[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PATCH COLO-Frame v16 29/35] COLO: Separate the process of
From: |
zhanghailiang |
Subject: |
[Qemu-devel] [PATCH COLO-Frame v16 29/35] COLO: Separate the process of saving/loading ram and device state |
Date: |
Fri, 8 Apr 2016 14:26:31 +0800 |
We separate the process of saving/loading ram and device state when do
checkpoint, we add new helpers for save/load ram/device. With this change,
we can directly transfer ram from master to slave without using
QEMUSizeBufferas as assistant, which also reduce the size of extra memory
been used during checkpoint.
Besides, we move the colo_flush_ram_cache to the proper position after the
above change.
Signed-off-by: zhanghailiang <address@hidden>
Signed-off-by: Li Zhijian <address@hidden>
Reviewed-by: Dr. David Alan Gilbert <address@hidden>
---
v16:
- Add Reviewd-by tag
v14:
- split two new patches from this patch
- Some minor fixes from Dave
v13:
- Re-use some existed helper functions to realize saving/loading
ram and device.
v11:
- Remove load configuration section in qemu_loadvm_state_begin()
---
migration/colo.c | 48 ++++++++++++++++++++++++++++++++++++++----------
migration/ram.c | 5 -----
migration/savevm.c | 4 ++++
3 files changed, 42 insertions(+), 15 deletions(-)
diff --git a/migration/colo.c b/migration/colo.c
index 61a5d16..8c05cdb 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -289,21 +289,37 @@ static int colo_do_checkpoint_transaction(MigrationState
*s,
goto out;
}
+ colo_send_message(s->to_dst_file, COLO_MESSAGE_VMSTATE_SEND, &local_err);
+ if (local_err) {
+ goto out;
+ }
+
/* Disable block migration */
s->params.blk = 0;
s->params.shared = 0;
- qemu_savevm_state_header(trans);
- qemu_savevm_state_begin(trans, &s->params);
+ qemu_savevm_state_begin(s->to_dst_file, &s->params);
+ ret = qemu_file_get_error(s->to_dst_file);
+ if (ret < 0) {
+ error_report("Save vm state begin error");
+ goto out;
+ }
+
qemu_mutex_lock_iothread();
- qemu_savevm_state_complete_precopy(trans, false);
+ /*
+ * Only save VM's live state, which not including device state.
+ * TODO: We may need a timeout mechanism to prevent COLO process
+ * to be blocked here.
+ */
+ qemu_savevm_live_state(s->to_dst_file);
+ /* Note: device state is saved into buffer */
+ ret = qemu_save_device_state(trans);
qemu_mutex_unlock_iothread();
-
- qemu_fflush(trans);
-
- colo_send_message(s->to_dst_file, COLO_MESSAGE_VMSTATE_SEND, &local_err);
- if (local_err) {
+ if (ret < 0) {
+ error_report("Save device state error");
goto out;
}
+ qemu_fflush(trans);
+
/* we send the total size of the vmstate first */
size = qsb_get_length(buffer);
colo_send_message_value(s->to_dst_file, COLO_MESSAGE_VMSTATE_SIZE,
@@ -575,6 +591,16 @@ void *colo_process_incoming_thread(void *opaque)
goto out;
}
+ ret = qemu_loadvm_state_begin(mis->from_src_file);
+ if (ret < 0) {
+ error_report("Load vm state begin error, ret=%d", ret);
+ goto out;
+ }
+ ret = qemu_loadvm_state_main(mis->from_src_file, mis);
+ if (ret < 0) {
+ error_report("Load VM's live state (ram) error");
+ goto out;
+ }
/* read the VM state total size first */
value = colo_receive_message_value(mis->from_src_file,
COLO_MESSAGE_VMSTATE_SIZE, &local_err);
@@ -607,8 +633,10 @@ void *colo_process_incoming_thread(void *opaque)
qemu_mutex_lock_iothread();
qemu_system_reset(VMRESET_SILENT);
vmstate_loading = true;
- if (qemu_loadvm_state(fb) < 0) {
- error_report("COLO: loadvm failed");
+ colo_flush_ram_cache();
+ ret = qemu_load_device_state(fb);
+ if (ret < 0) {
+ error_report("COLO: load device state failed");
qemu_mutex_unlock_iothread();
goto out;
}
diff --git a/migration/ram.c b/migration/ram.c
index 1f399a8..6edc039 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -2466,7 +2466,6 @@ static int ram_load(QEMUFile *f, void *opaque, int
version_id)
* be atomic
*/
bool postcopy_running = postcopy_state_get() >=
POSTCOPY_INCOMING_LISTENING;
- bool need_flush = false;
seq_iter++;
@@ -2501,7 +2500,6 @@ static int ram_load(QEMUFile *f, void *opaque, int
version_id)
/* After going into COLO, we should load the Page into colo_cache
*/
if (ram_cache_enable) {
host = colo_cache_from_block_offset(block, addr);
- need_flush = true;
} else {
host = host_from_ram_block_offset(block, addr);
}
@@ -2595,9 +2593,6 @@ static int ram_load(QEMUFile *f, void *opaque, int
version_id)
rcu_read_unlock();
- if (!ret && ram_cache_enable && need_flush) {
- colo_flush_ram_cache();
- }
DPRINTF("Completed load of VM with exit code %d seq iteration "
"%" PRIu64 "\n", ret, seq_iter);
return ret;
diff --git a/migration/savevm.c b/migration/savevm.c
index 26e95d2..4296eab 100644
--- a/migration/savevm.c
+++ b/migration/savevm.c
@@ -929,6 +929,10 @@ void qemu_savevm_state_begin(QEMUFile *f,
break;
}
}
+ if (migration_in_colo_state()) {
+ qemu_put_byte(f, QEMU_VM_EOF);
+ qemu_fflush(f);
+ }
}
/*
--
1.8.3.1
- [Qemu-devel] [PATCH COLO-Frame v16 18/35] COLO failover: Introduce state to record failover process, (continued)
- [Qemu-devel] [PATCH COLO-Frame v16 18/35] COLO failover: Introduce state to record failover process, zhanghailiang, 2016/04/08
- [Qemu-devel] [PATCH COLO-Frame v16 25/35] COLO: Update the global runstate after going into colo state, zhanghailiang, 2016/04/08
- [Qemu-devel] [PATCH COLO-Frame v16 30/35] COLO: Split qemu_savevm_state_begin out of checkpoint process, zhanghailiang, 2016/04/08
- [Qemu-devel] [PATCH COLO-Frame v16 32/35] net: Add notifier/callback for netdev init, zhanghailiang, 2016/04/08
- [Qemu-devel] [PATCH COLO-Frame v16 15/35] COLO: Add checkpoint-delay parameter for migrate-set-parameters, zhanghailiang, 2016/04/08
- [Qemu-devel] [PATCH COLO-Frame v16 08/35] COLO: Add a new RunState RUN_STATE_COLO, zhanghailiang, 2016/04/08
- [Qemu-devel] [PATCH COLO-Frame v16 31/35] filter-buffer: Accept zero interval, zhanghailiang, 2016/04/08
- [Qemu-devel] [PATCH COLO-Frame v16 29/35] COLO: Separate the process of saving/loading ram and device state,
zhanghailiang <=
- [Qemu-devel] [PATCH COLO-Frame v16 34/35] COLO: manage the status of buffer filters for PVM, zhanghailiang, 2016/04/08
- [Qemu-devel] [PATCH COLO-Frame v16 35/35] COLO: Add block replication into colo process, zhanghailiang, 2016/04/08
- [Qemu-devel] [PATCH COLO-Frame v16 33/35] COLO/filter: add each netdev a buffer filter, zhanghailiang, 2016/04/08
- [Qemu-devel] [PATCH COLO-Frame v16 22/35] COLO failover: Shutdown related socket fd when do failover, zhanghailiang, 2016/04/08
- Re: [Qemu-devel] [PATCH COLO-Frame v16 for-2.7 00/35] COarse-grain LOck-stepping(COLO) Virtual Machines for Non-stop Service (FT), Zhang Chen, 2016/04/08
- Re: [Qemu-devel] [PATCH COLO-Frame v16 for-2.7 00/35] COarse-grain LOck-stepping(COLO) Virtual Machines for Non-stop Service (FT), Dr. David Alan Gilbert, 2016/04/22