checkpoint process into migration zhanghailiang
2015-08-28 21:55 ` Eric Blake
2015-08-31 5:06 ` zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 07/34] migration: Integrate COLO checkpoint process into loadvm zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 08/34] COLO: Implement colo checkpoint protocol zhanghailiang
2015-08-27 10:40 ` Dr. David Alan Gilbert [this message]
2015-08-27 11:27 ` zhanghailiang
2015-08-27 12:43 ` Dr. David Alan Gilbert
2015-08-28 7:53 ` zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 09/34] COLO: Add a new RunState RUN_STATE_COLO zhanghailiang
2015-08-28 21:58 ` Eric Blake
2015-08-31 6:09 ` zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 10/34] QEMUSizedBuffer: Introduce two help functions for qsb zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 11/34] COLO: Save VM state to slave when do checkpoint zhanghailiang
2015-08-27 12:06 ` Dr. David Alan Gilbert
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 12/34] COLO RAM: Load PVM's dirty page into SVM's RAM cache temporarily zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 13/34] COLO VMstate: Load VM state into qsb before restore it zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 14/34] arch_init: Start to trace dirty pages of SVM zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 15/34] COLO RAM: Flush cached RAM into SVM's memory zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 16/34] COLO failover: Introduce a new command to trigger a failover zhanghailiang
2015-08-28 22:06 ` Eric Blake
2015-09-01 2:47 ` zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 17/34] COLO failover: Introduce state to record failover process zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 18/34] COLO failover: Implement COLO primary/secondary vm failover work zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 19/34] qmp event: Add event notification for COLO error zhanghailiang
2015-08-28 22:13 ` Eric Blake
2015-08-31 9:27 ` zhanghailiang
2015-08-31 15:07 ` Eric Blake
2015-09-01 1:08 ` zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 20/34] COLO failover: Don't do failover during loading VM's state zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 21/34] COLO: Add new command parameter 'forward_nic' 'colo_script' for net zhanghailiang
2015-08-28 22:24 ` Eric Blake
2015-08-31 10:57 ` zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 22/34] COLO NIC: Init/remove colo nic devices when add/cleanup tap devices zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 23/34] tap: Make launch_script() public zhanghailiang
2015-07-29 8:57 ` Jason Wang
2015-07-29 9:17 ` zhanghailiang
2015-07-29 9:24 ` Jason Wang
2015-07-29 9:43 ` zhanghailiang
2015-07-30 3:32 ` Jason Wang
2015-07-30 4:02 ` zhanghailiang
2015-07-29 9:19 ` Daniel P. Berrange
2015-07-29 9:37 ` Dr. David Alan Gilbert
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 24/34] COLO NIC: Implement colo nic device interface configure() zhanghailiang
2015-08-05 10:42 ` Dr. David Alan Gilbert
2015-08-05 11:54 ` Li Zhijian
2015-08-20 10:34 ` Dr. David Alan Gilbert
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 25/34] colo-nic: Handle secondary VM's original net device configure zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 26/34] COLO NIC: Implement colo nic init/destroy function zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 27/34] COLO NIC: Some init work related with proxy module zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 28/34] COLO: Handle nfnetlink message from " zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 29/34] COLO: Do checkpoint according to the result of packets comparation zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 30/34] COLO: Improve checkpoint efficiency by do additional periodic checkpoint zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 31/34] COLO: Add colo-set-checkpoint-period command zhanghailiang
2015-08-28 22:26 ` Eric Blake
2015-08-31 12:00 ` zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 32/34] COLO NIC: Implement NIC checkpoint and failover zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 33/34] COLO: Implement shutdown checkpoint zhanghailiang
2015-07-29 8:45 ` [Qemu-devel] [PATCH COLO-Frame v8 34/34] COLO: Add block replication into colo process zhanghailiang
2015-08-05 11:24 ` [Qemu-devel] [PATCH COLO-Frame v8 00/34] COarse-grain LOck-stepping(COLO) Virtual Machines for Non-stop Service (FT) Dr. David Alan Gilbert
2015-08-06 10:25 ` zhanghailiang
2015-08-12 8:20 ` zhanghailiang
2015-08-24 14:38 ` Dr. David Alan Gilbert
2015-08-25 7:03 ` zhanghailiang
2015-08-26 16:49 ` Dr. David Alan Gilbert
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20150827104051.GC2247@work-vm \
--to=dgilbert@redhat.com \
--cc=amit.shah@redhat.com \
--cc=arei.gonglei@huawei.com \
--cc=eddie.dong@intel.com \
--cc=lizhijian@cn.fujitsu.com \
--cc=peter.huangpeng@huawei.com \
--cc=qemu-devel@nongnu.org \
--cc=quintela@redhat.com \
--cc=yanghy@cn.fujitsu.com \
--cc=yunhong.jiang@intel.com \
--cc=zhang.zhanghailiang@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).