From nobody Fri Jan 2 22:07:18 2026 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 8CFF3E95A62 for ; Sat, 7 Oct 2023 17:11:09 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1344146AbjJGRLI (ORCPT ); Sat, 7 Oct 2023 13:11:08 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:37150 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1344021AbjJGRLD (ORCPT ); Sat, 7 Oct 2023 13:11:03 -0400 Received: from galois.linutronix.de (Galois.linutronix.de [193.142.43.55]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 23D7483; Sat, 7 Oct 2023 10:10:59 -0700 (PDT) Date: Sat, 07 Oct 2023 17:10:57 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linutronix.de; s=2020; t=1696698658; h=from:from:sender:sender:reply-to:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=rEKjWkFLMKb1KqE8X1Cj8gxs9qu9BQZVwHibia4c8QI=; b=k4v9dR4B81+um3HdySa2CnsSEXB3VfurPXy/Qv7TiKW5M24xL3Hytk432iNEgUzVxKpv2M sfqwT/e/mybsilrQaAT5J85/G/QfSjeCxxDJzfDqtvnSx5k9yWiXSVFTZ9oJHf8kfkLKZ9 kMUV+x0Hgh0zPztmrnpU0jwd3PHQOS5mTwx04qwTxDiQgg7415pI8oVGr3lUl4Zcl/FEPe nkwB3Y0q1+3YIN4OPq4LhE/pJQYB6A92V9mn8mNNfaue0TEs4VR1+D8zv6IYnkR5LQy6gU HOVuGGtFZFHt2pxjKzd8g2nl1sV8GNePmxWbmVdhMWZfLzloSVnlycCNsi9ldA== DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=linutronix.de; s=2020e; t=1696698658; h=from:from:sender:sender:reply-to:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=rEKjWkFLMKb1KqE8X1Cj8gxs9qu9BQZVwHibia4c8QI=; b=CyLOWe6PN45IXaKHFUwpwQ9qXqrGnv4wYl3e5QDr1XJt0a7GfhezY6P3eXY9Bxj/P7Q/ia QIlopJx8JTnWXpBA== From: "tip-bot2 for Waiman Long" Sender: tip-bot2@linutronix.de Reply-to: linux-kernel@vger.kernel.org To: linux-tip-commits@vger.kernel.org Subject: [tip: sched/core] intel_idle: Add ibrs_off module parameter to force-disable IBRS Cc: Waiman Long , Ingo Molnar , "Rafael J. Wysocki" , Linus Torvalds , x86@kernel.org, linux-kernel@vger.kernel.org In-Reply-To: <20230727184600.26768-5-longman@redhat.com> References: <20230727184600.26768-5-longman@redhat.com> MIME-Version: 1.0 Message-ID: <169669865746.3135.6564791936984807529.tip-bot2@tip-bot2> Robot-ID: Robot-Unsubscribe: Contact to get blacklisted from these emails Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org The following commit has been merged into the sched/core branch of tip: Commit-ID: aa1567a7e6440b8c3af4b0d8a8219d8fc5028c5f Gitweb: https://git.kernel.org/tip/aa1567a7e6440b8c3af4b0d8a8219d8fc= 5028c5f Author: Waiman Long AuthorDate: Thu, 27 Jul 2023 14:46:00 -04:00 Committer: Ingo Molnar CommitterDate: Sat, 07 Oct 2023 11:33:28 +02:00 intel_idle: Add ibrs_off module parameter to force-disable IBRS Commit bf5835bcdb96 ("intel_idle: Disable IBRS during long idle") disables IBRS when the cstate is 6 or lower. However, there are some use cases where a customer may want to use max_cstate=3D1 to lower latency. Such use cases will suffer from the performance degradation caused by the enabling of IBRS in the sibling idle thread. Add a "ibrs_off" module parameter to force disable IBRS and the CPUIDLE_FLAG_IRQ_ENABLE flag if set. In the case of a Skylake server with max_cstate=3D1, this new ibrs_off option will likely increase the IRQ response latency as IRQ will now be disabled. When running SPECjbb2015 with cstates set to C1 on a Skylake system. First test when the kernel is booted with: "intel_idle.ibrs_off": max-jOPS =3D 117828, critical-jOPS =3D 66047 Then retest when the kernel is booted without the "intel_idle.ibrs_off" added: max-jOPS =3D 116408, critical-jOPS =3D 58958 That means booting with "intel_idle.ibrs_off" improves performance by: max-jOPS: +1.2%, which could be considered noise range. critical-jOPS: +12%, which is definitely a solid improvement. The admin-guide/pm/intel_idle.rst file is updated to add a description about the new "ibrs_off" module parameter. Signed-off-by: Waiman Long Signed-off-by: Ingo Molnar Acked-by: Rafael J. Wysocki Cc: Linus Torvalds Link: https://lore.kernel.org/r/20230727184600.26768-5-longman@redhat.com --- Documentation/admin-guide/pm/intel_idle.rst | 17 ++++++++++++++++- drivers/idle/intel_idle.c | 11 ++++++++++- 2 files changed, 26 insertions(+), 2 deletions(-) diff --git a/Documentation/admin-guide/pm/intel_idle.rst b/Documentation/ad= min-guide/pm/intel_idle.rst index b799a43..39bd6ec 100644 --- a/Documentation/admin-guide/pm/intel_idle.rst +++ b/Documentation/admin-guide/pm/intel_idle.rst @@ -170,7 +170,7 @@ and ``idle=3Dnomwait``. If any of them is present in t= he kernel command line, the ``MWAIT`` instruction is not allowed to be used, so the initialization of ``intel_idle`` will fail. =20 -Apart from that there are four module parameters recognized by ``intel_idl= e`` +Apart from that there are five module parameters recognized by ``intel_idl= e`` itself that can be set via the kernel command line (they cannot be updated= via sysfs, so that is the only way to change their values). =20 @@ -216,6 +216,21 @@ are ignored). The idle states disabled this way can be enabled (on a per-CPU basis) from= user space via ``sysfs``. =20 +The ``ibrs_off`` module parameter is a boolean flag (defaults to +false). If set, it is used to control if IBRS (Indirect Branch Restricted +Speculation) should be turned off when the CPU enters an idle state. +This flag does not affect CPUs that use Enhanced IBRS which can remain +on with little performance impact. + +For some CPUs, IBRS will be selected as mitigation for Spectre v2 and Retb= leed +security vulnerabilities by default. Leaving the IBRS mode on while idlin= g may +have a performance impact on its sibling CPU. The IBRS mode will be turne= d off +by default when the CPU enters into a deep idle state, but not in some +shallower ones. Setting the ``ibrs_off`` module parameter will force the = IBRS +mode to off when the CPU is in any one of the available idle states. This= may +help performance of a sibling CPU at the expense of a slightly higher wake= up +latency for the idle CPU. + =20 .. _intel-idle-core-and-package-idle-states: =20 diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index 86ac9a4..dcda0af 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -68,6 +68,7 @@ static int max_cstate =3D CPUIDLE_STATE_MAX - 1; static unsigned int disabled_states_mask __read_mostly; static unsigned int preferred_states_mask __read_mostly; static bool force_irq_on __read_mostly; +static bool ibrs_off __read_mostly; =20 static struct cpuidle_device __percpu *intel_idle_cpuidle_devices; =20 @@ -1852,11 +1853,13 @@ static void state_update_enter_method(struct cpuidl= e_state *state, int cstate) } =20 if (cpu_feature_enabled(X86_FEATURE_KERNEL_IBRS) && - state->flags & CPUIDLE_FLAG_IBRS) { + ((state->flags & CPUIDLE_FLAG_IBRS) || ibrs_off)) { /* * IBRS mitigation requires that C-states are entered * with interrupts disabled. */ + if (ibrs_off && (state->flags & CPUIDLE_FLAG_IRQ_ENABLE)) + state->flags &=3D ~CPUIDLE_FLAG_IRQ_ENABLE; WARN_ON_ONCE(state->flags & CPUIDLE_FLAG_IRQ_ENABLE); state->enter =3D intel_idle_ibrs; return; @@ -2175,3 +2178,9 @@ MODULE_PARM_DESC(preferred_cstates, "Mask of preferre= d idle states"); * 'CPUIDLE_FLAG_INIT_XSTATE' and 'CPUIDLE_FLAG_IBRS' flags. */ module_param(force_irq_on, bool, 0444); +/* + * Force the disabling of IBRS when X86_FEATURE_KERNEL_IBRS is on and + * CPUIDLE_FLAG_IRQ_ENABLE isn't set. + */ +module_param(ibrs_off, bool, 0444); +MODULE_PARM_DESC(ibrs_off, "Disable IBRS when idle");