Re: [PATCH v2 2/2] xen/pci: Use APIC directly when APIC virtualization is supported by hardware
From: Konrad Rzeszutek Wilk
Date: Mon Oct 27 2014 - 14:39:57 EST
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 <boris.ostrovsky@xxxxxxxxxx>
> ---
> 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 <keir@xxxxxxx>
> + */
> +
> +#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 <xen/features.h>
> #include <xen/events.h>
> #include <asm/xen/pci.h>
> +#include <asm/xen/cpuid.h>
> #include <asm/i8259.h>
>
> 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@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/