From 94f5a43704129ca4995aa3385303c5ae225bde42 Mon Sep 17 00:00:00 2001 From: Liang Li Date: Mon, 2 Nov 2015 15:37:00 +0800 Subject: migration: defer migration_end & blk_mig_cleanup Because of the patch 3ea3b7fa9af067982f34b of kvm, which introduces a lazy collapsing of small sptes into large sptes mechanism, now migration_end() is a time consuming operation because it calls memroy_global_dirty_log_stop(), which will trigger the dropping of small sptes operation and takes about dozens of milliseconds, so call migration_end() before all the vmsate data has already been transferred to the destination will prolong VM downtime. This operation should be deferred after all the data has been transferred to the destination. blk_mig_cleanup() can be deferred too. For a VM with 8G RAM, this patch can reduce the VM downtime about 30 ms. Signed-off-by: Liang Li Reviewed-by: Paolo Bonzini Reviewed-by: Juan Quintela al3 Reviewed-by: Amit Shah al3 Signed-off-by: Juan Quintela al3 --- migration/block.c | 1 - 1 file changed, 1 deletion(-) (limited to 'migration/block.c') diff --git a/migration/block.c b/migration/block.c index f7bb1e0126..8401597bcc 100644 --- a/migration/block.c +++ b/migration/block.c @@ -750,7 +750,6 @@ static int block_save_complete(QEMUFile *f, void *opaque) qemu_put_be64(f, BLK_MIG_FLAG_EOS); - blk_mig_cleanup(); return 0; } -- cgit v1.2.1