From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757227Ab2DFMk0 (ORCPT ); Fri, 6 Apr 2012 08:40:26 -0400 Received: from mail-wi0-f178.google.com ([209.85.212.178]:46074 "EHLO mail-wi0-f178.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757210Ab2DFMkX (ORCPT ); Fri, 6 Apr 2012 08:40:23 -0400 From: Ido Yariv To: linux-kernel@vger.kernel.org, Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" Cc: Ravikiran Thirumalai , Shai Fultheim , Ido Yariv Subject: [PATCH 2/2] vsmp: Ignore IOAPIC IRQ affinity if possible Date: Fri, 6 Apr 2012 15:39:47 +0300 Message-Id: <1333715987-26059-2-git-send-email-ido@wizery.com> X-Mailer: git-send-email 1.7.7.6 In-Reply-To: <1333715987-26059-1-git-send-email-ido@wizery.com> References: <1333715987-26059-1-git-send-email-ido@wizery.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Ravikiran Thirumalai vSMP can route interrupts more optimally based on internal knowledge the OS does not have. In order to support this optimization, all CPUs must be able to handle all possible IOAPIC interrupts. Fix this by setting the vector allocation domain for all CPUs and by enabling this feature in vSMP. Signed-off-by: Ravikiran Thirumalai Signed-off-by: Shai Fultheim [ido@wizery.com: rebased, simplified, and reworded the commit message] Signed-off-by: Ido Yariv --- arch/x86/kernel/apic/probe_64.c | 10 ++++++++++ arch/x86/kernel/vsmp_64.c | 15 +++++++++++---- 2 files changed, 21 insertions(+), 4 deletions(-) diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c index 3fe9866..0f96703 100644 --- a/arch/x86/kernel/apic/probe_64.c +++ b/arch/x86/kernel/apic/probe_64.c @@ -29,6 +29,15 @@ static int apicid_phys_pkg_id(int initial_apic_id, int index_msb) } /* + * In vSMP, all cpus should be capable of handling interrupts, regardless of + * the apic used. + */ +static void fill_vector_allocation_domain(int cpu, struct cpumask *retmask) +{ + cpumask_setall(retmask); +} + +/* * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode. */ void __init default_setup_apic_routing(void) @@ -51,6 +60,7 @@ void __init default_setup_apic_routing(void) if (is_vsmp_box()) { /* need to update phys_pkg_id */ apic->phys_pkg_id = apicid_phys_pkg_id; + apic->vector_allocation_domain = fill_vector_allocation_domain; } } diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c index 61571fd..c91b08f 100644 --- a/arch/x86/kernel/vsmp_64.c +++ b/arch/x86/kernel/vsmp_64.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include @@ -95,6 +96,13 @@ static void __init set_vsmp_pv_ops(void) ctl = readl(address + 4); printk(KERN_INFO "vSMP CTL: capabilities:0x%08x control:0x%08x\n", cap, ctl); + /* + * If possible, let the vSMP foundation route the interrupt optimally */ + if (cap & ctl & BIT(8)) { + ctl &= ~BIT(8); + no_irq_affinity = 1; + } + if (cap & ctl & (1 << 4)) { /* Setup irq ops and turn on vSMP IRQ fastpath handling */ pv_irq_ops.irq_disable = PV_CALLEE_SAVE(vsmp_irq_disable); @@ -102,12 +110,11 @@ static void __init set_vsmp_pv_ops(void) pv_irq_ops.save_fl = PV_CALLEE_SAVE(vsmp_save_fl); pv_irq_ops.restore_fl = PV_CALLEE_SAVE(vsmp_restore_fl); pv_init_ops.patch = vsmp_patch; - ctl &= ~(1 << 4); - writel(ctl, address + 4); - ctl = readl(address + 4); - printk(KERN_INFO "vSMP CTL: control set to:0x%08x\n", ctl); } + writel(ctl, address + 4); + ctl = readl(address + 4); + printk(KERN_INFO "vSMP CTL: control set to:0x%08x\n", ctl); early_iounmap(address, 8); } -- 1.7.7.6