From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:56601) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eWi1z-0006az-Lx for qemu-devel@nongnu.org; Wed, 03 Jan 2018 07:20:28 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eWi1w-0001t3-Gw for qemu-devel@nongnu.org; Wed, 03 Jan 2018 07:20:27 -0500 Received: from mx1.redhat.com ([209.132.183.28]:59132) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1eWi1w-0001ry-4g for qemu-devel@nongnu.org; Wed, 03 Jan 2018 07:20:24 -0500 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.phx2.redhat.com [10.5.11.16]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 18623C0AE082 for ; Wed, 3 Jan 2018 12:20:23 +0000 (UTC) From: Peter Xu Date: Wed, 3 Jan 2018 20:20:04 +0800 Message-Id: <20180103122017.14794-1-peterx@redhat.com> Subject: [Qemu-devel] [PATCH v2 00/13] migration: cleanup migration_thread() List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Laurent Vivier , Juan Quintela , "Dr . David Alan Gilbert" , peterx@redhat.com v2 - add r-bs - the first patch: don't assert but error_report [Juan, Dave] - remove mig_* prefix in var names [Juan] - two more patches [Juan]: migration: remove notify in migrate_fd_error migration: remove some block_cleanup_parameters() - rename old_vm_running to vm_was_running [Juan] - rename vm_down_start_time to downtime_start [Juan] - set s->downtime conditionally in migration_calculate_complete() so that postcopy downtime can be kept [Juan] Firstly this series is something as a first attempt of me to cleanup some migration code. It may not be a good idea, but I still think it worth a try, so I posted it. Let me know if any of you don't like it, so I can stop. At least after the series the migration_thread() function can be far shorter and much easier for first-time readers AFAICT. For this single function, the most complexity part is quite a lot of local variables crossly referenced everywhere, and during the cleanup I do think the COLO part is hacky too. There can be some functional changes too in the future: - cancel migration as cleanup of QEMU quit - add migration locks to protect migration internal states - ... But for this single series, there is still no functional change yet. Please have a look. Any feedback is welcomed. Thanks, Peter Xu (13): migration: assert colo instead of check migration: qemu_savevm_state_cleanup() in cleanup migration: remove "enable_colo" var migration: split use of MigrationState.total_time migration: move vm_old_running into global state migration: introduce downtime_start migration: introduce migrate_calculate_complete migration: use switch at the end of migration migration: cleanup stats update into function migration: major cleanup for migrate iterations migration: put the finish part into a new function migration: remove some block_cleanup_parameters() migration: remove notify in fd_error migration/migration.c | 307 +++++++++++++++++++++++++++++--------------------- migration/migration.h | 22 ++++ 2 files changed, 200 insertions(+), 129 deletions(-) -- 2.14.3