From patchwork Thu Jan 25 05:46:37 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Fam Zheng X-Patchwork-Id: 865758 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=pass (mailfrom) smtp.mailfrom=nongnu.org (client-ip=2001:4830:134:3::11; helo=lists.gnu.org; envelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org; receiver=) Received: from lists.gnu.org (lists.gnu.org [IPv6:2001:4830:134:3::11]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by ozlabs.org (Postfix) with ESMTPS id 3zRrp263xfz9t2Q for ; Thu, 25 Jan 2018 16:54:17 +1100 (AEDT) Received: from localhost ([::1]:45578 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eeaUJ-0003ZO-5D for incoming@patchwork.ozlabs.org; Thu, 25 Jan 2018 00:54:15 -0500 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59293) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eeaNK-0006hq-Sa for qemu-devel@nongnu.org; Thu, 25 Jan 2018 00:47:03 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eeaNJ-00079j-TI for qemu-devel@nongnu.org; Thu, 25 Jan 2018 00:47:02 -0500 Received: from mx1.redhat.com ([209.132.183.28]:50932) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1eeaNJ-00078W-OY for qemu-devel@nongnu.org; Thu, 25 Jan 2018 00:47:01 -0500 Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.phx2.redhat.com [10.5.11.15]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id EF8817A16E; Thu, 25 Jan 2018 05:47:00 +0000 (UTC) Received: from lemon.usersys.redhat.com (ovpn-12-118.pek2.redhat.com [10.72.12.118]) by smtp.corp.redhat.com (Postfix) with ESMTP id E85925D6A8; Thu, 25 Jan 2018 05:46:59 +0000 (UTC) From: Fam Zheng To: qemu-devel@nongnu.org Date: Thu, 25 Jan 2018 13:46:37 +0800 Message-Id: <20180125054648.18747-4-famz@redhat.com> In-Reply-To: <20180125054648.18747-1-famz@redhat.com> References: <20180125054648.18747-1-famz@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.15 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.28]); Thu, 25 Jan 2018 05:47:01 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] [fuzzy] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PULL 03/14] coroutine-lock: convert CoQueue to use QemuLockable X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Peter Maydell Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Sender: "Qemu-devel" From: Paolo Bonzini There are cases in which a queued coroutine must be restarted from non-coroutine context (with qemu_co_enter_next). In this cases, qemu_co_enter_next also needs to be thread-safe, but it cannot use a CoMutex and so cannot qemu_co_queue_wait. Use QemuLockable so that the CoQueue can interchangeably use CoMutex or QemuMutex. Signed-off-by: Paolo Bonzini Message-Id: <20180116142316.30486-3-pbonzini@redhat.com> Reviewed-by: Fam Zheng Signed-off-by: Fam Zheng --- include/qemu/coroutine.h | 6 +++++- util/qemu-coroutine-lock.c | 12 +++++++----- 2 files changed, 12 insertions(+), 6 deletions(-) diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h index 8a5129741c..1e5f0957e6 100644 --- a/include/qemu/coroutine.h +++ b/include/qemu/coroutine.h @@ -183,7 +183,9 @@ void qemu_co_queue_init(CoQueue *queue); * caller of the coroutine. The mutex is unlocked during the wait and * locked again afterwards. */ -void coroutine_fn qemu_co_queue_wait(CoQueue *queue, CoMutex *mutex); +#define qemu_co_queue_wait(queue, lock) \ + qemu_co_queue_wait_impl(queue, QEMU_MAKE_LOCKABLE(lock)) +void coroutine_fn qemu_co_queue_wait_impl(CoQueue *queue, QemuLockable *lock); /** * Restarts the next coroutine in the CoQueue and removes it from the queue. @@ -271,4 +273,6 @@ void coroutine_fn qemu_co_sleep_ns(QEMUClockType type, int64_t ns); */ void coroutine_fn yield_until_fd_readable(int fd); +#include "qemu/lockable.h" + #endif /* QEMU_COROUTINE_H */ diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c index 846ff9167f..2a66fc1467 100644 --- a/util/qemu-coroutine-lock.c +++ b/util/qemu-coroutine-lock.c @@ -40,13 +40,13 @@ void qemu_co_queue_init(CoQueue *queue) QSIMPLEQ_INIT(&queue->entries); } -void coroutine_fn qemu_co_queue_wait(CoQueue *queue, CoMutex *mutex) +void coroutine_fn qemu_co_queue_wait_impl(CoQueue *queue, QemuLockable *lock) { Coroutine *self = qemu_coroutine_self(); QSIMPLEQ_INSERT_TAIL(&queue->entries, self, co_queue_next); - if (mutex) { - qemu_co_mutex_unlock(mutex); + if (lock) { + qemu_lockable_unlock(lock); } /* There is no race condition here. Other threads will call @@ -60,9 +60,11 @@ void coroutine_fn qemu_co_queue_wait(CoQueue *queue, CoMutex *mutex) /* TODO: OSv implements wait morphing here, where the wakeup * primitive automatically places the woken coroutine on the * mutex's queue. This avoids the thundering herd effect. + * This could be implemented for CoMutexes, but not really for + * other cases of QemuLockable. */ - if (mutex) { - qemu_co_mutex_lock(mutex); + if (lock) { + qemu_lockable_lock(lock); } }