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 initial work 'checkpoint-ready' <------------------------------ @ 'checkpoint-request' @ -----------------------------> Suspend (Only in hybrid mode) 'checkpoint-reply' <------------------------------ @ Suspend&Save state 'vmstate-send' @ -----------------------------> Send state Receive state 'vmstate-received' <------------------------------ @ Release packets Load state 'vmstate-load' <------------------------------ @ Resume Resume (Only in hybrid mode) Start Comparing (Only in hybrid mode) 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. 4) For now, we only support 'periodic' checkpoint, for which the Secondary VM is not running, later we will support 'hybrid' mode. Signed-off-by: zhanghailiang <zhang.zhanghaili...@huawei.com> Signed-off-by: Li Zhijian <lizhij...@cn.fujitsu.com> Signed-off-by: Gonglei <arei.gong...@huawei.com> Cc: Eric Blake <ebl...@redhat.com> --- v12: - Rename colo_ctl_put() to colo_put_cmd() - Rename colo_ctl_get() to colo_get_check_cmd() and drop the third parameter - Rename colo_ctl_get_cmd() to colo_get_cmd() - Remove useless 'invalid' member for COLOcommand enum. v11: - Add missing 'checkpoint-ready' communication in comment. - Use parameter to return 'value' for colo_ctl_get() (Dave's suggestion) - Fix trace for colo_ctl_get() to trace command and value both v10: - Rename enum COLOCmd to COLOCommand (Eric's suggestion). - Remove unused 'ram-steal' Signed-off-by: zhanghailiang <zhang.zhanghaili...@huawei.com> --- migration/colo.c | 183 ++++++++++++++++++++++++++++++++++++++++++++++++++++++- qapi-schema.json | 25 ++++++++ trace-events | 2 + 3 files changed, 208 insertions(+), 2 deletions(-) diff --git a/migration/colo.c b/migration/colo.c index 0ab9618..0ce2a6e 100644 --- a/migration/colo.c +++ b/migration/colo.c @@ -10,10 +10,12 @@ * later. See the COPYING file in the top-level directory. */ +#include <unistd.h> #include "sysemu/sysemu.h" #include "migration/colo.h" #include "trace.h" #include "qemu/error-report.h" +#include "qemu/sockets.h" bool colo_supported(void) { @@ -34,6 +36,100 @@ bool migration_incoming_in_colo_state(void) return mis && (mis->state == MIGRATION_STATUS_COLO); } +static int colo_put_cmd(QEMUFile *f, uint32_t cmd) +{ + int ret; + + if (cmd >= COLO_COMMAND_MAX) { + error_report("%s: Invalid cmd", __func__); + return -EINVAL; + } + qemu_put_be32(f, cmd); + qemu_fflush(f); + + ret = qemu_file_get_error(f); + trace_colo_put_cmd(COLOCommand_lookup[cmd]); + + return ret; +} + +static int colo_get_cmd(QEMUFile *f, uint32_t *cmd) +{ + int ret; + + *cmd = qemu_get_be32(f); + ret = qemu_file_get_error(f); + if (ret < 0) { + return ret; + } + if (*cmd >= COLO_COMMAND_MAX) { + error_report("%s: Invalid cmd", __func__); + return -EINVAL; + } + trace_colo_get_cmd(COLOCommand_lookup[*cmd]); + return 0; +} + +static int colo_get_check_cmd(QEMUFile *f, uint32_t expect_cmd) +{ + int ret; + uint32_t cmd; + + ret = colo_get_cmd(f, &cmd); + if (ret < 0) { + return ret; + } + if (cmd != expect_cmd) { + error_report("Unexpect colo command, expect:%d, but got cmd:%d", + expect_cmd, cmd); + return -EINVAL; + } + + return 0; +} + +static int colo_do_checkpoint_transaction(MigrationState *s) +{ + int ret; + + ret = colo_put_cmd(s->to_dst_file, COLO_COMMAND_CHECKPOINT_REQUEST); + if (ret < 0) { + goto out; + } + + ret = colo_get_check_cmd(s->rp_state.from_dst_file, + COLO_COMMAND_CHECKPOINT_REPLY); + if (ret < 0) { + goto out; + } + + /* TODO: suspend and save vm state to colo buffer */ + + ret = colo_put_cmd(s->to_dst_file, COLO_COMMAND_VMSTATE_SEND); + if (ret < 0) { + goto out; + } + + /* TODO: send vmstate to Secondary */ + + ret = colo_get_check_cmd(s->rp_state.from_dst_file, + COLO_COMMAND_VMSTATE_RECEIVED); + if (ret < 0) { + goto out; + } + + ret = colo_get_check_cmd(s->rp_state.from_dst_file, + COLO_COMMAND_VMSTATE_LOADED); + if (ret < 0) { + goto out; + } + + /* TODO: resume Primary */ + +out: + return ret; +} + static void colo_process_checkpoint(MigrationState *s) { int ret = 0; @@ -45,12 +141,28 @@ static void colo_process_checkpoint(MigrationState *s) goto out; } + /* + * Wait for Secondary finish loading vm states and enter COLO + * restore. + */ + ret = colo_get_check_cmd(s->rp_state.from_dst_file, + COLO_COMMAND_CHECKPOINT_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 */ + ret = colo_do_checkpoint_transaction(s); + if (ret < 0) { + goto out; + } + } out: if (ret < 0) { @@ -73,6 +185,31 @@ void migrate_start_colo_process(MigrationState *s) qemu_mutex_lock_iothread(); } +/* + * 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; + uint32_t cmd; + + ret = colo_get_cmd(f, &cmd); + if (ret < 0) { + /* do failover ? */ + return ret; + } + + switch (cmd) { + case COLO_COMMAND_CHECKPOINT_REQUEST: + *checkpoint_request = 1; + return 0; + default: + return -EINVAL; + } +} + void *colo_process_incoming_thread(void *opaque) { MigrationIncomingState *mis = opaque; @@ -93,7 +230,49 @@ void *colo_process_incoming_thread(void *opaque) */ qemu_set_block(qemu_get_fd(mis->from_src_file)); - /* TODO: COLO checkpoint restore loop */ + + ret = colo_put_cmd(mis->to_src_file, COLO_COMMAND_CHECKPOINT_READY); + if (ret < 0) { + goto out; + } + + while (mis->state == MIGRATION_STATUS_COLO) { + int request = 0; + int ret = colo_wait_handle_cmd(mis->from_src_file, &request); + + if (ret < 0) { + break; + } else { + if (!request) { + continue; + } + } + /* FIXME: This is unnecessary for periodic checkpoint mode */ + ret = colo_put_cmd(mis->to_src_file, COLO_COMMAND_CHECKPOINT_REPLY); + if (ret < 0) { + goto out; + } + + ret = colo_get_check_cmd(mis->from_src_file, + COLO_COMMAND_VMSTATE_SEND); + if (ret < 0) { + goto out; + } + + /* TODO: read migration data into colo buffer */ + + ret = colo_put_cmd(mis->to_src_file, COLO_COMMAND_VMSTATE_RECEIVED); + if (ret < 0) { + goto out; + } + + /* TODO: load vm state */ + + ret = colo_put_cmd(mis->to_src_file, COLO_COMMAND_VMSTATE_LOADED); + if (ret < 0) { + goto out; + } + } out: if (ret < 0) { diff --git a/qapi-schema.json b/qapi-schema.json index c9ff34e..85f7800 100644 --- a/qapi-schema.json +++ b/qapi-schema.json @@ -720,6 +720,31 @@ { 'command': 'migrate-start-postcopy' } ## +# @COLOCommand +# +# The commands for COLO fault tolerance +# +# @checkpoint-ready: SVM is ready for checkpointing +# +# @checkpoint-request: PVM tells SVM to prepare for new checkpointing +# +# @checkpoint-reply: SVM gets PVM's checkpoint request +# +# @vmstate-send: VM's state will be sent by PVM. +# +# @vmstate-size: The total size of VMstate. +# +# @vmstate-received: VM's state has been received by SVM. +# +# @vmstate-loaded: VM's state has been loaded by SVM. +# +# Since: 2.6 +## +{ 'enum': 'COLOCommand', + 'data': [ 'checkpoint-ready', 'checkpoint-request', 'checkpoint-reply', + 'vmstate-send', 'vmstate-size','vmstate-received', + 'vmstate-loaded' ] } + # @MouseInfo: # # Information about a mouse device. diff --git a/trace-events b/trace-events index 5565e79..39fdd8d 100644 --- a/trace-events +++ b/trace-events @@ -1579,6 +1579,8 @@ postcopy_ram_incoming_cleanup_join(void) "" # migration/colo.c colo_vm_state_change(const char *old, const char *new) "Change '%s' => '%s'" +colo_put_cmd(const char *msg) "Send '%s' cmd" +colo_get_cmd(const char *msg) "Receive '%s' cmd" # kvm-all.c kvm_ioctl(int type, void *arg) "type 0x%x, arg %p" -- 1.8.3.1