[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PATCH V12 08/19] ram/COLO: Record the dirty pages that SVM
From: |
Zhang Chen |
Subject: |
[Qemu-devel] [PATCH V12 08/19] ram/COLO: Record the dirty pages that SVM received |
Date: |
Mon, 3 Sep 2018 12:38:49 +0800 |
We record the address of the dirty pages that received,
it will help flushing pages that cached into SVM.
Here, it is a trick, we record dirty pages by re-using migration
dirty bitmap. In the later patch, we will start the dirty log
for SVM, just like migration, in this way, we can record both
the dirty pages caused by PVM and SVM, we only flush those dirty
pages from RAM cache while do checkpoint.
Signed-off-by: zhanghailiang <address@hidden>
Signed-off-by: Zhang Chen <address@hidden>
Signed-off-by: Zhang Chen <address@hidden>
Reviewed-by: Dr. David Alan Gilbert <address@hidden>
---
migration/ram.c | 43 ++++++++++++++++++++++++++++++++++++++++---
1 file changed, 40 insertions(+), 3 deletions(-)
diff --git a/migration/ram.c b/migration/ram.c
index a63315533e..df8202216f 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -3409,6 +3409,15 @@ static inline void
*colo_cache_from_block_offset(RAMBlock *block,
__func__, block->idstr);
return NULL;
}
+
+ /*
+ * During colo checkpoint, we need bitmap of these migrated pages.
+ * It help us to decide which pages in ram cache should be flushed
+ * into VM's RAM later.
+ */
+ if (!test_and_set_bit(offset >> TARGET_PAGE_BITS, block->bmap)) {
+ ram_state->migration_dirty_pages++;
+ }
return block->colo_cache + offset;
}
@@ -3626,7 +3635,7 @@ int colo_init_ram_cache(void)
RAMBlock *block;
rcu_read_lock();
- QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
+ RAMBLOCK_FOREACH_MIGRATABLE(block) {
block->colo_cache = qemu_anon_ram_alloc(block->used_length,
NULL,
false);
@@ -3639,10 +3648,29 @@ int colo_init_ram_cache(void)
memcpy(block->colo_cache, block->host, block->used_length);
}
rcu_read_unlock();
+ /*
+ * Record the dirty pages that sent by PVM, we use this dirty bitmap
together
+ * with to decide which page in cache should be flushed into SVM's RAM. Here
+ * we use the same name 'ram_bitmap' as for migration.
+ */
+ if (ram_bytes_total()) {
+ RAMBlock *block;
+
+ RAMBLOCK_FOREACH_MIGRATABLE(block) {
+ unsigned long pages = block->max_length >> TARGET_PAGE_BITS;
+
+ block->bmap = bitmap_new(pages);
+ bitmap_set(block->bmap, 0, pages);
+ }
+ }
+ ram_state = g_new0(RAMState, 1);
+ ram_state->migration_dirty_pages = 0;
+
return 0;
out_locked:
- QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
+
+ RAMBLOCK_FOREACH_MIGRATABLE(block) {
if (block->colo_cache) {
qemu_anon_ram_free(block->colo_cache, block->used_length);
block->colo_cache = NULL;
@@ -3658,14 +3686,23 @@ void colo_release_ram_cache(void)
{
RAMBlock *block;
+ RAMBLOCK_FOREACH_MIGRATABLE(block) {
+ g_free(block->bmap);
+ block->bmap = NULL;
+ }
+
rcu_read_lock();
- QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
+
+ RAMBLOCK_FOREACH_MIGRATABLE(block) {
if (block->colo_cache) {
qemu_anon_ram_free(block->colo_cache, block->used_length);
block->colo_cache = NULL;
}
}
+
rcu_read_unlock();
+ g_free(ram_state);
+ ram_state = NULL;
}
/**
--
2.17.GIT
- [Qemu-devel] [PATCH V12 01/19] filter-rewriter: Add TCP state machine and fix memory leak in connection_track_table, (continued)
- [Qemu-devel] [PATCH V12 01/19] filter-rewriter: Add TCP state machine and fix memory leak in connection_track_table, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 02/19] colo-compare: implement the process of checkpoint, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 03/19] colo-compare: use notifier to notify packets comparing result, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 04/19] COLO: integrate colo compare with colo frame, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 05/19] COLO: Add block replication into colo process, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 06/19] COLO: Remove colo_state migration struct, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 07/19] COLO: Load dirty pages into SVM's RAM cache firstly, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 08/19] ram/COLO: Record the dirty pages that SVM received,
Zhang Chen <=
- [Qemu-devel] [PATCH V12 09/19] COLO: Flush memory data from ram cache, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 10/19] qmp event: Add COLO_EXIT event to notify users while exited COLO, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 11/19] qapi/migration.json: Rename COLO unknown mode to none mode., Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 12/19] qapi: Add new command to query colo status, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 13/19] savevm: split the process of different stages for loadvm/savevm, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 14/19] COLO: flush host dirty ram from cache, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 15/19] filter: Add handle_event method for NetFilterClass, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 16/19] filter-rewriter: handle checkpoint and failover event, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 17/19] COLO: notify net filters about checkpoint/failover event, Zhang Chen, 2018/09/03
- [Qemu-devel] [PATCH V12 18/19] COLO: quick failover process by kick COLO thread, Zhang Chen, 2018/09/03