From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:44649) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1YnIc6-00039u-VY for qemu-devel@nongnu.org; Tue, 28 Apr 2015 23:24:44 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1YnIc2-0001Ki-8k for qemu-devel@nongnu.org; Tue, 28 Apr 2015 23:24:42 -0400 Received: from mx1.redhat.com ([209.132.183.28]:53822) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1YnIc2-0001Kd-0R for qemu-devel@nongnu.org; Tue, 28 Apr 2015 23:24:38 -0400 Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id t3T3ObS8018086 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=FAIL) for ; Tue, 28 Apr 2015 23:24:37 -0400 From: Fam Zheng Date: Wed, 29 Apr 2015 11:24:24 +0800 Message-Id: <1430277871-26761-1-git-send-email-famz@redhat.com> Subject: [Qemu-devel] [PATCH v4 0/7] aio: Support epoll by introducing qemu_poll abstraction List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Kevin Wolf , Paolo Bonzini , Stefan Hajnoczi v4: Rebase to 2.3 and rerun tests. timerfd+epoll shows clear improvement over current master when the virtio-console device attaches more fds to the main loop: syscall high # of fd low # of fd ------------------------------------------------- qemu.git(ppoll) 24 83 ppoll+epoll 49 82 timerfd+epoll 49 91 glib 20 82 (In high # of fd case, 3 activated but idle virtio-console devices are attached, which will add us hundereds of fds to poll) Although a new epoll_pwait1 syscall is still being worked on [1], the QEMU part to make use of that (if any) will base on this, so let's merge it first. [1]: http://www.spinics.net/lists/linux-api/msg08216.html v1 cover letter --------------- ppoll(2) doesn't scale as well as epoll: The elapsed time of the syscall is linear to the number of fd's we poll, which hurts performance a bit when the number of devices are many, or when a virtio device registers many virtqueues (virtio-serial, for instance). This series introduces qemu_poll, which is implemented with g_poll and epoll, decided at configure time with CONFIG_EPOLL. Fam Fam Zheng (7): poll: Introduce QEMU Poll API posix-aio: Use QEMU poll interface poll: Add epoll implementation for qemu_poll main-loop: Replace qemu_poll_ns with qemu_poll tests: Add test case for qemu_poll poll-glib: Support ppoll poll-linux: Add timerfd support Makefile.objs | 2 + aio-posix.c | 26 +++- async.c | 3 + include/block/aio.h | 6 + include/qemu/poll.h | 40 +++++++ include/qemu/timer.h | 13 -- include/qemu/typedefs.h | 4 +- main-loop.c | 35 +++++- poll-glib.c | 153 ++++++++++++++++++++++++ poll-linux.c | 310 ++++++++++++++++++++++++++++++++++++++++++++++++ qemu-timer.c | 28 ----- tests/Makefile | 2 + tests/test-poll.c | 272 ++++++++++++++++++++++++++++++++++++++++++ 13 files changed, 841 insertions(+), 53 deletions(-) create mode 100644 include/qemu/poll.h create mode 100644 poll-glib.c create mode 100644 poll-linux.c create mode 100644 tests/test-poll.c -- 1.9.3