Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753324Ab2KGQQM (ORCPT ); Wed, 7 Nov 2012 11:16:12 -0500 Received: from userp1040.oracle.com ([156.151.31.81]:31825 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751719Ab2KGQQL (ORCPT ); Wed, 7 Nov 2012 11:16:11 -0500 Date: Wed, 7 Nov 2012 11:03:01 -0500 From: Konrad Rzeszutek Wilk To: "Liu, Jinsong" Cc: Stefano Stabellini , Konrad Rzeszutek Wilk , "xen-devel@lists.xen.org" , "linux-kernel@vger.kernel.org" , Jan Beulich Subject: Re: [Xen-devel] [PATCH 1/2] Xen acpi pad implement Message-ID: <20121107160301.GF18615@phenom.dumpdata.com> References: <20121026201409.GF2708@phenom.dumpdata.com> <20121102164937.GC27213@konrad-lan.dumpdata.com> <20121106164303.GO9966@phenom.dumpdata.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) 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 Content-Length: 10668 Lines: 348 On Wed, Nov 07, 2012 at 12:58:19PM +0000, Liu, Jinsong wrote: > >>> > >>> If it is generic ACPI code, than it can depend only on ACPI. > >>> If it is ACPI code that contains X86 specific info, than it needs to > >>> depend on X86 too. > >> > >> No x86 specific so let's depend on ACPI. > > > > Huh? This feature is x86 specific isn't it? I mean it is in the ACPI > > spec, but only x86 does it right? > > OK, updated w/ ACPI & X86 dependency, and per other comments. Ok. Could you send all the patchset in one go. I am bit lost of which one to review as none of them seem to have the version information [vX: Did this yyy] right above your name or in the title: [PATCH vX 1/2]: yyyy anyhow, some comments below. > > Thanks, > Jinsong > > ================= > > >From d5f462cc8029863955a574f1541bece72aafb986 Mon Sep 17 00:00:00 2001 > From: Liu, Jinsong > Date: Thu, 8 Nov 2012 04:29:07 +0800 > Subject: [PATCH 1/2] Xen acpi pad implement The title should be: xen/acpi: ACPI PAD driver > > PAD is acpi Processor Aggregator Device which provides a control point > that enables the platform to perform specific processor configuration > and control that applies to all processors in the platform. > > This patch is to implement Xen acpi pad logic. When running under Xen > virt platform, native pad driver would not work. Instead Xen pad driver, > a self-contained and thin logic level, would take over acpi pad logic. > > When acpi pad notify OSPM, xen pad logic intercept and parse _PUR object > to get the expected idle cpu number, and then hypercall to hypervisor. > Xen hypervisor would then do the rest work, say, core parking, to idle > specific number of cpus on its own policy. > > Signed-off-by: Jan Beulich > Signed-off-by: Liu Jinsong > --- > drivers/xen/Makefile | 3 +- > drivers/xen/xen-acpi-pad.c | 186 ++++++++++++++++++++++++++++++++++++++ > include/xen/interface/platform.h | 17 ++++ > include/xen/interface/version.h | 15 +++ > 4 files changed, 220 insertions(+), 1 deletions(-) > create mode 100644 drivers/xen/xen-acpi-pad.c > > diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile > index 0e86370..3c39717 100644 > --- a/drivers/xen/Makefile > +++ b/drivers/xen/Makefile > @@ -10,7 +10,8 @@ CFLAGS_features.o := $(nostackp) > > dom0-$(CONFIG_PCI) += pci.o > dom0-$(CONFIG_USB_SUPPORT) += dbgp.o > -dom0-$(CONFIG_ACPI) += acpi.o > +dom0-$(CONFIG_ACPI) += acpi.o $(xen-pad-y) > +xen-pad-$(CONFIG_X86) += xen-acpi-pad.o > dom0-$(CONFIG_X86) += pcpu.o > obj-$(CONFIG_XEN_DOM0) += $(dom0-y) > obj-$(CONFIG_BLOCK) += biomerge.o > diff --git a/drivers/xen/xen-acpi-pad.c b/drivers/xen/xen-acpi-pad.c > new file mode 100644 > index 0000000..f02d3ff > --- /dev/null > +++ b/drivers/xen/xen-acpi-pad.c > @@ -0,0 +1,186 @@ > +/* > + * xen-acpi-pad.c - Xen pad interface > + * > + * Copyright (c) 2012, Intel Corporation. > + * Author: Liu, Jinsong > + * > + * This program is free software; you can redistribute it and/or modify it > + * under the terms and conditions of the GNU General Public License, > + * version 2, as published by the Free Software Foundation. > + * > + * This program is distributed in the hope it will be useful, but WITHOUT > + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or > + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for > + * more details. > + * > + * You should have received a copy of the GNU General Public License along with > + * this program; if not, write to the Free Software Foundation, Inc., > + * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. No address pls. > + * > + */ > + > +#include > +#include > +#include > +#include > +#include > +#include > + > +#define ACPI_PROCESSOR_AGGREGATOR_CLASS "acpi_pad" > +#define ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME "Processor Aggregator" > +#define ACPI_PROCESSOR_AGGREGATOR_NOTIFY 0x80 > + > +static DEFINE_MUTEX(xen_cpu_lock); > + > +static int xen_acpi_pad_idle_cpus(unsigned int idle_nums) > +{ > + struct xen_platform_op op; > + > + op.cmd = XENPF_core_parking; > + op.u.core_parking.type = XEN_CORE_PARKING_SET; > + op.u.core_parking.idle_nums = idle_nums; > + > + return HYPERVISOR_dom0_op(&op); > +} > + > +static int xen_acpi_pad_idle_cpus_num(void) > +{ > + struct xen_platform_op op; > + > + op.cmd = XENPF_core_parking; > + op.u.core_parking.type = XEN_CORE_PARKING_GET; > + > + return HYPERVISOR_dom0_op(&op) > + ?: op.u.core_parking.idle_nums; > +} > + > +/* > + * Query firmware how many CPUs should be idle > + * return -1 on failure > + */ > +static int acpi_pad_pur(acpi_handle handle) > +{ > + struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL}; > + union acpi_object *package; > + int num = -1; > + > + if (ACPI_FAILURE(acpi_evaluate_object(handle, "_PUR", NULL, &buffer))) > + return num; > + > + if (!buffer.length || !buffer.pointer) > + return num; > + > + package = buffer.pointer; > + > + if (package->type == ACPI_TYPE_PACKAGE && > + package->package.count == 2 && > + package->package.elements[0].integer.value == 1) /* rev 1 */ > + > + num = package->package.elements[1].integer.value; > + > + kfree(buffer.pointer); > + return num; > +} > + > +/* Notify firmware how many CPUs are idle */ > +static void acpi_pad_ost(acpi_handle handle, int stat, > + uint32_t idle_nums) > +{ > + union acpi_object params[3] = { > + {.type = ACPI_TYPE_INTEGER,}, > + {.type = ACPI_TYPE_INTEGER,}, > + {.type = ACPI_TYPE_BUFFER,}, > + }; > + struct acpi_object_list arg_list = {3, params}; > + > + params[0].integer.value = ACPI_PROCESSOR_AGGREGATOR_NOTIFY; > + params[1].integer.value = stat; > + params[2].buffer.length = 4; > + params[2].buffer.pointer = (void *)&idle_nums; > + acpi_evaluate_object(handle, "_OST", &arg_list, NULL); > +} > + > +static void acpi_pad_handle_notify(acpi_handle handle) > +{ > + int idle_nums; > + > + mutex_lock(&xen_cpu_lock); > + idle_nums = acpi_pad_pur(handle); > + if (idle_nums < 0) { > + mutex_unlock(&xen_cpu_lock); > + return; > + } > + > + idle_nums = xen_acpi_pad_idle_cpus(idle_nums) > + ?: xen_acpi_pad_idle_cpus_num(); > + if (idle_nums >= 0) > + acpi_pad_ost(handle, 0, idle_nums); > + mutex_unlock(&xen_cpu_lock); > +} > + > +static void acpi_pad_notify(acpi_handle handle, u32 event, > + void *data) > +{ > + switch (event) { > + case ACPI_PROCESSOR_AGGREGATOR_NOTIFY: > + acpi_pad_handle_notify(handle); > + break; > + default: > + pr_warn("Unsupported event [0x%x]\n", event); > + break; > + } > +} > + > +static int acpi_pad_add(struct acpi_device *device) > +{ > + acpi_status status; > + > + strcpy(acpi_device_name(device), ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME); > + strcpy(acpi_device_class(device), ACPI_PROCESSOR_AGGREGATOR_CLASS); > + > + status = acpi_install_notify_handler(device->handle, > + ACPI_DEVICE_NOTIFY, acpi_pad_notify, device); > + if (ACPI_FAILURE(status)) > + return -ENODEV; > + > + return 0; > +} > + > +static int acpi_pad_remove(struct acpi_device *device, > + int type) > +{ > + mutex_lock(&xen_cpu_lock); > + xen_acpi_pad_idle_cpus(0); > + mutex_unlock(&xen_cpu_lock); > + > + acpi_remove_notify_handler(device->handle, > + ACPI_DEVICE_NOTIFY, acpi_pad_notify); > + return 0; > +} > + > +static const struct acpi_device_id pad_device_ids[] = { > + {"ACPI000C", 0}, > + {"", 0}, > +}; > + > +static struct acpi_driver acpi_pad_driver = { > + .name = "processor_aggregator", > + .class = ACPI_PROCESSOR_AGGREGATOR_CLASS, > + .ids = pad_device_ids, > + .ops = { > + .add = acpi_pad_add, > + .remove = acpi_pad_remove, > + }, > +}; > + > +static int __init xen_acpi_pad_init(void) > +{ > + /* Only DOM0 and Xen4.2 or later support Xen acpi pad */ > + if (!xen_initial_domain() || > + !xen_running_on_version_or_later(4, 2)) > + return -ENODEV; Please restructure this. Just make it if (!xen_initial_domain()) return -ENODEV; if (!xen_running..) return -ENODEV; I know the compiler will end up with code that looks exactly the same but it is easier to read than the '!' and '||' logic. > + > + return acpi_bus_register_driver(&acpi_pad_driver); > +} > +subsys_initcall(xen_acpi_pad_init); > + > diff --git a/include/xen/interface/platform.h b/include/xen/interface/platform.h > index 4755b5f..5e36932 100644 > --- a/include/xen/interface/platform.h > +++ b/include/xen/interface/platform.h > @@ -324,6 +324,22 @@ struct xenpf_cpu_ol { > }; > DEFINE_GUEST_HANDLE_STRUCT(xenpf_cpu_ol); > > +/* > + * CMD 58 and 59 are reserved for cpu hotadd and memory hotadd, > + * which are already occupied at Xen hypervisor side. > + */ > +#define XENPF_core_parking 60 > +struct xenpf_core_parking { > + /* IN variables */ > +#define XEN_CORE_PARKING_SET 1 > +#define XEN_CORE_PARKING_GET 2 > + uint32_t type; > + /* IN variables: set cpu nums expected to be idled */ > + /* OUT variables: get cpu nums actually be idled */ > + uint32_t idle_nums; > +}; > +DEFINE_GUEST_HANDLE_STRUCT(xenpf_core_parking); > + > struct xen_platform_op { > uint32_t cmd; > uint32_t interface_version; /* XENPF_INTERFACE_VERSION */ > @@ -341,6 +357,7 @@ struct xen_platform_op { > struct xenpf_set_processor_pminfo set_pminfo; > struct xenpf_pcpuinfo pcpu_info; > struct xenpf_cpu_ol cpu_ol; > + struct xenpf_core_parking core_parking; > uint8_t pad[128]; > } u; > }; > diff --git a/include/xen/interface/version.h b/include/xen/interface/version.h > index 7ff6498..96d8d3d 100644 > --- a/include/xen/interface/version.h > +++ b/include/xen/interface/version.h > @@ -63,4 +63,19 @@ struct xen_feature_info { > /* arg == xen_domain_handle_t. */ > #define XENVER_guest_handle 8 > > +/* Check if running on Xen version (major, minor) or later */ > +static inline bool > +xen_running_on_version_or_later(unsigned int major, unsigned int minor) > +{ > + unsigned int version; > + > + if (!xen_domain()) > + return false; > + > + version = HYPERVISOR_xen_version(XENVER_version, NULL); > + if ((((version >> 16) == major) && ((version & 0xffff) >= minor)) || > + ((version >> 16) > major)) > + return true; > + return false; > +} > #endif /* __XEN_PUBLIC_VERSION_H__ */ > -- > 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/