d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
6682323732
freebsd-nq/sys/dev/acpica/acpi.c

4328 lines
118 KiB
C
Raw Normal View History

Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*-
* Copyright (c) 2000 Takanori Watanabe <takawata@jp.freebsd.org>
* Copyright (c) 2000 Mitsuru IWASAKI <iwasaki@jp.freebsd.org>
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
* Copyright (c) 2000, 2001 Michael Smith
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
* Copyright (c) 2000 BSDi
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
Fix SCM ID's.
2005-03-02 09:22:34 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
#include "opt_acpi.h"
Add more fine-grained kernel options for NUMA support. VM_NUMA_ALLOC is used to enable use of domain-aware memory allocation in the virtual memory system. DEVICE_NUMA is used to enable affinity reporting for devices such as bus_get_domain(). MAXMEMDOM must still be set to a value greater than for any NUMA support to be effective. Note that 'cpuset -gd' always works if MAXMEMDOM is enabled and the system supports NUMA. Reviewed by: kib Differential Revision: https://reviews.freebsd.org/D5782
2016-04-09 13:58:04 +00:00
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
#include <sys/param.h>
Extract eventfilter declarations to sys/_eventfilter.h This allows replacing "sys/eventfilter.h" includes with "sys/_eventfilter.h" in other header files (e.g., sys/{bus,conf,cpu}.h) and reduces header pollution substantially. EVENTHANDLER_DECLARE and EVENTHANDLER_LIST_DECLAREs were moved out of .c files into appropriate headers (e.g., sys/proc.h, powernv/opal.h). As a side effect of reduced header pollution, many .c files and headers no longer contain needed definitions. The remainder of the patch addresses adding appropriate includes to fix those files. LOCK_DEBUG and LOCK_FILE_LINE_ARG are moved to sys/_lock.h, as required by sys/mutex.h since r326106 (but silently protected by header pollution prior to this change). No functional change (intended). Of course, any out of tree modules that relied on header pollution for sys/eventhandler.h, sys/lock.h, or sys/mutex.h inclusion need to be fixed. __FreeBSD_version has been bumped.
2019-05-20 00:38:23 +00:00
#include <sys/eventhandler.h>
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
#include <sys/kernel.h>
Undo part of the tangle of having sys/lock.h and sys/mutex.h included in other "system" header files. Also help the deprecation of lockmgr.h by making it a sub-include of sys/lock.h and removing sys/lockmgr.h form kernel .c files. Sort sys/*.h includes where possible in affected files. OK'ed by: bde (with reservations)
2001-05-01 08:13:21 +00:00
#include <sys/proc.h>
Allow non-privilaged user to retrive battery or AC line information. Reviewed by: rwatson
2003-02-15 01:46:22 +00:00
#include <sys/fcntl.h>
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
#include <sys/malloc.h>
Add missing <sys/module.h> includes
2004-05-30 20:08:47 +00:00
#include <sys/module.h>
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
#include <sys/bus.h>
#include <sys/conf.h>
#include <sys/ioccom.h>
#include <sys/reboot.h>
#include <sys/sysctl.h>
#include <sys/ctype.h>
Add S4BIOS sleep (BIOS hibernation) and DSDT overriding support. - Add S4BIOS sleep implementation. This will works well if MIB hw.acpi.s4bios is set (and of course BIOS supports it and hibernation is enabled correctly). - Add DSDT overriding support which is submitted by takawata originally. If loader tunable acpi_dsdt_load="YES" and DSDT file is set to acpi_dsdt_name (default DSDT file name is /boot/acpi_dsdt.aml), ACPI CA core loads DSDT from given file rather than BIOS memory block. DSDT file can be generated by iasl in ports/devel/acpicatools/. - Add new files so that we can add our proposed additional code to Intel ACPI CA into these files temporary. They will be removed when similar code is added into ACPI CA officially.
2001-11-06 15:00:30 +00:00
#include <sys/linker.h>
Some fix for the recent apm module changes. - Now that apm loadable module can inform its existence to other kernel components (e.g. i386/isa/clock.c:startrtclock()'s TCS hack). - Exchange priority of SI_SUB_CPU and SI_SUB_KLD for above purpose. - Add simple arbitration mechanism for APM vs. ACPI. This prevents the kernel enables both of them. - Remove obsolete `#ifdef DEV_APM' related code. - Add abstracted interface for Powermanagement operations. Public apm(4) functions, such as apm_suspend(), should be replaced new interfaces. Currently only power_pm_suspend (successor of apm_suspend) is implemented. Reviewed by: peter, arch@ and audit@
2001-11-01 16:34:07 +00:00
#include <sys/power.h>
Use sbufs instead of snprintf for parsing debug strings.
2003-12-09 06:29:57 +00:00
#include <sys/sbuf.h>
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
#include <sys/sched.h>
Move the poweroff handler to a separate function. Make sure it is run on the boot processor (cpuid == 0). Some chipsets do not power off the system if the shutdown handler runs on an AP.
2004-03-19 07:05:01 +00:00
#include <sys/smp.h>
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
#include <sys/timetc.h>
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Extend the support for PCI-e memory mapped configuration space access: - Rename pciereg_cfgopen() to pcie_cfgregopen() and expose it to the rest of the kernel. It now also accepts parameters via function arguments rather than global variables. - Add a notion of minimum and maximum bus numbers and reject requests for an out of range bus. - Add more range checks on slot/func/reg/bytes parameters to the cfg reg read/write routines. Don't panic on any invalid parameters, just fail the request (writes do nothing, reads return -1). This matches the behavior of the other cfg mechanisms. - Port the memory mapped configuration space access to amd64. On amd64 we simply use the direct map (via pmap_mapdev()) for the memory mapped window. - During acpi_attach() just after loading the ACPI tables, check for a MCFG table. If it exists, call pciereg_cfgopen() on each subtable (memory mapped window). For now we only support windows for domain 0 that start with bus 0. This removes the need for more chipset-specific quirks in the MD code. - Remove the chipset-specific quirks for the Intel 5000P/V/Z chipsets since these machines should all have MCFG tables via ACPI. - Updated pci_cfgregopen() to DTRT if ACPI had invoked pcie_cfgregopen() earlier. MFC after: 2 weeks
2008-08-22 02:14:23 +00:00
#if defined(__i386__) || defined(__amd64__)
re-synchronize TSC-s on SMP systems after resume, if necessary The TSC-s are checked and synchronized only if they were good originally. That is, invariant, synchronized, etc. This is necessary on an AMD-based system where after a wakeup from STR I see that BSP clock differs from AP clocks by a count that roughly corresponds to one second. The APs are in sync with each other. Not sure if this is a hardware quirk or a firmware bug. This is what I see after a resume with this change: SMP: passed TSC synchronization test after adjustment acpi_timer0: restoring timecounter, ACPI-fast -> TSC-low Reviewed by: kib MFC after: 3 weeks Differential Revision: https://reviews.freebsd.org/D15551
2018-05-25 07:33:20 +00:00
#include <machine/clock.h>
Extend the support for PCI-e memory mapped configuration space access: - Rename pciereg_cfgopen() to pcie_cfgregopen() and expose it to the rest of the kernel. It now also accepts parameters via function arguments rather than global variables. - Add a notion of minimum and maximum bus numbers and reject requests for an out of range bus. - Add more range checks on slot/func/reg/bytes parameters to the cfg reg read/write routines. Don't panic on any invalid parameters, just fail the request (writes do nothing, reads return -1). This matches the behavior of the other cfg mechanisms. - Port the memory mapped configuration space access to amd64. On amd64 we simply use the direct map (via pmap_mapdev()) for the memory mapped window. - During acpi_attach() just after loading the ACPI tables, check for a MCFG table. If it exists, call pciereg_cfgopen() on each subtable (memory mapped window). For now we only support windows for domain 0 that start with bus 0. This removes the need for more chipset-specific quirks in the MD code. - Remove the chipset-specific quirks for the Intel 5000P/V/Z chipsets since these machines should all have MCFG tables via ACPI. - Updated pci_cfgregopen() to DTRT if ACPI had invoked pcie_cfgregopen() earlier. MFC after: 2 weeks
2008-08-22 02:14:23 +00:00
#include <machine/pci_cfgreg.h>
#endif
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
#include <machine/resource.h>
Add acpi_bus_alloc_gas() for allocating a memory or IO resource from its Generic Address Structure.
2003-11-15 19:18:29 +00:00
#include <machine/bus.h>
#include <sys/rman.h>
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
#include <isa/isavar.h>
Include isa/pnpvar.h and remove a duplicate copy of PNP_EISAID.
2004-06-29 01:33:35 +00:00
#include <isa/pnpvar.h>
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
Import ACPICA 20090521.
2009-06-05 18:44:36 +00:00
#include <contrib/dev/acpica/include/acpi.h>
#include <contrib/dev/acpica/include/accommon.h>
#include <contrib/dev/acpica/include/acnamesp.h>
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
#include <dev/acpica/acpivar.h>
#include <dev/acpica/acpiio.h>
Add support to read the _CLS entry if it's present. It is used by memory-mapped devices that are normally PCIe drives. Devices can then use the existing pci_get_class, etc. accessors to query this data. The ivar values are different enough from the existing ACPI and ISA values to not conflict. Reviewed by: jhb Obtained from: ABT Systems Ltd Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D8721
2016-12-16 10:40:00 +00:00
#include <dev/pci/pcivar.h>
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
#include <vm/vm_param.h>
Mark MALLOC_DEFINEs static that have no corresponding MALLOC_DECLAREs. This means that their use is restricted to a single C file.
2011-11-07 06:44:47 +00:00
static MALLOC_DEFINE(M_ACPIDEV, "acpidev", "ACPI devices");
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Hooks for the ACPI CA debugging infrastructure */
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
#define _COMPONENT ACPI_BUS
Match namespace cleanup changes in ACPI CA 20020217 update. Use ACPI_SUCCESS/ACPI_FAILURE consistently. The AcpiGetInto* interfaces are obsoleted by ACPI_ALLOCATE_BUFFER. Add AcpiBatteryIsPresent helper to determine whether a battery device is inserted. Add ACPI_ALL_DRIVERS to the list of debug layers, now that we own the namespace for this. Pr:
2002-02-23 05:21:56 +00:00
ACPI_MODULE_NAME("ACPI")
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
static d_open_t acpiopen;
static d_close_t acpiclose;
static d_ioctl_t acpiioctl;
static struct cdevsw acpi_cdevsw = {
Device megapatch 4/6: Introduce d_version field in struct cdevsw, this must always be initialized to D_VERSION. Flip sense of D_NOGIANT flag to D_NEEDGIANT, this involves removing four D_NOGIANT flags and adding 145 D_NEEDGIANT flags.
2004-02-21 21:10:55 +00:00
.d_version = D_VERSION,
Gigacommit to improve device-driver source compatibility between branches: Initialize struct cdevsw using C99 sparse initializtion and remove all initializations to default values. This patch is automatically generated and has been tested by compiling LINT with all the fields in struct cdevsw in reverse order on alpha, sparc64 and i386. Approved by: re(scottl)
2003-03-03 12:15:54 +00:00
.d_open = acpiopen,
.d_close = acpiclose,
.d_ioctl = acpiioctl,
.d_name = "acpi",
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
};
Add two new loader tunables 'hw.acpi.install_interface' and 'hw.acpi.remove_interface'. hw.acpi.install_interface lets you install new interfaces. Conversely, hw.acpi.remove_interface lets you remove OS interfaces from the pre-defined list in ACPICA. For example, hw.acpi.install_interface="FreeBSD" lets _OSI("FreeBSD") method to return 0xffffffff (or success) and hw.acpi.remove_interface="Windows 2009" lets _OSI("Windows 2009") method to return zero (or failure). Both are comma-separated lists and leading white spaces are ignored. For example, the following examples are valid: hw.acpi.install_interface="Linux, FreeBSD" hw.acpi.remove_interface="Windows 2006, Windows 2006.1"
2010-10-26 18:59:50 +00:00
struct acpi_interface {
ACPI_STRING *data;
int num;
};
Consolidate two identical copies of sysres_ids.
2018-01-09 16:42:24 +00:00
static char *sysres_ids[] = { "PNP0C01", "PNP0C02", NULL };
static char *pcilink_ids[] = { "PNP0C0F", NULL };
Diff reduction for style.
2004-06-30 15:10:02 +00:00
/* Global mutex for locking access to the ACPI subsystem. */
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
struct mtx acpi_mutex;
Convert from timeout(9) to callout(9).
2014-09-22 14:27:26 +00:00
struct callout acpi_sleep_timer;
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
Add new quirk code that disables problem BIOS versions. Remove old quirk code that was never really used. Print a message when disabling ACPI via a quirk. Allow the user to override the blacklist decision by setting hint.acpi.0.disabled="0". Add missing AcpiTerminate() calls; they are needed to clean up if bailing out after AcpiInitializeSubsystem().
2004-06-30 04:49:54 +00:00
/* Bitmap of device quirks. */
Diff reduction for style.
2004-06-30 15:10:02 +00:00
int acpi_quirks;
Add new quirk code that disables problem BIOS versions. Remove old quirk code that was never really used. Print a message when disabling ACPI via a quirk. Allow the user to override the blacklist decision by setting hint.acpi.0.disabled="0". Add missing AcpiTerminate() calls; they are needed to clean up if bailing out after AcpiInitializeSubsystem().
2004-06-30 04:49:54 +00:00
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
/* Supported sleep states. */
static BOOLEAN acpi_sleep_states[ACPI_S_STATE_COUNT];
Add a new device control utility for new-bus devices called devctl. This allows the user to request administrative changes to individual devices such as attach or detaching drivers or disabling and re-enabling devices. - Add a new /dev/devctl2 character device which uses ioctls for device requests. The ioctls use a common 'struct devreq' which is somewhat similar to 'struct ifreq'. - The ioctls identify the device to operate on via a string. This string can either by the device's name, or it can be a bus-specific address. (For unattached devices, a bus address is the only way to locate a device.) Bus drivers register an eventhandler to claim unrecognized device names that the driver recognizes as a valid address. Two buses currently support addresses: ACPI recognizes any device in the ACPI namespace via its full path starting with "\" and the PCI bus driver recognizes an address specification of 'pci[<domain>:]<bus>:<slot>:<func>' (identical to the PCI selector strings supported by pciconf). - To make it easier to cut and paste, change the PnP location string in the PCI bus driver to output a full PCI selector string rather than 'slot=<slot> function=<func>'. - Add a devctl(3) interface in libdevctl which provides a wrapper around the ioctls and is the preferred interface for other userland code. - Add a devctl(8) program which is a simple wrapper around the requests supported by devctl(3). - Add a device_is_suspended() function to check DF_SUSPENDED. - Add a resource_unset_value() function that can be used to remove a hint from the kernel environment. This is used to clear a hint.<driver>.<unit>.disabled hint when re-enabling a boot-time disabled device. Reviewed by: imp (parts) Requested by: imp (changing PCI location string) Relnotes: yes
2015-02-06 16:09:01 +00:00
static void acpi_lookup(void *arg, const char *name, device_t *dev);
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
static int acpi_modevent(struct module *mod, int event, void *junk);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
static int acpi_probe(device_t dev);
static int acpi_attach(device_t dev);
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
static int acpi_suspend(device_t dev);
static int acpi_resume(device_t dev);
Disable wake GPEs in the reboot path as well as poweroff path. This fixes "stray irq 9" messages on my Thinkpad. It may also help with general reboot consistency although the recent hang on reboot was solved by acpi_cpu.c rev 1.39.
2004-06-05 07:25:58 +00:00
static int acpi_shutdown(device_t dev);
bus_add_child: change type of order parameter to u_int This reflects actual type used to store and compare child device orders. Change is mostly done via a Coccinelle (soon to be devel/coccinelle) semantic patch. Verified by LINT+modules kernel builds. Followup to: r212213 MFC after: 10 days
2010-09-10 11:19:03 +00:00
static device_t acpi_add_child(device_t bus, u_int order, const char *name,
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
int unit);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
static int acpi_print_child(device_t bus, device_t child);
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
static void acpi_probe_nomatch(device_t bus, device_t child);
static void acpi_driver_added(device_t dev, driver_t *driver);
acpi: Fix stalled value returned by acpi_get_device() after device deletion Newbus device reference attached to ACPI handle is not cleared when newbus device is deleted with devctl(8) delete command. Fix that with calling of AcpiDetachData() from "child_deleted" bus method like acpi_pci driver does. MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D22902
2020-03-09 20:27:25 +00:00
static void acpi_child_deleted(device_t dev, device_t child);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
static int acpi_read_ivar(device_t dev, device_t child, int index,
uintptr_t *result);
static int acpi_write_ivar(device_t dev, device_t child, int index,
uintptr_t value);
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
static struct resource_list *acpi_get_rlist(device_t dev, device_t child);
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
static void acpi_reserve_resources(device_t dev);
Rework sysresource management. Instead of having each sysresource object hold its own values, pass them up to the parent (acpi0) and merge/uniq them on the way. After the namespace evaluation, acpi will reserve these resources and manage them via rman before bus_generic_probe() and bus_generic_attach(). This is necessary because some systems specify conflicting resources in separate sysresource objects. It's also cleaner in that the interface between sysresource and acpi is now merely the parent's resource list. This code handles the following cases: 1. Unique resource: add it to the parent via bus_set_resource(). 2. New wholly contained in old: discard new. 3. New tail overlaps old head: grow old head downward. AND/OR 4. New head overlaps old tail: grow old tail upward. Tested by: Pawel Worach <sajd_at_telia.com> Tested by: Radek Kozlowski <radek_at_raadradd.com> MFC after: 5 days
2004-08-23 16:28:42 +00:00
static int acpi_sysres_alloc(device_t dev);
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
static int acpi_set_resource(device_t dev, device_t child, int type,
Convert rman to use rman_res_t instead of u_long Summary: Migrate to using the semi-opaque type rman_res_t to specify rman resources. For now, this is still compatible with u_long. This is step one in migrating rman to use uintmax_t for resources instead of u_long. Going forward, this could feasibly be used to specify architecture-specific definitions of resource ranges, rather than baking a specific integer type into the API. This change has been broken out to facilitate MFC'ing drivers back to 10 without breaking ABI. Reviewed By: jhb Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D5075
2016-01-27 02:23:54 +00:00
int rid, rman_res_t start, rman_res_t count);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
static struct resource *acpi_alloc_resource(device_t bus, device_t child,
Convert rman to use rman_res_t instead of u_long Summary: Migrate to using the semi-opaque type rman_res_t to specify rman resources. For now, this is still compatible with u_long. This is step one in migrating rman to use uintmax_t for resources instead of u_long. Going forward, this could feasibly be used to specify architecture-specific definitions of resource ranges, rather than baking a specific integer type into the API. This change has been broken out to facilitate MFC'ing drivers back to 10 without breaking ABI. Reviewed By: jhb Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D5075
2016-01-27 02:23:54 +00:00
int type, int *rid, rman_res_t start, rman_res_t end,
rman_res_t count, u_int flags);
Implement BUS_ADJUST_RESOURCE() for the x86 drivers that sit between the Host-PCI bridge drivers and nexus.
2011-06-10 12:30:16 +00:00
static int acpi_adjust_resource(device_t bus, device_t child, int type,
Convert rman to use rman_res_t instead of u_long Summary: Migrate to using the semi-opaque type rman_res_t to specify rman resources. For now, this is still compatible with u_long. This is step one in migrating rman to use uintmax_t for resources instead of u_long. Going forward, this could feasibly be used to specify architecture-specific definitions of resource ranges, rather than baking a specific integer type into the API. This change has been broken out to facilitate MFC'ing drivers back to 10 without breaking ABI. Reviewed By: jhb Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D5075
2016-01-27 02:23:54 +00:00
struct resource *r, rman_res_t start, rman_res_t end);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
static int acpi_release_resource(device_t bus, device_t child, int type,
int rid, struct resource *r);
Add support for bus_delete_resource() and make acpi_bus_alloc_gas() do this before setting a new resource.
2005-03-22 20:00:57 +00:00
static void acpi_delete_resource(device_t bus, device_t child, int type,
int rid);
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
static uint32_t acpi_isa_get_logicalid(device_t dev);
static int acpi_isa_get_compatid(device_t dev, uint32_t *cids, int count);
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
static int acpi_device_id_probe(device_t bus, device_t dev, char **ids, char **match);
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
static ACPI_STATUS acpi_device_eval_obj(device_t bus, device_t dev,
ACPI_STRING pathname, ACPI_OBJECT_LIST *parameters,
ACPI_BUFFER *ret);
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
static ACPI_STATUS acpi_device_scan_cb(ACPI_HANDLE h, UINT32 level,
void *context, void **retval);
static ACPI_STATUS acpi_device_scan_children(device_t bus, device_t dev,
int max_depth, acpi_scan_cb_t user_fn, void *arg);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
static int acpi_isa_pnp_probe(device_t bus, device_t child,
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
struct isa_pnp_id *ids);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
static void acpi_probe_children(device_t bus);
Revert back to probing Host-PCI bridges in the order we encounter them in the tree rather than sorting them by their address on PCI bus 0. Reported by: kan
2008-04-07 18:35:11 +00:00
static void acpi_probe_order(ACPI_HANDLE handle, int *order);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
static ACPI_STATUS acpi_probe_child(ACPI_HANDLE handle, UINT32 level,
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
void *context, void **status);
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
static void acpi_sleep_enable(void *arg);
static ACPI_STATUS acpi_sleep_disable(struct acpi_softc *sc);
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
static ACPI_STATUS acpi_EnterSleepState(struct acpi_softc *sc, int state);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
static void acpi_shutdown_final(void *arg, int howto);
Enable fixed event at not only boot but also wakeup. Reported and patch tested by: Peter Dufault <dufault@hda.hda.com>
2001-01-10 18:01:51 +00:00
static void acpi_enable_fixed_events(struct acpi_softc *sc);
Reset clock after atrtc(4) is properly resumed.
2012-02-08 21:23:20 +00:00
static void acpi_resync_clock(struct acpi_softc *sc);
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
static int acpi_wake_sleep_prep(ACPI_HANDLE handle, int sstate);
static int acpi_wake_run_prep(ACPI_HANDLE handle, int sstate);
static int acpi_wake_prep_walk(int sstate);
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
static int acpi_wake_sysctl_walk(device_t dev);
static int acpi_wake_set_sysctl(SYSCTL_HANDLER_ARGS);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
static void acpi_system_eventhandler_sleep(void *arg, int state);
static void acpi_system_eventhandler_wakeup(void *arg, int state);
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
static int acpi_sname2sstate(const char *sname);
Fix style(9).
2009-04-30 17:45:43 +00:00
static const char *acpi_sstate2sname(int sstate);
Add new sysctl MIB (hw.acpi.supported_sleep_state) to indicate the list of supported sleep state. This should help people understand what following message means. acpi0: AcpiGetSleepTypeData failed - AE_NOT_FOUND MFC after: 3 days
2003-04-11 16:53:56 +00:00
static int acpi_supported_sleep_state_sysctl(SYSCTL_HANDLER_ARGS);
Add 3 new dynamic sysctl's to control the sleep states switched to on a power button, sleep button, or lid close event. The sysctl's use the ACPI sleep state names S0, S1, S2, S3, S4, S4B, and S5. Reviewed by: iwasaki
2001-01-13 21:28:57 +00:00
static int acpi_sleep_state_sysctl(SYSCTL_HANDLER_ARGS);
- Allow users to enable dumping Debug objects without ACPI debugger. Setting the new sysctl MIB "debug.acpi.enable_debug_objects" to a non-zero value enables us to print Debug object when something is written to it. - Allow users to disable interpreter slack mode. Setting the new tunable "debug.acpi.interpreter_slack" to zero disables some workarounds for common BIOS mistakes and enables strict ACPI implementations by the specification.
2010-03-09 19:02:02 +00:00
static int acpi_debug_objects_sysctl(SYSCTL_HANDLER_ARGS);
Some fix for the recent apm module changes. - Now that apm loadable module can inform its existence to other kernel components (e.g. i386/isa/clock.c:startrtclock()'s TCS hack). - Exchange priority of SI_SUB_CPU and SI_SUB_KLD for above purpose. - Add simple arbitration mechanism for APM vs. ACPI. This prevents the kernel enables both of them. - Remove obsolete `#ifdef DEV_APM' related code. - Add abstracted interface for Powermanagement operations. Public apm(4) functions, such as apm_suspend(), should be replaced new interfaces. Currently only power_pm_suspend (successor of apm_suspend) is implemented. Reviewed by: peter, arch@ and audit@
2001-11-01 16:34:07 +00:00
static int acpi_pm_func(u_long cmd, void *arg, ...);
Add ACPI PnP string. This affects devinfo(8) output with -v option.
2004-03-27 16:26:00 +00:00
static int acpi_child_location_str_method(device_t acdev, device_t child,
char *buf, size_t buflen);
static int acpi_child_pnpinfo_str_method(device_t acdev, device_t child,
char *buf, size_t buflen);
Extend the support for PCI-e memory mapped configuration space access: - Rename pciereg_cfgopen() to pcie_cfgregopen() and expose it to the rest of the kernel. It now also accepts parameters via function arguments rather than global variables. - Add a notion of minimum and maximum bus numbers and reject requests for an out of range bus. - Add more range checks on slot/func/reg/bytes parameters to the cfg reg read/write routines. Don't panic on any invalid parameters, just fail the request (writes do nothing, reads return -1). This matches the behavior of the other cfg mechanisms. - Port the memory mapped configuration space access to amd64. On amd64 we simply use the direct map (via pmap_mapdev()) for the memory mapped window. - During acpi_attach() just after loading the ACPI tables, check for a MCFG table. If it exists, call pciereg_cfgopen() on each subtable (memory mapped window). For now we only support windows for domain 0 that start with bus 0. This removes the need for more chipset-specific quirks in the MD code. - Remove the chipset-specific quirks for the Intel 5000P/V/Z chipsets since these machines should all have MCFG tables via ACPI. - Updated pci_cfgregopen() to DTRT if ACPI had invoked pcie_cfgregopen() earlier. MFC after: 2 weeks
2008-08-22 02:14:23 +00:00
static void acpi_enable_pcie(void);
Allow device hints to wire the unit numbers of devices. - An "at" hint now reserves a device name. - A new BUS_HINT_DEVICE_UNIT method is added to the bus interface. When determining the unit number of a device, this method is invoked to let the bus driver specify the unit of a device given a specific devclass. This is the only way a device can be given a name reserved via an "at" hint. - Implement BUS_HINT_DEVICE_UNIT() for the acpi(4) and isa(4) bus drivers. Both of these busses implement this by comparing the resources for a given hint device with the resources enumerated by ACPI/PnPBIOS and wire a unit if the hint resources are a subset of the "real" resources. - Use bus_hinted_children() for adding hinted devices on isa(4) busses now instead of doing it by hand. - Remove the unit kludging from sio(4) as it is no longer necessary. Prodding from: peter, imp OK'd by: marcel MFC after: 1 month
2008-11-18 21:01:54 +00:00
static void acpi_hint_device_unit(device_t acdev, device_t child,
const char *name, int *unitp);
Add two new loader tunables 'hw.acpi.install_interface' and 'hw.acpi.remove_interface'. hw.acpi.install_interface lets you install new interfaces. Conversely, hw.acpi.remove_interface lets you remove OS interfaces from the pre-defined list in ACPICA. For example, hw.acpi.install_interface="FreeBSD" lets _OSI("FreeBSD") method to return 0xffffffff (or success) and hw.acpi.remove_interface="Windows 2009" lets _OSI("Windows 2009") method to return zero (or failure). Both are comma-separated lists and leading white spaces are ignored. For example, the following examples are valid: hw.acpi.install_interface="Linux, FreeBSD" hw.acpi.remove_interface="Windows 2006, Windows 2006.1"
2010-10-26 18:59:50 +00:00
static void acpi_reset_interfaces(device_t dev);
Add ACPI PnP string. This affects devinfo(8) output with -v option.
2004-03-27 16:26:00 +00:00
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
static device_method_t acpi_methods[] = {
/* Device interface */
DEVMETHOD(device_probe, acpi_probe),
DEVMETHOD(device_attach, acpi_attach),
Disable wake GPEs in the reboot path as well as poweroff path. This fixes "stray irq 9" messages on my Thinkpad. It may also help with general reboot consistency although the recent hang on reboot was solved by acpi_cpu.c rev 1.39.
2004-06-05 07:25:58 +00:00
DEVMETHOD(device_shutdown, acpi_shutdown),
* Add a DEVMETHOD for acpi so that child detach methods get called. Add an acpi_cpu method for shutdown that disables entry to acpi_cpu_idle and then IPIs/waits for threads to exit. This fixes a panic late in reboot in the SMP case. * In the !SMP case, don't use the processor id filled out by the MADT since there can only be one processor. This was causing a panic in acpi_cpu_idle if the id was 1 since the data was being dereferenced from cpu_softc[1] even though the actual data was in cpu_softc[0] (which is correct). * Rework the initialization functions so that cpu_idle_hook is written late in the boot process. * Make the P_BLK, P_BLK_LEN, and cpu_cx_count all softc-local variables. This will help SMP boxes that have _CST or multiple P_BLKs. No such boxes are known at this time. * Always allocate the C1 state, even if the P_BLK is invalid. This means we will always take over idling if enabled. Remove the value -1 as valid for cx_lowest since this is redundant with machdep.cpu_idle_hlt. * Reduce locking for the throttle initialization case to around the write to the smi_cmd port. Add disabled code to write the CST_CNT. It will be enabled once _CST re-evaluation is tested (post 5.2R). Thank you: dfr, imp, jhb, marcel, peter Tested by: rwatson, Harald Schmalzbauer <h@schmalzbauer.de> Approved by: re (rwatson)
2003-11-19 20:27:06 +00:00
DEVMETHOD(device_detach, bus_generic_detach),
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
DEVMETHOD(device_suspend, acpi_suspend),
DEVMETHOD(device_resume, acpi_resume),
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/* Bus interface */
DEVMETHOD(bus_add_child, acpi_add_child),
DEVMETHOD(bus_print_child, acpi_print_child),
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
DEVMETHOD(bus_probe_nomatch, acpi_probe_nomatch),
DEVMETHOD(bus_driver_added, acpi_driver_added),
acpi: Fix stalled value returned by acpi_get_device() after device deletion Newbus device reference attached to ACPI handle is not cleared when newbus device is deleted with devctl(8) delete command. Fix that with calling of AcpiDetachData() from "child_deleted" bus method like acpi_pci driver does. MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D22902
2020-03-09 20:27:25 +00:00
DEVMETHOD(bus_child_deleted, acpi_child_deleted),
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
DEVMETHOD(bus_read_ivar, acpi_read_ivar),
DEVMETHOD(bus_write_ivar, acpi_write_ivar),
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
DEVMETHOD(bus_get_resource_list, acpi_get_rlist),
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
DEVMETHOD(bus_set_resource, acpi_set_resource),
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
DEVMETHOD(bus_get_resource, bus_generic_rl_get_resource),
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
DEVMETHOD(bus_alloc_resource, acpi_alloc_resource),
Implement BUS_ADJUST_RESOURCE() for the x86 drivers that sit between the Host-PCI bridge drivers and nexus.
2011-06-10 12:30:16 +00:00
DEVMETHOD(bus_adjust_resource, acpi_adjust_resource),
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
DEVMETHOD(bus_release_resource, acpi_release_resource),
Add support for bus_delete_resource() and make acpi_bus_alloc_gas() do this before setting a new resource.
2005-03-22 20:00:57 +00:00
DEVMETHOD(bus_delete_resource, acpi_delete_resource),
Staticize pnp methods, style fixes. Remove unused variable to unbreak kernel build.
2004-03-31 17:35:28 +00:00
DEVMETHOD(bus_child_pnpinfo_str, acpi_child_pnpinfo_str_method),
DEVMETHOD(bus_child_location_str, acpi_child_location_str_method),
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
DEVMETHOD(bus_activate_resource, bus_generic_activate_resource),
DEVMETHOD(bus_deactivate_resource, bus_generic_deactivate_resource),
DEVMETHOD(bus_setup_intr, bus_generic_setup_intr),
DEVMETHOD(bus_teardown_intr, bus_generic_teardown_intr),
Allow device hints to wire the unit numbers of devices. - An "at" hint now reserves a device name. - A new BUS_HINT_DEVICE_UNIT method is added to the bus interface. When determining the unit number of a device, this method is invoked to let the bus driver specify the unit of a device given a specific devclass. This is the only way a device can be given a name reserved via an "at" hint. - Implement BUS_HINT_DEVICE_UNIT() for the acpi(4) and isa(4) bus drivers. Both of these busses implement this by comparing the resources for a given hint device with the resources enumerated by ACPI/PnPBIOS and wire a unit if the hint resources are a subset of the "real" resources. - Use bus_hinted_children() for adding hinted devices on isa(4) busses now instead of doing it by hand. - Remove the unit kludging from sio(4) as it is no longer necessary. Prodding from: peter, imp OK'd by: marcel MFC after: 1 month
2008-11-18 21:01:54 +00:00
DEVMETHOD(bus_hint_device_unit, acpi_hint_device_unit),
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
DEVMETHOD(bus_get_cpus, acpi_get_cpus),
Add a bus method to fetch the VM domain for the given device/bus. * Add a bus_if.m method - get_domain() - returning the VM domain or ENOENT if the device isn't in a VM domain; * Add bus methods to print out the domain of the device if appropriate; * Add code in srat.c to save the PXM -> VM domain mapping that's done and expose a function to translate VM domain -> PXM; * Add ACPI and ACPI PCI methods to check if the bus has a _PXM attribute and if so map it to the VM domain; * (.. yes, this works recursively.) * Have the pci bus glue print out the device VM domain if present. Note: this is just the plumbing to start enumerating information - it doesn't at all modify behaviour. Differential Revision: D906 Reviewed by: jhb Sponsored by: Norse Corp
2014-10-09 05:33:25 +00:00
DEVMETHOD(bus_get_domain, acpi_get_domain),
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
/* ACPI bus */
DEVMETHOD(acpi_id_probe, acpi_device_id_probe),
DEVMETHOD(acpi_evaluate_object, acpi_device_eval_obj),
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
DEVMETHOD(acpi_pwr_for_sleep, acpi_device_pwr_for_sleep),
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
DEVMETHOD(acpi_scan_children, acpi_device_scan_children),
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
/* ISA emulation */
DEVMETHOD(isa_pnp_probe, acpi_isa_pnp_probe),
Use DEVMETHOD_END macro defined in sys/bus.h instead of {0, 0} sentinel on device_method_t arrays Reviewed by: cognet Approved by: cognet
2013-01-30 18:01:20 +00:00
DEVMETHOD_END
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
};
static driver_t acpi_driver = {
"acpi",
acpi_methods,
sizeof(struct acpi_softc),
};
Staticise devclasses and some unnecessarily global variables.
2002-01-08 06:46:01 +00:00
static devclass_t acpi_devclass;
Use EARLY_DRIVER_MODULE in the acpi bus. We need this to use EARLY_DRIVER_MODULE in child drivers on arm64. This should be a no-op on x86 as it has DRIVER_MODULE in the nexus driver making all later drivers attach in the last pass. Reviewed by: imp MFC after: 1 month Sponsored by: Innovate UK Differential Revision: https://reviews.freebsd.org/D23717
2020-02-17 15:32:21 +00:00
EARLY_DRIVER_MODULE(acpi, nexus, acpi_driver, acpi_devclass, acpi_modevent, 0,
BUS_PASS_BUS + BUS_PASS_ORDER_MIDDLE);
Unify on version 1 to be similar to the rest of the tree. After 5-stable branches, increment version on any API change visible to other modules.
2004-04-08 16:45:12 +00:00
MODULE_VERSION(acpi, 1);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
ACPI_SERIAL_DECL(acpi, "ACPI root bus");
Rework sysresource management. Instead of having each sysresource object hold its own values, pass them up to the parent (acpi0) and merge/uniq them on the way. After the namespace evaluation, acpi will reserve these resources and manage them via rman before bus_generic_probe() and bus_generic_attach(). This is necessary because some systems specify conflicting resources in separate sysresource objects. It's also cleaner in that the interface between sysresource and acpi is now merely the parent's resource list. This code handles the following cases: 1. Unique resource: add it to the parent via bus_set_resource(). 2. New wholly contained in old: discard new. 3. New tail overlaps old head: grow old head downward. AND/OR 4. New head overlaps old tail: grow old tail upward. Tested by: Pawel Worach <sajd_at_telia.com> Tested by: Radek Kozlowski <radek_at_raadradd.com> MFC after: 5 days
2004-08-23 16:28:42 +00:00
/* Local pools for managing system resources for ACPI child devices. */
static struct rman acpi_rman_io, acpi_rman_mem;
/tmp/m
2004-08-03 05:13:56 +00:00
#define ACPI_MINIMUM_AWAKETIME 5
Rework how the nexus(4) device works on x86 to better handle the idea of different "platforms" on x86 machines. The existing code already handles having two platforms: ACPI and legacy. However, the existing approach was rather hardcoded and difficult to extend. These changes take the approach that each x86 hardware platform should provide its own nexus(4) driver (it can inherit most of its behavior from the default legacy nexus(4) driver) which is responsible for probing for the platform and performing appropriate platform-specific setup during attach (such as adding a platform-specific bus device). This does mean changing the x86 platform busses to no longer use an identify routine for probing, but to move that logic into their matching nexus(4) driver instead. - Make the default nexus(4) driver in nexus.c on i386 and amd64 handle the legacy platform. It's probe routine now returns BUS_PROBE_GENERIC so it can be overriden. - Expose a nexus_init_resources() routine which initializes the various resource managers so that subclassed nexus(4) drivers can invoke it from their attach routine. - The legacy nexus(4) driver explicitly adds a legacy0 device in its attach routine. - The ACPI driver no longer contains an new-bus identify method. Instead it exposes a public function (acpi_identify()) which is a probe routine that the MD nexus(4) drivers can use to probe for ACPI. All of the probe logic in acpi_probe() is now moved into acpi_identify() and acpi_probe() is just a stub. - On i386 and amd64, an ACPI-specific nexus(4) driver checks for ACPI via acpi_identify() and claims the nexus0 device if the probe succeeds. It then explicitly adds an acpi0 device in its attach routine. - The legacy(4) driver no longer knows anything about the acpi0 device. - On ia64 if acpi_identify() fails you basically end up with no devices. This matches the previous behavior where the old acpi_identify() would fail to add an acpi0 device again leaving you with no devices. Discussed with: imp Silence on: arch@
2008-03-13 20:39:04 +00:00
/* Holds the description of the acpi0 device. */
static char acpi_desc[ACPI_OEM_ID_SIZE + ACPI_OEM_TABLE_ID_SIZE + 2];
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
SYSCTL_NODE(_debug, OID_AUTO, acpi, CTLFLAG_RD | CTLFLAG_MPSAFE, NULL,
"ACPI debugging");
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
static char acpi_ca_version[12];
SYSCTL_STRING(_debug_acpi, OID_AUTO, acpi_ca_version, CTLFLAG_RD,
acpi_ca_version, 0, "Version of Intel ACPI-CA");
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Add two new loader tunables 'hw.acpi.install_interface' and 'hw.acpi.remove_interface'. hw.acpi.install_interface lets you install new interfaces. Conversely, hw.acpi.remove_interface lets you remove OS interfaces from the pre-defined list in ACPICA. For example, hw.acpi.install_interface="FreeBSD" lets _OSI("FreeBSD") method to return 0xffffffff (or success) and hw.acpi.remove_interface="Windows 2009" lets _OSI("Windows 2009") method to return zero (or failure). Both are comma-separated lists and leading white spaces are ignored. For example, the following examples are valid: hw.acpi.install_interface="Linux, FreeBSD" hw.acpi.remove_interface="Windows 2006, Windows 2006.1"
2010-10-26 18:59:50 +00:00
/*
* Allow overriding _OSI methods.
*/
static char acpi_install_interface[256];
TUNABLE_STR("hw.acpi.install_interface", acpi_install_interface,
sizeof(acpi_install_interface));
static char acpi_remove_interface[256];
TUNABLE_STR("hw.acpi.remove_interface", acpi_remove_interface,
sizeof(acpi_remove_interface));
- Allow users to enable dumping Debug objects without ACPI debugger. Setting the new sysctl MIB "debug.acpi.enable_debug_objects" to a non-zero value enables us to print Debug object when something is written to it. - Allow users to disable interpreter slack mode. Setting the new tunable "debug.acpi.interpreter_slack" to zero disables some workarounds for common BIOS mistakes and enables strict ACPI implementations by the specification.
2010-03-09 19:02:02 +00:00
/* Allow users to dump Debug objects without ACPI debugger. */
static int acpi_debug_objects;
TUNABLE_INT("debug.acpi.enable_debug_objects", &acpi_debug_objects);
SYSCTL_PROC(_debug_acpi, OID_AUTO, enable_debug_objects,
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
CTLFLAG_RW | CTLTYPE_INT | CTLFLAG_NEEDGIANT, NULL, 0,
acpi_debug_objects_sysctl, "I",
- Allow users to enable dumping Debug objects without ACPI debugger. Setting the new sysctl MIB "debug.acpi.enable_debug_objects" to a non-zero value enables us to print Debug object when something is written to it. - Allow users to disable interpreter slack mode. Setting the new tunable "debug.acpi.interpreter_slack" to zero disables some workarounds for common BIOS mistakes and enables strict ACPI implementations by the specification.
2010-03-09 19:02:02 +00:00
"Enable Debug objects");
/* Allow the interpreter to ignore common mistakes in BIOS. */
static int acpi_interpreter_slack = 1;
TUNABLE_INT("debug.acpi.interpreter_slack", &acpi_interpreter_slack);
SYSCTL_INT(_debug_acpi, OID_AUTO, interpreter_slack, CTLFLAG_RDTUN,
&acpi_interpreter_slack, 1, "Turn on interpreter slack mode.");
Merge ACPICA 20140926.
2014-10-02 19:11:18 +00:00
/* Ignore register widths set by FADT and use default widths instead. */
static int acpi_ignore_reg_width = 1;
TUNABLE_INT("debug.acpi.default_register_width", &acpi_ignore_reg_width);
SYSCTL_INT(_debug_acpi, OID_AUTO, default_register_width, CTLFLAG_RDTUN,
&acpi_ignore_reg_width, 1, "Ignore register widths set by FADT");
Allow users to manually override quirks with the tunable "debug.acpi.quirks". Suggested by: Jung-uk Kim
2005-02-13 20:10:28 +00:00
/* Allow users to override quirks. */
TUNABLE_INT("debug.acpi.quirks", &acpi_quirks);
Redo r332918 with the ACPICA API and remove debug.acpi.suspend_deep_bounce. AcpiOsEnterSleep() was meant to implement this feature. Reviewed by: avg
2018-05-03 19:00:50 +00:00
int acpi_susp_bounce;
Add a sysctl, 'debug.acpi.suspend_bounce', that causes the system to bounce back in a simulated resume instead of entering the requested suspend state. This helps in testing drivers separately from the acpi suspend code. To test your drivers, set debug.acpi.suspend_bounce=1 and then run acpiconf -s3 (or 4). MFC after: 1 day
2007-05-25 05:26:21 +00:00
SYSCTL_INT(_debug_acpi, OID_AUTO, suspend_bounce, CTLFLAG_RW,
&acpi_susp_bounce, 0, "Don't actually suspend, just test devices.");
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
/*
* ACPI can only be loaded as a module by the loader; activating it after
* system bootstrap time is not useful, and can be fatal to the system.
Fix a typo.
2008-08-04 19:45:15 +00:00
* It also cannot be unloaded, since the entire system bus hierarchy hangs
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
* off it.
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
*/
static int
acpi_modevent(struct module *mod, int event, void *junk)
{
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
switch (event) {
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
case MOD_LOAD:
Add `if (!cold)' checkings for functions which is called via SYSINIT. Loading acpi.ko with kldload is disallowed, however some functions were executed unexpectedly. Approved by: re
2002-11-24 02:27:07 +00:00
if (!cold) {
printf("The ACPI driver cannot be loaded after boot.\n");
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (EPERM);
Add `if (!cold)' checkings for functions which is called via SYSINIT. Loading acpi.ko with kldload is disallowed, however some functions were executed unexpectedly. Approved by: re
2002-11-24 02:27:07 +00:00
}
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
break;
case MOD_UNLOAD:
Some fix for the recent apm module changes. - Now that apm loadable module can inform its existence to other kernel components (e.g. i386/isa/clock.c:startrtclock()'s TCS hack). - Exchange priority of SI_SUB_CPU and SI_SUB_KLD for above purpose. - Add simple arbitration mechanism for APM vs. ACPI. This prevents the kernel enables both of them. - Remove obsolete `#ifdef DEV_APM' related code. - Add abstracted interface for Powermanagement operations. Public apm(4) functions, such as apm_suspend(), should be replaced new interfaces. Currently only power_pm_suspend (successor of apm_suspend) is implemented. Reviewed by: peter, arch@ and audit@
2001-11-01 16:34:07 +00:00
if (!cold && power_pm_get_type() == POWER_PM_TYPE_ACPI)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (EBUSY);
Some fix for the recent apm module changes. - Now that apm loadable module can inform its existence to other kernel components (e.g. i386/isa/clock.c:startrtclock()'s TCS hack). - Exchange priority of SI_SUB_CPU and SI_SUB_KLD for above purpose. - Add simple arbitration mechanism for APM vs. ACPI. This prevents the kernel enables both of them. - Remove obsolete `#ifdef DEV_APM' related code. - Add abstracted interface for Powermanagement operations. Public apm(4) functions, such as apm_suspend(), should be replaced new interfaces. Currently only power_pm_suspend (successor of apm_suspend) is implemented. Reviewed by: peter, arch@ and audit@
2001-11-01 16:34:07 +00:00
break;
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
default:
break;
}
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (0);
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
Move the code to initialize ACPI-CA into a separate acpi_Startup() function that other modules can call to initialize ACPI-CA before the new-bus probe and change acpi_identify() to call it. Reviewed by: njl
2004-01-26 19:29:04 +00:00
* Perform early initialization.
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*/
Move the code to initialize ACPI-CA into a separate acpi_Startup() function that other modules can call to initialize ACPI-CA before the new-bus probe and change acpi_identify() to call it. Reviewed by: njl
2004-01-26 19:29:04 +00:00
ACPI_STATUS
acpi_Startup(void)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
Add new quirk code that disables problem BIOS versions. Remove old quirk code that was never really used. Print a message when disabling ACPI via a quirk. Allow the user to override the blacklist decision by setting hint.acpi.0.disabled="0". Add missing AcpiTerminate() calls; they are needed to clean up if bailing out after AcpiInitializeSubsystem().
2004-06-30 04:49:54 +00:00
static int started = 0;
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
ACPI_STATUS status;
int val;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Add an ACPI_FUNCTION_TRACE() to the newly-added acpi_Startup() routine to get the ACPI_DEBUG case (and LINT in particular) to build. Reviewed by: jhb, njl Approved by: jhb
2004-01-28 07:48:03 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
/tmp/m
2004-08-03 05:13:56 +00:00
/* Only run the startup code once. The MADT driver also calls this. */
Move the code to initialize ACPI-CA into a separate acpi_Startup() function that other modules can call to initialize ACPI-CA before the new-bus probe and change acpi_identify() to call it. Reviewed by: njl
2004-01-26 19:29:04 +00:00
if (started)
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
return_VALUE (AE_OK);
Move the code to initialize ACPI-CA into a separate acpi_Startup() function that other modules can call to initialize ACPI-CA before the new-bus probe and change acpi_identify() to call it. Reviewed by: njl
2004-01-26 19:29:04 +00:00
started = 1;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Merge ACPICA 20170728.
2017-07-28 22:23:29 +00:00
/*
* Initialize the ACPICA subsystem.
*/
if (ACPI_FAILURE(status = AcpiInitializeSubsystem())) {
printf("ACPI: Could not initialize Subsystem: %s\n",
AcpiFormatException(status));
return_VALUE (status);
}
Add tunables for disabling serialized method execution and disabling the new _OSI method. These can be used if these new features end up causing regression for users.
2004-03-18 18:42:22 +00:00
/*
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
* Pre-allocate space for RSDT/XSDT and DSDT tables and allow resizing
* if more tables exist.
Add tunables for disabling serialized method execution and disabling the new _OSI method. These can be used if these new features end up causing regression for users.
2004-03-18 18:42:22 +00:00
*/
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
if (ACPI_FAILURE(status = AcpiInitializeTables(NULL, 2, TRUE))) {
printf("ACPI: Table initialisation failed: %s\n",
AcpiFormatException(status));
return_VALUE (status);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
Add new quirk code that disables problem BIOS versions. Remove old quirk code that was never really used. Print a message when disabling ACPI via a quirk. Allow the user to override the blacklist decision by setting hint.acpi.0.disabled="0". Add missing AcpiTerminate() calls; they are needed to clean up if bailing out after AcpiInitializeSubsystem().
2004-06-30 04:49:54 +00:00
/* Set up any quirks we have for this system. */
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
if (acpi_quirks == ACPI_Q_OK)
Allow users to manually override quirks with the tunable "debug.acpi.quirks". Suggested by: Jung-uk Kim
2005-02-13 20:10:28 +00:00
acpi_table_quirks(&acpi_quirks);
Add new quirk code that disables problem BIOS versions. Remove old quirk code that was never really used. Print a message when disabling ACPI via a quirk. Allow the user to override the blacklist decision by setting hint.acpi.0.disabled="0". Add missing AcpiTerminate() calls; they are needed to clean up if bailing out after AcpiInitializeSubsystem().
2004-06-30 04:49:54 +00:00
Allow users to manually override quirks with the tunable "debug.acpi.quirks". Suggested by: Jung-uk Kim
2005-02-13 20:10:28 +00:00
/* If the user manually set the disabled hint to 0, force-enable ACPI. */
Add new quirk code that disables problem BIOS versions. Remove old quirk code that was never really used. Print a message when disabling ACPI via a quirk. Allow the user to override the blacklist decision by setting hint.acpi.0.disabled="0". Add missing AcpiTerminate() calls; they are needed to clean up if bailing out after AcpiInitializeSubsystem().
2004-06-30 04:49:54 +00:00
if (resource_int_value("acpi", 0, "disabled", &val) == 0 && val == 0)
acpi_quirks &= ~ACPI_Q_BROKEN;
if (acpi_quirks & ACPI_Q_BROKEN) {
printf("ACPI disabled by blacklist. Contact your BIOS vendor.\n");
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
status = AE_SUPPORT;
Add new quirk code that disables problem BIOS versions. Remove old quirk code that was never really used. Print a message when disabling ACPI via a quirk. Allow the user to override the blacklist decision by setting hint.acpi.0.disabled="0". Add missing AcpiTerminate() calls; they are needed to clean up if bailing out after AcpiInitializeSubsystem().
2004-06-30 04:49:54 +00:00
}
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
return_VALUE (status);
Move the code to initialize ACPI-CA into a separate acpi_Startup() function that other modules can call to initialize ACPI-CA before the new-bus probe and change acpi_identify() to call it. Reviewed by: njl
2004-01-26 19:29:04 +00:00
}
/*
Rework how the nexus(4) device works on x86 to better handle the idea of different "platforms" on x86 machines. The existing code already handles having two platforms: ACPI and legacy. However, the existing approach was rather hardcoded and difficult to extend. These changes take the approach that each x86 hardware platform should provide its own nexus(4) driver (it can inherit most of its behavior from the default legacy nexus(4) driver) which is responsible for probing for the platform and performing appropriate platform-specific setup during attach (such as adding a platform-specific bus device). This does mean changing the x86 platform busses to no longer use an identify routine for probing, but to move that logic into their matching nexus(4) driver instead. - Make the default nexus(4) driver in nexus.c on i386 and amd64 handle the legacy platform. It's probe routine now returns BUS_PROBE_GENERIC so it can be overriden. - Expose a nexus_init_resources() routine which initializes the various resource managers so that subclassed nexus(4) drivers can invoke it from their attach routine. - The legacy nexus(4) driver explicitly adds a legacy0 device in its attach routine. - The ACPI driver no longer contains an new-bus identify method. Instead it exposes a public function (acpi_identify()) which is a probe routine that the MD nexus(4) drivers can use to probe for ACPI. All of the probe logic in acpi_probe() is now moved into acpi_identify() and acpi_probe() is just a stub. - On i386 and amd64, an ACPI-specific nexus(4) driver checks for ACPI via acpi_identify() and claims the nexus0 device if the probe succeeds. It then explicitly adds an acpi0 device in its attach routine. - The legacy(4) driver no longer knows anything about the acpi0 device. - On ia64 if acpi_identify() fails you basically end up with no devices. This matches the previous behavior where the old acpi_identify() would fail to add an acpi0 device again leaving you with no devices. Discussed with: imp Silence on: arch@
2008-03-13 20:39:04 +00:00
* Detect ACPI and perform early initialisation.
Move the code to initialize ACPI-CA into a separate acpi_Startup() function that other modules can call to initialize ACPI-CA before the new-bus probe and change acpi_identify() to call it. Reviewed by: njl
2004-01-26 19:29:04 +00:00
*/
Rework how the nexus(4) device works on x86 to better handle the idea of different "platforms" on x86 machines. The existing code already handles having two platforms: ACPI and legacy. However, the existing approach was rather hardcoded and difficult to extend. These changes take the approach that each x86 hardware platform should provide its own nexus(4) driver (it can inherit most of its behavior from the default legacy nexus(4) driver) which is responsible for probing for the platform and performing appropriate platform-specific setup during attach (such as adding a platform-specific bus device). This does mean changing the x86 platform busses to no longer use an identify routine for probing, but to move that logic into their matching nexus(4) driver instead. - Make the default nexus(4) driver in nexus.c on i386 and amd64 handle the legacy platform. It's probe routine now returns BUS_PROBE_GENERIC so it can be overriden. - Expose a nexus_init_resources() routine which initializes the various resource managers so that subclassed nexus(4) drivers can invoke it from their attach routine. - The legacy nexus(4) driver explicitly adds a legacy0 device in its attach routine. - The ACPI driver no longer contains an new-bus identify method. Instead it exposes a public function (acpi_identify()) which is a probe routine that the MD nexus(4) drivers can use to probe for ACPI. All of the probe logic in acpi_probe() is now moved into acpi_identify() and acpi_probe() is just a stub. - On i386 and amd64, an ACPI-specific nexus(4) driver checks for ACPI via acpi_identify() and claims the nexus0 device if the probe succeeds. It then explicitly adds an acpi0 device in its attach routine. - The legacy(4) driver no longer knows anything about the acpi0 device. - On ia64 if acpi_identify() fails you basically end up with no devices. This matches the previous behavior where the old acpi_identify() would fail to add an acpi0 device again leaving you with no devices. Discussed with: imp Silence on: arch@
2008-03-13 20:39:04 +00:00
int
acpi_identify(void)
Move the code to initialize ACPI-CA into a separate acpi_Startup() function that other modules can call to initialize ACPI-CA before the new-bus probe and change acpi_identify() to call it. Reviewed by: njl
2004-01-26 19:29:04 +00:00
{
Rework how the nexus(4) device works on x86 to better handle the idea of different "platforms" on x86 machines. The existing code already handles having two platforms: ACPI and legacy. However, the existing approach was rather hardcoded and difficult to extend. These changes take the approach that each x86 hardware platform should provide its own nexus(4) driver (it can inherit most of its behavior from the default legacy nexus(4) driver) which is responsible for probing for the platform and performing appropriate platform-specific setup during attach (such as adding a platform-specific bus device). This does mean changing the x86 platform busses to no longer use an identify routine for probing, but to move that logic into their matching nexus(4) driver instead. - Make the default nexus(4) driver in nexus.c on i386 and amd64 handle the legacy platform. It's probe routine now returns BUS_PROBE_GENERIC so it can be overriden. - Expose a nexus_init_resources() routine which initializes the various resource managers so that subclassed nexus(4) drivers can invoke it from their attach routine. - The legacy nexus(4) driver explicitly adds a legacy0 device in its attach routine. - The ACPI driver no longer contains an new-bus identify method. Instead it exposes a public function (acpi_identify()) which is a probe routine that the MD nexus(4) drivers can use to probe for ACPI. All of the probe logic in acpi_probe() is now moved into acpi_identify() and acpi_probe() is just a stub. - On i386 and amd64, an ACPI-specific nexus(4) driver checks for ACPI via acpi_identify() and claims the nexus0 device if the probe succeeds. It then explicitly adds an acpi0 device in its attach routine. - The legacy(4) driver no longer knows anything about the acpi0 device. - On ia64 if acpi_identify() fails you basically end up with no devices. This matches the previous behavior where the old acpi_identify() would fail to add an acpi0 device again leaving you with no devices. Discussed with: imp Silence on: arch@
2008-03-13 20:39:04 +00:00
ACPI_TABLE_RSDP *rsdp;
ACPI_TABLE_HEADER *rsdt;
ACPI_PHYSICAL_ADDRESS paddr;
struct sbuf sb;
Move the code to initialize ACPI-CA into a separate acpi_Startup() function that other modules can call to initialize ACPI-CA before the new-bus probe and change acpi_identify() to call it. Reviewed by: njl
2004-01-26 19:29:04 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
if (!cold)
Rework how the nexus(4) device works on x86 to better handle the idea of different "platforms" on x86 machines. The existing code already handles having two platforms: ACPI and legacy. However, the existing approach was rather hardcoded and difficult to extend. These changes take the approach that each x86 hardware platform should provide its own nexus(4) driver (it can inherit most of its behavior from the default legacy nexus(4) driver) which is responsible for probing for the platform and performing appropriate platform-specific setup during attach (such as adding a platform-specific bus device). This does mean changing the x86 platform busses to no longer use an identify routine for probing, but to move that logic into their matching nexus(4) driver instead. - Make the default nexus(4) driver in nexus.c on i386 and amd64 handle the legacy platform. It's probe routine now returns BUS_PROBE_GENERIC so it can be overriden. - Expose a nexus_init_resources() routine which initializes the various resource managers so that subclassed nexus(4) drivers can invoke it from their attach routine. - The legacy nexus(4) driver explicitly adds a legacy0 device in its attach routine. - The ACPI driver no longer contains an new-bus identify method. Instead it exposes a public function (acpi_identify()) which is a probe routine that the MD nexus(4) drivers can use to probe for ACPI. All of the probe logic in acpi_probe() is now moved into acpi_identify() and acpi_probe() is just a stub. - On i386 and amd64, an ACPI-specific nexus(4) driver checks for ACPI via acpi_identify() and claims the nexus0 device if the probe succeeds. It then explicitly adds an acpi0 device in its attach routine. - The legacy(4) driver no longer knows anything about the acpi0 device. - On ia64 if acpi_identify() fails you basically end up with no devices. This matches the previous behavior where the old acpi_identify() would fail to add an acpi0 device again leaving you with no devices. Discussed with: imp Silence on: arch@
2008-03-13 20:39:04 +00:00
return (ENXIO);
Move the code to initialize ACPI-CA into a separate acpi_Startup() function that other modules can call to initialize ACPI-CA before the new-bus probe and change acpi_identify() to call it. Reviewed by: njl
2004-01-26 19:29:04 +00:00
This isn't functionally identical. In some cases a hint to disable unit 0 would in fact disable all units. This reverts r241856 Approved by: cperciva (implicit)
2012-10-22 13:06:09 +00:00
/* Check that we haven't been disabled with a hint. */
if (resource_disabled("acpi", 0))
return (ENXIO);
Rework how the nexus(4) device works on x86 to better handle the idea of different "platforms" on x86 machines. The existing code already handles having two platforms: ACPI and legacy. However, the existing approach was rather hardcoded and difficult to extend. These changes take the approach that each x86 hardware platform should provide its own nexus(4) driver (it can inherit most of its behavior from the default legacy nexus(4) driver) which is responsible for probing for the platform and performing appropriate platform-specific setup during attach (such as adding a platform-specific bus device). This does mean changing the x86 platform busses to no longer use an identify routine for probing, but to move that logic into their matching nexus(4) driver instead. - Make the default nexus(4) driver in nexus.c on i386 and amd64 handle the legacy platform. It's probe routine now returns BUS_PROBE_GENERIC so it can be overriden. - Expose a nexus_init_resources() routine which initializes the various resource managers so that subclassed nexus(4) drivers can invoke it from their attach routine. - The legacy nexus(4) driver explicitly adds a legacy0 device in its attach routine. - The ACPI driver no longer contains an new-bus identify method. Instead it exposes a public function (acpi_identify()) which is a probe routine that the MD nexus(4) drivers can use to probe for ACPI. All of the probe logic in acpi_probe() is now moved into acpi_identify() and acpi_probe() is just a stub. - On i386 and amd64, an ACPI-specific nexus(4) driver checks for ACPI via acpi_identify() and claims the nexus0 device if the probe succeeds. It then explicitly adds an acpi0 device in its attach routine. - The legacy(4) driver no longer knows anything about the acpi0 device. - On ia64 if acpi_identify() fails you basically end up with no devices. This matches the previous behavior where the old acpi_identify() would fail to add an acpi0 device again leaving you with no devices. Discussed with: imp Silence on: arch@
2008-03-13 20:39:04 +00:00
/* Check for other PM systems. */
if (power_pm_get_type() != POWER_PM_TYPE_NONE &&
power_pm_get_type() != POWER_PM_TYPE_ACPI) {
printf("ACPI identify failed, other PM system enabled.\n");
return (ENXIO);
}
Set the debug.acpi.acpi_ca_version sysctl even if ACPI support is not available.
2007-05-08 14:31:49 +00:00
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
/* Initialize root tables. */
if (ACPI_FAILURE(acpi_Startup())) {
printf("ACPI: Try disabling either ACPI or apic support.\n");
Rework how the nexus(4) device works on x86 to better handle the idea of different "platforms" on x86 machines. The existing code already handles having two platforms: ACPI and legacy. However, the existing approach was rather hardcoded and difficult to extend. These changes take the approach that each x86 hardware platform should provide its own nexus(4) driver (it can inherit most of its behavior from the default legacy nexus(4) driver) which is responsible for probing for the platform and performing appropriate platform-specific setup during attach (such as adding a platform-specific bus device). This does mean changing the x86 platform busses to no longer use an identify routine for probing, but to move that logic into their matching nexus(4) driver instead. - Make the default nexus(4) driver in nexus.c on i386 and amd64 handle the legacy platform. It's probe routine now returns BUS_PROBE_GENERIC so it can be overriden. - Expose a nexus_init_resources() routine which initializes the various resource managers so that subclassed nexus(4) drivers can invoke it from their attach routine. - The legacy nexus(4) driver explicitly adds a legacy0 device in its attach routine. - The ACPI driver no longer contains an new-bus identify method. Instead it exposes a public function (acpi_identify()) which is a probe routine that the MD nexus(4) drivers can use to probe for ACPI. All of the probe logic in acpi_probe() is now moved into acpi_identify() and acpi_probe() is just a stub. - On i386 and amd64, an ACPI-specific nexus(4) driver checks for ACPI via acpi_identify() and claims the nexus0 device if the probe succeeds. It then explicitly adds an acpi0 device in its attach routine. - The legacy(4) driver no longer knows anything about the acpi0 device. - On ia64 if acpi_identify() fails you basically end up with no devices. This matches the previous behavior where the old acpi_identify() would fail to add an acpi0 device again leaving you with no devices. Discussed with: imp Silence on: arch@
2008-03-13 20:39:04 +00:00
return (ENXIO);
Some fix for the recent apm module changes. - Now that apm loadable module can inform its existence to other kernel components (e.g. i386/isa/clock.c:startrtclock()'s TCS hack). - Exchange priority of SI_SUB_CPU and SI_SUB_KLD for above purpose. - Add simple arbitration mechanism for APM vs. ACPI. This prevents the kernel enables both of them. - Remove obsolete `#ifdef DEV_APM' related code. - Add abstracted interface for Powermanagement operations. Public apm(4) functions, such as apm_suspend(), should be replaced new interfaces. Currently only power_pm_suspend (successor of apm_suspend) is implemented. Reviewed by: peter, arch@ and audit@
2001-11-01 16:34:07 +00:00
}
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
if ((paddr = AcpiOsGetRootPointer()) == 0 ||
(rsdp = AcpiOsMapMemory(paddr, sizeof(ACPI_TABLE_RSDP))) == NULL)
Rework how the nexus(4) device works on x86 to better handle the idea of different "platforms" on x86 machines. The existing code already handles having two platforms: ACPI and legacy. However, the existing approach was rather hardcoded and difficult to extend. These changes take the approach that each x86 hardware platform should provide its own nexus(4) driver (it can inherit most of its behavior from the default legacy nexus(4) driver) which is responsible for probing for the platform and performing appropriate platform-specific setup during attach (such as adding a platform-specific bus device). This does mean changing the x86 platform busses to no longer use an identify routine for probing, but to move that logic into their matching nexus(4) driver instead. - Make the default nexus(4) driver in nexus.c on i386 and amd64 handle the legacy platform. It's probe routine now returns BUS_PROBE_GENERIC so it can be overriden. - Expose a nexus_init_resources() routine which initializes the various resource managers so that subclassed nexus(4) drivers can invoke it from their attach routine. - The legacy nexus(4) driver explicitly adds a legacy0 device in its attach routine. - The ACPI driver no longer contains an new-bus identify method. Instead it exposes a public function (acpi_identify()) which is a probe routine that the MD nexus(4) drivers can use to probe for ACPI. All of the probe logic in acpi_probe() is now moved into acpi_identify() and acpi_probe() is just a stub. - On i386 and amd64, an ACPI-specific nexus(4) driver checks for ACPI via acpi_identify() and claims the nexus0 device if the probe succeeds. It then explicitly adds an acpi0 device in its attach routine. - The legacy(4) driver no longer knows anything about the acpi0 device. - On ia64 if acpi_identify() fails you basically end up with no devices. This matches the previous behavior where the old acpi_identify() would fail to add an acpi0 device again leaving you with no devices. Discussed with: imp Silence on: arch@
2008-03-13 20:39:04 +00:00
return (ENXIO);
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
if (rsdp->Revision > 1 && rsdp->XsdtPhysicalAddress != 0)
paddr = (ACPI_PHYSICAL_ADDRESS)rsdp->XsdtPhysicalAddress;
else
paddr = (ACPI_PHYSICAL_ADDRESS)rsdp->RsdtPhysicalAddress;
AcpiOsUnmapMemory(rsdp, sizeof(ACPI_TABLE_RSDP));
/tmp/m
2004-08-03 05:13:56 +00:00
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
if ((rsdt = AcpiOsMapMemory(paddr, sizeof(ACPI_TABLE_HEADER))) == NULL)
Rework how the nexus(4) device works on x86 to better handle the idea of different "platforms" on x86 machines. The existing code already handles having two platforms: ACPI and legacy. However, the existing approach was rather hardcoded and difficult to extend. These changes take the approach that each x86 hardware platform should provide its own nexus(4) driver (it can inherit most of its behavior from the default legacy nexus(4) driver) which is responsible for probing for the platform and performing appropriate platform-specific setup during attach (such as adding a platform-specific bus device). This does mean changing the x86 platform busses to no longer use an identify routine for probing, but to move that logic into their matching nexus(4) driver instead. - Make the default nexus(4) driver in nexus.c on i386 and amd64 handle the legacy platform. It's probe routine now returns BUS_PROBE_GENERIC so it can be overriden. - Expose a nexus_init_resources() routine which initializes the various resource managers so that subclassed nexus(4) drivers can invoke it from their attach routine. - The legacy nexus(4) driver explicitly adds a legacy0 device in its attach routine. - The ACPI driver no longer contains an new-bus identify method. Instead it exposes a public function (acpi_identify()) which is a probe routine that the MD nexus(4) drivers can use to probe for ACPI. All of the probe logic in acpi_probe() is now moved into acpi_identify() and acpi_probe() is just a stub. - On i386 and amd64, an ACPI-specific nexus(4) driver checks for ACPI via acpi_identify() and claims the nexus0 device if the probe succeeds. It then explicitly adds an acpi0 device in its attach routine. - The legacy(4) driver no longer knows anything about the acpi0 device. - On ia64 if acpi_identify() fails you basically end up with no devices. This matches the previous behavior where the old acpi_identify() would fail to add an acpi0 device again leaving you with no devices. Discussed with: imp Silence on: arch@
2008-03-13 20:39:04 +00:00
return (ENXIO);
sbuf_new(&sb, acpi_desc, sizeof(acpi_desc), SBUF_FIXEDLEN);
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
sbuf_bcat(&sb, rsdt->OemId, ACPI_OEM_ID_SIZE);
sbuf_trim(&sb);
sbuf_putc(&sb, ' ');
sbuf_bcat(&sb, rsdt->OemTableId, ACPI_OEM_TABLE_ID_SIZE);
sbuf_trim(&sb);
sbuf_finish(&sb);
sbuf_delete(&sb);
AcpiOsUnmapMemory(rsdt, sizeof(ACPI_TABLE_HEADER));
Rework how the nexus(4) device works on x86 to better handle the idea of different "platforms" on x86 machines. The existing code already handles having two platforms: ACPI and legacy. However, the existing approach was rather hardcoded and difficult to extend. These changes take the approach that each x86 hardware platform should provide its own nexus(4) driver (it can inherit most of its behavior from the default legacy nexus(4) driver) which is responsible for probing for the platform and performing appropriate platform-specific setup during attach (such as adding a platform-specific bus device). This does mean changing the x86 platform busses to no longer use an identify routine for probing, but to move that logic into their matching nexus(4) driver instead. - Make the default nexus(4) driver in nexus.c on i386 and amd64 handle the legacy platform. It's probe routine now returns BUS_PROBE_GENERIC so it can be overriden. - Expose a nexus_init_resources() routine which initializes the various resource managers so that subclassed nexus(4) drivers can invoke it from their attach routine. - The legacy nexus(4) driver explicitly adds a legacy0 device in its attach routine. - The ACPI driver no longer contains an new-bus identify method. Instead it exposes a public function (acpi_identify()) which is a probe routine that the MD nexus(4) drivers can use to probe for ACPI. All of the probe logic in acpi_probe() is now moved into acpi_identify() and acpi_probe() is just a stub. - On i386 and amd64, an ACPI-specific nexus(4) driver checks for ACPI via acpi_identify() and claims the nexus0 device if the probe succeeds. It then explicitly adds an acpi0 device in its attach routine. - The legacy(4) driver no longer knows anything about the acpi0 device. - On ia64 if acpi_identify() fails you basically end up with no devices. This matches the previous behavior where the old acpi_identify() would fail to add an acpi0 device again leaving you with no devices. Discussed with: imp Silence on: arch@
2008-03-13 20:39:04 +00:00
snprintf(acpi_ca_version, sizeof(acpi_ca_version), "%x", ACPI_CA_VERSION);
return (0);
}
/*
* Fetch some descriptive data from ACPI to put in our attach message.
*/
static int
acpi_probe(device_t dev)
{
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
device_set_desc(dev, acpi_desc);
More BUS_PROBE_NOWILDCARD sweeping. Some devices here (if_ath_ahb and siba) resist easy conversion since they implement a great deal of their attach logic inside probe(). Some of this could be fixed by moving it to attach(), but some requires something more subtle than BUS_PROBE_NOWILDCARD.
2013-10-29 14:19:42 +00:00
return_VALUE (BUS_PROBE_NOWILDCARD);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
static int
acpi_attach(device_t dev)
{
struct acpi_softc *sc;
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
ACPI_STATUS status;
Select the highest valid (i.e., S3) sleep state for the default for the sleep button. Change the default for the lid switch to NONE. This can be overridden in /etc/sysctl.conf as desired.
2004-05-06 08:04:34 +00:00
int error, state;
Allow the ACPI subsystem to be disabled with a hint. Avoid fully initialising the ACPI namespace if we are attempting to avoid parts of it. This is a workaround for some systems that still crash the interpreter. Implement the ISA_IVAR_LOGICALID for ISA compatibility. Implement stubs for other PnP ID-related ivars.
2001-09-07 02:57:29 +00:00
UINT32 flags;
Select the highest valid (i.e., S3) sleep state for the default for the sleep button. Change the default for the lid switch to NONE. This can be overridden in /etc/sysctl.conf as desired.
2004-05-06 08:04:34 +00:00
UINT8 TypeA, TypeB;
Interpret new loader tunable "hw.acpi.verbose" to turn verbose mode on at boot time.
2002-10-31 20:23:41 +00:00
char *env;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
/tmp/m
2004-08-03 05:13:56 +00:00
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
sc = device_get_softc(dev);
sc->acpi_dev = dev;
CALLOUT_MPSAFE has lost its meaning since r141428, i.e., for more than ten years for head. However, it is continuously misused as the mpsafe argument for callout_init(9). Deprecate the flag and clean up callout_init() calls to make them more consistent. Differential Revision: https://reviews.freebsd.org/D2613 Reviewed by: jhb MFC after: 2 weeks
2015-05-22 17:05:21 +00:00
callout_init(&sc->susp_force_to, 1);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
error = ENXIO;
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
/* Initialize resource manager. */
acpi_rman_io.rm_type = RMAN_ARRAY;
acpi_rman_io.rm_start = 0;
acpi_rman_io.rm_end = 0xffff;
Give the ACPI I/O rman's unique description strings to make 'devinfo -u' output less confusing. MFC after: 3 days
2006-09-11 19:32:54 +00:00
acpi_rman_io.rm_descr = "ACPI I/O ports";
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
if (rman_init(&acpi_rman_io) != 0)
panic("acpi rman_init IO ports failed");
acpi_rman_mem.rm_type = RMAN_ARRAY;
Give the ACPI I/O rman's unique description strings to make 'devinfo -u' output less confusing. MFC after: 3 days
2006-09-11 19:32:54 +00:00
acpi_rman_mem.rm_descr = "ACPI I/O memory addresses";
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
if (rman_init(&acpi_rman_mem) != 0)
panic("acpi rman_init memory failed");
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
/* Initialise the ACPI mutex */
mtx_init(&acpi_mutex, "ACPI global lock", NULL, MTX_DEF);
/*
* Set the globals from our tunables. This is needed because ACPI-CA
* uses UINT8 for some values and we have no tunable_byte.
*/
- Allow users to enable dumping Debug objects without ACPI debugger. Setting the new sysctl MIB "debug.acpi.enable_debug_objects" to a non-zero value enables us to print Debug object when something is written to it. - Allow users to disable interpreter slack mode. Setting the new tunable "debug.acpi.interpreter_slack" to zero disables some workarounds for common BIOS mistakes and enables strict ACPI implementations by the specification.
2010-03-09 19:02:02 +00:00
AcpiGbl_EnableInterpreterSlack = acpi_interpreter_slack ? TRUE : FALSE;
AcpiGbl_EnableAmlDebugObject = acpi_debug_objects ? TRUE : FALSE;
Merge ACPICA 20140926.
2014-10-02 19:11:18 +00:00
AcpiGbl_UseDefaultRegisterWidths = acpi_ignore_reg_width ? TRUE : FALSE;
- Allow users to enable dumping Debug objects without ACPI debugger. Setting the new sysctl MIB "debug.acpi.enable_debug_objects" to a non-zero value enables us to print Debug object when something is written to it. - Allow users to disable interpreter slack mode. Setting the new tunable "debug.acpi.interpreter_slack" to zero disables some workarounds for common BIOS mistakes and enables strict ACPI implementations by the specification.
2010-03-09 19:02:02 +00:00
#ifndef ACPI_DEBUG
/*
* Disable all debugging layers and levels.
*/
AcpiDbgLayer = 0;
AcpiDbgLevel = 0;
#endif
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
Add two new loader tunables 'hw.acpi.install_interface' and 'hw.acpi.remove_interface'. hw.acpi.install_interface lets you install new interfaces. Conversely, hw.acpi.remove_interface lets you remove OS interfaces from the pre-defined list in ACPICA. For example, hw.acpi.install_interface="FreeBSD" lets _OSI("FreeBSD") method to return 0xffffffff (or success) and hw.acpi.remove_interface="Windows 2009" lets _OSI("Windows 2009") method to return zero (or failure). Both are comma-separated lists and leading white spaces are ignored. For example, the following examples are valid: hw.acpi.install_interface="Linux, FreeBSD" hw.acpi.remove_interface="Windows 2006, Windows 2006.1"
2010-10-26 18:59:50 +00:00
/* Override OS interfaces if the user requested. */
acpi_reset_interfaces(dev);
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
/* Load ACPI name space. */
status = AcpiLoadTables();
if (ACPI_FAILURE(status)) {
device_printf(dev, "Could not load Namespace: %s\n",
AcpiFormatException(status));
goto out;
}
Extend the support for PCI-e memory mapped configuration space access: - Rename pciereg_cfgopen() to pcie_cfgregopen() and expose it to the rest of the kernel. It now also accepts parameters via function arguments rather than global variables. - Add a notion of minimum and maximum bus numbers and reject requests for an out of range bus. - Add more range checks on slot/func/reg/bytes parameters to the cfg reg read/write routines. Don't panic on any invalid parameters, just fail the request (writes do nothing, reads return -1). This matches the behavior of the other cfg mechanisms. - Port the memory mapped configuration space access to amd64. On amd64 we simply use the direct map (via pmap_mapdev()) for the memory mapped window. - During acpi_attach() just after loading the ACPI tables, check for a MCFG table. If it exists, call pciereg_cfgopen() on each subtable (memory mapped window). For now we only support windows for domain 0 that start with bus 0. This removes the need for more chipset-specific quirks in the MD code. - Remove the chipset-specific quirks for the Intel 5000P/V/Z chipsets since these machines should all have MCFG tables via ACPI. - Updated pci_cfgregopen() to DTRT if ACPI had invoked pcie_cfgregopen() earlier. MFC after: 2 weeks
2008-08-22 02:14:23 +00:00
/* Handle MCFG table if present. */
acpi_enable_pcie();
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
* Note that some systems (specifically, those with namespace evaluation
* issues that require the avoidance of parts of the namespace) must
* avoid running _INI and _STA on everything, as well as dodging the final
* object init pass.
Allow the ACPI subsystem to be disabled with a hint. Avoid fully initialising the ACPI namespace if we are attempting to avoid parts of it. This is a workaround for some systems that still crash the interpreter. Implement the ISA_IVAR_LOGICALID for ISA compatibility. Implement stubs for other PnP ID-related ivars.
2001-09-07 02:57:29 +00:00
*
* For these devices, we set ACPI_NO_DEVICE_INIT and ACPI_NO_OBJECT_INIT).
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
* XXX We should arrange for the object init pass after we have attached
* all our child devices, but on many systems it works here.
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*/
Allow the ACPI subsystem to be disabled with a hint. Avoid fully initialising the ACPI namespace if we are attempting to avoid parts of it. This is a workaround for some systems that still crash the interpreter. Implement the ISA_IVAR_LOGICALID for ISA compatibility. Implement stubs for other PnP ID-related ivars.
2001-09-07 02:57:29 +00:00
flags = 0;
Rework the kernel environment subsystem. We now convert the static environment needed at boot time to a dynamic subsystem when VM is up. The dynamic kernel environment is protected by an sx lock. This adds some new functions to manipulate the kernel environment : freeenv(), setenv(), unsetenv() and testenv(). freeenv() has to be called after every getenv() when you have finished using the string. testenv() only tests if an environment variable is present, and doesn't require a freeenv() call. setenv() and unsetenv() are self explanatory. The kenv(2) syscall exports these new functionalities to userland, mainly for kenv(1). Reviewed by: peter
2002-04-17 13:06:36 +00:00
if (testenv("debug.acpi.avoid"))
Allow the ACPI subsystem to be disabled with a hint. Avoid fully initialising the ACPI namespace if we are attempting to avoid parts of it. This is a workaround for some systems that still crash the interpreter. Implement the ISA_IVAR_LOGICALID for ISA compatibility. Implement stubs for other PnP ID-related ivars.
2001-09-07 02:57:29 +00:00
flags = ACPI_NO_DEVICE_INIT | ACPI_NO_OBJECT_INIT;
/tmp/m
2004-08-03 05:13:56 +00:00
/* Bring the hardware and basic handlers online. */
Match namespace cleanup changes in ACPI CA 20020217 update. Use ACPI_SUCCESS/ACPI_FAILURE consistently. The AcpiGetInto* interfaces are obsoleted by ACPI_ALLOCATE_BUFFER. Add AcpiBatteryIsPresent helper to determine whether a battery device is inserted. Add ACPI_ALL_DRIVERS to the list of debug layers, now that we own the namespace for this. Pr:
2002-02-23 05:21:56 +00:00
if (ACPI_FAILURE(status = AcpiEnableSubsystem(flags))) {
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
device_printf(dev, "Could not enable ACPI: %s\n",
AcpiFormatException(status));
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
goto out;
Resolve conflicts arising from the ACPI CA 20020725 import.
2002-07-30 19:35:32 +00:00
}
Add ECDT (ACPI 2.0) support. This allows the EC to be enabled before the namespace has been evaluated. Machines with ACPI 2.0 expect this behavior and have AML which calls EC functions early in the boot process. If the ECDT is not available, fall back to original probe behavior. Other minor changes: * Add GPE bit and GLK usage to the device announcement * Always use the global lock in the ECDT case, but potentially downgrade to not using it if _GLK is 0 once the namespace is available. This is announced with "Changing GLK from 1 to 0" * Remove the acpi_object_list definitions which were earlier deprecated Ideas from: takawata
2003-07-20 00:48:38 +00:00
/*
* Call the ECDT probe function to provide EC functionality before
* the namespace has been evaluated.
Don't attach special devices in the order they appear in the AML tree. If the embedded controller exists before the sysresource devices, for example, it will be attached first. Instead, let the normal device order function work as we first desired. [1] There still remained a problem where we couldn't allocate resources in acpi0 that were passed up by the sysresource pseudo-devices. These devices had to probe/attach first to give their resources to acpi, then acpi would allocate them before probing/attaching other devices. To work around this, we attach them from acpi_sysres_alloc(). A better approach would be to implement multi-pass probe/attach in newbus but that's a much bigger task. Suggested by: jhb [1] Hardware from: Centaur Technologies MFC after: 1 week
2006-05-07 03:28:10 +00:00
*
* XXX This happens before the sysresource devices have been probed and
* attached so its resources come from nexus0. In practice, this isn't
* a problem but should be addressed eventually.
Add ECDT (ACPI 2.0) support. This allows the EC to be enabled before the namespace has been evaluated. Machines with ACPI 2.0 expect this behavior and have AML which calls EC functions early in the boot process. If the ECDT is not available, fall back to original probe behavior. Other minor changes: * Add GPE bit and GLK usage to the device announcement * Always use the global lock in the ECDT case, but potentially downgrade to not using it if _GLK is 0 once the namespace is available. This is announced with "Changing GLK from 1 to 0" * Remove the acpi_object_list definitions which were earlier deprecated Ideas from: takawata
2003-07-20 00:48:38 +00:00
*/
acpi_ec_ecdt_probe(dev);
/tmp/m
2004-08-03 05:13:56 +00:00
/* Bring device objects and regions online. */
Resolve conflicts arising from the ACPI CA 20020725 import.
2002-07-30 19:35:32 +00:00
if (ACPI_FAILURE(status = AcpiInitializeObjects(flags))) {
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
device_printf(dev, "Could not initialize ACPI objects: %s\n",
AcpiFormatException(status));
Resolve conflicts arising from the ACPI CA 20020725 import.
2002-07-30 19:35:32 +00:00
goto out;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Add 3 new dynamic sysctl's to control the sleep states switched to on a power button, sleep button, or lid close event. The sysctl's use the ACPI sleep state names S0, S1, S2, S3, S4, S4B, and S5. Reviewed by: iwasaki
2001-01-13 21:28:57 +00:00
/*
* Setup our sysctl tree.
*
* XXX: This doesn't check to make sure that none of these fail.
*/
sysctl_ctx_init(&sc->acpi_sysctl_ctx);
sc->acpi_sysctl_tree = SYSCTL_ADD_NODE(&sc->acpi_sysctl_ctx,
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
SYSCTL_STATIC_CHILDREN(_hw), OID_AUTO, device_get_name(dev),
CTLFLAG_RD | CTLFLAG_MPSAFE, 0, "");
Add new sysctl MIB (hw.acpi.supported_sleep_state) to indicate the list of supported sleep state. This should help people understand what following message means. acpi0: AcpiGetSleepTypeData failed - AE_NOT_FOUND MFC after: 3 days
2003-04-11 16:53:56 +00:00
SYSCTL_ADD_PROC(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
OID_AUTO, "supported_sleep_state",
CTLTYPE_STRING | CTLFLAG_RD | CTLFLAG_NEEDGIANT,
document some ACPI related sysctls. Submitted by: Oliver Pinter <oliver.pinter@hardenedbsd.org> Sponsored by: HardenedBSD Differential Revision: https://reviews.freebsd.org/D5263
2016-02-19 05:02:17 +00:00
0, 0, acpi_supported_sleep_state_sysctl, "A",
"List supported ACPI sleep states.");
Add 3 new dynamic sysctl's to control the sleep states switched to on a power button, sleep button, or lid close event. The sysctl's use the ACPI sleep state names S0, S1, S2, S3, S4, S4B, and S5. Reviewed by: iwasaki
2001-01-13 21:28:57 +00:00
SYSCTL_ADD_PROC(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
OID_AUTO, "power_button_state",
CTLTYPE_STRING | CTLFLAG_RW | CTLFLAG_NEEDGIANT,
document some ACPI related sysctls. Submitted by: Oliver Pinter <oliver.pinter@hardenedbsd.org> Sponsored by: HardenedBSD Differential Revision: https://reviews.freebsd.org/D5263
2016-02-19 05:02:17 +00:00
&sc->acpi_power_button_sx, 0, acpi_sleep_state_sysctl, "A",
"Power button ACPI sleep state.");
Add 3 new dynamic sysctl's to control the sleep states switched to on a power button, sleep button, or lid close event. The sysctl's use the ACPI sleep state names S0, S1, S2, S3, S4, S4B, and S5. Reviewed by: iwasaki
2001-01-13 21:28:57 +00:00
SYSCTL_ADD_PROC(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
OID_AUTO, "sleep_button_state",
CTLTYPE_STRING | CTLFLAG_RW | CTLFLAG_NEEDGIANT,
document some ACPI related sysctls. Submitted by: Oliver Pinter <oliver.pinter@hardenedbsd.org> Sponsored by: HardenedBSD Differential Revision: https://reviews.freebsd.org/D5263
2016-02-19 05:02:17 +00:00
&sc->acpi_sleep_button_sx, 0, acpi_sleep_state_sysctl, "A",
"Sleep button ACPI sleep state.");
Add 3 new dynamic sysctl's to control the sleep states switched to on a power button, sleep button, or lid close event. The sysctl's use the ACPI sleep state names S0, S1, S2, S3, S4, S4B, and S5. Reviewed by: iwasaki
2001-01-13 21:28:57 +00:00
SYSCTL_ADD_PROC(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
OID_AUTO, "lid_switch_state",
CTLTYPE_STRING | CTLFLAG_RW | CTLFLAG_NEEDGIANT,
document some ACPI related sysctls. Submitted by: Oliver Pinter <oliver.pinter@hardenedbsd.org> Sponsored by: HardenedBSD Differential Revision: https://reviews.freebsd.org/D5263
2016-02-19 05:02:17 +00:00
&sc->acpi_lid_switch_sx, 0, acpi_sleep_state_sysctl, "A",
"Lid ACPI sleep state. Set to S3 if you want to suspend your laptop when close the Lid.");
Add APM compatibility feature to ACPI. This emulates APM device node interface APIs (mainly ioctl) and provides APM services for the applications. The goal is to support most of APM applications without any changes. Implemented ioctls in this commit are: - APMIO_SUSPEND (mapped ACPI S3 as default but changable by sysctl) - APMIO_STANDBY (mapped ACPI S1 as default but changable by sysctl) - APMIO_GETINFO and APMIO_GETINFO_OLD - APMIO_GETPWSTATUS With above, many APM applications which get batteries, ac-line info. and transition the system into suspend/standby mode (such as wmapm, xbatt) should work with ACPI enabled kernel (if ACPI works well :-) Reviewed by: arch@, audit@ and some guys
2001-10-26 17:43:05 +00:00
SYSCTL_ADD_PROC(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
OID_AUTO, "standby_state",
CTLTYPE_STRING | CTLFLAG_RW | CTLFLAG_NEEDGIANT,
Add APM compatibility feature to ACPI. This emulates APM device node interface APIs (mainly ioctl) and provides APM services for the applications. The goal is to support most of APM applications without any changes. Implemented ioctls in this commit are: - APMIO_SUSPEND (mapped ACPI S3 as default but changable by sysctl) - APMIO_STANDBY (mapped ACPI S1 as default but changable by sysctl) - APMIO_GETINFO and APMIO_GETINFO_OLD - APMIO_GETPWSTATUS With above, many APM applications which get batteries, ac-line info. and transition the system into suspend/standby mode (such as wmapm, xbatt) should work with ACPI enabled kernel (if ACPI works well :-) Reviewed by: arch@, audit@ and some guys
2001-10-26 17:43:05 +00:00
&sc->acpi_standby_sx, 0, acpi_sleep_state_sysctl, "A", "");
SYSCTL_ADD_PROC(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
OID_AUTO, "suspend_state",
CTLTYPE_STRING | CTLFLAG_RW | CTLFLAG_NEEDGIANT,
Add APM compatibility feature to ACPI. This emulates APM device node interface APIs (mainly ioctl) and provides APM services for the applications. The goal is to support most of APM applications without any changes. Implemented ioctls in this commit are: - APMIO_SUSPEND (mapped ACPI S3 as default but changable by sysctl) - APMIO_STANDBY (mapped ACPI S1 as default but changable by sysctl) - APMIO_GETINFO and APMIO_GETINFO_OLD - APMIO_GETPWSTATUS With above, many APM applications which get batteries, ac-line info. and transition the system into suspend/standby mode (such as wmapm, xbatt) should work with ACPI enabled kernel (if ACPI works well :-) Reviewed by: arch@, audit@ and some guys
2001-10-26 17:43:05 +00:00
&sc->acpi_suspend_sx, 0, acpi_sleep_state_sysctl, "A", "");
Add new sysctl MIB (hw.acpi.sleep_delay) to specify the delay (in seconds) before ACPI sleep. Some machines might need this to sleep by Hot-key.
2002-08-25 06:13:53 +00:00
SYSCTL_ADD_INT(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
Minor sysctl cleanup. The RW flag means read|write and so it is redundant to add the RD flag. Also, the debug node does not need to be writable.
2006-06-10 08:04:38 +00:00
OID_AUTO, "sleep_delay", CTLFLAG_RW, &sc->acpi_sleep_delay, 0,
Improve the sleep_delay sysctl description by specifying which unit the number is in. PR: 159975 Submitted by: gcooper Approved by: re (kib) MFC after: 1 week
2011-09-13 15:57:29 +00:00
"sleep delay in seconds");
Add S4BIOS sleep (BIOS hibernation) and DSDT overriding support. - Add S4BIOS sleep implementation. This will works well if MIB hw.acpi.s4bios is set (and of course BIOS supports it and hibernation is enabled correctly). - Add DSDT overriding support which is submitted by takawata originally. If loader tunable acpi_dsdt_load="YES" and DSDT file is set to acpi_dsdt_name (default DSDT file name is /boot/acpi_dsdt.aml), ACPI CA core loads DSDT from given file rather than BIOS memory block. DSDT file can be generated by iasl in ports/devel/acpicatools/. - Add new files so that we can add our proposed additional code to Intel ACPI CA into these files temporary. They will be removed when similar code is added into ACPI CA officially.
2001-11-06 15:00:30 +00:00
SYSCTL_ADD_INT(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
Minor sysctl cleanup. The RW flag means read|write and so it is redundant to add the RD flag. Also, the debug node does not need to be writable.
2006-06-10 08:04:38 +00:00
OID_AUTO, "s4bios", CTLFLAG_RW, &sc->acpi_s4bios, 0, "S4BIOS mode");
Some small improvements of ACPI thermal driver. - Give a guaranteed minimum cooling run time to avoid too frequent cooling system On/Off switching. The minimum cooling run time can be specified by hw.acpi.thermal.min_runtime in sec. - Refine message printing (_AC-1 -> NONE). - Add verbose mode enable/disable capability by hw.acpi.verbose in bool. Reviewed by: acpi-jp@ folks
2001-10-29 18:09:43 +00:00
SYSCTL_ADD_INT(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
Minor sysctl cleanup. The RW flag means read|write and so it is redundant to add the RD flag. Also, the debug node does not need to be writable.
2006-06-10 08:04:38 +00:00
OID_AUTO, "verbose", CTLFLAG_RW, &sc->acpi_verbose, 0, "verbose mode");
By default, don't disable ACPI during reboot. This appears to hang some systems. Introduce a new sysctl "hw.acpi.disable_on_reboot" that allows users to re-enable the old behavior in case it's needed for some systems. We never disable in the power-off path. Original approach submitted by Alexander Logvinov <abuse@akavia.ru> with reworking by Jung-uk Kim and myself.
2006-06-11 20:31:41 +00:00
SYSCTL_ADD_INT(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
OID_AUTO, "disable_on_reboot", CTLFLAG_RW,
&sc->acpi_do_disable, 0, "Disable ACPI when rebooting/halting system");
Add a new sysctl, hw.acpi.handle_reboot. If set, acpi will attempt to perform the reboot action via the reset register instead of our legacy method. Default is 0 (use legacy). This is needed because some systems hang on reboot even though they claim to support the reset register. MFC after: 2 days
2006-07-29 21:46:16 +00:00
SYSCTL_ADD_INT(&sc->acpi_sysctl_ctx, SYSCTL_CHILDREN(sc->acpi_sysctl_tree),
OID_AUTO, "handle_reboot", CTLFLAG_RW,
&sc->acpi_handle_reboot, 0, "Use ACPI Reset Register to reboot");
Default to 5 seconds before sleeping to give some machines time to stabilize. This doesn't break anything on my laptop and some claim it helps them.
2003-08-08 03:19:22 +00:00
/*
Decrease sleep_delay default to 1 second now that the machines that required the 5 second delay have been fixed.
2004-05-29 05:34:04 +00:00
* Default to 1 second before sleeping to give some machines time to
Default to 5 seconds before sleeping to give some machines time to stabilize. This doesn't break anything on my laptop and some claim it helps them.
2003-08-08 03:19:22 +00:00
* stabilize.
*/
Decrease sleep_delay default to 1 second now that the machines that required the 5 second delay have been fixed.
2004-05-29 05:34:04 +00:00
sc->acpi_sleep_delay = 1;
Some small improvements of ACPI thermal driver. - Give a guaranteed minimum cooling run time to avoid too frequent cooling system On/Off switching. The minimum cooling run time can be specified by hw.acpi.thermal.min_runtime in sec. - Refine message printing (_AC-1 -> NONE). - Add verbose mode enable/disable capability by hw.acpi.verbose in bool. Reviewed by: acpi-jp@ folks
2001-10-29 18:09:43 +00:00
if (bootverbose)
sc->acpi_verbose = 1;
Follow up to r225617. In order to maximize the re-usability of kernel code in userland rename in-kernel getenv()/setenv() to kern_setenv()/kern_getenv(). This fixes a namespace collision with libc symbols. Submitted by: kmacy Tested by: make universe
2014-10-16 18:04:43 +00:00
if ((env = kern_getenv("hw.acpi.verbose")) != NULL) {
Always free the returned env pointer even it doesn't indicate we're in verbose mode. Found by: Coverity Prevent (via sam)
2005-03-31 19:07:26 +00:00
if (strcmp(env, "0") != 0)
sc->acpi_verbose = 1;
Interpret new loader tunable "hw.acpi.verbose" to turn verbose mode on at boot time.
2002-10-31 20:23:41 +00:00
freeenv(env);
}
Use AcpiReset() from ACPICA instead of rolling our own, which is actually incomplete. If FADT says the register is available, enable the capability by default. Remove the previous default value from acpi(4).
2010-10-13 00:21:53 +00:00
/* Only enable reboot by default if the FADT says it is available. */
if (AcpiGbl_FADT.Flags & ACPI_FADT_RESET_REGISTER)
sc->acpi_handle_reboot = 1;
AcpiGbl_FACS will not be defined when building using the reduced hardware model. This may be the case on ARM.
2015-05-06 14:14:14 +00:00
#if !ACPI_REDUCED_HARDWARE
Only enable S4BIOS by default if the FACS says it is available. The user can override this with a sysctl. Be sure to return the acpi_SetSleepState return value to userland.
2003-09-15 06:29:31 +00:00
/* Only enable S4BIOS by default if the FACS says it is available. */
There may not be an FACS table, check for this before accessing it. Sponsored by: The FreeBSD Foundation
2015-04-28 16:06:58 +00:00
if (AcpiGbl_FACS != NULL && AcpiGbl_FACS->Flags & ACPI_FACS_S4_BIOS_PRESENT)
Add new quirk code that disables problem BIOS versions. Remove old quirk code that was never really used. Print a message when disabling ACPI via a quirk. Allow the user to override the blacklist decision by setting hint.acpi.0.disabled="0". Add missing AcpiTerminate() calls; they are needed to clean up if bailing out after AcpiInitializeSubsystem().
2004-06-30 04:49:54 +00:00
sc->acpi_s4bios = 1;
AcpiGbl_FACS will not be defined when building using the reduced hardware model. This may be the case on ARM.
2015-05-06 14:14:14 +00:00
#endif
Only enable S4BIOS by default if the FACS says it is available. The user can override this with a sysctl. Be sure to return the acpi_SetSleepState return value to userland.
2003-09-15 06:29:31 +00:00
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
/* Probe all supported sleep states. */
acpi_sleep_states[ACPI_STATE_S0] = TRUE;
for (state = ACPI_STATE_S1; state < ACPI_S_STATE_COUNT; state++)
Merge ACPICA 20130117.
2013-01-17 23:56:43 +00:00
if (ACPI_SUCCESS(AcpiEvaluateObject(ACPI_ROOT_OBJECT,
__DECONST(char *, AcpiGbl_SleepStateNames[state]), NULL, NULL)) &&
ACPI_SUCCESS(AcpiGetSleepTypeData(state, &TypeA, &TypeB)))
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
acpi_sleep_states[state] = TRUE;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
Select the highest valid (i.e., S3) sleep state for the default for the sleep button. Change the default for the lid switch to NONE. This can be overridden in /etc/sysctl.conf as desired.
2004-05-06 08:04:34 +00:00
* Dispatch the default sleep state to devices. The lid switch is set
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
* to UNKNOWN by default to avoid surprising users.
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*/
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
sc->acpi_power_button_sx = acpi_sleep_states[ACPI_STATE_S5] ?
ACPI_STATE_S5 : ACPI_STATE_UNKNOWN;
sc->acpi_lid_switch_sx = ACPI_STATE_UNKNOWN;
sc->acpi_standby_sx = acpi_sleep_states[ACPI_STATE_S1] ?
ACPI_STATE_S1 : ACPI_STATE_UNKNOWN;
sc->acpi_suspend_sx = acpi_sleep_states[ACPI_STATE_S3] ?
ACPI_STATE_S3 : ACPI_STATE_UNKNOWN;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Select the highest valid (i.e., S3) sleep state for the default for the sleep button. Change the default for the lid switch to NONE. This can be overridden in /etc/sysctl.conf as desired.
2004-05-06 08:04:34 +00:00
/* Pick the first valid sleep state for the sleep button default. */
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
sc->acpi_sleep_button_sx = ACPI_STATE_UNKNOWN;
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
for (state = ACPI_STATE_S1; state <= ACPI_STATE_S4; state++)
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
if (acpi_sleep_states[state]) {
Select the highest valid (i.e., S3) sleep state for the default for the sleep button. Change the default for the lid switch to NONE. This can be overridden in /etc/sysctl.conf as desired.
2004-05-06 08:04:34 +00:00
sc->acpi_sleep_button_sx = state;
break;
}
Enable fixed event at not only boot but also wakeup. Reported and patch tested by: Peter Dufault <dufault@hda.hda.com>
2001-01-10 18:01:51 +00:00
acpi_enable_fixed_events(sc);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
* Scan the namespace and attach/initialise children.
*/
Remove disable_on_poweroff and our pre-sync shutdown handler. Disabling of GPEs is now done in acpi_shutdown() and so we no longer need the option of disabling ACPI in the poweroff case.
2004-06-14 04:37:45 +00:00
/* Register our shutdown handler. */
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
EVENTHANDLER_REGISTER(shutdown_final, acpi_shutdown_final, sc,
SHUTDOWN_PRI_LAST);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
* Register our acpi event handlers.
* XXX should be configurable eg. via userland policy manager.
*/
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
EVENTHANDLER_REGISTER(acpi_sleep_event, acpi_system_eventhandler_sleep,
sc, ACPI_EVENT_PRI_LAST);
EVENTHANDLER_REGISTER(acpi_wakeup_event, acpi_system_eventhandler_wakeup,
sc, ACPI_EVENT_PRI_LAST);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Flag our initial states. */
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
sc->acpi_enabled = TRUE;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
sc->acpi_sstate = ACPI_STATE_S0;
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
sc->acpi_sleep_disabled = TRUE;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Create the control device */
Change the group and the permissions on /dev/acpi, to make "acpiconf" work when called by members of the 'operator' group. They are already allowed to eg power off the system (via suid shutdown(8)), so they might as well be permitted to suspend it. Tested by: xmj@ Reviewed by: delphij@ MFC after: 2 weeks Relnotes: yes Sponsored by: DARPA, AFRL Differential Revision: https://reviews.freebsd.org/D16062
2018-07-02 14:15:30 +00:00
sc->acpi_dev_t = make_dev(&acpi_cdevsw, 0, UID_ROOT, GID_OPERATOR, 0664,
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
"acpi");
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
sc->acpi_dev_t->si_drv1 = sc;
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
if ((error = acpi_machdep_init(dev)))
Add APM compatibility feature to ACPI. This emulates APM device node interface APIs (mainly ioctl) and provides APM services for the applications. The goal is to support most of APM applications without any changes. Implemented ioctls in this commit are: - APMIO_SUSPEND (mapped ACPI S3 as default but changable by sysctl) - APMIO_STANDBY (mapped ACPI S1 as default but changable by sysctl) - APMIO_GETINFO and APMIO_GETINFO_OLD - APMIO_GETPWSTATUS With above, many APM applications which get batteries, ac-line info. and transition the system into suspend/standby mode (such as wmapm, xbatt) should work with ACPI enabled kernel (if ACPI works well :-) Reviewed by: arch@, audit@ and some guys
2001-10-26 17:43:05 +00:00
goto out;
Some fix for the recent apm module changes. - Now that apm loadable module can inform its existence to other kernel components (e.g. i386/isa/clock.c:startrtclock()'s TCS hack). - Exchange priority of SI_SUB_CPU and SI_SUB_KLD for above purpose. - Add simple arbitration mechanism for APM vs. ACPI. This prevents the kernel enables both of them. - Remove obsolete `#ifdef DEV_APM' related code. - Add abstracted interface for Powermanagement operations. Public apm(4) functions, such as apm_suspend(), should be replaced new interfaces. Currently only power_pm_suspend (successor of apm_suspend) is implemented. Reviewed by: peter, arch@ and audit@
2001-11-01 16:34:07 +00:00
/* Register ACPI again to pass the correct argument of pm_func. */
power_pm_register(POWER_PM_TYPE_ACPI, acpi_pm_func, sc);
Add a new device control utility for new-bus devices called devctl. This allows the user to request administrative changes to individual devices such as attach or detaching drivers or disabling and re-enabling devices. - Add a new /dev/devctl2 character device which uses ioctls for device requests. The ioctls use a common 'struct devreq' which is somewhat similar to 'struct ifreq'. - The ioctls identify the device to operate on via a string. This string can either by the device's name, or it can be a bus-specific address. (For unattached devices, a bus address is the only way to locate a device.) Bus drivers register an eventhandler to claim unrecognized device names that the driver recognizes as a valid address. Two buses currently support addresses: ACPI recognizes any device in the ACPI namespace via its full path starting with "\" and the PCI bus driver recognizes an address specification of 'pci[<domain>:]<bus>:<slot>:<func>' (identical to the PCI selector strings supported by pciconf). - To make it easier to cut and paste, change the PnP location string in the PCI bus driver to output a full PCI selector string rather than 'slot=<slot> function=<func>'. - Add a devctl(3) interface in libdevctl which provides a wrapper around the ioctls and is the preferred interface for other userland code. - Add a devctl(8) program which is a simple wrapper around the requests supported by devctl(3). - Add a device_is_suspended() function to check DF_SUSPENDED. - Add a resource_unset_value() function that can be used to remove a hint from the kernel environment. This is used to clear a hint.<driver>.<unit>.disabled hint when re-enabling a boot-time disabled device. Reviewed by: imp (parts) Requested by: imp (changing PCI location string) Relnotes: yes
2015-02-06 16:09:01 +00:00
if (!acpi_disabled("bus")) {
EVENTHANDLER_REGISTER(dev_lookup, acpi_lookup, NULL, 1000);
Attach ACPI children a bit later in attach(), specifically after performing any machine dependent initialization. This allows the MD code to set the interrupt routing model so that PCI interrupts are routed correctly when using an APIC or SAPIC for example.
2002-09-06 17:01:06 +00:00
acpi_probe_children(dev);
Add a new device control utility for new-bus devices called devctl. This allows the user to request administrative changes to individual devices such as attach or detaching drivers or disabling and re-enabling devices. - Add a new /dev/devctl2 character device which uses ioctls for device requests. The ioctls use a common 'struct devreq' which is somewhat similar to 'struct ifreq'. - The ioctls identify the device to operate on via a string. This string can either by the device's name, or it can be a bus-specific address. (For unattached devices, a bus address is the only way to locate a device.) Bus drivers register an eventhandler to claim unrecognized device names that the driver recognizes as a valid address. Two buses currently support addresses: ACPI recognizes any device in the ACPI namespace via its full path starting with "\" and the PCI bus driver recognizes an address specification of 'pci[<domain>:]<bus>:<slot>:<func>' (identical to the PCI selector strings supported by pciconf). - To make it easier to cut and paste, change the PnP location string in the PCI bus driver to output a full PCI selector string rather than 'slot=<slot> function=<func>'. - Add a devctl(3) interface in libdevctl which provides a wrapper around the ioctls and is the preferred interface for other userland code. - Add a devctl(8) program which is a simple wrapper around the requests supported by devctl(3). - Add a device_is_suspended() function to check DF_SUSPENDED. - Add a resource_unset_value() function that can be used to remove a hint from the kernel environment. This is used to clear a hint.<driver>.<unit>.disabled hint when re-enabling a boot-time disabled device. Reviewed by: imp (parts) Requested by: imp (changing PCI location string) Relnotes: yes
2015-02-06 16:09:01 +00:00
}
Attach ACPI children a bit later in attach(), specifically after performing any machine dependent initialization. This allows the MD code to set the interrupt routing model so that PCI interrupts are routed correctly when using an APIC or SAPIC for example.
2002-09-06 17:01:06 +00:00
Merge ACPICA 20101209.
2010-12-15 23:48:45 +00:00
/* Update all GPEs and enable runtime GPEs. */
status = AcpiUpdateAllGpes();
if (ACPI_FAILURE(status))
device_printf(dev, "Could not update all GPEs: %s\n",
AcpiFormatException(status));
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
/* Allow sleep request after a while. */
Convert from timeout(9) to callout(9).
2014-09-22 14:27:26 +00:00
callout_init_mtx(&acpi_sleep_timer, &acpi_mutex, 0);
callout_reset(&acpi_sleep_timer, hz * ACPI_MINIMUM_AWAKETIME,
acpi_sleep_enable, sc);
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
error = 0;
out:
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return_VALUE (error);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
static void
acpi_set_power_children(device_t dev, int state)
{
Use the correct device. Note this commit complements r274386. PR: 194884
2014-11-11 19:42:10 +00:00
device_t child;
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
device_t *devlist;
int dstate, i, numdevs;
if (device_get_children(dev, &devlist, &numdevs) != 0)
return;
/*
* Retrieve and set D-state for the sleep state if _SxD is present.
* Skip children who aren't attached since they are handled separately.
*/
for (i = 0; i < numdevs; i++) {
child = devlist[i];
dstate = state;
if (device_is_attached(child) &&
Use the correct device. Note this commit complements r274386. PR: 194884
2014-11-11 19:42:10 +00:00
acpi_device_pwr_for_sleep(dev, child, &dstate) == 0)
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
acpi_set_powerstate(child, dstate);
}
free(devlist, M_TEMP);
}
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
static int
acpi_suspend(device_t dev)
{
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
int error;
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
Temporarily revert the new-bus locking for 8.0 release. It will be reintroduced after HEAD is reopened for commits by re@. Approved by: re (kib), attilio
2009-08-20 19:17:53 +00:00
GIANT_REQUIRED;
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
error = bus_generic_suspend(dev);
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
if (error == 0)
acpi_set_power_children(dev, ACPI_STATE_D3);
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
return (error);
}
static int
acpi_resume(device_t dev)
{
Temporarily revert the new-bus locking for 8.0 release. It will be reintroduced after HEAD is reopened for commits by re@. Approved by: re (kib), attilio
2009-08-20 19:17:53 +00:00
GIANT_REQUIRED;
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
acpi_set_power_children(dev, ACPI_STATE_D0);
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
return (bus_generic_resume(dev));
}
Disable wake GPEs in the reboot path as well as poweroff path. This fixes "stray irq 9" messages on my Thinkpad. It may also help with general reboot consistency although the recent hang on reboot was solved by acpi_cpu.c rev 1.39.
2004-06-05 07:25:58 +00:00
static int
acpi_shutdown(device_t dev)
{
Temporarily revert the new-bus locking for 8.0 release. It will be reintroduced after HEAD is reopened for commits by re@. Approved by: re (kib), attilio
2009-08-20 19:17:53 +00:00
GIANT_REQUIRED;
Don't forget to pass shutdown events down to children first now that we handle them at the bus level too.
2004-06-05 09:56:04 +00:00
/* Allow children to shutdown first. */
bus_generic_shutdown(dev);
/tmp/m
2004-08-03 05:13:56 +00:00
/*
* Enable any GPEs that are able to power-on the system (i.e., RTC).
* Also, disable any that are not valid for this state (most).
*/
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
acpi_wake_prep_walk(ACPI_STATE_S5);
Disable wake GPEs in the reboot path as well as poweroff path. This fixes "stray irq 9" messages on my Thinkpad. It may also help with general reboot consistency although the recent hang on reboot was solved by acpi_cpu.c rev 1.39.
2004-06-05 07:25:58 +00:00
return (0);
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
* Handle a new device being added
*/
static device_t
bus_add_child: change type of order parameter to u_int This reflects actual type used to store and compare child device orders. Change is mostly done via a Coccinelle (soon to be devel/coccinelle) semantic patch. Verified by LINT+modules kernel builds. Followup to: r212213 MFC after: 10 days
2010-09-10 11:19:03 +00:00
acpi_add_child(device_t bus, u_int order, const char *name, int unit)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
struct acpi_device *ad;
device_t child;
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
if ((ad = malloc(sizeof(*ad), M_ACPIDEV, M_NOWAIT | M_ZERO)) == NULL)
return (NULL);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
resource_list_init(&ad->ad_rl);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
child = device_add_child_ordered(bus, order, name, unit);
if (child != NULL)
device_set_ivars(child, ad);
If a device_add_child fails (i.e. low memory situation), be sure to free the unused ivars also. Submitted by: pjd Obtained from: Coverity Prevent analysis
2005-03-27 03:37:43 +00:00
else
free(ad, M_ACPIDEV);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (child);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
static int
acpi_print_child(device_t bus, device_t child)
{
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
struct acpi_device *adev = device_get_ivars(child);
struct resource_list *rl = &adev->ad_rl;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
int retval = 0;
retval += bus_print_child_header(bus, child);
Use uintmax_t (typedef'd to rman_res_t type) for rman ranges. On some architectures, u_long isn't large enough for resource definitions. Particularly, powerpc and arm allow 36-bit (or larger) physical addresses, but type `long' is only 32-bit. This extends rman's resources to uintmax_t. With this change, any resource can feasibly be placed anywhere in physical memory (within the constraints of the driver). Why uintmax_t and not something machine dependent, or uint64_t? Though it's possible for uintmax_t to grow, it's highly unlikely it will become 128-bit on 32-bit architectures. 64-bit architectures should have plenty of RAM to absorb the increase on resource sizes if and when this occurs, and the number of resources on memory-constrained systems should be sufficiently small as to not pose a drastic overhead. That being said, uintmax_t was chosen for source clarity. If it's specified as uint64_t, all printf()-like calls would either need casts to uintmax_t, or be littered with PRI*64 macros. Casts to uintmax_t aren't horrible, but it would also bake into the API for resource_list_print_type() either a hidden assumption that entries get cast to uintmax_t for printing, or these calls would need the PRI*64 macros. Since source code is meant to be read more often than written, I chose the clearest path of simply using uintmax_t. Tested on a PowerPC p5020-based board, which places all device resources in 0xfxxxxxxxx, and has 8GB RAM. Regression tested on qemu-system-i386 Regression tested on qemu-system-mips (malta profile) Tested PAE and devinfo on virtualbox (live CD) Special thanks to bz for his testing on ARM. Reviewed By: bz, jhb (previous) Relnotes: Yes Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D4544
2016-03-18 01:28:41 +00:00
retval += resource_list_print_type(rl, "port", SYS_RES_IOPORT, "%#jx");
retval += resource_list_print_type(rl, "iomem", SYS_RES_MEMORY, "%#jx");
retval += resource_list_print_type(rl, "irq", SYS_RES_IRQ, "%jd");
retval += resource_list_print_type(rl, "drq", SYS_RES_DRQ, "%jd");
If flags are present, print them like ISA does. MFC after: 1 day
2004-10-13 07:27:21 +00:00
if (device_get_flags(child))
retval += printf(" flags %#x", device_get_flags(child));
Add a bus method to fetch the VM domain for the given device/bus. * Add a bus_if.m method - get_domain() - returning the VM domain or ENOENT if the device isn't in a VM domain; * Add bus methods to print out the domain of the device if appropriate; * Add code in srat.c to save the PXM -> VM domain mapping that's done and expose a function to translate VM domain -> PXM; * Add ACPI and ACPI PCI methods to check if the bus has a _PXM attribute and if so map it to the VM domain; * (.. yes, this works recursively.) * Have the pci bus glue print out the device VM domain if present. Note: this is just the plumbing to start enumerating information - it doesn't at all modify behaviour. Differential Revision: D906 Reviewed by: jhb Sponsored by: Norse Corp
2014-10-09 05:33:25 +00:00
retval += bus_print_child_domain(bus, child);
Print before the footer, not after.
2004-10-13 07:29:29 +00:00
retval += bus_print_child_footer(bus, child);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (retval);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
/*
* If this device is an ACPI child but no one claimed it, attempt
* to power it off. We'll power it back up when a driver is added.
*
* XXX Disabled for now since many necessary devices (like fdc and
* ATA) don't claim the devices we created for them but still expect
* them to be powered up.
*/
static void
acpi_probe_nomatch(device_t bus, device_t child)
{
Make the no driver stuff an ifdef.
2008-11-02 17:35:15 +00:00
#ifdef ACPI_ENABLE_POWERDOWN_NODRIVER
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
acpi_set_powerstate(child, ACPI_STATE_D3);
Make the no driver stuff an ifdef.
2008-11-02 17:35:15 +00:00
#endif
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
}
/*
* If a new driver has a chance to probe a child, first power it up.
*
* XXX Disabled for now (see acpi_probe_nomatch for details).
*/
static void
acpi_driver_added(device_t dev, driver_t *driver)
{
device_t child, *devlist;
int i, numdevs;
DEVICE_IDENTIFY(driver, dev);
Handle errors from device_get_children.
2008-08-23 16:38:20 +00:00
if (device_get_children(dev, &devlist, &numdevs))
return;
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
for (i = 0; i < numdevs; i++) {
child = devlist[i];
if (device_get_state(child) == DS_NOTPRESENT) {
Make the no driver stuff an ifdef.
2008-11-02 17:35:15 +00:00
#ifdef ACPI_ENABLE_POWERDOWN_NODRIVER
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
acpi_set_powerstate(child, ACPI_STATE_D0);
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
if (device_probe_and_attach(child) != 0)
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
acpi_set_powerstate(child, ACPI_STATE_D3);
Make the no driver stuff an ifdef.
2008-11-02 17:35:15 +00:00
#else
device_probe_and_attach(child);
#endif
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
}
}
free(devlist, M_TEMP);
}
Staticize pnp methods, style fixes. Remove unused variable to unbreak kernel build.
2004-03-31 17:35:28 +00:00
/* Location hint for devctl(8) */
static int
Style fix. Pointed out by: njl
2004-03-31 17:21:14 +00:00
acpi_child_location_str_method(device_t cbdev, device_t child, char *buf,
size_t buflen)
{
struct acpi_device *dinfo = device_get_ivars(child);
Populate the device info string with _PXM (proximity domain) information. This is primarily useful for debugging right now - it'll show up in devinfo. Reviewed by: jhb
2014-09-20 04:31:12 +00:00
char buf2[32];
int pxm;
if (dinfo->ad_handle) {
snprintf(buf, buflen, "handle=%s", acpi_name(dinfo->ad_handle));
if (ACPI_SUCCESS(acpi_GetInteger(dinfo->ad_handle, "_PXM", &pxm))) {
snprintf(buf2, 32, " _PXM=%d", pxm);
strlcat(buf, buf2, buflen);
}
} else {
Don't set the string "unknown" as a device's location_str Return an empty string when the location is unknown instead of the string "unknown". This ensures that all location entries are of the form key=val. Suggested by: imp Approved by: jhb (mentor) MFC after: 1 week Sponsored by: Intel Corporation Differential Revision: https://reviews.freebsd.org/D21326
2019-08-19 17:51:06 +00:00
snprintf(buf, buflen, "");
Populate the device info string with _PXM (proximity domain) information. This is primarily useful for debugging right now - it'll show up in devinfo. Reviewed by: jhb
2014-09-20 04:31:12 +00:00
}
Style fix. Pointed out by: njl
2004-03-31 17:21:14 +00:00
return (0);
}
Staticize pnp methods, style fixes. Remove unused variable to unbreak kernel build.
2004-03-31 17:35:28 +00:00
/* PnP information for devctl(8) */
acpi: Export functions required by upcoming acpi_iicbus driver.
2020-03-09 20:28:45 +00:00
int
acpi_pnpinfo_str(ACPI_HANDLE handle, char *buf, size_t buflen)
Style fix. Pointed out by: njl
2004-03-31 17:21:14 +00:00
{
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
ACPI_DEVICE_INFO *adinfo;
Style fix. Pointed out by: njl
2004-03-31 17:21:14 +00:00
acpi: Export functions required by upcoming acpi_iicbus driver.
2020-03-09 20:28:45 +00:00
if (ACPI_FAILURE(AcpiGetObjectInfo(handle, &adinfo))) {
Clean up our pnpinfo and location strings.
2004-07-13 18:59:49 +00:00
snprintf(buf, buflen, "unknown");
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
return (0);
}
acpi(4): Add _CID to PNP info string While a given ACPI device may have 0-N compatibility IDs, in practice most seem to have 0 or 1. If one is present, emit it as part of the PNP info string associated with a device. This could enable MODULE_PNP_INFO-based automatic kldload for ACPI drivers associated with a given _CID (but without a good _HID or _UID identifier). Reviewed by: imp, jhb Differential Revision: https://reviews.freebsd.org/D22846
2019-12-18 06:22:28 +00:00
snprintf(buf, buflen, "_HID=%s _UID=%lu _CID=%s",
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
(adinfo->Valid & ACPI_VALID_HID) ?
adinfo->HardwareId.String : "none",
(adinfo->Valid & ACPI_VALID_UID) ?
acpi(4): Add _CID to PNP info string While a given ACPI device may have 0-N compatibility IDs, in practice most seem to have 0 or 1. If one is present, emit it as part of the PNP info string associated with a device. This could enable MODULE_PNP_INFO-based automatic kldload for ACPI drivers associated with a given _CID (but without a good _HID or _UID identifier). Reviewed by: imp, jhb Differential Revision: https://reviews.freebsd.org/D22846
2019-12-18 06:22:28 +00:00
strtoul(adinfo->UniqueId.String, NULL, 10) : 0UL,
((adinfo->Valid & ACPI_VALID_CID) &&
adinfo->CompatibleIdList.Count > 0) ?
adinfo->CompatibleIdList.Ids[0].String : "none");
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
AcpiOsFree(adinfo);
Style fix. Pointed out by: njl
2004-03-31 17:21:14 +00:00
return (0);
}
acpi: Export functions required by upcoming acpi_iicbus driver.
2020-03-09 20:28:45 +00:00
static int
acpi_child_pnpinfo_str_method(device_t cbdev, device_t child, char *buf,
size_t buflen)
{
struct acpi_device *dinfo = device_get_ivars(child);
return (acpi_pnpinfo_str(dinfo->ad_handle, buf, buflen));
}
acpi: Fix stalled value returned by acpi_get_device() after device deletion Newbus device reference attached to ACPI handle is not cleared when newbus device is deleted with devctl(8) delete command. Fix that with calling of AcpiDetachData() from "child_deleted" bus method like acpi_pci driver does. MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D22902
2020-03-09 20:27:25 +00:00
/*
* Handle device deletion.
*/
static void
acpi_child_deleted(device_t dev, device_t child)
{
struct acpi_device *dinfo = device_get_ivars(child);
if (acpi_get_device(dinfo->ad_handle) == child)
AcpiDetachData(dinfo->ad_handle, acpi_fake_objhandler);
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
* Handle per-device ivars
*/
static int
acpi_read_ivar(device_t dev, device_t child, int index, uintptr_t *result)
{
struct acpi_device *ad;
if ((ad = device_get_ivars(child)) == NULL) {
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(child, "device has no ivars\n");
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (ENOENT);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* ACPI and ISA compatibility ivars */
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
switch(index) {
case ACPI_IVAR_HANDLE:
*(ACPI_HANDLE *)result = ad->ad_handle;
break;
case ACPI_IVAR_PRIVATE:
*(void **)result = ad->ad_private;
break;
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
case ACPI_IVAR_FLAGS:
*(int *)result = ad->ad_flags;
break;
Allow the ACPI subsystem to be disabled with a hint. Avoid fully initialising the ACPI namespace if we are attempting to avoid parts of it. This is a workaround for some systems that still crash the interpreter. Implement the ISA_IVAR_LOGICALID for ISA compatibility. Implement stubs for other PnP ID-related ivars.
2001-09-07 02:57:29 +00:00
case ISA_IVAR_VENDORID:
case ISA_IVAR_SERIAL:
case ISA_IVAR_COMPATID:
*(int *)result = -1;
break;
case ISA_IVAR_LOGICALID:
*(int *)result = acpi_isa_get_logicalid(child);
break;
Add support to read the _CLS entry if it's present. It is used by memory-mapped devices that are normally PCIe drives. Devices can then use the existing pci_get_class, etc. accessors to query this data. The ivar values are different enough from the existing ACPI and ISA values to not conflict. Reviewed by: jhb Obtained from: ABT Systems Ltd Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D8721
2016-12-16 10:40:00 +00:00
case PCI_IVAR_CLASS:
*(uint8_t*)result = (ad->ad_cls_class >> 16) & 0xff;
break;
case PCI_IVAR_SUBCLASS:
*(uint8_t*)result = (ad->ad_cls_class >> 8) & 0xff;
break;
case PCI_IVAR_PROGIF:
*(uint8_t*)result = (ad->ad_cls_class >> 0) & 0xff;
break;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
default:
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (ENOENT);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (0);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
static int
acpi_write_ivar(device_t dev, device_t child, int index, uintptr_t value)
{
struct acpi_device *ad;
if ((ad = device_get_ivars(child)) == NULL) {
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(child, "device has no ivars\n");
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (ENOENT);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
switch(index) {
case ACPI_IVAR_HANDLE:
ad->ad_handle = (ACPI_HANDLE)value;
break;
case ACPI_IVAR_PRIVATE:
ad->ad_private = (void *)value;
break;
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
case ACPI_IVAR_FLAGS:
ad->ad_flags = (int)value;
break;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
default:
don't include redunant \n in panic message
2002-08-12 22:36:18 +00:00
panic("bad ivar write request (%d)", index);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (ENOENT);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (0);
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
* Handle child resource allocation/removal
*/
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
static struct resource_list *
acpi_get_rlist(device_t dev, device_t child)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
struct acpi_device *ad;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
ad = device_get_ivars(child);
return (&ad->ad_rl);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Allow device hints to wire the unit numbers of devices. - An "at" hint now reserves a device name. - A new BUS_HINT_DEVICE_UNIT method is added to the bus interface. When determining the unit number of a device, this method is invoked to let the bus driver specify the unit of a device given a specific devclass. This is the only way a device can be given a name reserved via an "at" hint. - Implement BUS_HINT_DEVICE_UNIT() for the acpi(4) and isa(4) bus drivers. Both of these busses implement this by comparing the resources for a given hint device with the resources enumerated by ACPI/PnPBIOS and wire a unit if the hint resources are a subset of the "real" resources. - Use bus_hinted_children() for adding hinted devices on isa(4) busses now instead of doing it by hand. - Remove the unit kludging from sio(4) as it is no longer necessary. Prodding from: peter, imp OK'd by: marcel MFC after: 1 month
2008-11-18 21:01:54 +00:00
static int
acpi_match_resource_hint(device_t dev, int type, long value)
{
struct acpi_device *ad = device_get_ivars(dev);
struct resource_list *rl = &ad->ad_rl;
struct resource_list_entry *rle;
STAILQ_FOREACH(rle, rl, link) {
if (rle->type != type)
continue;
if (rle->start <= value && rle->end >= value)
return (1);
}
return (0);
}
/*
* Wire device unit numbers based on resource matches in hints.
*/
static void
acpi_hint_device_unit(device_t acdev, device_t child, const char *name,
int *unitp)
{
const char *s;
long value;
int line, matches, unit;
/*
* Iterate over all the hints for the devices with the specified
* name to see if one's resources are a subset of this device.
*/
line = 0;
Simplify the code a bit. Replace clumsy for(;;) { if (foo) break; ...} with simpler while (!foo) { ... }. Sponsored by: Netflix Differential Revision: https://reviews.freebsd.org/D13546
2017-12-20 19:14:16 +00:00
while (resource_find_dev(&line, name, &unit, "at", NULL) == 0) {
Allow device hints to wire the unit numbers of devices. - An "at" hint now reserves a device name. - A new BUS_HINT_DEVICE_UNIT method is added to the bus interface. When determining the unit number of a device, this method is invoked to let the bus driver specify the unit of a device given a specific devclass. This is the only way a device can be given a name reserved via an "at" hint. - Implement BUS_HINT_DEVICE_UNIT() for the acpi(4) and isa(4) bus drivers. Both of these busses implement this by comparing the resources for a given hint device with the resources enumerated by ACPI/PnPBIOS and wire a unit if the hint resources are a subset of the "real" resources. - Use bus_hinted_children() for adding hinted devices on isa(4) busses now instead of doing it by hand. - Remove the unit kludging from sio(4) as it is no longer necessary. Prodding from: peter, imp OK'd by: marcel MFC after: 1 month
2008-11-18 21:01:54 +00:00
/* Must have an "at" for acpi or isa. */
resource_string_value(name, unit, "at", &s);
if (!(strcmp(s, "acpi0") == 0 || strcmp(s, "acpi") == 0 ||
strcmp(s, "isa0") == 0 || strcmp(s, "isa") == 0))
continue;
/*
Tweak the way that the ACPI and ISA bus drivers match hint devices to BIOS-enumerated devices: - Assume a device is a match if the memory and I/O ports match even if the IRQ or DRQ is wrong or missing. Some BIOSes don't include an IRQ for the atrtc device for example. - Add a hack to better match floppy controller devices. Many BIOSes do not include the starting port of the floppy controller listed in the hints (0x3f0) in the resources for the device. So far, however, all the BIOS variations encountered do include the 'port + 2' resource (0x3f2), so adjust the matching for "fdc" devices to look for 'port + 2'. Reviewed by: imp MFC after: 3 days
2009-08-24 21:51:46 +00:00
* Check for matching resources. We must have at least one match.
* Since I/O and memory resources cannot be shared, if we get a
* match on either of those, ignore any mismatches in IRQs or DRQs.
Allow device hints to wire the unit numbers of devices. - An "at" hint now reserves a device name. - A new BUS_HINT_DEVICE_UNIT method is added to the bus interface. When determining the unit number of a device, this method is invoked to let the bus driver specify the unit of a device given a specific devclass. This is the only way a device can be given a name reserved via an "at" hint. - Implement BUS_HINT_DEVICE_UNIT() for the acpi(4) and isa(4) bus drivers. Both of these busses implement this by comparing the resources for a given hint device with the resources enumerated by ACPI/PnPBIOS and wire a unit if the hint resources are a subset of the "real" resources. - Use bus_hinted_children() for adding hinted devices on isa(4) busses now instead of doing it by hand. - Remove the unit kludging from sio(4) as it is no longer necessary. Prodding from: peter, imp OK'd by: marcel MFC after: 1 month
2008-11-18 21:01:54 +00:00
*
* XXX: We may want to revisit this to be more lenient and wire
* as long as it gets one match.
*/
matches = 0;
if (resource_long_value(name, unit, "port", &value) == 0) {
Tweak the way that the ACPI and ISA bus drivers match hint devices to BIOS-enumerated devices: - Assume a device is a match if the memory and I/O ports match even if the IRQ or DRQ is wrong or missing. Some BIOSes don't include an IRQ for the atrtc device for example. - Add a hack to better match floppy controller devices. Many BIOSes do not include the starting port of the floppy controller listed in the hints (0x3f0) in the resources for the device. So far, however, all the BIOS variations encountered do include the 'port + 2' resource (0x3f2), so adjust the matching for "fdc" devices to look for 'port + 2'. Reviewed by: imp MFC after: 3 days
2009-08-24 21:51:46 +00:00
/*
* Floppy drive controllers are notorious for having a
* wide variety of resources not all of which include the
* first port that is specified by the hint (typically
* 0x3f0) (see the comment above fdc_isa_alloc_resources()
* in fdc_isa.c). However, they do all seem to include
* port + 2 (e.g. 0x3f2) so for a floppy device, look for
* 'value + 2' in the port resources instead of the hint
* value.
*/
if (strcmp(name, "fdc") == 0)
value += 2;
Allow device hints to wire the unit numbers of devices. - An "at" hint now reserves a device name. - A new BUS_HINT_DEVICE_UNIT method is added to the bus interface. When determining the unit number of a device, this method is invoked to let the bus driver specify the unit of a device given a specific devclass. This is the only way a device can be given a name reserved via an "at" hint. - Implement BUS_HINT_DEVICE_UNIT() for the acpi(4) and isa(4) bus drivers. Both of these busses implement this by comparing the resources for a given hint device with the resources enumerated by ACPI/PnPBIOS and wire a unit if the hint resources are a subset of the "real" resources. - Use bus_hinted_children() for adding hinted devices on isa(4) busses now instead of doing it by hand. - Remove the unit kludging from sio(4) as it is no longer necessary. Prodding from: peter, imp OK'd by: marcel MFC after: 1 month
2008-11-18 21:01:54 +00:00
if (acpi_match_resource_hint(child, SYS_RES_IOPORT, value))
matches++;
else
continue;
}
if (resource_long_value(name, unit, "maddr", &value) == 0) {
if (acpi_match_resource_hint(child, SYS_RES_MEMORY, value))
matches++;
else
continue;
}
Tweak the way that the ACPI and ISA bus drivers match hint devices to BIOS-enumerated devices: - Assume a device is a match if the memory and I/O ports match even if the IRQ or DRQ is wrong or missing. Some BIOSes don't include an IRQ for the atrtc device for example. - Add a hack to better match floppy controller devices. Many BIOSes do not include the starting port of the floppy controller listed in the hints (0x3f0) in the resources for the device. So far, however, all the BIOS variations encountered do include the 'port + 2' resource (0x3f2), so adjust the matching for "fdc" devices to look for 'port + 2'. Reviewed by: imp MFC after: 3 days
2009-08-24 21:51:46 +00:00
if (matches > 0)
goto matched;
Allow device hints to wire the unit numbers of devices. - An "at" hint now reserves a device name. - A new BUS_HINT_DEVICE_UNIT method is added to the bus interface. When determining the unit number of a device, this method is invoked to let the bus driver specify the unit of a device given a specific devclass. This is the only way a device can be given a name reserved via an "at" hint. - Implement BUS_HINT_DEVICE_UNIT() for the acpi(4) and isa(4) bus drivers. Both of these busses implement this by comparing the resources for a given hint device with the resources enumerated by ACPI/PnPBIOS and wire a unit if the hint resources are a subset of the "real" resources. - Use bus_hinted_children() for adding hinted devices on isa(4) busses now instead of doing it by hand. - Remove the unit kludging from sio(4) as it is no longer necessary. Prodding from: peter, imp OK'd by: marcel MFC after: 1 month
2008-11-18 21:01:54 +00:00
if (resource_long_value(name, unit, "irq", &value) == 0) {
if (acpi_match_resource_hint(child, SYS_RES_IRQ, value))
matches++;
else
continue;
}
if (resource_long_value(name, unit, "drq", &value) == 0) {
if (acpi_match_resource_hint(child, SYS_RES_DRQ, value))
matches++;
else
continue;
}
Tweak the way that the ACPI and ISA bus drivers match hint devices to BIOS-enumerated devices: - Assume a device is a match if the memory and I/O ports match even if the IRQ or DRQ is wrong or missing. Some BIOSes don't include an IRQ for the atrtc device for example. - Add a hack to better match floppy controller devices. Many BIOSes do not include the starting port of the floppy controller listed in the hints (0x3f0) in the resources for the device. So far, however, all the BIOS variations encountered do include the 'port + 2' resource (0x3f2), so adjust the matching for "fdc" devices to look for 'port + 2'. Reviewed by: imp MFC after: 3 days
2009-08-24 21:51:46 +00:00
matched:
Allow device hints to wire the unit numbers of devices. - An "at" hint now reserves a device name. - A new BUS_HINT_DEVICE_UNIT method is added to the bus interface. When determining the unit number of a device, this method is invoked to let the bus driver specify the unit of a device given a specific devclass. This is the only way a device can be given a name reserved via an "at" hint. - Implement BUS_HINT_DEVICE_UNIT() for the acpi(4) and isa(4) bus drivers. Both of these busses implement this by comparing the resources for a given hint device with the resources enumerated by ACPI/PnPBIOS and wire a unit if the hint resources are a subset of the "real" resources. - Use bus_hinted_children() for adding hinted devices on isa(4) busses now instead of doing it by hand. - Remove the unit kludging from sio(4) as it is no longer necessary. Prodding from: peter, imp OK'd by: marcel MFC after: 1 month
2008-11-18 21:01:54 +00:00
if (matches > 0) {
/* We have a winner! */
*unitp = unit;
break;
}
}
}
Add a bus method to fetch the VM domain for the given device/bus. * Add a bus_if.m method - get_domain() - returning the VM domain or ENOENT if the device isn't in a VM domain; * Add bus methods to print out the domain of the device if appropriate; * Add code in srat.c to save the PXM -> VM domain mapping that's done and expose a function to translate VM domain -> PXM; * Add ACPI and ACPI PCI methods to check if the bus has a _PXM attribute and if so map it to the VM domain; * (.. yes, this works recursively.) * Have the pci bus glue print out the device VM domain if present. Note: this is just the plumbing to start enumerating information - it doesn't at all modify behaviour. Differential Revision: D906 Reviewed by: jhb Sponsored by: Norse Corp
2014-10-09 05:33:25 +00:00
/*
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
* Fetch the NUMA domain for a device by mapping the value returned by
* _PXM to a NUMA domain. If the device does not have a _PXM method,
* -2 is returned. If any other error occurs, -1 is returned.
Add a bus method to fetch the VM domain for the given device/bus. * Add a bus_if.m method - get_domain() - returning the VM domain or ENOENT if the device isn't in a VM domain; * Add bus methods to print out the domain of the device if appropriate; * Add code in srat.c to save the PXM -> VM domain mapping that's done and expose a function to translate VM domain -> PXM; * Add ACPI and ACPI PCI methods to check if the bus has a _PXM attribute and if so map it to the VM domain; * (.. yes, this works recursively.) * Have the pci bus glue print out the device VM domain if present. Note: this is just the plumbing to start enumerating information - it doesn't at all modify behaviour. Differential Revision: D906 Reviewed by: jhb Sponsored by: Norse Corp
2014-10-09 05:33:25 +00:00
*/
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
static int
acpi_parse_pxm(device_t dev)
Add a bus method to fetch the VM domain for the given device/bus. * Add a bus_if.m method - get_domain() - returning the VM domain or ENOENT if the device isn't in a VM domain; * Add bus methods to print out the domain of the device if appropriate; * Add code in srat.c to save the PXM -> VM domain mapping that's done and expose a function to translate VM domain -> PXM; * Add ACPI and ACPI PCI methods to check if the bus has a _PXM attribute and if so map it to the VM domain; * (.. yes, this works recursively.) * Have the pci bus glue print out the device VM domain if present. Note: this is just the plumbing to start enumerating information - it doesn't at all modify behaviour. Differential Revision: D906 Reviewed by: jhb Sponsored by: Norse Corp
2014-10-09 05:33:25 +00:00
{
Move VM_NUMA_ALLOC and DEVICE_NUMA under the single global config option NUMA. Sponsored by: Netflix, Dell/EMC Isilon Discussed with: jhb
2018-01-14 03:36:03 +00:00
#ifdef NUMA
Fix the NUMA build for non-x86 platforms. acpi_map_pxm_to_vm_domainid() is currently implemented only on x86. MFC after: 1 week
2018-06-04 14:56:02 +00:00
#if defined(__i386__) || defined(__amd64__)
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
ACPI_HANDLE handle;
ACPI_STATUS status;
int pxm;
handle = acpi_get_handle(dev);
if (handle == NULL)
return (-2);
status = acpi_GetInteger(handle, "_PXM", &pxm);
if (ACPI_SUCCESS(status))
return (acpi_map_pxm_to_vm_domainid(pxm));
if (status == AE_NOT_FOUND)
return (-2);
Fix the NUMA build for non-x86 platforms. acpi_map_pxm_to_vm_domainid() is currently implemented only on x86. MFC after: 1 week
2018-06-04 14:56:02 +00:00
#endif
Add a bus method to fetch the VM domain for the given device/bus. * Add a bus_if.m method - get_domain() - returning the VM domain or ENOENT if the device isn't in a VM domain; * Add bus methods to print out the domain of the device if appropriate; * Add code in srat.c to save the PXM -> VM domain mapping that's done and expose a function to translate VM domain -> PXM; * Add ACPI and ACPI PCI methods to check if the bus has a _PXM attribute and if so map it to the VM domain; * (.. yes, this works recursively.) * Have the pci bus glue print out the device VM domain if present. Note: this is just the plumbing to start enumerating information - it doesn't at all modify behaviour. Differential Revision: D906 Reviewed by: jhb Sponsored by: Norse Corp
2014-10-09 05:33:25 +00:00
#endif
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
return (-1);
}
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Reviewed by: wblock (manpage) Differential Revision: https://reviews.freebsd.org/D5519
2016-05-02 18:00:38 +00:00
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
int
acpi_get_cpus(device_t dev, device_t child, enum cpu_sets op, size_t setsize,
cpuset_t *cpuset)
{
int d, error;
d = acpi_parse_pxm(child);
if (d < 0)
return (bus_generic_get_cpus(dev, child, op, setsize, cpuset));
switch (op) {
case LOCAL_CPUS:
if (setsize != sizeof(cpuset_t))
return (EINVAL);
*cpuset = cpuset_domain[d];
return (0);
case INTR_CPUS:
error = bus_generic_get_cpus(dev, child, op, setsize, cpuset);
if (error != 0)
return (error);
if (setsize != sizeof(cpuset_t))
return (EINVAL);
CPU_AND(cpuset, &cpuset_domain[d]);
return (0);
default:
return (bus_generic_get_cpus(dev, child, op, setsize, cpuset));
}
Refactor out the _PXM -> VM domain lookup done in ACPI, in preparation for its use in upcoming code. This is inspired by something in jhb's NUMA IRQ allocation patchset. However, the tricky bit here is that the PXM lookup for a node may fail, requiring a lookup on the parent node. So if it doesn't exist, don't fail - just go up to the parent. Only error out of the lookup is the ACPI lookup returns an error. Sponsored by: Norse Corp, Inc.
2015-04-19 17:15:55 +00:00
}
/*
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
* Fetch the NUMA domain for the given device 'dev'.
Refactor out the _PXM -> VM domain lookup done in ACPI, in preparation for its use in upcoming code. This is inspired by something in jhb's NUMA IRQ allocation patchset. However, the tricky bit here is that the PXM lookup for a node may fail, requiring a lookup on the parent node. So if it doesn't exist, don't fail - just go up to the parent. Only error out of the lookup is the ACPI lookup returns an error. Sponsored by: Norse Corp, Inc.
2015-04-19 17:15:55 +00:00
*
* If a device has a _PXM method, map that to a NUMA domain.
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
* Otherwise, pass the request up to the parent.
* If there's no matching domain or the domain cannot be
* determined, return ENOENT.
Refactor out the _PXM -> VM domain lookup done in ACPI, in preparation for its use in upcoming code. This is inspired by something in jhb's NUMA IRQ allocation patchset. However, the tricky bit here is that the PXM lookup for a node may fail, requiring a lookup on the parent node. So if it doesn't exist, don't fail - just go up to the parent. Only error out of the lookup is the ACPI lookup returns an error. Sponsored by: Norse Corp, Inc.
2015-04-19 17:15:55 +00:00
*/
int
acpi_get_domain(device_t dev, device_t child, int *domain)
{
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
int d;
Refactor out the _PXM -> VM domain lookup done in ACPI, in preparation for its use in upcoming code. This is inspired by something in jhb's NUMA IRQ allocation patchset. However, the tricky bit here is that the PXM lookup for a node may fail, requiring a lookup on the parent node. So if it doesn't exist, don't fail - just go up to the parent. Only error out of the lookup is the ACPI lookup returns an error. Sponsored by: Norse Corp, Inc.
2015-04-19 17:15:55 +00:00
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
d = acpi_parse_pxm(child);
if (d >= 0) {
*domain = d;
Revert bus_get_cpus() for now. I really thought I had run this through the tinderbox before committing, but many places need <sys/types.h> -> <sys/param.h> for <sys/bus.h> now.
2016-05-03 01:17:40 +00:00
return (0);
Add a new bus method to fetch device-specific CPU sets. bus_get_cpus() returns a specified set of CPUs for a device. It accepts an enum for the second parameter that indicates the type of cpuset to request. Currently two valus are supported: - LOCAL_CPUS (on x86 this returns all the CPUs in the package closest to the device when DEVICE_NUMA is enabled) - INTR_CPUS (like LOCAL_CPUS but only returns 1 SMT thread for each core) For systems that do not support NUMA (or if it is not enabled in the kernel config), LOCAL_CPUS fails with EINVAL. INTR_CPUS is mapped to 'all_cpus' by default. The idea is that INTR_CPUS should always return a valid set. Device drivers which want to use per-CPU interrupts should start using INTR_CPUS instead of simply assigning interrupts to all available CPUs. In the future we may wish to add tunables to control the policy of INTR_CPUS (e.g. should it be local-only or global, should it ignore SMT threads or not). The x86 nexus driver exposes the internal set of interrupt CPUs from the the x86 interrupt code via INTR_CPUS. The ACPI bus driver and PCI bridge drivers use _PXM to return a suitable LOCAL_CPUS set when _PXM exists and DEVICE_NUMA is enabled. They also and the global INTR_CPUS set from the nexus driver with the per-domain set from _PXM to generate a local INTR_CPUS set for child devices. Compared to the r298933, this version uses 'struct _cpuset' in <sys/bus.h> instead of 'cpuset_t' to avoid requiring <sys/param.h> (<sys/_cpuset.h> still requires <sys/param.h> for MAXCPU even though <sys/_bitset.h> does not after recent changes).
2016-05-09 20:50:21 +00:00
}
if (d == -1)
return (ENOENT);
Refactor out the _PXM -> VM domain lookup done in ACPI, in preparation for its use in upcoming code. This is inspired by something in jhb's NUMA IRQ allocation patchset. However, the tricky bit here is that the PXM lookup for a node may fail, requiring a lookup on the parent node. So if it doesn't exist, don't fail - just go up to the parent. Only error out of the lookup is the ACPI lookup returns an error. Sponsored by: Norse Corp, Inc.
2015-04-19 17:15:55 +00:00
Add a bus method to fetch the VM domain for the given device/bus. * Add a bus_if.m method - get_domain() - returning the VM domain or ENOENT if the device isn't in a VM domain; * Add bus methods to print out the domain of the device if appropriate; * Add code in srat.c to save the PXM -> VM domain mapping that's done and expose a function to translate VM domain -> PXM; * Add ACPI and ACPI PCI methods to check if the bus has a _PXM attribute and if so map it to the VM domain; * (.. yes, this works recursively.) * Have the pci bus glue print out the device VM domain if present. Note: this is just the plumbing to start enumerating information - it doesn't at all modify behaviour. Differential Revision: D906 Reviewed by: jhb Sponsored by: Norse Corp
2014-10-09 05:33:25 +00:00
/* No _PXM node; go up a level */
return (bus_generic_get_domain(dev, child, domain));
}
Rework sysresource management. Instead of having each sysresource object hold its own values, pass them up to the parent (acpi0) and merge/uniq them on the way. After the namespace evaluation, acpi will reserve these resources and manage them via rman before bus_generic_probe() and bus_generic_attach(). This is necessary because some systems specify conflicting resources in separate sysresource objects. It's also cleaner in that the interface between sysresource and acpi is now merely the parent's resource list. This code handles the following cases: 1. Unique resource: add it to the parent via bus_set_resource(). 2. New wholly contained in old: discard new. 3. New tail overlaps old head: grow old head downward. AND/OR 4. New head overlaps old tail: grow old tail upward. Tested by: Pawel Worach <sajd_at_telia.com> Tested by: Radek Kozlowski <radek_at_raadradd.com> MFC after: 5 days
2004-08-23 16:28:42 +00:00
/*
* Pre-allocate/manage all memory and IO resources. Since rman can't handle
* duplicates, we merge any in the sysresource attach routine.
*/
static int
acpi_sysres_alloc(device_t dev)
{
struct resource *res;
struct resource_list *rl;
struct resource_list_entry *rle;
struct rman *rm;
Don't attach special devices in the order they appear in the AML tree. If the embedded controller exists before the sysresource devices, for example, it will be attached first. Instead, let the normal device order function work as we first desired. [1] There still remained a problem where we couldn't allocate resources in acpi0 that were passed up by the sysresource pseudo-devices. These devices had to probe/attach first to give their resources to acpi, then acpi would allocate them before probing/attaching other devices. To work around this, we attach them from acpi_sysres_alloc(). A better approach would be to implement multi-pass probe/attach in newbus but that's a much bigger task. Suggested by: jhb [1] Hardware from: Centaur Technologies MFC after: 1 week
2006-05-07 03:28:10 +00:00
device_t *children;
int child_count, i;
/*
* Probe/attach any sysresource devices. This would be unnecessary if we
* had multi-pass probe/attach.
*/
if (device_get_children(dev, &children, &child_count) != 0)
return (ENXIO);
for (i = 0; i < child_count; i++) {
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
if (ACPI_ID_PROBE(dev, children[i], sysres_ids, NULL) <= 0)
Don't attach special devices in the order they appear in the AML tree. If the embedded controller exists before the sysresource devices, for example, it will be attached first. Instead, let the normal device order function work as we first desired. [1] There still remained a problem where we couldn't allocate resources in acpi0 that were passed up by the sysresource pseudo-devices. These devices had to probe/attach first to give their resources to acpi, then acpi would allocate them before probing/attaching other devices. To work around this, we attach them from acpi_sysres_alloc(). A better approach would be to implement multi-pass probe/attach in newbus but that's a much bigger task. Suggested by: jhb [1] Hardware from: Centaur Technologies MFC after: 1 week
2006-05-07 03:28:10 +00:00
device_probe_and_attach(children[i]);
}
free(children, M_TEMP);
Rework sysresource management. Instead of having each sysresource object hold its own values, pass them up to the parent (acpi0) and merge/uniq them on the way. After the namespace evaluation, acpi will reserve these resources and manage them via rman before bus_generic_probe() and bus_generic_attach(). This is necessary because some systems specify conflicting resources in separate sysresource objects. It's also cleaner in that the interface between sysresource and acpi is now merely the parent's resource list. This code handles the following cases: 1. Unique resource: add it to the parent via bus_set_resource(). 2. New wholly contained in old: discard new. 3. New tail overlaps old head: grow old head downward. AND/OR 4. New head overlaps old tail: grow old tail upward. Tested by: Pawel Worach <sajd_at_telia.com> Tested by: Radek Kozlowski <radek_at_raadradd.com> MFC after: 5 days
2004-08-23 16:28:42 +00:00
rl = BUS_GET_RESOURCE_LIST(device_get_parent(dev), dev);
s/SLIST/STAILQ/ /imp/a\ pointy hat .
2005-03-18 11:57:44 +00:00
STAILQ_FOREACH(rle, rl, link) {
Rework sysresource management. Instead of having each sysresource object hold its own values, pass them up to the parent (acpi0) and merge/uniq them on the way. After the namespace evaluation, acpi will reserve these resources and manage them via rman before bus_generic_probe() and bus_generic_attach(). This is necessary because some systems specify conflicting resources in separate sysresource objects. It's also cleaner in that the interface between sysresource and acpi is now merely the parent's resource list. This code handles the following cases: 1. Unique resource: add it to the parent via bus_set_resource(). 2. New wholly contained in old: discard new. 3. New tail overlaps old head: grow old head downward. AND/OR 4. New head overlaps old tail: grow old tail upward. Tested by: Pawel Worach <sajd_at_telia.com> Tested by: Radek Kozlowski <radek_at_raadradd.com> MFC after: 5 days
2004-08-23 16:28:42 +00:00
if (rle->res != NULL) {
Use uintmax_t (typedef'd to rman_res_t type) for rman ranges. On some architectures, u_long isn't large enough for resource definitions. Particularly, powerpc and arm allow 36-bit (or larger) physical addresses, but type `long' is only 32-bit. This extends rman's resources to uintmax_t. With this change, any resource can feasibly be placed anywhere in physical memory (within the constraints of the driver). Why uintmax_t and not something machine dependent, or uint64_t? Though it's possible for uintmax_t to grow, it's highly unlikely it will become 128-bit on 32-bit architectures. 64-bit architectures should have plenty of RAM to absorb the increase on resource sizes if and when this occurs, and the number of resources on memory-constrained systems should be sufficiently small as to not pose a drastic overhead. That being said, uintmax_t was chosen for source clarity. If it's specified as uint64_t, all printf()-like calls would either need casts to uintmax_t, or be littered with PRI*64 macros. Casts to uintmax_t aren't horrible, but it would also bake into the API for resource_list_print_type() either a hidden assumption that entries get cast to uintmax_t for printing, or these calls would need the PRI*64 macros. Since source code is meant to be read more often than written, I chose the clearest path of simply using uintmax_t. Tested on a PowerPC p5020-based board, which places all device resources in 0xfxxxxxxxx, and has 8GB RAM. Regression tested on qemu-system-i386 Regression tested on qemu-system-mips (malta profile) Tested PAE and devinfo on virtualbox (live CD) Special thanks to bz for his testing on ARM. Reviewed By: bz, jhb (previous) Relnotes: Yes Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D4544
2016-03-18 01:28:41 +00:00
device_printf(dev, "duplicate resource for %jx\n", rle->start);
Rework sysresource management. Instead of having each sysresource object hold its own values, pass them up to the parent (acpi0) and merge/uniq them on the way. After the namespace evaluation, acpi will reserve these resources and manage them via rman before bus_generic_probe() and bus_generic_attach(). This is necessary because some systems specify conflicting resources in separate sysresource objects. It's also cleaner in that the interface between sysresource and acpi is now merely the parent's resource list. This code handles the following cases: 1. Unique resource: add it to the parent via bus_set_resource(). 2. New wholly contained in old: discard new. 3. New tail overlaps old head: grow old head downward. AND/OR 4. New head overlaps old tail: grow old tail upward. Tested by: Pawel Worach <sajd_at_telia.com> Tested by: Radek Kozlowski <radek_at_raadradd.com> MFC after: 5 days
2004-08-23 16:28:42 +00:00
continue;
}
/* Only memory and IO resources are valid here. */
switch (rle->type) {
case SYS_RES_IOPORT:
rm = &acpi_rman_io;
break;
case SYS_RES_MEMORY:
rm = &acpi_rman_mem;
break;
default:
continue;
}
/* Pre-allocate resource and add to our rman pool. */
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), dev, rle->type,
&rle->rid, rle->start, rle->start + rle->count - 1, rle->count, 0);
if (res != NULL) {
rman_manage_region(rm, rman_get_start(res), rman_get_end(res));
rle->res = res;
Move the message complaining about failed system resource allocations under bootverbose. Every example I've seen to date has been due to an ACPI system resource device reserving a range that overlaps with system memory (which ram0 attempts to reserve) or a local or I/O APIC (which apic0 attempts to reserve). These are always harmless but look scary to users. MFC after: 1 week
2015-04-06 17:39:36 +00:00
} else if (bootverbose)
Use uintmax_t (typedef'd to rman_res_t type) for rman ranges. On some architectures, u_long isn't large enough for resource definitions. Particularly, powerpc and arm allow 36-bit (or larger) physical addresses, but type `long' is only 32-bit. This extends rman's resources to uintmax_t. With this change, any resource can feasibly be placed anywhere in physical memory (within the constraints of the driver). Why uintmax_t and not something machine dependent, or uint64_t? Though it's possible for uintmax_t to grow, it's highly unlikely it will become 128-bit on 32-bit architectures. 64-bit architectures should have plenty of RAM to absorb the increase on resource sizes if and when this occurs, and the number of resources on memory-constrained systems should be sufficiently small as to not pose a drastic overhead. That being said, uintmax_t was chosen for source clarity. If it's specified as uint64_t, all printf()-like calls would either need casts to uintmax_t, or be littered with PRI*64 macros. Casts to uintmax_t aren't horrible, but it would also bake into the API for resource_list_print_type() either a hidden assumption that entries get cast to uintmax_t for printing, or these calls would need the PRI*64 macros. Since source code is meant to be read more often than written, I chose the clearest path of simply using uintmax_t. Tested on a PowerPC p5020-based board, which places all device resources in 0xfxxxxxxxx, and has 8GB RAM. Regression tested on qemu-system-i386 Regression tested on qemu-system-mips (malta profile) Tested PAE and devinfo on virtualbox (live CD) Special thanks to bz for his testing on ARM. Reviewed By: bz, jhb (previous) Relnotes: Yes Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D4544
2016-03-18 01:28:41 +00:00
device_printf(dev, "reservation of %jx, %jx (%d) failed\n",
Rework sysresource management. Instead of having each sysresource object hold its own values, pass them up to the parent (acpi0) and merge/uniq them on the way. After the namespace evaluation, acpi will reserve these resources and manage them via rman before bus_generic_probe() and bus_generic_attach(). This is necessary because some systems specify conflicting resources in separate sysresource objects. It's also cleaner in that the interface between sysresource and acpi is now merely the parent's resource list. This code handles the following cases: 1. Unique resource: add it to the parent via bus_set_resource(). 2. New wholly contained in old: discard new. 3. New tail overlaps old head: grow old head downward. AND/OR 4. New head overlaps old tail: grow old tail upward. Tested by: Pawel Worach <sajd_at_telia.com> Tested by: Radek Kozlowski <radek_at_raadradd.com> MFC after: 5 days
2004-08-23 16:28:42 +00:00
rle->start, rle->count, rle->type);
}
return (0);
}
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
/*
* Reserve declared resources for devices found during attach once system
* resources have been allocated.
*/
static void
acpi_reserve_resources(device_t dev)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
struct resource_list_entry *rle;
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
struct resource_list *rl;
struct acpi_device *ad;
struct acpi_softc *sc;
device_t *children;
int child_count, i;
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
sc = device_get_softc(dev);
if (device_get_children(dev, &children, &child_count) != 0)
return;
for (i = 0; i < child_count; i++) {
ad = device_get_ivars(children[i]);
rl = &ad->ad_rl;
Change acpi's handling of suballocating system resources to be a little simpler. It now can just use rman_is_region_manager() during acpi_release_resource() to see if the the resource is suballocated from a system resource. Also, the driver no longer needs MD knowledge about how to setup bus space tags and handles when doing a suballocation, but can simply rely on bus_activate_resource() in the parent setting all that up.
2007-03-21 15:39:11 +00:00
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
/* Don't reserve system resources. */
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
if (ACPI_ID_PROBE(dev, children[i], sysres_ids, NULL) <= 0)
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
continue;
STAILQ_FOREACH(rle, rl, link) {
/*
* Don't reserve IRQ resources. There are many sticky things
* to get right otherwise (e.g. IRQs for psm, atkbd, and HPET
* when using legacy routing).
*/
if (rle->type == SYS_RES_IRQ)
continue;
Don't try to reserve a resource that is already allocated. If the ECDT table is present, then the acpi_ec(4) driver will allocate its resources from nexus0 before the acpi0 device reserves resources for child devices. Reviewed by: jkim
2010-12-23 18:50:14 +00:00
/*
* Don't reserve the resource if it is already allocated.
* The acpi_ec(4) driver can allocate its resources early
* if ECDT is present.
*/
if (rle->res != NULL)
continue;
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
/*
* Try to reserve the resource from our parent. If this
* fails because the resource is a system resource, just
* let it be. The resource range is already reserved so
* that other devices will not use it. If the driver
* needs to allocate the resource, then
* acpi_alloc_resource() will sub-alloc from the system
* resource.
*/
resource_list_reserve(rl, dev, children[i], rle->type, &rle->rid,
rle->start, rle->end, rle->count, 0);
}
Change acpi's handling of suballocating system resources to be a little simpler. It now can just use rman_is_region_manager() during acpi_release_resource() to see if the the resource is suballocated from a system resource. Also, the driver no longer needs MD knowledge about how to setup bus space tags and handles when doing a suballocation, but can simply rely on bus_activate_resource() in the parent setting all that up.
2007-03-21 15:39:11 +00:00
}
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
free(children, M_TEMP);
sc->acpi_resources_reserved = 1;
}
static int
acpi_set_resource(device_t dev, device_t child, int type, int rid,
Convert rman to use rman_res_t instead of u_long Summary: Migrate to using the semi-opaque type rman_res_t to specify rman resources. For now, this is still compatible with u_long. This is step one in migrating rman to use uintmax_t for resources instead of u_long. Going forward, this could feasibly be used to specify architecture-specific definitions of resource ranges, rather than baking a specific integer type into the API. This change has been broken out to facilitate MFC'ing drivers back to 10 without breaking ABI. Reviewed By: jhb Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D5075
2016-01-27 02:23:54 +00:00
rman_res_t start, rman_res_t count)
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
{
struct acpi_softc *sc = device_get_softc(dev);
struct acpi_device *ad = device_get_ivars(child);
struct resource_list *rl = &ad->ad_rl;
Some BIOSes incorrectly use standard memory resource ranges to list the memory ranges that they decode for downstream devices rather than creating ResourceProducer range resource entries. The result is that we allocate the full range to the PCI root bridge device causing allocations in child devices to all fail. As a workaround, ignore any standard memory resources on a PCI root bridge device. It is normal for a PCI root bridge to allocate an I/O resource for the I/O ports used for PCI config access, but I have not seen any PCI root bridges that legitimately allocate a memory resource. Reviewed by: jkim MFC after: 1 week
2014-01-28 20:53:33 +00:00
ACPI_DEVICE_INFO *devinfo;
Convert rman to use rman_res_t instead of u_long Summary: Migrate to using the semi-opaque type rman_res_t to specify rman resources. For now, this is still compatible with u_long. This is step one in migrating rman to use uintmax_t for resources instead of u_long. Going forward, this could feasibly be used to specify architecture-specific definitions of resource ranges, rather than baking a specific integer type into the API. This change has been broken out to facilitate MFC'ing drivers back to 10 without breaking ABI. Reviewed By: jhb Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D5075
2016-01-27 02:23:54 +00:00
rman_res_t end;
acpica, pci_host_generic_acpi: redo pci_host_generic_acpi.c This is a major update for pci_host_generic_acpi.c, the current implementation has some gaps that are better fixed up in one go. The changes are to: * Follow x86 method of not adding PCI resources to PCI host bridge in ACPI code. This has been moved to pci_host_generic_acpi.c, where we walk thru its resources of the host bridge and add them. * Fixup code in pci_host_generic_acpi.c to read all decoded ranges and update the 'ranges' property. This allows us to share most of the code with generic implementation (and the FDT one). * Parse and setup IO ranges and bus ranges when walking the resources above. Drop most of the changes related to this from acpica code. * Add the ECAM memory area as mem resource 0. Implement the logic to get the ECAM area from MCFG (using bus range which we now decode), or from _CBA (using _BBN/bus range). Drop aarch64 ifdefs from acpica code which did part of this. * Switch resource activation to similar code as FDT implementation, this can be moved into generic implementation in a later pass. * Drop the mechanism of using the 7th bit of bus number as the domain, this is not correct and will work only in very specific cases. Use _SEG as PCI domain and use the bus ranges of the host bridge to provide start bus number. This commit should not make any functional change to dev/acpica/acpi.c for other architectures, almost all the changes there are to revert earlier additions in this file done for aarch64. Reviewed by: andrew Differential Revision: https://reviews.freebsd.org/D17791
2018-11-19 03:16:16 +00:00
int allow;
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
/* Ignore IRQ resources for PCI link devices. */
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
if (type == SYS_RES_IRQ &&
ACPI_ID_PROBE(dev, child, pcilink_ids, NULL) <= 0)
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
return (0);
Some BIOSes incorrectly use standard memory resource ranges to list the memory ranges that they decode for downstream devices rather than creating ResourceProducer range resource entries. The result is that we allocate the full range to the PCI root bridge device causing allocations in child devices to all fail. As a workaround, ignore any standard memory resources on a PCI root bridge device. It is normal for a PCI root bridge to allocate an I/O resource for the I/O ports used for PCI config access, but I have not seen any PCI root bridges that legitimately allocate a memory resource. Reviewed by: jkim MFC after: 1 week
2014-01-28 20:53:33 +00:00
/*
Expand r261243 even further and ignore any I/O port resources assigned to PCI root bridges except for the one known-valid case on x86 where bridges claim the I/O port registers used for PCI config space access. Tested by: Hilko Meyer <hilko.meyer@gmx.de> MFC after: 1 week
2014-06-25 20:30:47 +00:00
* Ignore most resources for PCI root bridges. Some BIOSes
Some BIOSes incorrectly use standard memory resource ranges to list the memory ranges that they decode for downstream devices rather than creating ResourceProducer range resource entries. The result is that we allocate the full range to the PCI root bridge device causing allocations in child devices to all fail. As a workaround, ignore any standard memory resources on a PCI root bridge device. It is normal for a PCI root bridge to allocate an I/O resource for the I/O ports used for PCI config access, but I have not seen any PCI root bridges that legitimately allocate a memory resource. Reviewed by: jkim MFC after: 1 week
2014-01-28 20:53:33 +00:00
* incorrectly enumerate the memory ranges they decode as plain
Expand r261243 even further and ignore any I/O port resources assigned to PCI root bridges except for the one known-valid case on x86 where bridges claim the I/O port registers used for PCI config space access. Tested by: Hilko Meyer <hilko.meyer@gmx.de> MFC after: 1 week
2014-06-25 20:30:47 +00:00
* memory resources instead of as ResourceProducer ranges. Other
* BIOSes incorrectly list system resource entries for I/O ranges
* under the PCI bridge. Do allow the one known-correct case on
* x86 of a PCI bridge claiming the I/O ports used for PCI config
* access.
Some BIOSes incorrectly use standard memory resource ranges to list the memory ranges that they decode for downstream devices rather than creating ResourceProducer range resource entries. The result is that we allocate the full range to the PCI root bridge device causing allocations in child devices to all fail. As a workaround, ignore any standard memory resources on a PCI root bridge device. It is normal for a PCI root bridge to allocate an I/O resource for the I/O ports used for PCI config access, but I have not seen any PCI root bridges that legitimately allocate a memory resource. Reviewed by: jkim MFC after: 1 week
2014-01-28 20:53:33 +00:00
*/
Expand r261243 even further and ignore any I/O port resources assigned to PCI root bridges except for the one known-valid case on x86 where bridges claim the I/O port registers used for PCI config space access. Tested by: Hilko Meyer <hilko.meyer@gmx.de> MFC after: 1 week
2014-06-25 20:30:47 +00:00
if (type == SYS_RES_MEMORY || type == SYS_RES_IOPORT) {
Some BIOSes incorrectly use standard memory resource ranges to list the memory ranges that they decode for downstream devices rather than creating ResourceProducer range resource entries. The result is that we allocate the full range to the PCI root bridge device causing allocations in child devices to all fail. As a workaround, ignore any standard memory resources on a PCI root bridge device. It is normal for a PCI root bridge to allocate an I/O resource for the I/O ports used for PCI config access, but I have not seen any PCI root bridges that legitimately allocate a memory resource. Reviewed by: jkim MFC after: 1 week
2014-01-28 20:53:33 +00:00
if (ACPI_SUCCESS(AcpiGetObjectInfo(ad->ad_handle, &devinfo))) {
if ((devinfo->Flags & ACPI_PCI_ROOT_BRIDGE) != 0) {
acpica, pci_host_generic_acpi: redo pci_host_generic_acpi.c This is a major update for pci_host_generic_acpi.c, the current implementation has some gaps that are better fixed up in one go. The changes are to: * Follow x86 method of not adding PCI resources to PCI host bridge in ACPI code. This has been moved to pci_host_generic_acpi.c, where we walk thru its resources of the host bridge and add them. * Fixup code in pci_host_generic_acpi.c to read all decoded ranges and update the 'ranges' property. This allows us to share most of the code with generic implementation (and the FDT one). * Parse and setup IO ranges and bus ranges when walking the resources above. Drop most of the changes related to this from acpica code. * Add the ECAM memory area as mem resource 0. Implement the logic to get the ECAM area from MCFG (using bus range which we now decode), or from _CBA (using _BBN/bus range). Drop aarch64 ifdefs from acpica code which did part of this. * Switch resource activation to similar code as FDT implementation, this can be moved into generic implementation in a later pass. * Drop the mechanism of using the 7th bit of bus number as the domain, this is not correct and will work only in very specific cases. Use _SEG as PCI domain and use the bus ranges of the host bridge to provide start bus number. This commit should not make any functional change to dev/acpica/acpi.c for other architectures, almost all the changes there are to revert earlier additions in this file done for aarch64. Reviewed by: andrew Differential Revision: https://reviews.freebsd.org/D17791
2018-11-19 03:16:16 +00:00
#if defined(__i386__) || defined(__amd64__)
allow = (type == SYS_RES_IOPORT && start == CONF1_ADDR_PORT);
#else
allow = 0;
#endif
if (!allow) {
Expand r261243 even further and ignore any I/O port resources assigned to PCI root bridges except for the one known-valid case on x86 where bridges claim the I/O port registers used for PCI config space access. Tested by: Hilko Meyer <hilko.meyer@gmx.de> MFC after: 1 week
2014-06-25 20:30:47 +00:00
AcpiOsFree(devinfo);
return (0);
}
Some BIOSes incorrectly use standard memory resource ranges to list the memory ranges that they decode for downstream devices rather than creating ResourceProducer range resource entries. The result is that we allocate the full range to the PCI root bridge device causing allocations in child devices to all fail. As a workaround, ignore any standard memory resources on a PCI root bridge device. It is normal for a PCI root bridge to allocate an I/O resource for the I/O ports used for PCI config access, but I have not seen any PCI root bridges that legitimately allocate a memory resource. Reviewed by: jkim MFC after: 1 week
2014-01-28 20:53:33 +00:00
}
AcpiOsFree(devinfo);
}
}
acpica: rework INTRNG interrupts On arm64 (where INTRNG is enabled), the interrupts have to be mapped with ACPI_BUS_MAP_INTR() before adding them as resources to devices. The earlier code did the mapping before calling acpi_set_resource(), which bypassed code that checked for PCI link interrupts. To fix this, move the call to map interrupts into acpi_set_resource() and that requires additional work to lookup interrupt properties. The changes here are to: * extend acpi_lookup_irq_handler() to lookup an irq in the ACPI resources * create a helper function acpi_map_intr() which uses the updated acpi_lookup_irq_handler() to look up an irq, and then map it with ACPI_BUS_MAP_INTR() * use acpi_map_intr() in acpi_pcib_route_interrupt() to map pci link interrupts. With these changes, we can drop the ifdefs in acpi_resource.c, and we can also drop the call for mapping interrupts in generic_timer.c Reviewed by: andrew Differential Revision: https://reviews.freebsd.org/D17790
2018-11-19 03:02:47 +00:00
#ifdef INTRNG
/* map with default for now */
if (type == SYS_RES_IRQ)
start = (rman_res_t)acpi_map_intr(child, (u_int)start,
acpi_get_handle(child));
#endif
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
/* If the resource is already allocated, fail. */
if (resource_list_busy(rl, type, rid))
return (EBUSY);
/* If the resource is already reserved, release it. */
if (resource_list_reserved(rl, type, rid))
resource_list_unreserve(rl, dev, child, type, rid);
/* Add the resource. */
end = (start + count - 1);
resource_list_add(rl, type, rid, start, end, count);
/* Don't reserve resources until the system resources are allocated. */
if (!sc->acpi_resources_reserved)
return (0);
/* Don't reserve system resources. */
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
if (ACPI_ID_PROBE(dev, child, sysres_ids, NULL) <= 0)
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
return (0);
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
/*
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
* Don't reserve IRQ resources. There are many sticky things to
* get right otherwise (e.g. IRQs for psm, atkbd, and HPET when
* using legacy routing).
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
*/
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
if (type == SYS_RES_IRQ)
return (0);
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
Don't pre-reserve resources for CPU devices when they are set. CPUs can use shared (RF_SHAREABLE) resources for the I/O port used for entering and exiting C states. If this I/O port is included in an ACPI system resource device, then this happens to still work, but if the port wasn't part of a system resource device, only the first CPU could allocate the I/O port and use C states since resource_list_reserve() was always allocating the resource from nexus0 without RF_SHAREABLE. By avoiding the reservation, the flags from the bus_alloc_resource() in the CPU driver (which include RF_SHAREABLE) are honored. PR: 236513 Reported by: stockhausen@collogia.de Sleuthing by: avg Reviewed by: avg MFC after: 2 weeks
2019-04-09 19:22:08 +00:00
/*
* Don't reserve resources for CPU devices. Some of these
* resources need to be allocated as shareable, but reservations
* are always non-shareable.
*/
if (device_get_devclass(child) == devclass_find("cpu"))
return (0);
Change acpi's handling of suballocating system resources to be a little simpler. It now can just use rman_is_region_manager() during acpi_release_resource() to see if the the resource is suballocated from a system resource. Also, the driver no longer needs MD knowledge about how to setup bus space tags and handles when doing a suballocation, but can simply rely on bus_activate_resource() in the parent setting all that up.
2007-03-21 15:39:11 +00:00
/*
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
* Reserve the resource.
*
* XXX: Ignores failure for now. Failure here is probably a
* BIOS/firmware bug?
Change acpi's handling of suballocating system resources to be a little simpler. It now can just use rman_is_region_manager() during acpi_release_resource() to see if the the resource is suballocated from a system resource. Also, the driver no longer needs MD knowledge about how to setup bus space tags and handles when doing a suballocation, but can simply rely on bus_activate_resource() in the parent setting all that up.
2007-03-21 15:39:11 +00:00
*/
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
resource_list_reserve(rl, dev, child, type, &rid, start, end, count, 0);
return (0);
}
static struct resource *
acpi_alloc_resource(device_t bus, device_t child, int type, int *rid,
Convert rman to use rman_res_t instead of u_long Summary: Migrate to using the semi-opaque type rman_res_t to specify rman resources. For now, this is still compatible with u_long. This is step one in migrating rman to use uintmax_t for resources instead of u_long. Going forward, this could feasibly be used to specify architecture-specific definitions of resource ranges, rather than baking a specific integer type into the API. This change has been broken out to facilitate MFC'ing drivers back to 10 without breaking ABI. Reviewed By: jhb Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D5075
2016-01-27 02:23:54 +00:00
rman_res_t start, rman_res_t end, rman_res_t count, u_int flags)
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
{
Add the start of INTRNG support for ACPI. This adds a new acpi_bus interface with a map_intr method. This is similar to the Open Firmware map_intr method and allows us to create the needed mapping from ACPI space to INTRNG space. Obtained from: ABT Systems Ltd Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D8617
2018-01-11 17:09:12 +00:00
#ifndef INTRNG
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
ACPI_RESOURCE ares;
Add the start of INTRNG support for ACPI. This adds a new acpi_bus interface with a map_intr method. This is similar to the Open Firmware map_intr method and allows us to create the needed mapping from ACPI space to INTRNG space. Obtained from: ABT Systems Ltd Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D8617
2018-01-11 17:09:12 +00:00
#endif
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
struct acpi_device *ad;
struct resource_list_entry *rle;
struct resource_list *rl;
struct resource *res;
Introduce a RMAN_IS_DEFAULT_RANGE() macro, and use it. This simplifies checking for default resource range for bus_alloc_resource(), and improves readability. This is part of, and related to, the migration of rman_res_t from u_long to uintmax_t. Discussed with: jhb Suggested by: marcel
2016-02-20 01:32:58 +00:00
int isdefault = RMAN_IS_DEFAULT_RANGE(start, end);
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
/*
* First attempt at allocating the resource. For direct children,
* use resource_list_alloc() to handle reserved resources. For
Typo.
2011-06-24 13:58:56 +00:00
* other devices, pass the request up to our parent.
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
*/
if (bus == device_get_parent(child)) {
ad = device_get_ivars(child);
rl = &ad->ad_rl;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
/*
* Simulate the behavior of the ISA bus for direct children
* devices. That is, if a non-default range is specified for
* a resource that doesn't exist, use bus_set_resource() to
* add the resource before allocating it. Note that these
* resources will not be reserved.
*/
if (!isdefault && resource_list_find(rl, type, *rid) == NULL)
resource_list_add(rl, type, *rid, start, end, count);
res = resource_list_alloc(rl, bus, child, type, rid, start, end, count,
flags);
Add the start of INTRNG support for ACPI. This adds a new acpi_bus interface with a map_intr method. This is similar to the Open Firmware map_intr method and allows us to create the needed mapping from ACPI space to INTRNG space. Obtained from: ABT Systems Ltd Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D8617
2018-01-11 17:09:12 +00:00
#ifndef INTRNG
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
if (res != NULL && type == SYS_RES_IRQ) {
- Defer BUS_CONFIG_INTR() on ACPI IRQ resources until the resources are actually used. For most ACPI devices this means deferring the call until bus_alloc_resource(). - Add a function acpi_config_intr() to call BUS_CONFIG_INTR() for an ACPI IRQ resource using the trigger mode and polarity information stored in the ACPI resource object. - Add a function acpi_lookup_irq_resource() to lookup the ACPI IRQ resource that corresponds to a specified rid and new-bus resource. - Have the ACPI PCI bridge driver call BUS_CONFIG_INTR() on interrupts that it routes through link devices. - Remove needactivate variable from acpi_alloc_resource() by changing the function not modify the flags variable but just mask off RF_ACTIVE when calling rman_reserve_resource(). Reviewed by: njl (1, an earlier version)
2004-06-23 17:21:02 +00:00
/*
* Since bus_config_intr() takes immediate effect, we cannot
* configure the interrupt associated with a device when we
* parse the resources but have to defer it until a driver
* actually allocates the interrupt via bus_alloc_resource().
*
* XXX: Should we handle the lookup failing?
*/
if (ACPI_SUCCESS(acpi_lookup_irq_resource(child, *rid, res, &ares)))
acpi_config_intr(child, &ares);
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
}
Add the start of INTRNG support for ACPI. This adds a new acpi_bus interface with a map_intr method. This is similar to the Open Firmware map_intr method and allows us to create the needed mapping from ACPI space to INTRNG space. Obtained from: ABT Systems Ltd Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D8617
2018-01-11 17:09:12 +00:00
#endif
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
/*
* If this is an allocation of the "default" range for a given
* RID, fetch the exact bounds for this resource from the
* resource list entry to try to allocate the range from the
* system resource regions.
*/
if (res == NULL && isdefault) {
rle = resource_list_find(rl, type, *rid);
if (rle != NULL) {
start = rle->start;
end = rle->end;
count = rle->count;
}
}
} else
res = BUS_ALLOC_RESOURCE(device_get_parent(bus), child, type, rid,
start, end, count, flags);
/*
* If the first attempt failed and this is an allocation of a
* specific range, try to satisfy the request via a suballocation
If an allocation for a specific resource range fails because it is not in a decoded range for an ACPI Host-PCI bridge, try to allocate it from the ACPI system resource range. If that works, permit the resource allocation regardless. MFC after: 1 week
2011-10-12 14:13:32 +00:00
* from our system resource regions.
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
*/
If an allocation for a specific resource range fails because it is not in a decoded range for an ACPI Host-PCI bridge, try to allocate it from the ACPI system resource range. If that works, permit the resource allocation regardless. MFC after: 1 week
2011-10-12 14:13:32 +00:00
if (res == NULL && start + count - 1 == end)
res = acpi_alloc_sysres(child, type, rid, start, end, count, flags);
return (res);
}
/*
* Attempt to allocate a specific resource range from the system
* resource ranges. Note that we only handle memory and I/O port
* system resources.
*/
struct resource *
Convert rman to use rman_res_t instead of u_long Summary: Migrate to using the semi-opaque type rman_res_t to specify rman resources. For now, this is still compatible with u_long. This is step one in migrating rman to use uintmax_t for resources instead of u_long. Going forward, this could feasibly be used to specify architecture-specific definitions of resource ranges, rather than baking a specific integer type into the API. This change has been broken out to facilitate MFC'ing drivers back to 10 without breaking ABI. Reviewed By: jhb Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D5075
2016-01-27 02:23:54 +00:00
acpi_alloc_sysres(device_t child, int type, int *rid, rman_res_t start,
rman_res_t end, rman_res_t count, u_int flags)
If an allocation for a specific resource range fails because it is not in a decoded range for an ACPI Host-PCI bridge, try to allocate it from the ACPI system resource range. If that works, permit the resource allocation regardless. MFC after: 1 week
2011-10-12 14:13:32 +00:00
{
struct rman *rm;
struct resource *res;
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
switch (type) {
case SYS_RES_IOPORT:
rm = &acpi_rman_io;
break;
case SYS_RES_MEMORY:
rm = &acpi_rman_mem;
break;
default:
return (NULL);
}
If an allocation for a specific resource range fails because it is not in a decoded range for an ACPI Host-PCI bridge, try to allocate it from the ACPI system resource range. If that works, permit the resource allocation regardless. MFC after: 1 week
2011-10-12 14:13:32 +00:00
KASSERT(start + count - 1 == end, ("wildcard resource range"));
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
res = rman_reserve_resource(rm, start, end, count, flags & ~RF_ACTIVE,
child);
if (res == NULL)
return (NULL);
rman_set_rid(res, *rid);
/* If requested, activate the resource using the parent's method. */
if (flags & RF_ACTIVE)
if (bus_activate_resource(child, type, *rid, res) != 0) {
rman_release_resource(res);
return (NULL);
}
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
return (res);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
static int
Implement BUS_ADJUST_RESOURCE() for the x86 drivers that sit between the Host-PCI bridge drivers and nexus.
2011-06-10 12:30:16 +00:00
acpi_is_resource_managed(int type, struct resource *r)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
Change acpi's handling of suballocating system resources to be a little simpler. It now can just use rman_is_region_manager() during acpi_release_resource() to see if the the resource is suballocated from a system resource. Also, the driver no longer needs MD knowledge about how to setup bus space tags and handles when doing a suballocation, but can simply rely on bus_activate_resource() in the parent setting all that up.
2007-03-21 15:39:11 +00:00
/* We only handle memory and IO resources through rman. */
switch (type) {
case SYS_RES_IOPORT:
Implement BUS_ADJUST_RESOURCE() for the x86 drivers that sit between the Host-PCI bridge drivers and nexus.
2011-06-10 12:30:16 +00:00
return (rman_is_region_manager(r, &acpi_rman_io));
Change acpi's handling of suballocating system resources to be a little simpler. It now can just use rman_is_region_manager() during acpi_release_resource() to see if the the resource is suballocated from a system resource. Also, the driver no longer needs MD knowledge about how to setup bus space tags and handles when doing a suballocation, but can simply rely on bus_activate_resource() in the parent setting all that up.
2007-03-21 15:39:11 +00:00
case SYS_RES_MEMORY:
Implement BUS_ADJUST_RESOURCE() for the x86 drivers that sit between the Host-PCI bridge drivers and nexus.
2011-06-10 12:30:16 +00:00
return (rman_is_region_manager(r, &acpi_rman_mem));
Change acpi's handling of suballocating system resources to be a little simpler. It now can just use rman_is_region_manager() during acpi_release_resource() to see if the the resource is suballocated from a system resource. Also, the driver no longer needs MD knowledge about how to setup bus space tags and handles when doing a suballocation, but can simply rely on bus_activate_resource() in the parent setting all that up.
2007-03-21 15:39:11 +00:00
}
Implement BUS_ADJUST_RESOURCE() for the x86 drivers that sit between the Host-PCI bridge drivers and nexus.
2011-06-10 12:30:16 +00:00
return (0);
}
static int
acpi_adjust_resource(device_t bus, device_t child, int type, struct resource *r,
Convert rman to use rman_res_t instead of u_long Summary: Migrate to using the semi-opaque type rman_res_t to specify rman resources. For now, this is still compatible with u_long. This is step one in migrating rman to use uintmax_t for resources instead of u_long. Going forward, this could feasibly be used to specify architecture-specific definitions of resource ranges, rather than baking a specific integer type into the API. This change has been broken out to facilitate MFC'ing drivers back to 10 without breaking ABI. Reviewed By: jhb Sponsored by: Alex Perez/Inertial Computing Differential Revision: https://reviews.freebsd.org/D5075
2016-01-27 02:23:54 +00:00
rman_res_t start, rman_res_t end)
Implement BUS_ADJUST_RESOURCE() for the x86 drivers that sit between the Host-PCI bridge drivers and nexus.
2011-06-10 12:30:16 +00:00
{
if (acpi_is_resource_managed(type, r))
return (rman_adjust_resource(r, start, end));
return (bus_generic_adjust_resource(bus, child, type, r, start, end));
}
static int
acpi_release_resource(device_t bus, device_t child, int type, int rid,
struct resource *r)
{
int ret;
Change acpi's handling of suballocating system resources to be a little simpler. It now can just use rman_is_region_manager() during acpi_release_resource() to see if the the resource is suballocated from a system resource. Also, the driver no longer needs MD knowledge about how to setup bus space tags and handles when doing a suballocation, but can simply rely on bus_activate_resource() in the parent setting all that up.
2007-03-21 15:39:11 +00:00
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
/*
Change acpi's handling of suballocating system resources to be a little simpler. It now can just use rman_is_region_manager() during acpi_release_resource() to see if the the resource is suballocated from a system resource. Also, the driver no longer needs MD knowledge about how to setup bus space tags and handles when doing a suballocation, but can simply rely on bus_activate_resource() in the parent setting all that up.
2007-03-21 15:39:11 +00:00
* If this resource belongs to one of our internal managers,
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
* deactivate it and release it to the local pool.
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
*/
Implement BUS_ADJUST_RESOURCE() for the x86 drivers that sit between the Host-PCI bridge drivers and nexus.
2011-06-10 12:30:16 +00:00
if (acpi_is_resource_managed(type, r)) {
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
if (rman_get_flags(r) & RF_ACTIVE) {
ret = bus_deactivate_resource(child, type, rid, r);
if (ret != 0)
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
return (ret);
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
}
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
return (rman_release_resource(r));
}
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
return (bus_generic_rl_release_resource(bus, child, type, rid, r));
Add acpi_bus_alloc_gas() for allocating a memory or IO resource from its Generic Address Structure.
2003-11-15 19:18:29 +00:00
}
Add support for bus_delete_resource() and make acpi_bus_alloc_gas() do this before setting a new resource.
2005-03-22 20:00:57 +00:00
static void
acpi_delete_resource(device_t bus, device_t child, int type, int rid)
{
struct resource_list *rl;
rl = acpi_get_rlist(bus, child);
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
if (resource_list_busy(rl, type, rid)) {
device_printf(bus, "delete_resource: Resource still owned by child"
" (type=%d, rid=%d)\n", type, rid);
return;
}
resource_list_unreserve(rl, bus, child, type, rid);
Add support for bus_delete_resource() and make acpi_bus_alloc_gas() do this before setting a new resource.
2005-03-22 20:00:57 +00:00
resource_list_delete(rl, type, rid);
}
Add acpi_bus_alloc_gas() for allocating a memory or IO resource from its Generic Address Structure.
2003-11-15 19:18:29 +00:00
/* Allocate an IO port or memory resource, given its GAS. */
Convert the acpi_bus_alloc_gas() and acpi_PkgGas() APIs to output the memory type. This is needed if the resource is to be released later. The RID is still also present, though less necessary since rman_get_rid() can be used to obtain it from the resource.
2005-02-05 22:28:36 +00:00
int
acpi_bus_alloc_gas(device_t dev, int *type, int *rid, ACPI_GENERIC_ADDRESS *gas,
Re-work Cx handling to be per-cpu and asymmetrical, fixing support on modern dual-core systems as well. - Parse the _CST packages for each cpu and track all the states individually, on a per-cpu basis. - Revert to generic FADT/P_BLK based Cx control if the _CST package is not present on all cpus. In that case, the new driver will still support per-cpu Cx state handling. The driver will determine the highest Cx level that can be supported by all the cpus and configure the available Cx state based on that. - Fixed the case where multiple cpus in the system share the same registers for Cx state handling. To do that, added a new flag parameter to the acpi_PkgGas and acpi_bus_alloc_gas functions that enable the caller to add the RF_SHAREABLE flag. This flag could also be useful to other callers (acpi_throttle?) in the tree but this change is not yet made. - For Core Duo cpus, both cores seems to be taken out of C3 state when any one of the cores need to transition out. This broke the short sleep detection logic. It is disabled now if there is more than one cpu in the system for now as it fixed it in my case. This quirk may need to be re-enabled later differently. - Added support to control cx_lowest on a per-cpu basis. There is still a generic cx_lowest to enable changing cx_lowest for all cpus with a single sysctl and for ease of use. Sample output for the new sysctl: dev.cpu.0.cx_supported: C1/1 C2/1 C3/57 dev.cpu.0.cx_lowest: C3 dev.cpu.0.cx_usage: 0.00% 43.16% 56.83% dev.cpu.1.cx_supported: C1/1 C2/1 C3/57 dev.cpu.1.cx_lowest: C3 dev.cpu.1.cx_usage: 0.00% 45.65% 54.34% hw.acpi.cpu.cx_lowest: C3 This work was done by Stephane E. Potvin with some simple reworking by myself. Thank you. Submitted by: Stephane E. Potvin <sepotvin / videotron.ca> MFC after: 2 weeks
2007-01-07 21:53:42 +00:00
struct resource **res, u_int flags)
Add acpi_bus_alloc_gas() for allocating a memory or IO resource from its Generic Address Structure.
2003-11-15 19:18:29 +00:00
{
Convert the acpi_bus_alloc_gas() and acpi_PkgGas() APIs to output the memory type. This is needed if the resource is to be released later. The RID is still also present, though less necessary since rman_get_rid() can be used to obtain it from the resource.
2005-02-05 22:28:36 +00:00
int error, res_type;
Add acpi_bus_alloc_gas() for allocating a memory or IO resource from its Generic Address Structure.
2003-11-15 19:18:29 +00:00
Convert the acpi_bus_alloc_gas() and acpi_PkgGas() APIs to output the memory type. This is needed if the resource is to be released later. The RID is still also present, though less necessary since rman_get_rid() can be used to obtain it from the resource.
2005-02-05 22:28:36 +00:00
error = ENOMEM;
Check for the address space type first before validating it. In particular, we want to return EOPNOTSUPP for FFixedHW no matter what the address. Submitted by: Bruno Ducrot
2005-02-17 19:00:14 +00:00
if (type == NULL || rid == NULL || gas == NULL || res == NULL)
Convert the acpi_bus_alloc_gas() and acpi_PkgGas() APIs to output the memory type. This is needed if the resource is to be released later. The RID is still also present, though less necessary since rman_get_rid() can be used to obtain it from the resource.
2005-02-05 22:28:36 +00:00
return (EINVAL);
Add acpi_bus_alloc_gas() for allocating a memory or IO resource from its Generic Address Structure.
2003-11-15 19:18:29 +00:00
Check for the address space type first before validating it. In particular, we want to return EOPNOTSUPP for FFixedHW no matter what the address. Submitted by: Bruno Ducrot
2005-02-17 19:00:14 +00:00
/* We only support memory and IO spaces. */
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
switch (gas->SpaceId) {
Add acpi_bus_alloc_gas() for allocating a memory or IO resource from its Generic Address Structure.
2003-11-15 19:18:29 +00:00
case ACPI_ADR_SPACE_SYSTEM_MEMORY:
Convert the acpi_bus_alloc_gas() and acpi_PkgGas() APIs to output the memory type. This is needed if the resource is to be released later. The RID is still also present, though less necessary since rman_get_rid() can be used to obtain it from the resource.
2005-02-05 22:28:36 +00:00
res_type = SYS_RES_MEMORY;
Add acpi_bus_alloc_gas() for allocating a memory or IO resource from its Generic Address Structure.
2003-11-15 19:18:29 +00:00
break;
case ACPI_ADR_SPACE_SYSTEM_IO:
Convert the acpi_bus_alloc_gas() and acpi_PkgGas() APIs to output the memory type. This is needed if the resource is to be released later. The RID is still also present, though less necessary since rman_get_rid() can be used to obtain it from the resource.
2005-02-05 22:28:36 +00:00
res_type = SYS_RES_IOPORT;
Add acpi_bus_alloc_gas() for allocating a memory or IO resource from its Generic Address Structure.
2003-11-15 19:18:29 +00:00
break;
default:
Convert the acpi_bus_alloc_gas() and acpi_PkgGas() APIs to output the memory type. This is needed if the resource is to be released later. The RID is still also present, though less necessary since rman_get_rid() can be used to obtain it from the resource.
2005-02-05 22:28:36 +00:00
return (EOPNOTSUPP);
Add acpi_bus_alloc_gas() for allocating a memory or IO resource from its Generic Address Structure.
2003-11-15 19:18:29 +00:00
}
If a register width is less than 8, assume the BIOS author thought it was in units of bytes and adjust accordingly. This is found at least on the Sony PCG-505BX.
2005-02-23 03:20:00 +00:00
/*
Instead of assuming units of bytes, it seems more likely that this is a bitfield.
2005-02-25 16:57:34 +00:00
* If the register width is less than 8, assume the BIOS author means
* it is a bit field and just allocate a byte.
If a register width is less than 8, assume the BIOS author thought it was in units of bytes and adjust accordingly. This is found at least on the Sony PCG-505BX.
2005-02-23 03:20:00 +00:00
*/
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
if (gas->BitWidth && gas->BitWidth < 8)
gas->BitWidth = 8;
If a register width is less than 8, assume the BIOS author thought it was in units of bytes and adjust accordingly. This is found at least on the Sony PCG-505BX.
2005-02-23 03:20:00 +00:00
Check for the address space type first before validating it. In particular, we want to return EOPNOTSUPP for FFixedHW no matter what the address. Submitted by: Bruno Ducrot
2005-02-17 19:00:14 +00:00
/* Validate the address after we're sure we support the space. */
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
if (gas->Address == 0 || gas->BitWidth == 0)
Check for the address space type first before validating it. In particular, we want to return EOPNOTSUPP for FFixedHW no matter what the address. Submitted by: Bruno Ducrot
2005-02-17 19:00:14 +00:00
return (EINVAL);
Convert the acpi_bus_alloc_gas() and acpi_PkgGas() APIs to output the memory type. This is needed if the resource is to be released later. The RID is still also present, though less necessary since rman_get_rid() can be used to obtain it from the resource.
2005-02-05 22:28:36 +00:00
bus_set_resource(dev, res_type, *rid, gas->Address,
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
gas->BitWidth / 8);
Re-work Cx handling to be per-cpu and asymmetrical, fixing support on modern dual-core systems as well. - Parse the _CST packages for each cpu and track all the states individually, on a per-cpu basis. - Revert to generic FADT/P_BLK based Cx control if the _CST package is not present on all cpus. In that case, the new driver will still support per-cpu Cx state handling. The driver will determine the highest Cx level that can be supported by all the cpus and configure the available Cx state based on that. - Fixed the case where multiple cpus in the system share the same registers for Cx state handling. To do that, added a new flag parameter to the acpi_PkgGas and acpi_bus_alloc_gas functions that enable the caller to add the RF_SHAREABLE flag. This flag could also be useful to other callers (acpi_throttle?) in the tree but this change is not yet made. - For Core Duo cpus, both cores seems to be taken out of C3 state when any one of the cores need to transition out. This broke the short sleep detection logic. It is disabled now if there is more than one cpu in the system for now as it fixed it in my case. This quirk may need to be re-enabled later differently. - Added support to control cx_lowest on a per-cpu basis. There is still a generic cx_lowest to enable changing cx_lowest for all cpus with a single sysctl and for ease of use. Sample output for the new sysctl: dev.cpu.0.cx_supported: C1/1 C2/1 C3/57 dev.cpu.0.cx_lowest: C3 dev.cpu.0.cx_usage: 0.00% 43.16% 56.83% dev.cpu.1.cx_supported: C1/1 C2/1 C3/57 dev.cpu.1.cx_lowest: C3 dev.cpu.1.cx_usage: 0.00% 45.65% 54.34% hw.acpi.cpu.cx_lowest: C3 This work was done by Stephane E. Potvin with some simple reworking by myself. Thank you. Submitted by: Stephane E. Potvin <sepotvin / videotron.ca> MFC after: 2 weeks
2007-01-07 21:53:42 +00:00
*res = bus_alloc_resource_any(dev, res_type, rid, RF_ACTIVE | flags);
Convert the acpi_bus_alloc_gas() and acpi_PkgGas() APIs to output the memory type. This is needed if the resource is to be released later. The RID is still also present, though less necessary since rman_get_rid() can be used to obtain it from the resource.
2005-02-05 22:28:36 +00:00
if (*res != NULL) {
*type = res_type;
error = 0;
Clean up resources properly if acpi_perf fails to attach. First, change acpi_bus_alloc_gas() to delete the resource it set if alloc fails. Then, change acpi_perf to delete the resource after releasing it if alloc fails. This should make probe and attach both fully restartable if either fails.
2005-03-27 22:38:28 +00:00
} else
bus_delete_resource(dev, res_type, *rid);
Convert the acpi_bus_alloc_gas() and acpi_PkgGas() APIs to output the memory type. This is needed if the resource is to be released later. The RID is still also present, though less necessary since rman_get_rid() can be used to obtain it from the resource.
2005-02-05 22:28:36 +00:00
return (error);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Include isa/pnpvar.h and remove a duplicate copy of PNP_EISAID.
2004-06-29 01:33:35 +00:00
/* Probe _HID and _CID for compatible ISA PNP ids. */
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
static uint32_t
Allow the ACPI subsystem to be disabled with a hint. Avoid fully initialising the ACPI namespace if we are attempting to avoid parts of it. This is a workaround for some systems that still crash the interpreter. Implement the ISA_IVAR_LOGICALID for ISA compatibility. Implement stubs for other PnP ID-related ivars.
2001-09-07 02:57:29 +00:00
acpi_isa_get_logicalid(device_t dev)
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
{
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
ACPI_DEVICE_INFO *devinfo;
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
ACPI_HANDLE h;
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
uint32_t pnpid;
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
Update code to work with 0619 dist * Use ACPI_BUFFER as the type for AcpiGetObjectInfo * Remove AcpiEnableEvent/AcpiClearEvent for ACPI_EVENT_FIXED (power/sleep buttons) as they are no longer needed * Change calls to use the new GPE functions * Add AcpiOs*Lock functions
2003-07-13 22:57:16 +00:00
/* Fetch and validate the HID. */
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
if ((h = acpi_get_handle(dev)) == NULL ||
ACPI_FAILURE(AcpiGetObjectInfo(h, &devinfo)))
return_VALUE (0);
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
pnpid = (devinfo->Valid & ACPI_VALID_HID) != 0 &&
devinfo->HardwareId.Length >= ACPI_EISAID_STRING_SIZE ?
PNP_EISAID(devinfo->HardwareId.String) : 0;
AcpiOsFree(devinfo);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return_VALUE (pnpid);
Allow the ACPI subsystem to be disabled with a hint. Avoid fully initialising the ACPI namespace if we are attempting to avoid parts of it. This is a workaround for some systems that still crash the interpreter. Implement the ISA_IVAR_LOGICALID for ISA compatibility. Implement stubs for other PnP ID-related ivars.
2001-09-07 02:57:29 +00:00
}
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
static int
acpi_isa_get_compatid(device_t dev, uint32_t *cids, int count)
Add integer value of _CID handling. If _CID is string, it will need more complicated handling to distinguish bus other than ISA. Submitted by: Paul Wankadia <junyer@gmx.net>
2003-03-06 14:40:15 +00:00
{
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
ACPI_DEVICE_INFO *devinfo;
Merge ACPICA 20121018.
2012-10-23 23:49:17 +00:00
ACPI_PNP_DEVICE_ID *ids;
Add integer value of _CID handling. If _CID is string, it will need more complicated handling to distinguish bus other than ISA. Submitted by: Paul Wankadia <junyer@gmx.net>
2003-03-06 14:40:15 +00:00
ACPI_HANDLE h;
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
uint32_t *pnpid;
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
int i, valid;
Add integer value of _CID handling. If _CID is string, it will need more complicated handling to distinguish bus other than ISA. Submitted by: Paul Wankadia <junyer@gmx.net>
2003-03-06 14:40:15 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
pnpid = cids;
Fix locking broken by recent _CID changes. Always be sure to unlock in the error case.
2003-12-23 18:47:31 +00:00
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
/* Fetch and validate the CID */
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
if ((h = acpi_get_handle(dev)) == NULL ||
ACPI_FAILURE(AcpiGetObjectInfo(h, &devinfo)))
return_VALUE (0);
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
if ((devinfo->Valid & ACPI_VALID_CID) == 0) {
AcpiOsFree(devinfo);
return_VALUE (0);
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
}
Add integer value of _CID handling. If _CID is string, it will need more complicated handling to distinguish bus other than ISA. Submitted by: Paul Wankadia <junyer@gmx.net>
2003-03-06 14:40:15 +00:00
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
if (devinfo->CompatibleIdList.Count < count)
count = devinfo->CompatibleIdList.Count;
ids = devinfo->CompatibleIdList.Ids;
for (i = 0, valid = 0; i < count; i++)
if (ids[i].Length >= ACPI_EISAID_STRING_SIZE &&
strncmp(ids[i].String, "PNP", 3) == 0) {
*pnpid++ = PNP_EISAID(ids[i].String);
valid++;
}
AcpiOsFree(devinfo);
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
return_VALUE (valid);
Add integer value of _CID handling. If _CID is string, it will need more complicated handling to distinguish bus other than ISA. Submitted by: Paul Wankadia <junyer@gmx.net>
2003-03-06 14:40:15 +00:00
}
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
static int
acpi_device_id_probe(device_t bus, device_t dev, char **ids, char **match)
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
{
ACPI_HANDLE h;
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
ACPI_OBJECT_TYPE t;
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
int rv;
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
int i;
h = acpi_get_handle(dev);
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
if (ids == NULL || h == NULL)
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
return (ENXIO);
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
t = acpi_get_type(dev);
if (t != ACPI_TYPE_DEVICE && t != ACPI_TYPE_PROCESSOR)
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
return (ENXIO);
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
/* Try to match one of the array of IDs with a HID or CID. */
for (i = 0; ids[i] != NULL; i++) {
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
rv = acpi_MatchHid(h, ids[i]);
if (rv == ACPI_MATCHHID_NOMATCH)
continue;
if (match != NULL) {
*match = ids[i];
}
return ((rv == ACPI_MATCHHID_HID)?
BUS_PROBE_DEFAULT : BUS_PROBE_LOW_PRIORITY);
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
}
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
return (ENXIO);
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
}
static ACPI_STATUS
acpi_device_eval_obj(device_t bus, device_t dev, ACPI_STRING pathname,
ACPI_OBJECT_LIST *parameters, ACPI_BUFFER *ret)
{
ACPI_HANDLE h;
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
if (dev == NULL)
h = ACPI_ROOT_OBJECT;
else if ((h = acpi_get_handle(dev)) == NULL)
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
return (AE_BAD_PARAMETER);
return (AcpiEvaluateObject(h, pathname, parameters, ret));
}
Add a new method to the PCI bridge interface, PCIB_POWER_FOR_SLEEP(). This method is used by the PCI bus driver to query the power management system to determine the proper device state to be used for a device during suspend and resume. For the ACPI PCI bridge drivers this calls acpi_device_pwr_for_sleep(). This removes ACPI-specific knowledge from the PCI and PCI-PCI bridge drivers. Reviewed by: jkim
2010-08-17 15:44:52 +00:00
int
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
acpi_device_pwr_for_sleep(device_t bus, device_t dev, int *dstate)
{
struct acpi_softc *sc;
ACPI_HANDLE handle;
ACPI_STATUS status;
char sxd[8];
handle = acpi_get_handle(dev);
/*
* XXX If we find these devices, don't try to power them down.
* The serial and IRDA ports on my T23 hang the system when
* set to D3 and it appears that such legacy devices may
* need special handling in their drivers.
*/
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
if (dstate == NULL || handle == NULL ||
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
acpi_MatchHid(handle, "PNP0500") ||
acpi_MatchHid(handle, "PNP0501") ||
acpi_MatchHid(handle, "PNP0502") ||
acpi_MatchHid(handle, "PNP0510") ||
acpi_MatchHid(handle, "PNP0511"))
return (ENXIO);
/*
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
* Override next state with the value from _SxD, if present.
* Note illegal _S0D is evaluated because some systems expect this.
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
*/
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
sc = device_get_softc(bus);
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
snprintf(sxd, sizeof(sxd), "_S%dD", sc->acpi_sstate);
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
status = acpi_GetInteger(handle, sxd, dstate);
if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) {
device_printf(dev, "failed to get %s on %s: %s\n", sxd,
acpi_name(handle), AcpiFormatException(status));
return (ENXIO);
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
}
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
return (0);
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
}
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
/* Callback arg for our implementation of walking the namespace. */
struct acpi_device_scan_ctx {
acpi_scan_cb_t user_fn;
void *arg;
ACPI_HANDLE parent;
};
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
static ACPI_STATUS
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
acpi_device_scan_cb(ACPI_HANDLE h, UINT32 level, void *arg, void **retval)
{
struct acpi_device_scan_ctx *ctx;
device_t dev, old_dev;
ACPI_STATUS status;
ACPI_OBJECT_TYPE type;
/*
* Skip this device if we think we'll have trouble with it or it is
* the parent where the scan began.
*/
ctx = (struct acpi_device_scan_ctx *)arg;
if (acpi_avoid(h) || h == ctx->parent)
return (AE_OK);
/* If this is not a valid device type (e.g., a method), skip it. */
if (ACPI_FAILURE(AcpiGetType(h, &type)))
return (AE_OK);
if (type != ACPI_TYPE_DEVICE && type != ACPI_TYPE_PROCESSOR &&
type != ACPI_TYPE_THERMAL && type != ACPI_TYPE_POWER)
return (AE_OK);
/*
* Call the user function with the current device. If it is unchanged
* afterwards, return. Otherwise, we update the handle to the new dev.
*/
old_dev = acpi_get_device(h);
dev = old_dev;
status = ctx->user_fn(h, &dev, level, ctx->arg);
if (ACPI_FAILURE(status) || old_dev == dev)
return (status);
/* Remove the old child and its connection to the handle. */
acpi: Fix stalled value returned by acpi_get_device() after device deletion Newbus device reference attached to ACPI handle is not cleared when newbus device is deleted with devctl(8) delete command. Fix that with calling of AcpiDetachData() from "child_deleted" bus method like acpi_pci driver does. MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D22902
2020-03-09 20:27:25 +00:00
if (old_dev != NULL)
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
device_delete_child(device_get_parent(old_dev), old_dev);
/* Recreate the handle association if the user created a device. */
if (dev != NULL)
AcpiAttachData(h, acpi_fake_objhandler, dev);
return (AE_OK);
}
static ACPI_STATUS
acpi_device_scan_children(device_t bus, device_t dev, int max_depth,
acpi_scan_cb_t user_fn, void *arg)
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
{
ACPI_HANDLE h;
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
struct acpi_device_scan_ctx ctx;
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
if (acpi_disabled("children"))
return (AE_OK);
if (dev == NULL)
h = ACPI_ROOT_OBJECT;
else if ((h = acpi_get_handle(dev)) == NULL)
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
return (AE_BAD_PARAMETER);
Update the interface for child drivers. Add acpi_scan_children, which allows a bus to re-enumerate its child handles and optionally replace them with new children, arranged to the bus's liking. (The current device space is flat with all devices immediately under acpi0). Add comments for each interface.
2004-07-15 16:29:08 +00:00
ctx.user_fn = user_fn;
ctx.arg = arg;
ctx.parent = h;
return (AcpiWalkNamespace(ACPI_TYPE_ANY, h, max_depth,
Merge ACPICA 20091112.
2009-11-16 21:47:12 +00:00
acpi_device_scan_cb, NULL, &ctx, NULL));
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
}
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
/*
* Even though ACPI devices are not PCI, we use the PCI approach for setting
* device power states since it's close enough to ACPI.
*/
acpi: Export functions required by upcoming acpi_iicbus driver.
2020-03-09 20:28:45 +00:00
int
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
acpi_set_powerstate(device_t child, int state)
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
{
ACPI_HANDLE h;
ACPI_STATUS status;
h = acpi_get_handle(child);
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
if (state < ACPI_STATE_D0 || state > ACPI_D_STATES_MAX)
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
return (EINVAL);
if (h == NULL)
return (0);
/* Ignore errors if the power methods aren't present. */
status = acpi_pwr_switch_consumer(h, state);
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
if (ACPI_SUCCESS(status)) {
if (bootverbose)
device_printf(child, "set ACPI power state D%d on %s\n",
state, acpi_name(h));
} else if (status != AE_NOT_FOUND)
device_printf(child,
"failed to set ACPI power state D%d on %s: %s\n", state,
acpi_name(h), AcpiFormatException(status));
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
Remove PCI_SET_POWERSTATE method from acpi.c and eradicate all PCI-specific knowledges from the file. All PCI devices enumerated in ACPI tree must use correct one from acpi_pci.c any way. Reduce duplicate codes as we did for pci.c in r213905. Do not return ESRCH from PCIB_POWER_FOR_SLEEP method. When the method is not found, just return zero without modifying the given default value as it is completely optional. As a side effect, the return state must not be NULL. Note there is actually no functional change by removing ESRCH because acpi_pcib_power_for_sleep() always returns zero. Adjust debugging messages and add new ones under bootverbose to help debugging device power state related issues. Reviewed by: jhb, imp (earlier versions)
2010-10-19 19:53:06 +00:00
return (0);
Turn ACPI and PCI devices off or to a lower power state in suspend and back on again in resume. Override the default of D3 with the value the BIOS specifies in _SxD, if present. Skip serial devices (PNP05xx) since they seem to hang when set to D3 and may require special driver support. Also, skip non-type 0 PCI devices (i.e., bridges) since our we don't yet save/restore their config space and that seems to be necessary. If this gives you trouble with suspend/resume, you can disable the new ACPI and PCI power behavior separately with these tunables & sysctls: debug.acpi.do_powerstate hw.pci.do_powerstate Approved by: imp (pci) Tested by: acpi@ (numerous)
2004-12-02 08:07:12 +00:00
}
Allow the ACPI subsystem to be disabled with a hint. Avoid fully initialising the ACPI namespace if we are attempting to avoid parts of it. This is a workaround for some systems that still crash the interpreter. Implement the ISA_IVAR_LOGICALID for ISA compatibility. Implement stubs for other PnP ID-related ivars.
2001-09-07 02:57:29 +00:00
static int
acpi_isa_pnp_probe(device_t bus, device_t child, struct isa_pnp_id *ids)
{
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
int result, cid_count, i;
uint32_t lid, cids[8];
Allow the ACPI subsystem to be disabled with a hint. Avoid fully initialising the ACPI namespace if we are attempting to avoid parts of it. This is a workaround for some systems that still crash the interpreter. Implement the ISA_IVAR_LOGICALID for ISA compatibility. Implement stubs for other PnP ID-related ivars.
2001-09-07 02:57:29 +00:00
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
Allow the ACPI subsystem to be disabled with a hint. Avoid fully initialising the ACPI namespace if we are attempting to avoid parts of it. This is a workaround for some systems that still crash the interpreter. Implement the ISA_IVAR_LOGICALID for ISA compatibility. Implement stubs for other PnP ID-related ivars.
2001-09-07 02:57:29 +00:00
/*
* ISA-style drivers attached to ACPI may persist and
* probe manually if we return ENOENT. We never want
* that to happen, so don't ever return it.
*/
result = ENXIO;
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Scan the supplied IDs for a match */
Add integer value of _CID handling. If _CID is string, it will need more complicated handling to distinguish bus other than ISA. Submitted by: Paul Wankadia <junyer@gmx.net>
2003-03-06 14:40:15 +00:00
lid = acpi_isa_get_logicalid(child);
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
cid_count = acpi_isa_get_compatid(child, cids, 8);
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
while (ids && ids->ip_id) {
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
if (lid == ids->ip_id) {
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
result = 0;
goto out;
}
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
for (i = 0; i < cid_count; i++) {
if (cids[i] == ids->ip_id) {
result = 0;
goto out;
}
}
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
ids++;
}
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
out:
Attach the device description for ISA devices on the ACPI bus. MFC after: 1 day
2004-10-12 21:33:08 +00:00
if (result == 0 && ids->ip_desc)
device_set_desc(child, ids->ip_desc);
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
return_VALUE (result);
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
}
Extend the support for PCI-e memory mapped configuration space access: - Rename pciereg_cfgopen() to pcie_cfgregopen() and expose it to the rest of the kernel. It now also accepts parameters via function arguments rather than global variables. - Add a notion of minimum and maximum bus numbers and reject requests for an out of range bus. - Add more range checks on slot/func/reg/bytes parameters to the cfg reg read/write routines. Don't panic on any invalid parameters, just fail the request (writes do nothing, reads return -1). This matches the behavior of the other cfg mechanisms. - Port the memory mapped configuration space access to amd64. On amd64 we simply use the direct map (via pmap_mapdev()) for the memory mapped window. - During acpi_attach() just after loading the ACPI tables, check for a MCFG table. If it exists, call pciereg_cfgopen() on each subtable (memory mapped window). For now we only support windows for domain 0 that start with bus 0. This removes the need for more chipset-specific quirks in the MD code. - Remove the chipset-specific quirks for the Intel 5000P/V/Z chipsets since these machines should all have MCFG tables via ACPI. - Updated pci_cfgregopen() to DTRT if ACPI had invoked pcie_cfgregopen() earlier. MFC after: 2 weeks
2008-08-22 02:14:23 +00:00
/*
* Look for a MCFG table. If it is present, use the settings for
* domain (segment) 0 to setup PCI config space access via the memory
* map.
acpica, pci_host_generic_acpi: redo pci_host_generic_acpi.c This is a major update for pci_host_generic_acpi.c, the current implementation has some gaps that are better fixed up in one go. The changes are to: * Follow x86 method of not adding PCI resources to PCI host bridge in ACPI code. This has been moved to pci_host_generic_acpi.c, where we walk thru its resources of the host bridge and add them. * Fixup code in pci_host_generic_acpi.c to read all decoded ranges and update the 'ranges' property. This allows us to share most of the code with generic implementation (and the FDT one). * Parse and setup IO ranges and bus ranges when walking the resources above. Drop most of the changes related to this from acpica code. * Add the ECAM memory area as mem resource 0. Implement the logic to get the ECAM area from MCFG (using bus range which we now decode), or from _CBA (using _BBN/bus range). Drop aarch64 ifdefs from acpica code which did part of this. * Switch resource activation to similar code as FDT implementation, this can be moved into generic implementation in a later pass. * Drop the mechanism of using the 7th bit of bus number as the domain, this is not correct and will work only in very specific cases. Use _SEG as PCI domain and use the bus ranges of the host bridge to provide start bus number. This commit should not make any functional change to dev/acpica/acpi.c for other architectures, almost all the changes there are to revert earlier additions in this file done for aarch64. Reviewed by: andrew Differential Revision: https://reviews.freebsd.org/D17791
2018-11-19 03:16:16 +00:00
*
* On non-x86 architectures (arm64 for now), this will be done from the
* PCI host bridge driver.
Extend the support for PCI-e memory mapped configuration space access: - Rename pciereg_cfgopen() to pcie_cfgregopen() and expose it to the rest of the kernel. It now also accepts parameters via function arguments rather than global variables. - Add a notion of minimum and maximum bus numbers and reject requests for an out of range bus. - Add more range checks on slot/func/reg/bytes parameters to the cfg reg read/write routines. Don't panic on any invalid parameters, just fail the request (writes do nothing, reads return -1). This matches the behavior of the other cfg mechanisms. - Port the memory mapped configuration space access to amd64. On amd64 we simply use the direct map (via pmap_mapdev()) for the memory mapped window. - During acpi_attach() just after loading the ACPI tables, check for a MCFG table. If it exists, call pciereg_cfgopen() on each subtable (memory mapped window). For now we only support windows for domain 0 that start with bus 0. This removes the need for more chipset-specific quirks in the MD code. - Remove the chipset-specific quirks for the Intel 5000P/V/Z chipsets since these machines should all have MCFG tables via ACPI. - Updated pci_cfgregopen() to DTRT if ACPI had invoked pcie_cfgregopen() earlier. MFC after: 2 weeks
2008-08-22 02:14:23 +00:00
*/
static void
acpi_enable_pcie(void)
{
acpica, pci_host_generic_acpi: redo pci_host_generic_acpi.c This is a major update for pci_host_generic_acpi.c, the current implementation has some gaps that are better fixed up in one go. The changes are to: * Follow x86 method of not adding PCI resources to PCI host bridge in ACPI code. This has been moved to pci_host_generic_acpi.c, where we walk thru its resources of the host bridge and add them. * Fixup code in pci_host_generic_acpi.c to read all decoded ranges and update the 'ranges' property. This allows us to share most of the code with generic implementation (and the FDT one). * Parse and setup IO ranges and bus ranges when walking the resources above. Drop most of the changes related to this from acpica code. * Add the ECAM memory area as mem resource 0. Implement the logic to get the ECAM area from MCFG (using bus range which we now decode), or from _CBA (using _BBN/bus range). Drop aarch64 ifdefs from acpica code which did part of this. * Switch resource activation to similar code as FDT implementation, this can be moved into generic implementation in a later pass. * Drop the mechanism of using the 7th bit of bus number as the domain, this is not correct and will work only in very specific cases. Use _SEG as PCI domain and use the bus ranges of the host bridge to provide start bus number. This commit should not make any functional change to dev/acpica/acpi.c for other architectures, almost all the changes there are to revert earlier additions in this file done for aarch64. Reviewed by: andrew Differential Revision: https://reviews.freebsd.org/D17791
2018-11-19 03:16:16 +00:00
#if defined(__i386__) || defined(__amd64__)
Extend the support for PCI-e memory mapped configuration space access: - Rename pciereg_cfgopen() to pcie_cfgregopen() and expose it to the rest of the kernel. It now also accepts parameters via function arguments rather than global variables. - Add a notion of minimum and maximum bus numbers and reject requests for an out of range bus. - Add more range checks on slot/func/reg/bytes parameters to the cfg reg read/write routines. Don't panic on any invalid parameters, just fail the request (writes do nothing, reads return -1). This matches the behavior of the other cfg mechanisms. - Port the memory mapped configuration space access to amd64. On amd64 we simply use the direct map (via pmap_mapdev()) for the memory mapped window. - During acpi_attach() just after loading the ACPI tables, check for a MCFG table. If it exists, call pciereg_cfgopen() on each subtable (memory mapped window). For now we only support windows for domain 0 that start with bus 0. This removes the need for more chipset-specific quirks in the MD code. - Remove the chipset-specific quirks for the Intel 5000P/V/Z chipsets since these machines should all have MCFG tables via ACPI. - Updated pci_cfgregopen() to DTRT if ACPI had invoked pcie_cfgregopen() earlier. MFC after: 2 weeks
2008-08-22 02:14:23 +00:00
ACPI_TABLE_HEADER *hdr;
ACPI_MCFG_ALLOCATION *alloc, *end;
ACPI_STATUS status;
status = AcpiGetTable(ACPI_SIG_MCFG, 1, &hdr);
if (ACPI_FAILURE(status))
return;
end = (ACPI_MCFG_ALLOCATION *)((char *)hdr + hdr->Length);
alloc = (ACPI_MCFG_ALLOCATION *)((ACPI_TABLE_MCFG *)hdr + 1);
while (alloc < end) {
if (alloc->PciSegment == 0) {
pcie_cfgregopen(alloc->Address, alloc->StartBusNumber,
alloc->EndBusNumber);
return;
}
alloc++;
}
#endif
acpica, pci_host_generic_acpi: redo pci_host_generic_acpi.c This is a major update for pci_host_generic_acpi.c, the current implementation has some gaps that are better fixed up in one go. The changes are to: * Follow x86 method of not adding PCI resources to PCI host bridge in ACPI code. This has been moved to pci_host_generic_acpi.c, where we walk thru its resources of the host bridge and add them. * Fixup code in pci_host_generic_acpi.c to read all decoded ranges and update the 'ranges' property. This allows us to share most of the code with generic implementation (and the FDT one). * Parse and setup IO ranges and bus ranges when walking the resources above. Drop most of the changes related to this from acpica code. * Add the ECAM memory area as mem resource 0. Implement the logic to get the ECAM area from MCFG (using bus range which we now decode), or from _CBA (using _BBN/bus range). Drop aarch64 ifdefs from acpica code which did part of this. * Switch resource activation to similar code as FDT implementation, this can be moved into generic implementation in a later pass. * Drop the mechanism of using the 7th bit of bus number as the domain, this is not correct and will work only in very specific cases. Use _SEG as PCI domain and use the bus ranges of the host bridge to provide start bus number. This commit should not make any functional change to dev/acpica/acpi.c for other architectures, almost all the changes there are to revert earlier additions in this file done for aarch64. Reviewed by: andrew Differential Revision: https://reviews.freebsd.org/D17791
2018-11-19 03:16:16 +00:00
}
Extend the support for PCI-e memory mapped configuration space access: - Rename pciereg_cfgopen() to pcie_cfgregopen() and expose it to the rest of the kernel. It now also accepts parameters via function arguments rather than global variables. - Add a notion of minimum and maximum bus numbers and reject requests for an out of range bus. - Add more range checks on slot/func/reg/bytes parameters to the cfg reg read/write routines. Don't panic on any invalid parameters, just fail the request (writes do nothing, reads return -1). This matches the behavior of the other cfg mechanisms. - Port the memory mapped configuration space access to amd64. On amd64 we simply use the direct map (via pmap_mapdev()) for the memory mapped window. - During acpi_attach() just after loading the ACPI tables, check for a MCFG table. If it exists, call pciereg_cfgopen() on each subtable (memory mapped window). For now we only support windows for domain 0 that start with bus 0. This removes the need for more chipset-specific quirks in the MD code. - Remove the chipset-specific quirks for the Intel 5000P/V/Z chipsets since these machines should all have MCFG tables via ACPI. - Updated pci_cfgregopen() to DTRT if ACPI had invoked pcie_cfgregopen() earlier. MFC after: 2 weeks
2008-08-22 02:14:23 +00:00
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
In total violation of at least 4 sections in the ACPI spec, some systems place device objects in \ (in this case, PCI links.) Work around this by starting our probe from \. To avoid attaching system scope objects, explicitly skip them. (I think it's an ACPI-CA bug that \_SB and \_TZ have device and thermal object types.) Thanks to pjd@ for testing. MFC after: 2 weeks
2005-01-08 09:10:20 +00:00
* Scan all of the ACPI namespace and attach child devices.
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*
In total violation of at least 4 sections in the ACPI spec, some systems place device objects in \ (in this case, PCI links.) Work around this by starting our probe from \. To avoid attaching system scope objects, explicitly skip them. (I think it's an ACPI-CA bug that \_SB and \_TZ have device and thermal object types.) Thanks to pjd@ for testing. MFC after: 2 weeks
2005-01-08 09:10:20 +00:00
* We should only expect to find devices in the \_PR, \_TZ, \_SI, and
* \_SB scopes, and \_PR and \_TZ became obsolete in the ACPI 2.0 spec.
* However, in violation of the spec, some systems place their PCI link
* devices in \, so we have to walk the whole namespace. We check the
* type of namespace nodes, so this should be ok.
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*/
static void
acpi_probe_children(device_t bus)
{
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
* Scan the namespace and insert placeholders for all the devices that
Call device_identify routines after doing the namespace walk. This is needed so that sysresource objects are created first to reserve all regions, then other devices can allocate from them. Otherwise, acpi_timer (the only ACPI device with an identify routine), would allocate its resources from the nexus, causing the later sysresource reserve to fail. Debugging by: Taku YAMAMOTO, Andrea Campi
2004-07-13 17:57:41 +00:00
* we find. We also probe/attach any early devices.
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*
* Note that we use AcpiWalkNamespace rather than AcpiGetDevices because
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
* we want to create nodes for all devices, not just those that are
* currently present. (This assumes that we don't want to create/remove
* devices as they appear, which might be smarter.)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*/
Updates to match the ACPI CA 20010816 import: - New debug macro (ACPI_DEBUG_PRINT), reducing debug-case code size. - New debug level/subsystem codes.
2001-08-26 22:50:15 +00:00
ACPI_DEBUG_PRINT((ACPI_DB_OBJECTS, "namespace scan\n"));
In total violation of at least 4 sections in the ACPI spec, some systems place device objects in \ (in this case, PCI links.) Work around this by starting our probe from \. To avoid attaching system scope objects, explicitly skip them. (I think it's an ACPI-CA bug that \_SB and \_TZ have device and thermal object types.) Thanks to pjd@ for testing. MFC after: 2 weeks
2005-01-08 09:10:20 +00:00
AcpiWalkNamespace(ACPI_TYPE_ANY, ACPI_ROOT_OBJECT, 100, acpi_probe_child,
Merge ACPICA 20091112.
2009-11-16 21:47:12 +00:00
NULL, bus, NULL);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Rework sysresource management. Instead of having each sysresource object hold its own values, pass them up to the parent (acpi0) and merge/uniq them on the way. After the namespace evaluation, acpi will reserve these resources and manage them via rman before bus_generic_probe() and bus_generic_attach(). This is necessary because some systems specify conflicting resources in separate sysresource objects. It's also cleaner in that the interface between sysresource and acpi is now merely the parent's resource list. This code handles the following cases: 1. Unique resource: add it to the parent via bus_set_resource(). 2. New wholly contained in old: discard new. 3. New tail overlaps old head: grow old head downward. AND/OR 4. New head overlaps old tail: grow old tail upward. Tested by: Pawel Worach <sajd_at_telia.com> Tested by: Radek Kozlowski <radek_at_raadradd.com> MFC after: 5 days
2004-08-23 16:28:42 +00:00
/* Pre-allocate resources for our rman from any sysresource devices. */
acpi_sysres_alloc(bus);
Use resource_list_reserve() to reserve I/O port and memory resources for ACPI devices even if they are not allocated by a device driver since the resources are in use and should not be allocated to another device.
2010-12-22 20:27:20 +00:00
/* Reserve resources already allocated to children. */
acpi_reserve_resources(bus);
Call device_identify routines after doing the namespace walk. This is needed so that sysresource objects are created first to reserve all regions, then other devices can allocate from them. Otherwise, acpi_timer (the only ACPI device with an identify routine), would allocate its resources from the nexus, causing the later sysresource reserve to fail. Debugging by: Taku YAMAMOTO, Andrea Campi
2004-07-13 17:57:41 +00:00
/* Create any static children by calling device identify methods. */
ACPI_DEBUG_PRINT((ACPI_DB_OBJECTS, "device identify routines\n"));
bus_generic_probe(bus);
Spelling fix.
2010-12-16 16:55:22 +00:00
/* Probe/attach all children, created statically and from the namespace. */
acpi: drop the second bus_generic_attach pass It is belived that that pass s not needed anymore. Specifically it is not required now for the reasons that were given in the removed comment. Discussed with: jhb MFC after: 4 weeks
2010-02-11 18:24:00 +00:00
ACPI_DEBUG_PRINT((ACPI_DB_OBJECTS, "acpi bus_generic_attach\n"));
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
bus_generic_attach(bus);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
/* Attach wake sysctls. */
Fix paste-o.
2004-05-28 06:29:30 +00:00
acpi_wake_sysctl_walk(bus);
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
ACPI_DEBUG_PRINT((ACPI_DB_OBJECTS, "done attaching children\n"));
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
return_VOID;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Clean up acpi_probe_order() a bit and clarify some comments.
2004-06-14 04:01:12 +00:00
/*
Revert back to probing Host-PCI bridges in the order we encounter them in the tree rather than sorting them by their address on PCI bus 0. Reported by: kan
2008-04-07 18:35:11 +00:00
* Determine the probe order for a given device.
Clean up acpi_probe_order() a bit and clarify some comments.
2004-06-14 04:01:12 +00:00
*/
Revert back to probing Host-PCI bridges in the order we encounter them in the tree rather than sorting them by their address on PCI bus 0. Reported by: kan
2008-04-07 18:35:11 +00:00
static void
Clean up acpi_probe_order() a bit and clarify some comments.
2004-06-14 04:01:12 +00:00
acpi_probe_order(ACPI_HANDLE handle, int *order)
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
{
- Give all clocks and timers on acpi0 the equal probing order. - Increase probing order for ECDT table to match HID-based probing. - Decrease probing order for HPET table to match HID-based probing. - Decrease probing order for CPUs and system resources. - Fix ACPI_DEV_BASE_ORDER to reflect the reality.
2012-02-07 20:54:44 +00:00
ACPI_OBJECT_TYPE type;
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
- Give all clocks and timers on acpi0 the equal probing order. - Increase probing order for ECDT table to match HID-based probing. - Decrease probing order for HPET table to match HID-based probing. - Decrease probing order for CPUs and system resources. - Fix ACPI_DEV_BASE_ORDER to reflect the reality.
2012-02-07 20:54:44 +00:00
/*
* 0. CPUs
* 1. I/O port and memory system resource holders
* 2. Clocks and timers (to handle early accesses)
* 3. Embedded controllers (to handle early accesses)
* 4. PCI Link Devices
*/
AcpiGetType(handle, &type);
if (type == ACPI_TYPE_PROCESSOR)
*order = 0;
else if (acpi_MatchHid(handle, "PNP0C01") ||
acpi_MatchHid(handle, "PNP0C02"))
*order = 1;
else if (acpi_MatchHid(handle, "PNP0100") ||
acpi_MatchHid(handle, "PNP0103") ||
acpi_MatchHid(handle, "PNP0B00"))
*order = 2;
else if (acpi_MatchHid(handle, "PNP0C09"))
*order = 3;
else if (acpi_MatchHid(handle, "PNP0C0F"))
*order = 4;
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
* Evaluate a child device and determine whether we might attach a device to
* it.
*/
static ACPI_STATUS
acpi_probe_child(ACPI_HANDLE handle, UINT32 level, void *context, void **status)
{
Add support to read the _CLS entry if it's present. It is used by memory-mapped devices that are normally PCIe drives. Devices can then use the existing pci_get_class, etc. accessors to query this data. The ivar values are different enough from the existing ACPI and ISA values to not conflict. Reviewed by: jhb Obtained from: ABT Systems Ltd Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D8721
2016-12-16 10:40:00 +00:00
ACPI_DEVICE_INFO *devinfo;
struct acpi_device *ad;
Merge ACPICA 20101209.
2010-12-15 23:48:45 +00:00
struct acpi_prw_data prw;
In total violation of at least 4 sections in the ACPI spec, some systems place device objects in \ (in this case, PCI links.) Work around this by starting our probe from \. To avoid attaching system scope objects, explicitly skip them. (I think it's an ACPI-CA bug that \_SB and \_TZ have device and thermal object types.) Thanks to pjd@ for testing. MFC after: 2 weeks
2005-01-08 09:10:20 +00:00
ACPI_OBJECT_TYPE type;
Import ACPI Dock Station support. Note that this is still very young. Additional detach implementaions (or maybe improvement) for other deivce drivers is required. Reviewed by: njl, imp MFC after: 1 week
2006-04-15 12:31:34 +00:00
ACPI_HANDLE h;
In total violation of at least 4 sections in the ACPI spec, some systems place device objects in \ (in this case, PCI links.) Work around this by starting our probe from \. To avoid attaching system scope objects, explicitly skip them. (I think it's an ACPI-CA bug that \_SB and \_TZ have device and thermal object types.) Thanks to pjd@ for testing. MFC after: 2 weeks
2005-01-08 09:10:20 +00:00
device_t bus, child;
Adjust a comment to clarify why \_SB_ and \_TZ_ are defined as device type in ACPICA. Reshuffle the code a bit to make sure this kludge only applies to these two specical cases and to make it cleaner.
2010-11-03 23:16:35 +00:00
char *handle_str;
Rework the previous change to honor MADT processor IDs when probing processor objects. Instead of forcing the new-bus CPU objects to use a unit number equal to pc_cpuid, adjust acpi_pcpu_get_id() to honor the MADT IDs by default. As with the previous change, setting debug.acpi.cpu_unordered to 1 in the loader will revert to the old behavior. Tested by: jimharris MFC after: 1 month
2012-05-23 13:45:52 +00:00
int order;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Adjust a comment to clarify why \_SB_ and \_TZ_ are defined as device type in ACPICA. Reshuffle the code a bit to make sure this kludge only applies to these two specical cases and to make it cleaner.
2010-11-03 23:16:35 +00:00
if (acpi_disabled("children"))
return_ACPI_STATUS (AE_OK);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Skip this device if we think we'll have trouble with it. */
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
if (acpi_avoid(handle))
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return_ACPI_STATUS (AE_OK);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
bus = (device_t)context;
Match namespace cleanup changes in ACPI CA 20020217 update. Use ACPI_SUCCESS/ACPI_FAILURE consistently. The AcpiGetInto* interfaces are obsoleted by ACPI_ALLOCATE_BUFFER. Add AcpiBatteryIsPresent helper to determine whether a battery device is inserted. Add ACPI_ALL_DRIVERS to the list of debug layers, now that we own the namespace for this. Pr:
2002-02-23 05:21:56 +00:00
if (ACPI_SUCCESS(AcpiGetType(handle, &type))) {
Adjust a comment to clarify why \_SB_ and \_TZ_ are defined as device type in ACPICA. Reshuffle the code a bit to make sure this kludge only applies to these two specical cases and to make it cleaner.
2010-11-03 23:16:35 +00:00
handle_str = acpi_name(handle);
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
switch (type) {
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
case ACPI_TYPE_DEVICE:
In total violation of at least 4 sections in the ACPI spec, some systems place device objects in \ (in this case, PCI links.) Work around this by starting our probe from \. To avoid attaching system scope objects, explicitly skip them. (I think it's an ACPI-CA bug that \_SB and \_TZ have device and thermal object types.) Thanks to pjd@ for testing. MFC after: 2 weeks
2005-01-08 09:10:20 +00:00
/*
* Since we scan from \, be sure to skip system scope objects.
Adjust a comment to clarify why \_SB_ and \_TZ_ are defined as device type in ACPICA. Reshuffle the code a bit to make sure this kludge only applies to these two specical cases and to make it cleaner.
2010-11-03 23:16:35 +00:00
* \_SB_ and \_TZ_ are defined in ACPICA as devices to work around
Consistently use padding `_' in the comment.
2010-11-08 19:52:14 +00:00
* BIOS bugs. For example, \_SB_ is to allow \_SB_._INI to be run
sys/dev: minor spelling fixes. Most affect comments, very few have user-visible effects.
2016-05-03 03:41:25 +00:00
* during the initialization and \_TZ_ is to support Notify() on it.
In total violation of at least 4 sections in the ACPI spec, some systems place device objects in \ (in this case, PCI links.) Work around this by starting our probe from \. To avoid attaching system scope objects, explicitly skip them. (I think it's an ACPI-CA bug that \_SB and \_TZ have device and thermal object types.) Thanks to pjd@ for testing. MFC after: 2 weeks
2005-01-08 09:10:20 +00:00
*/
Adjust a comment to clarify why \_SB_ and \_TZ_ are defined as device type in ACPICA. Reshuffle the code a bit to make sure this kludge only applies to these two specical cases and to make it cleaner.
2010-11-03 23:16:35 +00:00
if (strcmp(handle_str, "\\_SB_") == 0 ||
strcmp(handle_str, "\\_TZ_") == 0)
In total violation of at least 4 sections in the ACPI spec, some systems place device objects in \ (in this case, PCI links.) Work around this by starting our probe from \. To avoid attaching system scope objects, explicitly skip them. (I think it's an ACPI-CA bug that \_SB and \_TZ have device and thermal object types.) Thanks to pjd@ for testing. MFC after: 2 weeks
2005-01-08 09:10:20 +00:00
break;
Merge ACPICA 20101209.
2010-12-15 23:48:45 +00:00
if (acpi_parse_prw(handle, &prw) == 0)
AcpiSetupGpeForWake(handle, prw.gpe_handle, prw.gpe_bit);
Don't create a device_t object or parse current resources (via _CRS) for ACPI Device() objects that do not have any device IDs available via the _HID or _CID methods. Without a device ID a device driver cannot attach to the device anyway. Namespace objects that are devices but not of type ACPI_TYPE_DEVICE are not affected. A few BIOSes have also attached a _CRS method to a PCI device to allocate resources that are not managed via a BAR. With the previous code those resources are allocated from acpi0 directly which can interfere with the new PCI-PCI bridge driver (since the PCI device in question may be behind a bridge and its resources should be allocated from that bridge's windows instead). The resources were also orphaned and and would end up associated with some other random device whose device_t reused the pointer of the original ACPI-enumerated device (after it was free'd by the ACPI PCI bus driver) in devinfo output which was confusing. If we want to handle _CRS on PCI devices we can adjust the ACPI PCI bus driver to do that in the future and associate the resources with the proper device object respecting PCI-PCI bridges, etc. Note that with this change the ACPI PCI bus driver no longer has to delete ACPI-enumerated device_t devices that mirror PCI devices since they should in general not exist. There are rare cases when a BIOS will give a PCI device a _HID (e.g. I've seen a PCI-ISA bridge given a _HID for a system resource device). In that case we leave both the ACPI and PCI-enumerated device_t objects around just as in the previous code.
2011-06-17 21:19:01 +00:00
/*
* Ignore devices that do not have a _HID or _CID. They should
* be discovered by other buses (e.g. the PCI bus driver).
*/
if (!acpi_has_hid(handle))
break;
Adjust a comment to clarify why \_SB_ and \_TZ_ are defined as device type in ACPICA. Reshuffle the code a bit to make sure this kludge only applies to these two specical cases and to make it cleaner.
2010-11-03 23:16:35 +00:00
/* FALLTHROUGH */
case ACPI_TYPE_PROCESSOR:
case ACPI_TYPE_THERMAL:
case ACPI_TYPE_POWER:
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
Probe CPUs after the PCI hierarchy on i386, amd64, and ia64. This allows the cpufreq drivers to reliably use properties of PCI devices for quirks, etc. - For the legacy drivers, add CPU devices via an identify routine in the CPU driver itself rather than in the legacy driver's attach routine. - Add CPU devices after Host-PCI bridges in the acpi bus driver. - Change the ichss(4) driver to use pci_find_bsf() to locate the ICH and check its device ID rather than having a bogus PCI attachment that only checked for the ID in probe and always failed. As a side effect, you can now kldload ichss after boot. - Fix the ichss(4) driver to use the correct device_t for the ICH (and not for ichss0) when doing PCI config space operations to enable SpeedStep. MFC after: 2 weeks Reviewed by: njl, Andriy Gapon avg of icyb.net.ua
2008-03-10 22:18:07 +00:00
* Create a placeholder device for this node. Sort the
* placeholder so that the probe/attach passes will run
* breadth-first. Orders less than ACPI_DEV_BASE_ORDER
* are reserved for special objects (i.e., system
acpi: update stale comments about order of cpu devices probing These comments should have been updated in r203776 when the order was changed. Pointyhat to: avg MFC after: 3 days
2010-09-03 08:07:04 +00:00
* resources).
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*/
In total violation of at least 4 sections in the ACPI spec, some systems place device objects in \ (in this case, PCI links.) Work around this by starting our probe from \. To avoid attaching system scope objects, explicitly skip them. (I think it's an ACPI-CA bug that \_SB and \_TZ have device and thermal object types.) Thanks to pjd@ for testing. MFC after: 2 weeks
2005-01-08 09:10:20 +00:00
ACPI_DEBUG_PRINT((ACPI_DB_OBJECTS, "scanning '%s'\n", handle_str));
- Give all clocks and timers on acpi0 the equal probing order. - Increase probing order for ECDT table to match HID-based probing. - Decrease probing order for HPET table to match HID-based probing. - Decrease probing order for CPUs and system resources. - Fix ACPI_DEV_BASE_ORDER to reflect the reality.
2012-02-07 20:54:44 +00:00
order = level * 10 + ACPI_DEV_BASE_ORDER;
Don't attach special devices in the order they appear in the AML tree. If the embedded controller exists before the sysresource devices, for example, it will be attached first. Instead, let the normal device order function work as we first desired. [1] There still remained a problem where we couldn't allocate resources in acpi0 that were passed up by the sysresource pseudo-devices. These devices had to probe/attach first to give their resources to acpi, then acpi would allocate them before probing/attaching other devices. To work around this, we attach them from acpi_sysres_alloc(). A better approach would be to implement multi-pass probe/attach in newbus but that's a much bigger task. Suggested by: jhb [1] Hardware from: Centaur Technologies MFC after: 1 week
2006-05-07 03:28:10 +00:00
acpi_probe_order(handle, &order);
Rework the previous change to honor MADT processor IDs when probing processor objects. Instead of forcing the new-bus CPU objects to use a unit number equal to pc_cpuid, adjust acpi_pcpu_get_id() to honor the MADT IDs by default. As with the previous change, setting debug.acpi.cpu_unordered to 1 in the loader will revert to the old behavior. Tested by: jimharris MFC after: 1 month
2012-05-23 13:45:52 +00:00
child = BUS_ADD_CHILD(bus, order, NULL, -1);
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
if (child == NULL)
break;
Associate a device_t with an ACPI_HANDLE. This make AcpiWalkNamespace more useful. If ACPI-CA allowed null object handlers, we wouldn't need the placeholder function.
2004-06-13 17:29:35 +00:00
/* Associate the handle with the device_t and vice versa. */
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
acpi_set_handle(child, handle);
Associate a device_t with an ACPI_HANDLE. This make AcpiWalkNamespace more useful. If ACPI-CA allowed null object handlers, we wouldn't need the placeholder function.
2004-06-13 17:29:35 +00:00
AcpiAttachData(handle, acpi_fake_objhandler, child);
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
Disable devices that are not present; at a later stage we can then enable them if the device arrives. This should solve the problem where devices that have been disabled eg. in the BIOS show up with nonsense resources and hang the bootstrap process.
2001-09-15 04:14:31 +00:00
/*
* Check that the device is present. If it's not present,
* leave it disabled (so that we have a device_t attached to
* the handle, but we don't probe it).
Add a hack to get around PCI link devices that report "present" but not "functional" (i.e., if they are disabled). We should probe them anyway since we may enable them later. Tested by: thompsa MFC after: 3 days
2005-10-23 00:28:39 +00:00
*
* XXX PCI link devices sometimes report "present" but not
* "functional" (i.e. if disabled). Go ahead and probe them
* anyway since we may enable them later.
Disable devices that are not present; at a later stage we can then enable them if the device arrives. This should solve the problem where devices that have been disabled eg. in the BIOS show up with nonsense resources and hang the bootstrap process.
2001-09-15 04:14:31 +00:00
*/
Import ACPI Dock Station support. Note that this is still very young. Additional detach implementaions (or maybe improvement) for other deivce drivers is required. Reviewed by: njl, imp MFC after: 1 week
2006-04-15 12:31:34 +00:00
if (type == ACPI_TYPE_DEVICE && !acpi_DeviceIsPresent(child)) {
/* Never disable PCI link devices. */
if (acpi_MatchHid(handle, "PNP0C0F"))
break;
/*
* Docking stations should remain enabled since the system
* may be undocked at boot.
*/
if (ACPI_SUCCESS(AcpiGetHandle(handle, "_DCK", &h)))
break;
Disable devices that are not present; at a later stage we can then enable them if the device arrives. This should solve the problem where devices that have been disabled eg. in the BIOS show up with nonsense resources and hang the bootstrap process.
2001-09-15 04:14:31 +00:00
device_disable(child);
break;
}
Add support for attaching PnP-aware ISA drivers to ACPI. Always parse ACPI device resource settings (current resources only) and attach the resources to the device before probe/attaching.
2001-08-30 00:50:58 +00:00
/*
* Get the device's resource settings and attach them.
* Note that if the device has _PRS but no _CRS, we need
* to decide when it's appropriate to try to configure the
* device. Ignore the return value here; it's OK for the
* device not to have any resources.
*/
Add an interface to pass an argument to the resource parsing functions. This is just groundwork for changing sysresource behavior. PR: Submitted by: Reviewed by: Approved by: Obtained from: MFC after:
2004-03-31 17:23:46 +00:00
acpi_parse_resources(child, handle, &acpi_res_parse_set, NULL);
Add support to read the _CLS entry if it's present. It is used by memory-mapped devices that are normally PCIe drives. Devices can then use the existing pci_get_class, etc. accessors to query this data. The ivar values are different enough from the existing ACPI and ISA values to not conflict. Reviewed by: jhb Obtained from: ABT Systems Ltd Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D8721
2016-12-16 10:40:00 +00:00
ad = device_get_ivars(child);
ad->ad_cls_class = 0xffffff;
if (ACPI_SUCCESS(AcpiGetObjectInfo(handle, &devinfo))) {
if ((devinfo->Valid & ACPI_VALID_CLS) != 0 &&
devinfo->ClassCode.Length >= ACPI_PCICLS_STRING_SIZE) {
ad->ad_cls_class = strtoul(devinfo->ClassCode.String,
NULL, 16);
}
AcpiOsFree(devinfo);
}
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
break;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
}
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return_ACPI_STATUS (AE_OK);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Associate a device_t with an ACPI_HANDLE. This make AcpiWalkNamespace more useful. If ACPI-CA allowed null object handlers, we wouldn't need the placeholder function.
2004-06-13 17:29:35 +00:00
/*
* AcpiAttachData() requires an object handler but never uses it. This is a
* placeholder object handler so we can store a device_t in an ACPI_HANDLE.
*/
void
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
acpi_fake_objhandler(ACPI_HANDLE h, void *data)
Associate a device_t with an ACPI_HANDLE. This make AcpiWalkNamespace more useful. If ACPI-CA allowed null object handlers, we wouldn't need the placeholder function.
2004-06-13 17:29:35 +00:00
{
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
static void
acpi_shutdown_final(void *arg, int howto)
{
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
struct acpi_softc *sc = (struct acpi_softc *)arg;
Restore interrupt state after executing AcpiEnterSleepState().
2012-03-27 23:26:58 +00:00
register_t intr;
By default, don't disable ACPI during reboot. This appears to hang some systems. Introduce a new sysctl "hw.acpi.disable_on_reboot" that allows users to re-enable the old behavior in case it's needed for some systems. We never disable in the power-off path. Original approach submitted by Alexander Logvinov <abuse@akavia.ru> with reworking by Jung-uk Kim and myself.
2006-06-11 20:31:41 +00:00
ACPI_STATUS status;
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
Move the poweroff handler to a separate function. Make sure it is run on the boot processor (cpuid == 0). Some chipsets do not power off the system if the shutdown handler runs on an AP.
2004-03-19 07:05:01 +00:00
/*
Run the power off code directly instead of using indirection through smp_rendezvous() to ensure we run on the BSP. This reverts rev 1.128. Add a comment indicating that MI code should be the one that runs all shutdown functions on the BSP with the APs halted. This should work around problems in power off while waiting for the MI code to be improved.
2004-06-24 00:48:45 +00:00
* XXX Shutdown code should only run on the BSP (cpuid 0).
* Some chipsets do not power off the system correctly if called from
* an AP.
Move the poweroff handler to a separate function. Make sure it is run on the boot processor (cpuid == 0). Some chipsets do not power off the system if the shutdown handler runs on an AP.
2004-03-19 07:05:01 +00:00
*/
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
if ((howto & RB_POWEROFF) != 0) {
Move the call to AcpiEnterSleepStatePrep() to before we select the BSP (cpuid 0) as the processor. It mallocs some data and smp_rendezvous calls functions with locks held.
2004-04-26 05:25:06 +00:00
status = AcpiEnterSleepStatePrep(ACPI_STATE_S5);
if (ACPI_FAILURE(status)) {
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(sc->acpi_dev, "AcpiEnterSleepStatePrep failed - %s\n",
AcpiFormatException(status));
Move the call to AcpiEnterSleepStatePrep() to before we select the BSP (cpuid 0) as the processor. It mallocs some data and smp_rendezvous calls functions with locks held.
2004-04-26 05:25:06 +00:00
return;
}
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(sc->acpi_dev, "Powering system off\n");
Restore interrupt state after executing AcpiEnterSleepState().
2012-03-27 23:26:58 +00:00
intr = intr_disable();
Merge ACPICA 20120816.
2012-08-16 20:54:52 +00:00
status = AcpiEnterSleepState(ACPI_STATE_S5);
Restore interrupt state after executing AcpiEnterSleepState().
2012-03-27 23:26:58 +00:00
if (ACPI_FAILURE(status)) {
intr_restore(intr);
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(sc->acpi_dev, "power-off failed - %s\n",
AcpiFormatException(status));
Restore interrupt state after executing AcpiEnterSleepState().
2012-03-27 23:26:58 +00:00
} else {
Run the power off code directly instead of using indirection through smp_rendezvous() to ensure we run on the BSP. This reverts rev 1.128. Add a comment indicating that MI code should be the one that runs all shutdown functions on the BSP with the APs halted. This should work around problems in power off while waiting for the MI code to be improved.
2004-06-24 00:48:45 +00:00
DELAY(1000000);
Restore interrupt state after executing AcpiEnterSleepState().
2012-03-27 23:26:58 +00:00
intr_restore(intr);
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(sc->acpi_dev, "power-off failed - timeout\n");
Run the power off code directly instead of using indirection through smp_rendezvous() to ensure we run on the BSP. This reverts rev 1.128. Add a comment indicating that MI code should be the one that runs all shutdown functions on the BSP with the APs halted. This should work around problems in power off while waiting for the MI code to be improved.
2004-06-24 00:48:45 +00:00
}
Use AcpiReset() from ACPICA instead of rolling our own, which is actually incomplete. If FADT says the register is available, enable the capability by default. Remove the previous default value from acpi(4).
2010-10-13 00:21:53 +00:00
} else if ((howto & RB_HALT) == 0 && sc->acpi_handle_reboot) {
By default, don't disable ACPI during reboot. This appears to hang some systems. Introduce a new sysctl "hw.acpi.disable_on_reboot" that allows users to re-enable the old behavior in case it's needed for some systems. We never disable in the power-off path. Original approach submitted by Alexander Logvinov <abuse@akavia.ru> with reworking by Jung-uk Kim and myself.
2006-06-11 20:31:41 +00:00
/* Reboot using the reset register. */
Use AcpiReset() from ACPICA instead of rolling our own, which is actually incomplete. If FADT says the register is available, enable the capability by default. Remove the previous default value from acpi(4).
2010-10-13 00:21:53 +00:00
status = AcpiReset();
if (ACPI_SUCCESS(status)) {
Add reset register support. This is the only method to reboot some new systems (blade servers). On most systems, this is implemented as an IO write to the SMI port and the BIOS generates the actual reset. PR: kern/94939 Submitted by: dodell@ixsystems.com Reviewed by: jhb MFC after: 3 weeks
2006-03-29 06:30:47 +00:00
DELAY(1000000);
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(sc->acpi_dev, "reset failed - timeout\n");
Use AcpiReset() from ACPICA instead of rolling our own, which is actually incomplete. If FADT says the register is available, enable the capability by default. Remove the previous default value from acpi(4).
2010-10-13 00:21:53 +00:00
} else if (status != AE_NOT_EXIST)
device_printf(sc->acpi_dev, "reset failed - %s\n",
AcpiFormatException(status));
Add KERNEL_PANICKED macro for use in place of direct panicstr tests
2020-01-12 06:07:54 +00:00
} else if (sc->acpi_do_disable && !KERNEL_PANICKED()) {
By default, don't disable ACPI during reboot. This appears to hang some systems. Introduce a new sysctl "hw.acpi.disable_on_reboot" that allows users to re-enable the old behavior in case it's needed for some systems. We never disable in the power-off path. Original approach submitted by Alexander Logvinov <abuse@akavia.ru> with reworking by Jung-uk Kim and myself.
2006-06-11 20:31:41 +00:00
/*
* Only disable ACPI if the user requested. On some systems, writing
* the disable value to SMI_CMD hangs the system.
*/
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(sc->acpi_dev, "Shutting down\n");
Terminate ACPI subsystem on reboot. Some machines stuck on reboot if ACPI sleep/wakeup was executed.
2002-05-19 15:12:40 +00:00
AcpiTerminate();
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
}
Enable fixed event at not only boot but also wakeup. Reported and patch tested by: Peter Dufault <dufault@hda.hda.com>
2001-01-10 18:01:51 +00:00
static void
acpi_enable_fixed_events(struct acpi_softc *sc)
{
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
static int first_time = 1;
Enable fixed event at not only boot but also wakeup. Reported and patch tested by: Peter Dufault <dufault@hda.hda.com>
2001-01-10 18:01:51 +00:00
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
/* Enable and clear fixed events and install handlers. */
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
if ((AcpiGbl_FADT.Flags & ACPI_FADT_POWER_BUTTON) == 0) {
Fix a couple changes that were incorrect in updating for 0619. Only unlock the hardware mutex if it is held. Re-add calls to Enable/Clear fixed events. This is not known to have caused problems. Bug symptoms might have included instability after an aborted suspend attempt or power/sleep buttons not being enabled.
2003-08-15 02:10:38 +00:00
AcpiClearEvent(ACPI_EVENT_POWER_BUTTON);
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
AcpiInstallFixedEventHandler(ACPI_EVENT_POWER_BUTTON,
Prefer buttons defined in the AML over the ones in the FADT. Some systems define power/sleep buttons in both places but only deliver notifies to the ones defined in the AML. Also, reduce length of various function handler names. PR: Submitted by: Reviewed by: Approved by: Obtained from: MFC after:
2004-02-11 02:57:33 +00:00
acpi_event_power_button_sleep, sc);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
if (first_time)
Add necessary newlines.
2003-09-18 05:12:45 +00:00
device_printf(sc->acpi_dev, "Power Button (fixed)\n");
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
}
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
if ((AcpiGbl_FADT.Flags & ACPI_FADT_SLEEP_BUTTON) == 0) {
Fix a couple changes that were incorrect in updating for 0619. Only unlock the hardware mutex if it is held. Re-add calls to Enable/Clear fixed events. This is not known to have caused problems. Bug symptoms might have included instability after an aborted suspend attempt or power/sleep buttons not being enabled.
2003-08-15 02:10:38 +00:00
AcpiClearEvent(ACPI_EVENT_SLEEP_BUTTON);
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
AcpiInstallFixedEventHandler(ACPI_EVENT_SLEEP_BUTTON,
Prefer buttons defined in the AML over the ones in the FADT. Some systems define power/sleep buttons in both places but only deliver notifies to the ones defined in the AML. Also, reduce length of various function handler names. PR: Submitted by: Reviewed by: Approved by: Obtained from: MFC after:
2004-02-11 02:57:33 +00:00
acpi_event_sleep_button_sleep, sc);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
if (first_time)
Add necessary newlines.
2003-09-18 05:12:45 +00:00
device_printf(sc->acpi_dev, "Sleep Button (fixed)\n");
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
}
Enable fixed event at not only boot but also wakeup. Reported and patch tested by: Peter Dufault <dufault@hda.hda.com>
2001-01-10 18:01:51 +00:00
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
first_time = 0;
Enable fixed event at not only boot but also wakeup. Reported and patch tested by: Peter Dufault <dufault@hda.hda.com>
2001-01-10 18:01:51 +00:00
}
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
/*
* Returns true if the device is actually present and should
* be attached to. This requires the present, enabled, UI-visible
* and diagnostics-passed bits to be set.
*/
BOOLEAN
acpi_DeviceIsPresent(device_t dev)
{
Merge ACPICA 20180313.
2018-03-14 23:45:48 +00:00
ACPI_HANDLE h;
UINT32 s;
ACPI_STATUS status;
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
Merge ACPICA 20180313.
2018-03-14 23:45:48 +00:00
h = acpi_get_handle(dev);
if (h == NULL)
return (FALSE);
add snps IP uart support / genaralize UART This is an amalgam of a patch by Doug Ambrisko to generalize uart_acpi_find_device, imp moving the ACPI table to uart_dev_ns8250.c and advice by jhb to work around a bug in the EPYC 3151 BIOS (the BIOS incorrectly marks the serial ports as disabled) Reviewed by: imp MFC after: 8 weeks Differential Revision: https://reviews.freebsd.org/D16432
2018-08-19 21:10:21 +00:00
/*
Update comment for AMI00[12]0 override. The AML is even stupider than always returning 0. It will only return non-zero for an OS that reports itself as "Windows 2015", at least on the Threadripper board's AML that I've examined. Those AMLs also suggest we may need this quirk for AMI0030 as well. There may be other cases where we need to override the _STA in a generic way, so we should consider writing code to do that.
2018-10-25 17:17:11 +00:00
* Certain Treadripper boards always returns 0 for FreeBSD because it
* only returns non-zero for the OS string "Windows 2015". Otherwise it
* will return zero. Force them to always be treated as present.
* Beata versions were worse: they always returned 0.
add snps IP uart support / genaralize UART This is an amalgam of a patch by Doug Ambrisko to generalize uart_acpi_find_device, imp moving the ACPI table to uart_dev_ns8250.c and advice by jhb to work around a bug in the EPYC 3151 BIOS (the BIOS incorrectly marks the serial ports as disabled) Reviewed by: imp MFC after: 8 weeks Differential Revision: https://reviews.freebsd.org/D16432
2018-08-19 21:10:21 +00:00
*/
if (acpi_MatchHid(h, "AMDI0020") || acpi_MatchHid(h, "AMDI0010"))
return (TRUE);
acpi_{Device,Battery}IsPresent: restore pre-r330957 behaviour Specifically, assume that the device is present if evaluation of _STA method fails. Before r330957 we ignored any _STA evaluation failure (which was performed by AcpiGetObjectInfo in ACPICA contrib code) for the purpose of acpi_DeviceIsPresent and acpi_BatteryIsPresent. ACPICA 20180313 removed evaluation of _STA from AcpiGetObjectInfo. So, we added evaluation of _STA to acpi_DeviceIsPresent and acpi_BatteryIsPresent. One important difference is that the new code ignored a failure only if _STA did not exist (AE_NOT_FOUND). Any other kind of failure was treated as a fatal failure. Apparently, on some systems we can get AE_NOT_EXIST when evaluating _STA. And that error is not an evil twin of AE_NOT_FOUND, despite a very similar name, but a distinct error related to a missing handler for an ACPI operation region. It's possible that for some people the problem was already fixed by changes in ACPICA and/or in acpi_ec driver (or even in BIOS) that fixed the AE_NOT_EXIST failure related to EC operation region. This work is based on a great analysis by cem and an earlier patch by Ali Abdallah <aliovx@gmail.com>. PR: 227191 Reported by: 0mp MFC after: 2 weeks
2018-12-06 12:34:34 +00:00
status = acpi_GetInteger(h, "_STA", &s);
/*
* If no _STA method or if it failed, then assume that
* the device is present.
*/
Merge ACPICA 20180313.
2018-03-14 23:45:48 +00:00
if (ACPI_FAILURE(status))
acpi_{Device,Battery}IsPresent: restore pre-r330957 behaviour Specifically, assume that the device is present if evaluation of _STA method fails. Before r330957 we ignored any _STA evaluation failure (which was performed by AcpiGetObjectInfo in ACPICA contrib code) for the purpose of acpi_DeviceIsPresent and acpi_BatteryIsPresent. ACPICA 20180313 removed evaluation of _STA from AcpiGetObjectInfo. So, we added evaluation of _STA to acpi_DeviceIsPresent and acpi_BatteryIsPresent. One important difference is that the new code ignored a failure only if _STA did not exist (AE_NOT_FOUND). Any other kind of failure was treated as a fatal failure. Apparently, on some systems we can get AE_NOT_EXIST when evaluating _STA. And that error is not an evil twin of AE_NOT_FOUND, despite a very similar name, but a distinct error related to a missing handler for an ACPI operation region. It's possible that for some people the problem was already fixed by changes in ACPICA and/or in acpi_ec driver (or even in BIOS) that fixed the AE_NOT_EXIST failure related to EC operation region. This work is based on a great analysis by cem and an earlier patch by Ali Abdallah <aliovx@gmail.com>. PR: 227191 Reported by: 0mp MFC after: 2 weeks
2018-12-06 12:34:34 +00:00
return (TRUE);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
Merge ACPICA 20180313.
2018-03-14 23:45:48 +00:00
return (ACPI_DEVICE_PRESENT(s) ? TRUE : FALSE);
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
}
Match namespace cleanup changes in ACPI CA 20020217 update. Use ACPI_SUCCESS/ACPI_FAILURE consistently. The AcpiGetInto* interfaces are obsoleted by ACPI_ALLOCATE_BUFFER. Add AcpiBatteryIsPresent helper to determine whether a battery device is inserted. Add ACPI_ALL_DRIVERS to the list of debug layers, now that we own the namespace for this. Pr:
2002-02-23 05:21:56 +00:00
/*
* Returns true if the battery is actually present and inserted.
*/
BOOLEAN
acpi_BatteryIsPresent(device_t dev)
{
Merge ACPICA 20180313.
2018-03-14 23:45:48 +00:00
ACPI_HANDLE h;
UINT32 s;
ACPI_STATUS status;
Match namespace cleanup changes in ACPI CA 20020217 update. Use ACPI_SUCCESS/ACPI_FAILURE consistently. The AcpiGetInto* interfaces are obsoleted by ACPI_ALLOCATE_BUFFER. Add AcpiBatteryIsPresent helper to determine whether a battery device is inserted. Add ACPI_ALL_DRIVERS to the list of debug layers, now that we own the namespace for this. Pr:
2002-02-23 05:21:56 +00:00
Merge ACPICA 20180313.
2018-03-14 23:45:48 +00:00
h = acpi_get_handle(dev);
if (h == NULL)
return (FALSE);
status = acpi_GetInteger(h, "_STA", &s);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
acpi_{Device,Battery}IsPresent: restore pre-r330957 behaviour Specifically, assume that the device is present if evaluation of _STA method fails. Before r330957 we ignored any _STA evaluation failure (which was performed by AcpiGetObjectInfo in ACPICA contrib code) for the purpose of acpi_DeviceIsPresent and acpi_BatteryIsPresent. ACPICA 20180313 removed evaluation of _STA from AcpiGetObjectInfo. So, we added evaluation of _STA to acpi_DeviceIsPresent and acpi_BatteryIsPresent. One important difference is that the new code ignored a failure only if _STA did not exist (AE_NOT_FOUND). Any other kind of failure was treated as a fatal failure. Apparently, on some systems we can get AE_NOT_EXIST when evaluating _STA. And that error is not an evil twin of AE_NOT_FOUND, despite a very similar name, but a distinct error related to a missing handler for an ACPI operation region. It's possible that for some people the problem was already fixed by changes in ACPICA and/or in acpi_ec driver (or even in BIOS) that fixed the AE_NOT_EXIST failure related to EC operation region. This work is based on a great analysis by cem and an earlier patch by Ali Abdallah <aliovx@gmail.com>. PR: 227191 Reported by: 0mp MFC after: 2 weeks
2018-12-06 12:34:34 +00:00
/*
* If no _STA method or if it failed, then assume that
* the device is present.
*/
Merge ACPICA 20180313.
2018-03-14 23:45:48 +00:00
if (ACPI_FAILURE(status))
acpi_{Device,Battery}IsPresent: restore pre-r330957 behaviour Specifically, assume that the device is present if evaluation of _STA method fails. Before r330957 we ignored any _STA evaluation failure (which was performed by AcpiGetObjectInfo in ACPICA contrib code) for the purpose of acpi_DeviceIsPresent and acpi_BatteryIsPresent. ACPICA 20180313 removed evaluation of _STA from AcpiGetObjectInfo. So, we added evaluation of _STA to acpi_DeviceIsPresent and acpi_BatteryIsPresent. One important difference is that the new code ignored a failure only if _STA did not exist (AE_NOT_FOUND). Any other kind of failure was treated as a fatal failure. Apparently, on some systems we can get AE_NOT_EXIST when evaluating _STA. And that error is not an evil twin of AE_NOT_FOUND, despite a very similar name, but a distinct error related to a missing handler for an ACPI operation region. It's possible that for some people the problem was already fixed by changes in ACPICA and/or in acpi_ec driver (or even in BIOS) that fixed the AE_NOT_EXIST failure related to EC operation region. This work is based on a great analysis by cem and an earlier patch by Ali Abdallah <aliovx@gmail.com>. PR: 227191 Reported by: 0mp MFC after: 2 weeks
2018-12-06 12:34:34 +00:00
return (TRUE);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
Merge ACPICA 20180313.
2018-03-14 23:45:48 +00:00
return (ACPI_BATTERY_PRESENT(s) ? TRUE : FALSE);
Match namespace cleanup changes in ACPI CA 20020217 update. Use ACPI_SUCCESS/ACPI_FAILURE consistently. The AcpiGetInto* interfaces are obsoleted by ACPI_ALLOCATE_BUFFER. Add AcpiBatteryIsPresent helper to determine whether a battery device is inserted. Add ACPI_ALL_DRIVERS to the list of debug layers, now that we own the namespace for this. Pr:
2002-02-23 05:21:56 +00:00
}
Don't create a device_t object or parse current resources (via _CRS) for ACPI Device() objects that do not have any device IDs available via the _HID or _CID methods. Without a device ID a device driver cannot attach to the device anyway. Namespace objects that are devices but not of type ACPI_TYPE_DEVICE are not affected. A few BIOSes have also attached a _CRS method to a PCI device to allocate resources that are not managed via a BAR. With the previous code those resources are allocated from acpi0 directly which can interfere with the new PCI-PCI bridge driver (since the PCI device in question may be behind a bridge and its resources should be allocated from that bridge's windows instead). The resources were also orphaned and and would end up associated with some other random device whose device_t reused the pointer of the original ACPI-enumerated device (after it was free'd by the ACPI PCI bus driver) in devinfo output which was confusing. If we want to handle _CRS on PCI devices we can adjust the ACPI PCI bus driver to do that in the future and associate the resources with the proper device object respecting PCI-PCI bridges, etc. Note that with this change the ACPI PCI bus driver no longer has to delete ACPI-enumerated device_t devices that mirror PCI devices since they should in general not exist. There are rare cases when a BIOS will give a PCI device a _HID (e.g. I've seen a PCI-ISA bridge given a _HID for a system resource device). In that case we leave both the ACPI and PCI-enumerated device_t objects around just as in the previous code.
2011-06-17 21:19:01 +00:00
/*
* Returns true if a device has at least one valid device ID.
*/
acpi: Export functions required by upcoming acpi_iicbus driver.
2020-03-09 20:28:45 +00:00
BOOLEAN
Don't create a device_t object or parse current resources (via _CRS) for ACPI Device() objects that do not have any device IDs available via the _HID or _CID methods. Without a device ID a device driver cannot attach to the device anyway. Namespace objects that are devices but not of type ACPI_TYPE_DEVICE are not affected. A few BIOSes have also attached a _CRS method to a PCI device to allocate resources that are not managed via a BAR. With the previous code those resources are allocated from acpi0 directly which can interfere with the new PCI-PCI bridge driver (since the PCI device in question may be behind a bridge and its resources should be allocated from that bridge's windows instead). The resources were also orphaned and and would end up associated with some other random device whose device_t reused the pointer of the original ACPI-enumerated device (after it was free'd by the ACPI PCI bus driver) in devinfo output which was confusing. If we want to handle _CRS on PCI devices we can adjust the ACPI PCI bus driver to do that in the future and associate the resources with the proper device object respecting PCI-PCI bridges, etc. Note that with this change the ACPI PCI bus driver no longer has to delete ACPI-enumerated device_t devices that mirror PCI devices since they should in general not exist. There are rare cases when a BIOS will give a PCI device a _HID (e.g. I've seen a PCI-ISA bridge given a _HID for a system resource device). In that case we leave both the ACPI and PCI-enumerated device_t objects around just as in the previous code.
2011-06-17 21:19:01 +00:00
acpi_has_hid(ACPI_HANDLE h)
{
ACPI_DEVICE_INFO *devinfo;
BOOLEAN ret;
if (h == NULL ||
ACPI_FAILURE(AcpiGetObjectInfo(h, &devinfo)))
return (FALSE);
ret = FALSE;
if ((devinfo->Valid & ACPI_VALID_HID) != 0)
ret = TRUE;
else if ((devinfo->Valid & ACPI_VALID_CID) != 0)
if (devinfo->CompatibleIdList.Count > 0)
ret = TRUE;
AcpiOsFree(devinfo);
return (ret);
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
Add support to ACPI to manage its own resources. Previously, resource allocation was passed up to nexus. Now, we probe sysresource objects and manage the resources they describe in a local rman pool. This helps devices which attach/detach varying resources (like the _CST object) and module loads/unloads. The allocation/release routines now check to see if the resource is described in a child sysresource object and if so, allocate from the local rman. Sysresource objects add their resources to the pool and reserve them upon boot. This means sysresources need to be probed before other ACPI devices. Changes include: * Add ordering to the child device probe. The current order is: system resource objects, embedded controllers, then everything else. * Make acpi_MatchHid take a handle instead of a device_t arg. * Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
* Match a HID string against a handle
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
* returns ACPI_MATCHHID_HID if _HID match
* ACPI_MATCHHID_CID if _CID match and not _HID match.
* ACPI_MATCHHID_NOMATCH=0 if no match.
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*/
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
int
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
acpi_MatchHid(ACPI_HANDLE h, const char *hid)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
ACPI_DEVICE_INFO *devinfo;
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
BOOLEAN ret;
int i;
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
if (hid == NULL || h == NULL ||
ACPI_FAILURE(AcpiGetObjectInfo(h, &devinfo)))
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
return (ACPI_MATCHHID_NOMATCH);
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
acpi_MatchHid: use ACPI_MATCHHID_NOMATCH instead of FALSE Binary representation of both is the same (zero), but ACPI_MATCHHID_NOMATCH is better for consistency. MFC after: 4 days X-MFC with: r339754
2018-12-07 16:05:39 +00:00
ret = ACPI_MATCHHID_NOMATCH;
Fix acpi_MatchHid() to check the compatibility ID's if the hardware ID doesn't match. Submitted by: marcel
2003-12-26 15:42:13 +00:00
if ((devinfo->Valid & ACPI_VALID_HID) != 0 &&
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
strcmp(hid, devinfo->HardwareId.String) == 0)
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
ret = ACPI_MATCHHID_HID;
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
else if ((devinfo->Valid & ACPI_VALID_CID) != 0)
for (i = 0; i < devinfo->CompatibleIdList.Count; i++) {
if (strcmp(hid, devinfo->CompatibleIdList.Ids[i].String) == 0) {
Distinguish _CID match and _HID match and make lower priority probe when _CID match. Reviewed by: jhb, imp Differential Revision:https://reviews.freebsd.org/D16468
2018-10-26 00:05:46 +00:00
ret = ACPI_MATCHHID_CID;
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
break;
}
}
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
Catch up with ACPICA 20090903.
2009-09-11 22:49:34 +00:00
AcpiOsFree(devinfo);
Add support for multiple CIDs since _CID can contain a package of values. Implement this in acpi_MatchHid() and acpi_isa_get_compatid(). This should fix mouse support for some users. Move all users of AcpiGetObjectInfo() to use dynamic storage instead of a devinfo on the stack. This is necessary since ACPI-CA needs to allocate different sized arrays for the CompatList.
2003-12-18 03:25:22 +00:00
return (ret);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
/*
* Return the handle of a named object within our scope, ie. that of (parent)
* or one if its parents.
*/
ACPI_STATUS
acpi_GetHandleInScope(ACPI_HANDLE parent, char *path, ACPI_HANDLE *result)
{
ACPI_HANDLE r;
ACPI_STATUS status;
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Walk back up the tree to the root */
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
for (;;) {
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
status = AcpiGetHandle(parent, path, &r);
if (ACPI_SUCCESS(status)) {
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
*result = r;
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_OK);
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
}
/tmp/m
2004-08-03 05:13:56 +00:00
/* XXX Return error here? */
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
if (status != AE_NOT_FOUND)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_OK);
Match namespace cleanup changes in ACPI CA 20020217 update. Use ACPI_SUCCESS/ACPI_FAILURE consistently. The AcpiGetInto* interfaces are obsoleted by ACPI_ALLOCATE_BUFFER. Add AcpiBatteryIsPresent helper to determine whether a battery device is inserted. Add ACPI_ALL_DRIVERS to the list of debug layers, now that we own the namespace for this. Pr:
2002-02-23 05:21:56 +00:00
if (ACPI_FAILURE(AcpiGetParent(parent, &r)))
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_NOT_FOUND);
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
parent = r;
}
}
/*
* Allocate a buffer with a preset data size.
*/
ACPI_BUFFER *
acpi_AllocBuffer(int size)
{
ACPI_BUFFER *buf;
if ((buf = malloc(size + sizeof(*buf), M_ACPIDEV, M_NOWAIT)) == NULL)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (NULL);
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
buf->Length = size;
buf->Pointer = (void *)(buf + 1);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (buf);
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
}
Change to acpi_{Get,Set}Integer to provide both methods. Convert all callers to the new API. Submitted by: Mark Santcroos <marks@ripe.net>
2004-03-03 18:34:42 +00:00
ACPI_STATUS
Use an unsigned int instead of an int for the Get/Set Integer interface. Pointed out by: le
2004-03-09 05:41:28 +00:00
acpi_SetInteger(ACPI_HANDLE handle, char *path, UINT32 number)
Change to acpi_{Get,Set}Integer to provide both methods. Convert all callers to the new API. Submitted by: Mark Santcroos <marks@ripe.net>
2004-03-03 18:34:42 +00:00
{
ACPI_OBJECT arg1;
ACPI_OBJECT_LIST args;
arg1.Type = ACPI_TYPE_INTEGER;
arg1.Integer.Value = number;
args.Count = 1;
args.Pointer = &arg1;
return (AcpiEvaluateObject(handle, path, &args, NULL));
}
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
/*
* Evaluate a path that should return an integer.
*/
ACPI_STATUS
Use an unsigned int instead of an int for the Get/Set Integer interface. Pointed out by: le
2004-03-09 05:41:28 +00:00
acpi_GetInteger(ACPI_HANDLE handle, char *path, UINT32 *number)
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
{
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
ACPI_STATUS status;
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
ACPI_BUFFER buf;
Add OS layer ACPI mutex and threading support. - Temporary fix a bug of Intel ACPI CA core code. - Add OS layer ACPI mutex support. This can be disabled by specifying option ACPI_NO_SEMAPHORES. - Add ACPI threading support. Now that we have a dedicate taskqueue for ACPI tasks and more ACPI task threads can be created by specifying option ACPI_MAX_THREADS. - Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given caller's buffer unless AE_BUFFER_OVERFLOW occurs. Also CM battery's evaluations were changed to use acpi_EvaluateIntoBuffer(). - Add new utility function acpi_ConvertBufferToInteger(). - Add simple locking for CM battery and temperature updating. - Fix a minor problem on EC locking. - Make the thermal zone polling rate to be changeable. - Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case, entering Debugger is easier to investigate the problem rather than panic.
2001-12-22 16:05:41 +00:00
ACPI_OBJECT param;
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
if (handle == NULL)
handle = ACPI_ROOT_OBJECT;
Add a hack to acpi_EvaluateInteger() to handle the case of a method returning a Buffer that contains an Integer rather an an Integer directly. Submitted by: msmith Approved by: msmith
2001-09-06 23:16:55 +00:00
/*
* Assume that what we've been pointed at is an Integer object, or
* a method that will return an Integer.
*/
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
buf.Pointer = &param;
buf.Length = sizeof(param);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
status = AcpiEvaluateObject(handle, path, NULL, &buf);
if (ACPI_SUCCESS(status)) {
if (param.Type == ACPI_TYPE_INTEGER)
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
*number = param.Integer.Value;
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
else
status = AE_TYPE;
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
}
Add a hack to acpi_EvaluateInteger() to handle the case of a method returning a Buffer that contains an Integer rather an an Integer directly. Submitted by: msmith Approved by: msmith
2001-09-06 23:16:55 +00:00
/*
* In some applications, a method that's expected to return an Integer
* may instead return a Buffer (probably to simplify some internal
* arithmetic). We'll try to fetch whatever it is, and if it's a Buffer,
* convert it into an Integer as best we can.
*
* This is a hack.
*/
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
if (status == AE_BUFFER_OVERFLOW) {
Match namespace cleanup changes in ACPI CA 20020217 update. Use ACPI_SUCCESS/ACPI_FAILURE consistently. The AcpiGetInto* interfaces are obsoleted by ACPI_ALLOCATE_BUFFER. Add AcpiBatteryIsPresent helper to determine whether a battery device is inserted. Add ACPI_ALL_DRIVERS to the list of debug layers, now that we own the namespace for this. Pr:
2002-02-23 05:21:56 +00:00
if ((buf.Pointer = AcpiOsAllocate(buf.Length)) == NULL) {
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
status = AE_NO_MEMORY;
Add a hack to acpi_EvaluateInteger() to handle the case of a method returning a Buffer that contains an Integer rather an an Integer directly. Submitted by: msmith Approved by: msmith
2001-09-06 23:16:55 +00:00
} else {
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
status = AcpiEvaluateObject(handle, path, NULL, &buf);
if (ACPI_SUCCESS(status))
status = acpi_ConvertBufferToInteger(&buf, number);
Don't free the buffer if it wasn't actually allocated.
2003-09-04 15:55:41 +00:00
AcpiOsFree(buf.Pointer);
Add a hack to acpi_EvaluateInteger() to handle the case of a method returning a Buffer that contains an Integer rather an an Integer directly. Submitted by: msmith Approved by: msmith
2001-09-06 23:16:55 +00:00
}
}
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (status);
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
}
Add OS layer ACPI mutex and threading support. - Temporary fix a bug of Intel ACPI CA core code. - Add OS layer ACPI mutex support. This can be disabled by specifying option ACPI_NO_SEMAPHORES. - Add ACPI threading support. Now that we have a dedicate taskqueue for ACPI tasks and more ACPI task threads can be created by specifying option ACPI_MAX_THREADS. - Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given caller's buffer unless AE_BUFFER_OVERFLOW occurs. Also CM battery's evaluations were changed to use acpi_EvaluateIntoBuffer(). - Add new utility function acpi_ConvertBufferToInteger(). - Add simple locking for CM battery and temperature updating. - Fix a minor problem on EC locking. - Make the thermal zone polling rate to be changeable. - Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case, entering Debugger is easier to investigate the problem rather than panic.
2001-12-22 16:05:41 +00:00
ACPI_STATUS
Use an unsigned int instead of an int for the Get/Set Integer interface. Pointed out by: le
2004-03-09 05:41:28 +00:00
acpi_ConvertBufferToInteger(ACPI_BUFFER *bufp, UINT32 *number)
Add OS layer ACPI mutex and threading support. - Temporary fix a bug of Intel ACPI CA core code. - Add OS layer ACPI mutex support. This can be disabled by specifying option ACPI_NO_SEMAPHORES. - Add ACPI threading support. Now that we have a dedicate taskqueue for ACPI tasks and more ACPI task threads can be created by specifying option ACPI_MAX_THREADS. - Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given caller's buffer unless AE_BUFFER_OVERFLOW occurs. Also CM battery's evaluations were changed to use acpi_EvaluateIntoBuffer(). - Add new utility function acpi_ConvertBufferToInteger(). - Add simple locking for CM battery and temperature updating. - Fix a minor problem on EC locking. - Make the thermal zone polling rate to be changeable. - Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case, entering Debugger is easier to investigate the problem rather than panic.
2001-12-22 16:05:41 +00:00
{
ACPI_OBJECT *p;
Simplify some logic in converting a buffer to an integer.
2004-03-09 05:44:47 +00:00
UINT8 *val;
Add OS layer ACPI mutex and threading support. - Temporary fix a bug of Intel ACPI CA core code. - Add OS layer ACPI mutex support. This can be disabled by specifying option ACPI_NO_SEMAPHORES. - Add ACPI threading support. Now that we have a dedicate taskqueue for ACPI tasks and more ACPI task threads can be created by specifying option ACPI_MAX_THREADS. - Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given caller's buffer unless AE_BUFFER_OVERFLOW occurs. Also CM battery's evaluations were changed to use acpi_EvaluateIntoBuffer(). - Add new utility function acpi_ConvertBufferToInteger(). - Add simple locking for CM battery and temperature updating. - Fix a minor problem on EC locking. - Make the thermal zone polling rate to be changeable. - Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case, entering Debugger is easier to investigate the problem rather than panic.
2001-12-22 16:05:41 +00:00
int i;
p = (ACPI_OBJECT *)bufp->Pointer;
if (p->Type == ACPI_TYPE_INTEGER) {
*number = p->Integer.Value;
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_OK);
Add OS layer ACPI mutex and threading support. - Temporary fix a bug of Intel ACPI CA core code. - Add OS layer ACPI mutex support. This can be disabled by specifying option ACPI_NO_SEMAPHORES. - Add ACPI threading support. Now that we have a dedicate taskqueue for ACPI tasks and more ACPI task threads can be created by specifying option ACPI_MAX_THREADS. - Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given caller's buffer unless AE_BUFFER_OVERFLOW occurs. Also CM battery's evaluations were changed to use acpi_EvaluateIntoBuffer(). - Add new utility function acpi_ConvertBufferToInteger(). - Add simple locking for CM battery and temperature updating. - Fix a minor problem on EC locking. - Make the thermal zone polling rate to be changeable. - Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case, entering Debugger is easier to investigate the problem rather than panic.
2001-12-22 16:05:41 +00:00
}
if (p->Type != ACPI_TYPE_BUFFER)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_TYPE);
Add OS layer ACPI mutex and threading support. - Temporary fix a bug of Intel ACPI CA core code. - Add OS layer ACPI mutex support. This can be disabled by specifying option ACPI_NO_SEMAPHORES. - Add ACPI threading support. Now that we have a dedicate taskqueue for ACPI tasks and more ACPI task threads can be created by specifying option ACPI_MAX_THREADS. - Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given caller's buffer unless AE_BUFFER_OVERFLOW occurs. Also CM battery's evaluations were changed to use acpi_EvaluateIntoBuffer(). - Add new utility function acpi_ConvertBufferToInteger(). - Add simple locking for CM battery and temperature updating. - Fix a minor problem on EC locking. - Make the thermal zone polling rate to be changeable. - Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case, entering Debugger is easier to investigate the problem rather than panic.
2001-12-22 16:05:41 +00:00
if (p->Buffer.Length > sizeof(int))
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_BAD_DATA);
Add OS layer ACPI mutex and threading support. - Temporary fix a bug of Intel ACPI CA core code. - Add OS layer ACPI mutex support. This can be disabled by specifying option ACPI_NO_SEMAPHORES. - Add ACPI threading support. Now that we have a dedicate taskqueue for ACPI tasks and more ACPI task threads can be created by specifying option ACPI_MAX_THREADS. - Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given caller's buffer unless AE_BUFFER_OVERFLOW occurs. Also CM battery's evaluations were changed to use acpi_EvaluateIntoBuffer(). - Add new utility function acpi_ConvertBufferToInteger(). - Add simple locking for CM battery and temperature updating. - Fix a minor problem on EC locking. - Make the thermal zone polling rate to be changeable. - Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case, entering Debugger is easier to investigate the problem rather than panic.
2001-12-22 16:05:41 +00:00
*number = 0;
Simplify some logic in converting a buffer to an integer.
2004-03-09 05:44:47 +00:00
val = p->Buffer.Pointer;
Add OS layer ACPI mutex and threading support. - Temporary fix a bug of Intel ACPI CA core code. - Add OS layer ACPI mutex support. This can be disabled by specifying option ACPI_NO_SEMAPHORES. - Add ACPI threading support. Now that we have a dedicate taskqueue for ACPI tasks and more ACPI task threads can be created by specifying option ACPI_MAX_THREADS. - Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given caller's buffer unless AE_BUFFER_OVERFLOW occurs. Also CM battery's evaluations were changed to use acpi_EvaluateIntoBuffer(). - Add new utility function acpi_ConvertBufferToInteger(). - Add simple locking for CM battery and temperature updating. - Fix a minor problem on EC locking. - Make the thermal zone polling rate to be changeable. - Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case, entering Debugger is easier to investigate the problem rather than panic.
2001-12-22 16:05:41 +00:00
for (i = 0; i < p->Buffer.Length; i++)
Simplify some logic in converting a buffer to an integer.
2004-03-09 05:44:47 +00:00
*number += val[i] << (i * 8);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_OK);
Add OS layer ACPI mutex and threading support. - Temporary fix a bug of Intel ACPI CA core code. - Add OS layer ACPI mutex support. This can be disabled by specifying option ACPI_NO_SEMAPHORES. - Add ACPI threading support. Now that we have a dedicate taskqueue for ACPI tasks and more ACPI task threads can be created by specifying option ACPI_MAX_THREADS. - Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given caller's buffer unless AE_BUFFER_OVERFLOW occurs. Also CM battery's evaluations were changed to use acpi_EvaluateIntoBuffer(). - Add new utility function acpi_ConvertBufferToInteger(). - Add simple locking for CM battery and temperature updating. - Fix a minor problem on EC locking. - Make the thermal zone polling rate to be changeable. - Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case, entering Debugger is easier to investigate the problem rather than panic.
2001-12-22 16:05:41 +00:00
}
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
/*
* Iterate over the elements of an a package object, calling the supplied
* function for each element.
*
* XXX possible enhancement might be to abort traversal on error.
*/
ACPI_STATUS
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
acpi_ForeachPackageObject(ACPI_OBJECT *pkg,
void (*func)(ACPI_OBJECT *comp, void *arg), void *arg)
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
{
ACPI_OBJECT *comp;
int i;
Remove the ACPIIO_ENABLE and ACPIIO_DISABLE ioctls as well as all callers. These ioctls attempted to enable and disable the ACPI interpreter at runtime. In practice, it is not possible to boot with ACPI and then disable it on many systems and trying to do so can cause crashes, interrupt storms, etc. Binary compatibility with userland is retained. MFC after: 2 days
2004-08-18 05:48:24 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
if (pkg == NULL || pkg->Type != ACPI_TYPE_PACKAGE)
return (AE_BAD_PARAMETER);
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Iterate over components */
i = 0;
comp = pkg->Package.Elements;
for (; i < pkg->Package.Count; i++, comp++)
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
func(comp, arg);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_OK);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Add a new helper function for finding resources in resource buffers. Move the ACPI generic battery code into a new file.
2001-07-05 07:14:30 +00:00
/*
* Find the (index)th resource object in a set.
*/
ACPI_STATUS
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
acpi_FindIndexedResource(ACPI_BUFFER *buf, int index, ACPI_RESOURCE **resp)
Add a new helper function for finding resources in resource buffers. Move the ACPI generic battery code into a new file.
2001-07-05 07:14:30 +00:00
{
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
ACPI_RESOURCE *rp;
Add a new helper function for finding resources in resource buffers. Move the ACPI generic battery code into a new file.
2001-07-05 07:14:30 +00:00
int i;
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
rp = (ACPI_RESOURCE *)buf->Pointer;
Add a new helper function for finding resources in resource buffers. Move the ACPI generic battery code into a new file.
2001-07-05 07:14:30 +00:00
i = index;
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
while (i-- > 0) {
Remove trailing whitespace.
2004-12-27 05:36:47 +00:00
/* Range check */
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
if (rp > (ACPI_RESOURCE *)((u_int8_t *)buf->Pointer + buf->Length))
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_BAD_PARAMETER);
/* Check for terminator */
Catch up with ACPI-CA 20051021 import
2005-11-01 22:44:08 +00:00
if (rp->Type == ACPI_RESOURCE_TYPE_END_TAG || rp->Length == 0)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_NOT_FOUND);
Use ACPI_NEXT_RESOURCE instead of defining our own copy. The one provided with ACPI-CA is identical now.
2004-02-19 18:20:03 +00:00
rp = ACPI_NEXT_RESOURCE(rp);
Add a new helper function for finding resources in resource buffers. Move the ACPI generic battery code into a new file.
2001-07-05 07:14:30 +00:00
}
if (resp != NULL)
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
*resp = rp;
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_OK);
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
}
/*
* Append an ACPI_RESOURCE to an ACPI_BUFFER.
*
* Given a pointer to an ACPI_RESOURCE structure, expand the ACPI_BUFFER
* provided to contain it. If the ACPI_BUFFER is empty, allocate a sensible
* backing block. If the ACPI_RESOURCE is NULL, return an empty set of
* resources.
*/
#define ACPI_INITIAL_RESOURCE_BUFFER_SIZE 512
ACPI_STATUS
acpi_AppendBufferResource(ACPI_BUFFER *buf, ACPI_RESOURCE *res)
{
ACPI_RESOURCE *rp;
void *newp;
Remove the ACPIIO_ENABLE and ACPIIO_DISABLE ioctls as well as all callers. These ioctls attempted to enable and disable the ACPI interpreter at runtime. In practice, it is not possible to boot with ACPI and then disable it on many systems and trying to do so can cause crashes, interrupt storms, etc. Binary compatibility with userland is retained. MFC after: 2 days
2004-08-18 05:48:24 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Initialise the buffer if necessary. */
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
if (buf->Pointer == NULL) {
buf->Length = ACPI_INITIAL_RESOURCE_BUFFER_SIZE;
if ((buf->Pointer = AcpiOsAllocate(buf->Length)) == NULL)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_NO_MEMORY);
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
rp = (ACPI_RESOURCE *)buf->Pointer;
Catch up with ACPI-CA 20051021 import
2005-11-01 22:44:08 +00:00
rp->Type = ACPI_RESOURCE_TYPE_END_TAG;
- Some BIOSes use an Extended IRQ resource descriptor in _PRS for a link that uses non-ISA IRQs but use a plain IRQ resource in _CRS. However, a non-ISA IRQ can't fit into a plain IRQ resource. If we encounter a link like this, build the resource buffer from _PRS instead of _CRS. - Set the correct size of the end tag in a resource buffer. Tested by: Benjamin Lee <ben@b1c1l1.com> MFC after: 2 weeks
2013-04-22 15:51:06 +00:00
rp->Length = ACPI_RS_SIZE_MIN;
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
}
if (res == NULL)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_OK);
Remove the ACPIIO_ENABLE and ACPIIO_DISABLE ioctls as well as all callers. These ioctls attempted to enable and disable the ACPI interpreter at runtime. In practice, it is not possible to boot with ACPI and then disable it on many systems and trying to do so can cause crashes, interrupt storms, etc. Binary compatibility with userland is retained. MFC after: 2 days
2004-08-18 05:48:24 +00:00
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
/*
* Scan the current buffer looking for the terminator.
* This will either find the terminator or hit the end
* of the buffer and return an error.
*/
rp = (ACPI_RESOURCE *)buf->Pointer;
for (;;) {
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Range check, don't go outside the buffer */
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
if (rp >= (ACPI_RESOURCE *)((u_int8_t *)buf->Pointer + buf->Length))
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_BAD_PARAMETER);
Catch up with ACPI-CA 20051021 import
2005-11-01 22:44:08 +00:00
if (rp->Type == ACPI_RESOURCE_TYPE_END_TAG || rp->Length == 0)
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
break;
Use ACPI_NEXT_RESOURCE instead of defining our own copy. The one provided with ACPI-CA is identical now.
2004-02-19 18:20:03 +00:00
rp = ACPI_NEXT_RESOURCE(rp);
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
}
/*
* Check the size of the buffer and expand if required.
*
* Required size is:
* size of existing resources before terminator +
* size of new resource and header +
* size of terminator.
*
* Note that this loop should really only run once, unless
* for some reason we are stuffing a *really* huge resource.
*/
while ((((u_int8_t *)rp - (u_int8_t *)buf->Pointer) +
Catch up with ACPI-CA 20051021 import
2005-11-01 22:44:08 +00:00
res->Length + ACPI_RS_SIZE_NO_DATA +
ACPI_RS_SIZE_MIN) >= buf->Length) {
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
if ((newp = AcpiOsAllocate(buf->Length * 2)) == NULL)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_NO_MEMORY);
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
bcopy(buf->Pointer, newp, buf->Length);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
rp = (ACPI_RESOURCE *)((u_int8_t *)newp +
Move the resource pointer when we reallocate the buffer. Submitted by: "neckpain@nettaxi.com" <neckpain@nettaxi.com>
2001-08-03 08:38:11 +00:00
((u_int8_t *)rp - (u_int8_t *)buf->Pointer));
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
AcpiOsFree(buf->Pointer);
buf->Pointer = newp;
buf->Length += buf->Length;
}
Remove the ACPIIO_ENABLE and ACPIIO_DISABLE ioctls as well as all callers. These ioctls attempted to enable and disable the ACPI interpreter at runtime. In practice, it is not possible to boot with ACPI and then disable it on many systems and trying to do so can cause crashes, interrupt storms, etc. Binary compatibility with userland is retained. MFC after: 2 days
2004-08-18 05:48:24 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Insert the new resource. */
Catch up with ACPI-CA 20051021 import
2005-11-01 22:44:08 +00:00
bcopy(res, rp, res->Length + ACPI_RS_SIZE_NO_DATA);
Remove the ACPIIO_ENABLE and ACPIIO_DISABLE ioctls as well as all callers. These ioctls attempted to enable and disable the ACPI interpreter at runtime. In practice, it is not possible to boot with ACPI and then disable it on many systems and trying to do so can cause crashes, interrupt storms, etc. Binary compatibility with userland is retained. MFC after: 2 days
2004-08-18 05:48:24 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* And add the terminator. */
Use ACPI_NEXT_RESOURCE instead of defining our own copy. The one provided with ACPI-CA is identical now.
2004-02-19 18:20:03 +00:00
rp = ACPI_NEXT_RESOURCE(rp);
Catch up with ACPI-CA 20051021 import
2005-11-01 22:44:08 +00:00
rp->Type = ACPI_RESOURCE_TYPE_END_TAG;
- Some BIOSes use an Extended IRQ resource descriptor in _PRS for a link that uses non-ISA IRQs but use a plain IRQ resource in _CRS. However, a non-ISA IRQ can't fit into a plain IRQ resource. If we encounter a link like this, build the resource buffer from _PRS instead of _CRS. - Set the correct size of the end tag in a resource buffer. Tested by: Benjamin Lee <ben@b1c1l1.com> MFC after: 2 weeks
2013-04-22 15:51:06 +00:00
rp->Length = ACPI_RS_SIZE_MIN;
- Prevent the ACPI code from being loaded as a module other than at boot time. Loading as a module once the system is up and running doesn't make any sense. - Fix acpi_FindIndexedResource (it would only check the first resource), changes the calling interface. - Add a new helper function (acpi_AppendBufferResource) to help building buffers containing resources.
2001-07-30 08:59:43 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (AE_OK);
Add a new helper function for finding resources in resource buffers. Move the ACPI generic battery code into a new file.
2001-07-05 07:14:30 +00:00
}
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
acpi: Add an interface to obtain DSM information The Device Specific Method (_DSM) is on optional object that defines device specific controls. This will be useful for our power management controller in upcoming patches. More information can be found in ACPI spec 6.2 section 9.1.1 https://www.uefi.org/sites/default/files/resources/ACPI_6_2.pdf This patch had a minor modification changing ENOMEM to AE_NO_MEMORY after it got review and approval but before committing. Test Plan: Tested in my s0ix branch Reviewed by: kib Approved by: emaste (mentor) Differential Revision: https://reviews.freebsd.org/D17121
2018-10-22 03:29:54 +00:00
UINT8
acpi_DSMQuery(ACPI_HANDLE h, uint8_t *uuid, int revision)
{
/*
* ACPI spec 9.1.1 defines this.
*
* "Arg2: Function Index Represents a specific function whose meaning is
* specific to the UUID and Revision ID. Function indices should start
* with 1. Function number zero is a query function (see the special
* return code defined below)."
*/
ACPI_BUFFER buf;
ACPI_OBJECT *obj;
UINT8 ret = 0;
if (!ACPI_SUCCESS(acpi_EvaluateDSM(h, uuid, revision, 0, NULL, &buf))) {
ACPI_INFO(("Failed to enumerate DSM functions\n"));
return (0);
}
obj = (ACPI_OBJECT *)buf.Pointer;
KASSERT(obj, ("Object not allowed to be NULL\n"));
/*
* From ACPI 6.2 spec 9.1.1:
* If Function Index = 0, a Buffer containing a function index bitfield.
* Otherwise, the return value and type depends on the UUID and revision
* ID (see below).
*/
switch (obj->Type) {
case ACPI_TYPE_BUFFER:
ret = *(uint8_t *)obj->Buffer.Pointer;
break;
case ACPI_TYPE_INTEGER:
ACPI_BIOS_WARNING((AE_INFO,
"Possibly buggy BIOS with ACPI_TYPE_INTEGER for function enumeration\n"));
ret = obj->Integer.Value & 0xFF;
break;
default:
ACPI_WARNING((AE_INFO, "Unexpected return type %u\n", obj->Type));
};
AcpiOsFree(obj);
return ret;
}
/*
* DSM may return multiple types depending on the function. It is therefore
* unsafe to use the typed evaluation. It is highly recommended that the caller
* check the type of the returned object.
*/
ACPI_STATUS
acpi_EvaluateDSM(ACPI_HANDLE handle, uint8_t *uuid, int revision,
uint64_t function, union acpi_object *package, ACPI_BUFFER *out_buf)
{
ACPI_OBJECT arg[4];
ACPI_OBJECT_LIST arglist;
ACPI_BUFFER buf;
ACPI_STATUS status;
if (out_buf == NULL)
return (AE_NO_MEMORY);
arg[0].Type = ACPI_TYPE_BUFFER;
arg[0].Buffer.Length = ACPI_UUID_LENGTH;
arg[0].Buffer.Pointer = uuid;
arg[1].Type = ACPI_TYPE_INTEGER;
arg[1].Integer.Value = revision;
arg[2].Type = ACPI_TYPE_INTEGER;
arg[2].Integer.Value = function;
if (package) {
arg[3] = *package;
} else {
arg[3].Type = ACPI_TYPE_PACKAGE;
arg[3].Package.Count = 0;
arg[3].Package.Elements = NULL;
}
arglist.Pointer = arg;
arglist.Count = 4;
buf.Pointer = NULL;
buf.Length = ACPI_ALLOCATE_BUFFER;
status = AcpiEvaluateObject(handle, "_DSM", &arglist, &buf);
if (ACPI_FAILURE(status))
return (status);
KASSERT(ACPI_SUCCESS(status), ("Unexpected status"));
*out_buf = buf;
return (status);
}
Add a wrapper for evaluating _OSC methods. This wrapper does not translate errors in the first word to ACPI error status returns. Use this wrapper in the acpi_cpu(4) driver in place of the existing _OSC code. While here, fix a bug where the wrong count of words was passed when invoking _OSC. Reviewed by: jkim MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D6022
2016-04-20 20:55:58 +00:00
ACPI_STATUS
acpi_EvaluateOSC(ACPI_HANDLE handle, uint8_t *uuid, int revision, int count,
Optionally return the output capabilities list from _OSC. Both of the callers were expecting the input cap_set to be modified. This fixes them to request cap_set to be updated with the returned buffer. Reviewed by: jkim Differential Revision: https://reviews.freebsd.org/D6040
2016-04-22 17:51:19 +00:00
uint32_t *caps_in, uint32_t *caps_out, bool query)
Add a wrapper for evaluating _OSC methods. This wrapper does not translate errors in the first word to ACPI error status returns. Use this wrapper in the acpi_cpu(4) driver in place of the existing _OSC code. While here, fix a bug where the wrong count of words was passed when invoking _OSC. Reviewed by: jkim MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D6022
2016-04-20 20:55:58 +00:00
{
Optionally return the output capabilities list from _OSC. Both of the callers were expecting the input cap_set to be modified. This fixes them to request cap_set to be updated with the returned buffer. Reviewed by: jkim Differential Revision: https://reviews.freebsd.org/D6040
2016-04-22 17:51:19 +00:00
ACPI_OBJECT arg[4], *ret;
Add a wrapper for evaluating _OSC methods. This wrapper does not translate errors in the first word to ACPI error status returns. Use this wrapper in the acpi_cpu(4) driver in place of the existing _OSC code. While here, fix a bug where the wrong count of words was passed when invoking _OSC. Reviewed by: jkim MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D6022
2016-04-20 20:55:58 +00:00
ACPI_OBJECT_LIST arglist;
Optionally return the output capabilities list from _OSC. Both of the callers were expecting the input cap_set to be modified. This fixes them to request cap_set to be updated with the returned buffer. Reviewed by: jkim Differential Revision: https://reviews.freebsd.org/D6040
2016-04-22 17:51:19 +00:00
ACPI_BUFFER buf;
ACPI_STATUS status;
Add a wrapper for evaluating _OSC methods. This wrapper does not translate errors in the first word to ACPI error status returns. Use this wrapper in the acpi_cpu(4) driver in place of the existing _OSC code. While here, fix a bug where the wrong count of words was passed when invoking _OSC. Reviewed by: jkim MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D6022
2016-04-20 20:55:58 +00:00
arglist.Pointer = arg;
arglist.Count = 4;
arg[0].Type = ACPI_TYPE_BUFFER;
arg[0].Buffer.Length = ACPI_UUID_LENGTH;
arg[0].Buffer.Pointer = uuid;
arg[1].Type = ACPI_TYPE_INTEGER;
arg[1].Integer.Value = revision;
arg[2].Type = ACPI_TYPE_INTEGER;
arg[2].Integer.Value = count;
arg[3].Type = ACPI_TYPE_BUFFER;
Optionally return the output capabilities list from _OSC. Both of the callers were expecting the input cap_set to be modified. This fixes them to request cap_set to be updated with the returned buffer. Reviewed by: jkim Differential Revision: https://reviews.freebsd.org/D6040
2016-04-22 17:51:19 +00:00
arg[3].Buffer.Length = count * sizeof(*caps_in);
arg[3].Buffer.Pointer = (uint8_t *)caps_in;
caps_in[0] = query ? 1 : 0;
buf.Pointer = NULL;
buf.Length = ACPI_ALLOCATE_BUFFER;
status = AcpiEvaluateObjectTyped(handle, "_OSC", &arglist, &buf,
ACPI_TYPE_BUFFER);
if (ACPI_FAILURE(status))
return (status);
if (caps_out != NULL) {
ret = buf.Pointer;
if (ret->Buffer.Length != count * sizeof(*caps_out)) {
AcpiOsFree(buf.Pointer);
return (AE_BUFFER_OVERFLOW);
}
bcopy(ret->Buffer.Pointer, caps_out, ret->Buffer.Length);
}
AcpiOsFree(buf.Pointer);
return (status);
Add a wrapper for evaluating _OSC methods. This wrapper does not translate errors in the first word to ACPI error status returns. Use this wrapper in the acpi_cpu(4) driver in place of the existing _OSC code. While here, fix a bug where the wrong count of words was passed when invoking _OSC. Reviewed by: jkim MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D6022
2016-04-20 20:55:58 +00:00
}
Add a helper routine acpi_SetIntrModel() to call the _PIC method to set the interrupt model in use so that ACPI can properly route interrupts for machines using APIC's or SAPIC's.
2002-09-06 16:08:08 +00:00
/*
* Set interrupt model.
*/
ACPI_STATUS
acpi_SetIntrModel(int model)
{
Change to acpi_{Get,Set}Integer to provide both methods. Convert all callers to the new API. Submitted by: Mark Santcroos <marks@ripe.net>
2004-03-03 18:34:42 +00:00
return (acpi_SetInteger(ACPI_ROOT_OBJECT, "_PIC", model));
Add a helper routine acpi_SetIntrModel() to call the _PIC method to set the interrupt model in use so that ACPI can properly route interrupts for machines using APIC's or SAPIC's.
2002-09-06 16:08:08 +00:00
}
Extract the code to find and map the MADT ACPI table during early kernel startup and genericize it so it can be reused to map other tables as well: - Add a routine to walk a list of ACPI subtables such as those used in the APIC and SRAT tables in the MI acpi(4) driver. - Move the routines for mapping and unmapping an ACPI table as well as mapping the RSDT or XSDT and searching for a table with a given signature out into acpica_machdep.c for both amd64 and i386.
2009-09-23 15:42:35 +00:00
/*
* Walk subtables of a table and call a callback routine for each
* subtable. The caller should provide the first subtable and a
* pointer to the end of the table. This can be used to walk tables
* such as MADT and SRAT that use subtable entries.
*/
void
acpi_walk_subtables(void *first, void *end, acpi_subtable_handler *handler,
void *arg)
{
ACPI_SUBTABLE_HEADER *entry;
for (entry = first; (void *)entry < end; ) {
/* Avoid an infinite loop if we hit a bogus entry. */
if (entry->Length < sizeof(ACPI_SUBTABLE_HEADER))
return;
handler(entry, arg);
entry = ACPI_ADD_PTR(ACPI_SUBTABLE_HEADER, entry, entry->Length);
}
}
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
/*
* DEPRECATED. This interface has serious deficiencies and will be
* removed.
*
* Immediately enter the sleep state. In the old model, acpiconf(8) ran
* rc.suspend and rc.resume so we don't have to notify devd(8) to do this.
*/
ACPI_STATUS
acpi_SetSleepState(struct acpi_softc *sc, int state)
{
static int once;
if (!once) {
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(sc->acpi_dev,
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
"warning: acpi_SetSleepState() deprecated, need to update your software\n");
once = 1;
}
return (acpi_EnterSleepState(sc, state));
}
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
#if defined(__amd64__) || defined(__i386__)
Fix the problem acpi_sleep_force() hang. Suspending from callout cause the freeze in DEVICE_SUSPEND(). Suspend from acpi_task thread in stead. MFC after: 3 days
2012-05-29 05:09:40 +00:00
static void
acpi_sleep_force_task(void *context)
{
struct acpi_softc *sc = (struct acpi_softc *)context;
if (ACPI_FAILURE(acpi_EnterSleepState(sc, sc->acpi_next_sstate)))
device_printf(sc->acpi_dev, "force sleep state S%d failed\n",
sc->acpi_next_sstate);
}
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
static void
acpi_sleep_force(void *arg)
{
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
struct acpi_softc *sc = (struct acpi_softc *)arg;
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(sc->acpi_dev,
"suspend request timed out, forcing sleep now\n");
Fix the problem acpi_sleep_force() hang. Suspending from callout cause the freeze in DEVICE_SUSPEND(). Suspend from acpi_task thread in stead. MFC after: 3 days
2012-05-29 05:09:40 +00:00
/*
Convert from timeout(9) to callout(9).
2014-09-22 14:27:26 +00:00
* XXX Suspending from callout causes freezes in DEVICE_SUSPEND().
* Suspend from acpi_task thread instead.
Fix the problem acpi_sleep_force() hang. Suspending from callout cause the freeze in DEVICE_SUSPEND(). Suspend from acpi_task thread in stead. MFC after: 3 days
2012-05-29 05:09:40 +00:00
*/
if (ACPI_FAILURE(AcpiOsExecute(OSL_NOTIFY_HANDLER,
acpi_sleep_force_task, sc)))
device_printf(sc->acpi_dev, "AcpiOsExecute() for sleeping failed\n");
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
}
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
#endif
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
/*
* Request that the system enter the given suspend state. All /dev/apm
* devices and devd(8) will be notified. Userland then has a chance to
* save state and acknowledge the request. The system sleeps once all
* acks are in.
*/
int
acpi_ReqSleepState(struct acpi_softc *sc, int state)
{
Copy apm(4) emulation from sys/i386/acpica/acpi_machdep.c and install apm(8) and apm_bios.h on amd64.
2009-09-27 14:00:16 +00:00
#if defined(__amd64__) || defined(__i386__)
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
struct apm_clone_data *clone;
If there is any pending sleep request, disallow entering S5 state. Otherwise, bad things may happen. ;-)
2010-07-15 23:24:06 +00:00
ACPI_STATUS status;
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
if (state < ACPI_STATE_S1 || state > ACPI_S_STATES_MAX)
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
return (EINVAL);
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
if (!acpi_sleep_states[state])
return (EOPNOTSUPP);
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
Disable suspend when we're shutting down. This solves the "tell FreeBSD to shut down; close laptop lid" scenario which otherwise tended to end with a laptop overheating or the battery dying. The implementation uses a new sysctl, kern.suspend_blocked; init(8) sets this while rc.suspend runs, and the ACPI sleep code ignores requests while the sysctl is set. Discussed on: freebsd-acpi (35 emails) MFC after: 1 week
2015-10-01 10:52:26 +00:00
/*
* If a reboot/shutdown/suspend request is already in progress or
* suspend is blocked due to an upcoming shutdown, just return.
*/
if (rebooting || sc->acpi_next_sstate != 0 || suspend_blocked) {
Ignore the power button press event for resuming rather than starting shutdown. MFC after: 2 days
2012-05-22 05:18:30 +00:00
return (0);
}
Don't start the sleep state transition procedure while sleep is disabled or the system is in shutdown procedure. This should fix the problem which kernel never response to the sleep button press events after the message `suspend request ignored (not ready yet)'. MFC after: 3 days
2012-05-20 16:36:54 +00:00
/* Wait until sleep is enabled. */
while (sc->acpi_sleep_disabled) {
AcpiOsSleep(1000);
}
If there is any pending sleep request, disallow entering S5 state. Otherwise, bad things may happen. ;-)
2010-07-15 23:24:06 +00:00
ACPI_LOCK(acpi);
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
Don't start the sleep state transition procedure while sleep is disabled or the system is in shutdown procedure. This should fix the problem which kernel never response to the sleep button press events after the message `suspend request ignored (not ready yet)'. MFC after: 3 days
2012-05-20 16:36:54 +00:00
sc->acpi_next_sstate = state;
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
If there is any pending sleep request, disallow entering S5 state. Otherwise, bad things may happen. ;-)
2010-07-15 23:24:06 +00:00
/* S5 (soft-off) should be entered directly with no waiting. */
if (state == ACPI_STATE_S5) {
ACPI_UNLOCK(acpi);
status = acpi_EnterSleepState(sc, state);
return (ACPI_SUCCESS(status) ? 0 : ENXIO);
}
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
/* Record the pending state and notify all apm devices. */
STAILQ_FOREACH(clone, &sc->apm_cdevs, entries) {
clone->notify_status = APM_EV_NONE;
if ((clone->flags & ACPI_EVF_DEVD) == 0) {
selwakeuppri(&clone->sel_read, PZERO);
The lock associated with the /dev/apm knote is already held, so use KNOTE_LOCKED() instead of KNOTE_UNLOCKED(). Reported by: mav MFC after: 3 days
2010-06-08 21:27:05 +00:00
KNOTE_LOCKED(&clone->sel_read.si_note, 0);
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
}
}
Enter the sleep state immediately without waiting for timeout if devd(8) is not running such as the system in single user mode. MFC after: 1 week
2008-01-27 16:11:04 +00:00
/* If devd(8) is not running, immediately enter the sleep state. */
Check whether devd is running before calling resume notifier and reshuffle code to reduce unnecessary locking coverage.
2009-03-23 22:06:09 +00:00
if (!devctl_process_running()) {
Return errno value rather than boolean in this context. MFC after: 1 week
2008-01-28 01:57:48 +00:00
ACPI_UNLOCK(acpi);
If there is any pending sleep request, disallow entering S5 state. Otherwise, bad things may happen. ;-)
2010-07-15 23:24:06 +00:00
status = acpi_EnterSleepState(sc, state);
return (ACPI_SUCCESS(status) ? 0 : ENXIO);
Enter the sleep state immediately without waiting for timeout if devd(8) is not running such as the system in single user mode. MFC after: 1 week
2008-01-27 16:11:04 +00:00
}
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
/*
* Set a timeout to fire if userland doesn't ack the suspend request
* in time. This way we still eventually go to sleep if we were
* overheating or running low on battery, even if userland is hung.
* We cancel this timeout once all userland acks are in or the
* suspend request is aborted.
*/
callout_reset(&sc->susp_force_to, 10 * hz, acpi_sleep_force, sc);
ACPI_UNLOCK(acpi);
Check whether devd is running before calling resume notifier and reshuffle code to reduce unnecessary locking coverage.
2009-03-23 22:06:09 +00:00
/* Now notify devd(8) also. */
acpi_UserNotify("Suspend", ACPI_ROOT_OBJECT, state);
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
return (0);
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
#else
/* This platform does not support acpi suspend/resume. */
return (EOPNOTSUPP);
#endif
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
}
/*
* Acknowledge (or reject) a pending sleep state. The caller has
* prepared for suspend and is now ready for it to proceed. If the
* error argument is non-zero, it indicates suspend should be cancelled
* and gives an errno value describing why. Once all votes are in,
* we suspend the system.
*/
int
acpi_AckSleepState(struct apm_clone_data *clone, int error)
{
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
#if defined(__amd64__) || defined(__i386__)
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
struct acpi_softc *sc;
int ret, sleeping;
/* If no pending sleep state, return an error. */
ACPI_LOCK(acpi);
sc = clone->acpi_sc;
if (sc->acpi_next_sstate == 0) {
ACPI_UNLOCK(acpi);
return (ENXIO);
}
/* Caller wants to abort suspend process. */
if (error) {
sc->acpi_next_sstate = 0;
callout_stop(&sc->susp_force_to);
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(sc->acpi_dev,
"listener on %s cancelled the pending suspend\n",
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
devtoname(clone->cdev));
ACPI_UNLOCK(acpi);
return (0);
}
/*
* Mark this device as acking the suspend request. Then, walk through
* all devices, seeing if they agree yet. We only count devices that
* are writable since read-only devices couldn't ack the request.
*/
sleeping = TRUE;
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
clone->notify_status = APM_EV_ACKED;
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
STAILQ_FOREACH(clone, &sc->apm_cdevs, entries) {
if ((clone->flags & ACPI_EVF_WRITE) != 0 &&
clone->notify_status != APM_EV_ACKED) {
sleeping = FALSE;
break;
}
}
/* If all devices have voted "yes", we will suspend now. */
if (sleeping)
callout_stop(&sc->susp_force_to);
ACPI_UNLOCK(acpi);
ret = 0;
if (sleeping) {
if (ACPI_FAILURE(acpi_EnterSleepState(sc, sc->acpi_next_sstate)))
ret = ENODEV;
}
return (ret);
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
#else
/* This platform does not support acpi suspend/resume. */
return (EOPNOTSUPP);
#endif
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
}
Disable sleep requests for 5 sec after wakeup. This is needed for some Toshiba and Thinkpad laptops. Wakeup event is generated by power button or sleep button on some laptops but this also generates SCI interrupt, and shutdown the system as result. So this is introduced so that acpi driver ignore given requests for certain period.
2001-12-09 18:02:36 +00:00
static void
acpi_sleep_enable(void *arg)
{
Check whether devd is running before calling resume notifier and reshuffle code to reduce unnecessary locking coverage.
2009-03-23 22:06:09 +00:00
struct acpi_softc *sc = (struct acpi_softc *)arg;
/tmp/m
2004-08-03 05:13:56 +00:00
Convert from timeout(9) to callout(9).
2014-09-22 14:27:26 +00:00
ACPI_LOCK_ASSERT(acpi);
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
/* Reschedule if the system is not fully up and running. */
if (!AcpiGbl_SystemAwakeAndRunning) {
Convert from timeout(9) to callout(9).
2014-09-22 14:27:26 +00:00
callout_schedule(&acpi_sleep_timer, hz * ACPI_MINIMUM_AWAKETIME);
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
return;
}
sc->acpi_sleep_disabled = FALSE;
Check whether devd is running before calling resume notifier and reshuffle code to reduce unnecessary locking coverage.
2009-03-23 22:06:09 +00:00
}
static ACPI_STATUS
acpi_sleep_disable(struct acpi_softc *sc)
{
ACPI_STATUS status;
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
/* Fail if the system is not fully up and running. */
if (!AcpiGbl_SystemAwakeAndRunning)
return (AE_ERROR);
Check whether devd is running before calling resume notifier and reshuffle code to reduce unnecessary locking coverage.
2009-03-23 22:06:09 +00:00
ACPI_LOCK(acpi);
status = sc->acpi_sleep_disabled ? AE_ERROR : AE_OK;
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
sc->acpi_sleep_disabled = TRUE;
Check whether devd is running before calling resume notifier and reshuffle code to reduce unnecessary locking coverage.
2009-03-23 22:06:09 +00:00
ACPI_UNLOCK(acpi);
return (status);
Disable sleep requests for 5 sec after wakeup. This is needed for some Toshiba and Thinkpad laptops. Wakeup event is generated by power button or sleep button on some laptops but this also generates SCI interrupt, and shutdown the system as result. So this is introduced so that acpi driver ignore given requests for certain period.
2001-12-09 18:02:36 +00:00
}
Bring our local hack for wakeup back from sys/contrib/dev/acpica/Subsystem/Hardware/Attic/hwxface.c to the proper location after AcpiEnterSleepState(). - Wait for the WAK_STS bit - Evaluate the _WAK method and check result code
2001-03-07 15:22:14 +00:00
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
enum acpi_sleep_state {
ACPI_SS_NONE,
ACPI_SS_GPE_SET,
ACPI_SS_DEV_SUSPEND,
ACPI_SS_SLP_PREP,
ACPI_SS_SLEPT,
};
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
* Enter the desired system sleep state.
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
* Currently we support S1-S5 but S4 is only S4BIOS
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*/
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
static ACPI_STATUS
acpi_EnterSleepState(struct acpi_softc *sc, int state)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
Restore interrupt state after executing AcpiEnterSleepState().
2012-03-27 23:26:58 +00:00
register_t intr;
ACPI_STATUS status;
acpi: clear power button status bit after waking up... so that it is not confused for a new power off request. Learned from: Linux and ACPI specification Tested by: gjb MFC after: 12 days
2013-02-02 12:44:19 +00:00
ACPI_EVENT_STATUS power_button_status;
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
enum acpi_sleep_state slp_state;
Call AcpiLeaveSleepStatePrep() in interrupt disabled context (described in ACPICA source code). - Move intr_disable() and intr_restore() from acpi_wakeup.c to acpi.c and call AcpiLeaveSleepStatePrep() in interrupt disabled context. - Add acpi_wakeup_machdep() to execute wakeup MD procedures and call it twice in interrupt disabled/enabled context (ia64 version is just dummy). - Rename wakeup_cpus variable in acpi_sleep_machdep() to suspcpus in order to be shared by acpi_sleep_machdep() and acpi_wakeup_machdep(). - Move identity mapping related code to acpi_install_wakeup_handler() (i386 version) for preparation of x86/acpica/acpi_wakeup.c (MFC candidate). Reviewed by: jkim@ MFC after: 2 days
2012-06-01 15:26:32 +00:00
int sleep_result;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE_U32((char *)(uintptr_t)__func__, state);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
if (state < ACPI_STATE_S1 || state > ACPI_S_STATES_MAX)
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
return_ACPI_STATUS (AE_BAD_PARAMETER);
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
if (!acpi_sleep_states[state]) {
device_printf(sc->acpi_dev, "Sleep state S%d not supported by BIOS\n",
state);
return (AE_SUPPORT);
}
/* Re-entry once we're suspending is not allowed. */
status = acpi_sleep_disable(sc);
if (ACPI_FAILURE(status)) {
device_printf(sc->acpi_dev,
"suspend request ignored (not ready yet)\n");
return (status);
}
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
if (state == ACPI_STATE_S5) {
/*
* Shut down cleanly and power off. This will call us back through the
* shutdown handlers.
*/
shutdown_nice(RB_POWEROFF);
return_ACPI_STATUS (AE_OK);
}
hook userland threads suspend + resume into acpi suspend code Also, split power_suspend into power_suspend and power_suspend_early. power_suspend_early is called before the userland is frozen. power_suspend is called after the userland is frozen. Currently only VT switching is hooked to power_suspend_early. This is needed because switching away from X server requires its cooperation, so obviously X server must not be frozen when that happens. Freezing userland during ACPI suspend is useful because not all drivers correctly handle suspension concurrent with other activity. This is especially applicable to drivers ported from other operating systems that suspend all software activity between placing drivers and hardware into suspended state. In particular drm2/radeon (radeonkms) depends on the described procedure. The driver does not have any internal synchronization between suspension activities and processing of userland requests. Many thanks to kib for the code that allows to freeze and thaw all userland threads. Note that ideally we also need to park / inhibit (non-special) kernel threads as well to ensure that they do not call into drivers. MFC after: 17 days
2015-01-27 17:33:18 +00:00
EVENTHANDLER_INVOKE(power_suspend_early);
stop_all_proc();
Add event handlers for (ACPI) suspend/resume events. Suspend event handlers are invoked right before device drivers go into sleep state and resume event handlers are invoked right after all device drivers are waken up.
2011-04-14 22:17:39 +00:00
EVENTHANDLER_INVOKE(power_suspend);
Add an EARLY_AP_STARTUP option to start APs earlier during boot. Currently, Application Processors (non-boot CPUs) are started by MD code at SI_SUB_CPU, but they are kept waiting in a "pen" until SI_SUB_SMP at which point they are released to run kernel threads. SI_SUB_SMP is one of the last SYSINIT levels, so APs don't enter the scheduler and start running threads until fairly late in the boot. This change moves SI_SUB_SMP up to just before software interrupt threads are created allowing the APs to start executing kernel threads much sooner (before any devices are probed). This allows several initialization routines that need to perform initialization on all CPUs to now perform that initialization in one step rather than having to defer the AP initialization to a second SYSINIT run at SI_SUB_SMP. It also permits all CPUs to be available for handling interrupts before any devices are probed. This last feature fixes a problem on with interrupt vector exhaustion. Specifically, in the old model all device interrupts were routed onto the boot CPU during boot. Later after the APs were released at SI_SUB_SMP, interrupts were redistributed across all CPUs. However, several drivers for multiqueue hardware allocate N interrupts per CPU in the system. In a system with many CPUs, just a few drivers doing this could exhaust the available pool of interrupt vectors on the boot CPU as each driver was allocating N * mp_ncpu vectors on the boot CPU. Now, drivers will allocate interrupts on their desired CPUs during boot meaning that only N interrupts are allocated from the boot CPU instead of N * mp_ncpu. Some other bits of code can also be simplified as smp_started is now true much earlier and will now always be true for these bits of code. This removes the need to treat the single-CPU boot environment as a special case. As a transition aid, the new behavior is available under a new kernel option (EARLY_AP_STARTUP). This will allow the option to be turned off if need be during initial testing. I plan to enable this on x86 by default in a followup commit in the next few days and to have all platforms moved over before 11.0. Once the transition is complete, the option will be removed along with the !EARLY_AP_STARTUP code. These changes have only been tested on x86. Other platform maintainers are encouraged to port their architectures over as well. The main things to check for are any uses of smp_started in MD code that can be simplified and SI_SUB_SMP SYSINITs in MD code that can be removed in the EARLY_AP_STARTUP case (e.g. the interrupt shuffling). PR: kern/199321 Reviewed by: markj, gnn, kib Sponsored by: Netflix
2016-05-14 18:22:52 +00:00
#ifdef EARLY_AP_STARTUP
MPASS(mp_ncpus == 1 || smp_started);
thread_lock(curthread);
sched_bind(curthread, 0);
thread_unlock(curthread);
#else
Make SMP code path conditional at run-time.
2010-07-13 16:35:41 +00:00
if (smp_started) {
thread_lock(curthread);
sched_bind(curthread, 0);
thread_unlock(curthread);
}
Add an EARLY_AP_STARTUP option to start APs earlier during boot. Currently, Application Processors (non-boot CPUs) are started by MD code at SI_SUB_CPU, but they are kept waiting in a "pen" until SI_SUB_SMP at which point they are released to run kernel threads. SI_SUB_SMP is one of the last SYSINIT levels, so APs don't enter the scheduler and start running threads until fairly late in the boot. This change moves SI_SUB_SMP up to just before software interrupt threads are created allowing the APs to start executing kernel threads much sooner (before any devices are probed). This allows several initialization routines that need to perform initialization on all CPUs to now perform that initialization in one step rather than having to defer the AP initialization to a second SYSINIT run at SI_SUB_SMP. It also permits all CPUs to be available for handling interrupts before any devices are probed. This last feature fixes a problem on with interrupt vector exhaustion. Specifically, in the old model all device interrupts were routed onto the boot CPU during boot. Later after the APs were released at SI_SUB_SMP, interrupts were redistributed across all CPUs. However, several drivers for multiqueue hardware allocate N interrupts per CPU in the system. In a system with many CPUs, just a few drivers doing this could exhaust the available pool of interrupt vectors on the boot CPU as each driver was allocating N * mp_ncpu vectors on the boot CPU. Now, drivers will allocate interrupts on their desired CPUs during boot meaning that only N interrupts are allocated from the boot CPU instead of N * mp_ncpu. Some other bits of code can also be simplified as smp_started is now true much earlier and will now always be true for these bits of code. This removes the need to treat the single-CPU boot environment as a special case. As a transition aid, the new behavior is available under a new kernel option (EARLY_AP_STARTUP). This will allow the option to be turned off if need be during initial testing. I plan to enable this on x86 by default in a followup commit in the next few days and to have all platforms moved over before 11.0. Once the transition is complete, the option will be removed along with the !EARLY_AP_STARTUP code. These changes have only been tested on x86. Other platform maintainers are encouraged to port their architectures over as well. The main things to check for are any uses of smp_started in MD code that can be simplified and SI_SUB_SMP SYSINITs in MD code that can be removed in the EARLY_AP_STARTUP case (e.g. the interrupt shuffling). PR: kern/199321 Reviewed by: markj, gnn, kib Sponsored by: Netflix
2016-05-14 18:22:52 +00:00
#endif
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
Temporarily revert the new-bus locking for 8.0 release. It will be reintroduced after HEAD is reopened for commits by re@. Approved by: re (kib), attilio
2009-08-20 19:17:53 +00:00
/*
* Be sure to hold Giant across DEVICE_SUSPEND/RESUME since non-MPSAFE
* drivers need this.
*/
mtx_lock(&Giant);
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
slp_state = ACPI_SS_NONE;
Don't do sleep state transition if specified sleep state is not supported by the system.
2001-07-22 19:13:54 +00:00
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
sc->acpi_sstate = state;
Add device(power/sleep button and lid) wake function from sleeping state. This is required for some Thinkpad (and maybe VAIO) machines to wake the system up from sleep. Currently partially implemented, more complete implementation will come later.
2002-07-22 12:52:54 +00:00
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
/* Enable any GPEs as appropriate and requested by the user. */
acpi_wake_prep_walk(state);
slp_state = ACPI_SS_GPE_SET;
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
/*
* Inform all devices that we are going to sleep. If at least one
* device fails, DEVICE_SUSPEND() automatically resumes the tree.
*
* XXX Note that a better two-pass approach with a 'veto' pass
* followed by a "real thing" pass would be better, but the current
* bus interface does not provide for this.
*/
if (DEVICE_SUSPEND(root_bus) != 0) {
device_printf(sc->acpi_dev, "device_suspend failed\n");
goto backout;
}
slp_state = ACPI_SS_DEV_SUSPEND;
Yet another synch with minor changes in the ACPI CA 20011120 snapshot. We need to call AcpiEnterSleepStatePrep() before AcpiEnterSleepState().
2001-11-28 12:09:42 +00:00
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
status = AcpiEnterSleepStatePrep(state);
if (ACPI_FAILURE(status)) {
device_printf(sc->acpi_dev, "AcpiEnterSleepStatePrep failed - %s\n",
AcpiFormatException(status));
goto backout;
}
slp_state = ACPI_SS_SLP_PREP;
Yet another synch with minor changes in the ACPI CA 20011120 snapshot. We need to call AcpiEnterSleepStatePrep() before AcpiEnterSleepState().
2001-11-28 12:09:42 +00:00
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
if (sc->acpi_sleep_delay > 0)
DELAY(sc->acpi_sleep_delay * 1000000);
Add new sysctl MIB (hw.acpi.sleep_delay) to specify the delay (in seconds) before ACPI sleep. Some machines might need this to sleep by Hot-key.
2002-08-25 06:13:53 +00:00
stop and restart kernel event timers in the suspend / resume cycle I have a system that is very unstable after resuming from suspend-to-RAM but only if HPET is used as the event timer. The theory is that SMM code / firmware could be enabling HPET for its own uses and unexpected interrupts cause a trouble for it. Originally I wanted to solve the problem in hpet_suspend() method, but that was insufficient as the event timer could get reprogrammed again. So, it's better, for my case and in general, to stop the event timer(s) before entering the hardware suspend. MFC after: 4 weeks Differential Revision: https://reviews.freebsd.org/D15413
2018-05-21 20:23:04 +00:00
suspendclock();
Execute AcpiLeaveSleepStatePrep() for S1 and reduce code duplication. MFC after: 3 days
2012-06-01 17:00:33 +00:00
intr = intr_disable();
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
if (state != ACPI_STATE_S1) {
Call AcpiLeaveSleepStatePrep() in interrupt disabled context (described in ACPICA source code). - Move intr_disable() and intr_restore() from acpi_wakeup.c to acpi.c and call AcpiLeaveSleepStatePrep() in interrupt disabled context. - Add acpi_wakeup_machdep() to execute wakeup MD procedures and call it twice in interrupt disabled/enabled context (ia64 version is just dummy). - Rename wakeup_cpus variable in acpi_sleep_machdep() to suspcpus in order to be shared by acpi_sleep_machdep() and acpi_wakeup_machdep(). - Move identity mapping related code to acpi_install_wakeup_handler() (i386 version) for preparation of x86/acpica/acpi_wakeup.c (MFC candidate). Reviewed by: jkim@ MFC after: 2 days
2012-06-01 15:26:32 +00:00
sleep_result = acpi_sleep_machdep(sc, state);
acpi_wakeup_machdep(sc, state, sleep_result, 0);
acpi: after wakeup from a state > S1 re-enable SCI_EN with a direct write This hack is picked up from Linux, which claims that it follows Windows behavior. PR: amd64/174409 Tested by: Sergey V. Dyatko <sergey.dyatko@gmail.com>, KAHO Toshikazu <kaho@elam.kais.kyoto-u.ac.jp>, Slawa Olhovchenkov <slw@zxy.spb.ru> MFC after: 13 days
2013-02-02 12:42:07 +00:00
/*
* XXX According to ACPI specification SCI_EN bit should be restored
* by ACPI platform (BIOS, firmware) to its pre-sleep state.
* Unfortunately some BIOSes fail to do that and that leads to
* unexpected and serious consequences during wake up like a system
* getting stuck in SMI handlers.
* This hack is picked up from Linux, which claims that it follows
* Windows behavior.
*/
if (sleep_result == 1 && state != ACPI_STATE_S4)
AcpiWriteBitRegister(ACPI_BITREG_SCI_ENABLE, ACPI_ENABLE_EVENT);
acpi: clear power button status bit after waking up... so that it is not confused for a new power off request. Learned from: Linux and ACPI specification Tested by: gjb MFC after: 12 days
2013-02-02 12:44:19 +00:00
if (sleep_result == 1 && state == ACPI_STATE_S3) {
/*
* Prevent mis-interpretation of the wakeup by power button
* as a request for power off.
* Ideally we should post an appropriate wakeup event,
* perhaps using acpi_event_power_button_wake or alike.
*
* Clearing of power button status after wakeup is mandated
* by ACPI specification in section "Fixed Power Button".
*
* XXX As of ACPICA 20121114 AcpiGetEventStatus provides
* status as 0/1 corressponding to inactive/active despite
* its type being ACPI_EVENT_STATUS. In other words,
* we should not test for ACPI_EVENT_FLAG_SET for time being.
*/
if (ACPI_SUCCESS(AcpiGetEventStatus(ACPI_EVENT_POWER_BUTTON,
&power_button_status)) && power_button_status != 0) {
AcpiClearEvent(ACPI_EVENT_POWER_BUTTON);
device_printf(sc->acpi_dev,
"cleared fixed power button status\n");
}
}
Call AcpiLeaveSleepStatePrep() in interrupt disabled context (described in ACPICA source code). - Move intr_disable() and intr_restore() from acpi_wakeup.c to acpi.c and call AcpiLeaveSleepStatePrep() in interrupt disabled context. - Add acpi_wakeup_machdep() to execute wakeup MD procedures and call it twice in interrupt disabled/enabled context (ia64 version is just dummy). - Rename wakeup_cpus variable in acpi_sleep_machdep() to suspcpus in order to be shared by acpi_sleep_machdep() and acpi_wakeup_machdep(). - Move identity mapping related code to acpi_install_wakeup_handler() (i386 version) for preparation of x86/acpica/acpi_wakeup.c (MFC candidate). Reviewed by: jkim@ MFC after: 2 days
2012-06-01 15:26:32 +00:00
intr_restore(intr);
/* call acpi_wakeup_machdep() again with interrupt enabled */
acpi_wakeup_machdep(sc, state, sleep_result, 1);
call AcpiLeaveSleepStatePrep after re-enabling interrupts I want to do this change because this call (actually, AcpiHwLegacyWakePrep) does a memory allocation and ACPI namespace evaluation. Although it is not very likely to run into any trouble, it is still not safe to make those calls with interrupts disabled. witness(4) and malloc(9) do not currently check for a context with interrupts disabled via intr_disable and we lack a facility for doing that. So, those unsafe operations fly under the radar. But if intr_disable in acpi_EnterSleepState was replaced with spinlock_enter (which it probably should be), then witness and malloc would immediately complain. Also, AcpiLeaveSleepStatePrep is documented as called when interrupts are enabled. It used to require disabled interrupts, but that requirement was changed a long time ago when support for _BFS and _GTS was removed from ACPICA. The ACPI wakeup sequence is very sensitive to changes. I consider this change to be correct, but there can be fallouts from it. What AcpiHwLegacyWakePrep essentially does is writing a value corresponding to S0 into SLP_TYPx bits of PM1 Control Register(s). According to ACPI specifications that write should be a NOP as SLP_EN bit is not set. But I see in some chipset specifications that they allow to ignore SLP_EN altogether and to act on a change of SLP_TYPx alone. Also, there are a couple of accesses to ACPI hardware before the new location of the call to AcpiLeaveSleepStatePrep. One is to clear the power button status and the other is to enable SCI. So, the move may affect the interaction between then OS and ACPI platform. I have not seen any regressions on my test system, but it's a desktop. MFC after: 5 weeks
2018-06-01 09:44:23 +00:00
AcpiLeaveSleepStatePrep(state);
Call AcpiLeaveSleepStatePrep() in interrupt disabled context (described in ACPICA source code). - Move intr_disable() and intr_restore() from acpi_wakeup.c to acpi.c and call AcpiLeaveSleepStatePrep() in interrupt disabled context. - Add acpi_wakeup_machdep() to execute wakeup MD procedures and call it twice in interrupt disabled/enabled context (ia64 version is just dummy). - Rename wakeup_cpus variable in acpi_sleep_machdep() to suspcpus in order to be shared by acpi_sleep_machdep() and acpi_wakeup_machdep(). - Move identity mapping related code to acpi_install_wakeup_handler() (i386 version) for preparation of x86/acpica/acpi_wakeup.c (MFC candidate). Reviewed by: jkim@ MFC after: 2 days
2012-06-01 15:26:32 +00:00
if (sleep_result == -1)
Merge ACPICA 20120215.
2012-02-16 22:59:29 +00:00
goto backout;
Add ACPI S2-S4BIOS Suspend/Resume code. Some problems may remain. Reviewed by:iwasaki
2001-07-20 06:07:34 +00:00
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
/* Re-enable ACPI hardware on wakeup from sleep state 4. */
if (state == ACPI_STATE_S4)
AcpiEnable();
} else {
Merge ACPICA 20120816.
2012-08-16 20:54:52 +00:00
status = AcpiEnterSleepState(state);
Restore interrupt state after executing AcpiEnterSleepState().
2012-03-27 23:26:58 +00:00
intr_restore(intr);
call AcpiLeaveSleepStatePrep after re-enabling interrupts I want to do this change because this call (actually, AcpiHwLegacyWakePrep) does a memory allocation and ACPI namespace evaluation. Although it is not very likely to run into any trouble, it is still not safe to make those calls with interrupts disabled. witness(4) and malloc(9) do not currently check for a context with interrupts disabled via intr_disable and we lack a facility for doing that. So, those unsafe operations fly under the radar. But if intr_disable in acpi_EnterSleepState was replaced with spinlock_enter (which it probably should be), then witness and malloc would immediately complain. Also, AcpiLeaveSleepStatePrep is documented as called when interrupts are enabled. It used to require disabled interrupts, but that requirement was changed a long time ago when support for _BFS and _GTS was removed from ACPICA. The ACPI wakeup sequence is very sensitive to changes. I consider this change to be correct, but there can be fallouts from it. What AcpiHwLegacyWakePrep essentially does is writing a value corresponding to S0 into SLP_TYPx bits of PM1 Control Register(s). According to ACPI specifications that write should be a NOP as SLP_EN bit is not set. But I see in some chipset specifications that they allow to ignore SLP_EN altogether and to act on a change of SLP_TYPx alone. Also, there are a couple of accesses to ACPI hardware before the new location of the call to AcpiLeaveSleepStatePrep. One is to clear the power button status and the other is to enable SCI. So, the move may affect the interaction between then OS and ACPI platform. I have not seen any regressions on my test system, but it's a desktop. MFC after: 5 weeks
2018-06-01 09:44:23 +00:00
AcpiLeaveSleepStatePrep(state);
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
if (ACPI_FAILURE(status)) {
device_printf(sc->acpi_dev, "AcpiEnterSleepState failed - %s\n",
AcpiFormatException(status));
goto backout;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
}
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
slp_state = ACPI_SS_SLEPT;
Disable sleep requests for 5 sec after wakeup. This is needed for some Toshiba and Thinkpad laptops. Wakeup event is generated by power button or sleep button on some laptops but this also generates SCI interrupt, and shutdown the system as result. So this is introduced so that acpi driver ignore given requests for certain period.
2001-12-09 18:02:36 +00:00
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
/*
* Back out state according to how far along we got in the suspend
* process. This handles both the error and success cases.
*/
acpi: do not run resume/backout code when entering S0/S5 states This change adds (possibly redundant) early check for invalid state input parameter (including S0). Handling of S5 request is reduced to simply calling shutdown_nice(). As a result control flow of acpi_EnterSleepState is somewhat simplified and resume/backout half of the function is not executed for S5 (soft poweroff) request and invalid state requests. Note: it seems that shutdown_nice may act as nop when initproc is already initialized (to grab pid of 1), but init process is in "pre-natal" state. Tested by: Fabian Keil <fk@fabiankeil.de> Reviewed by: njl, jkim Approved by: rpaulo
2009-04-28 11:56:54 +00:00
backout:
stop and restart kernel event timers in the suspend / resume cycle I have a system that is very unstable after resuming from suspend-to-RAM but only if HPET is used as the event timer. The theory is that SMM code / firmware could be enabling HPET for its own uses and unexpected interrupts cause a trouble for it. Originally I wanted to solve the problem in hpet_suspend() method, but that was insufficient as the event timer could get reprogrammed again. So, it's better, for my case and in general, to stop the event timer(s) before entering the hardware suspend. MFC after: 4 weeks Differential Revision: https://reviews.freebsd.org/D15413
2018-05-21 20:23:04 +00:00
if (slp_state >= ACPI_SS_SLP_PREP)
resumeclock();
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
if (slp_state >= ACPI_SS_GPE_SET) {
acpi_wake_prep_walk(state);
sc->acpi_sstate = ACPI_STATE_S0;
}
Reorder resume procedures. DEVICE_RESUME() should be done before AcpiLeaveSleepState() because PCI config space evaluation can be occurred during control method executions. This should fix one of the hang up problems on resuming. MFC after: 3 days
2012-05-29 05:28:34 +00:00
if (slp_state >= ACPI_SS_DEV_SUSPEND)
DEVICE_RESUME(root_bus);
Call AcpiLeaveSleepStatePrep() in interrupt disabled context (described in ACPICA source code). - Move intr_disable() and intr_restore() from acpi_wakeup.c to acpi.c and call AcpiLeaveSleepStatePrep() in interrupt disabled context. - Add acpi_wakeup_machdep() to execute wakeup MD procedures and call it twice in interrupt disabled/enabled context (ia64 version is just dummy). - Rename wakeup_cpus variable in acpi_sleep_machdep() to suspcpus in order to be shared by acpi_sleep_machdep() and acpi_wakeup_machdep(). - Move identity mapping related code to acpi_install_wakeup_handler() (i386 version) for preparation of x86/acpica/acpi_wakeup.c (MFC candidate). Reviewed by: jkim@ MFC after: 2 days
2012-06-01 15:26:32 +00:00
if (slp_state >= ACPI_SS_SLP_PREP)
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
AcpiLeaveSleepState(state);
Reset clock after atrtc(4) is properly resumed.
2012-02-08 21:23:20 +00:00
if (slp_state >= ACPI_SS_SLEPT) {
re-synchronize TSC-s on SMP systems after resume, if necessary The TSC-s are checked and synchronized only if they were good originally. That is, invariant, synchronized, etc. This is necessary on an AMD-based system where after a wakeup from STR I see that BSP clock differs from AP clocks by a count that roughly corresponds to one second. The APs are in sync with each other. Not sure if this is a hardware quirk or a firmware bug. This is what I see after a resume with this change: SMP: passed TSC synchronization test after adjustment acpi_timer0: restoring timecounter, ACPI-fast -> TSC-low Reviewed by: kib MFC after: 3 weeks Differential Revision: https://reviews.freebsd.org/D15551
2018-05-25 07:33:20 +00:00
#if defined(__i386__) || defined(__amd64__)
/* NB: we are still using ACPI timecounter at this point. */
resume_TSC();
#endif
Reset clock after atrtc(4) is properly resumed.
2012-02-08 21:23:20 +00:00
acpi_resync_clock(sc);
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
acpi_enable_fixed_events(sc);
Reset clock after atrtc(4) is properly resumed.
2012-02-08 21:23:20 +00:00
}
If there is any pending sleep request, disallow entering S5 state. Otherwise, bad things may happen. ;-)
2010-07-15 23:24:06 +00:00
sc->acpi_next_sstate = 0;
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
Temporarily revert the new-bus locking for 8.0 release. It will be reintroduced after HEAD is reopened for commits by re@. Approved by: re (kib), attilio
2009-08-20 19:17:53 +00:00
mtx_unlock(&Giant);
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
Add an EARLY_AP_STARTUP option to start APs earlier during boot. Currently, Application Processors (non-boot CPUs) are started by MD code at SI_SUB_CPU, but they are kept waiting in a "pen" until SI_SUB_SMP at which point they are released to run kernel threads. SI_SUB_SMP is one of the last SYSINIT levels, so APs don't enter the scheduler and start running threads until fairly late in the boot. This change moves SI_SUB_SMP up to just before software interrupt threads are created allowing the APs to start executing kernel threads much sooner (before any devices are probed). This allows several initialization routines that need to perform initialization on all CPUs to now perform that initialization in one step rather than having to defer the AP initialization to a second SYSINIT run at SI_SUB_SMP. It also permits all CPUs to be available for handling interrupts before any devices are probed. This last feature fixes a problem on with interrupt vector exhaustion. Specifically, in the old model all device interrupts were routed onto the boot CPU during boot. Later after the APs were released at SI_SUB_SMP, interrupts were redistributed across all CPUs. However, several drivers for multiqueue hardware allocate N interrupts per CPU in the system. In a system with many CPUs, just a few drivers doing this could exhaust the available pool of interrupt vectors on the boot CPU as each driver was allocating N * mp_ncpu vectors on the boot CPU. Now, drivers will allocate interrupts on their desired CPUs during boot meaning that only N interrupts are allocated from the boot CPU instead of N * mp_ncpu. Some other bits of code can also be simplified as smp_started is now true much earlier and will now always be true for these bits of code. This removes the need to treat the single-CPU boot environment as a special case. As a transition aid, the new behavior is available under a new kernel option (EARLY_AP_STARTUP). This will allow the option to be turned off if need be during initial testing. I plan to enable this on x86 by default in a followup commit in the next few days and to have all platforms moved over before 11.0. Once the transition is complete, the option will be removed along with the !EARLY_AP_STARTUP code. These changes have only been tested on x86. Other platform maintainers are encouraged to port their architectures over as well. The main things to check for are any uses of smp_started in MD code that can be simplified and SI_SUB_SMP SYSINITs in MD code that can be removed in the EARLY_AP_STARTUP case (e.g. the interrupt shuffling). PR: kern/199321 Reviewed by: markj, gnn, kib Sponsored by: Netflix
2016-05-14 18:22:52 +00:00
#ifdef EARLY_AP_STARTUP
thread_lock(curthread);
sched_unbind(curthread);
thread_unlock(curthread);
#else
Make SMP code path conditional at run-time.
2010-07-13 16:35:41 +00:00
if (smp_started) {
thread_lock(curthread);
sched_unbind(curthread);
thread_unlock(curthread);
}
Add an EARLY_AP_STARTUP option to start APs earlier during boot. Currently, Application Processors (non-boot CPUs) are started by MD code at SI_SUB_CPU, but they are kept waiting in a "pen" until SI_SUB_SMP at which point they are released to run kernel threads. SI_SUB_SMP is one of the last SYSINIT levels, so APs don't enter the scheduler and start running threads until fairly late in the boot. This change moves SI_SUB_SMP up to just before software interrupt threads are created allowing the APs to start executing kernel threads much sooner (before any devices are probed). This allows several initialization routines that need to perform initialization on all CPUs to now perform that initialization in one step rather than having to defer the AP initialization to a second SYSINIT run at SI_SUB_SMP. It also permits all CPUs to be available for handling interrupts before any devices are probed. This last feature fixes a problem on with interrupt vector exhaustion. Specifically, in the old model all device interrupts were routed onto the boot CPU during boot. Later after the APs were released at SI_SUB_SMP, interrupts were redistributed across all CPUs. However, several drivers for multiqueue hardware allocate N interrupts per CPU in the system. In a system with many CPUs, just a few drivers doing this could exhaust the available pool of interrupt vectors on the boot CPU as each driver was allocating N * mp_ncpu vectors on the boot CPU. Now, drivers will allocate interrupts on their desired CPUs during boot meaning that only N interrupts are allocated from the boot CPU instead of N * mp_ncpu. Some other bits of code can also be simplified as smp_started is now true much earlier and will now always be true for these bits of code. This removes the need to treat the single-CPU boot environment as a special case. As a transition aid, the new behavior is available under a new kernel option (EARLY_AP_STARTUP). This will allow the option to be turned off if need be during initial testing. I plan to enable this on x86 by default in a followup commit in the next few days and to have all platforms moved over before 11.0. Once the transition is complete, the option will be removed along with the !EARLY_AP_STARTUP code. These changes have only been tested on x86. Other platform maintainers are encouraged to port their architectures over as well. The main things to check for are any uses of smp_started in MD code that can be simplified and SI_SUB_SMP SYSINITs in MD code that can be removed in the EARLY_AP_STARTUP case (e.g. the interrupt shuffling). PR: kern/199321 Reviewed by: markj, gnn, kib Sponsored by: Netflix
2016-05-14 18:22:52 +00:00
#endif
Initial suspend/resume support for amd64. This code is heavily inspired by Takanori Watanabe's experimental SMP patch for i386 and large portion was shamelessly cut and pasted from Peter Wemm's AP boot code.
2009-03-17 00:48:11 +00:00
hook userland threads suspend + resume into acpi suspend code Also, split power_suspend into power_suspend and power_suspend_early. power_suspend_early is called before the userland is frozen. power_suspend is called after the userland is frozen. Currently only VT switching is hooked to power_suspend_early. This is needed because switching away from X server requires its cooperation, so obviously X server must not be frozen when that happens. Freezing userland during ACPI suspend is useful because not all drivers correctly handle suspension concurrent with other activity. This is especially applicable to drivers ported from other operating systems that suspend all software activity between placing drivers and hardware into suspended state. In particular drm2/radeon (radeonkms) depends on the described procedure. The driver does not have any internal synchronization between suspension activities and processing of userland requests. Many thanks to kib for the code that allows to freeze and thaw all userland threads. Note that ideally we also need to park / inhibit (non-special) kernel threads as well to ensure that they do not call into drivers. MFC after: 17 days
2015-01-27 17:33:18 +00:00
resume_all_proc();
Add event handlers for (ACPI) suspend/resume events. Suspend event handlers are invoked right before device drivers go into sleep state and resume event handlers are invoked right after all device drivers are waken up.
2011-04-14 22:17:39 +00:00
EVENTHANDLER_INVOKE(power_resume);
Check whether devd is running before calling resume notifier and reshuffle code to reduce unnecessary locking coverage.
2009-03-23 22:06:09 +00:00
/* Allow another sleep request after a while. */
Convert from timeout(9) to callout(9).
2014-09-22 14:27:26 +00:00
callout_schedule(&acpi_sleep_timer, hz * ACPI_MINIMUM_AWAKETIME);
Check whether devd is running before calling resume notifier and reshuffle code to reduce unnecessary locking coverage.
2009-03-23 22:06:09 +00:00
/* Run /etc/rc.resume after we are back. */
if (devctl_process_running())
acpi_UserNotify("Resume", ACPI_ROOT_OBJECT, state);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return_ACPI_STATUS (status);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Reset clock after atrtc(4) is properly resumed.
2012-02-08 21:23:20 +00:00
static void
Add a function to reset system time after resuming, which will be used by amd64 shortly. It can be turned off by setting "debug.acpi.reset_clock" tunable to zero.
2009-03-23 22:12:33 +00:00
acpi_resync_clock(struct acpi_softc *sc)
{
/*
* Warm up timecounter again and reset system clock.
*/
(void)timecounter->tc_get_timecount(timecounter);
inittodr(time_second + sc->acpi_sleep_delay);
}
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
/* Enable or disable the device's wake GPE. */
int
acpi_wake_set_enable(device_t dev, int enable)
{
struct acpi_prw_data prw;
ACPI_STATUS status;
int flags;
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
/* Make sure the device supports waking the system and get the GPE. */
Catch up with ACPI-CA 20070320 import.
2007-03-22 18:16:43 +00:00
if (acpi_parse_prw(acpi_get_handle(dev), &prw) != 0)
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
return (ENXIO);
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
flags = acpi_get_flags(dev);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
if (enable) {
Merge ACPICA 20101209.
2010-12-15 23:48:45 +00:00
status = AcpiSetGpeWakeMask(prw.gpe_handle, prw.gpe_bit,
ACPI_GPE_ENABLE);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
if (ACPI_FAILURE(status)) {
device_printf(dev, "enable wake failed\n");
return (ENXIO);
}
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
acpi_set_flags(dev, flags | ACPI_FLAG_WAKE_ENABLED);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
} else {
Merge ACPICA 20101209.
2010-12-15 23:48:45 +00:00
status = AcpiSetGpeWakeMask(prw.gpe_handle, prw.gpe_bit,
ACPI_GPE_DISABLE);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
if (ACPI_FAILURE(status)) {
device_printf(dev, "disable wake failed\n");
return (ENXIO);
}
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
acpi_set_flags(dev, flags & ~ACPI_FLAG_WAKE_ENABLED);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
}
return (0);
}
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
static int
acpi_wake_sleep_prep(ACPI_HANDLE handle, int sstate)
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
{
struct acpi_prw_data prw;
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
device_t dev;
Update the new suspend/resume GPE methods to properly limit the GPE based on the destination sleep state. Add a method to restore the old state on resume. This is needed for the case of suspending to a very low state disabling a GPE (i.e. S4), resuming, and then suspending to a higher state (i.e. S3). This case should now keep the proper GPEs enabled.
2004-05-28 07:04:09 +00:00
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
/* Check that this is a wake-capable device and get its GPE. */
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
if (acpi_parse_prw(handle, &prw) != 0)
return (ENXIO);
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
dev = acpi_get_device(handle);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
/*
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
* The destination sleep state must be less than (i.e., higher power)
* or equal to the value specified by _PRW. If this GPE cannot be
* enabled for the next sleep state, then disable it. If it can and
* the user requested it be enabled, turn on any required power resources
* and set _PSW.
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
*/
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
if (sstate > prw.lowest_wake) {
Merge ACPICA 20101209.
2010-12-15 23:48:45 +00:00
AcpiSetGpeWakeMask(prw.gpe_handle, prw.gpe_bit, ACPI_GPE_DISABLE);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
if (bootverbose)
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
device_printf(dev, "wake_prep disabled wake for %s (S%d)\n",
acpi_name(handle), sstate);
} else if (dev && (acpi_get_flags(dev) & ACPI_FLAG_WAKE_ENABLED) != 0) {
acpi_pwr_wake_enable(handle, 1);
Update the new suspend/resume GPE methods to properly limit the GPE based on the destination sleep state. Add a method to restore the old state on resume. This is needed for the case of suspending to a very low state disabling a GPE (i.e. S4), resuming, and then suspending to a higher state (i.e. S3). This case should now keep the proper GPEs enabled.
2004-05-28 07:04:09 +00:00
acpi_SetInteger(handle, "_PSW", 1);
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
if (bootverbose)
device_printf(dev, "wake_prep enabled for %s (S%d)\n",
acpi_name(handle), sstate);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
}
Update the new suspend/resume GPE methods to properly limit the GPE based on the destination sleep state. Add a method to restore the old state on resume. This is needed for the case of suspending to a very low state disabling a GPE (i.e. S4), resuming, and then suspending to a higher state (i.e. S3). This case should now keep the proper GPEs enabled.
2004-05-28 07:04:09 +00:00
return (0);
}
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
static int
acpi_wake_run_prep(ACPI_HANDLE handle, int sstate)
Update the new suspend/resume GPE methods to properly limit the GPE based on the destination sleep state. Add a method to restore the old state on resume. This is needed for the case of suspending to a very low state disabling a GPE (i.e. S4), resuming, and then suspending to a higher state (i.e. S3). This case should now keep the proper GPEs enabled.
2004-05-28 07:04:09 +00:00
{
struct acpi_prw_data prw;
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
device_t dev;
Update the new suspend/resume GPE methods to properly limit the GPE based on the destination sleep state. Add a method to restore the old state on resume. This is needed for the case of suspending to a very low state disabling a GPE (i.e. S4), resuming, and then suspending to a higher state (i.e. S3). This case should now keep the proper GPEs enabled.
2004-05-28 07:04:09 +00:00
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
/*
* Check that this is a wake-capable device and get its GPE. Return
* now if the user didn't enable this device for wake.
*/
Update the new suspend/resume GPE methods to properly limit the GPE based on the destination sleep state. Add a method to restore the old state on resume. This is needed for the case of suspending to a very low state disabling a GPE (i.e. S4), resuming, and then suspending to a higher state (i.e. S3). This case should now keep the proper GPEs enabled.
2004-05-28 07:04:09 +00:00
if (acpi_parse_prw(handle, &prw) != 0)
return (ENXIO);
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
dev = acpi_get_device(handle);
if (dev == NULL || (acpi_get_flags(dev) & ACPI_FLAG_WAKE_ENABLED) == 0)
return (0);
Update the new suspend/resume GPE methods to properly limit the GPE based on the destination sleep state. Add a method to restore the old state on resume. This is needed for the case of suspending to a very low state disabling a GPE (i.e. S4), resuming, and then suspending to a higher state (i.e. S3). This case should now keep the proper GPEs enabled.
2004-05-28 07:04:09 +00:00
/*
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
* If this GPE couldn't be enabled for the previous sleep state, it was
* disabled before going to sleep so re-enable it. If it was enabled,
* clear _PSW and turn off any power resources it used.
Update the new suspend/resume GPE methods to properly limit the GPE based on the destination sleep state. Add a method to restore the old state on resume. This is needed for the case of suspending to a very low state disabling a GPE (i.e. S4), resuming, and then suspending to a higher state (i.e. S3). This case should now keep the proper GPEs enabled.
2004-05-28 07:04:09 +00:00
*/
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
if (sstate > prw.lowest_wake) {
Merge ACPICA 20101209.
2010-12-15 23:48:45 +00:00
AcpiSetGpeWakeMask(prw.gpe_handle, prw.gpe_bit, ACPI_GPE_ENABLE);
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
if (bootverbose)
device_printf(dev, "run_prep re-enabled %s\n", acpi_name(handle));
} else {
acpi_SetInteger(handle, "_PSW", 0);
acpi_pwr_wake_enable(handle, 0);
if (bootverbose)
device_printf(dev, "run_prep cleaned up for %s\n",
acpi_name(handle));
}
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
return (0);
}
static ACPI_STATUS
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
acpi_wake_prep(ACPI_HANDLE handle, UINT32 level, void *context, void **status)
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
{
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
int sstate;
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
/* If suspending, run the sleep prep function, otherwise wake. */
sstate = *(int *)context;
if (AcpiGbl_SystemAwakeAndRunning)
acpi_wake_sleep_prep(handle, sstate);
else
acpi_wake_run_prep(handle, sstate);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
return (AE_OK);
}
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
/* Walk the tree rooted at acpi0 to prep devices for suspend/resume. */
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
static int
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
acpi_wake_prep_walk(int sstate)
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
{
ACPI_HANDLE sb_handle;
if (ACPI_SUCCESS(AcpiGetHandle(ACPI_ROOT_OBJECT, "\\_SB_", &sb_handle)))
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
AcpiWalkNamespace(ACPI_TYPE_DEVICE, sb_handle, 100,
Merge ACPICA 20091112.
2009-11-16 21:47:12 +00:00
acpi_wake_prep, NULL, &sstate, NULL);
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
return (0);
}
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
/* Walk the tree rooted at acpi0 to attach per-device wake sysctls. */
static int
acpi_wake_sysctl_walk(device_t dev)
{
int error, i, numdevs;
device_t *devlist;
device_t child;
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
ACPI_STATUS status;
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
error = device_get_children(dev, &devlist, &numdevs);
If devclass_get_devices() returns success but a count of 0, free the pointer. If kernel malloc(0) returns a valid pointer, it needs to be freed. If it returns NULL, it's ok to free this also. Submitted by: pjd Reviewed by: imp, dfr Obtained from: Coverity Prevent
2005-05-20 05:00:43 +00:00
if (error != 0 || numdevs == 0) {
if (numdevs == 0)
free(devlist, M_TEMP);
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
return (error);
If devclass_get_devices() returns success but a count of 0, free the pointer. If kernel malloc(0) returns a valid pointer, it needs to be freed. If it returns NULL, it's ok to free this also. Submitted by: pjd Reviewed by: imp, dfr Obtained from: Coverity Prevent
2005-05-20 05:00:43 +00:00
}
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
for (i = 0; i < numdevs; i++) {
child = devlist[i];
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
acpi_wake_sysctl_walk(child);
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
if (!device_is_attached(child))
continue;
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
status = AcpiEvaluateObject(acpi_get_handle(child), "_PRW", NULL, NULL);
if (ACPI_SUCCESS(status)) {
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
SYSCTL_ADD_PROC(device_get_sysctl_ctx(child),
SYSCTL_CHILDREN(device_get_sysctl_tree(child)), OID_AUTO,
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
"wake", CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_NEEDGIANT, child, 0,
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
acpi_wake_set_sysctl, "I", "Device set to wake the system");
}
}
free(devlist, M_TEMP);
return (0);
}
/* Enable or disable wake from userland. */
static int
acpi_wake_set_sysctl(SYSCTL_HANDLER_ARGS)
{
int enable, error;
device_t dev;
dev = (device_t)arg1;
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
enable = (acpi_get_flags(dev) & ACPI_FLAG_WAKE_ENABLED) ? 1 : 0;
Attach per-device sysctls to allow users to set whether or not a given device can wake the system. For example: dev.root0.nexus0.acpi0.acpi_lid0.wake: 1 dev.root0.nexus0.acpi0.acpi_button0.wake: 1 dev.root0.nexus0.acpi0.pcib0.wake: 0 dev.root0.nexus0.acpi0.sio0.wake: 0
2004-05-28 06:28:55 +00:00
error = sysctl_handle_int(oidp, &enable, 0, req);
if (error != 0 || req->newptr == NULL)
return (error);
if (enable != 0 && enable != 1)
return (EINVAL);
return (acpi_wake_set_enable(dev, enable));
}
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
/* Parse a device's _PRW into a structure. */
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
int
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
acpi_parse_prw(ACPI_HANDLE h, struct acpi_prw_data *prw)
{
ACPI_STATUS status;
ACPI_BUFFER prw_buffer;
ACPI_OBJECT *res, *res2;
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
int error, i, power_count;
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
if (h == NULL || prw == NULL)
return (EINVAL);
/*
* The _PRW object (7.2.9) is only required for devices that have the
* ability to wake the system from a sleeping state.
*/
error = EINVAL;
prw_buffer.Pointer = NULL;
prw_buffer.Length = ACPI_ALLOCATE_BUFFER;
status = AcpiEvaluateObject(h, "_PRW", NULL, &prw_buffer);
if (ACPI_FAILURE(status))
return (ENOENT);
res = (ACPI_OBJECT *)prw_buffer.Pointer;
if (res == NULL)
return (ENOENT);
if (!ACPI_PKG_VALID(res, 2))
goto out;
/*
* Element 1 of the _PRW object:
* The lowest power system sleeping state that can be entered while still
* providing wake functionality. The sleeping state being entered must
* be less than (i.e., higher power) or equal to this value.
*/
if (acpi_PkgInt32(res, 1, &prw->lowest_wake) != 0)
goto out;
/*
* Element 0 of the _PRW object:
*/
switch (res->Package.Elements[0].Type) {
case ACPI_TYPE_INTEGER:
/*
* If the data type of this package element is numeric, then this
* _PRW package element is the bit index in the GPEx_EN, in the
* GPE blocks described in the FADT, of the enable bit that is
* enabled for the wake event.
*/
prw->gpe_handle = NULL;
prw->gpe_bit = res->Package.Elements[0].Integer.Value;
error = 0;
break;
case ACPI_TYPE_PACKAGE:
/*
* If the data type of this package element is a package, then this
* _PRW package element is itself a package containing two
* elements. The first is an object reference to the GPE Block
* device that contains the GPE that will be triggered by the wake
* event. The second element is numeric and it contains the bit
* index in the GPEx_EN, in the GPE Block referenced by the
* first element in the package, of the enable bit that is enabled for
* the wake event.
*
* For example, if this field is a package then it is of the form:
* Package() {\_SB.PCI0.ISA.GPE, 2}
*/
res2 = &res->Package.Elements[0];
if (!ACPI_PKG_VALID(res2, 2))
goto out;
prw->gpe_handle = acpi_GetReference(NULL, &res2->Package.Elements[0]);
if (prw->gpe_handle == NULL)
goto out;
if (acpi_PkgInt32(res2, 1, &prw->gpe_bit) != 0)
goto out;
error = 0;
break;
default:
goto out;
}
Move flags into a private ivar so it can't collide with device flags. Unify the code to disable GPEs with the enable code. Shutdown is handled the same way. ACPI now does all wake/sleep prep for child devices so now they no longer need to call external functions in the suspend/resume path. Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
/* Elements 2 to N of the _PRW object are power resources. */
power_count = res->Package.Count - 2;
if (power_count > ACPI_PRW_MAX_POWERRES) {
printf("ACPI device %s has too many power resources\n", acpi_name(h));
power_count = 0;
}
prw->power_res_count = power_count;
for (i = 0; i < power_count; i++)
prw->power_res[i] = res->Package.Elements[i];
Restructure the wake GPE API. Now there are three functions: acpi_wake_init: Evaluate _PRW and set the GPE type acpi_wake_set_enable: Enable or disable a device's GPE. acpi_wake_sleep_prep: Perform any last-minute changes to the device to prepare it for entering the given sleep state. Also, walk the entire namespace when transitioning to a sleep state, disabling any GPEs which aren't appropriate for the given state. Transition acpi_lid and acpi_button to the new API. This clears the way for non-ACPI-aware devices to wake the system (i.e. modems) and fixes a problem where systems power up after shutdown when a GPE is triggered.
2004-05-27 18:38:45 +00:00
out:
if (prw_buffer.Pointer != NULL)
AcpiOsFree(prw_buffer.Pointer);
return (error);
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
* ACPI Event Handlers
*/
/* System Event Handlers (registered by EVENTHANDLER_REGISTER) */
static void
acpi_system_eventhandler_sleep(void *arg, int state)
{
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
struct acpi_softc *sc = (struct acpi_softc *)arg;
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
int ret;
/tmp/m
2004-08-03 05:13:56 +00:00
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE_U32((char *)(uintptr_t)__func__, state);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
/* Check if button action is disabled or unknown. */
if (state == ACPI_STATE_UNKNOWN)
My previous commit introduced a spurious warning for the case where a switch (i.e. lid) is set to have an action of NONE. This is not an invalid state, so silently return. This fixes the warning: "acpi: request to enter state S6 failed (err 22)" Approved by: re
2007-06-30 17:27:31 +00:00
return;
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
/* Request that the system prepare to enter the given suspend state. */
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
ret = acpi_ReqSleepState(sc, state);
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
if (ret != 0)
Prefer device_printf() over printf() where ever possible.
2009-04-30 17:42:11 +00:00
device_printf(sc->acpi_dev,
"request to enter state S%d failed (err %d)\n", state, ret);
/tmp/m
2004-08-03 05:13:56 +00:00
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
return_VOID;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
static void
acpi_system_eventhandler_wakeup(void *arg, int state)
{
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
/tmp/m
2004-08-03 05:13:56 +00:00
ACPI_FUNCTION_TRACE_U32((char *)(uintptr_t)__func__, state);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
/tmp/m
2004-08-03 05:13:56 +00:00
/* Currently, nothing to do for wakeup. */
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
return_VOID;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
/*
* ACPICA Event Handlers (FixedEvent, also called from button notify handler)
*/
Work around a witness(4) panic introduced in r217238. Reported by: jh
2011-01-11 19:26:39 +00:00
static void
acpi_invoke_sleep_eventhandler(void *context)
{
EVENTHANDLER_INVOKE(acpi_sleep_event, *(int *)context);
}
static void
acpi_invoke_wake_eventhandler(void *context)
{
EVENTHANDLER_INVOKE(acpi_wakeup_event, *(int *)context);
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
UINT32
Prefer buttons defined in the AML over the ones in the FADT. Some systems define power/sleep buttons in both places but only deliver notifies to the ones defined in the AML. Also, reduce length of various function handler names. PR: Submitted by: Reviewed by: Approved by: Obtained from: MFC after:
2004-02-11 02:57:33 +00:00
acpi_event_power_button_sleep(void *context)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
struct acpi_softc *sc = (struct acpi_softc *)context;
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Work around a witness(4) panic introduced in r217238. Reported by: jh
2011-01-11 19:26:39 +00:00
if (ACPI_FAILURE(AcpiOsExecute(OSL_NOTIFY_HANDLER,
acpi_invoke_sleep_eventhandler, &sc->acpi_power_button_sx)))
return_VALUE (ACPI_INTERRUPT_NOT_HANDLED);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return_VALUE (ACPI_INTERRUPT_HANDLED);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
UINT32
Prefer buttons defined in the AML over the ones in the FADT. Some systems define power/sleep buttons in both places but only deliver notifies to the ones defined in the AML. Also, reduce length of various function handler names. PR: Submitted by: Reviewed by: Approved by: Obtained from: MFC after:
2004-02-11 02:57:33 +00:00
acpi_event_power_button_wake(void *context)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
struct acpi_softc *sc = (struct acpi_softc *)context;
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Work around a witness(4) panic introduced in r217238. Reported by: jh
2011-01-11 19:26:39 +00:00
if (ACPI_FAILURE(AcpiOsExecute(OSL_NOTIFY_HANDLER,
acpi_invoke_wake_eventhandler, &sc->acpi_power_button_sx)))
return_VALUE (ACPI_INTERRUPT_NOT_HANDLED);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return_VALUE (ACPI_INTERRUPT_HANDLED);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
UINT32
Prefer buttons defined in the AML over the ones in the FADT. Some systems define power/sleep buttons in both places but only deliver notifies to the ones defined in the AML. Also, reduce length of various function handler names. PR: Submitted by: Reviewed by: Approved by: Obtained from: MFC after:
2004-02-11 02:57:33 +00:00
acpi_event_sleep_button_sleep(void *context)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
struct acpi_softc *sc = (struct acpi_softc *)context;
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Work around a witness(4) panic introduced in r217238. Reported by: jh
2011-01-11 19:26:39 +00:00
if (ACPI_FAILURE(AcpiOsExecute(OSL_NOTIFY_HANDLER,
acpi_invoke_sleep_eventhandler, &sc->acpi_sleep_button_sx)))
return_VALUE (ACPI_INTERRUPT_NOT_HANDLED);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return_VALUE (ACPI_INTERRUPT_HANDLED);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
UINT32
Prefer buttons defined in the AML over the ones in the FADT. Some systems define power/sleep buttons in both places but only deliver notifies to the ones defined in the AML. Also, reduce length of various function handler names. PR: Submitted by: Reviewed by: Approved by: Obtained from: MFC after:
2004-02-11 02:57:33 +00:00
acpi_event_sleep_button_wake(void *context)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
struct acpi_softc *sc = (struct acpi_softc *)context;
Brutally deal with __func__ being 'const char *' on gcc-3.1.
2002-05-19 06:16:47 +00:00
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Work around a witness(4) panic introduced in r217238. Reported by: jh
2011-01-11 19:26:39 +00:00
if (ACPI_FAILURE(AcpiOsExecute(OSL_NOTIFY_HANDLER,
acpi_invoke_wake_eventhandler, &sc->acpi_sleep_button_sx)))
return_VALUE (ACPI_INTERRUPT_NOT_HANDLED);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return_VALUE (ACPI_INTERRUPT_HANDLED);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
/*
/tmp/m
2004-08-03 05:13:56 +00:00
* XXX This static buffer is suboptimal. There is no locking so only
* use this for single-threaded callers.
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*/
char *
acpi_name(ACPI_HANDLE handle)
{
/tmp/m
2004-08-03 05:13:56 +00:00
ACPI_BUFFER buf;
static char data[256];
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
/tmp/m
2004-08-03 05:13:56 +00:00
buf.Length = sizeof(data);
buf.Pointer = data;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Allow null handles to be passed into acpi_name().
2004-08-12 17:02:53 +00:00
if (handle && ACPI_SUCCESS(AcpiGetName(handle, ACPI_FULL_PATHNAME, &buf)))
/tmp/m
2004-08-03 05:13:56 +00:00
return (data);
return ("(unknown)");
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
/*
* Debugging/bug-avoidance. Avoid trying to fetch info on various
* parts of the namespace.
*/
int
acpi_avoid(ACPI_HANDLE handle)
{
Fix acpi_avoid() to call freeenv() on the original char * returned from getenv(). Reported by: joe Tested by: joe
2002-05-09 21:13:37 +00:00
char *cp, *env, *np;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
int len;
np = acpi_name(handle);
if (*np == '\\')
np++;
Follow up to r225617. In order to maximize the re-usability of kernel code in userland rename in-kernel getenv()/setenv() to kern_setenv()/kern_getenv(). This fixes a namespace collision with libc symbols. Submitted by: kmacy Tested by: make universe
2014-10-16 18:04:43 +00:00
if ((env = kern_getenv("debug.acpi.avoid")) == NULL)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (0);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Scan the avoid list checking for a match */
Fix acpi_avoid() to call freeenv() on the original char * returned from getenv(). Reported by: joe Tested by: joe
2002-05-09 21:13:37 +00:00
cp = env;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
for (;;) {
/tmp/m
2004-08-03 05:13:56 +00:00
while (*cp != 0 && isspace(*cp))
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
cp++;
if (*cp == 0)
break;
len = 0;
/tmp/m
2004-08-03 05:13:56 +00:00
while (cp[len] != 0 && !isspace(cp[len]))
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
len++;
Rework the kernel environment subsystem. We now convert the static environment needed at boot time to a dynamic subsystem when VM is up. The dynamic kernel environment is protected by an sx lock. This adds some new functions to manipulate the kernel environment : freeenv(), setenv(), unsetenv() and testenv(). freeenv() has to be called after every getenv() when you have finished using the string. testenv() only tests if an environment variable is present, and doesn't require a freeenv() call. setenv() and unsetenv() are self explanatory. The kenv(2) syscall exports these new functionalities to userland, mainly for kenv(1). Reviewed by: peter
2002-04-17 13:06:36 +00:00
if (!strncmp(cp, np, len)) {
Fix acpi_avoid() to call freeenv() on the original char * returned from getenv(). Reported by: joe Tested by: joe
2002-05-09 21:13:37 +00:00
freeenv(env);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
return(1);
Rework the kernel environment subsystem. We now convert the static environment needed at boot time to a dynamic subsystem when VM is up. The dynamic kernel environment is protected by an sx lock. This adds some new functions to manipulate the kernel environment : freeenv(), setenv(), unsetenv() and testenv(). freeenv() has to be called after every getenv() when you have finished using the string. testenv() only tests if an environment variable is present, and doesn't require a freeenv() call. setenv() and unsetenv() are self explanatory. The kenv(2) syscall exports these new functionalities to userland, mainly for kenv(1). Reviewed by: peter
2002-04-17 13:06:36 +00:00
}
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
cp += len;
}
Fix acpi_avoid() to call freeenv() on the original char * returned from getenv(). Reported by: joe Tested by: joe
2002-05-09 21:13:37 +00:00
freeenv(env);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (0);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
}
/*
* Debugging/bug-avoidance. Disable ACPI subsystem components.
*/
int
acpi_disabled(char *subsys)
{
Don't call freeenv() on a modified pointer. Submitted by: Alexander Kabaev <ak03@gte.com> Reviewed by: phk Pointy hat to: mux
2002-04-24 17:49:21 +00:00
char *cp, *env;
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
int len;
Follow up to r225617. In order to maximize the re-usability of kernel code in userland rename in-kernel getenv()/setenv() to kern_setenv()/kern_getenv(). This fixes a namespace collision with libc symbols. Submitted by: kmacy Tested by: make universe
2014-10-16 18:04:43 +00:00
if ((env = kern_getenv("debug.acpi.disabled")) == NULL)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (0);
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
if (strcmp(env, "all") == 0) {
Don't call freeenv() on a modified pointer. Submitted by: Alexander Kabaev <ak03@gte.com> Reviewed by: phk Pointy hat to: mux
2002-04-24 17:49:21 +00:00
freeenv(env);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (1);
Rework the kernel environment subsystem. We now convert the static environment needed at boot time to a dynamic subsystem when VM is up. The dynamic kernel environment is protected by an sx lock. This adds some new functions to manipulate the kernel environment : freeenv(), setenv(), unsetenv() and testenv(). freeenv() has to be called after every getenv() when you have finished using the string. testenv() only tests if an environment variable is present, and doesn't require a freeenv() call. setenv() and unsetenv() are self explanatory. The kenv(2) syscall exports these new functionalities to userland, mainly for kenv(1). Reviewed by: peter
2002-04-17 13:06:36 +00:00
}
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
/* Scan the disable list, checking for a match. */
Don't call freeenv() on a modified pointer. Submitted by: Alexander Kabaev <ak03@gte.com> Reviewed by: phk Pointy hat to: mux
2002-04-24 17:49:21 +00:00
cp = env;
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
for (;;) {
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
while (*cp != '\0' && isspace(*cp))
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
cp++;
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
if (*cp == '\0')
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
break;
len = 0;
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
while (cp[len] != '\0' && !isspace(cp[len]))
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
len++;
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
if (strncmp(cp, subsys, len) == 0) {
Don't call freeenv() on a modified pointer. Submitted by: Alexander Kabaev <ak03@gte.com> Reviewed by: phk Pointy hat to: mux
2002-04-24 17:49:21 +00:00
freeenv(env);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (1);
Rework the kernel environment subsystem. We now convert the static environment needed at boot time to a dynamic subsystem when VM is up. The dynamic kernel environment is protected by an sx lock. This adds some new functions to manipulate the kernel environment : freeenv(), setenv(), unsetenv() and testenv(). freeenv() has to be called after every getenv() when you have finished using the string. testenv() only tests if an environment variable is present, and doesn't require a freeenv() call. setenv() and unsetenv() are self explanatory. The kenv(2) syscall exports these new functionalities to userland, mainly for kenv(1). Reviewed by: peter
2002-04-17 13:06:36 +00:00
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
cp += len;
}
Don't call freeenv() on a modified pointer. Submitted by: Alexander Kabaev <ak03@gte.com> Reviewed by: phk Pointy hat to: mux
2002-04-24 17:49:21 +00:00
freeenv(env);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (0);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
Add a new device control utility for new-bus devices called devctl. This allows the user to request administrative changes to individual devices such as attach or detaching drivers or disabling and re-enabling devices. - Add a new /dev/devctl2 character device which uses ioctls for device requests. The ioctls use a common 'struct devreq' which is somewhat similar to 'struct ifreq'. - The ioctls identify the device to operate on via a string. This string can either by the device's name, or it can be a bus-specific address. (For unattached devices, a bus address is the only way to locate a device.) Bus drivers register an eventhandler to claim unrecognized device names that the driver recognizes as a valid address. Two buses currently support addresses: ACPI recognizes any device in the ACPI namespace via its full path starting with "\" and the PCI bus driver recognizes an address specification of 'pci[<domain>:]<bus>:<slot>:<func>' (identical to the PCI selector strings supported by pciconf). - To make it easier to cut and paste, change the PnP location string in the PCI bus driver to output a full PCI selector string rather than 'slot=<slot> function=<func>'. - Add a devctl(3) interface in libdevctl which provides a wrapper around the ioctls and is the preferred interface for other userland code. - Add a devctl(8) program which is a simple wrapper around the requests supported by devctl(3). - Add a device_is_suspended() function to check DF_SUSPENDED. - Add a resource_unset_value() function that can be used to remove a hint from the kernel environment. This is used to clear a hint.<driver>.<unit>.disabled hint when re-enabling a boot-time disabled device. Reviewed by: imp (parts) Requested by: imp (changing PCI location string) Relnotes: yes
2015-02-06 16:09:01 +00:00
static void
acpi_lookup(void *arg, const char *name, device_t *dev)
{
ACPI_HANDLE handle;
if (*dev != NULL)
return;
/*
* Allow any handle name that is specified as an absolute path and
* starts with '\'. We could restrict this to \_SB and friends,
* but see acpi_probe_children() for notes on why we scan the entire
* namespace for devices.
*
* XXX: The pathname argument to AcpiGetHandle() should be fixed to
* be const.
*/
if (name[0] != '\\')
return;
if (ACPI_FAILURE(AcpiGetHandle(ACPI_ROOT_OBJECT, __DECONST(char *, name),
&handle)))
return;
*dev = acpi_get_device(handle);
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
/*
* Control interface.
*
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
* We multiplex ioctls for all participating ACPI devices here. Individual
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
* drivers wanting to be accessible via /dev/acpi should use the
* register/deregister interface to make their handlers visible.
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
*/
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
struct acpi_ioctl_hook
{
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
TAILQ_ENTRY(acpi_ioctl_hook) link;
u_long cmd;
acpi_ioctl_fn fn;
void *arg;
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
};
static TAILQ_HEAD(,acpi_ioctl_hook) acpi_ioctl_hooks;
static int acpi_ioctl_hooks_initted;
int
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
acpi_register_ioctl(u_long cmd, acpi_ioctl_fn fn, void *arg)
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
{
struct acpi_ioctl_hook *hp;
if ((hp = malloc(sizeof(*hp), M_ACPIDEV, M_NOWAIT)) == NULL)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (ENOMEM);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
hp->cmd = cmd;
hp->fn = fn;
hp->arg = arg;
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
ACPI_LOCK(acpi);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
if (acpi_ioctl_hooks_initted == 0) {
TAILQ_INIT(&acpi_ioctl_hooks);
acpi_ioctl_hooks_initted = 1;
}
TAILQ_INSERT_TAIL(&acpi_ioctl_hooks, hp, link);
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
ACPI_UNLOCK(acpi);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (0);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
}
Remove trailing whitespace.
2004-12-27 05:36:47 +00:00
void
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
acpi_deregister_ioctl(u_long cmd, acpi_ioctl_fn fn)
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
{
struct acpi_ioctl_hook *hp;
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
ACPI_LOCK(acpi);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
TAILQ_FOREACH(hp, &acpi_ioctl_hooks, link)
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
if (hp->cmd == cmd && hp->fn == fn)
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
break;
if (hp != NULL) {
TAILQ_REMOVE(&acpi_ioctl_hooks, hp, link);
free(hp, M_ACPIDEV);
}
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
ACPI_UNLOCK(acpi);
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
static int
We no longer need to use d_thread_t, migrate to struct thread *.
2009-05-20 17:29:21 +00:00
acpiopen(struct cdev *dev, int flag, int fmt, struct thread *td)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (0);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
static int
We no longer need to use d_thread_t, migrate to struct thread *.
2009-05-20 17:29:21 +00:00
acpiclose(struct cdev *dev, int flag, int fmt, struct thread *td)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (0);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
static int
We no longer need to use d_thread_t, migrate to struct thread *.
2009-05-20 17:29:21 +00:00
acpiioctl(struct cdev *dev, u_long cmd, caddr_t addr, int flag, struct thread *td)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
struct acpi_softc *sc;
struct acpi_ioctl_hook *hp;
/tmp/m
2004-08-03 05:13:56 +00:00
int error, state;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
Initialize variables to fix kernel build on AMD64.
2004-08-03 14:14:53 +00:00
error = 0;
hp = NULL;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
sc = dev->si_drv1;
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
/*
* Scan the list of registered ioctls, looking for handlers.
*/
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
ACPI_LOCK(acpi);
/tmp/m
2004-08-03 05:13:56 +00:00
if (acpi_ioctl_hooks_initted)
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
TAILQ_FOREACH(hp, &acpi_ioctl_hooks, link) {
/tmp/m
2004-08-03 05:13:56 +00:00
if (hp->cmd == cmd)
break;
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
}
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
ACPI_UNLOCK(acpi);
/tmp/m
2004-08-03 05:13:56 +00:00
if (hp)
return (hp->fn(cmd, addr, hp->arg));
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Allow non-privilaged user to retrive battery or AC line information. Reviewed by: rwatson
2003-02-15 01:46:22 +00:00
/*
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
* Core ioctls are not permitted for non-writable user.
Allow non-privilaged user to retrive battery or AC line information. Reviewed by: rwatson
2003-02-15 01:46:22 +00:00
* Currently, other ioctls just fetch information.
* Not changing system behavior.
*/
Add implementation of the ACPI methods which hands them off to ACPI-CA. acpi_id_probe() returns NULL for no match or the ID string that matched if the driver should attach.
2004-06-29 19:00:36 +00:00
if ((flag & FWRITE) == 0)
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (EPERM);
Allow non-privilaged user to retrive battery or AC line information. Reviewed by: rwatson
2003-02-15 01:46:22 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
/* Core system ioctls. */
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
switch (cmd) {
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
case ACPIIO_REQSLPSTATE:
state = *(int *)addr;
if (state != ACPI_STATE_S5)
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
return (acpi_ReqSleepState(sc, state));
device_printf(sc->acpi_dev, "power off via acpi ioctl not supported\n");
error = EOPNOTSUPP;
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
break;
case ACPIIO_ACKSLPSTATE:
error = *(int *)addr;
error = acpi_AckSleepState(sc->acpi_clone, error);
break;
case ACPIIO_SETSLPSTATE: /* DEPRECATED */
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
state = *(int *)addr;
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
if (state < ACPI_STATE_S0 || state > ACPI_S_STATES_MAX)
return (EINVAL);
if (!acpi_sleep_states[state])
return (EOPNOTSUPP);
if (ACPI_FAILURE(acpi_SetSleepState(sc, state)))
error = ENXIO;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
break;
default:
/tmp/m
2004-08-03 05:13:56 +00:00
error = ENXIO;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
break;
}
Add ioctls to acpi_cmbat and acpi_acad. These use mike's acpi_register_ioctl(). Fix wrong AML method calling in acpi_cmbat.
2000-12-24 19:12:10 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (error);
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
static int
acpi_sname2sstate(const char *sname)
{
int sstate;
if (toupper(sname[0]) == 'S') {
sstate = sname[1] - '0';
if (sstate >= ACPI_STATE_S0 && sstate <= ACPI_STATE_S5 &&
sname[2] == '\0')
return (sstate);
} else if (strcasecmp(sname, "NONE") == 0)
return (ACPI_STATE_UNKNOWN);
return (-1);
}
Fix style(9).
2009-04-30 17:45:43 +00:00
static const char *
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
acpi_sstate2sname(int sstate)
{
static const char *snames[] = { "S0", "S1", "S2", "S3", "S4", "S5" };
if (sstate >= ACPI_STATE_S0 && sstate <= ACPI_STATE_S5)
return (snames[sstate]);
else if (sstate == ACPI_STATE_UNKNOWN)
return ("NONE");
return (NULL);
}
Add new sysctl MIB (hw.acpi.supported_sleep_state) to indicate the list of supported sleep state. This should help people understand what following message means. acpi0: AcpiGetSleepTypeData failed - AE_NOT_FOUND MFC after: 3 days
2003-04-11 16:53:56 +00:00
static int
acpi_supported_sleep_state_sysctl(SYSCTL_HANDLER_ARGS)
{
int error;
/tmp/m
2004-08-03 05:13:56 +00:00
struct sbuf sb;
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
UINT8 state;
Add new sysctl MIB (hw.acpi.supported_sleep_state) to indicate the list of supported sleep state. This should help people understand what following message means. acpi0: AcpiGetSleepTypeData failed - AE_NOT_FOUND MFC after: 3 days
2003-04-11 16:53:56 +00:00
/tmp/m
2004-08-03 05:13:56 +00:00
sbuf_new(&sb, NULL, 32, SBUF_AUTOEXTEND);
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
for (state = ACPI_STATE_S1; state < ACPI_S_STATE_COUNT; state++)
if (acpi_sleep_states[state])
sbuf_printf(&sb, "%s ", acpi_sstate2sname(state));
/tmp/m
2004-08-03 05:13:56 +00:00
sbuf_trim(&sb);
sbuf_finish(&sb);
error = sysctl_handle_string(oidp, sbuf_data(&sb), sbuf_len(&sb), req);
sbuf_delete(&sb);
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (error);
Add new sysctl MIB (hw.acpi.supported_sleep_state) to indicate the list of supported sleep state. This should help people understand what following message means. acpi0: AcpiGetSleepTypeData failed - AE_NOT_FOUND MFC after: 3 days
2003-04-11 16:53:56 +00:00
}
Add 3 new dynamic sysctl's to control the sleep states switched to on a power button, sleep button, or lid close event. The sysctl's use the ACPI sleep state names S0, S1, S2, S3, S4, S4B, and S5. Reviewed by: iwasaki
2001-01-13 21:28:57 +00:00
static int
acpi_sleep_state_sysctl(SYSCTL_HANDLER_ARGS)
{
char sleep_state[10];
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
int error, new_state, old_state;
Add 3 new dynamic sysctl's to control the sleep states switched to on a power button, sleep button, or lid close event. The sysctl's use the ACPI sleep state names S0, S1, S2, S3, S4, S4B, and S5. Reviewed by: iwasaki
2001-01-13 21:28:57 +00:00
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
old_state = *(int *)oidp->oid_arg1;
strlcpy(sleep_state, acpi_sstate2sname(old_state), sizeof(sleep_state));
Add 3 new dynamic sysctl's to control the sleep states switched to on a power button, sleep button, or lid close event. The sysctl's use the ACPI sleep state names S0, S1, S2, S3, S4, S4B, and S5. Reviewed by: iwasaki
2001-01-13 21:28:57 +00:00
error = sysctl_handle_string(oidp, sleep_state, sizeof(sleep_state), req);
if (error == 0 && req->newptr != NULL) {
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
new_state = acpi_sname2sstate(sleep_state);
if (new_state < ACPI_STATE_S1)
return (EINVAL);
Fix off-by-one bug. S5 state must be checked as well.
2009-04-30 18:00:53 +00:00
if (new_state < ACPI_S_STATE_COUNT && !acpi_sleep_states[new_state])
General sleep state change clean up. - Probe supported sleep states from acpi_attach() just once and do not call AcpiGetSleepTypeData() again. It is redundant because AcpiEnterSleepStatePrep() does it any way. - Treat UNKNOWN sleep state as NONE, i.e., "do nothing", and remove obscure NONE state (ACPI_S_STATES_MAX + 1) to avoid confusions. - Do not set unsupported sleep states as default button/switch events. If the default sleep state is not supported, just set it as UNKNOWN/NONE. - Do not allow sleep state change if the system is not fully up and running. This should prevent entering S5 state multiple times, which causes strange behaviours later. - Make sleep states case-insensitive when they are used with sysctl(8). For example, sysctl hw.acpi.lid_switch_state=s1 sysctl hw.acpi.sleep_button_state=none are now legal and equivalent to the uppercase ones.
2009-04-30 17:35:44 +00:00
return (EOPNOTSUPP);
if (new_state != old_state)
*(int *)oidp->oid_arg1 = new_state;
Add 3 new dynamic sysctl's to control the sleep states switched to on a power button, sleep button, or lid close event. The sysctl's use the ACPI sleep state names S0, S1, S2, S3, S4, S4B, and S5. Reviewed by: iwasaki
2001-01-13 21:28:57 +00:00
}
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
return (error);
Add 3 new dynamic sysctl's to control the sleep states switched to on a power button, sleep button, or lid close event. The sysctl's use the ACPI sleep state names S0, S1, S2, S3, S4, S4B, and S5. Reviewed by: iwasaki
2001-01-13 21:28:57 +00:00
}
Add devctl(4) notify support to ACPI. Various subsystems now notify userland whenever events occur. See the example in devd.conf below to see how to use it.
2003-10-25 05:03:25 +00:00
/* Inform devctl(4) when we receive a Notify. */
void
acpi_UserNotify(const char *subsystem, ACPI_HANDLE h, uint8_t notify)
{
char notify_buf[16];
ACPI_BUFFER handle_buf;
ACPI_STATUS status;
if (subsystem == NULL)
return;
handle_buf.Pointer = NULL;
handle_buf.Length = ACPI_ALLOCATE_BUFFER;
Catch up with ACPICA 20150717.
2015-07-22 16:26:17 +00:00
status = AcpiNsHandleToPathname(h, &handle_buf, FALSE);
Add devctl(4) notify support to ACPI. Various subsystems now notify userland whenever events occur. See the example in devd.conf below to see how to use it.
2003-10-25 05:03:25 +00:00
if (ACPI_FAILURE(status))
return;
snprintf(notify_buf, sizeof(notify_buf), "notify=0x%02x", notify);
devctl_notify("ACPI", subsystem, handle_buf.Pointer, notify_buf);
AcpiOsFree(handle_buf.Pointer);
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
#ifdef ACPI_DEBUG
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
/*
* Support for parsing debug options from the kernel environment.
*
* Bits may be set in the AcpiDbgLayer and AcpiDbgLevel debug registers
* by specifying the names of the bits in the debug.acpi.layer and
* debug.acpi.level environment variables. Bits may be unset by
* prefixing the bit name with !.
*/
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
struct debugtag
{
char *name;
UINT32 value;
};
static struct debugtag dbg_layer[] = {
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
{"ACPI_UTILITIES", ACPI_UTILITIES},
{"ACPI_HARDWARE", ACPI_HARDWARE},
{"ACPI_EVENTS", ACPI_EVENTS},
{"ACPI_TABLES", ACPI_TABLES},
{"ACPI_NAMESPACE", ACPI_NAMESPACE},
{"ACPI_PARSER", ACPI_PARSER},
{"ACPI_DISPATCHER", ACPI_DISPATCHER},
{"ACPI_EXECUTER", ACPI_EXECUTER},
{"ACPI_RESOURCES", ACPI_RESOURCES},
Resolve conflicts arising from the ACPI CA 20020815 import.
2002-08-29 01:52:27 +00:00
{"ACPI_CA_DEBUGGER", ACPI_CA_DEBUGGER},
Updates to match the ACPI CA 20010816 import: - New debug macro (ACPI_DEBUG_PRINT), reducing debug-case code size. - New debug level/subsystem codes.
2001-08-26 22:50:15 +00:00
{"ACPI_OS_SERVICES", ACPI_OS_SERVICES},
Resolve conflicts arising from the ACPI CA 20020815 import.
2002-08-29 01:52:27 +00:00
{"ACPI_CA_DISASSEMBLER", ACPI_CA_DISASSEMBLER},
Sort debugging levels and update the man page to match reality. Also update man page to reflect iasl(8) import.
2003-09-26 04:32:40 +00:00
{"ACPI_ALL_COMPONENTS", ACPI_ALL_COMPONENTS},
Updates to match the ACPI CA 20010816 import: - New debug macro (ACPI_DEBUG_PRINT), reducing debug-case code size. - New debug level/subsystem codes.
2001-08-26 22:50:15 +00:00
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
{"ACPI_AC_ADAPTER", ACPI_AC_ADAPTER},
{"ACPI_BATTERY", ACPI_BATTERY},
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
{"ACPI_BUS", ACPI_BUS},
Sync to my work in progress: - Reorder the acpi_* functions in a sensible fashion - Add acpi_ForeachPackageObject and acpi_GetHandleInScope - Use the new debugging layer/level names - Implement most of the guts of the acpi_thermal module; passive cooling isn't there yet, but active cooling should work. - Implement power resource handling (acpi_powerres.c) This compiles and mostly works, but my test coverage is small, so feedback is welcome.
2001-06-28 06:17:16 +00:00
{"ACPI_BUTTON", ACPI_BUTTON},
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
{"ACPI_EC", ACPI_EC},
{"ACPI_FAN", ACPI_FAN},
{"ACPI_POWERRES", ACPI_POWERRES},
Updates to match the ACPI CA 20010816 import: - New debug macro (ACPI_DEBUG_PRINT), reducing debug-case code size. - New debug level/subsystem codes.
2001-08-26 22:50:15 +00:00
{"ACPI_PROCESSOR", ACPI_PROCESSOR},
Update to synch with the 20010615 ACPI CA import. Add an ACPI subsystem mutex, and macros for handling it. Because it's not possible to differentiate between ACPI CA acquiring mutexes for internal use and for use by AML, and because AML in the field doesn't handle mutexes correctly, we can't use the ACPI subsystem's internal locking. In addition, we have other private data of our own to lock. Add initial locking to the ACPI driver code and the thermal module. These locks are currently inoperative. Pull some errant style back into line.
2001-06-29 20:29:59 +00:00
{"ACPI_THERMAL", ACPI_THERMAL},
Add support for quirks for acpi tables. Key off OEM vendor and revision. Sort acpi debug values. Change "disable" to "disabled" to match rest of the kernel. Remove debugging from acpi_toshiba since it was only used for probe/attach.
2004-03-03 03:02:17 +00:00
{"ACPI_TIMER", ACPI_TIMER},
Match namespace cleanup changes in ACPI CA 20020217 update. Use ACPI_SUCCESS/ACPI_FAILURE consistently. The AcpiGetInto* interfaces are obsoleted by ACPI_ALLOCATE_BUFFER. Add AcpiBatteryIsPresent helper to determine whether a battery device is inserted. Add ACPI_ALL_DRIVERS to the list of debug layers, now that we own the namespace for this. Pr:
2002-02-23 05:21:56 +00:00
{"ACPI_ALL_DRIVERS", ACPI_ALL_DRIVERS},
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{NULL, 0}
};
static struct debugtag dbg_level[] = {
Catch up to reworked debugging levels in latest Intel import.
2003-05-01 18:51:43 +00:00
{"ACPI_LV_INIT", ACPI_LV_INIT},
Updates to match the ACPI CA 20010816 import: - New debug macro (ACPI_DEBUG_PRINT), reducing debug-case code size. - New debug level/subsystem codes.
2001-08-26 22:50:15 +00:00
{"ACPI_LV_DEBUG_OBJECT", ACPI_LV_DEBUG_OBJECT},
Catch up to reworked debugging levels in latest Intel import.
2003-05-01 18:51:43 +00:00
{"ACPI_LV_INFO", ACPI_LV_INFO},
Add ACPI_LV_REPAIR debug level, available since ACPICA 20091214 (r200553). MFC after: 3 days
2012-03-22 17:01:25 +00:00
{"ACPI_LV_REPAIR", ACPI_LV_REPAIR},
Updates to match the ACPI CA 20010816 import: - New debug macro (ACPI_DEBUG_PRINT), reducing debug-case code size. - New debug level/subsystem codes.
2001-08-26 22:50:15 +00:00
{"ACPI_LV_ALL_EXCEPTIONS", ACPI_LV_ALL_EXCEPTIONS},
Resolve conflicts arising from the ACPI CA 20020815 import.
2002-08-29 01:52:27 +00:00
/* Trace verbosity level 1 [Standard Trace Level] */
Sort debugging levels and update the man page to match reality. Also update man page to reflect iasl(8) import.
2003-09-26 04:32:40 +00:00
{"ACPI_LV_INIT_NAMES", ACPI_LV_INIT_NAMES},
Updates to match the ACPI CA 20010816 import: - New debug macro (ACPI_DEBUG_PRINT), reducing debug-case code size. - New debug level/subsystem codes.
2001-08-26 22:50:15 +00:00
{"ACPI_LV_PARSE", ACPI_LV_PARSE},
{"ACPI_LV_LOAD", ACPI_LV_LOAD},
Resolve conflicts arising from the ACPI CA 20020815 import.
2002-08-29 01:52:27 +00:00
{"ACPI_LV_DISPATCH", ACPI_LV_DISPATCH},
Updates to match the ACPI CA 20010816 import: - New debug macro (ACPI_DEBUG_PRINT), reducing debug-case code size. - New debug level/subsystem codes.
2001-08-26 22:50:15 +00:00
{"ACPI_LV_EXEC", ACPI_LV_EXEC},
{"ACPI_LV_NAMES", ACPI_LV_NAMES},
{"ACPI_LV_OPREGION", ACPI_LV_OPREGION},
{"ACPI_LV_BFIELD", ACPI_LV_BFIELD},
{"ACPI_LV_TABLES", ACPI_LV_TABLES},
{"ACPI_LV_VALUES", ACPI_LV_VALUES},
{"ACPI_LV_OBJECTS", ACPI_LV_OBJECTS},
{"ACPI_LV_RESOURCES", ACPI_LV_RESOURCES},
{"ACPI_LV_USER_REQUESTS", ACPI_LV_USER_REQUESTS},
{"ACPI_LV_PACKAGE", ACPI_LV_PACKAGE},
Resolve conflicts arising from the ACPI CA 20020815 import.
2002-08-29 01:52:27 +00:00
{"ACPI_LV_VERBOSITY1", ACPI_LV_VERBOSITY1},
/* Trace verbosity level 2 [Function tracing and memory allocation] */
{"ACPI_LV_ALLOCATIONS", ACPI_LV_ALLOCATIONS},
{"ACPI_LV_FUNCTIONS", ACPI_LV_FUNCTIONS},
{"ACPI_LV_OPTIMIZATIONS", ACPI_LV_OPTIMIZATIONS},
{"ACPI_LV_VERBOSITY2", ACPI_LV_VERBOSITY2},
Updates to match the ACPI CA 20010816 import: - New debug macro (ACPI_DEBUG_PRINT), reducing debug-case code size. - New debug level/subsystem codes.
2001-08-26 22:50:15 +00:00
{"ACPI_LV_ALL", ACPI_LV_ALL},
Resolve conflicts arising from the ACPI CA 20020815 import.
2002-08-29 01:52:27 +00:00
/* Trace verbosity level 3 [Threading, I/O, and Interrupts] */
{"ACPI_LV_MUTEX", ACPI_LV_MUTEX},
{"ACPI_LV_THREADS", ACPI_LV_THREADS},
{"ACPI_LV_IO", ACPI_LV_IO},
{"ACPI_LV_INTERRUPTS", ACPI_LV_INTERRUPTS},
{"ACPI_LV_VERBOSITY3", ACPI_LV_VERBOSITY3},
/* Exceptionally verbose output -- also used in the global "DebugLevel" */
Resolve conflicts arising from the ACPI CA 20020611 import.
2002-07-09 17:54:02 +00:00
{"ACPI_LV_AML_DISASSEMBLE", ACPI_LV_AML_DISASSEMBLE},
{"ACPI_LV_VERBOSE_INFO", ACPI_LV_VERBOSE_INFO},
{"ACPI_LV_FULL_TABLES", ACPI_LV_FULL_TABLES},
{"ACPI_LV_EVENTS", ACPI_LV_EVENTS},
{"ACPI_LV_VERBOSE", ACPI_LV_VERBOSE},
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{NULL, 0}
};
static void
acpi_parse_debug(char *cp, struct debugtag *tag, UINT32 *flag)
{
char *ep;
int i, l;
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
int set;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
while (*cp) {
if (isspace(*cp)) {
cp++;
continue;
}
ep = cp;
while (*ep && !isspace(*ep))
ep++;
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
if (*cp == '!') {
set = 0;
cp++;
if (cp == ep)
continue;
} else {
set = 1;
}
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
l = ep - cp;
for (i = 0; tag[i].name != NULL; i++) {
if (!strncmp(cp, tag[i].name, l)) {
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
if (set)
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
*flag |= tag[i].value;
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
else
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
*flag &= ~tag[i].value;
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
}
cp = ep;
}
}
static void
- Updates for new constant naming in the ACPI CA 20010518 update. - Use __func__ instead of __FUNCTION. - Support power-off to S3 or S5 (takawata) - Enable ACPI debugging earlier (with a sysinit) - Fix a deadlock in the EC code (takawata) - Improve arithmetic and reduce the risk of spurious wakeup in AcpiOsSleep. - Add AcpiOsGetThreadId. - Simplify mutex code (still disabled).
2001-05-29 20:13:42 +00:00
acpi_set_debugging(void *junk)
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
{
Fix setting debug strings via sysctl. Also, clean up the way we print debug strings.
2004-05-06 02:05:45 +00:00
char *layer, *level;
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
if (cold) {
AcpiDbgLayer = 0;
AcpiDbgLevel = 0;
}
Add `if (!cold)' checkings for functions which is called via SYSINIT. Loading acpi.ko with kldload is disallowed, however some functions were executed unexpectedly. Approved by: re
2002-11-24 02:27:07 +00:00
Follow up to r225617. In order to maximize the re-usability of kernel code in userland rename in-kernel getenv()/setenv() to kern_setenv()/kern_getenv(). This fixes a namespace collision with libc symbols. Submitted by: kmacy Tested by: make universe
2014-10-16 18:04:43 +00:00
layer = kern_getenv("debug.acpi.layer");
level = kern_getenv("debug.acpi.level");
Fix setting debug strings via sysctl. Also, clean up the way we print debug strings.
2004-05-06 02:05:45 +00:00
if (layer == NULL && level == NULL)
return;
- Convert a lot of homebrew debugging output to use the ACPI CA debugging infrastructure. It's not perfect, but it's a lot better than what we've been using so far. The following rules apply to this: o BSD component names should be capitalised o Layer names should be taken from the non-CA set for now. We may elect to add some new BSD-specific layers later. - Make it possible to turn off selective debugging flags or layers by listing them in debug.acpi.layer or debug.acpi.level prefixed with !. - Fully implement support for avoiding nodes in the ACPI namespace. Nodes may be listed in the debug.acpi.avoid environment variable; these nodes and all their children will be ignored (although still scanned over) by ACPI functions which scan the namespace. Multiple nodes can be specified, separated by whitespace. - Implement support for selectively disabling ACPI subsystem components via the debug.acpi.disable environment variable. The following components can be disabled: o bus creation/scanning of the ACPI 'bus' o children attachment of children to the ACPI 'bus' o button the acpi_button control-method button driver o ec the acpi_ec embedded-controller driver o isa acpi replacement of PnP BIOS for ISA device discovery o lid the control-method lid switch driver o pci pci root-bus discovery o processor CPU power/speed management o thermal system temperature detection and control o timer ACPI timecounter Multiple components may be disabled by specifying their name(s) separated by whitespace. - Add support for ioctl registration. ACPI subsystem components may register ioctl handlers with the /dev/acpi generic ioctl handler, allowing us to avoid the need for a multitude of /dev/acpi* control devices, etc.
2000-12-08 09:16:20 +00:00
Fix setting debug strings via sysctl. Also, clean up the way we print debug strings.
2004-05-06 02:05:45 +00:00
printf("ACPI set debug");
if (layer != NULL) {
if (strcmp("NONE", layer) != 0)
printf(" layer '%s'", layer);
acpi_parse_debug(layer, &dbg_layer[0], &AcpiDbgLayer);
freeenv(layer);
}
if (level != NULL) {
if (strcmp("NONE", level) != 0)
printf(" level '%s'", level);
acpi_parse_debug(level, &dbg_level[0], &AcpiDbgLevel);
freeenv(level);
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
}
Fix setting debug strings via sysctl. Also, clean up the way we print debug strings.
2004-05-06 02:05:45 +00:00
printf("\n");
Initial FreeBSD OSPM (operating system power management) modules for ACPICA. Most of these are still works in progress. Support exists for: - Fixed feature and control method power, lid and sleep buttons. - Detection of ISA PnP devices using ACPI namespace. - Detection of PCI root busses using ACPI namespace. - CPU throttling and sleep states (incomplete) - Thermal monitoring and cooling control (incomplete) - Interface to platform embedded controllers (mostly complete) - ACPI timer (incomplete) - Simple userland control of sleep states. - Shutdown and poweroff.
2000-10-28 06:59:48 +00:00
}
/tmp/m
2004-08-03 05:13:56 +00:00
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
SYSINIT(acpi_debugging, SI_SUB_TUNABLES, SI_ORDER_ANY, acpi_set_debugging,
NULL);
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
static int
acpi_debug_sysctl(SYSCTL_HANDLER_ARGS)
{
Use sbufs instead of snprintf for parsing debug strings.
2003-12-09 06:29:57 +00:00
int error, *dbg;
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
struct debugtag *tag;
Use sbufs instead of snprintf for parsing debug strings.
2003-12-09 06:29:57 +00:00
struct sbuf sb;
The SYSCTL data pointers can come from userspace and must not be directly accessed. Although this will work on some platforms, it can throw an exception if the pointer is invalid and then panic the kernel. Add a missing SYSCTL_IN() of "SCTP_BASE_STATS" structure. MFC after: 3 days Sponsored by: Mellanox Technologies
2014-10-28 12:00:39 +00:00
char temp[128];
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
Use sbufs instead of snprintf for parsing debug strings.
2003-12-09 06:29:57 +00:00
if (sbuf_new(&sb, NULL, 128, SBUF_AUTOEXTEND) == NULL)
return (ENOMEM);
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
if (strcmp(oidp->oid_arg1, "debug.acpi.layer") == 0) {
tag = &dbg_layer[0];
dbg = &AcpiDbgLayer;
} else {
tag = &dbg_level[0];
dbg = &AcpiDbgLevel;
}
/* Get old values if this is a get request. */
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
ACPI_SERIAL_BEGIN(acpi);
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
if (*dbg == 0) {
Use sbufs instead of snprintf for parsing debug strings.
2003-12-09 06:29:57 +00:00
sbuf_cpy(&sb, "NONE");
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
} else if (req->newptr == NULL) {
for (; tag->name != NULL; tag++) {
Use sbufs instead of snprintf for parsing debug strings.
2003-12-09 06:29:57 +00:00
if ((*dbg & tag->value) == tag->value)
sbuf_printf(&sb, "%s ", tag->name);
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
}
}
Use sbufs instead of snprintf for parsing debug strings.
2003-12-09 06:29:57 +00:00
sbuf_trim(&sb);
sbuf_finish(&sb);
The SYSCTL data pointers can come from userspace and must not be directly accessed. Although this will work on some platforms, it can throw an exception if the pointer is invalid and then panic the kernel. Add a missing SYSCTL_IN() of "SCTP_BASE_STATS" structure. MFC after: 3 days Sponsored by: Mellanox Technologies
2014-10-28 12:00:39 +00:00
strlcpy(temp, sbuf_data(&sb), sizeof(temp));
Use sbufs instead of snprintf for parsing debug strings.
2003-12-09 06:29:57 +00:00
sbuf_delete(&sb);
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
The SYSCTL data pointers can come from userspace and must not be directly accessed. Although this will work on some platforms, it can throw an exception if the pointer is invalid and then panic the kernel. Add a missing SYSCTL_IN() of "SCTP_BASE_STATS" structure. MFC after: 3 days Sponsored by: Mellanox Technologies
2014-10-28 12:00:39 +00:00
error = sysctl_handle_string(oidp, temp, sizeof(temp), req);
/* Check for error or no change */
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
if (error == 0 && req->newptr != NULL) {
The SYSCTL data pointers can come from userspace and must not be directly accessed. Although this will work on some platforms, it can throw an exception if the pointer is invalid and then panic the kernel. Add a missing SYSCTL_IN() of "SCTP_BASE_STATS" structure. MFC after: 3 days Sponsored by: Mellanox Technologies
2014-10-28 12:00:39 +00:00
*dbg = 0;
kern_setenv((char *)oidp->oid_arg1, temp);
acpi_set_debugging(NULL);
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
}
MPSAFE locking * Serialize calls to acpi_alloc_resource(), acpi_release_resource(), acpi_Enable(), acpi_Disable(), and acpi_debug_sysctl(). * Acquire the ACPI mutex in acpi_register_ioctl(), acpi_deregister_ioctl(), and acpiioctl(). * Acquire the mutex while disabling subsequent requests to enter a sleep state in acpi_SetSleepState(). * Be sure to re-enable sleep requests and don't run resume methods when the current request fails. * Don't check if sleep requests are disabled in the ACPIIO_SETSLPSTATE ioctl. acpi_SetSleepState() does this for us. * Remove the acquisition of Giant from the struct cdevsw. * Remove the ACPI_USE_THREADS option.
2004-08-13 06:21:32 +00:00
ACPI_SERIAL_END(acpi);
Make debug.acpi.level and debug.acpi.layer sysctls that can be set with the strings found in acpi(4). Also make acpi_ca_version a string so it is more readable.
2003-09-26 21:22:10 +00:00
return (error);
}
/tmp/m
2004-08-03 05:13:56 +00:00
Mark more nodes as CTLFLAG_MPSAFE or CTLFLAG_NEEDGIANT (17 of many) r357614 added CTLFLAG_NEEDGIANT to make it easier to find nodes that are still not MPSAFE (or already are but aren’t properly marked). Use it in preparation for a general review of all nodes. This is non-functional change that adds annotations to SYSCTL_NODE and SYSCTL_PROC nodes using one of the soon-to-be-required flags. Mark all obvious cases as MPSAFE. All entries that haven't been marked as MPSAFE before are by default marked as NEEDGIANT Approved by: kib (mentor, blanket) Commented by: kib, gallatin, melifaro Differential Revision: https://reviews.freebsd.org/D23718
2020-02-26 14:26:36 +00:00
SYSCTL_PROC(_debug_acpi, OID_AUTO, layer,
CTLFLAG_RW | CTLTYPE_STRING | CTLFLAG_NEEDGIANT, "debug.acpi.layer", 0,
acpi_debug_sysctl, "A",
"");
SYSCTL_PROC(_debug_acpi, OID_AUTO, level,
CTLFLAG_RW | CTLTYPE_STRING | CTLFLAG_NEEDGIANT, "debug.acpi.level", 0,
acpi_debug_sysctl, "A",
"");
/tmp/m
2004-08-03 05:13:56 +00:00
#endif /* ACPI_DEBUG */
- Allow users to enable dumping Debug objects without ACPI debugger. Setting the new sysctl MIB "debug.acpi.enable_debug_objects" to a non-zero value enables us to print Debug object when something is written to it. - Allow users to disable interpreter slack mode. Setting the new tunable "debug.acpi.interpreter_slack" to zero disables some workarounds for common BIOS mistakes and enables strict ACPI implementations by the specification.
2010-03-09 19:02:02 +00:00
static int
acpi_debug_objects_sysctl(SYSCTL_HANDLER_ARGS)
{
int error;
int old;
old = acpi_debug_objects;
error = sysctl_handle_int(oidp, &acpi_debug_objects, 0, req);
if (error != 0 || req->newptr == NULL)
return (error);
if (old == acpi_debug_objects || (old && acpi_debug_objects))
return (0);
ACPI_SERIAL_BEGIN(acpi);
AcpiGbl_EnableAmlDebugObject = acpi_debug_objects ? TRUE : FALSE;
ACPI_SERIAL_END(acpi);
return (0);
}
Add two new loader tunables 'hw.acpi.install_interface' and 'hw.acpi.remove_interface'. hw.acpi.install_interface lets you install new interfaces. Conversely, hw.acpi.remove_interface lets you remove OS interfaces from the pre-defined list in ACPICA. For example, hw.acpi.install_interface="FreeBSD" lets _OSI("FreeBSD") method to return 0xffffffff (or success) and hw.acpi.remove_interface="Windows 2009" lets _OSI("Windows 2009") method to return zero (or failure). Both are comma-separated lists and leading white spaces are ignored. For example, the following examples are valid: hw.acpi.install_interface="Linux, FreeBSD" hw.acpi.remove_interface="Windows 2006, Windows 2006.1"
2010-10-26 18:59:50 +00:00
static int
acpi_parse_interfaces(char *str, struct acpi_interface *iface)
{
char *p;
size_t len;
int i, j;
p = str;
while (isspace(*p) || *p == ',')
p++;
len = strlen(p);
if (len == 0)
return (0);
p = strdup(p, M_TEMP);
for (i = 0; i < len; i++)
if (p[i] == ',')
p[i] = '\0';
i = j = 0;
while (i < len)
if (isspace(p[i]) || p[i] == '\0')
i++;
else {
i += strlen(p + i) + 1;
j++;
}
if (j == 0) {
free(p, M_TEMP);
return (0);
}
iface->data = malloc(sizeof(*iface->data) * j, M_TEMP, M_WAITOK);
iface->num = j;
i = j = 0;
while (i < len)
if (isspace(p[i]) || p[i] == '\0')
i++;
else {
iface->data[j] = p + i;
i += strlen(p + i) + 1;
j++;
}
return (j);
}
static void
acpi_free_interfaces(struct acpi_interface *iface)
{
free(iface->data[0], M_TEMP);
free(iface->data, M_TEMP);
}
static void
acpi_reset_interfaces(device_t dev)
{
struct acpi_interface list;
ACPI_STATUS status;
int i;
if (acpi_parse_interfaces(acpi_install_interface, &list) > 0) {
for (i = 0; i < list.num; i++) {
status = AcpiInstallInterface(list.data[i]);
if (ACPI_FAILURE(status))
device_printf(dev,
"failed to install _OSI(\"%s\"): %s\n",
list.data[i], AcpiFormatException(status));
else if (bootverbose)
device_printf(dev, "installed _OSI(\"%s\")\n",
list.data[i]);
}
acpi_free_interfaces(&list);
}
if (acpi_parse_interfaces(acpi_remove_interface, &list) > 0) {
for (i = 0; i < list.num; i++) {
status = AcpiRemoveInterface(list.data[i]);
if (ACPI_FAILURE(status))
device_printf(dev,
"failed to remove _OSI(\"%s\"): %s\n",
list.data[i], AcpiFormatException(status));
else if (bootverbose)
device_printf(dev, "removed _OSI(\"%s\")\n",
list.data[i]);
}
acpi_free_interfaces(&list);
}
}
Some fix for the recent apm module changes. - Now that apm loadable module can inform its existence to other kernel components (e.g. i386/isa/clock.c:startrtclock()'s TCS hack). - Exchange priority of SI_SUB_CPU and SI_SUB_KLD for above purpose. - Add simple arbitration mechanism for APM vs. ACPI. This prevents the kernel enables both of them. - Remove obsolete `#ifdef DEV_APM' related code. - Add abstracted interface for Powermanagement operations. Public apm(4) functions, such as apm_suspend(), should be replaced new interfaces. Currently only power_pm_suspend (successor of apm_suspend) is implemented. Reviewed by: peter, arch@ and audit@
2001-11-01 16:34:07 +00:00
static int
acpi_pm_func(u_long cmd, void *arg, ...)
{
int state, acpi_state;
int error;
struct acpi_softc *sc;
va_list ap;
error = 0;
switch (cmd) {
case POWER_CMD_SUSPEND:
sc = (struct acpi_softc *)arg;
if (sc == NULL) {
error = EINVAL;
goto out;
}
va_start(ap, arg);
state = va_arg(ap, int);
Remove trailing whitespace.
2004-12-27 05:36:47 +00:00
va_end(ap);
Some fix for the recent apm module changes. - Now that apm loadable module can inform its existence to other kernel components (e.g. i386/isa/clock.c:startrtclock()'s TCS hack). - Exchange priority of SI_SUB_CPU and SI_SUB_KLD for above purpose. - Add simple arbitration mechanism for APM vs. ACPI. This prevents the kernel enables both of them. - Remove obsolete `#ifdef DEV_APM' related code. - Add abstracted interface for Powermanagement operations. Public apm(4) functions, such as apm_suspend(), should be replaced new interfaces. Currently only power_pm_suspend (successor of apm_suspend) is implemented. Reviewed by: peter, arch@ and audit@
2001-11-01 16:34:07 +00:00
switch (state) {
case POWER_SLEEP_STATE_STANDBY:
acpi_state = sc->acpi_standby_sx;
break;
case POWER_SLEEP_STATE_SUSPEND:
acpi_state = sc->acpi_suspend_sx;
break;
case POWER_SLEEP_STATE_HIBERNATE:
acpi_state = ACPI_STATE_S4;
break;
default:
error = EINVAL;
goto out;
}
Update the suspend/resume user API while maintaining backwards compat. Improvements: * /etc/rc.suspend,rc.resume are always run, no matter the source of the suspend request (user or kernel, apm or acpi) * suspend now requires positive user acknowledgement. If a user program wants to cancel the suspend, they can. If one of the user programs hangs or doesn't respond within 10 seconds, the system suspends anyway. * /dev/apm is clonable, allowing multiple listeners for suspend events. In the future, xorg-server can use this to be informed about suspend even if there are other listeners (i.e. apmd). Changes: * Two new ACPI ioctls: REQSLPSTATE and ACKSLPSTATE. Request begins the process of suspending by notifying all listeners. acpi is monitored by devd(8) and /dev/apm listener(s) are also counted. Users register their approval or disapproval via Ack. If anyone disapproves, suspend is vetoed. * Old user programs or kernel modules that used SETSLPSTATE continue to work. A message is printed once that this interface is deprecated. * acpiconf gains the -k flag to ack the suspend request. This flag is undocumented on purpose since it's only used by /etc/rc.suspend. It is not intended to be a permanent change and will be removed once a better power API is implemented. * S5 (power off) is no longer supported via acpiconf -s 5 or apm -z/-Z. This restores previous behavior of halt/shutdown -p being the interface. * Miscellaneous improvements to error reporting Approved by: re
2007-06-21 22:50:37 +00:00
if (ACPI_FAILURE(acpi_EnterSleepState(sc, acpi_state)))
error = ENXIO;
Some fix for the recent apm module changes. - Now that apm loadable module can inform its existence to other kernel components (e.g. i386/isa/clock.c:startrtclock()'s TCS hack). - Exchange priority of SI_SUB_CPU and SI_SUB_KLD for above purpose. - Add simple arbitration mechanism for APM vs. ACPI. This prevents the kernel enables both of them. - Remove obsolete `#ifdef DEV_APM' related code. - Add abstracted interface for Powermanagement operations. Public apm(4) functions, such as apm_suspend(), should be replaced new interfaces. Currently only power_pm_suspend (successor of apm_suspend) is implemented. Reviewed by: peter, arch@ and audit@
2001-11-01 16:34:07 +00:00
break;
default:
error = EINVAL;
goto out;
}
out:
return (error);
}
static void
acpi_pm_register(void *arg)
{
Style and whitespace changes. Also, make the ivar functions non-inline since inlining failed due to the size of BUS_*
2003-08-28 16:06:30 +00:00
if (!cold || resource_disabled("acpi", 0))
Add `if (!cold)' checkings for functions which is called via SYSINIT. Loading acpi.ko with kldload is disallowed, however some functions were executed unexpectedly. Approved by: re
2002-11-24 02:27:07 +00:00
return;
s/hint.acpi.0.disable/hint.acpi.0.disabled/ Fix device hints entry for disabling acpi(4). This also should fix the arbitration with apm(4) when both drivers are enabled. Note that your /boot/device.hints needs to be updated if you want to stop auto-loading acpi.ko or disable acpi(4).
2002-08-30 11:11:07 +00:00
Add `if (!cold)' checkings for functions which is called via SYSINIT. Loading acpi.ko with kldload is disallowed, however some functions were executed unexpectedly. Approved by: re
2002-11-24 02:27:07 +00:00
power_pm_register(POWER_PM_TYPE_ACPI, acpi_pm_func, NULL);
Some fix for the recent apm module changes. - Now that apm loadable module can inform its existence to other kernel components (e.g. i386/isa/clock.c:startrtclock()'s TCS hack). - Exchange priority of SI_SUB_CPU and SI_SUB_KLD for above purpose. - Add simple arbitration mechanism for APM vs. ACPI. This prevents the kernel enables both of them. - Remove obsolete `#ifdef DEV_APM' related code. - Add abstracted interface for Powermanagement operations. Public apm(4) functions, such as apm_suspend(), should be replaced new interfaces. Currently only power_pm_suspend (successor of apm_suspend) is implemented. Reviewed by: peter, arch@ and audit@
2001-11-01 16:34:07 +00:00
}
Use NULL for SYSINIT's last arg, which is a pointer type Sponsored by: The FreeBSD Foundation
2018-05-18 17:58:09 +00:00
SYSINIT(power, SI_SUB_KLD, SI_ORDER_ANY, acpi_pm_register, NULL);
Reference in New Issue Copy Permalink