Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751688AbaJ0Sjy (ORCPT ); Mon, 27 Oct 2014 14:39:54 -0400 Received: from aserp1040.oracle.com ([141.146.126.69]:37248 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751204AbaJ0Sjw (ORCPT ); Mon, 27 Oct 2014 14:39:52 -0400 Date: Mon, 27 Oct 2014 14:39:39 -0400 From: Konrad Rzeszutek Wilk To: Boris Ostrovsky Cc: david.vrabel@citrix.com, stefano.stabellini@eu.citrix.com, jun.nakajima@intel.com, linux-kernel@vger.kernel.org, xen-devel@lists.xen.org Subject: Re: [PATCH v2 2/2] xen/pci: Use APIC directly when APIC virtualization is supported by hardware Message-ID: <20141027183939.GB13854@laptop.dumpdata.com> References: <1414425953-13987-1-git-send-email-boris.ostrovsky@oracle.com> <1414425953-13987-3-git-send-email-boris.ostrovsky@oracle.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1414425953-13987-3-git-send-email-boris.ostrovsky@oracle.com> User-Agent: Mutt/1.5.23 (2014-03-12) X-Source-IP: ucsinet21.oracle.com [156.151.31.93] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Oct 27, 2014 at 12:05:53PM -0400, Boris Ostrovsky wrote: > When hardware supports APIC/x2APIC virtualization we don't need to use pirqs > for MSI handling and instead use APIC since most APIC accesses (MMIO or MSR) > will now be processed without VMEXITs. > > As an example, netperf on the original code produces this profile > (collected wih 'xentrace -e 0x0008ffff -T 5'): > > 342 cpu_change > 260 CPUID > 34638 HLT > 64067 INJ_VIRQ > 28374 INTR > 82733 INTR_WINDOW > 10 NPF > 24337 TRAP > 370610 vlapic_accept_pic_intr > 307528 VMENTRY > 307527 VMEXIT > 140998 VMMCALL > 127 wrap_buffer > > After applying this patch the same test shows > > 230 cpu_change > 260 CPUID > 36542 HLT > 174 INJ_VIRQ > 27250 INTR > 222 INTR_WINDOW > 20 NPF > 24999 TRAP > 381812 vlapic_accept_pic_intr > 166480 VMENTRY > 166479 VMEXIT > 77208 VMMCALL > 81 wrap_buffer > > ApacheBench results (ab -n 10000 -c 200) improve by about 10% What type of NIC did you use for this? Thanks! > > Signed-off-by: Boris Ostrovsky > --- > arch/x86/include/asm/xen/cpuid.h | 91 ++++++++++++++++++++++++++++++++++++++ > arch/x86/pci/xen.c | 9 ++++ > 2 files changed, 100 insertions(+), 0 deletions(-) > create mode 100644 arch/x86/include/asm/xen/cpuid.h > > diff --git a/arch/x86/include/asm/xen/cpuid.h b/arch/x86/include/asm/xen/cpuid.h > new file mode 100644 > index 0000000..0d809e9 > --- /dev/null > +++ b/arch/x86/include/asm/xen/cpuid.h > @@ -0,0 +1,91 @@ > +/****************************************************************************** > + * arch-x86/cpuid.h > + * > + * CPUID interface to Xen. > + * > + * Permission is hereby granted, free of charge, to any person obtaining a copy > + * of this software and associated documentation files (the "Software"), to > + * deal in the Software without restriction, including without limitation the > + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or > + * sell copies of the Software, and to permit persons to whom the Software is > + * furnished to do so, subject to the following conditions: > + * > + * The above copyright notice and this permission notice shall be included in > + * all copies or substantial portions of the Software. > + * > + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR > + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, > + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE > + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER > + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING > + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER > + * DEALINGS IN THE SOFTWARE. > + * > + * Copyright (c) 2007 Citrix Systems, Inc. > + * > + * Authors: > + * Keir Fraser > + */ > + > +#ifndef __XEN_PUBLIC_ARCH_X86_CPUID_H__ > +#define __XEN_PUBLIC_ARCH_X86_CPUID_H__ > + > +/* > + * For compatibility with other hypervisor interfaces, the Xen cpuid leaves > + * can be found at the first otherwise unused 0x100 aligned boundary starting > + * from 0x40000000. > + * > + * e.g If viridian extensions are enabled for an HVM domain, the Xen cpuid > + * leaves will start at 0x40000100 > + */ > + > +#define XEN_CPUID_FIRST_LEAF 0x40000000 > +#define XEN_CPUID_LEAF(i) (XEN_CPUID_FIRST_LEAF + (i)) > + > +/* > + * Leaf 1 (0x40000x00) > + * EAX: Largest Xen-information leaf. All leaves up to an including @EAX > + * are supported by the Xen host. > + * EBX-EDX: "XenVMMXenVMM" signature, allowing positive identification > + * of a Xen host. > + */ > +#define XEN_CPUID_SIGNATURE_EBX 0x566e6558 /* "XenV" */ > +#define XEN_CPUID_SIGNATURE_ECX 0x65584d4d /* "MMXe" */ > +#define XEN_CPUID_SIGNATURE_EDX 0x4d4d566e /* "nVMM" */ > + > +/* > + * Leaf 2 (0x40000x01) > + * EAX[31:16]: Xen major version. > + * EAX[15: 0]: Xen minor version. > + * EBX-EDX: Reserved (currently all zeroes). > + */ > + > +/* > + * Leaf 3 (0x40000x02) > + * EAX: Number of hypercall transfer pages. This register is always guaranteed > + * to specify one hypercall page. > + * EBX: Base address of Xen-specific MSRs. > + * ECX: Features 1. Unused bits are set to zero. > + * EDX: Features 2. Unused bits are set to zero. > + */ > + > +/* Does the host support MMU_PT_UPDATE_PRESERVE_AD for this guest? */ > +#define _XEN_CPUID_FEAT1_MMU_PT_UPDATE_PRESERVE_AD 0 > +#define XEN_CPUID_FEAT1_MMU_PT_UPDATE_PRESERVE_AD (1u<<0) > + > +/* > + * Leaf 5 (0x40000x04) > + * HVM-specific features > + */ > + > +/* EAX Features */ > +/* Virtualized APIC registers */ > +#define XEN_HVM_CPUID_APIC_ACCESS_VIRT (1u << 0) > +/* Virtualized x2APIC accesses */ > +#define XEN_HVM_CPUID_X2APIC_VIRT (1u << 1) > +/* Memory mapped from other domains has valid IOMMU entries */ > +#define XEN_HVM_CPUID_IOMMU_MAPPINGS (1u << 2) > + > +#define XEN_CPUID_MAX_NUM_LEAVES 4 > + > +#endif /* __XEN_PUBLIC_ARCH_X86_CPUID_H__ */ > diff --git a/arch/x86/pci/xen.c b/arch/x86/pci/xen.c > index 1370716..d768371 100644 > --- a/arch/x86/pci/xen.c > +++ b/arch/x86/pci/xen.c > @@ -23,6 +23,7 @@ > #include > #include > #include > +#include > #include > > static int xen_pcifront_enable_irq(struct pci_dev *dev) > @@ -434,6 +435,14 @@ int __init pci_xen_init(void) > #ifdef CONFIG_PCI_MSI > void __init xen_msi_init(void) > { > + if (!disable_apic) { > + uint32_t eax = cpuid_eax(xen_cpuid_base() + 4); > + > + if (((eax & XEN_HVM_CPUID_X2APIC_VIRT) && x2apic_mode) || > + ((eax & XEN_HVM_CPUID_APIC_ACCESS_VIRT) && cpu_has_apic)) > + return; > + } > + > x86_msi.setup_msi_irqs = xen_hvm_setup_msi_irqs; > x86_msi.teardown_msi_irq = xen_teardown_msi_irq; > } > -- > 1.7.1 > -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/