[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[PULL 03/17] migration/ram: Optimize ram_save_host_page()
From: |
Dr. David Alan Gilbert (git) |
Subject: |
[PULL 03/17] migration/ram: Optimize ram_save_host_page() |
Date: |
Thu, 13 May 2021 18:37:23 +0100 |
From: Kunkun Jiang <jiangkunkun@huawei.com>
Starting from pss->page, ram_save_host_page() will check every page
and send the dirty pages up to the end of the current host page or
the boundary of used_length of the block. If the host page size is
a huge page, the step "check" will take a lot of time.
It will improve performance to use migration_bitmap_find_dirty().
Tested on Kunpeng 920; VM parameters: 1U 4G (page size 1G)
The time of ram_save_host_page() in the last round of ram saving:
before optimize: 9250us after optimize: 34us
Signed-off-by: Keqian Zhu <zhukeqian1@huawei.com>
Signed-off-by: Kunkun Jiang <jiangkunkun@huawei.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Message-Id: <20210316125716.1243-3-jiangkunkun@huawei.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
---
migration/ram.c | 39 +++++++++++++++++++--------------------
1 file changed, 19 insertions(+), 20 deletions(-)
diff --git a/migration/ram.c b/migration/ram.c
index 00b579b981..bb52bd97db 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -2013,6 +2013,8 @@ static int ram_save_host_page(RAMState *rs,
PageSearchStatus *pss,
int tmppages, pages = 0;
size_t pagesize_bits =
qemu_ram_pagesize(pss->block) >> TARGET_PAGE_BITS;
+ unsigned long hostpage_boundary =
+ QEMU_ALIGN_UP(pss->page + 1, pagesize_bits);
unsigned long start_page = pss->page;
int res;
@@ -2023,30 +2025,27 @@ static int ram_save_host_page(RAMState *rs,
PageSearchStatus *pss,
do {
/* Check the pages is dirty and if it is send it */
- if (!migration_bitmap_clear_dirty(rs, pss->block, pss->page)) {
- pss->page++;
- continue;
- }
-
- tmppages = ram_save_target_page(rs, pss, last_stage);
- if (tmppages < 0) {
- return tmppages;
- }
+ if (migration_bitmap_clear_dirty(rs, pss->block, pss->page)) {
+ tmppages = ram_save_target_page(rs, pss, last_stage);
+ if (tmppages < 0) {
+ return tmppages;
+ }
- pages += tmppages;
- pss->page++;
- /*
- * Allow rate limiting to happen in the middle of huge pages if
- * something is sent in the current iteration.
- */
- if (pagesize_bits > 1 && tmppages > 0) {
- migration_rate_limit();
+ pages += tmppages;
+ /*
+ * Allow rate limiting to happen in the middle of huge pages if
+ * something is sent in the current iteration.
+ */
+ if (pagesize_bits > 1 && tmppages > 0) {
+ migration_rate_limit();
+ }
}
- } while ((pss->page & (pagesize_bits - 1)) &&
+ pss->page = migration_bitmap_find_dirty(rs, pss->block, pss->page);
+ } while ((pss->page < hostpage_boundary) &&
offset_in_ramblock(pss->block,
((ram_addr_t)pss->page) << TARGET_PAGE_BITS));
- /* The offset we leave with is the last one we looked at */
- pss->page--;
+ /* The offset we leave with is the min boundary of host page and block */
+ pss->page = MIN(pss->page, hostpage_boundary) - 1;
res = ram_save_release_protection(rs, pss, start_page);
return (res < 0 ? res : pages);
--
2.31.1
- [PULL 09/17] exec: Relax range check in ram_block_discard_range(), (continued)
- [PULL 09/17] exec: Relax range check in ram_block_discard_range(), Dr. David Alan Gilbert (git), 2021/05/11
- [PULL 16/17] tests/qtest/migration-test: Use g_autofree to avoid leaks on error paths, Dr. David Alan Gilbert (git), 2021/05/11
- [PULL 17/17] tests/migration: introduce multifd into guestperf, Dr. David Alan Gilbert (git), 2021/05/11
- Re: [PULL 00/17] migration queue, Peter Maydell, 2021/05/13
- [PULL 00/17] migration queue, Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 01/17] migrate/ram: remove "ram_bulk_stage" and "fpo_enabled", Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 02/17] migration/ram: Reduce unnecessary rate limiting, Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 03/17] migration/ram: Optimize ram_save_host_page(),
Dr. David Alan Gilbert (git) <=
- [PULL 04/17] migration: Drop redundant query-migrate result @blocked, Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 05/17] util: vfio-helpers: Factor out and fix processing of existing ram blocks, Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 06/17] numa: Teach ram block notifiers about resizeable ram blocks, Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 07/17] numa: Make all callbacks of ram block notifiers optional, Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 08/17] migration/ram: Handle RAM block resizes during precopy, Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 10/17] migration/ram: Discard RAM when growing RAM blocks after ram_postcopy_incoming_init(), Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 09/17] exec: Relax range check in ram_block_discard_range(), Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 11/17] migration/ram: Simplify host page handling in ram_load_postcopy(), Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 13/17] migration/multifd: Print used_length of memory block, Dr. David Alan Gilbert (git), 2021/05/13
- [PULL 12/17] migration/ram: Handle RAM block resizes during postcopy, Dr. David Alan Gilbert (git), 2021/05/13