[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PULL 09/26] COLO: Flush memory data from ram cache
From: |
Jason Wang |
Subject: |
[Qemu-devel] [PULL 09/26] COLO: Flush memory data from ram cache |
Date: |
Mon, 15 Oct 2018 16:46:09 +0800 |
From: Zhang Chen <address@hidden>
During the time of VM's running, PVM may dirty some pages, we will transfer
PVM's dirty pages to SVM and store them into SVM's RAM cache at next checkpoint
time. So, the content of SVM's RAM cache will always be same with PVM's memory
after checkpoint.
Instead of flushing all content of PVM's RAM cache into SVM's MEMORY,
we do this in a more efficient way:
Only flush any page that dirtied by PVM since last checkpoint.
In this way, we can ensure SVM's memory same with PVM's.
Besides, we must ensure flush RAM cache before load device state.
Signed-off-by: zhanghailiang <address@hidden>
Signed-off-by: Li Zhijian <address@hidden>
Reviewed-by: Dr. David Alan Gilbert <address@hidden>
Signed-off-by: Jason Wang <address@hidden>
---
migration/ram.c | 37 +++++++++++++++++++++++++++++++++++++
migration/trace-events | 2 ++
2 files changed, 39 insertions(+)
diff --git a/migration/ram.c b/migration/ram.c
index 404c8f0..477853d 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -3974,6 +3974,39 @@ static bool postcopy_is_running(void)
return ps >= POSTCOPY_INCOMING_LISTENING && ps < POSTCOPY_INCOMING_END;
}
+/*
+ * Flush content of RAM cache into SVM's memory.
+ * Only flush the pages that be dirtied by PVM or SVM or both.
+ */
+static void colo_flush_ram_cache(void)
+{
+ RAMBlock *block = NULL;
+ void *dst_host;
+ void *src_host;
+ unsigned long offset = 0;
+
+ trace_colo_flush_ram_cache_begin(ram_state->migration_dirty_pages);
+ rcu_read_lock();
+ block = QLIST_FIRST_RCU(&ram_list.blocks);
+
+ while (block) {
+ offset = migration_bitmap_find_dirty(ram_state, block, offset);
+
+ if (offset << TARGET_PAGE_BITS >= block->used_length) {
+ offset = 0;
+ block = QLIST_NEXT_RCU(block, next);
+ } else {
+ migration_bitmap_clear_dirty(ram_state, block, offset);
+ dst_host = block->host + (offset << TARGET_PAGE_BITS);
+ src_host = block->colo_cache + (offset << TARGET_PAGE_BITS);
+ memcpy(dst_host, src_host, TARGET_PAGE_SIZE);
+ }
+ }
+
+ rcu_read_unlock();
+ trace_colo_flush_ram_cache_end();
+}
+
static int ram_load(QEMUFile *f, void *opaque, int version_id)
{
int flags = 0, ret = 0, invalid_flags = 0;
@@ -4150,6 +4183,10 @@ static int ram_load(QEMUFile *f, void *opaque, int
version_id)
ret |= wait_for_decompress_done();
rcu_read_unlock();
trace_ram_load_complete(ret, seq_iter);
+
+ if (!ret && migration_incoming_in_colo_state()) {
+ colo_flush_ram_cache();
+ }
return ret;
}
diff --git a/migration/trace-events b/migration/trace-events
index fa0ff3f..bd2d0cd 100644
--- a/migration/trace-events
+++ b/migration/trace-events
@@ -102,6 +102,8 @@ ram_dirty_bitmap_sync_start(void) ""
ram_dirty_bitmap_sync_wait(void) ""
ram_dirty_bitmap_sync_complete(void) ""
ram_state_resume_prepare(uint64_t v) "%" PRId64
+colo_flush_ram_cache_begin(uint64_t dirty_pages) "dirty_pages %" PRIu64
+colo_flush_ram_cache_end(void) ""
# migration/migration.c
await_return_path_close_on_source_close(void) ""
--
2.5.0
- [Qemu-devel] [PULL 00/26] Net patches, Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 01/26] filter-rewriter: Add TCP state machine and fix memory leak in connection_track_table, Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 09/26] COLO: Flush memory data from ram cache,
Jason Wang <=
- [Qemu-devel] [PULL 02/26] colo-compare: implement the process of checkpoint, Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 08/26] ram/COLO: Record the dirty pages that SVM received, Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 11/26] qapi/migration.json: Rename COLO unknown mode to none mode., Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 04/26] COLO: integrate colo compare with colo frame, Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 10/26] qmp event: Add COLO_EXIT event to notify users while exited COLO, Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 05/26] COLO: Add block replication into colo process, Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 14/26] COLO: flush host dirty ram from cache, Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 18/26] COLO: quick failover process by kick COLO thread, Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 03/26] colo-compare: use notifier to notify packets comparing result, Jason Wang, 2018/10/15
- [Qemu-devel] [PULL 15/26] filter: Add handle_event method for NetFilterClass, Jason Wang, 2018/10/15