From nobody Fri May 3 15:59:59 2024 Delivered-To: importer@patchew.org Received-SPF: pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) client-ip=208.118.235.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Authentication-Results: mx.zohomail.com; spf=pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org Return-Path: Received: from lists.gnu.org (208.118.235.17 [208.118.235.17]) by mx.zohomail.com with SMTPS id 1509447452422100.75352092873175; Tue, 31 Oct 2017 03:57:32 -0700 (PDT) Received: from localhost ([::1]:44730 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1e9UEY-0005AC-Kq for importer@patchew.org; Tue, 31 Oct 2017 06:57:26 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:58047) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1e9U6I-0006BC-Mw for qemu-devel@nongnu.org; Tue, 31 Oct 2017 06:48:55 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1e9U6F-0004ur-L9 for qemu-devel@nongnu.org; Tue, 31 Oct 2017 06:48:54 -0400 Received: from mail.ispras.ru ([83.149.199.45]:39952) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1e9U6F-0004uS-Cr for qemu-devel@nongnu.org; Tue, 31 Oct 2017 06:48:51 -0400 Received: from [127.0.1.1] (unknown [85.142.117.226]) by mail.ispras.ru (Postfix) with ESMTPSA id A44B554006A; Tue, 31 Oct 2017 13:48:50 +0300 (MSK) From: Pavel Dovgalyuk To: qemu-devel@nongnu.org Date: Tue, 31 Oct 2017 13:48:51 +0300 Message-ID: <20171031104851.3079.44019.stgit@pasha-VirtualBox> In-Reply-To: <20171031104741.3079.26584.stgit@pasha-VirtualBox> References: <20171031104741.3079.26584.stgit@pasha-VirtualBox> User-Agent: StGit/0.17.1-dirty MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable X-detected-operating-system: by eggs.gnu.org: GNU/Linux 3.x [fuzzy] X-Received-From: 83.149.199.45 Subject: [Qemu-devel] [RFCPATCH13/20] cpus: only take BQL for sleeping threads 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: dovgaluk@ispras.ru Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail: RSF_0 Z_629925259 SPT_0 From: Alex Benn=C3=A9e Now the only real need to hold the BQL is for when we sleep on the cpu->halt conditional. The lock is actually dropped while the thread sleeps so the actual window for contention is pretty small. This also means we can remove the special case hack for exclusive work and simply declare that work no longer has an implicit BQL held. This isn't a major problem async work is generally only changing things in the context of its own vCPU. If it needs to work across vCPUs it should be using the exclusive mechanism or possibly taking the lock itself. Signed-off-by: Alex Benn=C3=A9e Signed-off-by: Pavel Dovgalyuk --- cpus-common.c | 13 +++++-------- cpus.c | 10 ++++------ 2 files changed, 9 insertions(+), 14 deletions(-) diff --git a/cpus-common.c b/cpus-common.c index 59f751e..64661c3 100644 --- a/cpus-common.c +++ b/cpus-common.c @@ -310,6 +310,11 @@ void async_safe_run_on_cpu(CPUState *cpu, run_on_cpu_f= unc func, queue_work_on_cpu(cpu, wi); } =20 +/* Work items run outside of the BQL. This is essential for avoiding a + * deadlock for exclusive work but also applies to non-exclusive work. + * If the work requires cross-vCPU changes then it should use the + * exclusive mechanism. + */ void process_queued_cpu_work(CPUState *cpu) { struct qemu_work_item *wi; @@ -327,17 +332,9 @@ void process_queued_cpu_work(CPUState *cpu) } qemu_mutex_unlock(&cpu->work_mutex); if (wi->exclusive) { - /* Running work items outside the BQL avoids the following dea= dlock: - * 1) start_exclusive() is called with the BQL taken while ano= ther - * CPU is running; 2) cpu_exec in the other CPU tries to takes= the - * BQL, so it goes to sleep; start_exclusive() is sleeping too= , so - * neither CPU can proceed. - */ - qemu_mutex_unlock_iothread(); start_exclusive(); wi->func(cpu, wi->data); end_exclusive(); - qemu_mutex_lock_iothread(); } else { wi->func(cpu, wi->data); } diff --git a/cpus.c b/cpus.c index efde5c1..de6dfce 100644 --- a/cpus.c +++ b/cpus.c @@ -1127,31 +1127,29 @@ static bool qemu_tcg_should_sleep(CPUState *cpu) =20 static void qemu_tcg_wait_io_event(CPUState *cpu) { - qemu_mutex_lock_iothread(); =20 while (qemu_tcg_should_sleep(cpu)) { + qemu_mutex_lock_iothread(); stop_tcg_kick_timer(); qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); + qemu_mutex_unlock_iothread(); } =20 start_tcg_kick_timer(); =20 qemu_wait_io_event_common(cpu); - - qemu_mutex_unlock_iothread(); } =20 static void qemu_kvm_wait_io_event(CPUState *cpu) { - qemu_mutex_lock_iothread(); =20 while (cpu_thread_is_idle(cpu)) { + qemu_mutex_lock_iothread(); qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); + qemu_mutex_unlock_iothread(); } =20 qemu_wait_io_event_common(cpu); - - qemu_mutex_unlock_iothread(); } =20 static void *qemu_kvm_cpu_thread_fn(void *arg)