[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH COLO-Frame v13 32/39] COLO: Separate the process
From: |
Dr. David Alan Gilbert |
Subject: |
Re: [Qemu-devel] [PATCH COLO-Frame v13 32/39] COLO: Separate the process of saving/loading ram and device state |
Date: |
Wed, 27 Jan 2016 14:14:59 +0000 |
User-agent: |
Mutt/1.5.24 (2015-08-30) |
* zhanghailiang (address@hidden) wrote:
> 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 QEMUSizeBuffer 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>
> ---
> v13:
> - Re-use some existed helper functions to realize saving/loading ram and
> device.
> v11:
> - Remove load configuration section in qemu_loadvm_state_begin()
> ---
> include/sysemu/sysemu.h | 6 +++++
> migration/colo.c | 47 ++++++++++++++++++++++++++---------
> migration/ram.c | 5 ----
> migration/savevm.c | 66
> ++++++++++++++++++++++++++++++++++++++++++++++---
> 4 files changed, 104 insertions(+), 20 deletions(-)
>
> diff --git a/include/sysemu/sysemu.h b/include/sysemu/sysemu.h
> index 91eeda3..f553f8a 100644
> --- a/include/sysemu/sysemu.h
> +++ b/include/sysemu/sysemu.h
> @@ -133,7 +133,13 @@ void qemu_savevm_send_postcopy_ram_discard(QEMUFile *f,
> const char *name,
> uint64_t *start_list,
> uint64_t *length_list);
>
> +void qemu_savevm_live_state(QEMUFile *f);
> +int qemu_save_device_state(QEMUFile *f);
> +
> int qemu_loadvm_state(QEMUFile *f);
> +int qemu_loadvm_state_begin(QEMUFile *f);
> +int qemu_load_ram_state(QEMUFile *f);
> +int qemu_load_device_state(QEMUFile *f);
>
> typedef enum DisplayType
> {
> diff --git a/migration/colo.c b/migration/colo.c
> index fd87f34..4a99819 100644
> --- a/migration/colo.c
> +++ b/migration/colo.c
> @@ -290,21 +290,34 @@ static int
> colo_do_checkpoint_transaction(MigrationState *s,
> goto out;
> }
>
> + colo_put_cmd(s->to_dst_file, COLO_COMMAND_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_mutex_lock_iothread();
> - qemu_savevm_state_complete_precopy(trans, false);
> - qemu_mutex_unlock_iothread();
> -
> - qemu_fflush(trans);
> + 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;
> + }
>
> - colo_put_cmd(s->to_dst_file, COLO_COMMAND_VMSTATE_SEND, &local_err);
> - if (local_err) {
> + qemu_mutex_lock_iothread();
> + /* Only save VM's live state, which not including device state */
> + qemu_savevm_live_state(s->to_dst_file);
I'm a little worried that this is done with the lock held;
if the destination hung during this point and there was
enough data to fill the network buffers, couldn't we hang
and then we wouldn't be able to issue the colo_lost_heartbeat
command?
> + /* Note: device state is saved into buffer */
> + ret = qemu_save_device_state(trans);
> + if (ret < 0) {
> + error_report("save device state error\n");
> + qemu_mutex_unlock_iothread();
> goto out;
> }
> + qemu_fflush(trans);
> + qemu_mutex_unlock_iothread();
You can save an unlock by doing:
ret = ...
qemu_mutex_unlock_iothread();
if (ret < 0) {
> /* we send the total size of the vmstate first */
> size = qsb_get_length(buffer);
> colo_put_cmd_value(s->to_dst_file, COLO_COMMAND_VMSTATE_SIZE,
> @@ -573,6 +586,16 @@ 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_load_ram_state(mis->from_src_file);
> + if (ret < 0) {
> + error_report("load ram state error");
> + goto out;
> + }
> /* read the VM state total size first */
> value = colo_get_cmd_value(mis->from_src_file,
> COLO_COMMAND_VMSTATE_SIZE, &local_err);
> @@ -605,8 +628,10 @@ 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\n");
Unneeded \n
> vmstate_loading = false;
> qemu_mutex_unlock_iothread();
> goto out;
> diff --git a/migration/ram.c b/migration/ram.c
> index 8ff7f7c..45d9332 100644
> --- a/migration/ram.c
> +++ b/migration/ram.c
> @@ -2458,7 +2458,6 @@ 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++;
>
> @@ -2493,7 +2492,6 @@ 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);
> }
> @@ -2588,9 +2586,6 @@ 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 7ceec2d..7892070 100644
> --- a/migration/savevm.c
> +++ b/migration/savevm.c
> @@ -50,6 +50,7 @@
> #include "qemu/iov.h"
> #include "block/snapshot.h"
> #include "block/qapi.h"
> +#include "migration/colo.h"
>
>
> #ifndef ETH_P_RARP
> @@ -923,6 +924,10 @@ void qemu_savevm_state_begin(QEMUFile *f,
> break;
> }
> }
> + if (migration_in_colo_state()) {
> + qemu_put_byte(f, QEMU_VM_EOF);
> + qemu_fflush(f);
> + }
> }
>
> /*
> @@ -1192,13 +1197,20 @@ static int qemu_savevm_state(QEMUFile *f, Error
> **errp)
> return ret;
> }
>
> -static int qemu_save_device_state(QEMUFile *f)
> +void qemu_savevm_live_state(QEMUFile *f)
> {
> - SaveStateEntry *se;
> + /* save QEMU_VM_SECTION_END section */
> + qemu_savevm_state_complete_precopy(f, true);
> + qemu_put_byte(f, QEMU_VM_EOF);
> +}
>
> - qemu_put_be32(f, QEMU_VM_FILE_MAGIC);
> - qemu_put_be32(f, QEMU_VM_FILE_VERSION);
> +int qemu_save_device_state(QEMUFile *f)
> +{
> + SaveStateEntry *se;
>
> + if (!migration_in_colo_state()) {
> + qemu_savevm_state_header(f);
> + }
> cpu_synchronize_all_states();
>
> QTAILQ_FOREACH(se, &savevm_state.handlers, entry) {
> @@ -1957,6 +1969,52 @@ int qemu_loadvm_state(QEMUFile *f)
> return ret;
> }
>
> +int qemu_loadvm_state_begin(QEMUFile *f)
> +{
> + MigrationIncomingState *mis = migration_incoming_get_current();
> + Error *local_err = NULL;
> + int ret;
> +
> + if (qemu_savevm_state_blocked(&local_err)) {
> + error_report_err(local_err);
> + return -EINVAL;
> + }
> + /* Load QEMU_VM_SECTION_START section */
> + ret = qemu_loadvm_state_main(f, mis);
> + if (ret < 0) {
> + error_report("Failed to loadvm begin work: %d", ret);
> + }
> + return ret;
> +}
> +
> +int qemu_load_ram_state(QEMUFile *f)
> +{
> + MigrationIncomingState *mis = migration_incoming_get_current();
> + int ret;
> +
> + ret = qemu_loadvm_state_main(f, mis);
> + if (ret < 0) {
> + error_report("Failed to load ram state: %d", ret);
> + }
> + return ret;
> +}
This function does very little; why not just make qemu_loadvm_state_main
non-static and call that directly?
Dave
> +int qemu_load_device_state(QEMUFile *f)
> +{
> + MigrationIncomingState *mis = migration_incoming_get_current();
> + int ret;
> +
> + /* Load QEMU_VM_SECTION_FULL section */
> + ret = qemu_loadvm_state_main(f, mis);
> + if (ret < 0) {
> + error_report("Failed to load device state: %d", ret);
> + return ret;
> + }
> +
> + cpu_synchronize_all_post_init();
> + return 0;
> +}
> +
> void hmp_savevm(Monitor *mon, const QDict *qdict)
> {
> BlockDriverState *bs, *bs1;
> --
> 1.8.3.1
>
>
--
Dr. David Alan Gilbert / address@hidden / Manchester, UK
- Re: [Qemu-devel] [PATCH COLO-Frame v13 32/39] COLO: Separate the process of saving/loading ram and device state,
Dr. David Alan Gilbert <=