From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:48506) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZriYE-0005e5-0v for qemu-devel@nongnu.org; Thu, 29 Oct 2015 04:27:14 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ZriYA-0007lp-Ri for qemu-devel@nongnu.org; Thu, 29 Oct 2015 04:27:13 -0400 Received: from mga09.intel.com ([134.134.136.24]:32862) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZriYA-0007li-Ld for qemu-devel@nongnu.org; Thu, 29 Oct 2015 04:27:10 -0400 From: Liang Li Date: Thu, 29 Oct 2015 16:22:05 +0800 Message-Id: <1446106927-15490-1-git-send-email-liang.z.li@intel.com> Subject: [Qemu-devel] [RESEND 0/2] Optimze the env->tsc update operation List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: pbonzini@redhat.com, mtosatti@redhat.com, m.gibula@beyond.pl, Liang Li This patch aims for reducing the live migration downtime. It updates the env->tsc value when stopping kvmclock with a ligth weight interface, rather than using 'cpu_synchronize_all_states', which takes more time. For a VM with 4 CPUs, this path can help to reduce the VM downtime about 100us. Liang Li (2): kvmclock: use a light weight interface to update env->tsc. Revert "Introduce cpu_clean_all_dirty" cpus.c | 9 --------- hw/i386/kvm/clock.c | 26 +++++++++----------------- include/sysemu/cpus.h | 1 - include/sysemu/kvm.h | 8 -------- kvm-all.c | 5 ----- target-i386/kvm.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++ target-i386/kvm_i386.h | 1 + 7 files changed, 56 insertions(+), 40 deletions(-) -- 1.9.1